The role of subcellular compartments and structures in eukaryotic gene expression

MISSING IMAGE

Material Information

Title:
The role of subcellular compartments and structures in eukaryotic gene expression
Physical Description:
xii, 219 leaves : ill. ; 29 cm.
Language:
English
Creator:
Zambetti, Gerard Paul, 1958-
Publication Date:

Subjects

Subjects / Keywords:
RNA, Messenger   ( mesh )
Transcription, Genetic   ( mesh )
Gene Expression Regulation   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 206-218).
Statement of Responsibility:
by Gerard Paul Zambetti.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 22253045
ocm22253045
System ID:
AA00012891:00001


This item is only available as the following downloads:


Full Text














THE ROLE OF SUBCELLULAR COMPARTMENTS AND STRUCTURES
IN EUKARYOTIC GENE EXPRESSION












By

GERARD PAUL ZAMBETTI


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
1989















ACKNOWLEDGEMENTS

I would like to acknowledge the dedication of Drs.

Janet and Gary Stein to their work. Their commitment to

science has provided us with a well-equipped lab and the

funds to pursue our research. I would also like to extend

special thanks to the Stein's for their constant support and

guidance during my graduate studies. In addition, I would

like to thank Drs. Bert Flanegan, Tom Rowe and Dick Moyer

for their helpful advice as members of my graduate

committee.

My experiences with my fellow coworkers in the Stein's

lab have been very interesting, rewarding and fulfilling. I

would like to acknowledge and thank Marie Giorgio, Linda

Green, Farhad Marashi, Brian O'Donnell and Mary Beth Kennedy

for their technical guidance and assistance, and Ken Wright

and Joe Teixeira for their help dealing with the computers.

I thank Pat Naples for her appreciation and pledge of care

for "the coffee mug." I would like to thank Dave Collart,

Tim Morris and Charles Stewart for their friendship and for

their advice on raising bees, fishing, and golfing,

respectively. I would like to thank Andre van Wijnen for his

friendship, scientific expertise on and off the "Foos" ball

field, and for keeping an eye on Pat Naples and my coffee









mug. I would particularly like to thank the following

individuals for their tremendous inspiration: April, Emilie,

Ashley, Danny, Shannon, John, Aubrey, Stephan, Alan, Brooke,

Kate, Sarah and Nikki.

I would like to thank the members of the Department of

Immunology and Medical Microbiology. I will always remember

the football games in which Dr. Ed Siden led our team to an

undefeated record against the Pathology team starring Dr.

Ward Wakeland, and the late night "Nerf-ball" baseball games

in the hallway with Mark Labow, Paul Hermonat, Brian Masters

and Dick Wright.

Lastly, I would like to thank my family. The Lord has

truly blessed me with a loving and supportive family and I

fully appreciate all that they have done and sacrificed for

me during my studies.


iii















TABLE OF CONTENTS


PAGE

ACKNOWLEDGEMENTS........................................ ii

LIST Of FIGURES... ...................................... vi

LIST OF TABLES................................. .......... ix

KEY TO ABBREVIATIONS .....................................

ABSTRACT........................ .......................... xi

CHAPTERS

1) INTRODUCTION
General Background.. .............................. 1
Histone Genes and Proteins.......................... 2
Histone Gene Regulation: Transcriptional and Post-
Transcriptional Control........................ 3
Subcellular Compartments........................... 10
The Cytoskeleton: Microtubules, Intermediate
Filaments and Microfilaments........................ 12
Cytoskeletal-mRNA Interactions..................... 16
Nonmembrane-Bound and Membrane-Bound Polyribosomes:
The Signal Hypothesis........................... 19
Overview of Project............................... 22

2) GENERAL METHODS

Conditions for Enzyme Reactions.................... 28
Mammalian Cell Culture............................ 28
Plasmid DNA....................................... 30
DNA Isolation and Purification.................... 35
Isolation of Mammalian RNA........................ 40
Two-Dimensional Gel Electrophoresis and Immuno-
Blotting Analysis................................ 48
Northern Blot Analysis............................. 49
Sl Nuclease Protection Analysis.................... 52
In Vitro Nuclear Run-on Transcription Analysis..... 55
Radiolabeling DNA for Northern Blot Analysis....... 58
Site-Directed Mutagenesis.......................... 61
Transfection of DNA into Cells..................... 71
Establishment of Polyclonal and Monoclonal Stably
Transformed Cell Lines........................... 73









3) SUBCELLULAR LOCATION OF HISTONE mRNAs ON CYTOSKELETON
ASSOCIATED NONMEMBRANE-BOUND POLYSOMES IN HELA S3
CELLS
Introduction..... .................................... 75
Results.............................. .. ....... .... 77
Discussion............ .. ............ ............. 90

4) SUBCELLULAR LOCATION OF HISTONE mRNA PLAYS A ROLE IN
THE POSTTRANSCRIPTIONAL REGULATION OF HISTONE
GENE EXPRESSION
Introduction....................................... 96
Results ........ ................................ 97
Discussion.... ........ ......... .... .............. 126

5) DIFFERENTIAL ASSOCIATION OF MEMBRANE-BOUND AND
NONMEMBRANE-BOUND POLYSOMES WITH CELL CYTOSTRUCTURE
Introduction..................................... 130
Results............ ............. .................. 131
Discussion......................................... 146

6) HETEROGENEOUS PATTERN FOR CYTOSKELETAL-mRNA
INTERACTIONS
Introduction. ..................................... 153
Results........................................... 154
Discussion ........................................ 166

7) THE INFLUENCE OF THE CYTOSKELETON ON THE REGULATION
OF c-Fos GENE EXPRESSION
Introduction. .................. ................... 168
Results................. .......................... 171
Discussion ......................... ................ 190

8) SUMMARY AND FUTURE CONSIDERATIONS..................... 196

REFERENCES.............................................. 206

BIOGRAPHICAL SKETCH..................................... 219















LIST OF FIGURES


Figure Page

2-1 Outline of the cloning scheme for the construction of
the signal peptide-histone fusion gene............... 34

2-2 Outline of procedure for the isolation of subcellular
polysomal fractions................................. 45

3-1 Northern blot analysis of H3 histone, H4 histone and
HLA-B7 mRNA in subcellular fractions................. 81

3-2 Two-dimensional gel electrophoresis of cytoskeleton
and soluble associated proteins...................... 85

3-3 Northern blot analysis of H3 histone, H4 histone and
HLA-B7 mRNAs isolated from subcellular fractions of
cells treated with metabolic inhibitors.............. 89

3-4 Northern blot analysis of H3 histone, H4 histone and
HLA-B7 mRNAs associated with the cytoskeleton and
soluble fractions from HeLa cells treated with
metabolic inhibitors................................ 92

4-1 Schematic diagram of endogenous H3 histone mRNA ST519
and the beta-lactamase signal peptide-H3 histone
fusion mRNA ........................................ 100

4-2 Expression of the signal peptide-histone fusion gene
in HeLa cells....................................... 103

4-3 The subcellular localization of the signal peptide-
histone fusion mRNA................................. 106

4-4 The stability of the signal peptide-histone fusion
mRNA following inhibition of DNA synthesis.......... 109

4-5 Schematic diagram of wild type and mutated signal
peptide-histone fusion mRNAs....................... 112

4-6 Distribution of SPH3E1, SPH3Elalpha, and SPH3E1ATG
mRNAs in nonmembrane-bound and membrane-bound
polysomal fractions................................. 115

4-7 Effects of hydroxyurea treatment on SPH3E1 mRNA...... 119

vi















4-8 Effects of hydroxyurea treatment on SPH3E1ATG
mRNA ................................................. 121

4-9 Effects of hydroxyurea treatment on SPH3Elalpha
mRNA ................................................. 123

4-10 Quantitation of SPH3E1, SPH3Elalpha, SPH3E1ATG
mRNA during hydroxyurea treatment.................. 125

5-1 Northern blot analysis of the cytoskeleton and
soluble phase distribution of H3 histone, H4 histone
and HLA-B7 mRNAs in cytochalasin D treated cells.... 135

5-2 Northern blot analysis of the cytoskeleton and
soluble phase distribution of c-fos mRNA in
cytochalasin D, puromycin, CD/puro, and
CD/cycloheximide treated cells ..................... 140

5-3 Northern blot analysis of the cytoskeleton and
soluble phase distribution of nonmembrane-bound
and membrane bound polysomal mRNAs in cytochalasin D,
puromycin, CD/puro and CD/cycloheximide treated
cells................................................ 145

6-1 Cytochalasin D and puromycin cotreatment does not
release SPH3E1 or SPH3E1ATG- mRNA from the
cytoskeleton........................................ 157

6-2 The wild type signal peptide-histone fusion mRNA is
not released from the cytoskeleton in CD and
puromycin cotreated polyclonal HeLa cell cultures... 161

6-3 Endogenous membrane-bound polysomal mRNAs are
released from the cytoskeleton during cotreatment
with CD and puromycin in SPH3E1 and SPH3E1ATG-
expressing cell cultures........................... 163

7-1 Effects of cytochalasin D and puromycin treatment on
steady state levels of c-fos and chorionic
gonadotropin mRNA................................... 175

7-2 Densitometric analysis of steady state levels of
c-fos mRNA during cytochalasin D and puromycin
treatments......................................... 177


vii















7-3 Time course of c-fos mRNA accumulation during
cytochalasin D and puromycin treatments in
exponentially growing HeLa cells................... 181

7-4 Effect of cytochalasin D on transcription of c-fos
in HeLa cells...................................... 184

7-5 Effect of protein synthesis inhibition on
cytochalasin D induction of c-fos transcription in
HeLa cells........................... .... .............. 188


viii















LIST OF TABLES
Table Page

3-1 Quantity of histone and HLA-B7 mRNAs in the
subcellular fractions.................................. 82

4-1 QuAntitation of SPH3E1, SPH3Elalpha and SPH3E1ATG
mRNA in nonmembrane-bound and membrane-bound
polysome fractions ........ ....... ...... ............ 116

5-1 The percent of cytoskeleton and soluble phase
associated mRNAs in cells treated with
cytochalasin D..................................... 136

5-2 Percent distribution of nonmembrane-bound and
membrane bound polysomal RNAs in the Csk and Sol
fractions from CD, Puro, CD/puromycin and
CD/cycloheximide treated cells...................... 141

6-1 The percent of cytoskeleton associated mRNAs
isolated from cytochalasin D, puromycin and
CD/puromycin treated cells.......................... 158















KEY TO ABBREVIATIONS

BSA Bovine Serum Albumin

cpm Counts per minute

dCTP Deoxycytidine-5'-triphosphate

ddCTP Dideoxycytidine-5'-triphosphate

DNA Deoxyribonucleic acid

DNase Deoxyribonuclease

DTT Dithiothreitol

EDTA Ethylenediaminetetraacetic acid

Hepes N-2 hydroxyethyl piperazine-N'-2
ethanesulfonic acid

HU Hydroxyurea

MCi Microcurie

Mg Microgram

Al Microliter

Am Micrometer

AM Micromolar

OD Optical density

PVS Polyvinyl sulfate

RNA Ribonucleic acid

rpm Revolutions per minute

SDS Sodium dodecylsulfate

SSC Standard saline citrate buffer

Tris (hydroxymethyl)aminomethane

x















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


THE ROLE OF SUBCELLULAR COMPARTMENTS AND STRUCTURES
IN EUKARYOTIC GENE EXPRESSION

By

Gerard Paul Zambetti

December 1989


Chairperson: Janet Stein
Major Department: Immunology and Medical Microbiology


Histone mRNAs naturally reside on nonmembrane-bound

polysomes that are associated with the cytoskeleton. To

study whether the subcellular location plays a role in the

coupling of histone mRNA stability to DNA synthesis, a

signal peptide-histone fusion gene was constructed that

encodes a histone mRNA that is directed to membrane-bound

polysomes. The signal peptide-histone fusion mRNA when

localized in the foreign subcellular compartment is no

longer destabilized when DNA synthesis is inhibited. A

single nucleotide substitution in the signal peptide-histone

fusion gene by site-directed mutagenesis resulted in the

localization of the histone fusion mRNA on nonmembrane-bound

polysomes and restored the coupling of mRNA stability with









DNA synthesis. These results indicate that the subcellular

location of histone mRNA plays an important role in the

posttranscriptional regulation of histone gene expression.

We have determined that the association of histone mRNA

with the cytoskeleton is dependent on the integrity of actin

filaments. Disruption of the microfilaments with

cytochalasin D efficiently released histone mRNA from the

cytoskeleton into the soluble phase. In contrast, membrane-

bound polysomal mRNAs remained attached to the cytoskeleton

during cytochalasin D treatment. Subsequent studies have

demonstrated that membrane-bound polysomal mRNAs are

attached to the cytoskeleton at two different sites: 1) a

cytochalasin D sensitive site and 2) a puromycin sensitive

site.

Disruption of the cytoskeleton with cytochalasin D

resulted in a rapid and marked stimulation in transcription

of the c-fos proto-oncogene and an increase in c-fos mRNA

cellular levels. Transcription of several other genes was

unaffected by the cytochalasin D treatment. These results

suggest that the nucleus can respond to the structural

organization of the cytoskeleton and selectively adjust the

regulation of gene expression accordingly.


xii















CHAPTER 1

INTRODUCTION



General Background

Histones are small basic proteins that are involved in

the packaging of eukaryotic DNA into chromatin. It is

estimated that more than 2 meters of DNA are contained

within the limited area of the nucleus. The condensation of

the DNA to such a great extent is due primarily to its

interactions with the histone proteins. In addition to their

structural properties histones play a functional role in

controlling gene transcription. The way in which the histone

proteins are arranged along a region of DNA appears to

affect the transcriptional properties of that DNA domain.

The complex nature of the regulation of histone protein

synthesis at the transcriptional, posttranscriptional, and

posttranslational levels makes these genes an important

model for studying eukaryotic gene expression. However, one

of the most important reasons for studying the regulation of

histone gene expression is that the synthesis of the

majority of histone proteins (cell cycle dependent histones)

occurs during DNA replication. Elucidation of the mechanisms

coupling histone protein synthesis with DNA synthesis may











yield a better understanding of the processes controlling

proliferation, differentiation, cell aging and neoplasia.



Histone Genes and Proteins

There are five major classes of histone proteins (H4,

H2B, HIA, H3, and H1) ranging in molecular weight from

11,000 to 21,000 Daltons. They are basic proteins that have

been highly conserved throughout evolution. For example,

there are only two amino acids out of 102 which differ

between histone H4 from pea seedlings and calf thymus, which

suggests a critical role for these proteins in cell survival

(DeLange et al., 1969). The histone proteins are encoded by

a family of moderately repeated genes (Wilson and Melli,

1977; Kedes, 1979). In humans, the histone genes are

arranged in clusters that are not organized in simple tandem

repeats as found in the sea urchin and other lower

eukaryotic genomes (Heintz et al., 1981; Sierra et al.,

1982; Carozzi et al., 1984). The clusters are localized on

chromosomes 1, 6, and 12 and are interspersed with

repetitive elements (Green et al., 1984b; Collart et al.,

1985). It has been speculated that these repetitive elements

may play a role in recombination and gene duplication during

the evolution of this multi-gene family. The general

structure of the human histone genes is less complex than

many other eukaryotic genes since the histone protein coding

region is not interrupted by introns and the encoded mRNAs











are not polyadenylated (Adensik and Darnell, 1972). However,

the genes coding for histone proteins that are synthesized

in the absence of DNA synthesis (cell cycle independent

histones) generally contain introns and code for mRNAs that

are poly-adenylated (Engel et al., 1982; Harvey et al.,

1983; rush et al., 1985; Wells and Kedes, 1985). Subsequent

discussions will generalize what is true only for the cell

cycle dependent histones unless otherwise indicated.



Histone Gene Requlation: Transcriptional and Post-
Transcriptional Control


The stringent requirement to package newly replicated

DNA into chromatin, the invariable 1:1 molar ratio of

histone protein to DNA and the inability of most eukaryotic

cells to store unbound histone proteins implies a temporal

and functional coupling of histone protein synthesis with

DNA replication (Prescott, 1966; Allfrey et al., 1963).

Support for this hypothesis first came from studies on

synchronized cell cultures of Euplotes eurystomus in which

the synthesis of histone protein was found to occur only

during the S phase of the cell cycle, the time when DNA

replication takes place (Prescott, 1966). Furthermore, the

quantity of histone protein within the cell doubled during S

phase. The temporal coupling of histone protein synthesis

with DNA synthesis has since been confirmed and well

documented by a variety of direct experimental approaches











(Robbins and Borun, 1967; Spaulding et al., 1966; Stein and

Borun, 1972). In vitro translation studies using polysomal

RNA isolated from cells at various stages of the cell cycle

indicated that the histone mRNA was present exclusively

during S phase (Spalding et al., 1966; Gallwitz and Mueller,

1969; Jacobs-Lorena et al., 1972; Gallwitz and Briendl,

1972; Perry and Kelley, 1973; Borun et al., 1975).

Quantitation of histone mRNA content during the cell cycle

by Northern blot analysis, using radiolabeled histone cDNA

and genomic clones, demonstrated that histone mRNA steady

levels closely parallel the rates of histone protein

synthesis as well as the rates of DNA replication (Heintz et

al., 1983; Plumb et al., 1983a; Baumbach et al., 1984). The

timing of histone protein synthesis during DNA synthesis is

not coincidental and represents a functional relationship

between these two processes. This is supported by the

following observations: 1) there is a coordinate induction

of histone protein synthesis and DNA synthesis during

stimulation of quiescent cells to proliferate (DeLisle et

al., 1983; Green et al., 1984a); 2) inhibition of DNA

synthesis with compounds such as hydroxyurea or cytosine

arabinoside results in a loss of histone mRNA with a

concomitant decrease in histone protein synthesis (Breindl

and Gallwitz, 1973; Butler and Mueller, 1973; Borun et al.,

1975; Baumbach et al., 1984) and 3) terminal differentiation

is marked by a down regulation in histone protein synthesis











in conjunction with the decrease in proliferative activity

(Bird et al., 1985; Challoner et al., 1989; Stein et al.,

1989).

The expression of histone genes during the cell cycle

is regulated at both the transcriptional and

posttranscriptional levels. Analysis by hybrid selection of

RNA radiolabeled in synchronized HeLa cells demonstrates a

burst of 3- to 5- fold in transcription of the histone genes

within the first two hours of S phase, followed by a

reduction to a basal level that remains constant until the

following S phase (Plumb et al., 1983a). Similar results

were obtained when histone gene transcription during the

cell cycle was measured by nuclear run-on analysis (Sive et

al., 1984; Baumbach et al., 1987). Steady state levels of

histone mRNA, however, were elevated 15-20 fold during S

phase and closely paralleled the rates of DNA synthesis

(Plumb et al., 1983a; Heintz et al., 1983; Baumbach et al.,

1984). Maximal accumulation of histone mRNA occurred at 5

hours into S phase and returned to a basal level as DNA

synthesis ended. The timing and extent of the increase in

histone transcription can not account for the marked

increase in histone mRNA steady state levels and must be a

consequence of a posttranscriptional control mechanism.

Posttranscriptional regulation of histone gene expression is

further supported by DNA synthesis inhibitor studies. The

estimated half life of histone mRNA in exponentially growing











HeLa cells is 45-60 minutes (Sittman et al., 1983; Plumb et

al., 1983). Hydroxyurea, or any other compound or treatment

that inhibits DNA synthesis studied to date, results in a

rapid and selective destabilization of histone mRNA (half

life of 10-15 minutes) with little or no effect on histone

gene transcription (Sittman et al., 1983; Plumb et al.,

1983; Baumbach et al., 1987).

A long-standing observation concerning histone mRNA

turnover during inhibition of DNA synthesis is that the

mechanism is completely dependent on protein synthesis

(Butler and Mueller, 1973; Gallwitz, 1975; Stahl and

Gallwitz, 1977; Stimac et al., 1983; Baumbach et al., 1984;

Helms et al., 1984). Pretreatment of cells with inhibitors

of protein synthesis such as high salt, cycloheximide or

puromycin followed by the addition of a DNA synthesis

inhibitor prevents the destabilization of histone mRNAs and

actually results in an increase in their steady state levels

due to the continued transcription of the histone genes.

This observation led Butler and Mueller and others to

hypothesize that the histone proteins can autoregulate their

own synthesis through a feedback loop operative at the

posttranscriptional level (Butler and Mueller, 1973; Stein

and Stein, 1984; Wu and Bonner, 1985). The model predicts

that at the natural end of S phase or when DNA synthesis is

interrupted with inhibitors a transient accumulation of

unbound histone protein may arise due to a loss in high-











affinity binding sites for the newly synthesized histone

proteins. Unbound histone protein may then interact with

histone mRNA-containing polysomes resulting in a

conformational change in the mRNA that is more accessible to

ribonuclease. Growing support for this model comes from a

cell-free mRNA decay system in which the destabilization of

histone mRNA is accelerated when incubated in the presence

of histone proteins (Peltz and Ross, 1987). This result

appears to be specific for histone mRNA since the rates of

decay for gamma globin mRNA, c-myc mRNA, total poly(A)" RNA

or total poly(A)+ RNA are unaffected by the presence of

histone proteins.

Gene transfer studies demonstrated that the minimal

sequence necessary for coupling histone mRNA stability with

DNA replication lies within the 3' terminus of the mRNA

(Stauber et al., 1986; Levine et al., 1987; Pandey and

Marzluff, 1987). Removal of only 60 nucleotides from the 3'

end of a murine H3 histone mRNA releases the mutant mRNA

from destabilization during inhibition of DNA synthesis

(Levine et al., 1987). Conversely, the last 30 nucleotides

of histone H3 mRNA when fused to the 3' end of human gamma

globin mRNA are sufficient to confer instability to the

chimeric mRNA when DNA synthesis is inhibited, whereas wild

type gamma globin mRNA remains stable (Pandey and Marzluff,

1987). Contained within the minimal target sequence for

destabilization is a stem-loop structure which is common to











all known cell cycle dependent histone mRNAs (Hentschel and

Birnstiel, 1981). The presence of the stem-loop structure is

not sufficient by itself to couple mRNA stability with DNA

replication. The target sequences must be present in the

proper position at the 3' terminus of the mRNA to be

operative. Aberrant histone mRNA transcripts containing the

target sequences at an internal position are fully stable

during the inhibition of DNA synthesis (Alterman et al.

1985; Luscher et al., 1985; Pandey and Marzluff, 1987;

Levine et al., 1987). Furthermore, moving the stem-loop

structure 380 nucleotides downstream from the translation

stop codon abolishes the coupling of histone mRNA stability

with DNA replication despite the retention of the stem loop

structure at the 3'terminus (Graves et al., 1987).

An additional constraint on the histone

posttranscriptional regulatory mechanism is that translation

must proceed to within close proximity of the natural

termination codon for destabilization to occur during

inhibition of DNA synthesis (Graves et al., 1987). Frame-

shift mutations in histone mRNA resulting in the premature

termination of translation more than 300 nucleotides from

the 3' terminus stabilize the messenger RNA when DNA

synthesis is interrupted (Graves et al., 1987). Other frame-

shift mutants that allow translation to continue past the

natural UGA stop codon into the 3' untranslated region are

also stable when DNA synthesis is inhibited (Capasso et al.,











1987). Correcting for translation read through in these

mutants by inserting stop codons in all three reading frames

at the natural translation termination site restores the

coupling of mRNA stability to DNA replication.

Detailed studies on deletion mutants suggest that the

histone nascent polypeptide is not involved in the

posttranscriptional regulatory process (Capasso et al.,

1987). Systematic removal of approximately 100 nucleotide

blocks spanning nearly the entire protein coding region had

no effect on the destabilization of these mutant histone

mRNAs when DNA synthesis was inhibited. Consistent with this

result, the deletion of the ATG translation start codon

which prevents the synthesis of histone protein did not

prevent the destabilization of the mRNA during inhibition of

DNA replication. Furthermore, as discussed above, the

minimal target sequences containing the stem-loop structure

confer instability to gamma globin mRNA when DNA synthesis

is interrupted (Pandey and Marzluff, 1987). The globin-

histone chimeric mRNA, although regulated as wild type

histone mRNA, does not encode any histone protein sequences.

Destabilization of histone mRNA appears to occur in a

stepwise manner initiating at the 3' terminus (Ross et al.

1986). The first detectable histone mRNA degradation product

lacks 5 nucleotides from the 3' terminus (Ross and Kobs,

1986; Ross et al., 1986). This degradation product is

detectable in the in vitro mRNA decay system as well as in









10

whole cells (Ross et al., 1986). It is unclear whether this

product arises from a single endonucleolytic scission or

from multiple cuts removing one or more of the nucleotides

at a time. Degradation continues rapidly in a 3' to 5'

direction carried out by an exonuclease which is part of the

ribosome complex (Ross et al., 1987). These results are

consistent with the requirement for translation to proceed

near the 3' terminus for the destabilization of the histone

mRNA during inhibition of DNA synthesis.



Subcellular Compartments

The appearance of histone mRNA on polysomes within

minutes after transcription and the equally rapid transfer

of the newly synthesized histone proteins into the nucleus

where they complex with DNA imply that histone protein

synthesis may be compartmentalized in a region located just

outside the nucleus (Spaulding et al., 1966; Borun et al.,

1967; Robbins and Borun, 1967). The asymmetric distribution

of eukaryotic mRNAs in the cytoplasm supports this

possibility. Morphological studies initially demonstrated

that polyribosomes are concentrated in areas surrounding the

nucleus in mouse 3T3 cells (Fulton et al., 1980). Later,

localization of specific mRNAs by in situ hybridization

techniques provided direct evidence that mRNAs are

differentially distributed throughout the cell. Tubulin mRNA

is preferentially located in the periphery of the cytoplasm









11

whereas vimentin mRNA is concentrated in perinuclear regions

in chick myoblast and fibroblast cells (Lawrence and Singer,

1986). Furthermore, actin mRNA is concentrated in the

lamellipodia, structures involved in cell locomotion

(Lawrence and Singer, 1986). The requirement for actin

protein in the lamellipodia during cell movement (Wang,

1984) suggests a functional relationship between where the

mRNA is localized and where the encoded protein is needed

(Lawrence and Singer, 1986).

Localization of mRNAs in distinct areas of the

cytoplasm may occur due to an association of the mRNA with

an underlying structure. This structure could serve as an

anchor to concentrate mRNAs in particular regions of the

cytoplasm. Consistent with this reasoning is the existence

of a complex, proteinaceous structure in the cytoplasm of

most eukaryotic cells that is referred to as the

cytoskeleton. The cytoskeleton plays a role in cell shape,

cell motility and the intracellular transport of

macromolecules and organelles (Clarke and Spudich, 1977;

Goldman et al., 1978; and reviewed in Nielsen et al., 1983).

In addition, there is growing support for the hypothesis

that the cytoskeleton plays a key role in regulating the

translation process (Lenk et al., 1977; Lenk and Penman,

1979; Fulton et al., 1980; Cervera et al., 1981; van Verooij

et al., 1981; Nielsen et al., 1983; Howe and Hershey, 1984).











The Cytoskeleton: Microtubules. Intermediate Filaments
and Microfilaments


The cytoskeleton structure is an interconnected system

of three major classes of protein fibers, namely the

microtubules, intermediate filaments and microfilaments.

Each system represents a heterogenous population of

proteins. The microtubules, for example, are composed of 13

protofilaments of the well characterized alpha and beta

tubulin proteins (Amos and Klug, 1974). These filaments are

25 nm in diameter and appear to contain a hollow core.

Several microtubule associated proteins (MAPs) have also

been identified which appear to facilitate the assembly of

the microtubules and promote their interactions with other

cellular organelles (Sloboda et al., 1976, Bulinski and

Borisy, 1980; Bloom et al., 1984; Binder et al., 1985). It

has long been recognized that in mitotic cells the

microtubules reorganize into the mitotic spindle including

the astral, kinetochore and continuous fibers which mediate

the segregation of the chromosomes (Weber et al., 1975).

During interphase the role of the microtubules is less clear

and somewhat controversial. Studies on agents that cause the

depolymerization of the microtubules, such as colchicine and

colcemid, suggest that the microtubules play a role in cell

shape and the intracellular transport of macromolecules and

organelles (Hayden and Allen, 1984; Vale et al., 1985).











There are at least five major classes of protein that

constitute the intermediate filaments: 1) desmin, 2)

vimentin, 3) keratin, 4) neurofilaments and 5) glial

filaments. Biochemical and immunological studies

demonstrated that these proteins are generally expressed in

a celltype specific manner (for review see Lazarides,

1980). Desmin is an acidic protein with a molecular weight

of 53,000 Daltons and is predominantly found in intermediate

filaments from cardiac, smooth and skeletal muscle cells

(Lazarides and Hubbard, 1976; Hubbard and Lazarides, 1979).

The localization of desmin at the Z line in muscle suggests

a role for this protein in the coordination of the

contractile actions of the muscle fiber (Lazarides and

Hubbard, 1976; Izant and Lazarides, 1977).

Vimentin is found principally in intermediate filaments

from cells of mesenchymal origin. Immunological and

biochemical characterization of vimentin indicated that the

protein is 52,000 Daltons and widely conserved throughout

evolution (Brown et al., 1976; Bennett et al., 1978; Franke

et al., 1979). A characteristic feature of the vimentin

intermediate filaments is their sensitivity to colchicine

which erroneously led researchers to believe initially that

they were disaggregation products of the microtubules

(Goldman and Knipe, 1973; Holltrop et al., 1974; Blose and

Chacko, 1976). The collapse of vimentin into filamentous

bundles around the nucleus during colchicine treatment









14

suggests that these filaments are in close association with

the microtubules.

The keratin filaments, also known as tonofilaments, are

predominantly found in epithelial cells. The keratin

proteins are a family of approximately 30 distinct

polypeptides ranging in molecular weight from 40,000-65,000

Daltons (Steinert, 1975; Steinert and Gullino, 1976; Sun and

Green, 1978; Moll et al., 1982). The keratins can be clearly

divided into acidic and basic groups based on two-

dimensional electrophoretic analysis. Although only a few

keratin subtypes are expressed in any one epithelial cell,

there is always a coordinated synthesis of at least one

class each of the acidic and basic keratin subunits (Moll et

al., 1982; Woodcock-Mitchell et al., 1982; Fuchs and

Marchuk, 1983). The combination of keratin subtypes

expressed by an epithelial cell is characteristic of the

tissue origin as well as the differentiation state of the

cell (Roop et al., 1983; Sun et al., 1983). In addition,

some epithelially derived tumors express different keratin

subtypes than those found in the normal counterpart and

these differences may be useful in diagnostic tests

(Summerhayes et al., 1981; Osborn, 1983).

Little is known about the function and organization of

the intermediate filaments from neuronal and glial cells.

The major cytoskeletal element of neuronal axons and

dendrites are the neuronal filaments which are generally









15

composed of three large polypeptides with molecular weights

of 200,000, 150,000, and 68,000 Daltons (Hoffman and Lasek,

1975; Delacourte et al., 1980). Glial filaments are

specifically found in astrocytes and other types of glial

cells and consist of a single protein, the glial fibrillary

acidic protein (GFA), with a molecular weight of 51,000

Daltons (Schachner et al., 1977; Goldman et al., 1978).

Although the intermediate filament proteins from the

different cell types are heterogeneous in their chemical

composition, they assemble into similar morphological

structures consisting of filaments that are 10 nm in

diameter that are highly resistant to extraction with

detergents. It has been proposed that the intermediate

filaments contain a constant and variable region, analogous

to the structure of antibodies. The constant domain would

conserve the morphology whereas the variable domain would

confer functional specificity (Lazarides, 1980).

The microfilaments are the thinnest of the cytoskeletal

filament systems, with an average diameter of 6 nm, and are

composed primarily of actin protein. It has been estimated

that approximately 5-10% of total cellular protein is actin

which exists in an equilibrium between monomeric and

filamentous forms (Hartwig and Stossel, 1975).

Microfilaments from non-muscle cells contain two predominant

species of actin protein, namely beta- and gamma-actin. The

non-muscle actin proteins are quite similar to muscle











specific alpha-actin with respect to molecular weight

(42,000 Daltons), amino acid sequence and polymeric

structure (Tilney and Detmers, 1975; Hartwig and Stossel,

1975; Kane, 1975; Garrels and Gibson, 1976; Whalen et al.,

1976; Rubenstein and Spudich, 1977). These results suggest

that beta- and gamma-actin protein in non-muscle cells may

play a role in contraction processes such as cell motility ,

chromosome movement and exocytosis (Berl et al., 1973;

Huxley, 1973; Sanger, 1975).



Cytoskeletal-mRNA Interactions

Early studies using stereo-electron microscopic

analysis of whole cells indicated that polysomes are

associated with the cytoskeleton structure (Wolosewick and

Porter, 1977). Closer examination of the cell interior, by

removing membranes and soluble cytoplasmic proteins with

detergent, reveals that many of the ultrastructural features

and nearly all the polysomes are retained on the

cytoskeletal framework (Lenk et al., 1977). Furthermore, the

non-random topological distribution of the polysomes is

preserved in the cytoskeleton preparation (Fulton et al.,

1980). The association of the polysomes with the

cytoskeleton does not appear to be due to trapping since

monosomes and independent ribosomal subunits are efficiently

removed by the detergent extraction (Lenk et al., 1977;

Cervera et al., 1981; Ornelles et al., 1986). These results









17

indicate that actively translated mRNAs are associated with

the cytoskeleton and this association may be requisite for

protein synthesis. Consistent with this possibility, the

inhibition of host protein synthesis coincides with the

release of host mRNA from the cytoskeletal structure and the

attachment of viral mRNA to the cytoskeleton in poliovirus

infected and adenovirus infected cells (Cervera et al.,

1981; Lemieux and Beaud, 1982). During development,

maternally inherited mRNAs participate in protein synthesis

only at the time they become associated with the

cytoskeleton in sea urchin oocytes (Moon et al., 1983).

The region of the polysome involved in its attachment

to the cytoskeleton is not known. Previous results indicate

that the association of polysomal mRNA with the cytoskeleton

is through the mRNA or mRNP particle and not through the

ribosome (Lenk et al., 1977; Lenk and Penman, 1979; Cervera

et al., 1981; Lemieux and Beaud, 1982). Disruption of

polysomes by heat shock, high salt, sodium fluoride or

pactamycin treatment releases the ribosomal subunits but not

the mRNA from the cytoskeleton into the soluble phase (Lenk

et al., 1977; Cervera et al., 1981; van Venrooij et al.,

1981; Howe and Hershey, 1984). Immunofluorescence studies

indicate that the cap-binding protein is associated with the

cytoskeleton suggesting that the attachment of eukaryotic

mRNAs to the cytoskeleton may occur via the 5' mRNA cap

structure (Zumbe et al., 1982). Earlier studies suggest that











the association of eukaryotic mRNAs with the cytoskeleton

may occur through the poly-A tract (Milcarek and Penman,

1974). However, the association of uncapped poliovirus mRNAs

(Bonneau et al., 1985) and poly-A- histone mRNAs (Jeffery,

1984; Bonneau et al., 1985) indicates that the attachment

site might also exist in an internal region contained within

the 5' and 3' ends of the mRNA. Isaacs and Fulton (1987)

have presented evidence suggesting that myosin heavy chain

in embryonic chicken muscle cells is co-translationally

assembled into the cytoskeleton. The myosin nascent

polypeptide would therefore serve as an additional

cytoskeleton attachment site for the polysomal myosin mRNA.

Based on these results it appears that several mechanisms

may operate in the attachment of a specific polysomal mRNA

or mRNP to the cytoskeleton and these attachment sites may

vary depending on the species of mRNA.

The component of the cytoskeleton to which the mRNA is

attached is not known although the data accumulated to date

suggest that the microfilaments may be involved in the mRNA-

cytoskeleton association. Removal of tubulin from the

cytoskeleton by using cold calcium-containing buffers during

the isolation procedure has no effect on the attachment of

mRNA with the cytoskeleton; this result suggests that the

microtubules do not play a role in the association of mRNA

with the cytoskeleton (Lenk et al., 1977; Howe and Hershey,

1984; Ornelles et al., 1986). Disruption of vimentin









19

filaments with heat shock or indirectly with colcemid has no

effect on protein synthesis as determined by two-dimensional

electrophoresis indicating that translation is not dependent

on the integrity of the intermediate filaments (Welch and

Feramisco, 1985). Other reports suggest that the attachment

of mRNA to the cytoskeleton is mediated by the

microfilaments (Lenk et al., 1977; Howe and Hershey, 1984;

Ornelles et al., 1986). Treatment with cytochalasins, either

cytochalasin B or D, to disrupt the actin-containing

microfilaments results in the release of poly-A+ mRNA from

the cytoskeleton into the soluble phase. At least in the

case of cytochalasin D treatment, the release of poly-A+

mRNA from the cytoskeleton occurs in a dose dependent manner

and stoichiometrically reflects the degree of inhibition of

protein synthesis (Ornelles et al., 1986). These results

indicate that intact microfilaments are required for the

association of mRNA to the cytoskeleton; however, it can not

be concluded that the actin fibers directly bind the mRNA.



Nonmembrane-Bound and Membrane-Bound Polyribosomes:
The Signal Hypothesis


As described above, nearly 100 percent of the actively

translated polysomes are associated with the cytoskeleton.

Studies on vesicular stomatitis virus infected HeLa cells

demonstrated that membrane-bound polysomal mRNA coding for

the viral G glycoprotein and "free" polysomal mRNA coding









20
for the polymerase, nucleocapsid and matrix proteins are all

attached to the cytoskeleton during translation (Cervera et

al., 1981). The association of free polysomes with the

cytoskeleton has also been demonstrated by electron

microscopy (Wolosewick and Porter, 1979). Free polysomes are

generally the class of polysomes involved in the synthesis

of intracellular proteins whereas membrane-bound polysomes

are generally involved in the synthesis of exported proteins

that are destined for the cell surface or secretion (for

review see Emr et al., 1980). The association of free

polysomal mRNAs with the cytoskeleton suggests that the

polysome is anchored to a scaffold, albeit a dynamic

structure, and not freely diffusible within the cytoplasm.

The term "free polysome" is operationally defined and is a

misnomer when viewed in light of the biochemical and

microscopic studies on the cytoskeleton. The proper

reference to free polysomes is nonmembrane-bound polysomes.

The mechanism by which mRNAs assemble into membrane-

bound polysomes has been described in the signal hypothesis

(Blobel and Dobberstein, 1975). Many types of signal

peptides exist that target proteins to various subcellular

membrane compartments such as the endoplasmic reticulum,

mitochondria and chloroplasts in eukaryotes as well as the

cell membrane in prokaryotes (Date et al., 1980; Schatz and

Butow, 1983; Muller and Blobel, 1984; Cline, 1986). The work

presented here is focused on events that occur at the









21

endoplasmic reticulum and on what is generally known for the

synthesis of proteins that are localized on the cell surface

or secreted from the cell. Most cell surface or secreted

proteins are initially synthesized with a segment of amino

acid residues at the N-terminus that is not usually found in

the mature protein. This segment is commonly referred to as

a signal peptide. Although signal peptides are not highly

conserved in sequence, they are fairly conserved in

structure and based on the compilation of 277 signal

peptides ranging from bacteria to higher eukaryotes, a

consensus structure has been deduced (Watson, 1984). In

general, signal peptides are 25-30 amino acids in length.

They contain a charged amino acid within the first five

residues, which is followed by a stretch of at least nine

hydrophobic amino acids, and terminate at a small uncharged

amino acid such as alanine, serine or glycine. It is

believed that the hydrophobic domain of the signal peptide

facilitates the transfer of the protein across the lipid

environment of specific intracellular membranes whereas the

small, uncharged amino acid is thought to direct the

protease to cleave the peptide from the mature protein

(Hortin and Boime, 1981).

Presumably, translation of mRNA encoding exported

proteins is initiated on nonmembrane-bound polysomes.

Following the synthesis of approximately 70 amino acids, the

point at which the entire signal peptide has emerged from









22
the ribosome, the signal recognition complex (SRP) binds to

the signal peptide and arrests translation (Walter and

Blobel, 1981). The SRP is an 11S ribonucleic acid-protein

complex composed of six non-identical polypeptides and one

small cytoplasmic RNA molecule (Walter and Blobel, 1980;

Walter'and Blobel, 1982; Walter and Blobel, 1983). The RNA

molecule of SRP is 300 ribonucleotides in length and shares

approximately 80% homology with a middle repetitive human

DNA family (Ullu et al., 1982; Li et al., 1982). With the

SRP bound to the signal peptide, the complex can then

associate with the membrane of the endoplasmic reticulum

through the interaction with the SRP receptor which is also

known as the docking protein (Gilmore et al., 1982a; Gilmore

et al., 1982b; Meyer et al., 1982a; Meyer et al., 1982b).

The SRP then dissociates from the polysome with the

resultant release of the translational block. As translation

resumes the protein is co-translocated from the cytoplasm

into the endoplasmic reticulum. During translocation, a

signal peptidase which is located in the lumen of the

endoplasmic reticulum removes the signal peptide from the

protein (Maurer and Mckean, 1978; Walter et al., 1984).



Overview of Project

The overall aim of this work is to better define the

mechanisms) involved in coupling histone protein synthesis

with DNA replication. Specifically, we have focused our











attention on the posttranscriptional regulation of histone

gene expression. The evidence cited above suggests that the

cytoplasm of eukaryotic cells is a highly organized region

which contains subcellular compartments that concentrate

translation factors as well as specific mRNAs (Fulton et

al., 1980; Howe and Hershey, 1984; Lawrence and Singer,

1986; Ornelles et al., 1986). Traditionally, the idea of

subcellular compartments refers to organelles such as the

nucleus, endoplasmic reticulum, and mitochondria. For

example, a popular biochemical definition of

compartmentation is "a simple regulatory mechanism in which

the enzymes for catabolism and biosynthesis are physically

separated by membrane-bound cell organelles" (Rawn, 1983). A

broader definition of subcellular compartmentation includes

any specialized region of the cell regardless of the

presence of a membrane and may be extended to such

components and regions as 1) nonmembrane-bound and membrane-

bound polysomes, 2) cytoskeleton and soluble phase, and 3)

spatial location (i.e., near the nucleus versus near the

plasma membrane).

Histone mRNAs and the factors that are involved in

their selective degradation during the inhibition of DNA

synthesis may be localized in a particular subcellular

compartment. Consistent with this reasoning, the subcellular

compartment may influence the posttranscriptional control of

histone gene expression. Based on this possibility, we have











specifically asked the question: Does the subcellular

location of histone mRNA play a role in the destabilization

of histone mRNA during inhibition of DNA synthesis? To

address this question, we have applied a combination of

molecular and cellular approaches first, to determine the

natural location of histone mRNA and second, to alter the

subcellular compartment to study whether the foreign

location affects histone mRNA stability. The rationale for

this approach is that directing histone mRNA to an altered

subcellular location may physically remove the message from

the factors that are involved in the selective degradation

of histone mRNA during inhibition of DNA synthesis.

Our initial studies demonstrated that histone mRNAs are

translated predominantly on nonmembrane-bound polysomes that

are associated with the cytoskeleton. The subcellular

location of the mRNA was altered biologically by

constructing a histone fusion gene that contains sequences

coding for the signal peptide of the beta-lactamase gene of

E. coli pBR322 plasmid DNA. The encoded fusion histone mRNA

was directed to membrane-bound polysomes and was not

destabilized during inhibition of DNA synthesis.

Inactivation of signal peptide function by a single

nucleotide substitution in the ATG translation start codon

of the signal peptide resulted in the association of the

mutated fusion mRNA with nonmembrane-bound polysomes and

restored the coupling of the mRNA stability with DNA











replication. These results indicated that the subcellular

location plays a role in the posttranscriptional control of

histone gene regulation.

During the course of studying histone mRNA subcellular

localization it was necessary to examine histone mRNA-

cytoskeletal interactions in detail. Previously, Ornelles et

al. (1986) reported that total poly-A+ mRNA, as a

heterogenous population, is released from the cytoskeleton

in HeLa cells by cytochalasin D treatment. We demonstrated

by Northern blot analysis that cytochalasin D treatment of

HeLa cells readily releases histone mRNA, which is not poly-

adenylated, from the cytoskeleton into the soluble phase.

Surprisingly, mRNA coding for the cell surface HLA-B7 class

I antigen remained associated with the cytoskeleton during

cytochalasin D treatment. Subsequent studies demonstrated

that HLA-B7 mRNA expresses at least two distinct

cytoskeletal attachment sites: a cytochalasin D sensitive

site and a puromycin sensitive site. The puromycin sensitive

site appears to be a feature of membrane-bound polysomal

mRNAs and is most likely the nascent polypeptide and/or

ribosome associated with the protein substructure of the

endoplasmic reticulum.

Cytochalasin D and puromycin cotreatment, conditions

that efficiently release HLA-B7 membrane-bound polysomal

mRNA and histone nonmembrane-bound polysomal mRNA from the

cytoskeleton into the soluble phase, did not dissociate the











signal peptide-histone fusion mRNA from the cytoskeleton.

The nucleotide sequences of the signal peptide-histone

fusion mRNA encoding the signal peptide contain a

cytochalasin D and puromycin insensitive cytoskeletal

attachment site which is not expressed by histone mRNA or

HLA-B7 mRNA. These results provide evidence for the

heterogeneic pattern for cytoskeleton-mRNA interactions.

Lastly, we investigated the role of the cytoskeleton in

the regulation of gene expression by studying the effects of

disrupting the microfilaments with cytochalasin D on the

transcription and steady state mRNA levels of several genes.

We determined that the transcriptional and

posttranscriptional regulation of the c-fos proto-oncogene

is quite sensitive to the perturbation of the cytoskeleton

structure. Transcription of the c-fos gene is selectively

stimulated with a concomitant increase in c-fos mRNA steady

state levels within minutes of the addition of cytochalasin

D. Consistent with previous results, the transcription of

beta-actin was also elevated in cytochalasin D treated

cells. These findings suggests that the cell monitors the

organization of the cytoskeleton and selectively regulates

gene expression accordingly.















CHAPTER 2

GENERAL METHODS



Materials

[a-3P]dCTP (-3,000 Ci/mmol), [a- P]dATP (-3,000

Ci/mmol), [gamma-32P]ATP (-3,000 Ci/mmol) and [a- S]dATP

(>1,000 Ci/mmol) radionucleotides were purchased from

Amersham, Arlington Heights, IL; polyvinylsulfonic acid

(PVS), polyethylene glycol 8,000 (PEG), XAR-5 x-ray film

were obtained from Eastman Kodak Co., Rochester, NY;

ultrapure electrophoresis grade agarose and Zeta-probe nylon

membranes were from Bio Rad, Richmond, CA; formaldehyde

solution (37% w/w) was from Fisher Scientific, Fair Lawn,

NJ; nitrocellulose was from Schleicher and Schuell, Keene,

NH; RPI scintillator ppo-popop concentrated liquid

scintillator (Liquifluor) was from Research Products

International, Elk Grove Village, IL; formamide was from

Bethesda Reasearch laboratories (BRL), Gaithersburg, MD;

acrylamide, calf intestine alkaline phosphatase, nuclease

Sl, E. coli DNA polymerase I large fragment (Klenow enzyme)

and most restriction endonucleases were obtained from

Boehringer Mannheim Biochemicals (BMB), Indianapolis, IN; T4

DNA ligase was from New England Bio Labs (NEB), Beverly, MA;











T4 polynucleotide kinase (cloned) was from United States

Biochemical Corp. (USB), Cleveland, OH; cytochalasin D,

polyoxyethylene sorbitan monopalmitate (Tween 40), sodium

deoxycholate (NaDOC), sodium dodecyl sulfate (SDS), Triton

X-100, and DNase I (DN-EP) were purchased from Sigma, St.

Louis,'MO; lysozyme was purchased from Worthington

Biochemicals, Freehold, N.J.; fetal bovine, calf bovine and

horse sera were obtained from Gibco Laboratories, Grand

Island, NY or Flow Laboratories, McLean, VA; cycloheximide

was kindly provided by Upjohn, Kalamazoo, Michigan.



Conditions for Enzyme Reactions

The buffers and incubation conditions for each of the

restriction endonucleases and DNA modifying enzymes used in

this study were as described by the specific recommendations

of the manufacturers unless otherwise indicated.



Mammalian Cell Culture

HeLa Monolayer Cell Cultures

HeLa S3 suspension cells (human cervical carcinoma cell

line; ATCC ccl 2.2) were seeded into EMEM (Eagles-modified

minimum essential medium) containing 5% fetal calf serum, 5%

horse serum, 100 U/ml penicillin, 100 Ag/ml streptomycin and

1 mM glutamine at 3x106 cells per 10 cm tissue culture dish

and incubated at 370C under 5% CO2. Cells were maintained at

sub-confluent densities by splitting the cultures 1:10 into

fresh medium every 3 to 5 days.











HeLa Suspension Cell Cultures

HeLa S3 cells were grown and maintained in suspension

at 3-6x105 cells/ml in SMEM (Joklik-modified minimum

essential medium) supplemented with 7% calf serum, 100 U/ml

penicillin, 100 Cg/ml streptomycin and 1 mM glutamine in a

warm room at 370C. Monoclonal and polyclonal HeLa S3 cell

lines, which were isolated during G418 selection of

transfected monolayer cultures, were maintained in SMEM

containing 5% fetal calf serum, 5% horse serum, 100 U/ml

penicillin, 100 ig/ml streptomycin and 1 mM glutamine at 3-

6x105 cells/ml.



WI-38 Monolayer Cell Cultures

WI-38 cells (normal, human diploid fibroblast-like

cells; ATCC ccl 75) at passage 28, were grown as monolayers

in EMEM containing 10% fetal calf serum, 100 U/ml

penicillin, 100 ig/ml streptomycin and 1 mM glutamine at

37C under 5% CO2. Cell cultures were made quiescent by

maintaining cultures in EMEM containing 0.5% fetal calf

serum, 100 U/ml penicillin, 100 gg/ml streptomycin and 1 mM

glutamine at 370C under 5% CO2 for 48 hours.



Metabolic Inhibitor Treatments

Inhibition of DNA synthesis was usually performed with

1-5 mM hydroxyurea for 30 to 60 minutes. The drug was always

prepared fresh in IX spinner salts (Gibco) as a 100 mM stock

solution.











Protein synthesis was inhibited with 10 gg/ml

cycloheximide for varying lengths of time. Cycloheximide

stock solutions [500 gg/ml] were prepared fresh in IX

spinner salts. Protein synthesis was also inhibited with 0.4

mM puromycin for varying lengths of time. Puromycin stock

solutions [54.4 mg/ml] were prepared in dimethyl sulfoxide

(DMSO) and stored at -200C.

To study the dependence of protein synthesis on the

destabilization of histone mRNA during the inhibition of DNA

synthesis, we pre-treated cells with 10 gg/ml cycloheximide

for 5 minutes and then added 1 mM hydroxyurea for an

additional 30 minutes.

Disruption of the cytoskeleton was generally achieved

by treating the cells with 5 to 60 gg/ml cytochalasin D for

15-20 minutes at 370C. The cytochalasin D stock solutions

[5-10 mg/ml] were prepared in DMSO and stored at -200C.



Plasmid DNA

The human histone genes used in this study were

previously isolated by Sierra et al. from a lambda Ch4A

bacteriophage library (1982). The histone genes were

subsequently subcloned into Escherichia coli pBR322 plasmid

DNA and propagated in E. coli strain HB101 bacterium (Sierra

et al., 1982; Plumb et al., 1983; Stein et al., 1984).

Recombinant plasmids ST519 and FF435C each contain a

distinct genomic DNA fragment encoding a human H3 histone











gene. The recombinant plasmid FO108A contains an entire

human H4 histone gene which includes the promoter, coding

and 3' non-transcribed regions.

An HLA-B7 cDNA clone, pDP001, was generously supplied

by Dr. Sherman Weissman, Yale University, New Haven,

Connecticut (Sood et al., 1981). The pDP001 plasmid DNA was

constructed from the human cell line RPMI 4265 and contains

a 1400 bp insert coding for HLA-B7 in the PstI site of

pBR322 plasmid DNA. Therefore, the plasmid confers

tetracycline resistance and not ampicillin resistance to E.

coli HB101 bacteria.

The phCGa plasmid DNA was generously provided by Dr.

John Fiddes, California Biotechnology Inc., Mountain View,

California and Dr. John Nilson, Case Western Reserve

University, Cleveland, Ohio. The phCGa plasmid is a cDNA

clone containing a 621 bp fragment coding for the alpha

subunit of human chorionic gonadotropin inserted into the

HindIII site of pBR322 and confers ampicillin resistance

(Fiddes and Goodman, 1979).

The beta-lactamase signal peptide-globin fusion gene,

pSP125e, was kindly provided by Dr. Vishwanath Lingappa,

University of California at San Francisco. The pSP125e clone

contains the signal peptide from the beta-lactamase gene of

E. coli plasmid pBR322 fused in frame to the coding sequence

for chimpanzee alpha globin. This fusion gene was subcloned

into the pSP64 vector (Promega Biotech) as a HindIII-PstI









32
fragment and is under control of the SP6 promoter (Simon et

al., 1987).

To study the regulation of histone mRNA localized on

membrane-bound polysomes, we constructed a beta-lactamase

signal peptide-histone fusion gene as outlined in Figure 2-

1. The first intermediate, pSPpSTdeltaHH/E, involved the

placement of the E. coli pBR322 beta-lactamase signal

peptide-globin fusion gene from pSP125e under

transcriptional control of the cell cycle-dependent H3

histone regulatory region of pST519deltaH. The 840 bp

EcoRI/HindIII fragment from pST519deltaH (Marashi et al.,

1986) was isolated electrophoretically and blunt-ended using

the Klenow fragment of DNA polymerase I. The signal peptide-

globin fusion construct was digested with BglII restriction

endonuclease, which cuts once in the untranslated leader

region of the fusion gene, blunt-ended with Klenow fragment

and dephosphorylated using calf intestinal alkaline

phosphatase. The 840 bp blunt-ended fragment from pST519 was

ligated into the BglII blunt-ended site of pSP125e using T4

DNA ligase. The construction of the second intermediate,

pSPpSTdeltaHH/EpST, involved the substitution of the

chimpanzee globin coding sequences in pSPpSTdeltaHH/E with

the H3 histone coding sequences from pST519. The

pSPpSTdeltaHH/E DNA was digested with NcoI, which cuts at

the signal peptide-globin coding region junction, and

treated with calf intestinal phosphatase. The pST519 DNA was









Figure 2-1. Outline of the cloning scheme for the
construction of the signal peptide-histone fusion gene.

Restriction endonuclease sites are as follows: B,
BglII; E, EcoRI; H, HindIII; N, NcoI; P, PstI; S, SmaI.
Abbreviations: S.P., E. coli pBR322 beta-lactamase signal
peptide coding sequences; H3REG, transcriptional H3 histone
regulatory region including CAAT and TATAA consensus
sequences; CAP, histone H3 mRNA transcription start site; H3
struct, sequences coding for H3 histone protein; ATG,
translation start codon; TAA, translation stop codon; SP6,
bacteriophage SP6 promoter.













ATG
SI--S.P-- N-Globin--- nSP125e


,1


/2 Leader'

CAP
STATAA I H
T 795bp 125bp |20bp I


EcolHind
Isolate E/H fragment
Klenow


pST519AH
lr\


//2 Lead


SP6
CAP
~ TATAA j,.1 j
H3 1 1-- ,I S.P4GLOBIN I pSPpSTAHH/E


(ATG)
E CAP N H
kH3 Reg- H3 Struct.-- pST519


Nco I ,
Phosphatase

Ncol


Ligate


.SP



pS STAHH/EpST




3
HpaI
Isolate 1.1 Kb fragment
pUC8 Hinc Lgate


6
6 H3Cap H
HISeqa ATG T fp H pp, T/
43 Seq. ,I PP-" 4 -4H3 Seq. I I .IpSPpSTAHH/EpST


,SPH3


i-


40pEnhaAcer-

H3 SQ. 214H3 SEQ-?*p -
N*210bp "


EcoR I
EcoRI Isolate Enhancer cassette
Phosphatase

Ligate


H3 Struct.


P SPE
>F pSPH3E


SV0ES


t -









35

digested with NcoI, which cuts at the ATG translation start

codon, and ligated to NcoI/dephosphorylated pSPpSTdeltaHH/E

DNA using T4 DNA ligase. The signal peptide-histone fusion

gene was subcloned into pUC8 plasmid DNA by lighting the 1.1

kbp HpaI fragment from pSPpSTdeltaHH/EpST into the HincII

site of pUC8 to form pSPH3. The signal peptide-histone

fusion gene in pSPH3 was transcriptionally enhanced by

placing the SV40 viral enhancer element from pSVE108A

(Shiels et al., 1985) into the EcoRI site in the polylinker

of pSPH3 to form pSPH3El (containing one enhancer element)

and pSPH3E2 (containing 2-3 enhancer elements).



DNA Isolation and Purification

Plasmid DNA Preparations

Plasmid DNAs used in the experiments described here

were harbored in Escherichia coli strain HB101 bacteria (F-,

recA 13) unless otherwise indicated. Bacteria containing the

appropriate plasmid were grown in medium containing

antibiotics (usually 25-50 Ag/ml ampicillin or tetracycline)

as specified by the genotype of the plasmid or bacterium.

The following procedure is the routine method for

isolating pBR322 derived plasmid DNAs and includes an

amplification step with chloramphenicol for higher yields of

DNA (Maniatis et al., 1982a). Plasmids that are derived from

pUC8 DNA are also prepared as follows; however, due to their

high copy number the amplification step is unnecessary. The









36

standard protocol for the isolation of pUC8 related plasmid

DNAs is the same as described below except that the 500 ml

culture is inoculated directly and allowed to grow overnight

into the late log stage.

Starter cultures were prepared by inoculating 5 ml LB

mediumn(Luria-Bertani medium: 1% Bacto tryptone, 0.5% Bacto

yeast extract, 1% NaCl) supplemented with 0.1% glucose with

25-50 ul of bacterial glycerol stock and incubated overnight

at 370C. The 5 ml starter culture was transferred to 500 ml

pre-warmed LB medium supplemented with glucose and incubated

at 370C until the optical density at 590 nm (0.D.590) of the

culture reached 0.55 units. Plasmid DNA was amplified by

treating the culture with 200 gg/ml chloramphenicol

(prepared as a 20 mg/ml stock solution in 95% ethanol and

filter sterilized) at 370C for 12-16 hours. The bacteria

were centrifuged at 5,000 rpm, 40C for 10 minutes in a

Beckman JA-10 rotor and J2-21 centrifuge. The cell pellet

was resuspended in 10 ml solution I (50 mM glucose, 25 mM

Tris pH 8.0, 10 mM EDTA, 5 mg/ml lysozyme) and incubated at

room temperature for 5 minutes. The cells were lysed at 0C

with 20 ml solution II (1% SDS, 0.5 M NaOH) for 5 minutes.

The cell lysate was treated with 15 ml 5 M potassium

acetate, pH 4.8 and kept on ice for 5 minutes. The lysate

was transferred to a 50 ml polycarbonate tube and

centrifuged at 16,000 rpm for 20 minutes at 40C in Beckman

JA-20 rotor to pellet the bacterial DNA and cell debris. The









37

supernatant was transferred to a 150 ml Corex tube (Corning)

and nucleic acids were precipitated from solution with 30 ml

isopropanol at room temperature for 15 minutes. The DNA was

pelleted by centrifugation at 5,000 rpm in Beckman JS-7.5

rotor for 30 minutes at 200C. The pellet was briefly dried

under vacuum and resuspended in 8 ml 10:1 TE buffer (10 mM

Tris pH 8.0, 1 mM EDTA). The DNA solution was adjusted to a

final concentration of 1 mg/ml of CsC1 (refractive index of

1.3860) and 600 ig/ml ethidium bromide. The DNA solution was

then transferred to a 16 x 76 mm Quick-Seal tube (Beckman)

and centrifuged to equilibrium at 45,000 rpm in a Beckman

Type-50 Ti rotor and Beckman L5-65 ultracentrifuge for 36-48

hours at 200C. The supercoiled plasmid DNA was removed from

the cesium chloride gradient by puncturing the side of the

tube just below the DNA band with a syringe containing a 22

gauge needle. The DNA was extracted repeatedly with 1-

butanol to remove the ethidium bromide and to reduce the

volume of the solution. Cesium chloride was removed by

dialysis against 10:1 TE buffer at 40C. The DNA was

quantitated by absorbance at 260 nm and stored at -200C.

Routinely, the yield for pBR322 derived plasmids was 500-800

Ag DNA per 500 ml amplified culture.



M13 Bacteriophage DNA Preparations

M13 is a filamentous DNA bacteriophage that is specific

for male E. coli bacteria. M13 phage DNA is useful for site











directed mutagenesis and dideoxy DNA sequencing reactions

since it can be readily isolated as a double-stranded

covalently closed circular DNA replicative form (RF) from

the infected bacteria and as a circular, single-stranded DNA

genome (SS) from the mature phage particle. In addition, M13

phage DNA has been altered genetically to facilitate cloning

and subsequent manipulations. One such derivative, M13mpl8,

is small in size, maintained in high copy number

(approximately 200 RF molecules per cell), contains a

convenient polylinker for the insertion of foreign DNA (up

to 25 kb), and expresses beta-galactosidase which is

inactivated by the insertion of DNA into the polylinker

site.



Single-stranded templates. Single-stranded M13 DNA

templates were prepared from individual "plaques" which

formed on a lawn of E. coli strain JM103 bacteria

(delta(lac-proAB), supE / F'[traD36, proA+, proB+, lacIq

lacZdeltaM15]) that were infected with M13 bacteriophage or

transfected with M13 double stranded, covalently closed

circular DNA. A single plaque was removed from the agar

plate with a sterile pasteur pipette and placed in 2 ml YT

medium (0.8% Bacto-tryptone, 0.5% Bacto-yeast extract, 0.5%

NaC1). Two drops of an overnight JM103 bacterial culture

were added to the phage and incubated at 370C with agitation

for 6 hours. Longer incubation times result in deletions











within the phage DNA. The cell culture (1.5 ml) was

transferred to a microcentrifuge tube and spun in an

Eppendorf centrifuge at 12,000 x g for 10 minutes at 40C.

The supernatant (1.2 ml) was transferred to a sterile

microcentrifuge tube, 300 ul of 20% PEG, 2.5 M NaC1 were

added and the sample was incubated at room temperature for

15 minutes. The remaining 200-300 ul of supernatant were

removed and stored at -200C as a phage stock. The phage

suspension was centrifuged at 12,000 x g for 5 minutes at

room temperature and the supernatant was decanted. The

pellet was resuspended in 150 ul TES buffer (20 mM Tris-HCl

pH 7.5, 10 mM NaC1, 0.1 mM Na2EDTA). The sample was

extracted with phenol as follows: 1) added 50 ul phenol and

vortexed for 2 seconds; 2) incubated at room temperature for

5 minutes; 3) vortexed for 2 seconds; and centrifuged at

12,000 x g for 4 minutes at room temperature. DNA was

precipitated from 130 ul of aqueous phase with 5 ul 3 M

sodium acetate and 340 ul 95% ethanol on crushed dry ice for

15 minutes and collected by centrifugation at 12,000 x g,

for 15 minutes at 40C. The supernatant was discarded and the

pellet was washed in 70% ethanol at room temperature. The

sample was centrifuged at 12,000 x g, room temperature for 5

minutes. The DNA pellet was dried under vacuum, resuspended

in 80 ul TES buffer and stored at -20C.











Double-stranded templates. E. coli strain JM101

(delta(lac-proAB), supE / F' [proA+, DroB', lacIq,

lacZdeltaM15]) bacteria were grown overnight in 5 ml YT

medium at 370C with agitation. The overnight culture (1 ml)

was transferred to 100 ml prewarmed YT medium and incubated

at 37C until the O.D.590 reached 0.4 units. To assure

exponential growth the culture was again diluted 1:100 into

500 ml prewarmed YT medium and grown to an O.D.590 of 0.4

units. The culture was infected with 500 ul of M13

bacteriophage supernatant which had been prepared from an

isolated plaque as described above for the preparation of

M13 single-stranded DNA. The infected cells were incubated

at 370C with agitation for no more than 8 hours. The cells

were harvested and RF M13 DNA was isolated as described

above for the preparation of plasmid DNA. Routinely, the

yield was 300 Ag of supercoiled M13 DNA per 500 ml culture.



Isolation of Mammalian RNA

Isolation of Total Cellular RNA

The following procedure for the isolation of total

cellular RNA was developed by Dr Mark Plumb in the Stein's

laboratory primarily for work with HeLa cells. However, this

procedure has been successfully used without modification

for the isolation of total cellular RNA from human HL-60

cells, mouse 3T3-L1 cells, rat osteoblast primary cell

cultures, and Trypanosome cruzii cells.










Exponentially growing HeLa cells (5 x 107 cells) were

harvested by centrifugation at 1,500 rpm for 5 minutes at

370C in an IEC rotor. The cells were washed in phosphate-

buffered saline (PBS; 150 mM NaCl, 10 mM sodium phosphate pH

6.8) and resuspended in 3 ml lysis buffer (2 mM Tris-HCl pH

7.4, 1 mM EDTA) containing 5 gg/ml PVS. Cells were lysed in

the presence of 2.4% SDS and 88 pg/ml of proteinase K at

room temperature for 15 minutes. Lysates were adjusted to

300 mM NaCl, extracted with 5 ml Tris-saturated phenol and 5

ml CHCl3/isoamyl alcohol (IAA) (24:1, v/v) and centrifuged

at 3,000 rpm in an IEC rotor at room temperature. The

aqueous phase was then extracted with 5 ml CHCl3/IAA and

centrifuged at 3,000 rpm in an IEC rotor at room

temperature. The phenol/chloroform and chloroform

extractions were repeated as described above. Nucleic acids

were precipitated from the aqueous phase with 2.5 volumes of

95% ethanol at -20C in the presence of 53 mM potassium

acetate.

Nucleic acids were recovered from the ethanol

suspension by centrifugation at 12,000 x g and resuspended

in TCM buffer (10 mM Tris-Hcl, pH 7.4, CaCl2, 10 mM MgCl2)

for digestion with DNase I. The DNase I was dissolved in 20

mM Tris-HCl pH 7.4, 10 mM CaCl2 at 1 mg/ml, preincubated at

370C for 20 minutes and treated with 0.1 volume of

proteinase K [1 mg/ml] at 370C for 2 hours to remove

ribonuclease activity (Tullis and Rubin, 1980). The nucleic









42

acid samples were then digested with 0.1 mg/ml of proteinase

K treated DNase I for 20-30 minutes at 370C. Following DNase

I digestion, 0.05 volumes of 5 M NaCl and 0.25 volumes of

10% SDS were added to each sample and the RNA solutions were

repeatedly extracted with phenol and chloroform until the

interface between the aqueous and organic phases was clean.

RNA was precipitated from the aqueous phase with 2.5 volumes

of 95% ethanol, at -200C overnight. The RNA was collected at

12,000 rpm in a JA20 rotor at 40C for 30 minutes. The pellet

was resuspended in double distilled water, quantitated by

optical density at 260 nm and stored at -200C. Typical

yields of RNA were 400 gg per 90% confluent 10 cm plate of

HeLa S3 cells and 1 mg per 5 x 107 HeLa S3 cells with an

O.D.260/O.D.28g of 1.8 to 2.0.



Isolation of Nonmembrane-bound, Membrane-bound and Total
Polysomal RNA

The subcellular fractionation scheme used in these

studies is a modified procedure described by Venkatesan and

Steele (1972) and is outlined in Figure 2-2. Exponentially

growing HeLa cells (5 x 108 cells) were collected by

centrifugation at 1,500 rpm in an IEC rotor at 370C for 5

minutes. All subsequent procedures were carried out at 4C.

Cells were washed several times in PBS and resuspended in 25

ml RSB (10 mM Tris-HC1 pH 7.4, 10 mM NaC1, 2.5 mM MgCl2).

After incubation on ice for 20 minutes, the cells were











disrupted with 15 strokes in a tight fitting Dounce

homogenizer. The homogenate was then centrifuged at 2,000 x

g for 5 minutes. The supernatant supernatantt A) was removed

and maintained in an ice bath while the nuclear pellet

(pellet A) was resuspended in 9 ml TMN (20 mM Tris-HCl pH

7.5, 5 mM MgCl2, 25 mM NaC1, 3 mM dithiothreitol). The

nuclear suspension (pellet A) was adjusted to 1% Triton X-

100 and 1% NaDOC, vortexed briefly and centrifuged at 500 x

g for 10 minutes. The supernatant supernatantt C) from this

centrifugation step was removed and saved on ice for the

isolation of nucleus-associated rough endoplasmic reticulum

(NRER) polysomes. The supernatant from the centrifugation

step following homogenization supernatantt A) was

centrifuged at 12,000 rpm, 40C for 10 minutes in a Beckman

JA20 rotor. The resulting supernatant supernatantt B) was

removed onto ice for subsequent isolation of light

endoplasmic reticulum (LER)-associated polysomal RNA and

nonmembrane-bound polysomal RNA. The pellet (pellet B) was

resuspended in 9 ml TMN, adjusted to 1% Triton X-100 and 1%

NaDOC and centrifuged at 10,000 rpm, 40C for 10 minutes in a

JA20 rotor. The supernatant supernatantt D) was removed and

stored on ice for the isolation of mitochondria-associated

rough endoplasmic reticulum (MRER) polysomal RNA.

Exponentially growing HeLa cells (2.5 x 108 cells) were

collected by centrifugation for the isolation of total

polysomal RNA. The cell pellet was washed in PBS and



















Figure 2-2. Outline of procedure for the isolation of
subcellular polysomal fractions.
















HeLa Cells


homogenize
centrifuge


in RSB
2,000xg, 40C, 5'


supernatant A


centrifuge 11,400xg
10'



supernatant B pellet B

Triton/NaDOC
centrifuge 7,800x
10'
I 1


Triton/NaDOC
centrifuge 500xg
10'


discard pellet


discard pellet





centrifuge
200,000xg, 2 hrs
2M sucrose


Band at
interface


pellet


supernatant D super





centrifuge
200,000xg, 2 hrs
2M sucrose


pellet


natant C


centrifuge
200,000xg, 2 hrs
2M sucrose


pellet


LER Nonmembrane-bound


pellet A


NRER


MRER











resuspended in 15 ml TMN. The sample was adjusted to 1%

Triton X-100 and 1% NaDOC, disrupted with several strokes in

a tight fitting Dounce homogenizer and centrifuged at 12,000

x g at 40C for 10 minutes in a JA20 rotor. The supernatant

was removed and saved at 0C.

Each of the fractions (NRER, MRER, LER, nonmembrane-

bound, and total polysomal RNA) were layered onto a 3 ml 2 M

sucrose pad (prepared in RSB containing 100 uM spermidine)

and centrifuged at 40C for 2 hours at 48,000 x g in a

Beckman Type 50.2 Ti rotor (200,000 x g) without the brake.

The LER formed a cloudy band at the top of the sucrose pad

and was carefully removed by aspiration. The remaining

fractions were pelleted under these conditions and gently

resuspended in 3 ml TMN buffer. The samples were then

extracted extensively with phenol and chloroform as

described above for the isolation of total cellular RNA. The

RNA was precipitated from solution with 3 volumes of 95%

ethanol at -200C in the presence of 53 mM potassium acetate.

The RNA was digested with DNase I, extracted with phenol and

chloroform, and ethanol precipitated as described above for

the preparation of total cellular RNA. Typical yields were

1.5-2.0 mg MRER and NRER RNA, 2-3 mg nonmembrane-bound RNA,

1 mg LER RNA and 4-6 mg total polysomal RNA per liter of

exponentially growing HeLa cells.











Isolation of Cytoskeleton and Soluble Phase RNA

The cytoskeleton and soluble fractions were prepared as

previously described by Penman and coworkers (Cervera et

al., 1981). Exponentially growing HeLa cells (5 x 108 cells)

were harvested and washed in cold PBS. The cell pellet was

resuspended in 2 ml extraction buffer (10 mM Pipes pH 6.8,

100 mM KC1, 2.5 mM MgCl2, 0.3 M sucrose, 1 mM

phenylmethylsulfonyl fluoride) and Triton X-100 was

immediately added to a final concentration of 0.5%. The

sample was incubated at 0C for 3 minutes and then

centrifuged at 700 rpm in an IEC centrifuge for 3 minutes at

40C. The supernatant (SOL) was removed and stored on ice.

The pellet was resuspended in 8 ml RSB containing 1 mM

phenylmethylsulfonyl fluoride, 1% Tween 40 and 0.5% NaDOC.

The nuclei were stripped of cytoplasmic tags with 15 strokes

in a precision-bore stainless steel homogenizer with a

clearance of 0.002 inches and were pelleted at 2,000 rpm for

3 minutes at 40C in an IEC rotor. The supernatant (CSK) was

removed and saved at 0C for subsequent RNA isolation. Both

the CSK and SOL fractions were extracted in the presence of

1% SDS and 0.3 M NaCl with phenol and chloroform. Nucleic

acids were precipitated from the aqueous phase with 3

volumes of 95% ethanol at -200C in the presence of 53 mM

potassium acetate. The RNA was digested with DNase I, phenol

and chloroform extracted, and ethanol precipitated as

described above for the isolation of total cellular RNA.











Typical yields of RNA were 250-300 Mg from CSK and 100-150

Ag from SOL per 2.5 x 107 HeLa cells (50 ml exponentially

growing suspension culture).



Two-Dimensional Gel Electrophoresis and
Immunoblotting Analysis

The cytoskeleton and soluble fractions were isolated as

described above and each sample was dialyzed in 2 mM Tris-

HC1 pH 7.5 and treated with DNase I (20 gg/ml) and RNase A

(20 Ag/ml) in the presence of 0.5 mM MgC12 at 40C for 1

hour. The samples were dialyzed against 2 mM Tris-HCl pH 7.5

and then lyophilized. Two-dimensional gel electrophoresis

and immunoblotting analyses were performed by Dr. Warren

Schmidt (Department of Pathology, Vanderbilt University,

Nashville, TN). The protein preparations were focused in the

first dimension using equilibrium conditions as described by

O' Farrell (1975) with pH 3 to 10 or 5 to 7 ampholines. The

second dimension was electrophoresed on slab gels containing

a 3% stacking gel and 7.5% resolving gel as described by

Laemmli (1970). The proteins were then either stained with

Coomassie Brilliant Blue or electrotransferred to

nitrocellulose filters as described by Towbin et al. (1979).

Where indicated, immunotransfer analyses (Glass et al.,

1981) were performed with commercially available or other

(Schmidt et al., 1982) antibodies to actin, tubulin and

prekeratin.










Northern Blot Analysis

Aqarose-Formaldehyde RNA Denaturing Gel Electrophoresis

RNA samples were resolved electrophoretically in 1.5%

agarose-6% (w/v) formaldehyde horizontal slab gels and

running buffer (20 mM MOPS (morpholinepropane-sulfonic acid)

pH 7.0, 5 mM sodium acetate, 1 mM EDTA, 3.7% (w/v)

formaldehyde) as described by Rave et al. (1979). Routinely

10 gg RNA per sample were dried in a Savant Speed Vac

concentrator and resuspended in 3.2 ul double distilled

water (ddH0O), 5 ul formamide, 1 ul 10x MOPS (200 mM MOPS pH

7.0, 50 mM sodium acetate, 10 mM EDTA), 0.8 ul 37% (w/v)

formaldehyde and 2 ul sample buffer (0.5 ml formamide, 0.1

ml 10x MOPS, 80 ul 37% (w/v) formaldehyde and 0.32 ml 0.2%

bromophenol blue-90% glycerol). The samples were then heated

at 600C for 10 minutes, quick chilled to 0C and loaded onto

the gels. The 1.5% agarose-6% formaldehyde gels were

prepared by melting 3 g electrophoresis grade agarose in 144

ml ddHO2 plus 20 ml 10x MOPS buffer in a microwave oven

(Whirlpool) at full power for 3-4 minutes. The agarose

solution was cooled to approximately 650C (warm to the

touch) while stirring and then adjusted to 6% formaldehyde

with 36 ml 37% (w/v) formaldehyde. For minigels, 35 ml

agarose-formaldehyde solution was cast in a plexiglass gel

tray (10 cm x 6.3 cm) with an 8 well comb (1 mm x 5 mm per

well). The gel was electrophoresed at 40-45 milliamps until

the bromophenol blue migrated 9 cm (approximately 1.5









50

hours). For full size RNA gels, 200 ml agarose-formaldehyde

solution was cast in a plexiglass gel tray (24.5 cm x 19.8

cm) with 2: 20 well combs (1 mm x 6 mm per well). The gel

was electrophoresed at 100-120 volts until the bromophenol

blue migrated 11 cm (approximately 4 hours). The

formaldehyde gels were stained for 20 minutes in ethidium

bromide solution (1 Ag/ml ethidium bromide, 0.1 M ammonium

acetate) and destined for at least 2 hours in double

distilled water. RNA samples were then visualized by

fluorescence under ultraviolet light using a short

wavelength transilluminator (UVP, inc.; San Gabriel, CA).

The gels were usually photographed at f4.5 for 1/4 second

with Type 57 high speed Polaroid film and a Polaroid model

545 camera mounted on a stand above the transilluminator.



Transfer of RNA from Aqarose-Formaldehyde Gels to
Hybridization Filters

RNA samples were transferred to nitrocellulose (0.45

um) or nylon membrane hybridization filters in 20x SSC as

previously described by Thomas (1975). The gels were soaked

in 20x SSC (3 M NaCl, 0.3 M sodium citrate pH 7.0) for 20

minutes and were not stained with ethidium bromide. The

hybridization filters were uniformly wetted in ddH2O and

then 20x SSC. The gel was placed with the wells facing

downward onto a sheet of 3 MM paper (Whatman) and sponge

which had been soaked in 20x SSC. Strips of used X-ray films









51

were placed on areas of the sponge that were not covered by

the gel so as to direct the flow of 20x SSC only through the

gel. The filter was directly layered onto the gel and any

air bubbles that were trapped between the gel and filter

were carefully removed. The transfer set-up was completed by

placing two sheets of 3 MM paper followed by 2-3 boxes of

facial tissues (Kleenex brand) on top of the hybridization

filter. The buffer chamber was filled with 20x SSC and

transfer was carried out for 36 hours (note: complete

transfer is usually achieved within 14 hours). After

transfer, the filters were rinsed in 2x SSC for 10 minutes

to remove particles of agarose, air dried and baked in vacuo

at 800C for 2 hours. The filters were stored in sealed

hybridization bags at 40C.



Hybridization of Filters-Immobilized RNA (Northern Blot)

Filters (nylon or nitrocellulose) were prehybridized

for 3-4 hours in 0.5 ml hybridization buffer per cm2 filter

area at 430C for the detection of chorionic gonadotropin

alpha, HLA-B7 and c-fos mRNA and at 480C for histone mRNA.

Hybridization buffer consisted of 50% formamide, 5x SSC, 10x

Denhardt's (100x Denhardt's: 2% (w/v) ficoll 400, 2% (w/v)

polyvinylpyrrolidone), 1% SDS, 50 mM sodium phosphate pH

7.0, 20 gg/ml BSA, and 250 gg/ml E. coli DNA. The filters

were hybridized at the appropriate temperature for 36 hours

in 0.1 ml hybridization buffer per cm2 filter containing











1xl10 cpm/ml of each thermally denatured, 3P-radiolabeled

DNA probe. After hybridization, the filters were washed with

5x SSC, ix Denhardt's (0.5-1 ml/cm2 filter) for 10-15

minutes at room temperature to remove unbound probe. The

filters were then washed, in order, with the following

solutions (0.5 ml/cm2 filter area): 1) 5x SSC, Ix

Denhardt's; 2) 2x SSC, 0.1% SDS; 3) Ix SSC, 0.1% SDS; and 4)

0.1x SSC, 0.1% SDS at 600-650C for 30 minutes per wash.

After washing the filters were briefly air dried and then

exposed to preflashed Kodak XAR5 x-ray film at -70C with a

Cronex Lightning Plus intensifying screen (Dupont) for

varying lengths of time. Hybridization was quantitated by

scanning laser densitometric analysis of multiple exposures

of autoradiographs that were within the linear range of the

film.



S1 Nuclease Protection Analysis

Sl nuclease protection analysis was performed

essentially as described by Berk and Sharp (1977) and was

used in the work presented here to detect simultaneously

both endogenous H3 histone mRNA and signal peptide-histone

fusion mRNA in the same RNA sample. The Sl probe was

prepared by digesting 30 pg pSPH3E1, pSPH3Elalpha, or

pSPH3E1ATG- DNA with 30 units of SmaI at 300C for 3 hours.

The 450 bp SmaI fragment was isolated from a 0.8% agarose

minigel by the freeze-squeeze method. The solution









53

containing the SmaI fragment was butanol extracted to reduce

the volume and to remove the ethidium bromide. The fragment

was then extracted with phenol and chloroform and

precipitated in 2.5 volumes 95% ethanol at -200C overnight.

The DNA was collected by centrifugation at 12,000 x g at 40C

for 15"minutes and resuspended in 5 ul 10 mM Tris-HCl pH

8.0, 5 ul 10x CIP buffer (0.5 M Tris-HCl pH 8.0, 10 mM

MgCl2, 1 mM ZnCl2, 10 mM spermidine) and 38 ul ddH20. The

blunt-ended DNA was dephosphorylated with 0.12 units calf

intestinal alkaline phosphatase (CIP) at 370C for 15 minutes

and then at 550C for 15 minutes. A second aliquot of CIP was

added and the incubations at 370 and 550C were repeated. The

reaction was terminated with 40 ul ddH20, 10 ul 10x STE (100

mM Tris-HC1 pH 8.0, 1 M NaCI, 10 mM EDTA) and 5 ul 10% SDS

and heated at 680C for 15 minutes. The sample was extracted

with phenol and chloroform, adjusted to 0.25 M sodium

acetate and precipitated with 2.5 volumes of 95% ethanol at

-20C overnight. The dephosphorylated SmaI fragment was

pelleted at 12,000 x g, 40C for 15 minutes, vacuum dried and

resuspended in 12 ul Ix kinase buffer (66 mM Tris-HCl pH

9.5, 10 mM MgC12, 10 mM beta-mercaptoethanol, 2 mM

spermidine). The Sl fragment was phosphorylated with 50 uCi

[gamma-32 P]ATP and 1 ul polynucleotide kinase [30 units/ul]

at 370C for 30 minutes. The reaction was terminated with 0.5

ul 100 mM EDTA at 700C for 10 minutes. The DNA solution was

diluted with 1 ml elutip low salt buffer (20 mM Tris-HCl pH

7.4, 0.2 M NaC1, 1 mM EDTA) and passed over a Schleicher and









54

Schuell Elutip column. The column was washed with 5 ml low

salt buffer to remove unincorporated radiolabeled

ribonucleotide and the DNA fragment was eluted in 400 ul

high salt buffer (Tris-HC1 pH 7.4, 1 M NaCl, 1 mM EDTA). The

radiolabeled Sl probe was usually used immediately and any

excess was stored in high salt buffer at 4C.

Routinely, 10-25 gg RNA (30 ul) and 10 ul S1 probe were

co-precipitated with 100 ul 95% ethanol at -200C overnight.

The sample was centrifuged at 12,000 x g, 40C for 15-30

minutes. The pellet was vacuum dried and resuspended in 10

ul 5x hybridization buffer (0.2 M Pipes pH 6.4, 2 M NaC1, 5

mM EDTA) and 40 ul recrystallized formamide. The sample was

heated at 900C for 10 minutes and then incubated at 550C for

3 hours. Single-stranded nucleic acids were digested with Sl

nuclease by adding 400 ul ice cold Sl nuclease buffer (0.03

M sodium acetate pH 4.6, 0.25 M NaC1, 1 mM ZnSO4) and 900

units S1 nuclease at 370C for 30 minutes. The Sl digestion

reaction was terminated by organic extraction with 0.5

volume phenol and 0.5 volume chloroform and then 0.5 volume

chloroform alone. Nucleic acids were precipitated from the

aqueous phase with 1 ml 95% ethanol at -200C overnight.

The S1 nuclease digested samples were collected at

12,000 x g, 40C for 30 minutes. The pellets were vacuum

dried and resuspended in 5 ul loading buffer (80% formamide,

lx TBE [10x TBE: 500 mM Tris-HCl pH 8.3, 500 mM boric acid,

10 mM EDTA], 0.1% bromophenol blue, 0.1% xylene cyanol). The









55

samples were denatured by heating at 1000C for 2 minutes and

then quickly chilled at 0C to prevent reannealing prior to

loading onto the gel. Six percent polyacrylamide-8.3 M urea

(cross linked at 20:1 acrylamide:bis-acrylamide) gels with

dimensions of either 16 cm x 18 cm x 1.5 mm (1.5 mm x 6 mm

15 well comb or 1.5 mm x 4 mm 20 well comb) or 32.5 cm x 44

cm (0.5 mm x 8 mm 25 well comb) were prepared and pre-

electrophoresed for 30 minutes. The samples were loaded with

drawn-out capillary tubes or a Rainin Pipetteman (0-20 ul

capacity) using flat pipette tips. The large gels were

electrophoresed at 50 watts constant power (-1200 V, -45 A)

for 3-4 hours and the small gels were electrophoresed at 300

V (25-35 A) for 2-3 hours. The gels were placed on 2 sheets

of 3 MM Whatman paper, dried under vacuum at 800C for 1-2

hours and exposed to pre-flashed XAR5 x-ray film with a

Cronex Lightning Plus intensifying screen at -700C for

varying lengths of time. Sl analysis was quantitated by

scanning laser densitometry of multiple autoradiographic

exposures that were within the linear range of the film.



In Vitro Nuclear Run-On Transcription Analysis

The in vitro nuclear run-on transcription assays were

performed by Anna Ramsey-Ewing as described by Baumbach et

al. (1987). Cells were harvested by centrifugation and the

cell pellet washed twice in cold isotonic buffer (125 mM

KC1, 30 mM Tris-HCl pH 7.9, 5 mM MgCl2, 10 mM beta-









56
mercaptoethanol). Cells were disrupted by homogenization (6-

12 strokes) with a Dounce homogenizer, Wheaton type A

pestle. After >90% of the cells had been lysed, nuclei were

pelleted by centrifugation at 2000 rpm for 10 minutes in an

IEC centrifuge at 40C and resuspended in nuclei storage

buffer containing 40% glycerol (50 mM Tris-HCl pH 8.3, 5 mM

MgCl2, 0.1 mM EDTA). Nuclei were aliquoted and either snap

frozen in liquid nitrogen or used fresh in the in vitro

transcription reactions. Reactions typically contained 107

nuclei, 100 uCi alpha-32P-UTP (3000 Ci/mmole), 1 mM ATP,

0.25 mM GTP and CTP in a final volume of 130 ul and were

incubated for 30 minutes with intermittent shaking at 300C.

Radiolabeled RNAs were isolated by treatment of nuclei with

DNase I (100 Ag/ml) in the presence of 0.6 M NaC1, 50 mM

Tris-HCl pH 7.5, 20 mM MgCl2 for 15 minutes at room

temperature. The mixture was then incubated with proteinase

K (200 Ag/ml) for 30-60 minutes at 370C in the presence of

150 mM NaC1, 12.5 mM EDTA, 100 mM Tris-HCl pH 7.5 and 20 mM

MgC12. Sodium acetate (pH 5.5) was added to 0.2 M and

nucleic acids extracted several times by the hot phenol

method (Clayton and Darnell, Jr., 1983; Sherrer and Darnell,

1962). To the aqueous solution of 32P-labeled RNAs, 150 Ag

of yeast RNA and 2.5 volumes of 95% ethanol were added.

Precipitation was overnight at -200C. Radiolabeled

transcripts were resuspended in 10 mM Tris-HCl pH 8.0, 1 mM

EDTA and an aliquot of each sample was precipitated with 150









57

Mg yeast RNA and cold 10% TCA. TCA-precipitable counts were

determined by liquid scintillation spectrometry.

DNA excess hybridization has also been described by

Baumbach et al. (1987). In a preliminary experiment we

determined that 2 pg of H4 histone insert in pFO002 was at

least a two-fold DNA excess when hybridized to in vitro

nuclear run-on transcripts. Southern blots containing

restriction endonuclease digested DNA fragments or slot

blots containing linearized plasmid DNAs were prepared.

Southern blots on nitrocellulose or slot blots on Nylon were

prehybridized in 1 M NaCl, 20 mM Tris-HCl pH 7.4, 2 mM EDTA,

0.1% SDS, 5x Denhardt's, 250 Ag/ml E. coli DNA and 12.5 mM

sodium pyrophosphate at 650C for at least 6 hours.

Hybridizations were conducted at 650C for 72 hours in 1 M

NaCl, 20 mM Tris-HCl pH 7.4, 2 mM EDTA, 0.1% SDS, 2.5x

Denhardt's, 250 Ag/ml E. coli DNA with 32P-labeled

transcripts at 5x105-1X106 TCA-precipitable counts per ml of

hybridization solution. Blots were washed at 650C for 15

minutes in fresh prehybridization solution [0.13-0.14

ml/cm2], 1 hour in 2x SSC/0.1% SDS [0.5 ml/cm2], overnight

in 2x SSC/0.1% SDS [0.5 ml/cm2] and 1 hour in 0.2x SSC/0.1%

SDS [0.5 ml/cm2]. Air dried filters were autoradiographed

with XAR-5 or Cronex film and Cronex lightning Plus screens

at -70C for varying periods of time and developed with an

X-O-MAT X-ray film processor.











Radiolabeling DNA for Northern Blot Analysis

DNA used for Northern blot analysis was uniformally

labeled by nick translation as described by Maniatis et al.

(1982b) or by the random priming method of Roberts and

Wilson (1985). Radiolabeled probes prepared in this manner

are also suitable for other purposes such as Southern blot

analysis.



Random Priming Method

This protocol is very efficient for labeling small to

intermediate-sized DNA fragments (100-2000 bp in length) and

is the preferred method for work with purified DNA inserts.

Supercoiled plasmid DNA is labeled much less efficiently by

the random priming method and should be fragmented by

sonication prior to use. Routinely, specific activities of

1x109 cpm/gg DNA are achieved using as little as 20-100 ng

DNA. Material for oligolabeling DNA fragments was purchased

from Bethesda Research Laboratories as a kit.

The isotope (50 uCi [a-32P] dCTP) was vacuum dried and

resuspended in 10 ul of 2.5x reaction buffer (500 mM Hepes

pH 6.6, 12.5 mM MgCl2, 25 mM beta-mercaptoethanol, 125 mM

Tris-HCl pH 8, 1 mg/ml BSA, 7.5 mg/ml synthetic DNA

oligodeoxynucleotide primers, 50 uM each dGTP, dATP, dTTP).

The DNA fragment (100 ng) in 14 ul ddH2O was heated at 1000C

for 2 minutes and immediately chilled on ice. The denatured

DNA was added to the 1.5 ml microcentrifuge tube containing











the isotope and buffer and the priming reaction was

initiated with 5 units Klenow enzyme [5 U/ul]. The sample

was incubated at room temperature for at least 2 hours

(usually overnight). The reaction was terminated with 1 ul

250 mM EDTA and 175 ul TE buffer (10 mM Tris-HCl pH 8, 1 mM

EDTA) and extracted once with 200 ul chloroform/isoamyl

alcohol. Unincorporated radiolabeled deoxynucleotide was

removed from the DNA by chromatography over a (10 mm x 100

mm) Biogel A-15m (Bio Rad) column in TE buffer. Fourteen

fractions were collected; 800 ul in the first fraction and

200 ul in each subsequent fraction. Fractions containing

radiolabeled DNA were determined by Cerenkov counting in the

tritium channel with a Beckman LS-230 scintillation counter.

The peak fractions were pooled and accurately quantitated by

counting 2 ul DNA in 5 ml Triton-toluene based scintillation

cocktail (2000 ml toluene, 1066 ml Triton X-100, 134 ml

liquifluor) in the 32P channel. The samples were immediately

used for hybridization studies and excess probe was stored

at 4C.



Nick-Translation Method

Nick-translation was the common procedure for

uniformally labeling DNA fragments or supercoiled DNA before

the development of the random primer method. A disadvantage

to using the nick-translation procedure is that the system

must be optimized for efficiency for each lot of E. coli DNA









60
polymerase I enzyme. In addition, the specific activity (1-2

x 108 cpm/Ag DNA) achieved with the nick-translation

procedure is much lower than the random primer method (1 x

109 cpm/Ag DNA).

DNase I [1 mg/ml in 10 mM HCl] was activated by

incubating 1 ul of the enzyme in 9 ul DNase buffer (10 mM

Tris-HCl pH 7.5, 5 mM MgCl2, 1 mg/ml BSA) at 0C for 2

hours. The isotope (40-80 uCi [a-32p] dCTP) was dried in a

Savant Speed Vac centrifuge. The following components were

added, in order, to the isotope : 1) 12.25 ul ddHO2; 2) 2.5

ul 10x nick-translation buffer (500 mM Tris-HC1 pH 7.8, 100

mM beta-mercaptoethanol, 50 mM MgCl2); 3) 5 ul 10x dNTPs

(100 uM of each dATP, dGTP, dTTP); 4) 1.25 ul BSA [1 mg/ml];

5) 2.5 ul DNA [100 ng/ul]; and 6) 0.5 ul DNA polymerase I [5

U/ul]. The activated DNase I was diluted with 990 ul ice

cold ddH2O (1:1,000 dilution; 1 ng/ul final concentration)

and 1 ul of diluted enzyme was added to the nick-translation

mix. The sample was incubated at 140C for 45 minutes and the

reaction was then terminated with 20 ul 250 mM EDTA and 155

ul TE buffer. The sample was extracted once with 200 ul

chloroform and unincorporated radiolabeled

deoxyribonucleotides were removed by chromatography over a

Biogel A-15m column. The DNA was quantitated and immediately

used for hybridization.











Site-Directed Mutagenesis

In the past, researchers had to rely on nature to

produce the proper mutations to study how the normal

biological processes function. Zoller and Smith (1983) have

designed a powerful method for selectively introducing

single nucleotide mutations into specific sites of any

cloned, sequenced region of DNA (site-directed mutagenesis).

In addition, this procedure can be used to selectively

create deletion or insertion mutations in DNA fragments of

known sequence. The general approach of site-directed

mutagenesis is to incorporate an oligonucleotide containing

the appropriate mutation into the complementary strand of

recombinant M13 DNA. The heteroduplex is transfected into

bacteria and the resulting mutated bacteriophage are

isolated by various selection processes.

As discussed in Chapter 4, the signal peptide-histone

fusion mRNA is associated with membrane-bound polysomes and

is stable during inhibition of DNA synthesis. It was unclear

whether the uncoupling of the histone fusion mRNA stability

from DNA replication was due to the altered subcellular

location or to a change in mRNA structure, due to the

nucleotide sequence coding for the signal peptide. To

address these possibilities, we have introduced point

mutations into the signal peptide coding region with the

intention to inactivate signal peptide function. In doing

so, mRNA structure is perturbed as little as possible and











the mutated histone fusion mRNA, like endogenous histone

mRNA, associates with nonmembrane-bound polysomes.


Subcloning of the Signal Peptide-Histone Fusion Gene
into M13 DNA

The signal peptide-histone fusion gene (SPH3) was

subcloned into M13mpl8 RF DNA. SPH3 DNA was digested with

EcoRI and HindIII and the 1100 bp fragment was isolated from

a 0.8% agarose gel. The M13mpl8 DNA was digested with EcoRI

and HindIII and the 7196 bp fragment was isolated from a

0.8% agarose gel. The 1100 bp SPH3 DNA fragment (-50 ng) and

7196 bp M13 DNA fragment (-50 ng) were ligated in 10 ul Ix

ligation buffer (50 mM Tris pH 7.6, 10 mM MgCl2, 20 mM

dithiothreitol, 1 mM ATP and 50 .g/ml BSA) with 0.004 U T4

DNA ligase (NEB) at 220C overnight. The ligation mix was

transfected into competent JM101 bacteria, white plaques

were isolated, and M13/SPH3 recombinants were screened by

restriction endonuclease digestion analysis. Single-stranded

M13/SPH3 DNA was isolated from bacteriophage particles as

described above.



Synthesis and Processing of Oligonucleotides

The signal peptide encoded by the SPH3 histone fusion

gene was mutated by substituting leucine at amino acid

position 10 and proline at position 12 with histidine

residues. It was predicted that the introduction of two

positively charged amino acids into the hydrophobic region











of the signal peptide would disrupt its recognition by the

signal peptide recognition particle and would therefore,

prevent the translocation of the mutated histone fusion mRNA

to membrane-bound polysomes.

An oligonucleotide (SPa) containing two point mutations

was synthesized by Tom Doyle in the Department of

Microbiology and Immunology, University of Florida,

Gainesville, Florida. The oligonucleotide was composed of

the following sequence: 5' CAAAAAAGTGAATATGGGCGA 3', with

the mutated nucleotides underlined. The oligonucleotide was

received in -2.5 ml 37% ammonium hydroxide and was heated at

550C for 6 hours to remove the protecting amine group. The

sample was dried under a gentle stream of air in the fume

hood and resuspended in 1 ml ddH20. The sample was extracted

two times with 600 ul n-butanol and the oligonucleotide was

precipitated from the aqueous phase by adding 200 ul 0.5 M

ammonium acetate and 2 ml 95% ethanol and chilling at -700C

for 30 minutes. The sample was centrifuged at 12,000 x g at

40C for 30 minutes, washed in 80% ethanol and resuspended in

200 ul ddH20. The yield of oligonucleotide was estimated

spectrophotometrically and was found to be 2.64 mg

(A260/A80 of 2.0).

A second oligonucleotide (ATG-) was designed to destroy

the signal peptide ATG translation start codon of the SPH3

fusion gene and thereby completely block the synthesis of

any signal peptide. The SPH3ATG- mRNA retains the histone











translation start codon in its proper context, although

further downstream from the 5' terminus than endogenous H3

histone mRNA, and should therefore be translated on

nonmembrane-bound polysomes. In addition, SPH3ATG- mRNA

differs from SPH3 wild type signal peptide-histone fusion

mRNA at only one nucleotide. The sequence of the ATG-

oligonucleotide is as follows: 5' AATACTCAAACTCTTCC 3', and

was prepared and processed as described for the SPa

oligonucleotide. The yield of ATG- oligonucleotide was only

445 Ag (A260/A280 of 1.41).



Optimization of Conditions for Site-Directed Mutagenesis

It was necessary to optimize the conditions for the

hybridization of the oligonucleotide to M13 template so that

the synthesis of the second strand was primed by the

oligonucleotide from the correct site. Proper priming by the

oligonucleotide was tested by primer extension analysis.

After limited primer extension, the samples were digested

with a restriction endonuclease that recognized a site

downstream from the expected priming site. The number and

intensity of the fragments produced by the digestion

indicated the number of priming sites and the location of

the preferred site.

The oligonucleotide (20 pmol) was phosphorylated with 5

units T4 polynucleotide kinase in 100 mM Tris-HCl pH 8.0, 10

mM MgC12, 5 mM dithiothreitol, 20 uCi [gamma-32P] ATP at 370C











for 45 minutes. The reaction was terminated by heating at

650C for 10 minutes, and unincorporated radiolabeled

nucleotides were removed by chromatography through a

Sephadex G-25 (Pharmacia) column (10 mm x 100 mm) in 50 mM

ammonium bicarbonate (pH 7.8).

The phosphorylated oligonucleotide (10-30 molar excess

over template DNA) was added to 0.5 pmol of M13 recombinant

DNA and the volume was adjusted to 10 ul with ddHO2 and 1 ul

solution A (200 mM Tris-HCl pH 7.5, 100 mM MgCl2, 500 mM

NaCl, 10 mM dithiothreitol). The solution was heated at 550C

for 5 minutes and then cooled for 5 minutes at room

temperature. The primer extension reaction also contained 1

ul of 2.5 mM dCTP, 1 ul 2.5 mM dATP, 1 ul 2.5 mM dTTP, 1 ul

2.5 mM dGTP, 0.5 ul solution A and 1 unit Klenow enzyme. The

reaction proceeded at room temperature for 5 minutes and was

terminated by heating at 650C for 10 minutes. The sample was

diluted with 2 ul 10x PstI buffer (100 mM Tris-HCl pH 7.5, 1

M NaCI, 100 mM MgCl2, 1 mg/ml BSA) and 7 ul ddHO2 and then

digested with 1 ul PstI restriction endonuclease [5 U/ul] at

370C for 1 hour. The sample was heated at 1000C for 2

minutes, quickly cooled to 0C and diluted with 20 ul S1

loading buffer. The PstI digested samples (10 ul) were

resolved electrophoretically in 5% acrylamide (20:1;

acrylamide:bis), 7 M urea at 300 volts (-35 mA) for 1.5

hours. The gels were dried under vacuum at 800C for 1 hour

and analyzed by autoradiography. Primer extension analysis









66

revealed a single major band at -80 bp for each molar ratio

of oligonucleotide to template DNA studied. This was the

expected size fragment and suggested that under these

conditions, the oligonucleotide was priming the synthesis of

the second strand from the proper site. If several

predominant bands had been detected (i.e. multiple priming

sites), more stringent annealing conditions would have been

required.



Synthesis of the Mutated Strand

The oligonucleotide (1.3 Mg) was phosphorylated in 100

mM Tris-HCl pH 8, 10 mM MgCl2, 5 mM dithiothreitol, 10 uM

ATP with 15 units T4 polynucleotide kinase at 370C for 45

minutes. The reaction was terminated by heating at 650C for

10 minutes.

The phosphorylated oligonucleotide (13.3 pmole) was

annealed to M13/SPH3 recombinant DNA in Ix solution A by

incubating the sample at 550C for 5 minutes and then at room

temperature for 5 minutes. After hybridization, the DNA was

diluted with 10 ul solution C (1 ul solution B [0.2 M Tris-

HC1 pH 7.5, 0.1 M MgCl2, 0.1 M dithiothreitol], 1 ul 10 mM

dCTP, 1 ul 10 mM dTTP, 1 ul 10 mM dGTP, 0.5 ul 0.1 mM dATP,

1 ul 10 mM ATP, 1.5 ul [gamma-32P] dATP, 1.5 ul T4 DNA ligase

[1 U/ul], 2 ul ddH20) and the synthesis of the second strand

was initiated with 0.5 ul Klenow enzyme [100 U/16.7 ul] at

room temperature. After 5 minutes, 1 ul 10 mM dATP was added











and the incubation was continued at room temperature

overnight.



Enrichment for Covalently Closed Double-Stranded DNA

The in vitro synthesis of the entire second strand is

very inefficient due primarily to the large size of the M13

genome. The percent of covalently closed, double-stranded

DNA molecules is therefore, very small in comparison to the

partially double-stranded, incomplete DNA molecules. These

incomplete DNA molecules, when transfected into bacteria,

give rise to wild type plaque formation which complicates

the selection process for the mutated clones. Enrichment for

covalently closed, double-stranded DNA molecules by alkaline

sucrose gradient centrifugation greatly facilitates the

selection process.

The DNA was precipitated from the overnight extension

and ligation reaction by adding 30 ul ddH20, 50 ul 1.6 M

NaCl-13% PEG at 0C for 15 minutes. The DNA was pelleted at

12,000 x g at room temperature for 15 minutes. The pellet

was gently washed in 100 ul cold 0.8 M NaCl-6.5% PEG,

centrifuged at 12,000 x g for 30 seconds and resuspended in

180 ul TE buffer. The sample was denatured with 20 ul 2 N

sodium hydroxide and layered onto a 5%-20% alkaline sucrose

step gradient. The gradient was prepared by adding to a 0.5

in x 2 in centrifuge tube nitrocellulosee or polyallomer; do

not use Beckman Ultra-clear tubes which are incompatible











with the alkali conditions), in order, the following

solutions each containing 0.2 N NaOH, 1 M NaCl, 2 mM EDTA:

1) 1 ml 20% sucrose; 2) 1 ml 17.5% sucrose; 3) 1 ml 15%

sucrose; 4) 1 ml 10% sucrose; 5) 5% sucrose. The tube was

incubated at 40C for 2 hours. The samples were centrifuged

through the alkaline sucrose gradients in a Beckman SW 50.1

rotor at 37,000 rpm for 2 hours at 40C without the brake.

The samples were recovered from the gradients by puncturing

the bottom of the tube with a 21 gauge needle and collecting

7 drops per fraction in 1.5 ml microcentrifuge tubes

(usually -60 fractions). Fractions containing covalently

closed, double-stranded DNA were determined by Cherenkov

scintillation counting in the tritium channel, pooled and

neutralized with 1M Tris-citrate pH 5.



Isolation of Mutated M13 Recombinant Bacteriophage

The DNA was transfected into E. coli JM101 competent

cells, plated onto YT agar and incubated overnight at 370C.

Individual plaques were picked and bacteriophage were

subsequently prepared. Bacteriophage suspensions (2 ul) were

spotted onto a 0.45 micron nitrocellulose filter and

prehybridized in 6x SSC, 10x Denhardt's, 0.2% SDS (10 ml/100

cm2) at 670C for 1 hour. The filter was rinsed in 50 ml 6x

SSC for 1 minute at 210C and hybridized in 10 ml 6x SSC, 10x

Denhardt's containing 1x106 cpm radiolabeled oligonucleotide

at 210C for 1 hour. The filter was washed three times in 50

ml 6x SSC at 210C for a total of 10 minutes. The filter was











briefly dried under a heat lamp and placed against pre-

flashed XAR5 x-ray film without a screen for 1 hour at room

temperature. Under such low stringency each phage spot was

positive, including the negative control M13/SPH3 wild type

bacteriophage. The filter was re-washed at 350C in 6x SSC

for 5 minutes, air dried and exposed to pre-flashed XAR5

film for 1 hour. At the higher wash temperature many of the

phage spots were less intense whereas several remained

unchanged. When the filter was washed at 480C in 6x SSC for

5 minutes the negative control and nearly all the phage

spots completely disappeared. The bacteriophage (-5%) that

remained positive after the 480C wash were sequenced to

confirm the presence of the mutations.



DNA Sequencing

The site directed mutations in M13/SPH3 recombinant

DNAs were confirmed by Sanger's dideoxynucleotide chain

terminator DNA sequencing procedure using the "SequenaseTM

DNA Sequencing Kit" from United States Biochemical

Corporation, Cleveland, Ohio. The Sequenase enzyme is

derived from bacteriophage T7 DNA polymerase and has been

modified to efficiently use dideoxyribonucleotides, alpha-

thio deoxyribonucleotides and other nucleotide analogs that

are commonly used for sequencing. In addition, the Sequenase

enzyme expresses high processivity and low 3'-5' exonuclease

activity.










Annealing Template and Primer

The primer for synthesizing the complementary strand of

wild type and mutated M13/SPH3 recombinant DNA is a 20 bp

oligonucleotide which hybridizes to a site adjacent to the

polylinker. The annealing reaction was prepared by adding 1

ul priier (0.5 pmol), 7 ul single-stranded M13/SPH3

recombinant DNA (0.5-1.0 pmol) and 2 ul 5X sequencing buffer

(200 mM Tris-HCl pH 7.5, 100 mM MgCl2, 250 mM NaC1). The

sample was heated at 650C for 2 minutes and slowly cooled to

room temperature.



Labeling Reaction

The complementary strand was synthesized by incubating

the annealed sample with 1 ul 100 mM dithiothreitol, 2 ul

labeling nucleotide mix (1.5 uM dGTP, 1.5 uM dCTP, 1.5 uM

dTTP), 0.5 ul [a- S] dATP [10 uCi/ul] and 2 ul Sequenase

[1.5 U/ul] at room temperature for 5 minutes. Elongation was

terminated by mixing 3.5 ul aliquots of the labeling

reaction with 2.5 ul of one of the dideoxy nucleotide

solution (ddATP, ddCTP, ddGTP, ddTTP). The dideoxy

nucleotide solutions were 80 uM dATP, 80 uM dCTP, 80 uM

dGTP, 80 uM dTTP, 50 mM NaCl and 8 uM of the appropriate

dideoxyribonucleotide. The sequencing samples were diluted

with 4 ul stop solution (95% formamide, 20 mM EDTA, 0.05%

bromophenol blue and 0.05% xylene cyanol), heated at 750C

for 2 minutes and loaded onto a 6% acrylamide-8.3 M urea gel









71

(32.5 cm x 44 cm) with a 60 well shark toothed comb. The gel

was run at 75 W constant, ~2500 V, -40 mA for 3-5 hours

(Xylene Cyanol dye was 15 cm from the bottom of gel). The

gel was soaked in 10% acetic acid-12% methanol in the fume

hood for 1 hour, placed onto 2 sheets of 3 MM Whatman paper,

and dried under vacuum at 800C for 45 minutes. The gel was

exposed to preflashed XAR5 x-ray film at room temperature

for 12-18 hours.



Transfection of DNA into Cells

Introduction of Plasmid DNA into Bacterial Cells

An E. coli overnight cell culture was diluted 1:100

into 100 ml LB medium and incubated at 370C until the

optical density was approximately 0.37 at 590 nm. The cells

were chilled at 0C for 10 minutes and then centrifuged at

5,000 rpm, 40C for 5 minutes in a Beckman JA20 rotor. The

pellet was resuspended in 20 ml of cold CaC12 buffer (60 mM

CaCI2, 10 mM Pipes pH 7.0, 15% glycerol) and incubated at

0C for 30 minutes. The cells were collected by

centrifugation at 2,500 rpm for 5 minutes at 40C in a JA20

rotor, resuspended in 2.6 ml CaC12 buffer and divided into

100-200 ul aliquots. The competent cells were either used

immediately for transformation or stored in microcentrifuge

tubes at -80C.

For transformation, 100 ul competent E. coli cells were

mixed with 10 ul plasmid DNA (-10 ng) and incubated at OC

for 10 minutes. The cells were then heat shocked at 370C for











5 minutes, diluted with 0.9 ml prewarmed LB medium and

incubated at 370C for 1 hour. Different volumes of the

transfected bacterial cell culture (usually 10-200 ul) were

plated onto the appropriate medium dictated by the genotype

of the host and the introduced plasmid DNA.



Introduction of DNA into HeLa Cells

HeLa monolayer cell cultures were transfected with

plasmid DNA, according to Gorman et al. (1982), in a calcium

phosphate/DNA complex prepared as described by Graham and

van der Eb (1973). HeLa S3 cells growing exponentially in

suspension culture were plated into 20 ml completed EMEM

(EMEM supplemented with 5% fetal calf serum, 5% horse serum,

100 U penicillin, 100 gg/ml streptomycin and 1 mM glutamine)

at 3x106 cells per 10 cm tissue culture dish (Corning or

Falcon) and incubated overnight at 370C in 5% CO2. The

monolayer cell culture was refed with 10 ml completed EMEM

and incubated at 370C in 5% CO2 for 4 hours. The DNA/calcium

phosphate precipitate was prepared by adding 20 gg DNA in

500 ul 250 mM CaCl2 dropwise to 500 ul 2x Hebs buffer (50 mM

Hepes, 280 mM NaC1, 1.5 mM Na2HPO4, pH 7.120.05) while

vortexing the solution. The DNA/CaC12 precipitate was added

dropwise to the monolayer cell culture and incubation was

continued at 37C in 5% CO2 for 4 hours. The HeLa cells were

glycerol shocked by replacing the medium with 2 ml completed

EMEM containing 15% glycerol for 1 minute. The transfected










cells were then washed with 10 ml EMEM, refed with 20 ml

completed EMEM and incubated at 370C, 5% CO2.



Establishment of Polyclonal and Monoclonal Stably
Transformed Cell Lines

Polyclonal and monoclonal cell lines were isolated by

growing HeLa cell cultures that were co-transfected with

pSV2neo and the plasmid DNA of interest in medium containing

the antibiotic G418. The plasmid pSV2-neo, constructed by

Southern and Berg (1982), carries the bacterial

aminoglycoside phosphotransferase 3' (II) gene under the

control of the SV40 early promoter and confers resistance to

the aminoglycoside antibiotic G418. The rationale behind the

co-transfection method is that HeLa cells that are competent

to take up pSV2-neo DNA are equally competent to take up the

plasmid DNA of interest, and therefore, a certain percentage

of the G418 resistant cells contain both species of DNA.

Ideally, the selectable marker should be carried by the

vector that contains the gene which is to be studied.

HeLa cells in monolayer culture were co-transfected

with the plasmid DNA to be studied and pSV2-neo DNA (molar

ratio of 20:1, respectively) by the calcium phosphate

precipitation method as described above. After 36-48 hours

post-glycerol shock, the transfected cells were resuspended

in 5 ml Puck saline A [0.04% (w/v) KC1, 0.8% (w/v) NaC1,

0.035% (w/v) NaHCO3, 0.1% (w/v) glucose] containing 0.02%









74

(w/v) EDTA and 0.5 ml of the cell suspension was seeded into

20 ml completed EMEM containing 500 gg/ml biologically

active Geneticin (G418, Gibco) in 10 cm tissue culture

dishes. The cells were incubated at 370C, 5% CO, and refed

every 3-4 days with completed EMEM containing G418

antibiotic. Within 2-3 weeks, the majority of the cells had

died and individual colonies of G418 resistant HeLa cells

had formed. Polyclonal cell lines were established by

combining the G418 resistant colonies from a 10 cm dish as a

single, heterogeneous cell population (usually 20-40

colonies). For the establishment of monoclonal cell lines,

individual colonies were picked from the dish with sterile

pasteur pipettes and cultured as separate cell lines.















CHAPTER 3

SUBCELLULAR LOCATION OF HISTONE mRNAs ON CYTOSKELETON
ASSOCIATED NONMEMBRANE-BOUND POLYSOMES IN HELA S3 CELLS


INTRODUCTION

Histone biosynthesis occurs primarily during the S

phase of the cell cycle and is temporally as well as

functionally coupled with DNA replication (Allfrey et al.,

1963; Prescott, 1966; Spaulding et al., 1966; Robbins and

Borun, 1967; Stein and Borun, 1972). Early studies using in

vitro translation analysis of polysomal RNA isolated from S

phase HeLa cells provided an initial indication that histone

protein synthesis occurs on non-membrane-bound, light

polysomes (Jacobs-Lorena et al., 1972; Liautard and

Jeanteur, 1979; Gallwitz and Breindl, 1972; Borun et al.,

1967). In these studies it was demonstrated that

translatable, polysome-associated histone mRNAs have a

sedimentation coefficient of 8-10S and fluctuate in a cell

cycle-dependent manner. Histone mRNA was not detected by in

vitro translation studies on polyribosomes isolated from

HeLa cells in the G1 phase of the cell cycle (Borun et al.,

1967; Borun et al., 1975; Pederson and Robbins, 1970).

Consistent with this observation, the inhibition of DNA

replication by hydroxyurea or cytosine arabinoside treatment











resulted in the rapid disappearance of histone mRNAs from

these polysomes (Borun et al., 1967; Breindl and Gallwitz,

1974; Gallwitz, 1975).

The light polysomes on which translatable histone mRNAs

reside are presumably nonmembrane-bound ("free") polysomes

as predicted by the signal hypothesis (Blobel and

Dobberstein, 1975). Several lines of evidence suggest that

nonmembrane-bound polysomes are not freely diffusible within

the cytoplasm but are attached to the cytoskeleton. Penman

and co-workers (Lenk et al., 1977) have demonstrated that

nearly 100% of total cytoplasmic polysomes co-fractionate

with the cytoskeletal framework. In addition, vesicular

stomatitis virus encoded nonmembrane-bound and membrane-

bound polysomal mRNAs are associated with the cytoskeleton

in virally infected HeLa cells (Cervera et al., 1983).

Taken together, these observations provide a basis for

the possibility that histone mRNA-containing polysomes

and/or their subcellular location may play a role in the

regulation of histone gene expression. Such reasoning is

consistent with the appearance of translatable histone mRNAs

on polysomes within minutes following transcription (Borun

et al., 1967), the equally rapid transfer of newly

synthesized histones into the nucleus where they complex

with DNA (Robbins and Borun, 1967; Spaulding et al., 1966),

and the degradation of histone mRNAs and cessation of

histone protein synthesis in parallel with the inhibition of











DNA replication (Spaulding et al., 1966; Robbins and Borun,

1967; Borun et al., 1967; Baumbach et al., 1984). To address

the relationship between the localization of histone mRNAs

and the control of cellular histone mRNA levels, we have

used cloned human histone genes to examine the distribution

of histone mRNAs on nonmembrane-bound and membrane-bound

polysomes and the association of histone mRNA-containing

polysomes with the cytoskeleton. As predicted from previous

results of in vitro translation analysis and cDNA

hybridization studies, histone mRNAs were found to reside

predominantly on nonmembrane-bound polysomes. Both the

nonmembrane-bound polysomes containing histone mRNAs and the

membrane-bound polysomes containing the cell surface class I

HLA-B7 antigen mRNAs were found associated with the

cytoskeleton.



RESULTS

Localization of Histone mRNA and HLA-B7 mRNA in Subcellular
Fractions

To determine the intracellular locations of histone

mRNA and HLA-B7 mRNA, HeLa cells were first fractionated

into nonmembrane-bound and membrane-bound polysomal

components by the procedure outlined in Fig. 2-2. The

membrane-bound polysomes are represented by the mitochondria

associated rough endoplasmic reticulum (MRER) and the

nucleus associated rough endoplasmic reticulum (NRER)










fractions. The light endoplasmic reticulum (LER), which is

deficient in polysomes by definition, was also isolated

following this subcellular fractionation scheme. Histone and

HLA-B7 mRNA content in each of the subcellular fractions was

determined by Northern blot analysis. Briefly, RNA was

extracted from the samples, quantitated and electrophoresed

in agarose under denaturing conditions. The RNA was then

transferred to nitrocellulose filters for hybridization with
P-labeled (nick-translated) pFOO08A (H4 histone), pFF435C

(H3 histone) and pDP001 (cDNA clone of HLA-B7 class I

antigen) DNA probes. A typical screening of subcellular

fractions with these probes is represented by the

autoradiograph in Figure 3-1, and Table 3-1 summarizes the

numerical values derived from densitometric analysis of the

Northern blot. The HLA-B7 mRNA and histone mRNA species are

each present in total cellular RNA and total polysomal RNA

which were included as positive controls. Non-membrane-bound

polysomal RNA contains approximately 55% and 51% of the

total H4 and H3 histone mRNA, respectively. Surprisingly,

the LER fraction contains approximately 38% of the H4

histone mRNA and 41% of the H3 histone mRNA, which may be a

consequence of the method by which the fraction is isolated.

The LER and non-membrane-bound polysomal sample is

centrifuged through a 2 M sucrose pad under conditions where

the LER forms a loose band on top of the pad, while the

nonmembrane-bound polysomes are pelleted. The LER band is











removed by aspiration and occasionally material is removed

from within the sucrose pad. This material may be the source

of histone mRNA in the LER fraction, which could decrease

the true percentage of total histone message found on the

nonmembrane-bound polysomes. The HLA-B7 mRNA was located

almost exclusively on membrane-bound polysomes. Nearly 90%

of the HLA-B7 mRNA was collectively found on the NRER and

MRER, while less than 11% was measured in the LER and

nonmembrane-bound polysomal RNA fractions.



Histone mRNA and HLA-B7 mRNA Distribution Between Csk and
Sol Fractions

Existing within the cytoplasm of eukaryotic cells is an

intricate network of protein filaments, which is referred to

as the cytoskeleton. This structure can be prepared from

HeLa cells treated with non-ionic detergent as described by

Penman and coworkers (Cervera et al., 1981). We used this

procedure for the isolation of the cytoskeleton (Csk) and

soluble (Sol) fractions which were subsequently assayed for

mRNA content by Northern blot analysis.

Log phase, suspension grown HeLa S3 cells were

fractionated into a Csk phase and Sol phase as described in

Chapter 2. Although this fractionation procedure has been

characterized previously (Lenk and Penman, 1979), we

examined the protein heterogeneity of each fraction by two-

dimensional gel electrophoresis and immunoblotting

techniques in collaboration with Dr. Warren Schmidt. As seen









Figure 3-1. Northern blot analysis of H3 histone. H4 histone
and HLA-B7 mRNA in subcellular fractions.

The cytoskeleton and soluble fractions, as well as the
nonmembrane-bound and membrane-bound polysomal RNA
fractions, were isolated from HeLa cells by the procedures
described in Chapter 2 (Cervera et al., 1981; Venkatesan and
Steele, 1972). In addition, total polysomal RNA and total
cellular RNA were isolated (Plumb et al., 1983, Venkatesan
and Steele, 1972). RNA was extracted from each subcellular
fraction and 50 yg of RNA from each preparation was resolved
by 1.5% agarose, 6% formaldehyde gel electrophoresis. The
RNA was transferred to nitrocellulose and hybridized to
radiolabeled pFF435C (H3 histone and pFO108A (H4 histone)
probes, followed by hybridization to P-labeled (nick-
translated) pDP001 (cDNA of HLA-B7) probe. The hybridized
filters were then exposed to preflashed XAR5 x-ray film at -
700C for 72 hours. From left to right, lanes: 1) total
cellular RNA; 2) Csk RNA; 3) Sol phase RNA; 4) NRER; 5)
MRER; 6) nonmembrane-bound polysomal RNA; 7) LER; and 8)
total polysomal RNA.























H LA-B7








1 2 3 4 5 6 7 8









82

Table 3-1. Quantity of histone and HLA-B7 mRNAs in the
subcellular fractions.

H3 H4 HLA-B7
Cytoskeleton 90% 84% 97%
Soluble 10% 16% 3%

NRER 3% 3% 29%
MRER 6% 5% 60%
Free 51% 54% 5%
LER 40% 38% 6%

The distribution of histone and HLA-B7 mRNAs in the
NRER, MRER, LER and nonmembrane-bound polysomal fractions as
well as the cytoskeleton and soluble fractions was
determined by Northern blot analysis. The values have been
normalized with respect to the total yield of RNA within
each fraction.











in Figure 3-2, the cytoskeleton fraction clearly contained

actin as well as the major HeLa cytokeratins (Franke et al.,

1981). In addition, vimentin appeared to be enriched in the

cytoskeleton, which was confirmed by immunoblot analysis.

Extraction of the cytoskeleton required a cold, calcium

containing buffer and consequently the tubulins were

solubilized and therefore absent in the Csk fraction. The

tubulin proteins were clearly contained in the soluble

fraction as determined by immunoblot analysis. Additional

support for separation of cytoskeleton and soluble phases

comes from the observation that the bulk of the tRNA species

is in the soluble phase which is in agreement with

previously reported values (data not shown; Cervera et al.

1983). Although the HLA-B7 messenger RNA is found on

membrane-bound polysomes and the histone mRNA on

nonmembrane-bound polysomes, both species of messenger RNAs

are associated almost exclusively with the cytoskeletal

fraction. Nearly 97% of the HLA-B7 mRNA and approximately

85-90% of the H3 and H4 histone messages are retained in the

Csk (Fig. 3-1; Table 3-1).



The Effects of Metabolic Inhibitors on the Subcellular
Localization of Histone and HLA-B7 mRNA

Detectable levels of histone mRNA were observed in each

of the subcellular samples, including the Sol fraction and

the membrane-bound polysomal fractions. However, the HLA-B7









Figure 3-2. Two-dimensional gel electrophoresis of
cvtoskeleton and soluble associated proteins.

HeLa cells were fractionated into a cytoskeleton and
soluble phase, protein was extracted from each phase and
subsequently analyzed by two-dimensional gel electrophoresis
as described in Chapter 2. Proteins of the soluble fraction
(Sol), A, were separated using wide range (pH 3-7)
ampholines while cytoskeletal proteins (Csk), B, were
separated with narrow range (pH 5-7) ampholines to more
clearly identify cytoskeletal species. V: vimentin, a:
actin, and t: tubulins. Unlabeled arrows in B depict HeLa
cytokeratins of 54 Kd (pI 6.0); 52.5 Kd (pI 6.1); 46 Kd (pI
5.1); and 45 Kd (pI 5.7), which were spots highly reactive
with commercially prepared prekeratin or stratum corneum
antisera (Franke et al., 1981).









pH
7.0 6.4
T v


5.2 4.1
T T


5.4 5.0 4.4
T T V


A)








43K-


43K-


I A


CSK


SOL ...









86

mRNA was located almost exclusively in the cytoskeleton and

membrane-bound polysome fractions. The widespread

subcellular localization of histone mRNA may be due to non-

specific trapping of nonmembrane-bound polysomes in each of

the fractions during the isolation procedure. Alternatively,

this observation may reflect the presence of mRNAs encoding

variant histone proteins that could be targeted to the

different cytoplasmic compartments. Inhibition of DNA

synthesis with hydroxyurea results in a rapid

destabilization of cell cycle dependent histone mRNAs,

whereas cell cycle independent histone mRNAs remain stable

in the presence of this drug (Baumbach et al., 1984; Heintz

et al., 1983; DeLisle et al., 1983; Graves and Marzluff,

1984; Plumb et al., 1983). Conversely, inhibition of protein

synthesis with cycloheximide results in a 2-3 fold increase

in cytoplasmic levels of histone mRNAs (Butler and Mueller,

1973; Baumbach et al., 1984). Differential sensitivity of

histone mRNAs in the subcellular fractions to hydroxyurea or

cycloheximide treatment would favor the latter suggestion

that cell cycle dependent and independent histone gene

transcripts may selectively reside in specific cytoplasmic

compartments.

Treatment of exponentially growing HeLa cells with

hydroxyurea, as described in Chapter 2, resulted in a

complete and selective destabilization of histone mRNA in

the LER, nonmembrane-bound and membrane-bound polysomal











fractions (Figure 3-3). This is in sharp contrast to the

HLA-B7 mRNA which appeared to be fully stable under these

conditions. Cycloheximide treatment had a stabilizing effect

on the levels of histone mRNA in each subcellular fraction

(Figure 3-3). When compared with control HeLa cells,

cycloheximide treatment resulted in a 2.1 fold (H4) and a

2.4 fold (H3) increase in the levels of histone mRNA in the

nonmembrane-bound polysomal fraction. Similar increases were

also seen for levels of histone mRNAs in the LER and

membrane-bound polysomal fractions. Inhibition of protein

synthesis with cycloheximide followed by DNA synthesis

inhibition with hydroxyurea resulted in the stabilization of

histone mRNA in each subcellular fraction, whereas the

stability of HLA-B7 mRNA remained unaffected (Figure 3-3).

Similar results were obtained when these inhibitor studies

were expanded to include the cytoskeleton and soluble

fractions (Figure 3-4). Cycloheximide treatment resulted in

a 2-3 fold increase in the representation of H3 and H4

histone mRNA over untreated cultures in both the

cytoskeleton and soluble fractions. HLA-B7 mRNA appeared to

be only moderately stabilized in the presence of this drug.

When cells were treated with hydroxyurea, histone mRNA

levels were greatly reduced in both the cytoskeleton and

soluble fractions whereas the HLA-B7 mRNA was found in

comparable levels to those present in the untreated samples.

Consistent with previous data, cotreatment of HeLa cells









Figure 3-3. Northern blot analysis of H3 histone. H4 histone
and HLA-B7 mRNAs isolated from subcellular fractions of
cells treated with metabolic inhibitors.

Exponentially growing HeLa cells were incubated in the
presence of metabolic inhibitors and fractionated into
nonmembrane-bound and membrane-bound polysomes as described
in Chapter 2. RNA was extracted from each fraction and
assayed for H3 histone mRNA, H4 histone mRNA and HLA-B7 mRNA
content by Northern blot analysis as described in Figure 3-1
and Chapter 2. Upper panel from left to right, lanes: 1)
control (untreated), nonmembrane-bound polysomes; 2)
control, NRER; 3) control, MRER; 4) control, LER (omitted);
5) cycloheximide (Cy), nonmembrane-bound polysomes; 6) Cy,
NRER; 7) Cy, MRER; and 8) Cy, LER. Lower panel from left to
right, lanes: 1) hydroxyurea (HU), nonmembrane-bound
polysomes; 2) Hu, NRER; 3) Hu, MRER; 4) Hu, LER; 5) CY/HU,
nonmembrane-bound polysomes; 6) Cy/Hu, NRER; 7) Cy/Hu, MRER;
and 8) Cy/Hu, LER.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E7JEYIHGD_AM9AO5 INGEST_TIME 2012-12-07T22:58:22Z PACKAGE AA00012891_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES