On the mechanism of lambda prophage derepression

MISSING IMAGE

Material Information

Title:
On the mechanism of lambda prophage derepression
Physical Description:
ix, 70 leaves : ill. ; 29 cm.
Language:
English
Creator:
Crowl, Robert Mitchell, 1952-
Publication Date:

Subjects

Subjects / Keywords:
Bacteriophage lambda   ( mesh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1979.
Bibliography:
Includes bibliographical references (leaves 64-69).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Robert Mitchell Crowl.

Record Information

Source Institution:
University of Florida
Rights Management:
Permission granted to the University of Florida to digitize, archive and distributed this item for non-profit and educational purposes only. Any reuse of this item in excess of fair use requires permission of the copyright holder.
Resource Identifier:
oclc - 25174127
ocm25174127
System ID:
AA00020008:00001

Full Text











ON THE MECHANISM
OF LAMBDA PROPHAGE DEREPRESSION









By

ROBERT MITCHELL CROWL


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







UNIVERSITY OF FLORIDA


1979






















To my parents,
Dr. and Mrs. Robert H. Crowl














ACKNOWLEDGEMENTS

I wish to express my gratitude to Dr. Richard Boyce, my thesis

advisor, for his continuing support and friendship and for introducing

me to "Xology"; to Drs. Phil Laipis, Peter McGuire, and Neal Ingram

for their advice and encouragement during the course of this work;

to Dr. Koto at Kyushu University in Japan for sending me Xind5; and

to Dr. Harrison Echols at Berkeley for supplying me with anti-cl

serum.

Most importantly, I would like to express my deep appreciation

to Jody, my wife, for "putting hubby through" and for being a

constant source of happiness.













TABLE OF CONTENTS

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

LIST OF TABLES . . . . . . . . . . . .. iv

LIST OF FIGURES . . . . . . . . . . . v

ABBREVIATIONS USED . . . . . . . . . . .. vi

ABSTRACT . . . . . . . . . . . . .. vii

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

MATERIALS AND METHODS . . . . . . . . . .. 13

Bacterial and Phage Strains . . . . . . . .. . 13
Media . . . . . . . . . . . . .. 16
Chemicals . . .... ........................... 16
Enzymes and Proteins . . . . . . . . .. 17
Radioisotopes . . . . . . . . . . .. 17
Antisera . . . . . . . . . . . .. 17
Prophage Induction Assay . .... .................. 17
Preparation of [3H] Labeled X DNA . . . . . ... .17
Preparation of Cell Extracts
for cl Repressor Binding Assay . . . . . .. .18
DNA Binding Assay for cl Repressor . . . . . .. .19
SDS-PAGE . . . . . . . . . . . .. 19
Autoradiography . . . . . . . . . ... 21
Immunoprecipitation Assay for
the cl Repressor . . . . . . . . . .. 22

RESULTS . . . . . . . . . . . . .. 24

Studies on the Role of DNA
Degradation in X Derepression . . . . . ... .24
Studies on Xclind5 Prophage
Derepression . . . . . . . . . . .. 38

DISCUSSION . . . . . . . . . . . . .. 55

BIBLIOGRAPHY . . . . . . . . . . . .. 64

BIOGRAPHICAL SKETCH . . . . . . . . . .. 70















LIST OF TABLES


1. Bacterial strains . . . . . . .

2. Phage strains . . . . . . . .

3. Effect of MC on UV and thermal induction of X

4. Effect of A(xth-pncA) on UV induction of X.

5. Effect of protein synthesis inhibitors on
prophage derepression . . ... . ..

6. Effect of protease inhibitors on prophage
derepression. . ... . .. . . ..


. . . . 14
..* ......15
.315

. . . . 35

. . . . .37


. . . . .44


. . . . 51














LIST OF FIGURES

1. The early region of the A genome . . . . . .. 3

2. DNA-binding assay for the cl repressor . . . ... 20

3. Effect of recB21 on NA-induction of A . . . . .. 26

4. Effect of X exonuclease expression on
recB-dependent induction of the
RecA protein . . . . . . . . . . .. 29

5. Effect of uvrA6 on MC-induction of
the RecA protein . . . . . . . . . .. 32

6. Effect of uvrA6 on recB21 on
MC-induction of X . . . . . . . . .. 34
7. Comparison of cl and clind5 repressor
binding activities . . . . . . . . .. 40

8. UV dose-response for prophage derepression . . ... 42

9. Kinetics of UV-mediated prophage
derepression . . . . . . . . . . .. 43

10. Effects of recAl3 and lexA-l on
prophage derepression . . . . . . . .. 46

11. tif-mediated prophage derepression . . . . ... 47

12. NA-mediated clinds derepression:
effect of recB21 . . . . . . . . . .. 50

13. Proteolytic cleavage of clind5
repressor in UV irradiated lysogens . . . . .. 53











ABBREVIATIONS USED

BIS N,N'-methylene-bis-acrylamide

gME -mercaptoethanol

BSA bovine serum albumin

CAP chloramphenicol

CPM counts per minute

DTT dithiothreitol

EDTA ethylene diaminetetraacetic acid

Kd kilodaltons

MC mitomycin C

NA nalidixic acid

NEM N-ethylmaleimide

PAGE polyacrylamide gel electrophoresis

PFU placque forming units

POPOP p-bis [2-(5-phenyloxazoyl)]-benzene

PPO 2,5-diphenyloxazole

SDS sodium dodecyl sulfate

TEMED N,N,N',N'-tetramethylethylenediamine

TLCK tolyl leucylchloromethyl ketone

Tris Tris(hydroxymethyl)aminomethane

UV ultraviolet











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


ON THE MECHANISM
OF LAMBDA PROPHAGE DEREPRESSION

By

Robert Mitchell Crowl

June 1979

Chairman: Richard P. Boyce
Major Department: Biochemistry and Molecular Biology

Several treatments which cause DNA damage and/or cessation of DNA

synthesis elicit the derepression of X prophage in lysogenic strains

of E. coli, including exposure to ultraviolet (UV) light, mitomycin C

(MC), and nalidixic acid (NA). These same treatments also result in

the production of high levels of the bacterial recA gene product.

Existing evidence indicates that induction of the RecA protein is

requisite for X derepression. Biochemical studies have revealed that

the RecA protein contains an endopeptidase activity that specifically

cleaves the cAd repressor in vitro. This cleavage appears to be

identical to that observed in vivo in induced lysogens. The major

questions to which the studies in this dissertation are addressed

include the following: 1) What are the early events which occur in

the cell following inducing treatments that provoke induction of the

RecA protein and,consequently, derepression of X? 2) Is proteolytic

cleavage of the cl repressor the cause of derepression or the conse-

quence of some other primary derepression event?








At the time these studies were undertaken, the best existing

evidence that provided insight into the first question above was that

NA-mediated induction of the RecA protein requires functional recB and

recC genes. As the gene products of recB and recC, exonuclease V, are

responsible for DNA degradation following NA treatment, it was proposed

that a product of exonucleolytic activity is a necessary component in

the mechanism of RecA protein induction. This idea was subsequently

challenged in the literature on the basis of the evidence that UV-induc-

tion of the RecA protein is recB-independent and that intracellular

degradation of unmodified phage DNA via restriction does not result in

induction of the RecA protein.

Since the RecBC protein is a multifunctional enzyme, it seemed

possible that a function other than the exonuclease activity (such as

the ATPase activity) could be responsible for the induction response.

To test this, I determined whether expression of X genes controlling X

exonuclease, an exclusively exonucleolytic enzyme, could circumvent

the recB21 mutation in NA-mediated induction of the RecA protein. I

found that expression of X exonuclease could restore the missing

function for NA-mediated RecA protein induction in recB21 lysogens,

indicating that DNA degradation is necessary for induction of the RecA

protein, at least in the case of NA-induction.

To examine further the role of RecA-mediated proteolytic cleavage

in X derepression, I utilized the phage mutant, Xclinds, which is

sensitive to induction by very low doses of UV radiation. I found that

UV-mediated inactivation of clinds repressor requires a functional

recA gene; however, in contrast to XcI+ derepression, Xc1inds derepres-

sion does not require de novo synthesis of the RecA protein. Moreover,


viii









I determined that proteolytic cleavage is not necessary for the abolish-

ment of operator-binding activity of the clinds repressor in induced

lysogens. The finding that the clind repressor is inactivated by a

recA-dependent process that is independent of RecA-mediated cleavage

means that the recA gene product functions in some capacity other than

as a protease in the derepression of the Xclinds prophage.

The results described in this dissertation, in light of the recent

report that cl+ repressor binds to single-stranded gapped DNA, can be

accommodated by a model in which the cl+ repressor is inactivated by a

two-step mechanism: 1) reversible repressor binding to gapped DNA

and 2) irreversible proteolytic cleavage of the bound repressor. In

this model, the role of DNA degradation, at least in the case of some

inducing treatments, is to produce the gaps in the DNA. The role of

the recA gene product in this hypothetical mechanism may be two-fold:

1) to stabilize the gapped DNA structure, enabling the repressor to

bind, and 2) to catalyze the cleavage of the bound repressor. I

propose that the clind5 mutation circumvents the second derepression

step, and that an increased binding affinity of the mutant repressor

for gapped DNA is sufficient to remove it from its operator sites,

thus committing the Xclind prophage to the lytic developmental

pathway.













INTRODUCTION

After bacteriophage A infects its host Escherichia coli K12, one

of two possible modes of phage propagation ensues. In a fraction of

an infected bacterial population, X replicates autonomously, producing

many progeny phage per cell which are eventually released from the

cell upon lysis. This mode of phage propagation is called the lytic

developmental pathway. Alternatively, some of the bacterial cells

infected by X are lysogenized by the phage; that is to say, the X DNA

becomes integrated into a specific site on the E. coli genome and

remains in a dormant state, replicating passively as part of the bac-

terial chromosome. The integrated phage DNA is called a prophage, and

a bacterium harboring the X prophage is said to be "lysogenic for X."

If a X-lysogen is further infected by X, the superinfecting phage DNA

enters the cell; however the genetic information necessary for lytic

development is not expressed. Thus, X-lysogens are said to be "immune"

to superinfection by X. The lysogenic state is stable to the extent

that only once in about 104 cell generations, the prophage enters the

lytic deveopmental pathway. This frequency can be increased by exposing

a lysogenic culture to UV radiation. This results in the initiation of

the lytic pathway in almost every cell, culminating in lysis of the

culture. This phenomenon, which was discovered nearly thirty years ago

(Lwoff et al., 1950), is known as prophage induction.

The foundation for most, if not all, of the present understanding

of X-lysogeny and prophage induction was provided by genetic experiments,








described below, which were performed in Jacob's laboratory during the

1950's. In 1957, Kaiser isolated mutants of X which are defective in

lysogenization and showed that three genes are essential for this pro-

cess: cI, cll, and cIII. The cII and cll genes are required only for

establishing lysogeny, whereas the cI gene is essential for both the

establishment and maintenance of lysogeny (Echols and Green, 1971).

Although Xcl- mutants fail to lysogenize E. coli, they are subject to

the immunity specified by a XcI+ prophage. Kaiser and Jacob (1957)

crossed X with the related lysogenic phage 434, which displays a

different immunity specificity, and found that only a small segment of

the X genome is necessary for the expression of X-specific immunity.

This segment, called the immunity region, comprises about 3% of the

A genome. The immunity region includes the ci gene and a short length

of genetic material on either side of cl. Kaiser and Jacob (1957)

concluded that the cl qene determines both the ability to maintain the

prophage state and the immunity specificity of X. Jacob and Wollman

(1953) isolated the multiple mutant, Xvir, which is able to superinfect

and multiply in Xcl+ lysogens; that is, Xvir is not subject to cI+-

specified immunity. The mutations responsible for the phenotype of

Xvir map within the immunity region at sites which flank the cI gene

(denoted OR and 0L in Fig. 1). Based on the foregoing genetic evidence,

as well as other experiments, Jacob and Monod (1961) proposed that the

cI gene codes for a repressor that acts at operator sites within the

immunity region of X and inhibits the expression of phage functions

required for lytic development. Further support for this idea was

provided by Sussman and Jacob (1962) who isolated the temperature-

sensitive mutant, XcI857. At 30, XcI857 lysogenizes E. coli normally;



































imm434


' cf N rex
5 ------ ,-


cl cr :CU 0 P
I (x) yv)


The early region of the X genome.


This map shows the cl gene in relation to its operator
sites OR and 0L and to other X genes within the regula-
tory and replication region of X DNA. The N and cro
gene products function early in the lytic developmental
pathway. The immunity region is marked by the line
denoted imm434 which indicates the DNA segment which is
substituted in Ximm434. This map was taken from the
article by Eisen and Ptashne (1971)


Figure 1.


--------I-------I---<-1-------^-







incubating such a lysogen at 42 results in induction of the prophage.

The existence of both temperature-sensitive (Sussman and Jacob, 1962;

Lieb, 1966) and suppressor-sensitive mutations (Jacob et al., 1962)

indicated that the repressor is a protein.

The isolation of the cl protein (Ptashne, 1967a) eventually led

to the biochemical confirmation of the model proposed by Jacob and

Monod (1961) to explain how the ci gene product maintains lysogeny.

Ptashne (1967b) showed that the cI protein binds to X DNA but not to

Ximm434 DNA. Ximm434 is identical to \ except that it contains the

immunity region of phage 434 (Kaiser and Jacob, 1957). Ptashne and

Hopkins (1968) determined that the cI protein binds to two sites within

the immunity region, known as the operators, OR and 0L (see Fig. 1).

This specific binding inhibits the in vitro transcription of X DNA

initiating at the promoter sites, PR and PL (Steinberg and Ptashne,

1971; Wu et al., 1972).

Using the nitrocellulose membrane filter binding assay (see

MATERIALS AND METHODS) to monitor cl repressor-operator binding activity,

Pirrotta et al. (1971) purified about 40 mg of repressor from 4 Kg of

cells. The cI repressor binds as a dimer or a tetramer to its operator

with a dissociation constant of 3 x 10-14 mole liter-1 (Pirrotta,

1976). This tight binding protects the operator sites from nuclease

digestion, a property that allowed Maniatis and Ptashne (1973) to

isolate the operator DNA segments. The amino acid sequence of the

27,000 dalton cl repressor monomer was determined by standard protein

sequencing techniques (Sauer and Anderegg, 1978) and confirmed by

determining the nucleotide sequence of the cI gene (Sauer, 1978) by

the method of Maxam and Gilbert (1977). The nucleotide sequences of




5



the operator sites, OR and OL, have also been determined and three

repressor contact sites within each operator have been identified (see

Pirrotta, 1976).

In contrast to the detailed information available concerning the

cl repressor and its function in maintaining the prophage state of X,

very little is understood at the biochemical level about how the pro-

phage is induced. X can be induced by several treatments, in addition

to UV irradiation (Jacob and Wollman, 1953), that damage DNA and/or

inhibit DNA synthesis. These include exposure to mitomycin C (Otsuji

et al., 1959), nalidixic acid (Cowlishaw and Ginoza, 1970), and thymine

starvation (Sicard and Devoret, 1962). Expression of certain temper-

ature-sensitive mutations in genes involved in DNA synthesis also

results in X induction: lig (Gottesman et al., 1973) dna B (Noack and

Klaus, 1972), dna E, dna G (Schuster et al., 1973), and pol_ A (Blanco

and Pomes, 1977).

A induction requires two functionally separate events: derepression

(inactivation of the cl repressor) and excision of the prophage DNA

(Gingery and Echols, 1967). Tomizawa and Ogawa (1967) reported the

first comprehensive study on X induction at the level of derepression.

They developed a totally in vivo system, described below, in which the

cl repressor within the cell could be studied without the presence of

the prophage. When E. coli is infected with the deletion mutant Xb2

(which lacks the attachment region of the X genome), the phage cannot

integrate into the host genome but can remain as a plasmid. The cell

is then immune to superinfection by homologous phage and is said to be
abortivelyy lysogenized." The phage DNA is transmitted to only one

daughter cell during cell division and is eventually segregated out of







a population of bacteria. Tomizawa and Ogawa (1967) showed that non-

lysogenic segregants of cells abortively lysogenized with Xb2 contain

sufficient levels of cl repressor to remain immune to superinfection.

They found that when these cells are treated with UV radiation, their

immunity to superinfection disappears 30 minutes after irradiation.

They showed that the loss of immunity is inhibited when the cells are

incubated in the presence of CAP, suggesting that de novo protein syn-

thesis is required for derepression. These authors discovered that UV

irradiation does not result in the disappearance of cI-specified immunity

in cells containing a mutation in the recA gene. This result is con-

sistent with the finding that UV-induction of X prophage requires a

functional bacterial recA gene (Brooks and Clark, 1967; Hertman and

Luria, 1967). Tomizawa and Ogawa (1967) suggested that the cl regressor

is inactivated by a protein that is produced or activated in the cell

following UV treatment in the absence of the prophage, and that at least

one bacterial function (namely, the function of the recA gene) is

required for this process.

Shinagawa and Itoh (1973) used the membrane filter binding assay

to measure directly the inactivation of the cI repressor in induced

lysogens. They found that, after a lag period of about 10 minutes, the

operator-binding activity of the cl repressor decreases monotonically

and is negligible by 30 minutes after lysogenic cells have been exposed

to MC or UV radiation. In agreement with the results of Tomizawa

and Itoh (1967), these authors determined that inactivation of the cl

repressor in vivo requires a functional recA gene and de novo protein

synthesis (as determined by sensitivity to CAP).








The question of how the recA gene product is involved in X induc-

tion has been an intriguing one for over a decade. The first recA-

mutants were isolated independently in two laboratories as recombination

defective (Clark and Margulies, 1965) and DNA repair defective (Howard-

Flanders and Theriot, 1966) strains. Clark et al. (1966) found that the

recA- mutation results in abnormally high rates of DNA degradation both

in untreated cells and in cells exposed to UV radiation. On the basis

of genetic evidence, Willetts and Clark (1969) concluded that, in normal

cells, the recA gene product controls the exonuclease activity responsi-

ble for this DNA degradation. This function of the recA gene product

is presumably necessary for both general recombination and post-replica-

tion repair processes. The effect of the recA- mutation on X induction

implies that either this same function is an essential one in the pro-

phage derepression mechanism or that the recA gene product has more

than one function.

A functional recA gene is essential for the expression of a number

of UV-inducible responses in E. coli, an addition to prophage derepres-

sion, called SOS functions (Radman, 1975). These include mutagenesis

(Witkin, 1969), inhibition of cell division or cell filamentation

(Inouye, 1971), inducible DNA repair (Radman, 1975), and the production

of large amounts of a 40,000 dalton polypeptide, originally designated

protein X (Gudas and Pardee, 1975; West and Emmerson, 1977). Further

evidence which indicates that the recA gene product mediates the

expression of this complex set of responses is provided by an unusual

temperature-sensitive recA mutation, termed tif-I (Castellazzi et al.,

1972; Castellazzi et al., 1977). At elevated temperatures, the tif-l

mutation results in cell filamentation (Kirby et al., 1967), increased







mutagenesis (Witkin, 1974), induction of protein X (West and Emmerson,

1977), and X prophage induction in lysogenic strains (Kirby et al.,

1967). Unlike other temperature-sensitive mutations that result in

X induction (see p. 5 ), expression of the tif-l mutation produces no

observable pertubations of DNA structure or DNA synthesis (Kirby et al.,

1967).

Insight into a possible mechanism by which the recA gene product

mediates induction of the SOS functions was provided by the work of

Roberts and Roberts (1975) on X prophage derepression. Using an

immunoprecipitation technique, these authors followed the fate of 35S-

labeled cl repressor in induced lysogens by SDS-PAGE. They discovered

that exposure of a lysogenic culture to UV radiation or MC results in

a specific proteolytic cleavage of the cl repressor monomer. They

showed that cleavage does not occur in recA- lysogens or in lysogens

carrying the noninducible Xclind- prophage. They suggested that,

following inducing treatments, the recA gene product functions as a

protease or activates a protease that cleaves the cl protein. Witkin

(1976) proposed that all of the SOS functions are derepressed by a

recA-dependent proteolytic cleavage mechanism. This idea is supported

by the demonstration that the protease inhibitor, antipain, inhibits

tif-mediated X prophage induction, mutagenesis, and cell filamentation

(Meyn et al., 1977).

The construction of X transducing phage (XprecA) carrying the recA

gene (McEntee, 1976) led to the radiochemical isolation of the recA

gene product as a 40,000 dalton polypeptide (McEntee et al., 1976).

Little and Kleid (1977) demonstrated that trypsin and chymotrypsin

digestion patterns of the recA gene product (isolated from XprecA









infected cells) and protein X (isolated from NA-induced cells) are

identical. McEntee (1977) showed that mutations in recA that alter the

molecular weight (recAl2) or the isoelectric point (tif-l) of the recA

gene product, produce the same alterations in protein X. Thus, the recA

gene product is protein X.

Roberts et al. (1978) succeeded in purifying the protease activity

that appears to be responsible for cleaving the cl repressor in vivo.

They showed that this activity co-purifies with a 40,000 dalton poly-

peptide that appears homogeneous by SDS-PAGE. When this polypeptide is

coelectrophoresed with RecA protein, the two bands seem to be identical.

These authors demonstrated that antibody, prepared against the purified

protease, specifically precipitates the RecA protein. Moreover, they

determined that mutations in recA (tif-l and recAl) either enhanced or

eliminated the protease activity in the purified protein. Roberts et al.

(1978) concluded that the recA gene product is responsible for the

proteolytic cleavage of the cI repressor in vitro and in vivo.

RecA-mediated proteolytic cleavage of the cI repressor in vitro

requires the presence of ATP and Mg++. The requirement for ATP seems

to be a general one for intracellular proteases (Etlinger and Goldberg,

1977). A curious aspect of the protease activity associated with the RecA

protein is that high concentrations (on the order of 300 ig/ml) are

necessary to promote cleavage of the cl repressor (Roberts et al., 1978).

In vivo, high levels of the RecA protein also seem to be required for

the inactivation of the cI repressor videe infra).

Gudas and Pardee (1975) found that induction of protein X is

abolished by the lexA- mutation. The lexA" mutation results in a

phenotype similar to the recA- mutation in that lexA- cells are sensitive








to UV radiation and show high rates of DNA degradation; however the

lexA- mutation has no effect on recombination proficiency (Mount et al.,

1972). In addition, lexA- maps at 90 minutes on the recalibrated

E. coli genetic map (Bachman et al., 1976), whereas recA maps at 58

minutes. Thus, lexA codes for a different polypeptide than recA. Sev-

eral authors have proposed that the lexA gene encodes a repressor that

inhibits the transcription of the recA gene in uninduced cells (Gudas

and Pardee, 1975; Gudas and Mount, 1977; McEntee, 1977). This idea is

supported by the existence of mutations in lexA, described below, which

are analogous to repressor mutations in the cl gene of X and the i gene

of the lac operon (Jacob and Monod, 1961). 1) The spr- mutation, which

maps in lexA (Mount, 1977) results in a constitutive phenotype; that

is, high levels of the RecA protein are normally present in spr cells

(cited in Lloyd, 1978). 2) Expression of the thermosensitive mutation

in lexA, called tsl- (Mount et al., 1973), results in the increased

production of the RecA protein at elevated temperature (Gudas and

Pardee, 1975), and thus gives a conditionally constitutive phenotype.

3) The lexA- mutation (as stated above) results in a noninducible

phenotype; that is, increased production of the RecA protein does not

occur in lexA- cells treated with inducing agents (Gudas and Pardee,

1975).

The lexA- mutation not only blocks induction of the RecA protein,

but also inhibits induction of X prophage (Donch et al., 1970). This

indicates that high levels of RecA protein are necessary to inactivate

the cl repressor in induced lysogens. However, high levels of RecA

protein alone are insufficient to derepress X prophage, as expression

of the tsl- and spr mutations does not result in X induction (Mount







et al., 1973; Mount, 1977). This indicates that some element in addition

to the RecA protein is necessary to inactivate the cl repressor in vivo.

Several authors have suggested that a low molecular weight effector

is involved in X induction (Goldthwait and Jacob, 1964; Hertman and

Luria, 1967; Noack and Klaus, 1972). This is an attractive idea since

it is analogous to the mode of enzyme induction in the lac operon

(Jacob and Monod, 1961). The strongest support for this idea is provided

by the evidence that tif-mediated induction of X is enhanced by high

concentrations of adenine (Kirby et al., 1967). Gudas and Mount (1977)

proposed that a small effector molecule converts the RecA protein to an
"activated" protease. It is thought that the tif-l mutation circumvents

the requirement for an effector (McEntee, 1977).

As most treatments that induce x prophage also inhibit DNA replica-

tion, it has been proposed that the hypothetical effector is a precursor

to DNA synthesis that accumulates in the cell following cessation of

DNA replication (Goldthwait and Jacob, 1964; Noack and Klaus, 1972).

Lydersen and Pollard (1977) argued against this idea on the basis of

their evidence that the DNA synthesis inhibitor, hydroxyurea, which

blocks an early step in nucleotide biosynthesis, does not induce \ and

does not inhibit UV-mediated induction.

Gudas and Pardee (1975) proposed that a product of DNA breakdown

following inducing treatments is responsible for the initiation of

induction. This was based mainly on their finding that NA-mediated

induction of protein X requires functional recB and recC genes, which

code for exonuclease V. recBC-dependent DNA degradation following NA

treatment presumably occurs at a stalled replication fork (Gudas and

Pardee, 1976).






Little and Hanawalt (1977), on the other hand, challenged the idea

that DNA degradation is required for induction. They showed that UV-

mediated induction of protein X is independent of recB. They also

determined that intracellular degradation of unmodified phage DNA via

restriction is not sufficient to induce protein X. These authors con-

cluded that "no causal relationship exists between the production of DNA

fragments and induction of protein X."

The findings that the recA gene product is the UV-inducible protein

X and that this protein promotes the proteolytic cleavage of the X cl

repressor were the key breakthroughs which provided the impetus for the

experiments described in this dissertation. These developments directed

my interests to the following questions: 1) What are the early events

which occur in the cell following inducing treatments that provoke the

induction of the RecA protein and,consequently, the derepression of

X prophage? 2) Is proteolytic cleavage of the cl repressor the cause

of derepression or the consequence of some other primary derepression

event?













MATERIALS AND METHODS

Bacterial and Phage Strains

The strains of Escherichia coli K12 and bacteriophage X used in

this work are listed in Tables 1 and 2.

Lysogens were constructed by infecting cells with the appropriate

phage, streaking the surviving cells to single colonies, and testing

the colonies for sensitivity to infection by Xc190 and Xvir. Cells

which were resistant to Xc190 but sensitive to Xvir were picked as

possible lysogens. Lysogeny was confirmed by induction with either UV

or incubation at 42 (in the case where the prophage carried the c1857

allele).

RClO0 was constructed by crossing AB2437 with AB2470 and selecting

for colonies which survived infection with T6 phage and which grew on

media lacking proline and leucine. These colonies were then tested

for sensitivity to UV irradiation and compared to the parental strains.

RC100 showed extreme UV sensitivity, was unable to repair UV-irradiated

Xc190, and lysogenic derivatives were not induced with NA.

RC109 was constructed by P1 transduction. Pivir, grown on BW9109,

was used to infect RC100. Surviving cells were then selected for

resistance to 6-aminonicotinamide, indicating the absence of the pncA

gene (White et al., 1976).

RC9137 was constructed by crossing AB2437 with BW9109 and selecting

colonies which were resistant to T6 infection and which grew on media

lacking proline and leucine and containing 6-aminonicatinamide. These










Strain No.

ABI157


AB1886

AB2463

AB2494

AB2470

JMl 2

AB2437

W3350

Ql

Ymel

BW9109

RCI 00

RC109

RC9137


Table 1. Bacterial strains

Genotype

his-4 proA2 leu-6 thr-l argE3
rpsL31(Str') thi-1 ga1K2 lacY]
mtl-l xyl-15 ara-14 tsx-33 supE44

uvrA6 derivative of AB1157

recAl3 derivative of AB1157

lexA-1 derivative of AB1l57

recB21 derivative of ABl157

tif-l derivative of AB1157

Hfr J2 uvrA6 his-4 thi-l

gaIK2

thr-l leu-6 lacYl tonA21 supE44

supE57 supF58

A(xth-pncA) derivative of AB1157

uvrA6 recB21

uvrA6 recB21 A(xth-pncA)

uvrA6 A(xth-pncA)


Source or reference

Howard-Flanders et al. (1966)



Howard-Flanders et al. (1966)

Howard-Flanders and Theriot (1966)

Howard-Flanders (1968)

Emmerson (1968)

Castellazzi et al. (1972)

Howard-Flanders et al. (1966)

Weisberg and Gottesman (1969)

Shulman et al. (1970)

Yanofsky and Ito (1966)

White et al. (1976)
This work

This work

This work















Table 2. Phage strains


Strain

Xref or X+

Xc1857

Xc1857 redX314

Xc1857 redB113

Xc1857 Sam7

Ximm434 clts Sam7

Xclinds

Xc190

Xvir


Reference

Dove (1969)

Sussman and Jacob (1962)

Radding (1970)

Shulman et al. (1970)

Goldberg and Howe (1969)

Wu et al. (1972)

Horiuchi and Inokuchi (1967)

Kaiser (1957)

Jacob and Wollman (1953)







colonies were pooled and infected with Xc190 which had received
2
300 ergs/mm2 of UV radiation. Surviving colonies were then tested for

UV sensitivity.

The presence of the A(xth-pncA) marker was confirmed in RC109 and

RC9137 by demonstrating the absence of normal levels of exonuclease

III, using the assay of Milcarek and Weiss (1972).

Media

LB media contains 10 g Bactotryptone, 5 g yeast extract, and 5 g

NaCl per liter of deionized H20. LB-agar is LB media solidified with

1.5% Bactoagar. M9 contains 1 g NH4C1, 11 g Na2HPO4 7H20, 3 g KH2PO4

5 g NaCl, 1 m mole MgSO4, and 100 ilmole CaCl2 per liter of dionized

H20. M9-glucose is M9 containing 1% glucose. K-glucose is M9 contain-

ing 1% glucose, 1% decolorized casamino acids, and 1 pg/ml thiamine.

LSM contains per liter: 7 g Na2HPO4, 3 g KH2PO4, 1 g NaCl, 0.1 g MgCl2,

3 pmol FeCl2, and 100 pmol CaCl2, and 40 jimol or 200 wmol of MgSO4

CLSM is LSM with .2% glucose, 1 ig/ml thiamine, 100 jig each of threonine,

arginine, histidine, leucine, and 400 pg/ml proline.

Chemicals

Mitomycin C, nalidixic acid, chloramphenicol, neomycin, TLCK,

and 6-aminonicatinamide were obtained from Sigma; NEM, DTT, and "chicken

blood" DNA from Calbiochem; TEMED, gME, and bromophenol blue from

Eastman; acrylamide, BIS, SDS, and ammonium persulfate from Bio-Rad;

and PPO and POPOP from New England Nuclear; Coomassie blue from

Schwarz-Mann; Triton X-100 from Palmetto. All standard chemicals for

media and buffers were obtained from Scientific Products or from

Fisher.








Enzymes and Proteins

Lysozyme, DNase I, and ovalbumin were obtained from Sigma. RNase I

and BSA were purchased from Miles.

Radioisotopes

[ H] Thymidine and [14C] leucine were obtained from Schwarz-Mann;
[14C] mixed amino acids, Na235S04, and carrier-free H235SO4 from New

England Nuclear.

Antisera
Goat anti-serum to guinea pig IgG were purchased from Miles. Anti-

cl repressor prepared from guinea pig was a gift from Dr. Harrison Echols.

Prophage Induction Assay
Lysogens were grown in LB media to an OD590 = 0.2 0.4. After an

inducing treatment, the cells were collected by centrifugation (7,000 x g

for 5 minutes) and resuspended in pre-warmed (37) LB media. Following

2.5 to 3.0 hours at 37, a few drops of chloroform were added to each

sample and incubation was continued for 15 minutes. The lysates were

diluted in TM buffer (10 mM Tris, pH 7.4, 10 mM MgS04) and titered on
Q, or AB1157, using the standard soft-agar overlay method. The indicator

strain was grown overnight in LB media containing 0.2% maltose, and the

cells were collected and resuspended in TM buffer.
Preparation of [3H] Labeled X DNA

Strains W3350 (XcI857 Sam7) and W3350 (Ximm434 clts Sam7) were grown

in K-glucose at 30 to an OD590 = 0.5. The cultures were shifted to 42

for 20 minutes, deoxyadenosine was added to 250 pg/ml, and [3H] thymidine

(52Ci/mmo1e) was added to 10 pC/ml. After 3 hours of incubation at 37,

the cells were collected, and concentrated 100-fold in TM buffer (10 mM

Tris, pH 7.4, 10 mM MgS04). The cells were lysed by the addition of a








few drops of chloroform followed by shaking at 37 for 15 minutes. The

cell debris was moved by a low-speed centrifugation. The [3H] X phage

were purified by centrifugation in a CsCI step gradient. The gradient

consisted of three steps: 1.3 g/cc, 1.5 g/cc, 1.7 g/cc CsCI in 10 mM

Tris pH 7.4, 1 mM MgSO4. Centrifugation was performed in a Beckman

SW-50.1 rotor at 35K rpm for 75 minutes at 15. The phage band was

removed with a syringe through the side of the tube and dialyzed over-

night against TM buffer at 4.

The [3H] labeled phage DNA was extracted as described by Susskind

and Botstein (1975). SDS was added to the dialyzed phage at 0.5%, and

the suspension was heated at 65 for 15 minutes. After chilling the

preparation in ice, KC1 was added to 0.5 M. The suspension was kept on

ice for 2-3 hours, then centrifuged at 10K rpm for 15 minutes at 40.

The supernatant, containing the DNA, was dialyzed extensively against

10 mM Tris pH 7.4, 1 mM EDTA. From a 100 ml lysogenic culture, approx-

imately 250 pg (5 A260 units) of [3H] phage DNA could be obtained with

a specific activity of 1-2 x 105 vg/CPM.

Preparation of Cell Extracts
for cl Repressor Binding Assay

Samples (10 ml) of lysogenic cultures were added to 10 ml of ice-

cold S buffer (0.1 M NaCl, 10 mM EDTA, 10 mM NaN3). The cells were

collected by centrifugation, washed with cold TS buffer (50 mM Tris,

pH 7.4, 10% sucrose), resuspended in 0.1 ml TS buffer, and stored at

-20. The cells were thawed at room temperature, then placed in an

ice-water bath. Five microliters of a solution of lysozyme (4 mg/ml

in 0.25 M Tris, pH 7.4) and 2.5 p1 of 4 M NaCI were added, and the

samples were kept at 0 for 30 minutes. Cell lysis was completed by








a 90 second incubation at 37. The lysates were kept at 0-4 and

centrifuged at 16K rpm for 45 minutes in a Sorvall SS-34 rotor. The

supernatants were removed and assayed for cl repressor binding activity.

DNA Binding Assay for cl Repressor

The assay measures the ability of the cl repressor, which binds

to nitrocellulose filters, to retain [3H] labeled phage DNA (Chadwick

et al., 1970). The protocol used in these studies is a modification

of that described by Susskind and Botstein (1975). Samples to be

assayed were mixed with 230 pl of BA reaction mixture which was prepared

fresh by adding 1 to 1.5 pg of [3H] labeled X or Ximm434 DNA, 250 pg of

sonicated "chicken blood" DNA, and 50 pg of BSA per ml of BA buffer

(10 mM Tris, pH 7.4, 3.5 mM MgCl2, 1 mM EDTA, 50 mM KCI). After 3 to

5 minutes at room temperature, duplicate 0.1 ml samples were filtered

through Millipore HAMK binding assay filters with gentle suction. The

filters were washed with 230 pl of BA buffer, dried, and counted in a

scintillation cocktail containing 11.6 g PPO and 375 mg POPOP per

gallon of toluene.

Figure 2 shows the binding of X and Aimm434 DNA after incubation

with increasing amounts of crude extract prepared from AB1157(X ).

Since Ximm434 is identical to X except for a small region which includes

the operator sites for the cl repressor, the amount of Ximm434 DNA

bound represents nonspecific binding. The purpose of the unlabeled
"chicken blood" DNA in the reaction mixture is to keep the nonspecific

binding at a minimum level.

For the derepression experiments, the nonspecific binding was

subtracted from the amount of X DNA bound to give specific binding due

to the cl repressor. Each point represents the average of two filters.






















60-


C 50 X DNA
0
.40
< 40-
z
0

=30-


20-


10- Xi434DNA



12 3 45 10 20
Amount of repressor extract, /LI


Figure 2. DNA-binding assay for the cI repressor.

Lysogens were grown and extracts were prepared as described
in MATERIALS AND METHODS. The percentage of 3H-labeled
phage DNA bound was plotted as a function of volume of
extract added to the binding assay reaction mixture. The
amount of labeled DNA applied to each filter, that is 100%
of the input DNA, corresponds to 8.1 x 103 CPM.










For clarity, the data is presented as percent of the initial binding

activity. Initial cl repressor binding activity is defined as the

amount of X DNA specifically bound for a fixed volume of extract

prepared from a sample taken immediately after an inducing treatment.

In the experiments described in the RESULTS, this activity corresponds

to 3-7 x 103 CPM of 3H-DNA specifically bound.

SDS-PAGE

Slab gel electrophoresis was performed in a Bio-Rad model 220 ver-

tical slab electrophoresis apparatus using the discontinuous buffer

system of Laemmli (1970). The stacking gel was 4.5% acrylamide and

the separation gel was 10 or 15% acrylamide, as indicated. The ratio

of acrylamide to BIS was 30:0.8. Electrophoresis was for 1-2 hours at

60V (voltage regulated) then at 120V for 3 hours. After electro-

phoresis, the gel was fixed overnight in 40% methanol containing 7%

acetic acid. For staining, the gel was immersed in fixing solution

containing 0.125% Coomassie blue for 2 hours at room temperature. The

gel was destined in 10% methanol containing 7% acetic acid at 37 for

6 to 12 hours. Ovalbumin (43Kd) was used as a molecular weight marker.

Samples for electrophoresis were prepared as described by Little

and Hanawalt (1977). Suspended cells or antibody precipitates in 10 mM

Tris, pH 7.4 with 10% glycerol were mixed with an equal volume in 2X

sample buffer (10% glycerol, 2% SDS, 0.02% bromophenol blue, 10% ME,

0.125 M Tris, pH 6.8) and incubated at 100 for 3 minutes. Samples

(10-50 il) were applied to the wells of the slab gel with a 50 il

Hamilton syringe.








Autoradiography

Gels containing radioactive proteins were treated by the following

procedure. After fixing or destaining, the gel was immersed in 4%

glycerol for at least 2 hours at room temperature. The gel was then

dried down, supported on filter paper backing, under vacuum with heating

using a Bio-Rad Model 224 gel slab dryer. The dried gel was exposed to

Kodak X-Omat XRP-5 x-ray film at -20 or -70 for 3 to 25 days. The

gel and film were clamped flat in a light-tight Kodak x-ray exposure

holder between two glass plates. The exposed x-rays were developed in

a Kodak X-Omat RP Processor at the Shands Teaching Hospital.

Immunoprecipitation Assay for the cl Repressor

The protocol used in these studies is essentially that described

by Roberts and Roberts (1975). Lysogens were grown overnight in CLSM

with 0.2 mM MgSO4. Overnight cultures were diluted 100-fold in CLSM

with 40 pM MgSO4 and grown at 37 to an OD550 = 0.1. [ 35S] Sulfate was

added to 0.2 mCi/ml and the cultures were grown at 37 to an 0D550 = 0.5.

After UV irradiation, 0.5 ml samples were pipetted into 0.5 ml of chilled

non-radioactive AB1157 carrier cells which had been grown previously in

CLSM to an OD550 = 1.0, concentrated 40-fold, and supplemented to 0.01 M

NaN3., The cells were collected by centrifugation and stored at -20.

The frozen cell pellets were thawed in 25 p1 of L buffer (10 mM Tris,

pH 8.1, 5 mM MgCl2, 0.1 mM EDTA, 0.1 mM DTT, 50 mM KCl, 0.6 mg/ml lyso-

zyme). The cells were lysed by five cycles of freeze/thaw in a dry ice/

alcohol bath. DR buffer (50 pl) (10 mM Tris, pH 8.1, 10 mM MgCl2,

0.1 mM EDTA, 0.1 mM DTT, 10 mM KC1, 5% glycerol, 20 pg/ml DNase, and

10 pg/ml RNase A) was added and the lysates were incubated at room

temperature for 30 minutes. Cellular debris was removed by centrifugation







at 15K rpm for 30 minutes at 4 in a Sorvall SS-34 rotor. For all

subsequent steps, the extracts were kept at 0-4.

For antibody precipitation, the extracts were diluted with 0.1 ml

of AB buffer (50 mM Tris pH 7.4, 2.0 M KC1, and 2% Triton X-100). To

each sample, 1 pl of guinea pig anti-cl repressor serum was added. The

samples were kept on ice for 22-24 hours, then centrifuged at 15K rpm

for 20 minutes. According to Roberts et al. (1977), this centrifugation

step reduces non-specific precipitation and improves the background. To

precipitate the soluble repressor-antibody complex, 20 p1 of goat anti-

serum to guinea pig IgG was added. After standing overnight at 0, the

samples were centrifuged at 15K rpm for 30 minutes. The pellets were

washed 3 times with 0.5 ml of 50 mM Tris (pH 7.4), 1.2 M KCI, 1.2%

Triton X-100 and once with 0.5 ml of 50 mM Tris (pH 7.4), 0.1 M NaCl.

The pellets were then prepared for SDS-PAGE.












RESULTS

Studies on the Role of DNA Degradation
in X Derepression

Since a functional recB gene is required for RecA protein induction

in NA treated cells (Gudas and Pardee, 1975) and since induction of the

RecA protein seems to be required for X derepression (see INTRODUCTION,

p. 9), one might predict that NA-mediated X derepression would be

recB-dependent. This prediction is confirmed by the data in Figure 3

which shows that NA treatment does not result in induction of X prophage

in lysogenic strains carrying the recB21 mutation.

The evidence of Little and Hanawalt (1977), which suggested that

DNA degradation may not be involved in RecA protein induction, put in

doubt the idea of Gudas and Pardee (1975) that the role of recB in

NA-induction is to provide DNA breakdown products which trigger the

induction response. In an effort to resolve this issue, I sought to

re-examine the role of the recB gene product in NA-mediated X induction.

As the RecBC protein has an ATPase activity in addition to both

endo- and exonuclease activities (Goldmark and Linn, 1972), the missing

function for NA-induction of the RecA protein and X prophage could be

some function other than that of degrading DNA. I therefore tried to

ascertain whether X exonuclease, an exclusively exonucleolytic enzyme

(Little, 1967), could circumvent the recB21 mutation in NA-induction

of the RecA protein. To do this I determined if the RecA protein was

induced in recB21 lysogens carrying XcIind-857 prophage that were


























Figure 3. Effect of recB21 on NA-induction of X.

Lysogens were grown as described in MATERIALS AND
METHODS. Small volumes of a 10 mg/ml solution of
NA in 0.1 N NaOH were added to 1 ml samples of
lysogenic culture in LB media. After a 60 minute
incubation at 370, the cells were collected by
centrifugation and resuspended in fresh prewarmed
(37) LB media. The samples were incubated at 37
for 2.5 hours and production of free phage was
determined as described in MATERIALS AND METHODS.
(A) AB1886 (X+) uvrA6, (0) AB2470 (X+) recB21,
(0) RC100 (XW) uvrA6 recB21.


















7
10 .o.. ...






6
10-


u- 10
5- 0


o5-








5 10 50


Nalidixic Acid Concentration, u.g/ml








transiently heat-induced to produce X exonuclease prior to NA treat-

ment. This approach was prompted by similar experiments by others who

showed that transient prophage induction enhances recB-dependent cell

survival after UV or x-ray treatment (Braun and Gluck, 1977; Trgovcevic

and Rupp, 1974).

Figure 4 shows the effect of transient heat induction of Xc1857

prophage on RecA protein production in NA-treated recB21 lysogens.

Lanes 8-12 show that the amount of RecA protein increased with increasing

incubation time at 42. In control experiments, shown in lanes 3 and 4,

neither a 5 minute incubation at 42 nor NA treatment alone was

sufficient to induce the RecA protein. To investigate the role of X

exonuclease in this induction, the above experiments were repeated

using recB21 lysogens carrying prophage with mutations in the genes

controlling X exonuclease activity. The combined treatment of heat

induction and exposure to NA was not sufficient to induce the RecA

protein when the prophage carried mutations redX314 (lane 6) and redBll3

(lane 7). RedX314 and redBll3 are point mutations in the genes that

code for X exonuclease and 0 protein, respectively (Radding, 1970;

Shulman et al., 1970). X exonuclease acts at 5' phosphoryl termini on

double-stranded DNA molecules, and the only acid soluble products are

deoxyribonucleoside 5'-monophosphates (Little, 1967). The redX314

mutation decreases the activity of X exonuclease to less than 2% of

normal activity (Radding, 1970). Although the redBll3 mutation does not

affect the X exonuclease enzyme, it does alter the antigenic properties

of the B protein (Shulman et al., 1970). The only known activity of

the protein is to enhance the affinity of X exonuclease for DNA

in vitro (Radding and Carter, 1971). The foregoing evidence indicates















Figure 4. Effect of X exonuclease expression on recB-dependent
induction of the RecA protein.
Lysogens were grown at 30 to an 0D590 = 0.2 in M9-glucose
media containing required amino acids and thiamine. Samples
(1 ml) were incubated at 42 for various times before addition
of NA (50 pg/ml). At 30 minutes after addition of NA, 14C-
leucine was added to 5 iCi/ml and incubation was continued
for 60 minutes at 30. The cells were collected by centrifu-
gation and prepared for SDS-PAGE. This figure is a composite
picture of autoradiographs of two 10% slab gels. The RecA
protein is denoted as X. All wells except 6 and 7 show pro-
teins synthesized in strain AB2470 recB21 (Xc1857), unless
otherwise noted, that were treated as described below. 2
(1) untreated control; (2) UV irradiation with 300 ergs/mm;
(3) incubated at 42 for 5 minutes; (4) exposed to NA;
(5) incubated at 42 for 5 minutes followed by exposure to
NA; (6) AB2470 (Xc1857 redX314) treated as in well 5; (7)
AB2470 (XcI1857 redBll3) treated as in well 5; (8) incubated
at 42 for 1 minute, (9) 2 minutes, (10) 3 minutes, (11)
4 minutes, and (12) 5 minutes followed by exposure to NA.













-. -- - -^> '* -



xj t43K-U I U iIIII

---- -' "''''' ~ "'''> ^ w *"? %" g ''"?
w- 0 W W *
S2 3 4 5 6 7 8 9 10 12
I 3 4 5 6 7 8 9 10 11 12







that the activity of X exonuclease is responsible for the induction of

the RecA protein in NA treated recB21 lysogens. This means that DNA

degradation is necessary for NA-mediated induction of the RecA protein.

By inference, NA-mediated derepression of X prophage also requires DNA

degradation.

Although the above evidence indicates that recB-dependent DNA

degradation is required for NA induction of the RecA protein and X pro-

phage, this dependence may not apply to induction by other inducing

agents. For instance, lane 2 of Figure 4 shows that UV-mediated induc-

tion of the RecA protein is independent of the recB21 mutation, confirming

the results of Little and Hanawalt (1977). UV-induction of X, as one

might expect, is also recB-independent (Brooks and Clark, 1967).

It is possible that more than one pathway of DNA degradation can

result in induction. For instance, excision repair or post-replication

repair may result in sufficient DNA breakdown, which does not require

the RecBC enzyme, to provoke the induction response. If one assumes

this to be the case, one might predict that eliminating these alternate

pathways of degradation genetically would block induction of the RecA

protein and X prophage. To test this, I examined the effects of muta-

tions in genes responsible for various modes of DNA degradation on

induction of the RecA protein and X prophage.

Figure 5 shows that MC-induction of the RecA protein is blocked by

the uvrA6 mutation. This experiment was prompted by the finding that

MC-induced mutagenesis requires excision repair (Kondo et al., 1970).

Figure 6 shows that the uvrA6 mutation inhibits, but does not abolish,

MC-mediated X prophage induction. Figure 6 also shows that, although the

recB21 mutation alone has no inhibitory effect on MC-induction of X, the























Figure 5. Effect of uvrA6 on MC-induction of the
RecA protein.

Bacterial cultures were grown at 37 to an
0D590 = 0.2 in M9-glucose media containing
required amino acids and thiamine. Samples
(1 ml) were treated with MC (1 jig/ml) or
with UV (130 ergs/mm2). At 15 minutes fol-
lowing an inducing treatment, 14C-mixed
amino acids were added to 1 pCi/ml, and
incubation was continued at 37 for 30
minutes. Cells were collected and prepared
for SDS-PAGE. This is a picture of an auto-
radiograph of a 10% gel. The band denoted
"X" is the RecA protein. (1) RC100 uvrA6
recB21, untreated; (2) RC100, UV irradiated;
-T RC100, exposed to MC; (4) AB2470 recB21,
exposed to MC; (5) AB1886 uvrA6, exposed
to MC.













Iw -g !p fi -.-


U.--.a..f:
Am*


our*
mw
IF


b 4


2 3

























Figure 6. Effect of uvrA6 on recB21 on MC-induction of X.

Lysogens were grown as described in MATERIALS AND
METHODS. Small volumes of an aqueous solution of
MC (1 mg/ml) were added to 1 ml samples of lyso-
genic culture in LB media. After a 30 minute
incubation at 37, the cells were collected and
resuspended in prewarmed LB media. The samples
were incubated at 37 for 3 hours and the pro-
duction of free phage was determined as described
in MATERIALS AND METHODS. (A) AB1157 (X+);
(0) AB2470 (X+) recB21; (A) AB1886 (X+) uvrA6;
(0) RC100 (X+) uvrA6 recB21.
















8
108









I0
E






I0
I.-





1d0-



io5

105


Mitomycin C concentration, /.ug/ml


0.050.1 0.2 0.5 1.0














Table 3. Effect of MC on UV and thermal induction of X.


Treatment PFU/ml


RCl00 (X+)

7.2 x 105

1.4 x 106

4.4 x 108



1.8 x 108


RC100 (kc1857)

1.0 x 102

1.0 x 102


3.1 x 109



1.2 x 109


Lysogens were grown in LB media to an 0D590 = 0.2 0.4. MC
was added to 1 ig/ml. UV irradiation was at a dose of 130 ergs/mm2.
For thermal induction, the XcI857 lysogen, that had been grown at
30, was incubated at 42 for 20 minutes. The free phage produced
after 3 hours at 37 were titered as described in MATERIALS AND
METHODS.


Control


UV + MC

42 + MC








combination of recB21 with the uvrA6 mutation essentially eliminates

the production of progeny phage in MC-treated lysogens. Table 3 shows

that UV-induction of X and thermal induction of Xc1857 occur normally

in the recB21 uvrA6 strain, even in the presence of MC. Thus, the

effect of the two mutations is on derepression and not on the replica-

tion or maturation of the progeny phage in MC treated lysogens. The

above evidence is consistent with the suggestion that more than one

pathway of DNA degradation leads to induction. In the case of NA,

recB-dependent degradation at a stalled replication fork may lead to

induction. In the case of MC, DNA degradation due to excision repair

of MC lesions or RecBC exonuclease activity at replication forks which

are stalled at MC lesions may result in induction.

What pathway(s) of DNA degradation are responsible for UV-induction?

Braun (1976) showed that UV-induction of X prophage requires either

excision repair or replication of a damaged DNA template. He found that

when DNA replication is allowed to terminate prior to irradiation in a

dnaA- uvrB- lysogen, UV-induction of X does not occur.

It is possible that exonucleolytic activity at gaps in DNA, produced

by replicating past pyrimidine dimers, is responsible for triggering

X induction following UV irradiation. Kornberg (1974) suggested that

exonuclease III may act at such gaps. To determine if this enzyme plays

a role in UV-mediated X induction, I examined the effect of a deletion

mutation in xth (the gene that codes for exonuclease III) on X induction.

As indicated in Table 4, this mutation has no effect on UV-induction

of A prophage, even when combined with the uvrA6 and recB21 mutations.

Thus my attempts to block UV induction of X by genetically eliminating

various pathways of DNA degradation were unsuccessful.











Table 4. Effect of A(xth-pncA) on UV induction of X.


UV dose
ergs/mmn


AB1886 uvrA6 (X+)

2.9 x 105

5.4 x 107

1.3 x 108


PFU/ml

RC9137 uvrA6 A(xth-pncA)(X+)

4.6 x 105

2.3 x 108

2.2 x 108


RC109 uvrA6 recB21 A(xth-pncA)(X+)

4.5 x 105

2.8 x 108

3.6 x 108


Lysogens were grown in LB media to an 0D590 = 0.2-0.4. The cells were collected, resuspended
in PBS, and UV irradiated. Following the inducing treatment, the cells were collected again and
resuspended in prewarmed LB media. The free phage produced after 3 hours at 37 were titered as
described in MATERIALS AND METHODS.








Studies on Xclind5 Prophage Derepression

To examine further the role of RecA-mediated proteolytic cleavage

in the mechanism of X prophage derepression, I utilized the phage

repressor mutation, clind5 (Horiuchi and Inokuchi, 1967). The Xclind5

prophage is induced by relatively low doses of UV radiation (Tomizawa

and Ogawa, 1967). Noack and Klaus (1972) suggested that the clind5

repressor may be ultrasensitive to inactivation by the hypothetical

effector. An alternative possibility is that the mutant repressor might

be ultrasensitive to RecA-mediated proteolytic cleavage. The following

experiments were designed to distinguish between these two possible

explanations for the phenotype of Xclind5.

Derepression of Xclind5 prophage was studied using the nitro-

cellulose membrane filter binding assay to monitor the operator DNA-

binding activity of the cl repressor in induced lysogens. Figure 7

shows a comparison of clinds and cI+ repressor binding activities.

Three to four times more extract of Xclind5 lysogenic cells is needed

to obtain maximum repressor-operator binding activity, compared to

that of XcI+ lysogens. On the basis of in vivo evidence, the clind5

repressor appears to bind normally to its operator sites in the cell.

For instance, the frequency of lysogenization of Xclinds is the same

as that of XcI+ (Blanco and Pomes, 1977). Moreover, Alind5 lysogens

exhibit normal immunity to superinfection, and the spontaneous frequency

of induction of the Xclind5 prophage is essentially the same as that of

the XcI I+ prophage (Tomizawa and Ogawa, 1967). A possible reason for

the altered in vitro binding of clinds repressor will be discussed

later.





















Figure 7. Comparison of cl and clinds repressor
binding activities.

Lysogens were grown at 37 to an 0D590 = 0.8
in K-glucose. Cells were collected, extracts
were prepared, and cl repressor binding assays
were performed as described in MATERIALS AND
METHODS. The nonspecific binding to Ximm434
DNA (see Fig. 2),which was the same for both
repressors, has not been subtracted.














AB1157(X+)


ABI 157( Xinds)


5 10 15 20


Volume of cell extract, /ul


50

40

30

20

10








For the repressor inactivation experiments, the description of

which follows, 5 il of cl+ repressor extract and 20 pil of clinds repressor

extract were taken to monitor binding activity so that comparable initial

binding activities would be obtained.

Figure 8 shows that UV-mediated inactivation of cIinds repressor

binding activity occurs at much lower doses of UV radiation than those

required to cause cl + repressor inactivation. These results corroborate

the findings of Tomizawa and Ogawa (1967) that the immunity to super-

infection specified by the cI gene is lost in Aclinds lysogens at

approximately 10% of the UV dose required for loss of immunity in cI+

lysogens. Figure 9 shows that, at UV doses sufficient to give maximum

derepression, clind5 repressor is completely inactivated 10 minutes

following irradiation, whereas 30 minutes are required for maximum

inactivation of cl + repressor.

The 10 minute lag period observed for XcI+ derepression is thought

to reflect the requirement for de novo protein synthesis (Shinagawa

and Itoh, 1973), that is, induction of the RecA protein. To ascertain

whether protein synthesis is required for Xclind5 derepression, I

determined the effect of protein synthesis inhibitors on UV-mediated

cIind5 repressor inactivation. CAP and neomycin have been reported to

inhibit XcI+ derepression (Shinagawa and Itoh, 1973; Maenhaut-Michel

et al., 1978). Table 5 shows that these two inhibitors, both of which

inhibit cI + repressor inactivation, have no effect on the inactivation

of cIinds repressor.

The above results suggest that, in contrast to XcI+ derepression,

XcIinds derepression does not require de novo synthesis of the RecA

protein. One possible explanation for these results is that the clinds







































20 50 100


0 AB1157(X7)
*ABI 157(Xinds)














-e
400


0

0

4-
0l0
4100
0

. 80
>
.5
0 60
o-

S40
1_
0


C.
ffl 2
t_


-2
UV dose, ergs mm


UV dose-response for prophage derepression.


Lysogens were grown at 37 to an 0D590 = 0.8, chilled,
and UV irradiated with increasing doses. For each
dose, 10 ml samples were incubated at 37 for 30
minutes (X/cI+ lysogen) or 10 minutes (Xclind5 lysogen).
Cell extracts were prepared for each sample and assayed
for cI repressor binding activity as described in
MATERIALS AND METHODS. The data is presented as a
percentage of the unirradiated control versus UV dose.
One hundred percent corresponds to 9.8 x lO3 CPM of
[3H]-XDNA specifically bound, in the case of cl
repressor, and 4.5 x 103 CPM for clinds repressor.


200


Figure 8.


















100

80

60

40

20


ABII57 Xi+)
ABI157 (Xinds)


10 20 30


Time at 37C,min


Figure 9. Kinetics of UV-mediated prophage derepression.

Lysogens were grown at 37 to an 0D590 = 0.8 in
K-glucose, chilled, and UV irradiated with
400 ergs/mm2 (XcI+ lysogen) or 65 ergs/mm2
(Ac inds lysogen). During the post-irradiation
incubation at 37, samples were removed, processed,
and assayed for cl repressor binding activity as
described in MATERIALS AND METHODS.














Table 5. Effect of protein synthesis
inhibitors on prophage derepression.


Treatment


% of initial cI repressor


AB1157 (XcI+)


binding activity

AB1157 (Xclinds)


Control


UV + Chloramphenicol


UV + Neomycin


The
Figure 9.
10 minutes


experiments were done as described in the legend to
CAP (100 jig/ml) or neomycin (200 pg/ml) was added
prior to irradiation.








repressor may be inactivated by a mechanism that is completely independ-

ent of the "normal" recA/lexA pathway. To test this, I determined the

effects of mutations in recA and lexA on UV-mediated clinds repressor

inactivation. The data in Figure 10 show that the lexA-1 mutation,

which inhibits XcI+ derepression, does not block inactivation of clinds

repressor. As the lexA-1 mutation blocks RecA protein induction (Gudas

and Pardee, 1975), this finding confirms the suggestion that Xclinds

derepression does not require induction of the RecA protein. However,

Figure 10 also shows that UV-mediated inactivation of the clinds repressor

is abolished by the recAl3 mutation. This evidence indicates that,

although induction of the RecA protein is not necessary for Xclind5

derepression, a functional recA gene is required. This means that the

low, uninduced levels of functional RecA protein are essential to

inactivate clind5 repressor in UV irradiated lysogens.

As further support for the indication that the RecA protein

mediates Xclinds derepression, Figure 11 shows that expression of the

tif-l mutation results in the inactivation of clind5 repressor. The

kinetics of tif-mediated Xclinds derepression are comparable to those

of UV-mediated derepression. Figure 11 also shows that tif-mediated

clinds repressor inactivation occurs in the presence of neomycin,

indicating that de novo protein synthesis is not required.

The fact that XclIind5 lysogenizes E. coli normally (Blanco and

Pomes, 1977) indicates that uninduced levels of the RecA protein are

not sufficient to inactivate the clinds repressor. In the case of

tif-mediated derepression, an increase in temperature is necessary.

In the case of other inducing treatments, perhaps a product of DNA

degradation triggers a RecA-mediated inactivation of the repressor.












120- 0 ABII57(X)
ABI157(Xinds)
100- C AB2463 recAI3(X")
I I* AB2463recAI3(Xinds)
o \ AB24941exA-I (X*)
, 80- A AB2494 lexA-I (Xinds)

60-

S40-


I0 2005

10 20 30
Time at 37, min.
Figure 10. Effects of recAl3 and lexA-1 on prophage derepression.
The experiments were done as described in the legend to Figure 9.


















o JMIZ2(X)
* JM 12(Xinds)




^~0


10 20 30 40


Time at 42C, min.


Figure 11.


tif-mediated prophage derepression.


Lysogenic derivatives of strain JM12, which carries
the tif-l allele, were grown in K-glucose at 30 to
an OD590 = 0.8. At zero time, adenine was added to
100 v-g/ml and the culture was shifted to 42. At
various times samples were removed, processed,
and assayed for cl repressor binding activity as
described in MATERIALS AND METHODS. The dashed
lines represent the effect of neomycin (200 pg/ml)
which was added 30 minutes prior to the temperature
shift.







In support of this idea, Figure 12 shows that NA-mediated inactivation

of clinds repressor requires a functional recB gene. The evidence

presented in the previous section of the RESULTS suggests that this

reflects a requirement for recB-dependent DNA degradation.

If the idea were correct that DNA degradation products convert

normal RecA protein to an "activated" protease (Gudas and Mount, 1977),

one explanation for the foregoing results is that the clinds repressor

may be sensitive to proteolytic cleavage mediated by low concentrations

of the activated RecA protein. To test this, I determined the effect

of certain protease inhibitors on UV-mediated Xclinds derepression.

Several inhibitors have been reported to block XcI + derepression,

including antipain (Meyn et al., 1977), TPCK (Sedgewick et al., 1978),

TLCK (Radman et al., 1977), and NEM (Podsakoff et al., 1976). The data

in Table 6 show that, in contrast to cI+ repressor inactivation, neither

TLCK nor NEM inhibit inactivation of clinds repressor. This suggests

that proteolytic cleavage may not be required for Xclind5 derepression.

To determine directly the role of proteolytic cleavage in Aclinds

derepression, I utilized the immunoprecipitation assay of Roberts and

Roberts (1975) to follow the fate of [35S]-labeled clinds repressor
2
in UV irradiated lysogens. Irradiation at a dose of 20 ergs/mm ,

followed by a 10 minute incubation at 37, results in the complete

disappearance of the operator-binding activity of the clinds repressor

(see Fig. 8). The evidence presented in Figure 13 demonstrates that

following this inducing treatment, the clind5 protein remains essen-

tially intact. Thus RecA-mediated proteolytic cleavage cannot account

for the loss of the binding activity of the mutant repressor. At a
UV dose of 400 ergs/m2, which is sufficient to inactivate normal +
UV dose of 400 ergs/mm which is sufficient to inactivate normal cl






















Figure 12.


NA-mediated Xclinds derepression: effect of recB21.

Lysogens were grown at 37 to an 0D590 = 0.8 in K-glucose.
NA was added to 50 pg/ml at zero time and samples were
taken, processed, and assayed as described in MATERIALS
AND METHODS.




















* ABI157(Xinds)
A AB2470 recB21 (Xinds)


10 20 30
Time at 37C, min


I00

80-

60

40

20














Table 6. Effect of protease inhibitors
on prophage derepression.


% of initial cl repressor binding activity


ABI157 (Xci+)


AB1157 (Xclind5)


The experiments were done as described in the legend to
Figure 9. NEM was added to 10-4 M; TLCK was added to 10-3 M.


Treatment


Control


UV + NEM

UV + TLCK



















Figure 13. Proteolytic cleavage of clinds repressor
in UV irradiated lysogens.

Strain AB1157 (Xind s) was grown and
labeled with 0.2 mCi/ml H p5S04 as
described in MATERIALS AND METHODS. After
a 5 minute chase period with 7 mM Na2SO4,
the unirradiated control sample (1) was
taken and the remaining culture was treated
as described below. UV irradiated with
20 ergs/mm2 followed by incubation at 37
for 5 minutes (2), 10 minutes (3); UV
irradiated with 400 ergs/mm2 followed by
incubation at 37 for 5 minutes (4), 10
minutes (4), 15 minutes (6). Samples
(0.5 ml) were taken, processed, and pre-
pared for SDS-PAGE as described in
MATERIALS AND METHODS. This figure shows
the autoradiograph of a 15% polyacrylamide
slab gel. The 27,000 dalton cl polypeptide
is denoted as R. R' denotes the position
of the cleavage products (see Roberts and
Roberts, 1975).
























1 2 3


-v-- .~--


4 5 6











-- -


I
.|


.1 :

., -';


:*..* ..
' o -'


'U-^


'..*,


IN. ...




54



repressor (Fig. 8), proteolytic cleavage of the clind5 repressor is

observed. This cleavage follows faster kinetics than those of cl+

repressor cleavage (Roberts and Roberts, 1975). These results

indicate that, although the clind5 repressor is susceptible to RecA-

mediated proteolytic cleavage, this process is not essential for

Xclinds prophage derepression.












DISCUSSION

I have shown that, in NA treated lysogenic cells, which lack a

functional RecBC exonuclease, expression of X genes controlling X exo-

nuclease is necessary for the induction of the RecA protein. In

contrast to the RecBC enzyme, X exonuclease is an exclusively exonucleo-

lytic enzyme (Goldmark and Linn, 1972; Little, 1967). Since X exo-

nuclease-mediated DNA degradation provides the missing function of the

RecBC enzyme in cells carrying a mutation in the recB gene, I conclude

that DNA degradation in NA treated cells, normally mediated by the RecBC

enzyme, is essential for the induction of the RecA protein. As induction

of the RecA protein is requisite for X derepression, then by inference,

DNA degradation is necessary for NA-mediated A derepression.

In the case of inducing agents other than NA, the role of DNA

degradation in the induction process is not as clear. For instance,

UV and MC induction of the RecA protein and X prophage are recB-

independent. I have found that induction of the RecA protein in MC

treated cells requires excision repair; that is, the uvrA6 mutation

blocks MC-induction of the RecA protein (Figure 5). DNA degradation

following MC treatment is greatly reduced in cells carrying the uvrA6

allele compared to normal cells (Boyce and Howard-Flanders, 1964), thus

this is correlative evidence that DNA degradation may be involved in

MC induction of the RecA protein. Surprisingly, the uvrA6 mutation

does not abolish MC induction of A prophage; however, the combination

of the uvrA6 mutation with the recB21 mutation essentially eliminates








X induction in MC treated lysogens (Fig. 6). This evidence suggests

that more than one pathway of DNA degradation may lead to the induction

response.

If DNA degradation is the key event responsible for triggering

induction, two possibilities can be envisioned. As suggested by Gudas

and Pardee (1975), low-molecular weight DNA breakdown products may

serve as effectors in the induction mechanism. Alternatively, DNA

containing single-stranded gap regions, resulting from DNA degradation,

may be the key signal that provokes the induction response. Existing

evidence can be cited which supports each of these two possibilities.

In the former case, the finding that tif-mediated induction occurs

without any apparent alteration in DNA structure and is enhanced by high

concentrations of adenine (Kirby et al., 1967) provides a strong basis

for arguing against the idea that gapped DNA is important and supports

the idea of a low-molecular weight effector. Further evidence support-

ing the "small-molecule" hypothesis was provided by Villani et al.

(1978). These investigators found that inhibition of in vitro DNA

synthesis of UV irradiated 0X174 DNA results in an increased turnover

of nucleoside 5'-triphosphates to nucleoside 5'-monophosphates. They

determined that this turnover is due to the "proofreading" 3'-5' exo-

nuclease activity of the bacterial polymerases when these enzymes

encounter mismatched bases opposite pyrimidine dimers. In other words,

the polymerase activity presumably incorporates a nucleotide opposite

a dimer, which is subsequently removed by the exonuclease activity since

the base is not properly paired with a complementary base, and the con-

tinuation of these events results in what is called "idling." These

authors proposed that this "idling" reaction is responsible for the







accumulation of the hypothetical small molecule effector that leads

to the expression of the UV-inducible responses. If this idea is

correct, the idling reaction could account for my inability to eliminate

UV induction of X by genetically blocking other modes of DNA degrada-

tion in the strain RCl09A(xth-pncA) uvrA6 recB21 (Table 4).

If a nucleoside 5'-monophosphate produced by idling or by other

modes of DNA degradation does serve as an inducing signal, how might

this compound be distinguished from normal DNA precursors in the cell?

I propose that one possibility is that a modified (perhaps methylated)

DNA breakdown product serves as a specific chemical signal which triggers

induction. Since modification is a post-replication process, a modified

nucleotide could unequivocally be discerned as a product of DNA degrada-

tion. This idea would explain the results of Little and Hanawalt (1977)

who showed that degradation of unmodified phage DNA is not sufficient

to cause induction. However, no evidence to support this idea has yet

been obtained.

The possibility that discontinuities in DNA may be involved in the

mechanism of X derepression was first proposed by George and Devoret

(1971). This suggestion was based on their work on indirect prophage

induction. X prophage can be induced indirectly by mating unirradiated

F- lysogens with UV irradiated cells containing the F' plasmid (see

Borek and Ryan, 1973). George and Devoret (1971) showed that the

uvrA6 mutation in the nonlysogenic donor strain enhances indirect

induction, whereas the uvrA6 mutation in the recipient lysogen has no

effect. The UV damage which mediates indirect induction can be photo-

reactivated (Howard-Flanders et al., 1968), indicating that pyrimidine

dimers are responsible for the response. The fact that the uvrA6








mutation in the recipient has no effect on indirect induction indicates

that the dimers in the transferred DNA are not repaired by the excision

system. Rupp and Ihler (1968) showed that the F' plasmid is transferred

as a single-stranded DNA molecule which is then replicated in the

recipient cell. If this DNA contains dimers which are not repaired,

replication of the damaged template results in discontinuities or

single-stranded gaps, as suggested by Rupp and Howard-Flanders (1968).

George and Devoret (1971) proposed that UV-induction of X is the conse-

quence of the presence of these gaps in DNA.

The strongest evidence that indicates that gapped DNA, rather

than a low-molecular weight effector, is the inducing signal which

triggers X prophage derepression was provided by Sussman et al. (1978).

These investigators demonstrated that purified cI+ repressor binds to

single-stranded gapped DNA (prepared in vitro), but not to nicked or to

completely single-stranded DNA. Moreover, they determined that gapped

nonoperator DNA competes with operator DNA for binding to the cI

repressor. They showed that the repressor produced by the noninducible

XcIind prophage binds to gapped DNA with a 10-fold decreased affinity.

Based on this evidence, Sussman et al. (1978) proposed that cl repressor

binding to DNA repair intermediates, resembling the gapped DNA used in

their in vitro experiments, constitutes the primary derepression event.

These authors assign a secondary role of proteolytic cleavage which,

in their model, occurs as a consequence of the cl repressor binding to

gaps in the DNA.

In the experiments described in this dissertation, I have utilized

the phage mutation, cIind5, to examine further the role of proteolytic

cleavage in X derepression. I have shown that inactivation of the








operator binding activity of clind5 repressor in UV irradiated lysogens

requires neither induction of the RecA protein nor RecA-mediated proteo-

lytic cleavage. However, derepression of XcIind does require a

functional recA gene. The results of my experiments indicate that

uninduced levels of the RecA protein are essential for XcIind5 derepres-

sion, and that, in this case, the RecA protein functions in some capacity

other than as a protease in the derepression mechanism.

Since only purified RecA protein, ATP, and Mg++ are required to

promote cleavage of the cI+ repressor in vitro, Roberts et al. (1978)

concluded that proteolytic cleavage alone is responsible for X derepres-

sion. The evidence that protease inhibitors block X induction (Meyn

et al., 1977; Radman et al., 1977) suggests that proteolytic cleavage

is necessary for XcI+ derepression.

The two contending models of Roberts et al. (1975) and of Sussman

et al. (1978) may not be mutually exclusive and can be accomodated by

a model in which the cl + repressor is inactivated by a two-step mechanism,

as suggested by Shinagawa et al. (1977). These authors found that

proteolytic cleavage of the cI+ repressor in y-irradiated lysogens is

inhibited by CAP, even though the loss of cI repressor binding activity

in y-irradiated lysogens is insensitive to CAP. They proposed that the

first step of the derepression mechanism involves cI repressor binding

to DNA repair intermediates and is independent of protein synthesis;

the second step is proteolytic cleavage which does require de novo

synthesis of high levels of the RecA protein. My results are consistent

with this two-step model, assuming that the clinds mutation circumvents

the second step. In other words, I propose that mutant clinds repressor

has an increased binding affinity for gapped DNA, and that this binding








effectively removes the repressor from its operator sites. Thus,

induction of the RecA protein and RecA-mediated proteolytic cleavage

would not be necessary for XcIinds derepression.

As the RecA protein is known both to bind to single-stranded DNA,

(Gudas and Pardee, 1976) and to possess an endopeptidase activity

(Roberts et al., 1978), its role in normal AcI+ prophage derepression

may be two-fold: to protect the feature of the gapped DNA structure

enabling the repressor to bind at these sites, as suggested by

Sussman et al. (1978), and to catalyze the specific cleavage of the

bound repressor. In the case of XcIinds derepression, a functional

recA gene product may be needed only to stabilize gaps in the DNA. This

proposed role for the RecA protein in X derepression is consistent with

the fact that the RecA protein controls DNA degradation in irradiated

cells (Satta et al., 1979).

Assuming that the clinds repressor does indeed have a higher affin-

ity for gapped DNA than cI + repressor, the altered in vitro binding of

the clinds repressor (see Fig. 7) could be explained in the following

way. In the preparation of the extracts used for the binding assay (see

MATERIALS AND METHODS), no effort was taken to remove bacterial DNA

other than a 27,000 x g centrifugation, which would remove most but not

all of the DNA. Thus DNA fragments in the extract which would be

susceptible to degradation by endogenous nucleases could provide binding

sites for the mutant repressor. Moreover, the binding assay reaction

mixture contains a high concentration of sonicated "chicken blood" DNA

which would also be susceptible to limited breakdown by any nucleases

present in the added repressor extract. These nonoperator DNA species,

which may contain single-stranded gap regions could compete with labeled







operator DNA for clinds repressor binding, and to a lesser degree for
cl+ repressor binding. Therefore, the specific activity of the clinds

repressor would appear to be lower than that of cl + repressor. Further

in vitro experiments are needed to confirm this hypothesis.

The results of my experiments which indicate that DNA degradation

is necessary for X derepression (at least in the case of some inducing

agents) can also be integrated into the model discussed above. In the

case of NA-induction, the RecBC exonuclease may function by enlarging

discontinuities in newly replicated DNA at stalled replication forks,

creating gap structures* to which the cI+ repressor binds. In the

case of MC-induction, it is conceivable that either excision repair of

MC lesions in the DNA or RecBC exonuclease acting at stalled replication

forks produces the gaps. In the case of UV-induction, excision repair

of pyrimidine dimers or replication past these lesions might result in

gap structures to which the repressor binds.

Although the "gap" hypothesis seems to account for most of the

evidence presented in this dissertation, it falls short of explaining

tif-mediated prophage derepression. Inactivation of the clind5 repressor

occurs very rapidly in tif lysogens following incubation at 42 (Fig. 11),

even though expression of the tif mutation does not result in any

apparent alteration in DNA structure or DNA synthesis (Kirby et al.,

1967). Baluch and Sussman (1978) proposed that the altered RecA protein

in tif mutants has a higher affinity for single-stranded DNA at elevated



*For want of a better term, "gap structures" refers to single-stranded
regions of DNA of the size reported by Sussman et al. (1978) which
effectively compete for cI repressor binding.








temperatures and binds to transient gaps in the bacterial DNA which

occur during normal recombination and replication processes. They

suggest that these structures may not be detected by alkaline gradient

analysis or nuclease digestion, but would serve as binding sites for

the cl repressor. Although this is an attractive idea, no evidence

has yet been reported that supports or refutes this suggestion.

As X prophage induction is one of a number of UV-inducible (SOS)

responses in E. coli, the "gap" hypothesis can be extended to suggest

that the formation of single-stranded gaps in DNA is the key event

responsible for inducing all of the SOS functions (see INTRODUCTION,

p. 6). Specifically, I suggest that the lexA gene product, which is

thought to control the SOS responses through its regulation of the

recA gene as a repressor, binds to gapped DNA. I further suggest

that the recA gene is not derepressed by a proteolytic cleavage

mechanism, but rather that this gap binding effectively removes the

LexA repressor from its operator site. This would explain the results

of Sedgewick et al. (1978) who showed that induction of the RecA

protein is not inhibited by the protease inhibitor TPCK. In other

words, I propose that the LexA repressor is inactivated by the same

mechanism I propose for clinds derepression, i.e. that the Lex repressor

has a high affinity for gapped DNA analogous to the clinds repressor.

The mechanism of X prophage derepression proposed in this disser-

tation leads to several questions, including the following:

1) Does purified clinds repressor have a higher affinity for

gapped DNA compared to cl+ repressor?

2) Does the LexA repressor bind to gapped DNA with an affinity

comparable to clind5 repressor?








3) Does single-stranded gapped DNA enhance RecA mediated proteo-

lytic cleavage of the cl+ repressor in vitro?

4) What structural aspects of gapped DNA are important for cl

repressor binding?

5) In what oligomeric form does the repressor bind to gaps?

6) What is the amino acid sequence of the gap binding site of

the cI protein, how does this binding site relate to the region of the

cI protein that is essential for operator binding, and where is this

gap binding site with respect to the site of cleavage? (see Pabo et al.,

1979).

These and other questions can now be approached biochemically and

their answers should provide interesting insight into this problem.












BIBLIOGRAPHY

Bachmann, B., Low, K. B., and Taylor, A. L. (1976) Bact. Rev.
40, 116-167.

Baluch, J. and Sussman, R. (1978) J. Virology 26, 595-602.

Blanco, M. and Pomes, L. (1977) Molec. Gen. Genet. 154, 287-292.

Borek, E. and Ryan, A. (1973) Prog. Nucleic Acids Res. Mol. Biol.
13, 249-300.

Boyce, R. P. and Howard-Flanders, P. (1964) Z. Vererbungsl.
95, 345-350.

Braun, A. G. (1976) Nature 261, 164-166.

Braun, A. and Gluck, D. (1977) J. Bact. 131, 208-213.

Brooks, K. and Clark, A. J. (1967) J. Virology 1, 283-293.

Castellazzi, M., George, J., and Buttin, G. (1972) Molec. Gen.
Genet. 119, 139-152.

Castellazzi, M., Morand, P., George, J., and Buttin, G. (1977)
Molec. Gen. Genet. 153, 297-310.

Chadwick, P., Pirrotta, V., Steinberg, R., Hopkins, N., and Ptashne, M.
(1970) Cold Spring Harbor Symp. Quant. Biol. 35, 283-294.

Clark, A. J. and Margulies, A. D. (1965) Proc. Natl. Acad. Sci. U.S.A.
53, 451-459.

Clark, A., Chamberlin, M., Boyce, R., and Howard-Flanders, P. (1966)
J. Mol. Biol. 19, 442-454.

Cowlishaw, J. and Ginoza, W. (1970) Virology 41, 244-255.

Donch, J., Greenberg, J., and Green, M. (1970) Genet. Res. (Camb.)
15, 87-97.

Dove, W. (1969) Virology 38, 349-351.

Echols, H. and Green, L. (1971) Proc. Natl. Acad. Sci. U.S.A. 68,
2190-2194.

Eisen, H. and Ptashene, M. (1971) In Bacteriophage X, Hershey, A. D.
(ed.) pp. 239-245, Cold Spring Harbor, New York.







Emmerson, P. T. (1968) Genetics 60, 19-30.

Etlinger, J. D. and Goldberg, A. L. (1977) Proc. Natl. Acad. Sci.
U.S.A. 74, 54-58.

George, J. and Devoret, R. (1971) Molec. Gen. Genet. Ill, 103-119.

Gingery, R. and Echols, H. (1967) Proc. Natl. Acad. Sci. U.S.A.
58, 1507-1514.

Goldberg, A. R. and Howe, M. (1969) Virology 38, 200-202.

Goldmark, P. J. and Linn, S. (1972) J. Biol. Chem. 247, 1849-1860.

Goldthwait, D. and Jacob, F. (1964) C. R. Acad. Sci. 259, 661-664.

Gottesman, M., Hicks, M., and Gellert, M. (1973) J. Mol. Biol.
77, 531-547.

Gudas, L. J. and Mount, D. W. (1977) Proc. Natl. Acad. Sci. U.S.A.
74, 5280-5284.
Gudas, L. J. and Pardee, A. B. (1975) Proc. Natl. Acad. Sci. U.S.A.
72, 2330-2334.

Gudas, L. J. and Pardee, A. B. (1976) J. Mol. Biol. 101, 459-477.

Hertman, I. and Luria, S. E. (1967) J. Mol. Biol. 23, 117-133.

Horiuchi, T. and Inokuchi, H. (1967) J. Mol. Biol. 23, 217-224.

Howard-Flanders, P. (1968) Advan. Biol. Med. Phys. 12, 299-317.

Howard-Flanders, P., Boyce, R. P., and Theriot, L. (1966) Genetics
53, 1119-1136.

Howard-Flanders, P., Rupp, W. D., Wilkins, B. M., and Cole, R. S.
(1968) Cold Spring Harbor Symp. Quant. Biol. 33, 195-205.

Howard-Flanders, P. and Theriot, L. (1966) Genetics 53, 1137-1150.

Inouye, M. (1971) J. Bact. 106, 539-542.

Jacob, F. and Monod, J. (1961) J. Mol. Biol. 3, 318-356.

Jacob, F., Sussman, R., and Monod, J. (1962) C. R. Acad. Sci.
254, 4214-4216.
Jacob, F. and Wollman, E. L. (1953) Cold Spring Harbor Symp. Quant.
Biol. 18, 101-121.

Kaiser, A. D. (1957) Virology 3, 42-61.







Kaiser, A. D. and Jacob, J. (1957) Virology 4, 509-521.

Kirby, E. P., Jacob, F. and Goldthwait, D. A. (1967) Proc. Nati.
Acad. Sci. U.S.A. 58, 1903-1910.

Kondo, S., Ichikawa, H., Iwo, K., and Kato, T. (1970) Genetics 66,
187-217.

Kornberg, A. (1974) In DNA Synthesis, p. 94, San Francisco: Freeman.

Laemmli, U. K. (1970) Nature 227, 680-685.

Lieb, M. (1966) J. Mol. Biol. 16, 149-163.

Little, J. W. (1967) J. Biol. Chem. 242, 679-686.

Little, J. W. and Hanawalt, P. C. (1977) Molec. Gen. Genet. 150,
237-248.

Little, J. W. and Kleid, D. G. (1977) J. Biol. Chem. 252, 6251-6252.

Lloyd, R. G. (1978) J. Bact. 134, 929-943.

Lwoff, A., Siminovitch, L., and Kjeldgaard, N. (1950) Compt. Rend.
Acad. Sci. 231, 190.

Lydersen, B. K. and Pollard, E. C. (1977) J. Bact. 129, 698-701.

Maenhaut-Michel, G., Brandenburger, A., and Boiteux, S. (1978)
Molec. Gen. Genet. 163, 293-299.

Maniatis, R. and Ptashne, M. (1973) Nature 246, 133-136.

Maxam, A. and Gilbert, W. (1977) Proc. Nati. Acad. Sci. U.S.A.
74, 560-564.

McEntee, K. (1976) Virology 70, 221-222.

McEntee, K. (1977) Proc. Nati. Acad. Sci. U.S.A. 74, 5275-5279.

McEntee, K., Hesse, J. E., and Epstein, W. (1976) Proc. Natl. Acad.
Sci. U.S.A. 73, 3979-3983.

Meyn, M. S., Rossman, T., and Troll, W. (1977) Proc. Natl. Acad.
Sci. U.S.A. 74, 1152-1156.

Milcarek, C. and Weiss, B. (1972) J. Mol. Biol. 68, 303-318.

Mount, D. W. (1977) Proc. Nati. Acad. Sci. U.S.A. 74, 300-304.

Mount, D. W., Low, K. B., and Edmiston, S. J. (1972) J. Bact. 112,
886-893.







Mount, D. W., Walker, A. C., and Kosel, C. (1973) J. Bact. 116,
950-956.

Noack, D. and Klaus, S. (1972) Molec. Gen. Genet. 115, 216-224.

Otsuji, N., Sekiguchi, M., lijima, T., and Tagagi, Y. (1959) Nature
184, 1079-1080.

Pabo, C. 0., Sauer, R. T., Sturtevant, M., and Ptashne, M. (1979)
Proc. Natl. Acad. Sci. U.S.A. 76, 1608-1612.

Pirrotta, V. (1976) Current Topics in Microbiology and Immunology
74, 21-54.

Pirrotta, V., Ptashne, M., Chadwick, P., and Steinberg, R. (1971)
In Procedures in Nucleic Acid Research, Cantoni, G. L. and Davies,
D. R. (eds) Vol. 2, pp. 703-715, New York: Harper and Row.

Podsakoff, G. M., Ho, Y. L., Bullas, L. R., and Harris, T. W. (1976)
Virology 74, 531-534.

Ptashne, M. (1967a) Proc. Natl. Acad. Sci. U.S.A. 57, 306-313.

Ptashne, M. (1967b) Nature 214, 232-234.

Ptashne, M. and Hopkins, N. (1968) Proc. Natl. Acad. Sci. U.S.A.
60, 1282-1287.

Radding, C. M. (1970) J. Mol. Biol. 52, 491-499.

Radding, C. M. and Carter, D. M. (1971) J. Biol. Chem. 246, 2513-2518.

Radman, M. (1975) In Molecular Mechanisms for the repair of DNA,
Hanawalt, P. and Setlow, R. B. (eds.) pp. 355-367, New York: Plenum
Press.

Radman, M., Villani, G., Boiteux, S., Defais, M., Caillet-Fauquet, P.,
and Spadari, S. (1977) In Origins of Human Cancer, Watson, J. D.
and Hiatt, H. (eds.) pp. 903-922, New York: Cold Spring Harbor.

Roberts, J. W. and Roberts, C. W. (1975) Proc. Natl. Acad. Sci. U.S.A.
72, 147-151.

Roberts, J. W., Roberts, C. W., and Craig, N. L. (1978) Proc. Natl.
Acad. Sci. U.S.A. 75, 4714-4718.

Roberts, J. W., Roberts, C. W., and Mount, D. W. (1977) Proc. Natl.
Acad. Sci. U.S.A. 74, 2282-2287.

Rupp, W. D. and Howard-Flanders, P. (1968) J. Mol. Biol. 31, 291-304.

Rupp, W. D. and Ihler, G. (1968) Cold Spring Harbor Symp. Quant. Biol.
33, 647-650.








Satta, G., Gudas, L. J., and Pardee, A. B. (1979) Molec. Gen. Genet.
168, 69-80.

Sauer, R. T. (1978) Nature 276, 301-302.

Sauer, R. T. and Anderegg, R. (1978) Biochemistry 17, 1092-1100.

Schuster, H., Beyersmann, D., Mikolajczyk, M., and Schlicht, M.
(1973) J. Virology 11, 879-885.

Sedgewick, S. G., Levine, A., and Bailone, A. (1978) Molec. Gen.
Genet. 160, 267-276.

Shinagawa, H. and Itoh, T. (1973) Molec. Gen. Genet. 126, 103-110.

Shinagawa, H., Mizuuchi, K., and Emmerson, P. T. (1977) Molec. Gen.
Genet. 155, 87-91.
Shulman, M. J., Hallick, L. M., Echols, H., and Signer, E. R. (1970)
J. Mol. Biol. 52, 501-520.

Sicard, W. and Devoret, R. (1962) C. R. Acad. Sci. 255, 1417-1419.

Steinberg, R. and Ptashne, M. (1971) Nature (New Biol.) 230, 76-80.

Susskind, M. M. and Botstein, D. (1975) J. Mol. Biol. 98, 413-424.

Sussman, R. and Jacob, F. (1962) C. R. Acad. Sci. 254, 1517-1519.

Sussman, R., Resnick, J., Calame, K., and Baluch, J. (1978) Proc.
Natl. Acad. Sci. U.S.A. 75, 5817-5821.

Tomizawa, J. and Ogawa, T. (1967) J. Mol. Biol. 23, 247-263.

Trgovcevic, Z. and Rupp, W. D. (1974) Proc. Natl. Acad. Sci. U.S.A.
71, 503-506.

Villani, G., Boiteux, S., and Radman, M. (1978) Proc. Natl. Acad.
Sci. U.S.A. 75, 3037-3041.

Weisberg, R. and Gottesman, M. (1969) J. Mol. Biol. 46, 565-580.

West, S. C. and Emmerson, P. T. (1977) Molec. Gen. Genet. 151, 57-67.

West, S. C., Powell, K. A., Emmerson, P. T. (1975) Molec. Gen. Genet.
141, 1-8.

White, B. J., Hochhauser, S. J., Cintron, N. M., and Weiss, B. (1976)
J. Bact. 126, 1082-1088.

Willetts, N. S. and Clark, A. J. (1969) J. Bact. 100, 231-239.

Witkin, E. (1969) Mutat. Res. 8, 9-14.




69


Witkin, E. (1974) Proc. Natl. Acad. Sci. U.S.A. 71, 1930-1934.

Witkin, E. M. (1976) Bact. Revs. 40, 869-907.

Wu, A. M., Ghosh, S., Echols, H., and Spiegelman, W. G. (1972)
J. Mol. Biol. 67, 407-421.

Yanofsky, C. and Ito, J. (1966) J. Mol. Biol. 21, 313-334.












BIOGRAPHICAL SKETCH

Robert Mitchell Crowl received his Bachelor of Arts degree in

May, 1975, from Pfeiffer College in Misenheimer, North Carolina,

graduating magna cum laude with a major in Environmental Studies.

From September, 1975,until June, 1979, he pursued graduate studies

in the Department of Biochemistry and Molecular Biology at the Univer-

sity of Florida in Gainesville, Florida,under the direction of

Dr. Richard P. Boyce. After receiving his doctoral degree in June,

1979, the author plans to continue his training under the direction

of Dr. Harrison Echols in the Department of Molecular Biology at the

University of California at Berkeley funded by a postdoctoral fellow-

ship awarded by the Damon Runyon-Walter Winchell Cancer Fund.

The author was born in Akron, Ohio,on December 31, 1952. He is

married to the former Jody Ann Hash. They are expecting their first

child in August, 1979.








I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Ri'ehard P. Bo3Fce, Chairmah
Professor of Biochemistry
and Molecular Biology









I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.


Biochemistry and Molecular
Biology


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




Peter M. McGuire
Assistant Professor of
Biochemistry and Molecular
Biology








I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.




L. O'Neal Ingram
Associate Professor of
Microbiology








This dissertation was submitted to the Graduate Faculty of the College
of Medicine and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.

June 1979






Dean, College of Medicine