Identification of amino acid-starvation induced mRNAs in Fao rat hepatoma cells


Material Information

Identification of amino acid-starvation induced mRNAs in Fao rat hepatoma cells
Physical Description:
Shay, Neil Frank, 1954-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 25072934
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Figures
        Page vi
        Page vii
        Page viii
        Page ix
        Page x
    Chapter 1. Introduction
        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
    Chapter 2. Identification of mRNAs in Fao rat hepatoma cells induced by amino acid starvation
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        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
    Chapter 3. Monitoring of mRNA levels during amino acid deprivation of cells
        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
    Chapter 4. Characterization of a starvation-induced mRNA
        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
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
    Chapter 5. Conclusions and future directions
        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
    Biographical sketch
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
Full Text







To Betsy and Laura.


I would like to take this opportunity to thank my

committee members: Drs. Allen, Boyce, Ferl, and Laipis for

their time and guidance regarding this project, and

especially Drs. McGuire and Ostrer for their insight

provided in many helpful discussions. Thanks also go to my

comrades in the lab, Barrie Bode and Rohit Cariappa, for

their support and companionship, as well as Mary Handlogten

for being a willing teacher. Thanks go to Joy Doran, Xiao-

fang Deng, and Susie Zoltewicz for assistance with some of

the experiments presented here. A special thank you goes to

my friend, John Berceann, who has helped me in many ways

during my stay in the biochemistry department. Thanks go to

Dr. Harry Nick for the time he has spent in discussions

regarding this project, and a special thank you goes to Dr.

Michael S. Kilberg for his support of me and this project.

Special thanks go to my wife, Betsy Moore-Shay, and my

parents, Frank and Rose Shay, for their love and support

throughout the years.



ACKNOWLEDGEMENTS ....................................... iii

ABBREVIATIONS .......................................... v

LIST OF FIGURES ........................................ vi

ABSTRACT ............................................... ix


1 INTRODUCTION ...................................... 1


Introduction ...................................... 18
Materials and Methods ............................. 23
Results ........................................... 42
Discussion ........................................ 50

DEPRIVATION OF CELLS .............................. 55

Introduction ...................................... 55
Materials and Methods ............................. 58
Results ........................................... 71
Discussion ........................................ 95


Introduction ...................................... 102
Materials and Methods ............................. 105
Results ........................................... 119
Discussion ........................................ 146


REFERENCES ............................................. 161

BIOGRAPHICAL SKETCH .................................... 172


bp base pairs
BSA bovine serum albumin
cDNA complementary DNA
DNase deoxyribonuclease
DTT dithiothreitol
EDTA ethylenediaminetetraacetic acid
FBS fetal bovine serum
IPTG isopropylthio-B-D-galactopyranoside
kb kilobases
MOPS 3-(N-morpholino) propanesulfonic acid
PAGE polyacrylamide gel electrophoresis
RNase ribonuclease
SDS sodium dodecyl sulfate
SDS-PAGE polyacrylamide gel electrophoresis with
sodium dodecyl sulfate
SSC standard sodium citrate buffer
SSPE standard sodium phosphate/EDTA buffer
TBE Tris/boric acid/EDTA buffer
TE Tris/EDTA buffer
X-gal 5-chloro-4-bromo-3-indolyl B-D-

Figure pagie

1-1 Autoradiograms of Fao cellular protein for 6
hours with or without amino acids following
Two-Dimensional Polyacrylamide Gel
Electrophoresis .................................. 15

1-2 In vitro translation products following One-
Dimensional SDS-PAGE ............................. 17

2-1 Diagram illustrating the general protocol
involved in identification of an mRNA induced
by amino acid starvation ........................ 24

2-2 Diagram illustrating principle
of differential hybridization ................... 25

2-3 cDNA library construction ....................... 35

2-4 Hydroxyapatite chromatographic subtraction of
single-stranded cDNA ............................. 39
2-5 cDNA elution profile of hydroxyapatite
chromatography ................................... 45
2-6 Insert size characterization of cDNA libraries 48

2-7 Differential Hybridization ...................... 51

3-1 Subcloning of the Fao hepatoma (S-5)
ASI cDNA insert .................................. 73

3-2 Northern analysis of ASI mRNA induction ........ 74

3-3 Induction of ASI mRNA by amino acid starvation
of rat hepatoma cells ............................ 78

3-4 Abundance of several mRNAs during amino acid
starvation of rat Fao hepatoma cells ............ 80

3-5 Refeeding effects on ASI expression in
hepatoma cells ................................... 82
3-6 Quantitation of actin and ASI mRNA levels after
starvation of Fao hepatoma cells for 12 hours .. 83

3-7 Quantitation of ASI mRNA induction in amino
acid-free media with or without vitamins ....... 87

3-8 Effect of metabolic inhibitors on the induction
of ASI mRNA in Fao cells ........................ 88

3-9 Effect of amino acid analogs L-azetidine-2-
carboxylic acid and L-histidinol on ASI
mRNA induction ................................... 91

3-10 Amino acid specificity of ASI mRNA repression
in Fao cells ..................................... 92

3-11 Nuclear run-off assay for Fao cells incubated in
amino acid-free or -supplemented medium ........ 94

4-1 Molecular size determination of the ASI mRNA .. 121

4-2 Primer extension of the ASI mRNA ................ 124

4-3 Anchored polymerase chain reaction ............. 125

4-4 Rat tissue distribution of ASI mRNA ............ 127

4-5 Relative abundance of ASI mRNA in rat Fao
hepatoma cells and normal rat liver ............ 128

4-6 Localization of the ASI mRNA to cytoplasmic
ribosomes in rat liver ......................... 130

4-7 Sequencing strategy of the ASI cDNA ............ 133

4-8 ASI cDNA and putative protein sequence ......... 134

4-9 Map of the rat ASI cDNA sequence ............... 136

4-10 Predicted amino acid sequence of
the rat ASI protein ............................. 137

4-11 Predicted secondary structure of
the ASI protein ................................. 138

4-12 Predicted hydrophilic plot of
the rat ASI protein ............................. 139

4-13 In vitro transcription and translation ......... 142

4-14 In vitro translation of rat hepatoma ASI mRNA 143

4-15 Rat genomic Southern analysis of the ASI gene 144

4-16 Southern "Zoo" analysis of the ASI gene ........ 145


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



August, 1990

Chairman: Dr. Michael S. Kilberg
Major Department: Biochemistry and Molecular Biology

Total and subtracted lambda gtll cDNA libraries were

derived from rat Fao hepatoma cells which had been starved

for amino acids for 3 hours. These libraries were screened

using the method of differential hybridization or "plus-

minus" screening to identify cDNA clones corresponding to

mRNAs induced by amino acid starvation. One 632 bp cDNA

identified by the screening was characterized. The mRNA of

the ASI (Amino acid ptarvation-induced) gene was induced up

to three-fold after 12 hours of incubation in amino acid-

free Krebs-Ringer bicarbonate buffer. The relative

abundance of several other mRNAs such as actin and

glyceraldehyde-3-phosphate dehydrogenase decreased during

the same period of starvation. Both cycloheximide and

actinomycin D treatment during starvation blocked the

induction of the ASI mRNA. The half-life of the ASI mRNA in

the presence of actinomycin D was determined to be

approximately 7.5 hours in both the fed and starved

conditions. Incubation of cells in amino acid-supplemented

medium containing the proline analog L-azetidine-2-

carboxylate caused the ASI mRNA to be induced less than 2-

fold. The ASI mRNA was present in every rat tissue tested.

The corresponding full-length cDNA was sequenced, with no

significant homologies to any sequences contained in the

GenBank data base. Southern blot data showed the ASI gene

to be probably single copy in human, chicken, alligator, and

yeast. Southern analysis of several rodent DNAs tested

yielded a middle-repetitive pattern, believed to be due to a

rodent-specific repetitive element or "retroposon" present

in the ASI DNA sequence. The full-length protein coding

sequence, obtained by the anchored polymerase chain reaction

method, was ligated into a pGEM-9Z plasmid vector and

synthetic mRNA was made in vitro. When this mRNA was

translated in a rabbit reticulocyte in vitro translation

system, electrophoretic analysis showed the synthesis of an

approximately 22 kDa polypeptide, which compares well with

the 21.4 kDa predicted weight based on the amino acid

composition derived from the cDNA sequence.



Although examples of regulation by substrates are

plentiful in prokaryotic systems, there are few well

documented systems in mammalian cells or tissues that

illustrate regulation at the gene level by substrate

availability. The first and foremost example of nutrient

regulated transcription is the lac operon of E. coli,

elucidated primarily by Jacob and Monod (Pardee et al,

1959). The binding of lactose to a repressor protein

releases genetic repression of the lac operon, allowing

increased transcription of the lactose permease and B-

galactosidase. The protein products allow the cell to

respond to changes in lactose availability. When lactose

concentrations fall as the cell utilizes the sugar, binding

of the repressor protein to the DNA regulatory region is

increased, thus re-establishing the repression of the lac

operon gene expression. Also in bacteria, depletion of a

single amino acid leads to increased transcriptional

activity of those genes responsible for the enzymes in the

cognate pathway. For example, bacteria adapt to starvation

of L-tryptophan or L-histidine by elevating levels of

expression of all the enzymes responsible for the

biosynthesis of the appropriate amino acid (Miller and

Reznikoff, 1978).

In contrast to the specific regulation by the substrate

of the lactose or amino acid operons of E. coli, a more

general control mechanism is seen in the yeast Saccharomvces

cerevisiae. Expression of many genes encoding enzymes of

several different amino acid biosynthetic pathways increases

in response to amino acid starvation. The term general

control of amino acid biosynthesis, or "gcn", has been used

to describe the cross-pathway character of this response.

Over 30 enzymes in nine different amino acid biosynthetic

pathways have been shown to be regulated by starvation of

yeast cells for any one of at least 10 amino acids

(Hinnebusch, 1988). The biosynthetic pathways that are

regulated by substrate include tryptophan, arginine,

histidine, lysine, leucine, and methionine. The response to

amino acid starvation in yeast is rapid, with elevated

steady-state levels of controlled enzymes being established

in less than an hour when, for example, cells are shifted

from a rich to a minimal medium (Hinnebusch, 1986). Since

the control of the gcn response has been shown to be at the

level of transcription, and the genes controlled by the gcn

response are unlinked in the yeast genome, the general

control system has proved to be a useful model for the study

of coordinate regulation of unlinked genes in a eucaryotic

organism. In the last five years, it has become clear that

the coordinate regulation of the gcn system depends on a

short nucleotide sequence (5'-TGACTC-3') found in the

upstream region of each gcn-regulated structural gene (Arndt

et al., 1987).

It remains unclear what the exact mechanism is for

detecting the amino acid-starved condition in yeast.

Depletion of an amino acid pool is sufficient to derepress

the expression of amino acid biosynthetic enzymes, but it is

not a necessary condition to elicit the general control

response. This has been demonstrated in the mutant yeast

strain ilsl, which exhibits low levels of isoleucyl-transfer

RNA (tRNA) synthetase activity along with low levels of

isoleucyl-tRNA. With normal cellular concentrations of

isoleucine present, this cell line displays elevated

expression of enzymes in at least four amino acid

biosynthetic pathways that are normally derepressed by

isoleucine starvation in wild type cells (Messenguy and

Delforge, 1976). This derepression seen in the ilsl strain

is blocked by a mutation in the GCNI gene, a positive

regulatory gene that is required for gcn-mediated

derepression (Niederberger et Al, 1983). This suggests that

a reduction in the amount of charged tRNA or the reduced

rate of protein synthesis is more likely to be the detected

signal for derepression rather than the depletion of an

amino acid pool.

The HIS4 gene is a member of the histidine biosynthetic

pathway, and shows a three- to four-fold derepression in

response to amino acid starvation. This derepression is

completely dependent on the GCN4 gene (Hope and Struhl,

1986). GCN4 encodes a trans-acting transcription factor

that is the most positive regulator of transcription in the

gcn response. Not only is the increase in transcription of

the HIS4 gene dependent on the GCN4 protein, but two-thirds

of the basal level of transcription of HIS4 is GCN4

dependent as well. Deletion studies along with DNA

sequencing of the 5' end of the HIS4 gene have identified

six sets of 12 base pair sites containing the GCN4 binding

sequence. Along with those six sites is a single 14 base

pair region that is required for normal basal expression of

the HIS4 gene (Lucchini et _4l, 1984). Other well

characterized gcn-controlled regulatory regions are

contained within the HIS3 and TRP5 genes. The HIS3 gene has

7 sites in the 5' region where GCN4 may bind as well as an

additional region necessary for basal expression (Struhl,

1982). The TRP5 regulatory region has been shown to contain

two GCN4-associated binding regions along with a single

region essential for basal expression (Zalkin and Yanofsky,


Although GCN4 is known to be the most positive

regulator involved in the gcn-mediated response, it should

be noted that there are a total of 9 GCN genes involved in

the response, and a recessive mutation in any of the nine

will block derepression of enzymes regulated by general

control. There also exists a set of genes termed GCD, that

function to maintain repression under nonstarvation

conditions, probably by acting as negative regulators of the

GCN4 and other GCN proteins. GCNI, GCN2, and GCN3 serve as

positive regulators of GCN4; this regulation is at the level

of translation (Mueller et al., 1987). To confirm this, the

5' noncoding sequence of the GCN4 gene was replaced with the

transcriptional upstream activation site (UAS) of the GALl

gene. Although this construct now makes GCN4 transcription

dependent on the presence of galactose, GCN4-directed

derepression in response to starvation is normal (Mueller et

al., 1987). The mechanism of translational control of GCN4

is dependent on four upstream AUG codons. These start

codons are followed by one or two sense codons and then an

in-frame termination codon. Deletion of these short reading

frames in the 5' non-coding region results in constitutive

gcn-directed derepression, even though the abundance of the

GCN4 mRNA was essentially unchanged by the deletion (Thireos

et al., 1984). A heterologous transcript was constructed

from the 5' upstream GCN4 region followed by a hybrid GALl-

lacZ fusion gene. This heterologous gene was found to

exhibit the characteristic gcn derepression in response to

amino acid starvation (Mueller et Al., 1987). Point

mutations that disrupted the four AUG codons resulted in

constitutive derepression of the gene construct (Mueller and

Hinnebusch, 1986). The GCD1 gene is the major factor that

interacts with the 5' region of the GCN4 mRNA. The

translational control between the four upstream AUG codons,

the GCD1 gene, and three other positive regulators GCN1,

GCN2, and GCN3 all interact to control the production of

GCN4 protein, which in turn controls gcn-mediated


In contrast to the general control system of yeast that

has been worked out in great detail, only a few mammalian

systems have been identified that show regulation by

metabolites or substrates. One of the inherent difficulties

in studying substrate-dependent control in higher organisms

is the fact that in vivo a change in metabolite

concentration can cause changes in enzyme activity via a

complex hormonal or neural process rather than by direct

transcriptional or translational control by the substrate

itself (Morley et al., 1988). Resolution of this problem

must be accomplished by the use of in vitro studies to

corroborate any results seen in vivo.

Expression of 3-hydroxyl-3-methylglutaryl coenzyme

reductase (HMG CoA reducatase) has been found to be

regulated at the gene level by sterol (Chin et al., 1985).

Assays for mRNA abundance and mRNA transcription have shown

that the presence of an oxysterol acts to reduce the

transcription rate of the WMG CoA reductase gene (Goodridge,

1987). Sequences responsible for mediating the promotion

and inhibition of transcription have been identified (Osborn

et Al., 1985). At least two elements upstream from the

transcription start site are necessary for the basal

transcription of the HMG CoA reductase gene (Osborn et al.,

1987). These two regions are 85 bp and 30 bp upstream from

the initiation codon.

Glucose has been shown to have a direct effect on a few

genes, the best example being the two glucose-regulated

proteins (grp's). One of these proteins is a component of

the endoplasmic reticulum and may be serving some function

relating to the regulation of glycosylation of proteins in

this compartment (Munro and Pelham, 1986). Deprivation of

glucose causes a 10- to 20-fold increase in the steady-

state levels of mRNA for these proteins (Lin and Lee, 1984).

Pyruvate kinase of liver is a key enzyme in the

glycolytic pathway whose activity and mRNA levels in the

liver fluctuate according to dietary status. mRNA levels of

this gene are controlled by both glucose and insulin in

cultures of plated hepatocytes. Increases in glucose

concentration from 5mM to 40mM were shown to raise pyruvate

kinase mRNA levels about 20-fold, even while insulin levels

remain constant (Decaux et al., 1989).

Deprivation of D-glucose will cause an up-regulation of

the facilitative D-glucose transporter in rat brain glial

cells (Walker pt al., 1988). This regulation is seen both

as increased transport activity for D-glucose as well as a

4- to 6-fold increase in the abundance of the glucose

transporter mRNA. The induction is complete 6 to 12 hours

after the onset of glucose deprivation. The mechanism(s)

behind these examples is still unknown.

A variety of metals have also been shown to be able to

transcriptionally activate genes. Zinc, cadmium, copper,

and mercury all can activate one or more of the

metallothionein genes. This induction has been seen in

hepatic tissue (McCormick et al., 1981), in the kidney

(Swerdel and Cousins, 1982), and the induction has been

shown to be due to a transcriptional activation (Durnam and

Palmiter, 1981). The 5' regulatory region of the

metallothionein-IIA gene has been sequenced and

characterized (Karin et al., 1984) and in this case, there

are two "metal ion-responsive elements," or MRE's, located

about 140 and 40 bp upstream from the translational start

site. These sites are binding sites for proteins that

mediate the response to metal toxicity, and when these

proteins are bound to the MRE, then transcription of the

gene is enhanced.

Liver is the major site of gluconeogenesis in the

mammalian organism. As a result, transport of gluconeogenic

precursors by this tissue is a necessary process and, in

fact, is the rate-limiting step for hepatic gluconeogenesis

from alanine (Fafournoux et al., 1983). The general

properties of transport systems as well as the phenomenon of

adaptive regulation (Gazzola et al., 1972) have been

investigated. Adaptive regulation is a process by which

cells increase their transport rates for certain amino acids

when confronted with low extracellular concentrations of

these amino acids (Riggs and Pan, 1972; Gazzola et al.,

1972). This response is mediated primarily through

enhancing the activity of the amino acid transporter System

A, which was discovered over 25 years ago in the Ehrlich

ascites tumor cell (Oxender and Christensen, 1963). This

adaptive response in cultured hepatocytes and hepatoma cells

is first seen 60 to 90 minutes after cells are placed in an

amino acid-free medium (Kilberg et al., 1985). The

translation inhibitor cycloheximide (Handlogten et al.,

1982), the transcription inhibitor actinomycin D (Kelley and

Potter, 1978), and N-linked glycoprotein inhibitor

tunicamycin (Barber et al., 1983) prevent the increased

System A transport activity of adaptive regulation. The

prevention of stimulated System A transport in adaptive

regulation by tunicamycin suggests that System A is a

glycoprotein or, at least, that its activity is dependent on

the de novo synthesis of a glycoprotein.

Collectively, the data are consistent with a model of

adaptive regulation that proposes that a gene coding for a

plasma membrane protein, perhaps the System A carrier

itself, is more actively transcribed (derepression), and the

resulting mRNA transcripts are translated to make more

membrane protein (Handlogten et al., 1982). This hypothesis

is supported by kinetic experiments showing that during

adaptive regulation the Vmax of System A transport increases

rather than a change in Km for the substrate (Gazzola et

al., 1972), suggesting an increase in the number of
functional transporters. Furthermore, the fact that one can

isolate plasma membranes containing stimulated System A

activity from amino acid starved cells argues for additional

carrier synthesis (Fong et _l., 1990). Although there may

be other more complicated explanations for the inhibitor and

transport results, we believe the simpler working model of

increased transcription of the System A gene and production

of additional System A carrier protein to be a sound basis

for further experimention.

Along with the increase of System A transport seen

during amino acid deprivation, there is an adaptive

repression of the enhanced transport activity in the

presence of amino acid-containing medium (Gazzola et al.,

1972). Amino acids may inactivate stimulated System A

activity by either a protein synthesis-dependent

(repression) or protein synthesis-independent ("trans-

inhibition") mechanism (Kilberg et al., 1985; Bracy et al.,

1986). Although there are exceptions (Boerner and Saier,

1985; Englesberg and Moffett, 1986), there are also a number

of amino acids, the branched chain and aromatic amino acids

as examples, that have no regulatory effect on System A

activity when added back to amino acid-starved cells. Other

amino acids, such as histidine and glutamate exhibit an

inhibitory effect that appears to be due mostly to the

protein synthesis-independent mechanism called trans-

inhibition (Bracy et Al., 1986). It has been found that

only the amino acid substrates of System A act to repress

activity by the protein synthesis-dependent mechanism (Bracy

et Al., 1986), with alanine, asparagine, glycine, proline,
serine and threonine being the most effective repressors

found. This repression is blocked by both cycloheximide and

actinomycin D (Kilberg et al., 1985). When adequate levels

of amino acids are restored, the excess functional

transporters in the plasma membrane are presumably degraded,

and the elevated transcription rate slowed.

It is important to recognize that our lab has studied

the amino acid-dependent regulation of the System A amino

acid transporter, and we are interested in examples of amino

acid regulation of metabolic processes, and gene expression

in particular. It has been well documented that amino acids

exert a regulatory effect on the catabolism of protein in

hepatic tissue (Poso et Al., 1982), with seven amino acids

able to directly regulate the degradation, and alanine able

to co-regulate degradation. As concentrations of the amino

acids remain high, catabolism is blocked, but if the plasma

concentration of the controlling amino acids drops, then

degradation is favored. This phenomenon is seen as control

at the metabolic level, very different from what the

inhibitor studies suggest is happening with System A.

There exist few reports in the literature regarding the

induction of proteins responding specifically to amino acid

starvation. Levinson et al., (1980) observed that amino

acid starvation of cultured chick embryo cells induced the

biosynthesis of four proteins. Two hours after incubation

of cells in amino acid-free medium, proteins corresponding

to molecular weights of 89, 73, 35, and 27 kDa were induced.

Biosynthesis reached steady state within 8 hours, and began

to decay after 12 hours of culture in the amino acid-free


The activity of the methionyl-tRNA synthetase component

of the multi-enzyme aminoacyl-tRNA synthetase complex has

been shown to be increased upon methionine restriction

(Lazard et al., 1987). When methionine levels were dropped

from 100 AM to 1 AM in the culture medium of Chinese Hamster

ovary cells, a two-fold induction in enzyme activity

occurred. Antibodies specific to the methionyl-tRNA

synthetase were used to confirm that the increase in enzyme

activity was due to an increased amount of the enzyme. At

this time, it is not clear if this induction was

transcriptional or post-transcriptional.

Lacking strong examples of regulation by amino acids,

our laboratory has made it a goal to investigate the amino

acid-dependent regulation of System A, as well as the

cellular response to amino acid deprivation in general. To

identify proteins that might be up-regulated by amino acid

starvation, our lab has used radiolabelling and subsequent

two-dimensional gel analysis of rat liver proteins, and more

specifically, rat liver membrane proteins to identify

proteins whose biosynthesis is enhanced by amino acid

starvation, as we predict for the System A protein (Chiles

and Kilberg, 1987). The results of one such an experiment

are illustrated in Figures 1-la and 1-lb, which present two-

dimensional gel patterns of total cell protein from starved

and fed cells. The gel pattern of cellular proteins are

nearly identical at 6 hours of amino acid starvation.

Previous experiments have identified rat liver membrane

proteins of 66 and 73 kDa that were induced 2- to 3-fold by

amino acid deprivation (Chiles and Kilberg, 1987).

A second experiment, shown in Figure 1-2, illustrates

proteins made in vitro from mRNA purified from Fao hepatoma

cells that had been amino acid-fed or starved for three

hours. There is very little suggestion of any significant

differences in mRNA abundances in the time scale that our

model for System A suggests. The results from Figures 1-1

and 1-2 clearly indicate that there are few proteins induced

by starvation that may be seen by this type of gel analysis,

although it is known that only the most abundant cell

proteins are visible by these types of analyses. The

purpose of this project is to identify an mRNA and its

corresponding gene that is up-regulated by amino acid

starvation in the Fao hepatoma cell line.

(a) FED


Figure 1-1. Autoradiogram of Fao cellular protein obtained
from cells cultured for 6 hours with or without amino acids
following Two-Dimensional Polyacrylamide Gel
Electrophoresis. Cells were incubated in the presence of
S5-Methionine, total protein isolated and separated by pI
in the horizontal direction and by molecular weight in the
vertical direction. The gels were subsequently dried and
used for autoradiography (protein gel methods presented in
chapter 4). Figure 1-la shows cells incubated in the
presence of amino acids (MEM) while Figure 1-lb (next page)
shows cells incubated in the absence of amino acids (NaKRB).



Figure 1-1. continued.


W 1

"- 97"6




Figure 1-2. Autoradiograph of in vitro translation products
following One-Dimensional SDS-PAGE. Poly(A)* mRNA was
purified from Fao cells amino acid-starved and -fed for
three hours. Five micrograms of mRNA was then translated in
the presence of H3-Leucine using the rabbit reticulocyte
translation system (in yjir translation methods presented
in chapter 4). Proteins were size fractionated by
electrophoresis, the gel was then dried and used for
fluorography. Lanes on gel: 1 (+), amino acid-fed cells; 2
(-), amino acid-starved cells; 3 (M), Brome Mosaic Virus
mRNA, used as molecular weight markers; 4 (C), Control -- no
mRNA added to the translation mix.




To gain insight into the mechanisms by which higher

eucaryotic systems are regulated by availability of

substrate, we proposed to identify a cDNA clone

corresponding to a gene that is amino acid regulated by

using the methods of subtractive library production and

differential hybridization (Figures 2-1 and 2-2). The

procedure of differential hybridization uses cDNA probes

derived from two different populations of mRNA molecules in

contrast to other methods that derive probes from a single

gene or gene product such as specific DNA segments. This

differential, or "plus/minus" procedure is designed to

detect cDNA clones derived from mRNAs which are present

("plus") in one condition and absent or reduced ("minus") in

a second condition. The method is one of the less common

methods of identifying or cloning genes, compared to the

screening of cDNA expression libraries with antibodies, or

screening libraries with DNA probes, but nonetheless,

hundreds of genes have been identified by this method

(Sargent, 1987).

One of the first examples of differential hybridization

being used was from Williams and Lloyd (1979) in their study

of the slime mold, D. discoideum. A cDNA library, or "clone

bank", as they referred to it, was created from the

organism's mRNA, purified at the eighth hour of development

in liquid culture. The clone bank was screened using mRNA

obtained from the eighth hour ("plus") of development and

the beginning of the culture period, or zero hours of

development ("minus"), end-labelling the RNA with [ P]-ATP

to make it a radiolabelled probe. A number of clones

corresponding to high and medium abundance mRNAs were

obtained for which the relative level changed during the

first nine hours of development.

Dworkin and Dawid (1980a, 1980b) produced a similar

study examining changes in mRNA abundance during embryonic

development of the African clawed frog, Xenonus laevis. In

this work, several cDNA libraries were constructed from a

variety of stages in development, and they were screened by

radiolabelled first-strand cDNA using the method of colony

hybridization (Grunstein and Hogness, 1975). Dworkin and

Dawid identified many different clones showing a

differential pattern of expression both increasing and

decreasing throughout development.

Many other examples of differential hybridization as a

method for identifying regulated genes exist in the

literature. Some of these include clones from Saccharomyces

inducible by galactose (St. John and Davis, 1979);

identification of the T-cell receptor (Saito et al., 1984);

mRNAs under the influence of hormones such as

triiodothyronine (Magnuson et al., 1985), estradiol

(Masiakowski et al., 1982), and progesterone (Misrahi et

al., 1987). Other studies have identified mRNAs that are

sex-specific (Zurita et al., 1987), under the control of the

circadian clock in Neurospora (Loros Mt al., 1989), DNA

damage-inducible (Fornace et al., 1988), and transiently

induced during rat liver regeneration (Sobczak etAl.,


From these examples, it is clear that if specific

differences exist between two mRNA pools, then

identification of those differentially expressed mRNAs is

possible if there is a significant difference, and the mRNA

is not unusually rare. We believe that cells cultured in

amino acid-free and amino acid-supplemented media have

specific differences in their mRNA pools, one of the

differences between the two pools being a class of mRNAs for

which the abundance is increased when cells are cultured in

amino acid-free media. On the basis of inhibitor studies,

we believe that the System A gene in rat liver and hepatoma

cells is a member of this class of mRNAs (Kilberg, 1986).

To enhance the probability of finding clones for amino

acid regulated genes, of which the total number and

abundance is unknown, two things may be done. First, the

source of the mRNA should be a cell type or cell condition

that maximizes the abundance of the class of mRNA in

question. To this end, the Fao hepatoma cell line was

chosen because it is relatively easy to grow in cell

culture, and consistently shows a strong adaptive regulation

response. That is, System A transport activity in the Fao

cell is reproducibly three- to four-fold higher in the amino

acid starved condition compared to the cells supplemented

with amino acids. Furthermore, the Fao cell line does not

secrete appreciable amounts of albumin compared to

hepatocytes, suggesting that the mRNA abundance of albumin

is greatly reduced in the Fao as well, thus eliminating the

most abundant mRNA of the liver cell. Performing the

experiment in cell culture rather than in vivo allowed us to

separate the condition of starvation from hormonal changes

that would occur due to dietary differences in the rat. As

there was no mRNA known to be induced by amino acid

starvation in any mammalian system, we followed the enhanced

activity of the System A transporter as our indicator that

the cells were entering adaptive regulation. mRNAs for

which abundances increase during adaptive regulation are

assumed to increase along with the increase in System A-

mediated transport. This correlation between mRNA content

and related protein abundance has been documented in the

hormonally regulated gluconeogenic enzyme, PEPCK (Beale et

al., 1982). We chose to make a cDNA library using mRNA

purified from the amino acid-starved Fao cells displaying

the adaptive regulation response.

To further increase the probability of finding an mRNA

up-regulated by starvation, a second library was made with

amino acid-starved Fao cells as the source for mRNA, but in

this case, subtractive hybridization was used to increase

the effective abundance of induced clones in the library.

Subtraction is made possible by the fact that mRNA

populations can be readily hybridized to completion using

homologous cDNA (Sargent, 1987). If "induced" mRNA from

starved cells is used to generate first-strand cDNA

templates, then they may be hybridized to an excess amount

of "uninduced" mRNA purified from fed cells. Sequences in

common readily hybridize to one another, but cDNA sequences

enhanced in the induced condition will have fewer homologous

counterparts from the uninduced condition, and are therefore

less likely to hybridize. These molecules that remain

single stranded represent a population of cDNAs that are

more likely to contain induced sequences and can then be

used to generate a "subtracted" cDNA library. Such a

subtracted library was made and was screened using

differential hybridization to identify amino acid regulated


In either case, from the induced mRNA of the starved

Fao cells, or the subtracted cDNA, cDNA libraries were made

in the phage vector lambda gtl1. These libraries are

referred to as the "induced" library, and the "subtracted"

library, respectively. Lambda gtll was engineered as a cDNA

vector that would also express the gene product of the

cloned cDNA as a fusion protein, that is, a chimeric protein

consisting of the polypeptide corresponding to the cloned

cDNA linked to the Z. coli B-galactosidase protein, whose

gene was inserted into the vector (Huynh et al., 1985).

This construction would allow the expression libraries to be

utilized should antibodies be produced to amino acid-

starvation induced proteins. Once the libraries were

created, they were then screened by the "plus/minus" method,

and induced cDNA clones identified.

Materials and Methods

Cell culture

Fao Hepatoma cells were grown in Modified Eagle's

Medium (MEM) (pH 7.4) supplemented with 24 mM NaHCO, 2.5 mM

glutamine, 100 U/ml penicillin, 100 gg/ml streptomycin, 28.5

Ag/ml gentamicin, and 6% fetal bovine serum. Incubation was
at 37C with 5% CO2 added to the atmosphere. Cells were

grown to near confluence, prior to preparation of RNA,


Foo cells
S Isolate mRNA

Make cDNA library

S plate library

with AA-fed hybridize with
probe Oc.dulicate AA-starved probe


wash filters,
outoradiography identify differential

purify individual
gtl11 clone

prepare cDNA
I sert
RNA gel mae 32P
'k Northern prabe b2P

Figure 2-1. Diagram illustrating the general protocol
involved in identification of an mRNA induced by amino acid
starvation. Cartoon illustrates the preparation of a cDNA
library from amino acid-starved Fao cells, the library is
plated on an agar surface and duplicate nitrocellulose lifts
are prepared from the agar surface. The two lifts are
hybridized separately with cDNA probes made from mRNA
prepared from fed and starved cells. Differential
hybridization signals are identified, and the cDNA insert is
purified from the library clone responsible for the
differential pattern. The induced status of the mRNA is
confirmed using the purified cDNA as a probe on a Northern
blot containing lanes of mRNA prepared from fed and starved

M-F.d M-St~v.d
mRNA pooIa..~~

-~-~ ~

~~w- ~



- induced
by starvation

Amount of
Differential Probe:


M 0

Figure 2-2. Diagram illustrating principle of
differential hybridization. mRNA pools, represented by
"pie" graphs, depict a class of mRNAs induced by starvation
(marked with an asterisk). That difference in abundance is
reflected in the population of cDNA molecules (curved lines)
made from the two populations of mRNA. cDNAs corresponding
to the induced class are marked with asterisks. For a given
quantity of total cDNA made, there will be more cDNA from
the induced class in the starved case compared to the fed
case (bar graph). This difference will be reflected after
hybridization to equal amounts of DNA fixed to filters, as
seen by autoradiography.



transport assays, or passage to new flasks for continued


Amino acid uptake by Fao hematoma cells

To ascertain that cells in culture have responded

appropriately to amino acid starvation prior to preparation

of RNA, cells were assayed for System A transport activity.

Along with 150 mm tissue culture plates prepared for RNA

isolation, four extra plates of cells were prepared for this

assay. On the day of harvest, the cell media was removed

and replaced with sterile Na'-containing Krebs-Ringer

bicarbonate (NaKRB) buffer supplemented with antibiotics.

One half of the dishes received buffer supplemented with 3

mM of each of the six amino acids (alanine, asparagine,

glycine, proline, serine, and threonine) known to repress

System A activity to a basal level (Bracy et al., 1986).

After three hours of incubation in these media, cells in the

dishes were harvested for RNA purification by the method of

Chomczynski and Sacchi (1987). concurrently, the four extra

dishes were used to test the System A activity in the amino

acid-starved or amino-acid supplemented dishes.

Amino acid transport was measured by the method of

Gazzola et Al., (1981), except that 15 cm dishes were used

instead of cluster trays. Briefly, cells were incubated in

the presence of the radiolabelled amino acid AIB (200 pM)

with a Na*-containing or Na*-free buffer. Na+-dependent

transport for 2 minutes at 37"C was taken as the difference

between the uptake rate in the two buffers. Following the

transport assays, the cellular protein in the dishes was

precipitated with the addition of 10 ml of 10%

trichloroacetic acid (TCA). After incubation for 1 hour at

4C, 0.2 ml of the extract was placed in a scintillation

vial for determination of the radioactivity, whereas the

remainder of the extract is discarded. The precipitated

protein in each dish was solubilized with 5 ml of 0.2 M NaOH

containing 0.2% SDS and then measured by a modification of

the Lowry procedure (Kilberg et Al., 1983).

Handling of DNA and RNA

Methods used were generally as described in Maniatis

(1982). DNA and RNA concentrations were measured at 260 nm,

using 50 Ag/ml/A., and 42 Ag/ml/A2, respectively. Ethanol

precipitation of DNA was at -20"C for at least one hour in

the presence of 2.5 M ammonium acetate, pH 7.5, or 0.3 M

sodium acetate, pH 5.5, and two volumes of ethanol or one

volume of isopropanol. Precipitates were collected by

centrifugation, large volumes at 10,000g for at least 10

minutes, small volumes (in microfuge tubes) at 14,000g for

at least 10 minutes. RNA was precipitated for at least one

hour in the presence of 0.2 M sodium acetate, pH 4.0, and

two volumes of 100% ethanol or one volume isopropanol.

Precipitates were collected by centrifugation as described

for DNA.

Phenol/chloroform extraction refers to the

deproteination of a nucleic acid solution by emulsification

with 1 volume of phenol, followed by re-emulsification with

one volume of 1:1 (v/v) phenol/chloroform, and lastly one

volume of chloroform alone. Aqueous phases at each step

were recovered by centrifugation.

All solutions and materials that came in contact with

RNA were autoclaved or filter sterilized. Solutions, test

tubes, pipet tips, and other materials were pretreated

overnight with a solution of 0.01% diethylpyrocarbonate

(DEPC) prior to autoclaving to inactivate RNases.

RNA isolation

Total cellular RNA was prepared using the "Single-Step

Extraction" method of Chomczynski and Sacchi (1987). Cells

(or tissue) were solubilized in "Solution D", which contains

4 M guanidinium thiocyanate, 25 mM sodium citrate, pH 7.0,

and 0.5% sarcosyl. For a 150 mm tissue culture dish, 6 ml

of solution D were added, and for tissues, about 5 ml were

used per gram of tissue. After lysis of tissue culture

cells or homogenization of tissue in Solution D, the

solution is acidified by adding 0.1 volume of 2 M sodium

acetate, pH 4.0. An equal volume of water-saturated phenol

is added and mixed by inversion followed by the addition of

0.2 volumes of chloroform. After vigorous mixing for 10

seconds, the mixture is incubated on ice for 15 minutes

followed by a centrifugation for 20 minutes at 10,000g. The

aqueous phase containing the RNA is collected and the RNA is

precipitated by the addition of an equal volume of ice-cold

isopropanol followed by a 1 hour incubation at -20C. The

RNA is collected with a 10 minute centrifugation at 10,000g.

The pelleted RNA is resuspended in 0.3 volumes of the

original amount of Solution D. We then added the extra step

of extracting this volume with an equal amount of 1:1 (v/v)

water-saturated phenol/chloroform to further remove any

protein contaminants. The RNA was again precipitated with

an equal volume of isopropanol, and the RNA pelleted as

before. After rinsing the pellet with 80% ethanol, the RNA

was air-dried and resuspended in DEPC-treated water. The

absorbance of the RNA solution was then determined at 260

and 280 nm to determine RNA concentration and the relative

abundance of nucleic acid and protein. This 260:280 ratio

should be about 2.0 as an indication that the RNA is

reasonably pure. RNA obtained by these methods was size

fractionated by formaldehyde/agarose gel electrophoresis

(Davis et al., 1986), stained with ethidium bromide, and

densitometrically scanned to quantitate relative amounts of

18S and 28S ribosomal RNAs (complete descriptions of RNA

electrophoresis and densitometry are in Chapter 3 methods).

The ribosomal 28S:18S ratio should be about 2:1, indicating

that little degradation of RNA by ribonucleases has

occurred, and the RNA is of good quality to use for reverse

transcription (Han &t Al., 1987).

Polv(AI mRNA isolation

Poly(A)+mRNA was separated from other cellular RNA

using oligo-(dT) chromatography, essentially following the

method of Aviv and Leder (1972). Oligo-(dT) cellulose

(Sigma) columns were prepared by first washing 0.25 g

oligo-(dT) cellulose in a sterile plastic 12 ml tube with

DEPC-treated water. After hydration and gentle mixing, the

matrix was allowed to sediment, and the fines were removed

with the supernatant by pipetting. The matrix was then

resuspended in fresh DEPC-treated water and the slurry added

to a sterile disposable plastic 10 ml chromatography column

(BioRad no. 731-1550). After the matrix settled, it was

equilibrated with at least 10 ml of RNA loading buffer (RNA

loading buffer is 0.5 M NaCl, 10 mM Tris, pH 7.4). Total

RNA samples, dissolved in water and at concentrations at or

less than 1 mg/ml, were mixed with an equal volume of 2X

loading buffer, heated at 650C for 3 minutes and immediately

cooled on ice. The solution was then poured over the

column, allowed to flow through, and poly(A) RNA washed

through using 10 ml of loading buffer. The poly(A) mRNA

was eluted from the column using DEPC-treated water, with

the mRNA typically contained entirely within the first 2 ml

of eluent. The eluent was made 0.3 M in sodium acetate, pH

4.0, and precipitated overnight with 2.5 volumes of ethanol.

The RNA was then recovered by centrifugation. After one

round of purification, the poly(A)*mRNA obtained was

fractionated by formaldehyde-agarose gel electrophoresis and

stained with ethidium bromide and it was determined that the

samples still contained considerable amounts of ribosomal

RNA. Therefore, a second round of selection was always

done to purify the poly(A) mRNA to an acceptable level.

cDNA synthesis

To make cDNA for cloning into the bacteriophage vectors

lambda gtll, we followed Gubler and Hoffman (1983) as

modified by Davis et al. (1986) (Figure 2-3). First strand

cDNA was synthesized from 10 gg of poly(A)*mRNA using 10 Ag

of oligo(dT) primer and 40 units of AMV reverse

transcriptase (Seikagaiku) in 90 mM Tris, pH 8.7, 130 mM

KCA, 9 mM MgC12, the four dNTPs at 1 mM each, RNAsin

(Promega) at 1 unit/Al, and actinomycin D at 36 pg/l.

Fifty microcuries of a-[P]-dCTP (3000 Ci/mmol) were

included to quantitate the reaction yield and to visualize

the sizes of the products by autoradiography following

alkaline agarose gel electrophoresis (Maniatis, 1982).

There was no attempt made to quantitate the yield at

successive steps, but recovery of labelled cDNA was always

detected with a Geiger counter every time a precipitation

was done. It was found that at least 95% of the counts were

always present in the pelleted cDNA. After

phenol/chloroform extraction and precipitation using

ammonium acetate and ethanol, the first-strand cDNA-mRNA

duplex was dissolved in 50 Al of TE buffer (TE buffer is 10

mM Tris, pH 7.5, 1 mM EDTA). Two microliters of this volume

were removed and used for quantitation by liquid

scintillation spectrometry and electrophoresis. The

remainder was added to 50 Al of 2X second-strand buffer (2X

buffer is 40 mM Tris, pH 7.4, 1 mM MgCl2, 2 mM ammonium

sulfate, 200 mM KCI, BSA at 100 Ag/ml and the four dNTPs at

80 AM each). Two units of RNaseH and 11.5 units of DNA

Polymerase I were added, and this reaction was incubated for

1 hour at 12'C, followed by one hour at 22C. This reaction

mixture was extracted two times with chloroform, and the

cDNA was then precipitated twice with ammonium acetate and

ethanol to remove the unincorporated dNTPs from the cDNA.

Methylation of the cDNA at internal EcoRI sites was done

using 100 units of EcoR I methylase and 15 AM S-adenosyl

methionine in 100 mM Tris, 5 mM EDTA buffer (pH 8.0) for 20

minutes at 37C. Following a 10 minute incubation at 65"C

to inactivate the enzyme, the ends of the cDNA were blunted

or "polished" using 15 units of T4 DNA polymerase in the

presence of 100 AM dNTPs and 12 mM MgCl2 in the same

Tris/EDTA buffer used for the methylase reaction. This

mixture was incubated for 15 min at 37C, after which the

cDNA was subjected to an organic extraction followed by

ammonium acetate and ethanol precipitation. The cDNA,

resuspended in 10 mM Tris, pH 8.0, had phosphorylated DNA

linkers containing the EcoR I-specific sequence added to the

polished ends using 1,600 units of T4 ligase. The ligation

was allowed to incubate overnight at 15C. The cDNA was

then digested with the restriction endonuclease EcoR I to

generate cohesive EcoR I ends for cloning into the EcoR I

site of lambda gtl phage. The digested cDNA was size-

fractionated on a 5% polyacrylamide gel in 50 mM Tris, 50 mM

boric acid, 1 mM EDTA (TBE) buffer along with a parallel

lane of DNA size markers. All cDNA over 500 bp was excised

from the gel, electroeluted in TBE buffer, and then

concentrated using an Elutip-d mini column following the

protocol supplied by the manufacturer (Schleicher and

Schuell). Briefly, the small prepacked plastic column

allows DNA to be bound to the matrix in a low salt buffer,

the column is rinsed, then DNA eluted in a volume of 400 Al

with a high salt buffer containing 1 M NaCl. The cDNA was

then ethanol precipitated and resuspended in TE buffer. The

final yield of cDNA was estimated based on the specific

radioactivity of the first strand cDNA synthesized. Second

strand reactions are known to be nearly 100% efficient and

need not be monitored as closely as the first strand


Library construction

The cDNA, made as described above, was ligated into

dephosphorylated arms of the vector lambda gtll (Promega)

and packaged in vitro using the "Packagene" in vitro

packaging extract (Young and Davis, 1983) from Promega

Biotec, following the protocol supplied by the manufacturer.

Briefly, the cDNA inserts were ligated to the

dephosphorylated arms of the vector using T4 DNA ligase at

15C for four hours, followed by incubation with the

packaging extract for 2 hours at room temperature to create

infective phage. An aliquot of the packaging reaction was

taken and plated on F. coli host strain Y1088 to quantitate

the total number of plaque forming units, or "p.f.u" in the


Subtractive hybridization

Total cytoplasmic RNA was isolated from amino acid-

starved and amino acid-fed cells, poly(A)+mRNA selected by

oligo-(dT) chromatography, and first strand cDNA was made

from the poly(A)+mRNA from amino acid-starved cells using

AMV reverse transcriptase as described previously to obtain

a mRNA/cDNA duplex. The original mRNA template was then

hydrolyzed by adding NaOH to 0.4 M and heating at 50c for 1

hour followed by 65C for 15 minutes. This mixture was then

cDNA Library Construction




*j'f:-- cDNA:cDNA
gC.Ni "

methylated & linkere added

1 96I 1

EcoR I cut
"us gtli

ZInsrts gated
Into vector

Figure 2-3. cDNA library construction. Outline of
cDNA construction from poly(A)+ mRNA using the method of
Gubler and Hoffman. mRNA is reverse transcribed from mRNA
using an oligo(dT) primer and reverse transcriptase. A cDNA
duplex is made using RNaseH and DNA polymerase I. The cDNA
was tailed with EcoR I linkers, and ligated into the phage
vector lambda gtll.


neutralized with HCl, phenol-chloroform extracted,

precipitated by adding 0.4 volumes of 6M ammonium acetate

and 3 volumes of ethanol and then incubating in a dry

ice/ethanol bath for 45 minutes. The cDNA was collected by

centrifugation at 13,000g for 30 minutes. A TCA

precipitation analysis of the cDNA quantitated the single-

stranded cDNA made from the 10 Ag of starting mRNA. The

first-strand product was added to 20 Mg of mRNA derived from

amino acid-fed Fao cells, and lyophilized to dryness. Six

microliters of hybridization buffer (0.5 M sodium phosphate,

pH 6.5; 0.1% SDS; 5 mM EDTA) were added. The pellet was

resuspended and hybridized under mineral oil for 41 hours at

65C, to a Rt value of 1500, where Rt is the product of

RNA in moles of nucleotide/liter and time of hybridization

in seconds (Timberlake, 1980; Rowekamp and Firtel, 1980).

Unhybridized single-stranded cDNAs were separated from

cDNA/RNA hybrids by passage over a 15 cm long water jacketed

chromatography column containing approximately 10 ml of

hydrated DNA-Grade Bio-Gel HTP hydroxyapatite (BioRad). The

hybridization mixture was diluted to 0.01 M phosphate, the

mixture was then passed over a hydroxyapatite column

equilibrated to 0.01 M phosphate, pH 6.5, and kept at 65C,

followed by elution with 0.12 M phosphate buffer, pH 6.5,

containing 0.1% SDS with a continuous flow rate of

approximately 0.5 ml/minute, to selectively elute the

single-stranded cDNA from double-stranded hybrids which

remain bound to the column (Bernardi, 1965) (Figure 2-4).

The double-stranded molecules were removed from the column

by an elution with 0.5 M phosphate. The single-stranded

cDNA was collected, and was precipitated with ammonium

acetate and ethanol, resuspended in water, then phenol-

chloroform extracted, and precipitated. The resulting

pellet was air-dried, then dissolved in 40 4i of sterile

water. To this was added 30 Ml of random primer extension

buffer (0.67 M HEPES, 0.17 M Tris, 17 mM MgCl2, 33 mM B-

mercaptoethanol, 1.33 mg/ml BSA, 18 OD2 units/ml pd(N),

(random deoxynucleotide hexamers), pH 6.8), 5 Al of 1 mM

each dATP, dCTP, DGTP, and dTTP, 90 ng oligo-(dT) and 30

units of the Klenow Fragment of DNA Polymerase. This

mixture was allowed to incubate at room temperature for 18

hours, and was then subjected to an organic extraction and

precipitation as before. The resultant double-stranded cDNA

was then methylated at EcoR I sites, ends polished with T4

polymerase, linkers added and digested as described

previously. The tailed DNA was size-fractionated on a 5%

polyacrylamide gel and a section was cut from the gel

representing fragments 500 bp and greater. The gel slice

was electroeluted in 0.2X TBE, and the cDNA purified from

the buffer using an elutip-d column.

Differential screening

We essentially followed the protocols of Maniatis et

al. (1982) for phage library screening. Phage and host

bacteria E. coli Y1088 were cultured in LB media, plates

were 1.5% agarose and top agarose was 0.8% agarose.

Appropriate numbers of infective phage were mixed with 0.2

ml or 0.5 ml (for 100 mm or 150 mm plates, respectively) of

host cells and incubated at 37C for 20 minutes. These

cells were then mixed with molten (50"C) top agar; this

mixture then was poured over the surface of a 1.5% agarose

LB plate. After the top agarose hardened, plates were

incubated for 12 to 16 hours at 37"C. After chilling the

plates for one hour at 4'C, nitrocellulose filters were

placed on the plate until completely wetted, and then

carefully removed. Duplicate filters were laid on the plate

until wet, then allowed to sit for an additional two minutes

to equalize the amount of phage adhering to the two filters.

Filters were then treated as follows: air dried, laid on a

piece of 3 mm Whatman filter paper soaked with 0.1 M NaOH

and 1.5 M NaCl for 2 minutes, and then incubated for 3

minute on a similar Whatman filter soaked with a solution

containing 0.2 M Tris, pH 7.5; 1 M NaCl; 0.3 M sodium

citrate, pH 7.0; 0.26 M potassium phosphate; 2 mM EDTA.

Filters were again air-dried, then baked at 80"C for 2.5

hours in a vacuum oven to fix the denatured phage DNA to the

AA-Fed cells





tr IdI f VVVV

W uA.s. ds
so [phosphate]

Figure 2-4. Hydroxyapatite chromatographic subtraction
of single-stranded cDNA. Protocol outlined here was used to
isolate a fraction of cDNA from amino acid-starved Fao cells
enriched for induced sequences. First-strand cDNA from
starved cells was hybridized to an excess of mRNA from fed
cells. After hybridization, the induced, or single stranded
cDNAs were collected by passage over a hydroxyapatite
column. A hypothetical elution profile is shown. This cDNA
was primed with oligo-(dT) and random deoxynucleotide
hexamers, made double stranded with DNA polymerase, and then
further prepared as illustrated in Figure 2-3.

filters. Pairs of duplicate filters were hybridized with

"plus" or "minus" probes as outlined below.


For differential hybridization procedures between cDNA

radiolabelled probes and nucleic acids affixed to

nitrocellulose paper, we followed hybridization procedures

described by Davis et al., (1986). Briefly, filters

prepared as described above were prehybridized in 4X SSC (IX

SSC is 0.15 M sodium chloride, 0.015 M sodium citrate, pH

7), 0.02 M Tris, 0.02 M polyvinylpyrrolidone, 0.02 M bovine

serum albumin, 0.02 M Ficoll, 20 Ag/ml sheared salmon sperm

DNA, and 10% dextran sulfate. Following prehybridization of

at least one hour at 42C, radiolabelled first-strand cDNA

probe was added at 1 x 107 cpm/ml. First-strand cDNA probe

was made using the first-strand reaction of cDNA synthesis,

except that the sole source of dCTP in the reaction mix was

250 ACi of 3000 Ci/mmol a-UP dCTP. Hybridizations were

performed at 42C for 48 hours. After hybridization, blots

were washed two times in 2X SSC for 5 minutes at room

temperature, two times in 2X SSC, 1% SDS at 65C for 30

minutes, and then finally two times in 1X SSC for 30 minutes

at 65C. Blots were dried to dampness on filter paper,

wrapped in plastic wrap, and used for autoradiography with

Kodak XAR-5 film at -70C.

DNA purification from phage

cDNA inserts from phage were purified as outlined

below. Individual isolates of phage were mixed with E. coli

strain Y1088 and 0.4%, 50*C molten agarose, and then poured

onto a 1.5% agarose plate (for a 15 cm plate, 100,000 p.f.u.

of phage were mixed with 150 Al of a suspension of Y1088

concentrated 5-fold from an overnight liquid culture by

sedimentation and resuspension in 0.2 volumes of sterile SM

buffer). The plate was incubated overnight at 37C to

complete lysis. The top agar containing the phage was

collected by scraping, added to 5 ml of SM buffer (SM buffer

is 100 mM sodium chloride, 50 mM Tris, 8 mM magnesium

sulfate, 0.002% gelatin), and then mixed by rocking for one

hour at room temperature. This mixture was centrifuged for

10 minutes at 10,000g to separate the agarose from phage,

with the resulting supernatant centrifuged a second time to

insure that all traces of agarose have been removed.

Following an incubation at 37*C for one hour with 10 gg/ml

RNase I and 2 gg/ml DNaseI, whole phage are precipitated by

addition of 0.2 volumes of 20% polyethylene glycol (PEG-

8000, Sigma), 2.5 M NaCl, mixed, then incubated on ice for

1 hour. Following centrifugation at 10,000g for 20 minutes,

the pellet is saved, dried, and resuspended in 0.5 ml of SM

buffer. Five microliters of 10% SDS and 5 p1 of 0.5 M EDTA

are added to this mixture to lyse the phage and prevent

degradation of phage DNA by DNase. The mixture is then

subjected to a phenol/chloroform extraction as described

previously followed by ethanol precipitation.


We began by producing two lambda gtll libraries, the

first made from total poly(A)* mRNA isolated from amino

acid-starved, or induced Fao hepatoma cells. We refer to

this library as the "induced" library. A second library was

made from cDNA again derived from induced Fao cells,

however, uninduced sequences were subtracted away prior to

final library construction. We refer to this as the

"subtracted" library.

The induced library was derived from 10 Ag of poly(A)+

mRNA prepared from 3 hour amino acid-starved Fao cells

induced 3.2-fold for System A transport. Transport assays

performed on the batch of cells used for mRNA purification

found the sodium-dependent transport of AIB in the Fao cells

to be 1645 and 510 pmol AIB/mg protein/minute for the

starved and fed cells, respectively. TCA precipitation

analysis determined that 0.9 Ag of first-strand cDNA was

made from the 10 Mg of mRNA; this 9% yield is typical of

first-strand reactions (Gubler and Hoffman, 1983). From the

0.9 Mg of ss cDNA, about 1.8 Mg of ds cDNA was made.

Following complete processing and purification of the cDNA,

we recovered 600 ng of EcoR I tailed, cDNA product. As 1 ng

of DNA represents the equivalent of about 109 cDNA inserts

averaging 1 kb each, it was clear that there was more than

enough cDNA to prepare a library. Typical commercially

prepared libraries contain 500,000 to 5 X 106 recombinant

inserts (e.g. Clontech Laboratories, Palo Alto, CA.).

The cDNA inserts were then ligated to commercially

prepared dephosphorylated gtll arms (Promega) and packaged

in vitro as described previously. Five different packaging

mixes resulted in total numbers of p.f.u. ranging from

60,000 to 437,000. The total for the five mixes was 1.3 x

106, which compares favorably to commercially prepared

libraries. The collection of 1.3 X 106 infective gtl1 phage

is what is referred to as the induced library. This library

was then amplified using E. coli Y1090, following the

procedure outlined by Davis et al. (1986). The amplified

lysate contained 109 p.f.u./ml, and there was a total volume

of 6 ml.

The subtractive library was made from 10 gg of

poly(A) mRNA, and again, the mRNA used was a mixture of two

samples. One 5 Ag aliquot came from a cell preparation

described previously, in which System A transport was 3.2-

fold induced by starvation, and the second 5 Ag from cells

displaying an approximately 17-fold induction (593 vs. 34

pmol AIB/mg protein/minute starved vs fed, respectively.

First-strand cDNA was made as described previously using an

oligo-(dT) primer and AMV reverse transcriptase to extend

the cDNA product. Approximately 0.6 Ag of cDNA was made

from the mRNA, which was an acceptable yield for the first-

strand reaction (Davis et al., 1986). The cDNA/mRNA duplex

was then subjected to alkaline hydrolysis of the RNA as

described previously. The single-stranded cDNA was then

hybridized to a thirty-fold excess of non-induced mRNA,

derived from amino acid-fed Fao cells (System A transport 34

pmol AIB/mg protein/minute). After hybridization to a Rot

of 1500, the single-stranded fraction, representing

induced sequences was purified by hydroxyapatite

chromatography as described. Ninety nanograms, or 15%, of

the cDNA was collected in this fraction (Figure 2-5).

Following generation of 180 ng of ds cDNA, 66 ng of size-

selected (>500 bp) cDNA was obtained. Twenty five nanograms

of this subtracted cDNA was ligated to 500 ng of gtll arms,

and after packaging and titering, it was estimated that

there were 211,000 p.f.u. in this subtracted library.

To characterize the libraries, randomly selected clones

from both the induced and subtracted libraries were selected

and purified. DNA from these clones was purified, and

cleaved with the restriction endonuclease EcoR I to release

the cDNA insert from the gtll arms. These DNA samples were

size-fractionated by electrophoresis on a 1% agarose


:' 8 ,0 0 0 .. .................. . . . . . ................. ...........
o6,0 0 ..............................
N 4 0 o .... ... .... ... ]. .. .... ... .. .... ... .... ... .... ... i ... .... ... i .... ... .... ... .... ... .... ... ...

2, 0 .. ...... .. .."'. .,d s

0. 12M 0.50 M Phosphate

Figure 2-5. Hydroxyapatite Chromatography Profile.
Separation of single-stranded and double-stranded fractions
of cDNA and cDNA/mRNA hybrids used to prepare cDNA enriched
for sequences corresponding to induced mRNAs. After
hybridization to a Pt value of 1500, the hybridization mix
was diluted to 0.01 M phosphate and passed over the column
which was preequilibrated to 0.01 M phosphate, 650C. The
single-stranded fraction was eluted with 0.12 M phosphate.
cDNA eluted from the column was quantitated by liquid
scintillation counting of incorporated radioactivity.

TBE gel, and stained with ethidium bromide to visualize the

sizes of the inserts (Figure 2-6). The induced library had

an average insert size of 1100 400 bp (n=8), while the

inserts from the subtracted library were somewhat shorter,

the average being 900 300 bp (n=9) (Table 2-1). These

numbers compare well to commercial suppliers of cDNA phage

libraries, such as Clontech.

The induced library was produced first, and was the

first to be screened by the plus/minus method. As the limit

for detection of clones using reverse transcribed cDNA

probes is about 0.1% abundance of the total mRNA (Dworkin

and Dawid, 1980a), it was decided to screen 10,000 p.f.u.,

which is nearly 10-fold excess to the total number of p.f.u.

needed to be screened if every induced clone above that

detection level could be identified. That is,

theoretically, if you could only detect induced clones at

the 0.1% abundance level, then on the average, screening

1000 clones should present most of the clones of 0.1%

abundance or greater, and by screening ten times that many,

you are over 99% sure that every clone of abundance 0.1% or

higher will have been screened.

From the initial screen of the 10,000 plaques, 81 were

identified as potentially exhibiting an induced

hybridization signal, and those plagues were picked for

purification and verification. After the first round in

Table 2-1. Properties and Characteristics of
Subtracted cDNA libraries.



strand cDNA

subtracted cDNA
for enrichment of
induced clones

cDNA insert

p. f.u.

average insert

10 Ag

0.9 Mg

600 ng

1.3 X 10"

1100 400
(n = 8)

Induced and


10 Ag

0.6 jg

66 ng

2.1 X 10'

900 300
(n = 9)

Comparison of the two lambda gtl libraries made for
the purpose of differential screening. In both cases,
the source of the mRNA was amino acid-starved Fao
hepatoma cells.




Figure 2-6. Insert characterization of cDNA libraries.
Agarose gel analysis from the induced and subtracted Fao
lambda gtll libraries. DNA samples were prepared as
described and digested with EcoR I to release the cDNA
insert from the cloning site. Induced library, lanes 1-6:
six random clones showing insert sizes. lane 7: DNA
molecular size markers. Subtracted library, lanes 1-5:
five random clones, lane 6: molecular size markers are
wild type lambda DNA cleaved with Hind III and EcoR I.

which plaques were plated at 2,000 to 3,000 per 150 mm

plate, subsequent rounds used 100 mm plates for the 80

potential positives, and only about 100 p.f.u. were plated

per plate to insure clear separation of plaques, and thus

aid the purification of phage clones. Of the 80 clones

identified and purified, only two clones, numbers 1 and 58

(or I-1 and 1-58) maintained the differential hybridization

pattern through the three rounds of purification (Figure

2-7). cDNA inserts from these two clones were purified, and

used to prepare [E-P]-labelled probes to test amino acid-

fed and amino acid-starved mRNAs for abundance of the

corresponding mRNA using Northern Analysis (see Chapter 3).

The subtracted library was also screened by the method

of differential hybridization using first-strand cDNA probes

derived from amino acid-fed and amino acid starved cells.

It was decided to screen this library at a even higher level

of redundancy than the induced library, 50,000 p.f.u. being

chosen as an appropriate number to screen. After obtaining

the autoradiographs of the filters from the first round of

screening, 22 spots on the films were identified as

differentially hybridizing. The corresponding plaque or

area from the agar plate was found and picked for

purification and further screening. After three rounds of

screening for verification and purification, there were four

clones exhibiting a differential signal through all rounds

of screening. They were labelled S-1, S-3, S-5, and S-15.

Ultimately, only clone S-5 from this screening proved to be

an induced clone (see Chapter 3).


Two libraries were made, one from total mRNA, and a

second from subtracted first-strand cDNA. Both were derived

from amino acid-starved Fao hepatoma cells. The libraries

contained acceptable numbers of independent p.f.u., that is,

they should contain cDNA clones representative of mRNAs with

abundances down to at least 0.001%. A characterization of

the average insert size from the two libraries showed that

the inserts were of reasonable length, representing at least

a considerable fraction of a typical mRNA, from which the

full length could be identified using further library

screening and/or primer extension-APCR cloning.

First-strand cDNA probes were made from a mix of mRNA

prepared from different batches of cells. This was done in

an attempt to average out high or low inductions of the

mRNAs from each sample. Induction of System A transport in

the Fao cells used for the mRNA preparations ranged from

about 3-fold up to 17-fold, although the latter value may be

an overestimation of the induction due to an unusually low

sodium-independent transport determination.

Figure 2-7 Differential Hybridization. Top:
Autoradiograms illustrating a partially purified amino
acid starvation-induced cDNA clone. After picking
potential induced clones from the first round of
screening, the phage suspensions from each pick were
plated at a density of 50 to 100 plaques per 10 cm
dish, duplicate filter lifts made, and differentially
hybridized. After washing filters, an autoradiogram of
the filters was made. Autoradiogram shows enrichment
of an induced clone present at about 10% purity.
Arrows indicate location of induced signals. Clone
illustrated above is I-1 from the induced cDNA library.
Bottom: After 3 rounds of purification, the
potentially induced cDNA clone I-1 was purified.
Plaques picked from the second round of screening were
suspended in SM buffer and replated as in previous
round. Bottom photograph is of autoradiogram of
purified induced clone exhibiting a differential signal
despite presence of equal amounts of cDNA probe in both
conditions. As only a single plaque was picked from
the previous round, and every plaque on the plate shows
the differential signal, purification was assumed, and
confirmed by one additional round of screening (not


A .

5 k 0 0

ir 0 AO"* Ag +



*A 'A

The fact that only two clones were identified from

a screening of 10,000 p.f.u. from the induced library

and four clones from a screening of 50,000 clones from

the subtracted library suggests that there is not a

large number of easily detectable induced clones from

the condition of starvation at three hours. For

example, it would be expected that if there was just

one mRNA species that was induced, and its abundance

was at or above the 0.1% level in the induced state,

then its corresponding cDNA clone should be present at

about one in every 1,000 clones screened. Therefore,

in the 10,000 plaques screened in the induced library

one would expect the clone to appear about 10 times,

and from the subtracted library screening of 50,000

plaques, an induced clone would appear about 50 times.

Given that only 6 clones were identified from the two

libraries, one of the following may be true. First,

there may be no mRNA that is induced by starvation for

three hours for which the abundance is above the 0.1%

level. Second, it is possible that there are induced

clones for which the abundance is above the 0.1% level,

but the degree of induction (1- to 2-fold) may be

relatively weak and difficult to detect by this method.

There may, of course, be numerous mRNAs from the low

abundance class (< 0.1%) that are induced to a great

degree that remain undetected.

The reults from this differential screening are

consistent with the results of other work, also from

our own lab. We have attempted to detect amino acid

starvation-induced protein synthesis by combining

pulse-labelling techniques with two-dimensional

polyacrylamide gel elctrophoresis. only 3 liver

membrane proteins have been detected for which

synthetic rates appear to be increased by 2- to 3-fold

following amino acid deprivation of rat hepatocytes

(Chiles et al., 1987). Together these two approaches,

both designed to identify amino acid-regulated genes,

argue that gene control with respect to starvation of

mammalian cells is considerably different than that in

bacterial or yeast in which expression of a large

number of proteins is enhanced by several-fold or more.

Clearly, the fact that there is no mRNA yet

identified for which the cellular content is raised by

amino acid starvation leaves us without a positive

internal control to verify that the mRNA preparation

truly represents an "induced" condition. Our

assumption in doing this work has been that the

elevation of System A-mediated transport activity is an

indication that the cells are induced by starvation.

Chapter 3



The goal of any differential hybridization is to obtain

cDNA clones corresponding to mRNAs induced by a particular

stimulus or condition. Our goal, of course, was to obtain

clones that would allow us to learn more about the adaptive

regulation response seen in cells starved for amino acids.

It was hoped that the System A transporter cDNA might be one

of the clones identified in the screening, although it is

generally accepted that amino acid transporters are

relatively rare proteins, which probably corresponds to a

rare mRNA. This, if true, would make the identification of

the System A cDNA more difficult. In any case,

identification of induced clones allows us to characterize

the mRNA induction: the amount, the time course, and the

mechanism. As so few mammalian mRNAs have been shown to be

regulated by substrate, the investigation of any induced

clone allows us to determine if the regulation is primarily

transcriptional, as seen in the lactose operon of 9. coli

(Pardee et al., 1959) or if the control is mediated by

transcriptional and post transcriptional components, as

elucidated in the gcn response in Saccharomvces cerevisiae

(Hinnebusch, 1988). These systems are able to be regulated

by the presence of a single sugar or amino acid. In

mammalian systems, Poso et al., (1982) have demonstrated

direct repression of protein catabolism in hepatocytes by

individual amino acids. Although the regulation appears to

be a direct effect of the amino acids, it is probably due to

metabolic control and not a genetic mechanism. Our hope in

obtaining any cDNA clone that corresponds to a regulated

mRNA is that the mechanism, when fully understood, will

provide us with a better appreciation of how mammalian cells

respond to changes in nutrient levels.

The differential screening procedure described in

Chapter 2 was performed with mRNA isolated from hepatoma

cells incubated in the presence and absence of amino acids

for 3 hours. We investigated the changes in abundance of

our putative starvation-induced mRNA at various times after

amino acid removal, both shorter and longer periods than the

3 hours used for differential screening. Through the use of

cloned cDNAs available from our own and other laboratories,

we also were able to look at the changes in abundance of

several mRNA species that are typically used as controls,

that is, they are often seen to be unchanged during

experiments that change the abundance of more highly

regulated mRNAs. Monitoring our own clone as well as the

control mRNAs was done using Northern analysis. Northern

analysis is the name playfully given to electrophoretic

separation of RNA followed by blotting and hybridization

(Thomas, 1980), a variation of the method invented by

Southern (1975) for the analysis of DNA fragments. Northern

analysis allows one to identify the size and steady-state

levels of many RNA samples at any specific time. This makes

Northern analysis the method of choice to analyze content of

a specific RNA in cells at different times, under different

conditions, or from different tissues or cell types.

After finding an mRNA that was induced by amino acid

starvation, our next goal was to investigate the mRNA

induction in greater detail. Our current model for System A

transporter regulation proposes that the increase in

transport seen during adaptive regulation is due to

increased transcription of the System A gene (Kilberg,

1986). We performed the plus-minus screening to find an

mRNA which was induced primarily by increased gene

transcription. However, it is known that the cellular

concentration of many mRNAs is increased due to

stabilization against degradation (Theil, 1990). Tests

utilizing the inhibitor of translation, cycloheximide

(Pestka, 1971), and the inhibitor of transcription,

actinomycin D (Goldberg and Freidman, 1971), are necessary

along with assays measuring active transcription (Groudine

etAl., 1981) to determine accurately the mechanism of mRNA

induction. These experiments are presented in this chapter.

Materials and Methods

Plasmid DNA purification

To purify small amounts of plasmid DNA the "mini-prep"

method of Ish-Horowitz and Burke (1981) as described in

Davis et al., (1986) was used. Briefly, cells were pelleted

for 10-30 minutes at 1500g, then resuspended in a solution

of 50 mM glucose, 25 mM Tris, pH 8.0, and 10 mM EDTA.

Lysozyme was added to a final concentration of 2 mg/ml, and

the mixture was incubated at room temperature for 5 minutes.

A solution of 0.2 M NaOH and 1% SDS was added, the cells

were then incubated on ice for 5 minutes. A solution of

ice-cold 5 M potassium acetate, pH 4.8, was added and mixed,

and then the mixture was centrifuged at 13,000g at 40C for

10 minutes. The supernatant fraction, which contained the

plasmid, was filtered through gauze, and treated with 1

Ag/ml RNase A for one hour at 37C. The samples were

extracted with phenol-chloroform and the plasmid DNA

precipitated with ethanol or isopropanol.

The "Triton-Lysozyme" method of plasmid purification

from Davis et al., (1986) was used for large-scale

preparation of plasmid DNA from one liter cultures of

plasmid-containing bacteria. After pelleting cells as in

the mini-prep method, a solution of 10% (w/v) sucrose, 50 mM

Tris, pH 8.0 was used to resuspend the cells. Lysozyme was

added to the suspension to 3 mg/ml; this was then incubated

on ice for 10 minutes. A solution of 0.5 M EDTA was added

and mixed, followed by another incubation on ice for 5

minutes. At this point, Triton X-100 was added to 0.3%

(w/v) and mixed, and the mixture was incubated at 370C for

at least 2 minutes or until bacterial lysis became apparent.

The lysed cells were centrifuged in a Beckman Ti6O rotor at

200,000g (50,000 rpm) for 30 minutes to pellet lysed

bacteria. The supernatant, containing the plasmid DNA, was

RNase A treated as described above, then extracted with

phenol-chloroform and ethanol precipitated.

Use of DNA restriction enzymes

DNA restriction enzymes were used as recommended by the

supplier at concentrations of at least 1 unit of enzyme per

microgram of DNA to be digested. In the case of lambda

phage DNA or very large volumes of plasmid DNA, digestion

was typically carried out overnight, with up to 5 units of

enzyme per microgram of DNA.

DNA agarose gels

DNA was size-fractionated on 1% agarose gels using a 1X

TBE buffer, essentially as in Maniatis et al., (1982). DNA

samples were added to loading buffer containing 10% glycerol

with bromophenol blue and xylene cyanol, each at 0.05%

(w/v). After electrophoretic separation, DNA was stained

for 30 minutes in a 1 Mg/ml solution of ethidium bromide to

allow visualization of the DNA under ultraviolet light.

Electroelution of DNA fragments

The procedure from Davis et al., (1986) was used to

purify DNA fragments from agarose gels. Briefly, fragments

were cut from gels and placed in dialysis tubing with 1 to 3

ml of 0.2X TBE buffer depending on the size of the agarose

block cut from the gel. The tubing was clamped and placed

in an electrophoresis box with enough 0.2X TBE buffer to

completely cover the tubing. The voltage applied was 100 to

300 volts, depending on the size of the gel box used, and

electroelution was carried out for 1 to 3 hours, depending

on the size of the fragment in the agarose. Upon completion

of electroelution, the polarity on the gel box was reversed

and voltage was reapplied for 2 minutes to remove DNA

fragments from the inner surface of the dialysis tubing.

The fluid was removed from the dialysis bag by pipetting,

and the volume decreased to 0.4 ml using sec-butyl alcohol

extractions to remove water from the DNA solution. The

volume was phenol-chloroform extracted and precipitated

overnight with an equal volume of isopropanol. The DNA was

pelleted at 13,000g for 30 minutes, and the pellet rinsed

with 80% (v/v) ethanol, then air-dried. The pellet was

dissolved in 10 to 100 Al of TE, and 2 Ml of this solution

were run on a mini-gel along with molecular size markers of

known concentration to confirm purification of the insert

and to estimate the concentration of the insert.

RNA gel electrophoresis

RNA was size-fractionated on horizontal agarose gels

consisting of 1% (w/v) agarose, 0.04 M

morpholinopropanesulfonic acid (MOPS), pH 7.0, 0.01 M sodium

acetate, 2.2 M formaldehyde and 1 mM EDTA. RNA samples

(typically 1 to 20 Mg) were mixed with a loading buffer

containing the above buffer plus 50% (v/v) formamide, 0.5%

(w/v) bromophenol blue and 10% (v/v) glycerol. After

heating the samples for 15 minutes at 650C, 1 Al of a 1

mg/ml ethidium bromide solution was added to each sample,

mixed, and loaded on the gel. Typically, gels were run

overnight at 40 volts with buffer recirculation, although

shorter runs at higher voltages gave similar resolution.

RNA blotting

We used "Gene Screen" nylon paper for blotting and used

the following modification of a blotting protocol supplied

by the manufacturer (DuPont/NEN). Following

electrophoresis, gels were incubated in distilled water for

a total of 10 minutes with several changes of the water,

then incubated for 15 to 30 minutes in 50 mM NaOH, followed

by a 30 minute incubation in 100 mM Tris, pH 7. The gel was

then laid upside down on a wick made from 3 mm Whatman

paper, and the nylon membrane, which was previously

incubated in 1OX SSPE for 10 minutes, was applied to the

surface of the gel. The RNA was then transferred to the

nylon membrane by capillary action (Davis et al., 1986)

using 1OX SSPE as the transfer buffer and paper towels above

the nylon membrane to drive capillary flow. The next day,

the transfer was confirmed by inspection of the blot and gel

on an ultraviolet light box, then the RNA was covalently

cross-linked to the membrane by exposing the blot to

ultraviolet light for 3.5 minutes at a distance of 40 cm.

Membranes were stored damp, wrapped in plastic wrap at 40C

until needed.

Northern hybridization

Hybridizations used nylon membranes prepared as

described above and [mP]-labelled cDNA fragments to detect

relative mRNA abundance by autoradiographic analysis.

Radiolabelled probes were prepared as described below, and

added to the hybridization solution to achieve a

concentration of at least 106 cpm of probe per ml of

hybridization solution. Hybridization solution was 0.5 N

sodium phosphate, pH 7.2, 1 mM EDTA, 1% (w/v) Fraction V

bovine serum albumin (BSA) and 7% (w/v) SDS (Church and

Gilbert, 1984).

Prior to hybridization, membranes were pretreated as

follows: The membrane was washed in 20 mM Tris, pH 7.0-7.5

at room temperature. The Tris was heated to 950C, then

poured on the membrane which was incubated on a rocking

platform at room temperature for 10 minutes. Following this

wash, the blot was placed in 0.lX SSPE, 0.1% SDS for one

hour at 650C. This pretreatment enhances the hybridization

signal and allows for better detection by autoradiography

(Ecker and Davis, 1987). After hybridization overnight at

650C, membranes were washed 3 times at room temperature in

hybridization wash solution A containing 0.5% BSA, 40 mM

sodium phosphate, pH 7.2, 5% SDS, and 1 mM EDTA. Following

these washes, blots were washed for one hour at 650C in the

hybridization wash solution B: 40 mM sodium phosphate, pH

7.2, 1% SDS, and 1 mM EDTA. Membranes were then blotted dry

and subjected to autoradiography.

Blots previously hybridized were stripped for

rehybridization by boiling in a solution of 0.lX SSC, 1% SDS

for one hour. We found it useful to verify the stripping by

placing the stripped blot on film for one to two days to

confirm that no radioactivity was left on the filter.

Control probes

The B-actin, glyceraldehyde phosphate dehydrogenase,

histone H4 and Cu-Zn superoxide dismutase cDNA inserts were

obtained from the laboratory of Dr. H. S. Nick. Recombinant

plasmids containing coding sequences for these genes were

grown using B.coli host strain DH5a in liquid culture with

the appropriate antibiotic specific to each plasmid.

Plasmids and cDNA inserts were purified as described above.

Preparation of r-P]-labelled DNA probes

We used the "random-primer extension" method of

Feinberg and Vogelstein (1983) using a manufacturer's

reagent kit (BRL). We found that by using this kit, we

consistently obtained incorporation of radiolabel in excess

of 50% when labelling 50 to 100 ng of double-stranded DNA

insert. After boiling to denature the DNA, the insert was

incubated at room temperature for at least 2 hours in the

following solution: 20 AM of dATP, dGTP, and dTTP in 0.2 M

HEPES, 50 mM Tris, 5 mM MgCl2, 10 mM 2-mercaptoethanol, 0.4

mg/ml BSA, 5 OD2units/ml oligodeoxynucleotide primers

(hexamer fraction), pH 6.8. Fifty microcuries of [a-m-

P]dCTP, 3000 Ci/mmol, 10 MCi/gl and 3 units of Klenow

fragment were added in a final volume of 50 Al.

Incorporation of radiolabelled dCTP into DNA probe was

assayed as follows: two microliters of the probe mixture

were removed and diluted with 498 Al of TE buffer, then 10

Al of this dilution were pipetted into a 15 ml disposable

plastic test tube. Ten milliliters of ice-cold 10% (w/v)

TCA were added to the tube, mixed by inversion, and then the

mixture was filtered through a glass-fiber filter disc

(Whatman Type GFC). The tube was washed with 10 ml of

ethanol; this volume was filtered through the disc followed

by 10 ml of acetone. A second filter was directly spotted

with an additional 10 Al of the diluted probe mix. The two
filters were put into vials for liquid scintillation

counting. Total incorporation and percent incorporation

were determined by comparing the TCA-precipitated counts to

the total counts.

We found it unneccessary to purify the probe any

further for Northern hybridizations. After analysis of

incorporation, probes were boiled for 5 minutes to denature

the double-stranded DNA prior to adding the probe to the

hybridization solution. After hybridization, the probe

mixture was removed and saved frozen at -200C. The frozen

probe mix could be used up to two weeks later for a

subsequent hybridization to mRNAs of reasonable abundance.

Subcloning of cDNA inserts into plasmids

We essentially followed the method of Hanahan (1983).

A 100 ng aliquot of EcoR I-digested plasmid DNA was added to

a 3-fold molar excess of cDNA insert. The cDNA was ligated

to the plasmid for at least 2 hours at room temperature, or

overnight at 150C for blunt-end ligations, in a buffer

containing 50 mM Tris, pH 7.6, 10 mM MgCl2, 1 mM ATP, 1 mM

DTT, and 5% (w/v) polyethylene glycol-8000, which optimizes

ligation for subsequent transformation (King and Blakesley,


Transformation of bacteria

Ligation mixtures containing recombinant plasmids were

used to transform E. coli strain DH5a (Hanahan, 1983) which

was obtained as competent cells from Bethesda Research

Laboratories, Bethesda, Maryland. The protocol supplied

from the manufacturer was followed: incubation of 50 Al of

cells mixed with 2 Al of ligation mixture on ice for 45

minutes, followed by a 30 second incubation at 370C, then 2

more minutes on ice. The mixture was added to 1 ml of LB

medium and incubated at 370C for one hour. The cells in

this volume were pelleted with a one minute centrifugation

at 13,000g and the cell pellet was gently resuspended in 100

Al of LB. Cells were then spread on a plate of LB
containing 1.5% (w/v) agar, with 50 Ag/ml ampicillin, 0.025%

(w/v) IPTG, and 0.025% (w/v) X-gal. The plates were

incubated overnight at 370C. Because DH5a is ampicillin-

sensitive, only bacteria containing the plasmid will grow;

blue colonies indicate plasmid without an insert, white

colonies indicate a plasmid with insert.

Amino acid starvation of Fao and other cells in culture

The starvation-induced response of the ASI-l mRNA was

optimized as follows: Cells were trypsinized from a

confluent 150 cm2 culture flask and suspended in 50 ml of

MEM supplemented with 6% (v/v) fetal bovine serum (fbs). An

aliquot of the suspension was used to determine the number

of cells/mi using a "SPot lite" counting chamber (Hausser

Scientific, Blue Bell, Pennsylvania). After the cell

density was determined, 4 x 106 cells were plated into 10 cm

round plastic culture dishes; the total volume of MEM plus

FBS in each dish was brought up to 12 ml by adding extra

medium. The cells were cultured for 48 hours before

changing the medium to begin an experiment. Whenever new

medium was added to cells, the cells would be rinsed two

times in the new medium, aspirating off each rinse, followed

by the addition of the medium for the experiment. Initially

the amino acid-starved and -fed conditions were achieved by

using NaKRB with or without the following six amino acids:

alanine, glycine, serine, threonine, proline, and

asparagine, all at 3 mM. Later experiments used sodium-

containing Krebs-Ringer bicarbonate buffer (NaKRB) for the

amino acid-starved condition and MEM without FBS for the

amino acid-fed condition. Experiments using vitamins were

done by adding 10OX liquid vitamin stock (Sigma) to the

culture media. Inhibitors actinomycin D (25 AM),

cycloheximide (100 AM), L-azetidine-2-carboxylic acid (10

mM), and histidinol (10 mM), were added to 50 ml of culture

media 24 hours prior to their use to insure complete

dissolution into the media. Starvation and other time

course experiments were performed for time periods up to 12

hours. RNA was prepared from cells in culture by addition

of the denaturing guanidinium solution ("solution D")

directly onto cells adhering to the culture plate, and then

following the procedure described in Chapter 2.

Ouantitation of autoradiocrams

Relative abundances of mRNAs as determined by Northern

analyses were quantitated using the LKB Ultroscan XL laser

densitometer. Bands or spots on developed X-ray film were

scanned using the 2-dimensional scanning mode of the

densitometer, taking care to use exposures that produced

bands that were neither underexposed nor overexposed, so

that the exposures were in the linear range of detection.

Resultant intensities were given as absorbance units/mm2,

and values for various bands from an experiment were

expressed as a ratio to control values and expressed as

relative units.

Nuclear run-off transcription assay

To analyze active gene transcription, we used the

"nuclear runoff" method (Marzluff and Huang, 1985) as

described by Groudine et al, (1981) and modified by Ausubel

et al, (1987) with our own further modification involving

use of the "single-step" purification of RNA (Chomczynski

and Sacchi, 1987) to isolate radiolabelled mRNA from nuclei.

Fao cells were prepared as described previously,

plating 9 x 108 cells per 150 mm round culture dish 48 hours

prior to the start of an experiment. Upon starting an

experiment, MEM containing 6% fbs was replaced with either

MEM (no serum) for the amino acid-containing condition, or

NaKRB (no serum) for the amino acid-free condition. Dishes

were rinsed twice with one volume of the new medium before

adding a third volume to start the experiment. Two dishes

per condition were prepared to obtain adequate numbers of

cells for RNA labelling. At appropriate times, dishes were

removed from the incubator, culture medium drained and

replaced with ice-cold PBS, and the dish placed on a wet ice

slurry. Cells were scraped from the dish with a plastic

"Bondo" spreader. The PBS/cell suspension was collected

into a sterile 15 ml plastic tube and centrifuged for 5

minutes at 1000g to pellet the cells. To the pellet, 4 ml

of Nonidet P-40 (NP-40) lysis buffer (NP-40 lysis buffer is

10 mM Tris, pH 7.4, 10 mM NaCl, 3 mM MgCI2, and 0.5% NP-40)

were added dropwise while the tube was vortexing at half-

maximal speed. The tube was then vortexed for an additional

10 seconds. After sitting on ice for 5 minutes, the mixture

was centrifuged for 5 minutes at 10OOg to pellet the nuclei.

The pelleted nuclei were resuspended in NP-40 buffer and

centrifuged again as described above. The pellet from this

centrifugation was saved and resuspended in 200 Al of

glycerol storage buffer (glycerol storage buffer is 50 mM

Tris, pH 8.3, 40% glycerol, 5 mM MgCl2, and 0.1 mM EDTA).

Two hundred microliters of 2X reaction buffer (2X reaction

buffer is 10 mM Tris, pH 8.0, 5 mM MgCl2, 0.3 M KCl, 5 mM
DTT, and 1 mM each of ATP, CTP, and GTP) were added to this

volume along with 10 Al of [a-kP]UTP (800 Ci/mmol, 40

mCi/ml). After mixing, the nuclear suspension was incubated

in a shaking water bath at 300C for 30 minutes to allow

nuclear transcripts to elongate, incorporating the

radiolabelled UTP. After this incubation, RNA was purified

from the suspension using 4 ml of Solution D using the

method of Chomczynski and Sacchi (1987), described in detail

in Chapter 2. The resultant RNA pellet from this procedure

was dissolved in 100 Al of DEPC-treated water, and 5 Al were

removed and counted by liquid scintillation spectrometry to

determine the radioactivity incorporated into the

transcripts. Equal amounts of radioactive counts were used

for each hybridization, typically 5-10 x 106 cpm. An

appropriate volume of the radiolabelled-RNA was added to 1.5

ml of hybridization buffer (as described above) and mixed.

Then a nylon filter dot- or slot-blot containing 10 Mg of

each plasmid (either the ASI cDNA insert, the B-actin cDNA

insert or the pUC 19 plasmid serving as a control) was

placed in a microfuge tube containing the hybridization

solution. The tube was then incubated for 48-72 hours at

650C to allow hybridization of radiolabelled-mRNAs to their

corresponding cDNAs on the filter. After hybridization, the

filters were washed three times for 15 minutes each in 50 ml

of wash solution B (described above) at 650C. Washed

filters were subjected to autoradiography. In a cassette

containing Cronex "Lightning-Plus" intensifying screens,

typical exposure times of the Kodak XAR film were 1 to 7

days. Relative rates of transcription were determined using

the LKB Ultroscan densitometer to quantitate the amount of

hybridization between different conditions.


Confirmation of a starvation-induced mRNA

Although 3 bacteriophage cDNA-containing clones had

survived three rounds of purification exhibiting the

differential hybridization pattern that would be expected of

a starvation-induced clone, only one of those, clone S-5,

showed the differential, induced pattern as determined by

Northern analysis. Phage DNA was prepared from this clone

and 100 pg of this DNA were digested overnight using

150 units of the restriction enzyme EcoR I. The digest was

phenol-chloroform extracted, pelleted and taken up in TE

buffer. The digested DNA was size-fractionated on a 1%

agarose, TBE gel at 30 volts overnight, and the digested

phage arms and an approximately 450 bp insert were

detectable on the gel. The insert was cut from the gel and

electroeluted as described. The insert was subsequently

subcloned into the plasmid vector pGEM-3Z (Figure 3-1) as

described in the methods section of this chapter.

To confirm the induction of this clone, a [3P]-

labelled probe was made from the purified insert and

hybridized to a Northern blot containing two lanes of RNA,

one with RNA prepared from Fao cells starved for amino acids

for three hours, the second with RNA from cells prepared

identically, except that amino acids were always present.

To confirm an equal loading of RNA onto the two lanes of the

blot, a similarly radiolabelled-probe for the 9-actin mRNA

was prepared and added to the hybridization. When positive

results were obtained, a second hybridization was conducted

with another blot containing the same two conditions, but

from a second preparation of cellular RNA. From the two

experiments, the mRNA for S-5 was induced an average of 1.5-

fold by starvation, whereas the actin mRNA was quantitated

to be reduced to 0.65-fold of its fed level. The

combination of these two factors yielded a 2.3-fold level of

induction if expressed as a ratio of S-5 to actin. A third

experiment was performed to verify these results, and again

the results confirmed the first two trials. The average of

the three experiments was a 2.1-fold induction of S-5 as

compared to expression of B-actin (Figure 3-2). At this

point, confident that an induced clone had been identified,

we renamed clone S-5 as ASI, for Amino acid-Atarvation

induced clone.

12 3







Figure 3-1. Subcloning of the Fao hepatoma (S-5) ASI cDNA
insert. DNA purified from lambda clone S-5 was digested
with restriction endonuclease EcoR I and size-fractionated
on a 1% agarose gel. The 425 bp fragment was electroeluted
from the gel and ligated into the EcoR I site of plasmid
pGEM-3Z. Top, lambda DNA: lane 1, undigested wild type
DNA; lane 2, digested S-5 DNA; lane 3, molecular size
standards. Bottom, recombinant pGEM plasmid: lane 1,
linearized plasmid; lane 2, EcoR I digested plasmid; lane 3,
undigested plasmid; lane 4, molecular size standards.
Arrows mark location of 564 bp DNA standard from lambda EcoR
I/Hind III size markers.



> 1.0

0.0 t II
+ + (A.A.'s)


Figure 3-2. Northern analysis of ASI mRNA induction. ASI
aRNA abundance after 3 hours of hepatoma culture in amino
acid-free (NaKRB) or -supplemented (NaKRB + 6 amino acids)
media. A: abundance of ASI mRNA after 3 hours with or
without amino acids. B: Ratio of ASI mRNA abundance
compared to S-actin abundance, with or without amino acids.
Quantitation of mRNA abundance was determined using laser
scanning densitometry of three independent Northern

As the induction of ASI was small, a series of

experiments were undertaken to determine if under other

conditions, the induction of the ASI mRNA would be greater

than that observed at three hours. The first experiment

performed was to extend the period of amino acid-starvation

beyond three hours, and this again provided us with positive

results. When the starvation period was extended to longer

times, greater inductions were found. Numerous experiments

showed us that the magnitude of this induction was highly

variable, from 2-fold to 4- to 5-fold at its maximum. The

ratio during amino acid starvation of ASI to actin never

decreased below the starting ratio, and we found that the

increase would occur by 9 hours of starvation. Although

experiments were conducted in which cells were starved for

longer than 12 hours (data not shown), it was found that the

survival of Fao cells diminished significantly beyond this

time point. Therefore, we chose to limit our time course

experiments to 12 hours.

The variability in the induction of ASI proved to be

due to variation in the initial condition of the cells when

a starvation experiment was begun, with the age of the cells

on the plate and their density on the plate being major

factors in the cells' ability to produce the induction of

the ASI mRNA. Experiments were performed which varied the

number of cells plated onto the culture dishes and measured

the induction as a function of that variable, and the

following protocol produced a consistent induction: cells

were plated 48 hours prior to the beginning of an experiment

at a density of 4 x 106 cells per 10 cm culture dish in MEM

plus 6% FBS. After 48 hours of growth on the plates, the

MEM plus FBS was removed and the experimental culture media

were added to the plates. Dishes were rinsed twice with the

new culture medium before leaving a third volume of medium

on the cells to begin the experimental condition. Figure 3-

3 shows what we consider to be the optimized induction of

the ASI mRNA. From the initial time, the ASI mRNA abundance

increases from its starting level to an amount between 2-

and 3-fold higher, and if the induction is compared to the

level of actin mRNA during the time course, then the

relative level of the ASI mRNA was about 8-fold above its

starting value. This induction of ASI was complete by 9 to

12 hours. We note that the actin mRNA decreased as a result

of amino acid starvation; a typical change in actin was a

drop of one-half to one-fourth of its original abundance

after 12 hours of amino acid starvation.

Because we were not sure if the decrease in the actin

mRNA was a specific, unique response by this gene or

representative of a general phenomenon of mRNA decrease due

to starvation, we performed Northern analyses of Fao cells

starved for amino acids for up to 12 hours. RNA was

purified from cells at various times during the experiment.

The RNA samples were then used to prepare a blot of cells

starved for amino acids, fed with amino acids, and fed with

amino acids and supplemented with 6% fbs. This blot was

then probed with radiolabelled DNA probes that are typically

used by other laboratories in Northern analysis as control

probes -- that is, they are rarely changed to any great

extent by many cellular conditions. In addition to actin,

we used histone H4, glyceraldehyde 3-phosphate dehydrogenase

(GPD), and copper-zinc superoxide dismutase (SOD). We

observed that all four mRNAs decreased significantly due to

amino acid starvation, and remained relatively constant when

cultured in the presence of amino acids (Figure 3-5), except

for histone H4, which decreased even in the presence of

amino acids and did not remain constant unless fbs was added

to the medium (data not shown). By 2 to 4 hours, there are

slight decreases seen in mRNA abundances for all four of the

control probes, and by 6 to 9 hours, significant decreases

were seen. In the fed condition, the mRNA levels for actin,

GPD, and SOD remained constant throughout the experiment,

while the decrease in histone is seen after 4 hours and

decreased in a linear fashion through 12 hours.

ACT 404 t


0 3 6 12


Figure 3-3. Induction of ASI mRNA by amino acid starvation
of rat hepatoma cells. Fao cells were cultured in amino
acid-free media (NaKRB) for 12 hours, with RNA being
purified from the cells at 0, 3, 6, and 12 hours. Northern
analsis shows relative abundance of ASI and 8-actin (ACT)
mRNAs during amino acid starvation.

Another set of experiments was undertaken to examine

the reversibility of the induction of ASI-l by starvation.

Cells were prepared as described previously, and two groups

of Fao cells were maintained in either NaKRB (amino acid-

starved) or MEM (amino acid-fed) for 12 hours. After 12

hours, the cells from both conditions had their culture

medium withdrawn and replaced with amino acid-containing

MEM. These cells were cultured for an additional 24 hours.

RNA was prepared from both groups of cells at 0, 3, 6, 9,

and 12 hours of the starvation period, and then after 12 and

24 hours of the "re-fed" state, which are hours 24 and 36 of

the experiment from time zero. The Northern analysis is

shown in Figure 3-6. The ASI-I mRNA, after a slight

decrease from zero to 3 hours, increased approximately 3-

fold over the initial abundance after 9 to 12 hours of

starvation. Actin during this same time period decreased

well below its initial value. The actin and ASI-l mRNA

abundance was constant for the amino acid-fed cells

throughout the course of the experiment. Upon refeeding of

amino acids to the starved cells, two changes were noted.

The first change was a decrease of the ASI-l abundance,

although the ASI mRNA abundance did not drop to its initial

level. The second change was larger, an increase in the

actin mRNA of over 10-fold in the first 12 hours of

refeeding as compared to the abundance at the 12th hour




SOD 60,0400

H4 O

0 2 4 6 9 12 2 4 6 9 12


Figure 3-4. Abundance of several mRNAs during amino acid
starvation of rat Fao hepatoma cells. Fao cells were
cultured in amino acid-free ("starved") or -supplemented
("fed") culture medium for various time periods, and RNA was
then purified from the cells. Total cellular RNA was size-
fractionated, blotted, and used for hybridization with the
following cDNAs: ACT: B-actin, GPD: glyceraldehyde-3-
phosphate dehydrogenase, SOD: Cu-Zn superoxide dismutase,
H4: histone H4.

of starvation. The increase in actin due to refeeding

produced an mRNA abundance that was approximately 3-fold

above the initial abundance of the actin message.

A summary graph representing averaged values for the

ASI-l and actin mRNAs is shown in Figure 3-6. Data was

analyzed by laser scanning densitometry of Northern analyses

performed using mRNA prepared from amino acid-starved or

-fed Fao cells. Figure 3-7 presents results of four

separate Northern analyses which were performed to confirm

that changing the amino acid-fed cell medium from NaKRB

supplemented with six amino acids at 3 mM each to MEM, which

contains all of the amino acids, did not alter the basic


The only significant difference between NaKRB (plus the

six added amino acids) and MEM, except for their amino acid

differences, is that MEM contains vitamins essential for

cell growth, whereas NaKRB does not. The experiment shown

in Figure 3-7 involved the addition of a vitamin supplement

to NaKRB to bring the concentration of vitamins in that

medium up to the concentration in MEM, then starving cells

for amino acids using NaKRB with and without the vitamin

supplement. If the induction of ASI-1 occurred in both

instances, then the presence of vitamins in MEM could be

eliminated as the factor that repressed the induction of



ACT M an 40


0 3 6 9 12 24 36

3 6 9 12


Figure 3-5. Refeeding effects on ASI expression in hepatoma
cells. Fac cells were cultured in amino acid-free or -
supplemented medium. RNA was purified from the cells at the
indicated times for Northern analysis. After cells were
incubated for 12 hours in amino acid-free medium
("starved"), the culture medium was replaced with one
containing an amino acid supplement and then the cells were
incubated for an additional 24 hours ("re-fed").




0 12

0 12

Fig. 3-6. Quantitation of actin and ASI mRNA levels after
starvation of Fao hepatoma cells for 12 hours. Relative
abundances of ASI and actin mRNAs after 12 hours of amino
acid starvation are expressed as relative absorbance units
as determined by laser scanning densitometry of
autoradiographs of Northern analyses. ASI data are from 5
independent experiments, whereas the actin data are from 3
separate experiments.

ASI-l. The results presented in Figure 3-7 confirm that

amino acids rather than vitamins regulate the expression of

ASI-I. The induction of ASI-1, as compared to actin, was

3.2-fold (1.7) after 12 hours of culture in NaKRB plus

vitamins compared to MEM. The average induction in NaKRB

without vitamins at 12 hours of starvation was 4.3-fold

(0.8). After refeeding the amino acid-starved cells, the

relative abundance dropped to 1.5 0.4. Although the

average induction without vitamins was slightly higher, a

hypothesis test between the difference of those two means

using the t-distribution test (Merrington, 1941) showed that

the two numbers were statistically indistinguishable

experiments, all subsequent starvation experiments used

NaKRB for the amino acid-free medium and MEM as the amino

acid-containing medium.

Having characterized the time course of the induction

of the ASI-I mRNA, we proceeded to conduct a series of

experiments to investigate the nature of the ASI mRNA

induction. The inhibitor of transcription, actinomycin D,

and the translation inhibitor cycloheximide were used in

conjunction with Fao cells to follow the induction of ASI-l

by Northern analyses at various times after starvation.

Actinomycin D, used at 25 gM, has been shown to be an

effective inhibitor of mRNA synthesis (Goldberg and

Freidman, 1971), and cycloheximide, at 100 AN, an equally

effective inhibitor of protein synthesis (Pestka, 1971).

Using the two inhibitors at these concentrations, a series

of experiments were completed to determine if the induction

of the ASI-l mRNA was dependent on either active

transcription and/or translation. Northern analyses were

performed, and the autoradiograph from each hybridization

was scanned using laser scanning densitometry to normalize

each condition and time point to the initial time point

("t=0") from that experiment. A summary of the results of

this series of experiments is shown in Figure 3-8. As

presented previously in bar graph form (Fig. 3-6), the

induction of the ASI mRNA rose just over 2-fold by 9 hours

of starvation compared to a constant amount of the mRNA in

cells supplied with amino acids. Actinomycin D caused the

ASI mRNA to decay to approximately one-third of its initial

abundance at 9 hours in both the starved and fed cases. In

contrast, the presence of cycloheximide stabilized the ASI

mRNA abundance at its initial level. In both the starved

and fed cases, when cycloheximide was present, the relative

abundance of the ASI mRNA remained near its starting level.

The abundance of B-actin mRNA was also regulated by

presence or absence of amino acids, but in the opposite

manner to the ASI mRNA. Actin mRNA levels remained constant

in the presence of amino acids, but dropped to approximately

one-fourth of their initial value at 9 hours of amino acid

starvation. The presence of either actinomycin D or

cycloheximide caused the actin mRNA to drop in abundance to

about one-half of its initial value by 9 hours of culture.

There did not appear to be any significant difference

between the two inhibitors with respect to changes in actin

mRNA levels, in contrast to the change in abundance with or

without amino acids. The latter effect was statistically

significant and reproducible from experiment to experiment.

The metabolic inhibitors L-azetidine-2-carboxylic acid

and histidinol are amino acid analogs of proline and

histidine, respectively, that are incorporated into nascent

polypeptides, preventing any further elongation of the

growing polypeptide. In this respect, these compounds mimic

starvation for the two amino acids, as translation is

stalled when the proline or histidine analog is

incorporated. We chose to use the analogs as a supplement

to the fed condition, so that along with a normal level of

the amino acids proline and histidine, there would be

included, at 10 mM, either one of the two polypeptide chain

terminators. Fao cells were incubated as before, and after

48 hours of growth in round culture dishes the media was

removed and replaced with either NaKRB (amino acid-free),

MEM (amino acid-fed), or MEM plus either one of the






+vit -vit

Fig. 3-7. Quantitation of ASI mRNA induction in amino acid-
free media with or without vitamins. +AA: Fao cells
cultured for 12 hours in amino acid-supplemented media with
vitamins. -AA: Cells incubated for 12 hours in amino acid-
free media with (+vit) or without (-vit) vitamins.
RF: Cells were incubated for 12 hours in the absence of
amino acids and then transferred to media containing amino
acids and vitamins for an additional 12 hours in amino acid-
supplemented media with vitamins. Following Northern
analysis, the quantitation of autoradiographs was by laser
scanning densitometry.

2.0 2 0.

------ ---- --


0.0 2.0

0 3 6 9 0 3 6 9

Figure 3-8. Effect of metabolic inhibitors on the induction
of ASI mRNA in Fao cells. Amino acid starvation of Fao
cells was performed in the presence of the transcription
inhibitor actinomycin D (25 AM) or the translation inhibitor
cycloheximide (100 AM). Quantitation of a series of
Northern analyses was done using laser scanning
densitometry. Each time point is the average of 2 or 3
determinations and standard deviations where not shown are
contained within the symbol.

U-6 AA,/

0-3 2....

inhibitors at 10 mM. The cells were then cultured for 12

hours, at which time RNA was purified from the cells and

used for Northern analysis. The average of two identical

analyses showed that in response to amino acid deprivation

the ASI mRNA was induced 2.55-fold ( 0.30) over the fed

condition (Fig. 3-9). With the inclusion of L-azetidine-2-

carboxylic acid to MEM, the ASI mRNA was induced 1.81-fold

( 0.04) over the fed condition. Histidinol produced a

1.65-fold ( 0.23) elevation over the level in fed cells.

Additional experiments were pursued to investigate the

potential involvement of individual amino acids in the

repression of the ASI mRNA. Single amino acids at a

concentration of 10 mM were added to the amino acid-free

NaKRB as a "fed" condition versus NaKRB alone to determine

the amino acid specificity of repression for the ASI mRNA.

The amino acids alanine, arginine, glycine, histidine,

lysine, and proline were chosen for two reasons. First,

included in the group are some of the amino acids that are

known to have repressive activity on adaptive induction of

System A-mediated transport (Kilberg et al., 1985). Second,

all 20 amino acids were tested in a series of preliminary

experiments, and a few amino acids showed no effect on the

induction of the ASI mRNA; others had the ability to keep

the ASI mRNA levels nearly fully repressed. We chose six

amino acids and performed three independent Northern

analyses from three separate sets of cell cultures. The ASI

aRNA abundance was quantitated by densitometry of the

resulting autoradiographs. The ASI mRNA abundance in each

amino acid-supplemented media, as compared to fed cells was:

amino acid free, 2.10 0.21; alanine, 1.28 0.07;

arginine, 1.28 0.13; glycine, 2.00 0.31; histidine, 1.48

0.25; lysine, 1.99 0.43; and proline, 2.23 0.46. Most

of the induction values fell between those for the fed (MEM)

and the starved cells, although the induction in the

presence of proline was at least as great as that seen for

total starvation. Alanine and arginine proved to be the

most effective repressors of the induction of the ASI mRNA,

with only a 28% increase seen, compared to the nominal 100%

to 125% increase seen in the starved condition.

Nuclear run-offs were conducted to determine if active

transcription of the ASI gene changed during amino acid

starvation. Nuclei from amino acid-fed and -starved cells

were isolated as described in the methods section of this

chapter, then incubated in the presence of a-[mP]-UTP to

synthesize radiolabelled RNA probe. This probe was used

much like the first strand cDNA probes were used in

differential hybridization, except in this case, the

radiolabelled probe was hybridized to filter-bound cDNA

inserts corresponding to the ASI and actin mRNAs. The