V fim :1
1‘» .ufdfl‘...
37?. V.

3 a: a:

x, AM; u...”
. an

2.5%.”.
12.3”."

. 53,

.1. .1.
flsfiun.“..¢

..

.
9.
{may

(at;

213
u 51
flwvmm

iv

. 3. l: I
s. .1, .
it.) hv‘v

Ihnv;

 

‘7
«v». a‘Ai‘r H...o.-n,:wL-s
. . .

. r. 2,353.... .4. .
‘..:1.:.~.?u.;.d.é . A.

 

mists

 

3
9M3
swagsfi UBPA 9v
Michigan state
University

 

 

 

This is to certify that the

dissertation entitled

CHLOROPLAST DNA MUTATIONS: AN EXPERIMENTAL
EVALUATION

presented by
MONICA GUHA MAJ UMDAR

has been accepted towards fulfillment
of the requirements for

 

i
Ph . D- degree in .ELANLBIQLQGY l
1

Major professor

8 8 2003
Date I /

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DatoDuopGS-p. 1 5

CHLOROPLAST DNA MUTATIONS:
AN EXPERIMENTAL EVALUATION

By

Monica Guha Majumdar

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Plant Biology

2003

ABSTRACT

CHLOROPLAST DNA MUTATIONS: AN EXPERIMENTAL EVALUATION

Monica Guha Majumdar

Spontaneous mutation events may give rise to base substitutions as well as
insertions / deletions of bases. The rates at which these events occur can be considered to
be indicators of the fidelity and efficiency of the replication and repair machinery.
Mutation rates for the chloroplast DNA in plants have been obtained through
evolutionary observations rather than by experimental approaches. One of the goals of
this research was to experimentally assess the rates of different types of mutations
occurring in viva at selected target sites in the chNA of the unicellular green alga,
Chlamydomonas reinhardtii. The rates of point mutations in the chNA were assessed
for three different positions in the 16S rRNA gene, where changes confer spectinomycin
resistance accompanied by the loss of a restriction endonuclease cut site. Results showed
that the rates of forward mutation in the chNA to spectinomycin resistance ranged from
3 X 10'11 to 1.1 X 10's, and among the mutations recovered, transversions occurred at an

approximately four fold higher rate than transitions. Insertion/deletion events in the

chNA were examined by creating a reporter construct that carried a series of short
direct repeats representative of naturally occurring microsatellites. The reporter construct
contained an out—of-frame insertion of a (GAAA)6 tract in the rbcL gene of the
Chlamydomonas chNA which allowed the visualization and quantification of
insertion/deletion events, that were thought to have resulted from replication slippage.
Results showed that this site underwent slippage at fi‘equencies that ranged from 1-70 X
1045 and analyses of the slipped cells showed that only deletions had occurred at this site.
Once the mutation rates were quantified, a plasmid-based insertional mutagenesis system
was used to isolate putative mutator lines that showed high rates of chNA replication
slippage. One of the tagged lines has been confirmed to act as a chNA mutator. This

mutator line will be further analyzed to identify the gene disruption that has caused the

phenotype.

In addition to studies with the mutators in Chlamydomonas, experiments utilized
the plastome mutator of Oenothera. To examine whether the pm-activity repairs damage
caused by the mutagen 9-aminoacridine hydrochloride (9AA), a mutant plant A-E was
selected for molecular analyses. As a necessary prerequisite, the chNA was confirmed
as the site of the mutation induced by 9AA, through demonstration that the yellow sectors
in progeny derived from this mutant plant cosegregated with the chNA of the mutant
parent, and not with a mitochondrial DNA marker. Although 9AA acted as a weak
plastome mutagen, when applied to seeds with the plastome mutator genotype, it did not
synergistically increase the mutation frequency. Hence, the plastome mutator product

was not involved in repairing 9AA-induced mutations.

For my parents and Dipnath

iv

ACKNOWLEDGEMENTS

I would like to take this unique opportunity to acknowledge the people for whom
my PhD. has been possrble. First and foremost I would like to express my heartfelt
gratitude to Dr. Barbara Sears, my advisor, for her encouragement and support for all my
projects and her willingness to let me take those detours during my PhD. career that have
helped widen my scientific experience and have been a true source of inspiration to
continue on in the field of Plant Biology. I am also grateful for her contributions at the
lab bench during crosses and transformation experiments. I would also like to thank Dr.
Michele Fluck, my committee member who has inspired me by her scientific insights and
who has always been a source of true encouragement to me. I would also like to
gratefully acknowledge the generous help and guidance provided to me by committee

members Dr. Kenneth Keegstra and Dr. Gregg Howe throughout my PhD.

I owe a lot of gratitude to Lara Stoike Steben whose patience and selfless help
during my beginning years helped me develop many of the skills that I possess at the
bench as well as contributed to my understanding of science. I would also like to thank
Elaine Palucki from the bottom of my heart for her fiiendship, editorial contributions and
constant support at both scientific and personal aspects. Many thanks are also due to
several members of the Sears lab, namely Kristi Nichols, Elizabeth Graffy, Tseh-Ling
Chang, Siavash Jabbari, Brandi Foster, Latisha Miller, whose friendship and help have
contributed to the successfirl completion of my Ph.D. I am also indebted to Andrea
daRocha and Iffa Gaffoor for their continued friendship and help throughout my graduate

study. Finally I would like to thank my parents and my husband, Dipnath, for always

being there for me and providing me with all the support throughout my entire life and to

whom I owe any success that I may experience.

My research was possible due to the many teaching assistantships provided to me

by the Department of Biology and also for the financial support provided by the National

Science Foundation.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. x

LIST OF FIGURES ............................................................................................................ xi

CHAPTER 1

INTRODUCTION ............................................................................................................... 1
Chloroplasts: structure and function ........................................................................ 2
Chloroplast DNA mutations .................................................................................... 4
Mutation avoidance in the chNA .......................................................................... 7
Mutators: isolation, types and targets .................................................................... 15
Chlamydomonas as the experimental organism ..................................................... 21
References .............................................................................................................. 24

CHAPTER 2

CHLOROPLAST MUTATIONS INDUCED BY 9-AMINOACRIDINE
HYDROCHLORIDE ARE INDEPENDENT OF THE PLASTOME MUTATOR IN

OENOTHERA .................................................................................................................... 32
Abstract .................................................................................................................. 33
Introduction ............................................................................................................ 33
Materials and Methods ........................................................................................... 36
Results .................................................................................................................... 4O

9-aminoacridine hydrochloride dosage trials on wild-type ....................... 40
Mutant phenotypes from acridine treatment .............................................. 42
Cosegregation of mutant phenotype analyses using
molecular markers .......................................................................... 44
9-aminoacridine hydrochloride dosage effects on the
plastome mutator line .................................................................... 49
Discussion .............................................................................................................. 51
References .............................................................................................................. 54

CHAPTER 3

EXPERIMENTAL ASSAYS OF CHLOROPLAST DNA BASE SUBSTITUTIONS IN

CHLAMYDOMONAS REINHARDHI ............................................................................... 57
Abstract .................................................................................................................. 58
Introduction ............................................................................................................ 59
Materials and Methods ........................................................................................... 65
Results .................................................................................................................... 69

Estimation of rates of base substitution mutations in the

16S rRNA gene ............................................................................... 69
Analysis of base substitutions for each base .............................................. 72
Ratios of transition and transversion mutations ......................................... 75
Comparison of base changes in the 165 rRNA gene in

different organisms ........................................................................ 77

vii

Discussion .............................................................................................................. 78

References .............................................................................................................. 87
CHAPTER 4
CREATION OF A REPORTER CONSTRUCT TO ASSAY FOR CHLOROPLAST
DNA REPLICATION SLIPPAGE EVENTS .................................................................... 91
Abstract .................................................................................................................. 92
Introduction ............................................................................................................ 93
Materials and Methods ......................................................................................... 105
Results .................................................................................................................. 1 12
Introduction of slippage substrates into the chNA of
C. reinhardtii ............................................................................... 1 12
Rates of mutation at the microsatellites in the chNA
of the OF cells .............................................................................. 116
Molecular analysis of slippage events at the (GAAA)n
microsatellite site ......................................................................... 116
Slippage rates in a C. reinhardtii strain with a recombination
deficient background .................................................................... 122
Discussion ............................................................................................................ 127
References ............................................................................................................ 138
CHAPTER 5
ISOLATION OF CHLOROPLAST MUTATORS IN CHLAMYDOMONAS
REINHARDTII ................................................................................................................. 145
Abstract ................................................................................................................ 146
Introduction .......................................................................................................... 146
Materials and Methods ......................................................................................... 151
Results .................................................................................................................. 154
Construction of strains for nuclear transformation .................................. 154
Assay for isolating chloroplast DNA mutators
[tom the Zeocin-resistant trans formants ...................................... 157
Selection of true mutator lines ................................................................. 160
Discussion ............................................................................................................ 164
References ............................................................................................................ 166
CHAPTER 6
CONCLUSIONS ............................................................................................................. 168
APPENDIX
EXAMINING THE PATTERNS AND RATES OF REPLICATION SLIPPAGE IN
PLASMIDS OF ESCHERICHIA COLI ........................................................................... 173
Introduction .......................................................................................................... 174
Materials and Methods ......................................................................................... 176
Results .................................................................................................................. 178
Replication slippage in pOF7 plasmid ..................................................... 178
Replication slippage in aadA -OF cells .................................................... 181

viii

Discussion ............................................................................................................ 183
References ............................................................................................................ 1 84

ix

LIST OF TABLES

Table 2-1. Sector frequencies for wild-type seedlings containing plastome H treated with
O, 2 and 10 pg/ml concentrations of 9AA for different exposure times ............................ 41

Table 2-2. Crosses to establish the inheritance patterns of the 9AA-induced organelle
mutation in A-E .................................................................................................................. 45

Table 2-3. Effect of plastome mutator on the sector frequency in seedlings containing
plastome I treated with 9AA .............................................................................................. 50

Table 3-1. Frequencies of spontaneous base substitutions at Aat 11 target site in the
chloroplast DNA of Chlamydomonas reinhardtii .............................................................. 70

Table 3-2. Total numbers of the different types of base substitutions observed in each
experiment .......................................................................................................................... 74

Table 3-3. Comparison between the types, numbers and frequencies of base substitutions
among the three bases at the Aat 11 target site ................................................................... 76

Table 3-4. Spectrum of chNA base substitutions in wild-type Chlamydomonas
reinhardtii and the possible pathways by which each type is known to be repaired ......... 84

Table 4-1. Phenotypes of the wild type and the transformed cell lines of C. reinhardtii
under different conditions ................................................................................................ 115

Table 4-2. Rates of spontaneous mutations to photosynthetic competence in the OF strain
from three different trials ................................................................................................. 1 17

Table 4—3. Rates of spontaneous replication slippage events in the chloroplast DNA of
Chlamydomonas reinhardtii containing the OF construct in a recombination-deficient
background ...................................................................................................................... 128

Table 5-1. Number of true mutators obtained from screening the initial Zeocin—resistant
nuclear transformants ....................................................................................................... 159

Table 5-2. Phenotypes of two tetrad progenies from a cross between pm] and an OF
line .................................................................................................................................... 162

LIST OF FIGURES

Figure 1-1. Consequences of incorporation of 8-oxodGTP opposite template C and A,

and the roles of mutT, mutM and mutY gene products in 8-oxoG metabolism. ................. 9
Figure 2-1. A-B Electron micrographs of plastids with aberrant thylakoid structures and

mitochondria ..................................................................................................................... 43
Figure 2-2. PCR analysis of 16S rRNA—trnl spacer region ............................................... 47

Figure 2-3. Southern hybridization analysis using DNA from the mitochondria-emiched
fraction, extracted from the green and yellow sectors of the progeny and also fi‘om the
parents of the A-E X wt cross ............................................................................................ 48

Figure 3-1. Base substitutions at a particular region in the gene coding for 165 rRNA in
Chlamydomonas reinhardtii confer resistance to spectinomycin ...................................... 64

Figure 3-2. PCR amplification using the 16S-F and 16S-R primers followed by
restriction digestion with Aat H ......................................................................................... 71

Figure 3-3. Sequence alignment of wild-type and spectinomycin resistant colonies at the
Aat II target site showing the six different base changes, compiled fiom all three trials. .73

Figure 3-4. Alterations in the gene encoding 16S rRNA compiled from spectinomyicn-

resistant mutants of different organisms ............................................................................ 77
Figure 4-1. Model for replication slippage ....................................................................... 95
Figure 4-2. Overview of assay for detection of replication slippage in the chloroplast
DNA of the OF cells ........................................................................................................ 102
Figure 4-3. Sequences and amino acid predictions of a part of the rbcL coding region in
several Chlamydomonas lines .......................................................................................... 104
Figure 4-4. Diagrarnmatic representation of the transforming plasmids ........................ 107

Figure 4-5. Agarose gel showing the PCR amplified products from the wild-type, in-
flame and out-of-fi‘ame cells ............................................................................................ l 13

Figure 4—6. Agarose gel electrophoresis showing the PCR amplified products from the
slipped colonies with one or four repeat units deleted ..................................................... 119

Figure 4-7. Agarose gel showing the PCR amplified products from a subset of the
revertant colonies along with the wt and the OF cells ..................................................... 121

xi

Figure 4-8. Crosses to introduce the slippage construct into a chloroplast DNA subject to
reduced homologous recombination ................................................................................ 123

Figure 4-9. Agarose gel showing PCR amplified products from the meiotic progeny of
the cross CC3455 X OF, to check for the slippage insert ................................................ 125

Figure 4-10. Agarose gel showing PCR amplified products from the meiotic progeny of
the cross CC3455 X OF, to test for the presence of the delta N RecA gene ................... 126

Figure 4-11. Agarose gel showing the heteroplasmic condition of the slipped cell line # 3
containing chNAs with the original OF—size insert as well as a 4-repeat deleted
chNA ............................................................................................................................. 135

Figure 5-1. Diagramrnatic representation of the pSP124S plasmid used for the nuclear
transformation of the Chlamydomonas chF cells ......................................................... 150

Figure 5-2. Agarose gel picture showing the PCR amplified products from the wt, pOF7
and the cell wall-less OF cells ......................................................................................... 155

Figure 5-3. Agarose gel showing the pSP124S plasmid DNAs used for nuclear
transformation of the chF cells ..................................................................................... 156

Figure 5-4. Graphical representation of the comparisons of efficiency of nuclear
transformation in chF cells under different conditions ................................................ 158

Figure 5-5. Assay for isolating the chNA mutator strain ............................................. 161

Figure 5-6. Growth of tetrad progenies 11-1 to 11-4 obtained fiom the cross PM] X OF,
under different growth conditions .................................................................................... 163

Figure A-l. Insertion events seen in a plasmid containing a stretch of GAAA repeats
contained in E. coli .......................................................................................................... 179

Figure A-2. Replication slippage at GAA microsatellites in the 2“d codon position of the
aadA gene. ...................................................................................................................... 182

xii

CHAPTER 1

Introduction

Chloroplasts: structure and origin

Land plants and green algae are similar in that they possess chlorophyll
a/b—containing chloroplasts. Chloroplasts are double-membrane organelles that specialize
in photosynthesis. Chloroplasts possess their own genomes that are referred to as
chloroplast DNAs (chNAs) or plastomes. Chloroplasts have been proposed to be
descendants of cyanobacterial endosymbionts and it has been hypothesized that the
creation of functional plastids involved the migration of large numbers of genes from the
prokaryotic ancestral genome to the plant nucleus (Cavalier-Smith 1992; Gray 1992;
Wolfe et a1. 1992). Gene transfer requires a series of steps, which include the integration
of a copy of the plastid gene into the nucleus (duplication), acquiring of the proper gene
expression and transit signals by the nuclear copy, followed by subsequent loss of the
original genes from the chloroplast ancestor. The gene tqu encoding the chloroplast
translation factor is such an example, as it is found in the chNA of most green algae,
but is absent from land plant chNAs. Molecular and phylogenetic analyses have shown
that the nuclear genome of Arabidopsis contains a tqu-like gene of a prokaryotic origin,
whereas some lineages of green algae have tqu sequences in both the genomes. The
modem day chloroplast has transferred many of its genes to the nucleus and thus depends

on it for many of its essential functions (Wolfe et a]. 1992).

Chloroplast DNA: strudure and gene content
The chloroplast genomes are circular in structure and the genome sizes vary from
50- 300 kb among different algal and plant species (Gillham 1994). Chloroplast genomes

normally possess two unique sequence regions and two large inverted repeats (IRs). The

two repeat sequences are positioned on the circular molecule in an asymmetric fashion
such that it divides the two unique sequences into a small and a large unique sequence
region known as the small single copy (SSC) and the large single copy (LSC) regions
respectively. However, in the unicellular green alga, Chlamydomonas reinhardtii, the two
unique sequence regions are roughly the same size. The chNAs of some land plants
(such as rice and tobacco) and the green alga (C. reinhardtii) have been completely
sequenced and well characterized (Shinozaki et al. 1986; Hiratsuka et al. 1989; Maul et
a1. 2002). The plastid genomes contain genes that encode for essential components
required for chloroplast structure, photosynthesis and gene expression. However, none of
the plastid genomes sequenced so far contain genes that code for the enzymes required

for the replication and repair of the chloroplast DNAs.

Since chNA sequence data from different plants and algae have shown that the
genes for its replication and /or repair are absent from their respective plastid genomes, it
can be concluded that those genes reside in the nucleus and would possess appropriate
chloroplast targeting signals. The gene products in this case, however, would still retain
prokaryotic features reminiscent of their cyanobacterial ancestry. Alternatively, it my
also be possible that the genes responsible for chloroplast DNA replication and repair
may have been totally lost during the evolutionary process and the chNA nmy share
replication and repair enzymes with its eukaryotic nuclear counterpart. The nuclear
enzymes in the latter case may have acquired transit peptides and could firnction in

replication and/ or repair of both the genomes.

Chloroplast DNA mutations

Mutations are heritable changes in DNA sequences that are acted upon by natural
selection and contribute to genetic diversity. Mutations occur spontaneously or can be
induced by mutagens. The two major types of mutations that are observed in organisms
are base substitutions and insertions/ deletions. Base substitutions occur when one base is
exchanged for another in a DNA sequence. Base substitutions can arise naturally through
intracellular metabolism such as spontaneous alkylation, dearnination of bases or
modification by reactive oxygen species. Base substitutions can also result item the
incorrect insertion of nucleotides during DNA replication (Kunz et al. 1998). The second
type of spontaneous mutation that occurs frequently in organisms is the
duplication/deletion of bases in regions of the DNA that contain repeated sequences
(microsatellites). Replication slippage has been implicated to be the mechanism by which
such deletions or insertions are generated at these repeats (for a detailed description of
replication slippage refer to Figure 4-1). Prokaryotes and eukaryotes have strategies to
minimize both of these types of mutations. These include elimination of the DNA-
darnaging agents, a high accm'acy of DNA replication, and repair of the damage after it

has occurred

A low spontaneous mutation rate can be regarded as an indicator of efficient
replication and repair mechanisms in a given organism. In vivo mutation rates have been
well documented in both prokaryotes and eukaryotes through the use of experimental
analyses. In plants, the rates of mutations of nuclear and the chloroplast genomes have

been studied through evolutionary comparisons. The highly conserved nature of

chloroplast DNA sequences indicates that it has a lower mutation rate than does the
nucleus (Palmer 1985; Zurawski and Clegg 1987; Wolfe et al. 1987; Clegg et al. 1994).
These observations have led to the suggestion that the enzymes responsrble for the
maintenance of chloroplast replication and DNA repair have high levels of fidelity.
However such evolutionary comparisons do not always offer a true assessment of
mutation rates: a comparison of the chloroplast DNA sequences from the degenerate
chNAs of certain parasitic plants and a study of non-coding regions have both shown
that certain segments of the chloroplast DNA molecule may evolve at a much higher rate
(Wolfe et al. 1992; Gielly and Taberlet 1994). As a first step towards the understanding
the mutational mechanisms that govern the chloroplast DNA, I created reporter system
to experimentally assess the rates of base substitution and insertion/deletion events.
Furthermore, the experimental data obtained allowed me to compare the relative
fiequencies of occurrence of different types of base substitutions as well as replication

slippage events that occurred in the chNA

Assessment of relative mutation rates

Evolutionary comparisons of the chNA microsatellites between different species
of pine (Pinus) indicated that the length polymorphism occur at rates ranging from 3.2 to
7.9 X 10'5 per site per year (Provan et al. 1999). In contrast, another evolutionary study
has determined that nucleotide substitutions occur within a range of 1.1 to 3 X 10’9
substitutions per synonymous site per year in the chNA (Muse 2000). Studies with the
Oenothera plastome mutator have shown that within a 450-bp intergenic spacer region of

the chNA, variations in two chloroplast microsatellites occurred more frequently than

did base substitutions when different species of Oenathera were compared (Wolfson et
al. 1991). These suggest that the chNA replication/repair machinery may be less
efficient in handling insertion/ deletion events than base substitutions. However, these
deductions were based on evolutionary data and therefore do not reflect a true
comparison of mutation rates. The experimental data obtained from my studies allowed
me to directly compare the rates of spontaneous insertion/deletion mutations to the base
substitution events that occur in viva in the chNA and thereby provided an accurate

assessment of the relative rates of the two mutation events.

Overall goal for determining the spontaneous mutation rates for the chNA

In summary, the first part of my research project involved the creation of reporter
constructs to monitor and quantify the rates of different types of mutations that occurr in
the chNA of Chlamydamanas reinhardtii. Specifically, I compared the relative rates of
in viva base substitutions to replication slippage events for the chNA. In addition, the
analyses of the different mutants obtained from my study enabled me to assess the
individual frequencies of transition and transversion mutations in the chNA as well as
to quantify the frequencies of occurrence of duplication and deletion events occurring at
the microsatellites. These results were then compared to the conclusions from
phylogenetic analyses in order to see if my experimental data were consistent with the

evolutionary evidence.

Mutation avoidance in the chNA

Studies in Escherichia cali and Saccharamyces cerevisiae have shown that in
order to achieve a low rate of mutation, organisms must employ numerous mutation
avoidance strategies such as a high accuracy of DNA replication, with maintenance of
DNA in an error free state either through the elimination of DNA damaging agents or by
repairing the DNA damage after it has occmred (Schaaper and Dunn 1991; Kunz et al.
1998). The understanding of such mechanisms for the chloroplast DNA is still at a very
rudimentary stage and most of the studies have used in vitra enzyme assays. The
following is a brief summary of what is known regarding the mutation avoidance

machinery of the chloroplast.

Removal of DNA damaging agenm

Oxygen radicals are highly reactive and are generated in cells exposed to radiation
or chemicals. They also arise during the normal metabolic processes such as respiration
or photosynthesis in plants (Ohtsubo et al. 1998). Intracellular reactions producing active
oxygen species have been identified as major sources of DNA damage and as
contnbutors to spontaneous mutation (Schaaper and Dunn 1991). One of the deleterious
outcomes is the production of 8-oxoguanine (8-oxoG) which has ambivalent base pairing
properties due to its ability to pair effectively with both A and C during DNA synthesis.
Consequemly it is an intrinsically highly mutagenic base analog either as a constituent of
DNA (8-oxodG) or when present as a deoxynucleoside triphosphate (8-oxodGTP)
(Fowler et al. 2003). Specialized systems that are aimed at protecting the cell against the

effects of such oxidative stress exist: these are the MutT, MutY and the MutM systems of

E. cali that carry out related but separate aspects of 8-oxoG metabolism. The eukaryotic
homologs have also been identified for these genes. Figure 1-1 outlines the different
pathways that exist in bacteria for the prevention and repair of 8-oxoG mediated
mutagenesis. As seen in the figure, the MutT gene product hydrolyzes 8-oxodGTP to
prevent its use as a substrate by DNA polymerase; mutT deficient strains show an
increase in NT 9 C/G transversions resulting from the misincorporation of 8-oxodGTP
opposite adenine in the template (Fowler et al. 2003). Nuclear and chloroplast MutT

genes have not yet been identified in plants.

Maintenance of replication fidelity

The low chNA mutation rates indicated by evolutionary studies suggest that the
chloroplast possesses efficient DNA replication machinery. This section outlines certain
features of chNA replication that have been compiled from studies with different plants.
Aspects of replication of the chNA have been documented by microscopy, genetic
analyses and in-vitra replication assays in a range of plants including C. reinhardtii, pea,
soybean and Oenathera (Kolodner and Tewari 1975a and b; Wadell et al. 1984; Chiu and
Sears 1992; Heinhorst and Canon 1993). However, none of the genes have been isolated
or characterized from the respective nuclear genomes of any of the plant species
mentioned above. Based on prokaryotic and eukaryotic systems it is known that a basic
set of enzymes is necessary for the efficient replication of the chNA molecules. These
include the DNA polymerases, helicases, primases, topoisomerases, single stranded DNA

binding protein (SSB), and other accessory factors to the polymerases.

 

 

 

 

 

 

 

 

dGIP
:g: lam...
0M Incorporatrm Incorporation
Oxidation % d©w 1
—©— ET]
— c —
E/ d© MP + PPI E/
:3: Rephcation
-—C—
V ©
:9: o. :9: n...
mutation
Warden
_©_ _ r _
— C — —" A ‘—
GC—rIA
nutrition

 

Figure 1-1. Consequences of incorporation of 8-oxodGTP opposite template C and
A, and the roles of the mutT, mutM and trth gene products in 8-oxoG metabolism.

 

 

 

 

 

 

The boxed part of the pathway highlights the role of mutT in removal of the 8-

oxodGTP fiom the nucleotide pool and the non-boxed regions show the roles of
mutY and trth gene products in the repair of the damage caused due to
incorporation of 8-oxodGTP. (Adapted from Fowler et al. 2003)

 

The DNA polymerases from pea, spinach, soybean and Chlamydamanas
chloroplasts have been purified: all the chloroplast polymerases have an estimated
molecular mass between 90,000 and 120,000, belong to the y—class of DNA polymerases
and are resistant to the DNA polymerase (1 inhibitor, aphidicolin (Sala et al. 1980;
McKown and Tewari 1984-, Heinhorst et al. 1990; Wang et al. 1991). The DNA
polymerases fi'om chloroplasts of spinach and pea have moderate fidelity due to the

presence of associated 3’-)5’- exonucleases (Keim and Mosbaugh 1991; Gaikwad et al.

2002).

Both DNA topoisonrerase I and H activities have been observed from the
chloroplasts of algae and plants such as spinach, pea and cauliflower and
Chlamydamanas (Fukata et a1. 1991; Heinhorst and Cannon 1993; Woefle et a]. 1993;
Mukherjee et a1. 1994; Nielsen and Tewari 1998; Reddy et al. 1998). A DNA primase
activity has been reported from the chloroplasts of pea (Nielsen et al. 1991) and a DNA—
dependent helicase was purified from soybean chloroplasts (Cannon and Heinhorst
1990). Although the presence of these different replicative enzymes has been
demonstrated from chloroplasts, their direct roles in chloroplast DNA replication have

not been proven yet.

DNA repair path ways
One of the other aspects of the mutation avoidance machinery of the chNA is
the Presence of efficient DNA repair pathways that eliminate darmge that would

Otherwise result in high mutation frequencies. The types of DNA damage for which one

10

would anticipate that repair enzymes should exist include: mismatched bases, double-
strand breaks, and chemically modified bases. It is expected that plants possess protection
against all these kinds of damage, and since they possess three separate genomes, repair
enzymes are required for all of them The chNA lesions must be corrected by distinct
repair pathways whose components are encoded in the nucleus, since plastid genomes
that have been sequenced lack genes for repair enzymes (Britt 1996). Unlike the
replicative enzymes, however, few repair enzymes have been isolated from chloroplasts

of plants.

One of the few repair enzymes whose activity has been detected in the
chloroplasts of plants is the RecA protein. This protein is essential for both recombination
and repair activities in prokaryotes and eukaryotes. A honrolog of the E. cali RecA gene,
which has been cloned from Arabidapsis, has been hypothesized to play a role in the
repair of the chNA (Cerutti et al. 1992). Its role has been shown in C. reinhardtii by
transforming the chloroplasts of Chlamydamanas with a wild-type copy of the E. cali
gene and a dominant negative variant and then monitoring the recombination and repair
abilities in these different transformants (Cerutti et al. 1995). Results showed that the
transformant with the dominant negative gene had reduced efficiencies of repair of
chNA damage, thereby indicating a role for the native chloroplast RecA gene product in

repair of chloroplast DNA.

Plants require sunlight for photosynthesis but at the same time they are exposed to

the harmful effects of solar UV radiation (Gallego et al. 2000). UV-light results in

11

oxidative lesions as well as in the creation of cross-lirrks in DNA (cyclobutane
pyrimidine dimers and 6-4-photoproducts), the accumulation of which affects cell
viability (Gallego et al. 2000; Waterworth et al. 2002). A light-dependent repair pathway
exists for the repair of the above pyrimidine dimers: this process is called
photoreactivation, and is mediated by photolyase enzymes that directly reverse DNA
damage in an error-free manner (Waterworth et al. 2002). Chlamle DNAs of algae and
plants are not immune to the impact of UV-irradiation, accumulating pyrimidine dimers
and 6-4 photoproducts. However, surprisingly, little or no repair of the pyrimidine dimers
of the chloroplast DNAs in Arabidapsis thaliana has been observed (Chen et al. 1996). In
contrast C. reinhardtii has been shown to possess photolyase activity in both the
chloroplast and the nucleus; and a gene (PHRZ) encoding the chloroplast photolyase has

been identified (Small and Greimann 1977; Small 1987; Petersen and Small 2001).

Specialized systems that are aimed at protecting the cell against the effects of
reactive oxygen species are represented by the MutT, the MutY and MutM systerm of E.
cali (outlined in Figure 1-1). While MutT acts to remove the mutagenic 8-oxoG, the
MutY and MutM gene products act to repair the damage caused due to the incorporation
of 8-oxodGTP opposite template C and A and subsequent replication events. The MutM
and MutY proteins are glycosylases: MutM acts upon several modified pru‘ines including
8-oxoG and also initiates base excision repair of 8-oxoG when paired with C. If
urrrepaired, subsequent mispain'ng of 8-oxoG with dATP would lead to G/C -) T/A
transversions (Carbera et a]. 1988; Boiteux et al. 2002). MutY is also a glycosylase that

removes adenine from A-G, A-8-oxoG and A-C mispairings, and thus a defective mutY

12

allele results in enhanced production of G/C -) T/A mutations (Fowler et a]. 2003).
Therefore, analysis of a collection of spontaneously arising base substitution mutations
can provide an insight into the relative efficiencies of different mutation avoidance

pathways that exist in chloroplasts.

Both the prokaryotic and eukaryotic types of MutM gene have been isolated in
Arabidapsis: a structural homolog of the bacterial MutM named AtMMH has been cloned
(Ohtsubo et a]. 1998) and the homolog to the eukaryotic gene 0001 has also been found
in Arabidapsis (Dany and Tissier 2001). The presence of both types of DNA glycosylases
has led to the hypothesis that the prokaryotic-type protein, AtMMH, may be translocated
into the organelles, most possibly to the chloroplast, although it lacks any obvious
localization signals for the organelle. However, it should be noted that the existing
programs (such as chloroP) that are used to predict the transit peptide sequences for the
chloroplast are not 100% accurate and therefore it is possrble that the prokaryotic
homolog of the Arabidapsis MutM may possess a chloroplast transit sequence that was

not detected by the existing prograrm.

The mismatch repair system (MMR) is one of the major DNA repair systems that
is responsible for recognizing and correcting mismatched bases that are incorporated
during replication. In addition, the mismatch repair proteins recognize numerous types of
DNA damage and either initiate or directly participate in the cellular responses to such
damage (Harfe and Jinks-Robertson 2000). The mismatch repair system is thus critical

for maintaining the overall integrity of the genetic material. Proteins involved in the

13

MMR system have been identified in both prokaryotes and eukaryotes, and the basic
features of the mismatch system have been highly conserved during the process of
evolution. The bacterial MMR system involves three proteins: MutS, MutL and MutH, of
which MutS and MutL protein are active as homodimers (Modrich and Lahue 1996).
Most eukaryotic organisms characterized to date possess multiple MutS homologs (MSH
proteins) and multiple MutL homologs (MLH proteins). A MutH functional homolog has

been found in humans (Bellacosa et al. 1999; Harfe and Jinks-Robertson 2000).

In plants, the investigation of the MMR system has only recently begun and both
MutS and MutL homologs have been identified and partially characterized from some
plant species. Six genes encoding MutS homologs have been identified in the
A rabidapsis thaliana genome (Culligan and Hays 2000); while two MutS honrologs have
been identified from Zea mays (Horwath et al. 2002). In addition, partial cDNAs of
putative MMR genes were isolated from Triticum aestivum (Korzun et al. 1999), rice,
Brassica napus and soybean (Horwath et al 2002). Three MutL honrologs are also
encoded by the A. thaliana genome, although none has been characterized functionally
(Jean et al. 1999; Harfe and Jinks-Robertson 2000). None of the isolated MutS or MutL
homologs from plants has a known chloroplast transit peptide sequence, and thus they
may be responsrble for the mismatch repair of nuclear genomes only. However, as stated
earlier, the prograrm that predict chloroplast transit peptide sequences are not completely
accurate in identifying all types of transit sequences and hence it is also highly possrble
that one of the MutS and MutL homologs identified in plants could be translocated to the

chloroplasts. Genes for components of the mismatch repair system are of particular

14

interest because they contribute to the fidelity of maintenance of repeat lengths at

microsatellites.

The rurderstanding of the molecular mechanisms of chloroplast DNA replication
or repair is still at a very rudimentary stage, and very little is known about the nuclear
genes that encode for these proteins. Since the chNA sequence data from C. reinhardtii
and land plants have shown that the genes required for its replication and repair are
absent in their respective plastid genomes, one of the goals for my research project was to
establish a system for identifying the nuclear genes that contribute to the maintenance of

chNA replication fidelity and or participate in DNA repair.

Use of mutators in the isolation of genes required for DNA maintenance

One of the strategies to dissect the mechanisms that are responsible for the
maintenance of the E. cali and yeast genomes has been through the isolation and
characterization of strains that have enhanced spontaneous mutation rates (also known as
mutators). The mutator phenotypes arise from defects in genes that act to minimize
genetic instability (Kunz et al. 1998). In this way, E. cali and yeast strains have been
isolated that show high rates of replication errors or defects in DNA repair.
Characterizations of the defective gene(s) have shown that these loci define crucial genes

for the mutation avoidance pathways.

In a review of all the different isolated mutators in E. cali, Miller (1998) points

out that genes involved in mismatch repair (mutH, mutL, mutS), excision repair (ung

15

encoding an uracil DNA glycosylase) and oxidative repair (mutY, mutT, mutM) of the
bacterial genome were defined using the mutator approach. In addition, the mutator loci
also identified genes that are responsrble for the maintenance of replication fidelity in
bacterial cells, such as those that code for the proofreading subunit of the polymerases or
for factors that are associated with polymerases and contrrbute their processivity such as
the genes for B—sliding clamp protein and the clamp loader, 7 complex (Krishnaswamy et

al. 1993).

Mutator alleles have also been defined from yeasts: these include genes that are
involved in proofreading, nrismatch correction, base excision and nucleotide excision
repair activities. Exonucleolytic proofreading-deficient mutants of the two replicative
polymerases (6 and 6) exhibit a mutator phenotype with a mutation rate up to 130-fold
higher over the wild type (Kunz et a]. 1998). Mismatch repair is defective in the mutator
strains due to the inactivation of any one of the PMS], MSHZ, MSH3, MSH6 and MLHI
genes. However, some of the mutators produced characteristic mutation patterns that did
not overlap, suggesting that the substrates may be different for each mismatch repair
complex. For example, nrore insertion/deletion mutations were produced in the mshZ
strain in contrast to the msh6 strain that produced mainly base pair substitutions
(Marsischky et al. 1996). Inactivation of the genes coding for the nucleotide excision
repair (RADI), recombinational repair (RAD52) or base excision repair (APNI) also
resulted in enhanced rates of specific kinds of mutations when corrrpared to the mutation
rates of wild-type cells. In addition to the mutators that affect nuclear genomic stability in

yeast, a yeast mutator strain with a defect in the exonucleolytic proofreading domain of

16

the mitochondrial polymerase was also isolated which showed increased levels of

spontaneous mutation rates of mitochondrial DNA (Foury and Vanderstraeten 1992).

Previous studies with chloroplast mutators

Nuclear genes that affect chloroplast mutation rates (plastome mutators) should
also include loci that play a role in mutation avoidance and replication fidelity of the
chNA. To date, plastome mutators have been isolated and characterized from three
plant species. Most of these mutations that occur in those strains appear as chlorophyll-
deficient sectors (reviewed by Banner and Sears 1986). The plastome mutators
dramatically increase the rates of spontaneous mutation in the chNA without inducing
any nuclear gene mutations. The plastome mutators can be divided into two categories:
they either cause the same type of mutational phenotype repeatedly (such as the
albastrians mutator of barley and iajap mutator of maize) or they can induce a wide
variety of mutations such as the plastome mutator in Oenathera (Epp 1973; Epp and

Parthasarathy 1987) and chloroplast mutator in barley (Prina 1992).

Among the first class of plastome mutators, the iajap gene (Ij) fi'om maize has
been isolated, cloned and sequenced, but the sequence has not illuminated the function of
the gene product (Bryne and Taylor 1996; Han et al. 1992). Thus it is still unknown how
a defect in the nuclear gene (Ij) produces striped plants with poorly developed
chloroplasts in the white sectors and lacking chloroplast ribosomes. The albastrians
mutant in barley is well characterized physiologically: white seedlings contain

rudimentary plastids which are photosynthetically inactive, have only trace amounts of

17

chlorophylls and lack ribosomes (Hess et al. 1992; Hedtke et a1. 1999). Molecular

characterization of the albastrians gene has not yet been reported.

The second class of plastorrre mutators produces a wide range of phenotypic
change (such as chlorophyll deficiencies, morphological abnormalities, sterility,
resistance to atrazine and ultrastructural irregularities of the chloroplasts). The range of
mutational phenotypes obtained indicates that multiple loci in the chloroplast DNA are
susceptible to mutations (Chiu et al. 1990; Sears and Sokalski 1991). The chloroplast
mutator locus of barley (cpm), which belongs to this second class of mutators, has been
characterized for chNA alterations. In one instance, it was shown that cpm caused a
nucleotide substitution in the chloroplast-encoded psbA gene (Rios et al. 2003).
However, the cpm locus has not been isolated or further characterized, and hence its

made of action in inducing chNA mutation is still unknown.

At the molecular level, the plastome mutator (pm) allele of Oenathera is the best
characterized among all the known plastome mutators. The plastome mutator was
originally isolated by Mel Epp in 1973, and Oenathera plants homozygous for the pm
allele exhibit a 1000-fold higher frequency of chNA mutations in comparison to the
wild-type plants (Epp 1973; Sears and Sokalski 1991). Short direct repeats and stretches
of adenines were observed to be the preferential substrates for plastome mutator (Blasko
et al. 1988; Chiu et al. 1990; Chang et al. 1996; Stoike and Sears 1998). The plastome
mutator locus was thus hypothesized to code for a protein that is responsible for

preventing elevated slippage of the chNA during replication, and thus play a role in the

18

avoidance of insertion/deletion mutations of the chNA molecules. However, the
plastome mutator locus has not been identified nor has the genetic defect been

characterized biochemically.

Mutator targets and assessment of m utation rates at these sites

Because microbial systems are much more tractable for establishing correlations
between mutations and their phenotypes in comparison to higher plants, my experimental
assay was designed to be similar to assays in E. cali and yeast that have been successful
in isolating loci that are involved in mutation avoidance pathways. From such studies it
has been observed that specific mutators in most cases accelerate particular types of
mutation events. For example, analysis of mutations in a mutator for the exonucleolytic
proofreading domain of polymerases in yeast shows increases only in single base pair
events (substitutions, insertions and deletions), whereas a mutation of the yeast gene
encoding the proliferating cell nuclear antigen (PCNA) destabilizes mostly short direct
repeat stretches or nricrosatellite sequences (Kokoska et al. 1999; Kunz et al. 1998). Thus
the choice of a mutator target would define the types of mutators that the assay would
generate. With this in mind, one of my research objectives was to isolate mutators in C.
reinhardtii that would display replication slippage rates that are higher than those

observed in the wild-type cells.

Studies with the Oenathera plastome mutator in our lab have identified potential

target sites for the pm gene: these included regions of the chNA that contained stretches

of short direct repeats, with the smallest repeat unit involved being a single base (in a

19

stretch of tandernly repeated adenines) (Chang et al. 1996; Stoike and Sears 1998).
Therefore as a starting point, I looked for mutators that specifically caused increases in
the rates of changes in the repeat lengths of microsatellites resulting from
insertion/deletion events. For this purpose, the naturally occurring microsatellites could
not be used as they occur mostly in the non-coding regions of the genome, and changes in
repeat lengths at those sites would not result in a visible phenotype. Thus, a reporter
construct containing a microsatellite as a slippage substrate was first created and then
introduced into an essential chloroplast encoded gene for my assay, the details of which
are explained in Chapters 4 and 5. The type of slippage substrate that was chosen had
been known to vary in the Oenathera plastome mutator lines (Wolfson et al. 1991; Sears

et al. 1994) and similar sequences are found in the chNA of other plants.

Expected mutator loci

Several factors affect the frequency of replication slippage in bacteria and yeast
during the replication process: these include the presence of stable hairpin structures in
the DNA that lead to increased stalling of the replication fork, defects in proteins that
ensru'e the processivity of polymerases such as the single-stranded binding protein (SSB)
or the sliding clamp and clamp loader for polymerase, helicases, etc. In Salmonella
typhimurium, deletion rates of short direct repeat increased in a line with a mutant ssb
allele (Mukailrara and Enomoto 1997). Similar observations have been made in E. cali
and also in the case of a replication helicase mutant, hep, suggesting that replication
slippage is facilitated by the pausing of the replication complex (Bierne et al. 1997).

Following replication, the newly replicated DNA is scrutinized by a mismatch repair

20

system that recognizes and corrects mismatches resulting from the replication errors
(Schaaper 1993). Inactivation of the general mismatch correction system in bacteria and
yeast has been shown to be associated predominantly with an increase in
insertion/deletion mutagenesis, indicating that the mismatch repair system is crucial to
the maintenance of stability of the microsatellites in the respective genomes (Schaaper
and Dunn 1987; Yang et al. 1999). Using the mutator assay I hope to identify similar
gene(s) involved in DNA replication and/or repair that act to nurintain stability of the

microsatellites of the chNA.

Overall objective of my thesis project

In order to understand the molecular mechanisms of chloroplast DNA mutations, I
decided to experimentally assess the rates of base substitutions and replication slippage in
the chNA of the unicellular eukaryote, Chlamydamanas reinhardtii. Since studies in
bacteria and yeast have shown the potential of isolating mutators for the understanding of
replication and repair processes, a subsequent goal was to isolate a library of chloroplast
DNA mutators. Characterization of these mutator strains would be very useful in the
elucidation of different genes that are involved in mutation avoidance and rmintenance of
the chloroplast genome. Traits of Chlamydamanas, outlined in the following section,

made it particularly ideal for this research project.

Chlamydomonas as the aqrerimental organism
In recent years the green unicellular biflagellated alga Chlamydamanas

reinhardtii has emerged as a powerful model system for studying a wide range of topics

21

in cellular and molecular biology: genetics and biogenesis of organelles, especially
chloroplasts, photosynthesis, flagellar assembly and functions, cell wall synthesis, mating
reactions and carbon, sulfur, nitrogen metabolism (Rochaix 1995). For these reasons, it is
often referred to as the photosynthetic yeast. For my research, Chlamydamanas also
appeared to be an ideal experimental organism: firstly, the introduction of slippage
constructs into the chNA of C. reinhardtii can be accomplished because its chloroplast
DNA can be successfully transformed (Boynton et al. 1988). The transformation of
chNA through biolistics occurs through homologous gene replacement and several
markers are available for selection of the transformants (Goldschmidt-Clermont 1991).
Secondly, the reporter construct that will be used to monitor slippage events in the
chNA (details described in Chapter 4) offers a strong phenotypic selection, and this

would allow the slipped cells to be readily distinguishable from the non-slipped cells.

Determining the rates of spontaneous mutation frequencies for slippage events
and base substitutions requires the growing and handling of large numbers of cells.
Therefore, C. reinhardtii was ideal for this purpose as large numbers of cells could be
plated on few plates, allowing me to screen millions of cells in search of the rare
mutation event. Once the mutation rates were established in the appropriate reporter
strains, plasmid-based nuclear gene disruptions were performed to create the chloroplast
mutator phenotype. Procedures for the nuclear transformation of Chlamydamanas needed
for this part are well established (Kindle 1990), and dominant selectable markers are also
available for this purpose (Stevens et al. 1996). The isolated mutator phenotype was

confirmed through crosses and tetrad analysis, and future experiments using plasmid

22

rescue are being done to isolate the disrupted nuclear gene that caused the mutator
phenotype. The completed sequence of the nuclear genome of C. reinhardtii and the EST

database will then allow the recovery of the wild-type gene.

In summary, my studies provided the first experimental assessment of chNA
mutation rates that occur spontaneously in viva in C. reinhardtii. The frequencies of
different types of base substitution mutations obtained from analyzing the mutants
provided an insight into the relative frequencies of transition and transversion events that
occur in the chNA. In addition, the assessment of relative rates of base substitutions and
replication slippage events allowed us to make deductions about the efficiencies by which
they are generated and/or repaired in the chNA. Furthermore, I was able to establish a
mutator assay that was successful in the isolation of a chloroplast mutator in C.
reinhardtii that shows elevated levels of replication slippage. Characterization of the
disrupted gene in this mutator line will hopefully lead us to the identification of the first

gene involved in mutation avoidance of the chNA in C. reinhardtii.

23

REFERENCES

Bellacosa A, Cicchillitti L, Schepic F, Riccio A, Yeung AT (1999) MED], a novel
human methyl-CpG-binding endonuclease, interacts with DNA mismatch repair protein
NILHI. Proc Natl Acad Sci USA 96: 3969-3974

Bieme H, Seigneur M, Ehrlich SD, Michel B (1997) uer mutations enhance tandem
repeat deletion in the Escherichia cali chromosome via SOS induction of the RecF
recombination pathway. Mol Microbi0126: 557-567

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 Oenathera haakeri
strain Johansen. Curr Genet 14: 287-292

Boiteux S. Gellon L, Guibort N (2002) Repair of 8-oxoganine in Saccharamyces
cerevisiae: Interplay of DNA repair and replication mechanisms. Free Radic Biol Med
32(12): 1244-53

Bomer T, Sears BB (1986) Plastome mutants. Plant Mol Biol Reporter 4: 69-92

Boynton JE, Gillham NW, Harris EH, Hosler JP, Jolmson AM, Jones AR, Rarrdolph-
Anderson BL, Robertso D, Klein TM, Shark KB, Sanford JC (1988) Chloroplast
tramsformation of Chlamydamanas with high velocity microprojectiles. Science 240:
1534-1538

Britt AB (1996) DNA damage and repair in plants. Annu Rev Plant Physiol 47: 75-100
Byme M, Taylor WC (1996) Analysis of mutator induced mutations in the iajap gene of
maize. Mol Gen Genet 252: 216-220

Cannon GC, Heinhorst S (1990) Partial purification and characterization of a DNA
helicase from chloroplasts of Glycine max Plant Mol Biol. 1990 3: 457-464.

Carbera M, Nghiem Y, Miller JH (1988) mutM, a second mutator locus in Escherichia
cali that generates G:C -) T:A transversions. J Bacteriol 170(11): 5405-5407

24

Cavalier-Smith T (1992) The number of symbiotic origins of organelles. Biosystems 28:
91-106

Cerutti H, Osrmn M, Grandoni P, Jagendorf T (1992) A homolog of Escherichia cali Rec
A protein in plastids of higher plants. Proc Natl Acad Sci USA 89: 8068-8072

Cerutti H, Jolmson AM, Boynton JE, Gillham NW (1995) Inhibition of chloroplast DNA
recombination and repair by dominant negative mutants of Escherichia cali RecA. Mol
Cell Biol 15: 3003-3011

Chang TL, Stoike LL, Zarka D, Schewe G, Chiu WL, Jarrell DC, Sears BB (1996)
Characterization of primary lesions caused by the plastome mutator of Oenathera. Curr
Genet 30: 522-530

Chen JJ, Jiang CZ, Britt AB (1996) Little or no repair of cyclobutyl pyrimidine dimers is
observed in the organellar genomes of the young Arabidopsis seedling. Plant Physiol
1 l 1: 19-25

Chiu WL, Johnson EM, Kaplan SA, Blasko K, Sokalski MB, Wolfson R, Sears BB
(1990) 0enathera chloroplast DNA polymorphisms associated with plastonre mutator
activity. Mol Gen Genet 221: 59-64

Chiu WL, Sears BB (1992) Electron microscopic localization of the replication origins in
Oenathera chloroplast DNA. Mol Gen Genet 232: 33-39

Clegg MT, Gaut BS, Learn GH, Morton BR (1994) Rates and patterns of chloroplast
DNA evolution. Proc Natl Acad Sci USA 91: 6795-6801

Culligan KM, Hays JB (2000) Arabidopsis Mut S homologs-AtMSHZ, AtMSH3,
AtMSH6, and a novel AtMSH7- form three distinct protein heterodimers with different
specificities for mismatched DNA. Plant Cell 12: 991-1002

Dany AL, Tissier A (2001) A functional OGGI homologue from Arabidopsis thaliana.
Mol Genet Genomics 265(2): 293-301

25

Epp MD (1973) Nuclear gene-induced plastonre mutations in Oenathera haakerii 1.
Genetic analysis. Genetics 75: 465-483

Epp MD, Parthsarathy MV (1987) Nuclear gene-plastonre mutations in Oenathera
haakerii H. Phenotype description with electron microscopy. Amer J Bot 74: 143-151

Fowler RG, White SJ, Koyama C, Moore SC, Durm RL, Schaaper RM (2003)
Interactions arrrong the Escherichia cali mutT, mutM, and mutY damage prevention
pathways. DNA Repair 2: 159-173

Foury F, Vanderstraeten S (1992) Yeast mitochondrial DNA mutators with deficient
proof-reading exonucleolytic activity. EMBO J 11: 2717-2726

Fukuta H, Mochida A, Maruyarna N, Fukasawa H (1991) Chloroplast DNA
topoisomerase I from cauliflower. J Biochem 109: 127-131

Gaikwad A, Hop DV, Mukherjee SK (2002) A 70-kDa chloroplast DNA polymerase
from pea (Pisum sativum) that shows high processivity and displays moderate fidelity.
Mol Genet Genornics 267(1):45-56

Gallego F, Fleck 0, Li A, Wyrzykowska J, Tinland B (2000)AtRAD1, a plant
homologue of human and yeast nucleotide excision repair endonucleases, is involved in
M repair of UV damages and recombination. The Plant J 21: 507-518

Gielly L, Taberlet P (1994) The use of chloroplast DNA to resolve plant phylogenies :
noncodong versus rbcL sequences. Mol Biol Evol 11: 769-777

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

Goldschmidt-Clerrnont M (1991) Transgenic expression of aminoglycoside adenine
transferase in the chloroplast: a selectable marker of site-directed transformation of
Chlamydamanas. Nucl Acid Res 19: 4083-4089

Gray MW (1992) The endosymbiont hypothesis revisited. International Review of Cytol
141: 233-357

26

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

Harfe BD, Jinks-Robertson S (2000) DNA mismatch repair and genetic instability. Annu
Rev Genet 34: 359-399

Heinhorst S, Cannon GC, Weissbach A (1990) Chloroplast and mitochondrial DNA
polymerases from cultured soybean cells. Plant Physiol 92: 939-945

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

Hiratsuka J, Sharnida H, Whittier R, Ishibashi T, Sakamoto M, Mori M, Knodo C, Honji
Y, Sun CR, Meng BY, Li YQ, Kano A, Nishizawa Y, Hirai A, Shinozaki K, Sugiura M
(1989) The conrplete 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

Horwath M, Kramer W, Kunze R (2002) Structure and expression of the Zea mays mutS-
homologs Musl and Mus2. Theor Appl Genet 105: 423-430

Jean M, Pelletier J, Hilpert M, Belzile F, Kunze R (1999) Isolation and characterization
of AtMLHl , a MutL homologue from Arabidopsis thaliana. Mol Gen Genet 262: 633-
642

Keim CA, Mosbaugh DW (1991) Identification and characterization of a 3' to 5'
exonuclease associated with spinach chloroplast DNA polymerase. Biochemistry 30(46):
1 1 109-1 1 1 18

Kindle KL (1990) high frequency nuclear transformation of Chlamydamanas reinhardtii.
Proc Natl Acad Sci USA 87: 1228-1232

Kokoska RJ, Stefanovic L, Buermeyer Abrn Liskay RM, Petes TD (1999) A mutation of
the yeast gene encoding PCN A destabilizes both nricrosatellite and minisatellite DNA
sequences. Genetics 151: 511-519

27

Kolodner R, Tewari KK (1975a) Denaturation mapping studies on the circular
chloroplast deoxyribonucleic acid from pea leaves. J Biol Chem 250: 4888-4895

Kolodrrer R Tewari KK (1975b) Chloroplast DNA fi'om higher plants replicates by both
the Cairns and rolling circle mechanism. Nature 236: 708-711

Korzun V, Bomer A, Siebert R, Malyshev S, Hilpert M, Kunze R, Puchta H (1999)
Chromosomal location and genetic mapping of the mismatch repair gene homologs
MSHZ, MSH3 and MSH6 in rye and wheat. Genome 42: 1255-1257

Krishnaswamy S, Rogers JA, Isbell RJ, Fowler RG (1993) The high mutator activity of
the dnaQ49 allele of Escherichia cali is mediunr-dependent and results from both
defective 3 -) ' proofreading and methyl-directed mismatch repair. Mutat Res 288: 311-
319

Kunz BA Rarnachandran K, Vonarx EJ (1998) DNA sequence analysis of spontaneous
mutagenesis in Saccharamyces cerevisiae. Genetics 148: 1491-1505

Marsischky GT, filosi N, Kane MF, Kolodrrer R (1996) Redundancy of Saccharamyces
cerevisiae MSH3 and MSH6 in MSHZ-dependent mismatch repair. Genes Dev 10: 407-
420

Maul JE, Lilly JW, Cui L, dePamphilis CW, Miller W, Harris EH, Stern DB (2002) The
Chlamydamanas reinhardtii plastid chromosome: Islands of genes in a sea of repeats.
The Plant Cell 14: 2659-2679

McKown RL, Tewari KK (1984) Purification and properties of a pea chloroplast DNA
polymerase. Proc Natl Acad Sci USA 81: 2354-2358

Miller JH (1998) Mutators in Escherichia cali. Mutat Res 409: 99-106

Modrich P, Lahue R (1996) Mismatch repair in replication fidelity, genetic
recombination and cancer biology. Annu Rev Biochem 65: 101-133

Mukaihara T, Enomoto M (1997) Deletion formation between the two Salmonella
typhimurium flagellin genes encoded on the mini F plasmid: Escherichia cali ssb alleles

28

enhance deletion rates and change hot-spot preference for deletion endpoints. Genetics
145: 563-572

Mukherjee SK, Reddy MK, Kumar D, Tewari KK (1994) Pmification and
characterization of a eukaryotic type I topisomerase from pea chloroplast. J Biol Chem
269: 3793-3801

Muse SV (2000) Examining rates and patterns of nucleotide substitution in plants. Plant
Mol Biol 42: 25-43

Nielsen BL, Rajasekhar VK, Tewari KK (1991) Pea chloroplast DNA primase:
characterization and role in initiation of replication. Plant Mol Biol 16: 1019-1034

Ohtsubo T, Matsuda O, Iba K, Terashirna I, Sekiguchi M, Nakabeppu Y (1998)
Molecular cloning ofAtMMH, an Arabidopsis thaliana ortholog of the Escherichia cali
mutM gene, and analysis of functional domains of its product. Mol Gen Genet 259: 577-
590

Palmer JD (1985) Evolution of chloroplast and mitochondrial DNA in plants and algae.
Molecular evolutionary genetics (ed. MacIntyre RJ) Plenum Publ Corp, 131-240

Petersen J], Small GD (2001) A gene required for the novel activation of a class [1 DNA
photolyase in Chlamydamanas. Nucl Acids Res 29: 4472-4481

Prina AR (1992) A mutator nuclear gene inducing a wide spectrum of cytoplasmically
inherited chloroplast deficiencies in barley. Theor Appl Genet 85: 245-252

Provan J, Soranzo N, Wilson NJ, Goldstein DB, Powell W (1999) A low mutation rate
for chloroplast nricrosatellites. Genetics 153: 943-947

Rios D, Saione H, Robredo C, Acevedo A, Colombo N, Prina R (2003) Isolation and
molecular characterization of atrazine tolerant barley mutants. Theor Appl Genet 106:
696-702

Rochaix JD (1995) Chlamydamanas reinhardtii as the photosynthetic yeast. Annu Rev
Genet 29: 209-230

29

Sala F, Amineli AR, Parisi B, Spadari S (1980) A garmna-like DNA polymerase in
spinach chloroplasts. Eur J Biochem 112: 211-217

Schaaper RM, Dunn RL (1987) Spectra of spontaneous mutations in Escherichia cali
strains defective in mismatch correction: the nature of in vivo DNA replication errors.
Proc Natl Acad Sci U S A. 84(17): 6220-6224

Schaaper RM, Drmn R1 (1991) Spontaneous mutation in the Ecsherichia cali lac I gene.
Genetics 129: 317-326

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

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

Shinozaki K, Ohme M, Tanaka M, Wakasugi T, Hayashida N, Matsubayashi T, Zaita N,
Chiunwongse J, Obokata J, Yamaguchi-Shinozaki K, Ohto C, Torazawa K, Meng BY,
Sugita M, Deno H, KAmogashira T, Yarmda K, Kusuda J, TakaiwaF, KatoA, Tohdoh N,
Shimada H, Sugiura M (1986) The complete nucleotide sequence of the tobacco
chloroplast genome: its gene organization and expression. EMBO J 5: 2043-2049

Small GD, Greimann CS (1977) Photoreactivation and dark repair of ultraviolet light-
induced pyrimidine dimers in chloroplast DNA. Nucl Acids Res 4: 2893-2902

Small GD (1987) Repair systems for nuclear and chloroplast DNA in Chlamydamanas
reinhardtii. Mutat Res 181: 31-35

Stevens DR, Rochaix J-D, Purton S (1996) The bacterial phleomycin gene ble as a
dominant selectable marker in Chlamydamanas. Mol Gen Genet 251: 23-30

Stoike LL, Sears BB (1998) Plastame mutator-induced alterations arise in Oenothera
chloroplast DNA through template slippage. Genetics 149: 347-353

Wadell J, Wang XM, Wu M (1984) Electron microscopic localization of the chloroplast
DNA replicative origins in Chlamydamanas reinhardtii. Nucl Acids Res 12: 3843-3856

30

Wang ZF, Yang J, Nie ZQ, Wu M (1991) Purification and characterization of a garnma-
like DNA polymerase from Chlamydamanas reinhardtii. Biochemistry 30: 1127-1131

Waterworth WM, Jiang Q, West CE, Nikaido M, Bray CM (2002) Characterization of
Arabidopsis photolyase enzymes and analysis of their role in protection from ultraviolet-
B radiation] Exp Bot 53: 1005-1015

Woefle MA, Thompson RJ, Mosig G (1993) Roles of novobiocin-sensitive topisonrerases
in chloroplast DNA replication in Chlamydamanas reinhardtii. Nucl Acids Res 21: 4231-
4238

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among
plant mitochondrial, chloroplast and nuclear DNAs. Proc Natl Acad Sci USA 84: 9054-
9058

Wolfe KH, Morden CW, Palmer JD (1992) Function and evolution of a minimal plastid
genome ham 3 non-photosynthetic parasitic plant. Proc Natl Acad Sci USA 89: 10648-
10652

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

Yang Y, Karthikeyan R, Mack SE, Vonarx EJ, Kunz BA (1999) Analysis of yeast pmsI,
MM, and mlhI mutators points to differences in mismatch correction efficiencies
between prokaryotic and eukaryotic cells. Mol Gen Genet 261(4-5): 777-87

Zurawski G, Clegg MT (1987) Evolution of higher plant chloroplast DNA encoded
genes: implications for structure-function and phylogenetic studies. Annu Rev Plant
Physiol 38: 391-418

31

CHAPTER 2

Chloroplast mutations induced by 9-aminoacridine hydrochloride

are independent of the plastome mutator in Oenothera

Note. The content of this chapter has been accepted for publication in the Theoretical and

Applied Genetics as Guha Majumdar M, Baldwin S, and Sears BB 2003.

I would like to acknowledge that I have contributed to the molecular data
presented in this chapter, while the initial mutagenesis and phenotypic characterizations
were performed by Sue Baldwin. In specific, I have done all the DNA isolations, PCR
amplifications and Southern hybridizations that are outlined in the Chapter and also have

contributed fully to the writing and compilation of all figures and tables of the

manuscript.

32

Abstract

Oenothera plants homozygous for the recessive plastome mutator allele (pm)
show chloroplast DNA (chNA) mutation frequencies that are about 1000-fold higher
than spontaneous levels. The pm-encoded gene product has been hypothesized to have a
function in chNA replication, repair and! or mutation avoidance. Chemical mutagenesis
studies with the alkylating agent nitroso-methyl urea (NMU) showed a synergistic effect
of NMU on the induction of mutations in the pm line, suggesting an interaction between
the pm-encoded gene product and one of the repair systems that corrects alkylation
damage. To examine whether the pm-activity extends to the repair of damage caused by
non-alkylating mutagens, 9-aminoacridine hydrochloride (9AA) was tested for synergism
with the plastome mutator. A statistical analysis of the data reported here indicates that
the pm-encoded gene product is not involved in the repair of the 9AA-induced mutations.
However, the recovery of chlorotic sectors in plants derived from the mutagenized seeds

shows that 9AA can act as a mutagen of the chloroplast genome.

Introduction

Mutators serve as genetic tools for the study of mutation avoidance and repair
pathways (Miller 1998). Bacterial mutator genes encode proteins that contribute to the
fidelity of DNA replication (e. g. mut D and pal A in E. cali) as well as those that are
involved in DNA repair (such as genes of the mismatch repair system mut H, L and S in
E. cali). Studies with chloroplast mutator systems have sought similar characterizations

of loci involved in chloroplast DNA (chNA) replication and/ or repair pathways. One

33

such mutator locus is defined by the nuclear-encoded plastome mutator (pm) allele in

Oenothera.

The plastome mutator induces a high frequency of non-Mendelian chloroplast
mutations when the pm allele is homozygous (Epp 1973; Sears and Sokalski 1991). Due
to the high frequency of chNA mutations by the plastome mutator, the pm-encoded
gene product has been hypothesized to have a firnction in chNA replication and/or
repair. Studies by Blasko et al. (1988) and Chiu et al. (1990) showed restriction fiagment
length polymorphisms (RFLPs) in the chloroplast DNA (chNA) specific to the pm line.
Subsequently, mutations caused by the plastome mutator were shown to occur at chNA
nricrosatellite regions, resulting in the deletion or duplication of repeat(s) probably

through replication slippage (Chang et al. 1996; Stoike and Sears 1998).

In order to test whether a general defect in chloroplast DNA repair or mutation
avoidance is responsible for the plastome mutator activity, Sears and Sokalski (1991)
looked for synergistic effects of ultraviolet light (UV-light) and the chemical mutagen,
nitroso-methyl urea (NMU) in several pm lines. No synergism in mutation induction was
noted when the pm lines were irradiated with UV-light, allowing the conclusion that the
pm-encoded gene product is not involved in the repair of UV damage. In contrast, when
the potent plastome mutagen NMU is applied to the pm lines, extremely high frequencies
of mutation result, indicating a synergistic effect. Thus the alkylation damage caused by
NMU is probably repaired by a pathway that has the pm-encoded gene product as a

component (Sears and Sokalski 1991).

34

To examine whether the plastome mutator activity extends to the repair of
damage caused by non-alkylating mutagens, we chose to test 9-aminoacridine
hydrochloride (9AA) for synergism with the plastome mutator. 9AA intercalates between
the stacked nitrogen bases at the core of the DNA double helix (N asim and Brychcy
1979). The intercalated 9AA mimics base pairs and thus causes deletions and additions of
bases upon replication. In Salmonella (Hoffman et al. 1989; Kopsidas and MacPhee
1994) and E. coli (Thorms and MacPhee 1985; Gordon et al. 1991), 9AA causes single
base-pair deletions or duplications that result in frameshift mutations, with preference for
runs of mononucleotides (Skopek and Hutchinson 1984). Since an oligo-A stretch has
been found to be a plastome mutator target in Oenothera (Chang et. al. 1996), 9AA

seemed to be a good choice to test for potential synergism with the plastome mutator.

In plants such as barley and onion, the application of 9AA has been reported to
cause chlorotic mutations and growth retardation (D’Amato 1950; D’Amato 1952).
Therefore, as a first step in this study, our goal was to determine whether 9AA could
cause similar mutations in Oenothera and/or could act specifically as a chNA mutagen.
Once it was established that 9AA could cause mutations in the chNA, the next goal was

to determine whether it could act synergistically with the plastome mutator.

35

Materiab and Methods

Plant material

Homozygous plants of Oenothera haakeri strain Johansen containing plastomes I, II and
IV were constructed originally by Professor W. Stubbe (University of Dusseldorf), and
the lines were maintained by self-pollination. The Johansen strain containing either
plastome I or H was used as the green parent in crosses with the variegated mutant A-E.
Self-pollinations of the prm line result in the accumulation of mutations (Chiu et al.
1990). Hence, efforts were taken to obtain a prm line with a low background mutation
rate. These seeds were produced by crossing a +/pm line as the female parent with a
prm line as the pollen donor which should result in an equal mixture ofpm/+ and

prm seeds, with most plastids transmitted fiom the female parent.

9-aminoacridine hydrochloride mutagenesis

After a four-hour irnbibition period, seeds were surface-sterilized by incubating them in a
solution of 50% bleach and 0.1% SDS for 30 minutes, followed by one rinse with 0.01 N
HCl and several rinses with sterile water. To establish a dose cru've, 9AA (Sigma) was
added at specified final concentrations (see Results) for 0.5 hours, 8 hours, 16 hours and
32 hours. The seeds were swirled gently throughout the incubation period, and rinsed
several times before placing them in a beaker of sterile water where they were allowed to

germinate. Control seeds were swirled for corresponding time periods in water.

36

Maintenance and scoring of plants

All the lines were maintained by leaf tip cultures and by self pollinations as described by
Stubbe and Herrmann (1982). Seed germination medium was made according to Chiu et
al. (1990). For the mutagenesis experiments, seedlings were planted in multipots
(Hunrrnert Seed Co); the trays were placed under aquarium lights (GE) and were

checked periodically for chlorotic sectors and other developmental abnormalities.

Transmission Electron Microscopy
Preparation, sectioning, and microscopy of leaf samples were performed by the MSU

Center for Electron Optics.

Statistical Analyses

To discern whether the 9AA treatment significantly altered the mutation fiequency in the
mixed pm/pm, pm/+ line, the data contained in Table 2-3 were analyzed by applying the
Chi-Square goodness-of-fit test to the numbers of observed and expected mutations. If
the chemical and the genetic condition independently cause mutations, then the following

formula should predict the expected number of mutations:

[(CN) + (NP)] - [CP (V2 N)] = Expected

where N is the population size of the pm/pm, pm/+ seeds, C is the frequency of mutations
in the chemically treated wild-type plants, and P is the observed frequency of mutations

of pm/pm, pm/+ alone without any chemical treatment. CP would give the expected value

37

of both causing a mutation in the same plant, but is applicable to only half of the

population, as half of the seeds are prm.

Total DNA isolation

Total DNA was isolated from Oenothera leaves using a protocol adapted fi‘om David

Jarrell (unpublished results) and also from Stoike and Sears (1998).

Isolation of DNA fi'om a mitochondria-enriched fraction

For Southem analysis, DNA was isolated from a mitochondria-enriched fraction using a
protocol adapted from Sears et al. (1989). The leaf tissue from the plants was first ground
using a 10-fold vol/wt ratio of homogenization medium (50 mM Tris, 6% Sorbitol, 6 mM
EDTA, 0.1% BSA, 0.3% polyvinyl polypyrrolidone, 1 mM Ascorbic Acid and 3 mM
Cysteine) to leaf tissue. The mixture was filtered through two layers of cheese cloth and
one layer of nriracloth, and the filtrate was centrifuged in a Sorvall centrifuge using a
GSA rotor at 5900g for 6 nrinutes to remove chloroplasts and starch. The supernatant was
centrifuged at 25,000g to pellet the mitochondria The mitochondrial pellet was then
resuspended in 3 ml CTAB buffer and DNA was extracted using the protocol described

above.

38

PCR amplification

The oriB primers and conditions for amplifying the 16S rRNA - tmI region of the
chNA of Oenothera are descrrbed by Stoike and Sears (1998), except for the annealing
temperature, which was set at 48°C. The PCR products were run on a 1% agarose gel
along with a 123-hp DNA ladder (GIBCO BRL) and photographed using a Polaroid
camera. The gel picture was digitally scanned using a Hewlett Packard ScanJet 4C and

displayed using the Adobe Photoshop software (San Jose, CA).

Southern analysis

The DNA from the mitochondria-enriched fraction was digested with Barn HI (GIBCO
BRL) and then electrophoretically separated on a 0.8% agarose gel at 60 V for
approximately 8 hours. The DNA was transferred overnight to a nylon membrane
according to the method of Sarnbrook et al. (1989). The membrane was probed with a
radioactively labeled CMS-Sprite cosmid clone 651-12C2 (containing 12 random
mitochondrial DNA fragments from Phaseolus vulgaris) using the standard nick
translation method described by Sarnbrook et al. (1989); the 32P-dATP used for the
labeling was obtained from Andotek Sciences Co. The filter was washed according to the
method described by Sambrook et al. (1989), and then exposed to film (Kodak X-O-Mat)

at -80°C for 17 hours.

39

Results
9-aminoacridine hydrochloride dosage trials on wild type line

In Oenothera, different plastid DNA (plastome) types (I, H, 1H and IV) are
transnritted differentially in crosses and also have specific RFLPs associated with them
(Chiu et al. 1988; Chiu et al. 1990). Previous studies have shown that the different
plastome types are equivalent targets for the plastome mutator (Sears and Sokalski 1991;
Chang et al. 1996). The different plastonre types also responded similarly to preliminary
9AA dosage trials to determine the concentration and the time interval that would result
in the highest frequency of mutations. Complete data from the dosage trials from only
plastome H are reported here (Table 2-1) since the mutant plant (A-E) used for firrther

characterizations was isolated from one of these trials.

In two sets of experiments, 500 seeds were treated with 0, 2 or 10 ug/ml
concentrations of 9AA for 0.5 — 32 hours (Table 2-1). Mutant sectors were observed at
both the 2 and 10 ug/ml concentrations when the seeds were exposed for 16.5 hours. In
the preliminary trials, mutations were observed in seeds incubated for other periods of

time, but their most frequent occurrence was always after a 16-hour treatment.

40

Table 2-1. Sector frequencies for wild type seedlings containing plastome II treated with 0, 2 and
10 ug/rnl concentrations of 9AA for different exposure times. Each trial contained 500 seeds,

except for one treatment“, which contained 512 seeds.

 

9AA (pg/ml) Exposure time % Germination % Viability Mutation
(hours) frequencyb

Experiment

1 32 41.8 40.2 0

O

10 1" 100 55.9 0

10 6.5 72.6 53.4 0

10 16.5 93.2 68.2 3/318 (0.9%)

10 32 80.6 59.3 0

Experiment 2

0 32 92.0 63.9 0

2 1 90.4 83.8 0

2 6 90.2 83.3 0

2 16.5 75.2 93.6 3/352 (0.9%)

2 32 92.6 86.2 0

bMutation frequency = Number of seedlings with sectors
Number of viable seedlings

41

Mutant phenotypes from acridine treatment

The 9AA treatment produced a spectrum of pigment mutations in Oenothera, with
white, yellow and light green mutations being observed as both solid and mottled sectors
in leaves. Developmental anomalies such as abnormal leaves with fused midribs and
plants without trichomes were also observed in the 9AA treated seedlings, but these were
transient and never seen beyond the second set of leaves. Because so few mutants were
recovered, only one segregated into the germline and could be analyzed by crossing. This

mutant was labeled A-E and had pale yellow sectors.

Transmission electron microscopy

The mutant A-E was examined by transmission electron microscopy (TEM) to
characterize organelle ultrastructure. The plastids in the mutant sectors contained swollen
vesicles along with either no (Figure 2-1 A) or few grana stacks (Figure 2-1 B). In

contrast, the mitochondria in the mutant sector appeared normal (Figures 2-1 A and B).

Inheritance of the acridine-induced mutation

Initial inheritance studies showed non-Mendelian inheritance and vegetative
segregation of the A-E trait, indicating that the lesions caused by 9AA could reside either
in the chNA or mitochondrial DNA (mtDNA). In Oenothera, the differential pattern of
transnrission of an organelle mutation through a cross can be used to deterrrrine whether
the mutation lies in the chNA or mtDNA The mtDNA is inherited only from the

maternal parent (Brennicke and Schwemrnle 1984), whereas the chNA can be inherited

42

 

O
p
i O
‘
‘ O
I: v,‘
‘U.’a.a’

.rl

' 9

o .s‘. --.

 

r
‘
I-
-o-;

 

 

 

 

 

 

Figure 2-1. A—B Electron micrographs of plastids (P) with aberrant thylakoid structures and

nritochondria (M) from the mutant yellow sectors of A-E.

43

from both parents (Stubbe and Herrmann, 1982). Therefore, when the mutant plant is
used as the source of pollen, the progeny should not inherit the mutation if it is in the

mtDNA.

In Oenothera plants with non-Mendelian mutations, when a leaf containing a full
periclinal chimera occurs at a floral node, the germline of the adjacent flower consists
entirely of mutant organelles (Kutzelnigg and Stubbe 1974). However, the mutant plant
A-E had a partial periclinal chimera with both wild-type and mutant organelles present in
the gemrline. In crosses using mutant A-E as the maternal parent, 103 out of the 205
progeny inherited the mutant organelle, and most of the progeny were variegated (Table
2-2). In the reciprocal crosses with A-E as the male parent, no variegated progeny were

recovered.

Ca-segregation of the mutant phenotype analyses using molecular markers

Data from the TEM studies had indicated that the chloroplast was the affected
organelle in the 9AA treated mutant plant A-E, whereas the transmission patterns of the
mutation in reciprocal crosses suggested that the 9AA-induced nrutation could be in the
mtDNA. However, in Oenothera, when a weak plastonre type is used as the male parent
in a cross, the plastids may be transmitted only or predominantly from the maternal
parent (Chiu et al. 1988). Therefore it could not be determined with confidence which
organelle genome carried the 9AA-induced mutation. In order to establish whether the

mutation in A-E co—segregated with the chNA, the green and yellow sectors from

44

Table 2-2. Crosses to establish the inheritance pattern of the 9AA-induced organelle narration in
A-E. The table lists the number of seeds obtained and the fraction that gernrinated. The F,

progeny showed 100% viability, but varied in the inheritance of the A-E mutation.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Progeny phenotypes
a # of ‘V
Parents seeds Gain. % Progeny With
Fr mutant tissue
Maternal Paternal Mutant Variegated Green N
Variegated Green 396 37.6 2 68 79 149
(II) (1)
V2 mutant 46.5 76
V2 green
Variegated Green 100 32 2 17 13 32
(11) (IV)
‘/2 mtant 59.4 x
‘/2 green
Variegated Green 24 100 5 9 10 24
(11) (IV)
‘/2 mtant 58.3 at
V2 green
Green Varieg- 390 87.7 0 0 342 342
(I) ated
(II) 0 74
< ‘/2
mutant
> '/2
gm
Green Varieg— 109 91.7 0 O 100 100
(IV) ated
(II) 0 74
‘/2
mutant
‘/2 green

 

 

 

 

 

 

aA variegated plant with a heteroplasmic germline transmitted the A-E organelle nartation; the
other parent contributed the green wild-type plastids, with Roman numerals indicating the
plastome types.

45

the variegated progeny were dissected and DNA was extracted separately, so that a co-

segregation analysis could be conducted.

Chloroplast DNA marker: To determine whether the yellow sectors in the progeny fi'om
the cross (A-E X wild type) co-segregated with the physical markers on the chNA, PCR
anrplification using primers flanking the oriB region was performed with DNAs fi'om the
green and yellow sectors of the progeny as well as from the wild type and the mutant
parent (A-E). Figure 2-2 shows that the chNA marker in the yellow tissue in the
progeny plant (lanes 5 and 6) is identical to the band of plastome type (H) from the
mutant parent A-E plant (lane 2). The green tissue fi'om the progeny plant (lanes 3 and 4)
showed heteroplasmy as indicated by the presence of PCR products similar to those from
both the wild type and the mutant parent (lanes 1 and 2 respectively). A middle band was
also seen in the DNAs amplified from the green sectors; this third band is an
amplification artifact of the PCR reaction, also observed from a mixing experiment using

two different plastome types (not shown).

Mitochondrial DNA (mtDNA) timer: Southern analysis was conducted to test whether
the unique mtDNA band of the mutant parent A-E (indicated by an arrow in Figure 2-3)
cosegregated with the mtDNA of the yellow sectors in the progeny plants. For this
purpose, DNAs from the mitochondrial-enriched fractions of the same samples used for
the chNA analysis were extracted These DNAs were digested and then blotted onto a
membrane that was hybridized to a labeled mitochondrial cosmid probe (see Materials

and Methods). Southern blot analysis showed no difference between the yellow and the

46

MW 1 2 3 4 5 6
984—
861—

615—

 

Figure 2-2. PCR analysis of 16S rRNA-mt! spacer region. From left to right on the gel,
primers amplified through the 16S rRNA-rm] region from the following Oenothera
chNAs: lane I green wild—type (wt) parent with plastome type I; lane 2 mutant
variegated parent A-E, with plastome type 11; lanes 3 and 4 replicate sanples of green
tissue from a progeny of the cross, A—E X wt; lanes 5 and 6 replicate samples of yellow
sectors from a progeny of the cross, A-E X wt. The lane MW shows the DNA size
rmrkers (in bp) from the GIBCO—BRL 123—bp DNA ladder.

47

 

Figure 2-3. Southern hybridization analysis using DNA from the mitochondria-emiched
fraction, extracted from the green and yellow sectors of the progeny and also from the
parents of the A-E X wt cross. The numbered lanes contain Bam HI—digested DNAs from
the following plants: lane 1 green wild-type parent; lanes 2 and 3 green sectors of the
progeny from the cross; lane 4 yellow sectors from the progeny and lane 5 the mutant
variegated parent, A—E (5). The mtDNA of the two parents used for the cross is
distinguished by a single band (indicated by an arrow) that is exclusive to the rmtant
parent A-E.

48

green sectors from the progeny plant (lanes 2, 3 and 4), as all contained the band

exclusive to the mtDNA of the maternal parent, A-E (lane 5); a result which is consistent
with the strict maternal inheritance of the mtDNA. Although there was no RFLP between
the nrtDNA from the yellow and green sectors, the existence of a point mutation between
them cannot be ruled out. However, the clear co-segregation of the chNA PCR product

with the yellow sectors suggests that the A-E mutation is most likely on the chNA.

9—aminaacridine hydrochloride dosage eflects an the plastome mutator line

Seeds with a low background mutation rate were used to determine whether 9AA
acts in synergism with pm to induce mutations in the chNA, (see Materials and
Methods for details on how a pm line with low background mutation rate was produced).
As a control, wild type seeds containing the same plastome type as the pm seeds were
also treated with the chemical for comparison. Mutant sectors were observed in both the
9AA-treated wild type and the mixed pm seedlings (Table 2-3). Since the untreated pm-
line produces a high background of sectors, statistical methods were employed to assess
whether the higher number of seedlings with sectors in the 9AA-treated pm-line reflected
an additive or synergistic interaction of the two conditions (see Materials and Methods).
These analyses showed that the differences in the final mutation frequencies of the

treated and control pm/pm plants were not statistically significant (0.05 < p < 0.10).

49

Table 2-3. Effect ofplastome mutator on the sector frequency in seedlings containing plastome I

treated with 9AA. Each trial contained 500 seeds with an exposure time of 16 hours. Sectors were

 

 

 

 

 

 

 

 

 

 

 

scored by the four-leaf stage.
Nuclear pg/ml of % Germination % Viability No. of seedlings
background 9AA with sectors
(% mutation
frequencya)
Wild type 0 49.0 54.6 0
Wild type 2 47.2 66.5 2/157 (1.3 °/.)
Mixed +/pm and O 93.6 70.9 13/332 (3.9%)
pm/pm
Mixed +/pm and 2 93.4 62.5 22/292 (7.5%)
pm/pm

 

a% Mutation frequency determined using the formula described in Table 2-1.

50

Discussion

The study described in this paper was undertaken to determine whether the pm-
encoded gene product is involved in repairing the darmge caused by frameshifi
mutagens, represented by 9AA, in order to delineate the role of the plastome mutator in
the replication and lor repair of chNA. To determine whether 9AA acts as a plastome
mutagen in Oenothera, initial mutagenesis studies tested for conditions that would
produce the maximum number of mutations, identified by the appearance of chlorotic
sectors. Incubation of seeds in 9AA caused a low frequency of chlorotic sectors in the
seedlings. Dosage trials established that increased exposure times or concentrations of the
mutagen did not affect the germination or viability of the treated seedlings (Table 2-1),
nor did those conditions produce a higher frequency of chlorotic sectors. We are
perplexed about why the 32-hour treatment did not result in seedlings with sectors.
Perhaps, the slight drop in seedling viability reflects death of the affected plants. Under
any circumstance however, the low mutation rate indicates that the mutagen lacks good

penetration or potency.

The inheritance studies produced progeny with sectors indicating both vegetative
segregation and non-Mendelian inheritance of the mutation. Therefore the mutation was
located in either the chloroplast or mitochondrial genome. In order to clarify which
genome was targeted by the mutagen, several approaches were taken. TEM studies
showed that the ultrastructure of the mitochonchia in the mutated tissue was normal
whereas the plastids were affected in the yellow sectors (Figure 2-1). These observations

indicated that the chloroplast was the likely target organelle of 9AA, although instances

51

are known, where mtDNA mutations affect plastid development and ultrastructure (for

example, the NCS mutations in maize: Hunt and Newton 1991; Gu et al. 1993).

It is possible to distinguish between chloroplast and mitochondrial mutations in
Oenothera by crosses: mutations that are carried by the chNA can be inherited from
both parents while mtDNA shows purely maternal inheritance (Brennicke and
Schwenrrnle 1984). Initial data from crosses of one mutant (A-E) indicated maternal
inheritance of the 9AA-induced mutation (Table 2-2); however, since the mutation only

occupied a portion of the germ line, the data from the crosses remained inconclusive.

Because the crossing data were inadequate for establishing the target organelle of
9AA, co-segregation analysis using molecular markers was performed. As shown in
Figures 2-2 and 2-3, the mtDNA is inherited uniformly from the matemal parent, whereas
in contrast, the yellow sectors co-segregated with the chNA from the mutant parent (A-
E). The green sectors on the other hand contained chloroplasts from both the maternal
and paternal parents. These data are most consistent with the plastome as the site of the

mutation.

Since the data suggested that 9AA targets the chNA of Oenothera, its ability to
act synergistically with the plastome mutator was explored. In the wild-type line, 1.3% of
the seedlings showed mutant sectors after exposure to 9AA, whereas in the seeds that
were half prm and half heterozygote, a six-fold higher mutation frequency was

observed (Table 2-3). However, nrost of these mutations can be credited to the plastome

52

mutator, and our chi square analyses using the adjusted expected number of mutations
(See Materials and Methods), indicate no statistical significance. This lack of synergistic
effect implies that the pm-encoded gene product does not act upon the DNA lesions
caused by 9AA. Because studies with prokaryotes have shown that the methyl-directed
mismatch repair system (MMR) is involved in correcting 9AA-provoked mismatches
(Miller 1998; Skopek and Hutchinson 1984), we conclude that the pm-encoded gene

product is not likely a component of the chNA mismatch repair system.

53

REFERENCES

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 haakeri
strain Johansen. Curr Genet 14: 287-292

Brennicke A, Schwemrnle B (1984) Inheritance of mitochondrial DNA in Oenothera
berteriana and Oenothera adarata hybrids. Z Naturforsch 39: 191-192

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

Chiu WL, Stubbe W, Sears BB (1988) Plastid inheritance in Oenothera: organelle
genome modifies the extent of biparental plastid transmission. Curr Genet 13: 181-189

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

D’Amato F (1950) Mutazioni clorofilliane nell’oro indotte da derivati acridinici.
Caryologia 3: 211-220

D’Amato F (1952) Mutagenic activity of acridines. Caryologia 4:388-413

Epp MD (1973) Nuclear gene-induced plastome mutations in Oenothera haakeri 1.
Genetic analysis. Genetics 75: 465-483

Gordon A, Halliday J, Horsfad M, Glickrrran B (1991) Spontaneous and 9-
aminoacridine—induced fi'ameshifi mutagenesis: second site fiameshift mutation within
the N-ternrinal region of the lac! gene of Escherichia cali. Mol Gen Genet 227: 160-164

Gu J, Miles D, Newton KJ (1993) Analysis of leaf sectors in the NCS6 mitochondrial
mutant of maize. Plant Cell 8: 963-971

54

Hoffman GR, Freemer CS, Parente LA (1989) Induction of genetic duplications and
fiameshifi mutations in Salmonella typhimurium by acridines and acridine
mustards: dependence on covalent binding of the mutagen to DNA. Mol Gen

Genet 218: 377—383

Htmt MD, Newton KJ (1991) The NCS3 mutation: genetic evidence for the expression
of ribosomal protein genes in Zea mays mitochondria Embo J 5: 1045-1052

Kopsidas G, MacPhee DG (1994) Mutagenesis by 9-aminoacridine in Salmonella
typhimurium: inhibition by glucose and other PTS class A carbon sources. Mutat
Res 306: 111-1 17

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

Miller JH (1998) Mutators in Escherichia cali. Mutat Res 409: 99-106

Nasim A, Brychcy T (1979) Genetic effects of acridine compounds. Mutat Res 65: 261-
288

Sarnbrook J, Fritsch EF, Maniatis T (1989) Molecular cloning, A laboratory manual. 2"(1
edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York

Sears BB, Lirn PO, Holland N, Kirkpatrick BC, Klomparens KL (1989) Isolation and
characterization of DNA horn a Mycoplasmalike organism. Mol Plant Microbe
Interact 4: 175-180

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

Skopek TR, Hutchinson F (1984) Frameshift mutagenesis of lambda prophage by 9-
arninoacridine, proflavin and ICR-191. Mol Gen Genet 195: 418-423

Stoike LL, Sears BB (1998) Plastame mutator-induced alterations arise in Oenothera
chloroplast DNA through template slippage. Genetics 149: 347-353

55

Stubbe W, Herrmann RG (1982) Selection and maintenance of plastorrre mutants and
interspecific genome/plastome hybrids fi'om Oenothera. In: Edelrnan M, Hallick

RB, Chua NH (eds) Methods in chloroplast molecular biology. Elsevier, Amsterdam, pp
149-165

Thomas SM, MacPhee DG (1985) Frameshift mutagenesis by 9-aminoacridine and
ICR191 in Escherichia cali: effects of uvrB, recA and lexA mutations in plasmid
pKM101.Mutat Res 151: 49-56

56

CHAPTER 3

Experimental assays of chloroplast DNA base

substitutions in Chlamydomonas reinhardtii

57

Abstract

Base substitutions are minimized in organisms through error-free DNA
replication, removal of DNA damaging agents as well by removal of damaged
nucleotides by the DNA repair machinery. The absence of either of these mutation
avoidance pathways results in an increase of base substitution rates and occurrence of
specific mutational patterns. Comparisons of spontaneous base substitution rates and the
spectra of mutations between wild type and mutator strains have helped to identify the
role of each type of pathway in mutation avoidance. Our investigation provides the first
experimental assessment of spontaneous base substitution mutations occurring at a target
site in the chloroplast DNA of the green alga Chlamydamanas reinhardtii. In
Chlamydamanas mutations at two sites within a conserved region of the chloroplast-
encoded 16S rRNA gene have been reported to confer spectinomycin resistance; they also
result in the loss of an Aat II restriction endonuclease cut site. Rates for base substitution
events in the chloroplast DNA were obtained for the chloroplast DNA at this target site.
Our data showed that these mutations occur in the chloroplast DNA at a range of 3-1 800
per 1011 viable cells plated. Analysis of the different types of base changes that occurred
in the spectinomycin-resistant mutants also allowed us to quantify the relative
frequencies of transition and transversion mutations that occurred in the chloroplast
DNA. We report the isolation of four new mutations in Chlamydamanas reinhardtii at the
above conserved residues of the chloroplast I 6S rRNA gene, two of which are novel and
have not been reported from other organisms. In addition, all three bases underwent

changes that are predicted to disrupt a conserved stem structure in the 168 rRNA

58

molecule, thereby reinforcing the importance of this stem structure in the binding of

spectinomycin.

Introduction

Base substitutions can arise naturally through intracellular metabolism such as
spontaneous alkylation, deamination of bases or modification by reactive oxygen species.
Base substitutions can also result from the incorrect insertion of nucleotides during DNA
replication (Kunz et al. 1998). Base substitutions are minimized in several ways that
result in the low rates of occurrence of these mutations in organisms ranging fiom
prokaryotes to eukaryotes. These include a high accuracy of DNA replication,
elimination of DNA-damaging agents and the repair of DNA damage after it has
occurred Replication fidelity is achieved at two separate levels: accurate base selection
during the insertion of nucleotides and the editing or proofreading by the 3’ '9 5’
exonuclease that is associated with the DNA polymerase. A small fraction of replication
errors may evade the correction machinery; these errors are then rectified by the
mismatch correction systens that repair the newly synthesized DNA molecules (Yang et
al. 1999). If DNA replication errors still escape all the above-mentioned error avoidance

pathways, they give rise to spontaneous base substitution mutations.

Spontaneous mutations may also result fiom unrepaired DNA damage caused by

different agents. Intracellular reactions producing active oxygen species have been

identified as major sources of DNA damage and as contributors to spontaneous mutation

59

(Schaaper and Dunn 1991). Specialized systens that are aimed at repairing a single type
of base substitutions exist, such as the MutT system, the MutY and the MutM systems of
E. coli, which protect the cell against the effects of such oxidative stress. Characteristic
types of point mutations have been associated with the absence of each type of repair
pathway in E. coli (Schaaper and Dunn 1991', Fowler et al. 2002 and Boiteux et al. 2002).
The MutT gene product hydrolyzes 8-oxodGTP to prevent its use as a substrate by DNA
polymerases and mutT deficient strains show an increase in AA“ -) C/G transversions
resulting from the misincorporation of 8-oxodGTP opposite template A (Fowler et al.
2002). The MutM and MutY proteins are glycosylases; MutM acts upon several modified
purines including 8-oxoG, and also initiates base excision repair of 8-oxoG when paired
with C. If unrepaired, subsequent mispairing of 8-oxoG with dATP would lead to G/C -)
T/A transversions (Carbera et al. 1988; Boiteux et al. 2002). MutY is also a glycosylase
that removes adenine fi'om A-G, A-8-oxoG and A-C mispairings, and thus a defective
mutY allele results in enhanced production of G/C -) T/A mutations (Fowler et al. 2003).
Since different types of base substitutions may be produced when specific repair
pathways are absent, an analysis of a collection of mutants that arise spontaneously can
provide insight into the relative efficiencies of different mutation avoidance pathways

that exist in different organisms.

One of the general strategies that has been used to better understand the
mechanisms responsible for the generation of spontaneous mutations is by comparing the
mutation rates and patterns between wild type strains and mutator strains that have

enhanced spontaneous mutation rates. The rationale for this approach is that the mutator

6O

phenotypes are caused by defects in genes whose products act to minimize DNA
alterations. Mutations that arise spontaneously in wild-type cells represent those which
occur normally even in the presence of all the mutation avoidance machinery. To date,
spontaneous mutations have been extensively studied using experimental assays in both
prokaryotic systems such as E. coli (Schaaper and Dunn 1987) and eukaryotic systems
such as Saccharomyces cerevisiae (Yang et al. 1999). In plants, conclusions about the
spontaneous mutation rates for the nuclear and chloroplast genes have been deduced from
evolutionary comparisons rather than from experimental investigations (Wolfe et al.
1997, Clegg et al. 1994). Molecular evolutionary studies using DNA sequence data were
used to estirmte the rates of nucleotide substitution in the chloroplast DNA (chNA) of
plants: ranging from 1.1 to 3 X 10'9 substitutions per synonymous site per year (Muse
2000). The purpose of the current study was to quantify the rates of nucleotide
substitution in the chNA using an experimental approach. For this purpose, the rates
and types of base substitutions were observed at atarget region in the chNA of a model

system, Chlamydamanas reinhardtii.

For the mutation assay, a particular region in the chloroplast encoded 1 6S rRNA
gene was chosen because mutations at two sites (three different base changes in total) in
that gene were known to confer resistance to spectinomycin in C. reinhardtii. These
changes also eliminate an existing restriction endonuclease (Aat II) cut site, (Harris et al.
1994) and are therefore easily detectable both at the phenotypic and molecular levels
(Figure 3-1 A). This forward system was preferable for the study of base substitutions in

the chNA as it allowed us to detect multiple base substitution events within one

61

particular target site without a bias towards any particular base substitution. For a detailed

description of the assay refer to Figure 3-1 B and the Materials and Methods section.

Analysis of the different mutants obtained fi'om the above assay would allow us to
establish the baseline spectrum of base substitutions that naturally occur in the chNA of
C. reinhardtii. Since no prior studies exist for plants, our results will be compared to
similar mutational spectra obtained from both prokaryotic and eukaryotic systems. These
comparisons may provide us with an insight of whether the DNA replication machinery
of chloroplasts resembles that of prokaryotic or eukaryotic systerm. Since chloroplasts
have an endosymbiotic derivation from cyanobacteria (Gray 1993), it is expected that the
types of mutations and their abundance seen in the chloroplasts would be similar to that
observed in prokaryotes. However, all of the components of the chloroplast DNA
replication and repair machinery are encoded by the nucleus and therefore it may also

have acquired characteristics from the eukaryotic nucleus.

The relative abundance of the two different types of base substitutions (transitions
and transversions) that occur has been shown to differ among prokaryotes and
eukaryotes: for example, wild-type E. coli and yeasts have shown some differences in the
ratios of transition to transversion mutations (Schaaper and Dunn 1991; Kunz et al. 1998;
Yang et al. 1999). A study using an in vitro assay with the chloroplast DNA polymerase
from pea has reported that transversions were generated at a slightly higher frequency
than transitions (28 and 26 respectively) (Gaikwad et al. 2002). Analysis of the different

base substitutions generated from our assay would provide an insight into the relative

62

fiequencies of transitions and transversions that occur in vivo in the chNA and could

also be compared to similar ratios from the E. coli and yeast studies.

Different base substitutions arise due to different types of mismatches that are
generated during DNA replication or due to DNA damage. Furthermore, mismatches are
recognized and repaired at different efficiencies by the exonucleases (Keim and
Mosbaugh 1991), mismatch repair proteins (reviewed in Harfe and Jinks-Robertson
2000) and other repair pathways, such as the oxidative repair pathway (Fowler et al.
2002). Different base changes were reported to occur at different frequencies in E .coli
and yeasts (Schaaper and Dimn 1987; Yang et al. 1999) due to differences in efficiencies
of their mutation avoidance systems. Analysis of the different types of base substitutions
generated in our assay would allow us compare the types and frequencies of mismatches
that result in the different base substitutions in the chNA and other organisms. In
addition, we hoped that the mutation spectrum observed in the wild-type cells would
allow us to deduce the presence or absence of various chloroplast DNA replication and

repair components in C. reinhardtii.

63

 

AatII site

r———I

Wild type C. reinhardtii 1116 GAGGATG‘CGTCAAGTCA 1133

Known spec“ mutants ....... G ..........
....... C. . . . . . . . . .
......... A. . . . . . . .

A. Nucleotide sequence of wild-type C. reinhardtii (top line) and three spectinomycin-
rosistant mutants (lower lines. The Aat II recognition site is indicated. (Adapted from
Harris et al. 1994)

 

 

 

 

PCR amplification of DNA from spectinomycin-resistant mutants with primers
flanking the Aat [1 target site in the 16S rRNA game

 

=

léS-F fl l6S-R

Digestion of the PCR product with Aat II restriction mdonculease

 

 

 

 

l Aat n
._, _ Sequence
'_ E :5 PCR product
3 with 16S-F2
for exact base
_ change
—
-

 

 

 

 

 

 

Agarose gel electrophoresis to check for mutants that have lost the Aat 11 cut site in the target
region of the 16S rRNA gene

B. Experimental procedure to identify spectinomcyin-resistant rmtants of
Chlamydamanas reinhardtii that have undergone specific base changes at the target
site in the 16S rRNA gene. Oligonucleotides that flank the Aat 11 cut site are named
16S-F, l6S-F2 and l6S-R.

 

 

 

Figure 3-1. Base substitutions at a particular region in the gene coding for 16S rRNA in
Chlamydamanas reinhardtii confer resistance to spectinomycin.

64

Materiak and Methods

Strains and media

The Chlamydamanas reinhardtii strain used for all experiments was the wild-type CC125
(mt+) strain, which was obtained from the Chlamydamanas Genetics Center at Duke
University (Harris 1989). C. reinhardtii was grown in tris-acetate phosphate (TAP) media
(Gorman and Levine 1965) on either solid plates or liquid media when required. The
cultures were incubated under growth lights at room temperature (22 — 25 °C). The liquid
cultures were put on a rotary shaker under constant light at a speed of ~200 rpm. The
spectinomycin-resistant transformants were grown on TAP media supplemented with 100

ug/ml spectinomycin dihydrochloride (Sigma Chemical Co.).

Assay for base substitution mutations to spectinomycin resistance

The wild type (CC125, mt+) cells were grown in liquid TAP culture to late log phase (2-4
X 106 cells/ml) for about 7 days at room temperature under constant white lights on a
rotary shaker at a speed of ~200 rpm Cell counts were determined by using the
compound microscope and the standard hemocytometer method as described in the
Chlamydamanas sourcebook (Harris 1989). 250 ml of the cells was harvested by
centrifugation at 8000 X g in a Sorvall centrifuge using a GSA rotor at 4°C for 10 mins
under sterile conditions. The supernatant was discarded and the cells were resuspended in

a volume of TAP medium that would result in a final concentration of 2 X 108 cells/ml.

65

An aliquot of these cells were used for dilution platings to determine the number of
viable cells plated. 100 pl aliquots from the 10's, 106 and 10'7 dilutions were plated on
TAP plates and then grown under lights at room temperature. The number of colonies
that grew on these plates was then counted to assess the total number of viable cells that

were plated for the experiment.

The remaining volumes of the concentrated CC125 cells were then plated as 500 pl
aliquots on TAP agar supplemented with 100 ug/ml of spectinomycin. The TAP +
spectinomycin plates were put under growth lights on a shelf at room temperature. The
total number of spectinomycin resistant colonies that grew on these plates were counted,
and then transferred using sterile toothpicks and replated on TAP + spectinomycin plates

for further DNA analyses.

Total DNA isolation

Chlamydamanas total genomic DNA was prepared using a modified protocol adapted
from the ‘miniprep’ method described by Rochaix et al. (1988). The cells were scraped
from a plate using yellow pipette tips and resuspended into a 1.5 ml microfuge tube that
contained 0.5 ml of IX TEN buffer (1M TRIS, 0.5 M EDTA and 5 M NaCl). The cells
were then spun in a microcentrifuge (13,000 rpm) and the supernatant discarded; the
pellet was then resuspended by vortexing in 150 pl of sterile water. To these resuspended
cells, 300 pl of SDS-EB buffer was added (10% SDS, 5M NaCl, 0.5M EDTA and 1M
TRIS-HCl, pH 8.0). The cells were incubated on ice for 5 min and then one

phenolzchloroforrrr isoamyl alcohol (25:24: 1) extraction was done, with centrifugation for

66

10 min. The aqueous phase was transferred to a fresh microfuge tube and the DNA

isolations were then done according to the protocol described by Rochaix et al. (1988).

PCR amplification

DNAs extracted from all the spectinomycin-resistant colonies and also the wild type
CC125 cells were then used for PCR amplification. The 16S-F (5 ’-
CCGCACAAGCGGTGGATT-3’) and the 16S-R (5’-CCTTCCAGTAGCCCTACC-3’)
primers were used in the PCR amplification reactions as the forward and reverse primers
respectively to amplify a 580-bp fragment from the 16S rRNA gene. DNA (1-5ng) was
added to individual 100 u] reactions containing 200 arm] of each dideoxynucleotide, 2.5
units of Taq polymerase (GIBCO BRL), 10 pmoles of each primer and 1 X PCR reaction
buffer (GIBCO BRL). Conditions for the 30 cycles of PCR were as follows: denaturation
at 95°C for 2mins, 85°C for 5 mins, followed by 94°C for 40 sees; 30 secs at 48°C to
anneal the primers with the template and 30 secs at 72°C for DNA extension. The final
extension after all the cycles was done for 10 mins at 72°C. All the PCR reactions were

performed with a minicycler model PTC-150-16 (MJ Research Inc., Watertown, Mass).

Restriction digestion and Sequencing

The 580-bp PCR products were digested with the restriction enzyme Aat II (New
England Biolabs) at 37°C according to the manufacturer’s instructions. The digested
products were then run on a 1.5% Agarose gel with a 100-bp DNA ladder (Invitrogen
Life Technologies, CA). The gel was photographed using the BIORAD Alpha imager and

the image was displayed using the Quantity One software. The PCR products that had

67

lost the restriction endonuclease cut site were then further sequenced to determine the
exact base change that had occurred resulting in the loss of the restriction cut site. For
sequencing, the PCR products were cleaned by removing the excess primers, enzyme,
dNTPs using the Qiaquick PCR purification kit (Qiagen Inc.,CA) following the
manufacturer’s instructions. The concentrations of the purified PCR products were then
determined by running them on an agarose gel along with a standard (0.5 ug/ul) uncut
lambda DNA marker (GIBCO BRL). The purified PCR products were then sent for
sequencing to the Michigan State University Genomics Facility, East Lansing, MI
(wwwgenomjcsmsuedu), with an internal primer (16S-F2) provided for sequencing (5’-

CGTCAGCTCGTGCTGTG-3’).

Computer Analyses

The sequence data obtained from the sequencing facility were analyzed using the
Megalign Software from the DNAStar Program (Wisconsin, Madison). Using this
program, the wild-type sequence was used to align the data from the different
spectinomycin mutants lacking the Aat H cut site to determine the exact base substitution

that had occurred

Mutation Rate

To obtain the chNA base substitution rate, the total numbers of spectinomycin-resistant
colonies that had lost the Aat 11 cut site were determined for each experiment, and then
using the following formula, the mutation rate (u) was determined:

Number of spectinomycin resistant colonies with loss of Aat II site
Total number of viable cells plated (obtained from dilution plates)

68

Results

Estimation of rates of base substitution mutations at positions 1123 to 1125 in the 1 6S
rRNA gene

Specific base changes in the chloroplast encoded I 6S rRNA gene of C. reinhardtii
confer spectinomycin resistance and also eliminate the A at H endonuclease recognition
site. In order to determine the rates of spontaneous mutations in the chNA at the Aat H
target site, DNAs from the spectinomycin-resistant rrrutants were analyzed by PCR
amplification followed by restriction digestion of the PCR product. Since changes in
other genes coding for ribosomal proteins can also result in spectinomycin resistance
(Harris et al. 1994), the Aat H assay allows a quick determination of whether the target
site in the I 6S rRNA gene contains the mutation. 64 out of the total 121 spectinomycin-
resistant mutants isolated from three independent trials were altered at the Aat H target
site (Table 3-1). Figure 3-2 shows a subset of the spectinomycin resistant mutants
assayed. Among the seven mutants shown, two (m6 and m7) have lost the Aat H cut site.
Among the five spectinomycin-resistant mutants that had retained the cut site, two of
them (m3 and m4) have a small deletion in the 331-bp fragment. The rates of
spontaneous mutations at the target site of the chNA ranged from 3-1800 per 10‘ll

viable cells plated in the three independent experiments (Table 3-1).

69

 

 

 

 

 

 

 

 

 

Number of
Total number of spectinomycin
Rate of mutation
Experiment # Total number of spectinomycin resistant colonies
at the Act H target
cells plated (x 10’) resistant colonies that lost the Aat 11
site
cut site
1 1.1 20 20 1.8 x 10"
2 33.9 2 1 2.9 x 10'“
3 41.9 99 43 1.1 x 10‘“

 

 

Table 3-1. Frequencies of spontaneous base substitutions at the Aat H target site in the
chloroplast DNA of Chlamydamanas reinhardtii.

70

Digested with Aatll Undlg.

 

lm1 m2 m3 m4 m5m6 m7 wt wt wt

 

Figure 3-2. PCR amplification using the 16S—F and 16S-R primers followed by
restriction digestion with Aat H. Agarose gel showing digested and undigested PCR
products fiom the wild-type (wt) and digested PCR products fi‘om the spectinomycin
mutants (ml to m7).

7l

Analysis of base substitutions for each base in the A at 11 target site

The 64 spectinomycin mutants that were obtained from the above three trials were
further analyzed to determine the exact base changes that had occurred. For this purpose,
DNA from the mutants was sequenced through the target region of the 16S rDNA.
According to the sequence data obtained, all the base substitution changes could be
grouped into six different classes (Figure 3-3). In contrast to the previous report of
mutations at two bases (1123 and 1125) in the target Aat H site (Harris et a1. 1994), all
three bases (1123 to 1125) were found to be capable of base substitutions resulting in the
spectinomycin-resistance phenotype. Table 3-2 shows the detailed compilation of the
different changes at each base along with the numbers of mutants obtained hour each
trial. Since each trial did not produce mutations in all of the three bases in the Aat 11
target site (see Table 3-2), a range of mutation frequencies has been reported for each

base of the target site in Table 3-3.

Spectrum of base substitution mutations in the chNA of C. reinhardtii

DNA sequence analyses of the 64 mutants showed six different types of base
substitutions among which an NT 9 C/G change was the most frequent (Table 3-3).
However, the G/C 9 C/G type of base substitution occurred in each trial (Table 3-2) and

was the second most frequent type of base substitution overall.

72

 

 

Aat H site

 

 

 

 

I H

T G A C G T C A wild-type

T G A C T C A spec. mutant
T G A G T C A spec. mutant
r G C G T C A spec. mutant
T G A G T C A spec. mutant
T G A T C A spec. mutant
r G C G r C A spec. mutant

 

 

 

Figure 3-3. Sequence alignment of wild-type and spectinomycin resistant colonies at the Aat H
target site showing the six different base changes, compiled from all three trials. The new mutants
are represented in bold italics.

73

 

 

 

 

 

 

 

 

 

 

 

A9C A9G C9T C9A G9C G9T
ExpL-I 8/20 12/20
Expt-Z r/r
Bxpt-3 30/43 1/43 4/43 2/43 1/43 5/43

 

Table 3-2. Total nurrbers of the different types of base substitutions observed in each

experiment.

74

 

Ratios of transition to transversion mutations

Both transitions and transversions occurred at the target Aat H site. However,
analysis of the base substitutions at the above chNA target site showed that in all three
trials transversions occurred more fi'equently than transitions (Tables 3-2 and 3-3). In the
first experiment, transversions occurred at a slightly higher rate than transitions (12:8),
but in the third experiment transversions occurred eight times more frequently than
transitions (38 transversions in contrast to 5 transitions). Overall, transversion mutations

occurred nearly four times more frequently than transitions.

Comparison of the base changes in the 1 6S rRNA target site from spectinomycin-resistant
mutants in C. reinhardtii and other organisms

The chloroplast genes for the I 68 rRNAs are highly conserved at the sequence
level and most closely related to eubacterial sequences (Harris et al. 1994). Figure 3-4
shows a comparison of the different types of changes at the above region of the I 6S
rDNA observed in spectinomycin-resistant mutants from different organisms. Among the
six types of base substitutions reported in this study (Figure 3—3), four are new for C.
reinhardtii. Two of those are known to give spectinomycin-resistance in E. coli and
tobacco (Svab and Maliga 1991; Harris 1994). Two of the six mutations are novel, and
their isolation underscores the importance of all three bases in the binding of the

antibiotic to the I 68 rRNA molecule.

75

 

 

 

 

 

Wild type base A C G
Changesl C (tv) r (ts) C (tv)
Number of 30/64 12/64 14/64
mutants 0.072 0.73 1 .085
Frequency of
mutations (/10‘8)
Changes] G (ts) A (tv) r (tv)
Nurrber of 1/64 2/64 5/64
mutants 0.002 0.005 0.012
Frequency of
mutations (/10‘8)
Range of <1 in2.9X10'” <1 in2.9X10'” 2.9x 10'“to
mutation to 7.2 x 10'10 to 7.2 x 10'9 1.1 x 10‘8
frequencies in all
three experiments

 

 

 

 

 

 

 

Table 3-3. Comparison between the types, numbers and frequencies of base substitutions among
the three bases at the Aat H target site. lBase changes are represented as transitions (ts) or
transversions (tv).

76

 

AatII site

F'—_l
1184 GgGGArGaccTCAAGTCA 1201

Known spec“ mutant 8 ........ T .........

 

 

 

 

AatII site
r——-—I
Wild type C. reinhardtii 1116 GAGGATGACGTCAAGTCA 1133
Known spec“ mutants ....... G ..........
....... C. . . . . . . . . .
......... A. . . . . . . .
novel spec“ mutants ......... C ........
......... T. . . . . . . .
........ T. . . . . . . ..
........ A. . . . . . . ..

 

 

 

AatII site

r——'1

“xii 1:3 1 L_ T. L‘ tarp 1131 GAGGATGICCTCAAGTCA 1148

Known spec“ mutants ......... A ........
....... C. . . . . . . . . .
........ T. . . . . . . . .

 

 

 

 

. .. 1130 GAGGATGMGCCAAGTCA 1147
(naturally spec“ resistant)

 

 

 

~97: Tobacco, E.COii and C.reinhdrdrii
—————— : E.coli and C.reinhardtii

Figure 3-4. Alterations in the gene encoding 16S rRNA corrpiled from spectinomycin-resistant
mutants of different organisrm (Adapted from Harris et aL Genetics: 123: 281-292, 1989.)

77

Discussion
Mutations in the 16S rRNA gene that confer spectinomycin resistance

Spectinomycin is an aminoglycoside antibiotic that blocks protein synthesis on
prokaryotic-like rrbosomes (Frorrnn et al. 1987). Evidence exists for the binding of this
antibiotic to the RNA component of the small ribosomal subunit and several mutations in
the 16SrRNA molecule that disrupt the binding of the antibiotic have been obtained from
a wide range of organisms (Mark et al. 1983; Harris et al. 1989). Isolation of the
spectinomycin-resistant mutants that have changes in the 16S rRNA gene has therefore
helped to pinpoint the molecular basis of resistance to spectinomycin. Studies have
shown that a conserved stem structure in the 3’ end of the 16SrRNA molecule is involved
in the binding of the antibiotic because mutations that disnrpt the right arm of the stern
confer resistance in E. coli, as well as chloroplasts of Nicotiana and Chlamydamanas (De
Stasio et al. 1989; Harris et al. 1989). The three nucleotides that were assayed for
mutations in our study (1123 to 1125) are located in the right arm of the above-mentioned
stem structure that is conserved among organisms (Schwarz and Kossel 1980). Previous
studies with Chlamydamanas had identified mutations in two of the three bases (Figure
3-4), and our results provide additional evidence regarding the importance of all three
bases in the binding of the antibiotic to the stem loop. In addition, sorrre of the
spectinomycin-resistant mutants that had retained the Aat H cut site showed deletions in
the 331-bp fragment (bases 1455 to 1124) of the PCR product (Figrne 3-2) indicating that
deletions at the 3’end of the molecule are also able to confer resistance. Experiments are
currently being performed in the Sears laboratory to isolate more of these deletion

mutants in an effort to understand the basis of their resistance to spectinomycin.

78

Rates of chNA base substitutions

The primary objective of this study was to experimentally assess the rate of base
substitutions at a region of the 16S rDNA in C. reinhardtii. Results obtained from this
study indicate that base substitutions occur at low rates between 2.9 in 10'11 to 1.8 in 10’8
viable cells plated This range of low nucleotide substitution rates provides evidence for
the existence of a high fidelity of chNA replication and efficient repair rmchinery, even
though the genes for chloroplast DNA replication or repair have not yet been identified
through genomic analyses. In vitro studies with the pea chNA polymerase including its
3’ 9 5’ exonuclease had indicated that the error rate for the polymerase was
approximately 1 X 10*6 (Gaikwad et al. 2002). Therefore, we can hypothesize that the
lower rates of spontaneous base substitution mutations seen in vivo are due to the

involvement of post-replication repair systems that act to maximize chNA fidelity.

Mutation rates can either be measured by reversion of known mutations, as in the
Ames test, or by screening for forward mutations: the. latter strategy was used in our case
and mutations were detected by the appearance of spectinomycin-resistant C. reinhardtii
cells. Since the single chloroplast of C. reinhardtii contains around 100 copies of the
chNA, a cell will not be counted as a mutant until the number of mutant alleles has
increased to the point where the cell is phenotypically mutant (Birky and Walsh 1992).
Chloroplasts also have relaxed rather than stringent control of replication, such that DNA
molecules are randorrrly chosen for replication and are also partitioned randomly to the
daughter cells at cell division. In addition, they also undergo intracellular selection when

some chNA genomes replicate faster than some others in the same chloroplast

79

depending on their genotypes (Birky and Walsh 1992; Birky 1994; Birky 2001). Thus,
the mutation frequency is a function of all of the above complex mechanisms. Depending
on how early the mutation reached fixation, how fast the mutant alleles replicated and
were partitioned; the mutation frequency to spectinomycin resistance could vary in each
experiment. Since the mutation frequency varied among the different trials (3-1800 per
10“ viable cells plated), it remained to be seen whether the rates could go even higher if
more trials were done. However, similar experiments in the Sears laboratory
(unpublished results) using different substrates and different genes, to assay for
frequencies of base substitution mutations in the chNA failed to produce mutants even
when 109 viable cells were plated. These results provide confidence in my observations
that the spontaneous fi'equency of base substitution mutations for the chNA perhaps

does not occur at rates higher than 2 in 108 viable cells in C. reinhardtii.

None of the genes responsrble for the chNA mutation avoidance systems has
been isolated. One of the ways to identify them would be by isolating mutator lines that
show elevated fiequencies of base substitutions over spontaneous levels. Our study
provides the first experimental assessment of spontaneous rates of nucleotide
substitutions in the chNA and therefore, the next goal is to isolate C. reinhardtii lines
that will show base substitution rates higher than the observed spontaneous levels.
Characterization of these mutator genes would help us to identify the components

required for chNA replication fidelity and repair.

80

Are there any difi'erences in mutation frequencies among the three bases at the Aat 11
target site?

Since experiments 1 and 2 (Table 3-2) did not produce mutations at all three
nucleotides in the Aat H target site, data from trial 3 were examined to compare the
relative rates of base substitutions among the three nucleotide positions. Among the 43
mutations observed in the trial, 31 occurred at position 1123, while positions 1124 and
1125 each had six mutations. Therefore it appears that the adenine at position 1123 may
be prone to more mutations than the other bases. However, due to the relaxed replication
and partitioning of the chloroplast genomes, this difference in numbers may have been
skewed if the predominant mutation (A to C) at base 1123 was fixed early as a mutation,
or if the chNA with this mutation was preferentially replicated and partitioned to the

daughter chloroplasts compared to the mutations in bases 1124 and 1125.

Rates of transition and transversion mutations in the chNA

Transition and transversion mutations are generated at different frequencies in
prokaryotes and eukaryotes: a study in E. coli (Schaaper and Dunn 1991) has shown that
transition ruutations are generated at higher frequencies than transversions (175 and 118
respectively), whereas a similar study in yeast (Kunz et a1 1998; Yang et a1 1999)
indicated that transitions occur at a slightly lower fi‘equency than transversions (117 and
175 respectively). Our results show that among the chNA base substitutions obtained in
each trial, transversions occurred more frequently than transitions, and in overall there
were 51 transversions to 13 transitions. Relevant to this, an in vitro assay with the pea

chloroplast DNA polymerase also showed that transversions were generated at a slightly

81

higher frequency than transitions by the polymerase (Gaikwad et al. 2002). In this
respect, the mutational pattern of the chNA was more similar to that seen in eukaryotes.
This was sluprising as chloroplasts share a common ancestry with cyanobacteria (Gray

1993) and therefore we had expected that its mutational patterns would be similar to

prokaryotes.

Efl'ect of neighboring base composition on transition/transversion bias at bases
Evolutionary studies have shown that base substitutions are highly dependent
upon the composition of the flanking bases in both the coding and non-coding regions of
the chloroplast DNA (Morton 1995; Morton and Clegg 1995; Morton 1997; Morton et al.
1997). The types of substitutions were dependent on the composition of the two flanking
bases: substitutions in G+C rich environments showed a strong bias towards transitions,
while in A+T rich environments, transversions outnumbered transitions (Morton 1997).
Only one of the three bases in the target site fits the criterion of being in an A+T rich or
G+C rich context: the first of the three bases, wild type base “A” is surrounded by G+C,
but in contrast to the observations by Morton, this site showed a stronger bias towards
transversions than transitions (Table 3-3). However, since all trials did not produce
changes in the target base 1123 (Table 3-2), the results were interpreted from only one
trial and therefore, more data need to be gathered for this base as well as for other bases
showing similar G+C rich environments, in order to establish a general trend for chNA
mutations. Since evolutionary comparisons do not always reflect in viva chNA
conditions our experimental assay may provide a better tool to analyze whether the G+C

rich environments can bias the type of base substitution at a target site.

82

Spectrum of chNA base substitutions in wild-type C. reinhardtii

One of the purposes of this study was to establish the different types of
spontaneous base substitution that occur in a particular region of the chNA of C.
reinhardtii. The spectrum presented here (Table 34) represents a first essential step
towards the rmderstanding of the types of spontaneous mutations that are generated in
vivo in the chNA. A variety of base substitution events were recovered at the three
nucleotides in the target site. This forward mutational assay had a distinct advantage as a
mutational spectrum could be obtained without any bias towards a single base change in
contrast to reversion analyses (Kunz et al. 1998). As summarized in the introduction, the
roles of different repair pathways in minimizing specific base mutations have been
deduced in both prokaryotes and eukaryotes by comparing the mutational spectrum in
wild-type cells and cell lines lacking the particular repair pathway in question (Table 3-
4). Similar mutational analyses have not been done for chloroplasts, and therefore the
spectrum presented in Table 3-4 from our assay is the first representation of the different
types of mutations that can occur at a target site in the chNA. As mentioned earlier, our
intention is to isolate C. reinhardtii mutator lines and conrpare their mutation spectra,
thus providing a direct function for each mutator gene in the generation of specific base

substitutions.
Table 3-4 shows the number of different classes of mutants obtained from our

experiments. The predominance of the NT 9 0G type of mutations recovered could

mean that this type of mutation may be generated at relatively higher rates in the chNA

83

 

Original Mutation Class Occurrences Repair pathway

 

 

 

basepair (Out of 64) (genes involved)
A:T G:C Transition 1 Mismatch Repair
(mut H,L, S or homologs)
C:G Transversion 30 Oxidative Repair
(mut T)
C:G T:A Transition 12 Mismatch Repair

(mutH, L. SorMLHI, in

eukaryotes)

(Baross-Franscis et al. 2001)

 

 

 

A:T Transversion 2 Not Known
G:C C:G Transversion 14 Novel pathway in mice
MutY/MutM in bacteria? (Shin et
al. 2002)
T:A Transversion 5 Oxidative Repair
(mutY / mutM)

 

 

 

 

 

 

 

Table 3-4. Spectrum of chNA base substitutions in wild type Chlamydamanas reinhardtii and
the possible pathways by which each type is known to be repaired in different organisms.

84

or that it may also be repaired with a lower efficiency than any other type of base
substitution. One of the major mutagenic base lesions in DNA caused by exposure to
reactive oxygen species is 8-oxoguanine (8—oxoG), which has ambivalent base pairing
properties and is capable of pairing effectively with both A and C during DNA synthesis
(Dany and Tissier 2001; Fowler et al. 2002). The mutT, mutY and mutM error-prevention
systems in E. coli and their homologues in eukaryotes protect cells against the effects of
8-oxoG (Boiteux et a1. 2002; Fowler et al. 2002).The mutT mutator strains display a
specific increase in NT 9 C/G transversions (Fowler et al. 2002). In plants, chloroplasts
have an abundance of reactive oxygen species and hence their DNA may undergo
darmge by 8-oxoG at high rates. Since the NT 9 C/G type of transversion was observed
at a higher frequency than the other base substitutions, we believe that this may indicate
that the functional homolog of mutT gene product present in chloroplasts may be
insufficient to efficiently handle the oxidative damage that occurs spontaneously. It is
also possible that the chloroplasts may lack any mutT encoded repair enzyme, since no
mutT homolog has been isolated in plants, even for the nucleus. No mutT homolog has
been identified yet among the annotated genes in the Cyanobase (database for the
cyanobacterium, Synechocystis PCC6803) indicating that it may be absent in the ancestor
to the modern day chloroplasts. In contrast the nuclear homologs for the mutY and mutM

repair pathway have been isolated in plants.

In conclusion, our study presents the first experimental assay for determining the

rates of specific spontaneous base substitutions for the chNA. The mutational spectrum

observed at a target site in the chNA of C. reinhardtii has also been reported. In

85

addition, our studies have provided new information about the role of all three bases in

the Aat H target site in 1 6S rRNA for the binding of spectinomycin.

86

REFERENCES

Baross-Francis A, Makhani N, Liskay RM, Jirik FR (2001) Elevated mutant frequencies
and increased C:G9T:A transitions in Mlhl-l- versus PmsZ-/- murine small intestinal
epithelial cells. Oncogene 20(5): 619-625

Birky CW, Walsh JB (1992) Biased gene conversion, copy number and apparent
mutation rate differences within chloroplast and bacterial genomes. Genetics 130: 677-
683

Birky CW (1994) Relaxed and stringent genomes: Why cytoplasmic genes do not obey
Mendel’s laws. J Hered 85: 355-365

Birky CW (2001) The inheritance of genes in mitochondria and chloroplasts: Laws,
mechanisms, and models. Annu Rev Genet 35: 125-148

Boiteux S. Gellon L, Guibort N (2002) Repair of 8-oxoguanine in Saccharomyces
cerevisiae: Interplay of DNA repair and replication mechanisms. Free Radic Biol Med
32(12): 1244-53

Carbera M, Nghiem Y, Miller JH (1988) mutM, a second mutator locus in Escherichia
coli that generates G:C 9 T:A transversions. J Bacteriol 170(11): 5405-5407

Clegg MT, Gaut BS, Learn GH, Morton BR (1994) Rates and patterns of chloroplast
DNA evolution. Proc Natl Acad Sci USA 91: 6795-6801

Dany AL, Tissier A (2001) A functional OGGI homologue from Arabidopsis thaliana.
Mol Genet Genomics 265(2): 293-301

DeStasio EA, Moazed D, Noller HF, Dahlberg AE (1989) Mutations in 16S ribosomal
RNA disrupt antibiotic-RNA interactions. Embo J 4: 1213-1216

Fowler RG, White SJ, Koyama C, Moore SC, Dunn RL, Schaaper RM (2003)
Interactions among the Escherichia coli mutT, mutM, and mutY damage prevention
pathways. DNA Repair 2: 159-173

87

Fronrrn H, Edelrnan M, Aviv D, Galun E (1987) The molecular basis for rRNA-
dependent spectinomycin resistance in Nicotiana chloroplasts. EMBO J 6(11): 3233-3237

Gaikwad A, Hop DV, Mukherjee SK (2002) A 70-kDa chloroplast DNA polymerase
from pea (Pisum sativum) that shows high processivity and displays moderate fidelity.
Mol Genet Genomics 267(1):45-56

Gorman DA, Levine RP (1965) Cytochrome f and plastocyanin: their sequence in the
photosynthetic electron transport chain of Chlamydomonas reinhardtii. Proc Natl Acad
Sci USA 54: 1665-1669

Gray MW (1993) Origin and evolution of organelle genomes. Curr Opin Genet Dev 3(6):
884-890

Harfe BD, Jinks-Robertson S (2000) DNA mismatch repair and genetic instability. Annu
Rev Genet 34: 359-399

Harris EH (1989) The Chlamydomonas Sourcebook. A comprehensive guide to biology
and laboratory use. Academic Press San Diego

Harris EH, Burkhart BD, Gillham NW, Boynton JE (1989) Antibiotic resistance
mutations in the chloroplast I 6S and 23S rRNA genes of Chlamydomonas reinhardtii:

Correlation of genetic and physical maps of the chloroplast genome. Genetics 123: 281-
292

Harris EH, Boynton JE, Gillham NW (1994) Chloroplast ribosomes and protein
synthesis. Microbiol Rev 58(4): 700-754

Keim CA, Mosbaugh DW (1991) Identification and characterization of a 3' to 5'
exonuclease associated with spinach chloroplast DNA polymerase. Biochemistry 30(46):
11 109-1 1 1 18

Kunz BA, Ramachandran K, Vonarx EJ (1998) DNA sequence analysis of spontaneous
mutagenesis in Saccharomyces cerevisiae. Genetics 148: 1491-1505

88

Mark LG, Sigmund CD, Morgan EA (1983) Spectinomycin resistance due to a mutation
in an rRNA operon of Escherichia coli. J Bacteriol. 155(3): 989-994

Morton BR, Clegg MT (1995) Neighboring base composition is strongly correlated with
base substitution bias in a region of the chloroplast genome. J Mol Evol 41(5): 597-603

Morton BR (1995) Neighboring base composition and transversion/transition bias in a
comparison of rice and maize chloroplast noncoding regions. Proc Natl Acad Sci USA 92
(21): 9717—9721.

Morton BR (1997) The influence of neighboring base composition on substitutions in
plant chloroplast coding sequences. Mol Biol Evol 14(2): 189-194

Morton BR, Oberholzer VM, Clegg MT (1997) The influence of specific neighboring
bases on substitution bias in noncoding regions of the plant chloroplast genome. J Mol
Evol 45(3): 227-231

Muse SV (2000) Examining rates and patterns of nucleotide substitution in plants. Plant
Mol Biol 42: 25-43

Rochaix JD, Mayfield S, Goldschmidt-Clermont M, Erickson J (1988) Molecular
Biology of Chlamydomonas. In: Schaw C-H (ed) Plant Molecular Biology: A practical
approach. IRL Press Oxford

Schaaper RM, Dunn RL (1987) Spectra of spontaneous mutations in Escherichia coli
strains defective in mismatch correction: The nature of in vivo DNA replication errors.
Proc Natl Acad Sci USA 84: 6220-6224

Schaaper RM, Dunn RL (1991) Spontaneous mutation in the Escherichia coli lac] gene.
Genetics 129: 317-326

Schwarz ZS, Kossel H (1980) The primary structure of I 6S rDNA from Zea mays
chloroplast is homologous to E. coli 16S rRNA. Nature 283: 739-742

Shin CY, Ponomareva ON, Connolly L, Turker MS (2002) A mouse kidney cell line with
a G:C 9 C:G transversion mutator phenotype. Mutat Res 503: 69-76

89

Svab Z, Maliga P (1991) Mutation proximal to the tRNA binding region of the Nicotiana
plastid 16S rRNA confers resistance to spectinomycin. Mol Gen Genet 228: 316-319

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among
plant mitochondrial, chloroplast and nuclear DNA. Proc Natl Acad Sci USA 84: 9054-
9058

Yang Y, Karthikeyan R, Mack SE, Vonarx EJ, Kunz BA (1999) Analysis of yeast pmsl,
msh2, and thI mutators points to differences in mismatch correction efficiencies
between prokaryotic and eukaryotic cells. Mol Gen Genet 261(4-5): 777-87

90

 

CHAPTER 4

Creation of a reporter construct to assay for

chloroplast DNA replication slippage events

91

 

“F

Insertion or deletion of one or more bases occurs spontaneously at a higher
fi'equency in regions of DNA that contain repeated sequences (microsatellites).
Replication slippage has been implicated to be the mechanism by which these deletions
and insertions occur at the repeats. The few estimates that are available for rates of
replication slippage in the chloroplast DNA are based solely on evolutionary studies. This
study was undertaken to experimentally assess the rates of replication slippage that occur
in vivo in the chNA of Chlamydomonas reinhardtii. Since microsatellites occur
primarily in the non-coding regions of the chloroplast genome, their changes cannot be
measured experimentally as they do not produce any changes in the phenotype. Hence, a
reporter construct was created in the chNA of C. reinhardtii, in which a stretch of
GAAA repeats was inserted into a functional gene to monitor changes occurring at this
rrricrosatellite target site. Results showed that this site underwent slippage at frequencies
that ranged from 1 to 70 per 106 viable cells. Analysis of these slippage mutants showed
that deletions were the only type of slippage event observed, with deletion of one repeat
being the most common event. These observations lead us to believe that during chNA
replication in C. reinhardtii, the template strand is prone to more slippage than is the

daughter strand.

92

Introduction

Microsatellites

Microsatellite sequences, also called simple sequence repeats (SSR) are short
tandem repeats of DNA motifs, one to five bases long, commonly found in the genomes
of eukaryotes and some prokaryotes (Broun and Tanksley 1996; Zhu et a1. 2000). These
repeats, which are present in both the non-coding and coding regions of genomes (Toth et
al. 2000), exhibit a strong level of instability, undergoing additions or deletions of the
repeat units leading to variations in the length of the repeated stretches (Viguera et al.
2001a). Because of their variability and abundance, microsatellite loci have been used
extensively as genetic markers in evolutionary and ecological studies of natural
populations in eukaryotes (Vasquez et a]. 2000; Ishii and McCouch 2000). Genome
rearrangements involving unstable microsatellites have been associated with human
diseases: for example, a global microsatellite instability resulting from a defect in DNA
mismatch repair has been linked to hereditary nonpolyposis colorectal cancer (Marra and
Boland 1995) and instability of the mitochondrial DNA (mtDNA) has been associated

with several neuromuscular diseases (Brown and Wallace 1994; Pinder et al. 1998).

Replication slippage

The mechanism by which the number of repeats increases or decreases appears to
involve slippage during DNA replication (Zhu et a]. 2000). Replication slippage is a
recombination-independent process that occurs during synthesis of a daughter strand of

DNA and involves the following steps: (i) arrest of DNA synthesis within a direct repeat;

93

(ii) dissociation of the replicative polymerase fi'om the template followed by transient
dissociation of the template and the nascent daughter DNA strands; (iii) incorrect re-
association of the DNA strands; and (iv) resumption of DNA synthesis (Sia et al. 2000;
Bzymek and Lovett 2001; Viguera et al. 2001b). This results in the production of one or
more unpaired repeat units on either the template or the nascent strand, and if these
unpaired loops are not repaired, a second round of replication would result in a tract that
is shorter (if the unpaired repeats are on the template strand) or longer (if the unpaired
repeats are on the nascent strand) than the original tract (Sia et al. 2000). Figure 4-1
describes the above process of replication slippage through which alterations are

generated in the length of simple repetitive DNA sequences.

Factors that contribute to replication slippage

Error avoidance and correction are essential for reducing rates of replication
slippage at short direct repeats (Kirchner et al. 2000). Replication slippage has been
shown to be stimulated by blocking the progress of replication, and several DNA
secondary structures have been shown to contribute to replication slippage at repeats
(Pinder et al. 1998). At the level of DNA replication, several factors operate to reduce
replication slippage by preventing the formation of unpaired repeat loops on the template
or nascent strand (Xie et al. 1999). Proteins, such as the single-stranded DNA-binding
protein (SSB), bind specifically to ssDNA thereby melting hairpin barriers to polymerase
progression (reviewed in Chase 1984; Meyer and Laine 1990; Lohman and Ferrari 1994).

The interaction of SSB with the ssDNA increases the processivity and thus prevents

94

 

 

 

 

--'>
—rlrlllll——

1
35.1..
/ \
_.5..-, B __..

-{llllllll—— —EI%II[IJ——

$1 -_+

—Fllllll]—

 

 

 

 

 

 

 

 

 

 

 

Figure 4-1. Model firr replicat'nn slippage. D Refers to the repeat lmits in the terrplate
strand. whereas indicates the sarrn repeat lmits on the daung strand Slippage of the
daughter strarxi results in a duplicatim event (A), while slippage of the terrplate strand during
replication results in a delet'nn event (B).

Adaptedfi-om Hancock at al. 1996 and Kokoska a al. I 999)

95

replication slippage (Kelman et al. 1998; Viguera et al. 2001a). Mutations in the
replication fork SSB increase the fi‘equency of replication slippage events. Several other
mutations have been reported that induce repeat tract instability by disrupting the
processivity of the polymerase causing disassociation of the replication complex from the
DNA. In E. coli these include mutations in the genes coding for the B clamp and the
clamp-loading/recycling machinery of the DNA polymerase IH (Saveson and Lovett
1997), while in yeast, mutations in DNA replication genes coding for the polymerase
processivity factor PCN A, the flap endonuclease (R TH I /RAD27) and the large subunit of
the yeast clamp loader (RF CI ) were shown to confer repeat tract instability (Xie et al.

1999)

In addition to the above mentioned factors regulating the rates of slippage during
DNA replication, organisms possess mismatch repair (MMR) systems that help to correct
post-replication errors after slippage has occurred. The major system directed to repair
loops containing one or a few unpaired nucleotides is the bacterial mutHLS pathway and
a related but more complex pathway in eukaryotes (Xie et al. 1999). All eukaryotic
organisms possess multiple MutS homologs (MSH proteins) and multiple MutL
homologs (MLH proteins) with the active forms being heterodimers composed of two
different MSH proteins or two different MLH proteins (reviewed in Harfe and Jinks-
Roberston 2000). Mutations in these repair genes would therefore result in a phenotype

with increased repeat tract instability.

96

Analysis of microsatellite mutations in the chloroplast DNA
Chloroplast DNA microsatellites

Microsatellites have become established as highly polymorphic markers for
genetic studies of plants (Provan et al. 1997). Well-saturated microsatellite nraps have
been developed from a number of plant species such as rice, maize, barley, Arabidopsis,
soybean (Ishii and McCouch 2000), and they have targeted mostly the plant nuclear
genomes. Recently however, microsatellites have also been identified in chloroplast
genomes and have been found in all completely sequenced plant chloroplast genomes to
date as well as in hundreds of partial chloroplast sequences from plants such as Nicotiana
tabacum, Glycine soja, Oryza sativa, and Pinus thunbergii (reviewed by Provan et al.
2001). Chloroplast microsatellites have also been reported recently from the plastid
genome of Chlamydomonas reinhardtii, a unicellular green alga that is widely used as a
model system for chloroplast DNA studies (Maul et al. 2002). Studies with barley, rice
and pine have all shown that chloroplast microsatellites can be used to reveal much
higher levels of diversity than can be observed through chloroplast RFLPs (Provan et al.
2001). In addition, high levels of length variations have been reported between
chloroplast rrricrosatellites from different species or populations of plants such as pine
(Powell et al. 1995), rice (Provan et al. 1996; Ishii and McCouch 2000), the evening

primrose (Blasko et al. 1988; Wolfson et al. 1991) and maize (Doebley et al. 1987).

97

Replication slippage rates in the chloroplast DNA (chNA) and comparisons with
other genomes and organisms

The high allelic variability at the chNA rrricrosatellites has been interpreted to
be due to the occurrence of high rates of replication slippage resulting in changes in the
length of the repeats. This could be due to the fact that the chloroplast DNA replication
machinery is prone to “slippage” or that its post-replication repair machinery is unable to
repair all the unpaired tracts created by replication slippage at the nricrosatellites.
However, evolutionary comparisons of cpSSRs among different Pinus species indicated
that SSR length polymorphisms occur in chNA at rates ranging from 3.2 to 7.9 X 10'5
per site per year (Provan et al. 1999). As a comparison, an experimental study using the
nuclear rrricrosatellites in the chickpea plant (Cicer arrietinum L), showed that the
mutation rates at fifteen (TAA)n microsatellite loci in inbred populations occurred at
much higher frequencies, ranging from 1.0 X 10‘2 to 8.9 X 10'3 per locus in one species
and 2.9 x 10‘3 to 5.0 x 10'3 per locus in another species Udupa and Baum 2001). This
suggests that relative to the nucleus, chloroplasts may have more efficient mutation
avoidance machinery especially for handling replication slippage. However, since
slippage rates are known to depend on the type of slippage substrate assayed as well as

the length of the repeats in question, these studies may not necessarily be comparable.

An experimental study in Saccharomyces cerevisiae examined the relative
frequency of mutation of similar microsatellites introduced into the mitochondrial DNA
(mtDNA) and the nuclear DNA (Sia et a1. 2000). Results obtained from this study have

shown that slippage events at nuclear nricrosatellites occurred at a range between 1.2 X

98

10.5 to 1.3 X 10'5 per cell division. In contrast, the mitochondrial rates ranged from 4 X
10'6 to 2 X 10'10 per cell division. This comparison shows that the mtDNA microsatellites
had 10 to 10,000 times lower rates of mutations than did the nuclear microsatellites.
Thus, the mitochondria of yeast may possess more efficient DNA replication and repair
systems that work to reduce the replication slippage rates at mtDNA SSRs than the
respective nuclear counterparts. However, the same conclusion cannot be made for the
plants because nuclear SSR mutation rates were based using an experimental system with
inbred populations of chickpea (U dupa and Baum 2001) whereas the chNA SSR rates

were obtained through evolutionary comparisons in pine (Provan et al. 1999).

Comparisons of the available data suggest that the nuclear genomes in plants have
100-fold higher slippage rates when compared to yeast. In humans, the rates for SSR
length polymorphisms at nuclear microsatellites ranged from 6.2 X 10‘4 to 1.2 X 10'3 per
locus per gamete per generation (Weber and Wong 1993; Dib et al. 1996) while in mice
they occurred at a range of 10'3 to 10’5 (Dallas 1992). Similar studies with cultured
marmrlalian cell lines have shown that the rates of slippage at tetranucleotide repeats
(GAAA)17 occurred at a range of7.5 x 10‘6 to 1.3 x 10‘5per cell per generation (Lee et
al. 1999). Thus a wide range of mutation rates is seen in nuclear SSRs in animals, but in
general they occurred at a much lower frequency when compared to plant nuclear
slippage rates. It should be noted that the mutation rates for SSRs fiom plant nuclear
DNAs , yeast and animal nuclear DNA were based on experimental evaluations, whereas
the chNA rates were based on evolutionary comparisons. Since the chNA repeats

usually occurred in the non-coding regions of the genome, the mutational constraints

99

were different when compared to the nuclear repeats that were looked at in experimental
assays. In addition, the directionality of the length mutations is often not clear in
evolutionary comparison as the starting repeat units are not always well established.

(Olrmtead and Palmer 1994).

An experimental assay for assessing the rates of mutations for chNA SSRs

With the goal of establishing an experimental system to determine the mutation
rates for chNA microsatellites, we created a reporter construct that allows us to monitor
the rates of replication slippage in the chNA of Chlamydomonas reinhardtii. This
lmicellular green alga has been used as a model system for the studies of photosynthesis
and photoprotection (Dent et al. 2001), and it is ideal for this study because assessment of
mutation rates requires growing large numbers of individuals. Due to its size and

unicellular nature, large numbers of cells can be grown and placed under selection, while

occupying very little space.

An appropriate microsatellite locus had to be selected so that changes in length of
the repeat units could be easily monitored. For this purpose, the existing chNA
microsatellites ill Chlamydomonas were not ideal as they occur mostly in the non-coding
regions of the genome and hence a change in their length would not produce a visible
phenotype. Thus, a stretch of repeats was introduced into a chNA gene essential for
photosynthesis (rbcL) such that it created a disruption in its reading frame. Restoration of
photosynthetic competence could only be achieved by replication slippage events as

outlined in Figure 4-2. The cell lines that were thought to have undergone replication

100

slippage could be further verified by molecular techniques as well as the exact “type” of

slippage event (insertion or deletion) at the microsatellites ascertained.

Recombination at the microsatellites

Since replication slippage occurs at regions of short direct repeats, it would be
possible that homologous recombination (mediated by RecA) could give rise to the
insertion/deletion events that restore the photosynthetic competence in the lines
containing the out-of-franre rbcL insertions. An intermolecular recombination between
repeat units located on different chNA molecules would result in an unequal crossing
over, resulting in the formation of two different types of chNA molecules: one with
deletions of the original repeat tract, whereas another molecule would contain additional
repeat units. PCR screening of the cell lines at an early stage should be able to distinguish
between recombination and slippage events: a mixture of different sized PCR products
with both deletions and the reciprocal additions will be produced by recombination

whereas replication slippage would result in the forrmtion of only one sized product.

Choice of slippage substrate

Studies of insertion/deletion events at chNA microsatellites of Oenothera have
shown that several microsatellites contain or are bordered GAAA repeat units (Wolfson
et al. 1991; Sears et al. 1994). More importantly, this repeat is also one that has been

found to vary in the chNA of the Oenothera plastome mutator lines.

101

Grow OF cells to log phase

1

Plate OF cells on media selective for photosynthetic corrrpetence

Dilution plating of OF cells on non-selective media for viable cell counts

1

Select “slipped” cells by their ability to grow on non-selective media

PCR anplification using specific primers followed by agarose gel
electrophoresis to allow these DNA differences to be readily observed

 

 

 

 

 

rbchor2 rbchev P J I 1 l J [—J
_ a b c d
(j -
1 —
-

 

 

 

 

 

 

 

 

 

b T'r’TTT‘

:1 . . . 1 V —
° 1:13: j The PCR products would then be
d :[I i i i 1 I i i [l l verified by sequching.

C> PCRprimers D 4 bp repeat ‘

 

 

 

 

 

 

 

 

 

 

 

Figure 4-2. Overview of assay for detection of replication slippage in the
chloroplast DNA of the OF cells.

 

 

 

102

A survey of microsatellites from different eukaryotes has indicated that among the
tetranucleotide repeats observed in the eukaryotic genomes, the GAAA-type was highly
represented (Toth et al. 2000). Studies of nuclear genomes of tomato, swallows and
humans have also shown that the GAAA repeats are highly polymorphic (Broun and
Tanksley 1996; Lee et al. 1999; Twerdi et al. 1999). Therefore a 28-bp stretch made of
GAAA repeats was chosen to be introduced into the rbcL gene of C. reinhardtii for our
slippage assay. This insertion would create a premature stop codon in the reading frame
of the rbcL gene (Figure 4—3) and hence Chlamydomonas cells containing this out-of-
frarne (OF) construct would be photosynthetically incompetent. Replication slippage
events would restore the photosynthetic ability in these cells and thus a screenable

phenotype could be obtained for direct visualization of replication slippage events.

In sunrmary, our objective was to provide the first experimental estimate of the
rates of replication slippage for a particular chNA microsatellite region of
Chlamydomonas reinhardtii. Using the OF cell line with the rrricrosatellite reporter
construct, we could monitor the rates of changes at the (GAAA) repeat tract by screening
for photosynthetic competence. Recovery of the photosynthetic revertants provided us
with cells that had experienced insertions /deletions at the repeat tracts through
replication slippage. The exact change could be determined in these revertants by PCR
and sequencing. The rates for each type of slippage event obtained in our study could
then be compared with the experimentally established rates of microsatellite mutations of

plant nuclear as well as with data fiom other organisms.

103

Wild type
ATG GTT CCA CAA ACA GAA ACT AAA GCA GGT GCT GGA TTC AAA GCC GGT
I V P Q T E T K G G F K A G

An out-of—frane construct with SEVEN REPEATS inserted (6 GAAA and one
GAAA)

ATGGTTCCACAAACAGAAACAAAGAAAGAAAGAAAGAAAGAAAGAAACTAA
I V P Q T E T K K E R K K E R N STOP

AGCAGGTGCTGGATTCAAAGCCGGT

An in-fraale construct and a slipped cell line with on: (GAAA) REPEAT
deleted

ATGGTT CCACAAACAGAAACAAAGAAAGAAAGAAAGAAAGAAACTAAAGCAGGT
I V P Q T E T K K E R K K E T K A G

GCT GGA TTC AAA
A G F K

A slipped cell line with FOUR (GAAA) REPEATS deleted
ATGGTT CCACAAACAGAAACAAAGAAAGAAACI‘AAAGCAGGTGCTGGATTCAAA
I V P Q T E T K K E T K A G A G F K

GCC GGT
A G

Figure 4-3. Sequences and anrino acid predictions of a part of the rbcL coding region in several
Chlamydomonas lines.

The out of-fiame line has a 28-bp insertion in its rbcL gene, creating a stop codon that renders the
cells non-photosynthetic. The two types of slipped colonies are photosynthetically conpetcnt due
to restoration of the rbcL reading frame due to deletion of one or four GAAA repeat units. The
comparable sequence of the wild-type rbcL gene is shown at the top.

104

Materiak and Methods

Chlamydomonas strains, media and growth conditions

The Chlamydomonas reinhardtii strain used for biolistic transformation and all
replication slippage experiments was the wild-type CC125 (mt+). For crosses, the two
strains, CC67 (mt-, chNA with erythromycin resistance and wild type rbcL) and
CC3455 (mt+, chNA with wild-type rbcL and a dominant negative mutation of the E.
coli RecA) were used as parents at two different stages in the experiments. All of these
above strains were obtained fi'om the Chlamydomonas Culture Collection at Duke
University (Harris 1989). C. reinhardtii was grown in tris-acetate phosphate (TAP) media
(Gorman and Levine 1965) on either agar plates or liquid media when required. The
cultures were incubated under growth lights at room temperature (22 — 25 °C). The liquid
cultures were grown on a rotary shaker under constant light at a speed of ~200 rpm The
spectinomycin-resistant biolistic transforrmnts were grown on TAP media supplemented
with 100 ug/ml spectinomycin dihydrochloride (Sigma Chemical Co.). The in-frame
transformants (IF) were grown on TAP plates with agar and spectinomycin, under high
light conditions; whereas the out—of-fi'ame cells (OF) were grown on low-light shelves.
The photosynthetic revertants were selected on HS minimal medium (Sueoka high salt

medium, Harris 1988) under high light conditions.

Construction of transforming plasmids

The transforrrring plasmids pIF6 and pOF7 were constructed by Barbara Sears at Duke

University (at the Gillham and Boynton laboratory). All manipulations for the creation of

105

glow

the slippage substrates were carried out with a small3.9 kb subclone (BamI-H-EcoRI) of
the p67 EcoRI clone, containing the wild-type rbcL gene obtained from the
Chlamydomonas Genetics Center at Duke University (Harris 1988). A 24-bp insert of
GAAA repeats at the BspMI cut site in the 6'“ codon position created an in-frame
insertion of the rbcL gene, and this construct was carried in a plasmid called pIF6. A 28-
bp insertion at the same site with an additional GAAA repeat is carried by a plasmid
named pOF7. The oligonucletotide pairs used to generate the 24bp and the 28-bp

insertions were as follows:

I F— F : 5 ’ - GTTTCTTTCTTTCTTTCTTTCTTT - 3 '

IF-R: 3’ - GAAAGAAAGAAAGAAAGAAACAAA-S’

017- F : 5 ’ -GTTTCTTTCTTTCTTTCTTTCTTTCTTT- 3 ’
OF-R: 3’ - GAAAGAAAGAAAGAAAGAAAGAAACAAA-S’

An even smaller subclone containing the 1.2 kb EcoRV-EcoRI fiagment of the BamI-II-
EcoRI subclone of p67 was used for inserting the slippage substrate, while the aadA
expression cassette for the chNA was inserted into the Mfe I cut site of the BamHI-
EcoRI subclone of p67. Figure 4-4 shows the diagrammatic representations of pIF6 and

pOF7 plasmids that were then transformed into the GM 2163 bacterial strain.

Chloroplast transformations

The chloroplast transformations of Chlamydomonas were done using the biolistic
transformation method as described by Kindle et al. (1991) and Boynton and Gillham
(1993). The wild type c0125 strain was used as the recipient strain and the plasmids pIF6
and pOF7 were used to transform the cells. For the transformations, the cc125 cells were

grown under high light on a shaker at 200 rpm. to a cell density of about 2 X 106

106

 

 

 

inserts
rbcL atpA
‘ - - - - - ->
r i 1 r E -
BamHl Mfel EcoRV BJMI EcoRI

Figure 4-4. Diagramrnatic representation of the transforming plasmids (pIF6 and pOF 7).
The colored boxes represent the chloroplast encoded genes of C. reinhardtii, and the
arrows indicate the direction of their transcription. The two triangles represent the
insertions in the transforming plasmid construct: the aadA cassette was inserted between
two chloroplast genes and is the selectable marker, the IF and OF insertions were put in
plasmids pIF6 and pOF7 respectively. IF contains a 24-bp insertion at the 6th codon
position of the rbcL gene whereas OF has a 28-bp insertion in the same position. The
enzymes used for cloning are also indicated in the figure. The size of the BamHI—EcoRI
with the aadA gene is 5.5 kb. The insert was cloned into the BamHI-EcoRI site of the
pUC8 vector (size of the transforrrring plasmid = 8.1 kb).

107

cells/ml. The cells were then harvested by centrifugation and resuspended in TAP to a
final concentration of 2 X 108 cells/ml. A 1 ml aliquot of the cells was added to 1 ml of
melted sterile agar (42°C) and the two aliquots of 0.7 ml of this mixture were spread on
TAP plates. The M10 tungsten particles for bombardment were coated with the
transforming plasmid DNA (1 pg/ml) along with sterile aliquots of Soul of 2.5M CaClz
and 20 pl of 0. 1M sperrnidine. The particle bombardments were carried out using 1300
psi rupture membranes and each plate was exposed to a vacuum followed by a helium-
propelled shot. From each plate the cells were resuspended in 1.5 ml of TAP and then
spread on two plates of TAP and agar media supplemented with spectinomycin for direct
selection of the transformants. The plates were put under high light when the pIF6
plasmid was used for transformation, while the pOF7 transforrmnts were placed under

low light conditions.

Tests for photosynthetic competence

The in-frarne (IF), out-of-frame transformants (OF) and wild type cells were grown on
different types of media and under two different light conditions to compare their
photosynthetic competence. Cells were streaked on solid TAP (Gorlmn and Levine 1965)
or HS media (Harris 1988) and then placed on low light and high light shelves to assay

for photosynthetic competence. Similar tests were also performed with liquid media.
Assay for rates of replication slippage in the 0F cells

The OF cells were grown in liquid TAP culture to late log phase (2-4 X 106 cells/ml) for

about 7 days at room temperature under low light on a rotary shaker at a speed of ~200

108

r.p.m Cell counts were determined by using the compound microscope and the standard
hemocytometer method as described in the Chlamydomonas sourcebook (Harris 1988).
250 ml of the cells was sterilely harvested by centrifugation at 8000 X g in a Sorvall
centrifuge using a GSA rotor at 4°C for 10 mins. The supernatant was discarded and the
cells were resuspended in a volume of TAP medium that would result in a final

concentration of 2 X 108 cells/ml.

An aliquot of these cells were used for dilution platings to determine the number of
viable cells plated. 100 pl aliquots from the 10's, 10'6 and 10'7 dilutions were plated on
TAP plates and then grown under lights at room terrrperature. The number of colonies
that grew on these plates was then counted to assess the total number of viable cells that

were plated for the experiment.

The termining volumes of the concentrated OF cells were then plated as 500 pl aliquots
on HS minimal media with agar and supplemented with 100 ug/ml of spectinomycin.
These plates were put under high lights on a shelf at room temperature. The total number
of photosynthetically competent colonies that grew on these plates were counted, and
then transferred using sterile toothpicks to HS + spectinomycin plates for further DNA

analyses.
Total DNA isolation

Chlamydomonas total genomic DNA was prepared using as described in Chapter 3 of the

thesis.

109

PCR amplification

DNAs were extracted from a sarrrple of thecolonies that had reverted to photosynthetic
competence and were used for PCR arrrplifications, along with the DNAs from the
original OF and IF transformants and wild type cc125 cells. The rbcL 5 (5’-
GGCCCTTTCTATGCTCGACTG -3 ’) and rbcL mid (5’-CCGAATACGTTACCT AC -
3’) primers were used in the PCR arrrplification reactions as the forward and reverse
primers respectively to amplify ~560-bp fragments (+24/28)bp from the rbcL genes of
the wild type, OF, IF and revertant colonies. In some experiments in order to visualize the
size differences among the slipped colonies, a smaller PCR fragment was generated
(~180 bp) using a different set of primers, rbcL for2 (5’-
CTACGTAATCAGGTGTGTAG-3 ’) and rbcL rev (5’-
CCGGACAGATTAATTTTAGGA-3 ’). DNA (1-5ng) was added to individual 100p]
reactions containing 200 umol of each dideoxynucleotide, 2.5 units of Taq polymerase
(GIBCO BRL), 10 pmoles of each primer and 1 X PCR reaction buffer (GIBCO BRL).
Conditions for the 30 cycles of PCR were as follows: denaturation at 95°C for 2mins,
85°C for 5 nrins, followed by 94°C for 40 sees; 30 secs at 48°C to anneal the primers with
the template and 30 secs at 7 2°C for DNA extension. After all the cycles the final
extension was done for 10 mins at 72°C. All the PCR reactions were performed with a
nrinicycler model FTC-150-16 (MJ Research Inc., Watertown, Mass).

The PCR amplified products were run on a 1.5% agarose gel (for the 560-bp fragments)

or a 2% agarose gel (for the 180-bp fragments).

110

 

[hr

10

fol

A different set of primers was used to check for the presence of the deltaN RecA gene in
the Chlamydomonas strains obtained in crosses, generating a lkb PCR product. The
forward primer was (5’-GTGCCATGGGTTCGCTTTCACTGGATA—3 ’) and the reverse
primer was (5’-CTGGCATGCTTAAAAATCTTCGTTAGTTTC-3 ’. The PCR program
used for this reaction is as follows: denaturation at 95°C for 2 min, 85°C for 5 min,
followed by 94°C for 1 min; 2 min at 36°C to armeal the primers with the template and 2
rrrin at 68°C for DNA extension. The final extension after all the cycles was done for 10

mins at 72°C. The PCR products were run on a 1% agarose gel.

Sequencing and Computer Analyses

The 560-bp PCR products using the rbcLS and rbcL mid primers from several samples
were sequenced to determine the size of the rbcL segment in the region where the repeats
were inserted. The purified PCR products (Chapter 3 for details) were sent for sequencing
to the Michigan State University Genomics Facility, East Lansing, MI
(www.genonr_ics.m_su.edu) and the internal primer rbcL for2 was used to prime the

sequencing reaction. Computer analyses were done as described in Chapter 3.

Mutation Rate

To obtain the replication slippage rate, the total numbers of colonies that had reverted to
photosynthetic competence were determined for each experiment, and then using the
following formula, the mutation rate (u) was determined:

Number of photosynthetic revertinflolonies
Total number of viable cells plated (value obtained fi'om dilution plates)

111

Cr:

CC

811

Chlamydomonas crosses

Crosses were done as described by Harris (1988) and in order to increase the biparental
transmission of the chNA in the zygotes, a brief (2 min) UV treatment of the rrlt+ parent
was done before the crosses (Sager and Ramanis 1967). The meiotic progeny from the
crosses were dissected by Barbara Sears according to Harris (1988) and the resulting
lines were crossed to both mt+ and mt‘ gametes in microtiter plates in order to determine

their mating types.

Results

Introduction of replication slippage substrates into the chNA of C reinhardtii
Wild-type C. reinhardtii cells (CC125) were transformed using the biolistic
transformation method for chNA (Kindle et a1. 1991) with the plasmids pIF6 and pOF7
(Figure 4—4) that contained (GAAA), repeat stretches in the “in-flame” and “out-of-
fiame” versions respectively. The transformants were selected based on their resistance to
spectinomycin conferred by the presence of the aadA cassette in the transforming
plasmids. Several rounds of subcloning were required to obtain homoplasnric chNA
molecules all containing the repeat insertions in their rbcL genes. PCR amplification
confirmed the insertion of the slippage constructs in the rbcL gene of the transformants.
The PCR amplified products using the rbcL5 and rbchid primers from the IF and OF
strains were visualized on an agarose gel (Figure 4-5), and they were 24-bp and 28-bp

larger than the PCR product from wild-type cells (~580-bp). Thus, the

112

 

Figure 4-5. Agarose gel showing the PCR arrplified
products fiom the wild-type (WT), ill-frame (IF) and out-
of- frame (OF) cells. The primers lsed for arrplification

were the rbcL5 and rbchid primers, which generate
560-bp sized fragment for the wild-type cells and 584-bp
or 588-bp sized products for the IF and OF cells

respectively.

113

wild-type rbcL gene in the IF and OF strains was replaced by the in-fi'ame and out-of
frame constructs through homologous gene replacement. Since the OF product was only
4 bases larger than the IF product, the size differences were not clear on the agarose gel.
Therefore the IF and the OF lines were sequenced through the insertions in their
respective rbcL genes and the sequence alignments are shown in Figure 4-3. As seen in
the figure, the OF line had a 4-bp larger insertion than the IF line, and based on the
sequence information, it can be predicted that a premature stop codon is present in the

rbcL gene of the OF line that would make the cells non-photosynthetic.

Test for photosynthetic competence in the wild-type and transformed cells

The IF (in-frame) and the out—of-fi‘ame (OF) transforrmnts were then grown on
media with acetate under low light (non-selective condition), and without acetate under
high light (selective condition) to test for their photosynthetic competence. As a control,
wild type CC125 cells were grown under similar conditions. As shown in Table 4-1, the
IF and the CC125 cells were photosynthetically competent as they were able to grow on
media without acetate and also under high light conditions. In contrast, the OF cells were
not photosynthetically competent and died on media without acetate and also displayed a
high-light sensitive phenotype, similar to other non-photosynthetic mutants of C.

reinhardtii.

114

 

 

 

 

 

acetate + _ + _
light high high low low
Wild-type cells Growth Growth Growth Growth
(ccl 25)
ln-frame cells Growth Growth Grth Growth
(11“)
Out-of-framc No growth No growth Growth No Growth
cells (OF)

 

 

 

 

 

 

 

Table 4-1. Phenotypes of the wild-type (CC125) and the transformed cell lines (IF and OF) of C.
reinhardtii lmder different conditions.

The presence of acetate in the media provides an additional carbon somce and thus enables both
the photosynthetically conpetent and non-conpetent cell lines to grow on them

115

Rates of mutation at the (GAAA)... microsatellites in the chN A of the OF cells

To phenotypically monitor the replication slippage event at the (GAAA), repeats
in the rbcL gene, the OF cells were plated on media that would select for the restoration
of photosynthetic competence. To accurately calculate mutation rates, the number of
viable cells plated was determined fi'om dilution plating of the OF cells on a non-
selective medium. Procedures for determining the cell viability and mutation frequency
are described in the materials and methods section. Since the reading frame for the rbcL
gene in the OF line could only be restored through the elimination or addition of the
repeat units, the numbers of cells growing on the selective media would represent the
number of cells that had “slipped”. Replication slippage at the (GAAA)n microsatellite
region in the three different experiments ranged from 1-70 in 106 viable cells plated
(Table 4-2). A subset of the slipped cells fiom each experiment was then randomly

chosen for characterization of the changes at the chNA microsatellite locus.

Molecular analysis of slippage events at the (GAAA).. microsatellite site
Since the microsatellite reporter construct in the OF cells consisted of a 28-bp
insertion composed of a stretch of (GAAA)ll repeats, a loss of one repeat unit or an

insertion of two repeat units were two of the expected slippage events that could restore

116

 

 

 

 

 

 

 

 

 

 

Experiment Rates of # of # of Frequency of each
# reverison to photosynthetic colonies deletion event
photosynthetic revertants with
competence checked by deletions
(per 106 viable PCR due to
cells) replication
slippage
-l -4
repeat repeats
1 1.09 29 29 15 14
2 10.69 16 15* 9 6
3 69.75 62 62 34 38

 

 

Table 4-2. Rates of spontaneous mutations to photosynthetic competence in the OF strain from

three different trials.

* Out of the 16 colonies checked by PCR anplification for the type of slippage event, one had

undergone recombination while all others showed deletions of the repeats.

117

 

the reading frame of the disrupted rbcL gene. In addition, a deletion of 4 repeats (16 bp of
the original 28bp construct) or an insertion of 5 repeats (creating a 48 bp insert at rbcL)
could theoretically also restore photosynthetic competence in the OF cells. A subset of
the “slipped” cells with restored photosynthetic corrrpetence was chosen from each of the
three experiments to observe the types of slippage that had occurred and also to analyze

their respective frequencies of occrurence.

PCR amplification of the DNA fi'om the slipped cells, using rbcL-specific primers
(rbcL for2 and rbcL rev) showed that among all the “slipped” colonies demonstrating a
restoration of photosynthetic competence, deletions were the only type of slippage event
that was observed (Figures 4-6 and 4-7), except in one cell line (Line 3 of Figure 4-7). In
both the figures, the lane OF shows the PCR amplified product fiom the out-of-frame
cells that is 169-bp long including the 28-bp insertion at the sixth codon position in the
rbcL gene. Two different variants of the PCR amplified products were observed in the
slipped cell lines: both were smaller than the 169-bp fragment found in the OF cells, but
larger than the amplification product (141-bp) fiom the wild-type C. reinhardtii cells
(Lane WT). The slipped lines labeled 1 and 2 in Figure 4-6 have a PCR product that is
slightly lower than the band obtained from the OF cells, indicating that it had probably
undergone a deletion of a single 4-bp repeat unit. The PCR product in lane 3 in Figure 4-
6 is smaller than the ones that show a loss of one repeat unit (165-bp) but is larger than
the wild-type fragment. Since the only other deletion event that could have given rise to
the restoration of the photosynthetic competent phenotype was the loss of four repeat

units, the size of the PCR product was postulated to be 153-bp.

118

Slipped Colonies
MWl l 2 3I WT 0F

369-bp

246-bp

123-bp

 

-1 -1 -4

Figure 4-6. Agarose gel electrophoresis showing the PCR anplified products from
the slipped colonies with one or four repeat lmits deleted The slipped colonies are
nurrbered from 1 through 3. (—1) and (—4) refers to the nunber of repeats deleted in
each line with respect to the OF strain (which contained 7 repeats). MW indicates
the 123-bp DNA ladder, WT refers to the wild-type C. reinhardtii strain, and OF is
the original transforrrnnt with the 28-bp out-of-frame insertion in the rbcL gene.

119

From more than 100 revertant lines that were checked by PCR, only one showed a
product size higher than the PCR fi'agment from the OF cells. In that cell line, two PCR
products were seen in one lane, indicating the cells were heteroplasmic. In Figure 4-7,
Line 3 has an upper band, which is a higher sized product than the OF PCR product, and
a lower band that is slightly smaller sized than the 169-bp fragment seen in OF cells.
Most likely, this condition reflects a recombination event, with the two products
representing the reciprocal recombinant molecules generated due to recombination

between two chNA molecules with the GAAA repeats.

To verify the interpretation of events based on the size of the PCR products,
sequencing was performed on representatives of the two main size classes from the
slipped lines. As controls, sequencing was also done on the wild-type (cc125) and the OF
lines. PCR amplification was performed using primers that give a larger product (rbcL 5
and rbcL mid) from all the selected lines. An internal primer rbcL rev, was used to prime
sequencing reactions from the PCR products. Sequence comparisons confirmed my
interpretations of the PCR data. As predicted, of the two slipped colonies sequenced, one
showed a loss of a GAAA repeat unit whereas the other cell line had restored

photosynthetic competence due to the loss of four GAAA repeats.

120

Revertants Revertants
1 2 3 4 5 6

MWr———fiun OFr———————nMW

123bp

 

Figure 4-7. Agarose gel showing the PCR amplified products from a subset of the
revertant colonies along with the wild-type (WT) and the out-of-t'rame (OF) cells MW
represents the molecular weight marker (123-bp DNA ladder).

121

Slippage rates in a C. reinhardtii strain with a recombination-deficient background

A C. reinhardtii strain was constructed in which the chloroplast genome carried
both the rbcL gene with the 28-bp microsatellite insertion and a deltaN RecA gene (by
crosses described in Figure 4-8). Cross 1 was performed to move the chloroplast
transgenes into a rut strain. Since the mating type locus is a nuclear gene, half of the
progeny of cross 1 were mt' cells. Almost all of the progeny of cross I inherited the aadA
and the rbcLOF genes since chloroplast markers are transmitted predominantly fi'om the
maternal (mt+) parent. Cross 2 was designed so that a recombinant could be selected that
would have both the rbcLOF marker and a deltaN RecA gene in its chNA. The deltaN
RecA gene which is a dominant-negative mutation was introduced by Cerutti et al. (1995)
into the chNA of a C. reinhardtii strain. In the new strain, chNA recombination is
reduced five-fold and the cells are sensitive to chloroplast-targeted mutagens, indicating
that the E. coli gene is able to have a negative impact on the endogenous chloroplast

recombination and repair system.

The deltaN RecA strain used in the cross was first verified for the presence of the
RecA gene by PCR amplification using appropriate primers (described in Materials and
Methods). Since these cells also show reduced survival rates in the presence of a
mutagen, 5mM 5-fluorodeoxyuridine (FdUrd) relative to wild-type cells (Cerutti et a1.
1995), experiments were done to verify the high sensitivity of the deltaN RecA strain to
the above mutagen (data not shown). In C. reinhardtii about 5% of the zygotes produce
colonies have biparental inheritance of chNA. The deltaN RecA gene was carried in the

chNA of the mt+ parent, whereas the slippage construct was contained in the chNA of

122

 

 

 

Cross]: Transfer of the chNA slip construct of the OF line rrlt+ into a mt' background.

(mt‘) or x (mt') cc67
(chNA: spec-resistant, rbcL -OF) (chNA: spec-sensitive, wild type rbcL)
(aadA)

i

Select cell line that is mt' containing the aadA gene and rbcL-OF.

 

 

 

Cross 2. Introduction of the chNA slippage reporter construct into a strain with a
chloroplast reconirination-deficiency (containing a dominant negative E. coli RecA gene that
lacks the first 42 amino acids).

(mt‘) cc3455 x (mt') or
(chNA: deltaN RecA, spec-sensitive) (chNA: spec-resistant, rbcL-OF)
(wild type rbcL) (aadA)

1

Select a line containing the aadA gene, rbcL-OF and the deltaN RecA gene.

 

 

 

Figure 4-8. Crosses to introduce the slippage construct (GAAA repeats) into a chloroplast DNA

subject to reduced homologous reconirination.

123

mt' parent. The slippage construct included the aadA cassette that confers spectinomycin-
resistance to cells, and thus biparental progeny could be selected based on their resistance

to spectinomycin.

An initial cross in which the zygote progeny were immediately selected for
spectinomcyin resistance failed to produce progeny with the desired combination of
markers, so the mt+ parent was UV-irradiated immediately prior to mating to increase the
frequency of biparental progeny (Harris 1989). The 2-minute exposure of the rrlt+
gametes to UV-irradiation fi'om a hand-held lamp resulted in 15.3 % biparental zygotes
(data not shown) and tetrad analysis was performed to obtain individual meiotic progeny.
The dissected meiotic progeny were allowed to form colonies on non-selective media and
then spectinomycin-resistant cells were selected and checked for the presence of both the
slippage construct and the deltaN RecA gene in the chNAs by PCR amplification as

described in Materials and Methods.

The chNAs fi'om the progeny were amplified using the rbcL-specific primers to
check for the presence of the 28-bp insert in their rbcL gene. Figure 4-9 shows that all the
progeny from Cross 2 (a through q) that showed resistance to spectinomycin, also
contained the 28-bp out-of-frarne insertion in their rbcL gene. Figure 4-10 shows the PCR
amplified products fi'om a subset of these progenies fi'om Cross 2 (a-m) using the RecA
primers. Among the tetrad progenies showing resistance to spectinomycin, only progeny
c, g, f, j, k and m had the deltaN RecA gene in their chNAs. Combining the results from

figures 4-9 and 4-10 two colonies fi'om progeny “1” that contained both the

124

:\lci0tic progeny Meiotic Progelr) Mciotic

progeny

-u.

n “t 0 p q

 

Figure 4—9. Agarose gel showing PCR arrplified products fi'om the meiotic
progeny of the cross CC3455 X OF, to check for the slippage insert. The lane
wt refers to the PCR product from the wild-type Chlamydomonas (CC125).

125

+abcdwtgefhij+klm~

 

Figure 4-10. Agarose gel showing PCR arrplified products of the meiotic progeny
from the cross CC3455 X OF, to test for the presence of the delta N RecA gene. (+)
represents the parent CC3455 which is the positive control containing the delta N
RecA gene, whereas wild type CC125 is the negative control (-) .

126

slippage construct and the deltaN RecA gene were chosen for future replication slippage
experiments. Since these cell lines have a chloroplast recombination-deficiency, the data
obtained should allow us to determine the impact of recombination on insertion/deletion
events at the GAAA-repeat region. The results of Table 4-3 show that spontaneous
replication slippage conferring photosynthetic competence even in a recombination-
deficient background occurred at a range of 52-54 per 106 viable cells plated. These rates
do not differ fi'om those in a recombination-proficient background (1-70 per 106 viable

cells plated) of C. reinhardtii.

Discussion

Replication slippage has contributed significantly to genome evolution in
organisms as diverse as eubacteria, humans and plants (Hancock 1996). Replication
slippage affects the size of genomes in both ORFs and non-coding regions in bacteria
(Gur-Arie et al. 2000) and in eukaryotes (Metzgar et al. 2000). Polymorphisms of
microsatellite regions have made them useful as markers in genetic studies especially in
forensics and genome mapping (Udupa et al. 1999). In plants microsatellites have been
reported from the nuclear DNA of a number of species; examples include Arabidopsis,
soybean, corn, tomato, barley, and chickpea (Broun and Tanksley 1996; Udupa and
Baum 2001 ). Chloroplast genomes have also been shown to contain large numbers of
microsatellites or SSRs, with their presence documented in algae such as C. reinhardtii
(Maul et a]. 2002), in bryophytes such as Marchantia polymorpha (Powell 1995a), and

also in seed plants such as the gymnosperm Pinus and in angiosperms such as Glycine

127

 

 

 

 

 

 

 

Colony # # of viable cells plated # of cells that showed Rate of mutation
(x 1 0") photosynthetic to photosynthetic
competence competence (per
10‘ viable cells
plated)
1. 65 3360 51.7
2. 104.2 5 640 54.1

 

Table 4-3. Rates of spontaneous replication slippage events in the chloroplast DNA of

Chlamydomonas reinhardtii containing the OF construct in a recombination-deficient

background

128

 

max, Zea mays and Oryza sativa (Doebley et al. 1987; Hiratsuka et al. 1989; Powell et al.

1995a; Powell et al. 1995b; Provan et al. 1996).

Although the existence of microsatellites and their variability has been well
documented in the chloroplast genomes of many plant species, few studies have been
done to observe their rates of change. Phylogenetic comparisons have determined the
relative rates of insertion/deletion events at these chNA microsatellites in an
evolutionary context, and results have indicated that in comparison to nuclear DNA, the
chloroplast DNA has a low mutation rate for microsatellites (Provan et al. 1999; Udupa
and Baum 2001). The evolutionary studies have also indicated that insertion/deletion
events at the chNA nricrosatellites occur at a much higher rate than do base
substitutions (Clegg et a1 1994; Provan et al. 1999; Sears et al. 1994). However it should
be noted that the assessment of mutation rates at rrricrosatellites is more difficult to assay
than base substitutions, as one change often obscures the next. In addition, evolutionary
data is often unable to assess the directionality of repeat length changes at the
microsatellites. Our goal was to establish a reporter assay that would allow us to
determine the rates of replication slippage at the chNA nricrosatellites and also

ascertain the exact nature of the changes at the repeat regions (insertions/deletions).

In order to experimentally assess the rates of replication slippage in the chNA, a
model system for chloroplast studies was used: C. reinhardtii, the unicellular green alga,
was ideal for such studies as its chNA can easily be manipulated and also large

numbers of cells can be observed, utilizing very little space. Since the naturally occurring

129

microsatellites of the chNA are mostly present in the non-coding regions of the plastid
genome, a reporter construct was created to monitor replication slippage events in vivo
using a native chNA gene, rbcL, in order to provide a phenotypic selection for the
slippage event. The rbcL gene product is essential for the fixation of carbon fi‘om C02,
and an out-of-frame insertion in the gene would therefore disrupt its reading flame and
would result in the death of C. reinhardtii cells on media without an organic carbon
source (acetate). Therefore if a stretch of microsatellites was inserted into the rbcL gene,
disrupting its reading frame, the C. reinhardtii cells with this construct would not be able
to grow on media lacking acetate. Slippage in the chNA could then be phenotypically
observed by the restoration of photosynthetic competence in these cells: a deletion of one
or more repeat units or an insertion event due to replication slippage would result in the
above phenotype. However, since this assay required that the rbcL gene product would be
functional with the addition of extra codons, initial experiments were done to prove that
an in- frame insertion of additional codons in the rbcL gene of C. reinhardtii would allow
the cells to remain photosynthetically competent. Results showed that an iii-flame
insertion (IF) of 24-bp composed of five GAAA repeat in the chloroplast encoded rbcL
gene of C. reinhardtii did not affect the photosynthetic competence of the cells (Table 4-
1 and Figure 4-5). On the other hand, an out-of-frame insertion (OF) of 28-bp composed
of six of the same repeats resulted in cells that were unable to live on media without
acetate. Therefore, the phenotypic assay for monitoring replication slippage in vivo in the

chNA could be used for the OF cells.

130

 

Rates of replication slippage in the OF cells could then be calculated by
comparing the numbers of cells with restored photosynthetic ability with the original
numbers of cells plated In three experiments, the rates of insertion/deletion events in the
chNA at the GAAA-repeat site occurred fi'om 1 — 70 per 106 viable cells plated. There
is however a limitation in this experimental approach: the rates obtained in our study
represent only the slippage events that have restored the rbcL gene function and thus,
changes that would have occurred at the repeats without restoring the reading fiame do
not contribute to the rates. However, slippage events involving a single repeat have been
reported to be the most frequent in bacteria (Levirrson and Gutrnan 1987) and plants
(Provan et al. 1996). In comparison, my results have shown a nearly equal frequency of
deletions of one and four-repeat units, with one repeat deletion being slightly more

fi'equent.

Several factors affect the rates of replication slippage, these include: the size of
the repeat units, length of the microsatellite stretch, sequence context of the repeats
thermelves and finally the position of the repeat units in the genome (Blasko et al. 1988;
Clegg et al. 1994; Gragg et al. 2002; Harfe and Jinks-Robertson 2000; Yanrada et a1.
2002). To enable experiments to proceed that will determine whether different substrates
in the same position of the rbcL gene would slip at similar rates, I have created a C.
reinhardtii strain that has a stretch of adenines inserted at the same site. In addition, a
new reporter gene containing the same GAAA repeats is being constructed that can be
placed at different regions of the chNA to assess the effects of genome context in

replication slippage rates.

131

 

Features of replication that are intrinsic to chloroplast genomes can explain the
seventy- fold difference in the rates for replication slippage in the chNA among the
three different trials. Organelle DNA replication occurs in a relaxed rather than stringent
fashion: DNA is replicated randomly at every round. In such a case one type of chNA
molecule may be replicated more often than the others by chance alone (Birky and Walsh
1992; Birky 1994). In addition, the partitioning of the chNA molecules to daughter cells
is also under relaxed control, which means that during chloroplast division, the chNA
molecules are apportioned randomly between the daughter cells. In my experiments, a
higher fi'equency of replication slippage may have been observed if the initial slippage
event occurred early, while fewer slippage events may have been observed if an event
occurred late during the growth of the culture. In order to avoid a founder’s effect,
precautions were taken to exclude any pre-existing slippage present in the starter cells of
the assay: these included plating of the early pre-culture cells on media without acetate to
look for early revertants, and checking the sizes of their rbcL gene with PCR
amplification. Furthermore, the three-week delay in growth of the revertant
photosynthetic colonies on the selective plates indicates that preexisting revertants were

absent in these trials.

Results from all three experiments share a common feature: with only one
exception, all the slipped colonies had experienced deletion of one or forn' repeat units
(Table 4-2 and Figures 4-6 A and 4-6 B). Although deletion of one repeat unit was the
most corrrrnon slippage event (58 out of 106), deletion of four repeats was also frequent

(48 out of 106). Since the addition of two repeat units could have also restored the

132

 

photosynthetic competence, we conclude that the chloroplast DNA of Chlamydomonas
has a bias towards deletions rather than insertions by replication slippage. We know that
addition of two repeat units can occur spontaneously and can confer photosynthetic
competence because we have observed a 2-repeat addition twice in OF lines under long-
term maintenance even though none were seen in the experiments described here. These

observations give us confidence in concluding that the template strand is more prone to

 

:-
slippage than is the daughter strand in the Chlamydomonas chNA (Figure 4-1). A
deletion bias during replication slippage has been reported for other organisms, including
chNA of a number of plant species such as petunia and alfalfa (Aldrich et al. 1988),
mitochondrial microsatellites of yeast (Sia et a1. 2000) and nuclear genomes of animals i

such as snails (Weetman et al. 2002). The deletion bias observed in this report was
surprising because short dispersed repeats in the Chlamydomonas chloroplast genome
have proliferated to a great abundance (Maul et al. 2002). Since these repeats occur
mostly in the intergenic regions, and hence it is not known whether the deletion bias seen
in our experiments may reflect constraints related to function. Studies in E. coli and yeast
have shown that general differences in the types of repeats and their expansion exist
between coding and noncoding regions of the genome (Tautz et al. 1986', Gur-Arie et al.
2000; Metzgar et al. 2000). It has been hypothesized that this limitation of microsatellite
expansion in coding regions is due to selection against frameshifl mutations as well as
due to specific constraints such as maintenance of the correct distance between regulatory
elements or prevention of formation of nucleic acid secondary structures that would

affect gene expression (Metzgar et al. 2000).

133

In order to determine whether the frequency of duplication/deletion of short direct
repeats is affected by recombination in the OF lines, the rates of change were determined
in recombination-deficient backgrounds. My results have shown that spontaneous
insertion/deletion rates in these lines were similar to those in the recombination-proficient
background (Tables 4-2 and 43), indicating that the photosynthetically competent cells
arose mainly due to replication slippage and not recombination. Furthermore, screening
the photosynthetic revertants obtained fiom all three experiments using PCR
amplification enabled me to distinguish between recombination and slippage events as
the cause of restoration of photosynthetic competence. Among all the colonies screened,
only one colony showed apparent recombination products (Figure 4-7, line 3) where one
of the two bands was around 4-bp larger than the OF band and the lower band
represented the reciprocal recombinant molecule with a 4-bp deletion. In contrast, PCR
amplification of all the other photosynthetic revertants showed either a single smaller
sized band than OF, representing the slippage event in homoplasmic condition or in some
cases a heteroplasmic condition was observed in which the original size of the
microsatellite tract and a deleted product was seen (for example, Figure 4-11 lane 3).
Collectively, these observations have led me to conclude that the insertion/deletion events
at the microsatellites that resulted in the restoration of a photosynthetic phenotype were

almost solely due to replication slippage and not recombination.

134

 

Slipped colonies

 

W1 2 3 4 WT OF

269-bp

123—bp

 

Figure 4-11. Agarose gel showing the heteroplasmic condition of the slipped
cell line # 3 containing chNAs with the original OF -size insert as well as a
4—repeat deleted chNA.

MW refers to the wild—type product and theOF-lane contains the PCR
product from the OF line. The slipped colonies are numbered (tom 1 through
4.

135

An experimental assay in chickpea has indicated that replication slippage in the
nuclear DNA occurs at rates ranging fi'om 1 X 10'2 to 3.9 X 10’3 per locus per generation.
In contrast to this, the rates of slippage in the chNA from our experimental study with
Chlamydomonas are at a much lower frequency. Comparing these observations could
lead one to hypothesize that the chloroplast may possess more efficient replication or
repair machinery than the nucleus resulting in a lower slippage rate. However, it needs to
be stated that these studies include substrates that have different sequence contexts, and
therefore, a true comparison cannot be made between the mutation rates of these two

genomes.

In Chapter 3 of the thesis, rates of base substitutions at a particular target site in
the chNA were assayed experimentally and the results showed a frequency of 3 X 10'11
to 2 X 10““. Those values are 100-10,000 times lower than the frequency of chNA
replication slippage. A similar trend was also observed in phylogenetic comparisons of
chNA (Provan et al. 1999; Wolfe et al. 1987). It should be mentioned that the rates of
mutations in our studies were observed for two different regions of the chloroplast
genome: the base substitution rates were observed for the 168 rRNA gene that is within
the inverted repeat region (IR) while the slippage events occurred in the single copy
regions. Because the [Rs contain identical repeat units, copy correction can occur and act
to either eliminate or fix new mutations occurring in the IR region. If copy correction is
likely to occur one way or the other, the mutation rates of genes in the [Rs may be half of
that or two-fold higher than genes in the single copy regions. Under any circumstance,

my results show that base substitutions occurred at lOO-l0,000 times less often than did

136

replication slippage of the chNA. Hence, I believe that the mismatch repair machinery
of the chloroplasts may be able to handle base mismatches more efficiently than the
mismatches generated due to slippage. It could also be true that the replication machinery
of the chNA may have intrinsically low rates of processivity (provided by factors such
as SSB, Clamp loader, B-sliding clamp, etc.) that would stall the replicating polymerase

and generate a high slippage rate.

In conclusion, this study provides the first experimental assessment of the rates of
replication slippage in the chNA. Deletion events were by far the most frequent
slippage event observed in the OF cells containing the (GAAA),I microsatellites, and
slippage occurred at a range of 1 to 70 X 10‘. Only one recombination event was
observed. These observations lead me to believe that primarily the template strand is
susceptible to slippage during replication of the chNA in Chlamydomonas reinhardtii.
The reporter construct can now be used to monitor replication slippage events using
several different types of repeats to observe whether sequence context plays a role in
replication slippage rates in the chNA. These rates also provide a background rate for
insertion/deletion events at the chNA of C. reinhardtii, which is an essential piece of
information for isolation of mutator strains that show elevated levels of slippage in the
chNA. Isolation of the defective genes in these mutator strains will provide us with an

insight into genes that control chNA replication slippage rates in C. reinhardtii.

137

 

REFERENCES

Aldrich J, Cherney BW, Merlin E, Christopherson L (1988) The role of
insertions/deletions in the evolution of the intergenic region between psbA and trnH in
the chloroplast genome. Curr Genet 14: 137-146

Birky CW, Walsh JB (1992) Biased gene conversion, copy number and apparent
mutation rate differences within chloroplast and bacterial genomes. Genetics 130: 677-
683

Birky CW (1994) Relaxed and stringent genomes: Why cytoplasmic genes do not obey
Mendel’s laws. J Hered 85: 355-365

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

Boynton JE, Gillham NJ (1993) Chloroplast transformation in Chlamydomonas. Methods
Enzym01217: 510-536

Broun P, Tanksley SD (1996) Characterization and genetic mapping of simple repeat
sequences in the tomato genome. Mol Gen Genet 250: 39-49

Brown MD, Wallace DC (1994) Molecular basis of mitochondrial DNA disease. J
Bioenerg Biomembr 3: 273-89

Byzrnek M, Lovett ST (2001) Instability of repetitive DNA sequences: The role of
replication in multiple mechanisms Proc Natl Acad Sci USA 98(15): 8319-8325

Cerutti H, Johnson AM, Boynton JE, Gillham NW (1995) Inhibition of chloroplast DNA
recombination and repair by dominant negative mutants of Escherichia coli RecA. Mol
Cell Biol 15: 3003-3011

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

138

Chase JW, L’ltalien JJ, Murphy JB, Spicer EK, Williams KR (1994) Characterization of
the Escherichia coli SSB-113 mutant single-stranded DNA-binding protein. J Biol Chem
259: 805-814

Clegg MT, Gaut BS, Learn GH, Morton BR (1994) Rates and patterns of chloroplast
DNA evolution. Proc Natl Acad Sci USA 91: 6795-6801

Dallas JF (1992) Estimation of rrricrosatellite mutation rates in recombinant inbred strains
of mouse. Marrnn Genome 3: 452-456

Dent RM, Han M, Niyogi KK (2001) Functional genornics of plant photosynthesis in the
fast lane using Chlamydomonas reinhardtii. Trends Plant Sci 6: 364-371

Dib C, Faure S, Fizames C, Samson D, Drouot N, Vignal A, et al. (13 authors) (1996) A
comprehensive genetic map of the human genome based on 5264 microsatellites. Nature
380: 152-1 54

Doebley JF, MA DP, Renfi'oe WT (1987) Insertion/Deletion mutations in the Zea
chloroplast genome. Curr Genet 11: 617-624

Gragg H, Harfe BD, Jinks-Robertson S (2002) Base composition of mononucleotide runs
affects DNA polymerase slippage and removal of frameshift intermediates by mismatch
repair in Sachharomyces cerevisiae. Mol Cell Biol 22(24): 8576-8762

Gorman DA, Levine RP (1965) Cytochrome f and plastocyanin: their sequence in the
photosynthetic electron transport chain of Chlamydomonas reinhardtii. Proc Natl Acad
Sci USA 54: 1665-1669

Gumpel NJ, Rochaix JD, Purton S (1994) Studies on homologous recombination in the
green alga Chlamydomonas reinhardtii. Curr Genet 26: 43 8-442

Gur-Arie R, Cohen CJ, Eitan Y, Shelef L, Hallerman EM, Kashi Y (2000) Simple
sequence repeats in Escherichia coli: Abundance, composition, and polymorphism
Genome Res 10: 62-71

139

 

Hancock JM (1996) Simple sequences and the expanding genome. Bioessays 18(5): 421-
425

Hancock JM (2002) Genome size and the accumulation of simple sequence repeats:
implications of new data from genome sequencing projects. Genetica 115: 93-103

Harfe BD, Jinks-Robertson S (2000) Sequence composition and context effects on the
generation and repair of frameshift intermediates in mononucleotide runs in
Sachharomyces cerevisiae. Genetics 156: 571-578

Harfe BD, Jinks-Robertson S (2000) DNA mismatch repair and genetic instability. Annu
Rev Genet 34: 359-399

Harris EH (1989) The Chlamydomonas sourcebook. A comprehensive guide to biology
and laboratory use. Academic Press San Diego

Hiratsuka J, Sharnida H, Whittier R, Ishibashi T, Sakamoto M, Mori M, Knodo C, Honji
Y, Sun CR, Meng BY, Li YQ, Kano A, Nishizawa Y, Hirai A, Shinozaki K, 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

Ishii T, McCouch SR (2000) Microsatellites and microsynteny in the chloroplast
genomes of Oryza and eight other Grarnineae species. Theor Appl Genet 100: 1257-1266

Kelman Z, Yuzhakov A, Andjelkovic J, O’Donnell M (1998) Devoted to the lagging
strand- the x subunit of DNA polymerase HI holoenzyme contacts SSB to promote
processive elongation and sliding clamp assembly. EMBO J 17(8) 2436-2449

Kindle KL, Richards KL, Stern DB (1991) Engineering the chloroplast genome:
Techniques and capabilities for chloroplast trans forrmtion in Chlamydomonas
reinhardtii. Proc Natl Acad Sci USA 88: 1721-1725

Kirchner JM, Tran H, Resnick MA (2000) A DNA polymerase 8 mutant that specifically
causes +1 frameshifi mutations with homonucleotide runs in yeast. Genetics 155: 1623-
1632

140

Kokoska RJ, Stefanovic L, Buermeyer AB, Liskay RM, Petes TD (1999) A mutation of
the yeast gene encoding PCN A destabilized both microsatellite and minisatellite DNA
sequences. Genetics 151: 511-519

Lee JS, Hartford MG, Genova JL, Farber RA (1999) Relative stabilities of dinucleotide
and tetranucleotide repeats in cultured mammalian cells. Hum Mol Genet 8(13): 2567-
2572

Levinson G, Gutman GA (1987) High frequencies of short frameshifts in poly-CA/GT
tandem repeats borne by bacteriophage M13 in Escherichia coli K-12. Nucl Acids RES
15: 5323-5338

Lohman TM, Ferrari ME (1994) Escherichia coli single-stranded DNA-binding protein:
multiple DNA-binding modes and cooperatives. Annu Rev Biochem 63: 527-570

Loridon K, Cournoyer B, Goubely C, Depeiges A, Picard G (1998) Length
polymorphisms and allele structure of trinucleotide microsatellites in natural accessions
of Arabidopsis thaliana. Theor Appl Genet 97: 591 -604

Marra G, Boland CR (1995) Hereditary nonpolyposis colorectal cancer: the syndrome,
the genes, and historical perspectives. J Natl Cancer Inst 87(15): 1114-1125

Maul JE, Lilly JW, Cui L, dePamphilis CW, Miller W, Harris EH, Stern DB (2002) The
Chlamydomonas reinhardtii plastid chromosome: islands of genes in a sea of repeats.
Plant Cell 14: 2659-2679

Metzgar D, Bytof J, Wills C (2000) Selection against fi'ameshifi mutation limits
microsatellite expansion in coding DNA. Genome Res 10: 72-80

Meyer RR, LAine PW (1990) The single-stranded DNA-binding protein of Escherichia
coli. Microbiol Rev 54: 342-380

Olrnstead RG, Palmer JD (1994) Chloroplast DNA systematics: A review of methods and
data analysis. Amer J Bot 81(9): 1205-1224

141

Pinder DJ, Blake CE, Lindsey JC, Leach DRF (1998) Replication strand preference for
deletions associated with DNA palindromes. Mol Microbi0128(4): 719-727

Powell W, Morgante M, Andre C, McNicol JW, Machray JC, Doyle JJ, Tingey SV,
Rafalski JA (1995a) Hypervariable microsatellites provide a general source of
polymorphic DNA markers for the chloroplast genome. Curr Biol 5(9): 1023-1029

Powell W, Morgante M, McDevitt R, Vendramin GG, Rafalski JJ (1995b) Polymorphic
simple sequence repeat regions in chloroplast genomes: Applications to the population
genetics of pines. Proc Natl Acad Sci USA 92: 7759-7763

Provan J, Corbett C, Waugh R, McNicol JW, Morgante M, Powell W (1996) DNA
fingerprints of rice (Oryza sativa) obtained from hypervariable chloroplast simple
sequence repeats. Proc R Sco Lond 263: 1275-1281

Provan J, Corbett G, McNicol JW, Powell W (1997) Chloroplast DNA variability in wild
and cultivated rice (Oryza spp.) revealed polymorphic chloroplast simple sequence
repeats. Genome 40: 104-110

Provan J, Soranzo N, Wilson NJ, Goldstein DB, Powell W (1999) A low mutation rate
for chloroplast microsatellites. Genetics 153: 943-947

Provan J, Powell W, Hollingsworth PM (2001) Chloroplast microsatellites: new tools for
studies in plant ecology and evolution. Trends Ecol Evol 16(3): 142-147

Sager R, Ramanis Z (1967) Biparental inheritance of nonchromosomal genes induced by
ultraviolet radiation. Proc Natl Acad Sci USA 58: 931-935

Saveson CJ, Lovett ST (1997) Enhanced deletion formation by aberrant DNA replication
in Escherichia coli. Genetics 146: 457-470

Sears BB, Chiu WL, Wolfson R (1994) Replication slippage as a molecular mechanism
for evolutionary variation in chloroplast DNA due to deletions and insertions. In:
Tsunewaki K (ed) Plant Genome and Plastome. Kodansha Scientific Ltd Tokyo

142

Sia EA, Butler CA, Dominska M, Greenwell P, Fox TD (2000) Analysis of microsatellite
mutations in the mitochondrial DNA of Sachharomyces cerevisiae. Proc Natl Acad Sci
USA 97(1): 250-255

Tautz D, Trick M, Dover GA (1986) Cryptic simplicity in DNA is a major source of
genetic variation. Nature 322: 652-656

Toth G, Gaspari Z, Jurka J (2000) Microsatellites in different eukaryotic genomes: survey
and analysis. Genome Research 10: 967-981

Twerdi CD, Boyer JC, Farber RA (1999) Relative rates of insertions and deletion
mutations in a microsatellite sequence in cultured cells. Proc Natl Acad Sci USA 96:
2875-2879

Udupa SM, Robertson LD, Weigand F, Baum M, Kahl G (1999) Allelic variation at
(TAA)u microsatellite loci in a world collection of chickpea (Cicer arietinum L)
germplasm. Mol Gen Genet 261: 354-363

Udupa SM, Baum M (2001) High mutation rate and mutational bias at (TAA),
microsatellite loci in chickpea (Cicer arietinum L). Mol Genet Genornics 265: 1097-1103

Vasquez JF, Perez T, Albomoz J, Dominguez A (2000) Estimation of microsatellite
mutation rates in Drosophila melanogaster. Genet Res (Camb) 76: 323-326

Viguera E, Canceill D, Ehhrlich SD (2001a) Replication slippage involves DNA
polymerase pausing and disassociation. EMBO J 20(10): 2587-2595

Viguera E, Canceill D, Ehhrlich SD (2001b) In vitro replication slippage by DNA
polymerases from therrrrophilic organisms. J Mol Biol 312: 323-333

Weber J, Wong C (1993) Mutation of human short tandem repeats. Hum Mol Genet 2:
1 123-1 128

Weetrnan D, Hauser L, Carvalho GR (2002) Reconstruction of microsatellite mutation
history reveals a strong and consistent deletion bias in invasive clonal snails,
Potamopyrgus antipodarum. Genetics 162: 813-822

143

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among
plant mitochondrial, chloroplast and nuclear DNA. Proc Natl Acad Sci USA 84: 9054-
9058

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

Xie Y, Counter C, Alani E (1999) Characterization of the repeat-tract instability and
mutator phenotypes conferred by a T113 insertion in RF C1 , the large subunit of the yeast
clamp loader. Genetics 151: 499-509 E

Yarnada NA, Smith GA, Castro A, Roques CN, Boyer JC, Farber RA (2002) Relative
rates of insertion and deletion mutations in dinucleotide repeats of various lengths in
mismatch repair proficient mouse and mismatch repair deficient human cells. Mutat Res
499: 213-225

 

Zhu Y, Strassman JE, Queller DC (2000) Insertions, substitutions, and the origin of
microsatellites. Genet Res (Camb) 76: 227-236

144

CHAPTER 5

Isolation of chloroplast mutators in Chlamydomonas reinhardtii

145

Abstract

The isolation and characterization of mutator strains in prokaryotes and
eukaryotes have helped to identify essential genes that are involved in mutation
avoidance pathways. Since the genes involved in the maintenance of replication slippage
stability have not been identified in plants, this study was undertaken to isolate mutator
strains that would show an elevated frequency of replication slippage in the chNA of
Chlamydomonas reinhardtii. A plasmid based mutagenesis system was used to isolate
these mutators. To date, we have been able to identify one such mutator strain that will
now be analyzed to identify the gene disruption that has resulted in the mutator
phenotype. Utilizing this assay we could thus isolate a library of mutators, which would

help us identify components of the chNA mutation avoidance pathways.

Introduction

Mutator cells are those which display a higher incidence of spontaneous
mutations than a normal or a wild-type cell. Analysis of mutators has contributed to the
study of mutation avoidance and repair pathways, and they have often helped define new
pathways of mutagenesis (Miller 1998). Since mutator strains have increased levels of
mutation frequencies, it is expected that the mutator phenotypes arise due to defects in
genes whose products act to minimize genetic instability (Kunz et al. 1999). In bacteria
and yeasts, most of the knowledge about DNA repair has been obtained by the isolation

and analyses of the mutator strains. Mutator loci encode components of the mismatch

146

repair system, exonucleolytic proofreading domains of DNA polymerases, components
of oxidative repair pathways, etc (Schaaper and Dunn 1987; Miller 1998; Yang et al.

1999)

Chloroplast DNA replication has been well documented in several plant species
using microscopy, genetic analyses and in vitro experiments (reviewed by Heinhorst and
Cannon 1993). In addition DNA polymerases and their associated 3’to 5’ exonucleases
have also been purified from plants such as soybean and pea (Heinhorst et al. 1990;
Gaikwad et a1. 2002). However to date, none of the genes encoding for the polymerases
or other enzymes required during chloroplast DNA (chNA) replication has been found.
Although genes coding for the repair of the chNA has also not been reported in plants, a
few chNA repair enzymes have been isolated from plants: these include a bacterial
homolog of the RECA protein, which is essential in prokaryotes for both homologous
recombination and recombinational repair (Cerrutti et al. 1992) and certain DNA
photolyases that are important for repair of UV-induced darmges in the chNA (Small
and Greirnann 1987 ; Petersen and Small 2001). Furthermore, the evolutionary
comparisons of different plant species have reported that the chNA has a lower rate of
mutation than the nuclear and mitochondrial genomes (Wolfe et al. 1987), and this

suggests the presence of mutation avoidance pathways in the chloroplast.

The complete sequence data from the chloroplast genomes of several plants have

all demonstrated the lack of any of the genes involved in chNA replication or repair in

the chloroplast genome. Therefore the mutator genes that would induce higher rates of

147

 

mutations for the chloroplast DNA would be expected to be nuclear-encoded. Chloroplast
DNA mutators have been isolated from plants such as barley (albostrians and chloroplast
mutator or cpm), maize (iajap) and Oenothera (plastome mutator). The plastome mutator
of Oenothera has been the best characterized among all these mutators: when
homozygous it causes a 200-1000 fold increase in the spontaneous nmtation frequencies
of the chNA (Epp 1973; Sears and Sokalski 1991). The plastome mutator affected both .-
base substitutions and insertion/deletion events rates, although the latter type of mutations
was observed predominantly in the pm lines (Chiu et al. 1990; Chang et al. 1996). Further

analyses showed that these insertions/duplications occurred primarily at regions of short

 

“H

direct repeats, and hence elevated levels of replication slippage were assumed to be the
mechanism by which these changes occurred in the chNA substrates targeted upon by
the plastome mutator. This identity of the pm-encoded gene product is till unknown, but
based on its mutational spectrum it can be hypothesized that it may code for a gene

involved in mismatch repair or factors that affect the processivity of the chNA

polymerase.

Our objective was to utilize the mutator approach to isolate genes that are
involved in the repair and replication of chloroplast DNA. Initial mutator screens would
be done to look for strains that would show an increase in the frequency of replication
slippage of the chNA, and one of these could include the gene homologous to the
plastome mutator in Oenothera. For this purpose, microsatellites were chosen as the
substrates for the assay that monitor for a mutator activity; however, the naturally

occurring microsatellites could not be used for the assay since a change in repeat length

148

at the site would not result in a detectable phenotype. Therefore, the replication slippage
substrate was created in a reporter construct that would allow the repeat length changes to
be observed at phenotypic and molecular levels (See Chapter 4 for details).
Chlamydomonas reinhardtii was chosen as the experimental organism as it would be
possible to easily create a library of mutators that could define several loci involved in

the control of replication slippage frequencies in the chNA.

Rates of slippage in the chNA of wild—type C. reinhardtii strains occurred at a
range of 1 - 70 per 10‘5 viable cells (Chapter 4). Experiments were conducted to create
nuclear disruptions by random integration of the transforming plasmid pSP124S (Figure
5-1) which contains a chimeric gene composed of the coding sequence of the ble gene
from Streptoalloteichus hindustanus fused to the 5’ and 3’ untranslated regions of the C.
reinhardtii nuclear gene RBCSZ (Stevens et al. 1996). The ble gene confers resistance to
antibiotics such as phleomycin or related antrbiotics such as zeocin, and transformants
were screened by growing the cells on zeocin. Successful transformants were then
screened to identify lines that would show higher frequencies of replication slippage over
the wild-type. Since a true mutator will be defined by its ability to induce similar elevated
levels of slippage mutations at newly introduced chNA microsatellites, the “mutator”
phenotype of the putative strains were rechecked by introducing a fresh chNA slippage

substrate into their backgrounds. Once true mutators are identified, further

149

 

 

ble
S’RBCSZ I

3’RBCS2

pBlueScript

 

SacI Kpnl

Figure 5—1. Diagrammatic representation of the pSP124S plasmid used for the
nuclear transformation of the Chlamydomonas cw OF cells. The pSP124S plasmid
contains the bacterial gene, “ble”, which confers resistance to cells against the
antibiotics zeocin or phleomycin. The transforming plasmid also has the regulatory
regions (5’ and 3’ RBCSZ) from a nuclear gene that enables the expression of the
bacterial ble gene in Chlamydomonas cells. The whole size of the plasmid is ~ 4.3 kb
(insert is 1.3 kb and pBluescript is 3kb).

150

 

characterizations of the disrupted nuclear gene that led to the mutator phenotype would
allow us to identify nuclear genes involved in the maintenance of stability at the repeated

sequences of the chloroplast DNA in Chlamydomonas.

Materials and Methods

Construction of Chlamydomonas strains, growth and crosses

A Chlamydomonas cell line (OF, mt+) containing a reporter construct with slippage
substrate (used in assays for spontaneous replication slippage rates in Chapter 4) was
crossed with cc849, a cell wall-less strain, mt- to produce a strain that would contain the
chNA mutator target in a cell wall less background. The rrrt- parent, cc849 was obtained
from the Chlamydomonas culture collection at Duke University. The OF strain and c0849
were maintained on media as described in Chapters 3 and 4. The nuclear transformants
were selected on TAP media supplemented with agar and supplemented with IOug/ml
Zeocin(1nvitrogen, USA) and kept under low light conditions. The Chlamydomonas
strains were crossed using the protocol described in Chapter 4. The progeny from the

cross OF X cc849 were grown under low lights on TAP media

Selection for cell wall less phenotype in the meiotic progeny
The meiotic progeny that had the slippage substrate in the chNA and a cell wall less
phenotype was selected as the recipient strain for nuclear transformations. To select for

the cell wall less phenotype a modified protocol adapted from the Chlamydomonas

151

sourcebook (Harris 1984) was followed The cells were scraped off from the plate,
dissolved in one ml of TAP in an eppendorf tube, and then centrifuged at 13K rpm for 5
mins. The supernatant was discarded and cells resuspended in an appropriate volume of
TAP (depending on the size of the pellet) to obtain a concentrated cell suspension. For
each sample, equal volumes of the cell suspensions were added to two eppendorf tubes:
one tube contained one ml of sterile water, while the other tube contained a ml of
Triton/EDT A solution (Harris 1988). The cells were mixed well, and then cell counts
done using the hemocytometer as described in the Chlamydomonas sourcebook (Harris
1988). The efficiency of cell lysis was expressed as:

Marge cell count in water — me cell count in Triton/EDTA)
average cell count in water

Nuclear transformation of C. reinhardtii

The chF strain constructed was used as the recipient strain and 1 pg of digested or
undigested pSP124S plasmids were used for transformation. The nuclear transformation
was carried out using the glass bead method as described in Kindle (1990) and Stevens et
al. (1996), with a few modifications. Sheared salmon sperm DNA (long/ml) was used as
carrier DNA and the transformants were plated with corn starch for selection, on TAP

plates supplemented with Zeocin (Invitrogen, USA).

PCR amplification

PCR amplification was done using the primers rbcL for2 and rbcL rev using the same

conditions as described in Chapter 4.

152

 

 

Assay for isolating putative mutators among the Zeocin-resistant nuclear transformants
The transformants were replica plated on TAP media supplemented with Zeocin and kept
under low lights to check their viability and growth patterns on non-selective media; and
also on selective media containing HS (no acetate) and kept under high lights to select for
lines that grow many colonies (compared to the original OF strain that has no nuclear

disruptions and grown under similar conditions).

Restesting of the putative mutators to confirm their mutator phenotypes

Crosses were done with the putative mutator strains (photosynthetically competent) with
the original OF strain (photosynthetically incompetent) to reintroduce a new slippage
substrate into the nuclear disrupted background of the putative mutators. Crosses were
done according to the method described in Chapter 4. The meiotic progeny were
dissected and then each grown on three different growth conditions: TAP plates kept
under low lights, representing the non-selective condition: TAP plates supplemented with
Zeocin kept under low lights to identify the progeny that contains the nuclear ble gene;
HS plates kept under high lights to identify the progeny in which replication slippage has
occurred in high frequencies which could be phenotypically observed as green patches

(Figure 5-6).

153

Results

Construction of strains for nuclear transformation
The nuclear transformations were done using the glass bead method of
transformation as described by (Kindle 1990; Stevens et al. 1996) and for this purpose, an

appropriate cell wall-less recipient strain containing the slippage substrate in its chNA

 

was constructed by crosses (see Materials and Methods for details). The recipient strain, '."
cw OF, was selected by checking the meiotic progeny fi‘om the crosses using PCR

amplification with primers flanking the region of the rbcL gene containing the

microsatellites. Figure 5-2 shows that the cw OF cells contains the 28-bp insertion of ‘-

GAAA-repeats in its chloroplast encoded rbcL gene. The cell wall less phenotype was
also checked for in the same progeny cw OF, using the protocol described in Materials
and Methods. Nuclear transformations were then performed with this strain, using both
digested and undigested DNAs of pSP124S (Figure 5-3) and also in the presence and
absence of a carrier DNA (sheared salmon sperm DNA) to establish optimal

transformation conditions.

Conditions for nuclear transformation

cw OFwas used for nuclear transformation using the glass bead method (Kindle
1990) with pSP124S as the transforming plasmid. The transformation efficiencies
compared among the different transforming parameters used. The glass bead method for
nuclear transformation is explained in details in the Materials and Methods section.
Successfirl nuclear transformants were selected by plating them on TAP media containing

Zeocin(10ug/ml). Transformants were obtained from each type of treatment and the

154

CWOF CWOF
1 2 WT pOF7 3 4

 

Figure 5-2. Agarose gel picture showing the PCR anplified products
from the wild type (WT) , plasmid-pOF7, and the cell wall-less out-of-
frane cells (CW OF). The PCR amplified products in lanes numbered 1
and 2 were from the cw OF cells that were used as the starter culture.
The PCR products in lanes 3 and 4 were fiom the DNAs tlnt were
obtained from the final culture of cw OF cells used in the

transformation experiment

155

 

 

Dig. witthnI
1 2 3 Undig

MW

 

Figure 5—3. Agarose gel
showing the pSPlMS DNAs
used for nuclear
tnnsfornntion of the cw OF
cells. Tle Kpnl digested
DNAs are labelled froml
tlnrugh 3, and the undigested
DNA hne is labebd as
Undig. MW denotes a lane
with the rmbcular weight
standards (1 Kb DNA hdder
fiom Giico BRL).

 

 

156

results graphed in Figure 5-4. Results showed that both undigested and digested plasmid
DNA were able to integrate into the nuclear genome of the cw OF cells, but two of the
digested DNA samples (Dig-1 and Dig-2, Figure 5-4), produced Zeocin-resistant
transformants at higher frequencies than did the undigested DNAs. The presence of the
carrier DNA did not affect transformation frequencies when undigested DNAs were used,
but did produce higher efficiency when the Kpnl digested psP124S was used as the

transforming plasmid (Figure 5—4).

Assay for isolating chloroplast DNA mutators from the Zeocin-resistant transformants
The Zeocin-resistant transformants had the randomly integrated pSP124S plasmid
in their respective nuclear genomes as well as the slippage substrate in their chNA. The
presence of the 28-bp insertion in the rbcL gene of the transformants caused
them to be non-photosynthetic, as a result of which they were unable to live on media
without acetate (HS). As outlined in the overview of the experimental plan in Figure 5-5,
our next step was to identify the mutator lines that show increased numbers of cells with
restored photosynthetic competence. For this an assay was performed to check for rates
of restoration to photosynthetic competence in the different Zeocin-resistant
transformants. Out of the 1534 original transformints, only 617 were recovered when
they were replated on fresh TAP media supplemented with Zeocin, for maintenance. As
shown in Table 5-1, 44 among them displayed a phenotype in which the rates of
photosynethetic competence were higher than the original chF strain without any

nuclear disruption.

157

 

 

 

35

 

 

 

 

Efficiemy of .
transformation 25 i

108 F7 -
(per cells) 20 :

 

 

 

 

 

 

 

15
10- fl 5»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 5-4. Graphical representation of the comparisons of efficiency of

nmlear trans formation in cw OF cells under different conditions.

Dig. DNA refers to digested DNA with the nurrbers representing two different
nbes from which the respective DNAs were used; undigested DNA is ind'cated
by Undig. DNA on the graph, and CD refers to carrier DNA (sheared sahnon
sperm DNA). Mixed Dig. DNA refers to the condition when the digested DNAs
from the tubes were contined together.

158

 

 

# of initial # of # of “putative # of putative # showing a
Zeocin- transformants mutator” lines mutator lines “true
resistant that were obtained tested by crosses mutator”

transformants replica plated phenotype
1534 617 44 9 l

 

 

 

 

 

 

 

Table 5-1. Nunber of “true mutators” obtained flom screening the initial Zeocin-resistant nuclear
transforrmnts. Only a subset of the original transformants (non-photosynthetic) was used for
replica plating on the non-selective and selective rmdia (for photosynthetic competence). Among
the 617 colonies tested, 44 showed an elevated reversion flequency to photosynthetic
competence. Crosses were done to introduce a flesh chNA slippage substrate into the nuclear
background of nine of the putative rmrtators. One of the re-tested putative rrmtators showed a
“true” mutator phenotype.

159

 

Selection of true mutator lines

These 44 putative mutator strains may show a higher level of replication slippage
because of an earlier random event, or due to a nuclear disruption in an essential gene
that plays a role in the maintenance of the repeat tract stability of the chNA. In order to
test whether the nuclear gene disruption has caused the elevated level of replication
slippage in the chNA of a putative mutator line, a new chNA microsatellite substrate
was crossed into the nuclear backdrop of each of the putative mutators. To date, only nine

out of the 44 putative mutator lines have been crossed (Table 5—1).

In a sexual cross of Chlamydomonas, the meiotic progeny all contain the chNA
of the mating type + parent (Boynton et al. 1987) while the nuclear markers are inherited
in a Mendelian ratio. Since the line in which the putative mutators were isolated is a
mating type - parent, the new slippage substrate in the chNA is contributed by the
mating type + parent. Hence, almost all the progeny flom the cross would be
photosynthetically incompetent and their slippage rates similar to the original OF strain.
In contrast, the ble gene is a nuclear marker and would be present in only half of the
meiotic progeny. If the putative mutator lines were true mutators due to a disruption
caused by the pSP124S plasmid, then the presence of the Zeocin marker should increase
the slippage flequency, resulting in more cells that are photosynthetic. This condition is
illustrated in Figure 5-6, which shows one set of the four meiotic progenies labeled as 11-
2 through 11-4 flom a zygote. As summarized in Table 5-2, the progeny numbers 11-1

and 11-2 were resistant to Zeocin, and also were the only two which showed higher

160

 

 

 

Grow nuclear transformants (cw OF +pSP124S
on non-selective plates

Replica plate the cells
On nonselective media On selective media to test for for photosynthetic
for checking viability competence. Select for cell lines in which slippage

occurred at an elevated frequency and these will be
referred to as “putative mutator”

Retest “putative” mutator lines for mutator phenotype by re
introducing a new slippage substrate through crosses

 

Figure 5-5. Assay for isolating the chNA mutator strain

161

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tetrad TAP +Low TAP + Zeocin + HS + High Mating
progeny Light Low Light Light Type
(non-selective) (selective for the (selective for

nuclear marker, photosynthetic (nuclear

ble-gene) competence) marker)
11-1 Y Y Y +
11-2 Y Y Y +
11-3 Y N N -
11-4 Y N N -
14-1 Y Y Y -
14-2 Y N N +
14-3 Y Y Y -

 

Table 5-2. Phenotypes of two tetrad progenies flom a cross between pm! (with the putative

mutator phenotype, mt-) and an OF line (with the chNA slippage substrate, rm+). The

chloroplast DNA with the slippage substrate flom the mt+ parent (OF) is transmitted to all the

tetrad progeny flom the cross, whereas the nuclear rmrkers are transmitted in the regular

Mendelian ratio (2:2). Phenotypes of two sets of tetrad progenies flom the cross are shown here.

Y refers to growth on the medium, whereas N indicates that the cells died on this medium

162

 

 

 

 

 

 

 

 

 

 

 

Non- Zeocln Photo-
selective Resistant? synthetic?
Y Y Y
Y
N
N

 

 

 

 

A

 

B

 

11-1

11-2

11-3

11-4

C

Figure 5-6. Growth oftetrad progniss 11-1 to 11-4 obtained from the cross PMl X OF, unier
different growing contitbrs. Panel A shows the growth of the pmpnies on TAP (+ low light), a
non-selective condit'nn; panel B slnws the same eobnies plated on TAP + Zeocin (+bw light),
that Will only allow the gowth of Chlamydomonas cells with the ble gene, ani panel C shows the
00an on HS rmdia (kept unier high light), a coalition to sebct brplntosyntlttic

competence.

163

numbers of cells with restored photosynthetic ability, resulting in a green patch of

growth.

Discussion

Conditions were established for the successful transformation of the
Chlamydomonas strain, chF, which contained an out-of-flame slippage substrate in its
rbcL gene and a cell wall-less phenotype. Nuclear transformants were obtained flom all
the different transformation conditions used, including both digested and undigested
DNA (Figure 5-4). However, the digested DNA samples consistently produced a higher
number of transformants (per pg of DNA used), and therefore it seems that it may be
more optimal for integration into the nuclear genome. In contrast, results flom the nuclear
transformation experiment do not demonstrate a requirement for the carrier DNA in
increasing the transforrmtion flequencies using the glass bead method when undigested
DNA was used as the transforming plasmid. Form this experiment, it is seen that the most
optimal condition for producing the highest flequencies of nuclear transformation is
when the single digested plasmid DNA is used for transformation along with the carrier
DNA. Future experiments need to be done in order to firmly establish the optimal
conditions for glass bead transformantion for mu strain, as our conclusions are based on a

single experiment.

Out of the 600 Zeocin-resistant transformants that were selected flom the above

glass bead experiment, 44 had shown a “putative” mutator phenotype (Table 5-1).

164

 

However, when these were retested for verification of their mutator, only one (pm I ) out
of the nine tested thus far, had a true mutator phenotype. During the growing of the chF
cells for nuclear transforrmtion, few cells may have undergone slippage with a reversion
to photosynthetic competence. These could be mistaken as putative mutator lines if they
also were also among the successful nuclear flansfornmnts, and this could be a reason
why our assay shows a high number of false positives (8 out of nine putative mutators IF,
retested). The true mutator nature was established in the me strain by re-introducing a
new chNA slippage substrate against the nuclear background of the putative mutator,

which would correlate the gene disruption in the putative mutator strain with the

 

“a.

increased slippage flequency phenotype of the chNA. This true mutator line (me) is
now being analyzed to isolate and identify the disrupted nuclear gene using the plasmid
rescue method. This would hopefully lead to the identification of the first gene involved

in the maintenance of stability of chNA microsatellites.

165

REFERENCES

Boynton JE, Harris EH, Burkhart BD, Iamerson PM, Gillham NW (1987) Transmission
of mitochondrial and chloroplast genomes in crosses of Chlamydomonas. Proc Natl Acad
Sci U S A 84(8): 2391-2395

Cannon GC, Heinhorst S (1990) Partial purification and characterization of a DNA
helicase flom chloroplasts of Glycine max. Plant Mol Biol 3: 457-464

Cerutti H, Osman M, Grandoni P, Jagendorf AT (1992) A homolog of Escherichia coli r
RecA protein in plastids of higher plants. Proc Natl Acad Sci U S A 89(17): 8068-8072

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

 

*1!

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

Epp MD (1973) Nuclear gene-induced plastome mutations in Oenothera hookeri I .
Genetic analysis. Genetics 75: 465-483

Gaikwad A, Hop DV, Mukherjee SK (2002) A 70-kDa chloroplast DNA polymerase
flom pea (Pisum sativum) that shows high processivity and displays moderate fidelity.
Mol Genet Genomics 267(1):45-56

Kindle KL (1990) high flequency nuclear transformation of Chlamydomonas reinhardtii.
Proc Natl Acad Sci USA 87: 1228-1232

Kunz BA, Ramachandran K, Vonarx EJ (1998) DNA sequence analysis of spontaneous
mutagenesis in Saccharomyces cerevisiae. Genetics 148: 1491-1505

Miller JH (1998) Mutators in Escherichia coli. Mutat Res 409: 99-106

166

Petersen JL, Small GD (2001) A gene required for the novel activation of a class II DNA
photolyase in Chlamydomonas. Nucl Acids Res 29(21): 4472-4481

Schaaper RM, Dunn RL (1987) Spectra of spontaneous mutations in Escherichia coli
strains defective in mismatch correction: the nature of in vivo DNA replication errors.
Proc Natl Acad Sci U s A. 84(17): 6220-6224

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

Small GD, Greirnann CS (1977) Photoreactivation and dark repair of ultraviolet light-
induced pyrimidine dimers in chloroplast DNA. Nucleic Acids Res 4 (8): 2893-2902

Stevens DR, Rochaix J-D, Purton S (1996) The bacterial phleomycin gene ble as a
dominant selectable marker in Chlamydomonas. Mol Gen Genet 251: 23-30

Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among
plant mitochondrial, chloroplast and nuclear DNA. Proc Natl Acad Sci USA 84: 9054-
9058

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

Yang Y, Karthikeyan R, Mack SE, Vonarx EJ, Kunz BA (1999) Analysis of yeast pmsI ,
mshZ, and mlhl mutators points to differences in mismatch correction efficiencies
between prokaryotic and eukaryotic cells. Mol Gen Genet 261(4-5): 777-787

167

 

CHAPTER 6

Conclusions

168

Base substitutions and insertion/ deletion events are minimized in organisms
through error-flee DNA replication, successful inactivation of DNA damaging agents, as
well as by the recognition and repair of the DNA damage after it has occurred Mutation
rates of organisms are usually considered as indicators about whether a high degree of
fidelity exists during replication, and also as a testament to the efficiency of the DNA

repair mchinery. Rates of base substitutions and insertion/deletion events have been

 

ay—
measured in organisms through experimental investigations as well as by evolutionary

studies. Spontaneous mutation rates that have been obtained through experimental

evaluations provide a baseline for selecting mutator strains. The isolation and

characterization of mutator strains in prokaryotes and eukaryotes have helped to identify J

essential genes that are involved in their mutation avoidance pathways.

For the chloroplast DNA, investigations to examine the rates of base substitutions
and replication slippage tlu'ough experimental analyses are lacking. Current estimates of
chNA mutation rates are based solely on phylogenetic comparisons: these studies have
indicated that the chNA mutation rates are slower than their nuclear counterpart, with
synonymous base substitutions occurring at a range of 1.1 to 3 X 10'9 synonymous base
substitutions per site per year (Wolfe et al. 1987). This low rate of mutation has been
hypothesized to be due to efficient replication and repair systems of the chNA. In order
to have confidence in the existence of a slow molecular clock for the chNA sequences,

the true rates of mutations occurring in vivo need to be examined experimentally.

169

Our investigation provided the first experimental assessment of base substitution
mutations occurring at a target site in the chloroplast DNA of the green alga,
Chlamydomonas reinhardtii. Our data showed that these mutations occurred in the
chloroplast DNA at a range of 3 in 1011 to 1 in 108 viable cells plated. Analyses of the
different types of base changes allowed us to conclude that transversions occurred about
four times more flequently in the chNA than transitions. In addition, the spectrum of
mutations that occur spontaneously in the chNA provided information regarding the
existence of different repair pathways that act to minimize their forrmtion or are involved

in their repair.

Rates of insertion/ deletion events at microsatellites were also assayed for a
particular region in the chNA of C. reinhardtii. Replication slippage has been identified
as the mechanism by which such additions or deletions occur at the repeated sequences.
Compared to the base substitution events, replication slippage at a stretch of GAAA
repeats occurred 100 —10,000 times more flequently, at a rate of 1 to 70 per 106 cells
plated. Furthermore, the chNA showed a strong bias towards deletion events, indicating
that the template strand is prone to slip more during chNA replication, at least at this
particular site. In contrast, similar substrates carried on plasmids in E. coli showed a
higher flequency of slippage, and insertions of repeat units occurred predominantly
(Appendix A). This result is particularly interesting as the chNA is known to have a
cyanobacterial ancestry and hence it was expected that its replication pattern would
mirror that of prokaryotes. Experiments are now being conducted in our laboratory to

determine the flequency and type of replication slippage in the cyanobacterium

170

Synechocystis. If the chNA slippage event resembles that of the cyanobacteria, it would
suggest that the mutational machinery that maintains stability at the chNA repeats have

remnants of its prokaryotic ancestry.

Mutation rates are dependent on the types of sequences assayed for, as well as on
their respective positions in the genome and may also vary depending on the flanking a.
sequences. Future experiments need to be conducted to introduce different substrates for

both slippage and nucleotide substitutions in order to get a true assessment of the range of

 

mutations in the chNA. In addition, the presence of copy correction in the chNA, - '-
especially between sequences in the inverted repeat region, should be considered while

comparing the rates between different target loci assayed for mutations.

Since the genes involved in the maintenance of replication slippage stability or
those that repair the mismatches due to slippage have not been identified in plants, we
decided to isolate mutator strains that would show an elevated flequency of replication
slippage in the chNA of Chlamydomonas reinhardtii. The rationale for this approach
was that such a mutator phenotype would be caused due to deletion in an essential gene
that is otherwise involved in the maintenance of the repeat lengths in wild-type cells. A
plasmid-based mutagenesis system was used for gene disruptions. To date, we have been
able to identify one putative mutator strain in which a nuclear gene disruption has
occurred that appears to control the rates of slippage in the chNA containing the
microsatellites. The mutator strain is currently being analyzed to identify the gene

disruption that has resulted in the mutator phenotype. Utilizing this assay it should be

171

possrble to isolate a library of mutators, which would help us identify components
involved in the maintenance of stability of chNA microsatellites. In this process we
could unravel the different components of the replication machinery that act to prevent
slippage at repeats, or identify genes involved in the mismatch repair system of the
chNA of plants. Since none of these genes have yet been identified or isolated in plants,

this library should present us with a unique opportunity for gene discovery.

172

 

APPENDIX

Examining the patterns and rates of replication slippage in

plasmids of Escherichia coli

173

Introduction

Insertion/ deletion events that occur mostly in regions of DNA containing short
direct repeats (microsatellites) are major sources of mutations. Microsatellites are
distributed throughout the genomes of both prokaryotes and eukaryotes (Metzgar et al.
2000; Gut-Arie et al. 2000). Eukaryotic microsatellites have been shown to have a high _
rate of mutation, as a result of which they have become a primary source of polymorphic

markers in studies of population genetics (Metzgar et al. 2000). The addition or deletion

 

of repeat units occurs mostly by a mechanism known as replication slippage (Levinson i
and Gutrmn, 1987). In bacteria such as Escherichia coli, the study of repetitive DNA
sequence instability has provided insights into the molecular mechanisms by which cells

avoid these insertion/deletion events at microsatellites.

My experimental assay of replication slippage of a chNA microsatellite region
(Chapter 4 of thesis) has shown that the chNA microsatellites undergo mutations much
more flequently than do base substitutions (100-10,000 times higher). In addition, the
chNA microsatellites have a deletion bias, as almost all the “slipped” colonies
contained a deletion of one or more repeat units. Therefore, since the same replication
slippage substrate used in the chNA studies was also available to us on a plasmid
contained in E. coli (GM2163), we decided to test for the patterns of slippage events
occurring in the bacterial cells. Since chloroplasts have been proposed to be descendants
of cyanobacterial endosymbionts (Gray 1992), it could be possible that the rates and

patterns seen in our study with the chNA (Chapter 4) would be similar to those in E.

174

coli. However, since this gene has no phenotypic selection in bacteria, estimation of

slippage rates would be difficult.

In order to assess rate of slippage in bacteria, the first goal was to create a reporter
construct to monitor the rates and patterns of replication slippage in bacteria. The aadA
gene is a eubacterial gene that confers resistance to the antibiotic, spectinomycin, in
bacterial cells when present (Shaw et al. 1993). The aadA gene was chosen and a stretch
of GAAA repeats was inserted to create an out-of-flame insertion. Cells containing this
out-of-flame construct would thus be sensitive to spectinomycin, and therefore die on LB
media containing the antibiotic. Restoration of resistance to the antibiotic will be a result
of changes at the repeats, and by counting the number of cells with restored resistance,
the rate of slippage in bacteria can be assessed. PCR amplification of the slipped colonies
will then allow us to observe the different types of slippage events that would occur in
bacteria Since the same slippage substrate will be put in two different genes on the
plasmid (aadA and rbcL), it could be deduced if genome position affects the rates and
patterns of slippage events. Fmthermore, a different type of slippage substrate (a l4-bp
duplication) was introduced into the aadA gene at a different position to determine
whether gene context and slippage substrate play a role in the replication slippage pattern

and rates.

175

Materiab and Methods

Construction of plasmid strains for the replication slippage assay

One of the plasmids pOF7 used for the assay had been used previously, and details are
outlined in Chapter 4. The aadA-OF plasmid was made by creating appropriate primers
that would contain the microsatellite of interest along with anchor sequences on either
direction that would anneal to the wild type aadA sequences carried in a pUC 8
expression vector (p699, Sears lab). The out of flame forward primer used for PCR
amplification had the following sequence: 5’
AGGTATGCGTCCATGGAAAGAAAGAAAGAAAGAAAGAAAGAAAG
AAAGCTCGTGAAGCGGT 3’. The reverse primer used for PCR amplification was
aadA R, which had the following sequence: 5’ AAT CT C GCT CTC TCC AGG 3.
The PCR product had Bcl I and Nco I endonuclease cut sites on either side to facilitate

directional cloning into the vector.

To create a different type of replication slippage substrate, a 14-bp duplication was
generated in the aadA gene at the Bcl I cut site using restriction digestion and cloning
techniques. The plasmid p699 containing the wild type rbcL gene and the aadA gene was
used to create the new aadA -Dup plasmid using the Oligonucleotides Bcl I F and Bcl I R
to create a l4-bp duplication at the Bcl I cut site of the aadA gene that resulted in a
disruption of its reading flame.

Bcl I F: GAT CAA CGA CCT TT

Bell R: GAT CAA AGG TCG TT

176

Replication slippage assays

The pOF7 colonies were grown under no selection in liquid media and then plated on
non-selective media (LB-agar with ampicillin). Several assays were undertaken to
establish the rates of replication slippage in the aadA -OF cells. For one assay, the aadA-
OF cells were grown in LB media overnight and then planed on media containing
spectinomycin to select for revertants with resistance to the antibiotic. For viable cell
counts, cells were diluted and plated on media without selection (LB agar with
ampicillin). The replication slippage rates were determined by comparing the numbers of
colonies that grew on spectinomycin with the numbers of colonies on LB and arrrpicillin.
A second type of assay was done which required plating the aadA -OF cells on a non-
selective media at several dilutions as well as time points in order to quantify colony
forming units. For this purpose, the aadA-OF cells were grown on LB media to a certain
cell density such that the viable cells could be counted at dilution of 104, 10‘5 and 10“.
Aliquots of the culture were plated at the above three dilutions on multiple plates of non-
selective media, and the plates were placed in a 37 Cincubator. A subset of the plates
representing each dilution was removed at l-horn‘ intervals, and then a layer of top agar
containing spectinomycin or ampicillin was poured over the non-selective media The
number of colonies that grew on the LB+ arnpicillin media represented the numbers of
viable cells plated at that particular time point, and the number of colonies that grew on
the LB + spectirrorrrycin plates represented the numbers of colonies that underwent

slippage to restore a spectinomcyin-resistance phenotype.

177

 

 

 

PCR amplification

The slipped colonies using the pOF7 plasmid were checked by PCR amplification using
the primers rbcL for2 and rbcL rev and the PCR program described in Chapter 4. The
slipped colonies in the aadA-OF cell line were checked using psbA F (5’-
AATGTGCTAGGTAACTAA-3 ’) and aadA R (5’-AATCTCGCTCTCTCCAGG—3 ’)

using the same PCR program as in Chapter 4.

Results

Replication slippage in the GM 2163 cells containing the pOF 7 plasmid

The pOF7 plasmid contains the GAAA stretch of repeats in the rbcL gene of
Chlamydomonas reinhardtii, which has no phenotype in bacterial cells. In order to
evaluate replication slippage events at this site, the bacteria were grown and plated under
non-selective conditions, allowing for mutations to accumulate in the process. Single
colonies were then randomly selected (see Materials and Methods for details) and
checked using PCR amplification to assess the changes that had occurred at the GAAA-
stretch within the rbcL gene carried on a plasmid (pUC8 derivative). Figure A-l , shows
an agarose gel picture with the PCR amplified products containing the microsatellite
region of the rbcL gene flom the randomly selected colonies. Among the 22 colonies
flom which DNA was amplified, 10 colonies showed higher sized products than that
were present in the original strain. Thus, it appears that replication slippage occurs

flequently in bacteria, and that the repeat units show a

178

PCR amplified products from colonies grown without any selection MW

 

123M)

 

Figure A—l. Insertion events seen in a plasmid containing a stretch of GAAA repeats contained in
E. coli. Agarose gel showing the PCR anplified products from random colonies of bacterial cells
containing the pOF7 plasmid. MW denotes the lane that contains the molecular weight standard

(123-prNA ladder, Gibco BRL).

179

bias towards amplification. Estimation of direct rates at this microsatellite region in pOF7

was not possible due to the lack of any phenotypic selection.

Selection of a site within the aadA gene to insert a slippage substrate that could be
assessed for frequencies in bacterial cells

In order to assess the rates of replication slippage in bacterial cells, the aadA gene
that confers resistance to spectinomycin was chosen for the insertion of slippage
substrates. The principle was to insert an out-of-flarne slippage substrate into the aadA
gene, which will result in cells that are spectinomycin-sensitive, and a slippage event
would subsequently restore the reading flame resulting in the restoration of
spectinomycin resistance. At first the Bcl I cut site at the 66m codon position of the aadA
gene was chosen for inserting two different types of slippage substrates, 3. GAAA stretch
of repeats and a 14-bp duplication, to assess for replication slippage flequencies in E.
coli. My results showed that this region of the aadA gene is not amenable for accepting
additional codons and hence both these substrates could not be inserted at that position of
the aadA gene for assessing their slippage flequencies in E coli. A different region (the
second codon position) of the aadA gene was then chosen for the insertion of the GAAA
stretch of repeats to assess the rates of replication slippage events. This region as shown
in the results below, proved to be flexible in accepting additional codons generated by the
insertion of an in-flarne slippage substrate, which still retained the spectinomycin

resistance.

180

Estimation of rates and patterns in the aadA gene containing the GAAA microsatellites

The aadA -OF cells containing the out-of-flarne insertion were assayed for rates
and patterns of replication slippage that would occur in the microsatellites (for details, see
Materials and Methods section). Figure A-2 shows PCR products flom a subset of the
slipped colonies, which had a spectinomycin-resistant phenotype. All of the slipped
colonies (12 were analyzed) were heteroplasmic, containing the original sized repeats as
well as a larger band representing an insertion event. Therefore, the GAAA repeats in the
aadA gene also preferentially undergo expansions as none of the randomly chosen

colonies showed any smaller sized product.

Results flom the assays to estimate the rates of replication slippage by comparing
the numbers of viable cells plated to the numbers of cells with restoration of the
antibiotic-resistance phenotype obtained flom several trials failed to establish a consistent
rate for the replication slippage events. From the first type of assay for rates, 1 in 100
cells appeared to have undergone slippage (trial 1), whereas in the second trial, 1 in 10
cells showed slippage (data not shown). Form the overlay experiments to assay for rates
of slippage; nearly every colony produced mutant cells showing spectinomycin
resistance. This could be due to the very high mutation rates, along with the high copy
number of plasmids and their short generation time within a bacterial cell as a result of
which, the exact replication slippage rates could not be assessed. However, these data
leave no doubt that replication slippage in the bacterial plasmids show a very high

flequency of replication slippage.

181

 

Slipped Slipped

 

Figure A-2. Replication slippage at GAAA microsatellites in the 2nd codon position of the aadA
gene. Agarose gel showing the PCR arrplified products from the slipped lines that are resistant to
the antibiotic; IF and OF are the original strains containing an in—flame or an out—of-flame
insertion of GAAA repeats, respectively; along with the wild—type strain without any GAAA

repeats (Wt).

182

Discussion

I observed that replication slippage occurred in plasmids being replicated within
bacteria, with and without the presence of any selection pressure: GAAA repeats present
in the rbcL gene in pOF7 and in the aadA gene in aadA -OF both showed replication
slippage. In both genes the GAAA microsatellites showed only expansions of the repeat
units and among all the samples tested by PCR, none showed any deletion events. This
aspect of replication slippage in bacteria is in contrast to the scenario seen for the same
type of repeat in the chNA of Chlamydomonas reinhardtii (Chapter 4), which only
showed deletions of the repeat units. Therefore, it appears that according to the model
proposed by Levinson and Gutman (1987), in E. coli plasmids, during replication at a
GAAA repeat region, the daughter strand preferentially undergoes slippage, as a result of
which duplications of the repeat units occur. In contrast in chloroplasts, the template
strand of the replicating chNA molecule must be more prone to slippage. Experiments
are now being conducted in our laboratory to introduce similar slippage substrates into a
cyanobacterial genome in order to observe the types of slippage events for comparisons

with the chNA.

183

REFERENCES
Gur-Arie R, Cohen CJ, Eitan Y, Shelef L, Hallerman EM, Kashi Y (2000) Simple

sequence repeats in Escherichia coli: Abundance, composition, and polymorphism
Genome Res 10: 62-71

Gray MW (1992) The endosymbiont hypothesis revisted. Intnl Rev Cytol 141: 23-357

Levinson G, Gutman GA (1987) Slipped strand mispairing: A major mechanism for
DNA sequence evolution. Mol Biol Evol 4(3): 203-221

Metzgar D, Bytof J, Wills C (2000) Selection against flameshifl mutation limits
microsatellite expansion in coding DNA. Genome Res 10: 72-80

Shaw KJ, Rather PN, Hare RS, Miller GH (1993) Molecular genetics of arrrinoglycoside
resistance genes and familial relationships of the aminoglycosied—modifying enzymes.
Microbiol Reviews 57(1): 138-163

184

lllgl‘jlljglgljlllfllilrll