[m

,... "Mm?" w»...
I..w;1 W |n' 1'11“ 111'
. ' 1.1171 ‘ .Ufl'dk In ‘2'“.

“.111 1’11””

N“

:4" n ‘Y‘u'vflluu 1101‘. "71"!13119' If” _3‘ 1H1 4:4; J:
W .. ~ -» m, c

£1133 -
1:11? 211 Mufti?“ H; . z. p4 1.199% '5' 3'1 “f": .
151Fmb‘vraj1‘1‘rgnl 5911‘? 31”" Am ".1, j

$.04)?“ (I (”1' $35.21;; V‘s: JUL‘HLSI‘ L ’%
d ,I

Ww 1‘11; rum-1"; '._‘ -“"' 3" "0.5V"

luf‘ I
'2‘; "E11111 PqflLm'Th 111W 1 11”, 'nwupg $312k“ 1.5:].15'1

.1. 144’ I? ‘1‘13115'L‘11Lgmm7331 . "1.53
”Mm .19?“
ii?“ Rfl ‘ I??"“2_‘ s.’

  
 
 
    
 
 
  

.

    
    
 
 

      
   

  

        

    
  
 
 
    
   
   
 
 

l

      

- 1m 11111945114111 1
up... ' L i, 5‘: 1'1“. 1
Mg, ”[111? 111255.513"?§}“

 
  
  

  
  

    
    
  

   
  
    
   
   
        
    
   
     
 
 
 
  
      

     
 

 

               

   
 

 
     
 

*é . 5 a;
"'1 ‘1 “#2.
$33!!

4.2; :2- 1“: yrf'l-‘xi’dfggr I; 4’4
.; J, , '

u
‘v

  
 

1 r!
T‘ 1 1: 3‘ 3
(111,1 t1? "11;; “5‘1“?“ ., x t , X-
M A fit" . ”71’.“ 'iv‘fi’i‘ ”"33- .1! r; ‘.‘.:o'.;:_. - .. .
- I; f '4 a}??? «"¥ ‘5‘“? 4311133311119 $3311: 3% 33:3:13‘3’11g9zggn a.j:-'<:.a~;*—-:‘
u. z .' ' 3 ‘11 'I 1;“... ‘sn’rétg'fl :‘fiffbiui'i’l 11'; 14:”11’M. "’5:1ll.'§!T ”fixnl'f"1y{;i;.;g'
.1 ‘7 3‘: 1 1“ “11:11:11“ '- 1§5 1“!“ '::‘.O?}":.E 11_ of I_ :T. K1335? :fi ‘1 ‘ “~31 #38147
»- f: .142; ""11“; , :15: 9 1w — wafifigg}; M; “.43; mill. 1:11;: *L; 3715? , ES: ”.51;
r . ‘ . ‘1’: . \ «.a.‘ ,1 z ‘ . " .‘ ",2 . .l.‘ . y.‘ "n‘ "?:J, .4 b .
L {1‘ ; 11:11:13? "11., I $2211. -1?5o,31w?:§‘m!:;ggi ' *fil.‘_,;}fi£1;1§x;§¥;£;$gl; §jL<'}E;-1E;P
.l.‘ . 1hfig~¢5iig ’1?“ \' -'; v: i. “Fr; 5- L ..'..' -rvu: "-
"“- " I ‘ " 1‘; 93'; 1n! "- A 5.
0- ' ¢§¢ . ‘ . 4:.“ q.‘ .1 ..
-: ' ;_ ' g . glint :L ' 1 1m; 1111:1141 . .
" 1 :1 . 1. ~ 3.‘ . ,6
' ' ‘17; .325}; 51.3,}!- , g:
,3 - 1 p ;. .I,; Kath. . :3 ‘
- . - ~ 2.2; c a . . .1
_. I t if Hahn} 31;; ¢ .1. -:.‘:';1'
'- - ' .. .,.}¢€:§.:'n3§‘ 5
2' ' "’ . :i?~§e:aizzr
{:14

   
  
    
 
  

     

    

 
 
 
 
  
   
  

     
     
  

  

  
   
  
  
 
 
     

 

  
  

 

 

 
 

- ' _ “v 4: ‘ .
g -. 2!‘ = .g-‘z* 33.1%; ,.y,9::j'-§§§ $3;
"‘3 1.4%“ .5” jJ ‘. “"-.:i:‘£?‘.- fit “-33" f
:f' :1": .1141“? r; ugh: .} ’n ’1‘; ' " ~'~£'I‘J.»“.!‘,-‘v ’5. "buy-‘- 133 ..
a ' ““131 ‘3"“imun~ ' ‘ * ""-:"”?a‘:;~“ " 13%;.»
1"“ " é"! { £1.11,” {'1’ ' ‘ . 3.7!:
x.‘ aw .fi' ' ‘ ' 1 I}. .1 .'
1:11;. “3&1. mu: ”1;! _ 3??
‘3 ‘1' .. "H .7 L. .5."-
-,V 7:: . 'l’"’;'1:'§.( .: . I:
a | O
:" u‘

 

 

‘- - 1.‘._"|v
gym ' ar-

.,~.<-

 
 
 
 
   

 

s4.
‘ 24‘.

7

iii; 7.. .,.'
:ti‘ ‘ :7 'éJl I o
313711.12‘1“.
9,1,3 1M

 

n
.
u

a..."

 
 
 
 
   
    
     
   

,‘—’

'5? ’

.251: 5:41.?)
-1.“ '
.._ ’HM. i&;.'a:-)6
. ._ . I ..
l‘ u c

- a

I)

. «4‘

gm- :1“,

‘1 {31 x‘:
113' ‘1’!»
39g.)

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Kinds and spectra of mutations formed when a shuttle
vector containing adducts of benzo(a)pyrene-7,8-diol-
9,10-epoxide or l-nitrosopyrene replicates in mammalian
cells

presented by

Jia-Ling Yang

has been accepted towards fulfillment
of the requirements for

Doctoral degree in Biochemistry

 

 

Major profesligr/k

Date 1-8-88

 

MSU is an Affirmatiw Action /Equal Opportunity Institution 0— 12771

 

 

 

 

‘}V153I_] RETURNING MATERIALS:
Place in book drop to
maintains remove this checkout from
”In. your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

 

 

KINDS AND SPECTRA OF MUTATIONS FORMED WHEN A SHUTTLE VECTOR CONTAINING
ADDUCTS OF BENZO[a]PYRENE-7,8-DIOL-9,lO-EPOXIDE OR l—NITROSOPYRENE
I REPLICATES IN MAMMALIAN CELLS

by

Jia—Ling Yang

A DISSERTATION

Submitted to
Michgan State University
in part fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Biochemistry

1988

ABSTRACT

KINDS AND SPECTRA 0F HUTATIONS FORMED NHEN A SHUTTLE VECTOR CONTAINING
ADDUCTS OF BENZOla]PYRENE-7,8-DIOL-9,lO-EPOXIDE OR l-NITROSOPYRENE
REPLICATES IN NAHNALIAN CELLS

by

Jia-Ling Yang

Polycyclic aromatic hydrocarbons such as benzo[a]pyrene (BP) and
nitrated derivatives such as l-nitropyrene (l-NP) are widespread
environmental contaminants produced by incomplete combustion. The
principal reactive metabolite of BP is the anti isomer of (i)BP-7,8-
dial-9,10-epoxide (BPDE) which binds to DNA mainly at the N2 position
of guanine. The reactive metabolite of l-NP, l-nitrosopyrene (l-NOP)
is further reduced and binds covalently to DNA at the C8 position of
guanine. Since the kinds of tumors formed by the two parent
carcinogens differ and since the DNA adducts differ, they may cause
mutations by different mechanisms.

To examine this question, a shuttle vector (p3AC), carrying a
bacterial suppressor tRNA target gene (supfi), was treated with BPDE.
The BPDE-treated plasmid was introduced into a monkey cell line (C057)

and allowed to replicate. Progeny plasmids were then isolated,

purified, and introduced into bacteria carrying an amber mutation in
the B-galactosidase gene to detect those carrying mutations in the M
gene. The background frequency of supfi mutants obtained was too high
to detect BPDE-induced mutants, but electrophoretic analysis showed
that mutants derived from treated plasmid contained predominantly point
mutations, whereas those derived from untreated plasmid contained
predominantly gross alterations. A second shuttle vector carrying sgpfi
in a different position (le89) was obtained and the human 293 cell
line was used. With this system the background frequency of supfi
mutants was lowered to 1.4 x 10'4- The number of adducts required to
reduce the bacterial transformation activity of the plasmid was the
same for either carcinogen, but the frequency of supfi mutants formed
per adduct in 293 cells was 3.5-fold higher for BPDE than for l-NOP.
DNA sequence analysis of independent mutants derived from BPDE- or I-
NDP-treated plasmids showed that 60 out of 86 BPDE-induced mutants and
51 out of 60 l-NOP-induced mutants contained base substitutions. The
majority involved G-C pairs, with 60% of these representing G'C -+ T°A
transversions. Examination of the location of the point mutations
among the 85 base pairs in the structure of the sgpfi tRNA revealed that
the progeny of the l—NDP-treated plasmid had five hot spots, and those
of the BPDE-treated plasmid had eight hot spots, only two of these hot

spots were common for both agents.

To
my parents, Chin-Chieh and Shu-Huey Yang;

and
my husband, Jiunn-qu Lee

ii

ACKNOWLEDGMENT

I would like to express my appreciation to Dr. Veronica H. Maher
for her constant advice during my entire graduate training and for
serving as the director of this research. I wish to thank Dr. J.
Justin McCormick for his support and advice as well as the other
members of my graduate committee, Drs. Susan E. Conrald, Jerry B.
Dodgson and Jon Kaguni for their advice and invaluable time.

I also wish to express my gratitude to Dr. Dennis Fry for his many
discussion regarding this project and for his technical advice as well
as Bernard Schroeter for his constant assistance. I also owe thanks
to numerous individuals whose constant encouragement and generous
friendship supported me through this endeavor. They include Hecky and
Jeffery Howell, Hilliam Parks, Becki Corner, Yen-Yun Hang, Dominic Ho,
Carol Patterson, Lisa Cantu, Lonnie Nilam, Daniel Wilson, Peter Hurlin,
and Helen Palmer.

I wish to thank my parents Chin—Chieh and Shu-Huey Yang for their
love, financial support and encouragement. To my husband Jiunn-qu
Lee, I would like to express my appreciation for his joining me at
Michigan State University, even though it meant extending the time
required for his graduate studies as well as sharing with me the care

of our daughter Nancy.

TABLE OF CONTENTS

Page
LIST OF TABLES .................................................... vii
LIST OF FIGURES ................................................... viii
ABBREVIATIONS ..................................................... x
INTRODUCTION ...................................................... 1

CHAPTER I.

LITERATURE REVIEW

A. Role of Polycyclic Aromatic Hydrocarbons and Their
Nitrated Derivatives in Mutagenesis and Carcinogenesis.. 8

1.

01-th

Metabolic activation of benzo[a]pyrene
and l-nitropyrene .................................. 9

. DNA lesions produced by reactive metabolites of

benzo[a]pyrene and l-nitropyrene ................... 19

. Mutagenesis of reactive derivatives of benzo[a]pyrene,

1- nitropyrene, and dinitropyrenes .................

. Carcinogenesis of benzo[a]pyrene, benzo[a]pyrene

diolepoxides, l-nitropyrene, and dinitropyrenes.... 25

. The specific kinds of mutations induced by

BPDE and l-NOP ..................................... 26

B. Systems Presently Available for Investigating
Mutagenesis in Mammalian Cells at the Molecular Level... 28

 

l. Endogenous mammalian gene systems .................... 29
a. Studies with the hprt gene ........................ 30
b. Studies with the aprt gene ........................ 32
c. Studies with the dflf; gene ........................ 33
2. Integrated exogenous gene systems .................... 34
a. Studies with the gpt gene ......................... 35
b. Studies with the supF gene ........................ 36
3. Viral vector systems ................................. 37
4. Transiently replicating shuttle vector systems ....... 39

a. High background mutant frequency of transfected
vector DNA ........................................ 40

b. Systems to minimize the high background

frequency of mutants .............................. 41
5. Stably replicating shuttle vector systems ............ 46
References ................................................. 52

iv

CHAPTER II ESTABLISHMENT OF A SUITABLE SHUTTLE VECTOR
SYSTEM TO STUDY MOLECULAR MUTAGENESIS .................. 63

A. The Shuttle Vector Mutagenesis System ................... 63
B. Determination of Experimental Conditions to Maximize

the Yields of Progeny Plasmid from Mammalian cells.... 76
C. Reducing the Background Mutant Frequency ................ 76

References ................................................. 86

CHAPTER III KINDS OF MUTATIONS FORMED WHEN A SHUTTLE VECTOR
CONTAINING ADDUCTS OF BENZO[a]PYRENE-7,8-DIOL-

9,10-EPOXIDE REPLICATES IN C057 CELLS ................. 87
Summary .................................................... 88
Introduction ............................................... 90
Materials and Methods and Results

Cells and plasmids ....................................... 92
Preparation of plasmid DNA containing BPDE adducts ....... 92
Transfection of COS cells and assay of progeny
plasmid for mutations .................................. 93
Characterization of mutant plasmids by gel
electrophoresis and sequencing ......................... 93
Discussion ................................................. 101
Acknowledgements ........................................... 104
References ................................................. 105

CHAPTER IV KINDS OF MUTATIONS FORMED WHEN A SHUTTLE VECTOR
CONTAINING ADDUCTS OF (1)43 ,8a-DIHYDROXY-9a ,IOa
-7,8,9,lO-TETRAHYDROBENZO[a]PYRENE REPLICATES

IN HUMAN CELLS ......................................... 108
Summary .................................................... 109
Introduction ............................................... 111
Materials and Methods

Cells and plasmids ....................................... 113
Formation of BPDE adducts on the plasmid ................. 113
Transfection and rescue of replicated plasmid ............ 114
Bacterial transformation and mutant identification ....... 114
Characterization of mutants .............................. 115
Results
Characterization of BPDE-treated plasmid ................. 116
Frequency of mutants induced by BPDE treatment .......... 116
Characterization of the mutations formed in 293 cells.... 123
Mutational hot spots for BPDE-induced mutations .......... 126
Discussion ................................................. 131
Acknowledgements ........................................... 135
References ................................................. 136

CHAPTER V KINDS AND SPECTRUM OF MUTATIONS INDUCED BY
l-NITROPYRENE ADDUCTS DURING PLASMID

REPLICATION IN HUMAN CELLS .............................. 139
Summary .................................................... 140
Introduction ............................................... 142
Meterials and Methods

Cells and plasmid ........................................ 145
Formation of 1-NOP adducts on the plasmid ................ 145
Transfection and rescue of replicated plasmid ............ 146
Bacterial transformation and mutant identification ....... 146
Characterization of mutants .............................. 147
Determination of sites of carcinogen-induced adducts ..... 147
Results
Characterization of l-NOP-treated plasmid ................ 149
Frequency of mutants induced by l-NOP adducts ............ 149
Nature and location of the specific mutations
induced by l-NOP-treated plasmid in 293 cells .......... 156
Sites of carcinogen-induced adducts ...................... 159
Discussion ................................................. 168
Acknowledgements ........................................... 173
References ................................................. 174

vi

LIST OF TABLES

Table Page
Chapter II

1.
2.
3.

Comparison the frequency of spontaneous supfi mutants
generated during replication of p3AC or p2189 in C057 cells.... 83

Frequency of supF mutants generated during replication

of untreated p2189 in C057 cells ............................... 84
Frequency of spontaneous supF mutants observed during

replication of p2189 in 293 cells .............................. 85

Chapter III

1. Frequency of supF mutants obtained with BPDE-treated or

untreated p3AC that had replicated in C057 cells ............... 94
2. Characterization of the kinds of mutations generated in

p3AC during replication in C057 cells .......................... 96
3. Analysis of the kinds of base changes found in the supF

gene of the mutant p3ACs analyzed .............................. 97
Chapter IV
1. Analysis of mutants obtained by transformation of E. golj

with progeny of p2189 generated during replication
in 293 cells ................................................... 120

. Analysis of sequence alterations generated in the supF

gene by replication BPDE-treated or untreated p2189
in 293 cells ................................................... 124

3. Kinds of base substitutions generated in the supF

gene by replication of BPDE-treated or untreated

p2189 in 293 cells ............................................. 125
Chapter V
1. Analysis of mutants obtained by transformation of E. golj

2.

with progeny of l-NOP-treated p2189 generated during

replication in 293 cells ....................................... 157
Analysis of sequence alterations generated in the supF

gene by replication l-NOP-treated or untreated p2189

in 293 cells ................................................... 158

. Kinds of base substitutions generated by replication of

I-NOP-treated or untreated le89 in 293 cells .................. 160

vii

LIST OF FIGURES

Figure page
Chapter I
1. Structure of the major BPDE-DNA adduct and l-NOP-DNA adduct.... 11
2. Mechanism of enzymatic activation of BP to
BP 7,8-diol-9,10-epoxides ...................................... 14
3. Structure and nomenclature of four isomers of
benzo[a]pyrene diolepoxide ..................................... 16
4. The reductive metabolism of l-nitropyrene ...................... 18
5. Structure of guanine-, adenine- and cytosine-BPDE adducts ...... 22
Chapter II
. Structure of p3AC and pZ189 .................................... 65
. The shuttle vector mutagenesis system .......................... 68

Ch 0'! «h th—I

7.

. Digestion of Dpnl on plasmid with the bacterial

methylation pattern ............................................ 70

. Agarose gel electrophoresis analysis of plasmids containing

mutations at the supF gene ..................................... 73

. DNA sequencing analysis of a plasmid containing mutations

in the supF gene ............................................... 75

. The relationship between the yield of p3AC generated

during replication in C057 cells and the amount of
input plasmid .................................................. 78
Time course experiment of replication of p3AC in C057 cells.... 80

Chapter III

1.

Distribution of mutants in the sgpF gene of p3AC ............... 99

Chapter IV

1.

(A) Number of BPDE residues bound per plasmid as a function of
concentration of BPDE. (8) Relative frequency of transformation
of bacteria to ampicillin resistance by plasmid p2189 modified
with BPDE compared to untreated plasmid ........................ 118

. Frequency of sgpfi mutants as a function of the number of

BPDE residues per plasmid ...................................... 122

. Location of independent point mutations in the supF tRNA

gene of pZ189 .................................................. 128

. Location of the BPDE-induced hot spots on the cloverleaf

structure of the sgpF tRNA ..................................... 130

viii

Chapter V

l.

0% 01 -h w
0

Number of l-NOP adducts bound per plasmid as a function of
concentration of l-NOP in the presence of ascorbic acid ........ 151

. Relative frequency of transformation of bacteria to

ampicillin resistance by plasmid p2189 containing

l—NOP adducts compared to untreated plasmid .................... 153
. Frequency of supf mutants as a function of the number

of 1-NOP adducts per plasmid ................................... 155

Location of independent point mutations in the sgpF tRNA

gene of p2189 .................................................. 162

. Location of the l-NOP-induced hot spots on the cloverleaf

structure of the supF tRNA ..................................... 164

. Relative frequency of BPDE-guanine or l-NOP-guanine adducts

in the supF gene of pZ189 and location of BPDE- or
1-NOP-induced base substitution within the tRNA coding
sequence of p2189 .............................................. 167

ix

AAAF

aprt
BP
BPDE

BPV
BrdU
ghf:
DNP

EBNA-l

EBV
EMS
ENU

92L
DELL
HSV
IPTG

L—N
MNU

ABBREVIATIONS

2-(N-acetoxy-N-acetylamino)fluorene

ampicillin gene

adenine phosphoribosyltransferase gene
benzo[a]pyrene

(il-73 ,80 -dihydroxy-9a ,10a -epoxy-7,8,9,10-tetrahydro
benzo[a]pyrene

bovine papillomavirus

bromodeoxyuridine

dihydrofolate reductase gene

dinitropyrene

Epstein-Barr virus nuclear antigen 1

Epstein-Barr virus

ethyl methanesulfonate

N-ethyl-N-nitrosourea

galactokinase gene

xanthine (guanine) phosphoribosyltransferase gene
hypoxanthine phosphoribosyltransferase gene
herpes simplex virus

isopropyl B-D-thiogalactoside

a gene coding for the repressor of the lactose operon
Lesch-Nyhan

N-methyl-N-nitrourea

I-NOP
l-NP
PAH
(6-4) py-C
PY‘PY
sgpfi
SV40
LL

UV
X-Gal
XP

l-nitrosopyrene

l-nitropyrene

polycyclic aromatic hydrocarbon

(6-4) pyrimidine-cytosine
pyrimidine-pyrimidine

a gene coding for a tyrosine suppressor tRNA
simian virus 40

thymidine kinase gene

ultraviolet

5—bromo-4-chl oro-3-indolyl B -D-gal actoside

xeroderma pigmentosum

xi

INTRODUCTION

The recent discovery that some human cancers are associated with
single point mutation in the [gt oncogene family (Reddy gt _1., 1982;
Tabin gt gl., 1982; Sukumar gt gl., 1983; Fujita gt g1., 1984) and that
specific oncogenes can be activated through specific mutational changes
by chemical carcinogens (Marshall gt 91.,1984; Vousden gt g1., 1986)
enhanced interest in the molecular mechanisms by which mutations arise
and in the role of mutation in initiating genetic disease and cancer.
Attempts to deduce the nature of the molecular machanisms by which
carcinogens induce mutations in mammalian cells have been hampered by
the inability to isolate and analyze newly mutated genes at the
sequence level. However, the development of shuttle vectors, i.e.,
plasmids carrying a defined target gene and capable of replicating in
mammalian cells and also in bacteria, provide a solution to this
problem (Calos e_t_ 51., 1983; Razzaque .e_t g1., 1983; Sarkar gt 11.,
1984; Lebkowski _t _l., 1985; Seidman gt gl., 1985; Bredberg gt al.,
1986; Drinkwater and Klinedinst, 1986; Hauser gt _1., 1986; Lebkowski
gt fl“ 1986). After fixation of mutations by replication of the
vector in mammalian cells, mutants can be efficiently detected in
bacteria and amplified for subsequent molecular analysis.

The shuttle vector approach was first demonstrated in simian

cells, using vectors based on simian virus 40 (SV40). The use of these

vectors has been limited by a high spontaneous mutation frequency (1%

2

to 10%) associated with the transfection process rather than arising
during replication, and also by an inability of the vectors to
replicate efficiently in mammalian cells (Razzaque gt ._1., 1983;
Ashman and Davidson, 1984; Lebkowski gt at, 1984; Miller gt Q”
1984; Razzaque gt g1., 1984; Chakrabarti gt g1., 1985). Both of these
problems were solved by the discovery by Calos and her associates
(Lebkowski gt _1., 1984) that the human cell line 293 (Graham gt gl.,
1977) replicates SV40-based shuttle vectors extremely' well with a
relatively low spontaneous mutation frequency. Another approach taken
by Seidman gt a. (1985) to minimize the high spontaneous mutation
frequency was to decrease the possibility of recovering mutants
containing gross rearrangements by the judicious design of a shuttle
vector in which the target gene is strategically located between two
genes essential for recovery of the vector in bacteria, i.e., the gene
coding for ampicillin resistance and the bacterial origin of
replication.

The well-studied carcinogen, (jg-78,80—dihydroxy-9a,10cz-epoxy-
7,8,9,lO-tetrahydrobenzo[a]pyrene (BPDE) is a major reactive metabolite
of the widely distributed enviromental carcinogen benzo[a]pyrene (BP),
an agent produced primarily by incomplete combustion processes
(Weinstein gt g1., 1976). l-Nitrosopyrene (l-NOP) is a partially
reduced metabolite of 1-nitropyrene (I-NP), a carcinogen which accounts
for 20% to 30% of the "direct-acting" mutagenicity of diesel emission
particles (Rappaport gt l., 1980; Rosenkranz, 1982; Schuetzle, 1983).
Both BPDE and l-NOP have been shown to form tumors in animals (Buening

g1 g1., 1978; Wislocki gt 91., 1986) and to induce mutations in
bacteria (Mermelstein fl g1., I981; Eisenstadt et 1., 1982; Howard gt

3

gl., 1983; Heflich gt gl., 1985) and in mammalian cells (Yang gt gl.,
1980; Patton gt .g1., 1986). However, the' mutational specificity
induced in mammalian cells by these two compounds has not been
determined. Both BPDE and l-NOP form covalently bound adducts in DNA
pricipally with guanine (Weinstein gt g1., 1976; Meehan gt gl., 1977;
Heflich g at, 1985). l-NOP binds to guanine at the C8 position
(Heflich gt g1., 1985), whereas BPDE forms its pricipal guanine adduct
at the N2 position (Weinstein gt _l., 1976). Since these guanine
adducts are located in very different position in the DNA helix and
only the N2 position is involved in the base pairing part of the
molecule, the mechanism of molecular mutagenesis by these compounds may
be different.

This thesis was undertaken (1) to investigate the specific kinds
of mutations at the sequence level induced by two structurally-related
carcinogens, i.e., BPDE and l-NOP, when a shuttle vector replicates in
mammalian cells; (2) to determine their location in the DNA sequence of
the target gene and to see if these two compounds induce the same
mutational spectrum; (3) to investigate if these two compounds have the
same biological effectiveness, i.e., to compare the ability of their
DNA adducts to interfere with bacterial transformation and to induce
mutations during replication of plasmids in human cells.

Chapter I of the thesis reviews the background literature bearing
on these questions. Chapter 11 describes fundamental work I carried
out to set up the shuttle vector system for such mutagenesis studies.
Chapter III consists of a manuscript published in the Notes section of
the March 1987 issue of Molecular and Cgllular Biology 7:1267-1270

inhich describes my work on mutations formed when p3AC containing

4

adducts of BPDE replicated in C057 monkey cells. Chapter IV consists
of a manuscript published in the June 1987 issue of Proceeding of the
National Academy of Sciences U.5.A. 84:3787-3791 which describes the
research I carried out to determine the specific kinds of mutations
induced when vector pZ189 containing BPDE adducts replicated in 293
cells, as well as their location in the gggfiy gene. Chapter V
describes comparable work carried out with vector p2189 treated with 1-
NOP and compares the results obtained with l-NOP with those I found
using BPDE. The format used for Chapter V is that of a manuscript to
be submitted to Molecmr and Cellular Biology as soon as I have
completed experiments comparing the rate of excision repair of l—NOP

and BPDE-induced DNA adducts in the 293 cell line.

REFERENCES

Ashman, C. H. and Davidson, R. L. (1984), High spontaneous mutation
frequency in shuttle vector sequences recovered from mammalian cellular
DNA, Mol. Cell. Biol., 5, 2266-2272.

Bredberg, A., Kraemer, K. H. and Seidman, M. M. (1986), Restricted
ultraviolet mutational spectrum in a shuttle vector propagated in
xeroderma pigmentosum cells, Proc. Natl. Acad. Sci. USA 8;, 8273-8277.

Buening, M. K., Wislocki, P. G., Levin, H., Yagi, H., Thakker, D. R.,
Akagi, H., Koreeda, M., Jerina, D. M. and Conney, A. H. (1978),
Tumorigenicity of the optical enantiomers of the diastereomeric
benzo[a]pyrene 7,8-diol-9,10-epoxides in newborn mice: Exceptional
activity of (i)-7 B ,8 a -dihydroxy-9 a ,10 a -epoxy-7,8,9,10-tetrahydro
benzo[a]pyrene, Proc. Natl. Acad. Sci. USA 15, 5358-5361.

Calos, M. P., Lebkowski, J. S. and Botchan, M. R. (1983), High mutation
frequency in DNA tranfected into mammalian cells, Proc. Natl. Acad.
Sci. USA 89, 3015-3019.

Chakrabarti, 5., Joffe, 5., and Seidman, M. M. (1985), Recombination
and deletion of sequences in shuttle vector plasmids in mamalian
cells, Mol. Cell. Biol., 5, 2265-2271.

Drinkwater, N. R. and Klinedinst, D. K. (1986), Chemically induced
mutagenesis in a shuttle vector with a low-background mutant frequency,
Proc. Natl. Acad. Sci. USA 8;, 3402-3406.

Eisenstadt, E., Warren, A. J., Porter, J., Atkins, D. and Miller, J. H.
(1982), Carcinogenic: epoxides of benzo[a]pyrene and
cyclopenta[cd]pyrene induce base substitutions via specifh:
transversions, Proc. Natl. Acad. Sci. USA 19, 1945-1949.

Fujita, J., Yoshida, 0., Yuasa, Y., Rhim, J. 5., Hatanaka, M. and
Aaronson, S. A. (1984), Ha-tg§_ oncogenes are activated by somatic
alterations 'hi human urinary tract tumors, Nature (London) 392, 464-

466.

Graham, F. L., Smiley, J., Russell, W. C. and Mairn, R. (1977),
Characteristics of a human cell line transformed by DNA from human
adenovirus type 5, J. Gen. Virol. 36, 59-74.

Hauser, J., Seidman, M. M., Sidur, K. and Dixon, K. (1986), Sequence
specificity of point mutations induced during passage of a UV-
irradiated shuttle vector plasmid in monkey cells, Mol. Cell. Biol. 6,
277-285.

Heflich, R. H., Howard, P. C. and Beland, F. A. (1985), 1-
Nitrosopyrene: an intermediate in the metabolic activation of 1-
nitropyrene to a mutagen in Salmonella tyghimurjgm TA1538, Mutat. Res.,
112, 25-32.

6

Howard, P. C., Heflich, R. H., Evans, F. E. and Beland, F. A. (1983),
Formation of DNA adducts in vitro and in Salmongllg typhimurium upon
metabolic reduction of the environmental mutagen 1-nitropyrene, Cancer
Res., 66, 2052-2058.

Lebkowski, J. 5., DuBridge, R. B., Antell, E. A., Greisen, K. S. and
Calos, M. P. (1984), Transfected DNA is mutated in monkey, mouse, and
human cells, Mol. Cell. Biol. 5, 1951-1960.

Lebkowski, J. 5., Clancy, 5., Miller, J. H. and Calos, M. P. (1985),
The lagl shuttle: Rapid analysis of the mutagenic specificity of
ultraviolet light in human cells, Proc. Natl. Acad. Sci. USA gg, 8606-
8610.

Lebkowski, J. 5., Miller, J. H. and Calos, M. P. (1986), Determination
of DNA sequence changes induced by ethyl methanesulfonate in human
cells, using a shuttle vector system, Mol. Cell. Biol. 6, 1838-1842.

Marshall, C. J., Vousden, K. H. and Phillips, 0. H. (1984), Activation
of c-Ha-tgg-l proto-oncogene by in vitro modification with a chemical
carcinogen, benzo[a]pyrene dial-epoxide, Nature (London) 6t6, 586-589.

Meehan, T., Straub, K. and Calvin, M. (1977), Benzo[a]pyrene diol
epoxide covalently binds to deoxyguanosine and deoxyadenosine in DNA,
Nature (London) Z62, 725-727.

Mermelstein, R., Kiriazides, D. K., Butler, M., McCoy, E. C. and
Rosenkranz, H. 5. (1981) , The extraordinary mutagenicity of
nitropyrenes in bacteria, Mutat. Res. 66, 187-196.

Miller, J. H., Lebkowski, J. 5., Greisen, K. S. and Calos. M. P.
(1984), Specificity of mutations induced in transfected DNA by
mammalian cells, EMBO J. 6, 3117-3121.

Patton. J. D., Maher, V. M. and McCormick, J. J. (1986), Cytotoxic and
mutagenic effects of l-nitropyrene and l-nitrosopyrene in diploid human
fibroblasts, Carcinogenesis 1, 89-93.

Rappaport, 5. M., Wang, Y. Y., Wei, E. T., Sawyer, R., Watkins, B. E.
and Rappaport, H. (1980), Isolation and identification of a direct-
acting mutagen in diesel-exhaust particulates, Environ. Sci. Tech. 16,
1505-1509.

Razzaque, A., Mizusawa, H. and Seidman, M. M. (1983), Rearrangement and
mutagenesis of a shuttle vector plasmid after passage in mammalian
cells, Proc. Natl. Acad. Sci. USA 66, 3010-3014.

Razzaque, A., Chakrabarti, 5., Joffe, 5. and Seidman, M. (1984),
Mutagenesis of a shuttle vector plasmid in mammalian cells, Mol. Cell.
Biol. 1, 435-441.

7

Reddy, E. P., Reynolds, R. K., Santos, E. and Barbacid, M. (1982), A
point mutation is responsible for the acquisition of transforming

properties by the T24 human bladder carcinoma oncogene, Nature (London)
666, 149-152.

Rosenkranz, H. S. (1982), Direct-acting mutagens in diesel exhausts:
magnitude of the problem, Mutat. Res. 161, 1-10.

Sarkar, 5., Dasgupta, U. and Summers, W. C. (1984), Error-prone
mutagenesis detected in mammalian cells by a shuttle vector containing
the 6666 gene of Escherighia 9011, Mol. Cell. Biol. 6, 2227-2230.

Schuetzle, D. (1983), Sampling of vehicle emissions for chemical
analysis and biological testing, Environ. Health Perpect 61, 65-80.

Seidman, M. M., Dixon, K., Razzaque, A., Zagursky, R. J. and Berman, M.
L. (1985), A shuttle vector plasmid for studying carcinogen-induced
point mutations in mammalian cells, Gene 66, 233-237.

Sukumar, J., Notario, V., Martin-Zanca, D. and Barbacid, M. (1983),
Induction of mammary carcinomas in rats by nitrosomethylurea involves
activation of H-tgg-l locus by single point mutations, Nature (London)
666, 658-661.

Tabin, C. J., Bradley, S. C., Bargmann, C. I. and Heinebrg, R. A.
(1982), Mechanism of activation of a human oncogene, Nature (London)
666, 143-149.

Vousden, K. H., 805, J. L., Marshall, C. J. and Phillips, 0. H. (1986),
Mutations activating human c-Ha-rasl protooncogene (HRASI) induced by
chemical carcinogens and depurination, Proc. Natl. Acad. Sci. USA 66,
1222-1226.

Weinstein, I. 8., Jeffrey, A. M., Jennette, K. H., Blobstein, S. H.,
Harvey, R. G., Harris, C., Autrup, H., Kasai, H. and Nakanishi, K.
(1976), Benzo[a]pyrene diol epoxides as intermediates in nucleic acid
binding in vitro and in vivo, Science 166,592-595.

 

Wislocki, P. G., Bagan, E. 5., Lu, A. Y. H., Dooley, K. L., Fu, P. P.
Han-Hsu, H., Beland, F. A. and Kadlubar, F. F. (1986), Tumorigenicity
of nitrated derivatives of pyrene, benzo[alanthracene, chrysene and
benzo[a]pyrene in new mouse assay, Carcinogenesis 1, 1317-1322.

Yang, L. L., Maher, V. M. and McCormic J. J. (1980), Error-free
excision of the cytotoxic, mutagenic N -deoxyguanosine DNA adduct
formed in human fibroblasts by (1)43 ,80 -dihydroxy-9a ,loa -epoxy-
7,8,9,10-tetrahydrobenzo[a]pyrene, Proc. Natl. Acad. Sci. USA, 11,
5933-5937.

CHAPTER I

LITERATURE REVIEN

A. Role of Polycyclic Aromatic Hydrocarbons and Their Nitrated
Derivatives in Mutagenesis and Carcinogenesis

Polycyclic aromatic hydrocarbons (PAH) are environmental
contaminants which are metabolized in a wide variety of species to
reactive electrophiles that bind to cellular contituents and lead to
mutations and cancer (Miller and Miller, 1974; Sims and Grover, 1974).
Examples include pyrene, fluoranthene, benzo[a]anthracene, chrysene,
benzo[e]pyrene, benzo[h,g,i]pyrene, coronene, and benzo[a]pyrene (BP).
The most extensively studied PAH is BP which is ubiquitous in the
environment because it results from the incomplete combustion of
organic materials. It has been estimated that about 1300 tons of the
carcinogen BP are released into the environment of the United State
each ,year from sources such as heat and power' generation, refuse
buring, and coke production (National Academy of Science Reports, USA,
1972).

The nitrated PAH are also found in the environment, e.g., in
diesel emission particles, fly ash, photocopier fluids, cigarette
smoke, and emissions of fuel burners (Rosenkranz gt 61., 1980; McCoy
and Rosenkranz, 1982; Rosenkranz, 1982; Tokiwa gt 61., 1985).
Examples include 1,3-, 1,6-, and 1,8-dinitropyrene (DNP), 7-
nitrobenzo[a]anthracene, 6-nitrochrysene, 4-nitropyrene, 6-
nitrobenzola]pyrene, and l-nitropyrene (l-NP). The predominant

nitroarene in environmental samples appears to be l-NP which is

9
produced in various combustion processes, including incomplete burning
of diesel (Rosenkranz, 1982). l-NP has been shown to account for 20%
to 30% of the "direct-acting" mutagenicity of diesel emission particles
(Rappaport gt 61., 1980; Rosenkranz, 1982).

The principal reactive metabolite of BP is the gum isomer of
(i)BP-7,8,-diol-9,10-epoxide (BPDE) which binds to DNA mainly at the N2
position of guanine (Figure 1a), with minor binding to adenine
(Weinstein gt 61., 1976; Meehan gt 61., 1977). The reactive metabolite
of l-NP, formed by nitroreduction to the partially reduced intermediate
metabolite, l-nitrosopyrene (l-NOP), followed by fUrther reduction to
N-hydroxy-l-aminopyrene, binds covalently to DNA predominantly the C8
position of guanine (Heflich gt 61., 1985) (Figure 1b). The kinds of
tumors formed by the parent carcinogens differ and since the DNA
adducts they form with guanine are located in different positions in
the DNA helix, the molecular mechanisms of mutagenesis by these two
carcinogens and their biological effectiveness may chffer

significantly.

1. Metabolic activation of benzo[a]pyrene and l-nitropyrene

Neither BP nor NP bind directly to DNA, RNA or proteins. They need
to be activated to reactive metabolites through enzymatic metabolism.
BP requires metabolic activation by the microsomal enzymes to exert
several biological effects including cytotoxicity, mutagenicity, and
carcinogenicity. BP was found to be metabolized to arene oxides,
phenols, quinones, dihydrodiols, and diolepoxides (Yang gt 61., 1978).
These metabolites were resolved by high-pressure liquid chromatography

(Yang gt 61., 1978). The pathways of BP metabolism catalyzed by rat

10

Figure 1. Structure of the major BPDE-DNA adduct and l-NOP-
DNA adduct. a. BPDE-Nz-guanine adduct. From
Weinstein gt gt. (1976), Science 166, 592-595.
b. 1-NOP-C8-guanine adduct.

12

liver* microsomes are summarized in Figure Z (Yang gt ._1., 1977).
Studies of the metabolism-induced binding of BP and its metabolites to
DNA have provided evidence that 7,8-diol-9,10-epoxides of this
carcinogen are responsible for most of these bound adducts (Borgen gt
g1., 1973; Sims gt gfl,, 1974). The ultimate metabolites of BP, the
7,8-diol-9,10-epoxides, exist as a pair of diastereomers in which the
7-hydroxyl group is either gig (510 form isomer) or ttggg (ggti form
isomer) to the 9,10 epoxide. Each diastereomeric 7,8-diol-9,10-epoxide
can be resolved into a pair of optical enantiomers (Yagi gt g1., 1977).
The structures of these four compounds are shown in Figure 3. The form
studied in this thesis (Chpater III and IV) is the racemic mixture of
the gm isomers, i.e., (1)48, 8a-dihydroxy-9a,loa-epoxy-7,8,9,10-
tetrahydrobenzo[a]pyrene which I abbreviate as BPDE.

In mammalian cells and bacteria, l-NP is predominantly activated
metabolically through nitroreduction. Reduction of the nitro moiety of
NP appears to be a critical step in its metabolic activation since
mutation assays conducted with nitroreductase-deficient strains of
Sglmongllg typhimgrium show much lower reversion frequencies than
those obtained with nitroreductase-proficient tester strains (Wang gt
g1., 1980; McCoy gt _l., 1981). The scheme proposed by Heflich gt g1.
(1985) for the reductive metabolism of l-NP to a derivative capable of
binding to DNA is illustrated in Figure 4. The rate limiting step is
enzymatic reduction to an intermediate product, l-NOP (Heflich et al.,
1985). 'The observation that l-NOP forms the same DNA adduct as l-NP
and is more mutagenic than l-NP in the Salmonella typhimurium strain
'DA1958 (Heflich gt g1., 1985; Beland gt gl., 1986) as well as in human
di poid fibroblasts (Patton gt g1., 1987), suggests that l-NOP is an

13

Figure 2. Mechanism of enzymatic activation of BP to BP
7,8-diol-9,10-epoxides. Abbreviations: MFO,
mixied-function oxidases; EH, epoxide
hydratase. From Yang gt g1. (1977), Science
166, 1199-1201.

14

.2..I9.Qo§ 3:0:733—65

+ Hm 09.nooo...o.u
Ox Illa :
. .. HM....I
a... V

3...-..35..\\

\2&\O(Z
02
-HH
1
coucmu 5...?

.mo_=uo_oEo5mE 2 mEnEm

7...

.0... 6.7.2.1. 3:23...» conic—0.02.3

 

3:8-.935 3:... “a .03.: /o O..=2
x
O... a 093000-33
V o: + :o

:9 £5.” 510:

TI V 0

10¢: .o . a

33.3. 0‘24\ .6

1&012

cm.“

15

Figure 3. Structure and nomenclature of four isomers of
benzo[a]pyrene diolepoxide. From Yang gt g1.
(1978), in: Polycyclic Hydrocgrbons ggg
6ggggt, Vol. I (Gelboin and Ts’0, eds.),

Academic Press, pp. 205-231.

16

 

r-7. t-8-dihvdtoxv- t-9. l O'OXY'
7.8.9.10-terranvorooenzola lpvrene

dict epoxide I

arm isomer

1:1]: Bil-dhvdroxv-Sfl. 100-900“-
7.8.9.10-tenanv0robenzoia lpyreno

I z ’7 (lamination-9c. IDs-epoxy-
7.8.9.lO-tetranvomoenzoia 'pvrene

dlot 600102 2

n7, t-8-di'vvdtoxy- 09. lO-oxy-
7.8.9.10-teuanydrobenzola )pyrena

diol epoxide II
syn isomer

( :17c8II-dihvdroxy-9c iOa-ODO‘W
7.8.9.10-tetranydrobenzolalpyrm

l: 170.8t-dihvdroxv-9ii, Ion-epoxy-
7.8.9.10-tettanyorobenzoia Iowan.

d|OI epoxide l

17

Figure 4. The reductive metabolism of 1-nitropyrene.
From Patton gt g1. (1987), in: Polynuclear
Aromgtic Hydrocarbons: A decade of progress,
(Cooke and Dennis, eds.), Battelle Press, pp.

678-698.

18

35.2430 mlov PUDOO< <20

 

 

 

wzmm>aoz__>_<l_
I>XOmQ>IIZ mzmm>d0momtzic

@ 0

 

mzwm>momtzip

 

z@@@ A .. N z@@@ A .. ...

IO\ x: =
o

@

 

wzmm>domtzl mo ZO_._.<>_._.U<

19
intermediate in the mutagenic activation of l-NP. ‘Those studies
indicate that the critical first step, enzymatic reduction to l-NOP, is
followed by a subsequent reduction to N-hydroxy-l-aminopyrene and the
acid-catalyzed formation of a nitrenium intermediate which reacts
readily with DNA. l-NOP is the metabolite of l-NP studied in this
thesis (Chapter V).

2. DNA lesions produced by reactive metabolites of benzo[a]pyrene and
l-nitropyrene

It is now 'widely accepted that the covalent interactions of
carcinogens with cellular macromolecules, particularly DNA, are
essential initial steps in the process of mutagenesis and
carcinogenesis (Miller and Mliier, 1974; Rajalakshmi gt g1., 1982;
Weisburger and Williams, 1982). Studies by Sims gt, l. (1974),
Ivanovic gt g1. (1976), Meehan gt g1. (1976), Weinstein gt g1. (1976),
and Jeffrey gt g1, (1977) have indicated that the ggtt isomer of the
7,8-diol-9,10-epoxide of BP is the molecular species responsible for
the major covalent binding of metabolized forms of BP to nucleic acids
in hamster embryo cells, bovine, and human bronchial explants. Meehan
gt g1. (1976) showed that treated with DNA in ytttg, 92% of the total
stable covalent adducts formed at deoxyguanosine, 5% at deoxyadenosine,
3% at deoxycytidine, and 0% at deoxythymidine. These investigators
also studied the asymmetric binding to double-stranded DNA of the two
stereoisomers of the 7,8-diol-9,10-epoxide of BP and showed that the
(+)-enantiomers contribute most of the binding, e.g., 90% of
deoxyguanosine adducts corrresponded to adduct formation with the (+)

and and only 10% were with the (-) enantiomers. In contrast, only one

20

deoxyguanosinie adduct was formed using BP itself activated by a
microsomal system. Straub gt ,gl. (1977) further isolated seven
distinct products from the reaction of the racemic mixture of the ggti
isomer of the 7,8-diol-9,10-epoxide (BPDE) with thymus DNA and
determined the structure of six of these products. The structure of
these guanine, adenine, and cytosine adducts is shown in Figure 5a, b,
and c. Each of these structures represents two diastereomers (each of
the two enantiomers of BPDE reacted with the enantiomeric
deoxyribonucleoside). Those adducts are formed between the C-10
position of the BPDE and the N2 position of guanine, the N4 position
of adenine, and the N4 position of cytosine. The major adduct
corresponds to the binding of BPDE through the C-10 position to the N2
position of guanine.

Minor adducts of BPDE at the 06 and N7 position of guanine have
been identified (Osborne gt g1., 1978; 1981). The minor adduct at the
N7 position of guanine (Figure 5d) was spontaneously lost from the DNA
within a few hours and resulted in the formation of alkali-labile
apurinic sites (Osborne and Merrifield, 1985). It was found that the
06 and N7 guanine products were derived mainly from reaction of the
(-)-enantioner of the ggti isomers of the 7,8-diol-9,10-epoxide of BP
(Osborne gt _l_., 1981). Unstable BPDE adducts which cause single
strand breaks were detected in supercoiled DNA at neutral pH (Gamper gt
gl., 1977), but this strand scission represents less than 1% of the DNA
modification by BPDE.

Howard gt g1. (1983) identified one major and two minor adducts of
1-NP when it was reacted with calf thymus DNA in the presence of a

metabolic activation system. The major adduct was characterized as

Figure 5.

21

Structure of guanine-, adenine-, and cytosine-BPDE
adducts. a. Nz-guanine adduct. b. N4-guanine adduct.

c. N4-cytosine adduct. d. N7-guanine adduct. Figure 5a,
b, and c adapted from Straub e_t_ _l_. (1977), Proc. Natl.
Acad. Sci. USA: 16, 5285-5289. Figure 5d adapted from
Osborne and Merrifield (1985), Chem.-Biol. Interactions:

§_3_, 183-195.

22

~ (inn
0 0 o
”0““.0

OH HO

O.
"’OH
NH

Q0

\

6

OH

HO
H0
0300;: m... %%

2*

l :10 H,N N

23
N-(deoxyguanosin-B-yl)-1-aminopyrene (Figure 1). The minor adducts
appear to be decomposition products of the major adduct. When calf
thymus DNA was treated with l-NOP in the presence of ascorbic acid as a
source of reduction, the same major DNA adduct was formed as that
formed by l-NP in the presence of a metabolizing system (Heflich gt
g1., 1985). 16 2119 studies of the formation of l-NP and l-NOP adducts
showed that both compounds formed the same, single major DNA adduct,
N-(deoxyguanosin-B-yl)-1-aminopyrene in bacteria (Howard gt g1., 1983;
Heflilch gt g1., 1985) as well as in CHO cells (Heflich gt g1., 1986)
and in human diploid fibroblasts (Beland gt g1., 1986; Patton gt gl.,

1937).

3. Mtagenicity of reactive derivatives of benzo[a]pyrene, 1-
nitropyrene, and dinitropyrenes.

The inherent mutagenic activity of BP derivatives has been
exanmined in §almonellg typhimurium strains and in the Chinese hamster
cell line V79. 0f more than 30 BP metabolites and derivatives tested
for mutagenic activity, the BPDE were the most potent mutagenic
compounds in strains TA98 and TAIOO and in V79 cells (Newbold and
Brookes, 1976; Wislocki gt gl., 1976; Wood gt gl., 1976). Substantial
differences in the mutagenic activities of the optically pure (+)- and
(-)-enantiomers of the ggtt or gyg isomers of the 7,8-diol-9,10-epoxide
have been reported by Wood gt g1. (1977). In V79 cells, (+)ggt1-isomer
was the most mutagenic for the four isomers. In contrast,
(-)§m-isomer was a more potent mutagen than the other three optical

isomers in Salmonellg typhimgrium.

Some nitropyrenes or dinitropyrenes are powerful mutagens, while

24
others are devoid of activity. The result depends on the type of
bacteria or mammalian cells used for mutagenesis assay. Although 1-,
3-, and 6-nitropyrene and the three dinitropyrenes (DNP), 1,3-, 1,6-,
and 1,8-DNP are direct-acting mutagens in the: mutagenicity assay,
1,6-DNP and 1,8-DNP have more potential to induce revertants than does
l-NP. The mutagenicity of l-NP is greatly reduced when tested in
bacterial strains deficient in nitroreductase, suggesting that this
enzyme is required for expression of 1-NP-induced mutagenicity. In
contrast, 1,6-DNP and 1,8-DNP did not depend on the classical
introreductase for maximal activity (Mermelstein gt gl., 1981;
Rosenkranz and Mermelstein, 1983). In studies of the mutagenicity of
nitropyrenes in cultured mammalian cells, I-NP was active in inducing
thioguanine resistant mutants in human HepGZ hepatoma cells, whereas
1,6-DNP and 1,8-DNP were inactive (Eddy gt gl., 1986). Conversely, in
Chinese hamster V79 cells, (Takayama gt gl., 1983), CHO cells (Li and
Dutcher, 1983), or lung fibroblasts (Nakayatsu gt g1., 1982), 1,6-DNP
and 1,8-DNP exhibited maximal activity while l-NP showed almost no
activity. These findings suggest that the diverse enzymes involved in
nitroreductions in various systems differ in their specificity and
catalytic mechanism. Eddy gt g1. (1986) suggested that the biological
activity of nitropyrenes and dinitropyrenes was correlated with the
number of electrons involved in the rate-determining step of
nitroreduction, e.g., for l-NP, one electron transfer was involved in
the reduction of the first nitro function; for 1,6-DNP and 1,8-DNP, two

electrons were tranferred (Klopman gt g1., 1984).

25

4. Carcinogenicity of benzo[a]pyrene, benzo[a]pyrene diolepoxides, 1-
nitropyrene, and dinitropyrenes

Benzo[a]pyrene forms tumors predominantly in the lung (Levin gt
gt., 1978; Wislocki gt l., 1986). The induction of pulmonary tumors
in the newborn mouse demonstrated that BPDE is ultimate carcinogenic
metabolite of BP (Kapitulnik gt gt., 1977; 1978). Buening gt gt.
(1978) further tested the tumorigenic activity of each of the four
optically pure isomers of the 7,8-diol-9,10-epoxides of BP on newborn
unice. 'They found that the (+)ggtt-isomer' had exceptional
tumorigenicity, whereas BP and the other three optically pure isomers
had no or little activity at a total dose of 7 or 14 nmol used.
Bresnick gt gt. (1977) reported that BPDE is the most potent inducers
of mouse skin epidermal hyperplasia of some 20 BP derivatives tested.
In another different tumor model, the initiation-promotion system of
tumorigenesis on mouse skin, used to evaluate the carcinogenic
potential of BP and its derivatives, BPDE was less active than BP as
tumor initiators (Slaga gt gt., 1976; 1977). This result may reflect
the fact that BPDE is an extremely unstable compound. Slaga gt gt.
(1979) further analyzed the skin tumor-intiating activity of the four
isomers. They found that the (+)ggtt-isomer was the most active tumor
initiator (60% as active as BP) among the four isomers. When the
(+)ggtt-isomer was given in fractionated doses, the skin
tumor-intiating activity increased and was comparable to that of BP.
These results, together with the data of the mutagenesis studies with
V79 cells (Huberman gt _t., 1976; Wood gt gt., 1977), indicate that the
(+)ggt1-isomer is the ultimate carcinogenic metabolite of BP in

mammalian cells.

26

Ohgaki gt gt. (1982) showed that l-NP produced subcutaneous
tumors at site of injection in 8 out of 17 male F344/Duer rats.
However, Ohgaki reported later (quoted in a meeting report by Serres
and Matsushima, 1986) that the batch of 1-NP used for the previous
study had been contaminated with 0.2% 1,3-DNP, 0.3% 1,6-DNP, and 0.3%
1,8-DNP, the three dinitropyrenes were found to produce subcutaneous
tumors at site of injection in ten out of ten rats. No tumors were
found when the l-NP experiment was repeated using a highly purified
sample. The poor carcinogenicity of l-NP in BALB/C mice was also
reported by Tokiwa at that same meeting. The weak mutagenicity and
carcinogenicity of l-NP in rodents may related to the low ability of
rodent cells to carry out the reduction of l-NP to a reactive
metabolite. In contrast, Hirose gt gt. (1984) showed the ability of 1-
NP to induce malignant fibrous histiocytomas in ten out of 31 male and
nine out of 32 female animals at sites of injection, and mamary gland
tumors in 15 out of 32 femal newborn Sprague-Dawley derived CD rats.
The ability of NP to induce lung tumors in A/J mice has also been shown
by El-Bayoumy gt gt. (1984). Beland and his coworkers have reported
that i.p. injection of l-NOP at total doses of 2800 nmols per mouse
induced 21% - 28% liver tumors in new born mouse, Liver tumors did not
occur in females, and lung tumors were not induced in both male and
female animals (Wislocki gt l., 1986). These investigators also
reported that l-NOP caused liver tumors in 45% of the male and 9% of
the females at 700 nmols per mouse by i.p. injection (Wislocki gt gt.,

1986) .

27

5. The specific kinds of mutations induced by BPDE and l-NOP

Several investigators have examined the kinds of mutations induced
by BPDE in bacteria, but few have studied this question in mammalian
cells. In the Salmonellg reverse mutation assay, BPDE caused a much
higher mutagenic response in a missense mutation tester strain (TA100)
than in a frameshift mutation tester strain (TA98) (Malaveille gt gt.,
1977; Wood gt _t., 1977; Fahl gt _t., 1981; Aust gt _t., 1984). The
most extensive studies of the specific kinds of base changes induced by
BPDE in 6. ggtt was carried out by Eisenstadt gt gt. (1982) who showed
that BPDE significantly induces G-C - T-A transversions in the tggt
gene of 6. ggtt by genetic mapping. Using bacterial plasmid
transformation and recombinant DNA techniques, Mizusawa gt gt. (1981a)
treated a bacterial plasmid carrying a marker gene with BPDE and
transferred this treated plasmid into various 6. Q11 recipients to
study the effect of BPDE on plasmid survival in bacteria and on the
frequency of mutants. Agarose gel electrophoresis analysis of
restriction endonucleotide flggll digested mutant plasmids showed that
94% did not involve detectable alterations in the size of the DNA
fragments, indicating that BPDE is a point mutagen. Limited data on
the specific kinds of mutations induced at DNA sequence level by BPDE
in bacteria plasmid have been reported. Mizusawa gt gt. (1981b) found
that two BPDE-induced mutants exhibited T-A insertions and one
contained a G-C base pair deletion. Chakrabarti gt ,gt., (1984)
reported that three out of five BPDE-induced mutant plasmids had G-C
base pair deletions, one contained a G-C - T-A transversion, and one
had a A°T - G-C transition.

The specific kinds of mutations induced at the DNA sequence level

28

in mammalian cells have never been studied. Kfing and Brookes (1984)
showed that BPDE induced mutations in V79 Chinese hamster cells were
predominantly point mutations as detected by DNA hybridization. The
data from Aust gt gt. (1984) also indicate that BPDE is a point
mutation mutagen because it does not cause gross alterations expected
to completely inactivate the gene coding for elongation factor 2, which
is involved in diphtheria toxin resistance.

Both 1-NP and 1-NOP induce revertants in §glmgngllg typhimgrigm
frameshift tester strains TA1537, TA1538, and TA98 (Mermelstein gt gt.,
1981; Heflich gt gfl,, 1985). No revertant has been found in strain
TA1535 (a base substitution indicator) and only some revertants have
been found in strain TAIOO (a missense mutation indicator) when treated
with l-NP. Little is known about the kinds of mutations induced by 1-

MP or 1-NOP in either 6. cgli or mammalian cells.

8. Systems Presently Available for Investigating Mutagenesis in
Mammalian Cells at the Molecular Level

The molecular mechanisms responsible for mutational changes in
mammalian calls are not well understood. In this decade, the
development of recombinant DNA techniques has enabled researchers to
identify mutational changes in genes by direct nucleic acid analysis.
These molecular mutagenesis systems resolve the difficulties of
previous indirect mutational analysis, i.e., detection of phenotypic
changes or gene products. Some of these systems for studies of the
molecular nature of spontaneous and induced mutations in mammalian
cells have been reviewed by Lehmann (1985) and Thacker (1985). During

the past two years, still more elegant systems for such molecular

29
studies of mutations have been developed (see below), making this an
exciting era in which to investigate the specific kinds of mutations at
the sequence level caused by specific mutagens and carcinogens.

The recent applications of these recombinant DNA techniques to
analysis of spontaneous and induced mutations in cultured mamalian
cells include: 1) mutations in endogenous mammalian genomic genes, 2)
mutations in bacterial genes integrated into the mammalian genome, 3)
mutations in viral vectors, 4) mutations in transiently replicating
shuttle vectors, and 5) mutations in stably replicating shuttle

vectors.

1. Endogenous mammalian gene systems

In these systems, the particular genes studied are those for which
cloned cDNA sequences have been isolated. The entire genomic DNA is
digested with restiction enzymes and the DNA fragments are
electrophoresed. Southern blotting of these restiction fragments,
followed by hybridization with the cloned cDNA or other probes, permits
analysis of mutations in the structural genes. Deletions or insertions
of genetic material result in alteration in the size of hybridizing
restriction fragments. Base alteration mutations localized to
restriction endonucleoase recognition sites in the particular gene can
also be detected, since the nucleotide alterations caused by the
mutations lead to loss or gain of the recognition sites. Mutant
endogenous genes can be further analyzed by sequencing the base changes
causing the alterations of the recognition sites (Nalbantoglu gt gt.,
1987).

The amount and the size of the mRNA in the mutant cells can also

30
be analysed by transfer of the RNA from mutant cells onto filter
papers, followed by hybridization with the cloned cDNA probe (Northern
blotting). Changes in the blotting pattern can reveal alterations in

gene expression.

a. Studies with the hypoxanthine phosphoribosyltransferase (DELL) gene
The successful production of cDNA. clones of 'the hypoxanthine
phosphorisyltransferase (hart) gene (Brennand gt gt., I982; Konecki gt
l., 1982; Jolly {gt .gt., 1983) has enabled molecular studies of
mutations in this locus. The mouse gene is ~33kb and contains nine
exons varying in length from 18 bases to 593 bases (Melton gt gt.,
1984). These is close homology of the HPRT protein sequence of mouse,
hamster and human (Konecki gt gt., 1982) and the organization of mouse
and human DD£L gene (Yang gt gt., 1984), suggesting strong evolutionary
conservation of this gene. In the process of characterizing the ngtt
gene of a mouse cell line, Melton gt gt. (1984) reported that the
spontaneous ngtt mutant used for seletion of an overproducing
revertant contains a single base transition of G -+ A in exon 8.
Fuscoe gt gt. (1983) analysed ten spontaneous and nine UV-induced hgtt'
mutants of V79 Chinese hamster cells. Only one spontaneous and one UV-
induced mutant exhibited large deletions in the DELL gene when
hybridized to 1.1 kb mouse cDNA or 1.3 kb hamster cDNA probe. Fenwick
gt gt. (1984) also studied a hgtt mutant of Chinese hamster cells and
its ngtt+ revertants with a cloned cDNA for ngtt. This mutant line may
contain a point mutation because no alterations in restriction patterns
were observed. Some of the spontaneous revertants produced a much

stronger hybridizing signal, indicating that the gene had been

31
amplified up to ten to twenty times. This resulted in a corresponding
overproduction of mRNA. Overproduction of the defective gene product
was presumably responsible for phenotypic reversion.

In a related study, King and Brookes (1984) found that each of
eleven 8-azaguanine resistant, cross-reacting material-negative Chinese
hamster cell mutants induced by BPDE showed no detectable alteration in
gene structure when six different restriction enzyme digested mutant
DNA fragments were probed with the 1.3 kb hamster cDNA. However, four
of these mutants showed much less amount of 6m mRNA than did wild
type cells. Two of these four mutants also showed a size reduction by
about 300 bases. The truncated mRNA was suggested to arise from
incorrect processing, perhaps by point mutation of’ a splice site.
Their results suggest that BPDE functions primarily as a point mutagen.

In humans, complete deficiency of HPRT enzyme leads to the Lesch-
Nyhan (L-N) syndrome and partial deficiency is associated with gouty
arthritis. This fact has also drawn attention to this gene. Using
mouse cDNA probes, Yang g _t. (1984) examined DNA from cells of 28
untreated L-N patients. These DNA samples were digested with four
different restriction enzymes and analysed by Southern blotting using a
full-lenth mouse ggtt cDNA probe. These investigators found that 23
L-N patients out of the 28 surveyed did not show major 11th gene
alterations. Assuming that the entire human gene is organized
similarly to the mouse gene, and using cDNA probes encoding only parts
of the gene, it was possible to show that each of the five L-N patient
with an altered gm; gene carried a different mutation. Three had
deletions at different positions, one had alterations in exons 4, 5

and/or 6, and one had a suggested duplication involving in exons 2 and

32

b. Studies with the Adenine Phosphoribosyltransferase (gggt) gene

Adenine phosphoribosyltransferase (ggtt) gene is a particularly
attractive locus for analysing endogenous mutational events. Since
this gene is located on autosomal chromosome, the interpretation of
molecular data is complicated by the presence of two copies of the
gene. However, a strain hemizygous for ,ggtt has been identified
(Bradley and Letovanec, 1982; Nalbantoglu _e_t gt., 1983), making it
possible to collect single-step spontaneous and induced mutants and
allow mutagenesis studies in this locus. Furthermore, the relatively
small size of ggtt gene facilitated its cloning from the genomic DNA of
CHO cells (Lowy gt gt., 1980): a 3.8 kb 6gmHI fragment contains all the
sequences necessary to 'transform .APRT’ CHO cells to APRT+. 'This
allows, in many instances, localization of mutational events in 1129 to
restriction endonuclease sites and rough mapping of deletion or
insertion termini (Nalbantoglu gt fl“ 1983). Another advantage of
using gp_rt as a model for the study of gene variation in mammalian
cells is that it is a true endogenous gene with the low mutational rate
inherent to such loci (4 x 10'8 per cell per generation; Nalbantoglu gt
gt., 1987). However, an odd feature of these loci is that the rodent
ggtt genes are not good probes for the human ggtt gene in Southern blot
hybridizations (Stambrook gt gt., 1984). Furthermore, the DNA sequence
analyses of these loci were limited to mutants which lost restriction
endonuclease sites. This limitation eliminates the possibility that
mutations might exist outside these regions, and thus the true

Inutational spectrum at the DNA sequence of this locus cannot be

33
obtained.

Meuth and his colleagues (Meuth and Arrand, 1982; Nalbantoglu gt
gt., 1983: Goncalves gt gt., 1984) analyzed mutations induced in the
hamster gpr_t gene by ethyl methanesulfonate (EMS). These studies
showed that most mutants induced by EMS in CHO cells have unaltered
restriction fragment patterns, suggesting that EMS is a point mutagen.
Most of the spontaneous mutations were also point mutations, but a
small number of deletions and an insertion were also detected. A
number of these mutations were localized to restriction endonuclease
recognition sites in the gm gene, since the nucleotide alterations
produced by the mutations led to the loss or' gain of ‘the site.
Nalbantoglu gt gt. (1987) further cloned those spontaneous ggtt genes
and sequenced the base changes causing the alterations of the sites.
0f the 12 mutant genes analyzed, five contained single base pairs
transitions (both G-C -+ A-T and A-T -+ G-C), two exhibited
transversions (G-C - T-A and A-T -+ T.A), one showed multiple
mutations, i.e., a G°C -+ T°A transition next to a single base pair
insertion, and four contained more complex changes, involving small

deletions or duplications.

c. Studies with the dihydrofolate reductase (d_h_fii_') gene

As for DELL. the 666: genes have proved to be considerably larger
than their mRNA: in the mouse "31 kb (Crouse gt gt., 1982), Chinese
hamster "25 kb (Carothers gt gt., 1983), and human ~30 kb (Chen gt fl.,
1984). The genes from these species are similar in being split into
six exons with conservation of intron-exon junctions, but having some

divergence in intron length and sequence. Several different-sized mRNA

34
coding for DHFR have been found in mouse, hamster, and human cells
(Crouse gt gt., 1982; Carothers gt gt., 1983; Morandi gt gt., 1982).

Graf and Chasin (1982) analyzed Y-irradiation induced mutations
at the M locus in a presumed heterzygote CHO cells using a mouse
cDNA probe. Two out of nine mutants analyzed contained altered
restriction patterns consistent with a deletion, insertion or
rearrangement.

In these endogenous mammalian gene systems, it is obviously useful
to select cell lines lacking one copy of autosomal genes to rapidly and
unambiguously identify mutations in the remaining copy (e.g., ggtt and
d_hfi seletion). But the use of these systems to investigate point
mutations is difficult. The use of a sufficiently large number of
restriction enzymes to detect if a base substitution or frameshift is
involved is very time-consuming and expensive, espacially for a large
gene. Cloning of mutants into bacteriophage makes it possible to
investigate mutation at the level of base alterations by sequencing.
However, this method is not widely used because of the time and labor

involved.

2. Integrated exogenous gene systems

The above mentioned difficulties of analyzing native genes have
been circumvented by introducing small cloned genes as targets for
mutagenesis. These genes can be integrated into the mammalian cell
genome or introduced as autonomously replicating genes. The use of
genes, coding for selectable markers,and integrated into genomic DNA
for mutation studies has the advantage that the genes behave like

chromosomal DNA with respect to replication and mutagenesis. The

35
target genes, e.g., xanthine (guanine) phosphoribosyltransferase (ggt),
ggpt, and thymidine kinase (t3), are carried in vectors, such as a
SV40-based plasmid (Ashman and Davidson, 1984; Tindall gt gt., 1984) or
phage vector (Glazer gt gt ., 1986) or retroviral vector (Ashman and
Davidson, 1987). In most of these systems (M and t6), the target
genes are transfected into a deficient cell line (TK' or HPRT’) and
integrated into genomic DNA. Transformants containing the integrated
gene are treated with mutagens and selected in a special medium for the
mutation assay. For example, Perucho gt gt. (1980) and Robins gt gt.
(1981) have used the Herpes simplex virus (HSV)-tt gene to transform
TK-deficient mouse, human, or rat cells, and studied mutations of HSV-
tk transformants by selection in medium containing bromodeoxyuridine
(BrdU). They reported that most of the spontaneous mutation consisted
of various-sized deletions of the introduced HSV-t6 gene. This
phenomenom may be the result of the instability of newly-introduced

genes (Scangos gt _t., 1981).

a. Studies with the ggt gene

Thacker gt gt. (1983) reported that most of mutants they detected
after introducing vector DNA into genome were due to complete loss of
the integrated target gene. They then established a hamster V79 cell
line containing a single ggt gene and expressing ggt stably even when
maintained for generations in non-selective conditions. These cells
showed a spontaneous ggt‘ mutant frequency of 5 x 10'4. The frequency
of mutations induced in the ggt gene by x-rays and EMS was about 20-
fold high than that found with the endogenous DELL gene in V79 cells.

However, the target gene of the majority of these gpt' mutants also

36
proved to be completely deleted.

Ashman and Davidson (1987) have introduced a retroviral shuttle
vector containing the ggt gene into a .bpnt' mouse cell line by
infection. They isolated a stable cell line which contains a single
copy of the vector integrated into chromosomal DNA in a proviral form.
Cells with spontaneous mutations in the ggt gene were selected by their
resistance to 6-thioguanine and then were fused with C05 cells for
recovery of the mutant genes. The majority (29/43) of the mutants
contained deletions with 19 out of 29 containing a deletion of three
base pairs which resulted in the "in frame" loss of an aspartic acid
codon. Eleven mutants contained single base substitutions, viz., six
transversions and five transitions. These investigators found no
obvious preference for any particular type of base substitution among
the spontaneous point mutants sequenced. They also studied mutations
induced by EMS and BrdU. They reported that both chemicals induced
mutants and that based on the restriction enzyme analysis, the mutants

contained 90% putative point mutations.

b. Studies with the 6626 gene

Glazer fl gt., (1986) developed a lambda phage vector carrying
the neomycin resistance gene which allows selection for stable
integration of the vector into the mammalian genome. The vector also
carries a small gene, 6666, as a target for mutations. They
transfected this vector into a mouse L-cell line and established stable
lines with multiple copies of this lambda phage vector integrated into
mouse genomic DNA. They efficiently rescued viable phage from high

molecular weight mouse cellular DNA using lambda _i_n_ vitrg packaging

37

extracts and found a negligible background of mutant phage (0/54,605).
With this system, they exposed mouse cells carrying multiple copies of
the lambda vector in the genome to 254 nm ultraviolet (UV) light. Of
78,510 phage were rescued, eight SUDE mutants were found. DNA sequence
analysis of the mutants suggests that the primary site of UV
mutagenesis in mammalian cells was at pyrimidine-cytosine sequences,
and that the most frequent mutation was a G°C -+ A'T transition.

These recently developed integrated vector systems (Glazer gt gt.,
1986; Ashman and Davidson, 1987) allow one to recover the integrated
mutated genes and the integrated gene behaves somewhat like chromosomal
DNA with respect to replication. The major advantage of these systems
over the endogenous gene systems is that the small defined target genes
(6666; ggt) enable rapid sequencing. Thus the molecular mechanism

induced by mutagens at the base changes level can be easily studied.

3. Viral vector systems

Animal virus, such as HSV and SV40, can themselves be used as the
target for mutations in mammalian cells. Those viruses are useful in
such a study because they replicate as minichromosomes in mamalian
host cells and the mutants can be rescued from the cells, mapped and
sequenced. Dasgupta and Summers (1978) have made use of the genetic
system involving the TK locus of HSV-1 to reveal that an "error-prone"
inducible UV-reactivation phenomenon exists in mammalian cells.
Although they did not determine the molecular nature of the mutants at
that times, it is posible to analyse those t3" mutants at the molecular
level at present.

Bourre and Sarasin (1983) used SV40 virus as a biological probe to

38
study the mutagenic effect of UV-irradiation at the molecular level.
They studied reversion of the tsA58 locus (a thermosensitive SV40
mutant of the large T antigen) in SV40 virus which had been UV-
irradiated i vitro. Monkey kidney cells were infected with UV-

 

irradiated tsA58 virus. Revertants of tsA58 were selected by growth of
the virus at the nonpermissive temperature. Then the reversion sites
were analysed by using the marker rescue technique and DNA sequencing.
Of 16 mutations sequenced, all were localized opposite a possible UV-
induced pyrimidine-pyrimidine lesion, suggesting targeted mutagenesis
of UV. The advantage of this system is that the spontaneous reversion
rate of the tsA58 is low ("10'5). One major disadvantage of using this
system is that frameshift mutations cannot be detected because under
their conditions, any frameshift mutation leads to the production of an
inactive T antigen which is lethal for the virus.

Using the same system, Gentil gt 11., (1986) studied mutations
induced by 2-(N-acetoxy-N-acetylamino)fluorene (AAAF). In more than
60% of independently isolated mutants, the molecular analysis of AAAF-
induced revertants revealed a hot spot for a base substitution mutation
not locaized opposite a major DNA lesion. These investigators
hypothesized that a specific DNA structure, stablized by an AAAF
adduct, is responsble for the hot spot. In a bacterial forward
mutation system, 90% of AAAF-induced mutations were frameshifts
(Koffel-Schwatz g_t_ l., 1984). The inability of the viral assay to be
used to detect frameshift mutations eliminates the chance of making
comparisons with the AAAF-induced mutations data in the prokaryotic

system.

39

4. Transiently Replicating Shuttle Vector Systems

Mutagenesis studies involving endogenous genes are not ideal for
analyzing mutations at the base-pair changes level. Shuttle vectors,
however, provide rapid and powerful tools for the study of mutagenesis
at the DNA sequence level in cultures mamalian cells. A shuttle
vector is a plasmid DNA that is capable of replicating in mamalian
cells and in bacteria. In many instances, such shuttle vectors contain
the origin of replication and large T-antigen from SV40 which allows
transient replication in mammalian cells, the bacteria origin of
replication for propagation in bacteria, a drug resistant gene (e.g.,
the gene for ampicillin resistance), and a variety of bacteria genes as
targets for mutagenesis, e.g., supfi, galfi (a gene coding for a
galactokinase), and lag1_ (a gene coding for the repressor' of the
lactose operon). The vectors are transfected into permissive mammalian
cells and allowed to replicate in the nucleus for several days.
Mutagenesis takes place during this time. The progeny vectors are then
extracted and separated from cellular DNA, and used to transform an
appropriate indicator bacteria strain for mutant identification at the
target gene. The mutation in the defined target genes can be easily
analyzed by restriction mapping, genetic techniques (only for _l__a_cI
gene), and DNA sequencing.

One particular advantage of the transiently replicating systems
over the stably replicating systems (see part 8.5) is that vectors can

be treated with mutagens in vitro and allowed to replicate in a variety

 

of cells which themselves have been treated or not treated with
mutagens. Therefore, transiently replicating vectors are particularly

useful for indirect mutagenesis studies, such as studies of SOS

40

inducible error-prone mutagenesis in mammalian cells.

a. High background mutant frequency of transfected vector DNA

A major problem recognized with the early developed shuttle vector
systems was the very high "spontaneous" mutant frequency in the
transfected vectors (0.2% - 10%) (Calos gt gt., 1983; Razzaque gt gt.,
1983; Ashman and Davidson, 1984; Lebkowski gt gl., 1984; Sarkar gt gt.,
1984). This mutant frequency was several orders of magnitude higher
than the spontaneous rate for the same genes in E. 9911 of 10'5 - 10'6
(Miller gt gl., 1977). It was also much higher than the spontaneous
mutation rate for genes in mammalian cells of 10'6 - 10'8 (Lewin gt
gt., 1980).

Studies by Lebkowski gt g1. (1984) indicated that the high
mutation frequency was due to damage incurred in vector DNA early after
its entry into the mamalian cells, rather than during replication,
since even nonreplicating molecules were subject to the effect. Other
evidence in support of this hypothesis comes from the finding by
Razzaque gt g1. (1984) that the mutation frequency did not increase
during the replication period. It has been shown that transfection of
DNA by CaPO4 coprecipitation, DEAE-dextran, protoplast fusion, or
electroporation, all led to elevated mutation frequencies (Calos Lt
g1., 1983; Razzaque gt gt., 1983; Lebkowski gt g1., 1984). The target
gene itself was not causing the high mutation frequency (Calos gt l.,
1983). Both point mutations and gross rearrangements were detected
(Calos gt gt., 1983; Lebkowski gt gt., 1984). It has been suggested
that intracellular DNA damage is the cause of the formation of

mutations in transfected DNA. If newly transfected DNA were subject to

41
attack by endonucleases, exonucleases, and ligases which are present in
the mammalian nucleus, deletions could readily be generated (Miller Q
_a_l_., 1984; Razzaque gt _a_l_., 1984)

Genetic mapping and DNA sequencing of the spontaneous point
mutants obtained in the 135,1 gene during replication in C057 cells
showed that the majority of the point mutations were base
substitutions (Miller gt gt., 1984). Of 93 independent base
substitutions obtained in El gene during passage through C057 cells,
all occurred at 6°C base pairs and were approximately equally divided
between 6°C -. A'T transitions and G-C -+ T-A transversions. These

investigators suggested that this specificity resulted from deamination

of cytosine and depurination of guanine.

b. Systems to minimize the high background frequency of mutants

Several groups have made efforts to design other transiently
replicating vector systems to achieve a lower background mutation
frequency than that observed previously. Lebkowski gt 11., (1984)
introduced a series of vectors into a variety of mamalian cells to
characterize further the mutational fate of transfected DNA. They
found that high mutation frequency appears to be the characteristic
outcome of transfection of DNA into mammalian cells. However, they
observed an approxiomately 10-fold lower mutation frequency when a
SV40-based vector, pJYMib, or a BK-based vector (8K is a human
papovavirus closely related to SV40), pBKib, were passaged in a human
embroynic kidney cell line, designated 293. They also noted that
replication of transfected vectors was more efficient in 293 cells.

Their finding that efficient replication of shuttle vectors in 293

42

cells was accompanied by a relatively low background frequency allowed
a shuttle vector system to be used successfully in this thesis for the
study of mutations induced by BPDE and l-NOP (see Chapter IV and V).
Using 293 cells, Calos and her coworkers have reported on 1gg1 mutants
induced by UV (Lebkowski gt al., 1985) and EMS (Lebkowski gt 11.,
1986). Lebkowski gt Q1. (1985) reported that UV predominantly induces
point mutation mutants. The frequency of point mutations was
increased S-fold with 50 J/m2 dose of UV. Using genetic mapping
techniques to analyze nonsense mutations and DNA sequencing techniques
to locate the nfissense mutations, they observed that UV specifically
induced G-C —+ A'T transitions (81%). More than 90% of the UV-induced
transitions involved pyrimidine-pyrimidine (py-py) sequences. Since
the two major photoproducts caused by UV, the cyclobutane pyrimidine
dimer and the (6-4) pyrimidine-cytosine (py-C) photoproduct, occur at
py-py sequences, the location of the majority of the mutation opposite
to these sequence argues that the UV mutagenesis in the human cells was
targeted to premutational lesions. The base substitution specificity
observed closely resembles that observed with UV light in _E. 9311
(Coulondre and Miller, 1977; Miller, 1985), suggesting that human and
bacterial cells respond similarly to damage from UV light. However,
frameshifts were seen in _E. gilt (60%), but occurred very rarely in
human cells (4%).

The base-pair changes induced in 293 cells by EMS *were also
determined using the identical 1gg1 shuttle system. ENS induced
predominantly point mutations which occurred at a frequency six times
above background. Ninety-eight percent (53/54) of the 1gg1 mutations

derived from EMS-treated human cells were G-C _. A-T transitions.

43

These results are similar to the data obtained for 1gg1 in E. ggli in
which 98% (680/691) of the mutations were also 6°C -+ A~T transitions
(Coulondre and Miller, 1977). The major mutagenic lesion caused by
alkylation agents is the O5-alkyl guanine (Singer gt g1., 1986). If
the 05-alkyl group is not removed by repair enzymes before replication,
mispairing of the 05-alkyl guanine with thymine can form, leading to a
full G-C -¢ A°T transition during the next round of replication. These
results also demonstrated targeted mutagenesis operating in mammalian
cells.

Seidman gt g1., (1985) constructed a plasmid so as to minimize the
recovery of these background deletions and to facilitate the detailed
characterization of the induced mutations. Their vector, p2189,
contains the origin of replication-and large T-antigen gene from SV40,
as well as the small target gene, gggfi, strategically located between
two genes essential for recovery of the plasmid in E. ggli, i.e., the
gene for ampicillin resistance and the bacterial origin of replication.
Mutations containing base substitutions and small deletions or
insertions in the region of ggpfi gene can be recovered. However, large
deletions and rearrangements in the region of _sggfi gene that extend
into either of the two flanking regions which are made up of genes
essential for recovery of the plasmid in bacteria are not recovered.
When this vector was allowed to replicate in a variety of mammalian
cells, it exhibited a low relatively low ‘frequency' of' spontaneous
deletions (Seidman gt_g1., 1985; Bredberg gt g1., 1986; Hauser gt g1.,
1986). The small size of the tRNA target gene permits rapid
sequencing of the entire gene in a single opteration, and thus all

kinds of mutations that eliminate the function of the suppressor tRNA

44
can be characterized. It has been shown that the gggf gene is highly
responsive to base changes. Among the 85 nucleotides making up the
tRNA structure, base changes at any one of at least 63 positions
results in a detectable phenotypic change (Celis and Piper, 1982;
Bredberg gt gt., 1986; Glazer gt g1., I986; Hauser gt al., 1986; Yang
gt g1., 1987a; 1987b).

Seidman gt gl. (1985) compared the replication efficiency of p2189
in the CV1 monkey cell line to that of SV40 viral DNA. The results
indicated that as much plasmid pZ189 DNA accumulated in the cells as
SV40 viral DNA. High efficiency of replication of p2189 has also been
observed in SV40 transformed human fibroblast cells, including XP
cells (Berdberg gt g1. 1986). This high efficiency of replication of
p2189 facilitates the recovery of plasmid DNA following transfection
with plasmids partially inactivated by mutagen treatment. In addition,
use of this vector allows a more detailed characterization of molecular
defect that may underlie individual sensitivities to DNA damaging
agents leading to increased risk of cancer and other consequences of
mutagenesis. For example, elegant studies to reveal the UV-induced
mutagenesis in mammalian cells have been done using this shuttle
vector. Hauser gt gt, (1986) treated the vector with UV light at a
dose of 500 J/mz, then transfected this UV-irradiated vector into CV1
monkey cells. They assayed the mutations that formed during the vector
replication in CV1 cells for 2 days. They found a 20-fold increase of
mutant frequency above background. Most of the UV-induced mutations
were point mutations (95%), primarily base substitutions. UV-induced
base substitutions all involved 6°C base pairs, despite the preference

for photoproduct formation at TT sites. These investigators suggested

45

that the preference for G-C -+ A-T transition among UV-induced mutation
reflects the fact that DNA polymerases reponsible for these mutations
prefer to insert adenine opposite photoproducts in the DNA. They also
reported that hot spots for UV mutagenesis did not correspond to hot
spots for UV-induced photoproduct formation when determined by a DNA
synthesis arrest assay. This conclusion was later supported by related
studies by Brash gt g1. (1987).

In an extended study, the UV-irradiated vector was treated with L.
ggli photolyase prior to transfection so that pyrimidine cyclobutane
dimers were removed selectively (Protic-Sabljic gt 11., 1986). Removal
of 90% of pyrimidine cyclobutane dimers decreased the mutagenic
activity frequency by 80%, suggesting that UV-induced cyclobutane
pyrimidine dimers are mutagenic in mammalian cells. There were
significantly fewer tandem double-base changes and 6°C -+ A-T
transitions after photoreactivation of tha DNA. These two alterations
may indicate that cyclobutane dimers are principally responsible for
these types of mutational changes and that (6-4) py-C photoproducts may
be more likely to cause single base transversions.

Using this vector, Bredberg gt 31., (1986) studied UV mutagenesis
in normal human cells as well as cells derived from xeroderma
pigmentosum (XP) patients (group A) whose cells have a profound defect
in DNA repair and are almost totally unable to excise UV photoproducts
and who are extremely sensitive to sunlight-induced skin cancer
(Robbins gt g1., 1974). The point mutation frequency increased up to
100-fold with UV dose. Mutations were infrequent at potential TT dimer
sites. Ninety-three precent of the mutant plasmids from XP cells

showed G-C -+ T-A transition compared to 73% in the normal cells. A

46
significantly lower frequency of transversion mutations was observed
with the XP cells (6%), compared to the normal cells (25%). These
investigators concluded that the 6°C -’ A°T transition (at sites other
than TT dimer sites) present in the XP spectrum may, therefore, be of
greater importance in UV-induced cutaneous carcinogenesis than are the
transversion and multiple base substitution mutations.

In the case of UV—irradiation, the high frequency of 6°C -+ A°T
transition in the normal human cells was similar to that seen in CV1
cells after in yitgg plasmid treatment with UV (Hauser gt 1t., 1986).
It was also similar to that observed using the 293 cells-1111 shuttle
system (Lebkowski gt 11., 1985) and in the mouse cells-lambda phage
integrated system (Glazer gt 11., 1986). It has been suggested that
the lesions that initiate the 6°C -+ A°T transition are dipyridine
photoproducts in which as A is inserted opposite a C by the polymerase

(Bredberg 1t 11., 1986; Hauser gt 11., 1986).

5. Stably Replicating Shuttle Vector Systems

Another appoach to solve the problem of the high spontaneous
mutation frequency of transfected shuttle vectors in mammalian cells
and the inefficient replication of some SV40 shuttle vectors is to use
stably replicating shuttle vector system. If most of the mutations
incurred by transiently replicating vectors are early events targeted
to intracellular damage during the transfection process, then stably
replicating vectors could offer a solution to the problem. The
initital establishment of stable vectors via transfection would be
expected to lead to some mutation of the incoming DNA. However, since

this mutation frequency is about 1% - 0.01%, it should be relatively

47

easy to clone cell lines which carry only wild-type vector DNA. If no
appreciable mutagenesis occurs during subsequent vector replication,
the mutation frequency of vector DNA purified from such cell lines
should remain low. In other words, stable vector system potentially
allow one to clone cells containing vector molecules which did not
suffer any damage to their DNA during the process of transfection.
This procedure is not possible in cells transfected with transiently
replicating vectors because vector replication prevents long-term
survival of these cells.

In contrast to transiently replicating vectors, stably replicating
vectors replicate in the nuclei of permissive cells in synchrony with
DNA systhesis period of the cell cycle of the transfected cells. They
generally attain a fairly stable copy number per cell, but the number
is very much lower than that achieved by transiently replicating
vectors. Immortalized cell lines transfected with stably replicating
vectors are generally able to carry the vectors as extrachomosomal DNA
indefinitely. Examples of stable replicating vectors are those derived
from the bovine papillomavirus (BPV) and the human herpesvirus Epstein-
Barr (EBV).

Unlike vectors based on EBV, vectors based on BPV show high
mutation frequency, even in clonal lines (Ashman and Davidson, 1985).
The mutations, mainly rearrangements, may due to the inherent
instability of BPV vectors (Dimaio e_t_ 11., 1982; Schenborn gt fl“
1985). Therefore, BPV-based vectors were found to be unsuitable for
mutagenesis studies.

Autonomously replicating vectors have been derived from EBV by

Sugden and colleagues (Yates gt 11., 1984; 1985; Sudgen gt 11., 1985).

48

These vectors contain a gig-acting element (1113, the origin of
replication) of EBV that permits replication and stable maintenance of
recombinant plasmids in the nuclei of permisive cells; the tr111-acting
gene coding for the EBV nuclear antigen 1 (EBNA-l) that is required for
activation of the origin; a gene coding for hygromycin resistance for
selsction in mammalian cells; and portions of plasmid p8R322 for
selection and replication of the shuttle vector in bacteria. Sudgen gt
11. (1985) showed that these vectors replicate stably at copy numbers
of 5-100 per cell in wide variety of human and other primate cell lines
as long as hygromycin in maintained. The high efficiency with which
EBV-derived vectors can be introduced, selected and maintained as
plasmids in a variety' of' mammalian cells indicated that they can
provide a powerful tool for mutagenesis studies in mammalian cells.

Drinkwater and Klinedinst (1986) have constructed a shuttle vector
based on the EBV-derived vector of Sugden g 11.(1985) for studying
mutagenesis in EBV-transformed human lymphoblastoid cell lines. Their
vector also contains the 1.1kb HSV t1 gene as the target for
mutagenesis. This vector originally did not contain the EBNA-l gene so
that it could replicate only in cells containing the EBNA-l antigen.
After introduction of this vector into an EBV-transformed human
lymphoblastoid cell line by electroporation, approximately 2% of the
transfected cells became hygromycin resistant. Plasmid DNA isolated
from cells which had been grown in the presence of hygromycin for ten
population doublings posttransfection contained mutations in the target
HSV t_k gene at a frequency of 6 x 10's. This frequency was 4-fold
higher than that observed fOr the vector which only replicated in E.

coli. However, it is 2-fold lower 'than the background frequency

49
obtained in the best transiently replicating system.

These plasmid-containing cells were treated with N-ethyl-N-
nitrosourea (ENU). The frequency of mutations in the HSV 1:1 gene
increased up to 15-fold with lmM ENU. Drinkwater and Klinedinst (1986)
showed that the induction of mutations by ENU in the plasmid-encoded
HSV t1 gene and in the cellular gene for HPRT follows a similar dose-
response. Characterization of HSV t1 by restiction mapping showed that
30% of the induced mutants from populations of cells contained
detectable deletions (>30bp). This result indicates that ENU is a
principal base substitution mutagen. The proportion of deletions was
large than would be expected. This may indicate that vector containing
deletion mutations via the transfection process still existed in the
transfected populations. The problem could be avoided by using cloned
plasmid-containing cell lines which have a very low background of
deletions as target for studies of induced mutagenesis.

Calos and coworkers (DuBridge gt 11., 1987) contructed other EBV-
derived vectors containing the 1112, the gene coding for EBNA-l, the
gene for hygromycin resistance, and sequences of p8R322 for selection
and replication in bacteria as does those in vectors of Sugden gt 11.
(1985). Their vectors also contain the 1111 gene as the target for
mutation, and the SV40 origin of replication. These vectors were
transfected into 293 cells and hygromycin resistant colonies were
selected. These EBV vectors carrying the bacterial 1111 gene were than
established in 293 cells and later returned to g. 1111 for rapid
detection and analysis of 1111 mutations. The background frequency of
1111 mutants derived from vectors that had stably maintained in the

populations of 293 cells was 2-5 x 10‘4. These values are comparable

50
to mutant frequency of 4 x 10‘4 obtained with a SV40 vector transiently
replicating in 293 cells.

In contrast, the majority (16/18) of clonal cell lines created by
establishment of the 1111-E8V vector show spontaneous mutant frequency
of less than 10'5 and therefore are suitable for studies of induced
mutation. Their results also indicate that the high mutation frequency
which is seen generally in transiently replicating plasmids, and in
pooled populations or sporadic isolated clones with stably replicating
vectors, is the result of transfection, not replication.

A cloned vector-containing cell line which carries approximately
ten copies of EBV-1111 vectors and has a spontaneous background of 6.4
x 10'6 was treated with lmM of N-methyl-N-nitrourea (MNU) (DuBridge gt
11., 1987). Vector DNA was harvested and assayed for 1111 mutations.
The observed induced mutation frequency of 2.1 x 10'3 is more than 300-
fold above the background frequency. The MNU-induced 1111 mutations
were subjected to genetic analysis. They found 34 independent nonsense
mutations among the 225 1111 mutations (15%).

0f the 34 nonsense mutations, 33 resulted from 6°C —9 A°T
transition, the other one being 6°C -» T°A transversion. These
investigators also sequenced 27 of the nonsuppressible mutations
induced by MNU. Of the 27 mutations, 26 represent 6°C —+ A°T
transition, the single exception being an A°T -+ 6°C transition. The
mutagenic lesion for 6°C —+ A°T transition is presumably addition of
the alkyl group to the 06 position of guanine, leading to mispairing
with thymine at the first round replication and fixtion of the full
transition at the next round (Drake and Baltz, 1976). The alkylations

of the 04 position of thymine could also potentially cause mispairing,

51
leading to A°T -+ 6°C transition (Singer gt 11., 1986).

These studies have also provided support for the mutational theory
of' cancer in connection with the experiments of Barbacid and his
colleagues (Sukumar g 11., 1983; Zarbl g g“ 1985). In those
studies, MNU was used to induce manlnary from most of the animals
contained activated Ha-111 genes. In 48 out of 48 cases, the altered
111 genes contained a 6°C -9 A°T transition in the second base of the

code for amino acid 12.

REFERENCES

Ashman, C.H. and Davidson, R.L. (1984), High sponstaneous mutation
frequency in shuttle vector sequences recovered from mammalian cellular
DNA, Mol. Cell. Biol., 1, 2266-2272.

Ashman, C.R. and Davidson, R.L. (1985), High spontaneous mutation
frequency of BPV shuttle vector, Somat. Cell Mol. 6enet., 2, 499-504.

Ashman, C.R. and Davidson, R.L. (1987), Sequence analysis of
spontaneous mutations in a shuttle vector gene integrated into
mammalian chromosomal DNA, Proc. Natl. Acad. Sci. USA, 81, 3354-3358.

Aust, A.E., Drinkwater, N.R., Debein, K., Maher, V.M. and McCormick,
J.J. (1984), Comparison of the frequency of diphtheria toxin and
thioguanine resistance induced by a series of carcinogens to analyze
their mutational specificities in diploid human fibroblasts, Mutat.
Res., 125, 95-104.

Beland, F.A., Ribovich, M., Howard, P.C., Heflich, R.H., Kurian, P.
and Milo, 6.E. (1986), Cytotoxicity, cellular transformation and DNA
adducts in normal human diploid fibroblasts exposed to
l-nitrosopyrene, a reduced derivative of the environmental
contaminant, 1-nitropyrene, Carcinogenesis, 1, 1279-1283.

Borgen, A., Darvey, H., Castagnoli, N., Crocker, T.T., Rasmussen, R.E.
and Hang, I.Y. (1973), Metabolic conversion of benzo[a]pyrene by
Syrian hamster liver microsomes and binding of metabolites to
deoxyribonucleic acid, J. Med. Chem., 11, 502.

Bourre, F. and Sarasin, A. (1983), Targeted mutagenesis of SV40 DNA
induced by UV light, Nature (London), 191, 68-70.

Bradley, R.E.C. and Letovanec, D. (1982), High frequency' nonrandom
mutation event at the adenine phosphoribosyltransferase (1p_rt) locus
of sib-selected CHO variants heterozygous for 111t, Somatic Cell
6enet., 1, 51-66.

Brash, D.E., Seetharam, S., Kraemer, K.H., Seidman, M.M. and Berdberg,
A. (1987), Photoproduct frequency is not the major determinant of UV
base substitution hot spots or cold spots in human cells, Proc. Natl.
Acad. Sci. U.5.A., 11, 3782-3786.

Bredberg, A., Kraemer, K.H. and Seidman, M.M. (1986), Restricted
ultraviolet mutational spectrum in a shuttle vector propagated in
xeroderma pigmentosum cells, Proc. Natl. Acad. Sci. USA, 11, 8273-8277.

Brennand, J., Chinault, A.C., Konecki, D.S., Melton, D.N. and Caskey,
C.T. (1982), Cloned cDNA sequence of the hypoxanthine/guanine
phosphoribosyltransferase gene from a mouse neuroblastoma cell line
found to have amplified genomic sequences, Proc. Natl. Acad. Sci. USA,
12, 1950-1954.

52

53

Bresnick, E., McDonald, F., Yagi, H., Jerina, D.M., Lewin, H., Hood,
A.H., and Conney; A.H. (1977), Epidermal hyperplasia. after ‘topical
application of benzo[a]pyrene, benzo[a]pyrene diol epoxides, and other
metabolites, Cancer Res., 11, 984.

Buening, M.K., Wislocki, P.6., Levin, H., Yagi, H., 'Thakker3 D.R.,
Akagi, H., Koreeda, M., Jerina, D.M. and Conney, A.H. (1978),
Tumorigenicity of the optical enantiomers of the diastereomeric
benzo[a]pyrene 7,8-diol-9,10-epoxides in newborn mice: Exceptional
activity of (1)43 ,8a -dihydroxy-9a ,loa-epoxy-7,8,9,10-tetrahydro
benzo[a]pyrene, Proc. Natl. Acad. Scix U.5.A., 11, 5358-5361.

Calos, M.P., Lebkowski, J.S. and Botchan, M.R. (1983), High mutation
frequency in DNA transfected into mammalian cells, Proc. Natl. Acad.
Sci. USA, 11, 3015-3019.

Carothers, A.M., Urlaub, 6., Ellis, N. and Chasin, L.A. (1983),
Structure of the dihydrofolate reductase gene in Chinese hamster ovary
cells, Nucleic Acids Res., 11, 1997-2012.

Celis, J.E. and Piper, P.H. (1982), Compilation of mutant suppressor
tRNA sequences, Nucleic Acids Res., 11, R83-R91.

Chakrabarti, S., Mizusawa, H. and Seidman, M. (1984), Survival and
mutagenesis of bacterial plasmids with localized carcinogen adducts,
Mutat. Res., 111, 127-137.

Chen, M.-J., Schimada, T., Moulton-Davis, A., Cline, A., Humphries,
R.K., Maizel, J. and Nienhuis, A.H. (1984), The functional human
dihydrofolate reductase gene, J. Biol. Chem., 111, 3933-3943.

Coulondre, C. and Miller, J.H. (1977), Genetic studies of the 111
repressor IV. Mutagenic specificity in the 1111 gene of E§1hgrichi1
ggli, J. Mol. Biol., 111, 577-606.

Crouse, 6.F., Simonsen, C.C., McEwan, R.N. and Schimke, R.T. (1982),
Structure of amplified normal and variant dihydrofolate reductase genes
in mouse sarcoma 5180 cells, J. Biol. Chem., 211, 7887-7897.

Dasgupta, U.B. and Summers, N.C. (1978), Ultraviolet reactivation of
herpes simplex virus is mutageneic and inducible in ammmalian cells,
Proc. Natl. Acad. Sci. USA, 11, 2378-2381.

Dimaio, D., Treisman, R. and Maniatis, T. (1982), Bovine papillomavirus
vector that propagates as a plasmid in both mouse and bacterial cells,
Proc. Natl. Acad. Sci. USA, 11, 4030-4034.

Drake, J.H. and Baltz, R.H. (1976), The biochemistry of mutagenesis,
Ann. Rev. Biochem., 11, 11-37.

Drinkwater, N. and Klinedinst, D.K. (1986), Chemically induced
mutagenesis in a shuttle vector with a low-background mutant frequency,
Proc. Natl. Acad. Sci. USA, 11, 3402-3406.

54

DuBridge, R.B., Tang, P., Hsia, H.C., Leong, P.-M., Miller, J.H. and
Calos, M.P. (1987), Analysis of mutation in human cells by using an
Epstein-Barr virus shuttle vector, Mol. Cell. Biol., 1, 379-387.

Eddy, E.P., McCoy, E.C., Rosenkranz, H.S. and Mermelstein, R. (1986),
Dichotomy in the mutagenicity and genotoxicity of nitropyrenes:
Apparent effect of the number of electrons involved in it
nitroreduction, Mutat. Res., 111, 109-111.

Eisenstadt, E., Harren, A.J., Porter, J., Atkins, D. And Miller, J.H.
(1982), Carcinogenic epoxides of benzo[a]pyrene and
cyclopenta[cd]pyrene induce base substitutions via specifh:
transversions, Proc. Natl. Acad. Sci. USA, 12, 1945-1949.

El-Bayoumy, K., Hecht, S.S., Sacki, T. and Stoner, 6.0. (1984),
Tumori genicity and metabolism of l-nitropyrene in A/J mice,
Carcinogenesis, 1, 1449-1452.

Fahl, R.E., Scarpeili, 0.6 and Gill, K. (1981), Relationship between
benzo[a]pyrene-induced DNA base modification and frequency of reverse
mutations in mutant strains of Salmonglla typimgrjum, Cancer Res., 11,
3400-3406.

Fenwick, R.6., Fuscoe, J.C. and Caskey, C.T. (1984), Amplification
versus mutation as a mechanism for reversion of an fiGPB! mutation,
Somat. Cell Mol. 6enet., 11, 71-84.

Fuscoe, J.C., Fenwick, R.6., Ledbetter, D.H. and Caskey, C.T. (1983),
Deletion and amplification of the H6PRT locus in Chinese hamster cells,
Mol. Cell Biol., 1, 1086-1096.

6amper, H.B., Tung, A. S.-C., Straub, K., Bartholomew, J.C. and
Calvin, M. (1977), DNA strand scission by benzo[a]pyrene diol
epoxides, Science, 111, 671-674.

6entil, A., Margot, A. and Sarasin, A. (1986), 2-(fl-acetoxy-fl-
acetylamino)fluorene mutagenesis in mammalian cells: Sequence-specific
hot spot, Proc. Natl. Acad. Sci. USA, 11, 9556-9560.

Glazer, P.M., Sarkar, S.N. and Summers, N.C. (1986), Detection and
analysis of UV-induced mutations in mammalian cell DNA using a lambda
phage shuttle vector, Proc. Natl. Acad. Sci. U.S.A., 11, 1041-1044.

Goncalves, 0., Drobetsky, E. and Meuth, M. (1984), Structural
alterations of the 1p_rt locus induced by deoxyribonucleoside

triphosphate pool imbalances in Chinese hamster ovary cells, Mol. Cell
Biol., 1, 1792-1799.

Graf, L.H. and Chasin, L.A. (1982), Direct demonstration of genetic
alterations at the dihydrofolate reductase locus after gama
irradiation, Mol. Cell. Biol., 1, 93-96.

55

Hauser, J., Seidman, M.M., Sidur, K. and Dixon, K. (1986), Sequence
specificity of point mutations induced during passasge of a
UV-irrated shuttle vector plasmid in monkey cells, Mol. Cell. Biol.,
1, 277-285.

Heflich, R.H., Howard, P.C. and Beland, F.A. (1985), I-Nitrosopyrene:
An intermiate in the metabolic activation of l-nitropyrene to a

mutagen in 11lmgngll1 t1111111111 TA1538, Mutat. Res., 111, 25-32.

Heflich, R.H., Fullerton N.F. and Beland, F.A. (1986), An examination
of the weak mutagenic response of l-nitropyrene in Chinese hamster
ovary cells, Mutat. Res., 111, 99-108.

Hirose, M., Lee, M.-S., Hang, C.Y. and King, C.H. (1984), Induction of
rats mammary gland tumors by l-nitropyrene, a recently recognized
environmental mutagen, Cancer Res., 11, 1158-1162.

Howard, P.C., Heflich, R.H., Evans, F.E. and Beland, F.A. (1983),
Formation of DNA adducts 11 vitro and in 11M typhimgrjum upon
metabolic reduction of the environmental mutagen l-nitropyrene, Cancer
Res., 11, 2052-2058.

Huberman, E., Sachs, L., Yang, S.K. and 6elboin, H.V. (1976),
Identification of mutagenic metabolites of benzo[a]pyrene in mammalian
cells, Proc. Natl. Acad. Sci. USA, 11, 607-612.

Ivanovic, V., 6eacintov, N.E. and Neinstein, 1.8. (1976), Cellular
binding of benzo[a]pyrene to DNA\ characterized by low temperature
fluorescence, Biochem. Biophys. Res. Commun., 11, 1172-1179.

Jeffrey, A.M., Weinstein, 1.8., Jennette, K.H., 6rzeskowiak, K.,
Nakanishi, K., Harvey, R.6., Autrup, H. and Harris, C. (1977),
Structures of benzo[a]pyrene-nucleic acid adducts formed in human and
bovine bronchial explants, Nature (London), 211, 348-350.

Jolly, D.H., Okayama, H., Berg, P., Esty, A.C., Filpula, D., Bohlen,
P., Johnson, 6.6., Shively, J.E., Hunkapillar, T. and Friedman, T.
(1983), Isolation and characterization of a full-length expressible
cDNA for human hypoxanthine phosphoribosyltransferase, Proc. Natl.
Acad. Sci. USA, 11, 477-481.

Kapitulnik, J., Levin, H., Conney, A.H., Yagi, H. and Jerina, D.M.
(1977), Benzo[a]pyrene 7,8-dihydrodiol is more carcinogenic than
benzo[a]pyrene in newborn mice, Nature (London), 111, 378-380.

Kapitulnik, J., Hislocki, P.6., Levin, H., Yagi, H., Jerina, D.M. and
Conney, A.H. (1978), Tumorigenicity studies with diol-epoxides of
benzo[a]pyrene which indicate that (1)-trans-7 B ,8 a -dihydroxy-9 a ,
10C!-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene is an ultimate carcinogen
in newborn mice. Cancer Res., 11, 354-358.

King, H.H.S. and Brookes, P. (1984), On the nature of the mutations
induced by the diolepoxide of benzo[a]pyrene in mammalian cells,
Carcinogenesis, 1, 965-970.

56

Klopman, 6., Tonucci, D.A., Holloway, M. and Rosenkranz, H.S. (1984),
Relationship between polarographic reduction potential and
mutagenicity of nitroarenes, Mutat. Res., 111, 139-144.

Koffel-Schwartz, N., Verdier, J.H., Bichara, M., Freund, A.M., Daune,
M.P. and Fuchs, R.P.P. (1984), Carcinogen-induced mutation spectrum in
wild-type, nxrA, and umuC strain of £§chgr11hj1 1111, J. Mol. Biol.,
111, 33-51.

Konecki, D.S., Brennand, J., Fuscoe, J.C., Caskey, C.T. and Chinault,
A.C. (1982), Hypoxanthine-guanine phosphoribosyltransferase genes of
mouse and Chinese hamster: construction and sequence analysis of cDNA
recombinants, Nucleic Acids Res., 11, 6763-6775.

Lebkowski, J.S., DuBridge, R.B., Antell, E.A., Greisen, K.S. and
Calos, M.P. (1984), Transfected DNA is mutated in monkey, mouse, and
human cells, Mol. Cell. Biol., 1, 1951-1960.

Lebkowski, J.S., Clancy, S., Miller, J.H. and Calos, M.P. (1985), The
1111 shuttle: Rapid analysis of the mutagenic specificity of
ultraviolet light in human cells, Proc. Natl. Acad. Sci. USA, 11, 8606-
8610.

Lebkowski, J.S., Miller, J.H. and Calos, M.P. (1986), Determination of
DNA sequence changes induced by ethyl methanesulfonate in human cells:
Using a shuttle vector system, Mol. Cell. Biol., 1, 1838-1842.

Lehmann, A.R. (1985), Use of recombinant DNA techniques in cloning DNA
repair genes and in the study of mutagenesis in mammalian cells, Mutat.
Res., 111, 61-67.

Levin, H., Hood, A.H., Hislocki, P.6., Chang, R.L., Kapitulnik, J.,
Mah, H.D., Yagi, H., Jerina, D.M. and Conney, A.H. (1978), Mutagenicity
and carcinogenicity of benzo[a]pyrene and benzo[a]pyrene derivatives,

in: Polygyclig Hydrgcarbons 1nd C1ncer, Vol I. 6elboin, H.V. and Ts’0,
P.0. (eds.) Academic Press, New York, NY, pp. 189-202.

Lewin, B. (1980), in: Gene Expresszn; Egkaryotic Chromosomes, Vol. 2,
2nd Ed., Hiley, New York.

Li, A.P. and Dutcher, J.S. (1983), Mutagenicity of mono-, di-, and
tri-nitropyrenes in Chinese hamster ovary cells, Mutat. Res., 112,
387-392.

Lowy, 1., Pellicer,A., Jackson, J.E., Sim, 6.-K., Silverstein, S. and
Axel, R. (1980), Isolation of transforming DNA: cloning the hamster
111t gene, Cell 12, 817-823.

Malaveille, C., Kuroki, T., Sims, P., Grover, P.L. and Bartsch, H.
(1977), Mutagenicity of isomeric dial-epoxides of benzo[a]pyrene and
benzo[a]anthracene in 11 tyghimgrigm TA98 and TA100 and in V79 Chinese
hamster cells, Mutat. Res., 4, 313-326.

57

McCoy, E.C., Rosenkranz, H.S. and Mermelstein, R. (1981), Evidence for
the existence of a family of bacterial nitroreductases capable of

activating nitrated polycyclics to mutagens, Environ. Mutagenesis, 1,
421-427.

McCoy, E.C. and Rosenkranz, H.S. (1982), Cigarette smoking may yield
nitroarenes, Cancer Lett., 11, 9-13.

Meehan, T., Straub, K. and Calvin, M. (1976), Elucidation of
hydrocarbon structure in an enzyme-catalyzed benzo[a]pyrene-poly(6)
covalent complex, Proc. Natl. Acad. Sci. USA, 11, 1437-1441.

Meehan, T., Straub, K. and Calvin, M. (1977), Benzo[a]pyrene diol
epoxide covalentiy binds to deoxyguanosine and deoxyadenosine in DNA,
Nature (London), 212, 725-727.

Melton, D.H., Konecki, D.S., Brennand, J. and Caskey, C.T. (1984),
Structure, expression, and mutation of the hypoxanthine
phosphoribosyltransferase gene, Proc. Natl. Acad. Sci. USA, 11, 2147-
2151.

Mermelstein, R., Kiriazides, D.K., Butler, M., McCoy, E.C. and
Rosenkranz , H . S . (1981) , The extraordinary mutagenicity of
nitropyrenes in bacteria, Mutat. Res., 11, 187-196.

Meuth, M. and Arrand, J.E. (1982), Alterations of gene structure in
ethyl methanesulfonate-induced mutants of mammalian cells, Mol. Cell.
Biol., 2, 1459-1462.

Miller, E.C. and Miler, J.A. (1974), Biochemical mechanisms of

chemical carcinogenesis, in: lhg Molegular Biglogy of C1ncgr, Busch, H.
(ed.), Academic Press, New York, NY, pp. 377-402.

Miller, J.H., 6anem, D., Lu, P. and Schmitz, A. (1977), Genetic studies
of the 111 repressor; 1. Correlation of mutational sites with specific

amino acid residues: Construction of a colinear gene-protein map, J.
Mol. Biol., 111, 275-302.

Miller, J.H., Lebkowski, J.S., Greisen, K.S. and Calos, M.P. (1984),
Specificity of mutations induced in transfected DNA by mammalian
cells, EMBO J., 1, 3117-3121.

Miller, J.H. (1985), Mutagenic specificity of ultraviolet light, J.
Mol. Biol., 111, 45-68.

Mizusawa, H., Lee, C.-H. and Kakefuda, T. (1981a), Alteration of
plasmid DNA-mediated transformation and mutation induced by covalent
binding of benzo[a]pyrene-7,8-dihydrodioT-9,10-oxide in Eggherighia
11li, Mutat. Res., 11, 47-57.

Mizusawa, H., Lee, C.-H., Kakefuda, T., McKenney, K., Shimatake, H.
and Rosenberg, M. (1981b), Base insertion and deletion mutations
induced 'hi an plasmid by benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide,
Proc. Natl. Acad. Sci. USA, 11, 6817-6820.

58

Morandi, C., Masters, J.N., Mottes, M. and Attardi, 6. (1982), Multiple
forms of human dihydrofolate reductase message RNA, J. Mol. Biol., 111,
583-607.

Nakayatsu, M., Sakamoto, H., Hakabayashi, K., Terada, M., Sugimura, T.
and Rosenkranz, H.S. (1982), Potent mutagenic activity of nitropyrenes
on Chinese hamster lung cells with diphtheria toxin resistance as a
selective marker, Carcinogenesis, 1, 917-922.

Nalbantoglu, J., Goncalves, O. and Meuth, M. (1983), Structure of
mutant allels at the 1111 locus of Chinese hamster ovary cells, J. Mol.
Biol., 111, 575-594.

Nalbantoglu, J., Phear, 6. and Meuth, M. (1987), DNA sequence analysis
of spontaneous mutations at the aprt locus of hamster cells, Mol. Cell.
Biol., 1, 1445-1449.

National Academy of Science Reports, U.S.A. (1972), P1rt11ulate
P1|y1y1111 Org1ni1 flattgr, Commettee on biologic effects of atmospheric
pollutants, Div. Med. Sci., Natl. Res. Counc., Washington, D.C., pp.
30.

Newbold, R.F. and Brookes, P. (1976), Exceptional mutagenicity of
benzo[a]pyrene diol epoxides in cultured mammalian cells, Nature
(London), 111, 52-54.

Ohgaki, H., Matsukura, N., Morino, K., Kawachi, T., Sugimura, ‘T.,
Morita, K., Tokiwa, H. and Hirota, T. (1982), Carcinogenicity in rats
of the mutagenic compounds l-nitropyrene and 3-nitrofluoranthene,
Cancer Lett., 11, 1-7.

Osborne, M.R., Harvey, R.6. and Brookes, P. (1978), The reaction of

trans-7 , 8-di hydroxy- anti -9,10-epoxy- ,8,9,10-tetrahydrobenzo[a]pyrene
with DNA involves attack at the N -position of' guanine moieties,
Chem.-Biol. Interactions, 11, 123-130.

Osborne, M.R., Jacobs, S., Harvey, R.6. and Brookes, P. (1981), Minor
products from the reaction of (+) and (-)benzo[a]pyrene-anti-
diolepoxide with DNA, Carcinogenesis, 1, 553-558.

Osborne, M. and Merrifield, K (1985), Depurination of benzo[a]pyrene
diolepoxide treated DNA, Chem.-Biol. Interactions, 11, 183-195.

Patton, J.D., Maher, V.M. and McCormick, J.J. (1987) Cytotoxicity,
mutagenicity, and transformation of diploid human fibroblasts by 1-

nitropyrene and l-nitrosopyrene, in: Polynucle1r Ar1m1tic Hydzocarbons:

A 1gc111 of 1rogrg11, Cooke, M. and Dennis, A.J. (eds.), Battelle
Press, Columbus, pp. 678-698.

Perucho, M., Hanahan, D. and Higler, M. (1980), Genetic and physical
linkage of exogeneous sequences in transformed cells, Cell, 11, 309-
317.

59

Protit-Sabljit, M., Tuteja, N., Munson, P.J., Hauser, J., Kraemer, K.H.
and Dixon, K. (1986), UV light-induced cyclobutane pyrimidine dimers
are mutagenic in mammalian cells, Mol. Cell. Biol., 1, 3349-3356.

Rajalakshmi, S., Rao, P.M. and Sarma, D.S.R. (1982), Chemical
carcinogenesis: Interactions of carcinogens. with nucleic acid, in:
11111:, Becker, F.F. (ed.), Vol. 1, 2nd ed., pp. 241-333.

Rappaport, S.N., Hang, Y.Y., Nei, E.T., Sawyer, R., Nalkins, B.E. and
Rappaport, H. (1980) Isolation and identification of a direct-acting
mutagen in diesel-exhaust particulates, Environ. Sci. Tech., 11, 1505-
1509.

Razzaque, A., Mizusawa, H. and Seidman, M.M. (1983), Rearrangement
and mutagenesis of a shuttle vector plasmid after passage in mammalian
cells, Proc. Natl. Acad. Sci. USA, 11, 3010-3014.

Razzaque, A., Chakrabarti, S., Joffe, S. and Seidman, M. (1984),
Mutagenesis of a shuttle vector plasmid in mammalian cells, Mol. Cell.
Biol., 1, 435-441.

Robins, D. M., Riply, S., Henderson, A. S. and Axel, R. (1981),
Transforming DNA integrates into the host chromosome, Cell, 11, 29-39.

Robbins, J.H., Kraemer, K.H., Lutzner, M.A., Festoff, B.H. and Coon,
H.6. (1974), Xeroderma pigmentosum -— an inherited deseade with sun

sensitivity, multiple cutaneous neoplasms, and abnormal DNA repair,
Ann. Intern. Med., 11, 221-248.

Rosenkranz, H.S., McCoy, E.C., Sander, D.R., Butler, M., Kiriazides,
P.K. and Mermelstein, R. (1980), Nitropyrenes: Isolation,
identification, and reduction of mutagenic impurities in carbon black
and toners, Science, 111, 1039-1043.

Rosenkranz, H.S. (1982), Direct-acting mutagens in diesel exhauusts:
Magnitude of problem, Mutat. Res., 111, 1-10.

Rosenkranz, H.S. and Mermelstein, R. (1983), Mutageneicity and
genotoxicity of nitroarenes. All nitro-containing chemicals were not
created equal, Mutat. Res., 111, 217-267.

Sarker, S., Dasgupta, U.B. and Summers, H.C. (1984), Error-prone
mutagenesis detected in mammalian cells by a shuttle vector containing
the 1111 gene of E§chgrichi1 111i, Moi. Cell. Biol., 1, 2227-2230.

Scangos, 6.A., Huttner, K.M., Juricek, D.K. and Ruddle, F.H. (1981),
Deoxyribonucleic acid-mediated gene transfer in mammalian cells:

Molecular analysis of unstable transformants and their progresion to
stability, Mol. Cell. Biol., 1, Ill-120.

Schenborn, E.T., Lund, E., Mitchen, J.L. and Dahlberg, J.E. (1985),
Expression of a human Ul RNA gene introduced into mouse cells via
Bovine Papillomavirus DNA vectors, Mol. Cell. Biol., 1, 1318-1326.

60

Seidman, M.M., Dixon, K., Razzaque, A., Zagursky, R.J. and Berman,
M.L. (1985), A shuttle vector plasmid for studying carcinogen-induced
point mutations in mammalian cells, Gene, 11, 233-237.

Serres, F.J. and Matsushima, T. (1986), Meeting Report: Mutagenesis
and carcinogenesis by nitropyrenes and cancer chemotherapeutics,
Mutat. Res., 111, 3-8.

Sims, P. and Grover, P.L. (1974), Epoxides in polycyclic aromatic
hydrocarbon metabolism and carcinogenesis, Advan. Cancer Res., 21, 165-
274.

Sims, P., Grover, P.L., Swaisland, A., Pal, K. and Hewer, A. (1974),
Metabolic activation of benzo[a]pyrene proceeds by a diol epoxide,
Nature (London), 111, 326-328.

Singer, 8 , Spangler, S.J., Fraenkel-Conrat, H. and Kusmierek, J.T.
(1986), 04-methyl, -ethyl, or -isopropyl substituents on thymidine in
poly (dA-dT) all lead to transition upon replication, Proc. Natl. Acad.
Sci. USA, 11, 28-32.

Slaga, T.J., Viaje, A., Berry, D.L. and Bracken, H. (1976), Skin tumor
initiating ability of benzo[a]pyrene4,5-,7,8-, and 7,8-diol-9,10-
epoxides and 7,8-diol, Cancer Lett., 1, 115-122.

Slaga, T.J., Bracken, R.N., Viaje, A., levin, H., Yagi, H., Jerina,
D.M. and Conney, A.H. (1977), Comparison of the tumor-initiating
activities of benzo[a]pyrene arene oxides and diol epoxides. Cancer
Res., 11, 4130-4133.

Slaga, T.J., Bracken, R.J., 6eason, 6., Levin, H., Yagi, H., Jerina,
D.M. and Conney, A.H. (1979), Marked differences in the skin
tumor-initiating activities of the optical enantiomers of the
diastereomeric benzo[a]pyrene 7,8-diol-9,10-epoxides, Cancer Res., 11,
67-71.

Stambrook, P.J., Dush, M.K., Trill, J.J. and Tischfield, J.A. (1984),
Cloning of a functional human adenine phosphoribosyltransferase (APRT)
gene: Identification of a restriction fragment length polymorphism and
preliminary analysis of DNAs from APRT-deficient families and cell
mutants, Somat. Cell Mol. 6enet., 11, 359-367.

Straub, K.M., Meehan, T., Burlingame, A.L. and Calvin, M. (1977),
Identification of the major adducts formed by reaction of
benzo[a]pyrene diol epoxide with DNA 11 x1t11, Proc. Natl. Acad. Sci.
USA, 11, 5285-5289.

Sugden, B., Marsh, K. and Yates, J. (1985), A vector that replicates as
a plasmid and can be efficiently selected in B-lymphoblasts transformed
by Epstein-Barr virus, Mol. Cell. Biol., 1, 410-413.

61

Sukumar, S., Notario, V., Hartin-Zanca, D. and Barbacid, M. (1983),
Induction of mammary carcinomas in rats by nitrosomethylurea involves
activation of H-r11-1 locus by single point mutations, Nature (London),
195. 658-661.

Takayama, S., Tanaka, M., Katoh, Y., Terada, M. and Sugimura, T.
(1983), Mutagenicity of nitropyrenes in chinese hamster V79 cells,
Gann., 11, 338-341.

Thacker, J., Debenham, P.6., Stretch, A. and Hebb, M.B.T. (1983), The
use of a cloned bacterial gene to study mutation in mammalian cells,
Mutat. Res., 111, 9-23.

Thacker, J. (1985), The molecular nature of mutations in cultured
mammalian cells: A review, Mutat. Res., 111, 431-442.

Tindall, K.R., Stankowski, L.F., Machanoff, R. and Hsie, A.H. (1984),
Detection of deletion mutations in pSVngt-transformed cells, Mol. Cell
Biol., 1, 1411-1415.

Tokiwa, H., Nakagawa, R. and Horikawa, K. (1985), Mutagenic/
carcinogenic agents in indoor pollutants; the dinitropyrenes generated
by kerosine heaters and fuel gas and liquified petroleum gas burners,
Mutat. Res., 7, 39-47.

Hang, C.Y., Lee, H.S., King, C.H. and Harner, P.0. (1980), Evidence
for nitroaromatics as direct-acting mutagens of airborne particulates,
Chemosphere, 1, 83-87.

Heinstein, I.B., Jeffrey, A., Jennett, K., Blobstein, H., Harvey,
R.6., Harris, C., Autrup, H., Kasai, H. and Nakanishi, K. (1976),
Benzo[a]pyrene diol epoxides as intermediates in nucleic acid binding
and 11 vjv1, Science, 111, 592-595.

Heisburger, J.H. and Hilliams, 6.M. (1982), Metabolism of chemical
carcinogenesis, in: 1111gr, Becker, F.F. (ed.) Vol. 1, 2nd Ed., pp241-
333.

Hislocki, P.6., Hood, A.H., Chang, R.L., Levin, H., Yagi, H.,
Hernandez, 0., Jerina, D.M. and Conney, A.H. (1976), High mutagenicity
and toxicity of diol epoxide derived from benzo[a]pyrene, Biochem.
Biophys. Res. Commun., 11, 1006-1012.

Hislocki, P.6., Bagan, E.S., Lu, A.Y.H., Dooley, K.L., Fu, P.P. Han-
Hsu, H., Beland, F.A. and Kadlubar, F.F. (1986), Tumorigenicity of
nitrated derivatives of pyrene, benzo[a]anthracene, chrysene and
benzo[a]pyrene in newborn mouse assay, Carcinogenesis, 1, 1317-1322.

Hood, A.H., Hislocki, P.6., Chang, R.L., Levin, H., Lu, A.Y.H., Yagi,
H., Hernandez, 0., Terina, D.M. and Conney, A.H. (1976), Mutagenicity
and cytotoxicity of benzo[a]pyrene benzo-ring epoxides, Cancer Res.,
11, 3358-3366.

62

Hood, A.H., Chang, R.L., Levin, H., Yagi, H., Thakker, D.R.,
Jerina, D.M. and Conney, A.H. (1977), Differenced in mutagenicity of
the optical enantiomers of the diasteriomeric benzo[a]pyrene
7,8-diol-9,lO-epoxides, Biochem. Biophys. Res. Commun., 11, 1389-1396.

Yagi, H., Thakker, D.R., Hernandez, 0., Koreeda, M. and Jerina, D.M.
(1977), Synthesis and reactions of the highly mutagenic
7,8-diol-9,10-epoxides of the carcinogen benzo[a]pyrene, J. Am. Chem.
Soc., 11, 1604.

Yang, J.-L., Maher, V.M. and McCormick, J.J. (1987a), Kinds of
mutations formed when a shuttle vector containing adducts of
benzo[a]pyrene-7,8-diol-9,10-epoxide replicates in COS7 cells, Mol.
Cell. Biol., 1, 1267-1270.

Yang, J.-L., Maher, V.M. and McCormick, J.J. (1987b), Kinds of

mutations formed when a shuttle vector containing adducts of (1)-7B ,8a
-dihydroxy-9a ,100 -epoxy-7,8,9,lO-tetrahydrobenzo[a]pyrene replicates

in human cells, Proc. Natl. acad. Sci. U.S.A., 11, 3787-3791.

Yang, S.K., Deutsch, J. and 6elboin, H.V. (1978), Benzo[a]pyrene
metabolism: activation and detoxification, in: P1|xcxc|j1 Hydrggar1ons

111111111, Vol I, 6elboin, H.V. and Ts’O, P.O.P. (eds.), Academic
Press, New York, NY, pp. 205-231.

Yang, S.K., McCourt, D.H., Leutz, J.C. and 6elboin, H.V. (1977),
Benzo[a]pyrene diol epoxides: Mechanism :of enzymatic formation and
optically active intermediates, Science, 111, 1199-1201.

Yang, T.P., Patel, P.I., Cinault, A.C., Stout, J.T., Jackson, L.6.,
Hilderbrand, B.M. and Caskey, C.T. (1984), Molecular evidence for new
mutation at the h1rt locus in Lesch-Nyhan patients, Nature (London),
111, 412-414.

Yates, J., Harren, N., Reisman, D. and Sugden, B. (1984), A 111 acting
element from the Epstein-Barr viral genome that permits stable
replication of recombinant plasmids in latently infected cells, Proc.
Natl. Acad. Sci. USA, 11, 3806-3810.

Yates, J., Harren, N. and Sugden, B. (1985), Stable replication of
plasmids derived from Epstein-Barr virus in various mammalian cells.
Nature (London), 111, 812-815.

Zarbl, H., Sukumar, S., Arthur, A.V., Martin-Zanca, D. and Barbacid, M.
(1985), Direct mutagenesis of Ha-111-1 oncogenes by u-nitroso-u-
methylurea during initiation of mammary carcinogenesis in rats, Nature
(London), 111, 382-385.

CHAPTER II

Establishment of a Suitable Shuttle Vector System
to Study Molecular Mutagenesis

A. The Shuttle Vector Mutagenesis System

Two shuttle vectors, p3AC and p2189, were used in my studies for
the thesis. Vector p3AC was constructed and supplied by Summers and
his colleagues (Sarkar gt _1., 1984); p2189 was constructed by Seidman
and his colleagues (Seidman g l., 1985) and supplied to me by Dr.
Kenneth Kraemer of the National Cancer Institution. The structure of
the vectors is shown in Figure 1. Both vectors contain the origin of
replication and large T-antigen gene from SV40 virus which allows them
to replicate in mammalian cells, the gene coding for ampicillin
resistance, the bacterial origin of replication from p8R322 or pBR327
plasmid, and the same target gene, 1111, coding for a tyrosine
supressor tRNA. The major difference between these two vectors is that
the 1111 gene in p2189 is strategically located between two genes
essential for recovery of the plasmid in bacteria (the gene for
ampicillin resistance and the bacterial origin of replication).
Therefore, plasmids which contain large deletions in the 11% gene
which extend into either of these two essential genes are not
recovered for mutant analysis, which lowers the apparent frequency of
background mutants.

The mammalian cell line used in my earliest studies was COS7,
monkey kidney cells transformed with SV40 virus containing a defective

origin. These cells produce high levels of T-antigen, which

63

64

Figure 1. Structure of p3AC and p2189. p2189 adapted from Hauser gt
11. (1986), Mol. Cell. Biol. 1: 277-285. Abbreviations: Amp,
ampillin resistance gene ( B-lactamase); Ori, origin of

replication; RI, 1511“ site; t and T, small and large T antigen.

65

   

p8R322 Ori

AHoe II B V Born HI

 

66

facilitates replication of the SV40-based shuttle vector (6luzman,
1981). In my later experiments the host cells used were a human cell
line, designated 293, which was developed by transforming human
embryonic kidney cells with adenovirus 5 DNA fragments (Graham gt 11.,
1977). Substitution of the 293 cell line for C057 cells as the host
cells for my studies was suggested to me by Dr. Calos of Stanford
University who observed that shuttle vectors exhibited a lower
background in 293 cells than in other host cell lines (Lebkowski gt
11., 1984). Dr. Calos supplied me with the 293 cells.

Figure 2 diagrams the shuttle vector mutagenesis system I
employed. Shuttle vector is exposed to reactive forms of carcinogens

1_ vitro. The carcinogen-treated or untreated plasmid is transfected

 

into mammalian cells using the DNA-calcium-phosphate coprecipitation
technique (Graham and Van der Eb, 1973; Chu and Sharp, 1981). The
plasmid enter the nucleus of the cells where DNA replication and
mutagenesis taken place. 48 hr after DNA transfection, progeny plasmid
is extracted from the mammalian cells and separated from large
molecular weight cellular DNA by the procedure of Hirt (1967). In
order to distinguish independent mutants with identical mutations from
putative siblings derived from a single event, progeny plasmid obtained
from the cells in each dish is kept separate from the rest and assayed
separately. Progeny plasmid is purified by phenol extraction, digested
with RNase A, followed by proteinase K, purified again by phenol
extraction, precipitated with ethanol, resuspended in buffer, and
treated with a restriction enzyme 1111 to destroy any DNA that did not
replicate inside the mammalian cells. As shown in Figure 3, if any

input plasmid remains unreplicated, it will be destroyed by 11111

67

Figure 2. The shuttle vector mutagenesis system. The
triangles on the shuttle vector represent

carcinogen residues.

68

A
I... '

 
 
 
   

O
MUTAGEN V

MAMMALIAN

‘ ’1...

TRANSFECTION

   

V 4

SHUTTLE VECTOR

 

 

RESCUE

fl
1

PROCENY PLASMIDS

TR A NSI'OR NATION
I 1:. sang

TRA NSFORMATION

ISOLATE MUTANT PLASMIDS FROM
WHITE TRANSFORMANTS

 

 

 

 

 

 

CRARACTERIZE MUTANT PLASMIDS
BY GEL ELECTROPHORESIS

J

SEQUENCE PUTA’I‘IVE POINT MUTATIONS

l

ANALYZE THE SPECIFIC LOCATION 01’
DNA CHANGES

69

Figure 3. Digestion of QQQI on plasmid with the bacterial methylation
pattern. Qpfll digests DNA with methylated adenine on the 5’
GATC 3’ sequences. The thicker circles represent plasmids
that have replicated in mammalian cells. The thiner circles

are the input plasmids that was first prepared from bacteria.

71
because it has a methylation pattern which differs from that of
plasmid which has replicated in mammalian cells. Progeny plasmid is
then assayed to see which contains mutations in the sgpfi gene.

To detect M mutants, the Dpnl resistant progeny plasmid is
introduced into indicator bacteria by the procedure of Hanahan (1983).
The indicator bacteria g. £911 SY204 (Sarkar gt _1., 1984) carries an
amber mutation in the B-galactasidase gene so that bacterial
transformants containing a plasmid with functioning sgLF gene which
suppresses the amber mutation can form blue clonies on 5-bromo-4-
chloro-3-indolyl B-D-galactoside (X-Gal) plates containing ampicillin.
Bacterial transformants containing plasmid with a mutated, inactive
sggfi gene form white colonies. Only those bacteria that have taken up
a plasmid can form colonies on these agar plates because the medium
contains ampicillin. The bacteria used for transformation with
plasmid are ampicillin sensitive unless they receive a plasmid
containing the gene coding for ampicillin.

The white colonies are isolated and the mutant plasmid DNA from
each of these colonies is then amplified, extracted, and analyzed by
gel electrophoresis for altered DNA mobility (gross alterations).
Plasmid with a normal gel pattern are considered to contain putative
point mutations in the sgpfi gene and are further characterized by DNA
sequencing. Examples of an agarose gel pattern and a DNA sequencing

pattern are shown in Figures 4 and 5.

72

Figure 4. Agarose gel electrophresis analysis of plasmids containing
mutations at the sgpfi gene. Plasmid DNA was prepared from
overnight bacterial culture and isolated from E. go_l_i_ by
rapid alkli-lysis procedure. Plasmid DNA was applied and run
on 0.8% agarose gel. The gel was stained with ethidium
bromide (0.5 ug/ml) and the gel was photographed using
transmitted UV light. Lane Z shows wild-type p2189. Lanes
2, 3-8, 10, 11 show sgpfi mutants without altered gel
mobility. Lane 1 shows a sgpfi mutant with gross deletions.
Lane 9 shows a sgLF mutant with gross deletions. Lane M
shows flindIII-digested lambda DNA as molecular weight

markers.

73

122 M34 567891011

 

74

Figure 5. DNA sequencing analysis of a plasmid containing
mutations in the sgpfi gene. Part of the sequence
of a $1915 mutant and of the wild type DNA showing
the 93 G - A and the 104 T -» C.

75

ild type

W

.G‘CACTTGA

Mutant

_
0
3
1

120 ‘-

110 _

 

C

04 T—.

A

l

100 _

~

93 G—aA

90—

80—

—70

60—

—60

 

50—

76
B. Deter-ination of Experimental Conditions to Maximize the Yields of
Progeny Plasmid from Mammalian Cells.

It is critical to rescue a sufficient amount of the progeny
plasmid from mammalian cells to be able to analyze the frequency of
mutants and have enough mutants to sequence. At the begining of my
studies, I determined the optimal yield of the progeny plasmids by
transfecting 0.1 ug to 20 pg of p3AC DNA into ”10 x 106 COS7 cells,
using the CaP04 coprecipitation method of Chu and Sharp (1981). After
the plasmid was rescued, the yield of progeny plasmid was analyzed by
Southern blot analysis (Southern, 1975) (Figure 6). The results showed
that approximately the same yield of progeny plasmid was generated
using 1 to 20 pg of plasmid for transfection. Therefore, 1 pg of p3AC
was used for.further experiments.

To further optimize the yield of progeny plasmid, a time course
experiment was conducted. One ug of plasmid was transferred into ~10 x
106 C05 cells, and the progeny plasmid was harvested at 45 hr, 49hr,
and 57 hr after transfection. The Southern blot analysis (Figure 7)
showed that the yield of plasmid harvested from COS cells at 49 hr was
increased over the yield at 45 hr, but the yeild after 57 hr was not

further increased.

C. Reducing the Background Mutant Frequency.

In the study of p3AC replicating in COS7 cells, the ngl resistant
progeny plasmid obtained from more than ten COS7 transfection
experiments was assayed for mutations in the sgpfi gene. Among a total
number of 21,300 bacterial transformants analyzed, 77 were sypfi

mutants. Therefore, the mutant frequency was 36 x 10‘4. This mutant

77

Figure 6. The relationship between the yield of p3AC generated during
replication in COS7 cells and the amount of input plasmid.
Various amount of p3AC were transfected into COS7 cells as
indicated above lane 1-6. l/20th of the total volume of each
DNA sample extracted from COS7 cells was digested with 0231
and gggRI (lane 1-6) and applied to the gel. Blot
hybridization analysis was carried out by the procedure of
Southern (1975), but the probe was S35-nick translated p3AC.
Lanes 8-10 represent 0.] ng, 1 ng, 10 ng of LQRI-digested
input p3AC used as markers, respectively. Lane 7 contains 10
ng of QpflI-digest p3AC. This enzyme is used to distinguish
the plasmids replicated in g. £911 from those replicated in

COS7 cells.

78

——+—§. L11 sv204a,(

COS7

'r__

 

79

Figure 7. Time course experiment of replication of p3AC in COS7 cells.
1/20th of the total volume of each DNA sample extracted from
COS7 cells at various times (as indicated above lanes 1-6 and
lanes 14-16) was digested with Qpnl (lanes 4—6) or with Qpnl
plus figgRI (lanes 14—16), or was not digested with enzymes
(lanes 1-3). Various amounts of input p3AC were used as
markers (the amounts are indicated above lanes 7-13). Lanes
8-10 shows that p3AC was not digested by enzymes . Lanes
11-13 shows that p3AC was digested with figQRI. Lane 7 shows
that p3AC was digested with 0231. Blot hybridization
analysis was carried out by the method of Southern (1975).
The probe was 32P-nick translated p3AC.

80

_+_ouéogr7zon .1

coli SY204

DNA FROM

g.

l.— DNA FROM
COS7

a+¢ .skm
c+x .zme
o+¢ .gme
a .mco~
m .mzfi

a .m=H.o
use“

me“
m=H.o

a .mcefi
a .gkm

a .gme

a .gme
35m

:me

:me

12 13 14 15 16

11

10

 

81

frequency was somewhat lower than that obtained in the earliest
shuttle vector systems developed, but higher than the frequency
reported by Sarkar gt g1. (1984) (i.e. 23 x 10'4). It proved to be too
high to observe an increase in mutant frequency induced by carcinogens
(see Chapter 111). At the conclusion of my study of the frequency and
kinds of mutants induced when BPDE-treated plasmid (p3AC) replicated in
COS7 cells, I tried to find out if I could lower the mutant frequency,
as well as increase the yield of plasmid, by transfecting a variety of
mamalian cell lines, including SV40-transformed normal human cells
(GM637), SV40 transformed xeroderma pigmentosum cells (XP12R0), and an
human embryonic kidney cell line, 293 transformed with the adenovirus-5
DNA fragments.

Plasmid p3AC was transferred into these cell lines and 48 hr
later, progeny plasmid was rescued from the cells and the yield was
assayed by Southern blot analysis. The results showed that GM637 cells
contained an abundant amount of autonomously replicated SV40 viral DNA,
but little p3AC DNA. The XPlZRD cells also produced little yield of
plasmid. Only the 293 cells produced yields comparable to those I had
obtained from COS7 cells.

The second strategy to lower the background frequency in my
studies was to use another vector, p2189, in which the target gene is
placed between the gene for ampicillin resistance and the origin for
replication in E ggli. Tests with that plasmid showed a low
background frequency of mutants, as well as a reduced recovery of
spontaneous gross deletions produced during replication in monkey CV1
cell line (Seidman gt gt., 1985; Hauser gt gt., 1986). I then applied

this vector in the mutagenesis assay, and determined the background

82
mutant frequency as well as the yield of progeny plasmid when p2189
replicated in either COS7 or 293 cells. The results were compared with
what I had found with p3AC.

Comparison of the frequency of spontaneous gggfi mutants produced
during replication of these two plasmids in COS7 cells is shown in
Table 1. The spontaneous mutant frequency of vector p2189 was 3.6 fold
lower than that of vector p3AC. The yield of progeny plasmids of p2189
recovered from COS7 cells was approximately ten fold higher than that
of p3AC (data not shown). These data indicate that p2189 is better for
investigating mutations in induced in the gggfi gene during replication
in mammalian cells than is p3AC.

I also compared the yield of the two plasmid in 293 cells. A very
poor recovery yield of p3AC progeny plasmids during replication in 293
cells was obtained. In contrast, the yield of p2189 progeny plasmids
during replication in 293 was very high. The yield of p2189 progeny
plasmids was five times higher than that obtained in COS7 cells and the
background mutant frequency (1 x 10‘4, Table 2) was ten times lower
than that observed in COS7 cells (Table 3). Agarose gel analysis of
mutants derived from p2189 showed that gross deletion/insertion
mutations were greatly reduced; only 16% compared to 78% derived from
p3AC replication in COS7 cells (see Chapter III and IV). Therefore,
293 cells proved to be particularly good for the p2189 shuttle vector

mutagenesis studies.

83

 

 

Table 1. Comparison of the frequency of spontaneous
sugF mutants generated during replication of
p3AC or p2189 in C057 cells.

Number Number Number of

Vector of of sugF sgpF mutants

transformants mutants x 104

p3AC 21,300 77 36

p2189 34,710 36 10

 

84

Table 2. Frequency of supF mutants generated during
replication of untreated pZ189 in COS7 cells.

 

 

 

Transfection Number Number of
experiment of gggfi
number transformants mutants

1 15,470 14

2 1,190 0

3 4,090 2

4 4,210 7

5 5,325 7

6 1,885 2

7 2,540 4

Total 34,710 36

 

Average frequency = 10.3 x 10'4

85

Table 3. Frequency of spontaneous supF mutants observed

during replication of p2189 in 293 cells.

 

 

Transfection Number Number Frequency of
experiment of of gggfi gggfi mutants
number transformants mutants x 104

1 68,435 4 0.58

2 25,410 2 0.79

3 10,425 1 0.96

4 3,375 0 -—

5 3,235 0 -—

6 13,560 2 1.50

7 10,590 1 0.94

8 11,310 3 2.65

9 2,735 0 -

10 5,455 1 1.80

 

Total 154,530 14 0.91

 

REFERENCE

Chu, A. and Sharp, P. A. (1981), SV40 DNA transfection of cells in
suspension: analysis of the efficiency of transcription and
translation of T-antigen, Gene 13, 197-202.

Gluzman, Y. (1981), SV40-transformed simian cells supported the
replication of early SV40 mutants, Cell 23, 175-182.

Graham, F. L., Smiley, J., Russell, H. C. and Mairn, R. (1977),
Characteristics of a human cell line transformed by DNA from human
adenovirus type 5, J. Gen. Virol. 36, 59-74.

Hanahan, D. (1983), Studies on transformation of Escherichjg cglj with
plasmids, J. Mol. Biol. tgg, 557-580.

Hauser, J., Seidman, M. M., Sidur, K. and Dixon, K. (1986), Sequence
spdcificity of point mutations induced during passage of a UV-
irradiated shuttle vector plasmid in monkey cells, Mol. Cell. Biol. g,
277-285.

Hirt, B. (1967), Selective extraction of polyoma DNA from infected
mouse cell cultures, J. Mol. Biol. gt, 365-369.

Lebkowski, J. S., DuBridge, R. B., Antell, E. A., Greisen, K. S., and
Calos, M. P. (1984), Transfected DNA is mutated in monkey, mouse, and
human cells, Mol. Cell. Biol. 5, 1951-1960.

Sarkar, S., Dasgupta, U. and Summers, N. C. (1984), Error-prone
mutagenesis detected in mammalian cells by a shuttle vector containig
the supF gene of Escherichig coli, Mol. Cell. Biol. 5, 2227-2230.

Seidman, M. M., Dixon, K., Razzaque, A., Zagursky, R. J. and Berman, M.
L. (1985), A shuttle vector plasmid for studying carcinogen-induced
point mutations in mammalian cells, Gene 28, 233-237.

Southern, E. M. (1975), Detection of specific sequences among DNA
fragments separated by gel electrophoresis, J. Mol. Biol. 28, 503-517.

86

CHAPTER III

Kinds of Mutations Formed When a Shuttle Vector

Containing Adducts of Benzo[a]pyrene-7,8-diol-9,10-epoxide
Replicates in COS7 Cells

Jia-Ling Yang, Veronica M. Maher, and J. Justin McCormick

Carcinogenesis Laboratory, Fee Hall
Department of Microbiology and Department of Biochemistry

Michigan State University, East Lansing, MI 48824-1316

87

88
SUMMARY

We have investigated the kinds of mutations induced when a shuttle
vector containing covalently bound residues of the (iI-7B ,Ba-
dihydroxy-9cz ,10 a -epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE)
replicates in the monkey kidney cell line COS7. The target for
detecting mutations was the ZOO-base pair gene for a tyrosine
suppressor tRNA (gggfi), inserted at the EggRI site in shuttle vector
p3AC (Sarkar gt gt., Mol. Cell Biol. 4:2227-2230, 1984). When
introduced by transformation, a functioning gggfi gene in progeny
plasmid recovered from COS7 cells allows suppression of a lgg; amber
mutation in the indicator Escherichia coli host. Treatment of p3AC
with BPDE caused a linear increase in the number of BPDE residues bound
per plasmid. Untreated plasmids and plasmids containing 6.6 BPDE
residues were transfected into COS7 cells, and the progeny were assayed
for mutations in the gggfi gene. The frequency of mutants generated
during replication of the BPDE-treated plasmids was not higher than
that from untreated plasmids, but the two populations differed markedly
in the kinds of mutations they contained. Gel electrophoresis analysis
of the size alterations of 77 mutant plasmids obtained with untreated
DNA and 45 obtained with BPDE-treated DNA showed that the majority of
the mutant progeny of untreated plasmids exhibited gross alterations,
principally large deletions. In contrast, the majority of the mutants
generated during replication of the BPDE-treated plasmids contained
only minor alterations, principally point mutations. Sequence analysis
of progeny of untreated plasmids containing putative point mutations

showed insertions and deletions of bases and a broad spectrum of base

89
substitutions; in those from BPDE-treated plasmids, all base

substitutions involved guanosine-cystosine pairs.

90
INTRODUCTION

As part of a study of the mechanisms of carcinogenesis, we are
investigating at the sequence level the specific kinds of mutations
induced in mammalian cells by carcinogens, including (1)43 ,80 -
dihydroxy-9cz,10<1-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE), a
major reactive metabolite of the widely distributed environmental
carcinogen benzo[a]pyrene (26). Several investigators have examined
the kinds of mutations induced by BPDE in bacteria, but not in
mammalian cells. For example, Eisenstadt et al. (8) treated nucleotide
excision repair-deficient Escherichia ggli with BPDE and genetically
analyzed a large number of nonsense mutations in the lggl gene for the
kinds of base substitutions induced. The majority proved to be G°C-+
T-A. Chakrabarti gt _l. (6) treated plasmid with BPDE, isolated a
specific fragment in the gene for tetracycline resistance, ligated it
back into the complementary part of unmodified plasmids, and then
transformed E. ggli with the chimeric plasmid containing the localized
patch of BPDE adducts to determine their effect on plasmid survival and
mutagenesis. They found that the majority of the mutations did not
involve major alterations in the target gene. Sequence analysis of
five mutants showed that three had G°C pair deletions, one had a G-C-4
T-A transversion, and one had a A-T -s G-C transition. Mizusawa gt
g1. (18) conducted a similar type of study, treating plasmids with BPDE
and transfecting them into various E. _c_ol_i strains with different
repair capacities to study the effects of BPDE on plasmid survival in

the various recipients and on the frequency of mutants (gglK+ —-> galls).

Sequence analysis was not carried out because of the large size of the

91

gene, but gel electrophoresis analysis of the mutant plasmids taken
from the bacteria showed that the majority (15 of 16) did not involve
detectable alterations in the size of the DNA fragment of interest. In
a related study involving a gglK -+ gg1K+ selection system and a 230-
base pair (bp) transcription termination sequence as the target for
mutations, sequence analysis of the DNA from 15 gglK+ colonies revealed
three with mutations in the region of interest (19). Two of these
showed insertion of TVA into a cluster of T°A base pairs; the third
showed a deletion of G-C from a cluster of G-C base pairs.

Indirect evidence suggests that the mechanisms of mutagenesis by
certain carcinogens in mammalian cells can differ from that in bacteria
(2). Therefore, we treated a shuttle vector with tritium-labeled BPDE
and transfected it into the monkey kidney cell line COS7 to allow
replication to occur. Mutant progeny plasmids were identified by

transforming an E.

ggl_i_ indicator host and analyzed for mutations in
the target gene, the 200-bp tyrosine suppressor tRNA (gggfi). He found
that the frequency of gross alterations formed in the progeny of BPDE-
treated plasmid was significantly lower than with untreated plasmid,
and DNA sequence analysis showed that, unlike the case for the
untreated plasmid, the majority of the point mutations obtained with
BPDE-treated plasmid were base substitutions, and all of these involved

a G-C base pair transition or transversion.

92
MATERIALS, METIKDS, AND RESULTS

Cells and plasmids. The 6.6 kbp shuttle vector used p3AC,
constructed by Sarkar gt g1. (23) and provided by N.C. Sumners,
contains parts of a p8R322 plasmid including the origin of replication
and the gene for ampicillin resistance (gmg), the BamHI and 1_1p_aII
fragments of the early region of simian virus 40 DNA, and the 200-bp
sggfi gene, which serves as the target gene for mutagenesis and
subsequent sequencing. The ampicillin-sensitive indicator bacterial
host, E. go_l_i SY204, carries an amber mutation in the B-galactasidase
gene (23). The eucaryotic host of the shuttle vector, COS7 simian
cells (10), were grown in Eagle’s minimal essential supplemented with
0.2mM L-serine, 0.2mM L-aspartic acid, 1M sodium pyruvate and 10%
fetal calf serum (Gibco Laboratories, Grand Island, NY).

Preparation of plasmid DNA containing BPDE adducts. Plasmid were
prepared by an alkaline lysis procedure (16) and purified by ethidium
bromide-CsCl density centifugation. A small volume (1-16 pl) of
generally tritiated BPDE (692 mCi/mmole, 0.3 mg/ml in tetrahydrofuran,
96% pure [Midwest Research Institute, Kansas City, M0.]), was diluted
in anhydrous acetone immediately before use and added to 200 pl of a
0.5-mg/ml solution of DNA in 10 mM Tris hydrochloride-1 mM EDTA buffer,
pH 8.0. The mixture, protected from light, was incubated at room
temperature for 2 h. Unbound BPDE was removed by three successive
ethanol precipitations, and the moles of BPDE residues bound per mole
of p3AC was calculated from the A250 profile of the DNA and the
specific activity. The number of BPDE adducts per molecule of plasmid
was proportional to the concentration of BPDE, with 5 pM giving 6.6

93
BPDE residues per plasmid.

Transfection of COS7 cells and assay of progeny plasmid fur gggf
mutations. COS7 cells in suspension were transfected with untreated
plasmid or plasmid containing 6.6 BPDE, using the DNA-calcium
phosphate coprecipitation method of Chu and Sharp (7). After 48 h,
plasmid DNA was extracted by the procedure of Hirt (13), purified with
phenol, treated with RNase A (50 pg/ml) at 37°C for 1 h, followed by
proteinase K (100 pg/ml) at 50°C for 2 h, and then extracted with
phenol-chloroform, precipitated with ethanol, and further purified by
drop dialysis (26). The purified plasmid was treated with 0931 to
digest any input plasmid and then was used to transform SY204 bacterial
cells to ampicillin resistance, using the method of Hanahan (11).
Transformants were selected on Luria-Bertani (LB) broth plates
containing 5-bromo-4-chloro-3-indolyl- fii-D-galactoside (X-Gal) (40
mg/liter), an inducer, isopropyl- B-D-thiogalactoside (20 mg/liter),
and ampicillin (50 mg/liter). Cells containing plasmids with a
functioning ggpfi gene which suppresses the amber mutation can form blue
colonies on X-Gal plates, whereas cells containing plasmids with a
mutated, inactive ggLF gene form white colonies. The frequency of
mutants was determined by dividing the number of white colonies by the
total number of colonies. The results are shown in Table l.

Charaterization of mutant plasmids by gel electrophoresis and
sequencing. Bacterial cells from white colonies were restreaked on
fresh X-Gal plates containing isopropyl- B-D-thiogalactoside and
ampicillin to confirm their phenotype, and their DNA was extracted,
purified, and analyzed by electrophoresis on 0.8% agarose gels for

altered DNA mobility (gross alterations). Plasmids with normal agarose

94

Table 1. Frequency of supF mutants obtained with BPDE-treated or
untreated p3AC that had replicated in C057 cells.

 

 

Treatment No. of No. of Mutant
transformants mutants frequency
None 21.300 77 36 x 10-4

BPDE 13,500 45 33 x 10-4

 

95

gel patterns were digested with EggRl and analyzed by electrophoresis
on 6% polyacrylamide gels for changes in the size of the ggpj gene.
The data for 77 mutants from untreated p3AC and the 45 mutants obtained
from the progeny of BPDE-treated plasmid are sumarized in Table 2.
The majority of the mutants obtained after transfection of COS7 cells
with untreated p3AC exhibited gross rearrangements, predominantly
deletions but also large insertions. In contrast, the majority of the
mutant plasmids obtained with BPDE-treated DNA exhibited only minor
alterations, and most of these were putative point mutations since they
did not show any visible alteration in size on agarose or
polyacrylamide gels.

Plasmids with a normal polyacrylamide gel pattern were considered
to contain putative point mutations in the gggfi gene. Seven
unambiguously independent mutants of this type derived from untreated
plasmids and nine derived from BPDE-treated plasmids were sequenced,
using a modification of the dideoxyribonucleotide method of Sanger gt
g. (22). DNA was prepared and purified through CsCl gradients as
described above and denatured by alkali to generate single-strand DNA
templates as described previously (28). Polymerization from a pBR322
EgQRI site primer was carried out with the Klenow fragment of DNA
polymerase I. [355] a-dATP (0345; New England Nuclear, Boston, Mass.)
and buffer gradient-denatured polyacrylamide gels were used for greater
resolution (3). The results are shown in Table 3 and Fig. 1. Among
the untreated controls, three of the seven mutants showed a single base
substitution G -» A, G -+ T, and C -+ T; two had two base
substitutions; and two had several changes, including single base

insertions and deletions. In contrast, five of the nine mutants from

96

Table 2. Characterization of the kinds of mutations generated in p3AC

during replication in COS7 cells.

 

No. (%) of mutants

 

 

 

 

Characterization of No-treatment BPDE-treated
mutant p3AC vectors (controls) (6.6 adducts/p3AC)
MINOR ALTERATION
Putative point 12 (15.6%) 15 (33.3%)
mutations
Small deletions or 5 (6.5%) 8 (17.8%)
insertions
(delete or insert <30bp)
GROSS ALTERATION
Deletions (delete ~200bp) 25 (32.5%) 12 (26.7%)
Large deletions 24 (31.1%) 4 (8.9%)
(delete >1 kba)
Large insertions 11 (14.3%) 6 (13.3%)
(insert >1 kb)
TOTAL MUTANTS 77 (100%) 45 (100%)

 

akb, Kilobase

97

Table 3. Analysis of the kinds of base changes found in the supF
gene of the mutant p3ACs analyzeda.

 

 

 

 

Untreated p3AC BPDE-treated p3AC
Type of
change Times % of Times % of
occurring total occurring total
G-C —+ T°A 2 14 6 46
G-C -» A°T 5 36 3 23
C°G -+ G°C 1 7 2 15
A-T -+ 6°C 1 7 0 0
A°T -+ C-G 2 14 0 0
Insert an A 2 14 1 8
Insert a T 0 O l 8
Delete a C 1 7 0 O

 

aOnly those p3ACs containing putative point mutations were

analyzed at the sequence level.

98

Figure 1. Distribution of mutants in the gggfi gene of p3AC. The DNA
strand shown is the strand synthesized during the DNA
sequencing reaction, using the gggRl rightward-sequencing
primer. It corresponds to the tRNA sequence. CI-C7 refers
to mutants obtained with untreated plasmids; B1-B9 refers to

those from treated plasmids.

99

GDLD

.a

o<<<wuuHo<<<<ubppu<u~<uu<uu¢uuUUUHHUUH<<oUHPmo<QQUHhu<ouw<UHouuop<uw<<<bup
cc“ cm~ c- cam > cog

< 8..
<5: 91.]
E

b H <
<

u<o<ua<oow<<<uwooumcouuquoooohochwhuu<~p<uu¢<<<po<uuom<uo<ooo<<m<ouuuppua
on co > oe om om o~ _

< _
I<z~t 28mm <szmai—

mm
mm
mm
mm
mm

cu
Hu
mu
mu

Nu

am
mm
mm
Hm

flu
Nu
mu

100
BPDE-treated plasmids showed a single base substitution; three of nine
had two base substitutions; and one had three base substitutions. All
of these substitutions involved G-C base pairs. One mutant had a base
substitution, but also the insertion of a nucleotide; one had just the
insertion of a nucleotide. The l0cation of each change in the gene can

be seen Fig. 1.

101
DISCUSSION

For comparative purposes, it is useful to calculate the mutant
analysis data of Table 1 and 2 on the basis of the frequency of each
type of mutation. This calculation shows that the frequency of mutant
progeny from untreated plasmids carrying minor alterations was 8 x
10'4, but for the BPDE-treated plasmids it was 17 x 10". This
indicated that BPDE induces point mutations, which supports the
findings of King and Brookes (14), who studied the kinds of mutations
induced in V79 Chinese hamster cells as detected by DNA hybridization,
and those of Aust gt gt. (2), who showed in human diploid fibroblasts
that BPDE does not cause the kinds of mutations (deletions,
rearrangements) expected to completely inactivate the gene coding for
elongation factor 2, which is involved in diphtheria toxin resistance.
That all base substitutions observed with progeny of BPDE-treated
plasmid involved G-C pairs is in keeping with the hypothesis that BPDE
adducts were involved since this carcinogen predominantly binds to
guanine (26,27).

A similar analysis of the data in Tables 1 and 2 for the frequency
of mutants carrying gross alterations shows that for the progeny of
untreated plasmid this value was 28 x 10'4, with 23 x 10'4 being
deletions and 5 x 10'4 being insertions. This result is not
surprising. Evidence from several groups of investigators indicates
that, when simian virus 40-based plasmids are introduced into mammalian
cells, they are subjected to strand breaking, deletions, duplications,
recombination, and more complex rearrangements, including the insertion

of sequences from the mammalian cell genome (1,4,5,15,17,20,21).

102

However, with the progeny of BPDE-treated plasmids, the frequency of
mutants carrying such gross alterations was significantly lower than
background, only 16 x 10", with 12 x 10"4 being deletions, compared to
23 x 10'4 in the control. (The frequency of insertions was about the
same in both populations.) The failure to recover the background level
of plasmids carrying deletions suggests that the presence of BPDE
adducts interferes in some way with the process that produces these
gross rearrangements. Another explanation is that BPDE adducts
increase size of the deletions so that they extend into the gmp gene.
Such plasmids would not be recovered in our assay.

Gross rearrangements were rarely seen when we transformed E. ggli
SY204 directly with BPDE-treated or untreated p3AC without first
transfecting COS7 cells (data not shown). This result agrees with the
results of Chakrabarti gt gt. (6) who did not observe rearrangements
following transformation of E. gglt with plasmid. Neither did Glazer
gt gl. (9) find rearrangements in the gggE gene when it was integrated
into a mammalian cell chromosome and the host cell was exposed to
mutagens.

One important advantage of using the gng gene as the target for
mutation studies at the sequence level is its small size. Another is
that the tRNA gene is responsive to base substitutions in many
different positions (9,12,23). Recently Seidman gt g1. (24) reported
construction of a shuttle vector, p2189, in which the §_ugE gene is
strategically located between the origin of replication of the plasmid
in E. ggli and the gene for ampicillin resistance so that the
possibility of recovering mutants containing these spontaneous gross

rearrangements is greatly decreased. This plasmid will make it easier

103
to study the types of minor alterations induced by various mutagens
(12) and we are currently using it to obtain such data for BPDE.
Nevertheless, use of p3AC in the present study made it possible for us
to detect the effect of BPDE on the processes responsible for

generating the spontaneous gross alterations.

104

ACKNOWLEDGEMENTS

We are grateful to Dr. William C. Summers for providing us with
the plasmid and the host cells and for many fruitful discussions during
this work and to Dr. Saumyendra Sarkar for his helpful advice on the
project. We thank Bernard Schroeter for his technical assistance and
Carol Howland for typing the manuscript.

This research was supported in part by Public Health Service Grant
CA21253 from the National Cancer Institute and by a Grant from the

Women’s Auxiliary of the Veterans of Foreign Wars.

10.

11.

12.

REFERENCES

. Ashman, C. H., and R. L. Davidson. 1984. High spontaneous

mutation frequency shuttle vector sequences recovered from
mammalian cellular DNA. Mol. Cell. Biol. 4:2266-2272.

. Aust, A. E., N. R. Drinkwater, K. Debien, V. M. Maher, and J. J.

McCormick. 1984. Comparison of the frequency of diphtheria toxin
and thioguanine resistance induced by a series of carcinogens to
analyze their mutational specificities in diploid human
fibroblasts. Mutat. Res. 125:95-104.

. Biggen, M. D., T. 1.3.5 Gibson, and G. F. Hong. 1983. Buffer
5

gradient gels and label as aid to rapid DNA sequence
determination. Proc. Natl. Acad. Sci. U.S.A. 80:3963-3965.

. Calos, M. P., J. S. Lebkowski, and M. R. Botchan. 1983. High

mutation frequency in DNA transfected into mammalian cells. Proc.
Natl. Acad. Sci. U.S.A. 80:3015-3019.

. Chakrabarti, S., S. Joffe, and M. M. Seidman. 1985. Recombination

and deletion of sequences in shuttle vector plasmids in mammalian
cells. Mol. Cell. Biol. 5:2265-2271.

. Chakrabarti, S., H. Mizusawa, and M. Seidman. 1984. Survival and

mutagenesis of bacterial plasmids with localized carcinogen
adducts. Mutat. Res. 126:127-137.

. Chu, A., and P. A. Sharp. 1981. SV40 DNA transfection of cells in

suspension: analysis of the efficiency of transcription and
translation of T-antigen. Gene 13:197-202.

. Eisenstadt, E., A. J. Harren, J. Porter, 0. Atkins, and J. H.

Miller. 1982. Carcinogenic expoxides of benzo[a]pyrene and
cyclopental[cd]pyrene induce base substitutions via specific
transversions. Proc. Natl. Acad. Sci. U.S.A. 79:1945-1949.

. Glazer, P. M., S. N. Sarkar, and H. C. Summers. 1986. Detection

and analysis of UV-induced mutations in mammalian cell DNA using a
lambda phage shuttle vector. Proc. Natl. Acad. Sci. U.S.A.
83:1041-1044.

Gluzman, Y. 1981. SV40-transformed simian cells support the
replication of early SV40 mutants. Cell 23:175-182.

Hanahan, D. 1983. Studies on transformation of Egghgrigfitg coli
with plasmids. J. Mlol. Biol. 166:557-580.

Hauser, J., M. M. Seidman, K. Sidur, and K. Dixon. 1986. Sequence
specificity of point mutations induced during passage of a UV-
irradiated shuttle vector plasmid in monkey cells. Mol. Cell.
Biol. 6:277-285.

105

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

106

Hirt, B. 1967. Selective extraction of polyoma DNA from infected
mouse cell cultures. J. Mol. Biol. 26:365-369.

King, H. H. S., and P. Brookes. 1984. On the nature of the
mutations induced by the diolepoxide of benzo[a]pyrene in mammalian
cells. Carcinogenesis 5:965-970.

Lebkowski, J. S., R. B. DuBridge, E. A. Antell, K. S. Greisen, and
M. P. Calos. (1984). Transfected DNA is mutated in monkey, mouse,
and human cells. Mol. Cell. Biol. 4:1951-1960.

Maniatis, T., Fritsch, E. F., and Sambrook, J. 1982. Molecular
Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold
Spring Harbor, New York.

Miller, J. H., J. S. Lebkowski, K. S. Greisen, and M. P. Calos.
1984. Specificity of mutations induced in transfected DNA by
mammalian cells. EMBO J. 3:3117-3121.

Mizusawa, H., C.-H. Lee, T. Kakefuda. 1981. Alteration of plasmid
DNA-mediated transformation and mutations induced by covalent
binding of benzo[a]pyrene-7,8-dihydrodiol-9,10-oxide in Escherichia
goli. Mutat. Res. 82:47-57.

Mizusawa, H., C.-H. Lee, T. Kakefuda, K. McKennedy, H. Shimatake,
and M. Rosenberg. 1981. Base insertion and deletion mutations
induced in an Escherichig coli plasmid by benzo[a]pyrene-7,8-
dihydrodiol-9,IO-epoxide. Proc. Natl. Acad. Sci. U.S.A. 78:6817-
6820.

Razzaque, A., S. Chakrabarti, S. Joffe, and M. Seidman. 1984.
Mutagenesis of a shuttle vector plasmid in mammalian cells. Mol.
Cell. Biol. 4:435-441.

Razzaque, A., H. Mizusawa, and M. M. Seidman. 1983. Rearrangement
and mutagenesis of a shuttle vector plasmid after passage in
mammalian cells. Proc. Natl. Acad. Sci. U.S.A. 80:3010-3014.

Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing
with chain-terminating inhibitors. Proc. Natl. Acad. Sci. U.S.A.
74:5463-5467.

Sarkar, S., U. B. Dasgupta, and H. C. Summers. 1984. Error-prone
mutagenesis detected in mammalian cells by a shuttle vector
containing the sgpF Gene of EscherichiLcoli. Mol. Cell. Biol.
4:2227-2230.

Seidman, M. M., K. Dixon, A. Razzaque, R. J. Zagursky, and M. L.
Berman. 1985. A shuttle vector plasmid for studying carcinogen-
induced point mutations in mammalian cells. Gene 38:233-237.

Silhavy, T. J., M. L. Berman., and L. H. Enquist. 1984.
Experiments in Gene Fusions. Cold Spring Harbor Laboratory, Cold
Spring Harbor, New York.

107

26. Heinstein, I. B., A. M. Jeffrey, K. N. Jennette, S. H. Blobstein,

27.

28.

R. G. Harvey, C. Harris, H. Autrup, H. Oasai, and K. Nakanishi.
1976. Benzo[a]pyrene diol epoxides as intermediates in nucleic
acid binding in vitro and in vivo. Science 193:592-595.

 

Yang, L. L., V. M. Maher, and J. J. Mc ormick. 1980. Error-free
excision of the cytotoxic, mutagenic N deoxyguanosine DNA adduct
formed in human fibroblasts by (1)43 ,8 a -dihydroxy-9a ,10a-
epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene. Proc. Natl. Acad. Sci.
U.S.A. 77:5933-5937.

Zagursky, R. J., K. Baumeister, N. Lomax, and M. L. Berman. 1985.
Rapid and easy sequencing of large linear double-stranded DNA and
supercoiled plasmid DNA. Gene Anal. Techn. 2:89-94.

CHAPTER IV

Kinds of Mutations Formed Hhen a Shuttle vector
Containing Adducts of
(1)48 ,8 a -dihydroxy-9 a ,10 a -epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene

Replicates in Human Cells

Jia-Ling Yang, Veronica M. Maher, and J. Justin McCormick

Carcinogenesis Laboratory, Fee Hall
Department of Microbiology and Department of Biochemistry

Michigan State University, East Lansing, MI 4B824-l3l6

108

109
SUMMARY

He have investigated the kinds of mutations induced when a
shuttle vector containing covalently bound residues of (_+_)-7B ,BQ-
dihydroxy-9a ,lOa -epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE)
replicates in human cells. A human embryonic kidney cell line, 293,
was used as the eucaryotic host. The target gene for mutation
analysis, gggE, codes for a tyrosine supppressor tRNA and is
strategically located between the origin of replication of the plasmid
in Escherichia ggli and the gene for a selectable marker, so that the
possibility of recovering gggE mutants containing gross rearrangements
is low. The frequency of gggE mutants obtained when untreated plasmid
replicated in 293 cells was 1.4 x 10'4. The frequency with BPDE-
treated plasmid increased linearly as a function of the number of
adducts, with 16 adducts per plasmid giving 38 x 10'4. Polyacrylamide
gel and agarose gel electrophoresis analysis of 137 plasmids. with
mutations in the gggE gene indicated that 70% (21/30) from untreated
plasmids contained deletions or insertions or showed altered gel
mobility, whereas only 28% (30/107) of those devired from BPDE-treated
plasmid contained such alterations. Of the 86 unequivocally
independent mutants derived from BPDE-treated plasmids that were
analyzed by sequencing, the majority (60/86) exhibited base
substitutions. Mutants exhibiting frameshifts (insertions or deletions
of one, two, or four base pairs) were also found, but they were a
minority (ll/86). In the progeny of BPDE-treated plasmids 61/71 base
substitutions observed were transversions, with 45/61 G-C. -+ T°A.

Examination of the location of BPDE-induced mutations among the 85 base

110
pairs in the structure of the tRNA revealed that 30% of the base
substitutions occurred at two sites and 44% of the rest occurred at
five other hot spots. Only 20% of all these base changes involved a
site in which a guanine containing a BPDE adduct is predicted to be

labile-—i.e., a guanine that has a pyrimidine to its 5’ side.

111
INTRODUCTION

The molecular mechanisms responsible for the induction of
mutations in mammalian cells are not well understood. Attempts to
deduce the nature of such mechanisms from the products of the
mutagenic events have been hampered up until now by an inability to
isolate and analyze newly-mutated genes at the sequence level.
However, several elegant systems have recently been devised (l-lO)
that allow rescue from mammalian cells of DNA containing mutations in
target genes whose sequences are known. He have adapted one of these
systems to investigate the kinds of mutations induced when DNA
containing covalently-bound residues of (i)-7B,8a-dihydroxy-9a,l0a
-epoxy-7,8,9,lO-tetrahydrobenzo[a]pyrene (BPDE), a major reactive
metabolite of the widely distributed environmental carcinogen
benzo[a]pyrene (ll), replicates in human cells. Our study of
BPDE-induced mutation is part of a larger study of the molecular
mechanisms of mutagenesis and carcinogenesis in human cells.

He treated a shuttle vector containing a small defined target
gene, gggE, which codes for a tyrosine suppressor tRNA. Use of this
particular gene offers the advantage that base substitutions at any one
of a large number of positions among the 85 nucleotides making up the
tRNA structure results in a detectable phenotypic change (6,7,10,l2).
Use of the human cell line 293 as the eukaryotic host for replicating
the plasmid offers the advantage of a very low spontaneous background
frequency of mutants (13). Hith this system we have found that BPDE
induces point mutations, principally base substitutions, as a linear

function of the number of adducts per plasmid. DNA sequence analysis

112
of the mutations generated in human cells with BPDE-treated plasmids
has allowed us to determine the type of structural alterations induced,
as well as their specific location in the target gene. Our results
indicate that the majority of the base substitutions observed did not
occur at sites predicted by Lobanenkov gt g1. (14) to be particularly
subject to depurination-—i.e., ones in which the guanine containing a

BPDE adduct has a pyrimidine adjacent to it at the 5’ side.

113
MATERIALS AND METHODS

Cells and Plasmid. A human embryonic kidney cell line, 293,
transformed by adenovirus-5 DNA fragments (l5) served as the
eucaryotic host. It was obtained from Dr. Michele Calos (Stanford
University, Stanford, CA). The cells were grown in Eagle’s minimal
essential medium supplemented with 0.2mM L-serine, 0.2mM L-aspartic
acid, lmM sodium pyruvate and 10% fetal calf serum (GIBCO). The
ampicillin sensitive indicator bacterial host, Eschgrichja ggli SY204,
carrying an amber mutation in the fi-galactasidase gene, was obtained
from William C. Summers (Yale University, New Haven, CT) and has been
described (3). The 5504-base-pair shuttle vector used, le89,
constructed by Seidman gt gt. (4), contains the gene for a tyrosine
suppressor tRNA, gan, which is flanked by two genes essential for
recovery of the plasmid in E. ggLi, i.e., the gene for ampicillin
resistance and the bacterial origin of replication, as well as the
origin of replication and large tumor antigen gene from simian virus
40. It was provided by Kenneth Kraemer (National Cancer Institute).

Formation of BPDE Adducts on the Plasmid. Plasmid was prepared
by using an alkaline lysis procedure and purified by ethidium
bromide/CsCl density centrifugation (16). A small volume (l-4 pl) of
generally tritiated BPDE (692 mCi/mmole, 0.3 mg/ml in tetrahydrofuran,
96% pure; Midwest Research Institute, Kansas City, MO; 1 Ci . 37 GBq)
was added to 200 pl of an 0.5 mg/ml solution of DNA in lO mM Tris-HCl,
l mM EDTA buffer, pH 8.0. The nfixture was protected from light and

incubated at room temperature for 2 hr. Unbound BPDE was removed by

three successive precipitations with ethanol and the number of moles of

114
BPDE residues bound per mole of p2189 was calculated from the UV250
absorption profile of the DNA and the specific activity.

Transfection and Rescue of Replicated Plasmid. Human 293 cells
(0.5-l x l05) were plated into a series of l50-nIn-diameter dishes.
After 24 hr later they were transfected with 6 pg of plasmid by using
calcium phosphate coprecipitation but omitting the glycerol shock
(17). After 48 hr, low molecular weight DNA (plasmid) was extracted
and separated from cellular DNA by the procedure of Hirt (l8), purified
with buffered phenol, and treated with RNase A (50 pg/ml) at 37°C for l
hr, followed by proteinase K (lOO pg/ml) at 50° C for 2 hr, then
extracted with phenol/chloroform, precipitated with ethanol, and
further purified by drop dialysis (l9). Care was taken to keep the DNA
extracted from the target cells in each individual lSO-mm dish
separate from the next so that we could know if any mutants arose from
the same population (putative siblings). Before the plasmid DNA from
each Hirt supernatant was used to transform the indicator bacteria, it
was treated with Opal (20 units) to digest any original input plasmid.
This restriction enzyme digests any DNA that still has the bacterial
methylation pattern generated when the plasmid was first prepared.

Bacterial Transformation and Mutant Identification. .E» gglt
SY204 cells were transformed to ampicillin resistance (20) and selected
on Luria-Bertani broth agar plates containing 5-bromo-4-chloro-3-
indolyl 13-D-galactoside (X-Gal) (40 mg/liter), an inducer, isopropyl
8-D-thiogalactoside (IPTG) (20 mg/liter), and ampicillin (50
mg/liter). Cells containing plasmids with a functioning sggE gene,
which suppresses the amber mutation, can form blue colonies on X-Gal

plates, whereas cells containing plasmids with a mutated, inactive supF

115
gene form white or light-blue colonies. Each colony was restreaked on
fresh plates to confirm the phenotype.

Characterization of Mutants. Plasmid DNA from cells from white or
light-blue colonies was amplified, extracted, dissolved in buffer, and
analyzed by electrophoresis on 0.8% agarose gels for altered DNA
mobility (gross alterations). Plasmids without evidence of' gross
alterations were amplified further and purified by CsCl centrifugation
and analyzed by a secondary bacterial transformation to ensure that the
cells’ inability to utilize X-Gal actually reflected inactivation of
the gggE gene (see Egggttg) If it did, these mutants were sequenced
using the dideoxyribonucleotide method of Sanger gt g1. (2l) with the
following modifications. Plasmids were denatured with alkali to
generate single-strand templates (22) and polymerization from a p8R322
EggRI site primer was carried out with the Klenow fragment of DNA
polymerase 1. Both were purchased from Boehringer Mannheim.
3Y’s-labeled adenosine 5'-[a-thio]triphosphate (New England Nuclear,
034$) and buffer gradient polyacrylamide gels were used for greater

resolution (23).

116
RESULTS

Characterization of BPDE-treated plasmids. le89 was treated with
BPDE at various concentrations and assayed for the number of residues
bound per plasmid and for loss of ability to transform SY204 cells to
ampicillin resistance. As shown in Fig. l, there was a linear
relationship between the number of BPDE-induced adducts per plasmid and
the BPDE concentration used, and the decrease in transforming activity
was directly proportional to the number of BPDE residues bound. Nine
residues per le89 (l.6 per l03 bp) were required to lower the
transforming activity to 37% of the control.

Frequency of gggE Mutants Induced by BPDE Treatment. BPDE-treated
and untreated plasmids were transfected into the human cells and
allowed to replicate. The progeny plasmids were rescued and introduced
into Et ggli SY204 cells to identify those carrying mutations in the
gggE gene. Fourteen transformation experiments using progeny of
untreated plasmid yielded 48 white or light-blue colonies out of
2l8,750 transformants, an apparent background frequency of 2.2 x l0'4.
However, when plasmid from these colonies was retested by a secondary
transformation study in SY204, l7 proved not to have a mutant gng
gene, indicating that the inability of the bacteria to utilize X-Gal
resulted from a separate mutation in the bacteria, not from the lack of
an active suppressor tRNA. The frequency of these non-gggE mutants was
0.8 x 10". Therefore, the background frequency of gggE mutants among
the progeny of plasmids replicating in the human cells was actually l.4
x l0’4. Agarose gel and DNA sequence analysis of these mutants showed

that 70% of the spontaneous supF mutants contained deletions or

117

Figure 1. (A) Number of BPDE residues bound per plasmid as a function
of concentration of BPDE. The symbols (OandA) simply
indicate that the plasmid was treated with BPDE in two
separate experiments. (B) Relative frequency of
transformation of bacteria to ampicillin resistance by
plasmid le89 modified with BPDE compared to untreated
plasmid. The error bars indicate the SEM of four or five

determinations.

BPDE(pM)

118

 

 

 

 

 

$11.11 I I I .38
O 1
— r+4 --4
N_ m :1“-
9_ :
_ :9
9"" 71-0—19 :
LO— f '2'“
--l-+—I :(
IIILL L L l '—
OO O O O O
9 (O <I' N -

(%) AONBIOHdB
NOIiVWEOdSNVEi EALLV'ISH

 

 

lllllTTllllTTIFTTT
\

d

/.
I
IS 20

 

 

O
N

— 04 -9
_. <12 \4 -L0
\
.. 1 ‘ _
IIILLLILILIILILILINO
l0 0 l0 0

GIWSV'ld 83d
GNOOB SBOGISEH 30d8

BPDE RESIDUES BOUND
PER PLASMID

BPDE (pM)

119
insertions or had altered gel mobility (gross alteration). Therefore,
the background frequency of point mutation mutants was only 0.4 x lO‘4
(9/2l8,750) (Table l).

As shown in Table l and Fig. 2, the frequency of ggpj mutants
generated during replication of BPDE-treated plasmid in 293 cells
increased as a linear function of BPDE residues per le89, reaching 38
x 10'4 for plasmids containing an average of l6 adducts. Agarose gel
and DNA sequence analysis of these mutants showed that none of the
plasmids giving high frequencies (i.e., 22.9 x 10'4 or 37.8 x 10'4)
exhibited deletions and only one contained an insertion (20 bp) (Table
1). Therefore, the frequency of point mutations in the progeny of the
le89 treated with BPDE at the two highest concentrations was 60 and 95
times higher than background, respectively. Even the M mutants
derived from plasmids exposed to lower doses of BPDE—i.e., those
containing 2.0 to 8.7 residues per plasmid, exhibited a much lower
proportion of deletion, insertion or altered gel mobility than did
those derived from untreated plasmid.

Evidence from two kinds of control experiments indicated that the
gggE mutants observed were generated in the human cells rather than in
the bacteria. First, aliquots of plasmid isolated from human cells and
treated with 9931 before being used to transform bacteria to ampicillin
resistance gave a high yield of transformants. But when an aliquot of
the same Hirt supernatant plasmid preparation was treated with flggl,
the number of transformants was drastically reduced-e.g., from l200 to
l. Mggl digests plasmids containing the mammalian cell replication
pattern. A second proof comes from experiments in which the indicator

bacteria were transformed directly, using the original input

120

.qn N so — mo coNNLNN=N so .coNNopuu .=o_N:N.NNa:m+

.aa cN - N. Na =o_NLNN=_.

.Na cm. - N No =o_aa_oo.

.Am ass—cu .Louacoe==. ANN—Nae: _cm tween—a m=.zogN mucausm «Nos» Nap; o gas—cu c. «use an cauN>Nc o ceapou g. guess:
ogu N. Nucaaae »:_oa No coNNoNcw ash .A. ess—ou. Nucoaaasu uu>cmmno ogu Nos.“ mucous: u:.oa up :oNNuacN sec» vans—:upauo
.Aan cm—A. poo unocaaa co ops—N.> co.uacaup<a¢

.Auxuu ummv ocom mama ecu mo co.aa>.~ua=. mos; vow—smog

Nau-x NNNNonNNwm cu NNN—Naac. N.——wu osu gage ogzmca cu :ONNngoumcuca asavcouom a Na taxman. an: acauam goon sac» t_mNa—ac

 

N.NN __ N o p. N_\o N.NN NN..N\N. N o.N_
N._N N. _ o N. N_\o N.NN NNN.N\N_ N N.N.
N.N. N o o N _.\N N.N_ o_N.N\N_ N N.N
..e N c N __ ._\N N.o oNN.NN\N_ N N.N
e.. N_ o N N. NN\N N.N cNo..N\NN N _.N
o.. N. a __ NN NN\N N.N NNN.NN\N. N o.N
..o N N N. NN NN\N N., NNN.N_N\_N N. o

 

Id.

co. x +N:o_uaaze .Nco.ucomc. .NcoNNu—uc kucmsawm cocNeaxo .o: Aoop x. Nucascouncagu mucoENLaaxu uNENapq

 

.co.aaase N:.oa ;N.: .62 :NN: .oz Nocmm mama \rcN».—.noe Nucaaam \«Nocauam :cNuuuumcaca can
«c.6a mumw zu.3 .oz uNENNNQ pom voLmNPN Maud mo uuaw N—ou snazc Nausvv<
Co Nucwaamcu Nacauae pouch nu.) NcNENapm mucoscogu go .62

cmucoaaom mo :o.~a~.gmuuagazu

 

.N—Nmu mow
:_ :o_NNo,_gmc u:_c:v cmumcwcmo mm—Na Co acomogg cu.3 NAdu «u we coNNNELcumcugu an vocNauno mucauam so ".mapac< .— opaap

121

Figure 2. Frequency of gng mutants as a function of the number of BPDE
residues per plasmid. Untreated plasmids or plasmids
containing BPDE residues were transfected into 293 cells and
allowed 48 hr for replication. The progeny plasmids were
rescued, treated with Opal, and transferred into SY204
indicator bacteria by DNA transformation. The frequency of
transformants with a non-functioning gng gene were
identified by colony color and secondary transformation (see
text). The symbols used correspond to those shown in Fig. l.
The error bars refer to SEM of the sggE mutant frequencies
obtained from a series of individual human cell transfection

experiments made with each set of treated plasmid.

FREQUENCY OF
SUP F MUTANTS (x I04)

122

 

4O |flll|+l

 

(N
O
Tllllllll|llll

l\)
O

 

.. ”/4
Og/én'o

 

x2.
llllIllllIllllIllll

 

I l 1

IS 20

BPDE RESIDUES BOUND
PER PLASMID

 

123
BPDE-treated plasmid rather than progeny. Untreated plasmid gave a
background frequency of l white colony per 104 total colonies; plasmid
containing the highest number of BPDE residues, 15.8, gave only 2 per
l04. He did not determine what fraction of the latter were true gggE
mutants.

Characterization of the Mutations Formed in 293 Cells. DNA
sequence analysis of the M gene was carried out on 25 mutant
progeny from untreated plasmids and 93 mutant progeny from BPDE
treated plasmids (Table l). Some of the ll8 mutants sequenced could
have been siblings. However, at least l07 of them, 21/25 from untreated
le89 and 86/93 from BPDE-treated le89, represented unequivocally
independent mutants. This is because either their alteration was unique
or they had not been derived from the same set of transfected human
cells. Therefore, in analyzing the frequency of specific kinds of
mutations (Tables 2 and 3) we included only the unequivocally
independent mutants. The data showed that 62% (13/2l) of the mutants
from the control contained deletions or insertions of more than 4 bp,
compared to only l7% (l5/86) from the BPDE-treated plasmids (Table 2).
In the latter, the majority of the deletions were found with mutants
produced when the plasmid contained only 2 BPDE-induced adducts (Table
1). Table 2 shows the frequency of specific types of point mutations
as well as deletions and insertions of more than 4 bp.

The frequency at which a particular transversion or transition
occurred is shown in Table 3. The majority of the changes were single
base substitutions. Transversions of a G-C pair to a T-A pair
predominated, 6/7 (86%) for the untreated, 45/7l (63%) for the progeny

of BPDE-treated plasmid. Transitions occurred much less frequently.

124

Table 2. Analysis of sequence alterations generated in the supF gene

by replication BPDE-treated or untreated le89 in 293 cells.

 

No. of times occurring

 

Sequence alterations observed Untreated BPDE-treated

 

Single base substitution 3 5l

Two base substitutions

Tandem O 3
520 bases apart 2 3
>20 bases apart 0 3
Deletions
Single G°C pair 2 5
Single A°T pair l O
Tandem base pairs 0 2
4-20 base pairs 4 3
>20 base pairs 7 ll
Insertions
Single A°T pair 0 4
520 base pairs 2 l

Total Sequenced 2l 86

 

125

Table 3. Kinds of base substitutions generated in the gggE gene by
replication of BPDE-treated or untreated p2189 in 293 cells.

 

No. of mutations observed

 

 

Base change Untreated BPDE-treated
Transversions

G-C -+ T;A 6 45

G-C —+ C-G 1 l3

A°T -+ T°A 0 3

A°T -+ C-G 0 0
Transitions

G-C -» A°T 0 6

A°T -+ 8°C 0 4

Total 7 71

 

126
This was the case for mutants obtained with p2189 containing a low,
medium, or high number of BPDE induced adducts. There was no
relationship between adduct numbers and the location and types of base
substitution mutations.

Mutational Hot Spots for BPDE Induced Mutations. Fig. 3 shows
the spectrum of BPDE-induced point mutations. Two strong hot spots were
found, position l09 and 123. Both of these sites are located at the
middle base pair of GGG triplets of the tRNA gene, and the sites occur
opposite each other at the stem of the dihydrouracil loop on the
cloverleaf structure of the gng tRNA (Fig. 4). All the base
substitutions at position 109 were transversions (lO G-C -+ T°A, l G;C
-+ C°G), 8/l0 base substitutions at position l23 were G-C -4 T-A
transversions. As can be seen from Fig. 3, five less prominent hot
spots were found at positions ll2, 133, l39, l60, and l64. The
majority of these also were transversions. Their location on the
cloverleaf structure of the gng tRNA are indicated in Fig. 4. Six out
of 7 of the BPDE-induced hot spots were located at G-C pairs of the
gng gene, and one involved an A-T pair in the center of a G-C-rich
region.

The locations of the 10 point mutations obtained from untreated
plasmid were distributed over the 85 base pairs of the tRNA gene. Only
1 out of 7 base substitution mutations was located at a hot spot for

BPDE-induced substitutions (position 109).

127

Figure 3. Location of independent point mutations in the gggE tRNA gene
of le89. The DNA strand shown is the 5’-+ 3’ strand
synthesized during the DNA sequencing reaction using the
EggRI rightward-sequencing primer. The point mutations
observed in the progeny of untreated le89 are placed above
the tRNA sequence; those from BPDE—treated plasmid are placed
below the sequence. A rectangle represents a deleted base;
the caret shows the location of an inserted adenine. Every
10th residue and the anticodon triplet are underlined. The
tandem deletion of CT, shown at position 133 - l34, could

have occurred anywhere between positions l3l and l34.

128

N a
N N N N
m “N N N N N N
N N N NN N N
NNN NN NNNNNLN.
NNNNNNNNNNNNNNHNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNHNNNNN NNN N
com om. omN oNN QNN om. ov—
a N NNN N a
<zNIN New aTII
N
N N
N N N
N. < <
- N N
. N N
. N«
. N .
NN Nam NN NNN NN NN NN 33......

 

wmwfiogoofiwmuuoouoéwwwfi589%wauu<t<uo<<<§38852325589 :8

o p < h Nzowcaacoam
INNNL PENN; INZNN-2NI.._

129

Figure 4. Location of the BPDE-induced hot spots on the cloverleaf
structure of the gpj tRNA. The circles indicate the two
very strong hot spots, each of which exhibited l0 or ll base
substitution mutations out of a total of 64. The squares
indicate the 5 less-hot spots, each of which exhibited 4 or 5
base substitutions, and the hexagon indicates a site at which

3 base changes occurred.

two
base
ares
Ir 5

II'ch

 

 

 

 

C
I

G

G IIO

8

GA

130

E’Ics-c
I00 G-C
T-A
G-C
G-C
c-c
I I T G C (I: I I I C
G A A c;

C
C-G GT
TT
CC

A-T
G C

 

 

 

 

I3OA_T

C
T

 

 

C

TA

 

 

CCA3

A |80

A
A

I

I70

A

T

 

 

 

G

 

I60T=

A

T G
C

|50

A

T

A

 

 

 

 

C

131
DISCUSSION

The data in Tables 1-3 indicate that BPDE causes a decrease in the
frequency of gross rearrangements, deletions, and insertions in
plasmids rescued from the human cells and a corresponding increase in
the frequency of point mutations. It is known that transfection of
shuttle vectors into mammalian cells results in a high frequency of
spontaneous deletions and insertions (1,2,13,24-28). Hhen rodent or
monkey cell lines are used, the frequencies can exceed 1% (13). Since
high backgrounds can obscure the induction of mutations by a specific
mutagen, it is important to reduce this effect. Calos and her
associates (13) have shown that use of the 293 cell line results in a
significantly lower background of lggt mutants compared to monkey cell
lines. We also found that the frequency of spontaneous gng mutants
formed when le89 replicated in COS7 monkey cells was higher than in
293 cells (data not shown).

Since the observed gross rearrangement mutations are hypothesized
to involve nucleolytic degradation of the plasmid, ligation and,
perhaps, homologous recombination (13,27), Seidman and coworkers (4)
engineered the p2189 plasmid so that the §_ugE gene is strategically
located between two genes essential for replicating in the bacterial
host. This means that plasmids containing gross rearrangements-—e.g.,
deletions that extend from gng into either of the neighboring genes,
are not recovered in the indicator bacteria. The combined use of the
_sgLF gene in pZ189 and 293 cells for our study resulted in the low
background frequency we observed, 1.4 x 10'4 with 0.4 x 10'4 consisting

of point mutations. The fact that the progeny of BPDE-treated plasmid

132
rescued from human cells exhibited a much lower frequency of gross
rearrangements, deletions, and insertions than did the control suggests
that the presence of BPDE adducts interferes with the cellular
processes that cause gross alterations.

The data in Table 3 indicate that 90% of the BPDE-induced base
substitutions involved G-C base pairs. This result correlates with the
reported high percentage of binding of BPDE to guanine (29). The
predominant site for adduct formation by BPDE in double-stranded DNA is
the N2 position of guanine (ll), but binding to the N-7 position of
guanine can occur (30,31). The N2 guanine adduct is stable, but the
N-7 guanine adduct can result in depurination, leaving an alkali labile
site. Loeb and Preston (32) have suggested that apurinic or
apyrimidinic sites are the lesions responsible for transversions. If
so, the principal potentially mutagenic BPDE lesion should be the
unstable N-7 guanine adduct. Information bearing on the question has
recently been provided by Lobanenkov gt 1. (14) who examined the
sequence specificity of single-strand breakage induced by treating
double-stranded DNA with BPDE. They showed that scission at dG
residues is sequence specific and that cleavage and/or the generation
of alkali sensitive apurinic sites occurs preferentially at triplets in
which the middle dG has a pyrimidine on the 5’ side. In their study,
all of the TGA, TGT, TGG, CGT, and CGG sites exhibited a cleavage.
Cleavage at the middle dG when a purine was at the 5’ side was much
rarer, with 5’-AGT-3’ being cleaved, but not 5’-GGA-3’ or 5’-GGC-3’,
and cleavage at the dG of 5’-AGC-3’ required a C to the 5’ of A.
Cleavage of the middle dG of a GGG triplet was very rare. They

concluded that cleavage at sequence-specific sites results from the

133
nonrandom formation of unstable BPDE adducts.

Although they did not study the mutagenic effects of BPDE, these
investigators hypothesized that the marked effect of neighboring bases
on the frequency of G-specific DNA cleavage may be responsible for
BPDE-induced mutational hot spots. Analysis of our data on the
specific location of the base substitution mutations in the sequence of
the 5ng gene (Fig. 3) indicates that this is not the case. Neither of
our two major hot spots was a G with a pyrimidine located 5’ to it. In
fact, of all the base substitutions we observed that involved a G°C
pair, only 20% (13/64) occurred at sites predicted by the study of
Lobanenkov gt _a_t. (14) to be the location of an unstable guanine
adduct.

In view of the above arguments, our data support the hypothesis
that the N2 guanine adduct is the principal potentially mutagenic
BPDE-induced lesion. The reason for the seven hot spots shown in Fig.
4 cannot be that only those areas of the gene respond to base changes,
since the data obtained in the present study, combined with that of
previously published studies (6,7,10,12), show that a change in any one
of at least 63 of the 85 bases in the tRNA structure will result in a
phenotypic change. It should be noted that, except for position 123,
the mutagenic hot spots found in the gggE tRNA gene after BPDE
treatment differ from those found with UV treatment (6). This is not
surprising, since UV damage involves pyrimidines rather than purines.

Our finding that BPDE treatment of p2189 and its subsequent
replication in human cells results in point mutations, predominantly
base substitutions, agrees with the results of a comparative genetic

marker study by Aust gt g1. (33) and the Southern blot analysis of King

134

and Brookes (34). Our data showing that the majority of the
transversions were G-C -+ T-A support the findings of Eisenstadt gt g1.
(35) in bacteria. They used genetic crosses to analyze a large number
of nonsense mutations in the 1gg1 gene for the kinds of base
substitutions induced by BPDE in E. gglt. Our studies also agree with
the more limited data reported earlier (36-38) on the kinds of
mutations formed when BPDE-treated plasmids replicate in bacteria or in
monkey cells (28).

As shown in Table 3, 79% (48/61) of the transversions obtained
with progeny of BPDE-treated plasmids involved a change to a T°A. The
majority (45/48) were G-C -+ T-A. This preference supports the
hypothesis that the DNA polymerase involved preferentially inserts
adenine across from a noninstructional base (39), but transversions to
C-G were also found (13/61). In their study with E. ggli, Eisenstadt
gt g1. (35) who found similar results, invoked the purine pairing model
proposed by Topal and Fresco (40) to explain how such transversions
could occur. Such purine-purine mispairing may also be occurring in

mammalian cells.

135
ACKNOHLEDGEMENT

He thank Drs. Michael M. Seidman, and Hilliam C. Summers for
helpful advice. The excellent technical assistance of Bernard Schroeter
is gratefully acknowledged. This research was supported in part by
Grant CA21253 from the National Cancer Institute and by a grant from

the Ladies Auxiliary to the Veterans of Foreign Hars.

10.

11.

12.

13.

14.

15.

16.

117.

REFERENCES

Razzaque, A., Mizusawa, H. & Seidman, M. M. (1983) Ergg. Ngtl.
Acgd. Sgi. USA 80, 3010-3014.

Calos, M. P., Lebkowski, J. S. & Botchan, M. R. (1983) Eroc.
Ngtl. Acad. Sci. USA 80, 3015-3019.

Sarkar, S., Dasgupta, U. & Summers, H. C. (1984) M91, Cgll.
Biol, 4, 2227-2230.

Seidman, M. M., Dixon, K., Razzaque, A., Zagursky, R. J. &
Berman, M. L. (1985) Gene 28, 233-237.

Lebkowski, J. S., Clancy, S., Miller, J. H. & Calos, M. P.
(1985) Proc. Natl. Acad. Sci. USA 82, 8606-8610.

Hauser, J., Seidman, M. M., Sidur, K. & Dixon, K. (1986) M91.
Cell. Biol. 6, 277-285.

Glazer, P. M., Sarkar, S. N. & Summers, H. C. (1986) Proc.
Ngtl. Acad. Sci. USA 83, 1041-1044.

 

Lebkowski, J. S., Miller, J. H. & Calos, M. P. (1986) M91.
Eell. Biol. 6, 1838-1842.

Drinkwater, N. R. & Klinedinst, D. K. (1986) Erog. Ngtl. Acad.
Sci. USA 83, 3402-3406.

Bredberg, A., Kraemer, K. H. & Seidman, M. M. (1986) Eroc.
Ngtl. Acad. Sci. USA 83, 8273-8277.

Heinstein, I.B., Jeffrey, A. M., Jennette, K. H., Blobstein, S H.,
Harvey, R. G., Harris, C., Autrup, H., Oasai, H. & Nakanishi, K.
(1976) Science 193, 592-595.

Celis, J. E. & Piper, P. H. (1982) Nucleic Agids Res. 10,
R83-R9l.

Lebkowski, J. S., DuBridge, R. B., Antell, E. A., Greisen, K. S. &
Calos, M. P. (1984) Mol. Cell. Biol. 4, 1951-1960.

Lobanenkov, V. V., Plumb, M., Goodwin, G. H. & Grover, P. L.
(1986) Carcinogenesis 6, 1689-1695.

Graham, F. L., Smiley, J., Russell, H. C. & Nairn, R. (1977)_Jt
Gen, Virol. 36, 59-74.

Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molggglgr
Clonjng: A Laboratory Manual (Cold Spring Harbor Laboratory, Cold
Spring Harbor, NY), pp86-94.

Graham, F. L. & Van der Eb, A. J. (1973) Virology 52, 456-467.

136

18.
19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.
35.

36.

137
Hirt. B. (1967) J. Mol. Biol. 26, 365-369.
Silhavy, T. J., Berman, M. L., 8 Enquist, L. H. (1984)
Expgrimgnts in Gene Fusions (Cold Spring Harbor Laboratory, Cold
Spring Harbor, NY), p182.
Hanahan, D. (1983) J. Mol. Biol. 166, 557-580.

Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Erog. Natl.
Acad. Sci. USA 74, 5463-5467.

Zagurski, R. J., Baumeister, K., Lomax, N. & Berman, M. L. (1985)
fieng Anal. Tgchn. 2, 89-94.

Biggen, M. D., Gibson, T. J. & Hong, G. F. (1983) 2:09, Natl,
Acad. Sgi. USA 80, 3963-3965.

Ashman, C. H. 8 Davidson, R. L. (1984) Mol. Cgll, Biol. 4,
2266-2272.

Miller, J. H., Lebkowski, J. S., Greisen, K. S. 8 Calos, M. P.
(1984) EMBO J. 3, 3117-3121.

 

Razzaque. A., Chakrabarti, S., Joffe, S. & Seidman, M. M. (1984)
M91. Cell. Biol. 4, 435-441.

Chakrabarti, S. H., Joffe, S. & Seidman, M. M. (1985) figl, Cgll.
ngl. 5, 2265-2271.

Yang, J.-L., Maher, V. M. & McCormick, J. J. (1987) N01. Cell.
3191; 7, 1267-1270.

Osborne, M. R., Jacobs, S., Harvey, R. G. & Brooks, P. (1981)
garginggengsis 2, 553-558.

Osborne, M. & Merrifield, K. (1985) Chem.-Biol. Interact. 53,
183-185.

Osborne, M., Harvey, R. G. & Brookes, P. (1978) thm,-Bigl.
Intgract. 20, 123-130.

Loeb, L. A. & Preston, B. D. (1986) Ann. Rev. nget. 20,
201-230.

Aust, A. E., Drinkwater, N. R., Debien, K., Maher, V. M. &
McCormick, J. J. Mutat. Res. 125, 95-104.

King, H. H. S. & Brookes, P. (1984) Cargigoggnesia 5, 965-970.

Eisenstadt, E., Harren, A. J., Porter, A., Atkins, D. & Miller, J.
H. (1982) Proc. Natl. Acad. Sci. USA 79, 1945-1949.

Mizusawa, H., Lee, C.-H. & Kakefuda, T. (1981) Ngtat. Res. 82,
47-57.

37.

38.

39.

40.

138

Mizusawa, H., Lee, C.-H., Kakefuda, T., McKennedy, K., Shimatake,

H. & Rosenberg, M. (1981) Proc. Natl. Acad, Sgi, USA 78,
6817-6820.

Chakrabarti, S., Mizusawa, H. & Seidman, M. (1984) Mutat. Res.
126, 127-137.

Strauss, B. S., Rabkin, S., Sagher, S. & Moore, P. (1982)
Siochimie 64, 829-838.

Topal, M. D. & Fresco, J. R. (1976) Science 263, 285-289.

 

CHAPTER V

Kinds and Spectrum of Mutations Induced by

l-Nitrosopyrene Adducts

during Plasmid Replication in Human Cells

Jia-Ling Yang, Veronica M. Maher, and J. Justin McCormick

Carcinogenesis Laboratory - Fee Hall
Departments of Microbiology and Biochemistry

Michigan State University, East Lansing, MI 48824-1316

139

140
SUMMARY

l-Nitropyrene (l-NP) has been shown in bacterial assays to be the
principal mutagenic agent in diesel emission particulate. It has also
been shown to be mutagenic in human fibroblasts and carcinogenic in
animals. To investigate the kinds of' mutations induced by this
carcinogen, and compare them with those induced by a structurally
related carcinogen, (i)-7B,8a-dihydroxy-9a,100-epoxy-7,8,9,10-tetra
hydrobenzo[a]pyrene (BPDE) (Yang Q _1. Proc. Natl. Acad. Sci. USA,
84:3787,1987), we treated a shuttle vector with tritiated
l-nitrosopyrene (l-NOP), a carcinogenic, mutagenic intermediate
metabolite of l-NP which forms the same DNA adduct as the parent
compound, and introduced it into human cells for DNA replication to
take place. The treated plasmid, pZ189, carrying a bacterial
suppressor tRNA target gene gagE, was allowed 48 hr to replicate in the
human cells. Progeny plasmids were then rescued, purified, and
introduced into bacteria carrying an amber mutation in the
-galactosidase gene in order to detect those carrying mutations in the
aggE gene. The frequency of mutants increased in direct proportion to
the number of DNA-l-NOP adducts formed per plasmid. At the highest
level of adduct formation tested, the frequency of gggE mutants was 26
times higher than the background frequency of 1.4 x 10'4. DNA
sequencing of 60 unequivocally independent mutants derived from
l-NOP-treated plasmid indicated that 80% contained a single base
substitution, 5% had two base substitutions, and 4% had small
insertions or deletions (one or two base pairs), and only 11% showed a

deletion or insertion of four or more base pairs. Sequence data from

141

25 gLF mutants derived from untreated plasmid showed 64% contained
deletions of 4 or more base pairs. The majority (83%) of the base
substitutions in mutants from l-NOP-treated plasmid were transversions,
with 73% of these being G~C —+ T°A. This is very similar to what we
found previously in this system using BPDE, but each carcinogen
produced its own spectrum of mutations in the gupE gene. 0f the five
hot spots for base substitution mutations produced in the gng gene
with l-NOP, two were the same as seen with BPDE-treated plasmid.
However, the three other hot spots were cold spots for BPDE-treated
plasmid. Conversely, the four other hot spots seen with BPDE-treated
plasmid were cold spots for 1-NOP-treated plasmid. Comparison of the
two carcinogens for the frequency of SggE mutants induced per DNA
adduct showed that 1-NOP adducts were 3.5 times less effective than
BPDE adducts.

142
INTRODUCTION

Many carcinogens have been found to induce tumors only at
specific sites in the body of the treated animal. For example, i.p.
injection of benzo[a]pyrene, a widely distributed polycyclic aromatic
hydrocarbon formed by incomplete combustion, forms tumors predominantly
in the lung (24,39), whereas the structurally-related nitro aromatic
compound 1-nitropyrene (l-NP) primarily forms liver (39) or mamary
tumors (14). One explanation for such specificity is that the
carcinogen is metabolized into a reactive form only in specific target
organs. Another is that the carcinogen is capable of inducing only
certain kinds of’ mutations and that a specific change in DNA is
required in a particular organ to bring about the necessary alteration
leading ultimately to tumorigenesis. The latter hypothesis is
supported by recent studies on the specific mutational changes needed
to activate specific oncogenes (7,37,43). Therefore, we and others
(3,5,8,9,16-18,21,22,30,32,40,41) have begun to study the specific
kinds of mutations induced by carcinogens.

Attempts to deduce the nature of the molecular mechanisms by
which carcinogens induce mutations in mammalian cells have previously
been hampered by the inability to isolate and analyze newly mutated
genes at the sequence level. But the development of shuttle vectors,
i.e., plasmids carrying a defined target gene and capable of
replicating in mammalian cells and also in bacteria, provide a
solution to this problem (8,9,21,32,33). He are using the shuttle
vector p2189, containing the sgpfi gene which codes for a tyrosine

suppressor tRNA, to investigate at the DNA sequence level the kinds of

143

mutations induced when DNA containing covalently bound carcinogen
residues replicates in human cells. The advantage of using the aggE
gene as the target for mutation studies at the sequence level is its
small size and the fact that it is highly responsive to base changes.
Examination of data from a number of studies, including those in refs.
3,5,6,17,18,30,40,41 indicate that a change in any one of at least 63
of the 85 bases which make up the tRNA structure will result in a
mutant phenotype. Use of the human cell line 293 as the eukaryotic
host for replicating the plasmid offers the advantage of a background
mutant frequency of 1.4 x 10'4 (41) which is low enough to allow one to
observe an increase in mutant frequency induced by carcinogen treatment
of the plasmid.

Using this system, we recently determined the specific kinds of
mutations induced by the nmjor reactive metabolite of benzo[a]pyrene,
(1)43 ,8 a-dihydroxy-9a , 100 -epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene
(BPDE) (41), which has been shown to form lung tumors when i.p.
injected into newborn mice (24). In the present study we investigated
the mutagenic specificity of 1-nitrosopyrene (1-NOP), a partially
reduced metabolite of 1-NP, which forms the same DNA adducts as the
parent compound (1,29), induces mutations in bacteria (12) and
mamalian cells (13,28,29), and forms liver tumors in animals (39).
Both I-NOP and BPDE form covalently bound adducts in DNA principally
with guanine (12,19,26,38). 1-NOP binds only at the C8 position
(1,12,13,29), whereas BPDE forms its principal guanine adduct at the
N2 position (38). Since these guanine adducts are located in very
different positions in the DNA helix and only one, the N2 position, is

involved in the base pairing part of the molecule, we were particularly

144
interested in comparing the specific kinds of mutations induced by
these two agents, their location in the target gene, as well as their
biologic effectiveness, i.e., their ability to induce mutations when
the treated plasmid replicates in human cells and to interfere with
bacterial transformation. The results indicate that the adducts formed
by the two carcinogens are equivalent in decreasing bacterial
transformation and the kinds of mutations induced by 1-NOP adducts are
similar to those induced by BPDE. However, the frequency of mutants
induced per DNA adduct is 3.5 times lower for l-NOP than for BPDE, and
each compound exhibits a specific spectrum of base changes in the

target gene.

145
MATERIALS AND METIINJS

Cells and Plasmid. The human embryonic kidney cell line, 293, was
grown in modified Eagle’s minimal essential medium containing 10% fetal
calf serum (Grand Island Biological Co., Grand Island, NY) as described
previously (41). The ampicillin sensitive indicator bacterial host
was, E. _c_cfli SY204, which carries an amber mutation in the 3-
galactosidase gene (32). The 5.5 kbp shuttle vector, p2189, contains
the _s_upE tyrosine suppressor tRNA gene flanked by two genes essential
for recovery in E. 9m, i.e., the ampicillin gene and the bacterial
origin of replication (33). It also carries the origin of replication
and large T-antigen gene from simian virus-40.

Formation of l-NOP adducts on the plasmid. Plasmid DNA, prepared
using an alkaline lysis procedure (25) and purified by ethidium
bromide-CsCl density centrifugation, was resuspended in 190 pl of
helium—purged sodium citrate (10 M) buffer, pH 5.0, at a
concentration of 300 pg/ml. A 5 pl aliquot of a freshly prepared
solution of ascorbic acid (20 pM in H20) was added to provide needed
reduction of the 1-NOP (12), followed by 5 pl of a stock solution of
tritiated 1-NOP (specific activity 217 mCi/nlnole, purity 99%) dissolved
in dimethylsulfoxide at a concentration of 0.05 to 0.08 M. The
radiolabeled l-NOP was supplied by Dr. F. A. Beland of the National
Center for Toxicological Research, Jefferson, AK. The samples were
inmediately mixed and incubated at 37°C for 2 hr. Unbound 1-NOP was
removed by phenol-chloroform extraction and three successive ethanol
precipitations. The moles of l-NOP residues bound per mole of plasmid

was calculated from the A250 absorption profile of the DNA and the

146
specific activity.

Transfection and rescue of replicated plasmid. The human cell
line 293 was plated into a series of 150 mm-diameter dishes at 1 x 106
cells per dish. After 24 hr the cells in each dish were transfected
with 6 pg of plasmid using a modification of the CaPO4 coprecipitation
technique as described (41). After 48 hr, the cells from each dish
were harvested separately and progeny plasmid was extracted and
separated from cellular DNA as described (15,41). The DNA was purified
with buffered phenol, treated with RNase A (50 pg/ml) at 37°C for 1 hr,
followed by proteinase K (100 pg/ml) at 50°C for 2 hr, then extracted
with phenol-chloroform, precipitated with ethanol, and further purified
by drop dialysis (34). In order to distinguish between independent
mutants with identical mutations and putative siblings derived from a
single event, the progeny plasmid obtained from each dish of cells was
maintained and assayed separately. Before progeny plasmid was used to
transform indicator bacteria, it was treated with 20 units of Opal, a
restriction enzyme which digests any input DNA that still has the
bacterial methylation pattern generated when the plasmid was first
prepared in bacteria. As a control, an aliquot was also treated with
Opal, which digests plasmid generated during replication in human
cells, and tested for any residual transforming ability. The results
showed that >99.9% of the purified plasmid was derived from material
that had replicated in the human cells.

Bacterial transformation and mutant identification. The progeny
plasmid DNA was assayed for mutant gapE genes by transforming E. gall
SY204 to ampicillin resistance and selecting the transformants on

plates containing ampicillin (50 mg/ml), X-Gal, and an inducer of the

147

B-galactosidase gene, as described (41). Transformants containing
plasmids lacking a functioning gapE gene, which is needed to suppress
the amber mutation in the B-galactosidase gene of the bacteria, could
be identified because they form white or light blue colonies rather
than dark blue colonies on X-Gal plates. Each white or light blue
putative mutant colony was restreaked on fresh plates to confirm the
phenotype.

Characterization of mutants. Plasmid DNA from bacteria in these
white or light blue colonies was amplified, extracted, dissolved in
buffer, and analyzed by electrophoresis on 0.8% agarose gels for
altered DNA mobility (gross alteration). Plasmids without evidence of
gross alterations were amplified further and purified by CsCl
centrifugation and analyzed by a secondary bacterial transformation to
ensure that the observed inability of the bacteria to utilize X-Gal
was the result of inactivation of the §_ng gene, rather than a
mutation in the B-galactosidase gene of E. w (41). Putative sppE
mutants were sequenced using the dideoxyribonucleotide method (31)
modified as follows. Plasmids were denatured with alkali to generate
single-stranded templates (42) and polymerization from a p8R322 E_c_pRl
site primer was carried out using the Klenow fragment of DNA polymerase
I. The primer and polymerase were purchased from New England Biolabs
(Beverly, MA). 355- adATP (New England Nuclear, NEN-O34S, Boston, MA)
and buffer gradient denatured polyacrylamide gels (2) were used for
greater resolution of the sequencing gel.

Determination of sites of carcinogen-induced adducts. The
positions of l-NOP- or BPDE-adducts in the gapE gene of carcinogen-
treated plasmid were determined by the ta Map DNA polymerase-stop

148
assay of Moore and Strauss (27). Briefly, double-stranded plasmid
containing l-NOP- or BPDE adducts (10 to 70 adducts per molecule of
plasmid) was denatured and annealed with the pBR322 EngI site primer.
The length of the DNA from the primer site to the end of the gapE gene
is ~230 nucleotides, so that the average number of adducts per strand
of the gapE gene was 0.2 to 1.5. The polymerization reaction were then
carried out as for the sequencing reaction except that the
dideoxynucleotides were omitted. DNA from the four dideoxy sequencing
reactions, carried out on an untreated template, was electrophoresed on
the same gel to serve as DNA size markers. The relative intensities of
the bands on the autoradiography were determined by a laser
densitometer (LKB 2222-010, Ultroscan XL) and were corrected for
position in the gene by taking into account the number of 35S-labeled
adenine bases that would be present in each length of newly synthesized

DNA.

149
RESULTS

Characterization of l-NOP-treated plasmid. Vector p2189 was
treated with various concentrations of tritium labeled l-NOP in the
presence of ascorbic acid as the reducing agent. 'The number of
residues bound per plasmid was determined and their ability to
interfere with transformation of bacterial cells to ampicillin
resistance was determined using the transformation method of Hanahan
(11). As shown in Figure 1, the number of l-NOP-induced adducts per
molecule of plasmid increased linearly with the concentration of l-NOP
used. Figure 2 shows that the transforming activity' of' modified
plasmid decreased in direct proportion to the number of l-NOP residues
bound. Approximately seven l-NOP residues bound were required to lower
the transforming activity of the treated plasmid to 37% of the
untreated control plasmid.

Frequency of mutants induced by l-NOP adducts. Plasmid containing
various levels of l-NOP adducts and untreated plasmid were introduced
into human cells by transfection and allowed to replicate for 48 hr.
The progeny plasmid were harvested and assayed for mutations in the
gapE gene by introducing them into E. gpli SY204 indicator bacteria.
As shown in Figure 3, there was a linear increase in the frequency of
gapE mutants as a function of the number of l-NOP-induced adducts per
plasmid. At the highest level of l-NOP adduct formation tested, i.e.,
63 adducts per pZ189, the frequency of gapE mutants was 35.8 x 10'4,
which is 26 times higher than that the background frequency of 1.4 x
10-4.

Agarose gel and DNA sequence analysis of mutants obtained with 1-

150

Figure 1. Number of 1-NOP adducts bound per plasmid as a function of
concentration of l-NOP in the presence of ascorbic acid (0.5
uM). The symbols (CandA) indicate that the plasmid was

treated with l-NOP in two separate experiments.

I-NITROSOPYRENE RESIDUES
BOUND PER PLASMID

70
so
50
4o
30
20
I o

o

151

 

l
l

IIII[IIFIrIIII[IIII]]

 

—. —

 

 

lllIllllIllllIllllIl

O 5 IO I5 20

NITROSOPYRENE (OM)

152

Figure 2. Relative frequency of transformation of bacteria to
ampicillin resistance by plasmid p2189 containing l-NOP
adducts compared to untreated plasmid. The error bars

indicate the SEM of four determinations.

153

 

O
[\J
T I
<1
.1

Ill I r II
I—-.—I
/
H—I
I-—-.-i
I l I

<3.
/

[[Tllr.4

I I I IIIIII

1 1TIIIIII

 

“I

5.

lllllllrl

A
A
\

RELATIVE TRANSFORMATION
EFFICIENCY (%)
5
O
I TTITI I
I—N
I)!
4

I IJIIIIIL I JIIIIIII

 

 

 

I (5.2! l_ I I. I I I I I I I IL I
O 20 4O 60

l-NITROSOPYRENE RESIDUES
BOUND PER PLASMID

154

Figure 3. Frequency of gpLF mutants as a function of the number of
l-NOP adducts per plasmid. Untreated plasmids or plasmids
containing 1-NOP adducts were transfected into 293 cells and
allowed 48 hr for replication. The progeny plasmids were
rescued, treated with Owl, and introduced into indicator
bacteria by DNA transformation. The frequency of
transformants with a nonfunctioning gene was identified by
colony color and secondary transformation (see text). The
error bars refer to SEM of the gapE; mutant frequencies
obtained from a series of individual human cell transfection

experiments made with each set of treated plasmid.

FREQUENCY OF
SUP F MUTANTS (x I04)

155

 

4O I|I|I|IrI|I|

/

DYIIIIIIILIIII
O 20 4O 60

l-NITROSOPYRENE RESIDUES
BOUND PER PLASMID

 

I;
20:" i T.
l o:— 1/ -

 

156

NOP-treated plasmid revealed that the majority of them contained point
mutations. The data are summarized in Table 1. Included in the table
are data from 14 experiments in which human cells were transfected with
untreated plasmid to determine the background frequency of gapE mutants
and to generate control mutants for analysis. Eight of these control
transfection experiments with untreated plasmid accompanied studies
with BPDE-treated plasmid, six accompanied studies with l-NOP-treated
plasmid. The analysis of all 31 of the mutants obtained was reported
in our earlier paper showing the mutagenic effect of BPDE (41), but the
data have been included here in Table 1 for comparative purposes.

The mutants derived from plasmids containing the two highest
levels of adducts contined mainly point mutations. Only 4.5% (3/67)
exhibited altered gel mobility, and only 10% of those mutants that were
sequenced (5/50) proved to contain deletions or insertions of four or
more base pairs. In contrast, 70% of the control mutants (21/30)
contained such mutations. Calculation of the fraction of mutants
containing point mutations (Table 1, column 10) showed that the
frequencies derived from plasmid carrying the two highest levels of 1-
NOP adducts were respectively 35 and 75 times higher than the control.

Nature and location of the specific mutations induced by l-
MOP-treated plasmid in 293 cells. He examined the nature of the
point mutations induced in plasmids carrying l-NOP adducts, as well as
the location of the point mutations in the target gene. As shown in
Table 2, sequence analysis of 60 unequivocally independent mutants
derived from l-NOP-treated plasmid indicated that 88.3% contained point
mutations, i.e., 80% (48/60) contained a single base substitution, 5%
(3/60) contained two base substitutions and 3.3% (2/60) had deletion

157

.aa N No — No coNacmNNN so .co.uopmu .:o_a=u.amn:mu

.NN NN - N. No NNNNLNNNNN
.NN NNN - N No NNNNNNNNN

.mm Nazpou .couacwe:=v ANNNNaos pom magma—a achcgN Nacauas omega Napa m czapou :. «as» an wouN>Nu a ema—cu c. Lassa:
ace N. Nucmuae N:.oa No NNNNuNNN ogp .mN csapouv Nucosamcm um>NoNao use Noe.» Nucaazm NNNNN No capuuasu sou» ammo—:o—auu
.maa ompAv _ma oNoNNaN co NNNNNN> coNuNNauN<a

.maxmu mmmm ocoo mama agu No :oNua>NuuN=N seem wag—smug

_Nu-x NNNNNNNNNE cu NNNNNnNNN N.N—mu mgu maze weamca ca =NNNNsNoNN=~NN Ngavcouom a an waxmmma Na: menus: guaw sec» uNeNapaa

 

N.NN NN N N NN _N\N N.NN NNN.NN\NN N N.NN
N.N_ N. N N NN NN\_ N.NN NNN.N_\NN N N.NN
N.N N N N NN NN\N N.N NNN.NN\N_ N N.NN
N.N N N N N N\N N.N NNN.NN\N N N.N
N.N N N NN NN NN\N N.N NNN.NNN\NN N. N

 

No— x NNNoNNNN:e «Nee—NNNNNN umchampwu vmocosawm uNNNENxm .o: Aeop xv Nacaegoumcagu NN=¢ENNmaxo vNeNaNa

oco.uaase ucNoa gNNx .02 ca.) .6: chmm mama \nNNNN—noe Nucaaam \amucause coNNuoumcagu Nan
NNNNN mama guNz .cz vNeNa—a Nam vogmupa mama No mama N—ou essag Nauavu<
No Nucmzamga Nacauzs paaoN cuNz NcNeNupa mucosaoga No .oz

umucmauom No zoNNNNNquuasagu

 

.NN—uu mow c. NNNNNuNNNNL
chcac umumcmcwm NNNNN vaNmNN-Noz-_ No acuuoga Nu.) mmam am No :oNNNELoumcasu Na uocNauaa magmas: mo N.Nmpa:< .— opnap

158

Table 2. Analysis of sequence alterations generated in the supfi gene

by replication l-NOP-treated or untreated p2189 in 293 cells.

 

No. of times occurring

 

Sequence alterations Untreated l-NOP-treated

 

Single base substitution 3 48

Two base substitutions

Tandem O 2
520 bases apart 2 0
>20 bases apart 0 l
Deletions
Single G'C pair 2 1
Single A°T pair 1 0
Tandem base pairs 0 0
4-20 base pairs 4 3
>20 base pairs 7 2
Insertions
Single A°T pair 0 l
520 base pairs 2 2

Total sequenced 21 60

 

159
or insertion of one base pair. These data contrast with the control
mutants where only 38% showed such point mutations. Table 3 shows the
specific kinds of base substitutions induced in the ,supf, gene by
replication of l-NOP-treated plasmid in 293 cells. The majority
(45/54) of the changes were transversions, with 33 out of 45 being G-C
-+ T°A. The majority (87%) of base changes (47/54) involved G-C pairs.

The specific location of l-NOP-induced point mutations (spectrum)
is shown in Figure 4. Three strong hot spots were found at positions
109, 123 and 127. All of the base substitutions at these three hot
spots were G-C pair transversions (l6 G-C -+ T°A, 4 G-C -+ C-G). Two
less prominent hot spots were seen at positions 156 and 159 with
transitions occurring twice as frequently as transversions (6 G-C-+
A°T, 3 G'C -+ T°A). All five hot spots were located at the stem of the
cloverleaf structure of the supfi tRNA (Figure 5). All seven of the A-T
pair base substitutions were located in the loops of the tRNA
cloverleaf structure.

Sites of l-NOP or BPDE-adducts. To determine whether there was a
correlation between the positions and frequencies of the carcinogen-
induced mutations and the sites and frequencies of carcinogen-DNA
adducts in the supfi gene, we carried out the DNA synthesis-stop assay
of Moore and Strauss (27), in which bulky adducts interfere with DNA
replication. The template DNA used in the assay contained 10 to 70
l-NOP or BPDE adducts per plasmid, which represents 0.2 to 1.5 adducts
per strand of the M gene, if binding to guanine is essentially
random. This protocol for estimating the percentage of carcinogen
adducts formed at particular sites on the gene assumes first, that the

density of the bands in a sequencing gel, adjusted for extent of

160

Table 3. Kinds of base substitutions generated by replication of

l-NOP-treated or untreated p2189 in 293 cells.

 

No. of mutations observed

 

 

Base change Untreated l-NOP-treated
Transversions

G-C -» T-A 6 33

G-C —+ C-G l 8

A-T - T-A 0 3

A°T - C-G 0 l
Transitions

G-C -4 A-T 0 6

A°T -+ 6°C 0 3

Total 7 54

 

161

Figure 4. Location of independent point mutations in the sgpfi tRNA gene
of p2189. The DNA strand shown is the 85 nucleotides making
up the tRNA structure. The point mutations observed in the
progeny of l-NOP treated plasmid are placed below the
sequence. The rectangle represents a deleted guanine; the
caret shows the location of an inserted thymidine. Every

tenth residue and the anticodon triplet are underlined.

m2

4
_4 4
.444

4
4

4
4
4 FL.

i—444

l—444

h
h

4

o
o o
004

U
4444400

h

<

PPHPHUU

44444

0 .r 04. nu

483988858240839:88»20.855838433meN04o84mg880480850830

Om.

n:4m

D

ON.

4

.u4

00.

b

On.

0v. .

G

On.

0

H

ON.

0: 09.
4 h

allll42m: ||1_
2.8m

82.27%: -_

n
mzochNcoqm

163

Figure 5. Location of the l-NOP-induced hot spots on the cloverleaf
structure of the supF tRNA. The circles indicate the three

strong hot spots. The squares indicate the 2 less prominent

hot spots.

164

 

 

 

165
355-cszTP incorporated, is proportional to the number of DNA molecules
of a particular length, and second, that the length reflects the chance
of adduct-inddced premature termination of the polymerization.

The gel pattern of the DNA bands obtained in this assay for DNA
containing 1-NOP or BPDE residues corresponded to positions one
nucleotide before virtually every cytosine residue in the DNA
sequencing standard lane, indicating that DNA synthesis was terminated
one band away from each guanine in the template. No evidence of any
interference with polymerization was obtained with untreated template
and bands corresponding to positions one nucleotide away from any base
other than guanine were not seen. The pattern of bands did not vary
significantly with the numbers of l-NOP- or BPDE-adducts per molecule
of plasmid. Presumably, these bands were generated by a stop of the
Klenow fragment of DNA polymerase I at the sites of bulky adducts. The
relative intensities of the bands in the 85 nucleotides making up the
tRNA of the 3’ to 5’ template using the p8R322 figgRI site primer are
shown in Figure 6. The frequency of DNA adducts, as estimated with
this assay, ranged from 0% (position 131) to 6.6% (position 143) for 1-
NOP-adducts and 0% (position 131 and 152) to 12.9% (position 110) for
BPDE-adducts. As shown in Figure 6, the l-NOP-guanine adducts were
most frequently located at position 143, a position at which no
mutation was observed. The frequency of l-NOP-guanine adducts located
at positions 109, 114, 118, 142, 149, 163, and 169 was approximately
the same ("3%), but only at 109 did mutation occur and, in fact,
position 109 represents a very strong mutational hot spot for l-NOP.
This apparent lack. of correlation between the frequency' of' adduct

formation and mutation frequency was also found with BPDE.

166

Figure 6. Relative frequency of l-NDP-guanine or BPDE-guanine adducts
in the M gene of pZ189 and location of l-NOP- or BPDE-
induced base substitutions within the tRNA coding sequence of
p2189. Klenow fragment of DNA polymerase I was used with
EQRI rightward sequencing primer to determine polymerase-
stop sites on the 3’ to 5’ strand of l-NOP— or BPDE-treated
supfi gene (panel shown above the supfi gene sequence). The
relative intensities of the bands on the autoradiography were
determined by densitometer and were corrected for position in
the gene by taking into account the number of 358—labeled
adenine bases that would be present in each length of newly
synthesized DNA. The base substitutions shown below the sgpfi
tRNA gene were G-C base changes on the 5’ to 3’ SELF gene

using the figQRI rightward site primer.

167

i325-

<
< < <
<<< <

4Q
4469

4

44444

4

4444466

(54444444444

p...

444444

c

umuzn:_-mc¢m

NNNNNN_-NN2-_

.m <9ufiuu<uuuuuhHuuh<<o9hhdo<<uohhu<m9h<uhuuucHu~<<<huhumu<uu<ucu<m<uuuuuu<uw90hhwuochmu em

_
NNN

N

_
NNN

om" o¢~

_
NNN

N

.m huuhuohucouu<dou<hhuc<<uuphou<<upua<hw<uucumu<hhh<o<chuhcupuuuphhcouuouhuouu<<uuuu<uu .m

 

 

__:____ __ _

___+_

 

 

_. _ N

L moz-~

 

 

::

 

:: _1 _

_

_—

_

N

: __.

. mama

 

168
DISCUSSIGI

The data in Fig. 1 indicate that the number of l-NOP adducts per
molecule of plasmid increased linearly with the concentration of l-NDP
used in the presence of ascorbic acid. The amount bound per applied
concentration of carcinogen was three times higher than what we found
previously with BPDE (41), but unlike the direct-acting BPDE, l-NOP
requires reducing agents such as ascorbic acid for activation (12).
Once the l-NOP adducts formed, they were very similar to BPDE adducts
in their ability to interfere with the processes involved in bacterial
transformation, i.e., approximately seven l-NOP adducts per plasmid
(Figure 2) and nine BPDE adducts per plasmid (41) were sufficient to
lower the transforming activity of the treated plasmid to 37% of that
of untreated plasmid.

The data in Table 3, indicating that l-NOP adducts caused base
substitutions mainly at 6°C base pairs (87%) just like BPDE does (41),
strongly suggust that the mutagenesis was targeted to sites where
adducts occur. Targeted mutagenesis is also indicated for l-NOP-
induced base substitutions involving A°T base pairs. Our results
(Table 3) show that 13% of the base changes involved A°T base pairs,
and Kinouchi and 0hnishi (19) recently demonstrated with bacterial
nitroreductases that a minor adduct formed by l-NP involved
deoxyadenosine. We consider the base substitutions induced by these
two carcinogens to be targeted to adducts rather than to apurinic
sites, as suggested by Loeb and Preston (23). The argument for this
conclusion for BPDE was given in our earlier paper (41). In the case

of l-NOP, apurinic sites are highly unlikely since this carcinogen

169
does not form unstable adducts.

The majority of the l-NOP or BPDE-induced base changes we observed
were G°C -+ T°A transversions. It may be that, as suggested by Strauss
gt a1. (35), the DNA polymerase in the human 293 cell line
preferentially inserts an adenine nucleotide opposite a nonstructional
base containing a bulky adduct ("A rule"). However, 0°C -+ C-G
transversions were also found (15% of the base substitutions). Another
explanation for the predominance of 6°C -+ T-A transversions, and the
one which we favor, is that with some frequency, purine-purine base
pairing occurs when the guanine carries a l-NOP residue (or a BPDE
residue). Eisenstadt gt al. (10) invoked the model of Topal and Fresco
(36), in which such mispairing occurs by having the guanine carrying a
BPDE residue at the N2 position be oriented in the m position and
pair' with an incoming adenine in the rare tautomeric imino form.
However, Kennard and her colleagues (4) recently reported that stable
Ganti°Asyn base pairs are formed in a synthetic deoxydodecamer and are
accommodated in the DNA double helix with little or no disruption of
the local or global conformation. Their model does not require the
formation of the very rare enol or imino tautomeric forms. l-NOP binds
predominantly to the 08 position of guanine which is not involved in
base pairing. The presence of this bulky residue may direct guanine
in the normal _a_n_t_i configuration to pair with an incoming adenine
nucleotide triphosphate with adenine adopting the syn configuration.
This would result in a stable base pair.

Using the polymerase-stop assay (Figure 6), we did not find a high
correlation between the frequency of l-NOP or BPDE adduct sites and the

frequency of base substitution mutations at these sites. Although the

170

"A rule" or/and the purine-purine mispairing mechanism may explain G-C
-+ T°A transversions, they cannot easily explain the location of the
mutational hot spots we observed with either carcinogen. This lack of
correlation is not caused by the inability of changes in these sites to
cause detectable phenotypic changes in the tRNA gene since base
substitutions at most of these positions have been detected previously.
Fuchs and his coworkers proposed the ”mutation-prone sequences" model
to explain the lack of correlation between the sites of
acetylaminofluorine-induced frameshift mutations in a bacterial
plasmid and the sites at, which T4 DNA polymerase exonuclease III
activity was blocked by acetylaminofluorine residues. Similarly, Brash
_t _l. (3) proposed the "pass/fail" model to explain the lack of
correlation of mutation frequency with UV-induced photoproduct
frequency at different sites in the §_up£ gene. Such ”mutation-prone
sequences"- or ”pass site"-determined mutational hot spots, rather than
mere DNA lesion—determined hot spots, also fit our findings, but do not
explain why some sites are very hot for mutation occurrence.

Although both l-NDP and BPDE induced point mutations, with single
base substitutions predominating and with G-C -+ T°A transversions
being the most common mutation, the spectrum of mutations in the supfi
tRNA gene induced by l-NOP differed from that induced by BPDE. The
spectrum of the two carcinogens exhibited two hot spots in common
(positions 109 and 123). However, three l-NOP-induced hot spots were
cold spots for BPDE (positions 127, 156, and 159), and four BPDE-
induced hot spots were cold spots for l-NOP (positions 112, 139, 160,
and 164). These differences are not likely to be ‘the result of

differential binding of the two carcinogens at those positions, at

171

least not as determined by the DNA polymerase-stop assay. For example,
adducts were formed at position 139 more frequently by l-NOP than by
BPDE, but this site was a mutational hot spot for BPDE not for l-NOP.
These differences may, as suggested by Brash gt gl. (3), reflect the
fact that structural features differ between hot spots and non-hot
spots in DNA sequences and that the ability of carcinogen-modified
nucleotides to alter DNA structure at specific sites differs for l-NOP
and BPDE.

This explanation is consistent with our finding that when the
mutagenicity of these two structurally-related carcinogens is compared
on the basis of equal numbers of adducts per plasmid, l-NOP is 3.5
times less effective than BPDE in inducing mutations during the
replication of the modified plasmid in the human host cells. This
difference in mutagenic effectiveness of these two carcinogens may
reflect the intrinsic difference in the nature of the adducts formed.
However, it may also reflect a difference in the rate of excision l-
NOP- or BPDE-adducts from the plasmid by the host 293 cells.
Unpublished data from this laboratory indicate that diploid human
fibroblasts can excise l-NOP-induced adducts faster than BPDE-induced
adducts (V. M. Maher, J. D. Patton and J. J. McCormick, manuscript in
preparation). Therefore, the ability of 293 cells to remove 1-NOP- or
BPDE-adducts from DNA is currently under investigation.

It may be important to point out that, unlike Seidman, Kraemer and
their collaborators who studied UV-induced _stILF mutations in p2189
replicating in SV40-transformed human cells (3,5), or Hauser e_t_ fl.
who studied this in a monkey cell line (18), we did not observe

multiple mutations occurring in a single mutant plasmid. The reason

172
for this significant difference in results between UV induced lesions
and adducts formed by bulky carcinogens is also currently under

investigation in our laboratory.

l 73
ACMLEDGDIENTS

We thank Dr. Frederick A. Beland of the National Center for
Toxicological Research for providing us with radiolabeled l-NOP and for
his helpful advice with the ascobic acid reaction procedures. The
excellent technical assistance of Bernard Schroeter is gratefully
acknowledged. This research was supported in part by Grant CA21253
from the National Cancer Institute and by Contract #87-2 from the
Health Effect Institute (HEI), an organization jointly fUnded by the
U.S. Environmental Protection Agency (EPA) (Assistance Agreement
x812059) and automotive manufacturers. It is currently under review by
the Institute. The contents of this article do not necessarily
reflect the views of the HEI, nor do they necessarily reflect the

policies of EPA, or automotive manufacturers.

10.

11.

REFERENCES

Beland, F. A., M. Ribovich, P. C. Howard, R. H. Heflich, P.
Kurian, and G. E. Milo. 1986. Cytotoxicity, cellular
transformation and DNA adducts in normal human diploid
fibroblasts exposed to l-nitrosopyrene, a reduced derivative of

the environmental contaminant, l-nitropyrene. Carcinogenesis
7:1279-1283.

Biggen, M. D., T. Gibson, and G. F. Hong. 1983. Buffer
gradient gels and S label as aid to rapid DNA sequence
determination. Proc. Natl. Acad. Sci. USA 80:3963-3965.

Brash, D. E., S. Seetharam, K. H. Kraemer, M. M. Seidman, and
A. Bredberg. 1987. Photoproduct frequency is not the major
determinant of UV base substitution hot spots or cold spots in
human cells. Proc. Natl. Acad. Sci. USA 84:3782-3786.

Brown, T., H. N. Hunter, G. Kneale, and 0. Kennard. 1986.
Molecular structure of the G-A base pair in DNA and its

implications for the mechanism of transversion mutations. Proc.
Natl. Acad. Sci. USA 83:2402-2406.

Bredberg, A., K. H. Kraemer, and M. M. Seidman. 1986.
Restricted ultraviolet mutationsl spectrum in a shuttle vector

propagated in xeroderma pigmentosum cells. Proc. Natl. Acad.
Sci. USA 83:8273-8277.

Celis J. E., and P. H. Piper. 1982. Compilation of mutant
suppressor tRNA sequences. Nucleic Acids Res. 10: R83-R91.

Dandekar, S., S. Sukumar, H. Zarbl, L. J. T. Young, and R. D.
Cardiff. 1986. Specific activation of the cellular Harvey::g§
oncogene in dimethylbenzanthracene-induced mouse mammary
tumors. Mol. Cell. Biol. 6:4104-4108.

Drinkwater, N. R., and D. K. Klinedinst. 1986. Chemically
induced mutagenesis in a shuttle vector with a low-background
mutant frequency. Proc. Natl. Acad. Sci. USA 83:3402-3406.

DuBridge, R. B., P. Tang, H. C. Hsia, P.-M. Leong, J. H.
Miller, and M. P. Calos. 1987. Analysis of mutation in human

cells by using an Epstein-Barr virus shuttle system. Mol. Cell.
Biol. 7:379-387.

Eisenstadt, E., A. J. Harren, J. Porter, D. Atkins, and J. H.
Miller. 1982. Carcinogenic epoxides of benzo[a]pyrene and
cyclopenta[cd]pyrene induce base substitutions via specific
transversions. Proc. Natl. Acad. Sci. USA 79:1945-1949.

Hanahan, D. 1983. Studies on transformation of Escherichia coli
with plasmids. J. Mol. Biol. 166:557-580.

174

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

175

Heflich, R. H., P. C. Howard, and F. A. Beland. 1985.
l-Nitrosopyrene: An intermediate in the metabolic activation of
l-nitropyrene to a mutagen in Salmonella typhimurium TA1538.
Mutat. Res. 149:25-32.

Heflich, R. H., N. F. Fullerton, and F. A. Beland. 1986. An
examination of the weak mutagenic respose of l-nitropyrene in
Chinese hamster ovary cells. Mutat. Res. 161:99-108.

Hirose, M., M.-S. Lee. C. Y. Hang, and C. M. King. 1984.
Induction of rat mammary gland tumors by l-nitropyrene, a

recently recognized environmental mutagen. Cancer Res.
44:1158-1162.

Hirt, B. 1967. Selective extraction of polyma DNA from infected
mouse cell cultures. J. Mol. Biol. 26:365-369.

Gentil, A., A. Margot, and A. Sarasin. 1986. 2-(N-acetoxy-N-
acetylamino)fluorene mutagenesis in mammalian cells:
Sequence-specific hot spot. Proc. Natl. Acad. Sci. USA
83:9556-9560.

Glazer, P. M., S. N. Sarkar, and H. C. Summers. 1986. Detection
and analysis of UV-induced mutations in mamalian cell DNA
using a lambda phage shuttle vector. Proc. Natl. Acad. Sci. USA
83:1041-1044.

Hauser, J., M. M. Seidman, K. Sidur, and K. Dixon. 1986.
Sequence specificity of point mutations induced during passage
of a UV-irradiated shuttle vector plasmid in monkey cells. Mol.
Cell. Biol. 6:277-285.

Kinouchi, T., and Y. 0hnishi. 1986. Metabolic activation of
l-nitropyrene and 1,6-dinitropyrene by nitroreductases from
Bacteroides fragilis and distribution of nitroreductase
activity in rats. Microbiol. Immunol. 30:979-992.

Koffel-Schwartz, N., J. M. Verdier, M. Bichara, A. M. Freund,
M. P. Daune, and R. P. P. Fuchs. 1984. Carcinogen-induced
mutation spectrum in wild-type, gytA, and gmgc strain of
Escherichia coli. J. Mol. Biol. 177:33-51.

Lebkowski, J. S., S. Clancy, J. H. Miller, and M. P. Calos.
1985. The lacl shuttle: Rapid analysis of the mutagenic
specificity of ultraviolet light in human cells. Proc. Natl.
Acad. Sci. USA 82:8606-8610.

Lebkowski, J. S., J. H. Miller, and M. P. Calos. 1986.
Determination of DNA sequence changes induced by ethyl

methanesulfonate in human cells, using a shuttle vector system.
Mol. Cell. Biol. 6:1838-1842.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

176

Levin, H., A. H. Hood, P. G. Hislocki, R. L. Chang, J.
Kapitulnik, H. D. Mah, H. Yagi, 0. M. Jerina, and A. H. Conney.
1978. Mutagenicity and carcinogenicity of benzo[a]pyrene and
benzo[a]pyrene derivatives. in: Bg1ygyg11g__flygrggg;tgg§__gng
tgnggn, Vol I. Gelboin, H. V. and Ts’0, P. 0. (eds) Academic
Press, pp189-202.

Loeb, L. A., And 8. 0. Preston. 1986. Mutagenesis by
apurinic/apyrimidinic sites, Ann. Rev. Genet. 20:201-230.

Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. in:

Mglgcglar cloning: A 1W. Cold Spring Harbor
Laboratory, Cold Spring Harbor, H.V. pp. 86-94.

Meehan, T., K. Straub, and H. Calvin. 1977. Benzo[a]pyrene diol
epoxide covalently binds to deoxyguanosine and deoxyadenosine
in DNA. Nature (London) 269:725-727.

Moore, P., and B. S. Strauss. 1979. Sites of inhibition of in
vitrg DNA synthesis in carcinogen- and UV-treated Xm74 DNA.
Nature (London) 278:664-666.

Patton, J. D., V. M. Maher, and J. J. McCormick. 1986.
Cytotoxic and mutagenic effects of l-nitropyrene and

l-nitrosopyrene in diploid human fibroblasts. Carcinogenesis
7:89-93.

Patton J. D., V. M. Maher, and J. J. McCormick. 1987.
Cytotoxicity, mutagenicity, and transformation of diploid human
fibroblasts by 1-nitropyrene and 1-nitrosopyrene. in:
Polvnuclegr Aromgtic Hydrocgrbons: A deggde of progress.
Cooke, M., and Dennis, A. J. (eds.). Battelle Press, Columbus,
pp.678-698.

Protic-Sabljic, M., N. Tuteja, P. J. Munson, J. Hauser, K. H.
Kraemer, and K. Dixon. 1986. UV light-induced cyclobutane
pyrimidine dimers are mutagenic in mammalian cells. Mol. Cell.
Biol. 6:3349-3356.

Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing
with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA
74:5463-5467.

Sarkar, S., U. B. Dasgupta, and H. C. Summers. 1984.
Error-prone mutagenesis detected in mammalian cells by a
shuttle vector containing the gggfi gene of Esgherighia coli.
Mol. Cell. Biol. 4:2227-2230.

Seidman, M. M., K. Dixon, A. Razzaque, R. J. Zagursky, and M.
L. Berman. 1985. A shuttle vector plasmid for studying
carcinogen-induced point mutations in mammalian cells. Gene
38:233-237.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

177

Silhavy, T. J., M. L. Berman., and L. H. Enquist. 1984.
Experiments in gene fusions. Cold Spring Harbor Laboratory,
Cold Spring Harbor, H.V.

Strauss, B., S. Rabkin, D. Sagher, and P. Moore. 1982. The role
of DNA polymerase in base substitution mutagenesis on
noninstructional templates. Biochimie 64:829-838.

Topal, M. D., and J. R. Fresco. 1976u Complementary base
pairing and the origin of substitution mutations. Nature
(London) 263:285-289.

Vousden, K. H., J. L. Bob, C. J. Marshall, and D. H. Phillips.
1986. Mutations activating human c-Ha-ml protooncogene
(HRASI) induced by chemical carcinogens and depurination. Proc.
Natl. Acad. Sci. USA 83:1222-1226.

Heinstein, I. B., A. M. Jeffrey, K. H. Jennette, S. H.
Blobstein, R. G. Harvey, C. Harris, H. Autrup, H. Kasai, and K.
Nakanishi.1976.Benzo[a]pyrene diol epoxides as intermediates

in nucleic acid binding in vitrg and in vjvg. Science
193:592-595.

Hislocki, P. G., E. S. Bagan, A. Y. H. Lu. K. L. Dooley, P. P.
Fu, H. Han-Hsu, F. A. Beland, and F. F. Kadlubar. 1986.
Tumorigenicity of nitrated derivatives of pyrene,
benzo[a]anthracene, chrysene and benzo[a]pyrene in newborn
mouse assay. Carcinogenesis 7:1317-1322.

Yang, J.-L., V. M. Maher, and J. J. McCormick. 1987a. Kinds of
mutations formed when a shuttle vector containing adducts of
benzo[a]pyrene-7,8-diol-9,10-epoxide replicates in COS7 cells.
Mol. Cell. Biol. 7:1267-1270.

Yang, J.-L., V. M. Maher, and J. J. McCormick. 1987b. Kinds of
mutations formed when a shuttle vector containing adducts
of(_+_)-7B ,80 -dihydroxy-9a ,100 -epoxy-7,8,9,10-tetrahydro
benzo[a]pyrene replicates in human cells. Proc. Natl. Acad.
Sci. USA 84:3787-3791.

Zagursky, R. J., K. Baumeister, N. Lomax, and M. L. Berman.
1985. Rapid and easy sequencing of large linear double-stranded
DNA and supercoiled plasmid DNA. Gene Anal. Technol. 2:89-94.

Zarbl, H., S. Sukumar, A. V. Arthur, D. Martin-Zanca, and M.
Barbacid. 1985. Direct mutagenesis of Ha-Lag-l oncogenes by

N-nitroso-N-methylurea during initiation of mammary

carcinogenesis in rats. Nature (London) 315:382-385.