,..§§,..3¥<z .3. .2, V . .
. .. . .3111 1.2..
....,,....y.....a¢finm.~ $3.16, “5: .

.. :y 3.
‘6... 3... . . ..

. ~ I \.
2.; , Lawfitquaan i..." $1
1‘ 3s .2 2 . v .
.... 4 \ 5&3,qu .1 u

1;» 13 1 .

II»... «I It: I. c.
v..\ .rv :0 v) .61 .4
I!) 55.1.41 .3 A f

a? 3 s .. . . xv... . ‘
3 2 . .
. an .5 Eu. 3.84.. V , . . Swrfi... .
n k . , ‘3; P iguflufifi . 6.26.? .é.
(I 1...» T; .._.. . . V
I \N( , V‘lx .‘<“..'» I
.L.‘ . .. a 3
575%.?

- :3 .‘
1.5::

a. Annifluhys
., r , ~

. . c 1.?
£2. 0‘“... :2 v‘l ‘ V :anrfixfllv‘éaxa:
V . u? ' . . V vcvvucr...
N < N. . ’ {$.32
. Lair-SW :53);
$.54 .81. . a Mfu-KI‘LQA :-
.; 3.5 :33. , v61!- )\ I.L I:
....u he. or: . :1.
19.3. . 3 ,

x.
.q
:

u
’t.
.

r

[9

|
l
1

b
.

:11: A.
« Iv (In 5
th‘t.
l

1“
r‘q"
t
g

1
:1
b

§II.V.:¢.

c

. ...
r.
“V.
fig
{V

. . i

4

33.. x6.

5:.
.3:
‘fl.
L

‘4

n’. 1.; :v 69.»
a l‘s... .19. 9525...); n
, 6.. 1:1) riurnovo.
.5!.AA¢O ti 1!: !
If 1.."

.
..

,~

..

{r
.;.I.

\

v?"
.,- ‘
.
t‘.‘
gfi I

L4: 3...:

Ethwtju 34...».

{8% .v!!::.
g

,1
t

5:.
Ii:

. S
if. .. «hub», 23
’7‘... .lW‘n‘fi‘.

.IL . .

p
s'v
,
.31
z.
1
’5’ ‘ I ’

.u...:.
.11:

P,»
H“

ul. .3 5.2.23! 5. 1 $8
V :57-..” dub“: 52.4.. Jr. .wsrlve‘o 9

..: y - .91 l . .
. . h Ialvova 0} V .v f~c|7..qu.|.u
. . i . vii-fl.‘
. , 3.123;»: .
‘li . l V}
ilk!!! 3

4—:
35‘:

1g

1." . .11....
‘.'V..b.§
ovuly c

.1

It:
oiifififi it
VIP. (4H)

09!. '6. 131.46}?
9‘ t .1...» I...

V v OK .. . :0 vi... ‘i‘tvi.

E chlvvti .Xihiu . A: V a titty 0ft} 9: a): I ..v..1-Vf¢.?n.€l!v
_V r :2rlv’io 3‘4. AIchaIJIQ .IFDIIA Yolk;
. . at. “11.01:?!” . .10).!!! ..r.1.fi!fihflfl s.)
.. r x .ttlwtvrlfiv .E.
rt 1 .- IV:
1:33“. 5. v} V
I ..v 0033!“...
a

.w 3%. .

In: }. it pt:

.\|rkm a:
1 3......30v
ui.bk-\. r I 3|...
L...\v.-. A

1..

v1.1".
. 1.. v 3 .
v1: .5125?!
.1 4: 30a). . V Al};
13.1.5-1: an xa.:>.vv14.0:i‘vflvntxv‘ll
.l i. I . . {9:33.}... nb‘iili
\ V . :vlrvv
|.\a {OuEVK-cérx
. {Hf-53.)}!Itlfi '4 {in}.
l li‘wl‘. .0 .
v, :5: V
. .fitcxll‘Nl

. u .r . ‘D‘ll
Eb»...

.yll‘ v! |
.itlaalfi?l.
kctx‘kbal
7.. 1-; IX 5‘12va 5. .l (.l"!>!it’
. (fair: ~ A .L?.Vl|£v'vf. ; . 4-!!Fvililtlfl'u'
llil:-l!l'lV-1

2
I. . 3.
vv....~.:.¢L

 

. Vt» l.- nfi’). .30:
.w , .»..4.‘b|
II

 

 

III I a:pn,fo.)r..|m.:r v-A» b

LII'VIIVVKIIIII. 3.: van. .

THESJS

lllll||l|||||||||||l|||lllllllllllllllllllllllllllllllllllll

31293 01706 9802

This is to certify that the

dissertation entitled

The Plastome Mutator of Oenothera: A Molecular and
Biochemical Examination of the Mutation Process

 

presented by
Tseh-Ling Chang

has been accepted towards fulfillment
of the requirements for

Ph.D. degreeinBOtany and Plant Pathology

 

 

Barbara E . _ MEL——
Major professor

Date WI", 19‘} 7

 

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIE'EARY
‘ Mlchlgan State
University

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/96 chIRCJDateDue.p6&p.14

THE PLASTOME MUTATOR OF OENOTHERA:
A MOLECULAR AND BIOCHEMICAL EXAMINATION
OF THE MUTATION PROCESS

By

Tseh-Ling Chang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Botany and Plant Pathology
and
Genetics Program

1997

ABSTRACT
THE PLASTOME MUTATOR OF OENOTHERA: A MOLECULAR AND
BIOCHEMICAL EXAMINATION OF THE MUTATION PROCESS

By

Tseh-Ling Chang

Oenotheraplants homozygous for a recessive allele at the plastome mutator (pm)
locus show non-Mendelian mutation frequencies that are lOOO-fold higher than
spontaneous levels. Characterization of RF LP sites in a collection of mutants shows that
insertion-deletion hot spots in the pm lines are defined by tandem direct repeats,
implicating replication slippage or misalignment during recombination. To search for
other DNA lesions that would not be visible as restriction fragment length
polymorphisms, PCR—amplification products of the psbB gene were digested with a
restriction endonuclease, denatured, and examined for single strand conformational
polymorphisms (SSCP). Among 21 mutants, one 4-bp insertion and one point mutation
were identified in psbB. The discovery that the plastome mutator can cause base
substitutions as well as repeat-mediated insertions and deletions points to a likely defect
in a component of the chNA replication machinery.

An in vitro system has been developed to investigate the chloroplast replication
machinery of Oenothera. Few differences were observed between the pm- and wild-type

chloroplast extracts provided with activated DNA. However, biochemical analyses show

that KCl is more essential for DNA syntheses by the pm-extracts than the wild-type
extract. In a different assay, phagemids carrying a pm deletion hot spot were used to
produce single stranded DNA as a template to examine DNA synthesis from a specific
primer. Variation in extension of the primer were observed with the chloroplast
replication extracts from the pm- and wild-type lines. The in vitro differences on the
single stranded template allow topoisomerase and helicase to be ruled out as candidates

for the genetic lesion which results in plastome mutator activities.

‘1?) Mir-75mg,
My fiebvecfflusfiand

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to Professor Barbara Sears for her
encouragement and assistance throughout this research. It is certain that this work would

not have been possible without her.

I would also like to thank my guidance committee members Professor Karen Friderici,
Professor John Ohlrogge and Professor Kenneth Keegstra for their generous support and

constructive suggestions during the period of this study.

Special thanks are given to late Professor James Asher for his help in establishing the
SSCP technique, my college classmate Hao-Ping Chen who is a graduate student at
Cambridge University for technical advice on chromatography, and Professors. Gordon
Cannon and Sabine Heinhorst at University of Southern Mississippi for their support and

fruitful discussion.

Kindly assistance and suggestions for my presentation from Monica Guhamajumdar

and other colleagues are highly appreciated.

Finally, loving thanks is offered to my parents and my husband for their sacrifice and

patience.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... viii
LIST OF FIGURES .......................................................................................................... x

CHAPTER 1. GENERAL INTRODUCTION

Organization of plastid genome ............................................................................. 1
DNA replication in chloroplast .............................................................................. 3
In vitro replication system of chloroplast .............................................................. 6
Plastome mutators: an overview ............................................................................ 8
The plastome mutator of Oenothera ...................................................................... 9

CHAPTER 2. CHARACTERIZATION OF DNA LESIONS CAUSED

BY THE PLASTOME MUTA TOR
Introduction .......................................................................................................... 12
Materials and Methods ......................................................................................... 20
Results ............................................ '. ..................................................................... 25
Discussion ............................................................................................................ 30

CHAPTER 3. ANALYSIS OF PLASTOME MUTATOR PREFERRED
TARGETS IN A SECOND TYPE OF PLASTOME

Introduction .......................................................................................................... 34
Materials and Methods ......................................................................................... 35
Results .................................................................................................................. 36
Discussion ............................................................................................................ 48

CHAPTER 4. ESTABLISHMENT OF AN IN VITRO REPLICATION SYSTEM TO
ASSESS DNA REPLICATION IN CHLOROPLASTS OF THE
OENOTHERA PLASTOME MUT A TOR

Introduction .......................................................................................................... 5 5
Materials and Methods ......................................................................................... 56
Results .................................................................................................................. 62
Discussion ............................................................................................................ 75

vi

CHAPTER 5. A IN VITRO REPLICATION SYSTEM OF OENOT HERA
CHLOROPLASTS DOES NOT SUPPORT PREVIOUSLY

IDENTIFIED “ORI” ORIGIN
Introduction .......................................................................................................... 80
Materials and Methods ......................................................................................... 83
Results .................................................................................................................. 88
Discussion ............................................................................................................ 89
CHAPTER 6. SUMMARY AND CONCLUSION ........................................................ 94

APPENDIX A ESTABLISHMENT OF THE PRIMED-SSDNA IN VITRO
REPLICATION SYSTEM IN OENOTHERA

Introduction .......................................................................................................... 99
Materials and Methods ....................................................................................... 100
Results and Discussion ...................................................................................... 100

APPENDIX B THE PLASTOME MUT A T OR OF OENOT HERA DOES
NOT AFFECT COPY NUMBERS OF PLASTOMES

IN THE m-PLANTS
Rationale ............................................................................................................. 106
Materials and Methods ....................................................................................... 107
Results and Discussion ...................................................................................... 109

APPENDIX C CLONING OF PM-PREFERRED TARGETS INTO PHAGEMIDS

Introduction ........................................................................................................ 1 17
Materials and Methods ....................................................................................... 117
Results and Discussion ..................................................................................... 118
LIST OF REFERENCES ............................................................................................... 122

vii

LIST OF TABLES

Chapter 2

Table 2.1 Primers for PCR and sequencing ........................................................ 24
Chapter 3

Table 3.1 Primers for PCR .................................................................................. 37

Table 3.2 Summary of pm-induced DNA alterations

in Oenothera pm-I stocks ............................................................. 38

Chapter 4

Table 4.1 Purification of the chloroplast DNA polymerase ............................... 64

Table 4.2 Requirement for in vitro DNA synthesis ............................................ 66
Chapter 5

Table 5.1 Plasmids used in site specific initiation experiments ......................... 84
Appendix A

Table A. 1. DNA synthesis activity of different primer/template ratio
at room temperature ................................................................... 101

Table A2. DNA synthesis activity of different primer/template ratios
at 70°C ....................................................................................... 102

Table A3. DNA synthesis activity of different primer/template ratios
at 100°C ..................................................................................... 103

Table A4. DNA synthesis activity of different primer/template ratios
with alkali treatment ................................................................. 104

viii

Appendix B

Table B.1 DNA content of plastids in pm-plants .............................................. 111
Appendix C
Table C.l Phagemids constructed for propagation of ssDNA .......................... 119

ix

Chapter 1
Figure 1.1
Chapter 2

Figure 2.1

Figure 2.2
Figure 2.3
Figure 2.4

Figure 2.5

Figure 2.6
Figure 2.7
Chapter 3

Figure 3.1

Figure 3.2
Figure 3.3

Figure 3.4

Figure 3.5

LIST OF FIGURES

Physical map of Oenothera chloroplast DNA ...................................... 2

Simplified sequence comparison of the hypervariable region
of orf2280 from the three Oenothera lines (C1, C2 and D),

with the similar region from Nicotiana tabacum (N. t) ............... 13
Sequence comparisons of sites responsible for RFLPs ..................... 14
ori region ........................................................................................... 17
Model of replication slippage ............................................................ 19
Sequence comparison of oligo-A stretches between the rpll4

and rpll6 genes ........................................................................... 21
SSCp analysis of psbB gene .............................................................. 27
DNA sequence of psbB gene segments ............................................. 28

A) and B) PCR analysis of the intergenic spacer region

of 16S rRNA-trnl .......................................................................... 40
A), B), C) and D) PCR analysis of orf2280 variable region 41
A) and B) PCR analysis of the Hind fragment of Bam 3b ................ 44
PCR (A, B) and restriction digestion (C, D) analyses of the

intergenic spacer region of the rplI 6-rplI 4-rps8 genes ............... 46
SSCP analysis of the psbB gene on the MDE gel ............................ 49

Chapter 4

Figure 4.1 Elution profile of DNA synthesis activity ........................................... 63
Figure 4.2 Duration of DNA synthesis ................................................................. 68
Figure 4.3 Activated DNA requirement ............................................................... 69
Figure 4.4 Dependence of DNA synthesis on MgCl2 .......................................... 71
Figure 4.5 Processive DNA replication (photo) ................................................... 73

Figure 4.6 Quantification of DNA synthesis through secondary structures ......... 74

Chapter 5

Fig. 5.1. A) Origins of plastome replication mapped on the vicinity
of the 23S rRNA and 16S rRNA genes; B.) Approximate
locations of putative replication origins relative
to rplI6 genes 82
Fig. 5.2 Site specific initiation ............................................................................ 90
Appendix B

Fig. B.1 Southern blot of undigested DNA (no treatment) hybridized
with (A) chNA probe and (B) nuclear probe .......................... 110

Fig. B 2 Southern blot of undigested DNA (caylase treated) hybridized with
(A) chNA probe and (B) nuclear probe ................................. 114

xi

Chapter 1

GENERAL INTRODUCTION

Organization of the plastid genome

Plastids are autonomously replicating organelles in photosynthetic eukaryotes.
Since the discovery of the plastid genome (plastome) more than three decades ago,
knowledge of the organization and structure of this molecule has accumulated
dramatically (reviewed by Kirk and Tilney-Bassett, 1978; Palmer,l991; Shimada and
Sugiura, 1991). Despite the fact that plastids with different shapes and functions exist in
various tissues, the DNA composition is the same in all of them while differential
expression of plastome genes distinguishes them from each other.

Plastids contain multiple copies of plastomes subdivided among nucleoids that are
probably the units for segregation and division. The plastome molecules of higher plants
are closed circular double stranded DNAs, with unit sizes that vary from 120- to 200- kbp
(Palmer, 1991). Monomeric- or oligomeric units of plastomes have been observed in pea
(Kolodner and Tewari, 1975), spinach (Deng et a1. 1989), and watermelon (Bendich and
Smith, 1990). Except for few species (e.g. legumes), most plastome molecules in higher
plants contain two exact inverted repeats which are 20 to 30 kb in size and separated by a
large and a small single copy region, as exemplified by the plastid genome map of

Oenothera in Figure 1.1. The entire nucleotide sequences of the plastomes from the

psgDv

Oenothera hookeri

Chloroplast DNA
150 kb

 

Figure 1.1 Circular map of Oenothera chloroplast DNA. Relative
positions of genes used in this dissertation are as shown. Inverted
repeat regions are represented by bold lines.

liverwort Marchantia polymorpha (Ohyama et al., 1986), tobacco (Shinozaki et al.,
1986), rice (I-Iiratsuka et al., 1989), and maize (Maier et al., 1995) have been determined.
Each plastome encodes 120 to 140 genes that are involved in photosynthetic electron
transfer activities and its own transcription and translation. Often, a complex multi-
subunit protein in organelles is encoded by genes from both nuclear and chloroplast
genomes. The nuclear encoded precursors are transported into plastids through the
guidance of transit peptides, they associate with proteins encoded in plastids, which then
become functional. The best-known example is the ribulose bisphosphate carboxylase
(rubisco) with its large subunit encoded by the plastome while the small subunit is
encoded by the nuclear genome. Generally, plastomes are highly conserved in genome
size and gene order. The main causes of length variation among plastomes include
changes in the amount of repeated DNA, intron content, gene content, and other classes
of deletion-insertion events. The majority of spontaneous length mutations are 1~10 bp in
size (Palmer, 1991). In monocots, DNA rearrangement and amplification/diminishment
of specific fragments were found to be induced in plastomes during callus induction and

plant regeneration in cell/tissue culture (Day and Ellis, 1984; Fukuoka et a1. 1994).

DNA replication in chloroplasts

Part of this section is based on reviews by Heinhorst and Cannon (1993) and
Gillham (1993).

The most popular model of chloroplast DNA replication was proposed by
Kolodner and Tewari (1975). Based on the electron microscopic (EM) examination of

replicating chNA in pea and maize, they postulated that chNA replication in higher

plants involves the Cairns and the rolling circle models. It was interpreted that chNA
replication begins with the Cairns mechanism in which DNA replication is initiated by
two replication displacement loops (D-loops). The two origins expand towards each
other to form the Cairns fork. Replication subsequently moves bidirectionally until the
two ends meet each other at a site that is 180° around the DNA molecule from the
initiation site. The DNA gaps left could either be sealed by DNA ligase or serve as the
initiation site for rolling circle replication, through which new plastome molecules are
synthesized through amplification. The mode of chNA replication in pea has been
investigated in vitro (Reddy et a1. 1994) and both types of replication intermediates
(sigma structure of the rolling circle mechanism and theta structure of the Cairn
mechanism) have been observed. Evidence obtained from other organisms seem to
support this model to some degree. The replication intermediates of chNA from
Euglena, Chlamydomonas, tobacco and Oenothera have been examined with electron
microscopy and their replication D-loops were mapped (Koller and Delius, 1982; Ravel-
Chapuis et a1. 1982; Waddell et al. 1984; Chiu and Sears, 1992). In Oenothera,
replication intermediates of the displacement loops fi'om unidirectional synthesis of one
strand of DNA were observed (Chiu and Sears, 1992), and in some cases, two D-loops
were observed on the same molecule. The replication intermediates observed in Euglena
(Koller and Delius, 1982; Ravel-Chapuis et a1. 1982) and Chlamydomonas (Waddell et al.
1984) are mainly double-stranded arms resulting from bidirectional synthesis of two
nascent strands, which suggested that discontinuous synthesis begins immediately after

D-loop initiation (Heinhorst and Cannon, 1993), or the replication origins are too close to

discern from each other. After applying the technique of pulsed-field gel electrophoresis,
Bendich and Smith (1990) suggested that the long linear molecules they observed may
have resulted as the intermediates of rolling circle replication. The existence of plastome
oligomers observed by Deng et a1. (1989) might also seem to support the concept of
rolling circle replication, except that no tail longer than 1.5 unit of the plastome length
was discovered. Rather, frequent recombination events were proposed to explain the
multimers observed in the chloroplasts of spinach.

Most of the proteins that are involved in chNA replication appear to be encoded
in the nucleus. Evidence came from the observation that in white plastids of iojap
maize, which lack chloroplast ribosomes and therefore do not contain any chloroplast-
encoded proteins, the plastome DNA is nevertheless replicated (Walbot and Coe, 1979).
Additionally, Rapp and Mullet (1991) showed that an inhibitor of chloroplast genome
transcription (tagetitoxin) did not impair chNA synthesis during leaf development).
Furthermore, in wheat, up to 80% of the chloroplast genome apparently can be deleted
without affecting replication of the mutant plastomes (Day and Ellis, 1984). This line of
evidence provides indirect proof that chNA replication is not largely dependent on
plastome encoded proteins.

The abundance of plastomes and plastids in higher plant cells differ
developmentally and tissue-specifically (Lamppa and Bendich, 1979; Possingham and
Lawrence, 1983). Light seems to be an important factor stimulating synthesis of
plastomes. Plastome numbers vary from 3300 to 12,000 copies per cell under dark and
light grth conditions respectively in Nicotiana tabacum (Weissbach et a1. 1985). It

was estimated that in the leaf meristems of wheat, there are about 1000 plastome copies

per plastid before chloroplasts divide (Boffey and Leech, 1982), and the plastome number
drops to ca. 60 per plastid in mature leaves. Conceivably, plastomes replicate rapidly at
an early stage of leaf development, and then plastid division outpaces plastome
replication resulting in reduction of plastome copy numbers per plastid. It has been
suggested by Bendich (1987) that increases in the amount of chNA content occur in
order to provide sufficient templates for expression of chloroplast encoded genes.
Consequently, an adequate quantity of proteins will be made to meet the demand for
elevated photosynthesis rates during leaf development. In older leaves of barley, the
reduction of chNA copy numbers per plastid may also be caused by degradation of
chNA in addition to redistribution of plastomes (Baumgartner et a1. 1989). However,
the copy numbers of plastome per plastid in epidermal cells of spinach does not change

during leaf development (Lawerence and Possingham, 1986).

In vitro replication systems of chloroplast

Several in vitro replication systems have been constructed in different laboratories
to study the process and regulation of DNA synthesis within the chloroplast. Thus far,
the goal has been to establish a system that could support site-specific initiation of
chNA replication.

In earlier work, crude extracts with DNA synthesis activity were obtained directly
from isolated intact or disrupted chloroplasts of maize (Zirnmermann and Weissbach,
1982), Marchantia polymorpha (Tanaka et al., 1984), and Petunia hybrida (de Haas et

al., 1987), with no further purification. DNA synthesis in crude extracts from maize

depended on exogenously added plasmid containing plastome inserts, but no site specific
initiation was observed. The same situation was observed in the crude chloroplast extracts
from Marchantia polymorpha. In the crude extracts fiom Petunia chloroplasts,
deoxynucleotide incorporation does not need exogenous DNA. In these systems,
prevalent endogenous DNA and nuclease may have inhibited DNA replication and site
specific initiation to some extent (Heinhorst and Cannon, 1993).

Recently, more refined in vitro replication systems of chloroplasts have been
established following additional purification steps (Gold et a1. 1987; Meeker et al. 1988;
Carrilo and Bogorad, 1988; Heinhorst et al. 1989). Replication extracts having DNA
polymerase activity are prepared by passing triton-disrupted chloroplasts through a
diethyl-aminoethyl (DEAE) cellulose and a heparin column. Most endogenous chNA
and DNA nucleases are removed at this step. Further chromatography steps may be
performed in some cases for better purification.

A crude chloroplast replication system from Chlamydomonas reinhardtii was
developed by Wu et a1. (1986). A high salt extract of the thylakoid membrane fiaction
was isolated following the procedure developed by Orozoco et a1 (1985), to study site
specific transcription of spinach extracts in vitro. A soluble protein fraction essential for
chNA replication was added to the high salt extract to support this in vitro replication
system. With increasing amount of inhibitors (ddCTP) of chNA replication, Wu et a1.
(1986) showed that chNA replication begins close to a site that was mapped by D-
looping as a replication origin (ori A). Similarly, Meeker et a1. (1988) showed that the
partially purified pea chloroplast DNA polymerase initiated DNA replication at the

vicinity of Orz' regions mapped through EM studies. In maize, the in vitro replication

system (Carrillo and Bogorad, 1988) pointed to a site-specific initiation from a preferred
template which did not match the D-loop sites observed by EM (Kolodner and Tewari,
1975 a & b). The inconsistency between the in vitro and in vivo studies was also observed
in studies of tobacco in which a single replication origin was mapped by electron
microscopy (Takeda et al. 1992) while an extra replication origin was mapped with both

the in vitro and in viva tools in another study (Lu et a1. 1996).

Plastome mutators : an overview

Plastome mutators are nuclear loci that greatly enhance the naturally low
spontaneous rate of plastome mutation. According to the classification of Kirk and
Tilney-Bassett (1978), these nuclear mutants can be grouped into two classes, based on
the spectrum of mutant phenotypes induced. One example of the narrow-spectrum group
is iojap of maize (Shumway and Weier 1967; Walbot and Coe 1979; Thompson et a1.
1983, Han et a1. 1992), in which affected leaves always have longitudinal white stripes of
the same phenotype. The wide-spectrum group is exemplified by chm of Arabidobsis
(Redei and Plurad, 1973; Mourad and White, 1992; Mmmz-Zapater et a1, 1992;
Sakamoto et a1. 1996), cpm of barley (Prina, 1992, 1996), and the plastome mutator (pm)
of Oenothera (Epp 1973; Chiu et a1. 1990; Johnson and Sears 1990a & 1990b; Johnson et
a1. 1991; Sears and Sokalski 1991; Chang et a1. 1996). Mutant phenotypes vary in these
plastome mutators, e. g. leaf deformation, and varying degree of pigmentation deficiencies
of leaves. In all cases, the mutator activity is due to a recessive nuclear allele, capable of
causing plastid defects only when it is homozygous. The mutations caused by the mutator

are transmitted independently of the nuclear background. Variable features of the various

mutators suggest that they are likely to affect genes of different functions. The iojap
locus has been cloned and sequenced (Han et al. 1992), but its function remains
unknown. The chm-locus of Arabidopsis has three alleles, and two reports correlate
mitochondrial DNA rearrangements with chm activity ( Martinez-Zapeter et a1. 1992;
Sakamoto et al., 1996). Conceivably, the chm is a mutator of mitochondria conferring an
impact on both mitochondria and plastids (Mourad and White 1992; Sakamoto et a1.
1996). Analysis of mutants induced by the cpm mutator of barley implied that it may
have been possibly involved in activation of transposable elements. However, no

molecular evidence was provided in that report (Prina, 1996).

The plastome mutator of Oenothera

The plastome mutator of Oenothera hookeri strain Johansen was isolated by Epp
(1973) through EMS mutagenesis. It has a lower penetrance than does chm, but it still
causes variegated sectors to appear in about 35% of all pm /pm plants (Epp 1973; Sears
1983; Epp et al. 1987; Sears and Sokalski 1991). Genetic analysis showed that the
plastome mutator is encoded by a recessive nuclear gene. This nuclear gene is capable of
elevating the frequency of plastome mutations 200-1000 times higher than the
spontaneous levels (Epp 1973; Sears and Sokalski 1991). Restriction fragment length
polymorphism (RF LP) analyses revealed the association of the plastome mutator activity
with chNA alterations in pm lines (Chiu et a1. 1996). Sequence characterizations
indicated that all of these variable sites are A-T rich (about 80%) and are surrounded by

direct repeats (Chang et a1. 1996). These pm-preferred targets are in all respects similar to

10

the sites where replication slippage events tend to occur fi'equently in prokaryotes
(Levison and Gutman, 1988). A replication slippage model therefore has been proposed
to explain the mode of action of the plastome mutator (Chang et al. 1996). Although most
evidence collected in our laboratory points to a likely defect in a component of the
chNA replication machinery, the real role of the plastome mutator in DNA metabolism
remains elusive. In this research, molecular and biochemical approaches have been
undertaken to study the plastome mutator of Oenothera from different angles.

As reported in Chapter 2, to search for other mutations that would not be visible
as RFLPs, PCR-amplification products of several photosynthetic electron transfer related
genes were digested with restriction endonucleases, denatured, and examined for single
strand conformational polymorphisms (SSCPs). Variations thus identified were
sequenced.

As discussed in Chapter 3, all of the previous variants recovered from the
plastome mutator were in plants that carried plastome type I. To investigate whether the
plastome mutator impacts different plastome types differentially, PCR amplification was
performed with newly derived pm/pm-IV plants, to search for polymorphisms in
previously identified mutation sites.

Chapter 4 describes an in vitro replication system that was adapted to assess DNA
synthesis from segments of cloned Oenothera chNA. Additionally, a primed-template
replication strategy was designed to investigate the chNA replication machinery from
pm-plants. Plastome fragments containing previously identified pm-hot spots were cloned
into phagemids to produce ssDNA as templates for replication experiments. Additionally,

in order to observe in vitro site specific initiation in Oenothera chloroplasts, several

11

clones containing the putative origins of replication that were identified in chNA from
Oenothera (Chiu and Sears, 1992) or other sources, (Meeker et al. 1988; Wu et al. 1986;
Gold et a1. 1987; Hedrick et al. 1993 and Lu et a1. 1996) were examined with this newly
developed in vitro system.

In Appendix A, a comparison of chNA content between the pm- and wild-type
plants was conducted to study the impact of the plastome mutator on efficiency of
chNA replication. The effect of the caylase treatment on the preparation of total DNA

from Oenothera is also discussed in that context.

Chapter 2

CHARACTERIZATION OF DNA LESIONS CAUSED BY THE
PLAS TOME MU TA TOR OF OENOT HERA

The main content of this chapter is from Chang et al. (1997)

Introduction

Initial studies of chNA in pm-induced mutants used restriction endonucleases that
cut chNA infiequently, and no RFLPs were observed (Epp et al. 1987). In contrast,
Chiu et a1. (1990) used frequently cutting enzymes and showed that chNA
polymorphisms were found at a high frequency in the plastome mutator lines of
Oenothera. These RF LP sites consisted of deletions ranging from 50-500 bp.
Examination of these pm-hot spots in natural lines revealed little variability relative to
what was observed in pm-plants (Chiu et a1. 1990). Sequence characterization of one
RFLP site in a collection of pm-plants showed that different copies of numbers of 24—bp
tandem repeats were deleted in two pm lines (Figure 2.1; Blasko et a1. 1988), suggesting
a repeat-mediated mutagenesis mechanism for the plastome mutator. Chang et a1. (1996)
examined the same fragment in another set of pm mutants, and all plants were found to
harbor deletions relative to the progenitor chNA. No duplications were observed at this
site. At another site where direct repeats exist in chNA, a deletion event eliminated one
copy of a 29 bp tandem repeat in the non-coding spacer of rps8-rplI4 in one member of

the pm collection (Figure 2.2A; Chang et a1. 1996). The plants also included a variant

l2

 

l3

 

m. I

 

0m» a IB

 

041m.) 5 I! I! 13

O-j-(D) E I! IE IE I! IS

 

Figure 2.1. Simplified sequence comparison of the hypervariable region of
orf2280 (Figure from Blasko et a1. 1988) from the three Oenothera lines
(C1, C2 and D), with the similar region from Nicotiana tabacum (N. t.). The
first 16-bp of the 24-bp repeats are represented by black boxes, while the
stippled box indicates the remaining 8-bp. The bars within the repeat units
displays the only nucleotides difference between the near-repeat and the
tandem repeats.

 

Figure 2.2. Sequence comparisons of sites responsible for RFLPs (Chang et al.
1996). In'the regions of variability, sequences of the wild-type Dusseldorf line
(wt) and mutant (pm) lines are aligned. Duplicated/deleted regions are framed by
boxes; dashes indicate the absence of a comparable sequence. Direct and
inverted repeats are indicated with arrows, dashed arrows indicate sequences of
G[A],,T[A]n. (A) The Cornell-2 pm-line has a deletion of one copy of a 29-bp
tandem repeat found in the wild-type between the rps8 and rplI4 genes (Wolfson
et a1. 1991). (B) The Cornell-1 pm-isolate has a 15-bp exact duplication relative
to wild-type in the fourth largest BamHI fragment located between trnG and trnR
genes.

15

3 2:5

 

 

 

 

 

 

 

 

AIIII All
A. I I I I A IIIII A IIIII
<OH<H<<E<<P<<O<E <<H<<DH<HUE<<H <<H<<OE<HUHF<<P UEUE<E<F<UH<<E<<FUE< Em
<OH<H<<E<<H<<O<E ................. <<H<<OH<HUE<<H UEE<E<H<UH<<E<<BE< :5
V AIIIIII A Tl
IIIV IV I AIIIIII I
Al Al I

m

<0UH<H<<EUHOE<EUH<H<<EUE<<0E0 <HH<FOHUOH<E<<<U<H<<<O Em
<OUH<H<<EUHOE<EUH<H<<EUE<<UEO <E<HOPUOF<E<<<U<F<<<D “3

 

 

 

 

 

 

 

 

A A AIA ........
<2 <Eo<oe<<<<<<§<<e<o<e<<<oofi ................................. L <<<<uE<E :5
<H< <EU<OH<<<<<<<H<<H<O<B<<<OO <EO<OH<<<<<<<H<<H<O<H<<<OOE <<<<UE<E “3
I
A ........ A. .............. v

16

with a unique 15-bp tandem duplication in another intergenic spacer (Figure 2.2B). Both
inverted and direct repeats are found in the neighborhood of these indels, including 5- and
7- bp direct repeats at either end of the 15 base duplication. Larger deletions were
identified in the 16S rRNA-trnI spacer, and these also occurred between direct repeats
(Figure 2.3; Chang et a1. 1996; Stoike 1997). Conceivably, these direct repeats mediated
the duplication through replication slippage as explained below.

The replication slippage model was proposed by Albertini et a1. (1982) to explain
sites of deletion/duplication in prokaryotes (Figure 2.4). They pointed out that as DNA
is unwound at the replication fork, the unreplicated and unpaired DNA could physically
slip out of position. Slippage of the template strand during replication would result in
deletion, while slippage of the daughter strand would yield duplications. Short direct
repeats facilitate the mispairing of the template and daughter strands. Secondary
structures were also likely to be involved in stabilizing the slipped strand. Preferred
substrates for replication slippage include regions where DNA tends to be single
stranded, sequences that are AT rich, and sites that have a simple alternating sequence
(Albertini et a1. 1982; Levinson and Gutrnan 1987).

If replication slippage is occurring in the plastome mutator lines at elevated
frequencies, we predicted that we would see expansion/contraction of oligo-A tracks in
chNA. Therefore, the mutability of two poly A tracks in pm plants was checked.
Single adenine additions were observed in three of thirteen plants in a 13-bp track, but no
variation was found in a nearby 11-base poly-A track (Figure 2.5; Chang et al. 1996).
The differential susceptibility of the two oligo-A tracks to the plastome mutator-induced

slippage could be due to the requirement of a substrate > llbp or differences in the

14)
Cl

pm8

l-D
CI

pm8

l-D

pm8

l7

-AAAAAA--K

CAAGTCGGTAGG ATCCCCTITI'GGACG‘I'CCCCA'I'GCCCITI'CCGCGCGGGGTAGCA G TCCCCGCGCCC'ITI'CCGCGCT

 

 

 

 

 

worcoorxoo Erccjjcjcm'rj jjjjj G:GACGT T' "' "' J\
CAAGTCGGTAGGm
moreoorxooF—chccrmomcor )

 

I...O'COOOOOOOOCUOICCCOOCO> ———-—-—--

————>
GGGGGGCATG GGGGCGAAAAAAGGAAGGAGGGGGAGOGGWTCTCTCGCFFITGACATAGCAGCGGGCCCCGG

 

 

 

 

 

9 =
TGGGAGGCCCGCACGACGACGACGATTAGCI'CA‘I'I'GGTAGGATCCCC'I'ITI'GGACGTCCCCATGCCCI'ITCCGCGCO
L CCCCATGCCCI l ICCGCGCG

 

 

 

 

{ CCCCATGCCCTI'TCCGCGCG

 

C'0'00.'."°."“."..'00> -——_--—--—-—-———9

ED ____,
GGGTAGCATGGGGGCGAAAAAAGGAAGTAAAATAAGGAGGC’ITI'GACATAGCAGCGGGCCCCGGTGGGAG GCCCGC
GGGTAGCAT®GGGCGAAAAAAGGAA GTAAAATAAGGA GGCITTGACATAGCAGCGGGCCCCGGTGGGAG GCCCGC

 

GGGTAGCAT§GGGGCGAAAAAAGGAAGTAAAATAAGGAGGCTI'I‘GACATAGCAGCGGGCCCCGGTGGGAG GCCCGC

 

 

 

a» % in” 3”” >1. a» mo»

ACGACACGACGA'ITAGATI'AGCTCATI‘GGTAGGACGACGATI'AGCI'CATI'GGTAOGACGACGATI'AGCTCATTGGTA G

ACGACACGACGATI'AGA'ITAGCTCA‘ITGGTAGGEEEACGAI [AECTCATTGGTAEGACGACEAI iAGCT:ATT§GTZ§®

ACGACACGACGATTAGATI'AGCTCATTGGTAGGA CGACGA'I'TAGCTCA'ITGGTAGGACGACGATTAGCI'CATI'GGTAGG

 

’”’ ,»» A A A44 A - A4_A_A_A -—-—-———
V—Vfi v

V—v—v—V—v

W
ACGACG A'I'I'AGCTCGTI'OGTA'ITGGTAGGATCCCCITI'I'GGACGTTGACATAG GAGCGG ATGACAT AGGAGCGGGCCCC

 

 

 

ACGACG ATl'AGCI'CG'ITGGTA‘ITGGTAGGATCCCC’I'ITI'GGACGTI'GACATAG GAGCGG ATGACAT AGGAGCGGGCCCC

 

——————— 9 ”’ ,” ”’ ”’ "’ ”, ," > ”~fldfld‘”‘g—.

AGCGGGAGT CCCGCACGACGACGACACGACGACGACGA‘ITAGCT C GTTGGTA'ITGGTA GOATCCCC'ITITGGACGTT

 

 

I I
\

AGCGGGAGT CCCGCACGACGACGACACGACGACGACGATTAGCT C O'ITGGTA'ITGGTA GGATCCCCITI'TGGACGTT

 

GGGAGCGGATGACATAGGAGCGGGCC CCAGCGGGAGTCCCGC ACGACGACGACACGACG/ ACGACGATI'AGCTCATI'GG

 

 

 

GGGAOCGGATGACATAGGAGCGGGCC CCAGCGGGAGTCCCGC ACGACGACGACACGACG ACGACGATI'AGCI'CATI'GG
GGGAGCGGATGACATAGGAGCGGGCC CCAGCGGGAGTCCCGC ACGACGACGACACGACG ACGACGATI’AGCTCA'ITGG

-—D in» am»
TAGG ACGACG ATTAOC’TCATI'GGTAGG ATTAGCTCAGTGTTAGAG'ITAGAG CGGGCCCCAGTGGGAGGCCCGCACAAC
’ ATFAGCTCAGTGTTAGAGTI'AGAG CGGGCCCCAGTGGGAGGCCCGCACAAC
TAGG ACGACG A‘I'TAGCTCATI'GGTAGG ATTAGCTCAGTG'ITAGAG‘ITAGAG CGGGCCCCAGTGGGAGOCCCGCACAAC
TACO ACGACG ATTAGCTCATI‘GGTAGG ATTAGCTCAGTGTI'AGAG'ITAGAG CGGGCCCCAGTGGGAGGCCCGCACAAC

 

 

 

Figure 2.3 Sequence comparisons of the intergenic spacer of 16S rRNA-trnl fi'om pm-
induced mutants (Stoike, 1998; Chang et a1. 1996). I-D: wild type Dfisseldorf line. pm-
variants: C1 (Cornell-1), C2 (Cornell-2), pm8. Different families of repeat units are
represented by various arrowheads. The borders of the deletions are framed by boxes.

 

 

Figure 2.4 Model of replication slippage. Boxes represent direct repeats and the direction
of replication is indicated by arrowheads. A.) Slippage occurs during replication due to
misalignment between repeat units of template and nascent DNA. B.) After another round
of replication, deletion is recovered from slippage of template strand while duplication is
obtained from slippage of nascent strand.

l9

 

 

 

 

immfimxmmfimmfixw

 

flflflnflnHMnUGHHMUNHHMUHHu

 

"g
.8535

+

 

 

 

v .N .5

 

   

 

 

   

 

\\\\\\\\\\\\\\\\\\\\\\\k\\\\8

 

“HUDNNUU

 

 

 

HHNHHUNH
552.5

+

EV

 

 

 

 

 

2V

 

_:c
:oomm:

IS]

20

context of the oligo-A stretches. For example, in the analyses conducted thus far, one
short repeat sequence G[A],,T[A]n, has been located in or near most of the pm-target sites
(marked with dashed line and arrowhead in Figures 2.2 and 2.5).

Although we believe our evidence supports the interpretation that direct repeat-
mediated replication slippage occurs at an elevated fi'equency in plants carrying the
Oenothera plastome mutator, unequal inter- or intramolecular recombination could be an
alternative explanation for the insertion and deletion events in pm-derived plants as
postulated for some examples of repeat-mediated evolutionary change in chNA (Lin et
a1. 1984, Wallis et a1. 1989, Palmer 1991).

During this analysis, we realized that it was possible that the plastome mutator
causes more types of mutations than we had thus far recognized. Since our initial work
focused on RFLPs, small indels or base substitution would have been overlooked.
Analysis of SSCPs provides great sensitivity for revealing single base substitutions. Thus,
it is the technique of use to screen for mutations in the present study. New variegated
mutations were selected and a subset of plastome genes required for photosynthetic

electron transfer were assayed for single strand conformational polymorphisms (SSCPs).

Materials and methods

Plant material. The plastome mutator seeds were obtained originally from Prof. W.
Stubbe (University of Dusseldorf), who had perpetuated the line by self pollination of
progeny descended directly from the original E-15-7 mutant of Oenothera hookeri strain

Johansen (Epp 1973).

21

.N.N 0.53m 5 8 03 30:80: .350 333 .3 neg...“ 2a magma <-om=O .vocmm—a 2a ANMEQ
can maxi .NéoEoDv 8c: Em Ba 23— .? Ho cembogv on: boEmuQQ 25-33 EB.“ macaw ENS 28 V23 =3§on 368. <
.boa 95 mo mooaoscom .GmE .3 Ho wSfiUV macaw “SE 23 V23 2: 5223 8:323 <-ow=o Co 53888 cocoacom .md oSmE

AI I I I
0<0<0 <<<<<<<<<< H9500: <H<<<0 <<<<<<<<<<<<<< H0<<0H<<0<<<H0<<<008<<0EH Em
0<0<0 <<<<<<<<<<< 9.0.5.00: <H<<<0 <<<<<<<< .. <<<<< H0<<0H<<0<<<P0<<<00P<<0EH «3

III II

.Inlll AMI.

 

 

 

 

 

 

 

 

22

As explained by Chiu et a1. (1990), plants of the strain Johansen are all descended
from a single rosette collected in 1927 (Cleland 193 5). The line that was mutagenized
by Epp during his work at Cornell University, and from which Epp isolated the pm
allele, is referred to as the Cornell line. Except for mutations caused by the EMS
treatment, it should be identical to the wild-type Dusseldorf line of strain J ohansen. The
Cornell-1 and Cornell-2 lines contain photosynthetically-competent plastids, with a wild-
type phenotype, although they contain chNA RFLPs relative to each other and relative
to the Diisseldorf line (Chiu et al. 1990).

The pm7 and pm 73:: mutations were isolated from the same plant, and have been
included in previous investigations (Johnson and Sears 1990a,b, Johnson et a1. 1991;
Chiu et al. 1990). These plastome I mutants have been maintained in leaftip culture in
the Sears laboratory since 1983; they have an MC, pm/pm+ nuclear background. The
pm/pm7 line was obtained by self pollination of the original pm7 plant; within this line,
the pm7 plastids have been kept for seven years in a pm/pm background, in
vegatatively-propagated leaf tip cultures. These three lines were compared because the
plastomes should be identical except for mutations caused by pm-activity. The 1993
field plants that were tested for new mutations were obtained by self-pollination of a
single pm/pm plant from our 1992 field. These plants carried the Cornell-1 type
plastids, and are descended directly from sibs of the plants analyzed by Epp et a1. (1987).
New mutations induced by the plastome mutator were recognized as light green or white
sectors in young pm/pm seedlings; these plants were selected and transplanted to the field.
During the growing season, leaves were sampled for DNA extraction from 18 different

plants containing pigment deficient sectors.

23

Preparation of nucleic acids. Plant total DNA extractions were performed as

described by Fang et a1. (1992).

Primer design, PCR amplification, restriction digestion, and SSCP analysis.
Primers were designed based on sequence conservation of coding regions between
tobacco and rice plastomes (Table 2.1). Oligonucleotides were synthesized at the
Macromolecular Structure Facility at Michigan State University (East Lansing). PCR
was performed with a Tempcycler 11 model 1105 (Coy Corporation, Grass Lake, Mich.)
or a MiniCycler Model PTC-150-16 (MJ Research, Inc. Watertown, MA), using
nucleotides from Perkin Elmer Cetus Corp. (Norwalk, CT), and Taq DNA polymerase
from Boehringer Mannheim Biochemicals (Germany). PCR reactions were done in a
100u1 total volume and overlaid with mineral oil. The amplification conditions consisted
of 3 min at 94°C followed by 30 cycles of 1 min at 94°C and 1.5 min at 40°C and 2 min
at 72°C, with a final extension for 10 min at 72°C, followed by prolonged incubation at
4°C. Amplification products were transferred to fresh tubes, precipitated and washed
with ethanol, and dissolved in water. The amplification products were digested
individually with restriction enzyme Alu I (rbcL), HaeIII (psaA), T aq I (psaB), Sau3A
(psbB), MspI (psz). The pst+ F product is only 350 bp in size, and therefore it does
not require restriction digestion. Visualization of single strand conformational
polymorphisms (SSCPs) through MDE (JT Baker) gels and silver staining were

performed as described by Morell et a1 (1993).

24

Table 2.1 Primers for PCR and sequencing

CDDNA region

rbcL

psaA

PsaB

psbB

psz

prE+prF

N.A.

Name*
rbcLF
rbcLR

psaAF
psaAR

psaBF
psaBR

psbBF
psbBR

psbB3F

pszF
pszR

pst+pstF
pst+pst R

universalF

Sequence

5 ’ -GTCACCACAAACAGA-3 ’
5’-CTCCAI I I GCAAGC-3’

5’-CATTTCTCAAGAAC-3’
5 ’ -CCTACTGCAATAATTC-3 ’

5 ’ -GGTTTAGCCAAGGC-3 ’
5 ’ -GAAAAGTGAGCTAA-3 ’

5 ’ -CCTTGGTATCGTGT-3 ’
5 ’ -GTAGTTGGATCTCC-3 ’

5’-GAATTAGATCGCGC-3 ’

5 ’ -GACTGGTTACG-3 ’
5 ’ -CAAATCCTGCTGCA-3 ’

5 ’ -ATGTCTGGAAGCAC-3 ’
5-TTATCGTTGGATGAAC-3 ’

5 ’ -GTAAAACGACGGCCAGT-3 ’

*The direction noted for PCR primers F-forward (5’), R-reverse (3’) corresponds to
gene transcription for psbB gene and orf2280; all others correspond to the
directionality of data entry from the tobacco chloroplast genome (Shinozaki et a1.
1986). The direction of the universal primer was defined by the manufacturer

(Gibco BRL).

25

Cloning and sequencing. The amplification product of the entire psbB gene, and
the third largest fragment resulting from Sau3A digestion of the PCR product, were
obtained from wild-type and mutant nucleic acids. The products were run on a 1.5%
agarose gel to separate them from primers and other fragments, extracted from the gel
with Qiaex (Qiagen Inc., Chatsworth, Calif), cloned by conventional methods (Sambrook
et a1. 1989) in plasmid pUCl8 (Norrander et a1. 1983) in E. coli strain DH50L. Plasmids
were purified using the Wizard mini prep kit (Promega Co., WI). Sequence data were
obtained using 35S-labeled dATP from New England Nuclear/Dupont (Wilmington, DE).
The preparation of double strand DNA and dideoxy sequencing (Sanger et al. 1977) with
Sequenase, were performed as outlined in the U. S. Biochemical (Clevand, Ohio)

instruction manual. Primers are indicated in Table 2.1.

Results

The psbB gene is a site of pm—induced mutation in two albino isolates from 1993
field

DNAs from 18 new albino pm-mutants and the pm 7, pm 73s, and pmpm7 lines were
amplified with primers specific for the psbB, rbcL, psaA, psaB, psz and pst+ pst
genes (Table 2.1). As indicated in the materials and methods section, all but the latter
amplification products were digested with a restriction enzyme before denaturation and
electrophoresis on an MDE gel. Most of the samples were a mixture of wild-type and

mutant chloroplast DNA because nucleic acids were extracted from variegated leaves that

26

carried newly arisen mutations. In principle, any mutation in the segment of DNA being
amplified, should be visualized as an extra band preceding or following the normal band
(Morell et al. 1993). A mutant that has completely sorted out will give single rather than
multiple bands, but the mobility will differ from the wild-type band. Since each band on
the gel is composed of single stranded DNA, a mutational change will be reflected in two
bands of DNA simultaneously. As shown in Figure 2.6, SSCPs were found at the psbB
gene from an albino mutant “93-b45” and the leaftip culture line “pmpm 7”.

The PCR amplification products of the psbB gene were sequenced directly using
internal primer psbB3F, but a frame-shift in mutant 93-b45 rendered the sequencing gel
unreadable at the site of mutation because the sample contained a mixture of wild type
and mutant DNA. Consequently, the PCR products from wild-type and mutant tissues
were cloned and the plasmid insert was sequenced. As shown in Figure 2.7, a 4-bp
insertion (TTTC) was found in mutant 93-b45 at position 84-87 of the Oenothera
sequence. This mutation causes a frame-shift in psbB at a position equivalent to the
455th codon of the 501 codon tobacco gene, and thus is likely to be the reason for the
albino phenotype. Through direct sequencing of the PCR product from pm/pm 7, a
single base transition (C/G -> T/A) was found at position 150 of the Oenothera
sequence in Figure 2.3. The observation that the mutation has sorted out and the wild-
type sequence is lost is probably due to a founder’s effect during subculturing and

maintenance of the leaftip culture line.

 

 

27

 

 

F1gure2.6. SSO’ amlysis ofpsngene (Grmget a1. 1996). PCRproductswere
digestedwithSLm3A dantmedarxlseparatedonanh/DEgel. [armoontainthe
following samples: 1. pm7ss, 2. pnpn7, 3. pm7, 4. 93—b45, 5. 93-a31, 6. 93-028, M
1kb narker (Gibco). Arrowheads point tobandsof altered nobility.

28

Figure 2.7. DNA sequence of psbB gene segments (Chang et al. 1996). Sequences from
Oenothera wild-type (OEN), 93-b45 (B45) and pm/pm7 (pp7) were aligned with the
homologous region from chNA of Nicotiana tabacum (N17). The numbering at the top
left above the sequences refers to the location of the tobacco chNA sequence (Shinozaki
et al. 1986); the wild-type Oenothera plastome I sequence numbers are shown
underneath the sequence data. Dashes indicate the absence of a comparable sequence.
In the tobacco sequence, bases that differ between the wild-type Oenothera sequence
and tobacco are underlined. The 4-bp insertion and the base substitution are framed by
boxes. The position of a stop codon created due to the frameshift in mutant 93-45b is

marked by the diamonds. Direct repeats are indicated with arrows.

NIT
OEN
B45
PP7

NIT
OEN
B45
PP7

NIT
OEN
B45
PP7

NIT
OEN
B45
PP7

NIT
OEN
B45
PP?

29

76256

C'ITTG AAATC QGATG GTGTT TTTCG TAGCA QTCCA AGQGG ITGGT
CTTTG AAATC GGATG GTG'I'I' TTTCG TAGCA ATCCG AGAGG CTGGT
C'I'ITG AAATC GGATG GTG'IT TTTCG TAGCA ATCCG AGAGG CTGGT
CTI'TG AAATC GGATG GTGTT TTTCG TAGCA ATCCG AGAGG CTGGT

76301'" —> —>-> —>—>—>

rerr moo GCATG orroo moo moo rorrorro ----- 'ITC’ITC
rerr moo GCATG orroo moo roroo rorrorro ----- 'I'I‘CTTC
'I'l‘ACT moo GCATG orroo moo roroo Torrorro -i'rorro
rerr moo GCATG orroo moo roroo rorrorro ----- 'ITC'I‘TC
89

76342 _,

GGACA oAm GGCAI ooroo TAGAA ooxro Trvo AGATG m
GGGCA oAm GGCAC ooroo TAGAA oooro Trvo AGATG ’ITl'I'l‘
GGGCA CA'ITT oovo ooroo TAGAA ooorc 'ITCAG AGATG T'ITI’I‘
GGGCA oAm oovo ooroo TAGAA oooro 'I’I‘CAG AGATG m

—_. 134

76387 «M

ooroo TAT’I‘G ACCC A’oAm AGATQ CICAA GrooA Amo GAGCA
ooroo TA'ITG ACCC o GATTT GGATA CGCAA GTGGA AT'ITG GAGCA
ooroo TA'I‘TG ACCC o oAm GGATA CGCAA GTGGA Amo GAGCA
oo'roo TATTG ACC.GAT1'1‘ GGATA CGCAA GTGGA Amo GAGCA

--—-----+ -------——r 134

76432 —>

TTCCA AA AAC TTGGA GATC

TTCCA AA AAC TTGGA GATC

TTCCA AA AAC TTGGA GATC

TTCCA AA AAC TTGGA GATC
198

Fig. 2.7

30

Discussion

RFLP analysis has been used to detect mutations induced by the plastome mutator in
previous studies (Chiu et al. 1990). Due to its limited resolution, only DNA alterations
that range from 50-500 bp were detectable. In order to carefully assess the spectrum of
mutations caused by the plastome mutator, I sought to utilize a method which is sensitive
enough to reveal changes as subtle as a single base. To accomplish this, the SSCP
technique was exploited to observe variation of PCR amplification products from
photosynthetic electron transfer related genes. Restriction endonucleases digestion were
used to digest the products of PCR reactions before gel electrophoresis, in order to
increase the resolution of the SSCP analysis.

Using this procedure to assess PCR amplification products from 18 new albino
mutants and three older lines that had all been derived from the same plant, one insertion
mutation (B45) among a series of direct repeats, and a point mutation (PP7) were
obtained in the psbB gene, as shown in Figure 2.7. The four-base insertion of TTTC
occurred in the midst of a series of TTC tandem repeats. This 4-bp insertion causes a
frameshift mutation, beginning with the 45 5th codon which is altered to specify Leu
instead of Gly. Due to the frameshift, a truncated protein should result, which would be
19 amino acids shorter than the wild-type polypeptide, and which would also contain 28
missense amino acids at its carboxy end. Unfortunately, the B45 plastome mutation was
not recovered from the field, so this prediction cannot be tested. The TTTC insertion can
be explained if slippage of the daughter strand occurred during replication, and resulted in

an imprecise duplication. The second mutation identified in the psbB gene was

31

recovered in a pm-homozygous line, pm/pm 7. This base substitution is not present in
the original pm7 mutant (lane 3 of Fig. 2.6), and would not alter the amino acid
sequence. Thus, it is not responsible for the albino phenotype of the leaftip culture.
Most likely, many other mutations have happened during the seven years of vegetative
propagation of the pmpm7 line. Although the base substitution in pmpm7 did not
involve any direct repeats, its recovery is consistent with our previous finding that NMU,
a mutagen that causes point mutations, has a synergistic effect with the plastome mutator
(Sears and Sokalski 1991). The recovery of this point mutation in the plastome mutator
line indicates that pm-induced mutations are more variable than we had previously
realized. At this time, the frequency of occurrence of point mutations relative to indels
(insertion/deletion) is unclear, but further SSCP analysis of albino mutants should clarify
the likelihood of occurrence of different types of mutations.

The involvement of the direct repeat motif in deletion/insertion mutations strongly
points to replication slippage as being a major mutation mechanism in the plastome
mutator line. In the plastome, likely examples of replication slippage have been
documented in evolutionary comparisons (Treier et a1. 1989, Wolfson et a1. 1991, Madsen
et a1. 1993, and Sears et a1. 1995), but in pm/pm plants, the frequency of replication-
slippage must be greatly elevated.

Several lines of evidence have linked DNA arrangements among tandem repeats to
replication errors. Frequencies of deletion formed between tandem direct repeats are
increased to as much as 100 times in yeast with a temperature-sensitive DNA polymerase
delta and the spectrum of mutations induced are also different than in the wild-type strain

(Tran et al. 1995). The alpha-catalytic subunit of E. coli DNA polymerase III elevated

If“: ‘ El‘flnfifia-§i E."

32

frameshift mutations dramatically when exonuclease was removed to eliminate proof-
reading activities (Mo and Schaaper, 1996). Additionally, it has been shown that the
mutation frequencies of regions containing short direct repeats depend on the orientation
of replication (Rosche et a1. 1995; Trinh and Sinden, 1991). In these studies, inserts
containing asymmetric palindrome sequences with respect to the flanking direct repeats
were designed, such that different misaligned intermediates could form on the leading
and lagging strands during replication. Evidence obtained from these studies has
suggested that deletions occur preferentially on the lagging strand during replication-
associated events, although data from others led to a different conclusion (Westo-Hafer
and Berg, 1991). The misaligned replication intermediate of the replication slippage
model*has been demonstrated in E. coli by Lovett and F eshenko (1996). A dysfunctional
methyl-directed mismatch repair (MMR) pathway increases deletion frequency of 101-
bp tandem repeats that differ from each other slightly (mismatches at four bases).
Normally, such repeats would be removed by the MMR pathway during or soon after
replication, because replication slippage would result in a misaligned-heterduplex
intermediate that would be recognized by the MMR machinery.

Mismatch repair is known to repair mutations caused by replication slippage in both
prokaryotic and eukaryotic systems (Radman and Wagner, 1986; Strand et a1. 1993;
Karren and Bignami, 1994; Kinzler and Vogelstein, 1996; Lovett and Feschenko, 1996).
Conceivably, plastome mutator activity could be due to a defect in MMR. However, the
MMR pathway rarely repairs mismatches longer than 8 bp (Radman and Wagner, 1986),
and the repeat units of most of the pm-target sites are much longer. (Chiu et a1. 1990;

Chang et a1. 1996) Thus, it is unlikely that the plastome mutator is deficient in mismatch

33

repair. Rather, it is likely that some protein involved in the replication/or recombination/
repair machinery is impaired.

A defect in chNA replication or repair has long been considered to be the likely
cause of the high fi‘equency of non-Mendelian mutation due to the plastome mutator
(Epp 1973; Sears 1983; Chiu et al. 1990, Sears and Sokalski 1991). The completely
recessive nature of the pm-allele indicates that it is probably a null allele, since a
defective, but present polypeptide would probably increase the mutation level in the
heterozygote. Thus, we deduce that the pm-homozygote probably lacks a polypeptide
that participates in chNA metabolism. DNA helicase, which unvvinds the helix, is a
possible candidate for the pm gene product (Lahaye et al. 1991), as is the single-stranded
DNA—binding protein (SSB), which stabilizes single stranded DNA during replication and
repair (Van Dyck et a1. 1992). Increased stalling of the replication complex due to
absence of a helicase or SSB protein could increase the frequency of slippage and /or
affect the proofreading function of the DNA polymerase. Absence of the 3’-> 5’
exonuclease subunit of DNA polymerase would also result in many types of mutations
since proofreading would be eliminated (Johnson et a1. 1995).

Given the genetic traits and our deductions about possible functions of the pm-gene
product, in vitro assays of chNA replication and other aspects of chNA metabolism

are needed to discriminate among these possibilities.

Chapter 3

ANALYSIS OF THE PLASTOME M U TA TOR PREFERRED
TARGETS IN A SECOND TYPE OF PLASTOME

Introduction

According to Stubbe’s genome categorization (1959), the Johansen strain from
which the pm -allele was isolated contains genome type A and plastome type I. In
addition to its native plastome type, the AA nuclear background is compatible to varying
degrees with plastomes type II, III and IV (Kutzelnigg and Stubbe, 1974). Differences in
inheritance patterns (Chiu et al. 1988) and photosynthetic rates (Glick and Sears, 1993)
among various plastome types in a constant nuclear background were observed in our
laboratory. It has been reported by Sears and Sokalski (1991) that albino sectors arose in
pm ~p1ants with plastome IV at a lower frequency than in pm-plants with plastome I. In
order to study more carefully the impact of the plastome mutator on different plastome
types, 'several previously identified pm -hot spots from plastome I were reexamined in
our newly developed collections of pm-induced mutants from plastomes I and IV. The
investigation of pm -1 plants was performed by Lara Stoike, and is described elsewhere.
The observation of the impact of the plastome mutator on plastome IV is presented and

discussed below.

34

35

Materials and Methods

Plant Materials The wild type Johansen strain that carries plastome type IV was
constructed by Prof. Stubbe who transferred the chloroplasts from O. atrovirens into the
AA—Johansen nuclear background (Chiu et al. 1988). This line was maintained by self-
pollination, and the source of seeds was field accession Nr. 95-4. To obtain seeds with a
newly restored pm/pm genotype, plants with an albicans/percurvans genotype with
plastome IV were used as the recipient for pollen from a Johansen line that was
homozygous for the pm allele. Because the albicans-percurvans chromosomes form a
circle of 14 in meiosis, only two types of meiotic product are produced (Cleland, 1972).
Furthermore, because the albicans genome carries a pollen-lethal factor, and the
percurvans genome carries an egg-lethal allele, the progeny of this cross are all of the
genotype albicans-Johansenp‘“, a heterozygote with plastome type IV. In the F1, a circle
also forms at meiosis. Due to the pollen-lethal allele carried by the albicans strain and
predominance of the Johansen genome during megaspore competition (Cleland, 1972),
99% of the F2 progeny are homozygous for the Johansen genome and hence the pm
allele (accession Nr. 96-23). Nineteen pm /pm plants from the 96 field were identified as

having mutant sectors, and were screened for DNA alterations.

Preparation of nucleic acids. It has been a problem to obtain good quality of
templates for PCR amplification from Oenothera due to the excess amount of
polysaccharide existing in the nucleic acid preparation (See Appendix B). Therefore,
extraction of total nucleic acid was performed by the conventional protocol described by

Fang et al. (1992), and was further purified by caylase digestion to remove the remaining

36

starch, as outlined by Rether et al.(1993). Nucleic acid samples were dissolved in a
solution containing 50mM KoAc, lOmM EDTA, 5 g/ml RNAse and 0.5 mg/ml Caylase
M3, then subjected to incubation at 37°C for overnight. After the caylase digestion, the
pH of the samples was increased by adding 1M Tris-HCl (pH 8) to the final concentration
of 50mM, followed by the conventional procedure of phenol/chloroform extraction and
alcohol precipitation. Ethanol was replaced by ethoxyethanol to remove the digested

polysaccharide more efficiently.

Other procedures. SSCP analysis, agarose gel electrophoresis, PCR
amplification, programming and chemicals were described in detail in Chapter 2.
Restriction digestions were performed following the manufacturer’s instructions. Primers

for PCR amplification are listed in Table 3.1.

Results:

No variation in the intergenic spacer region of 168 rRNA-(ml was observed

The intergenic spacer region of 16SrRNA-trn1 genes was previously identified as
a hypervariable region among the wild type plastomes (Sears et a1. 1996). In the Sears
pm-collection, all mutants had deletions at this site (Table 3.2; Chang et a1. 1996). Due to
inverted repeats in this region, strong secondary structures are capable of forming and
appear to be highly susceptible to the plastome mutator in plastome I (Stoike, 1997): The
mutability of this region in plastome IV to the plastome mutator was examined in the

present study. The intergenic spacer region was amplified from DNA isolated fiom plants

 

37

TABLE 3.1. Primers for PCR and sequencing

CpDNA region

psbB

Bam3b

rpll4-rps8

orf2280

l6S rRNA-trnl

N.A.

Name*

psbBF
psbBR

psbB3F

ch3F
ch3R

chPL14F
chPL16R

chPSSF

ch 12F
ch 12R

Cprl 6F ’
trnIA’

universalF

Sequence

See Table 2.1
See Table 2.1

5’-GAATTAGATCGCGC-3’

5 ’ -GAATGGATTCAAAGAG-3 ’
5 ’-AATTTGCGTCCAATAGG-3 ’

5’-TTCTCGAGCCCCACT-3’
5 ’ -AAATGCCTATACGAATCAA-3 ’

5’-TCTATTCATGTCAACAI I 103 ’

5’-CAACCTC'I'TTCAGAT-3’
5 ’ -ATTCCAG'ITTGAGAG-3 ’

5’-TCGTAACAAGGTAGCCGTAC-3’
5 ’-CGACGCAATTATCAGGGGC-3 ’

5’-GTAAAACGACGGCCAGT—3’

*The direction noted for PCR primers F-forward (5’), R-reverse (3’) corresponds to
gene transcription for psbB gene and orf2280; all others correspond to the
directionality of data entry from the tobacco chloroplast genome (Shinozaki et al.
1986). The direction of the universal primer was defined by the manufacturor

(Gibco BRL).

«and Q gum .NmIEIm .ZIEm .Nflpla £39 a .Nfim 8588 3035.33 mfilowwfld 80¢ 838::

38 28 new? 9 @2388 2a mac: GOV .Nflzlofiolo EB .99 .39qu ADV floflumlmwd 05 was: anew 025 80¢ €3QO
929-315 :85 <-ow=o 3-3 .4: .«o onm 98 608% final—N5 cote—ow 3mm $50an Md.mg @055 we moocozuom
“memo.— .«o 598:: E8 28 .<ZQ game 8589 5:85—qu Mom mo 36 2: 688a 2a 3168 a 9 26 303882...

38

 

 

 

 

 

 

 

 

 

 

 

 

32 an 8N4 E... E4 BNN x N 32 x _ a E 2.5
32 an 8N4 2a :2 EN x N 32 x 2 a ME 33
32 B 8N< Ba :2 EN x N 32 x _ a a 3%
an: an 8N< ea :2 EN x N 32 x _ E a N23.
32 n3 8N< use :2 EN x N 32 x I E E :3
32 an R2 ENN x N 32 x _ rue: 58L N23
3: an :2 SN x N as: x _ raga swan men
32 an :2 EN x N 32 x _ was: $81.33
32 8 :2 EN x N as: x n as: :EE Nam
E B 82 a 32 x _ a .<.z No
B: an 8N< e5 :2 EN x N a a a G
82 an 2: EN x N 32 x I 85 8E Q
m <
233.2% 28-35: a: ES 6:3 «as -ofi SNNE 8:
mo beam

 

floogméi Eofiefis 5 282.283 <75 3265-89 no bafifism .N.m «Bah

39

of the pm /pm IV -collection by primers Cprl6F’and trnIA’ (Table 3.1), which flank the
polymorphic region. The PCR amplification products are about 1 kb in size. Despite the
many repeat units found in this area in plastome IV, no polymorphism was observed
(Figure 3.1), which is in contrast to the highly variable phenotypes of the same region in

plastome I.

orlZ280 did not show variability in the pm -IV derived lines of Oenothera

Two variable regions in orf2280 of the pm -IV mutant collection were surveyed
for variation. In one region (orf228OA) of other closely related Oenothera species,
extensive stretches of 21 and 24 bp related repeats alternate with each other (N imzyk et
a1. 1993). Region B is located downstream from region A. High frequencies of deletions
have been observed in that region from our old pm -I stocks (Table 3.2), although an
exact sequence composition remains to be determined (D. Jarrell, unpublished data). The
PCR amplification products produced by primers OF3 and 0R2 are about 600 bp in size
for the wild type Dusseldorf line and 350 bp to 600 bp in our pm -I collection due to loss
of several c0pies of repeat units. With the same pair of primers, region A of orf2280 was
screened across the new pm -IV collection by PCR amplification. Their product sizes
were estimated on a 2% agarose gel (Figure 3.2A). Region A of orf2280 in plastome IV is
about 450 bp in size which is smaller than that of plastome I. However, no alterations
were observed among all of the samples examined.

Region B of orf2280 contains a set of eight tandem 24-bp repeats in plastome I-D
(Blasko et a1. 1988). Deletion of three to five copies of the repeat units in this region had

occurred in all members of the pm-I mutant collection (Table 3.2; Chang et a1. 1996).

40

A)

M12345678910WC1M

 

B)

14 M111213 1516171819WM

 

Figure 3. 1. A) and B) PCR analysis of the intergenic spacer region of

16S rRNA-trnI on the 1.5% agarose gels. The primers cpr16F’ and trnIA’
amplified a segment of this hypervariable region from the following 0
enothera plastomes: A) pm-IV mutants: 95-1 (1), 95-2 (2), 95-3 (3), 95-4
(4), 95—5 (5), 95-6 (6), 95-7 (7), 95-8 (8), 95-9 (9), 95-10 (10); Wild type D
iisseldorf line (W), Comell-I (Cl). B) pm-IV mutant 95-14 (14), 95-11

(1 1), 95-12 (12), 95-13 (13),), 95-15 (15), 95-16 (16), 95-17 (17), 95-18
(18), 95-19 (19), Wild type Diisseldorf line (D). The size markers (M)
(From Gibco 1kb DNA ladder) in gel A are 1018 bp and 517 bp r
espectively, in gel B are 1636-, 1018-, 517- and 396 bp respectively.

41

Figure 3. 2. A), B), C) and D) PCR analysis of orf2280 on the 1.5% agarose gels.

A) and B) The primers OF3 and 0R2 amplified the variable region A of orf2280 (David
Jarrell, unpublished data) from the following Oenothera plastomes: A) pm-IV mutants:
95-1 (1), 95-2 (2), 95-3 (3), 95-4 (4), 95-5 (5), 95-6 (6), 95-7 (7), 95-8 (8), 95-9 (9), 95-10
(10); Wild type Dusseldorf line (W), Cornell-1 (C1 from 93-field), Cornell-2 (C2). B)
pm-IV mutants: 95-11 (11), 95-12 (12), 95-13 (13),), 95-14 (14), 95-15 (15), 95-16 (16),
95-17 (17), 95-18 (18), 95-19 (19), Wild type Dusseldorf line (D), Cornell-1 (Cl from
93-field), Cornell-2 (C2). The size markers (M) (Gibco 1kb DNA ladder) are 517-, 396-,
344-, 298-bp respectively. C) and D) The primers ch12F and ch12R amplified the
variable region B of orf2280 (Blasko et al. 1988) from the following Oenothera
plastomes: C) pm-IV mutants: 95-1 (1), 95-2 (2), 95-3 (3), 95-4 (4), 95-5 (5), 95-6 (6),
95-7 (7), 95-8 (8), 95-9 (9), 95-10 (10); Wild type Dusseldorf line (D), Cornell-1 (C1
from 93-field), Cornell-2 (C2). D) pm-IV mutants: 95-11 (11), 95-12 (12), 95-13 (13),),
95-14 (14), 95-15 (15), 95-16 (16), 95-17 (17), 95-18 (18), 95-19 (19), Wild type
Dusseldorf line (D), Cornell-1 (Cl from 93-field), Cornell-2 (C2). The size markers (M)
(Gibco 1kb DNA ladder) are 517-, 396-, 344-, 298-bp respectively.

42

A)

M12345678910WC1C2M

 

B)

M111213141516171819WC1C2 M

 

M 123 4 5 6 78 910111213WC2pp7

can“ “um—nun...

 

M 1 11 1213141516171819 W C1pp7M

 

Figure 3.2

43

Using primers ch12F and ch12R listed in Table 3.1, the PCR amplification products
of the wild type Diisseldorf line and pm-I plants were between 260 - 330 bp (Chang et a1.
1996; Table 3.2). To assess the mutability of this region in pm-IV plants, PCR
amplifications were performed with the same primers and products sizes were examined
by agarose gel electrophoresis. Including the wild type control, all of the PCR products
were about 350 bp, migrating slower than those of the wild-type Dfisseldorf line (Figure
3.2B). This is consistent with previous observations that plastome IV contains more
copies of the 24-bp repeats (Sears et al. 1995). No variability was observed across our

pm-IV collection.

The plastome mutator did not induce polymorphisms in the variable Hinfl
segment of the fourth largest BamHI fragment

The fourth largest BamHI fragment (Bam 3b) had been observed as an RFLP in
comparisons of pm -lines (Sears 1983; Chiu et al. 1990), and had been found to have two
polymorphic regions (Kaplan 1987). One of these was examined by subcloning and
sequencing a variable Hinfl fragment from Bam clones of the plastome I lines:
Diisseldorj; Cornell-I, and Cornell-2 lines. Subsequently, primers ch3F and ch3R
were designed for amplification and direct sequencing of all of the members of our pm-I
collection. One line (the original Cornell-1 isolate) showed a 15 bp exact duplication in a
non-coding region corresponding to the spacer between the trnG and trnR genes on the
tobacco chNA map (Fig. 2.2B; Table 3.2; Chang et al. 1996). PCR amplification was
performed to investigate variability of this region in the pm-IV mutant collection. As

shown in Figure 3.3, the amplification products were about 500 bp in size. No mobility

44

A)

12 3 45 67 8 91011 121314WC1C2M

 

B)

11516171819WC1C2M

 

Figure 3.3. A) and B) PCR analysis of the Hian fragment of

Bam 3b on the 1.5% agarose gels The primers ch3F and ch3R
amplified this Hinfl variable segment (Chang et al. 1996) from
the following Oenothera plastomes: A) pm-IV mutants: 95-1 (1),
95-2 (2), 95-3 (3), 95-4 (4), 95-5 (5), 95-6 (6), 95-7 (7), 95-8 (8),
95-9(9),95-10(10)95-11(11),95-12(12),95-13(13),95-14
(14); Wild type Dusseldorf line (W), Cornell-1 (C1), Cornell-2
(C2). B) pm-IV mutants: 95-15 (15), 95-16 (16), 95-17 (17),
95-18 (18), 95-19 (19), Wild type Dusseldorf line (W), Cornell-I
(C1), Cornell-2 (C2). The size markers (M) (From Gibco 1kb
DNA ladder) are 517-, 396—, 344-bp, respectively.

45

differences of PCR amplification products were found in this region in the pm-IV

collection.

The intergenic spacer region of the rpll6-rplI4-rps8 genes remains the same
in pm lines

The rplI6-rpll4-rps8 region was previously identified as a pm -target in pm -I
plants (Chang et a1. 1996): one of two 29-bp repeats was deleted (Fig. 2.2A; Table 3.2).
Plastome types II, III, and IV contain only one copy of this 29-bp segment. Presumably, a
29-bp duplication occurred in the plastome I progenitor after its evolutionary divergence
(Wolfson et al., 1991). Because the plastome mutator is able to induce both insertions and
amplifications in pm-plants (Chang et al. 1996), I sought to determine if the one copy
29-bp fragment in plastome IV was amplified to two copies due to the effect of the
plastome mutator. The whole intergenic spacer region was amplified with primers
chPL16R and chPSSF (Table 3.1), and examined with agarose gel electrophoresis
(Figure 3.4A). PCR amplification products from pm-IV plants were about 900 bp in size,
migrating slightly faster than the products from the plastome-I plants due to their intrinsic
difference in sequence (Wolfson et al. 1991). No other variation was observed across the
pm-IV samples when they were compared with the wild type control. To better visualize
the migration pattern of these products, samples were digested with the Bell restriction
enzyme. The Bell enzyme restricted this intergenic spacer region into three fragments
with approximate sizes of 450-, 300-, and 150- bp (See Figure 3.4B; Wolfson et a1.
1991). The largest fragment of plastome IV is shorter than that of plastome I because it

lacks the 29-bp duplication. The middle fragment from type IV samples ran slower than

 

46

Figure 3.4 PCR (A, B) and restriction digestion (C, D) analyses of the intergenic spacer
region of the rplI6-rpII4-rps8 genes on the 1.5% agarose gels The primers chPL16R
and cpPRS8F amplified this variable intergenic region (Wolfson et al. 1991) from the
following Oenothera plastomes: A) pm—IV mutants: 95-1 (1), 95-2 (2), 95-3 (3), 95-4 (4),
95-5 (5), 95-6 (6), 95-7 (7), 95-8 (8), 95-9 (9), 95-10 (10), Wild type Dfisseldorf line (W),
Cornell-I (C1). B) pm-IV mutant: 95-11 (11), 95—12 (12), 95-13 (13), 95-14 (14), 95-15
(15), 95-16 (16), 95-17 (17), 95-18 (18), 95-19 (19), Wild type Dusseldorf line (D),
Cornell-1 (C1). The size marker (M) (Gibco 1kb DNA ladder) is 1018-bp. In gels C and
D, the products of PCR amplification from gels A and B were digested with endonuclease
BclI. The sample order in gels C and D correspond to that in gels A and B respectively.
The size markers (M) are 517-, 396-, 344-, 298-, 220-75- bp respectively. Fragments a, b,
c are approximately 450, 300 and 150-bp in size respectively.

47

A)

M12345678910WC1M

   
 

B)

M111213141516171819WC1M

 

C)
M12345678910WC1M

“'nufimmuuuumflflw
mummmfiuuunaw...

 

D)
M111213141516171819WM

FA‘

'~~«~*~-~~“nm“‘”
uwvwwaevmmm ,...

 

Figure 3.4

48

type I samples due to a 1-bp insertion in a 13-bp poly-A stretch and the expansion of an
11-bp poly-A track to 19 bp in type IV plastome. Other previously identified base
substitutions in this region between the two plastome types would not have been resolved

on an agarose gel.

The hypervariable region of the psbB gene remains intact in pm-IV isolates

SSCPs were found at the psbB gene of plastome I from an albino mutant “93-b45”
and the leafiip culture line “pmpm7” (Figures 2.3; Chang et al. 1996). The polymorphic
region of the psbB gene was then cloned and sequenced. Based on the sequence data,
primer psbB4F was designed to be used with primer psbBR to amplify the surrounding
156 bp of the variable region (Table 3.1). PCR amplification was performed with these
two primers to screen the pm-IV collection. The product sizes were assessed on an
agarose gel and further examined by SSCP analysis. As presented in Figure 3.5A and

3.5B, no variation was observed among the samples.

Discussion

Several pm -hot spots were mapped previously by RFLP analysis (Chiu et al.
1990) and characterized by sequencing (Blasko et al. 1988; Chang et al. 1996). All of
them occurred in regions surrounded by inverted or direct repeats. These pm-induced
mutations probably resulted from misalignment of repeat units during recombination or
replication. Several observations have led us to favor replication slippage as an
explanation for the pm-mutations (See Chapter 2). Sequence analysis of one l3-bp oligo-

A stretch revealed a one-bp insertion in several pm-I isolates, indicating that replication

49

B4512 3 45 678 910111213141516171819

   

   

Figure 3. 5. SSCP analysis of the psbB gene on the MDE gel. This
variable region (Chang et a1. 1996) was amplified from plastomes of
pm-IV plants with primers psbB4F and psbBR. The products of PCR
amplification were denatured before sample loading. Samples order is
as follow: pm—IV mutants: 95-1 (1), 95-2 (2), 95-3 (3), 95-4 (4),

95-5 (5), 95-6 (6), 95-7 (7), 95-8 (8), 95-9 (9), 95-10 (10) 95-11 (11),
95-12 (12), 95-13 (13), 95-14 (14), 95-15 (15), 95-16 (16), 95-17 (17),
95-18 (18), 95-19 (19), The sample pm93-b45 (B45) in which a 4-bp
insertion was recovered (Chang et al. 1996), was amplified as a size
markers (156-160 bp).

50

slippage had probably occurred. In addition, two SSCPs were recovered in pm-I lines.
One of them is an imprecise 4-bp duplication at a site containing stretches of 3-bp tandem
repeats.

Prior to my study, all of the observations on the pm-induced mutations had been
performed with its native plastome type I. Besides type I, the AA nuclear genotype of
pm-isolates is also compatible with other plastomes (Stubbe, 1959). In the present study,
pm-lines with plastome type IV were used to investigate the impact of pm on a non-native
plastome. The data obtained by Sears and Sokalski (1991) indicated that the mutation
frequencies of the pm-I, II and III plants are about the same, but pm-IV plants had much
lower mutation frequencies. However, S. Baldwin (1995) found no statistically
significant differences in the spontaneous appearance of white sectors in her comparison
of pm-I and IV lines. Therefore, doubts had arisen about the genetic composition of the
pm -line carrying plastome IV that was analyzed by Sears and Sokalski.

Because of the inconsistent results obtained from independent studies in our
laboratory, we sought to carefully assess the impact of the plastome mutator of
Oenothera on plastome types I and IV. Several RFLP sites recovered from pm-I plants
were reexamined in our new pm-I and pm-IV collection. To be assured that all of the
mutations observed are newly induced and not inherited from a previous generation, the
plastome mutator genotype was newly restored as described in the materials and methods
section. In the following discussion, the investigation of the pm-IV lines will be
discussed. Parallel studies on the plastome mutator effect on plastome type I were

performed by Lara Stoike in her thesis (1998).

51

Six previously identified RFLPs were examined in the pm-IV collection: the
intergenic spacer of 16S rRNA and trnI genes, two polymorphic areas in orf2280, the
Barn 3b fragment, the rpII 6-rplI 4-rps8 intergenic spacer and the variable sequence of the
psbB gene. None of them show polymorphisms in plastome IV. Similarly, in the
examination of a new pm -I collection was performed by L. Stoike (Ph.D thesis, 1997),
except for the intergenic spacer of 168 rRNA-rm], no variation was observed. However,
in the subsequent generation, variations could be discerned in both variable regions of the
orf2280 region (Stoike, 1997). Whether the plastome mutator has the same effect on
plastome IV remains unknown. Examination of the same kind of collection of second
generation plants with plastome IV would answer this question.

The absence of variability in the 16S rRNA and (m! intergenic spacer in pm-IV
plants is intriguing. Many direct and inverted repeats exist in this region in both plastome
types I and IV (Sears et al. 1996), and it is a hotspot for evolutionary variation in the
Oenotheras (Sears et a1. 1996; Gordon et al. 1982). According to the analysis by the GCG
program, a strong secondary structure is capable of forming at a site that suffers a high
deletion rate in pm-I plants, (Stoike, 1998). The same region is also present in plastome
IV, but no deletion was observed in my studies, contradicting the postulation that the
potential secondary structure is associated with high frequencies of deletion. The
discovery that no variation occur in the intergenic spacer of the 16S rRNA and trnI may
implicate an important difference in the physiological function of this region in plastome
IV. Alternatively, physiological differences between plastids carrying plastome IV and
plastome I may render the 16S rRNA-trnI region differentially susceptible to the

plastome mutator.

52

In prokaryotes, the deletion frequencies of the same secondary structure can differ
by 1000 fold although they are just a single-bp apart in their positions on a plasmid (Das
et a1, 1987). A precisely controlled position effect on mutation frequencies was
suggested, although the basis for this effect is still unclear (Leach, 1996). Perhaps the
differing susceptibility of the spacer region between plastome I and IV is regulated
similarly. Conceivably, the secondary structure may form less readily in plastome IV due
to other adjacent sequences.

Data obtained by D. Jarrell (personal communication) and Chang et al. (1996),
showed that in both the variable regions of orf2280, DNA alterations exist in every pm-
isolate from our established pm-I mutant collection (Table 3.2). orf2280 is the largest
open reading frame found in the plastome. Studies of tomato orf2280 showed that it
might encode a protein expressed in the chromoplasts (Richards et al. 1991).
Nevertheless, the real fiinction of orf2280 remains elusive. Screening of the two variable
region of orf2280 in the pm-IV collection revealed no variation, which is the same in the
new pm-I stock. However, for the pm-I plants, polymorphisms start segregating out in
the subsequent generation of heterozygous (pm/+). DNA deletions in this area occur in an
“in frame” manner, which possibly would not impair the function of the orf2280 protein.

The two kinds of repeat units in orf2280, which are 21-24 hp respectively, could
also serve as substrates for recombination (Blasko et a1. 1988; Nimzyk et al. 1993).
Therefore, both replication slippage and recombination mechanisms could be responsible
for mutations in this area, and render it more susceptible to DNA alteration. It is
interesting to note that the two most variable regions: “intergenic spacer of 16 rRNA and

trnI genes” and “orf2280” are both located in the inverted regions of chloroplast where

53

presumably recombination occurs at high fi'equencies (Palmer, 1991). Introducing a
dominant Rec’ gene into plastomes of pm-plants would allow the contribution of
recombination to be tested. The fact that in pm-I plants, deletions occurred at relatively
high frequencies in orf 2280 and the rRNA operon, but not in most other intergenic
spacers that contain repeats implies that gene expression may also be involved in
mutagenesis of this open reading frame (Wierdle et al. 1996; Francino et al. 1996). For
transcription, DNA must be unwound , and ssDNA can form hairpins. When confronted
with secondary structures, the replication machinery of the plastome may “slip”.

A 15-bp duplication exists in the Bam 3b region in one member of our historic
pm -I stocks (Table 3.2; Chang et al. 1996). The amplification of this 15-bp segment from
a single copy to a duplicate appears to have been mediated by replication slippage
between a S-bp repeat. The size of this repeat element would have provided inadequate
homology to be involved in a recombinational misalignment. This varying segment was
reexamined in both new pm -I and IV plants and no variation in this area was observed
in the present study.

One copy of a 29-bp duplication in the intergenic spacer region of the rpl] 6-
rplI4-rps8 genes was eliminated in one of our old pm I-stocks (Table 3.2; Chang et al.
1996). This 29—bp region exists as only one copy in the wild-type plastome IV. Therefore,
an amplification would have had to occur for a difference to be detected. However, no
variation was observed in either of the new pm -collections.

An SSCP has been observed from our pm -I isolate as described in Chapter 2 and
Chang et al. (1996). This polymorphic region of the psbB gene contains four copies of 3-

bp direct repeats that offer a typical substrate for replication slippage in both prokaryotic

54

and eukaryotic systems (Radman and Wagner, 1986; Kinzler and Vogelstein, 1996). If
chloroplast recombination resembles that in prokaryotes, the 12-bp would be small to
mediate a misaligned recombination event. The psbB gene encodes the cp47, a
chlorophyll binding protein of photosystem II. A severe mutation in this gene would be
detrimental to the assembly of PSII (Richards et al. 1991), which would give it a selective
disadvantage. In the plastome IV collection being analyzed, slippage events did not occur
in the variable region of the psbB gene. Thus, despite its being a previously observed
target of replication slippage, no variation was observed in this set of pm -IV plants.

The fact that no polymorphisms were observed in the newly developed generation
of pm-isolates raises questions about the frequencies and segregation of pm -induced
mutations. The genetic data (Epp, 1974; Sears and Sokalski, 1991) showed that white
sectors start showing up soon after the pm -seedlings germinate, indicating mutations are
induced and segregated out immediately after the plastome mutator genotype is
regenerated. However, none of the specific spots investigated in this study were hit by it,
or segregated out right away. We had hoped that examining the known mutations sites
would provide information that could be used to analyze possible mechanisms involving
in the plastome mutator. However, the investigation has shown that the different

plastome types may have varying susceptibilities to the action of the plastome mutator.

Chapter 4

ESTABLISHMENT OF AN IN VITRO REPLICATION SYSTEM TO
ASSESS DNA SYNTHESIS IN CHLOROPLASTS OF
THE OENOT HERA PLASTOME MUTATOR

Introduction

In a few plants, chloroplast DNA replication has been studied using in vitro
methods. In early studies, crude extracts with DNA synthesis activity were obtained fi'om
isolated intact or disrupted chloroplasts of maize (Zimmermann and Weissbach, 1982),
Marchantia polymorpha (Tanaka et al., 1984), and Petunia hybrida (de Haas et al., 1987)
with no further purification. In crude extracts from maize and Marchantia polymorpha,
DNA synthesis occurred when exogenous plasmids containing chloroplast DNA inserts
were added. However, no specific initiation site was detected. In the crude extracts from
Petunia chloroplasts, the incorporation of deoxynucleotides into nucleic acids was
observed, without the addition of exogenous DNA. In these systems, the assays were
probably complicated by the presence of endogenous DNA and nucleases (Heinhorst and
Cannon, 1993).

Recently, more refined systems of in vitro replication have been developed in
Chlamydomonas reinhardtii (Wu et al., 1986), maize (Carrillo and Bogorad, 1988), pea
(Meeker et al. 1988), soybean (Hedrick et a1. 1993), and tobacco (Lu et a1. 1996). In these

systems, crude extracts were purified through chromatographic steps and fractions

55

56

containing DNA synthesis activity were collected for in vitro studies. Site specific
initiations of replication were observed. In addition to DNA polymerase, RNA
polymerase, helicase and topoisomerase were detected in these extracts (Hedrick et al.
1993; Reddy et a1. 1994, and personal communication with Gordon Cannon).

As explained in Chapter Two, plastome mutator activity may result fi'om defects
in chloroplast DNA polymerase, topoismerase, helicase, SSB and proteins involved in an
over-expressed S.O.S.-bypass pathway. Therefore the goals of the work described in this
chapter were to develop an in vitro chloroplast replication system and to compare the
DNA synthesis properties of extracts isolated from wild-type and pm-plants. In order to
recover most of the relevant proteins in the chloroplast extracts while removing factors
that inhibit or disturb the interpretation of results, isolated chloroplasts were disrupted
and DNA polymerase and associated proteins were subsequently purified by DEAE
cellulose chromatography.

In this study, procedures were adapted to allow me to study in vitro DNA
synthesis using an Oenothera chloroplasts extract. In the following context, parallel
comparisons of basic features of this system between the wild-type and pm-plants will be
described. Plastome fragments containing previously identified pm-hot spots have been
cloned into phagemids (Appendix C) to produce ssDNA as templates for the examination

of DNA synthesis through primer extension.

Materials and methods
Plant materials. Both wild-type and the pm/pm plants are derived from

Oenothera hookeri strain Johansen with plastome type I. The wild-type Dusseldorf line

57

of strain J ohansen was described in Chapter Two. The pm/pm line was generated from
self pollination of a variant of the Dusseldorf I line (field accession number: 96-40). Leaf
tip cultures were maintained on the modified medium as described by Chiu et al. (1990).
In order to obtain sufficient rapidly growing leaf materials, 7-10 day old leaftip cultures
of Oenothera were harvested, providing a minimum of 80 g of plant material, from
which chloroplasts were isolated. This stage of grth allowed leaves to be sampled
before the developmental stage of expansion, in order to obtain strong replication activity

in the chloroplast extracts.

Isolation of chloroplasts from Oenothera. (On the day that chloropalst were
isolated, Oenothera leaf tip cultures were harvested, and homogenized for 7 seconds in
6-10 volumes of homogenization buffer containing 50mM Tris-HCl, pH 7.5, 6% sorbitol,
6 mM sodium ascorbic acid, 0.15% (w/v) polyvinyl pyrrolidone (pvp), 0.1% BSA, 3mM
cysteine and 5 mM 2-mercaptoethanol. The debris was removed from the homogenate by
pouring it through multiple layers of cheese cloth and 2 layers of Miracloth. Intact cells,
starch, and nuclei were separated from the chloroplasts by differential centrifugation.
Plastids were pelleted at 6000xg for 4 min, and subsequently washed with a solution
containing: 50mM Tris-HCl, pH 7.5, 6% sorbitol, 6mM sodium ascorbic acid, 0.1%
BSA, 3mM cysteine and SmM 2-mercaptoethanol. The final chloroplast pellets were

stored at -80°C for subsequent use.

Preparation of chloroplasts extracts. This protocol is adapted and

modified from the Cannon-Heinhorst Laboratory (Hedrick et al. 1993). Purified

58

chloroplasts from Oenothera were resuspended in a minimal volume of buffer A
containing 50mM Tris-HCl, pH 8.0, 100mM (NH4)ZSO4, 25% glycerol, 10 mM 2-
mercaptoethanol, lmM EDTA, 0.3 mM p-sulfonylfluoride (PTSF), 0.3 mM phenyl
methyl sulfonyl fluoride (PMSF), and 1 mM s-aminocaproic acid. Triton X-100 was
added to a final concentration of 2.0 % and the solution was incubated and stirred on ice
for 1 hour. The crude extracts were centrifuged at 9000 rpm in a DuPont Sorvall SS34
rotor for 30 min. The clear supernatant was loaded at a speed of 0.5ml/min onto a DE52
(Whatman Ltd. England) column pretreated with HCl and NaOH, equilibrated with buffer
A in 0.5M (NH4)2SO4. Before loading the sample, ten volumes of buffer A with 100 mM
(NH,,)2SO4 were loaded at 3ml/min to wash and equilibrated the column. After the
sample loading was frnished, the column was washed with three volumes of buffer A
with 100mM (NH,,)2SO4 at a rate of 2ml/min. The replication extracts were then eluted at
2ml/min by buffer A with 500 mM (NH4)ZSO4. Fractions showing DNA polymerase
activity were collected together and dialyzed against 100 mM (NH,)ZSO4 buffer A with
0.1 % Triton X-100 and 0.1 mM EDTA. After dialysis, the replication extracts were
concentrated to one tenth volume with centricon 10 concentration tubes (Amicon,

Beverly, MA), and stored at -80°C for subsequent experiments.

DNA polymerase assays. The frnal volume of a standard DNA polymerase
assay was 50 1.11. It contained 50 mM Tris-HCl, pH 8.0, 5 mM MgC12, 125 mM KCl, 65
ug/ml activated DNA (DNAse treated calf thymus DNA; purchased from Sigma

Biosciences, St. Louis), 33uM of each deoxynucleotide with 3H-dTTP at 300-1000

59

cpm/pmole, 1 mM DTT and 140 ug/rnl BSA (USB, Cleveland, Ohio). Aphidicolin (fiom
Sigma Biosciences) was dissolved in DMSO at 20ug/ml. The final concentration of
aphidicolin in a standard replication assay was 0.4 ug/ml. NTPs and dNTPs were
purchased from Gibco BRL (Gaithersburg, MD). and 3H-dTTP was obtained from
Amersham (Arlington Heights, IL). The reaction began with the addition of 5 ul of the
extract, followed by incubation at 37°C for 30 min. The quantity of radioactive
incorporation into acid-insoluble materials was measured by transferring the entire
reaction to a DEAE cellulose papers (Whatman, DEl l) and washing once with 0.5 M
NazHPO4 for 10 min, HZO/l min and 95% ethanol/1min respectively. Liquid scintillation

counting was performed to measure the precipitated radioactivity afier the filters were

dry.

Quantification of protein concentrations. Protein concentrations were

quantified using the BioRad Protein Assay kit following the manufacturer’s instructions.

Cloning of chNA substrates into phagemids and preparation of ssDNA.

Two clones containing the hypervariable region between the 16S rRNA and 1m]
genes were obtained using the following procedures. The intergenic region was PCR
amplified with primer cpr16F (5’-TCGTAACAAGGTAGCCGTAC-3’) and trnIA’ (5’-
CGACGCAATTATCAGGGGC-3’). The amplification products were separated from the
primers by agarose gel electrophoresis. The fragments of interest were then extracted

from the gel with Qiaex (Qiagen Inc., Chatsworth, Calif), followed by cloning with

60

conventional methods (Sambrook et al. 1989) into the SmaI site of pBluescript SK (Short
et al. 11988). The pBluescript SK vectors are derived fi'om pUC19, and contain a
replication origin of fl filamentous phage. Depending on the orientation of the fl origin,
pBluescript SKs are designated as (+) or (-). With the infection of helper phage, the (+)
strain allows the recovery of the DNA sense strand of the IacZ gene while the (-) strain
produces the antisense strand. E. coli strain XLl-Blue (Bullock et a1. 1987) was used as
the host for the constructions.

R408, an f1 helper phage (Russel et a1. 1986) was used to infect transformed
E.coli strain XLl-blue for propagation of ssDNA. The preparation of single strand
phagemid DNA was performed as outlined in Sambrook et al. (1989). An overnight
culture of the bacterial strain containing recombinant phagemid was infected with R408
at a final concentration of 2 x 10 7 pfu/ml, incubated for 18 hours at 37°C with strong
agitation (300 rpm/min). The overnight culture was transferred to a centrifuge tube and
centrifuged at 12,000g for 5 min at 4°C. The supernatant was transferred to a fresh tube
and precipitated in 4% (w/v) PEG 6000 and 3% NaCl for one hour, followed by
resuspension in 10 mM Tris-HCl (pH 8). Three phenol/chloroform extractions were
performed to remove proteins from the preparation. The upper, clean aqueous phase was
transferred to a clean tube and precipitated with 1/10 volume of sodium acetate and two
volumes of absolute ethanol at -20°C for one hour, followed by centrifugation at 12,000g
for 20 min at 4°C for 10 min. The precipitate was washed with 70% ethanol and

resuspended in TE buffer for subsequent use.

61

Analysis of products of DNA synthesis by alkaline gel electrophoresis.

Reaction mixtures (100 pl) contained 50 mM Tris-HCl (pH 8.0), 4 mM MgC12,
125mM KCl, 30uM dNTPs, lmM DTT, 7.0 ug BSA, 0.5 ug single stranded recombinant
phagemid DNA primed with 4 p mole SK primer labeled with y-3ZP (Andotek, Ca, 3000
Ci /mole). The SK primer (5’-CGCTCTAGAACTAGTGGATC-3’) complementary to
the antisense strand of the lacZ gene on pBluescript SK (-) phagemid was synthesized at
NBI Inc. (Plymouth, MN). The reaction mixtures were incubated at 30°C for 15 and 30
min. The reactions were terminated by the addition of EDTA and SDS to final
concentrations of 10 mM and 0.1 % respectively, extracted twice with phenol/chloroform
and precipitated with ethanol. The precipitates were resuspended in 30mM NaOH and 1
mM EDTA for electrophoresis on a 1.5 % alkaline agarose gel (25cm x 15cm x 0.5cm)
containing 30mM NaOH and lmM EDTA at 40 V for overnight. After electrophoresis,
the gel was washed in distilled water, dried under vacuum, exposed at -80°C to Kodak X-
Omat AR X-ray film, or at room temperature to a storage phospho screen (Molecular
Dynamics). Quantification of signals on the gel was performed by scanning the
autoradiographs using the Image Quant Program (Molecular Dynamics), with data
manipulation through Microsoft Excel. Background was measured in an adjacent lane,

and was subtracted automatically.

62

Results

Properties of the replication extracts from Oenothera chloroplasts

Elution profile: Following the procedures described in the materials and
methods section, chloroplasts of Oenothera were isolated by differential centrifugation
and disrupted with triton. The crude extracts of chloroplasts were loaded onto a DEAE
cellulose column, DNA binding proteins were step-eluted with the high salt buffer
containing 0.5M (NH4)ZSO,. Fractions containing DNA polymerase were selected based
on their ability to incorporate 3H-dT’TP into acid insoluble materials when DNase treated
calf thymus DNA was supplied as substrate.

The clution profile of fractions displaying DNA polymerase activity from the
Oenothera chloroplasts of wild-type and pm/pm plants is shown in Figure 4.1. With the
step elution procedure, chlor0plast DNA polymerase was recovered as a single peak from
both the wild-type and pm-plants. The maximum level of 3H-dTTP incorporation is
significantly higher (5-10 times) than the background level (Figure 4.1 fraction 4 vs.
fraction 1) even before the concentration step. The heights of peaks varied in different
preparations, possibly as a result of slightly differing tissue culture conditions or minor
variations in the purification process for each preparation. However, the peak activity
showed up in approximately the same fractions in every preparation, regardless of the
batch or the lines of plant materials used. As shown in the representative data in Table
4.1, the DEAE cellulose column afforded a five to seven fold purification of the DNA
synthesis activity. The enhanced specific activity of DNA polymerase afier
chromatographic purification steps (Table 4.1 and Figure 4.1) may be due to higher

concentration of proteins involved in replication activity relative to other proteins, or a

63

cowoom £5508 Ea £38me E wontomow 3 Bo 353 803 @983 ommuoabom <75
638:8 803 316.8 coumuofioofi 02820:: wcgmfioo maouomm .bmzuom £888? <75 mo 2&on noufim fiv .wE

.on E3: cocoa."—

Ir
8
+

‘9. V. N. o
o o o

uonerodroou! H8

°°.
o

 

 

 

 

 

:3 ....... 83 22,1:

 

8E2“. 5:5

64

Table 4.1 Purification of the chloroplast DNA polymerase

 

Protein Activity specific activity purification
(ug) (*units) units/ug
A. Wild type
Crude extract 202.8 42.5 0.2 1
DEAE cellulose 10.4 15 1.4 7
B pm
Crude extract 169.6 72.8 0.4 1
DEAE cellulose 14.8 33.8 2.2

 

Due to differing culture conditions and the amount of plant materials, DEAE resins used,
the extent of purification of extracts for each batch is not exactly the same as each other.
*A unit of activity is defined as 1 nmol of deoxynucleotide incorporated into acid-
insoluble material per hour.

65

reduced amount of replication inhibitors present in the crude extracts, e.g., nuclease and
endogenous nucleic acids (Heinhorst and Cannon, 1993). The extent of purification in
each preparation differs due to slight variations in experimental conditions. No important
differences were observed between wild-type and pm-plants in terms of elution profile

and the extent of purification of replication-active proteins.

gm : The purity of the replication extract of Oenothera chloroplasts was
tested using aphidicolin, which is known to be a strong inhibitor of the nuclear DNA
polymerase or (Sala et al. 1980; Reviewed by Heinhorst et al. 1989) and other nuclear
polymerases to varying extents, while not inhibiting the 7 type DNA polymerases of
organelles. Therefore, it is commonly used to test for nuclear contamination in organelle
replication extracts. The diagnostic concentration of aphidicolin used in this study (20
ug/ml) does not allow synthesis of either leading or lagging strands by the nuclear
enzyme. As shown in Table 4.2, after the addition of aphidicolin, the incorporation rate of
dNTP remains at 90-95 % of the original value. Thus, DNA synthesis activity in our
system is predominantly resistant to aphidicolin. This line of evidence indicates that the
DNA polymerase from chloroplasts is a 7 type and is responsible for almost all of the

DNA synthesis activity observed.

Other features: As summarized in Table 4.2, this system is completely
dependent on MgCl2 and exogenous DNA. The addition of KCl has a minor effect on the

replication activity on the wild type plants while the pm-plants are more affected by the

66

Table 4.2 Requirement for in vitro DNA synthesis

 

 

Reaction conditions DNA synthesis activity, %

VVlld type pm-plants
Complete 100 100
Plus aphidicolin 9_5 90
Minus replication extracts 2 2
Minus calf thymus DNA, activated 4 3
Minus Mg++ 2 3
Minus KCI 7_3_ _3_Z
Minus dNTPs 9 9
Plus rNTPs 86 87
Plus ATP 76 83
-40 primer <1 N.A.
M13 template 16 NA
~40 primer+ M13 template 117 NA

 

67

concentration of KCl. The replication extract is dependent on deoxynucleotides, but does
not need ribonucleotides or ATP alone for the incorporation of dTTP into nucleic acid. In
fact, the ribonucleotides seem to slightly inhibit deoxynucleotide incorporation. These
basic features are very similar to systems characterized from other organisms, including
Chlamydomonas (Wu et al. 1986), soybean (Heinhorst et al. 1989), and maize (Carrillo et
al. 1988). There are no differences observed between the wild-type and pm-plants for

these basic features, except for the KC] requirement.

Kinetics of PHI-TTP incorporjation: The incorporation activity of in vitro DNA
synthesis was quantified during a time course. Because of the slight differences in each
preparation, the incorporation activity is presented as a percentage of the highest activity
in each set of experiments. The comparison of the incorporation activity between the wild
type and pm-plants is shown in Figure 4.2. No significant differences exist between the
two types of samples in their overall incorporation rate during the entire two hour
incubation. At the end of the time course experiments, both of them were still at the peak
of the incorporation activity, which is consistent with observations fiom other systems
(Zimmermann and Weissbach, 1982; Tanaka et al. 1984; Wu et a1. 1986; de Hass et al.

1987) where the incorporation activities last longer than one hour.

Substrate concentration: Initially, the concentration of activated DNA used was
65 pg/ml which is the same amount used in the Cannon-Heinhorst Laboratory for the
DNA replication extract assay. However, later observations indicated that reduction in the

amount of activated DNA led to higher activity. As shown in Figure 4.3, the

68

douoom $5808 98 mfitouna
E 333:0 3 Bo @038 883 888%.. 42288985 828225 no 358 08C. 2858? <75 mo 539:5 NV .wE

E E .05: 20:33:.

ON? 0: oo_. om cm on cm on ov om om or o

.1 2 e 2 _ _ L _ _ p _ 2
_ _ _ _ 2 a _ _ _ N _ 2 o

 

ID
0
parerodroou! Hit

N we

 

 

 

 

Ea - - -0- - - o9: n__>>I|O|

 

 

69

._ .5302 3052: Ea 2222:
E 3:550 3 35308 33 £858? <ZQ no 5326880 22223 .«o 80am 4:082:62 <75 vo~m>m8< mé .wE

A. 553 <zo 8:268 co 2.8 :2

00¢ oom CON oo_. o
_ w l. O

“F-

_

l 000—.

 

l ooom

l ooom

 

uogrerodroou! d _|_ lp-Hg

l ooov

70

incorporation of dTTP continued to climb when the amount of activated DNA was
decreased to 39 ug/ml. The amount of activated DNA required to reach the peak activity
was not followed further because of the slow increment in incorporation activity with
decreasing DNA content. Final concentration of activated DNA of 39 ug/ml or less was

used for later experiments.

MgCl_2 optimum: As described in the previous section, MgCl2 is essential for in
vitro DNA synthesis by the chloroplast replication extracts. The influence of Mg++
concentration on [3H]-dTTP incorporation in replication extracts from both the wild type
and pm-plants has been investigated in this study. As shown in Fig. 4.4, an optimum
concentration of Mg++ at 0.5 mM was observed for extracts from both the wild type and
pm-plants. As explained before, due to the slight differences of the replication extracts
from each preparation, the incorporation of [3H]-dTl"P is expressed as percentage of the

peak activity.

Variations in primer extension were observed with the replication extracts
from wild-type and pm-plants

To observe the products of primer extension using the Oenothera chloroplast
replication extracts from wild-type and pm-plants, I used an end-labeled primed-ssDNA
that would enable me to observe both quantitative and qualitative aspects of DNA
synthesis. Although the in vitro replication system is purified through one DEAE
cellulose column to remove endogenous nucleic acids and it is dependent on exogenous

substrates for replication, extensive contaminating DNA and RNA appear to be present

71

on

p
q

mm

cm

-0—

.couoom 30508 ES “”3588 E Bum—So ma 35308 33
538385 02820:: no 523588 £032 mo 83:: .NGwE no £858? <75 no 0052039 TV .wfi

:5 .~_omsao.o=oo
2 S

.0-

 

 

Ea ....:l 25 gall:

 

 

un-

 

 

fio
Nd
m6
v.0
m.o

% ‘paieJodJoou! Hg

72

based on ethidium bromide fluorescence in agarose gel electrophoresis (data not shown).
Thus, to avoid competition with the contaminating nucleic acids, an end-labeled primer
was chosen which would be complementary to the vector but not to the chloroplast
genome. The oriB segment was chosen for this assay because it is a deletion hot spot of
the plastome mutator and contains numerous secondary structures was chosen for this
assay. It was cloned into phagemid to produce ssDNA as described in the materials and
methods section and appendix Cl. DNA polymerase has been reported to be very
sensitive to secondary structures. It will stall at secondary structures that have formed in
the template when no 888 is available to provide assistance (Williams and Kaguni,
1995). Thus, if SSB is limiting or non functional in the extracts, we expect to see DNA
synthesis pausing at those secondary structures.

Fig. 4.5 shows the products synthesized by the primed-chlorOplast extracts after
15 and 30 min incubation. Three main pausing sites (bands 1, 2 and 3) are apparent:
(approximately 530-, 300-, 150-bp). For each extract, signals obtained from both time
points (15 and 30 min) have approximately the same pattern of distribution.

For the wild type, the most intense signal is in a band of about 530 bp in size (Fig.
4.5, bands 1 and 2), indicating that the strongest pause site is at this position. The relative
intensity of the bands was then quantified using a Phospho Imager, and these values are
depicted in Fig. 4.6. Less radioactivity accumulates in the smaller bands, which also
represent pause sites due to secondary structure.

For the extract from the pm-line, the results are also consistent between the two
time points. Although discrete bands are observed for DNA segments of similar size

between wild type and pm-extracts, the 530 bp band is a less abundant product in the pm-

73

.muofiofi Ea 2&6me 05 E contomow ma gnu 38mm“ cam—8:“ an no 3282
28 Budget So? 32:55 .58 cm Ho 2 Sm @2385 use 8me 5:82 ofi 9 @036 0.83
muoabxo-Em 93 093.232 of. 68368 voucubmoEEméoEma a :o ammonium <29 m4. 05mE

.3 .2 .90. .2

 
 

85 2:5

.380me can
we .wE E :39? £255 28 m .m J 3% 3:3 8 wficaommoboo 818m Son .wowfiwoufi 825 58 cm “a
c2385 Baum Sow 3:850 mfinwmm .8503? $388 :wzohu mmmoflamm <ZQ mo noumomqud 9v .mE

 

com o

‘5‘;
Z

.I‘

- oov

y .. . .....:,.{.\.....}...7w.... .

.5
9!.

n J"- al.

74

.Jt-l 4 ..‘ol - an. ‘-

3 .uhu!?a hWJ.5€.. a
4.

D

.‘r
?
)0
'I

:5

 

 

9.: Ea - - - - 95 2:5]:

 

 

(IA

‘ a , .1f..<)s.

smufigs to qifiuans eAnele .

 

- com

oowr

ooov

75

sample (bandl; Fig. 4.5). More radioactivity is distributed among smaller products. The

quantification is also depicted in Fig. 4.6.

Discussion

. Previous studies have indicated that the plastome mutator of Oenothera causes
elevated mutation frequencies in the chloroplast genome (Epp 73; Sears and Solkalski,
1991). Sequence characterization of mutations recovered from pm-plants have led us to
the hypothesis that replication slippage is the consequence of the defective product in the
plastome mutator phenotype. As discussed in Chapter Two, possible candidates for the
defective product in the plastome mutator include DNA polymerase, SSB, helicase,
topoisomerase as well as proteins involved in an over-expressed S.O.S. system or the
mismatch repair mechanism. In the present report, the in vitro replication systems from
Oenothera chloroplasts of both wild-type and pm-plants were compared and
characterized, in search of differences in the replication machinery of chloroplasts in pm-
plants.

In vitro DNA synthesis is not inhibited by aphidicolin at 20 ug/ml. Since
aphidicolin is a strong inhibitor of nuclear DNA polymerase 0. from both plants and
animals (Sala et al. 1980; Misumi and Weissbach, 1982), the DNA synthesis we observed
should be due to organelle DNA polymerase y. In addition to chloroplasts, the other
source for 7 type DNA polymerase is mitochondria. However, contamination by the
mitochondrial enzyme is unlikely, due to the differential centrifugation used to initially

isolate the chloroplasts.

76

The incorporation of [3H]-d'ITP into acid insoluble materials with our system is
highly dependent on the addition of exogenous DNA, MgClz, and dNTPs, which are
similar to other systems. rNTPs and ATP had slight inhibitory effects on the in vitro
DNA synthesis, consistent with the data from soybean (Hedrick et al. 1993) and maize
(Zimmermann and Weissbach, 1982). In some other systems, both rNTP and ATP have
very little stimulating effect on DNA replication (Tanaka et al. 1984; Wu et al. 1986; de
Haas et al. 1987; Canillo and Bogorad, 1988). These observation indicate that RNA
primers are not necessary to stimulate DNA synthesis when activated DNA is provided as
the template.

Despite the finding a few minor variations of the chloroplast replication extracts
from Oenothera and other plants, no major differences were observed between the wild-
type and pm-plants regarding the aforementioned features of the replication extracts. One
exception is the KCl requirement. As indicated in Table 4.2, the presence of KCl in the
reaction mixtures is more critical for the extract from the pm-plants than for the wild type
extract. In chloroplast replication extracts from pea and spinach (Sala et al. 1980;
McKown and Tewari, 1984), KCl is necessary for high levels of DNA synthesis. In
extracts from liverwort (Tanka et al. 1984), soybean (Hedrick et al. 1993), and the wild
type Oenothera, KCI is relatively unimportant. Observations from other in vitro
replication systems maybe helpful in interpreting these results. Williams and Kaguni
(1995) found that the addition of E. coli SSB lowers the KCl Optimum of DrOSOphila pol
1. Both KCl and SSB contribute to the destabilization of secondary structures, thus

improving the processivity of pol 7. If similar interactions occur for chloroplast

77

replication proteins and DNA, the lower nucleotide incorporation efficiency of pm-
extracts in the absence of KCI may indicate that a defect in SSB exists in pm-
chloroplasts. Although the specific components in the Oenothera chloroplast extracts
were not analyzed in this work, data obtained from similar chloroplast systems, e. g., pea
(Reddy et al., 1994) and soybean (Hedrick et al. 1993; Gordon Cannon and Sabine
Heinhorst, personal communication) have indicated that in addition to DNA polymerase
y, primase, topoisomerase, helicase, and RNA polymerase and many DNA binding
proteins are also present in the replication extracts after chromatography. Therefore, it is
reasonable to suspect that a protein with SSB traits may exist in the replication extracts, a
deficiency in the pm-extracts may make the KCl requirements for DNA synthesis
different from the wild type.

The distribution of radioactive DNA products (Fig. 4.5) was quantified to analyze
DNA synthesis. Most of the products of wild-type extracts results from pausing at the top
band while the products of primer extension from pm-extracts were more broadly
distributed. Thus, DNA synthesis in the wild type-system appears to be more processive
than that with pm-system. This tendency was even more clear when the data were
integrated and quantified. As shown in Fig. 4.6, most of the signals from the wild-type
accumulate at the top band (peaks 1) while in the parallel experiments with pm-material,
signals were evenly distributed along the entire lane. IN repeating this experiments, I
tested equal amount of pm and wild-type specific activities, and still observed the
disparity in the banding patterns. When the amount of replication extract were doubled,

no change was observed in the bands, indicating that the substrates are not limiting, and

78

that the larger bands are not a consequence of reinitiation of synthesis. These data
indicate that the replication extracts from the pm-line have difficulty extending the
primers. This is observed despite the fact that the pm-extracts added to the reaction
mixtures were twice as active in deoxynucleotide incorporation in the initial assay with
calf thymus DNA than were the wild-type extracts. The two differences in the products of
DNA synthesis indicate that either DNA polymerase or SSB may be involved in mutation
activities of the plastome mutator since those two components of the replication
machinery are sufficient for replication of ssDNA phage.

Several lines of evidence in bacteria have linked DNA rearrangements among
tandem repeats to the DNA polymerase. Frequencies of deletion formed between tandem
direct repeats are increased to as much as 100 times in yeast having a temperature-
sensitive DNA polymerase delta and the spectrum of mutations induced is also different
than in the wild-type strain (Tran et al. 1995). The alpha-catalytic subunit of E. coli
DNA polymerase III elevates frameshift mutations dramatically when the exonuclease
subunit is absent and therefore eliminate proof-reading activities (Mo and Schaaper,
1996)

The in vitro chloroplast replication system developed here could be utilized
further to investigate the roles of DNA polymerase and SSB in the mutagenesis process
of the plastome mutator. Data obtained in the present study implicated the chloroplast
DNA polymerase or SSB as potential candidates for the pm defect. However, the
involvement of other accessory proteins in the replication complex could not be

completely excluded since the replication extracts were only partially purified. A closer

79

scrutiny of these individual protein through experiments designed specifically for each of

them is recommended.

Chapter 5

AN IN VIT R0 REPLICATION SYSTEM OF OENOT HERA
CHLOROPLASTS DOES NOT SUPPORT INITIATION
OF REPLICATION AT THE PREVIOUSLY
IDENTIFIED “0R1 ” REGIONS

Introduction

Electron microscopy has been used to observe replicative intermediates of the
plastome from pea (Kolodner and Tewari, 1975), Chlamydomonas (Waddle et a1. 1984),
Oenothera (Chiu and Sears, 1992) and tobacco (Takeda et al. 1992). Based on their
observations of DNA spreads fiom pea and maize, Kolodner and Tewari (1975) proposed
that the replication of the plastome initiates from two displacement (D) loops located
about 7kb apart on complementary strands. As replication proceeds, the two D-loops
move toward each other in a unidirectional manner. After the two D-loops pass each
other, bidirectional semicontinuous DNA synthesis begins. DNA replication terminates
when the two Cairns forks meet at a site approximately 180° from the starting point.
Based on their observation of lariat structures, Kolodner and Tewari proposed that rolling
circle replication takes place subsequently to further amplify the plastome. Recently, both
the Cairns type- and rolling circle- replications have been observed with an in vitro
replication system from tobacco. However, no other electron microscopic reports of DNA

spreads have indicated the presence of lariats. Two pairs of putative replication origins

80

81

(oriA and oriB, Figure 5.1) were mapped in plastomes I and IV of Oenothera by electron
microscopy (Chin and Sears, 1992). Each pair of origins is located in the inverted repeat
region flanking the 16S rRNA gene. The relatively imprecise localization has not yet been
refined, since no in vitro replication system has been developed for Oenothera
chloroplasts.

As an alternative to electron microscopy, partially purified chloroplast replication
extracts from pea (McKnown and Tewari, 1984), Chlamydomonas reinhardtii (Wu et al.
1986; Waddell et al. 1984), maize (Gold et al. 1987), soybean (Hedrick et al. 1993) and
tobacco (Takeda et al. 1992; Lu et al. 1996) have been used to identify chloroplast DNA
replication origins through observation of site specific initiation. Presumably, these
replication extracts contain proteins essential for site specific initiation of chNA
replication, including DNA polymerase and other accessory proteins (Hedrick et al. 1993;
Reddy et al. 1994). In my experiments, I sought to use the newly developed Oenothera
chloroplast in vitro replication system to examine several sites that had been identified as
putative replication origins of the plastome. (Figure 5.1). Two of the putative origins
tested (Figure 5.1A) are located in or near the rRNA operon in Oenothera (Chiu and
Sears, 1992), pea (Meeker et al. 1988 and Reddy et al. 1994), and tobacco (1996). The
other potential origin (Figure 5.1B) maps to the vicinity of the rplI6 gene region in
Chlamydomonas reinhardtii (Wu et al. 1986), soybean (Hedrick et al. 1993) and maize

(Gold et al. 1987), but not in Oenothera.

82

 

23S 16S

 

 

 

 

+ + H + +
TA Pu P103 Ta CA
I I l I
p0j32 p0ri32; pOar2
B
[ rpuo J I rle I rps

 

1 11
CAS M
pOjorZ; pOarl

Figure 5.1 A) Putative origins of plastome replication mapped in the vicinity of the 23S
rRNA and 16S rRNA genes in pea (P,, P“; Meeker et al. 1988), Oenothera (0A, 03; Chiu
and Sears, 1992), and tobacco (TA and TB Liu et al. 1993). Bold black line represents the
inverted repeats of plastome in tobacco and Oenothera . Thin line extended to the left end
represents the small single copy region for tobacco and Oenothera . Despite the fact that
the inverted repeat does not exist in pea, the relative locations of origins are very similar
among the three species in question. Tn and Pa represent of plastome replication from
tobacco and pea, respectively. The location of the 5 kb insert containing oriB in clones
p0ri32, p0ar2 and a 5.5 kb insert countaining partial 23S rRNA gene in clone p0j32,
respectively, are indicated below the ori map. B.) Approximate locations of putative
replication origins relative to ml] 6 genes for Chlamydomonas (Wu et al. 1988), soybean
(Hedrick et al. 1993) and maize (Carrillo and Bogorad. 1988) are indicated by arrowhead.
The region covered by the insert of plasmid p0jor2, pOarl is shown below with a bold
line.

83

Materials and methods

Plant materials Both wild-type and the pm/pm plants are Oenothera hookeri
strain J ohansen with plastome type I. The wild-type Dusseldorf line of strain Johansen is
described in Chapter Two. The pm/pm line is generated from self pollination of a variant
of the Dfisseldorf line of plastome I (field accession number: 96-40). Leaf tip cultures

were maintained on a medimn described by Chiu et a1 (1990).

Plasmid construction and preparation Plasmids used in this study are listed in
Table 5.1. The 5 kb EcoRI fiagment of the 16S rRNA-(ml spacer from the Dusseldorf line
was constructed in pBR328 (Sears et a1. 1996) and transferred into pBluscript‘, where it
was renamed as p0ri32 in the present study. This region contains a putative origin of
replication (orz’ B) identified by electron microscopy (Chiu and Sears, 1990). Plasmid
p0ar2 contains the same region from plastome IV in pBR328 (Sears et al. 1996). Plasmid
p0j32 was constructed in pBR322. It covers the 5’ end of the 23S rRNA region where one
of the replication origins from pea and tobacco were observed (unpublished data, Figure
5.1A). Plasmid pOj108 is constructed in pBR322. It contains the fifth largest BamHI
fragment from the 0. johansen plastome which is similar in size to the insert in p0j32
(unpublished). The 2.5 Kb EcoRI fragment of the rpl16-rpII4-rps8 region from Cornell-1
and plastome IV were obtained as clones in pUC18 as strains p0jor2, pOarl respectively
(Wolfson et al. 1991). The region containing the rplI6 gene has been mapped as one of
the D-loops (ori A) in the plastome of Chlamydomonas reinhardtii (Waddle et al. 1984),
it also supports in vitro dNTP incorporation into acid insoluble material in

Chlamydomonas (Lou et al. 1987; Wu et al. 1986) maize (Gold et a1. 1987; Carrillo and

84

Table 5.1 Plasmids used in site specific initiation experiments

 

Plasmid Features

A. Intergenic region of 16S rRNA-trnI genes
(Putative origin in Oenothera, tobacco and pea)

p0ri32 Plastome I
pBluescript' vector of 3-2
pOar2 Plastome IV
pBR328 vector of B57

B. Intergenic region of 23S rRNA-16S rRNA genes
(Putative origin in tobacco, pea)

p0j32 Plastome I
pOj 108 a random insert of the same size as p0j32
pBR322 vector of p0j32and pOj108

C. The vicinity of rpl16-rpll4-rps8 genes

(Putative origins in chalymydomonas, maize, soybean)
p0jor2 Plastome Cl

pOarl Plastome IV

pUC18 vector of p0jor2 and pOarl

 

85

Bogorad 1988) and soybean (Hedrick et al. 1993). Plasmid DNA was prepared following

the standard mini prep procedure described by Sambrook et a1 (1989).

Isolation of chloroplasts from Oenothera Using a procedure described by Chiu
et al. (1990) Oenothera leaf tip cultures (ca. 80 g) were harvested 7-10 days after being
transferred into fresh medium, and homogenized in 6-10 volumes of homogenization
buffer containing: 50mM Tris-HCl, pH 7.5, 6% sorbitol, 6 mM sodium ascorbic acid,
0.15% (w/v) polyvinyl pyrrolidone (pr), 0.1% BSA, 3mM cysteine and 5 mM 2-
mercaptoethanol. The debris from the homogenate was removed by pouring through
multiple layers of cheese cloth and two layers of Miracloth. Intact cells, starch, lipids and
nuclei were removed from the chloroplasts by differential centrifugation. Plastids were
pelleted at 6000xg for 4 min and subsequently washed with the same basic medium
lacking PVP. The final chloroplast pellets were stored at -80°C for subsequent use. When
the plastid pellet was observed with phase microscopy, <20% of the plastids were
brightly luminescent (i.e. intact). However, 80% of the organelles that were pelleted were
extremely small and not at all luminescent. These are most likely to be composed of

rapidly—dividing proplastids.

Preparation of chloroplast extracts This protocol has been adapted and
modified from the Cannon-Heinhorst Laboratory at the University of Mississippi (
Hedrick et al. 1993). Purified chloroplasts from Oenothera were resuspended in a
minimal volume of buffer A containing: 50mM Tris-HCl, pH 8.0, 100mM (NH4)ISO,,

25% glycerol, 10 mM 2-mercaptoethanol, lmM EDTA, 0.3 mM p-sulfonylfluoride

86

(PTSF), 0.3 mM phenyl methyl sulfonyl fluoride (PMSF), and 1 mM e-aminocaproic
acid. Triton X-100 was added to a final concentration of 2.0 % and the solution was
incubated and stirred on ice for 1 hour. The crude extracts were centrifuged at 9000 rpm
in a Sorvall SS34 rotor for 30 min. The clear supernatant was loaded onto a 10 ml DE52
(Whatman Ltd. England) column pretreated with HCl and NaOH and equilibrated with
buffer A in 0.5M (NH4)ZSO4. Before loading the sample, ten volumes of buffer A with
100 mM (N H02804 was loaded to wash and equilibrate the column. After sample loading
was finished, three volumes of buffer A with 100mM (NH,)ZSO4 was loaded to wash the
column. The replication extracts were then eluted by buffer A with 500 mM (NH4)2SO,.
Fractions were tested for activity using the system described in chapter 4 and activated
calf thymus DNA. Fractions showing DNA polymerase activity were pooled together and
dialyzed against 100 mM (NH.,)ZSO4 buffer A with 0.1 % Triton X-100 and 0.1 mM
EDTA. After dialysis, the replication extracts were concentrated to one tenth volume with
Centricon 10 concentration tubes at 5000 x g (Amicon, Beverly, MA) and stored at -80°C

for subsequent experiments.

DNA polymerase assays. Each reaction for a standard DNA polymerase assay
was performed in a 50 ul volume, containing 50 mM Tris-HCl, pH 8.0, 5 mM MgC12, 125
mM KCl, 39-65 ug/ml activated DNA (DNAse treated calf thymus DNA), 30uM of each
of the four deoxynucleotides with 3H-d'l‘TP at BOO-1000 cpm/pmole, 1 mM DTT and 140
ug/ml BSA. Aphidicolin was dissolved in DMSO at 20ug/ml. The final concentration of

aphidicolin in a standard replication assay is 0.4 ug/ml. The reaction began with the

87

addition of 5 ul of the enzyme sample, followed by the incubation at 37 °C for 30 min.
The quantity of radioactivity incorporated into acid-insoluble material was measured by
transferring the entire reaction to a DEAE cellulose paper (Whatman, DEl l), and
washing once with 0.5 M NazHPO4 for 10 min, H20 for 1 min and 95% ethanol for 1min.
Liquid scintillation counting was performed to measure the precipitated radioactivity after

the filters were dry.

Quantification of protein concentrations Protein concentrations were

quantified by the protein assay kit (BioRad) following the manufacturer’s instructions.

DNA replication initiation assay The standard replication initiation assay was
adapted from that of Hedrick et al. (1993). Each 50 ul assay contained 50 mM Tris-HCl
pH 8.0, 12 mM MgC12, lOOuM each GTP, CTP, UTP, 2mM ATP, 20 uM each dATP,
dCTP, dGTP, 2uM 3H-TTP (1000-3000 cpm/pmole), 0.1 mg/ml acetylated BSA and 20
ng/ul substrate DNA. After the addition of chloroplast extract (1-10ug protein/ul
reaction), the initiation experiments were incubated at 25°C for 1 hour. The quantity of
radioactivity incorporated into acid-insoluble material was measured as outlined in the

DNA polymerase assay.

88

Results

The replication extracts of Oenothera chloroplasts do not support dNTP
incorporation within either the 23S rRNA-16S rRNA-1m] or the rplI6-rpll4-rps8
regions.

Plastome regions examined in this study are shown in Figure 5.1. As explained in
the materials and methods section, two regions were investigated: the vicinity of the 23S
rRNA and 16S rRNA, and the intergenic region of the rp116-rpII4-rps8 genes. In addition
to these areas, the appropriate vector DNA was also used as a control substrate for each
initiation experiment. The replication extracts from both wild type and pm-plants were
assessed for site specific initiation, as described in the materials and methods section.

The ori B site was chosen for the preliminary assays because this region appears
to be one of the replication origins in pea and tobacco, and its complete sequence has
been determined and analyzed (Sears et a1, 1996, Homung et al. 1996; Stoike and Sears,
1997). In my assay, the ONE regions from both plastome types I and IV were examined,
and neither of them stimulate the DNA synthesis activity of the replication extracts
(Figure 5.2; p0ri32 and pOar32).

The S’-end flanking the 23S rRNA gene was mapped as one of the replication
origins in both tobacco and pea, but not in Oenothera Wigure 5.1). Preliminary
observations of Lara Stoike (personal communication) suggested that this area might be a
putative replication origin for the Oenothera plastome. Therefore, plasmid p0j32
containing this region was included in the present study. Additionally, plasmid p0j108
with a plastome insert similar in size to that of p0j32 was used as a control substrate. No

significant level of DNA synthesis was observed in this set of experiments. However, the

89

vector pBR322 control seemed to promote a high level of DNA synthesis in extracts from
both the wild type and pm-plants (Figure 5.2).

The intergenic region of the rpll 6-ribosomal protein Operon is thought to contain
the replication origins in maize (Carillo and Bogorad, 1988), Chlamydomonas (Wu et al.
1988) and soybean (Hedrick et al. 1993), as determined by in vitro analysis. Therefore, I
was interested to test whether this region may function as an origin in the Oenothera
plastome. For this purpose, plasmids were used that carry this portion of the rp116 operon
from the Cornell-1 “plastome (clone p0jor2) and wild type-plastome IV (clone pOarl).
Although the Comell-l line originated from the pm line, our studies have indicated its
chNA is identical to the wild-type in this region. Thus, this clone was used instead of a
much larger clone containing this fragment from the wild-type-I (Dusseldorf line), where
a risk would have existed that a higher background would have been obtained.
Additionally, pUC18, in which the p0jor2 and pOarl were constructed, was used as a
control to observe the background incorporation level. Although the overall DNA
synthesis in these clones seemed higher than all other parallel experiments, no significant
levels of 3H-dTTP incorporation is observed when these results are compared with the

control vector.

Discussion

A partially purified in vitro replication system of Oenothera chlorOplasts was
described in Chapter IV. With similar procedures, in vitro replication systems have been
developed from pea (McKnown and Tewari, 1984), Chlamydomonas reinhardtii (Wu et

al. l986),maize (Gold et al. 1987), soybean (Hedrick et a1. 1993) and tobacco (Lu et al.

90

.8308 $232: 98 flaw—038 2: E “Contact 3 So @350 203 $585qu 63.23820
905250 88% muombxo meadow—mo." 05 53» @8on 203 26 oEmb BaoEwfim 8:252 «953 SmfiozmQ
mcgficoo moan—9:2 283E Bogota .mwombxo :38:qu «85050 05 .«o ooaouomoa 839:3. Nd 03E

 

 

 

 

 
 

 

 

 

 

 

 

 

 

m
n
8 8 9
n b S
o m .m w m. .m w m m. m
m H U a 8 Z R U ..n... Z
L 1 I“! o
.. 8m
.. 89
.. 82
-. 08m
.. 8mm
-- o8m
E I 82 2304

 

c2325 2:83 2a

INdO ‘uoumodsoaul-Hc

91

1996) for the observation of site specific initiation of replication. In these systems, the
presence of DNA polymerase was used as the standard to define the replication activity in
the chromatographic purification steps. However, other replication related proteins are
also present in the extracts. For example, in the replication extracts of soybean
chloroplasts, activities of gyrase, topoisomerase, helicase, and RNA polymerase were
also detected in addition to the DNA polymerase activity (Hedrick et a1, 1993). Similarly,
in the high salt extracts of pea chloroplasts, the fractions containing activities of primase,
helicase and topoisomerase overlap with the ones that have the DNA polymerase activity
(Reddy et al. 1994). Since a similar preparation procedure was followed, the replication
extracts obtained from Oenothera chloroplasts should contain most of the essential
components for site specific initiation. In this chapter, this newly developed in vitro
replication system from Oenothera was used to investigate the in vitro function of
putative origins of Oenothera previously identified using the electron microscope (Chiu
and Sears, 1992), and some potential ori sites that had been recognized in other
organisms (McKnown and Tewari, 1984; Wu et al. 1986; Gold et al. 1987; Hedrick et al.
1993; Lu et al. 1996).

Despite the fact that the in vitro replication system from Oenothera chloroplasts
actively incorporated 3H—dTI'P into acid insoluble materials from primed, single stranded
templates (Table 4.2), it did not support site specific initiation of any of the potential
replication origins (Figure 5.2). Although in vitro replication data does not always
coincide with the in vivo observation (Carrillo and Bogorad, 1988; Hedrick et al. 1993),
it is surprising that no incorporation activity was observed in the oriB region of the

Oenothera plastome (Chiu, 1990). The oriB of Oenothera was mapped to the intergenic

92

spacer of 168-238 rRNA, which is consistent with the oriA of tobacco (Lu et al. 1996)
and oril of pea (Meeker et al. 198 8) (Figure 5.1). Evidence from Chlamydomonas (Wu et
al. 1986) and maize (Gold et a1. 1987) showed that a putative origin of replication
supporting in vitro replication could be delimited to a fragment that is as little as 450 bp.
However, the replication activity was not detected if the essential cis-elements of
initiation were missing. For the oriA fragment of tobacco. both sides of the BamHI
restriction site within the EcoRI fragment containing oriA were needed for significant
replication activity in vitro (Lu et al. 1996). The two clones flanking each side of the
BamHI site did not support in vitro replication alone (Nielsen et al. 1993). Thus, the lack
of incorporation of nucleotides with clones p0ri32 and pOarl of the Oenothera plastome
could be due either to the absence of the essential cis-elements capable of supporting
initiation of replication, or alternatively, the trans-acting factors for site specific initiation
may not have been c0purified with fractions containing active DNA polymerase. Another
possibility is that the accuracy of measurement of D-loop sites in Oenothera plastomes
may have been affected by the use of a small circular size marker ((pl74) in EM work to
estimate the long fragments resulting from restriction digestions (Chiu and Sears, 1992).

The rplI 6—rpII4-rps8 region also did not support in vitro replication. This
observation is consistent with the failure to identify this region as a D-loop in the EM
studies (Chiu and Sears, 1992).

Both the in vivo and in vitro data from pea (Kolodner and Tewari, 1975; Meeker
et al. 1988) and tobacco (Lu et al. 1996) indicated that the two putative origins flank the

23S rRNA (Figure 5.1) while in Oenothera, the two D-loops are located on each side of

93

the 16S rRNA gene. Perhaps using more clones overlapping the rRNA operon can provide
a complete answer for the ONE puzzle of Oenothera. Additionally, a thorough
examination of the kinds of proteins present in the replication extracts is recommended to

assure the reliability of the data obtained from the in vitro replication system of

Oenothera chloroplasts.

Chapter 6

SUMMARY AND CONCLUSIONS

The plastome mutator of Oenothera is capable of causing elevated fiequencies
of non-Mendelian mutations that are ZOO-1000 times higher than spontaneous levels.
Several pm-hot spots have been mapped, and sequence characterization of these pm-
induced RFLP sites showed that all of them occurred in regions that contain direct/
inverted repeats. Subsequently, replication slippage was suggested as being responsible
for the mutations caused by the plastome mutator. Evidence obtained from the insertion
mutations recovered from a poly-A track of several pm-isolates provided further support
for this model. The objective of this dissertation was to examine the mutagenesis process
of the plastome mutator in several ways. One goal was to accomplish more thorough
examination of the types of mutation induced by the plastome mutator and their
frequencies of occurrence in different plastome types. A second goal was the
development of an in vitro replication system to investigate the replication enzymes of
pm-chloroplasts biochemically. The effect of the plastome mutator on plastome copy
numbers was examined using Southern blotting experiments.

Due to the limited resolution of RFLPs, only mutations ranging from 50-500 bp
were recovered from pm-plants in the initial studies (Chiu et a1. 1990). It was unknown

whether the plastome mutator causes a broader spectrum of mutations. Therefore, a more

94

95

sensitive screening technique was needed for a thorough examination of the types of
mutations induced by the plastome mutator. The single stranded conformational
polymorphism (SSCP) technique served this role. Combined with PCR amplification and
restriction endonuclease digestion, the SSCP technique was used to screen six
photosynthetic electron transfer-related genes from 37 pm-plants having chlorophyll
deficiencies. Two mutations were recovered from the psbB gene. One of these was a 4-bp
frameshift insertion, that looked like a typical product of replication slippage. A truncated
protein would be generated due to shifting of the coding frame of the psbB gene, and
hence could have been responsible for the albino phenotype. Another mutation recovered
by SSCP analysis was also located in the psbB gene. It is not surrounded by repeats
sequences, nor by stretches of a single oligonucleotide. Sequence characterization of this
area revealed no features of preferred substrates for replication slippage. Nevertheless,
this line of evidence is consistent with our previous finding that the plastome mutator has
a synergistic effect with NMU, a mutagen that predominantly induces point mutation. It
also pointed to the likelihood that the plastome mutator may have caused more varieties
of mutations than those which we had previously observed. The discovery that the
plastome mutator can cause base substitutions as well as repeat-mediated insertions and
deletions points to a likely defect in a component of the chNA replication machinery.
The plastome mutator of Oenothera carries genotype A and plastome type I. It is
compatible with plastome type II, III and IV to differing degrees. Evidence obtained by
Sears and Sokalski (1991) suggested that the plastome mutator causes a lower frequency
of chlorophyll deficient mutants in plastome IV, but subsequent work in the Sears

laboratory showed no significant differences between plastome type I and IV (Baldwin,

96

1995). In order to carefully assess the impact of the plastome mutator on different
plastome types, previously identified RFLPs were examined among newly-derived pm-IV
mutants. A parallel experiment with plastome I was performed by Lara Stoike, who found
that the oriB region is targeted almost immediately, but deletions at other sites can be
observed only in a second plant generation. With plastome IV, no activity was observed
at the DNA level, although an equivalent frequency of albino sectors was observed. The
finding that the super hot spot “oriB” in plastome I is not affected by the plastome
mutator in the plastome IV background is intriguing. The very strong secondary
structure corresponding to frequent deletions in pm-I is also present in pm-IV. This line of
evidence indicates that factors other than the secondary structures are also involved in
mutagenesis of the plastome mutator. To make the analyses perfectly parallel, the
plastome IV plants need to be carried through a second generation.

Before my project was initiated, all of the pm-induced mutations had been
examined long after they were induced. Because the results pointed to the involvement of
replication slippage, questions arose about the integrity of the chloroplast replication in
the plastome mutator. Therefore, in order to initiate an investigation of the replication
enzymes of pm and to observe the induced events in vitro, the establishment of an in vitro
replication system was necessary.

With slight modification of the system fiom the Cannon and Heinhorst Laboratory
(Hedrick et al. 1993) and some technical improvements specific to Oenothera plant
materials, an in vitro replication system was developed from Oenothera chloroplasts. For
the first time, the activity of such extracts from Oenothera reached a workable activity 5-

10 times higher than the background level. This system is able to incorporate [3H]-dTTP

97

into acid insoluble materials and the incorporation activity was stable for longer than two
hours. In this system, DNA synthesis is dependent on exogeneously added substrates and
nucleotides. MgCl2 is essential for DNA synthesis, but KCI is apparently not critical for
DNA synthesis of an extract prepared from wild-type chloroplasts. However, it is
interesting to note that the presence of KCI seemed important for DNA synthesis in the
pm-system. It has been reported that the addition of SSB to DNA polymerase lowered the
requirement of Drosophila mitochondrial DNA polymerase for KCI. Further
examinations designed specifically for assessment of SSB composition in Oenothera
chloroplasts will indicate whether a defective SSB gene is the plastome mutator we have
sought for.

A primed-template replication strategy was designed to observe processive DNA
replication. Plastome fragments containing previously identified pm-hot spots have been
cloned into phagemids to produce ssDNA as templates for replication experiments. The
processivity of pm-replication extracts is apparently lower than that of the wild-type
extracts. In the experiment with wild-type extracts, DNA synthesis stalled at a strong
secondary structure while the pm extracts stall earlier, at the numerous weaker secondary
structures present in the template. As a consequence, pm-DNA polymerase appears to
have difficulty processing DNA replication. A primed-template replication system needs
two main proteins: DNA polymerase and SSB. Since DNA polymerase tends to stall at
the secondary structures if SSB is not present at the replication fork (William and Kaguni,
1995), the discovery that pm-replication extracts is less processive indicates that either

the DNA polymerase or the SSB is not functioning well in the pm-extract.

98

To investigate the role of the plastome mutator on copy number control, copy
numbers of plastomes were compared between the wild-type and pm-plants. As shown in
appendix B, my preliminary data revealed insignificant differences of plastome copy
numbers between the wild-type and pm-plants. Evidence obtained from this experiment
indicated that copy number of the plastome in pm-plants might not be affected by the
plastome mutator. In prokaryotes, the prevalent mechanism controlling copy numbers is
through the initiator protein or less often, through other replication related proteins, as
discussed in appendix A. Since plastome copy numbers are normal in pm-plants, we
suggest that the initiator protein of plastome replication is not a candidate of the plastome
mutator .

Studying the mechanism of mutagenesis of the plastome mutator of Oenothera
has been a long term project in our laboratory. My thesis research has enhanced our
understanding of the plastome mutator in several ways. From a molecular aspect, the
spectrum of the types of mutations induced by the plastome mutator was found to be
broader than what was previously known. Other data obtained from biochemical analyses
indicated that the plastome mutator is likely to involve a defect in a chloroplast DNA
polymerase or SSB. These observations suggest a future direction for the study of the
plastome mutator. Experiments should focus on assessing the functions of these two

proteins in pm-plants.

APPENDIX A

EFFICIENCY OF IN VIT R0 REPLICATION USING DOUBLE-
STRANDED AND SINGLE STRANDED TEMPLATES.

In order to start DNA replication, an initiator recognizes the replicator (origins)
(Huberman 1995). With help from other accessory proteins at the replication fork, a
primer or nick is made, then replication proceeds. A defect in any of these proteins would
render the replication mechanism malfunctional. As explained in Chapter 2, evidence
obtained in our laboratory has suggested that the plastome mutator is involved in DNA
replication in pm-chloroplasts. To be able to characterize the plastome mutator defect
biochemically, I hoped to assay the frequency of errors using an in vitro replication
system. Defects in a helicase or topoisomerase would be visualized with a ds-template,
while defects in a single stranded binding protein would be observed using a ss-template.
Thus, a comparison of DNA replication with ds- and ss-templates would facilitate a better
understanding of the accessory proteins involving DNA replication.

In this section, a heterologous primed-template replication system consisting of
the replication extracts from soybean chloroplasts and the substrates from clones
containing fragments of the Oenothera plastome are described. The goal was to use the
abundant soybean material to prepare a chloroplast replication extract to test the potential
substrates for the in vitro replication system that I was going to develop. The intention
was to test both dsDNA and ssDNA templates for their utility in assessing properties of
the chloroplast replication extracts. Several attempts have been made to overcome the

thermodynamic barrier of double stranded DNA lacking an origin of replication.

99

100

Materials and methods
The details of the preparation of the soybean extracts and the conditions of DNA

polymerase assays described in Hedrick et al. (1993).

Isolation of chloroplasts from soybean cells During my three-month visit to Dr.
Gordon Cannon’s Laboratory at the University of Southern Mississippi, the SB-l cell line
of Glycine max (L.) Mer. V. Corsoy was incubated photomixotrophically on KT medium

(Horn et al. et al. 1983), and harvested for chloroplasts (Hedrick et al. 1993)

Results:

Impact of varying ratios of primer/ds-template on replication activities were
tested at room temperature with undenatured dsDNA

To start testing the possibility of using a primed-dsDNA replication system,
undenatured dsDNA was used as a substrate. The template used in this test contains the
2.5 kb EcoRI fragment of the rplI 6-rplI4-rps8 gene region from plastome IV in pUCl8
(Wolfson et al. 1991). The primer B827 is complementary to the 3’end of the rplI 6 gene.
0.5 ug of the template was used in each assay, with the molecular ratios of
primer/template adjusted to 10:1 and 100:1 accordingly. A summary of the results from
this test is listed in Table A.l, including controls. Even a 10-100 fold excess of primers to

template molecules does not enable priming for the replication of this dsDNA template.

 

100

Materials and methods
The details of the preparation of the soybean extracts and the conditions of DNA

polymerase assays described in Hedrick et al. (1993).

Isolation of chloroplasts from soybean cells During my three-month visit to Dr.
Gordon Cannon’s Laboratory at the University of Southern Mississippi, the SB-l cell line
of Glycine max (L.) Mer. V. Corsoy was incubated photomixotrophically on KT medium.

(Horn et al. et a1. 1983), and harvested for chloroplasts (Hedrick et al. 1993)

Results:

Impact of varying ratios of primer/ds-template on replication activities were
tested at room temperature with undenatured dsDNA

To start testing the possibility of using a primed-dsDNA replication system,
undenatured dsDNA was used as a substrate. The template used in this test contains the
2.5 kb EcoRI fragment of the rplI 6—rplI4-rps8 gene region from plastome IV in pUC18
(Wolfson et al. 1991). The primer BS27 is complementary to the 3’end of the rplI 6 gene.
0.5 ug of the template was used in each assay, with the molecular ratios of
primer/template adjusted to 10:1 and 100:1 accordingly. A summary of the results from
this test is listed in Table A. 1 , including controls. Even a 10-100 fold excess of primers to

template molecules does not enable priming for the replication of this dsDNA template.

101

Table A.l DNA synthesis activity of different primer/template ratio at room
temperature

 

Relative concentrations DNA synthesis activity (%)
primer template

10 l 1

100 1 1

100 H20 1

H20 1 ' 1

H20 H20 (blank control) 1

activated DNA (DNase digested calf thymus DNA, 3ug) 100

 

Impact of template denaturation on replication activities

In ssDNA sequencing reactions, samples are heated up to 65°C, followed by slow
cooling to anneal the primer and the template. To apply this approach to my assay,
samples were denatured at 70°C for 10 minutes in the presence of primers, and slowly
cooled to room temperature for 30 min before the replication experiments. In addition to
the primer/template amounts used in the first set of experiments, a ratio of 1000/1 was
also used. The results of replication activity for each experiment are summarized in Table
A2.

The replication activity was slightly increased in the experiments in which the
ratio of primer/template was 100/1. However, it is insignificant compared to the level of
replication when activated DNA was the substrate of replication (Table A2). The
decrease of template concentration did not affect the replication efficiency. Heating

samples to 7 0°C before replication did not seem helpful for initiation of replication.

102

Table A.2 DNA synthesis activity of different primer/template ratios at 70°C

 

Relative concentrations DNA synthesis activity (%)
primer template

10 1 1
100 l 2
10 0.1 1
10 0.01 2
10 0.001 1
100 H20 1
H20 1 1
H20 H20 (blank) 1
activated DNA (3 ug) 100

 

Impact of boiling and alkali treatments of double stranded template on replication
activity

Two clones and two primers were used for the following experiments. Both
clones pOarl (Wolfson et al. 1991) and pOjr2 (Sears et al. 1996) were described in
Chapter 5. Primer BS49 is complementary to the 3’ end of the 16S rRNA gene in the
clone B54. The universal forward primer is at the 5’ end of the multiple cloning sites of
pUC18 (Yanisch et al. 1985). This primer is used for annealing to clone pOarl. The
boiling and alkali treatments that are both known to denature dsDNA (Sambrook et a1.

1989), were tested.

A) 100°C treatment
Samples of template and primer were denatured at 100°C for 10 min, slowly

cooled to room temperature for another 30 minutes, then left on ice before being added to

 

103

Table A.3 DNA synthesis activity of different primer/template ratios after 100°C
treatment

 

Relative concentrations DNA synthesis activity (%)
primer template

pOarl 10 1 2
100 1 2
100 H20 2
H20 1 2

pOar2 10 1 2
100 1 2
100 H20 2
H20 1 2
H20 H20 (blank control) 1
activated DNA (3ug) 100

 

the replication reaction. The results of replication activity obtained for each experiment
are summarized in Table A.3. This 100°C boiling treatment did not boost replication

activity.

B. The alkali treatment and the primed-ssDNA replication system

The alkali treatment renders dsDNA susceptible to DNA polymerase in
sequencing reactions (Lim and Pene, 1988). Therefore, it was chosen to test the materials
used in the present study. Templates were denatured in 0.2M NaOH and 0.2mM EDTA at
37°C for 30 minutes, precipitated by NaOAc and ethanol, then resuspended for use in the

replication reaction. Because the double stranded templates are denatured into two

104

interlocked ssDNAs after the treatment, the ratios of primer/template were reduced to 1/1
and 10/1 accordingly. The levels of nucleotide incorporation in each experiment are
summarized in Table A4. The replication levels were slightly increased in the 10/1
(primer/template) reaction for both samples, indicating the initiation of replication after
annealing of synthetic primers on alkali-denatured dsDNA.

In the last set of preliminary tests of this replication extract, the M13 ssDNA and
the -40 primer adjacent to the cloning site (U SB) were used for the replication study. The
level of replication of this primed-ssDNA system was 10—15 times higher than all other

treatments tested (Table A4).

Table A.4 DNA synthesis activity of different primer/template ratios with alkali
treatment

 

Relative concentrations - DNA synthesis activity (%)
primer template

pOarl 1 1 3
10 1 4
10 H20 2
H20 1 2

pOar2 1 1 3
10 1 6
10 H20 2
H20 1 2

M13 1 1 29
1 0 3
O 1 3
H20 H20 (blank) 1

activated DNA (3ug) 100

 

105

Conclusion

Double stranded DNA is an inert substrate for replication unless it contains the
origin of replication (Kornberg, 1991). Therefore, single stranded DNA is usually the
template of choice for biochemical assays of DNA replication when the initiation at the
origins is not the focus of study or when DNA synthesis on the lagging strand is under
study (O’Donnell and Kornberg, 1985; Nielsen et al. 1991; Willams and Kaguni, 1995;
Tuteja et al. 1996). In this report, several attempts have been made to break the
thermodynamic barrier of native dsDNA. Nevertheless, data obtained from the tests still
suggested that ssDNA is a better substrate for annealing of an in vitro-synthesized
oligonucleotide to prime DNA replication. Thus, the primed-ssDNA replication system

has been the system of choice for the research performed in Chapter 4.

105

Conclusion

Double stranded DNA is an inert substrate for replication unless it contains the
origin of replication (Kornberg, 1991). Therefore, single stranded DNA is usually the
template of choice for biochemical assays of DNA replication when the initiation at the
origins is not the focus of study or when DNA synthesis on the lagging strand is under
study (O’Donnell and Kornberg, 1985; Nielsen et al. 1991; Willams and Kaguni, 1995;
Tuteja et al. 1996). In this report, several attempts have been made to break the
thermodynamic barrier of native dsDNA. Nevertheless, data obtained from the tests still
suggested that ssDNA is a better substrate for annealing of an in vitro-synthesized
oligonucleotide to prime DNA replication. Thus, the primed-ssDNA replication system

has been the system of choice for the research performed in Chapter 4.

APPENDIX B

THE PLASTOME MU TA TOR OF OENOT HERA DOES NOT AFFECT
PLASTOME COPY NUNIBERS

Rationale

In wheat (Day and Ellis, 1984), spinach (Lawrence and Possingham 1986), barley
(Baumgartner et al. 1988) and rice (Cuzzoni et al. 1995), plastome copy number changes
in response to different physiological conditions and developmental stages of the plant
(Lamppa and Bendich, 1979; Lawerence and Possingham, 1986). It has been postulated
that templates for replication are randomly chosen, and new chNAs are partitioned into
new chloroplasts in a random fashion, regardless of the genotype (Berky, 1994).

Several lines of genetic and physical evidence indicate that plastome copy number
control is “relaxed” rather than “strict” (reviewed by Birky, 1994), similar to the
maintenance of high c0py number plasmids in prokaryotes. In prokaryotes, the prevailing
mechanism for control of plasmid copy number is through inhibition of initiator protein
binding to the replication origin (reviewed by Khan, 1996). However, other studies point
out that mutations in the replication complex also can impact copy number control. For
example, a DNA polymerase I mutation in Rhizobium leguminosarum reduces the levels
of resident plasmids (Crank et al. 1994).

As discussed in previous chapters, replication slippage has been postulated as the
mechanism responsible for the plastome mutator activity (Sears et al. 1995 ; Chang et al.

1996; Stoike and Sears 1997). Several proteins that are involved in movement of the

106

 

107

replication fork have been considered as potential candidates for the plastome mutator
product, e.g. DNA polymerase, helicase, topoisomerase, and SSB (Chang et al. 1996).
Investigations with other organisms have indicated that defects in these proteins may not
only affect replication fidelity, but may also impact replication efficiency (Meyer et a1.
1979; Nowicka et a1. 1994; Vincent et al. 1996). Therefore, I sought to investigate the
impact of the plastome mutator on chNA replication, by comparing the relative amount

of chNA in wild-type and pm-plants.

Materials and Methods

Plant materials The origin of the plastome mutator and plant materials used in
this study have been described in Chapter II. The pm/pm plants were obtained by self
pollination of albicans-Johansenpm heterozygote carrying plastome type IV. Due to the
pollen-lethal allele carried by the albicans strain and the megaspore competition between
the Johansen and albicans strains, 99% of the progeny are homozygous for the pm allele.
The pm/+ heterozygote background was obtained by crossing the pm/pm plants from the
94 field season (field accession 94-23) to the +/+ Johansen strain with plastome type IV
(field accession 94-4). Because the plastome mutator of Oenothera creates some
pigmentation deficiencies, it was conceivable that these phenotypes (rather than the
nuclear genotype) could have a secondary effect of altering the chNA content in the
chloroplasts. Thus, leaf samples were carefully chosen, to enable all possible
comparisons. In addition, because plastome copy numbers change in different
developmental stages, leaves were harvested at the same size (10 cm) and age (4 weeks

old). The one exception is the sample pm50-2, which was from a middle size (5cm)green

 

 

 

 

108

plant, purposely sampled to observe the developmental effect. In total eight samples were
harvested, each representing a unique combination of nuclear background and phenotype.
These include one wild-type plant 95-4 (1), two pm/pm plants: pp93 (green) (2), pp92
(variegated) (3), and five pm/+ plants: pm50-4 (green) (4), pm50-3 (light green) (5),

pm50-2 (green) (6), pm50-1 (variegated) (7), pm69 (variegated) (8).

Preparation of nucleic acids. Extractions of total plant DNA were performed as
described by Fang et al. (1992). Half of the total nucleic acids from each sample as then
subjected to further purification by caylase digestion as follows (Rether et al.1993): DNA
samples were dissolved in a solution (500ul) containing 50mM KoAc, lOmM EDTA, 5
g/ml RNAse and 0.5 mg/ml Caylase M3, then subjected to overnight incubation at 37°C.
After the caylase digestion, the pH of the samples was increased by adding 1M Tris-HCl
(pH 8) to a final concentration of 50mM, followed by the conventional procedure of
phenol/chloroform extraction and alcohol precipitation. However, ethanol was replaced

by ethoxyethanol to remove digested polysaccharides more efficiently.

Southern blotting and quantification of chloroplast DNA content. Southern
blotting was performed following the conventional procedure described by Maniatis et al.
(1982), with 3’zP-labeled dATP from New England Nuclear/Dupont. Two clones
containing DNA fragments from Oenothera were used as probes for blotting. The
plastome probe contains a 2.4 kb fragment of the petA gene, (Johnson and Sears, 1990)

and the nuclear probe has a random nuclear genomic insert of 8 kb (Sears et al. 1997).

109

Approximately the same amount of sample (lug/each sample) was loaded twice onto
each of two separate gels, and hybridized with the nuclear and chloroplast probes (0.5
ug/each probe) respectively. After hybridization, the blots were exposed to the storage
phosphor screen (Molecular Dynamics) at room temperature. Signals on the gel were
quantified using Image Quant Program (Molecular Dynamics), with data manipulation

through Microsoft Excel.

Results
Southern blotting results are shown in Figure B1 (no caylase treatment) and
Figure B2 (caylase-treatment). Signals obtained from these two gels were calculated and

drawn using Excel software as displayed in Table l and Figure 3.

Measurements of plastome content in the “no-caylase treatment”
experiments. Southern blotting results are shown in Figure 1A (gel A) and Figure 1B
(gel B). Eight samples were loaded twice on each gel, providing two replicates. Gel A
was hybridized to the chloroplast probe while the gel B was hybridized to the nuclear
probe. As displayed in Table 1, signals from the two replicates (set 1 and set 2) on each
gel were listed and averaged. The averages obtained from the chloroplast blot (C) were
divided by the averages obtained from the nuclear-blot (N) to normalize the quantity of
hybridization on a per-cell basis (C/N). No pronounced variation in plastome contents
was observed among the samples, except that sample 3 (variegated plant pp92) has a
lower C/N ratio (3.75) compared to all other samples that together have an average of

5.35.

110

A
12345678 C 12345678

...:: m . "a. «I

B

12345678 C12345678
WW ""5“ 33““ “W

   

Figure B.1 Southern blot of undigested DNA hybridized with (A) chNA
probe and (B) nuclear probe. Total DNA was extracted by the conventional
method, and no caylase treatment was performed. Duplicate samples of total
DNA were loaded into two separate gels, then hybridized with (A) a plastid
probe and (B) a nuclear probe respectively, as described in materials and
methods. Lanes contain the following samples: one wild-type plant (1) 95-4;
two pm/pm plants: (2) pp93 (green), (3) pp92 (variegated); five pml+ plants:

(4) pm50-4 (green), (5) pm50-3 (light green ), (6) pm50-2 (green and middle
size), (7) pm50-1 (variegated) and (8) pm69 (variegated). C (control): the same
probes used in each experiment were loaded into the agarose gels to use as a
control for hybridization.

111

.>H 093 68833 “62:8 853 =< .303on3
En cc €882.33 25.5 S .38 2%? Es :8qu SEE. @ A :86 :35 225. E .EBE Ewen
E ”$ch iSm o>m Avofiwoific mafia. Amv A583 mama. ANV ”3233 33:5 95 £13 CV Eda 090. 25 one ”Bozo.“
mm mam—9:5 “5.0.862 Amv 98 Geo 228m 5 £385: ofimfiwm .mEoEComxo cowwobéwflbwo 05 80¢ 8:58 £385:
Esoam Amv Locum #58365 ems—>8 o: HEB 83 05 8on onEHoHow 968 93 AZV ADV go 833 mm: A<V Locum .968
Enema $205: 6.3» ADV Enema “Emma 05 wEEZe 3 35328 203 £32 <75 233m mo 32%, 322250: 2:.
.womfigm 28 gm: 82$ 3w 258 2: mo 8:823“: 95 2t 8on wommmfio mfiqmmm .mEtoE Eonusom 3 @83on mm

96 320:: 93 ADV mwsmfim 2: mo Eofioo <75 Am: 98 2V 35$ .mfiaEéQ E 26526 mo EoEoo <75 mm 2an

112

ofw
wow
00.0

ow.“g
no.5
No.0
«5.0

A79\n:

uwa‘
omgw
mmxw
meg“
wugw
mngw
omgw
veg“

AVA\H¢

¢N¢vmw
mammmm
wawmmm
;%g4
woamfim
mwumom
mommwm
ummnoa

wmmum><

omovmo
«mofiov
oNoon
omowm
¢No¢mm
wmmoom
hwmwwu
owwwmm

MWMIIII

mmnmom
Nmfimm¢
owomcm
;«44
nomovm
ommmwm
onvvw
mwoauw
Numm

 

g 329:2

cmuoao
ommmmv
Nofiomm
omowm
ooowmv
mmmmmm
mmmwvw

wHHonv
mmummm
vafiwm
;$g4
wmowwm
vmmwmm
a¢mm~m
mvfiwvm
Humm

 

oaomou
mamawv
nuomvm
:<4z
Nmoomm
ONNmNm
ovmwwo

moauwoa owmomm

 

um>< N umm

i 3232

 

H 3m

om¢owom
cmmmmmv
muvmovm
JQQA
ammmmma
wmmwoom
nommuhw
vawHHo

mwauw><

aooowmm
Nawvv¢m
omommum
wmmoma
¢5NNONN
nomammfi
ovwomov
wmomwov

mwmuw><

 

 

cmmwuwm wumuomm
nmmomwv mofimmmv
Hfiwwmwm wmmmuwm
.<:7H .«a7g
ummummfi Hmwmmwfi
ovoumwm Hoofimmm
ammnofim vamwwmw
mwmmamo Nofiw¢wm
N 2% H :5
6V “£30820
ofiouoom wfiummom
ouwmfioa mmfiuumm
aomnumfi mofiwowfl
mmmmfim mocvufi
«wmmfiwm «umfifiufi
QV¢mamH “mucoflfi
mmnoomv mmamwvm
wmuooom wmovomm
N «mm H uwm
6v “£38020

 

 

{an Asmoam
iamérémam
Em Afiwvmémfim
\Sm Amomémam

\sm ‘ETomam
Em\fim A>vmomm
Em\6m vammmm

+\+ .3313
mmboamw dEmZ

:2 €38
iamfrémam
Em Afiwvmémfim
>3 chmémam

\Em ‘vavomam
EACEQ Atmmmm
Em\8m vamamm

+\ + $3-3
mmxuoamw \mEaZ

musflmufim E mvwmaa mo 33:8 <29 fim 3an

HNCO‘d‘lOQBw

HNCOfi‘LOOBUJ

113

Measurements of plastome content in caylase-treated samples The Southern
blots displayed in Figure B. 2 were handled exactly the same as described above except
that the DNA had been treated with caylase. The signals obtained from both gels were
integrated and are listed in Table 1. The green tissue samples have similar plastome
contents in all nuclear backgrounds. Except that the green plant pm50-4 shows a lower
plastome content (C/N: 4.8) relative to all other green samples (average C/N: 7.36).
Overall, plastome content seems higher in variegated plants, (pm50-1, pm69 and pp92)
compared to the green plants (pp93, pm50-4 and pm50-2). In other words, the main
differences among them depend more on their phenotypes (variegated vs. green), rather
than the nuclear backgrounds (+/+ vs. pm/+ vs. pm/pm) from which these materials are
sampled. The relative differences among samples are more obvious in the caylase-treated

samples than in these that were not treated with caylase,

Discussion

A defect in the replication complex has been proposed to cause elevated
frequencies of mutations in pm-plants (Sears et al. 1995; Chang et al. 1996). Thus, the
impact of the plastome mutator on plastome replication was investigated through
comparing the copy numbers of plastomes between the wild-type and pmeplants.

In this experiment, one set of samples followed a standard procedure for DNA
extraction in the Sears Laboratory, while a replicate set was treated with caylase in
addition to the conventional method of DNA extraction. Without the additional caylase

treatment, no obvious differences of chNA levels were observed in various nuclear

114

A

42163785 C 42163785

,,- ,. . ‘ , ."_ “if," '1 \
.. .. .V .‘ ‘. . . n. -..
e." -. watt” ““r'ff'" ’57»;
I . : v.“ r ,
r ' :u ;_ , --
b . - . .

   

.4216-3'785 C42163785

   

Figure 82 Southern blot of undigested DNA with (A) chNA probe and (B)nuclear
probe. Total DNA was extracted by the traditional method, followed by caylase
treatment. Duplicate samples of total DNA were loaded into two separate gels, then
hybridized with (A) a plastid probe and (B) a nuclear probe respectively. Lanes contain
the following samples: one wild type plant (1) 95-4; two pm/pm plants: (2) pp93 (green),
(3) pp92 (variegated); five pm/+ plants: (4) pm50-4 (green), (5) pm50-3 (light green), (6)
pm50-2 (green and middle size), (7) pm50-1 (variegated), (8) pm69 (variegated), and C
(control). The same probes used in each experiment were loaded into the agarose gels to
use as a control (C) for hybridization.

115

backgrounds, except for one variegated plant pp92, in which lower plastome copy
number was obtained. In the caylase-treated samples (Figure B.3), the chNA copy
numbers in variegated plants tended to be higher than in the green lines, regardless of the
nuclear background. The lowest chNA content was in a green pm/+ plant pm50-4,
which is inconsistent with the data from the set that was not treated with caylase. The
different results between samples that were not treated with caylase that were otherwise
replicates indicates that the variations were probably not due to the unusual physiological
condition of the plant line in use, e.g. altered plastid numbers per cell (Epp and
Parthasarathy, 1987). Rather, the effect of caylase treatment or handling errors maybe
responsible for the differences.

Carbohydrates that are isolated along with the nucleic acid preparations from
Oenothera pose an obstacle for accurate pippeting and quantification of DNA. Digestion
of the carbohydrates with caylase (Rether et a1. 1993) appeared to be a procedure that
would improve our purification of DNA. The pippeting of DNA samples becomes much
easier due to the reduced viscosity of samples after caylase treatment. In this report, the
effect of caylase treatment was tested with the Southern blotting experiments. The
relative strengths of signals among samples were stronger in the caylase-treated samples
than in the non-treated samples. The divergent results obtained from the different
treatments suggest that caylase treatment may have unmasked the interference caused by
large amount of carbohydrates present in the DNA preparation. However the possibilities
that the observed effects of the caylase treatment originated from pippeting errors or other

handling mistakes during sample preparation, can not be excluded.

116

The evidence displayed here is still preliminary. Despite the variable results
gained fiom the two sets of Southern blots, neither of them reveal any pronounced
distinction of chNA copy numbers among different nuclear backgrounds. Therefore, it
seems unlikely that the plastome mutator of Oenothera plays a direct role in copy-

nurnber control of plastomes.

APPENDIX C

CLONING OF PM-HOT SPOTS INTO THE PHAGEMID
PBLUESCRIPT SK

Introduction

Due to the uncertainty of the precise location of the replication origins, a primed-
replication reaction was designed to study the replication machinery of Oenothera
chloroplasts. In E. coli, double stranded DNA is an inert substrate for primed-replication
(Kornberg, 1991), and similar results were observed in my chloroplast replication extracts
(Appendix A). Therefore, a primed-ssDNA replication was developed as a system to
study DNA replication of an extract prepared from Oenothera chloroplasts. The ultimate
goal for this cloning project was to observe replication slippage events in vitro.

Several fragments containing previously identified pm-hot spots were cloned into
the phagemid pBlueScript SK for further usage. Stretches of tandem repeats that were
designed based on the preferred targets of the replication slippage mechanism, were
synthesized in vitro and inserted into the 3’end of the rbcL gene of Chlamydomonas
(Sears, unpublished data). The 1kb fragment containing this rbcL end and part of the 5’-

end of ath gene, was also cloned into the phagemids for further manipulation.

Materials and Methods
Vectors used are pBluescript SK (+/-) (Short et al. 1988) derived from pUC19.

The pBluescript SKs contain a replication origin of f1 filamentous phage. Depending on

117

118

the orientation of the f1 origin, pBluescript SKs are designated as (+) or (-). With the
infection of helper phage, the (+) strain allows the recovery of the sense strand of the lacZ
gene while the (-) strain produces the antisense strand. E. coli strain XLl-Blue (Bullock et
a1. 1987) was used as the host for constructions. In all cases, the inserts were transferred
from other constructed clones containing replication slippage targets. Based on the known
physical maps of these previously constructed clones, restriction enzymes were selected
and the digested fragments were separated by electrophoresis on agarose gels. The
fragments of interest were then extracted fi'om the gel. with Qiaex (Qiagen Inc.,
Chatsworth, Calif), followed by cloning using conventional methods (Sambrook et al.

1989)

Results

Cloning results are summarized in Table C. 1. Ten clones were obtained. All of the
insert fragments contained previously recognized pm-hot spots (Chang et al. 1996), or in
vitro designed replication slippage targets (Sears, unpublished data). The phagemids
pBlueScript SK (+) and (-) were used as the cloning vectors. ssDNA produced from
phagemid has the advantage of being stable, and can be easily maintained within bacteria
(Sambrook et al. 1989). Additionally, the two transcription promoters- T3 and T7
flanking both sides of the multiple cloning sites of the phagemid pBluescript SK may be
of use to produce RNAs from one strand or the other for future research. Therefore, (that
vector was chosen for this cloning project. As a record of all of these clones, their

features and origins are described in this appendix.

 

119

Table C.l Phagemids constructed for propagation of ssDNA

 

Insert (origin, size, flanking restriction

 

 

Name“ sites) & vector [(+) or (-)] of use feature of interest
pORIF rRNAs spacer; + hypervariable region in the
pORIR 5kb; EcoRI + intergenic region of

1 6S rRNA-trnl
pRPF rpll6-rpll4-rps8 spacer; + 2 copies of a 29-bp repeat
pRPR 2.5 kb; EcoRI + oligo-A stretches
p2280F orf2280 coding region; + one set of 8 tandem 24-bp
p2280R 450 bp; RsaI-HincII - repeats
pRB3F interruppted coding region + in vitro synthsized multiple
pRB4F of rbcL gene; 1kb; EcoRI- + copies of CI I I (preferred-
pRB4R EcoRV - targets of replication
pRB6F - slippage)

" The direction designated for clones. F-forward, R-reverse corresponds to gene
transcription for 16S rRNA (Sears et al. 1996), rpll6 (Wolfson et al. 1991), and orf2280
(Blasko et al. 1988) genes of Oenothera plastome and rbcL gene of Chlamydomonas
plastome (Goldschmidt—Clermont and Rahire, 1986) respectively.

pORIF+ and pORIR+ were both constructed in pBlueScript SK (+), with the
same inserts arranged in opposite directions. The insert fragment was from the
hypervariable region between the 16S rRNA and trnI genes from the Dusseldorf strain
containing plastome I (Sears et al. 1996). When the host E. coli is infected with the helper
phage R408, the sense strand of 16S RNA gene could be generated from pORIF+ while

the antisense strand could be produced from pORIR+.

120

pRPF+ and pRPR+ were both cloned in pBlueScript SK (+). They have the same
inserts but in different directions. The insert DNA was from the intergenic spacer region
of rplI6-rpII4-rps8 of the Cornell-I variant of plastome I (Figure 5.1B; Wolfson et al.
1991). The ssDNA propagated fi'om pRPF is the sense strand of the rpII6 gene while the
ssDNA from pRPR is the antisense strand.

p2280F+ and p2280F- were cloned in pBlueScript SK (-), (+) respectively, with
the insert containing one set of 8 tandem repeats in orf2280 from plastome I (Blasko et a1.
1988). With the infection of the helper phage R408 to its host E. coli, the p2280F+
produces the sense strand of the orf2280 while the p2280R— produces the antisense strand.

The inserts in pRB3F+, pRB4F+, pRB4F- and pRB6F- are 1kb fragments which
cOntain part of the S’end of the rbcL gene and part of the 3’ end of the ath gene from
plastome of Chlamydomonas. The reading frame of the rbcL gene was interrupted by
several stretches of tandem repeats, that had been synthesized in vitro and inserted at a
BspMI cut site (Sears, unpublished data).

M is cloned in pBlueScript SK (+), it contains a monomer insert of 5’-
gtttCTTTCTTTCTTTCTTTCTTT-3’, at the S’end of the rbcL gene. Propagation of
pRB3R would generate the sense strand of the rbcL gene.

DRB4R-, DRB4F+ were cloned in pBlue Script SK (-), (+) respectively. They
contain two units of the same insert in pRB3F+ at the 5’end of the rbcL gene. The sense
strand of the rbcL gene can be produced when the host strain of pRB4F+ is infected with
the helper phage R408 while the antisense strand would be produced from pRB4F-.

M was cloned in pBlueScript SK(-). It contains a similar repeat sequence at

the S’end ofthe rbcL gene. The repeat is 5’-GTTTCTTTCI I'ICI I IC'I I ICI I ICI I I-

121

3’. With the help from the helper phage R408, the antisense strand of rbcL gene could be

generated from pRB6F-.

 

 

LIST OF REFERENCES

Albertini AM, Hofer M, Calos MP, Miller JH (1982) On the formation of spontaneous
deletions: the importance of short sequence homologies in the generation of large
deletions. Cell 29: 319-328.

Baumgartner BJ, Rapp J C, Mullet JE (1989) Plastid transcription activity and DNA copy
number increase early in barley chloroplast development. Plant Physiol 89: 101 l-
101 8.

Baldwin S (1995) MSc Thesis, Michigan State University, East Lansing, MI.

Bendich (1987) Why do chloroplasts and mitochondria contain so many copies of their
genome BioEssays 6: 279-282.

Bendich AJ, Smith SB (1990) Moving pictures and pulsed-field electrophoresis show
linear DNA molecular from chloroplast and mitochondria. Curr Genet 17:
421-425.

Berky CW, Jr. (1994) Relaxed and strigent genomes: why cytoplasmic genes don’t obey
Mendel’s Laws. Journal of Heredity 85: 355-365.

Blasko K, Kaplan SA, Higgins KG, Wolfson R, Sears BB (1988) Variation in copy
number of a 24-base pair tandem repeat in the chloroplast DNA
of Oenothera hookeri strain Johansen. Curr Genet 14: 287-292.

Boffey SA and Leech RM (1982). Chloroplast DNA levels and the control of chloroplast
division in light-grown wheat leaves. Plant Physiol. 69: 1387-1391.

Bramhill D, Kornberg A (1988) Duplex opening by dnaA protein at novel sequences in
initiation of replication at the origon of the E. coli chromosome. Cell 52:
743-755.

Carrillo N, Bogorad L (1988) Chloroplast DNA replication in vitro: site specific initiation
from preferred templates. Nucleic Acid Research 16: 5603-5620.

Chang T-L, Stoike LL, Zarke D, Schewe G, Chiu W-L, Jarrell DC and Sears BB (1996)
Characterization of primary lesions caused by the plastome mutator of Oenothera.
Curr Genet 30: 522-530.

Chiu W-L, Stubbe W, Sears BB (1988) Plastid inheritance in Oenothera: Organelle
genome modifies the extent of biparental plastid transmission. Curr Genet 13:

1 81-1 89.

Chiu W-L, Johnson EM, Kaplan SA, Blasko K, Sokalski M, Wolfson R, Sears BB (1990)
Oenothera chloroplast DNA polymorphisms associated with plastome mutator
activity. Mol Gen Genet 221: 59-64.

Chiu W-L and Sears BB (1992) Electron microscopic localization of replication
origins in Oenothera chloroplast DNA. Mol Gen Genet 232: 33-39.

Cleland RE (1935) Cyto-taxonomic studies on certain Oenothera from California. Proc
Amer Phil Soc 75:339-429.

122

 

123

Cleland RE (1972) Oenothera, cytogenetics and evolution. Academic Press, London and
New York.

Crank SF and Downie JA (1994) Isolation of a DNA polymerase I (pol A) mutant of
Rhizobium leguminosarum that has significantly reduced levels of an Ian-group
plasmid. MGG 243: 119-123.

Cuzzoni E, Giordani C, Stampacchia O, Bolchi A, Malcevshi A, Ottonello S, Ferretti L
and Sala F (1995) Presence of a chloroplast DNA sequence in an autonomous
circular DNA molecule in cultured rice cells (Orysa sativa L.).

Plant Cell Physiol. 36 (4): 717-725 .

Das Gupta, U., Weston-Hafer, K. And Berg, DE. (1987) Genetics 115: 41-49.

Day, A. And Ellis, T.H.N. (1984) Chloroplast DNA deletions associated with wheat
plants regenerated from pollen: possible basis for maternal inheritance of
chloroplast. Cell 39: 359-368.

de Hass JM, Kool AJ, Overbeeke N, van Brug W, Nijkamp JJ (1987) Characterization
of DNA synthesis and chloroplast DNA replication initiation in a Petunia hybrida
chloroplast lysate system. Current Gnetics 12:377-3 86.

Deng X-W, Wing RA, Gruissem W (1989) The chloroplast genome exists in multimeric
forms. PNAS 86: 4156-4160.

Dunford R and Walden RM (1991) Plastid genome structure and plastid-related transcript
levels in albino barley plants derived from anther culture. Curr.Genet. 20:
339-347 Epp MD (1973) Nuclear gene-induced plastome mutations in
Oenothera Hookeri 1. Genetic analysis. Genetics 75:465-483.

Epp.MD and Parthasarathy (1987) Nuclear gene-induced plastome mutations in
Oenothera hookeri II. Phenotypic description with electron microscopy.

Amer. J. Bot. 74: 143-151.

Epp MD, Erion JL, Sears BB, Stubbe W (1987) The plastome mutator of Oenothera
continues to fimction as originally described. Mol Gen Genet 206: 515-518.

Fang G-W, Harnmar S, Grumet R (1992) A quick and inexpensive method for removing
polysaccharides from plant genomic DNA. Biotechniques 13: 52-55.

F unnell BE, Baker TA and Kornberg (1986) Complete enzymatic replication of plasmids
containing the origin of the Escherichia coli chromosome. JBC 261: 5616-5624.

Francino MP, Chao L, Riley MA, Ochman H (1996) Asymmetries generated by
transcription-coupled repair in enterobacterial genes. Science 272: 107-109 .

Fukuoka H, Kawata M and Takaiwa F (1994) Molecular changes of organelle DNA
sequences in rice through dedifferentiation, long term culture, or the
morphogenesis process. Plant Molecular Biology 26: 899-907.

Gilham NW (1994) Organelle genes and genomes. Oxford University Press.

Glick RE, Sears BB (1994) Genetically programmed chloroplast dedifferentiation as a
consequence of plastome-genome incompatibility in Oenothera.

Plant Physiology 106: 367-373.

Gold B, Carrillo N, Tewari KK, Bogorad L (1987) Nucleotide sequence of a preferred
maize chloroplast genome template for in vitro DNA synthesis.
Proc. Natl. Acad. Sci. USA 84: 194-198.

124

Goldschmidt—Clermont M and Rahire M (1986) Sequence, evolution and fidderential

expression of the two genes encoding variant small subunit and ribulose
. bisphosphate carboxylase/oxygenase in Chlamydomonas reinhardtii. J. Mol. Biol
1 91 : 421 -432

Han CD, Coe EH, Martienssen (1992) Molecular cloning and characterization of iojap
(i1), a pattern striping gene of maize EMBO J 11: 4037-4046.

Harrak H, Lagrange T, Bisanz-Seyer C, Lerbs-Mache S, Mache R (1995) The expression
of nuclear gene genes encoding plastid ribosomal proteins precedes the expression
of chloroplast genes during early phase of chloroplast development. Plant Physiol
108:685-692.

Hedrick LA, Heinhorst S, White MA, Cannon G (1993) Analysis of soybean chloroplast
DNA replication by two-dimensional gel electrophoresis. Plant Mol Biol
23: 779-7 92.

Heinhorst S, Cannon GC (1993) DNA replication in chloroplasts. Journal of Cell Science
104: 1-9.

Hiratsuka J, Shimada H, Whittier R, Ishibashi T, Sakamoto M, Mori M, Kondo C, Honji
Y, Sun C-R, Meng B-Y, Li Y-Q, Kanno A, Nishizawa Y, Hirai A, Shinozaki K
and Sugiura M. (1989) The complete sequence of the rice (Oryza sativa)
chloroplast genome: Intermolecular recombination between distinct tRNA genes
accounts for a major plastid DNA inversion during the evolution of the cereals.
Mol. Gen. Genet. 217: 185-194.

Huberman JA (1995) Prokaryotic and eukaryoteic replicons. Cell 82: 535-542.

Johnson EM and Schnabelrauch LS and Sears BB (1991) A plastome mutation affects
processing of both chloroplast and nuclear DNA-encoded plastid proteins. Mol
Gen Genet 225: 106-112.

Johnson EM, Sears BB (1990a) Membrane composition and physiological activity of
plastids from an Oenothera plastome mutator-induced chloroplast mutant. Plant
Physiol 92: 254-261.

Johnson EM and Sears BB (1990b) Structure and expression of cytochrome f in an
Oenothera plastome mutant. Curr Genet 17: 529-534.

Johnson E.R., Kovvali GK, Prakash L, Prakash S (1995) Requirement of the yeast
RTHl 5’ to 3’ exonuclease for the stability of simple repetitive DNA.

Science 269: 238-239.

Kaplan S A (1987) Characterization of chNA fiom wild type and mutant plastids of
Oenothera hookeri str. Johansen. MSc Thesis, Michigan State University,

East Lansing, MI.

Karran P, Bignami (1994) DNA damage tolerance, mismatch repair and genome
instability. Bioassay 16: 833-838.

Khan SA (1996) Mechanism of replication and copy number control of plastmids
in gram-positive bacteria. Genetic Engineering 18: 1 83-201.

Kinzler KW, Vogelstein B (1996) Lessons from hereditary colorectal cancer 87: 159-170.

Kirk JTO and Tilney-Bassett RAE (1978) THE PLASTIDS: THEIR CHEMISTRY,
STRUCTURE, GROWTH AND INHERITANCE. Elservier/North-Holland
Biomedical Press, New York.

 

125

Kolodner R, Tewari KK (1975a) Chloroplast DNA from higher plants replicates by both
the Cairns and the rolling circle mechanism. Nature 256: 708-711.

Kolodner R, Tewari KK (1975b) Presence of displacement loops in the covalently closed
circular chloroplast deoxynucleic acid from higher plants JBC 250: 8840-8847.

Kroymann J, Schneider and Zetsche (1995) Opposite regulation of the copy number and
the expression of plastid and mitochondrial genes by light and acetate in the green
flagellate chlorogonium. Plant Physiol. 108: 1641-1646.

Kutzelnigg H, Stubbe W (1974) Investigation on plastome mutants in Oenothera. 1.
General considerations. Subcell Biochem 3: 73-89.

Lahaye A, Stall H, Thines-Sempoux D, and Foury F (1991) PIF 1: a DNA helicase in
yeast mitochondria. EMBO 10: 997-1007.

Lamppa GK, Bendich AJ (1979) Changes in chloroplast DNA levels during development
of pea. Plant Physiol. 64:126-130.

Lawrence ME and Possingham JV (1986) Microspectrofluorometric measurement of
chloroplast DNA in dividing and expanding leaf cells of Spinacia oleracea. Plant
Physiol 81: 708-710.

Leach DRF (1996) Cloning and characterization of DNAs with palindromic sequences.
Genetic Engineering, 18: 1-11.

Leech RM and Pyke KA (1988) Chloroplast division in higher plants with particular
reference to wheat. In Division and Segregation of organelles (Boffey SA and
Lloyd D, eds.), Soc. Exp. Biol. 35: 39-62.

Levison G and Gutrnan GA (1987) Slipped-strand mispairing: a major mechanism for
DNA sequence evolution. Mol Biol Evol 4: 203-221.

Lin L-F, Sperle K, Stemberg N (1984) Model for homologous recombination during
transfer of DNA into mouse L cells: Role for DNA ends in the recombination
process. Molecular and Cellular Biology 4: 1020-1034.

Lovett ST, F eschenko VV (1996) Stabilization of divergent tandem repeats by mismatch
repair: Evidence for deletion formation via a misaligned replication intermediate.
PNAS 93: 7120-7124.

Lu Z, Kunnimalaiyaan M, Nielsen B (1996) Characterization of replication origins
flanking the 23S rRNA gene in tobacco chloroplast. Plant Mol
Biol 32: 693-706.

Madsen CS, Ghivizzani SC and Hauswirth WW (1993) In vivo and in vitro evidence for
slipped mispairing in mammalian mitochondria. PNAS 90: 7671-7675.

Martinez-Zapater JM, Gil P, Somerville CR (1992) Mutations at the Arabidopsis CHM
locus promote rearrangement of the mitochondrial genome. The Plant Cell vol 4:
889-899.

Meeker R, Nielsen B, Tewari KK (1988) Localization of replication origins in pea
chloroplast DNA. M01 and Cellular Biol 8: 1216-1223.

MO J Y and Schaaper RM (1996) Fidelity and error specificity of the alpha-catalytic
subunit of E.coli DNA polymerase-III. JBC 271: 18947-18953

Morell R, Friedman TB, Asher JH (1993) A plus-one frameshift mutation in PAX3 alters
the entire deduced amino acid sequence of the paired box in a Wardenburg
syndrome type I (WSI) family. Human Mol Genet 2: 1487-1488.

 

126

Mourad GS, JA White (1992) The isolation of apparently homoplastidic mutants induced
by a nuclear recessive gene in Arabidopsis thaliana. Theor Appl Genet 84:
906-914.

Nakata T, Takebe D I (1971) Plating of isolated tobacco mesophyll protoplasts on agar
medium. Planta 99: 12-20.

Nielsen BL, Raj asekhar VK, Tewari KK (1991) Pea chloroplast DNA primase:
characterization and role in initiation of replication. Plant Molecular Biology 16:
1019-1034.

Nimzyk R, Schondorf T, Hachtel W (1993) In-frame length mutations associated with
short tandem repeats are located in unassigned open reading frames of Oenothera
chloroplast DNA. Curr Genet 23: 265-270.

Norrander, J ., T. Kempe, J. Messing (1983). Construction of improved M13 vectors
using oligo deoxynucleotide mutagenesis. Gene 26: 101.

Nowicka A, Kanabus M, Sledziewskagoj ska E, Clesla Z (1994) Different UMU C
requirements for generation of different kinds of UV-induced mutations in
Escherichia-coli. MGG 243 (5): 584-592.

O’Donnell ME and Kornberg A (1985) Dynamics of DNA polymerase III holoenzyme of
Escherichia coli in replication of a multiprirned template. JBC 260: 12875-12883.

Ohyama K, Fukuzawa H, Kohchi T, Sano T, Sano S, Umesono K, Shiki Y, Takeuchi M,
Chang Z, Aota S, Inokuchi H and Ozeki H (1986) Chloroplast genes organization
deduced from complete sequence of liverwort Marchantia polymorpha
chloroplast DNA. Nature (London) 322:572—574.

Orozco, EM, Jr., Mullet JE, Chua N-H ( 1985) An in vitro system for accurate
transcription initiation of chloroplast genes. Nucleic Acids Res. 13: 1283-1302.

Palmer JD (1991) Plastid chromosomes: structure and evolution. In: Bogorad L and Vasil
I (eds.) The Molecular Biology of Plastids, Cell Culture and Somatic Cell
Genetics of Plants, vol 7a, Acad. Press, N.Y. pp.5-53.

Possingham and Rose (1976) Chloroplast replication and chloroplast DNA synthesis in
spinach leaves. Proc R Soc Lond B 193: 295-305.

Prina AR (1992) A mutator nuclear gene inducing a wide spectrum of cytoplasmically
inherited chlorophyll deficiencies in barley. Theor Appl Genet 85: 245-251.

Prina AR (1996) Mutator-induced cytoplasmic mutants in barley: Genetic evidence of
activation of a putative chloroplast transposon. The Journal of Heredity 5:
385-391.

Radman M, Wagner R (1986) Mismatch repair in Escherichia coli. Ann Rev Genet 20:
523-538.

Rapp J C and Mullet JE (1991) Chloroplast transcription is required to express the nuclear
gene rch and cab. Plastid DNA copy number is regulated independently. Plant
Mol Biol 17: 813-823.

Reddy M, Choudhury NR, Kumar D, Mukherj ee SK, Tewari KK (1994) Characterization
and mode of in vitro replication of pea chloroplast OriA sequences. Eur. J.
Biochem. 220: 933-941.

Redei GP (1973) Extra-chromosomal mutability determined by a nuclear gene locus in
Arabidopsis. Mutat Res 18: 149-162.

127

Rether B, Delmas G, Laouedj A (1993) Isolation of polysaccharide-free DNA from
plants. Plant Mol Biol Reporter 11: 333-337.

Richards CM, Hinman SB, Boyer CD, Hardison RC (1991) Survey of plastid RNA
abundance during tomato fruit ripening: the amount of RNA from the ORF 2280
region increase in chromoplasts. Plant Mol Biol 17: 1179-1188.

Rosche WA, Trinh TQ, Sinden RR (1995) Differential DNA secondary structure-
mediated deletion mutation in the leading and lagging strands. Journal of
Bacteriology 177: 4385-4391

Russel M, Kidd S, Kelley R (1986) An improved filamentous helper phage for generating
single-stranded plasmid DNA Gene 45: 333-338

Sakamoto W, Kondo H, Murata M, Motoyoshi F (1996) Altered mitochondrial gene
expression in a maternal distorted leaf mutant of Arabidopsis induced by
chloroplast mutator. The Plant Cell 8: 1377-1390.

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning, 2nd ed. Cold Spring
Harbor Laboratory Press. N.Y.

Sanger F, Milden S, Coulson AR (1977) DNA sequencing with chain-terminating
inhibitors. PNAS USA 74: 5463-5467.

Sears BB (1983) Genetics and evolution of the chloroplast. In: Gustafsson P (ed) Stadler
Symposium XV, University of Missouri Press Columbia, Mo pp. 1 19-139.

Sears BB, Sokalski MB (1991) The Oenothera plastome mutator: effect of UV irradiation
and nitroso-methyl urea on mutation frequencies. Mol Gen Genet 229: 245-252.

Sears BB, Chiu W-L, Wolfson R (1995) Replication slippage as a molecular mechanism
for evolutionary variation in chloroplast DNA due to deletions and insertions. In:
Tsunewaki K (ed) Plant Genome and Plastomes: Their Structure and Evolution.
Kodasha Scientific Ltd, Tokyo pp. 139-146.

Sears BB, Stoike L and Chiu W-L (1996) Proliferation of direct repeats near the
Oenothera chloroplast DNA origin of replication. Mol Biol Evol 13: 850-863.

Sears et al. 1997 Molec. and Physiol. Plant Path.

Shimada and Sugiura (1991) Fine structural features of the chloroplast genome:
comparison of the sequenced chloroplast genomes. Nucleic Acid Research 19:
983-995.

Shinozaki K, Ohme M, Tanaka M, Wakasugi T, Hayashida N, Matsubayashi T, Zaita
N,Chunwangse J, Obkata J, Yamaguchi- Shinozaki-K, Ohto C, Torazawa K,
Meng B-Y, Sugita M, Deno H, Kamagoshira T, Yamada K, Kusuda J, Takaiwa F,
Kato A, Tohdoh N, Shimada H and Sugiura M (1986) The complete nucleotide
sequence of the tobacco chloroplast genome: Its gene organization and
expression. EMBO J 5: 2043-2049.

Shumway L K, Weier T E (1967) The chloroplast structure of iojap maize. Am J Bot 54:
773-780.

Stoike L (1997) Ph.D. Thesis Michigan State University, East Lansing, MI.

Strand M, Prolla TA, Liskay RM, Petes TD (1993) Destablization of tracts of simple
repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature
365: 274-276.

Stubbe W (1959) Genetische analyse des Zusamnerbewirkens von genome and plastome
bei Oenothera. Z Vererbungsl 90: 228-298.

128

Takeda Y, Hirokawa H, Nagata T (1992) The replication origin of proplastid DNA in
cultured cells of tobacco. Mol Gen Genet 232: 191-198.

Thompson D, Walbot V, Coe EH Jr (1983) Plastid development in iojap and mutator-

' affected maize plants. Am J Bot 70: 940-950.
Trinh TO and Sinden RR (1991) Preferential DNA secondary structure mutagenesis in
the lagging strand of replication in E. col. Nature 352: 544-547

Treier M, Pfeifle C, Tautz D (1989) Comparison of the gap segmentation gene hunchback
between Drosophila melanogaster and Drosophila virilis reveals novel modes of
evolutionary change. EMBO 8: 1517-1525.

Tuteja N, Phan TN, Tewari KK (1996) Purification and characterization of a DNA
helicase from pea chloroplast that translocates in the 3 ’-to-5’ direction. Eur. J.
Biochem 238: 54-63.

Van Dyck E, F oury F, Stillman B, Brill SJ (1992) A single stranded DNA binding
protein required for mitochondrial DNA replication in S. cerevisiae is
homologous to E.coli SSB. EMBO 11: 3421-3430.

Vincent SD, Mahdi AA, Lloyd RG (1996) The recG branch migration protein
Escherichia-coli dissociates R-loops. Journal of Molecular Biology 264: 713-721.

Walbot V, Coe Jr EH (1979) Nuclear gene iojap conditions a programmed change to
ribosome-less plastids in Zea mays. PNAS 76: 2760-2764.

Wallis JW, Chrebet G, Brodsky G, Rolfe M, Rothstein R (1989) A hyper-recombination
mutation in S. cerevisiae identifies a novel eukaryotic topoisomerase. Cell 58:
409-41 9.

Wierdle M, Greene CN, Datta A, Jinks-Robertson S (1996) Destablization of simple

repetitive DNA sequences by transcription in yeast. Genetics 143: 713-721

Williams AJ and Kaguni LS (1995) Stimulation of Drosophila mitochondria DNA
polymerase by single-stranded DNA-binding protein. JBC 270: 860-865.

Wolfgang Z, Weissbach A (1982) Deoxyribonucliec acid synthesis in isolated
chloroplasts and chloroplast extraxt of maize. Biochemistry 21: 3334-3343.

Wolfson R, Higgins KG, Sears BB (1991) Evidence for replication slippage in the
evolution of Oenothera chloroplast DNA. Mol Biol Evol 8: 709-720.

Wu M, Lou JK, Chang DY, Chang CH, Nie ZQ (1986) Structure and fucnction of a
chloroplast DNA replication origin of Chlamydomounas reinhardtii. Proc. Natl
Acad. Sci. USA 83: 6761-6765.

U V

111 (113W

   

7.

”11111131“