Transcriptional and post-transcriptional regulation of the CIT1 gene in Saccharomyces cerevisiae


Material Information

Transcriptional and post-transcriptional regulation of the CIT1 gene in Saccharomyces cerevisiae
Physical Description:
Lawson, Sobomabo D
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 50281687
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
    Materials and methods
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
    Summary and discussion
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
    Biographical sketch
        Page 182
        Page 183
        Page 184
Full Text







I dedicate this work to my family, especially to my wife and best friend,
Alaro Lawson. Without her love and support it would have been impossible for
me to complete this work. To my children, Banimi and Emi, who had to deal with
my many absences. Finally to my parents, who laid the foundation for my


I wish to thank the University of Florida to have given me the opportunity

to attend graduate school.

A special thanks goes to Dr. Alfred S. Lewin for allowing me to work in his

laboratory, but more important than that was his genuine care and concern for

me and all other members of the lab. His mentorship has been invaluable and

shall remain part of my scientific career.

I thank all the members of my committee for accepting that duty and giving

me the guidance to improve my ability. I would like to pay special thanks to Dr.

Henry V. Baker, who first taught me a lot about yeast, for the many innumerable

ways he has helped me over the years.

To the members of Lewin's laboratory, I thank you all for making my stay

there a little less tedious. To Mr. James Thomas Jr., I thank you for all the help

you have given me. To Dr. Lynn C. Shaw, the Macintosh specialist, without your

help it would have been impossible to get all my figures ready in the last days.

To Mr. Bruce W. Ritching, I thank you for help editing part of the dissertation.

To my family, especially my wife Alaro Lawson, without you this would not

have been possible.


ACKNOWLEDGMENTS ...................

ABSTRACT .............................

INTRODUCTION ........................

Utility of Baker's Yeast ...............
The Citrate Synthase System .........
Transcriptional Regulation in Eukaryotes
Glucose Repression ................
m RNA Stability .....................


. . . . . . . . . . . . iii

. . . . . v i

. . . . . 1

(rrnwfh r.nnrlitinn. nnrl Marlin

Yeast Transformation
Construction of 5' (DIS
Construction of InternE
Heterologous Fusion

mw ~ ~ n . . . . . . . . . . . . . . . .

TAL) and 3' (PROXIMAL) Deletions .........
Il D e letions .............................

Cloning of Oligonucleotides .................................. 38
Measurement of 13-Galactosidase Level in CITI-lacZ Fusion ........ 39
R N A Isolation ............................................. 40
Ribonuclease Protection Assay ............................... 42
Northern A analysis ......................................... 44
Primer Extension Analysis ................................... 46
In Vivo Footprinting ........................................ 47
Preparation of Single-Stranded DNA ........................... 51
Bandshift Assay/In Vitro Footprinting Analysis ................... 52
Messenger RNA Stability (5' UTR deletion) Assay ................ 55
Introduction of Stop Codon at the Fifth Amino Acid Position in the
C IT 1 G ene ............................................... 56

R E S U LT S ..................................................... 63

Analysis of 5' (Distal) and 3' (Proximal) Deletions ................. 63
Internal Deletions Show Several Putative UASs .................. 73
There are Multiple UAS Elements ............................. 77
Evidence for URS Element .................................. 84
Steady-State mRNA Levels Correlate with Enzyme Assay .......... 88


Band Shift Assay and In Vitro Footprint Analysis .................. 91
In Vivo Footprint Analysis .................................. 103
CIT1 mRNA is More Stable in Cells Grown in Ethanol Than in
Cells Grown in Glucose ..................... .......... 108
CITI::IacZ Fusion mRNA Has a Similar Decay Rate As Full-Length CIT1
...... ...................................... 118
CIT1 mRNA From Cells Grown in YPD and YPE Media Have
Identical 5' Mature Ends ................................... 127
The Glucose-Dependent Instability Element Lies Within the CIT1
C oding Region ........................................... 130
Sequences Within the 5' Terminus of CIT1 mRNA Confer
Nonsense-Mediated Decay ................................. 136

SUMMARY AND DISCUSSION ................................... 140

Cis-acting Elem ents ....................................... 142
Nutrient Requirement on the Expression of CIT1 ................ 151
HAP2/HAP3/HAP4 Independent Expression of CIT1 ............. 153
m R NA Stability ........................................... 157
Future G oals ............................................ 164

BIBLIO G RA PHY ............................................... 167

BIOGRAPHICAL SKETCH ....................................... 182

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Sobomabo D. Lawson

May, 1995

Chairperson: Dr. Alfred S. Lewin
Major Department: Molecular Genetics and Microbiology

Citrate synthase catalyzes the condensation of oxaloacetate with acetyl-

CoA to form citrate, a reaction that takes place in both the tricarboxylic acid cycle

and the glyoxylate pathway. The tricarboxylic acid cycle occurs in the inner

compartment of the mitochondrion. In yeast Saccharomyces cerevisiae,

mitochondrial citrate synthase is encoded by CIT1, a nuclear gene.

This gene, like most other genes encoding mitochondrial enzymes, is

severely repressed when yeast is grown in the presence of glucose. These

genes become fully derepressed when glucose is totally utilized or when cells are

grown in ethanol, a nonfermentable carbon source. To understand at what stage

this regulation occurs, I examined transcriptional and post-transcriptional control

of CIT1. Deletions of the 5' non-coding region of the gene, in a CITI::IacZ hybrid,

identified several upstream activating sequences (UASs) that positively affected

the transcription of the gene. Near the distal (5') end of the insert there is also an

upstream repressing sequence (URS) that affects the transcription. Another

URS element was found at the proximal end of the 5' non-coding region that

affected transcription only in glucose medium. These results suggest a

combinatorial regulation of CIT1 by different factors in a carbon source-

dependent and -independent manner. A second level of carbon source-

dependent control was the regulation of the stability of the mRNA in the different

growth media. In ethanol medium, the half-life of CIT1 mRNA was more than

twice that in glucose medium. Deletion analysis showed that the first 78

nucleotides of the 5' coding region of the mRNA contain a sequence element that

is necessary and sufficient for glucose-dependent mRNA degradation.

Introduction of a stop codon at the fifth amino acid position also caused rapid

degradation of the hybrid mRNA, showing that it contained all the elements that

were sufficient to confer nonsense-mediated decay.

These results demonstrate an example of metabolic regulation of gene

expression occurring at two levels, transcription and mRNA stability, each

contributing a portion of the overall regulation.


Utility of Baker's Yeast

Baker's yeast, Saccharomyces cerevisiae, is a eukaryote that has been

studied extensively. It is a unicellular organism with 16 chromosomes of

approximately 12.5 Mb of DNA (Olson, 1992). Yeast has a generation time of

about 90 minutes to several hours, depending on whether it is grown in a

complex medium or a minimal medium, and on the carbon source provided in the

medium. There are several advantages of using this organism as a model

system to dissect eukaryotic functions: 1) It is amenable to genetic manipulation

because the genome size is small compared to higher eukaryotes. 2) It can

survive with a haploid genome making it possible to dissect metabolic pathways

using mutations. 3) Yeast is a facultative anaerobe, making respiratory

functions dispensable. 4) Many genes, particularly those involved in

transcriptional regulation, have mammalian homologs that are structurally and

functionally well conserved. These include RNA polymerase II (Nonet et al.,

1987), GCN4/JUN (Struhl, 1988), TATA binding proteins (TBP) (Cormack et al.,

1994; Reddy and Hahn, 1991; Gill and Tjian, 1991), and CP1/HAP2-3 (Chodosh

et al., 1988b). The gene products of these homologs can complement each


other in in vitro assays. In vivo studies have shown that JUN can complement a

gcn4 mutation in yeast (Struhl, 1988). Yeast is particularly useful for studying

mitochondrial enzymes because it is a facultative anaerobe: mutations

eliminating aerobic energy metabolism are not lethal but can be identified by their

inability to grow on glycerol plates.

The Citrate Svnthase System

Citrate synthase catalyzes the first committed step of the tricarboxylic acid

cycle (TCA), the condensation of oxaloacetate and acetyl CoA. The role of this

pathway in cellular metabolism is twofold. First, the TCA cycle provides the

carbon skeletons used in many biosynthetic pathways such as the synthesis of

glutamate and aspartate. Second, the cycle is oxidative, generating NADH,

which drives the synthesis of ATP. Two different nuclear genes code for

isozymes of citrate synthase (Suissa et al., 1984; Kim et al., 1986; Rosenkrantz

et al., 1986; Rickey and Lewin, 1986). They are CIT1 and CIT2, encoding the

mitochondrial and peroxisomal pathway enzymes, respectively (Lewin et al.,

1990). The CIT2 gene product is involved in the glyoxylate pathway that

produces carbon skeletons for other biosynthetic pathways. The difference in the

cellular location of these two enzymes lies in the N-terminal mitochondrial

targeting sequence on Citlp that is lacking in Citp2 (Rosenkrantz et al., 1986). In

strains in which the CIT1 gene is disrupted, cells still grow on non-fermentable

carbon sources, and some citrate synthase activity is found in the mitochondrial


fraction, suggesting that there may be a cryptic mitochondrion targeting

sequence in the Cit2p (Rickey and Lewin, 1986; Rosenkrantz et al., 1986). The

activity of Citl p is severely repressed when cells are grown in a glucose medium.

When glucose is depleted, or when cells are grown in a non-fermentable carbon

source, the enzyme level increases (derepression) (Hoosein and Lewin, 1984).

This increase in enzyme activity (derepression) correlates with an increase in

steady-state mRNA levels (Kim et al., 1986), and a greater amount of

translatable mRNA (Hoosein and Lewin, 1984). The appearance of increased

mRNA in the derepressed state suggested that regulation of the CIT1 gene may

occur at the transcriptional level. The other possibility is that the increase could

be due to increased stability of the message. Differential stability of mRNA due

to environmental or cellular signals has been demonstrated for other messages

encoded by SP011, SP012, and SP013 genes (required for sporulation) of

yeast (Surosky and Esposito, 1992; Surosky et al., 1994); interleukin 3 (IL-3)

(Wodnar-Filipowicz and Moroni, 1990), and 9E3 mRNA, which encodes an

inflammatory mediator (Stoeckle and Hanafusa, 1989). The SPO transcripts are

much less stable in vegetative growth than they are when cells are in meiosis.

The IL-3 and 9E3 mRNAs are made more stable by calcium ionophores and

serum, respectively (Wodnar-Filipowicz and Moroni, 1990; Stoeckle and

Hanafusa, 1989).

Our goal is to dissect how the CIT1 gene is regulated at both the

transcriptional and mRNA stability level. Since little is known about the

mechanism of mRNA decay, we hope that the results of this study may give us


an insight about the cis and trans elements that are involved in messenger RNA

degradation. The CIT1 gene was chosen as a model to study gene regulation

because of its strategic position in cellular metabolism. As stated earlier, cellular

respiration requires citrate synthase because it catalyzes the first step in the TCA

cycle. The reactions of the TCA cycle are required to generate the reducing

power needed in the electron transport chain, which reduces molecular 02 and is

coupled to the production of ATP. Carbon skeletons are also generated from the

reactions of the TCA cycle for amino acid biosynthesis. The first indication of

unique regulation of citrate synthase came from the work of Satrustegui and

Machado (1977) who showed that induction was not inhibited by cycloheximide

following aeration of an anaerobic culture. This observation suggested that the

precursor for citrate synthase was already present in cultures growing under

repressed growth conditions, and induction does not require de novo protein

synthesis. Understanding how this gene is regulated differently from the related

CIT2 gene may serve as a paradigm as to how cells can regulate genes serving

the same function in different cellular locations.

Transcriptional Regulation in Eukaryotes

The expression of many genes is controlled at the level of transcription.

For this reason understanding how genes are transcriptionally regulated is one of

the fundamental goals in molecular biology.


There are two classes of transcriptional regulatory elements in eukaryotes,

cis- and trans-acting elements (Struhl, 1989). The promoter and enchancer

elements of mammalian genes or the upstream activating sequences (UAS) of

yeast genes constitute cis-acting elements. Upstream repressing sequences

(URS) in yeast are also cis-acting elements. The promoter is made up of the

transcriptional initiation site, the TATA sequences and other proximal elements,

such as SpIl sites (Struhl, 1989). The transcriptional initiation site defines the

first nucleotide incorporated into the newly synthesized mRNA. Many genes

have a single initiation site, but there are some genes that have several initiation

sites, especially in yeast (Fay et al., 1981; Hahn et al., 1985; Repetto and

Tzagoloff, 1990). In yeast, when the distance between the TATA element and

the initiation site is experimentally varied, transcription still starts at defined

positions (Chen and Struhl, 1985). This fact contrasts with mammalian genes in

which changing the position of the TATA element forces the transcription to start

approximately 25-30 bp downstream from the new TATA site. This result would

suggest that the start of mRNA transcription is sequence-dependent in yeast,

whereas it is distance-dependent in mammals. TATA elements are always

situated upstream but near the initiation sites and are found in most class II

transcribed genes. Class II genes constitute those genes that encode proteins

and are transcribed by RNA polymerase II. In yeast, the TATA sequence is

located between 40-120 bp upstream from the start site (Brent, 1985; Chen and

Struhl, 1985). Although the TATA sequence is required for transcription initiation

of most genes, approximately 20% of eukaryotic genes have neither the


conserved classical TATAAA sequence nor is it required for transcription

initiation. The PGK gene of yeast and the terminal deoxytransferase gene of

mammalian cells are examples of genes that do not require a TATA sequence

(Ogden et al., 1986). This means that there are other cis-elements necessary for

transcription initiation which have yet to defined. There are also different classes

of the TATA sequence. The HIS3 gene contains a TATA element that is involved

only in constitutive transcription (.Tc) and another TATA sequence that is involved

in regulated expression (TR) (Harbury and Struhl, 1989; Chen and Struhl, 1988).

The work of Chen and Struhl (1985) showed that sequences downstream of the

TATA elements are also important for proper transcription initiation, since

mutations surrounding the start site move the site to a different location.

Mutations at certain positions in the sequence TATAAA discriminated between

the GCN4 and GAL4 as transcriptional activators, suggesting that the

mechanisms of activation or the accessory factors required for activation are

different for the various activators. The basic transcriptional machinery proteins

including RNA polymerase II and TFIIA through F bind to the TATA sequence in

a sequential manner to initiate transcription (Buratowski et al., 1989).

Enhancers are sequences that increase the transcription of genes when

bound to their cognate factors (Dynan, 1989). They may be situated up to 50 kb

from the initiation site and still affect transcription in either orientation and can

function whether present upstream or downstream from the transcription unit.

Enhancers are modular in nature, made up of identical or a mixture of different

enhanson elements that usually work synergistically (Dynan, 1989). Enhanson is


defined as the minimum discrete sequence that binds to a transcriptional factor.

Yeast cells lack enhancers such as those of mammalian cells but instead contain

upstream activating sequences (UAS) that bind activating proteins (Guarente,

1987; Struhl, 1993). Unlike enhancers elements, UASs can only function when

placed upstream from the transcriptional start site (Guarente and Hoar, 1984).

The UASs usually function in either orientation when situated between 20-1500

bp from the TATA element. They are usually 9-30 bp in length and may or may

not have a dyad symmetry. Those with dyad symmetry usually bind to

homodimers or heterodimers of a specific activator. Therefore, UAS elements

resemble proximal promoter elements such as Spl sites more than they do true

enhancers. Operators or upstream repressing sequences (URS) bind to

negative acting trans-acting factors and repress gene expression. These

elements usually lie between the UAS and the TATA elements and prevent

activation by activators, but have also been shown to lie up to 2 kb upstream or

downstream of the mRNA initiation site (Brand et al., 1985) and still affect

transcription. Binding to UAS or URS may be mutually exclusive if they have

overlapping sequences or they may be independent of one another.

Trans-acting factors constitute the second class of transcriptional

regulatory elements. These are proteins that bind to DNA at specific sites. The

RNA polymerase II and ancillary proteins such as TFIIA thru F make up the basic

transcriptional machinery. The RNA polymerase II of yeast has 12 subunits

(Thuriaux and Sentenac, 1992); the largest subunit (220 kD) is encoded by the

gene RPBI. This subunit is similar to the largest mammalian subunit and shares


similarities with the largest subunit of the yeast RNA polymerase III and the P'

subunit of the E. coli RNA polymerase. In yeast, the carboxy terminal domain

(CTD) of Rpbl1p has a set of seven amino acids (PTSPSYS) repeated 26 times.

In the largest subunit of mammalian RNA polymerase II, the CTD is repeated 52

times (Corden et al., 1985). The Rpbl1p is essential for viability, and deletion of

the CTD is lethal (Nonet et al., 1987). A deletion that left only 11 or 12 of the

repeat units allowed viability but caused a cold-sensitive phenotype (Nonet et al.,

1987). Some of the subunit polypeptides found in RNA polymerase II are shared

by RNA polymerase I and Ill.

The first factor that binds to the DNA and allows for the formation of a

competent transcription complex is the TFIID. In mammalian cells, TFIID

consists of the TATA-binding protein (TBP) and the TBP-associated factors

(TAFs) whereas in yeast only a single protein, TBP, has been identified to carry

out this function. In a DNase I protection assay in vitro, the TBP protects

approximately 19 bp, suggesting that sequences beyond the TATA element are

necessary for proper functioning (Buratowski et al., 1989). Ironically, TBP alone

is not sufficient to form a DNA/protein complex in a bandshift assay except when

TFIIA is present in the complex (Buratowski et al., 1989). The TBP does not

have any recognizable DNA binding motif, but it has a highly basic C-terminus.

There are two direct repeats at the C-terminus, separated by a stretch of basic

residues. The yeast TBP (yTBP) and the human (hTBP) are functionally

interchangeable in an in vitro transcription assay, but the hTBP cannot

complement a strain deleted for the SPT15 gene which encodes TBP (Gill and


Tjian, 1991; Cormack et al., 1991). Cell viability could be restored by a hybrid

protein, if the C-terminal domain was derived from yeast (Cormack et al., 1991).

This would suggest that the species specificity determinants lie in this region. To

determine the exact amino acids(s) responsible for the species specificity,

Cormack et al. (1994) selected for a hTBP/yTBP hybrid protein, which could

support faster yeast growth. The starting hybrid contained a human C-terminal

domain which had been shown to support growth at a very slow pace. This

selection identified three independent mutants that changed arginine 231 to

lysine. Interestingly, lysine occupies an identical position in the native yTBP. In

addition, mutation at this position in an otherwise intact hTBP supported growth

of yeast. After the initial binding by TBP, other factors bind to the TATA and

surrounding sequences (Buratowski et al., 1989). The order in which the basic

transcription factors come into the preinitiation complex was shown by bandshift

assay to be as follows: TFIID, TFIIA, TFIIB, RNA polymerase II, TFIIE, then

TFIIF/H complex (Buratowski et al., 1989). Assembly of these factors forms the

preinitiation complex. Transition to the initiation phase is preceded by

phosphorylation of the CTD of polymerase II by TFIIH factor (Lu et al., 1992).

Other factors playing significant roles in transcriptional regulation of genes

include activating, repressing, and inducing factors (Struhl, 1989; Guarente,

1992). These factors are required for proper regulation of individual genes. The

most studied of these secondary factors are the activator proteins. These

proteins usually have a modular structure, each one of the modules being

capable of functioning independently. The "domain swap" experiment with LexA


and Gal4 by Brent and Ptashne (1985) clearly illustrated that transcriptional

activators such as GAL4 have DNA-binding domains and activator domains, but

other activators such as glucocorticoid receptor and Hapl p also have ligand

binding domains that regulate the activators (Picard et al., 1988; Kim et al., 1990;

Chandler et al., 1983).

Many of the DNA-binding domains have identifiable structural motifs

involved in DNA binding. These include the: 1) helix-turn-helix motif (Pabo and

Lewis, 1982; Sauer et al., 1982), 2) zinc-finger domain (Laughon and Gelsteland,

1984), 3) leucine-zipper motif (Landshultz et al., 1988), and 4) P3-sheet motif

(Guarente, 1992). The helix-turn-helix motif is most commonly found among

prokaryotic DNA-binding proteins such as the A Cro and A repressor proteins

(Pabo et al., 1982) and CAP protein (Sauer et al., 1982). The helix-turn-helix

proteins usually have one a-helix followed by a turn, then a second a-helix. The

second helix is usually called the recognition helix because it fits into the major

groove of a B-form DNA, while the first helix seats above the groove. Many

homeotic gene proteins of Drosophila, such as the antennapediaa and engrailed

(McGinnis et al., 1984a; McGinnis et al., 1984b), also have a helix-turn-helix

motif similar to that described above; hence they are commonly referred to as the

homeodomains. Homeotic genes are defined as genes which when mutated

convert one body part into another. Yeast regulatory proteins having similar

structure are the al and al mating type regulators (Porter and Smith, 1986).

The zinc-finger motif was first discovered in the TFIIIA protein, a Xenopus

5S DNA-binding protein. One unit of zinc-finger motif usually consist of about 30


amino acid residues, containing the sequence pattern Cys-X2or4-Cys-Xi12-His-X3or5-

His. Binding of the zinc ion is coordinated by the 2 cysteine and histidine

residues (Miller et al., 1985). The yeast Adrl p, an activator of ADH2 gene, also

has a sequence composition similar to the zinc-finger motif. In another type of

zinc-finger, the ion binding is coordinated by four cysteine residues instead of 2

cysteine and 2 histidines. The yeast regulatory proteins Gal4p and Hapl p are

examples of this class of zinc-finger (Laughon and Gelsteland, 1984; Pfeifer et

al., 1987b; Kim et al., 1990). The Gal4p activates the GAL1-, -7, and -10 genes

that are required for galactose utilization by yeast (Braum et al., 1986). The

Haplp regulates several yeast genes such as CYC1 (Guarente et al., 1984;

Pfeifer et al, 1987a), CYC7 (Pfeifer et al, 1987b; Prezant et al., 1987), COX5A

(Trueblood et al., 1988), and CYT1 (Schneider and Guarente, 1991). An unusual

property of Haplp is its recognition of nonidentical UASs (Prezant et al., 1987).

The affinity for the different binding sites varies, allowing for flexibility in

regulation. The Haplp requires heme for activation (Pfeifer et al., 1987; Kim et

al., 1990).

A third common class of binding domain is the leucine-zipper found in

activators such as Gcn4p, avianjun, AP1, Myc, Fos, and C/EBP (Landshultz et

al., 1988). The zipper region of the protein has about 30 amino acids and a

leucine residue at every seventh position. This region of the protein is involved in

dimerization, either homologous or heterologous, necessary for DNA binding.

Located N-terminal to the zipper region is usually a stretch of basic residues that

actually binds to the DNA. This basic region can bind to DNA by itself if there is


a disulfide bond allowing dimerization. Additionally, there are other activators

which do not have an easily identifiable DNA-binding motif.

All of the transcriptional activators also have an activation domain.

Perhaps the most well characterized activation domain is the acidic activation

domain. These activator domains contain many negatively charged amino acids;

hence are often referred to as the "acidic-activation" domains (AAD). Studies by

Giniger and Ptashne (1987) in which a synthetic peptide was used with a

predicted amphipatic a-helix (AH) and net negative charge in conjunction with

the GAL4 DNA-binding domain, showed it was competent to activate

transcription in vivo, though only when over-expressed. Cloning of random

oligonucleotides from E. coli that could support activation resulted in sequences

with net negative charge (Ma and Ptashne, 1987). A gradual reduction in

activation potential was observed when some of these acidic residues were

removed from the Gcn4p (Hope et al., 1988). Together these results strongly

suggested the need for an acidic activation domain. However, as recently shown

(Leuther et al., 1993; Hoy et al., 1993), the ability to activate does not require

acidity or net negative charge. Rather, the most important criteria to function as

an activator was the ability to form a P3-pleated sheet. Replacement of the

negatively charged residues with non-charged or positively charged residues will

still support activation. The Gal4p and Gcn4p were shown to form a 3-sheet

under near physiological conditions (Hoy et al., 1993). Other defined activation

domains in mammalian cells are rich in glutamine, e.g. Spl (Courey and Tjian,

1988), or proline amino acid residues. These factors are usually found in


mammalian cells and do not function in yeast cells as the acidic activation

domains do.

As stated earlier, one of the hallmarks of enhancers is their ability to

regulate a gene from a distance, sometimes up to 50 kb. The UAS elements in

yeast usually lie within a few hundred bases of the TATA box. The question has

been, how do protein factors that bind to these sequence effect activation? The

current model to explain the ability of trans-factors to activate at a distance is that

DNA sequences between these activators and the basic transcription machinery

"loop-out", allowing protein-protein interaction between these factors (Hofmann et

al., 1989; Ptashne, 1986). Beside the transcriptional activators and proteins of

the basic transcriptional machinery, there are other intermediary factors,

commonly called adaptors, co-activators, or mediators (Kelleher et al., 1990;

Pugh and Tjian, 1990; Berger et al., 1992; Struhl, 1993). These factors do not

usually have DNA-binding domains but carry out their function by direct protein-

protein interaction. Their exact mode of action is not known but is believed to

involve either strengthening the interaction of the activator and the basic

machinery or enabling the specific activators to gain better access to the

chromatin (Berger et al., 1992). One such factor identified is the ADA2 gene

product (Berger et al., 1992). Mutation in this gene suppresses the lethal effect

of the overexpression of Gal4p-VP16 in yeast. The activity of Gal4p-VP16 and

GCN4 activators are reduced in an ada2 mutant, while no effect is exhibited by a

Gal4p-Hap4p activator. This result suggests that these factors are not universal;

rather, they are specific for a particular class of activators. The GCN5 gene


product is required for normal levels of transcription by GCN4 and HAP2/3/4

heterotrimeric activators (Georgakopuolus and Thireos, 1992) and is considered

to be an adaptor that may enhance the activity to different activators.

Glucose Repression

When glucose is present in the growth medium, the products of a large

number of genes are severely repressed in S. cerevisiae. These genes include

those necessary for alternative carbon source metabolism (Perlman and Mahler,

1974; Carlson and Bostein, 1982; Denis et al., 1981), gluconeogenesis (Sedivy

and Fraenkel, 1985; Scholer and Schuller, 1994), TCA cycle enzymes (Polakis et

al., 1965; Hoosein and Lewin, 1984; Lombardo et al., 1992; Roy and Dawes,

1987; Repetto and Tzagoloff, 1990), and respiration chain cycle proteins (Polakis

and Bartley, 1965; Perlman and Mahler, 1974; Szekely and Montgomery, 1984;

Mueller and Getz, 1986; Guarente and Mason, 1983; Wright and Poyton, 1990).

The extent of repression varies from about 1000-fold for the GAL genes

(Johnston et al., 1994) to about 5-fold for some genes encoding mitochondrial

proteins (Szekely and Montgomery, 1984).

The phenomenon of repression of many genes when glucose is present in

the growth medium has been observed in some prokaryotes such as E. coil

(Magasanik, 1962) and B. subtilis (Rosenkrantz et al., 1985) and other

eukaryotes such as S. pombe (Hoffman and Winston, 1991). In E. coli, the

repression is mediated by cAMP (Magasanik, 1962). In the presence of glucose,


the cAMP level is significantly reduced; but after the glucose has been depleted,

the level of cAMP increases. This increase facilitates binding of cAMP to

catabolite activation protein (CAP). Activated CAP, cAMP-CAP, binds to the

promoters of the glucose repressed genes to activate transcription. However,

the role of cAMP in glucose repression in S. cerevisiae in not as clear.

Matsumoto et al. (1982, 1983) isolated yeast strains that required exogenously

supplied cAMP for growth. In these strain, the galactokinase enzyme encoded

by (GAL1), which is glucose repressible, was still derepressed in the presence of

high levels of cAMP. Measurement of the level of cAMP present in different

media containing glucose or other non-fermentable carbon sources showed that

the level of cAMP was higher in the depressed state. These experiments

suggest that cAMP may not play any role in mediating the effect of glucose.

Cyclic AMP in eukaryotes is postulated to act by activation of protein

kinases which control the phosphorylation of various critical proteins and thereby

modulate the activity of the proteins. If so, cAMP may be involved in the

regulation of one glucose repressible gene, ADH2. The ADH2 encodes an

isozyme of alcohol dehydrogenase, which converts ethanol to acetaldehyde in

the ethanol utilization pathway. The Adrl p, encoded ADR1 gene, is a positive

transcriptional activator of ADH2 that binds to UAS1 (Denis and Young, 1983).

The protein contains several potential sites for phosphorylation by cAMP-

dependent protein kinase (cAPK) (Hartshorne et al., 1986). It has been

demonstrated in vitro that yeast cAPK can phosphorylate one of these putative

phosphorylation sites (Ser-230) (Cherry et al., 1989). A class of ADR1 mutant


(ADR1c) was isolated that partially relieved the glucose repression of ADH2.

These mutants alter the phosphorylation site and reduce its efficiency of

phosphorylation (Cherry et al., 1989). Also, when the regulatory subunit for

adenylate catalase (BCY1) was mutated, which allowed for unregulated

expression of the catalytic subunit, ADH2 expression was severely reduced. This

observation suggests that the regulation of the ADH2 gene is mediated, partially,

through the modulation of ADR1 activity by cAMP. However, a more recent

study (Denis et al., 1992) shows that other ADR1c mutants which can still be

phosphorylated also partially relieved glucose repression. It is believed that the

ADR1c mutations may block the binding of a repressor to Adrl p or alter the

structure of Adrl p so that transcriptional activation regions become unmasked.

This would mean that the level of phosphorylation plays little role in the regulation

of Adrl p.

Another eukaryotic organism exhibiting glucose repression is the yeast

Schizosaccharomyce pombe. In Schizosaccharomyce pombe, the gene for

fructose- 1,6-biphosphate (fbpl) is glucose repressible. The work of Hoffman and

Winston (1991) showed that mutation in git2- (cyrl) caused constitutive

expression of fbpl. The git2+ gene codes for adenylate cyclase, which converts

ATP to cAMP; cAMP is required for the activation of cAPK. When cAMP was

exogenously added it caused reduction in the mRNA level of fbpl. However,

even this study showed that some of the git2- mutants that allowed constitutive

expression of fbpl did not reduced the expression level when exogenous cAMP

was added to the growth medium. In addition, levels of cAMP did not show any


significant difference in repressing and derepressing media. Although the

genetic evidence strongly implicates a role for cAMP in mediating glucose

repression, this effect may be indirect at best.

Among the best studied of the glucose repressible genes are the GAL,

GAL7, and GAL10 genes of Saccharomyces cerevisiae, which encode

galactokinase, galactotransferase, and UDP-galactose epimerase, respectively.

These proteins are required for galactose utilization. Their regulation shows that

glucose mediated repression of genes is complex and occurs at several levels.

The regulation of the GALl gene, for example, occurs at three levels. First,

glucose reduces the level of functional inducer, galactose, in the cell by

repressing transcription of the galactose transporter-galactose permease, which

is encoded by the GAL2 gene (Braum et al., 1986; Tschopp et al., 1986) and

inactivating preexisting permease molecules, thereby preventing any transport of

inducer into the cell (Holzer and Matern, 1977). The reduction in the inducer

levels reduces function of the activator Gal4p. The second mechanism of

glucose repression of the GAL genes involves inhibition of the transcriptional

activator Gal4p (Flick and Johnston, 1990). The inhibition is due to reduction in

the expression of GAL4 (Johnston et al., 1994) and inhibition of Gal4p function

by the inhibitory domain "ID". The inhibitory domain constitutively inhibits the

transcription of a heterologous activator in glucose and glycerol media.

However, when the glucose response domain (GRD) is present, activation

occurs in a glycerol medium but not in a glucose medium (Stone and Sadowski,

1993). Repression of the GAL4 gene is mediated by the Migl p, which binds to


URS sequences in the GAL4 promoter (Johnston et al., 1994). Other

transcriptional activators that share structural similarity with the Gal4p include

Leu3p (Zhou et al., 1987), Prplp (Schmitt et al., 1990), Put3p (Marczak and

Brandiss, 1991), and Lac9p (Salmeron, Jr. and Johnston, 1986) of K. lactis. A

third mechanism of glucose regulation involves URS sequences which mediate

repression of genes by binding to repressor proteins. The GALl gene has a

URSGAL located between the UASGAL and the TATA sequence (Finley et al., 1990;

Flick and Johnston, 1992). The Migl p binds to URSGAL sequences to inhibit

Gal4p activation.

There are several other genes that are required for glucose regulation.

They are either needed to relieve repression or to maintain repression. Studies

by Rose et al. (1991) showed that the products of HXK1 and HXK2 are required

for glucose repression of many genes. The gene products of HXK1 and HXK2

phosphorylate hexose sugars, but how they mediate their effect is not known.

The SNF1 gene encodes a protein kinase that is required for derepression of the

SUC2 (invertase) gene (Calenza and Carlson, 1986). Mutation in SNF1 also

causes defects in derepression of SDHI (succinate dehydrogenase), ICL1

(isocitrate lyase), and MDHI (malate dehydrogenase) genes. The SNF1 gene is

believed to exert its effect by modifying transcriptional activators that bind to the

UAS of SUC2. The target for this kinase is probably the Snf2p/Snf5p/Snf6p

complex, which is a transcriptional activator (Laurent et al., 1992; Laurent and

Carlson, 1992). The exact role of this protein may be to disrupt nucleosomes

and allow the subsequent entry of gene specific transcriptional activators.


Another set of genes that is required to maintain repression are SSN6 and TUP1

genes (Keleher et al., 1992). They are transcriptional repressors interacting with

gene-specific factors to mediate their effect. In contrast to the negative roles

SSN6 and TUP1 play in regulating many glucose repressible genes, they have a

positive effect on CYC1 expression via the HAP1 transcriptional activator (Zhang

and Guarente, 1994).

Another well characterized glucose repressible gene is the CYC1. It

encodes iso-1 -cytochrome c, which is involved in the electron transport chain of

respiration. The CYC1 gene has two UASs, UAS1 and UAS2. Regulation at

UAS1 occurs via the Haplp after it has been bound by heme (Guarente et al.,

1984; Kim et al., 1990). Glucose regulation of CYC1 occurs at the UAS2 site

through a multisubunit protein called the Hap2p/Hap3p/Hap4p, encoded by

HAP2, HAP3, and HAP4 genes, respectively (Guarente et al., 1984; Pinkham

and Guarente, 1985; Pinkham et al., 1987; Hahn and Guarente, 1988; Forsburg

and Guarente, 1989). Mediation of glucose repression on CYC1 expression

occurs by repressing the transcription of HAP4. The Hap4p has the activation

domain of this multisubunit complex (Forsburg and Guarente, 1989); therefore, in

a glucose medium reduced synthesis of this activator causes reduction of CYC1

expression. Mutations in any one of the genes that encode the transcriptional

activator protein reduce the expression of many genes involved in the Krebs

Cycle such as the genes encoding lipoamide dehydrogenase (Bowman et al.,

1992), aconitase (Gangloff et al., 1990), and dihydrolipoyl transsuccinylase

(Repetto and Tzagoloff, 1990). These genes also have the consensus binding


site for the Hap2p/Hap3p/Hap4p complex. The CIT1 gene also has the

consensus binding site for this activator, but deletion of this sequence or

mutation of any one gene encoding the proteins does not severely impact

expression (this study). Other mitochondrial genes also known to be regulated

by this transcriptional activator complex include COX5A (Trueblood et al., 1988),

COX6 (Trawick et al., 1989; Trawick et al., 1992), and HEM1 (Keng and

Guarente, 1987). The COX5A and COX6 genes encode subunits Va and VI,

respectively, of cytochrome c oxidase, and HEM1 encodes 6-aminoluvilinate


The other interesting feature about the consensus binding site for the

Hap2p/Hap3p/Hap4p is the presence of the CCAAT-box sequence at the core of

the consensus sequence. This sequence is also present in many mammalian

promoters and functions as promoter. The CCAAT-box also binds a multisubunit

activator, CP1A and CP1B (Chodosh et al., 1988a; 1988b). Using bandshift

assay and DNase I protection assays, Chodosh et al. (1988b) showed that the

Hap proteins and CP1 proteins bind to and protect similar DNA sequences. In a

bandshift assay, they showed that Hap2p can substitute for CP1B and Hap3p

can substitute for CP1A in binding DNA at each cognate sequence. Although

these two sets of proteins have evolved to regulate different activities, they still

bind similar DNA sequences and can complement each other.


mRNA Stability

The steady-state level of a given species of mRNA is a function of both

the rate of transcription and the rate of decay of the message. Although much is

known about how genes are transcribed and the factors involved, little is known

about how mRNA is targeted for decay and the mechanisms of decay. The

degradation of mRNA provides the cell another level of control and a very

powerful means of gene regulation. The wide difference in the half-life of

different messages (Herrick et al., 1990) would suggest that there are specific

degradation pathways for different messages. Apart from decay of normal

mRNA, aberrant mRNAs, such as unspliced mRNA or those RNA containing

premature termination signals are rapidly removed (Leeds et al., 1991; Peltz et

al., 1993). This is necessary to prevent the assembly of the translational

apparatus on a message that would not produce a productive protein or one that

might even be detrimental to the cell. A wide variety of external and cellular

signals such as oxygen, iron, glucose and light have been shown to affect the

level of mRNA stability (reviewed in Brawerman, 1993). The half-lives of mRNAs

vary greatly in different cell types. In E.cofi mRNA half-lives ranges from about

20 seconds to about 50 minutes (Blundel et al., 1972). In yeast, it range from

about 1 minute to almost 100 minutes for some messages (Herrick et al., 1990).

In mammalian cells, mRNA half-life could range anywhere from 15 minutes to

over 24 hours (Gordon et al., 1988; Shyu et al., 1990). Since all mRNAs do not

have the same half-life, there must be features unique to each mRNA or class of


mRNA that causes them to follow a particular pathway for decay; this feature

may constitute a cis-element(s) inherent in each mRNA. The search for cis-

elements that are involved in regulating the decay of mRNA has so far revealed

sequences that usually confer instability rather than stability (Heaton et al.,

1992). No sequence has yet been shown to confer increased stability.

The structural determinants for mRNA instability seem to be present

throughout the message, especially for a eukaryotic mRNA. Although the 5' cap

structure on a eukaryotic message has not been shown to directly affect stability

of any mRNA, it is believed that it could serve a protective role, because the 5'-5'

phosphodiester bond is intrinsically resistant to ribonucleases. This putative

protective role of the 5' cap structure was shown by Muhlrad et al. (1994). These

workers showed that in the degradation pathway of MFA2 mRNA, decapping of

the message always takes place before the decay intermediates could be

detected. The 5' untranslated region (UTR) of eukaryotes has not been shown

to directly affect mRNA stability, except in cases where translation is required for

degradation and the 5' UTR controls the translation of that message. In contrast

to eukaryotes, prokaryotes have stem-loop structures at the 5' termini of their

messages that affect their decay rate (Emory et al., 1992; Bouvet and Belasco,

1992; DiMari and Bechhoffer, 1993). For example, in the ompA mRNA of,

the presence of the stem structure is critical for maintaining the normal half-life of

approximately 14 minutes. Insertion of up to 3 nucleotides to the 5' end of the

terminal hairpin structure causes dramatic decrease in the half-life of the OmpA

mRNA (Emory et al., 1992). Also, removal of the Shine-Delgano sequences from


the second single-stranded region of the leader reduced the half-life (Emory et

al., 1992). This would imply that ribosome binding may also protect the mRNA.

Hence translation of the message may be required for stability.

In eukaryotes, the coding regions of several genes, including c-myc

(Willington et al., 1993), MATaI (Caponigro et al., 1993), and STE3 (Heaton et

al., 1992) have instability elements, that promote rapid decay of the messages

they encode. The putative instability element of MATal was localized to a 65

nucleotide sequence that has a 5' and a 3' portion (Caponigro et al., 1993). The

3' portion is necessary and sufficient to decrease the half-life of an otherwise

stable mRNA, but the decay is further stimulated when the 5' portion is included

in the fusion. The 5' portion contains some rare codons which when replaced

with more common codons, increased the half-life of the chimeric mRNA. This

result suggests that rare codons may cause the ribosome to stall on the message

which may lead to an initial endonucleolytic cleavage followed by an exonuclease


The 3' untranslated region (3' UTR) of many genes contains sequences

that cause their rapid decay. These include STE3 (Heaton et al., 1992), MATa1

(Caponigro et al., 1993), MFA2 (Muhlrad and Parker, 1992), c-myc (Willington et

al., 1993), and transferring receptor (TfR) gene (Klausner et al., 1993). The

sequence of the STE3, MFA2 and MATa1 that cause rapid decay have not been

well characterized.

Interestingly, the TfR mRNA, regulated by iron, has several well

conserved stem-loop structures that are observed in widely divergent species.


These stem-loop structures at the 3' UTR are called the iron response element

(IRE) and can bind a protein called the IRE-binding protein (IRE-BP) (Klausner et

al., 1993). The IRE-BP also has aconitase activity. When iron is scarce, the

IRE-BP binds to the IRE of TfR mRNA and increases its half-life. There is also

an uncharacterized sequence called the rapid turnover determinant overlapping

the IRE. Point mutations within the IRE eliminating IRE-BP binding still caused

rapid decay of the mRNA, whereas a deletion mutation removing the IRE-BP

binding site and presumably the rapid turnover determinant slowed the mRNA

decay rate.

The 3' UTR of c-myc and lymphokines and proto-oncogenes contain the

sequence AUUUA, usually repeated several times, referred to as AU-rich

elements (ARE), that cause rapid decay of an mRNA containing it. However,

recent studies by Zubiaga and coworkers (1995) show that the pentanucleotide

AUUUA is not sufficient to confer destabilization upon heterologous mRNA.

Instead a consensus nanonucleotide of UUAUUUAUU is required for the

destabilization phenotype. This sequence, when present in multiple copies,

increases the decay rate.

The mechanism of mRNA decay in eukaryotic cells is beginning to be

elucidated. Parker and coworkers (Muhlrad and Parker, 1992; Decker and

Parker, 1993; Muhlrad et al., 1994) demonstrated that deadenylation of MFA2

mRNA is the first step before decay occurs, for some mRNAs. The

deadenylation occurs in two stages: an initial deadenylation and a terminal

deadenylation. The rate of the initial deadenylation of various mRNAs


determines their half-life. Messages with longer half-life have an initial

deadenylation rate significantly lower than those with a short half-life (Xing et al.,

1993). This result suggests that the rate limiting step for this class of mRNA is

the deadenylation step. To define direction of decay, whether 5' -+ 3' or 3' -+ 5', a

poly(G) sequence was inserted into the 3' UTR of a test mRNA. The poly(G)

forms a secondary structure that slows decay in either direction. When the fate

of the mRNA containing the poly(G) track was followed using a poly C probe,

decay was found to proceed in a 5' --+ 3' direction. In an xrnl mutant (XRN1

encodes the major 5' -+ 3' exonuclease in yeast), full length mRNA was seen for

a much longer time, indicating that, after deadenylation, this exonuclease is

responsible for degrading the RNA to mononucleotides. Using an antibody to the

5' cap structure Muhlrad et al. (1994) were able to show that the cap structure is

removed before exonuclease digestion.

The use of mutations that result in premature translation termination in

several genes have identified some genes in yeast that are involved in the rapid

decay of mRNAs with premature nonsense codons. Two such genes are UPF1

and UPF3 (Leeds et al., 1991). In a wild-type strain, most mRNAs containing

premature termination signals have a decay rate up to 12 times faster than

normal mRNA (Peltz et al., 1993), but in either upfl and upf3 mutants some of

these messages are selectively stabilized without affecting the turnover of the

other message (Leeds et al., 1991; Leeds et al., 1992). The nonsense mutations

that caused the rapid decay are always located within the first two-thirds of the

coding region. If the mutation is near the 3' end of the gene, the half-life is quite


similar to the wild-type mRNA. This finding suggested that some sequences

downstream of the nonsense codon may be required for the rapid decay. By

introducing nonsense mutations throughout the PGK1 gene, Peltz et al. (1993)

were able to show that a "downstream element" was necessary to cause rapid

decay. This "downstream element" functions in an orientation dependent


Other genes required for rapid degradation of specific mRNAs include the

UME2 and UME5 genes (Surosky et al., 1994). These genes are required for

rapid decay of meiosis specific genes in a medium containing glucose. How

these trans factors target specific mRNA for decay is not known, but they could

serve as molecular tags that designate the mRNA for decay when bound at their

recognition site.


Growth Conditions and Media

All E. coli strains were cultivated in Luria-Bertani media (1% Bacto-

tryptone, 0.5% Bacto-yeast extract, 1% NaCI pH 7.5). Yeast cells were grown in

either complex media (1% Bacto-peptone, 1% Bacto-yeast extract)

supplemented with 2% dextrose or 2% ethanol, or synthetic dextrose (SD)

(0.67% yeast nitrogen base without amino acids, 2% dextrose). All plates were

supplemented with 1.5% agar.

Yeast Transformation

Yeast transformation was routinely done either by the method described

by Ito et al., (1983) or the colony method (Baker, 1991). Several colonies were

picked from a YPD plate that was no more than two days old and resuspended in

1 ml 1 X TEL solution (10 mM Tris-HCI, pH 7.5; 1 mM EDTA; 100 mM lithium

acetate pH 7.5) in a microcentrifuge tube. The suspension was left at room

temperature for 1 minute then centrifuged at 12,000 rpm in an Eppendorf

centrifuge for 10 seconds. The supernatant was decanted and cellls were


resuspended in 100 pI 1 X TEL solution. Then 50 pg denatured salmon sperm

DNA plus 10 pg of transforming DNA was added. The cells were incubated with

the DNA at 30C with gently shaking for 15 minutes, subsequently 700 pI 40%

PEG 4000 (polyethylene glycol)/TEL solution was added to the mixture, which

was vortexed vigorously for a few seconds. Cells were transferred to a 30C

heat block and incubated for additional 15 minutes without shaking. At the end,

they, were incubated at 42C for 15 minutes. Cells then were collected by

centrifugation in an Eppendorf centrifuge for 30 seconds, the supernatant was

decanted and cells were resuspended in 200 pI TE pH 7.5 and plated 100 pI per

petrie plate on appropriate selective media and incubated at the appropriate

temperature. Transformants were usually obtained in 4-5 days. Putative

transformants were restreaked on the same selective media and incubated for

another 3 days. Transformants were usually screened for their auxotrophic

markers by streaking on minimal media on which they should not grow.

Construction of 5' (DISTAL) and 3' (PROXIMAL) Deletions

The construction of these plasmids was started by Timothy Rickey

(Rickey, 1988). All CITI-lacZ constructions originated from two plasmids,

pSH18-8 and YcpZ2. pSH18-8 contains approximately 800 bp of the upstream

sequences and 26 codons sequence of the CIT1 gene cloned into Smal site of

pUC18. The YcpZ-2 plasmid was used to make the lacZ fusion constructs to

study the effect of promoter deletions. YcpZ-2 was made by inserting the CEN4


sequence from pBM150 between the ARS1 sequence and the polylinker of

pMC1790. The pMC1790 vector was provided by Dr. M. J. Casadaban (1979).

The YcpZ-2 vector contains the lacZ gene without a promoter or the first 22

nucleotides of the coding sequence, an E. coil origin of replication, and the bla

(13-lactamase) gene which conferred ampicillin resistance. There is a TRP1 gene

which served as a selectable marker in yeast, and the CEN4 and ARS1

sequences which allowed the plasmid to be maintained in a stable form and

replicate in yeast, respectively.

A schematic of the strategy used to generate the 5' deletions is depicted in

Figure 1. Plasmid pSH18-8 was cleaved with Smal, followed by a Bal31

exonuclease digestion according to the manufacturer's recommendation

(Boehringer Mannheim). Following the Bal31 digestion, the DNA was treated

with BamHI, to release the yeast DNA bearing the sequential deletions upstream

from the transcriptional start site of CIT1, and run on 1 % agarose gel. Selected

fragments which had varying degrees of deletion were ligated to the YCpZ-2

vector which had been digested by Smal and BamHl. The end points of the

deletions were determined by sequencing, using the Sanger dideoxy sequencing


Generation of 3' proximall) deletions and subsequent subcloning into the

YCpZ-2 vector were done in two steps. First, pSH1 8-8 was digested with EcoRV

which cuts at a unique site 111 bp upstream from the major transcriptional start

site, followed by a Bal31 exonuclease digestion for varying length of time. The

EcoRV site is 11 bp upstream of the putative TATA element. Since the Bal31

Figure 1. Construction of the 5' deletions. The first step was to digest
plasmid pSH18-8 with Smal, followed by treatment with Bal 31nuclease. The
nuclease treated DNA was digested with BamHI which released the yeast DNA.
This was then ligated to YCpZ-2 vector which had been linearized with BamHI
and Smal. The thin line represents CITI1 sequences located upstream of the
coding region. The straight line hatch marks represents CITI1 coding region and
the crossed hatch marks represents pUC18 vector sequences. In the second
vector, the filled box represents the E. co/i /acZ gene and the stippled box
represents yeast TRP1 gene.

EcoRV +

IIII111111111 II I i7111111/

Small and Bal 31

EcoRV +1
I I n

EcoRV +1

1111111114m/WM WA,7



EcoRV +1 BamHl1
I 1 ,,, ,,, ,,, ,,,

EcoRV +1 BamH1

EcoR1 Smal BamH1

TRP\\\ La .- .....

..... .... ....... llll lllll ll ..... .... ....






Lac Z


digestion was likely to remove this element, which is essential for proper

initiation, it was necessary to restore this sequence. The nuclease-treated DNA

was digested with BamHI and run on a 1% agarose gel. This generated two

fragments, a small fragment which consisted of CIT1 sequences from the EcoRV

site to the twenty-sixth codon, and a large fragment that consisted of vector

sequences and partially deleted CIT1 upstream sequences. A 300 bp EcoRV-

BamHI fragment from pSH18-8 containing the essential CIT1 sequences, was

ligated to the nested set of deleted DNA fragments. This generated deletion

clones that started at the EcoRV site and extended upstream, but retained the

TATA sequence, the transcription initiation site, and the coding region. Figure 2

is a diagrammatic representation of how these 3' deletions were generated.

Construction of Internal Deletions

Internal deletions of the promoter region were constructed to determine

the relative contribution of each of these regions to high level expression of CIT1

gene. The technique used to make these constructs is called recombinant circle

polymerase chain reaction (RCPCR) (Jones and Howard, 1991). Two primers

were used to prime PCR which extended in opposite directions on the template.

The standard polymerase chain reaction consisted of 10 pmoles each of the

primers, 3.4 ng of template DNA (pSL123), 200 moles of four dNTP's, 2.5 mM

magnesium chloride; 10 mM Tris-HCI, pH 8.3; 30 mM potassium chloride; and

2.5 U Ampli-Taq DNA polymerase (Cetus Corporation). Twenty cycles of

Figure 2. Construction of the 3' deletions. First, pSH18-8 was digested with
EcoRV which is 111 bp from the transcriptional start site. The DNA was then
treated with Bal 31. Following the nuclease treatment, the DNA was digested
with BamHI, this released CIT1I sequences that contained sequences essential
for proper transcription initiation. To restore these sequences, an EcoRV/BamHI
fragment from the original plasmid (pSH1 8-8) was ligated to the Bal 31 treated
DNA. A Smal/BamHI fragment from the nuclease digested DNA was then
subcloned into YcpZ-2 vector.


EcoRV +1

/1 77777777

IEcoRV and Bal 31

I ...... I I I I I I I IIIIV /A






Small '

Small EcoRV +1

coRV +1 BamH1

EcoR1 Sima BamH1

Lac Z






amplification were performed for each reaction following initial denaturation at

95C for 5 minutes. Each cycle consisted of 95C denaturation for 1 minute,

annealed at the TH for each pair of primers for 1 minute, and 72C extension for 3

minutes. After 20 cycles of amplification an additional 10 minutes of extension

was performed to enable most products to have a common end. The 5' ends of

each primer pair were complementary to each other by four to ten nucleotides.

After the PCR reaction the products were phenol/chloroform extracted once,

precipitated, resuspended in water and used to transform competent E. coil by

electroporation (SURE strain, Stratagene). Upon transformation, circular

molecules were generated by the host E. coil by homologous recombination at

the termini of the PCR products because of their complementarity. The template

used was pSL123. Plasmid pSL123 has approximately 750 bp EcoRI fragment,

exercised from p5-498 and subcloned into pBluescript KS+ at the EcoRI site.

This EcoRI fragment contained all of the CIT1 sequences in p5-498 plasmid.

The CIT1 sequences consisted of upstream sequences, the TATA element, and

178 nucleotides long transcription unit, which included the first 26 codons.

Many transformants were usually obtained, therefore initial screening for

deletion mutants was done by the colony hybridization (Grunstein and Hogness,

1975). During the screening, 32P radiolabeled probes were prepared from the

mutant primer whose sequence should now be continuous in the recombinant

clone but discontinuous on the parent plasmid. The hybridization temperature

used for each screening was the TH of the oligonucleotide. Deletion mutants

were further characterized by restriction digestion, then positive clones were


sequenced to determine the exact deleted region. Because the deletions were

made in a pBluescript plasmid, the EcoRI fragment was recloned in the original

yeast/E. co/i shuttle plasmid it was obtained, replacing the full-length insert. After

ligation and subsequent transformation, recombinant plasmids were sequenced

to determine their orientation. Four regions were deleted that span -370 to -252,

-245 to -216, -200 to -160, and -370 to -160 (+1 indicates the start site for

transcription). All recombinant plasmids were transformed into S150-2B strain,

and 13-galactosidase activity was determined as described below. The primers

used were AL60/AL61 to delete -200 to-160; MS41/MS42 to delete -245 to -216;

MS43/MS44 to delete -370 to -252; and AL61/MS44 to delete -370 to -160.

Annealing temperature for each primer pair was: 1). AL60/61 at 51 C, 2).

MS41/MS42 at 47C, 3). MS43/MS44 at 55C, and 4). AL61/MS44 at 510C.

Heterologous Fusion

To show that sequences upstream of the putative TATA element of CIT1

could function as a UAS (upstream activating sequence), a 400 bp EcoRI-EcoRV

fragment from p5-498 was subcloned into plCZ312 (generous gift from Dr.

Meyers). The CIT1 upstream sequences were obtained from pSL123 plasmid

described earlier. The pSL123 plasmid was digested with EcoRI-EcoRV, which

released a 400 bp fragment that contains all of the putative CIT1 UAS and

recovered the DNA by the Spin-Bind method (Costar) after running on a 1 %

agarose gel. This fragment was ligated to plCZ312 plasmid that had been cut


with Smal- Xhol enzymes and filled-in with all four dNTP's using 2 U of Klenow

enzyme. The plCZ312 plasmid has the CYC1 UAS1 and UAS2, the TATA

element, and three nucleotides of CYC1 coding sequences fused to lacZ gene.

It also has E. coli replication origin, ampicillin resistance gene, and the URA3

selectable marker in yeast, but there is no yeast replication origin. Therefore,

the plasmid can be maintained only if it integrates into the yeast chromosome. In

order to direct the integration, the plasmid was digested at a unique Stul site

within the URA3 gene. Stul digested plasmid DNA was transformed into 1-7A

and JP16-8A(hap2::URA3) strains, and transformants were plated on SD(2%)

with 20 pg/ml histidine, 2.5 pg/ml adenine, and 20 pg/ml leucine and incubated at

30C. Several single colonies from each transformation were isolated and re-

streaked on similar plate and incubated at 30C again. Because multiple,

tandem integration events could occur, the number of integration of each

transformant was determined by Southern blot analysis. Yeast chromosomal

DNA was isolated by the mini-prep method and digested 10 pg with Sadcl

enzyme. The digested DNA was run on a 0.8% agarose gel in TBE (0.89 mM

Tris; 0.89 mM borate; 0.005 mM EDTA) and electrophoretically transferred onto

Zeta-bind nylon membrane (Bio Rad) using half strength TBE. Nucleic acid was

fixed onto the membrane by heating in a vacuum at 80C for 2 hours.

Prehybridization and hybridization were performed according to the manufacturer

(Biorad). The membrane was hybridized with 32P radio-labelled probe prepared

from YIp56 plasmid (gift from Dr. H. Baker's laboratory), which contains the

URA3 gene, using the random primer method using a kit from United States


Biochemical, Inc. Probe was denatured by boiling and added at 100,000 cpm

per milliliter of hybridization solution. Hybridization was carried out at 60C for 16

hours with shaking. Membrane was washed according to the manufacturer's

recommendation and exposed to X-ray film. Transformants with more than single

integration event were identified by the presence of a plasmid-length band.

Cloning of Oligonucleotides

Oligonucleotides corresponding to different upstream regions of CIT1

were subcloned into the plCZ312 vector to test their ability to either activate or

repress transcription. The oligonucleotides used were synthesized at the DNA

Synthesis Core, University of Florida. Their sequences and location in the gene

are given in Table 3. Approximately 5 pg of complementary oligonucleotides

were annealed in 10 mM Tris-HCI, pH 8.0; 5 mM MgCl2; 20 mM NaCI; by boiling

for 5 minutes then slowly cooled to room temperature. To clone the region

between -200 to -160, AL86 and AL87 oligonucleotides were annealed, and

AL84 and AL85 oligonucleotides were annealed for the -245 to -216 region. After

annealing 1 pg of each annealed oligonucleotide was end labeled with 1 mM

ATP using T4 polynucleotide kinase (1 U) in 500 mM Tris-HCI, pH 7.6; 100 mM

MgCl2; 5 mM DTT at 37C for 60 minutes. At the end of the reaction the enzyme

was heat inactivated by incubating at 75C for 10 minutes. Salt was removed

from the sample by precipitating with absolute ethanol, then the sample was

resuspended in 10 pl water. Approximately 4 pl of each annealed oligonucleotide


was ligated with plCZ312 vector which had already been digested with Smal and

Xhol restriction enzymes. This digestion removed the UAScycl. Ligation was

performed overnight at 16C in 1X ligase buffer (66 mM Tris-HCI, pH 7.6; 6.6 mM

MgCl2; 10 mM DTT; 66 pM ATP) with 1 U of T4 ligase. Ligation mixture was

used to transform into competent E. coil cells. Transformants were analyzed by

isolating plasmid and digesting with the Sphl restriction enzyme. Recombinants

were subsequently confirmed by sequencing using the Sequenase kit (US

Biochemical). DNA from confirmed recombinants was linearized with Stul

enzyme and transformed into yeast and plated on the appropriate selective agar

plate. Digestion of the DNA directs integration at the URA3 locus. The number

of integration was determined by performing Southern analysis as described


Measurement of 13-Galactosidase Level in CITI-lacZ Fusion

Cells were grown to either early logarithmic phase (OD60o ~ 1.0) or

stationary phase (OD60o > 20.0) and 10 ml of culture was harvested by

centrifugation at (2,780 X g) in a Sorvall (Dupont) desktop centrifuge for 5

minutes. The supernatant was decanted and the pellet was resuspended in 10

ml water. The cells were centrifuged again at (2,780 X g) in the Sorvall clinical

centrifuge for 5 minutes,and the supernatant was decanted. The pellet was

resuspended in 1 ml 10 mM Tris-HCI, pH 7.4 plus 1 pl 100 mM PMSF and

transferred the cells into round bottom 13 ml centrifuge tube (Starstedt). To


disrupt the cells, 4 mm diameter acid-washed glass beads were added to the

suspension until the glass beads reached the meniscus of the liquid. The

mixture was vortexed vigorously for 45 seconds. The tube was cooled on ice for

at least 1 minute; then vortexed again for additional 45 seconds. The lysate was

then transferred to fresh microcentrifuge tube and centrifuged in an Eppendorf

centrifuge for 1 minute at 4C. The supernatant was transferred to a fresh

microcentrifuge tube and incubated in a -70C freezer for at least 15 minutes.

The lysate was then set on ice to thaw and centrifuged again in a microcentrifuge

for 1 minute at 4C. The supernatant was transferred into a fresh microcentrifuge

tube and stored at -70C 13-galactosidase activity of each lysate was

determined by the method of Craven et al (1965). Hewellet Packard kinetics

program in model 8452A spectrophotometer was used to determine the reaction

rate. Specific activity from each sample is reported as nanomoles of o-

nitrophenyl-P3-galacotopyranoside (ONPG) hydrolyzed per minute per milligram of

protein. Protein concentrations were determined by the method of Lowry et al


RNA Isolation

RNA for northern analysis and ribonuclease protection assays was

routinely prepared as described by Schmitt et al (1990). Cell cultures were

grown to early logarithmic phase (OD600oo ~ 1.0) and harvested by centrifugation in

the Beckman J2-21 centrifuge in a JA-20 rotor at 10,000 rpm for 45 seconds.


Then the supernatant was removed and cells were resuspended in 400 pl AE

buffer ( 50 mM sodium acetate; 20 mM EDTA, pH 5.3). Cells were transferred to

a microcentrifuge tube, then added one-tenth volume (40 pl) 10% SDS was

added to each tube and vortexed for about 30 seconds. A 1.2 volume (480 pl)

prewarmed (65C) phenol/chloroform (1:1 ratio),equilibrated with AE buffer was

added, and the mixture was vortexed for 30 seconds. The tube was incubated in

a 65C water bath for 5 minutes with occasional vortexing, then cooled down by

setting in a dry-ice ethanol bath for approximately 10 seconds. The aqueous

phase was separated from the organic phase by centrifugation at 2500 X g in an

Eppendorf centrifuge at room temperature for 20 minutes. The organic phase

was discarded. An equal volume of prewarmed (65C) phenol/chloroform was

added again and the extraction repeated as above. A final extraction was

performed with an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1).

The aqueous phase was transferred to a fresh microcentrifuge tube. RNA was

precipitated by adding one-tenth volume 3 M sodium acetate, pH 5.3 plus 2.5

volume absolute ethanol and incubated at -20C for 30 minutes. The ethanol

pellet was recovered by centrifuging for 15 minutes in a microcentrifuge at 4C.

The supernatant was decanted, and the pellet was washed with 1 ml 70%

ethanol. The pellet was dried under vacuum and resuspended in 50 pI water.

RNA concentration was determined in spectrophotometer (Hewellet Packard

model 8452). Samples were stored at -70C until needed. When isolating RNA

for half-life or rate of decay determination, an RNA polymerase II temperature

sensitive mutant strain was usually used to isolate the RNA. Transcription was


stopped by adding equal volume of prewarmed (48C) medium to the culture,

and immediately transferring the culture to a 36C water bath. Chemical

inhibitors, such as thiolutin (generous gift from Dr. S. Kadin, Pfizer Inc. Groton

CT.) (Jimenez et al., 1973) and 1,10-phenanthroline (Santiago et al., 1986), were

also used to stop transcription in separate experiments. Thiolutin was dissolved

in DMSO and used at a final concentration of 3 pg/ml and 1,10-phenanthroline

was prepared in ethanol at 10 mg/ml and used at a final concentration of 100


Ribonuclease Protection Assay

To show that the P3-galactosidase activity from the different deletion

constructs reflects the steady-state mRNA level, ribonuclease protection assays

were performed on total yeast RNA isolated from yeast strains harboring

selected deletion constructs. A radiolabeled cRNA from the lacZ gene was

prepared from pSLOOl plasmid. pSLOOl plasmid was constructed by subcloning

a EcoRV/Clal fragment from p5-498 into pBluescript KS+ cut with the same

enzymes. This EcoRV/Clal fragment contains an 817 bp of the 5' coding region

of lacZ gene and 287 bp of CIT1 sequences which include the first 178

nucleotides of CIT1 RNA and the TATA element. To prepare the probe, the

plasmid was linearized with Ddel restriction enzyme, which cuts within the lacZ

gene, and transcribed with a T3 RNA polymerase at 37C for 1 hour. This

generated a probe that was approximately 300 nucleotides long. At the end of


transcription 1 pjl RQ1 DNase I (1 U/pIl) (Promega Corporation) was added, and

the sample was incubated at 37C for 15 minutes to digest the template DNA.

The volume was adjusted to 100 pI with water and extracted once with

phenol/chloroform/isoamyl alcohol. The aqueous phase was transferred,

extracted once again with chloroform and subsequently transferred to a fresh

tube, and an equal volume of 5 M ammonium acetate, pH 5.3 plus 2.5 volumes

ethanol were added. The transcript was incubated at -20C for 30 minutes to

precipitate. It was then centrifuged in an Eppendorf centrifuge at 12,000 rpm for

15 minutes to collect the precipitate. After decanting the supernatant, the pellet

was resuspended in 100 p1 2.5 M ammonium acetate and precipitation was

repeated two additional times to remove the unincorporated nucleotides. The

pellet was washed once with 70% ethanol, dried in vacuum and resuspended in

100 pl hybridization buffer. One microliter of the transcript was analyzed in a

scintillation counter. To hybridize, approximately 250,000 to 500,000 cpm per

probe was added to 10 pg of precipitated total RNA, and the volume was

adjusted to 30 p1 with the hybridization buffer. The mixture was heated at 85C

for 15 minutes, then quickly transferred to a 45C heat block and hybridized

overnight. Three hundred and fifty microliters of RNase digestion buffer

containing 40 pg/ml RNase A plus 2 pg/ml RNase T1 was added to each sample,

which was then incubated at 30C for 60 minutes to digest unhybridized

transcript. The RNase digestion was stopped by treatment with 2.5 pl 20 pg/ml

proteinase K; 20 pl 10% SDS at 37C for 15 minutes. The sample was extracted

once with equal volume phenol/chloroform/isoamyl alcohol and the aqueous


phase was transferred to a fresh microcentrifuge tube containing one microliter of

10 pg/pl yeast tRNA plus 1 ml absolute ethanol. Precipitation was carried out at -

20C for 30 minutes. Samples were centrifuged in an Eppendorf centrifuge for

15 minutes at 4C. The supernatant was decanted and the pellet washed once

with 70% ethanol, dried in vacuum and resuspended in 10 pl RNA sample buffer

(95% Formamide; 0.0025% bromophenol blue; 0.0025% xylene cyanol).

Samples were heated at 75C for 5 minutes and loaded in a 6% (19:1)

polyacrylamide gel. The bromophenol blue dye ran about two-thirds the length of

the gel before electrophoresis was stopped. The gel was dried and exposed to

X-ray film. Relative 32p content for each sample was quantitated by exposing

the dried gel to a Phosphor-lmager screen (ABI).

Northern Analysis

For northern analysis, 10-15 pg of total RNA was precipitated with one-

tenth volume 3 M sodium acetate, pH 5.3 plus 2.5 volumes of absolute ethanol

and centrifuged in a microcentrifuge for 15 minutes, and the supernatant was

decanted. The pellet was washed with 70% ethanol and dried in vacuum. Two

microliters of water were used to resuspend the pellet and 8.8 pl of sample mix

was added to the sample. Sample mix consisted of 50% formamide; 0.22 M

formaldehyde; 1 X MOPS (0.2 M MOPS; 0.05 M sodium acetate; 0.001 M

EDTA); 40 pg/pl ethidium bromide. The sample was heated at 65C for 15

minutes, chilled on ice for few minutes, then 1 pl dye mix was added and the


sample loaded on a 1.2% agarose gel containing 0.22 M formaldehyde/1 X

MOPS buffer. The running buffer consisted of 0.22 M formaldehyde/ 1X MOPS

buffer. The gel was run at 150 V until the dye ran to the bottom of the gel. This

usually took about 6 hours. The buffer was recirculated with a peristaltic pump to

prevent formation of a pH gradient. At the end of the run, the gel was

photographed with Polaroid film on a UV transilluminator to determine the

integrity of the RNA. The gel was then soaked in a 20 X SSC (SSC is 0.15 M

sodium chloride/0.015 M sodium citrate) for 15 minutes. RNA transfer onto

Hybond N+ nylon membrane by capillary action using 20X SSC for approximately

15 hours at room temperature. At the end of the transfer, the RNA was cross

linked to the membrane in a Stratalinker (Stratagene) set on auto-crosslink. The

membrane was then rinsed with 2 X SSC. The rapid hybridization solution

(Amersham) was used as recommended by the manufacturer for hybridizations

and prehybridizations. Prehybridization was performed with 50 pi of rapid

hybridization solution per square centimeter of membrane at 60C for at least 30

minutes. After prehybridization, 100,000 200,000 cpm of probe was added per

milliliter of hybridization solution, and hybridization was performed at 60C for at

least 2 hours. Washes were done with 2X SSC/0.1 % SDS at room temperature

for 15 minutes once and changed to 1 X SSC/0.1% SDS and repeated wash

twice at 60C for 20 minutes. The membrane was then air dried, wrapped in

Saran Wrap and exposed to X-ray film. An intensifying screen was used to boost

weaker signals. For quantitative results the membrane was exposed to a

Phosphor-lmager screen (ABI). If stripping was necessary, the membranes were


routinely stripped by adding 0.5% SDS at boiling temperature, then set at room

temperature until the solution cooled to room temperature. After stripping, the

membrane was then re-exposed to X-ray film to make sure that the were no

residual bands from previous hybridization before subsequent hybridizations was

performed on the membrane.

Primer Extension Analysis

Two oligonucleotides were used in the primer extension analysis to map

the 5' ends of the chromosomal-initiated CIT1 mRNA and the CITI::IacZ fusion

mRNA that were being transcribed from the plasmid. The oligonucleotides were

first end labeled using T4 polynucleotide kinase as recommended by the

manufacturer, New England Biolabs. An end labeled oligomer (250,000 cpm)

and 50 pg of total yeast RNA were first precipitated together using ethanol. The

pellet was resuspended in 30 pl hybridization buffer (40 mM PIPES, pH 6.4; 1

mM EDTA, pH 8.0; 0.4 M sodium chloride; 80% Formamide). This mixture was

heated at 85C for 10 minutes, then immediately transferred to 25C heat block

and hybridized overnight. Following hybridization, 150 pl of water plus 20 pi 3 M

sodium acetate, pH 5.2 were added to the sample. Nucleic acid was

precipitated with 2.5 volumes ethanol. The pellet was washed with 70% ethanol

and allowed to air dry. Pellet was resuspended in 10 pl sterile distilled water, 4 pl

5X Reverse transcription buffer (250 mM Tris-HCI, pH 7.9; 375 mM potassium

chloride; 15 mM magnessium chloride), 2 pl 0.1 M DTT; 2 pl 10 mM each all four


deoxynucleotides(dNTP's), 1 pIj RNasin (26 U/jpl) (Promega Corporation), and 1

pI Superscript II reverse transcriptase (200 U/pl) (Life Technologies). The

sample was incubated at 37C for 90 minutes, then 1 pl 0.5 M EDTA was added

to stop the reaction. The sample was then treated with 1 pl 5 mg/ml RNase A to

digest unhybridized RNA at 37C for 30 minutes. The mixture was extracted

once with phenol/chloroform/isoamyl alcohol, then adjusted to a final

concentration of 2.5 M ammonium acetate and precipitated with 2.5 volume

ethanol. The pellet was first resuspended in 4 pl TE pH 8.0, then 6 pl formamide

loading buffer was added. The sample was heated at 95C for 3 minutes and

loaded on a 5% Longer Ranger gel (AT Biochem) and run until the lower dye had

run two-thirds the length of the gel.

To identify the start sites, the same end labeled primer (AL41) was used to

prime DNA sequencing reactions on pCSB plasmid, which consists of an EcoRV-

EcoRV CIT1 fragment that includes the TATA element and 5' half of the coding

region. AL215 was used to prime plasmid specific transcript. The sequencing

reaction was performed as recommended by the manufacturer (US Biochemical).

In Vivo Footprinting

In vivo footprinting was performed essentially as described by Giniger et al

(Giniger et al., 1985) with some modification. One liter of culture was grown in

either YPE or YPD to early logarithmic growth phase (OD60o). Cells were

harvested by centrifuging in a JA-10 rotor (Beckman) at 5,000 rpm at room


temperature for 5 minutes. The supernatant was decanted and the pellet

resuspended in 10 ml of growth medium. The suspensions were then divided

into ten 1 ml aliquots in an Oakridge centrifuge tubes. Two microliters of

concentrated dimethyl sulfate (DMS) was added to each aliquot, which were held

at room temperature for varying amounts of time, ranging from 2 minutes to 10

minutes. At the end, 40 ml ice cold TEN (10 mM Tris-HCI, pH 8.0; 1 mM EDTA;

and 40 mM sodium chloride) solution was added to each to stop the reaction.

Cells were centrifuged at 5,000 rpm in a JA- 20 rotor at 4C for 5 minutes and

supernatant was decanted. Pellets were resuspended in 1 ml 1 M sorbitol/0.1 M

EDTA plus 2 l P113-mercaptoethanol. To form spheroplasts, 200 pl 5 mg/ml

mureinase (1509 BGX units/g, US Biochemical) was added to each cell

suspension. These were incubated at 37C with gentle shaking until

spheroplasts were formed. This usually took approximately 30-40 minutes.

Spheroplast formation was determined by adding 50 pl of cell suspension to 500

pl 0.1% SDS and measuring the change in absorbance at OD6oo. Reduction in

absorbance by 90% was considered an acceptable level before DNA isolation

was performed. Spheroplasts were collected by centrifuging for 1 minute,

decanting the supernatant and resuspending the pellet in 1 ml 50 mM Tris-HCI

pH 8.0/20 mM EDTA. The sample was divided into two halves and transferred

into microcentrifuge tubes. Fifty microliters of 10% SDS were added to each half

which was incubated at 65C for 30 minutes to lyse spheroplasts. Two hundred

microliters 5 M potassium acetate, pH 8.0 were added to each and samples were

incubated on ice for 60 minutes. Samples were centrifuged sequentially for 10


minutes and 5 minutes in a microcentrifuge and the supernatants of both

centrifugations were pooled. Isopropanol (700 pl) was added to each and

centrifuged for 20 seconds to collect theDNA pellet. The pellet was rinsed with

95% ethanol, decanted supernatant and allowed to air dry. Each pellet was

resuspended in 300 pI of TE pH 8.0. The divided samples were pooled and

treated with 10 pul 10 pg/pl RNase A at 37C for at least 2 hours to digest RNA.

Aliquots (3 pI) of 1 M spermidine, pH 7.0 were added to each sample until a DNA

precipitate appeared. It usually took about 2-3 aliquots to precipitate DNA.

Samples were set on ice for 15 minutes and centrifuged 20 seconds to collect

DNA. Supernatants were decanted and pellets were allowed to air dry. The

DNA pellet was dissolved by adding 300 pl 3 M ammonium acetate to the pellet

and incubating at 65C for at least 4 hours. Absolute ethanol (750 pl) was added

and each sample was held at -70C for 15 minutes to precipitate the DNA. DNA

was collected by centrifuging for 15 minutes in a microcentrifuge and decanting

the supernatant. The pellets were rinsed with 70% ethanol, dried in vacuum and

resuspended in 200 pl water. The DNA was digested with either Accl or EcoRV

to analyze the coding strand or the noncoding strand, respectively. After

digestion, the DNA was precipitated and washed three times to remove residual

salts. First, one-half volume of 7.5 M ammonium acetate was added to the

sample plus 1.5 volume ethanol. Samples were set in a dry-ice ethanol bath for

10 minutes to precipitate, then centrifuged for 15 minutes in a microcentrifuge

and the supernatants were decanted. Pellets were resuspended in 200 pl TEN

(10 mM Tris-HCI, pH 8.0; 1 mM EDTA; 100 mM sodium chloride), then 100 pl 7.5


M ammonium acetate plus 400 pl ethanol was added. This precipitation was

repeated and the pellet was washed in 1 ml 70% ethanol. To cleave the DNA,

10 pl concentrated (10 M) piperidine (Fisher Biotechnology) was added to the

sample to make a final concentration of 1 M. The sample was transferred to a

screw cap tube and incubated at 95C for 30 minutes. At the end of the

incubation, the tube was chilled on ice for few minutes and lyophilized overnight

in the Speed Vac (Savant). The pellet was resuspended in 250 pl 0.3 M sodium

acetate, pH 5.3. Then 3 volumes of ethanol was added and the sample was

placed in a dry-ice ethanol bath. DNA was collected by centrifugation in an

Eppendorf centrifuge. The pellet was resuspended in 200 pl 0.3 M sodium

acetate, pH 5.3, and the ethanol precipitation was repeated. The final pellet was

rinsed with 70% ethanol and dried in vacuum. Each pellet was resuspended in 5

pl of sample dye and loaded on a 6% polyacrylamide/50% urea gel. The gel was

run at 60 watts constant power in TBE buffer (0.89 M HCI; 0.89 M borate; 0.005

M EDTA) until the bromophenol blue dye reached the bottom of the gel. The gel

was picked up with a Hybond N+ (Amersham) membrane precut to the size of

the gel and presoaked in TBE, the transfer buffer. The DNA was transferred onto

the nylon membrane by electroblotting, using the Genesweeper instrument

(Hoefer Scientific) as recommended by the manufacturer. After the transfer, the

nucleic acid was crosslinked to the membrane on a transilluminator for 10

minutes. Prehybridization was performed with 25 ml of hybridization solution,

which consisted of 7% SDS; 0.5 M sodium phosphate, pH 7.4; 1% BSA; 1 mM

EDTA, at 60C for at least 60 minutes. Radiolabeled DNA probe was then added


and hybridization carried out overnight at 60C with shaking. Radiolabeled probe

was prepared by primer extension on a single-stranded DNA template with a

complementary oligonucleotide using the Klenow fragment of DNA polymerase I.

After hybridization, the membrane was washed three times at the hybridization

temperature. The wash solution consisted of 1 % SDS; 40 mM sodium

phosphate, pH 7.4; 1 mM EDTA.

Preparation of Single-Stranded DNA

Single-stranded DNA was prepared from pSL123 and pSL123R, to serve

as a template in generating probes for in vivo footprint analysis. These two

plasmids contain the upstream sequences of p5-495 in opposite orientations at

the EcoRI site on a phagemid plasmid. The single stranded DNA generated from

either plasmid would be complementary to either the coding (pSL123R) or

noncoding (pSL123) strands of CIT1. An isolated colony of XLI Blue strain

containing the plasmid was grown in 2.5 ml of super broth containing 12.5 pg/ml

of tetracycline and 100 pg/ml of ampicillin overnight at 37C with vigorous

shaking. Two and half milliliters of the overnight culture was added to 50 ml

super broth in a 500 ml flask and grown until the OD60o reached 0.3. VCS-M13

helper phage was added at an MOI (multiplicity of infection) of 20:1 and

incubation continued for an additional 8 hours. The culture was heated at 65C

for 15 minutes and centrifuged at 9,500 rpm in JA20 rotor at room temperature.

The supernatant was transferred to a new tube and centrifuged again as above.


One-fourth volume of 3.5 M ammonium acetate, pH 7.5; 20% polyethylene glycol

(PEG) 8000 was added to the spheroplast. The tube was inverted several times

to mix the sample, which was held at room temperature for 45 minutes. The

pellet was collected by centrifugation sequentially at 9,500 rpm at room for 20

minutes and 1 minute; discarding the supernatant each time. The pellet was

resuspended in 15 ml TE, pH 8.0 and 7.5 ml phenol/chloroform (1:1) was added.

The mixture was vortexed for 1 minute and centrifuged at 11,500 rpm for 5

minutes in the JA20 rotor at room temperature. The aqueous phase was

transferred to a fresh tube and the extraction repeated until no interphase was

present. This usually took four extractions to accomplish. Then 10 ml chloroform

was added and the sample vortexed 1 minute and centrifuged at 11,500 rpm in

JA20 rotor for 5 minutes. The aqueous phase was transferred to a fresh tube and

one-third volume 7.5 M ammonium acetate (final concentration, 2.5 M) was

added. 2.5 volumes absolute ethanol were added and the sample was incubated

on ice for 40 minutes to precipitate. The sample was then centrifuged at 9,500

rpm in the JA20 rotor at 4C. The supernatant was decanted, and the pellet was

dried and then resuspended in 400 pl TE pH 8.0.

Bandshift Assayl/In Vitro Footprinting Analysis

To determine if sequences upstream of the TATA element were involved

in direct protein/DNA interaction, bandshift assay and in vitro DNase I protection

assay were performed to identified such regionss. For the bandshift assay, a


single end radiolabeled probe from the upstream sequence was prepared and

incubated with an extract of yeast cells at room temperature for 20 minutes. It

was then run on a 4% nondenaturing polyacrylamide (40:1) gel in TBE at room

temperature at 100 V, until the bromophenol blue dye had migrated to the

bottom. The gel was dried under vacuum at 80C and exposed to X-ray film.

The cell extract for the assay was prepared as follows: A cell culture was grown

to early logarithmic phase (OD60o ~ 1.0) and harvested by centrifugation at 7,500

rpm in a JA10 rotor (Beckman) for 10 minutes at 4C. The cell pellet was

resuspended in 10 ml extraction buffer(0.2 M Tris-HCI pH 8.0; 0.4 M ammonium

sulfate; 10 mM magnesium chloride; 1 mM EDTA; 20% glycerol). Cells were

washed by resuspension in the same buffer and repeated centrifugation. Three

milliliters of extraction buffer plus 2 mM PMSF; 0.5 mM DTT; and 1 pg/ml

pepstatin were added per 1 g of wet weight of cells. Cells were disrupted by

passing through a French Pressure Cell at 20,000 psi three times. The

homogenate was then centrifuged at 17,000 rpm (35,000 X g) for 45 minutes in

the JA 20 rotor at 4C. The supernatant was aliquoted into microcentrifuge tubes

and stored at -70C. Protein concentrations were determined as described

earlier. Two probes were used for the bandshift assays: (1) -406 to -216

fragment and (2) -245 to -111 fragment. These two probes together span the

entire region of the upstream sequences of p5-498, which contains all of the

presumptive CIT1 UAS. The probes were prepared by using oligonucleotides

AL82/AL85 (Table 3) to generate the -406 to -216 fragment in a PCR reaction,

and primers AL84/AL104 to make the -245 to -111 probe. Standard PCR


reaction was performed as described above. The annealing temperatures for

AL82/AL85 and AL84/104 were 47C and 51 C, respectively. One of each pair of

primers was end labeled with T4 polynucleotide kinase before the PCR reaction

to make sure that only one end was labeled. The two probes were used instead

of a single probe that encompasses the entire region, because preliminary

experiments showed that a single fragment alone would not migrate into a 4%

(39:1) nondenaturing polyacrylamide gel very well. Similar PCR products as

described earlier or double stranded oligonucleotides were used for competition

assay. The oligonucleotides were annealed by mixing equimolar amounts of

each strand in 10 mM Tris-HCI, pH 8.0/5 mM MgCl2. The reaction mixture was

boiled for 5 minutes and allowed to slowly cool to room temperature.

To identify the exact sequences involved in the bandshift assay, a DNase I

protection assay was performed on both sets of probes. After the standard

bandshift assay, 1 pl 20X DNase buffer and 1 pl 1 U/pl RQ1 DNase I (Promega

Corporation), were added to the reaction mixture and permitted to digest for 45

seconds. The reaction was then stopped by adding 1 pl 0.5 M EDTA. The

sample was loaded into a 4% (39:1) polyacrylamide gel and run as described

above. At the end of the run, the wet gel was then exposed to an X-ray film

overnight at room temperature. The bands were then excised and DNA eluted

onto a DEAE cellulose membrane in an agarose gel. The DNA was recovered

from the DEAE membrane by incubating it at 65C with high NET (1.0 M sodium

chloride; 0.1 mM EDTA; 20 mM Tris-HCI, pH 8.0) for 45 minutes. The eluate was

transferred into a fresh microcentrifuge tube and precipitated with 1 ml absolute


ethanol at -70C for 30 minutes. Pellet was resuspended in 200 pl water and 500

pI absolute ethanol were added. DNA was precipitated at -70C again for 30

minutes and centrifuged at maximum speed for 15 minutes. The supernatant

was decanted and the pellet rinsed with 1 ml 70% ethanol. The pellet was dried

in a Speed-Vac and resuspended in 5 p1 of sequencing dye solution. The sample

was then loaded on a 6% polyacrylamide gel (19:1) and run at 1000 V until the

bromophenol blue had migrated two-thirds of the length of the gel. Control

reactions were performed by digesting naked DNA with DNase I. Sequence

ladder was generated by using the primer that was end labeled in a dideoxy

sequencing reaction. The gel was dried and exposed to X-ray film.

Messenger RNA Stability (5' UTR deletion) Assay

The 5' untranslated region and the coding region on p5-498 plasmids were

individually dissected to determine their effects on the stability of the CIT1::IacZ

fusion mRNA. I used RCPCR to delete these regions essentially as described

above for the pSL123 plasmid. After the deletion, the EcoRI fragment was

subcloned into p5-498, which was then transformed into yeast strains. I used

primers AL189 and AL190 pair to delete the 5' untranslated region. The first

nucleotide, adenine, of the major transcriptional start site was retained. The rest

of the sequences remained essentially the same. A similar strategy was also

used to delete the coding region present in the CITI::IacZ fusion. The primers

used were AL205 and AL206 (Table 3). In this construct, the deletion was


created in such a way as to retain the first codon of CIT1, which was joined in the

proper reading frame to the lacZ gene. After the deletion, these constructs were

sequenced to determine their new adjourning sequences and to be certain that

no other mutations were introduced in this region during the PCR reaction. The

annealing temperature for ALl 89/190 was 53C and AL205/AL206 was 63C.

The schematic of how transcription was terminated is presented in Figure 3.

Introduction of Stop Codon at the Fifth Amino Acid Position in the CIT1 Gene

To introduce a stop codon at the fifth position on the CIT1 portion of the

CITI:: lacZ fusion plasmid, the TransformerTM Site-Directed Mutagenesis Kit

(Clontech) was used as recommended by the manufacturer. Briefly, two primers

were designed called the Selection primer and the Mutagenic primer, to prime

synthesis of a complementary strand by T4 DNA polymerase after initial

denaturation of the template, pSL123R. The selection primer introduced a

mutation at a unique Scal site that converted that recognition site into a new

unique Stul site. This enabled screening of putative mutants easily by digestion

of the putative recombinants with the newly created Stul site. The nicks on the

newly synthesized complementary strand were sealed by T4 DNA ligase. The

ligated DNA was then used to transform an E. coil strain that is deficient in DNA

mismatch repair, a mutS strain. This allowed the amplification of both mutated

and unmutated plasmid. Transformants were grown in LB broth for several

generations, then I isolated plasmid DNA. The pool of DNA obtained was


digested by Scal, which linearized wild type plasmid but left plasmids with a

mutated Scal site in the circular form. This pool of DNA was then used to

transform E. coil a second time. Because the linearized wild type plasmids were

not as efficient as the circular mutated plasmids in transforming E. coli, the

majority of the transforming DNA had now lost the Scal site, but had obtained the

Stul site. Those clones with the Stul site were sequenced by the dideoxy

sequencing method to identify those that had also acquired the T to G

transversion at position +389. Clones that were verified to have the correct

mutations were subsequently subcloned into the p5-498 plasmid, at the EcoRI


Table 1. E. co/i Strains.

Name Genotype
HBII0 supE44, ara14, galK2, lacY1, proA2, rpsL20, xyl-5, mtl-1,
recA13, A(mcrC-mrr), HsdS-(r-m-)
C600 e14-(mcrA), supE44, thi-1, thr-1, leuB6, lacY1, tonA21
BMH71-18 mutS thi, supE, A(lac-proAB), [mutS::TnIOJ[F' proAB,
XL 1-Blue recA1, endA1, gyrA96, thi-1, hsdR17, supE44, relAl, lac,
~~________ [F'proAB, laclqZAM15, TnlO(tetf)]
SURE e14-(mcrA), A(mcrCB-hsdSMR-mrr)171, sbcC, recB, recJ,
umuC::Tn5(kanr), uvrC, supE44, lac.gyrA96,relA1, thi-
~~________1,endA 1 [F'proAB, lacPZIAM 15, Tn 1O, (tel')]

Table 2. Yeast Strains.

Name Genotype Source
S150-2B MATa, his3A200, leu2-3, 112, trpl-289, ura3- H. Fukuhara
1-7A MATa, adel-100, his4-519, leu2-3, 2-112, P. Srere
JP16-8B MATa, adel-100, ade2, leu2-3, leu2-112, P. Srere
SHY40 MATa, adel-100, leu2-3,2-112, ura3-52, J. Pinkham
SLF401 MATa, adel-100, his4-519, leu2-3, 2-112, L. Guarente
ura3-52, hap4::LEU2______
Z118 MATa, ade2, leu2-3,112, his3A200, rpbl-1, R. Young
ura3-52 ______
Z118URA3 MATa, ade2, leu2-3, 112, his3A200, rpbl-1, This work
~______ura3-52, trpl::URA3______

Table 3. Oligonucleotides Used in this Research.

Name Gene Position Sequence 5' to 3'
160 to-151 G
_______ __~C
-371 to -288

Table 3 continued

Name Gene Position Sequence 5' to 3'
to 117
178 to 190
-216 to -207 G
_____-252 to -243 CT

Table 4. Plasmids Used in this Research.

Designation Construction
pSHI18-8 Approximately 970 bp of CIT1 sequence, consisting of 78 bp
of coding region and sequences further upstream in pUC18
(New England Biolab) cut with Smal
YCpZ-2 A yeast/E coil shuttle vector used to carry all deletion
constructs. See Rickey (1988)
p5-498 Approximately 670 bp of CIT1 sequence contained in
pSH18-8 following exonuclease digestion was subcloned in
YCpZ-2 cut with BamHI/Smal.
pSLOO1 A 1.1 kb EcoRV/Clal from p5-498 that consists of 817 bp of
5' lacZ sequence and approximately 290 bp of CIT1
sequences from -111 to +78, subcloned into pBluescript KS+
(Strategene) cut with EcoRV/Clal.
pSL123 Approximately 670 bp EcoRI fragment from p5-498
consisting entirely CIT1 sequence sucloned into pBluescript
KS+ cut with EcoRI.
pSL123R Similar to pSL123, the insert is in reverse orientation.
pGEM-Actin A 563 bp internal Clal fragment from ACT1 gene subcloned
into pGEM 4 (Promega) cut with same enzyme (generous
gift from Dr. R. Butow).
plCZ312 Generous gift from Dr. Alan Myer.
YISL101 Approximately 380 bp of CIT1 5' upstream sequence from
pSL123 subcloned into plCZ312 cut with Smal/Xhol.
YISL101R Similar to YISL101 R; the insert is in reverse orientation.
YISLI 11-139X Double stranded oligonucleotide corresponding to sequences
between -139 to -111 was annealed and subcloned into
plCZ312 digested with Xhol.
YISLA1 -99 Similar to p5-498 except that sequences between 1 to 99 bp
of CIT1 sequence have been deleted.
YISLA100-178 Similar to p5-498 except that sequences between 100 to 178
Sbp of CIT1 sequence have been deleted.

Table 4 continued.

Designation Construction
YISLSTOP Similar to p5-498 except that a T to G transversion mutation
was introduced at position 114 of the CIT1 sequence.


Analysis of 5' (Distal) and 3' (Proximal) Deletions

Studies by Hoosein and Lewin (1984) showed that there were increased

amounts of CIT1 translatable mRNA in cells grown in glucose medium as they

approach stationary growth phase and begin to utilize ethanol as a primary

carbon source. Logarithmic-growth phase cells grown in ethanol contained more

CIT1 mRNA than log phase cells from a glucose culture. This observation

suggested that the increased steady-state level of CIT1 mRNA may be due to an

increased rate of transcription or to a decreased rate of turnover in ethanol

medium. My initial focus was to determine the boundaries on the sequences

upstream of the transcriptional initiation site that may contribute to the high-level

expression of CIT1 in ethanol medium or glucose depleted medium. The

decision to characterize the UAS (upstream activating sequence) elements was

based in part on the presence of a perfect match to the nine nucleotide

(TNATTGGT) consensus binding site for the trimeric protein

Hap2p/Hap3p/Hap4p. This protein complex has been well characterized and

shown to be required for high-level expression of genes encoding proteins in the

mitochondrial electron transport chain (Trueblood et al., 1988; Trawick et al.,


1990), TCA cycle genes (Bowman et al., 1992; Gangloff et al., 1990; Repetto and

Tzagaloff, 1990), and heme biosynthesis (Keng and Guarente, 1987).

The strategy employed to identify regulatory elements involved fusion of a

fragment of DNA from the CIT1 with the E. col lacZ gene. The CIT1 segment

contained some of the coding sequences and the transcriptional regulatory unit,

which consisted of the TATA element, transcriptional start site and the putative

UAS elements. Sequential deletion of the putative UAS element was performed

in a manner that deletion progressed toward or away from the transcriptional

start site. Deletions that progressed toward the transcriptional start site were

designated 5' (distal) deletions (Figure 1), and those that progressed away from

the transcriptional start site were designated 3' proximall) deletions (Figure 2).

The lacZ gene was used as a reporter gene for several reasons. First, deletion

of the CIT1 gene in yeast causes slower growth (Kispal et al., 1988), which may

lead to pleiotropic effects on other metabolic processes; therefore, the native

CIT1 gene on the chromosome was left intact. Second, there is another citrate

synthase isozyme encoded by the CIT2 gene (Kim et al., 1986; Lewin et al.,

1990), which partially compensates for CIT1 deletion (Kim et al., 1986). This

makes assaying for citrate synthase activity of a CIT1 gene on a plasmid

impractical. Therefore, in order to determine the effect of the upstream sequence

deletions on CIT1 gene expression, the reporter gene was used. The gene for 13-

galactosidase was used, because there is no similar activity in yeast; therefore,

any enzyme activity detected would be from the plasmid carrying the gene and

under the control of the CIT1 promoter elements. The vector plasmid, YCpZ-2


(Rickey, 1988) contains CEN4 and ARS1 sequences which were necessary for

the plasmid to replicate and be maintained at approximately a single copy per

yeast cell. It also has a TRP1 marker for selection in yeast, an origin of

replication for E. co/i and the bla, ampicillin resistance, gene for selection in E.

coil. Recombinant plasmids were transformed into the S150-2B yeast strain and

selected for transformants on SD (2%) medium containing 10 pg/ml histidine, 20

pg/ml leucine, and 5 pg/ml uracil. Total cellular extract was prepared from at

least two different isolates of each transforming plasmid and its P3-galactosidase

activity determined in triplicate as described in materials and methods.

Figure 3 shows the specific activities obtained from the selected deletion

mutants. CIT1 mutants were named according to the end-points of the deletion

relative to the major transcriptional start site. Deletions from the 5' end of the

gene that extended toward the transcriptional start site did not show any

significant reduction in specific activity until they extended beyond position -498.

Hence, the activity of this clone was designated wild-type level, and all other

clones were compared to it to assess the effect of deletion. There was about a

27-fold induction when the p5-498 clone carrying strain was grown in an ethanol

medium compared to when grown in glucose. This is somewhat higher than the

induction of citrate synthase activity usually observed in yeast, suggesting that 13-

galactosidase may be more stable in yeast than citrate synthase. However, it

should be noted that the level of induction of citrate synthase between

derepressing and repressing media varies considerably between strains. Further

removal of sequences to position -245 caused only a slight decrease in 13-

Figure 3. 13-Galactosidase Activity of 5' and 3' Deletions in Complex
Medium. (A). The top line represents the CIT1 sequence from which all the 5'
and 3' deletions were derived. The number of each deletion construct represents
the deletion end-point. Yeast cells were grown in complex media supplemented
with glucose (YPD) or ethanol (YPE) to early logarithmic phase. Glucose and
ethanol were added to 2% weight/volume. The specific activity of 13-
galactosidase is presented as nanomoles of ONPG hydrolyzed per minute per
milligram of protein in the lysate. Each value represents the average of triplicate
assay from two to three transformants and differed from each by no more than
10%. (B). A diagramatic interpretation of the results. The hatched areas
represent putative UAS and the dotted area represents a putative URS region.






-3 72




"1 r
-172 -111

7 -111


-11 1

,-800 AT: G
I I I I 1 I
;-400:S ^






-00 '

-800 OJ /


ATG Specific Activity

70 2017

57 1938

9 1183

23 626

163 2933

138 1797

104 670

8 232

1 46











6 c




galactosidase level expressed from this clone, p5-245, in a derepressing

medium, but produced about 20% reduction in a repressing, glucose, medium.

Consequently, there was a higher fold induction of the p5-245 clone than the p5-

498 clone. Removal of additional 19 base pairs (p5-227) caused a severe

reduction in 13-galactosidase level expressed in a repressing medium while

activity was reduced only about 50% in a derepressing medium. The overall

induction, YPE/YPD, of p5-227 was 131-fold, which was almost seven times the

fold induction for p5-498. In clone p5-168 the specific activity in glucose and

ethanol was reduced to about 33% of wild-type level. Therefore, the level of

induction of ethanol versus glucose media was about the same as the wild-type


Deletions beginning from the 3' or proximal end of the upstream sequence

resulted in a range of specific activities from 2.5 times greater than the wild-type

insert in a repressing medium (YPD) to a barely detectable level. In the

derepressing medium, YPE, the activity ranged from 150% of wild-type level to

about 2% of the wild-type level. In clone p3-139, the sequence from -139 to -111

was deleted. In this clone, specific activities were consistently higher than wild-

type in cells from both media, suggesting that an upstream repressing sequence

(URS) may have been removed. Rosenkrantz and his colleagues (1994) also

found that removing sequences in this region caused an increase in 13-

galactosidase level expressed from a CITI::IacZ fusion gene. A deletion that

stopped at position -172 had a specific activity that was nearly 90% of wild-type

in the YPE medium but almost 2.5 times greater than wild-type in cells grown in


the YPD medium. Removal of an additional 35 bp, clone p3-217, reduced

specific activity by two-thirds in YPE, but still maintained a slight increase (104

units/mg protein) over wild-type level in YPD. Clone p3-252 expressed 13-

galactosidase level that was approximately 10% of wild-type in both YPD and

YPE media. Further deletion to position -372 produced barely detectable

enzyme activity in cells carrying this clone in YPD medium and only about 2% of

the wild-type level in YPE medium. These 5' and 3' deletions showed that there

are three regions that caused increase of CIT1 expression in YPD and YPE

media. These regions include sequences between -372 to -252, -245 to -216

and -200 to -160. In addition sequences between -139 and -111 have a

repressing effect that is most pronounced in the YPD medium.

There have been two reports in yeast (Kim et al., 1986; Gangloff et al.,

1990) and one in B. subtilis (Rosenkrantz et al., 1985) which showed that

addition of glutamate to cultures grown in a minimal medium repressed the

activity of citrate synthase and aconitase. 13-galactosidase levels expressed in

yeast cells bearing the deletion constructs were identical, whether cells were

grown in SD(2%), or SD(2%) medium supplemented with glutamate (Rickey,

1988). However, it was observed that the 13-galactosidase levels expressed from

most of these deletion constructs were substantially higher in a minimal medium

with 2% dextrose than in the YPD (2%) medium. The results of the specific

activities obtained from cells grown in the SD medium harboring the deletion

constructs are presented in Figure 4. The wild-type clone produced 12 times

more 13-galactosidase activity in SD(2%) (897 units/mg protein) than in YPD(2%)

Figure 4. 13-Galactosidase Activity of 5' and 3' Deletions in Minimal
Medium. Sections (A) and (B) are as described in Figure 3. P3-galactosidase
assays were performed from cultures grown in SD(2%) as described in figure 3.


ATG Specific Activity
um SD(2%)




-245 __1________________________

-168 I

I 72
-172 -111



, -400
'I \\\\\N\\\ N -\N NNX






pITAA +1


1 /Z







L- o


1 ,


(^ +1^
r --- A t


(70 units/mg protein). This large effect, was lost when sequences up to -245

were deleted. Deletion constructs beginning from the 3' end of the CIT1 upstream

sequences showed significant reduction in 13-galactosidase expressed from them

when deletion extended up to position -252. There was no increase in specific

activity from clone p3-139 compared to the wild-type clone. This would suggest

that the sequences between -139 and -111 are involved in regulation of CIT1 in

response to glucose but not in response to minimal medium. There are two

putative core GCN4 consensus-binding sites that lie between -374 and -369 and

-268 and -263, which may be responding to growth in a minimal medium. GCN4

is a transcriptional activator involved in the regulation of genes in the amino acid

biosynthesis pathway (reviewed in Hinnebusch, 1988). Reduction in expression

from clone p5-227 to 5% of wild-type and clone p3-372 to less than 1% of wild-

type suggest that all the necessary sequences for regulation in a minimal

medium lie in this region. In clone p5-227, the two putative GCN4 binding sites

were deleted, whereas in clone p3-372 the proximal site was deleted and the

distal site was disrupted.

The results of these 5' and 3' deletions showed that several regions,

between sequences -370 and -252, -245 and 216 and -200 and -160 in the

upstream sequence, that contribute to transcriptional regulation of this gene. It is

clear that these sequences have activating functions because when deleted they

reduced specific activities, but the sequence between -139 and -111 has a

repressing effect in response to glucose. The sequences necessary for

regulation in a minimal medium lie between -370 and -227.


Internal Deletions Show Several Putative UASs

The results of the exonuclease deletions of the CIT1 nontranscribed

region suggested that there was more than a single element which may be

involved in the regulation of this gene. In yeasts and other eukaryotes,

regulatory elements such as UASs and proximal regulatory elements, such as

Spl and AP1 sites, may lie in close proximity to each other. Therefore, there is

increased likelihood that more than one regulatory element may be removed

during exonuclease digestion. To test for this, individual regions were dissected

from the upstream sequences by inverse PCR. The regions deleted include

positions: 1) -160 to -200, 2) -216 to -245, 3) -252 to -370, and 4) -160 to -370.

These regions were chosen for the internal deletion analysis because the results

of the directional deletions from the proximal and distal ends suggested that they

may be important in regulating this gene.

In clone pA1 60-200, the sequences between -160 and -200 were deleted.

This region of CIT1 upstream sequences has the consensus binding site for the

Hap2p/Hap3p/Hap4p transcriptional activator protein at -192 to -185. Several

laboratories have shown that the Hap2p/Hap3p/Hap4p activator is involved in

regulation of many genes for mitochondrial proteins, including CYC1 (Olesen et

al., 1987; Guarente and Mason, 1983), HEM1 (Keng and Guarente., 1987),

COX6 (Trueblood et al., 1988), KGD2 (Repetto and Tzagoloff, 1990), LPD

(Bowman et al., 1992), and ACI01 (Gangloffet al., 1990). Removal of this region

reduced the specific activity of 13-galactosidase by one-third, to approximately


658 units per milligram protein in the derepressing medium, YPE (Figure 5).

However, in the YPD medium the specific activity obtained for this deletion was

100 units per milligram protein, which is higher than the wild-type level. The

second region deleted were sequences from -245 to -216 with respect to the

transcriptional start site. Specific activity was reduced to 3.5 and 664 units/mg

protein in glucose and ethanol media, respectively. This represented a 20-fold

reduction in the glucose medium but only a 3-fold reduction in the ethanol


The third region deleted were the sequences from -370 to -252, and the

plasmid was named pA252-370. Removal of this region, which was 119 bp in

length, drastically reduced the specific activity of P3-galactosidase. In a

derepressing medium the specific activity was reduced to 79 units per milligram

protein, which was approximately 25-fold lower than the wild-type level.

However, in a repressing medium the reduction was more severe, lowering

activity nearly 50-fold relative to the wild-type level in a similar medium. The

fourth internal deletion constructed encompassed all the other three deletions

previously described, from positions -370 to -160. There was approximately 24

units per milligram protein of specific activity detected in clone pA160-370 in a

depressing medium, which was nearly 85-fold lower than the wild-type activity in

a similar medium. In YPD medium the specific activity for this construct was only

0.7 units, which is 100-fold lower than the wild-type activity. The fold induction in

a derepressing medium versus a repressing medium was 34 times in cells

harboring clone pA160-370. This level of induction reflects the very low activity

Figure 5. 3-Galactosidase Activity of Internal Deletion Constructs. (A).
Internal deletions were constructed by using inverse PCR with primers that
surround the region of interest. The CIT1 sequences present in the wild-type
clone was used as the template in the PCR. The HAP symbol represents the
region on the CITI1 sequence that has the consensus site for the transcriptional
activator HAP2/3/4. p-galactosidase activity represents the average from
triplicate assay from at least two transformants. (B). A schematic interpretation
of regions with UAS activity.


-800 -600 -400 -200 +1. ATG



I I-1


-498 -245 -216

-498 -370 -252
-498 -370 -160
-498 -370 -160

Specific Activity

70 2017 26

100 658 7

3.5 664 190

1.4 79 56

0.7 24 34



from this construct in YPD, confirming that the response to glucose can be

mediated by sequences outside this region (e.g. the URS from -139 to -111).

The results obtained with p5-245 and pA252-370 seem at odds because the

region deleted from the pA252-370 clone was also deleted from the p5-245

clone, yet activities remain relatively high in p5-245, especially in derepressing

media. The sequences between -498 and -370 were present in pA252-370, but

were deleted from clone p5-245. This result suggests that the -498 to -370

region may contain a URS.

There are Multiple UAS Elements

Upstream activating sequences have, by definition, the ability to activate

the transcription of their cognate gene or a heterologous gene in an orientation

independent manner. The activation may show regulation in the heterologous

context similar to that in the native gene. To show that the whole upstream

sequence present in p5-498 and segments of it have either a UAS or URS

function, the entire region or smaller regions were subcloned into plasmid

plCZ312 whose UAS elements were removed by digestion with restriction

enzymes Xhol and Smal. The important features of this plasmid were the

presence of the CYC1 promoter elements UAS1 and UAS2, the TATA element,

the transcriptional start site, and only three nucleotides of the coding region fused

to the lacZ gene. UAS2 has seven of eight nucleotides of the consensus

sequence that binds to the Hap2p/Hap3p/Hap4p transcriptional activator


complex. Use of this plasmid would allow direct comparison of the putative

UASCITI to UAS2cycl regulation by the Hap2/3/4 activator complex mentioned

earlier, since they both have the consensus site for the activator.

The plCZ312 plasmid was digested with Xhol and Smal enzymes and

filled-in with the Klenow fragment of E co/i DNA polymerase I in the presence of

5 mM of all four deoxynucleoside triphosphates. The CIT1 sequence was

generated by cutting p5-498 plasmid with EcoRI and EcoRV enzymes, filling-in

with Klenow enzyme and recovering a 388 bp fragment that contained the entire

regulatory region. The CIT1 fragment was cloned into the plCZ312 vector and

both forward- and reverse-orientation recombinants were recovered. These

clones were designated YISL101 and YISL101R. These plasmids and the intact

plCZ312 plasmid were transformed into 1-7A and JP16-8B (hap2). JP16-8B was

a derivative of 1-7A by insertion of the URA3 gene at the HAP2 locus to disrupt

the gene (Pinkham and Guarente, 1985). 13-galactosidase activities from each

transformant was determined as described in materials and methods.

The results of these experiments are shown in Figure 6 and 7. The

YISL101 R clone produced approximately the same level of specific activity as

plCZ312 in both repressing and depressing media. However, with YISL101 the

specific activity was nearly one-half the amount of UAScycI in YPD. The

difference in specific activities between YISL101 and YISL101 R was observed in

a repressing medium but was less pronounced after glucose has been depleted

in stationary phase (Figure 6) or when cells were grown in ethanol (Figure 7).

When the UASCIT containing plasmid was transformed into a hap2 strain, the

Figure 6. UAS Activity of CIT1 5' Untranscribed Region. The entire 5'
untranscribed region of CITI present in the "wild-type" clone, p5-498, was
subcloned before a CYCI::IacZ fusion deleted of its native UAS. 13-galactosidase
assays were performed from cultures grown in YPD. 1-7A is a wild-type yeast
strain and JB16-8B is a hap2 mutant derivative of 1-7A. Activities are presented
as described in the legend to Figure 3. ND= P-galactosidase assay was not

Specific Activity




-4JAS1-UAS -----TATA CYC1::lacZ
>,, .>







"y ,

43 93

,bc UAScnr 9.-,- TATA, CYC:.:IacZ

SUASc-r, T
-TATA- CYC1::lacZ
UA'" f ""A -y>:ea





101 367

70 303





Figure 7. UAS Activity of Various CIT1 Upstream Sequences. A 41-mer
representing sequences from position -200 to -160 and a 30-mer representing
sequences from position -245 to -216 were subcloned before a CYCI::IacZ
fusion without its native UAS. P3-galactosidase assays were performed as
described in Figure 3.


-UAS1-UAS2-, TATA CYC1::lacZ
,7 >




Specific Activity





I-> TATA CYC1:.I:lacZ

(41 -mer) -TATA CYC1::lacZ

(3 0 m er) -, TATA- CYC1::lacZ






specific activity of YISL101 was slightly higher than the specific activity of

YISL101R (Figure 6). But UAScycl driven expression was very low during

logarithmic phase growth and was only about 66% of the UASCITn after glucose

had been depleted. While the expression from the UASCITIn in YISL101 was

reduced 42% in shifting from the HAP2 to the hap2, strain the activity of the

UAScycI dropped more than 99% in the hap2 mutant. The results indicate that,

while the hap2 mutation has an effect on the UAScycl, this effect is much less

significant than on the CYC1 gene.

The three internal regions deleted from CIT1 upstream (see Figure 5)

showed that they each contributed to the transcriptional regulation of the gene.

In order to test their contribution to activation these sequences were individually

subcloned into plCZ312 vector whose UAS1 and UAS2 had been removed as

described earlier. Only two of these regions were successfully cloned. The

YISL160 clone was derived by annealing two complementary oligonucleotides,

AL86 and AL87, and lighting the result into the vector. These oligonucleotides

span the region -200 to -160, which includes the Hap2p/Hap3p/Hap4p binding

site. YISL216 was cloned by annealing AL84 and AL85 which spans -245 to -

216. I also attempted to clone the region that encompasses the region between -

370 to -252 but was not successful.

The results from these experiments are presented in Figure 7. YISL160

produced specific activity of 47 units per miligram of protein in a culture that was

grown to logarithmic phase in the YPD medium. Cultures harvested from YPE

medium had 1252 units/mg protein. The specific activity of YISL216R was 718


units per milligram of protein in a depressing medium and 94 units per milligram

of protein in YPD. These results show that the regions encompassing -245 to -

216 and -200 to -160 have an activating function. The -245 to -216 region has

greater activation potential than the -200 to -160 region in YPD. However, in the

YPE medium, the -200 to -160 region showed greater expression than the -245

to -216 region, suggesting that this region contains part of the glucose

responding promoter element.

Evidence for URS Element

Under both repressing and derepressing conditions, the level of 3-

galactosidase expressed from clone p3-139 was higher than the wild-type clone

(see Figure 3). 13-galactosidase levels were more than twice as high under

repressing conditions, yet under derepressing conditions the levels were only

50% higher than the wild-type clone. This suggested that there may be a URS

element between -139 to -111 of the CIT1 sequence. To determine the potential

negative regulatory capability of this region, I cloned it into the reporter plasmid

plCZ312. To accomplish the cloning, plCZ312 was linearized with Xhol which

cuts downstream of the two UAS elements described earlier (see introduction),

and the region of CIT1 from -139 to -111 was inserted. Recombinants

designated YISL111-139X were sequenced using AL45 primer to determine the

orientation of insertion. Only recombinants in the forward orientations were

recovered and subsequently transformed into a yeast strain. If the -139 to -111


region has a URS function, the P-galactosidase levels expressed from such a

construct should be lower than the levels expressed from intact plCZ312.

Specific activities from this construct are presented in Figure 8. In a

repressing medium, the activity was reduced approximately 50% compared to

plCZ312 level. The same level of reduction was also seen from cultures grown

in YPD to stationary phase, when most of the glucose has been depleted and

thus represents a depressed state. However, when the cells were grown in

ethanol-supplemented complex medium there was no significant reduction in an

specific activity. This suggested that either repression of transcription occurs

only in a glucose-containing medium or that the UAS activity in a YPE medium

masks the repressing effect. However, there was still about 10% reduction of

specific activity in YPE. Even in the context of CIT1 sequences, the increase

after this region was removed was greater in a YPD medium at logarithmic phase

than it was in a YPE medium. Similar observations were made by Rosenkrantz

and coworkers (1994). However, it is possible that increasing the distance

between the UAS elements and the transcriptional start site of the CYC1 gene

could decrease the level of expression. This could come about by placing the

UAS site and the TATA site on opposite sites of the DNA, thereby hindering

proper contacts between these factors to allow activation. Insertion of an

unrelated oligonucleotide of similar length at the same site in the plCZ312 vector

should distinguish a distance effect and specific sequence effect.

Figure 8. URS Activity of -139 to -111 Region of the CIT1 Gene. The -139 to
-111 region from CIT1 upstream sequence was subcloned into the CYCI::IacZ
fusion downstream of the native CYC1 UAS. 13-galactosidase assays were
performed as described in Figure 3.


Specific Activity






-- TATA CYC1::IacZ

,-UAS1-UAS2( URScrr) ",-e TATA CYC1::IacZ
^ 'p













Steady-State mRNA Levels Correlate with Enzyme Assay

The P3-galactosidase activities from the various promoter deletions strongly

suggested transcriptional regulation of the CIT1 gene affecting the steady-state

mRNA level. In order to correlate the enzyme activities with the steady-state

mRNA levels, total yeast RNA was isolated from selected strains and the level of

lacZ specific message was determined by ribonuclease protection assay.

Radiolabeled complementary RNA (cRNA) specific to the lacZ gene was

generated from plasmid pSL00l using T3 RNA polymerase. The actin message

from ACT1 gene was used as an internal control to correct for possible loading

differences. The cRNA for the actin mRNA was transcribed from pGEM-actin

plasmid, a generous gift from Dr. R. Butow, using SP6 RNA polymerase.

The results are shown in Figure 9. Figure 9a shows the autoradiogram of

the ribonuclease protection assay, and Figure 9b shows the quantitative result.

The quantitative results were obtained by calculating the ratio of lacZ mRNA

versus ACTI mRNA for each sample. Each sample was then compared to the

p5-498 mRNA level to measure their relative level of expression. Overall, the

amount of lacZ mRNA from each deletion paralleled the 13-galactosidase activity

observed for each construct. However, the YPE/YPD ratios were lower for the

steady-state mRNA levels than enzyme levels for each of the constructs tested.

Liao et al (1991), also found that in certain yeast strains the YPE/YPD ratios for

citrate synthase activity were as much as four times higher than the steady-state

mRNA produced from the CIT1 gene in identical media. Surprisingly, the mRNA

Figure 9. Steady state mRNA level of selected deletion constructs. (A) 15
pg of total RNA was hybridized in solution to radiolabeled cRNA probes for lacZ
and ACTI genes simultaneously. After hybridization, samples were digested
with RNase A and RNase T1 and resolved on 6% Long Ranger gel (AT
Biochem). Lane M is end labeled RNA molecular weight marker (1.77 to .155
kb) (Life Technologies). Lanes 1-8 are samples from selected deletion
constructs as shown. In lane 9, RNA was isolated from strain that does not
harbor any plasmid carrying the lacZ gene. In lane 10, RNA was isolated from
an isogenic strain that carry a plasmid bearing TPI::IacZ fusion (Courtesy Dr. H.
Baker). (B) Graphical representation of the net cpm of each sample obtained by
exposing the gel to a Phosphor-lmager screen (Molecular Dynamics).


D E D E D E D E E E Medium
0.400- E
) 0) r0 r- Go CO 1O O. S 0
C C- n No 0 L N, P0asmid
L CL. L. Q. a Q. CL Q- c C.


B 1000



< 400


p3-139 p3-227 p5-498 p5-245



level for p3-139 was slightly lower than the p5-498 level both in repressing and

derepressing media even though a higher level of P-galactosidase activity was

detected (Figure 3). At present, no explanation satisfactorily accounts for this

discrepancy between the two methods of determining transcriptional efficiency.

Measuring P-galactosidase activities reflects both transcriptional and translational

effects and may possibly amplify small differences in RNA level. However,

Rosenkrantz et al., (1994) showed that 13-galactosidase levels increased when

the sequences between this region were deleted. This would suggest that the

enzyme assay may be more reliable than the quantitative result of the steady-

state mRNA transcribed from the same plasmids. Lane 9 of Figure 9a was RNA

isolated from a yeast strain without the plasmid construct that has the lacZ gene.

This confirms that there is no other gene in yeast that hybridizes to the E coli

lacZ probe. The sample in lane 10 (Figure 9a) was isolated from a yeast strain

transformed with pES90 plasmid, a generous gift from Dr. H. Baker's laboratory.

pES90 plasmid has the triose-phosphate isomerase (TPI1) gene fused to the

lacZ gene.

Band Shift Assay and In Vitro Footprint Analysis

To map the sequences that are involved in regulation of the CIT1 gene by

an independent method, both bandshift assays and in vitro footprint analysis

were performed. The bandshift assays were performed to see if there are

proteins from total yeast extract that can bind to a DNA fragment containing the


CIT1 upstream region. Footprint analysis, using DNase I, was used to identify

the sequences that bind to the proteins.

The CIT1 upstream sequence present in the wild-type construct, p5-498,

was divided into two regions to perform these assays because preliminary

assays showed that the DNA fragment of the entire upstream was too large to

migrate into the 4% polyacrylamide gel used.

Extracts from yeast grown in YPD, YPE or SD (2% glucose) media were

used for the binding assays with the different fragments. The results of the

binding experiments are shown in Figures 10 and 11. Each fragment was shifted

in response to crude yeast extract. For fragment -245 to -111 (Figure 10), two

shifts were observed: band "A" at all concentrations of extract and band "B"

(lower band) appearing only at high levels of extract. The appearance of a

second band at high protein concentration may mean that the affinity between

the protein and DNA is low, requiring a high concentration of the protein to be

present before binding is detected. Unlike fragment -245 to -111, the second

fragment, extending from position -406 to -216, gave only one shifted band even

at high extract level (Figure 11). There was no difference in binding patterns

observed amongst the extracts of cultures from YPD, YPE or synthetic medium

supplemented with 2% glucose (Figure 11).

To show that the binding observed with the different fragments was a

specific interaction, a competition reaction was performed with unlabeled DNA

fragments. The unlabeled DNA used was either identical to the labeled probe or

from other region of the CIT1 upstream sequence. Figure 12 shows the result of

Figure 10. Bandshift of -245 to -111 fragment. Approximately 12 pg of crude
yeast extract from glucose-grown cells was incubated with approximately 2 fmole
of end-labeled 135 bp CITI1 DNA fragment. There was no extract in lanes 1 and
11. Lanes 2 through 10 had increasing amounts (1 pI to 9 pl) of extract.