!"!"#"
$%!&$"%$%
'
(
)
((''(*+!!!&#%
!
"#
$$%
&'
(#'
'
)((*$
'!+,-(
.
/#(,
012,%,'
( 3,4
!('!
,
5
,
&,6,,72 8&9:&;96:9:,
((#(
(##.
(
#
<'#
(
#,".
'
.(
(
(''
(
(,
777.
#
(.(
''
(
,
('
.''
(
#
#'
(
#.
(
'
(
,=
##
(
(((
#.
(
(
#
'>'
(
(
'
,
7 7 . # (. ,, #
,##(.(.
(
#
(
,$
.
(#
(
#
(
'
'
(#(45
,
7777.'
((.#
'
(.
(
&?'
(
>
85'
('
,@
(
(
'
.#
'##
85,
7 7A . #
(. ( 85 ) 7A 42 ''
'
''
,'
(B5#
(
.'
'
85,42
#.
(
(''
,-(
.1''
#.
(
'
42,
!
#
##85
(#
85
"#
$%
#$&'()#
#*+',)-
# C201%
722866%6
72 8&9:&;96:9:
;%:*(
DD,1,DEF;%:+
I think emotions should rule us.
I have good reasons why.
Logic, you see can be faulty.
Emotions never lie.
Reality brings us doubts.
We know an emotion’s real.
But I want logic to rule us.
It’s just the way I feel.
List of Papers
This thesis is based on the following papers, which are referred to in the text
by their Roman numerals.
I
II
III
IV
Nilsson, A.I., Koskiniemi, S., Eriksson, S., Kugelberg, E., Hinton, J.C., Andersson, D.I. (2005) Bacterial genome size reduction by experimental evolution. Proc. Natl. Acad. Sci.,
102(34):12112–6
Koskiniemi, S., Andersson, D.I., (2009) Variation in spontaneous deletion rates at different locations of the Salmonella typhimurium chromosome. (Manuscript)
Koskiniemi, S., Andersson, D.I., (2009) Translesion DNA polymerases are required for spontaneous deletion formation in
Salmonella typhimurium.
Proc. Natl. Acad. Sci.
106(25):10248-53
Koskiniemi, S., Hughes, D., Andersson, D.I., (2009) Effect of
translesion DNA polymerases, endonucleases and RpoS on mutation rates in Salmonella typhimurium. (Submitted manuscript)
Reprints were made with permission from the respective publishers.
Contents
Introduction...................................................................................................11
Salmonella enterica..................................................................................11
The bacterial genome ...............................................................................12
The eukaryotic cell as a growth niche ......................................................13
Reductive evolution..................................................................................13
Chromosome organization .......................................................................14
Mechanisms of reductive evolution .........................................................15
Homologous recombination.................................................................15
Illegitimate recombination...................................................................17
Mechanisms of deletion formation ......................................................19
DNA breaks and deletion formation....................................................20
Mutagenesis and DNA repair ...................................................................21
Origins of change.................................................................................21
Fidelity of DNA replication ............................................................22
Sources of DNA damage.................................................................23
Maintaining genome integrity..............................................................24
Base excision repair (BER).............................................................24
Nucleotide excision repair (NER)...................................................25
Methyl directed mismatch repair ....................................................26
Mutators...............................................................................................28
Cellular responses to stress..................................................................29
The adaptive mutation controversy .....................................................29
Present investigations....................................................................................32
Reductive evolution can be a rapid process in laboratory settings...........32
Mechanism of genome reduction .............................................................33
Mechanism of deletion formation ............................................................34
Large variation in deletion rate at different chromosomal locations........36
Fitness effects of large chromosomal deletions .......................................37
Mutation rates are remarkably robust.......................................................38
Concluding remarks .................................................................................39
Future perspectives .......................................................................................41
Acknowledgements.......................................................................................43
References.....................................................................................................45
Abbreviations
DNA
NHEJ
S. typhimurium
E. coli
BER
NER
ORF
SSB
ssDNA
dsDNA
Pol III holoenzyme
AP site
PFGE
PCR
Deoxyribonucleic acid
Non-homologous end joining
Salmonella typhimurium
Escherichia coli
Base excision repair
Nucleotide excision repair
Open reading frame
Single strand binding protein
Single stranded DNA
Double stranded DNA
The replicative polymerase with all subunits
An apurinic or apyrimidic site in the DNA
Pulsed field gel electrophoresis
Polymerase chain reaction
Genes
recA
recBCD
ruvAB
recFOR
ligD
dnaE
dnaQ
dnaB
dnaN
dnaQ
holC
dinB
polB
umuDC
samAB
tag
alkA
ung
uvrABC
mutSHL
Major homologous recombination protein
Exonuclease V
Holliday junction solving helicase
Homologous recombination proteins
ATP-dependent DNA ligase
Pol III polymerase subunit
Pol III editing subunit
Pol III helicase subunit
Pol III beta-clamp subunit
Pol III clamploader subunit
Pol III clamploader subunit
Translesion DNA polymerase Pol IV
Translesion DNA polymerase Pol II
Translesion DNA polymerase Pol V
Translesion DNA polymerase SamAB
3-Methyl adenine DNA glycosylase
3-Methyl adenine DNA glycosylase
Uracil DNA glycosylase
NER enzymes
Methyl directed mismatch repair enzymes
lexA
rpoS
katEG
sodC
lacI
lacZ
Repressor of the SOS-response
Alternative transcription sigma factor S
Catalase enzymes
Superoxide dismutase
Repressor of the lac-operon
β-galactosidase enzyme
Introduction
Evolution is essential for our understanding of biology, from the understanding of the beginning of life on earth to modern issues such as the development of antibiotic resistance. Evolution, i.e. changes in allele frequencies,
can occur by a number of processes including mutation, migration, drift and
by natural selection. The latter requires in its simplest form two elements,
genetic variability in a population and selection among this variation. Mutation is one of the major driving forces of evolution as it creates genetic diversity that allows organisms to adapt to the constantly changing environment through natural selection. The fundamental process of evolution by
natural selection was first recognized by Darwin when he described it in his
“Origin of the Species” 150 years ago (Darwin 1859). Natural selection acts
on individuals with variable genetic traits, selecting those that are most fit
for survival in each specific situation. The opposite of natural selection and
the second major evolutionary process is chance (i.e. genetic drift). Whereas
natural selection fixes the fittest individual for survival in a population,
chance will fix any random individual irrespective of fitness. The two keyplayers of evolution, natural selection and chance, are often battling for
precedence in life. In this thesis, I have studied how the forces of evolution
form bacterial genomes and what factors affect this process. To this end I
have used Salmonella enterica serovar typhimurium as a model organism
throughout my work.
Salmonella enterica
Salmonella enterica serovar typhimurium, hereafter referred to as S. typhimurium, is a rod shaped gram-negative bacteria belonging to the class of γproteobacteria (McClelland, Sanderson et al. 2001). Unlike its close relative
Escherichia coli (E. coli), that is mainly found as a human commensal, Salmonella enterica never colonizes the human gut without causing infection
(Winfield and Groisman 2003). Salmonella enterica serovar typhi is the most
virulent of the Salmonella species and rare in that it can only colonize humans. It is the cause of typhoid fever, with an annual global burden of 16
million cases and 600,000 deaths (Parry, Hien et al. 2002).
S. typhimurium on the other hand is not specific to humans and causes a
disease similar to typhoid fever in mice (Mastroeni and Sheppard 2004). In
11
humans S. typhimurium causes a milder disease called salmonellosis with
symptoms such as diarrhea, fever, and abdominal cramps which develop 12
to 72 hours after infection and may last for up to 7 days (Salyers and Whitt
1994). Infection with Salmonella ssp is through the fecal/oral route, mostly
by ingestion of contaminated food or water (Oosterom 1991).
Together with Escherichia coli, Salmonella enterica is one of the two
main workhorses of bacterial genetics and with its fully sequenced genome
and the extensive knowledge of its regulatory networks, S. typhimurium is
one of the most convenient model organisms to use (Neidhardt 1996).
The bacterial genome
Genetic material can be acquired by cells via horizontal gene transfer or through
amplification of the cellular DNA. In contrast DNA can be lost from cells by
deletions (Fig 1). As a consequence bacterial genomes are not fixed but are continuously changing in a population with regard to size and gene content.
Fig. 1
Horizontal
gene transfer
Amplification
Forces affecting
chromosome
composition
Deletion
Genome sizes vary to a large extent between bacterial species, from 0.16 kbp
for the smallest genome known today to up to 13 kbp for the larger genomes
(Casjens 1998; Nakabachi, Yamashita et al. 2006; Schneiker, Perlova et al.
2007). In spite of this size variation, bacterial genomes of all sizes are mostly
tightly packed with genes and contain little non-coding genetic material.
This raises a multitude of questions. For example how can certain bacteria
sustain life when they contain 100 times less genetic material than their larger relatives? From the opposite perspective, if 200 genes are enough to
uphold life, why do organisms with larger genomes carry all the additional
baggage? The answer to these questions is found in the environment where
our microscopic companions reside. In general, intracellular bacteria harbor
smaller genomes than their free-living relatives. To understand why, a closer
look at the intracellular lifestyle of bacteria is needed.
12
The eukaryotic cell as a growth niche
Inside a host cell bacteria living freely in the cell cytoplasm have unlimited
access to most nutrients required for growth, thereby relaxing selection to
maintain many of the bacterial metabolic genes. Relaxed selection allows
these genes to become inactivated by mutation and subsequently lost without
affecting cell fitness or the cells chance to overcome extinction by natural
selection. Furthermore, intracellular bacteria usually encounter population
bottlenecks during transfer from one host to another. By chance any individual cell, irrespective of its fitness in the intracellular environment, might be
selected to establish the new population. These population bottlenecks allow
deleterious mutations such as deletions to become fixed in the population by
chance, which would not have been possible in an environment where natural selection predominates (Andersson and Kurland 1998). The third factor
affecting the genomic content of intracellular bacteria is the lack of horizontal gene transfer. Inside the eukaryotic cell the individual bacteria are
unlikely to receive genetic material from other bacteria, except their closest
relatives, which contain more or less identical genetic material. This results
in a ratchet where the DNA that is lost cannot be regained as the cell cannot
reconstruct the deleted regions through lateral transfer (Moran 1996).
Reductive evolution
A small genome size was previously believed to reflect an evolutionary
primitive state from which free-living species with larger genomes originated (Wallace and Morowitz 1973). The sequencing of many bacterial genomes in the last decade has proved this hypothesis wrong. Instead, sequencing data implies that the smaller genomes of intracellular pathogens and
endosymbionts evolved from larger free-living species (Andersson and Kurland 1998). It is now believed that low rates of gene influx combined with a
mutation bias towards deletions and relaxed selection for certain genes
causes a gradual shrinkage in genome size. Furthermore, in organisms with
small population sizes, recurrent bottlenecks and low rates of recombination,
even slightly deleterious mutations may become fixed in the population. This
phenomenon, known as Mullers ratchet, (Muller 1964; Felsenstein 1974) has
been extensively studied in the genus Buchnera which includes some of the
smallest genomes known to date (Gil, Sabater-Munoz et al. 2002). Two lineages of the Buchnera species, Buchnera aphidicola (Ap) an endosymbiont of
Acyrthosiphon pisum, and Buchnera aphidicola (Sg) an endosymbiont of
Schizaphis graminum, have undergone 200-250 million years (My) of endosymbiotic evolution since their divergence from a free-living gammaproteobacterial ancestor (Moran and Mira 2001). Reconstruction of the genome of the free-living gamma-proteobacterial ancestor suggests that
13
Buchnera lost many of its genes through fixation of large deletions soon
after entering an endosymbiotic lifestyle (Moran and Mira 2001). Interestingly, comparison of the two Bucnera genomes that diverged 50 My ago
revealed that no rearrangements, duplications or horizontal gene transfers
had occurred from the time of their divergence (Tamas, Klasson et al. 2002).
In comparison, free-living relatives belonging to the same gammaproteobacterial family, such as S.typhimurium and E.coli are highly dynamic
and contain several large rearrengements of their chromosomes (Rocha
2008). This evolutionary stasis in Buchnera species over a period of 50 My
has been ascribed to the lack of the major homologous recombination protein
RecA. As a result of this observation it has been inferred that early gene
losses in Buchnera species occurred through a RecA-dependent mechanism,
before the two species diverged (Tamas, Klasson et al. 2002).
The tempo and mode of gene loss has been widely debated throughout the
literature. Are genes generally lost as large fragments of DNA or are they
eradicated by slow erosion of the genetic material? A beautiful illustration of
the latter can be seen in the genome of the obligate intracellular parasite
Ricketsia prowazekii, which is notable for its high number of pseudogenes.
Up to 27% of the R. prowazekii chromosome consists of non-coding genetic
material and it has been speculated that the high fraction of pseudogenes
represent remnants of ancient genes that are in the process of elimination
from the genome. A detailed analysis of the pattern of inactivation in R.
prowazekii has revealed that deletions are far more common than insertions
and are, on average, much larger in size (Andersson and Andersson 1999).
Examples of the former, where genome reduction occurred through large
deletions removing several genes at once, are also found. For example an
analysis of the total gene content of natural isolates of E. coli shows that
large deletions occasionally become fixed within lineages (Ochman and
Jones 2000). Likewise, at least 25 long deletions, in one case with the removal of as many as 16 ORFs at once, has been identified in Mycobacterium
tuberculosis (Kato-Maeda, Rhee et al. 2001). Also for endosymbionts, reductive evolution through formation of large deletions has been inferred
(Ochman and Moran 2001). In conclusion, the loss of DNA from a genome
can occur through either large deletions or through slow erosion, although
the relative importance of these two routes will vary among different bacterial species (Mira, Ochman et al. 2001).
Chromosome organization
In contrast to the size and gene content of bacterial genomes the overall gene
order and orientation of the bacterial chromosome remains relatively unchanged, even between different species. For example when the genome of
E. coli is compared to that of Salmonella enterica, synonymous mutations
14
are almost saturated in the genome and a difference of more than 10% is
seen on the protein level. In comparison, aside from insertions and deletions,
the organization of the two genomes is almost identical (Rocha 2008). So
what are the forces that preserve the gene order of bacterial chromosomes
and why are some changes allowed when others are not? One possible answer to these questions is the operon theory, where genes present in an operon are kept together for regulatory purposes. Operons could also be maintained due to horizontal gene transfer, since transfer of just one of the genes
in an operon would be non-selective for the recipient if genes in an operon
code for different parts of a certain pathway. In contrast, receiving a complete operon could enable the recipient to access a new growth niche
(Lawrence and Roth 1996)). Another possible explanation for the lack of
large chromosomal rearrangements is the symmetry of replication (equal
distance from the origin of replication to the terminus for each replichore)
(Liu, Liu et al. 2006). If a large deletion spans a portion of one half of the
chromosome but not the other, one replichore would be longer than the
other, i.e. the two replichores will not finish replication at the same time,
possibly resulting in problems with decatenation and segregation (reviewed
in (Rocha 2008)).
Mechanisms of reductive evolution
Homologous recombination
For a long time reductive evolution was thought to occur mainly through
homologous recombination between long sequence repeats. A key enzyme in
homologous recombination is the RecA protein, which binds and filaments
on single stranded (ss) DNA (Story, Weber et al. 1992). As replication in
bacterial cells often stalls or stops at different types of lesions or DNA
breaks, a repair system designed to retain the genetic information is necessary. When a DNA break is encountered, break-processing enzymes target
the break and prepare it for homologous recombination. A double stranded
(ds) DNA break is rapidly degraded by the RecBCD enzyme complex,
which possesses high affinity for dsDNA ends. These enzymes degrade the
dsDNA until they reach a recognition site known as chi (χ). At chi the specificity of the enzyme is altered so that only the 5’ DNA strand is degraded.
Simultaneously the enzyme recruits RecA to filament on the created 3’ DNA
overhang (Fig. 2). Filamented RecA can now bind to the double stranded
sister chromatid and search for homology. When homology is reached, RecA
mediates strand transfer and recruits factors for polymerization from the 3’
DNA end, which repair the degraded DNA. The junction that remains after
the strand exchange event will eventually be resolved by the Holliday junction resolving enzymes RuvABC (Kowalczykowski, Dixon et al. 1994).
15
Fig. 2
3’
5’
5’
3’
DSGap
5’
3’
3’
5’
RecB
5’
3’
RecC
RecD
chi
5’
3’
RecA
chi
5’
3’
5’
3’
5’
3’
Replication
restart
3’
5’
5’
3’
Any single stranded (ss) DNA break is rapidly coated with single strand
binding protein, SSB. For RecA to be able to filament on the ssDNA the
SSB protein must be removed. To this end, RecFOR enzymes assist in the
removal of SSB and in the loading of RecA on the ssDNA. RecA then
searches for homology and performs the rest of its task as described for
dsDNA breaks, resulting in a repaired ssDNA break (Fig. 3)
(Kowalczykowski, Dixon et al. 1994).
16
Fig. 3
3’
5’
5’
3’
SSGap
5’
3’
3’
5’
SSB
5’
3’
3’
5’
RecF
RecA
RecO
RecR
5’
3’
3’
5’
5’
3’
3’
5’
5’
3’
3’
5’
5’
3’
Replication
restart
Illegitimate recombination
In eukaryotes, the most common mechanism to repair broken DNA is direct
ligation of the DNA ends, a process referred to as non-homologous end joining (NHEJ). NHEJ requires approximation of the two DNA ends (i.e. bringing the two DNA ends in proximity of each other), a process facilitated by
the DNA-end-binding protein Ku, followed by sealing of the DNA by a specialized ligase (Lig 4 in eukaryotes) (Schar, Herrmann et al. 1997; Teo and
Jackson 1997). This type of recombination (NHEJ) is often referred to as
illegitimate recombination or non-conservative recombination, as it in contrast to homologous recombination will not repair the DNA legitimately, i.e.
without alterations. Recently, these specialized ligases (LigD in eukaryotes)
and the DNA-end-binding protein Ku were found in some species of bacteria, e.g. Mycobacterium, Pseudomonas, Bacillus and Agrobacterium species
(Weller, Kysela et al. 2002; Zhu and Shuman 2005; Moeller, Stackebrandt et
al. 2007; Zhu and Shuman 2007).
17
As many bacteria lack the specialized ligase required for NHEJ, another
mechanism for illegitimate recombination in bacteria has been suggested,
which involves DNA gyrase (Ikeda, Moriya et al. 1981; Ikeda, Shiraishi et
al. 2004). As a replication fork unwinds the DNA, negative supercoiling
ahead of the fork must be relieved and as a result positive supercoiling accumulates in the wake of the replication fork. It is essential for the cell to
convert this positive supercoiling back to its negative form to condense the
DNA and establish the correct balance of constrained and unconstrained
superhelicity (Liu and Wang 1987; Cook, Ma et al. 1992). To this end the
bacterial cell utilizes topoisomerase II, also known as DNA gyrase. The topoisomerase reaction is distinguished by a reaction cycle where an active
tyrosine residue attacks the phosphodiester bond of a DNA strand, creating a
break in the DNA and forming a covalent bond between the 5’ end of the
broken DNA and the tyrosine residue of the topoisomerase (Fig. 4) (Wang
2002). Type I topoisomerases will only break one of the DNA strands,
thereby allowing the other strand to pass through before resealing the DNA
(Champoux 2001).
Fig. 4
However, for type II topoisomerases such as DNA gyrase, the break consists
of two staggered cuts where both 5’ ends remain linked to the tyrosine residues of the enzyme. This complex is called the cleavage complex and once it
has been formed, the enzyme allows dsDNA to pass through the gap changing the DNA topology before resealing the gap (Fig. 4) (Wang 1998; Wang
2002). The model for gyrase-mediated illegitimate recombination proposed
by Ikeda and his coworkers involves two separate gyrase molecules covalently bound to DNA. The two DNA gyrases then meet and undergo subunit
exchange resulting in the formation of recombinant DNA (Fig. 5) (Ikeda,
Shiraishi et al. 2004).
18
Fig. 5
Mechanisms of deletion formation
The most studied mechanism of deletion formation is that of nonconservative recombination between long sequence repeats. Several studies
have used artificial template on plasmids or on the chromosome to elucidate
possible mechanisms of deletion formation (Lovett, Gluckman et al. 1994;
Bierne, Seigneur et al. 1997; Bierne, Vilette et al. 1997; Bzymek and Lovett
2001). The slipped mispairing model is conceptually related to the copychoice model of recombination (Lederberg 1955) which was initially proposed to explain frame-shift mutations (Streisinger, Okada et al. 1966). The
key feature of this model is the slippage of a growing DNA chain from one
sequence repeat to another and the use of the tip of the growing DNA chain
as a primer for further DNA synthesis (Fig. 6a) (Weston-Hafer and Berg
1989; Trinh and Sinden 1991; Lovett, Hurley et al. 2002). This type of slippage is easy to envision as an event occurring inside a replication fork, but
studies of transposon excision have revealed that this type of slippage can
also occur between distant sites (Egner and Berg 1981; Weston-Hafer and
Berg 1989; Gordenin, Malkova et al. 1992). Another model, the break and
join model, introduces a break between the repeats by specific or nonspecific enzymes. Subsequent processing of the resulting DNA ends by exonucleases could then expose the repeats, which could anneal and the resulting molecule could be repaired by DNA synthesis and ligation (Fig. 6b). The
two models make different predictions with regard to the genetic material
from which the illegitimate recombination product molecule is formed. The
19
break and join model predicts the deleted DNA to be absent whereas according to the slipped mispairing model, the deleted DNA is present in the cell.
Both of these models require some sequence homology at the deletion endpoints and experimental setups have estimated how the length of the repeat
affects the rate of non-conservative recombination. For RecA independent
events, the rate of recombination is constant at homologies >50bp whereas
for RecA dependent events homologies >150bp are required for the highest
rates of recombination (Lovett, Hurley et al. 2002). Interestingly, experimental data shows that the formation of large spontaneous deletions on the
chromosome of S. typhimurium (Paper I) (Nilsson, Koskiniemi et al. 2005)
and E.coli (Mashimo, Kawata et al. 2003) frequently occurs between very
short (<10bp) or no repeats. Thus, several processes might be at play in
spontaneous deletion formation on the bacterial chromosome.
DNA breaks and deletion formation
The mechanism of deletion formation between short or no direct DNA repeats might still be unsolved but several factors affecting deletion formation
have been elucidated both for short-homology independent illegitimate recombination (gyrase mediated illegitimate recombination) and the shorthomology dependent illegitimate recombination described above. Many of
these factors can be directed back to one fact, the presence of stalled or broken replication forks in the cell. For gyrase mediated illegitimate recombination, an increase in recombination is seen when cells are treated with
oxolinic acid. Oxolinic acid is an antibiotic belonging to the quinolone antibiotics as it targets DNA gyrase directly. It locks the gyrase enzyme in its
open conformation, resulting in an increase in DNA breaks and increased
deletion formation (Ikeda, Moriya et al. 1981; Ikeda, Aoki et al. 1982; Shi20
mizu, Yamaguchi et al. 1995). For short-homology dependent illegitimate
recombination, one process affecting deletion formation is DNA replication.
A wide variety of mutants with impaired processivity of replication were
shown to have an increased deletion rate. For example, mutations in the
normal replicative polymerase Pol III (dnaE) (Bierne, Vilette et al. 1997;
Saveson and Lovett 1997) and other enzymes of the Pol III holoenzyme;
dnaQ (the editing subunit), dnaB (the helicase), dnaN (the beta-clamp) and
dnaX and holC (clamp loader complex) increased deletion rates (Saveson
and Lovett 1997). Moreover, mutants lacking one of two helicases, rep or
uvrD (DNA helicase II) have a decreased speed of replication and more illegitimate recombination (Shiraishi, Imai et al. 2005; Shiraishi, Imai et al.
2006). In eukaryotes, special DNA conformations, known as non-B DNA,
have been associated with deletion endpoints (Bacolla, Jaworski et al. 2004).
Specific sequence motifs, such as inverted repeat structures, have been
shown to undergo structural changes during replication, resulting in such
non-B DNA conformations, e.g hairpins or cruciforms (Bowater and Wells
2001; Majumdar and Patel 2002). Such structures could function as blockage
for the replication fork, resulting in stalling and breakage of the fork, which
could explain why non-B DNA is associated with deletion endpoints in eukaryotes. Finally, the presence of stalled replication of forks and DNA
breaks has been shown more directly to result in deletion formation (Kong
and Masker 1994; Bierne, Ehrlich et al. 1997; Michel, Ehrlich et al. 1997).
How slow processivity or stalling of the replication fork could mediate deletion between sites that are hundreds of kilobases apart is however still unknown.
Mutagenesis and DNA repair
Mutation plays a fundamental role in evolution but it is also a potential threat
to the viability and survival of cells. An inactivating mutation in an essential
gene results in cell death, and for unicellular organisms such as bacteria, the
loss of that individual from the population. As a consequence all characterized species harbor mechanisms to limit and prevent the accumulation of too
many mutations and a resulting mutational meltdown.
Origins of change
Spontaneous genomic mutation rates are rather similar among unicellular
microbes. A survey conducted by Jan Drake (1991) compared mutation rates
in bacteria, phage and fungi with very different genome sizes and found that
the mutation frequencies were similar for all species examined, approximately 0.003 mutations / genome / generation (Drake 1991). This would
21
suggest that the rate of mutation and the rate of repair have been optimized
during evolution.
Fidelity of DNA replication
DNA replication is the largest contributor to maintaining genomic stability.
Faithful replication of genetic material allows for unaltered inheritance of
essential characteristics from parent to offspring. Nevertheless, every time a
genome is duplicated, there is a possibility of base misincorporation. Even
the most accurate DNA polymerases make errors as they replicate the genome and these mistakes form part of the mutation frequencies observed in
all organisms. The replicative DNA polymerase III (Pol III) normally replicates the bacterial chromosome in S. typhimurium and E. coli. Pol III is a
complex multicomponent enzyme where the core holoenzyme consists of 10
different subunits. The holoenzyme includes both proofreading and exonuclease activities which limit the amount of errors produced (reviewed in
(Nohmi 2006)). Factors that cause incorporation of the wrong base include
damaged or modified bases. In S. typhimurium and E. coli specialized polymerases exist to replicate past various types of DNA lesions and damaged
bases. These polymerases are known as the translesion DNA polymerases,
and they are capable of replication past lesions that the normal replicative
polymerase Pol III cannot pass (Ohmori, Friedberg et al. 2001). In this process the translesion DNA polymerases produce errors in their wake, which
has led to the suggestion that the majority of mutations are the result of error-prone replication by the translesion DNA polymerases (Wagner and
Nohmi 2000; Nohmi 2006). Five DNA polymerases are found in S. typhimurium and E. coli; the normal replicative polymerase Pol III, the major
gap-filling polymerases Pol I and Pol II and the translesion DNA polymerases Pol IV and Pol V. Despite its proofreading activity Pol II is also
counted as a translesion DNA polymerase as it is involved in error-prone
translesion synthesis across various damages in the template DNA (reviewed
in (Nohmi 2006)). In Salmonella a sixth polymerase SamAB, a homologue
of Pol V, is encoded on the virulence plasmid pSLT (Nohmi, Hakura et al.
1991). The translesion DNA polymerases are each specialized in bypassing
different types of lesions although a high degree of redundancy can be seen
(reviewed in (Nohmi 2006)). The current model of translesion synthesis
proposes that the replicative DNA polymerase Pol III stalls at damaged bases
and is replaced by one of the SOS induced polymerases to enable bypass of
the lesion. However, recent findings suggest that the normal replicative polymerase Pol III itself has the ability to bypass several lesions in an errorprone manner (Fujii and Fuchs 2004; Kokubo, Yamada et al. 2005). So how
are these polymerases coordinated when multiple polymerases possess the
ability to bypass the same lesions? The beta-subunit of Pol III holoenzyme
(HE), hereafter referred to as the beta-clamp, appears to have a key role in
coordinating polymerase switching as it interacts with all the polymerases at
22
the same site (Lopez de Saro, Georgescu et al. 2003; Burnouf, Olieric et al.
2004). Interestingly, a hierarchy appears to exist (Pol III > Pol I > Pol II >
Pol IV = Pol V, in S. typhimurium) among the polymerases in their propensity to gain access to the beta-clamp of the replication fork (reviewed in
(Nohmi 2006)). As the beta-clamp is a homodimer two polymerases can
bind to the replication fork simultaneously, one for leading strand synthesis
and the other for lagging strand synthesis (Indiani, McInerney et al. 2005).
Recent evidence however, suggests that three polymerases are bound to the
fork simultaneously, where the third polymerase is held in reserve
(McInerney, Johnson et al. 2007). This illustrates the replication fork as a
highly dynamic complex that is able to frequently switch between polymerases with little effort.
Sources of DNA damage
DNA can be damaged both by endogenous and external factors. An example
of the former is spontaneous decomposition of DNA in the form of depurination or depyrimidination. This occurs when the N-glycosyl bond of the deoxyribnonucleotides is exposed to hydrolytic attack, resulting in a apurinic
/apyrimidic (AP) site. An AP site is non-coding and the passing polymerase
could, in theory, insert any base at that position. However, the AP sites are
difficult to pass and in E. coli adenine (A) is inserted with high frequency
opposite the AP site as the A/AP pair is thermodynamically more stable than
others. Insertion of an A base opposite the AP site results in GC to TA or AT
to TA transversions (Loeb and Preston 1986). Transversions are mutations
that change a purine to a pyrimidine and vice versa and result in more dramatic structural changes, thought to be toxic. In contrast, a purine to purine
or a pyrimidine to pyrimidine exchange are called transitions and considered
less deleterious than transversions. Deamination is another type of spontaneous DNA damage, that occurs when the exocyclic amino group of the deoxyribonucleotides is hydrolyzed. Mutations caused by deamination of cytosine,
5-methylcytosine, adenine and guanine are mostly transitions and in the case
of guanine, the correct base is inserted (Kreutzer and Essigmann 1998).
With the exception of endogenous DNA damage bacterial cells are continuously exposed to various damaging agents from the environment, such as
UV-light, acidic or alkaline pH, temperature shifts and ionizing radiation.
External damaging agents will result in a variety of base damage, the most
studied being the formation of pyrimidine dimers caused by exposure to UVlight. Bacteria living in an aerobic environment need protection from oxidative damage due to the production of hydroxyl free radicals (OH.) that can
deprotonate any DNA position causing a wide variety of base damage. The
best-characterized hydroxyl-radical induced DNA lesions are 8oxoguanosine (8-oxo-dG) and cis-thymine glycol. Thymine glycol is toxic in
dsDNA as it blocks the replication fork and is mutagenic in ssDNA where it
causes AT to GC transitions. 8-oxo-dG on the other hand, can mispair with
23
adenine during replication resulting in GC to TA transversions (Cheng, Cahill et al. 1992).
Maintaining genome integrity
Only a limited subset of DNA damage is repaired by direct reversal. Instead,
most DNA damage is repaired by excision of the damaged or inappropriate
base followed by replacement with the correct one. To this end bacterial
cells possess a cellular response, referred to as excision repair, which can be
further divided into subclasses according to the mechanism of action.
Base excision repair (BER)
The excision of damaged bases is initiated by the action of DNA glycosylases, which catalyze the hydrolysis of the N-glycosylic bonds linking bases
to the phosphate backbone of DNA. Excision repair initiated by DNA glycosylases is referred to as base excision repair (BER), as the excised elements
consist of free bases (Fig. 7a) (Duncan, Hamilton et al. 1976). Removal of
the base however, results in another form of DNA damage, an apurinic /
apyrimidic (AP) site. The repair of AP-sites require another set of enzymes
known as AP-endonucleases which will target the 3’ or 5’ phosphodiesterbonds next to the AP-site, nicking the phosphate backbone of the DNA. Subsequently, exonucleases can enter the DNA at the nick and remove the APsite from the phosphate backbone, resulting in repair by replication and subsequent ligation (Lindahl 1979). Each DNA glycosylase acts on very specific
damaged bases, e.g. Tag and AlkA only recognize and excise 3methyladenine whereas Ung removes uracil residues that are incorporated
instead of thymine residues (Lindahl, Ljungquist et al. 1977; Yamamoto,
Katsuki et al. 1978; Steinum and Seeberg 1986). As a consequence E. coli
and S. typhimurium possess a wide variety of DNA glycosylases all with
their own specificity for different damaged bases.
24
Fig. 7a
Fig. 7b
Base excission repair
Guanine
O
N
O
-
O
-
O
O
O
-
NH
O
O
H2C
O
O
N
O
O
O
P
O
NH
+
O
H 2C
-
OH
N
O
O
N
O
O
O
Adenine
P
O
N
O
O
H2C
N
NH2
-
O
N
O
P
O
N
O
O
H2C
N
O
-
O
N
NH2
-
O
N
O
O
O
-
O
P
O
O
H2C
N
O
O
-
P
O
N
P
O
O
NH2
O
H2C
OH
Adenine
OH
-
O
O
P
N
N
NH2
N
O
O
H2C
N
O
O
O
O
UvrC
endonuclease
N
NH2
O
-
O
N
O
N
H2C
Adenine
O
P
Cytosine
P
NH
O
Adenine
P
NH2
N
O
Thymine
H
-
O
O
N
O
HC
H2C
O
-
N
O
Uracil
O
NH2
O
N
N
O
H2C
Cytosine
P
H2C
O
P
O
Uracil-DNA
glycosylase
O
N
O
NH2
N
O
O
O
O
N
H 2C
Uracil
O
P
O
-
O
N
N
N
O
H2C
NH2
O
O
N
NH2
N
Cytosine
O
P
N
H2C
O
O
NH2
Guanine
O
N
N
N
O
H2C
Cytosine
O
P
O
-
NH2
N
Guanine
O
N
N
N
O
H2C
Nucleoide excission repair
Guanine
N
N
O
O
-
O
P
O
O
Nucleotide excision repair (NER)
Like base excision repair, nucleotide excision repair (NER) is a multistep
process that results in the formation of a gap in the DNA duplex that needs
to be repaired by DNA polymerases and DNA ligase. In contrast to BER,
NER excises the damaged bases, e.g. pyrimidine dimers or photoproducts, as
intact nucleotides, or parts of oligonucleotide fragments (Grossman, Caron et
al. 1988). The excision itself produces a gap in the DNA duplex that can be
repaired by replication and ligation (Fig. 7b). A fundamental property of
NER is the ability of the repair machinery to locate and recognize damaged
bases. In E. coli and S. typhimurium two proteins collaborate to mediate this
task. The UvrA protein in its dimeric form (UvrA2) binds DNA containing
various forms of base damage (Grossman, Caron et al. 1988). However,
these DNA-UvrA2 complexes are very unstable and it has been suggested
that the role of UvrA in NER is that of a “molecular matchmaker” as it recruits UvrB to the site of damage before dissociating from the DNA (Lin and
Sancar 1992). UvrB-DNA complexes are, on the contrary, highly stable. The
third factor involved is the damage specific endonuclease, UvrC, which possesses high affinity for the UvrB-DNA complexes. The UvrC protein is constitutively expressed at low levels and possesses the endonuclease activity
required to incise the DNA (Orren and Sancar 1989). When over-expressed
from a plasmid in the absence of DNA damage UvrC has been shown to
25
attack undamaged DNA (Branum, Reardon et al. 2001). UvrC nicks the
DNA on both sides of the damaged base and an oligonucleotide fragment is
removed. The small ssDNA gap created can then be repaired by DNA polymerases and DNA ligase (Fig. 7c).
Fig. 7c
UvrB
UvrAUvrA
TT
UvrC
TT
UvrC
TT
UvrC
TT
DNA helicase II
UvrC
TT
TT
DNA
polymerase I
26
TT
Methyl directed mismatch repair
One of the most important contributors to mismatched base pairs in the bacterial cell is replication. In this case the correct base is located on the parental strand of the newly replicated DNA and to maintain fidelity the newly
synthesized strand must be repaired.
Fig. 8
To this end the cell possesses a
MutS
mechanism to separate between parental and newly synthesized DNA,
DNA methylation. After replication,
the newly synthesized DNA strand is
rapidly
methylated
by
Dammethylase at GATC consensus sequences. The short window of time
MutL
between replication and methylation
allows the cell to distinguish between
newly synthesized DNA and parental
MutH
DNA. During this time the MutS
enzyme of the methyl directed mismatch repair system scans the newly
synthesized DNA for errors. When
MutS recognizes a mispaired base
pair it recruits the MutL and MutH
proteins to form a complex, which is
believed to facilitate bending of the
DNA. The bent DNA brings the nearest GATC sequence into contact with
DNA
the MutH protein which nicks the
helicase II
unmethylated DNA strand at the
GATC sequence. Thus, the state of
methylation determines the specificity
Exo I
RecJ
of repair (Modrich 1991). DNA heliExo X
case II then unwinds the newly synthesized DNA strand between the
GATC site and the mismatch. As the
GATC site can be located either 3’ or 5’ to the mismatch, the excised DNA
will be degraded by the proper exonuclease depending on the direction (Exo
VII and RecJ for 5’ and Exo I for 3’) (Grilley, Griffith et al. 1993). When the
mismatched DNA has been removed, DNA polymerase can re-synthesize the
remaining gap to complete the repair process (Fig. 8).
27
Mutators
Strains with an increased mutation rate compared to the standard (i.e. the
median rate for the population) rate of mutation of a particular species, are
referred to as mutators. A mutator phenotype can be generated in several
ways, but the most common route is inactivation of one of the cellular repair
pathways. Inactivating mutations in, for example, mutS, mutH or mutL have
been shown to increase mutation rates approximately 200-fold in E. coli
(Schaaper and Dunn 1991). Mutators are a source of genetic diversity in a
population and it has been assumed that a subpopulation of mutators is always present. However, in the absence of variable environments and strong
novel selection, mutators do not generally benefit from their high mutation
rate. As beneficial mutations only arise at very low frequencies (Denamur
and Matic 2006), most mutations accumulating in these cells are detrimental.
If the selection pressure would change however, the mutators have a higher
chance of carrying or acquiring a mutation that is beneficial in the new selective environment. It is well documented that mutators have a selective advantage over non-mutators during strong selection and in variable environments (Mao, Lane et al. 1997; Miller, Suthar et al. 1999; Giraud, Matic et al.
2001). For example during the evolution of 12 lineages of E. coli in minimal
M9 media supplemented with glucose, mutators arose in 3 of those lineages
after 10000 generations (Cooper, Rozen et al. 2003). Furthermore, constitutive mutS- mutators were shown to initially have an advantage compared to
the wild type in colonizing the mouse gut. However, when the two strains
had undergone 400 generations of preadaptation in the mice gut, the mutator
no longer possessed an advantage in colonizing as the beneficial mutations
had already been acquired by both wild type and mutS- strains (Giraud,
Matic et al. 2001). Selection of constitutive mutators and their role in adaptation has been well documented in natural isolates where mutators are frequently found among both pathogenic and commensal bacteria (reviewed in
(Giraud, Radman et al. 2001)). Mutator bacteria succeed in short-term adaptation as they can accumulate adaptive mutations faster than wild type bacteria, allowing them to take over the growth niche. After selection is relieved
however, the mutators will have accumulated mutations that may be neutral
in the present environment but may decrease their fitness upon a shift to a
new environment. However, mutator bacteria are not necessarily doomed.
Mutator phenotypes can be reversed either by back mutation or by replacing
the mutated gene by means of horizontal gene transfer (Denamur, Lecointre
et al. 2000). Alternatively, the mutators could acquire suppressor mutations
of the mutator phenotype, thereby preventing further accumulation of disadvantageous mutations (Trobner and Piechocki 1984).
28
Cellular responses to stress
When bacterial cells encounter stressful situations such as nutrient starvation, low pH, oxidative stress, or the consequences of such stress i.e. DNA
damage, response systems in the cell are activated to protect the cell from
damage.
Upon DNA damage a response system in the cell, known as the SOSresponse, is activated (George, Devoret et al. 1974; Radman 1975). Damaged DNA is rapidly processed and filamentation of RecA on ssDNA mediates proteolytic cleavage of the LexA repressor of the SOS-response
(Moreau 1985). Induction of the LexA regulon alters the expression of approximately 40 genes under LexA control. Among these genes are the translesion DNA polymerases (Fernandez De Henestrosa, Ogi et al. 2000; Quillardet, Rouffaud et al. 2003). Under non-induced conditions these polymerases are present at relatively low levels but their expression is increased
10-20 fold during the SOS-response (Delmas and Matic 2006). It is assumed
that this increase facilitates replication across various types of DNA lesions
including the DNA damage that may have activated the stress response to
begin with. LexA also regulates the expression of the UvrA and UvrB proteins of the nucleotide excision repair pathway described above, which can
then remove and repair any modified nucleotides causing the stress.
Also, conditions such as oxidative stress, nutrient starvation or low pH,
induce a cellular stress response that is regulated by the alternative transcription sigma factor RpoS. Induction of Rpos changes the expression of more
than 300 genes to varying degrees (Patten, Kirchhof et al. 2004; Weber, Polen et al. 2005). Among these genes, many proteins involved in the protection and repair of DNA are found, e.g. one of the TLS DNA polymerases Pol
IV (dinB) and the catalase genes katE and katG, which protect against the
formation of oxygen free radicals by catalysing the formation of water and
oxygen from hydrogen peroxide (Loewen 1984; Loewen, Triggs et al. 1985;
Layton and Foster 2003). The sodC gene is also up-regulated during RpoS
induction and is involved in protection against superoxide, as the gene product superoxide dismutase breaks down superoxide ions (Fridovich 1989;
Imlay and Imlay 1996).
The adaptive mutation controversy
A key to survival for microorganisms such as bacteria is their ability to adapt
to a rapidly changing environment. Adaptation requires a phenotypic change,
which will bestow the individual with this new phenotype with higher fitness
in the changed environment. Such phenotypic change can be the result of
altered regulation of a certain pathway but for adaptation to be hereditary a
genetic change is required, i.e. a mutation. The classical neo-Darwinian view
of evolution postulates that mutations occur randomly and at a rate that is
29
essentially independent of selection. The fact that at least some spontaneous
mutations arise before and independently of selection was demonstrated
early by the classic experiments of Luria and Delbruck (1943) and Lederberg
(1952) (Luria and Delbruck 1943; Lederberg and Lederberg 1952). Luria
and Delbruck studied the origin of phage-resistant mutants that appeared
when a culture of E. coli was plated in the presence of bacteriophage T1.
Differences in the number of mutants which arose in independent cultures
led them to conclude that the mutations were rare spontaneous events and
occurred prior to selection (Luria and Delbruck 1943). However, the lethal
selections they used made it impossible to detect mutations inducible by
selective conditions. Several researchers (Shapiro 1984; Cairns, Overbaugh
et al. 1988; Hall 1990; Hall 1990) pointed out this deficiency and described
genetic systems where selective conditions seemed to increase the mutation
rate (Shapiro 1984; Cairns, Overbaugh et al. 1988; Hall 1990; Hall 1990).
The Cairns system has been analyzed in greatest detail. In this system Lacmutant cells, (Lac- due to of a frameshift mutation in the lacI region of a
lacIZ gene fusion), incubated on lactose-containing medium accumulated
mutations, allowing them to grow on lactose, over several days (Cairns,
Overbaugh et al. 1988) (Cairns and Foster 1991). From these observations
several models have been presented, suggesting that starvation for lactose
induces a stress which in turn increases the mutation rate of the cells, resulting in the increased number of Lac+ revertants observed after several days.
One specific model system suggests that the increased mutagenesis requires
induction of the RpoS regulon and over-expression of the translesion DNA
polymerase Pol IV (dinB), found under the control of LexA and RpoS. The
model suggests that a subpopulation of cells under stress will enter a hypermutable state and that these cells will accumulate mutations at a high rate
until a beneficial mutation restoring growth occurs. For cells to enter the
hypermutable state two factors are required simultaneously: induction of the
general RpoS stress response by any stress inducing factor and induction of
Pol IV expression by spurious SOS-induction. Induction of Pol IV expression by two pathways simultaneously would then increase the mutation rate
of the cell by an error-prone break repair process (Fig. 9a) (Ponder, Fonville
et al. 2005). It is easy to interpret the observation of increasing Lac+ cells at
later time points as a global or localized increase in mutation rate. If, however, even a subset of the population grows, new opportunities for mutation
are provided and any pre-existing variant with the smallest ability to grow,
can improve. In fact, for many of the systems where stress induced
mutagenesis have been proposed, the evidence that exists favors models
where a subpopulation of cells grow under selection and a stepwise improvement occurs during the selection period (Cairns, Overbaugh et al.
1988; Taddei, Halliday et al. 1997; Andersson, Slechta et al. 1998; Wrande,
Roth et al. 2008). Thus, an alternative explanation for the accumulation of
mutants on lactose in the lac-system suggests that if a subpopulation of the
30
plated cells carry a pre-existing duplication of the leaky lac-gene, that subpopulation can grow slowly on the lactose medium. Further amplification of
this pre-existing duplication increases growth on lactose further. The increased copy number of the leaky lac-gene increases the number of target
genes for which a beneficial -1 frameshift mutation can revert the lac-gene
back to wild type function. After reversion there is no selection to retain the
amplification and the cells will revert back to one copy of lac (Fig. 9b) (reviewed in (Roth, Kugelberg et al. 2006)).
Fig. 9a
-1 frameshift
Fig. 9b
DSB
-1 frameshift
SOS-induction -> Pol IV
Homologous recombination
Growth on lactose
(+)
lacI lacZ
Stress
High fidelity
repair
+
RpoS-induction
Pol IV
n
0 frameshift
n
+++
+++
+++
Stress-induced point mutations
31
Present investigations
Reductive evolution can be a rapid process in laboratory
settings
Obligate parasites and endosymbionts possess the smallest bacterial genomes seen today, but they are believed to have originated from free-living
ancestors with larger genomes (Gil, Sabater-Munoz et al. 2002). The tempo
and mode of reductive evolution has been widely debated throughout the
literature. An attempt to calculate some approximate rate of early genome
reduction in Buchnera species established that the initial gene loss must have
occurred faster than one gene / 5-10 million years (My), a number that was
calculated for the last 50 My of separate evolution for two lineages of
Buchnera species (Tamas, Klasson et al. 2002).
In paper I we set out to study genome reduction in a laboratory setting
and set up an experimental system where we could partly mimic the intracellular lifestyle of endosymbiotic and parasitic bacteria. By growing cells on
rich medium plates, hematin agar, we mimicked the nutrient rich environment bacteria would encounter inside host cells. As population bottlenecks
are associated with fixation of deleterious mutations (e.g. deletions) transferring cells with one-cell bottlenecks could mimic the extensive genetic drift
the intracellular bacteria typically encounter inside host. The lack of horizontal gene transfer inside a host represents a ratchet that only allows the reductive process of endosymbiontic genomes to go in one direction. To simulate
this a single colony was always picked during passaging to minimalize horizontal gene transfer among the bacteria.
Fig. 10
32
The evolution experiment was performed using two strains with either a low
wild type or high (mutS-) mutation rate. 60 lineages of S.typhimurium LT2
wild type and the mutS- mutant each were grown on hematin agar plates.
Every day one random colony from each lineage was re-streaked onto a new
plate (Fig. 10). This was repeated for 300 and 60 cycles, for wild type and
mutS- mutant respectively, resulting in a total of ~7500 and ~1500 generations of growth. Cells were continuously (every 30 cycles) checked for
chromosomal rearrangements by PFGE but for the wild type strain no rearrangements could be seen even after 300 cycles, ~7500 generations. Using
the mod-gal assay the deletion rate of the mutS- mutant was determined to be
50-fold higher than that of the wild type. The deletions found in the mutSstrain varied from 173bp to 1,2 kbp in size and from the four deletions found
in the mutS- after ~1500 generations of growth, an initial rate of gene loss
could be calculated. Assuming a generation time of one day, the initial rate
of gene loss could be calculated to 2,5bp /cell /generation for the mutS- mutant and 0,05bp /cell /generation for a wt strain. These rates most likely represent the maximum rate of genome reduction, as subsequent deletions
probably would occur at a lower rate due to the reduction in amount of deletable DNA. However, on an evolutionary time scale these data indicate that
genome reduction could occur during a very short time, approximately 1Mbp
in 50,000 years if a generation time of one day is assumed. This is an extremely high rate of gene-loss compared to the estimation of the initial rate
of genome reduction in the Buchnera species, 1 gene / 50 My. These calculations are obviously associated with a considerable level of uncertainty since
many factors such as the size of the bottleneck and the degree of horizontal
gene transfer will affect the rate of DNA loss. However, for the Buchnera
species there is reason to believe that reductive evolution initially occurred at
high rates, to slow down as the genomes shrunk and the amount of nonessential genes decreased (Tamas 2002).
Mechanism of genome reduction
How genome reduction occurs has been a subject of debate during the last
decades (Moran and Mira 2001). Are genes lost in big pieces of DNA or
slowly eroded by smaller deletions? According to our results in paper I genome reduction can initially occur through large genomic deletions, up to
200kb at the time. Furthermore genes appearing to be essential in gene
knock-out studies (Knuth, Niesalla et al. 2004), can be lost in these large
deletions with no impairment of growth. It has been proposed that genome
reduction during evolution of the intracellular parasites and endosymbionts
was mainly carried out through RecA- dependent homologous recombination. This was used as an argument for the evolutionary stasis in B.aphdicola
the last 50million years as these strains long were the only sequenced bacte33
ria lacking RecA (Tamas, Klasson et al. 2002). However, analysis of the
endpoints for the deletions analyzed in paper I indicated little (<15bp) or no
homology at the deletion endpoints. Actually, 5/6 deletions found could not
have been formed through RecA dependent homologous recombination, as
this requires at least 25bp of homology (Shen and Huang 1986). These results suggest that many large spontaneous deletions are not formed through
RecA mediated homologous recombination, implying that RecA plays a less
important role in reductive evolution than previously thought.
Mechanism of deletion formation
If large spontaneous deletions are not formed between long direct repeats,
how are they generated? As S. typhimurium does not appear to posses the
enzymes required for NHEJ, another mechanism must be involved in the
formation of these deletions. Genes involved in DNA repair and replication
have been shown to influence deletion formation and the presence of DNA
breaks, or stalled replication forks have been shown to increase deletion
formation in several systems (Kong and Masker 1994; Bierne, Ehrlich et al.
1997; Bierne, Vilette et al. 1997; Michel, Ehrlich et al. 1997; Saveson and
Lovett 1997). In paper III we set out to study the mechanism of spontaneous deletion formation on the bacterial chromosome. Our aim was to investigate the role of four translesion DNA polymerases found in S.typhimurium,
in spontaneous deletion formation. We studied spontaneous deletion formation in a specific region of the S. typhimurium chromosome. In the moaA-gal
region of the S.typhimurium chromosome two operons, moaA and gal are
separated by approximately 40 kb. The moaA gene encodes a protein involved in molybdate biosynthesis. Molybdate is a co-factor of nitrate reductase and inactivation of nitrate reductase turns cells chlorate resistant as they
no longer convert chlorate to the toxic chlorite under anaerobic growth conditions (Miller and Amy 1983). Also, any inactivating mutation in any of the
gal genes, makes cells unable to grow on galactose, which can be seen as
white colony appearance on MacConkey agar plates containing galactose.
Finally we included a kanamycin resistance gene between the two operons.
By selecting for cells that are chlorate resistant and screening for white
kanamycin sensitive colonies, we could measure the deletion rate (Fig. 11)
(Nilsson, Koskiniemi et al. 2005). We found that 90% of all spontaneous
deletions required the four translesion DNA polymerases, Pol II, Pol IV, Pol
V and SamAB for their formation. Conversely, increased expression of these
polymerases, via SOS induction or artificial overproduction from an arabinose inducible plasmid increased the rate of deletion formation about 10- to
30-fold. To our knowledge this is the first direct evidence provided showing
that deletion formation is dependent on the translesion DNA polymerases.
Previous reports suggest an involvement of the normal replicative Pol III in
34
deletion formation between short and long tandem repeats (Trinh and Sinden
1993; Lovett and Feschenko 1996) and in illegitimate recombination (Dutra,
Sutera et al. 2007). Furthermore, an important role for the translesion polymerases, in particular Pol IV, has been shown for frameshift mutations (Kim,
Maenhaut-Michel et al. 1997) and in deletion formation between long tandem repeats, but the mechanism of frameshift formation and RecA depedent
homologous recombination is different from that observed for large deletions
with no or little homology at deletion endpoints (Lovett, Hurley et al. 2002;
Tippin, Pham et al. 2004).
Fig. 11
837339
848198
mod
kan
gal
Inactivation results in
Chlorate
resistance
Kanamycin
sensitivity
White appearance on
MacConkey agar plates
So how would these translesion DNA polymerases contribute to deletion
formation in the cell? One possible mechanism is by interference at the replication fork. Since the translesion DNA polymerases have a lower processivity at the replication fork compared to the normal replicative polymerase
Pol III (Wagner, Fujii et al. 2000), an excess of these polymerases at the fork
could result in frequent replication pausing. To test this hypothesis we used
the TUNEL assay to measure the number of DNA breaks in the cell. The
TUNEL assay uses terminal deoxyribonucleotidyltransferase, an enzyme
that adds on fluorescent nucleotides to free 3’OH DNA ends and as a consequence the relative increase in DNA breaks can be measured as an increase
in fluorescence. We found that cells with a constitutively de-repressed
LexA-regulon had a 10-fold increase in DNA breaks. The increase in breaks
was not mediated by the translesion DNA polymerases but by three endonucleases, UvrBC and Cho, under LexA-control. Earlier reports confirm that
over-expression of UvrC, in the absence of DNA damage resulted in an increase in DNA breaks (Branum, Reardon et al. 2001). Interestingly, removal
of the three endonucleases under LexA-control resulted in a concomitant
reduction in deletion rate. This indicates that two separate factors are limiting deletion formation, the number of DNA breaks and the amount of translesion DNA polymerases in the cell. In conclusion we suggest that the translesion DNA polymerases do not increase spontaneous deletion formation by
35
an increase in DNA breaks, but possibly use the breaks as template for extension of misaligned DNA ends thereby allowing deletion formation between sites with little or no homology.
Large variation in deletion rate at different
chromosomal locations
Even though bacterial chromosomes are highly dynamic in their nature the
overall gene organization of the chromosome remains relatively unaltered
between different species. For example, comparison of the genomes of two
closely related bacteria, E. coli and S. typhimurium, revealed that while synonymous substitutions were nearly saturated between these two genomes and
amino acid changes occurred up to 10%, the organization of the chromosome remained almost the same (Rocha 2008). One of the factors affecting
chromosomal rearrangements is the presence of supercoiling and other
macro domains in the chromosome. The presence of supercoiling domains
has been shown to confer mechanistic constraints on chromosomal rearrangements, such as deletions, creating boundaries across which rearrangements rarely occur (Charlebois and St Jean 1995). In paper II we constructed a genetic tool, the “deletometer”, to study how chromosomal location will affect large chromosomal rearrangements in the form of deletions.
The deletometer can be used to study spontaneous chromosomal deletion
rates at various chromosomal locations using the same genetic screen. The
deletometer consists of a transposable backbone (Tn10dtet) carrying the lac
operon, the moaA gene and a chloramphenicol resistance gene. The gene
product of the moaA gene is involved in the molybdate biosynthesis pathway
and loss of this gene renders cells chlorate resistant. Cells lacking the lac
operon will no longer be able to grow on lactose, which can be seen as white
colony appearance on MacConkey agar plates containing lactose. Finally,
cells lacking a chloramphenicol gene are not able to grow on plates containing chloramphenicol. Using these selections and screens the deletometer
could be used to study deletion rates at various chromosomal locations, by
plating cells on MacConkey agar plates containing lactose and screening for
chlorate resistant white colonies that were sensitive to chloramphenicol.
We measured deletion rates at 12 chromosomal locations using the deletometer and found that the spontaneous deletion rate varied 44-fold between
different chromosomal locations. As the rate of deletion formation is likely
to appear higher when the deletometer is surrounded by more deletable material, we corrected for this effect by normalizing the deletion rate at each
location to the experimentally defined deletable region. The deletable regions were mapped as the largest deletion found at each location using
pulsed field gel electrophoresis (PFGE) and PCR. After this normalization,
36
deletion rates varied more than a 100-fold between different chromosomal
locations. The highest deletion rates were found with three individual deletometers all located around 2 Mbp of the S.typhiumurium chromosome, implying a potential hotspot for deletion formation in this region. This could be
expected if some of the deletometers were flanked by perfect direct repeats,
such as IS-elements. However, no longer (>25bp) direct repeats were found
in the regions with higher deletion rates than those with lower deletion rates.
Differences in the rate of deletion formation at different chromosomal locations could affect the trajectory of reductive evolution. As deletions are generally considered deleterious, they will be fixed in a population by chance
due to high drift and small population bottlenecks. However, if deletion rates
at one chromosomal location are 100-fold higher as compared to other regions, the chance of fixing deletions in this region is likely to increase, affecting the rate of reductive evolution in that specific region.
Fitness effects of large chromosomal deletions
Chromosomal rearrangements, such as large deletions are prone to alter the
highly conserved organization of the chromosome. Bacterial chromosomes
are selected for specific gene order and orientation and large rearrangements
in the genome organization are often deleterious for the cells (Rocha 2008).
Using the deletions found with the deletometer, in paper II we set out to
study how different large deletions at various chromosomal locations would
affect cellular fitness. Fitness was measured as growth rate during exponential growth in rich LB media and poorer M9-media supplemented with glycerol. The isolated deletions ranged from 1-200 kbp in size but no negative
correlation between fitness (measured as exponential growth rate) and deletion size could be seen. Instead the gene content of the deletion seemed to
affect fitness more than the actual size. Surprisingly, some of the deletions,
ranging from 18- to 38 kbp in size, increased the exponential growth rate of
the cells with up to 10%. Why removal of certain genetic elements would
increase the growth rate of the cells is at present unclear but one explanation
could be that loss of certain genetic elements that are unused in both rich and
minimal growth media, reduces energetic costs of DNA, RNA and protein
synthesis, freeing resources for synthesis of other required gene products.
Another alternative explanation could be that the deleted regions contain
sequences that posses a direct inhibitory effect on growth at these conditions,
as has been observed for anti-virulence genes in bacterial pathogens
(Maurelli 2007).
37
Mutation rates are remarkably robust
The classical neo-Darwinian view of evolution postulates that mutations
arise randomly and at a rate that is independent of selection. It has however
been suggested that bacteria can change both the rate and specificity of mutation formation in response to various stresses. One specific model-system
proposes that this putative stress-induced mutagenesis requires both induction of the error-prone DNA polymerase IV under control of the DNA damage-inducible LexA regulon (SOS response) and an increase in levels of the
starvation-induced sigma-factor RpoS (Galhardo, Hastings et al. 2007). In
paper IV we have examined this idea by determining whether a simultaneous induction of the LexA-mediated SOS response and an increase in RpoS
levels will cause an increase in mutation rates. We have studied the impact
of these two different stress responses on four different types of mutations:
resistance to the antibiotics nalidixic acid and rifampicin caused by amino
acid changes in gyrA and rpoB, respectively, resistance to chlorate caused by
any inactivating mutation in the chlA-G genes and reversion to Lac+ caused
by -1 frameshift mutations in the lacI gene of the the lacIZ gene fusion.
We observed that constitutive de-repression of the LexA regulon by a
lexA(def) mutation increased mutation rates for all types of mutations. However, the increase was due to the translesion DNA polymerases for only one
type of mutations. For rifampicin- and nalidixic acid resistance mutations
and Lac+ reversion no decrease in mutation rates was seen in a strain lacking
all four translesion DNA polymerases in a lexA(def) background. In contrast,
for chlorate resistance mutations the mutation rate was decreased to wild
type levels when inactivating all four translesion DNA polymerases in a
lexA(def) background. Chlorate resistance can be acquired by a wide variety
of mutations in a large number of genes whereas rifampicin and nalidixic
acid resistance mutations are acquired through a limited subset of basepair
substitutions in specific genes. Our results imply that the translesion DNA
polymerases do contribute to the increased mutagenesis seen when derepressing the LexA-regulon, but only to some types of mutations. For other
types of mutations, such as base pair substitutions, the translesion DNA polymerases do not contribute considerably to mutagenesis, at least not at the
polymerase levels conferred when the LexA regulon is de-repressed by a
lexA(def) mutation. In contrast to the translesion DNA polymerases, three
endonucleases found under LexA-control, decreased mutation rates to wild
type levels for all types of mutations in a lexA(def) background. In paper III
we showed that these endonucleases under LexA-control, increase the
amount of ssDNA gaps in the cell when overexpressed constitutively in a
lexA(def) mutant. These results indicate that an increase in DNA breaks/gaps
increases the mutation rate, through any polymerase at the fork in an errorprone gap repair process. Error-prone break repair has been described earlier, but as a mechanism where the translesion DNA polymerase Pol IV re38
pairs DNA breaks in an error-prone manner, resulting in an increased
mutagenesis in the cell (Ponder, Fonville et al. 2005). Here we show that the
DNA breaks per se appear to increase mutagenesis, with or without the translesion DNA polymerases.
With regard to the potential involvement of RpoS in mutagenesis, neither
an increase in RpoS levels conferred by artificial over-expression from a
plasmid nor long-term stationary phase incubation or slow growth caused an
increase in any of the four mutation rates measured, alone or in combination
with over-expression of the translesion DNA polymerases. In conclusion,
mutation rates are remarkably robust and no combination of growth conditions, induction of translesion polymerases or increased RpoS expression
could confer an increase in mutation rates higher than the moderate increase
caused by de-repression of the LexA regulon alone.
Concluding remarks
Bacteria frequently encounter rapidly changing environments and their ability to adapt is the key for their survival. An understanding of the mechanisms and rates of genetic change is necessary if we want to gain insight into
the future of our microscopic companions. Microorganisms are masters of
adaptation, and to co-exist we must find ways to keep up, for example by
limiting their rate of adaptation to antibiotics. In this thesis my work has
focused on the changes in the bacterial chromosome and which factors affect
this change.
It is fascinating to see how fast selection, or the lack of selection, can alter
the composition of a bacterial genome. In paper I we showed that during
conditions of high genetic drift, genome reduction could occur at very high
rates. Furthermore, in paper II we show that large spontaneous deletions can
actually increase fitness of the cells in a specific environment. Thus, it is
possible that reductive genome evolution can be driven by selection as well.
This is an example of where genetic drift (i.e. chance) and natural selection
could result in essentially the same outcome, i.e. gene loss. Yet another example of how little we know about the forces affecting chromosome composition is seen in studies where the aim has been to establish the minimal gene
set required for life (Koonin 2000; Gil, Silva et al. 2004; Knuth, Niesalla et
al. 2004; Glass, Assad-Garcia et al. 2006). In paper I we find that many
genes found to be essential in such knock-out studies (Knuth, Niesalla et al.
2004), were easily disposable in large deletions, with no obvious affects on
bacterial fitness.
Even though selection is often the factor deciding which changes are
fixed in a population and which are not, the rates of mutation will play an
important role in environments where purifying selection is low. In paper II
we find that the rate of deletion formation can vary at least 100-fold depend39
ing on the chromosomal region. Differences like this are likely to affect the
trajectory of reductive evolution, as the probability of fixing frequent deletions is likely to increase.
Another aspect of adaptation is how it occurs mechanistically. How are
large chromosomal rearrangements formed in the cell and what are the
sources of mutation? In paper I we observed that most large chromosomal
deletions were not generated by homologous recombination, an important
finding since homologous recombination has been suggested as the major
path to genome reduction (Tamas, Klasson et al. 2002). In paper III we
follow up this finding and show that large spontaneous deletions require the
translesion DNA polymerases and DNA gaps for their formation. The importance of stalled replication forks and DNA gaps in deletion formation has
been established previously (Kong and Masker 1994; Bierne, Ehrlich et al.
1997), but the forces acting on the stalled or broken forks to mediate deletions have not been identified. The same two factors, DNA breaks and translesion DNA polymerases have been described to induce an error-prone
break repair process that will repair the DNA, but with errors (Ponder, Fonville et al. 2005; Lovett 2006). In paper IV we found that mutation rates
increased with increased numbers of DNA breaks, but even though the translesion DNA polymerases did contribute to mutagenesis, they were only
responsible for a subset of mutations. In conclusion, restart of a stalled or
collapsed replication fork per se seems to increase mutagenesis in the form
of basepair substitution, whereas deletions require the translesion DNA polymerases at the restarting fork. It will be interesting to see how and why
restart of a replication fork results in increased mutagenesis.
40
Future perspectives
From the results of this thesis several interesting follow-up projects have
emerged.
At first it will be interesting to determine why deletion rates vary more than a
100-fold between different chromosomal locations. One explanation could be
the presence of supercoiling domains that affect the rate of recombination and
deletion formation in the cells. To test this the level of supercoiling of the cells
could be altered. Generally the DNA of E. coli cells is more negatively supercoiled than that of its close relative S. typhimurium. It is possible to change the
supercoiling of S. typhimurium to that of E. coli and see how that would affect
the rates of deletion formation (Champion and Higgins 2007). Will the rates
change uniformly over the chromosome, or will only some regions be affected? Another possible explanation is that regions with highly transcribed
genes have higher rates of deletion formation as the DNA in these regions is
more accessible for DNA damaging agents that could generate DNA breaks.
To test this, the deletometer could be transferred to a gene known to have low
expression but that is in close proximity to the original region with high deletion rate. If the hypothesis is true, the deletion rate should be lower at this new
location. The third possible reason is the presence of DNA structures such as
hairpins or cruciforms in the DNA. Cruciforms and hairpins can function as
roadblocks for the replication fork and the presence of such cruciforms could
therefore increase deletion rate locally. This hypothesis could be tested in a
mutant lacking the hairpin specific endo- and exonuclease SbcCD. SbcCD
cleaves and degrades these structures resulting in more DNA gaps, which have
been shown increase the frequency of deletion formation. Mutants lacking this
enzyme should have lower deletion rates in regions where cruciforms are influencing deletion formation.
Secondly it will be interesting to determine why some large chromosomal
deletions increase cellular fitness. Initially, we need to find out the exact
endpoints of these deletions. Possibly the deletions contain genes that are
highly expressed but not required for growth in that specific environment.
Alternatively, the region could contain specific DNA sequences that inhibit
growth as seen for anti-virulence genes in pathogens. It will also be important to find out if all the deletions that increase exponential growth rate also
out-compete the wild type during the whole growth cycle.
41
Thirdly, we still do not know which types of deletion endpoints are generated in the large spontaneous chromosomal deletions. It would be interesting
to compare the deletion endpoints in cells with and without the four translesion DNA polymerases present. This could give us clues to whether the translesion DNA polymerases contribute to short-homology dependent illegitimate recombination or short-homology independent illegitimate recombination, or both.
Finally, it would be interesting to find out why stalled and broken replication
forks increase the formation of deletions and mutations. Are restarted replication forks somehow different from normal replicative replication forks and
if they are, how are they different?
42
Acknowledgements
I am not a master with words nor someone who is known to extensively
show my gratitude, so for all of you that have helped me through the years,
here are my thanks to you.
First of all I would like to thank my supervisor Dan Andersson without
whom this thesis would never have come to be. Thank you for all the inspiration and guidance along my path to becoming an independent researcher, I
do not think anyone else could have done a better job. I am deeply grateful
for the time I have spent under your supervision.
I would also like to thank my co-supervisor Diarmaid Hughes for all of your
critical and constructive opinions on whatever the project. Together you and
Dan added up to the perfect balance of enthusiasm and structure. You are
greatly appreciated.
I would also like to thank my co-authors and other nice scientific people that
I have had the opportunity to meet over the years. Special thanks to John
Roth for all the great suggestions about my projects and for communicating
one of the PNAS papers. And the people in John's lab, Drew, Amiko, Semarhy, Doug, Jacob and Eric, it has been a pleasure to get to know you.
Thanks to Jim Sawitzke for many tips and tricks about linear transformations. And to Kelly Hughes, Sue Lovett, Stanley Malloy, Paul Rainey, Martin Marinus and many more for inspiring me to become a better researcher
and for all good comments on my projects over the years.
I am grateful to all the past members of the DA-lab: To Annika for taking
care of me when I first started and for all the support since then. You were
my lab supervisor, but you soon became one of my dearest friends. Wille,
who has brought fun and energy to my world of science. You are outstanding in intensity and those who have not met you do not know what hard
work really is. Elisabeth, I counted the days to you're return to Stockholm as
your presence always made the lab brighter. Sophie, the one who knew everything, you were and still are a role-model in science for me. Cissi, who
always had great suggestions and time to answer stupid questions. Therese,
for all the great times in the gym and for discussions on every perspective of
life. And finally to Caroline who became one of my best friends from the
43
first day in the lab and have remained by my side for all these years, you are
my safety vent.
Thanks also to all the present members of the DA-lab, without whom I don’t
think I would have survived. You feed me and help me to remain sane, but
most of all, the lab is a better place with you people in it. Special thanks to
Anna who has supported me like no one else. You are truly one of a kind and
I am sure that nothing I have done during this time would have been the
same without you. Maria, for the joyful singing and the atmosphere you
bring to the lab. Thank you for being you. Linus for always making me feel
brilliant and creative even when I’m not. You are the hub that our lab circles
around. Chris, for all the fun times in and outside of the lab and for being a
great friend. You bring joy to my world. Peter, who began as my student and
soon became the master. Thank you for all the mind ravelling discussions
and for being special. To Song who is outstanding and scientifically way
past the rest of us already. It has been a pleasure to get to know you. To
Jocke for all the great scientific discussions in and outside of the lab, you are
my problem solver. And to the newcomers; Amira, Hervé, Marlen, Erik and
Hava, with hopes of good times to come and wishing you the best of luck in
the future.
I would also like to thank the people of B9:3, D7:3 and many others at IMBIM for great times and fun Iphab pubs and Imbim days. And the secretaries especially Barbro, Marianne and Rehné for your help with all the paper
work.
Special thanks to Chris and Anna for helping me with the language as I
wrote this thesis and to Erik or helping me with the cover.
Special thanks to Lena for taking care of me when I first arrived in Uppsala
and for you're continued support throughout all the years that followed.
I would also like to thank Jojjo for being my friend and for supporting me
over the years. You are the greatest.
And last but not least, I would like to thank my family. My parents for supporting me every step of the way. To know you are always there for me is
what keeps me going. My sister for inspiring me to get into the world of
science in the first place. You are my dearest friend and the one I turn to
when troubled. Thank you for being there. My brother and Janette for supplying me with my own safe haven. Thank you for providing a home full of
joy and atmosphere when needed. And finally to my Tomas without whom I
would never have lasted this long. I love you.
44
References
Andersson, D. I., E. S. Slechta and J. R. Roth (1998). "Evidence that gene
amplification underlies adaptive mutability of the bacterial lac operon."
Science 282(5391): 1133-5.
Andersson, J. O. and S. G. Andersson (1999). "Insights into the evolutionary
process of genome degradation." Curr Opin Genet Dev 9(6): 664-71.
Andersson, S. G. and C. G. Kurland (1998). "Reductive evolution of resident
genomes." Trends Microbiol 6(7): 263-8.
Bacolla, A., A. Jaworski, J. E. Larson, J. P. Jakupciak, N. Chuzhanova, S. S.
Abeysinghe, C. D. O'Connell, D. N. Cooper and R. D. Wells (2004).
"Breakpoints of gross deletions coincide with non-B DNA conformations."
Proc Natl Acad Sci U S A 101(39): 14162-7.
Bierne, H., S. D. Ehrlich and B. Michel (1997). "Deletions at stalled replication forks occur by two different pathways." Embo J 16(11): 3332-40.
Bierne, H., M. Seigneur, S. D. Ehrlich and B. Michel (1997). "uvrD mutations enhance tandem repeat deletion in the Escherichia coli chromosome
via SOS induction of the RecF recombination pathway." Mol Microbiol
26(3): 557-67.
Bierne, H., D. Vilette, S. D. Ehrlich and B. Michel (1997). "Isolation of a
dnaE mutation which enhances RecA-independent homologous recombination in the Escherichia coli chromosome." Mol Microbiol 24(6): 1225-34.
Bowater, R. P. and R. D. Wells (2001). "The intrinsically unstable life of
DNA triplet repeats associated with human hereditary disorders." Prog Nucleic Acid Res Mol Biol 66: 159-202.
Branum, M. E., J. T. Reardon and A. Sancar (2001). "DNA repair excision
nuclease attacks undamaged DNA. A potential source of spontaneous mutations." J Biol Chem 276(27): 25421-6.
Burnouf, D. Y., V. Olieric, J. Wagner, S. Fujii, J. Reinbolt, R. P. Fuchs and
P. Dumas (2004). "Structural and biochemical analysis of sliding
clamp/ligand interactions suggest a competition between replicative and
translesion DNA polymerases." J Mol Biol 335(5): 1187-97.
45
Bzymek, M. and S. T. Lovett (2001). "Instability of repetitive DNA sequences: the role of replication in multiple mechanisms." Proc Natl Acad Sci
U S A 98(15): 8319-25.
Cairns, J. and P. L. Foster (1991). "Adaptive reversion of a frameshift mutation in Escherichia coli." Genetics 128(4): 695-701.
Cairns, J., J. Overbaugh and S. Miller (1988). "The origin of mutants." Nature 335(6186): 142-5.
Casjens, S. (1998). "The diverse and dynamic structure of bacterial genomes." Annu Rev Genet 32: 339-77.
Champion, K. and N. P. Higgins (2007). "Growth rate toxicity phenotypes
and homeostatic supercoil control differentiate Escherichia coli from Salmonella enterica serovar Typhimurium." J Bacteriol 189(16): 5839-49.
Champoux, J. J. (2001). "DNA topoisomerases: structure, function, and
mechanism." Annu Rev Biochem 70: 369-413.
Charlebois, R. L. and A. St Jean (1995). "Supercoiling and map stability in
the bacterial chromosome." J Mol Evol 41(1): 15-23.
Cheng, K. C., D. S. Cahill, H. Kasai, S. Nishimura and L. A. Loeb (1992).
"8-Hydroxyguanine, an abundant form of oxidative DNA damage, causes G---T and A----C substitutions." J Biol Chem 267(1): 166-72.
Cook, D. N., D. Ma, N. G. Pon and J. E. Hearst (1992). "Dynamics of DNA
supercoiling by transcription in Escherichia coli." Proc Natl Acad Sci U S A
89(22): 10603-7.
Cooper, T. F., D. E. Rozen and R. E. Lenski (2003). "Parallel changes in
gene expression after 20,000 generations of evolution in Escherichiacoli."
Proc Natl Acad Sci U S A 100(3): 1072-7.
Darwin, C. (1859). On the Origin of Species by Means of Natural Selection,
or the Preservation of Favoured Races in the Struggle for Life. London,
John Murray
Albemarle street.
Delmas, S. and I. Matic (2006). "Interplay between replication and recombination in Escherichia coli: impact of the alternative DNA polymerases."
Proc Natl Acad Sci U S A 103(12): 4564-9.
Denamur, E., G. Lecointre, P. Darlu, O. Tenaillon, C. Acquaviva, C. Sayada,
I. Sunjevaric, R. Rothstein, J. Elion, F. Taddei, M. Radman and I. Matic
46
(2000). "Evolutionary implications of the frequent horizontal transfer of
mismatch repair genes." Cell 103(5): 711-21.
Denamur, E. and I. Matic (2006). "Evolution of mutation rates in bacteria."
Mol Microbiol 60(4): 820-7.
Drake, J. W. (1991). "A constant rate of spontaneous mutation in DNAbased microbes." Proc Natl Acad Sci U S A 88(16): 7160-4.
Duncan, J., L. Hamilton and E. C. Friedberg (1976). "Enzymatic degradation
of uracil-containing DNA. II. Evidence for N-glycosidase and nuclease activities in unfractionated extracts of Bacillus subtilis." J Virol 19(2): 338-45.
Dutra, B. E., V. A. Sutera, Jr. and S. T. Lovett (2007). "RecA-independent
recombination is efficient but limited by exonucleases." Proc Natl Acad Sci
U S A 104(1): 216-21.
Egner, C. and D. E. Berg (1981). "Excision of transposon Tn5 is dependent
on the inverted repeats but not on the transposase function of Tn5." Proc
Natl Acad Sci U S A 78(1): 459-63.
Felsenstein, J. (1974). "The evolutionary advantage of recombination." Genetics 78(2): 737-56.
Fernandez De Henestrosa, A. R., T. Ogi, S. Aoyagi, D. Chafin, J. J. Hayes,
H. Ohmori and R. Woodgate (2000). "Identification of additional genes belonging to the LexA regulon in Escherichia coli." Mol Microbiol 35(6):
1560-72.
Fridovich, I. (1989). "Superoxide dismutases. An adaptation to a paramagnetic gas." J Biol Chem 264(14): 7761-4.
Fujii, S. and R. P. Fuchs (2004). "Defining the position of the switches between replicative and bypass DNA polymerases." Embo J 23(21): 4342-52.
Galhardo, R. S., P. J. Hastings and S. M. Rosenberg (2007). "Mutation as a
stress response and the regulation of evolvability." Crit Rev Biochem Mol
Biol 42(5): 399-435.
George, J., R. Devoret and M. Radman (1974). "Indirect ultravioletreactivation of phage lambda." Proc Natl Acad Sci U S A 71(1): 144-7.
Gil, R., B. Sabater-Munoz, A. Latorre, F. J. Silva and A. Moya (2002). "Extreme genome reduction in Buchnera spp.: toward the minimal genome
needed for symbiotic life." Proc Natl Acad Sci U S A 99(7): 4454-8.
47
Gil, R., F. J. Silva, J. Pereto and A. Moya (2004). "Determination of the core
of a minimal bacterial gene set." Microbiol Mol Biol Rev 68(3): 518-37.
Giraud, A., I. Matic, O. Tenaillon, A. Clara, M. Radman, M. Fons and F.
Taddei (2001). "Costs and benefits of high mutation rates: adaptive evolution of bacteria in the mouse gut." Science 291(5513): 2606-8.
Giraud, A., M. Radman, I. Matic and F. Taddei (2001). "The rise and fall of
mutator bacteria." Curr Opin Microbiol 4(5): 582-5.
Glass, J. I., N. Assad-Garcia, N. Alperovich, S. Yooseph, M. R. Lewis, M.
Maruf, C. A. Hutchison, 3rd, H. O. Smith and J. C. Venter (2006). "Essential
genes of a minimal bacterium." Proc Natl Acad Sci U S A 103(2): 425-30.
Gordenin, D. A., A. L. Malkova, A. Peterzen, V. N. Kulikov, Y. I. Pavlov,
E. Perkins and M. A. Resnick (1992). "Transposon Tn5 excision in yeast:
influence of DNA polymerases alpha, delta, and epsilon and repair genes."
Proc Natl Acad Sci U S A 89(9): 3785-9.
Grilley, M., J. Griffith and P. Modrich (1993). "Bidirectional excision in
methyl-directed mismatch repair." J Biol Chem 268(16): 11830-7.
Grossman, L., P. R. Caron, S. J. Mazur and E. Y. Oh (1988). "Repair of
DNA-containing pyrimidine dimers." Faseb J 2(11): 2696-701.
Hall, B. G. (1990). "Directed evolution of a bacterial operon." Bioessays
12(11): 551-8.
Hall, B. G. (1990). "Spontaneous point mutations that occur more often
when advantageous than when neutral." Genetics 126(1): 5-16.
Ikeda, H., K. Aoki and A. Naito (1982). "Illegitimate recombination mediated in vitro by DNA gyrase of Escherichia coli: structure of recombinant
DNA molecules." Proc Natl Acad Sci U S A 79(12): 3724-8.
Ikeda, H., K. Moriya and T. Matsumoto (1981). "In vitro study of illegitimate recombination: involvement of DNA gyrase." Cold Spring Harb Symp
Quant Biol 45 Pt 1: 399-408.
Ikeda, H., K. Shiraishi and Y. Ogata (2004). "Illegitimate recombination
mediated by double-strand break and end-joining in Escherichia coli." Adv
Biophys 38: 3-20.
Imlay, K. R. and J. A. Imlay (1996). "Cloning and analysis of sodC, encoding the copper-zinc superoxide dismutase of Escherichia coli." J Bacteriol
178(9): 2564-71.
48
Indiani, C., P. McInerney, R. Georgescu, M. F. Goodman and M. O'Donnell
(2005). "A sliding-clamp toolbelt binds high- and low-fidelity DNA polymerases simultaneously." Mol Cell 19(6): 805-15.
Kato-Maeda, M., J. T. Rhee, T. R. Gingeras, H. Salamon, J. Drenkow, N.
Smittipat and P. M. Small (2001). "Comparing genomes within the species
Mycobacterium tuberculosis." Genome Res 11(4): 547-54.
Kim, S. R., G. Maenhaut-Michel, M. Yamada, Y. Yamamoto, K. Matsui, T.
Sofuni, T. Nohmi and H. Ohmori (1997). "Multiple pathways for SOSinduced mutagenesis in Escherichia coli: an overexpression of dinB/dinP
results in strongly enhancing mutagenesis in the absence of any exogenous
treatment to damage DNA." Proc Natl Acad Sci U S A 94(25): 13792-7.
Knuth, K., H. Niesalla, C. J. Hueck and T. M. Fuchs (2004). "Large-scale
identification of essential Salmonella genes by trapping lethal insertions."
Mol Microbiol 51(6): 1729-44.
Kokubo, K., M. Yamada, Y. Kanke and T. Nohmi (2005). "Roles of replicative and specialized DNA polymerases in frameshift mutagenesis: mutability
of Salmonella typhimurium strains lacking one or all of SOS-inducible DNA
polymerases to 26 chemicals." DNA Repair (Amst) 4(10): 1160-71.
Kong, D. and W. Masker (1994). "Deletion between direct repeats in T7
DNA stimulated by double-strand breaks." J Bacteriol 176(19): 5904-11.
Koonin, E. V. (2000). "How many genes can make a cell: the minimal-geneset concept." Annu Rev Genomics Hum Genet 1: 99-116.
Kowalczykowski, S. C., D. A. Dixon, A. K. Eggleston, S. D. Lauder and W.
M. Rehrauer (1994). "Biochemistry of homologous recombination in Escherichia coli." Microbiol Rev 58(3): 401-65.
Kreutzer, D. A. and J. M. Essigmann (1998). "Oxidized, deaminated cytosines are a source of C --> T transitions in vivo." Proc Natl Acad Sci U S A
95(7): 3578-82.
Lawrence, J. G. and J. R. Roth (1996). "Selfish operons: horizontal transfer
may drive the evolution of gene clusters." Genetics 143(4): 1843-60.
Layton, J. C. and P. L. Foster (2003). "Error-prone DNA polymerase IV is
controlled by the stress-response sigma factor, RpoS, in Escherichia coli."
Mol Microbiol 50(2): 549-61.
Lederberg, J. (1955). "Recombination mechanisms in bacteria." J Cell
Physiol Suppl 45(Suppl. 2): 75-107.
49
Lederberg, J. and E. M. Lederberg (1952). "Replica plating and indirect
selection of bacterial mutants." J Bacteriol 63(3): 399-406.
Lin, J. J. and A. Sancar (1992). "(A)BC excinuclease: the Escherichia coli
nucleotide excision repair enzyme." Mol Microbiol 6(16): 2219-24.
Lindahl,
T.
(1979).
"DNA
glycosylases,
endonucleases
for
apurinic/apyrimidinic sites, and base excision-repair." Prog Nucleic Acid
Res Mol Biol 22: 135-92.
Lindahl, T., S. Ljungquist, W. Siegert, B. Nyberg and B. Sperens (1977).
"DNA N-glycosidases: properties of uracil-DNA glycosidase from Escherichia coli." J Biol Chem 252(10): 3286-94.
Liu, G. R., W. Q. Liu, R. N. Johnston, K. E. Sanderson, S. X. Li and S. L.
Liu (2006). "Genome plasticity and ori-ter rebalancing in Salmonella typhi."
Mol Biol Evol 23(2): 365-71.
Liu, L. F. and J. C. Wang (1987). "Supercoiling of the DNA template during
transcription." Proc Natl Acad Sci U S A 84(20): 7024-7.
Loeb, L. A. and B. D. Preston (1986). "Mutagenesis
apurinic/apyrimidinic sites." Annu Rev Genet 20: 201-30.
by
Loewen, P. C. (1984). "Isolation of catalase-deficient Escherichia coli mutants and genetic mapping of katE, a locus that affects catalase activity." J
Bacteriol 157(2): 622-6.
Loewen, P. C., B. L. Triggs, C. S. George and B. E. Hrabarchuk (1985).
"Genetic mapping of katG, a locus that affects synthesis of the bifunctional
catalase-peroxidase hydroperoxidase I in Escherichia coli." J Bacteriol
162(2): 661-7.
Lopez de Saro, F. J., R. E. Georgescu, M. F. Goodman and M. O'Donnell
(2003). "Competitive processivity-clamp usage by DNA polymerases during
DNA replication and repair." Embo J 22(23): 6408-18.
Lovett, S. T. (2006). "Replication arrest-stimulated recombination: Dependence on the RecA paralog, RadA/Sms and translesion polymerase, DinB."
DNA Repair (Amst) 5(12): 1421-7.
Lovett, S. T. and V. V. Feschenko (1996). "Stabilization of diverged tandem
repeats by mismatch repair: evidence for deletion formation via a misaligned replication intermediate." Proc Natl Acad Sci U S A 93(14): 7120-4.
Lovett, S. T., T. J. Gluckman, P. J. Simon, V. A. Sutera, Jr. and P. T. Drapkin (1994). "Recombination between repeats in Escherichia coli by a recA50
independent, proximity-sensitive mechanism." Mol Gen Genet 245(3): 294300.
Lovett, S. T., R. L. Hurley, V. A. Sutera, Jr., R. H. Aubuchon and M. A.
Lebedeva (2002). "Crossing over between regions of limited homology in
Escherichia coli. RecA-dependent and RecA-independent pathways." Genetics 160(3): 851-9.
Luria, S. E. and M. Delbruck (1943). "Mutations of Bacteria from Virus
Sensitivity to Virus Resistance." Genetics 28(6): 491-511.
Majumdar, A. and D. J. Patel (2002). "Identifying hydrogen bond alignments
in multistranded DNA architectures by NMR." Acc Chem Res 35(1): 1-11.
Mao, E. F., L. Lane, J. Lee and J. H. Miller (1997). "Proliferation of mutators in A cell population." J Bacteriol 179(2): 417-22.
Mashimo, K., M. Kawata and K. Yamamoto (2003). "Roles of the RecJ and
RecQ proteins in spontaneous formation of deletion mutations in the Escherichia coli K12 endogenous tonB gene." Mutagenesis 18(4): 355-63.
Mastroeni, P. and M. Sheppard (2004). "Salmonella infections in the mouse
model: host resistance factors and in vivo dynamics of bacterial spread and
distribution in the tissues." Microbes Infect 6(4): 398-405.
Maurelli, A. T. (2007). "Black holes, antivirulence genes, and gene inactivation in the evolution of bacterial pathogens." FEMS Microbiol Lett 267(1):
1-8.
McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L.
Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H.
Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston and R. K.
Wilson (2001). "Complete genome sequence of Salmonella enterica serovar
Typhimurium LT2." Nature 413(6858): 852-6.
McInerney, P., A. Johnson, F. Katz and M. O'Donnell (2007). "Characterization of a triple DNA polymerase replisome." Mol Cell 27(4): 527-38.
Michel, B., S. D. Ehrlich and M. Uzest (1997). "DNA double-strand breaks
caused by replication arrest." Embo J 16(2): 430-8.
Miller, J. B. and N. K. Amy (1983). "Molybdenum cofactor in chlorateresistant and nitrate reductase-deficient insertion mutants of Escherichia
coli." J Bacteriol 155(2): 793-801.
51
Miller, J. H., A. Suthar, J. Tai, A. Yeung, C. Truong and J. L. Stewart
(1999). "Direct selection for mutators in Escherichia coli." J Bacteriol
181(5): 1576-84.
Mira, A., H. Ochman and N. A. Moran (2001). "Deletional bias and the
evolution of bacterial genomes." Trends Genet 17(10): 589-96.
Modrich, P. (1991). "Mechanisms and biological effects of mismatch repair." Annu Rev Genet 25: 229-53.
Moeller, R., E. Stackebrandt, G. Reitz, T. Berger, P. Rettberg, A. J. Doherty,
G. Horneck and W. L. Nicholson (2007). "Role of DNA repair by nonhomologous-end joining in Bacillus subtilis spore resistance to extreme dryness, mono- and polychromatic UV, and ionizing radiation." J Bacteriol
189(8): 3306-11.
Moran, N. A. (1996). "Accelerated evolution and Muller's rachet in endosymbiotic bacteria." Proc Natl Acad Sci U S A 93(7): 2873-8.
Moran, N. A. and A. Mira (2001). "The process of genome shrinkage in the
obligate symbiont Buchnera aphidicola." Genome Biol 2(12): RESEARCH0054.
Moreau, P. L. (1985). "Role of Escherichia coli RecA protein in SOS induction and post-replication repair." Biochimie 67(3-4): 353-6.
Muller, H. J. (1964). "The Relation of Recombination to Mutational Advance." Mutat Res 106: 2-9.
Nakabachi, A., A. Yamashita, H. Toh, H. Ishikawa, H. E. Dunbar, N. A.
Moran and M. Hattori (2006). "The 160-kilobase genome of the bacterial
endosymbiont Carsonella." Science 314(5797): 267.
Neidhardt (1996). Escherichia coli and Salmonella: Cellular and molecular
biology. Washington D.C., ASM press.
Nilsson, A. I., S. Koskiniemi, S. Eriksson, E. Kugelberg, J. C. Hinton and D.
I. Andersson (2005). "Bacterial genome size reduction by experimental evolution." Proc Natl Acad Sci U S A 102(34): 12112-6.
Nohmi, T. (2006). "Environmental stress and lesion-bypass DNA polymerases." Annu Rev Microbiol 60: 231-53.
Nohmi, T., A. Hakura, Y. Nakai, M. Watanabe, S. Y. Murayama and T. Sofuni (1991). "Salmonella typhimurium has two homologous but different
umuDC operons: cloning of a new umuDC-like operon (samAB) present in a
52
60-megadalton cryptic plasmid of S. typhimurium." J Bacteriol 173(3): 105163.
Ochman, H. and I. B. Jones (2000). "Evolutionary dynamics of full genome
content in Escherichia coli." Embo J 19(24): 6637-43.
Ochman, H. and N. A. Moran (2001). "Genes lost and genes found: evolution of bacterial pathogenesis and symbiosis." Science 292(5519): 1096-9.
Ohmori, H., E. C. Friedberg, R. P. Fuchs, M. F. Goodman, F. Hanaoka, D.
Hinkle, T. A. Kunkel, C. W. Lawrence, Z. Livneh, T. Nohmi, L. Prakash, S.
Prakash, T. Todo, G. C. Walker, Z. Wang and R. Woodgate (2001). "The Yfamily of DNA polymerases." Mol Cell 8(1): 7-8.
Oosterom, J. (1991). "Epidemiological studies and proposed preventive
measures in the fight against human salmonellosis." Int J Food Microbiol
12(1): 41-51.
Orren, D. K. and A. Sancar (1989). "The (A)BC excinuclease of Escherichia
coli has only the UvrB and UvrC subunits in the incision complex." Proc
Natl Acad Sci U S A 86(14): 5237-41.
Parry, C. M., T. T. Hien, G. Dougan, N. J. White and J. J. Farrar (2002).
"Typhoid fever." N Engl J Med 347(22): 1770-82.
Patten, C. L., M. G. Kirchhof, M. R. Schertzberg, R. A. Morton and H. E.
Schellhorn (2004). "Microarray analysis of RpoS-mediated gene expression
in Escherichia coli K-12." Mol Genet Genomics 272(5): 580-91.
Ponder, R. G., N. C. Fonville and S. M. Rosenberg (2005). "A switch from
high-fidelity to error-prone DNA double-strand break repair underlies
stress-induced mutation." Mol Cell 19(6): 791-804.
Quillardet, P., M. A. Rouffaud and P. Bouige (2003). "DNA array analysis
of gene expression in response to UV irradiation in Escherichia coli." Res
Microbiol 154(8): 559-72.
Radman, M. (1975). "SOS repair hypothesis: phenomenology of an inducible
DNA repair which is accompanied by mutagenesis." Basic Life Sci 5A: 35567.
Rocha, E. P. (2008). "The organization of the bacterial genome." Annu Rev
Genet 42: 211-33.
Roth, J. R., E. Kugelberg, A. B. Reams, E. Kofoid and D. I. Andersson
(2006). "Origin of mutations under selection: the adaptive mutation controversy." Annu Rev Microbiol 60: 477-501.
53
Salyers and Whitt (1994). Bacterial pathogenesis. Washington D.C., ASM
press.
Saveson, C. J. and S. T. Lovett (1997). "Enhanced deletion formation by
aberrant DNA replication in Escherichia coli." Genetics 146(2): 457-70.
Schaaper, R. M. and R. L. Dunn (1991). "Spontaneous mutation in the Escherichia coli lacI gene." Genetics 129(2): 317-26.
Schar, P., G. Herrmann, G. Daly and T. Lindahl (1997). "A newly identified
DNA ligase of Saccharomyces cerevisiae involved in RAD52-independent
repair of DNA double-strand breaks." Genes Dev 11(15): 1912-24.
Schneiker, S., O. Perlova, O. Kaiser, K. Gerth, A. Alici, M. O. Altmeyer, D.
Bartels, T. Bekel, S. Beyer, E. Bode, H. B. Bode, C. J. Bolten, J. V. Choudhuri, S. Doss, Y. A. Elnakady, B. Frank, L. Gaigalat, A. Goesmann, C.
Groeger, F. Gross, L. Jelsbak, L. Jelsbak, J. Kalinowski, C. Kegler, T.
Knauber, S. Konietzny, M. Kopp, L. Krause, D. Krug, B. Linke, T. Mahmud, R. Martinez-Arias, A. C. McHardy, M. Merai, F. Meyer, S. Mormann,
J. Munoz-Dorado, J. Perez, S. Pradella, S. Rachid, G. Raddatz, F. Rosenau,
C. Ruckert, F. Sasse, M. Scharfe, S. C. Schuster, G. Suen, A. Treuner-Lange,
G. J. Velicer, F. J. Vorholter, K. J. Weissman, R. D. Welch, S. C. Wenzel, D.
E. Whitworth, S. Wilhelm, C. Wittmann, H. Blocker, A. Puhler and R. Muller (2007). "Complete genome sequence of the myxobacterium Sorangium
cellulosum." Nat Biotechnol 25(11): 1281-9.
Shapiro, J. A. (1984). "Observations on the formation of clones containing
araB-lacZ cistron fusions." Mol Gen Genet 194(1-2): 79-90.
Shen, P. and H. V. Huang (1986). "Homologous recombination in Escherichia coli: dependence on substrate length and homology." Genetics
112(3): 441-57.
Shimizu, H., H. Yamaguchi and H. Ikeda (1995). "Molecular analysis of
lambda bio transducing phage produced by oxolinic acid-induced illegitimate recombination in vivo." Genetics 140(3): 889-96.
Shiraishi, K., Y. Imai, S. Yoshizaki and H. Ikeda (2005). "Rep helicase suppresses short-homology-dependent illegitimate recombination in Escherichia
coli." Genes Cells 10(11): 1015-23.
Shiraishi, K., Y. Imai, S. Yoshizaki, T. Tadaki, Y. Ogata and H. Ikeda
(2006). "The role of UvrD in RecET-mediated illegitimate recombination in
Escherichia coli." Genes Genet Syst 81(4): 291-7.
54
Steinum, A. L. and E. Seeberg (1986). "Nucleotide sequence of the tag gene
from Escherichia coli." Nucleic Acids Res 14(9): 3763-72.
Story, R. M., I. T. Weber and T. A. Steitz (1992). "The structure of the E.
coli recA protein monomer and polymer." Nature 355(6358): 318-25.
Streisinger, G., Y. Okada, J. Emrich, J. Newton, A. Tsugita, E. Terzaghi and
M. Inouye (1966). "Frameshift mutations and the genetic code. This paper is
dedicated to Professor Theodosius Dobzhansky on the occasion of his 66th
birthday." Cold Spring Harb Symp Quant Biol 31: 77-84.
Taddei, F., J. A. Halliday, I. Matic and M. Radman (1997). "Genetic analysis of mutagenesis in aging Escherichia coli colonies." Mol Gen Genet
256(3): 277-81.
Tamas, I., L. Klasson, B. Canback, A. K. Naslund, A. S. Eriksson, J. J.
Wernegreen, J. P. Sandstrom, N. A. Moran and S. G. Andersson (2002). "50
million years of genomic stasis in endosymbiotic bacteria." Science
296(5577): 2376-9.
Teo, S. H. and S. P. Jackson (1997). "Identification of Saccharomyces cerevisiae DNA ligase IV: involvement in DNA double-strand break repair."
Embo J 16(15): 4788-95.
Tippin, B., P. Pham and M. F. Goodman (2004). "Error-prone replication
for better or worse." Trends Microbiol 12(6): 288-95.
Trinh, T. Q. and R. R. Sinden (1991). "Preferential DNA secondary structure mutagenesis in the lagging strand of replication in E. coli." Nature
352(6335): 544-7.
Trinh, T. Q. and R. R. Sinden (1993). "The influence of primary and secondary DNA structure in deletion and duplication between direct repeats in
Escherichia coli." Genetics 134(2): 409-22.
Trobner, W. and R. Piechocki (1984). "Selection against hypermutability in
Escherichia coli during long term evolution." Mol Gen Genet 198(1): 177-8.
Wagner, J., S. Fujii, P. Gruz, T. Nohmi and R. P. Fuchs (2000). "The beta
clamp targets DNA polymerase IV to DNA and strongly increases its processivity." EMBO Rep 1(6): 484-8.
Wagner, J. and T. Nohmi (2000). "Escherichia coli DNA polymerase IV
mutator activity: genetic requirements and mutational specificity." J Bacteriol 182(16): 4587-95.
55
Wallace, D. C. and H. J. Morowitz (1973). "Genome size and evolution."
Chromosoma 40(2): 121-6.
Wang, J. C. (1998). "Moving one DNA double helix through another by a
type II DNA topoisomerase: the story of a simple molecular machine." Q
Rev Biophys 31(2): 107-44.
Wang, J. C. (2002). "Cellular roles of DNA topoisomerases: a molecular
perspective." Nat Rev Mol Cell Biol 3(6): 430-40.
Weber, H., T. Polen, J. Heuveling, V. F. Wendisch and R. Hengge (2005).
"Genome-wide analysis of the general stress response network in Escherichia coli: sigmaS-dependent genes, promoters, and sigma factor selectivity." J Bacteriol 187(5): 1591-603.
Weller, G. R., B. Kysela, R. Roy, L. M. Tonkin, E. Scanlan, M. Della, S. K.
Devine, J. P. Day, A. Wilkinson, F. d'Adda di Fagagna, K. M. Devine, R. P.
Bowater, P. A. Jeggo, S. P. Jackson and A. J. Doherty (2002). "Identification
of a DNA nonhomologous end-joining complex in bacteria." Science
297(5587): 1686-9.
Weston-Hafer, K. and D. E. Berg (1989). "Palindromy and the location of
deletion endpoints in Escherichia coli." Genetics 121(4): 651-8.
Winfield, M. D. and E. A. Groisman (2003). "Role of nonhost environments
in the lifestyles of Salmonella and Escherichia coli." Appl Environ Microbiol 69(7): 3687-94.
Wrande, M., J. R. Roth and D. Hughes (2008). "Accumulation of mutants in
"aging" bacterial colonies is due to growth under selection, not stressinduced mutagenesis." Proc Natl Acad Sci U S A 105(33): 11863-8.
Yamamoto, Y., M. Katsuki, M. Sekiguchi and N. Otsuji (1978). "Escherichia coli gene that controls sensitivity to alkylating agents." J Bacteriol
135(1): 144-52.
Zhu, H. and S. Shuman (2005). "A primer-dependent polymerase function of
pseudomonas aeruginosa ATP-dependent DNA ligase (LigD)." J Biol Chem
280(1): 418-27.
Zhu, H. and S. Shuman (2007). "Characterization of Agrobacterium tumefaciens DNA ligases C and D." Nucleic Acids Res 35(11): 3631-45.
56
,
+,-
*(.
/' *
**
/' *
0,--
,
+0'''
')--12
--*
3-42*
)02'
*)'*
'556-
+
',--
/'
*
1789
'0#02-')*
'
*:6-
+',--
/' *
;1<
)'
(-')
1''1
'
(
)
((''(*+!!!&#%
