p1.

. ill.
6.5...

 

 

.’ 1.3.13 ,5.
FILL:

 

'1)

299?

This is to certify that the
dissertation entitled

SURVIVAL AND EXPERIMENTAL EVOLUTION OF
ESCHERICHIA COLI UNDER FREEZE-THAW STRESS

presented by

SEAN CHRISTIAN SLEIGHT

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Microbiology and Molecular
Genetics

 

 

Vac/Rafi ‘5- M

 

Major Professor’s Signature

Maj (M (M07

 

Date

MSU is an affinnative-action, equal-opportunity employer

—.-.-._.-.-.-.-.-.-._.-._.-.-._.-._._._.-._.-.—.-.-.-.-._.-

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DaleDue.indd-pi1

SURVIVAL AND EXPERIMENTAL EVOLUTION OF ESCHERICHIA COLI
UNDER FREEZE-THAW STRESS

By

Sean Christian Sleight

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Microbiology and Molecular Genetics

2007

ABSTRACT

SURVIVAL AND EXPERIMENTAL EVOLUTION OF ESCHERICHIA COLI
UNDER FREEZE-THAW STRESS

By
Sean Christian Sleight

This dissertation presents experiments on survival and evolution of
Escherichia coli under freeze-thaw (FT) stress. The initial chapter describes the
survival experiments necessary to identify conditions and suitable progenitors for
the freeze-thaw-growth (FTG) evolution experiment. The following chapters
illustrate this FTG evolution experiment as well as the phenotypic and genetic
differences of the FTG-evolved lines relative to their progenitors. The FTG-
evolved lines consisted of two groups. One group is derived from a single clone
from each of the 12 populations of the long-term evolution experiment (Lenski et
al. 1991) at the 20,000 generation timepoint. The other group is derived from
three replicates of the ancestor from this long-term evolution experiment.

Chapter 1 illustrates that the 12 long-term lines are more susceptible to
repeated FT cycles relative to the ancestor, but have no significant changes to
prolonged freezing. There is significant variation among the long-term lines in FT
survival and this variation corresponds with differences in fitness. It was
hypothesized that these 12 long-term lines would have a greater potential for
evolutionary adaptation to the FT component of the FTG cycle and thus were
chosen as progenitors for the FT G evolution experiment along with three

replicates of the ancestor.

Chapter 2 describes the FTG evolution experiment and various phenotypic
measurements of the FTG-evolved lines relative to their progenitors.
Competition experiments demonstrated evolutionary adaptation to FTG
conditions that can be attributed to both improved FT survival and a shorter
transition to growth after the FT cycle. Growth curve experiments measure the
lag phase and growth rate after a FT cycle and stationary phase.

Chapter 3 describes the candidate gene sequencing and insertion
sequence (IS) fingerprinting analysis performed to identify genetic changes in
FTG-evolved lines. Multiple FTG-evolved lines acquired mutations in the cIs
gene and uspA/B intergenic region. In selected evolved lines, these mutations
are beneficial in FTG conditions under the genomic context that they arose in.
Mutations in the 0/3 gene contribute to maintaining membrane fluidity after a FT
cycle. The IS150 insertion in the intergenic region of uspA/B disrupts uspB

transcription in FTG conditions, but the exact physiological benefit is unknown.

This dissertation is dedicated to the recovery of Dr. Thomas Whittam from his
iHness.

iv

ACKNOWLEDGEMENTS

I would like to thank my advisor Dr. Richard Lenski for his tremendous
support and advisement, allowing me to join his lab, and for obtaining funding for
my research. He gave me the independence necessary to develop my critical
thinking skills that are essential for performing scientific research, yet provided
guidance whenever I needed it that no doubt steered my research in the right
direction and improved its quality. Dr. Lenski is a top-notch scientist as
evidenced by his elegant long-term evolution experiment, critical analysis skills,
and brilliant writing. As a mentor, I could not ask for better and will always be
grateful for this opportunity.

I’d also like to thank Dr. Dominique Schneider, who took the time to advise
me via e-mail the best way to perform the insertion sequence analysis and how
to make isogenic constructs. These experiments were key to discovering the
genetic adaptations of my evolved lines and their significance. Dr. Schneider is
essentially an unofficial committee member and made wonderful suggestions
that helped guide my research.

I want to also thank all my committee members, Drs. Thomas Whittam,
Mike Thomashow, Thomas Schmidt, and John Breznak. Their insightful
questions, comments, and suggestions at committee meetings tremendously
helped shape and advance the bulk of my research, not to mention the time and
effort needed to read my dissertation proposal and this dissertation.

There are many past and present members of the Lenski lab that I want to

thank for their help in the lab, helpful discussions, friendship, and suggestions at

lab meeting presentations: Neerja Hajela, Christian Orlic, Nicholas Wigginton,
Gabe Yedid, Zachary Blount, Mike Wiser, Drs. Bob Woods, Kristina Hillesland,
Chris Borland, Christopher Marx, Susi Renold, Tim Cooper, Dusan Misevic,
Elizabeth Ostrowski, Jeff Barrick, and all of Neerja’s lab assistants over the
years. I especially want to thank Neerja Hajela for all her help in the lab and
making lab supplies needed for my research, Christian and Nick for helping me
with experiments vital to my research, Bob Woods for advising me on techniques
for making isogenic constructs and helpful discussions, Kristina Hillesland and
Chris Bor1and for reading and critiquing my dissertation proposal, and Zachary
Blount for reading a postdoctoral fellowship proposal draft and helpful
discussions. I also need to thank Jeff Landgraf at RTSF for his help with my RT-
PCR experiments and analysis, Jorge Rodrigues for his help with fluorescence
microscopy (live-dead count) experiments, the Department of Microbiology and
Molecular Genetics, and the NASA Astrobiology Institute for funding my
research.

I want to thank my entire family for their continued support and fun
conversations about my research and science in general. Last, but certainly not
least, I’d like to thank my amazing girlfriend Malia Moren. Not only has she
continued to support my graduate studies endeavor, but is always there for me
through the rough patches. And I almost forgot to thank my cat Sy for keeping

my lap warm while I finish writing this dissertation.

vi

TABLE OF CONTENTS

LIST OF TABLES ..................................................................... viii
LIST OF FIGURES .................................................................... ix
CHAPTER 1: CHANGES IN FREEZE-THAW SURVIVAL FOLLOWING
EVOLUTION IN A BENIGN ENVIRONMENT ................................... 1
Abstract .......................................................................... 1
Background ..................................................................... 2
Materials and Methods ....................................................... 5
Results and Discussion ...................................................... 9
Conclusions ..................................................................... 13
Figures ........................................................................... 16
Tables ............................................................................ 21

CHAPTER 2: ADAPTATION TO CYCLES OF FREEZING, THAWING,

AND GROWTH: PHENOTYPIC EVOLUTION ................................... 22
Abstract ........................................................................... 22
Background ...................................................................... 23
Materials and Methods ........................................................ 29
Results ............................................................................ 36
Discussion ........................................................................ 46
Figures ............................................................................ 52

CHAPTER 3: ADAPTATION TO CYCLES OF FREEZING, THAWING,

AND GROWTH: GENETIC EVOLUTION ......................................... 64
Abstract ........................................................................... 64
Background ...................................................................... 65
Materials and Methods ........................................................ 69
Results ............................................................................ 81
Discussion ........................................................................ 97
Figures ............................................................................ 108
Tables .............................................................................. 125

REFERENCES ............................................................................ 128

APPENDIX 1: Supplementary figures and tables for Chapter 2 ............... 143

APPENDIX 2: Supplementary figures and tables for Chapter 3 ............... 158

vii

LIST OF TABLES

Table 1. ANOVA testing for heterogeneity among the evolved lines

in mortality rates under the repeated freeze-thaw regime ....................... 21

Table 2. ANOVA testing for heterogeneity among the evolved lines

in mortality rates under the prolonged freezing regime ........................... 21

Table 3. IS changes in FTG-evolved lines relative to progenitors ...........
Table 4. Mutations in cls and uspA/B genes ......................................
Table 5. cls mutations in FTG-evolved lines through time .....................

Table 6. uspA/B mutations in FTG-evolved lines through time ...............

Table 7. 2-way ANOVA testing for differences in the cls allele,
progenitor or evolved background, and interaction effects in

Iaurdan anisotropy after freeze-thaw treatment ...................................

Table 8. 2-way ANOVA testing for differences in the cls allele,
progenitor or evolved background, and interaction effects in Iaurdan

anisotropy after stationary phase .....................................................

Table A1 .1. Phenotypic measurements between individual

FTG-evolved lines relative to their progenitors ....................................

Table A1.2. Statistical tests for FTG-evolved groups and their

progenitors .................................................................................

Table A2.1. Primer pairs used to PCR amplify candidate genes ............

Table A2.2. Primer pairs used to PCR amplify IS element probes for

RFLP analysis .............................................................................

Table A2.3. Primer pairs used to PCR amplify sequences adjacent to

IS elements .................................................................................

Table A2.4. IS band differences between individual FTG-evolved lines

relative to their progenitors .............................................................

Table A2.5. Laurdan anistropy differences between individual

FTG-evolved clones, their progenitors, and isogenic strains ..................

viii

. 125

. 125

126

126

127

. 127

156

157

165

166

167

. 168

. 169

LIST OF FIGURES

Figure 1. Survival of ancestral E. coli strain under the freeze-only
(prolonged freezing) and repeated freeze-thaw cycle regimes ............... 16

Figure 2. Comparison of mortality rates between the evolved E. coli
and their ancestor under the repeated freeze-thaw regime ................... 17

Figure 3. Comparison of mortality rates between the evolved E. coli
and their ancestor under the prolonged freeze-only regime .................. 18

Figure 4. Heterogeneity of mortality rates among the 12 evolved
lines under the repeated freeze-thaw regime .................................... 19

Figure 5. Correlation between freeze-thaw mortality and fitness in
the benign environment ............................................................... 20

Figure 6. Evolutionary histories of freeze-thaw-growth (FTG)
populations ............................................................................... 52

Figure 7. Schematic illustration showing how the apparent duration
of lag phase is corrected for FT survival to calculate the actual
duration of the physiological lag prior to growth ................................. 53

Figure 8. Relative fitness of FTG evolved groups A and B and their
progenitors over the two-day FTG cycle .......................................... 54

Figure 9. Relative fitness of the 15 FTG evolved lines and their
progenitors ................................................................................ 55

Figure 10. Decomposition of overall fitness gains under the FTG
regime into changes in (A) FT survival and (B) growth performance ...... 56

Figure 11. Growth dynamics of the FTG-evolved lines and their
progenitors following (A) FT treatment and (B) stationary phase at
37°C ........................................................................................ 58

Figure 12. Lag phases following the FT treatment for two
FTG-evolved lines and their progenitors .......................................... 60

Figure 13. Lag phases following stationary phase at 37°C for two
FTG-evolved lines and their progenitors .......................................... 62

Figure 14. Evolutionary histories of individual freeze-thaw-growth (FTG)
populations ................................................................................ 108

ix

Figure 15. Southern hybridization of HincII-digested genomic DNA
with the IS150 probe ..................................................................... 109

Figure 16. Schematic representation of mutations in cls ....................... 110

Figure 17. Schematic representation of IS150 insertions in uspA/B
intergenic region ........................................................................... 111

Figure 18. Relative fitness components of the following clones:
A-2 cls' construct, A-2 cls+ progenitor, A-2/FTG cls+ construct, and
A-2/FTG cls' evolved clone .............................................................. 112

Figure 19. Growth dynamics following (A) FT treatment and (B)
stationary phase at 37°C ................................................................ 115

Figure 20. Laurdan anisotropy differences following (A) FT treatment
and (B) stationary phase ................................................................. 117

Figure 21. Relative fitness components of the following clones:
ancestor (uspA/B -IS150), evolved clone AncB/FTG uspA/B +IS 150,
and the constructed clone AncB/FTG uspA/B -IS 150 ............................ 119

Figure 22. Growth dynamics following (A) FT treatment and (B)
stationary phase at 37°C ................................................................ 122

Figure 23. Relative uspB transcription differences using Real-Time
PCR ........................................................................................... 124

Figure A1 .1. Relative freeze-thaw survival of the 15 FTG-evolved lines
and their progenitors ...................................................................... 144

Figure A1 .2. Comparison of mortality rates between individual
FTG-evolved lines and their progenitors under the repeated
freeze-thaw regime ......................................................................... 145

Figure A1.3. Comparison of mortality rates between the FTG-evolved
groups and their progenitors under the repeated freeze-thaw regime ....... 146

Figure A1.4. Relative growth fitness after freezing and thawing of the
15 FT G-evolved lines and their progenitors ......................................... 147

Figure A1.5. FTG timecourse for two FTG-evolved lines and their
progenitors .................................................................................. 148

Figure A1.6. Growth timecourse after stationary phase for two
FTG-evolved lines and their progenitors ............................................. 150

Figure A1.7. Fluorescence microscopy of the Evolved Group A
(AncB/FI’ G) clone and the ancestor at different timepoints in the
FTG cycle .................................................................................. 152

Figure A1.8. Fluorescence microscopy of the Evolved Group B

(A+1/FTG) clone and its progenitor (A+1) at different timepoints in the

FTG cycle .................................................................................. 154
Figure A2.1. cls and uspA/B isogenic constructs and reconstructs ......... 159

Figure A2.2. Southern hybridization of EcoRV-digested genomic DNA
to the IS1 probe ........................................................................... 160

Figure A2.3. Southern hybridization of EcoRV ~digested genomic DNA
to the IS3 probe ........................................................................... 161

Figure A2.4. Southern hybridization of EcoRV-digested genomic DNA
to the IS186 probe ........................................................................ 162

Figure A2.5. Southern hybridization of EcoRV-digested genomic DNA
to the IS 186 probe (second blot) ....................................................... 163

Figure A2.6. Relative uspA transcription differences using Real-Time
PCR .......................................................................................... 164

xi

CHAPTER 1

CHANGES IN FREEZE-THAW SURVIVAL FOLLOWING
EVOLUTION IN A BENIGN ENVIRONMENT

BMC Evolutionary Biology 2006, 6:104

ABSTRACT

In order to study the dynamics of evolutionary change, 12 populations of
E. coli B were serially propagated for 20,000 generations in minimal glucose
medium at constant 37°C. Correlated changes in various other traits have been
previously associated with the improvement in competitive fitness in the selective
environment. This study examines whether these evolved lines changed in their
ability to tolerate the stresses of prolonged freezing and repeated freeze-thaw
cycles during adaptation to a benign environment. All 12 lines that evolved in the
benign environment for 20,000 generations are more sensitive to freeze-thaw
cycles than their ancestor. The evolved lines have an average mortality rate of
54% per daily cycle, compared to the ancestral rate of 34%. By contrast, there
was no significant difference between the evolved lines and their ancestor in
mortality during prolonged freezing. There was also some variability among the
evolved lines in susceptibility to repeated freeze-thaw cycles. Those lines that
had evolved higher competitive fitness in the minimal glucose medium at 37°C
also had higher mortality during freeze-thaw cycles. This variability was not
associated, however, with differences among lines in DNA repair functionality
and mutability. The consistency of the evolutionary declines in freeze-thaw

tolerance, the correlation between fitness in glucose medium at 37°C and

mortality during freeze-thaw cycles, and the absence of greater declines in
freeze-thaw survival among the hypermutable lines all indicate a trade-off
between performance in minimal glucose medium at 37°C and the capacity to
tolerate this stress. Analyses of the mutations that enhance fitness at 37°C may

shed light on the physiological basis of this trade-off.

BACKGROUND

Most research in evolution pursues the comparative method, in which the
present-day patterns of organismal diversity are examined in order to infer
historical processes of change (Harvey and Pagel 1991 ). Research in
paleontology allows a more direct examination of the past (Gould 1989), but
fossil data are limited in certain respects, including the inability to measure the
performance abilities of organisms. A third approach for studying evolution is to
perform long-term studies, either observational (Grant 1999) or experimental
(Bell 1997), that allow one to observe evolution in action across many
generations. In recent years, bacteria and viruses have become especially
popular for experimental evolution, owing to their rapid generations that allow
studies to run for hundreds or even thousands of generations (Lenski et al. 1991;
Bennett and Lenski 1993; Burch and Chao 2000; Elena and Lenski 2003;
MacLean et al. 2004; Wichman et al. 2005).

In a long-term experiment, Lenski and colleagues have propagated 12
populations of E. coli for mOre than 20,000 generations at 37°C in a minimal-salts
medium supplemented with glucose (Lenski et al. 1991; Lenski and Travisano

1994; Lenski 2004). The dynamics of both phenotypic (Sniegowski et al. 1997;

Lenski et al. 1998; Cooper and Lenski 2000; Rosen and Lenski 2000; Cooper et
al. 2001a; de Visser and Lenski 2002; Novak et al. 2006) and genomic (Cooper
et al. 2001 b; Cooper et al. 2003; Lenski et al. 2003; Crozat et al. 2005; Rosen et
al. 2005; Woods et al. 2006; Pelosi et al. 2006) evolution have been
characterized in a variety of ways. During the 20,000 generations, the bacteria
have genetically adapted to their selective environment, such that their mean
fitness relative to the ancestor increased by about 70%, based on direct
competitions (Cooper and Lenski 2000). Interestingly, four of the 12 populations
evolved defects in their DNA repair mechanisms, which caused them to become
hypermutable (Sniegowski et al. 1997; Cooper and Lenski 2000). The evolving
bacteria also increasingly became ecological specialists, in the sense that their
performance in some, but not all, other test environments tended to decline
(Lenski et al. 1998; Cooper and Lenski 2000; Cooper et al. 2001a; Pelosi et al.
2006). The parallel trajectories between increasing fitness in the selective
environment and declining performance in other environments suggest that most
of the decline in other environments is the result of pleiotropic side-effects of the
same mutations that produce adaptation in the selective environment (Cooper
and Lenski 2000). The fact that the four populations that became mutators do
not show much more specialization is also consistent with this interpretation
(Cooper and Lenski 2000).

Stressful environments, such as prolonged freezing or repeated freeze-
thaw cycles, may reveal other performance tradeoffs in the evolved lines. In fact,

freezing and thawing impose several interconnected stresses including

dehydration, hyperosmotic stress, ice formation, oxidative stress, and low
temperature (Cox and Heckly 1973; Calcott 1985; Gao and Critser 2000). The
acute responses of bacterial cells to freezing and thawing, including the effects of
prior exposure to cold and other stresses on survival (Thammavongs et al. 1996;
Jeffreys et al. 1998; Panoff et al. 1998; Wouters et al. 1999a; Wouters et al.
1999b; Drouin et al. 2000; Leebanon and Drake 2001), are reasonably well
understood. Many other variables also contribute to whether bacteria survive
freezing and thawing, including their nutritional status and growth phase as well
as the cooling rate employed (Calcott 1985; Gao and Critser 2000; Packer et al.
1965). Freezing and thawing E. coli cells without an exogenously supplied
cryoprotective agent, such as glycerol, severely decreases their viability (Calcott
and MacLeod 1975; Calcott and Gargett 1981). Loss of viability is proportional to
the number of freeze-thaw cycles that cells experience (Packer et al. 1965;
lngraham 1987; Calcott et al. 1983). Therefore, the elapsed time that cells are
frozen generally influences viability less than the processes of freezing and
thawing.

By contrast, much less is known about how and why different bacterial
strains and species vary in their capacity to survive these stresses (Calcott 1985;
Drouin et al. 2000; Young et al. 1964; Scher et al. 1964). In this study, we
examine how evolutionary adaptation by populations of E. coli to serial
propagation on a minimal glucose medium at a constant temperature of 37°C
affected survival during prolonged freezing and repeated freeze-thaw cycles in

the absence of cryoprotectant. In particular, we test whether there was an

evolutionary trade-off such that adaptation to this benign environment led to
correlated losses in survival capacity under these stresses. We also evaluate
whether replicate lines that evolved under the same regime show heterogeneous
changes in their stress responses. One of our motivating interests in this
research is to identify strains and conditions suitable for a future experiment that
will investigate evolutionary adaptation to repeated freeze-thaw-growth cycles.
We want to identify conditions in which some survival is possible, but where there
is sufficient mortality to impose strong selection. Also, evidence for heritable
variation among lines in survival under these extreme conditions would indicate

the potential, at least, for evolutionary adaptation in that future experiment.
MATERIALS AND METHODS

Long-term evolution experiment and bacterial strains

The long-term evolution experiment is described in detail elsewhere
(Lenski et al. 1991; Lenski 2004). In brief, 12 populations were founded using an
E. coli B ancestor and then propagated at 37°C for 20,000 generations (3,000
days) in Davis minimal medium supplemented with glucose at 25 mg/L (DM25).
The populations were diluted 100-fold daily into fresh medium, and their re-
growth allowed about 6.6 (= I092 100) generations per day. The source strain,
REL606, cannot grow on arabinose (Ara'), and it was used to start six
populations; the other six were started with a spontaneous Ara+ mutant, REL607,
of the source strain. The Ara marker is selectively neutral in the experimental

environment (Lenski et al. 1991). The 12 evolved lines used in this study were

isolated as single clones at generation 20,000, after which they have been stored

in glycerol at —80°C.

Methods for measuring survival under freeze-only and freeze-thaw regimes

Culture media and experimental pre-conditioning

Bacteria used for testing were removed from storage in the freezer,
inoculated into LB (Luria-Bertani) medium, and incubated for 24 h at 37°C. A
culture was then diluted 1:10,000 into DM25 and incubated for 24 h at 37°C. A
final 1:100 dilution was made into fresh DM25 and incubated for 24 h at 37°C, at
which time cells were well into stationary phase, having exhausted the glucose in
the first 8 h or so. Hence, all cells in the experiments were physiologically
acclimated to the same environmental conditions before their survival was

measured during repeated freeze-thaw cycles or during prolonged freezing.

Measuring cell density

Colony forming units (CFU) were used to estimate viable cell densities. A
pilot study was performed to find a range of serial dilutions that would allow
accurate cell counts as populations declined during repeated freeze-thaw cycles.
For the evolved lines that were most sensitive, no dilution was made on the final
day of the freeze-thaw experiment, so that no line fell below the detection limit of
~10 cells/mL. Diluted or undiluted cell cultures were plated on tetrazolium-
arabinose (TA) indicator agar (Lenski et al. 1991) and incubated at 37°C for 24 h

before counting.

Freeze-thaw regime

After pre-conditioning, 1 mL of a stationary-phase culture was transferred
into each of three replicate vials, which were then immediately placed in a -80°C
freezer. Each day the tubes were kept frozen for 22.5 h and allowed to thaw at
room temperature (about 22°C) for 1.5 h. Viable cell density was measured after
the desired number of freeze-thaw cycles. For example, in order to measure
survival after 1 day, the vials were frozen for 22.5 h and thawed for 1.5 h before
measuring cell density. To measure survival after 7 days, the vials were
subjected to seven freeze-thaw cycles and the cell density was measured after

the last thaw.

Freeze-only regime

After pre-conditioning, 1 mL of a stationary-phase culture was transferred
into each of three replicate vials and placed in a —80°C freezer. After the desired
duration, a vial was thawed for 1.5 h and viable cell density measured. For
example, to measure the freeze-only survival after 1 day, vials were frozen for
22.5 h and thawed for 1.5 h before measuring cell density. To measure the
freeze-only survival after 7 days, vials were frozen for 166.5 h (= 7 days minus
1.5 h) and thawed for 1.5 h. Thus, the elapsed time at -80°C is varied
experimentally, and only a single thaw occurs regardless of duration under this

regime.

Calculations and statistical methods

The first experiment was performed to measure and compare the survival
of the ancestral strain under two different 28-day regimes, one with daily freeze-

thaw cycles and the other representing the freeze-only regime. In each case,

linear regression was performed on the logo-transformed viable cell densities to
estimate the daily mortality rate and determine whether it differed significantly
from zero. Data from all 28 days, including day 0, were used to calculate the
mortality rate under the freeze-thaw regime, whereas that initial value was
excluded when calculating the mortality rate during the freeze-only regime in
order to adjust for effect of the single episode of thawing that was experienced in
all subsequent days.

The second experiment compared the survival of the ancestral and
evolved lines under the freeze-only and freeze-thaw regimes. It also allowed us
to test for heterogeneity in survival rates among the 12 independently evolved
lines. This experiment lasted 10 days; given the temporal constancy of the
mortality rates observed in the first experiment, we estimated viable cell densities
on days 0 and 10 only. For each assay, we calculated the percentage survival
as s = n10 / no, where no and mo denote initial and final densities, respectively.
For the freeze-thaw regime, we then computed the mortality rate per day as
mFT = 1 — 8“ 1°. For the freeze-only regime, we took into account that there were
nine days of sustained freezing and one freeze-thaw cycle. Therefore, the
freeze-only mortality rate was calculated as mm = 1 — (s /(1 — mFT))1/9. For
example, if the mortality rate per freeze-thaw cycle were 30%, then one would
expect 8 = 0.7 under the freeze-only regime even with perfect survival during the
other nine days. If there were also 2% daily mortality during the constant
freezing, then one would expect overall survival under the freeze-only regime to

be s = 0.7 x (0.98)9 z 0.58. Also, to preserve the statistical independence of the

replicate mpo estimates, each one was calculated using a unique paired estimate
of my.

For each mortality rate parameter, mpr or mFo, we compared the evolved
lines with the ancestral strain as follows. We first computed the mean of the
three replicate assays for each evolved line, and then we computed the grand
mean and standard deviation from the individual means of the 12 independently
evolved lines. For the ancestor, we computed the mean and standard deviation
over the six assays (three each for the two marker states). We performed a two-
tailed t-test with unequal variances, which allows for the fact that the evolved
lines may have diverged from one another as well from their common ancestor.
For the freeze-thaw and prolonged freezing mortality rates, we also tested for
heterogeneity among the evolved lines by performing a one-way analysis of
variance (ANOVA), with the three replicate assays per line providing statistical

replication.
RESULTS AND DISCUSSION

Effects of freeze-only and freeze-thaw regimes on survival of the ancestor

Figure 1 shows the survival trajectories for the E. coli B ancestral strain
under the -—80°C freeze-only and freeze-thaw regimes over the course of 28
days. With daily freeze-thaw cycles, the density of viable cells declined by about
five orders of magnitude. As evidenced by the log-linear trajectory (r2 = 0.97, p <
0.0001 ), the bacteria experienced a neany constant mortality rate of 34.4% killed

per freeze-thaw cycle.

It is quite clear that most of this mortality was caused by the repeated
bouts of freezing and thawing, as opposed to the time that cells spent frozen,
because the cumulative mortality under the freeze-only regime was far less. In
fact, over the entire 28 days at —80°C, with one thaw, the viable population size
declined by only 35.2%. This decline almost exactly matches the cell death
observed after one day in the freeze-thaw regime (Fig. 1), implying that no further
death occurred during the other 27 days at constant -—80°C. In fact, the slope of
the cell-survival trajectory from day 1 to day 28 under the freeze-only regime was
not significantly different from zero (p = 0.3483). Thus, there was little or no
mortality, even over several weeks, beyond that caused by the single freeze-
thaw cycle that was a necessary part of the survival assay procedures for all
samples, regardless of how long they had spent at —80°C.

Therefore, the ancestral strain used for the long-term evolution experiment
at 37°C is quite hardy with respect to prolonged freezing at —80°C. However, it is
much more sensitive to repeated cycles of freezing and thawing. In the next
section, we examine whether the lines that previously evolved for 20,000
generations in a benign environment became less tolerant of either prolonged

freezing or repeated freeze-thaw cycles.

Effects of freeze-only and freeze-thaw regimes on survival of the evolved

lines in comparison with the ancestor

We performed 10-day experiments under both freeze-only and freeze-
thaw regimes using the 12 lines that evolved by serial propagation on a minimal

glucose medium at constant 37°C for 20,000 generations and their ancestor.

10

Each evolved line had three replicates, while the ancestor was replicated six—fold
(three each for the Ara' and Ara" marker variants). Figure 2 compares the
average mortality rates of the evolved lines and their ancestor under the freeze-
thaw regime. The ancestor experienced a mortality rate of 34.0% per day under
the freeze-thaw regime, a value almost identical to our first experiment. By
contrast, all 12 evolved lines experienced greater mortality, with an average rate
of 53.7% per day. The difference in freeze-thaw mortality rates between
ancestral and evolved rates is highly significant (two-tailed t-test with unequal
variances, p < 0.0001).

However, we observed no significant difference between the evolved lines
and their ancestor in mortality rates during prolonged freezing at —80°C (Fig. 3;
two-tailed t-test with unequal variances, p = 0.7689). Mortality rates under this
regime were calculated, as described in the Materials and Methods, such that
they correct for the effect of one cycle of freezing and thawing. The average
mortality rates estimated for the ancestor and evolved lines under the freeze-only
regime were 5.8% and 6.6% per day, respectively. The ancestral value is
somewhat higher than estimated in the first experiment, but it is still much lower
than the mortality rates measured during repeated freeze-thaw cycles in either
experiment.

Thus, the 37°C-evolved lines as a group remained about as robust as their
ancestor to the effects of prolonged freezing at —80°C. However, the evolved
lines are much more sensitive than their ancestor to the effects of repeated

freeze-thaw cycles, with the average mortality rate increasing from 34.0% to

11

53.6% per daily cycle. In the section that follows, we examine variation among

the evolved lines in their freeze-thaw sensitivity.

Heterogeneity among the evolved lines in freeze-thaw and prolonged

freezing survival

Figure 4 shows the mortality rate per daily freeze-thaw cycle for each of
the 12 lines that independently evolved at 37°C. The 95% confidence intervals
were calculated by using the three replicate assays performed for each line.
Estimated mortality rates vary from 40.2% to 78.0% per day. An analysis of
variance confirms that variation among the evolved lines is highly significant
(Table 1, p < 0.0001). Moreover, Figure 5 shows there is a significant correlation
between competitive fitness measured in the benign selective environment
(Cooper and Lenski 2000) and mortality rates measured in the freeze-thaw
regime (r = 0.5882, n = 12, two-tailed p = 0.0442). This correlation provides
further support for the trade-off between fitness in the benign environment of
serial propagation on glucose minimal medium at constant 37°C and survival
during repeated freeze-thaw cycles. In the prolonged freezing regime, an
ANOVA shows there is also significant variation in the ability to survive prolonged
freezing among the long-term lines (Table 2, p < 0.0001). Thus, while there is
heterogeneity among these lines in both repeated freeze-thaw and prolonged
freezing survival, there is no directional trend across these lines as a group.

The four lines that evolved defects in DNA repair, and which have much
higher mutation rates (A-2, A-4, A+3, and A+6; Sniegowski et al. 1997; Cooper

and Lenski 2000), do not have consistently higher freeze-thaw mortality rates, as

12

a group, than do the other evolved lines. Also, there is no significant correlation
between competitive fitness in the benign environment and survival under the
freeze-only treatment (r = 0.0756, n = 12, two-tailed p = 0.8154), consistent with
the absence of any significant difference between the ancestor and evolved lines
as a group in this stress-related capacity.

The variation among lines that evolved in the benign selective
environment indicate that they acquired different sets of mutations that
differentially affect their sensitivity to freeze-thaw cycles. The differences
between lines are not associated with variation in DNA repair function and
mutability, which implies that pleiotropic effects of the mutations that were
selected for their beneficial effects during the evolution experiment in minimal
glucose medium at 37°C are responsible for the increased susceptibility to
freezing and thawing. If, alternatively, mutations that harmed freeze-thaw
survival had accumulated by neutral drift, then we would expect significantly
higher susceptibility among the four lines with defective DNA repair functions

(Lenski and Cooper 2000; Funchain et al. 2000).

CONCLUSIONS

E. coli B cells experience little mortality during prolonged freezing at
—80°C, even in the absence of added cryoprotectant. However, they are much
more susceptible to repeated freezing and thawing, consistent with eariier
studies (Packer et al. 1965; lngraham 1987; Calcott et al. 1983). Twelve lines
that previously evolved for 20,000 generations in a benign environment,

consisting of serial transfer in a minimal glucose medium at constant 37°C, all

13

became more susceptible to freeze-thaw mortality than was their ancestor.
Moreover, those evolved lines with higher fitness gains in the benign selective
environment also tended to have greater susceptibility to freeze-thaw cycles,
further supporting the trade-off in performance between these environments.
However, the variation among the evolved lines was not associated with
differences in DNA repair function and mutability that arose during the evolution
experiment (Sniegowski et al. 1997; Cooper and Lenski 2000). Therefore,
increased susceptibility to freeze-thaw cycles in the evolved lines probably
reflects pleiotropic effects of mutations that were beneficial to the bacteria during
evolution in the benign environment with minimal glucose medium and constant
37°C. In any case, significant variation among the lines in their freeze-thaw
survival implies that there is genetic variation for this trait, such that it can be
selected, which will be a focus of our future research.

Future research directions include evolving E. coli populations under
freeze-thaw—growth cycles. The growth phase will allow populations to recover
from mortality caused by freezing and thawing; selection should favour a
reduction in freeze-thaw mortality, a faster transition to growth after thawing, or
both. Another future direction includes identifying individual mutations
responsible for increased susceptibility to freezing and thawing in the lines
evolved In the benign environment of constant 37°C, or increased resistance in
lines evolved under the freeze-thaw-growth regime. Finally, phenotypic and
genetic analyses of freeze-thaw resistance could be extended to natural isolates

of E. coli. Comparisons between food-borne pathogens and commensals would

14

be of particular interest because the ability to survive and recover from freezing

and thawing might be an important adaptation of some food-borne pathogens.

15

FIGURES

*1
an
I-O-OO
lOl
DOI
0
O
O

o
I—O—l

;
F—O—I

Loom Viable c." Density (CFUlmL)
on
I——O—-I

N U
i—e—r

o Prolonged Freezing
0 Repeated Freeze-Thaw Cycles

‘

0 5 10 15 20 25 30
Days

Figure 1. Survival of ancestral E. coli strain under the freeze-only
(prolonged freezing) and repeated freeze-thaw cycle regimes. Each point is
the log1o-transformed viable cell density (CFU/mL) averaged over six replicates.
Closed circles show the repeated freeze-thaw regime, and open circles show the
prolonged freeze-only regime. Error bars are 95% confidence intervals and,
when not visible, are smaller than the corresponding symbol.

16

0.7 ,

l
l

0.4

 

 

0.3

WPerDay

0.2

0.1

 

 

 

 

0.0

 

 

Evolved Ancestor

Figure 2. Comparison of mortality rates between the evolved E. coli and
their ancestor under the repeated freeze-thaw regime. The height of each
bar shows the mean mortality rate per day for the evolved lines or their ancestor,
measured over ten daily freeze-thaw cycles. For the evolved bacteria, the mean
is calculated over all 12 lines, with three assays for each line. For the ancestor,
the mean is calculated over two marked variants (Ara+ and Ara'), again with three
assays for each one. See the Materials and Methods section for the mortality
rate calculation. Error bars are 95% confidence intervals based on the number of
lines (evolved) or total assays (ancestor).

17

0.16
0.14
0.12
0.10
0.08

if
g 0.03

0.04

 

 

 

0.02

 

 

 

 

 

 

0.00

 

EvolVed Ancestor
«0.02

Figure 3. Comparison of mortality rates between the evolved E. coli and
their ancestor under the prolonged freeze-only regime. The height of each
bar shows the mean mortality rate per day for the evolved lines or their ancestor,
measured over ten days at —80°C. For the evolved bacteria, the mean is
calculated over all 12 lines, with three assays for each line. For the ancestor, the
mean is calculated over two marked variants (Ara+ and Ara‘), with three assays
for each one. See the Materials and Methods section for the mortality rate
calculation, which includes an adjustment for the mortality caused by thawing in
the final day. Error bars are 95% confidence intervals based on the number of
lines (evolved) or total assays (ancestor).

18

.o .o .o
1‘ 3 2
'I'

.o
8

E
h}—c
l—-I—l

 

Mortallty Per Freeze-Thaw Cycle
' ‘2
Ir—J——o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.54 - if“ .............................................................. _ ......
0.40 . 1
m-
0.44 -
0.8 «
A-1 A-2 A-3 A4 A-5 A-6 A+1 A+2 A+3 A+4 A+5 A+6
Evolved Line

Figure 4. Heterogeneity of mortality rates among the 12 evolved lines
under the repeated freeze-thaw regime. The height of each bar shows the
mean mortality rate per day for one of the evolved lines, measured over ten daily
freeze-thaw cycles, calculated from three assays for each line. The x-axis value
(34%) is the estimated mortality rate of the ancestor; the dashed line shows the
average mortality rate for the 12 evolved lines. Error bars are 95% confidence
intervals calculated using the replicate assays for each line. See Table 1 for the
statistical analysis testing for variation among the evolved lines.

19

8

030 . r I 0.5002
n I 12 e
p I 0.0442

0.75 .

 

 

P
V
O

.0
8
e

.0
On
an

e

'8

Mortallty Per Freeze-Thaw Cycle
'8
O

'2

0.40 ' 0

035 . . . , , .
1.55 1.60 1.65 1.70 1.75 1.80 1.85 1.90

Relative Fitness in Benign Environment

Figure 5. Correlation between freeze-thaw mortality and fitness in the
benign environment. Each point shows the mortality rate per freeze-thaw cycle
measured in this study, and the relative fitness measured previously (Cooper and
Lenski 2000) in the benign environment, for one of the 12 evolved lines.

20

TABLES

 

 

 

 

Source df SS MS F p
Line 11 0.3906 0.03551 88.67 < 0.0001
Error 24 0.0096 0.00040

 

 

 

 

 

 

 

Table 1. ANOVA testing for heterogeneity among the evolved lines in
mortality rates under the repeated freeze-thaw regime. Mortality rates were
estimated with three-fold replication for each of the 12 evolved lines. The very
low p-value indicates significant heritable variation among these lines in their
survival capacity under the freeze-thaw regime.

 

 

 

 

Source df SS MS F p
Line 11 0.1912 0.01738 22.06 < 0.0001
Error 24 0.0189 0.00079

 

 

 

 

 

 

 

Table 2. ANOVA testing for heterogeneity among the evolved lines in
mortality rates under the prolonged freezing regime. Mortality rates were
estimated with three-fold replication for each of the 12 evolved lines. The very
low p-value indicates significant heritable variation among these lines in their
survival capacity under the prolonged freezing regime.

21

CHAPTER 2

ADAPTATION TO CYCLES OF FREEZING, THAWING, AND GROWTH:
PHENOTYPIC EVOLUTION

Physiological and Biochemical Zoology 2007, in press.

ABSTRACT

Fifteen populations of Escherichia coli were propagated for 150 freeze-
thaw—growth (FTG) cycles in order to study the phenotypic and genetic changes
that evolve under these stressful conditions. Here, we present the phenotypic
differences between the evolved lines and their progenitors as measured by
competition experiments as well as growth curves. Three FTG lines evolved
from an ancestral strain that was previously used to start a long-term evolution
experiment, while the other twelve FTG lines are derived from clones that had
previously evolved for 20,000 generations at constant 37°C. Competition
experiments indicate that the former FTG group improved their mean fitness
under the FTG regime by about 90% relative to their progenitor, while the latter
FTG group gained on average about 60% relative to their own progenitors.
These increases in fitness result from both improved survival during freezing and
thawing and more rapid recovery to initiate exponential growth after thawing.
This shorter lag phase is specific to recovery after freezing and thawing. Future
work will seek to identify the mutations responsible for evolutionary adaptation to
the FTG environment, and use them to explore the physiological mechanisms

that allow increased survival and more rapid recovery.

22

BACKGROUND

Most physiological and genetic studies of adaptation to stressful
environments focus on the proximate mechanisms that promote survival and
growth. By contrast, research on the evolutionary adaptation of organisms to
stressful environments examines how new stress responses evolve, or existing
stress responses are reshaped, by selection. One immediate question is
whether a population can survive the stress and persist on an evolutionary
timescale, or whether it will go extinct, and many factors influence this outcome:
the frequency of stress conditions, the conditions for growth between bouts of
stress, and so on. Assuming that a population survives, then there exists the
potential for evolutionary changes that improve survival and recovery under the
same or similar stresses in the future.

The direct study of evolutionary changes can be accomplished by
observing populations of bacteria or other organisms with suitably short
generations while they are propagated in controlled and reproducible laboratory
environments (Rose 1984; Lenski et al. 1991; Bennett et al. 1992; Bennett and
Lenski 1999; Wichman et al. 1999, 2000; Elena and Lenski 2003; Lenski 2004;
Riehle et al. 2005; Chippindale 2006; Herring et al. 2006; Leu and Murray 2006;
Schoustra et al. 2006; Zeyl 2006). Advantages of using bacteria in experimental
evolution include their rapid generations and large populations, as well the
capabilities of establishing replicate populations from the same ancestral clone
and reviving ancestral and derived cells stored at different times in an

experiment. Also, bacteria reproduce asexually, and thus stable genetic markers

23

can be used to distinguish ancestral from derived genotypes during competition
experiments to measure their relative fitness. Moreover, a wealth of genetic,
biochemical, and physiological information exists for model species, including
Escherichia coli, that can enrich and inform analyses built around evolution
experiments.

In a long-term evolution experiment, Lenski and colleagues have studied
the dynamics of phenotypic and genomic evolution in twelve initially identical
populations of E. coli while they were propagated on a glucose-supplemented
minimal medium at 37°C for more than 20,000 generations (Lenski et al. 1991;
Lenski and Travisano 1994; Cooper and Lenski 2000; Rozen and Lenski 2000;
Schneider et al. 2000; Cooper et al. 2003; Lenski et al. 2003; Pelosi et al. 2006;
Woods et al. 2006). Bennett and colleagues performed related experiments in
which bacteria derived from this long-term study were used to found several sets
of new populations that continued to evolve in the same medium but under
different thermal regimes, including constant lower and higher temperatures as
well as in temporally fluctuating environments (Bennett et al. 1992; Lenski and
Bennett 1993; Leroi et al. 1994; Travisano et al. 1995; Mongold et al. 1996, 1999;
Bennett and Lenski 1999; Cullum et al. 2001; Riehle et al. 2001, 2003, 2005).
The latter studies have shown adaptation that is often temperature specific, with
frequent but not universal tradeoffs in performance at other temperatures. There
is evidence from these studies, as well as others, to suggest that rates of
evolutionary adaptation to stressful environments may often exceed

corresponding rates in non-stressful environments (Parsons 1987; Hoffman and

24

Parsons 1993; Lenski and Bennett 1993; Bennett and Lenski 1997; Rutherford
and Lindquist 1998; Queitsch et al. 2002).

In this study, we examine the evolutionary adaptation of E. coli
populations to a regime of alternating days of freezing without added
cryoprotectant at -80°C and thawing at room temperature, followed by growth at
the benign temperature of 37°C. In preliminary work we found that repeated
freeze-thaw (FT) cycles cause mortality such that, without intervening
opportunities for growth, populations go extinct. Therefore, the opportunity for
growth every other day allows populations to recover and multiply, including the
generation of new mutations that might allow improved survival during or faster
recovery following FT cycles. Our aims here are to report the basic experimental
design, provide evidence for adaptation to this regime, and determine whether
adaptation occurs by improved survival, faster recovery, or both. Also, our study
shows how evolutionary history can influence evolutionary changes because the
populations were founded by strains that had different histories prior to this
experiment. We will conclude by outlining our future plans to identify the genetic
and physiological bases of the adaptations reported here.

Before turning to this evolution experiment, we now provide a brief
overview of bacteria in nature that experience freezing and thawing conditions,
their physiological responses to freezing and thawing, and a preliminary
experiment that we performed to identify the conditions for our evolution
experiment. Many different bacterial species live in environments subject to

freezing and thawing, including Siberian permafrost (Rivkina et al. 2000), Arctic

25

tundra (Eriksson et al. 2001), and Antarctic ponds (Mountfort et al. 2003). Some
of these bacteria live as parts of mat communities that include diatoms and green
algae as well as cyanobacteria. Metabolic activity and growth have been
reported for some permafrost bacteria at temperatures as low as —10°C
(Bakermans et al. 2003). Comparative evidence suggests that evolutionary
adaptation to these freeze-thaw environments has involved changes in cold-
shock or cold-active proteins (Cloutier et al. 1992; Hebraud and Potier1999;
Siddiqui and Cavicchioli 2006), membrane fluidity (Russell 1990; Hebraud and
Potier 1999, Ponder et al. 2005), and enzymes that inhibit or promote ice
nucleation (Muryoi et al. 2004; Walker et al. 2006). In general, however, little is
known about the time scale over which such changes have occurred, the number
of mutations needed to allow survival starting from a freeze-thaw sensitive
progenitor, whether survival during freezing or recovery upon thawing is the more
important target of selection, and other such issues that can be explored by
experimental evolution.

Freezing and thawing are multi-faceted stresses that involve not only low
temperature but also potentially lethal intracellular ice formation and dehydration
resulting from hyperosmotic shock (Mazur 1984; Gao and Crister 2000). Also,
freezing and thawing cause oxidative damage (Cox and Heckly 1973) and can
thus be mutagenic (Grecz et al. 1980; Calcott and Gargett 1981 ). In the face of
all these potentially damaging effects, many bacteria nonetheless can protect
themselves or otherwise recover under the right circumstances (Calcott 1985).

Several variables have been shown to affect whether bacteria can survive

26

freezing and thawing, including nutritional status, growth phase, and cooling rate
(Calcott 1985; Gao and Crister 2000). For example, bacteria harvested in
stationary phase survive freezing and thawing much better than those harvested
during exponential growth (Souzu 1989). Furthermore, pre-exposure to cold,
osmotic, and other stresses can increase subsequent FT survival, presumably
owing to changes in levels of particular proteins or other cell components that
provide cross-tolerance (Drouin et al. 2000; Thammavongs et al. 1996; Panoff et
al. 2000).

Various approaches could be pursued to investigate the molecular genetic
basis of FT tolerance. The traditional genetic approach screens sets of insertion
mutants to identify those that are deficient in FT survival, but not in their ability to
grow under permissive conditions, presumably owing to knock-outs of genes
whose products are specifically required for FT survival or recovery. Analyses of
changes in mRNA and protein levels might allow the identification of other genes
whose expression levels are altered during freezing, thawing, or subsequent
recovery. An evolutionary approach, such as the one we have pursued, differs
because it allows the study of improvements in the stress response, and not just
the basis of the tolerance as it presently exists in a particular strain or species.
Indeed, there may be many different types of evolutionary adaptations that could
improve FT survival and recovery, some of which might involve genes other than
those already known to be important for this stress response. By studying the
evolution of multiple experimental lines, including ones started with different

ancestral strains, one can observe the uniformity or diversity of responses to any

27

given stress. Also, the use of a mesophilic organism such as E. coli, instead of a
psychrophilic species that is already cold-adapted, may increase the opportunity
to see new evolutionary adaptations to FT conditions on an experimental
timescale. In a previous study by another group, Lactobacillus delbrueckii were
found to evolve greater FT tolerance after serial propagation in milk with
intermittent freezing and thawing (Monnet et al. 2003).

We performed a pilot experiment to isolate cryotolerant mutants of E. coli
that could better survive repeated freezing and thawing. The ancestor of the
long-term experimental populations was subjected to daily FT cycles without
added nutrients, but no viable cells remained after 40 days. Therefore, for this
study we introduced a growth phase after the FT cycle to allow populations to
recover and, potentially, to evolve. Elsewhere, we have also reported that 12
populations of E. coli that evolved in a minimal salts medium with glucose at
37°C for 20,000 generations all became more sensitive to repeated FT cycles
than their ancestor, although the extent of this change varied among the derived
lines (Sleight et al. 2006). These findings demonstrate genetic variation in FT
tolerance, and they also indicate that some of the mutations responsible for
improved fitness in that warm environment are detrimental under FT conditions.
Also, because these lineages are pre-adapted to the growth conditions at 37°C,
they are interesting progenitors to start an evolution experiment under the FTG
regime. We can compare the adaptation to the FTG regime of the 37°C-evolved

lines and their progenitors, with our a priori expectations that the 37°C-evolved

28

lines have more scope for improvement with respect to FT cycles but less

opportunity to improve further during the growth phase.

MATERIALS AND METHODS

Culture media and experimental pre-conditioning

The bacterial clones used in our experiments, including those evolved in
our experiment as well as their progenitors, are kept in long-term storage at
-80°C with glycerol added as a cryoprotectant. However, in the FTG evolution
experiment and in our assays of FT survival and relative fitness under the FT G
regime, we used medium without added glycerol in order to investigate the
evolution of changes in survival and recovery. To ensure that cells were in
comparable physiological states at the start of assays of FT survival and
competitive fitness, cells were removed from the freezer; inoculated into LB
(Luria-Bertani) medium independently for each replicate assay and incubated at
37°C for 24 h; then diluted 10,000-fold into Davis minimal medium supplemented
with glucose at 25 mg/L (DM25) and incubated at 37°C again for 24 h, at which

point these cells were used to start all assays.

Long-term evolution experiment and bacterial strains

The long-term evolution experiment at constant 37°C is described in detail
elsewhere (Lenski et al. 1991; Lenski 2004). In brief, 12 replicate populations
evolved for 20,000 generations (3,000 days) starting from two variants of the
same ancestral strain of E. coli B. One variant (REL606) cannot grow on

arabinose, while the other (REL607) is a spontaneous Ara+ mutant; six

29

populations were founded from each type. The Ara marker is neutral in the long-
term experimental environment (Lenski et al. 1991), which consists of daily 1:100
transfers into flasks that contain DM25 and incubation with shaking at 37°C. The
dilution and subsequent re-growth allow about 6.6 (= I092 100) generations per

day.
Freeze-thaw-growth evolution experiment

Figure 6 gives an overview of our freeze-thaw-growth (FTG) evolution
experiment. The 15 FTG populations include three replicates, designated group
A, founded by the Ara+ variant of the original ancestor, and twelve populations
founded by clones sampled from each of the long-term populations at generation
20,000, designated group B. These 15 populations evolved for 150 two-day FTG
cycles, which equals some 1,000 generations based on the 100-fold dilution and
re-growth in alternating days; in fact, somewhat more generations occurred
because additional cell growth also offset death during the freeze-thaw cycle in
alternating days. To start the evolution experiment, 1 mL of stationary-phase
DM25 culture was transferred into a freezer tube and put in a -80°C freezer for
22.5 h. The tubes were then thawed at room temperature (~22°C) for 1.5 h, after
which time the contents were diluted 1:100 into fresh DM25 and incubated at
37°C without shaking for 24 h. The use of unshaken tubes, rather than shaken
flasks, during the growth phase represents a small departure from the methods
used in the long-term evolution experiment. This change allowed us to handle
more cultures simultaneously. Any effect on oxygen levels is small, owing to the

low sugar concentration used in our experiments and the resulting low population

30

density (about 3-5 x 107 cells per mL in stationary phase, roughly two orders of
magnitude below the density in most research with E. COII). In a previous study,
there was no discernible difference in relative fitness between shaken flasks and
unshaken tubes under conditions similar to those used in our study (Travisano
1997)

Thus, the populations in our evolution experiment experienced cycles of a
day of freezing and thawing that alternated with a day in which they grew in the
same medium and at the same temperature as in the long-term evolution
experiment. Populations and clones were sampled and stored every 100
generations (30 days) in freezer vials with glycerol for future study. During the
long-term evolution experiment at 37°C, unique mutations were substituted at
certain loci in each population (Woods et al. 2006); we have confirmed by
sequencing the presence of at least one such mutation in each group B line,
which demonstrates they were all derived from their intended progenitors (S. C.

Sleight and R. E. Lenski, unpublished data).

FTG competition experiments

We performed competitions to measure the extent of evolutionary
adaptation of the FTG-evolved lines relative to their progenitors. Competitions
were performed in the exact same environment as used in the FTG evolution
experiment. For group A, the FTG-evolved lines competed directly against the
ancestor with the opposite Ara-marker state. For group B, the FTG-evolved lines
and their 37°C-evolved progenitors each competed separately against the

original ancestor bearing the opposite marker state.

31

The densities of two competitors were estimated by serially diluting
cultures on tetrazolium-arabinose (TA) indicator plates, except for one FTG
evolved line and its progenitor that form poor colonies on TA medium and whose
densities were estimated from minimal glucose plates. Plates were incubated for
24 h at 37°C, and the resulting number of colony forming units was used to
estimate the corresponding density. Ara+ and Ara‘ cells form white and red
colonies, respectively, when grown on TA plates. As a control for any effect of
the Ara marker, the Ara+ and Ara’ ancestral variants also competed in each
batch of competition experiments; the Ara marker is neutral in the FTG regime,
as was previously shown under the growth conditions of the long-term
experiment.

Prior to starting a competition, each competitor was separately
conditioned as described previously. Equal volumes of each competitor were
then mixed, and 1 mL was transferred into a tube and placed in a —80°C freezer.
After 22.5 h, the tube was thawed at room temperature for 1.5 h. Then 0.1 mL
was diluted into 9.9 mL of DM25 and incubated at 37°C for 24 h. Measurements
of the densities of the competitors were taken at the start (before freezing), after
thawing, and again after the growth phase that completes the full FTG cycle.
From these data, we estimated the relative fitness of the two competitors over
the course of the complete FTG cycle by using the initial and final samples, as
described below. In addition, we partitioned the overall FTG fitness into two
components that reflect FT survival and subsequent growth by using the middle

sample, also as described below.

32

the single ancestor of group A where the mean was based on the replicate

assays. Details of the statistical analyses are given below.

Growth curves

We obtained separate growth curves, based on optical density (OD), for
the evolved clones and their progenitors in order to estimate the durations of their
lag phase prior to commencing growth as well as their doubling times during the
growth phase. These growth parameters were first calculated for each line by
averaging nine replicate growth trajectories for that line. Group means were then
calculated by averaging the estimates for each line in the group, except for the
single ancestor of group A whose parameters were estimated directly from the
nine replicates. OD measurements were made at the 420-nm wavelength on 0.2
mL cultures in 96-well microtiter plates incubated at 37°C. The data were
collected following either a FT cycle or stationary phase at 37°C, with a
comparison between the two curves permitting an estimate of the effect of the FT
cycle on the time required to achieve exponential growth following dilution into

fresh DM25 medium, as described below.

Calculations of lag phase duration and doubling time

Calculations of the duration of lag phase used the general method given
by Lenski et al. (1994) with modifications as described here. OD values from
growth curve experiments were standardized by dividing by the initial OD
(measured immediately after the FT cycle before dilution, and divided by 100 to
take the dilution factor into account) and Iogz-transformed in order to express

changes as doublings. The transformed data were then plotted and inspected to

34

 

identify the window of exponential growth, and linear regression was performed
on those exponential-phase data. The doubling time was calculated as the
inverse of the slope of the regression. The “apparent” lag phase was then
estimated by extrapolating the exponential growth back in time until the
regression intersected the initial OD measurement (Fig. 7). Note, however, that
this apparent lag phase does not take into account any contribution of dead cells
to OD, and so it over-estimates the actual duration of the lag phase. We
corrected for this mortality by independently measuring the FT survival of the
clone, then transforming the proportion surviving to a logz scale; that proportion is
generally below one, and hence the value is negative on a log scale, with more
negative values corresponding to higher mortality. As shown schematically in
Figure 7, the logz-transformed survival value was plotted as a horizontal line on
the graphs showing the exponential-phase growth data, where the intersection of
this line with the extrapolated regression then provides an estimate of the actual

duration of the physiological lag (i.e., the duration corrected for FT mortality).

Statistical methods

We performed t-tests to compare performance measures between the
FTG evolved lines and their progenitors. Paired tests were always used when
comparing B group evolved lines with their progenitors, owing to their unique
relationships. Paired tests were also used to compare the A group evolved lines
with the common ancestor when the relevant assays for each evolved line were
paired with particular assays for the ancestor. Othenrvise, we used a t-test for

comparing a single specimen, the ancestor, with a sample, the three evolved

35

identify the window of exponential growth, and linear regression was performed
on those exponential-phase data. The doubling time was calculated as the
inverse of the slope of the regression. The “apparent” lag phase was then
estimated by extrapolating the exponential growth back in time until the
regression intersected the initial OD measurement (Fig. 7). Note, however, that
this apparent lag phase does not take into account any contribution of dead cells
to OD, and so it over-estimates the actual duration of the lag phase. We
corrected for this mortality by independently measuring the FT survival of the
clone, then transforming the proportion surviving to a I092 scale; that proportion is
generally below one, and hence the value is negative on a log scale, with more
negative values corresponding to higher mortality. As shown schematically in
Figure 7, the Iog2-transformed survival value was plotted as a horizontal line on
the graphs showing the exponential-phase growth data, where the intersection of
this line with the extrapolated regression then provides an estimate of the actual

duration of the physiological lag (i.e., the duration corrected for FT mortality).

Statistical methods

We performed t-tests to compare performance measures between the
FTG evolved lines and their progenitors. Paired tests were always used when
comparing B group evolved lines with their progenitors, owing to their unique
relationships. Paired tests were also used to compare the A group evolved lines
with the common ancestor when the relevant assays for each evolved line were
paired with particular assays for the ancestor. Othenrvise, we used a t-test for

comparing a single specimen, the ancestor, with a sample, the three evolved

35

group A lines (Sokal and Rohlf 1981, pp. 229-231). This test assumes equal
standard deviations for the distributions from which the single specimen and the
sample are drawn; in fact, the ancestor, being a homogenous type, should have
a lower standard deviation than the independently derived lines, which makes
our inferences conservative. We used one-tailed tests for those hypotheses with
clear a priori expectations about the direction of the outcome; othenrvise, two-
tailed tests were used conservatively, including those cases with potentially
opposing effects. For example, we expect overall fitness and its components to
increase for all FTG evolved lines when measured in the FTG environment. We
also expect the lines in group B, which previously evolved in the same medium at
37°C and suffered correlated losses in FT survival, to show less improvement in
growth rate, but greater gains in FT survival, than the lines in group A.
Supporting data and supplementary analyses

Appendix 1, which can be found at the back of this dissertation, provides
some supporting data and supplementary analyses relevant to this Chapter. All
information pertaining to the Materials and Methods is included in the figure
legend if not already described.

RESULTS

Evolutionary adaptation to the FTG regime

Figure 8 shows that the lines evolved under the FTG regime improved in
their overall fitness when measured under that same regime. The three lines in
group A increased their fitness, on average, by 89% relative to the ancestral

strain. The twelve lines in group B improved relative to their 37°C-evolved

36

progenitors by 60%, on average. Notice that the progenitors of group B were
themselves much more fit than the common ancestor under the FTG regime; this
result indicates that their improvements in growth during 20,000 generations at
37°C more than offset their losses in FT survival. The fitness gains in groups A
and B relative to their own progenitors were both significant (one-tailed paired t-
tests: group A, 2 df, p = 0.0034; group B, 11 df, p < 0.0001). Also, the
proportional gains in group A, relative to their ancestor, were significantly greater
than those observed in group B, relative to their progenitors, during the FTG
evolution (two-tailed t-test with unequal variances, 13 df, p = 0.0417), although B
had a final mean fitness that was significantly higher than A (two-tailed t-test with
unequal variances, 13 df, p = 0.0003).

Figure 9 illustrates the extent of variation among the 15 evolved lines and
their progenitors in overall fitness under the FTG regime. The three group A
lines improved to a similar extent, and their fitness gains were all highly
significant (p < 0.0001, based on one-tailed t-tests comparing six replicate
assays for an evolved line and two assays for the ancestor that were paired with
each evolved line). The group B lines showed much more variability, reflecting
differences in their progenitors’ fitness as well as their subsequent gains. Eleven
of the B lines improved significantly (all p < 0.05, based on one-tailed t-tests
comparing six replicate assays for each evolved line with six replicate assays for
its own progenitor), while the gain in line A+5 was not significant. The A+5
progenitor had the highest FT survival of all group B progenitors; therefore, it had

less scope to improve in this respect.

37

The fitness values for the B progenitors, which previously evolved at
constant 37°C for 20,000 generations, vary by about two-fold under the FTG
regime (Figure 9), whereas they vary much less, although significantly, in the
benign environment where they evolved (Lenski et al. 1991; Lenski and
Travisano 1994; Cooper and Lenski 2000). There is no significant correlation
between the FTG fitness values measured here and the fitness values measured
previously, on the same generational samples (Cooper and Lenski 2000), in that
benign environment (r = —0.1673, 10 df, two-tailed p = 0.6033). Although the
medium per se is the same, the ancestral and FTG environments are so
dissimilar overall that performance in one does not predict performance in the
other. However, we recently showed that, while all the B progenitors have
greater sensitivity to FT mortality than the common ancestor, the extent of their
increased sensitivity was significantly correlated with their fitness gains in the
benign environment (Sleight et al. 2006). Thus, prior adaptation by the B
progenitors reduced the scope for adaptation to the growth period of the FTG
regime, while increasing potential adaptation to the FT treatment.

The overall adaptation to the FT G regime can be partitioned into
improvements in FT survival and subsequent growth performance. Figure 10A
shows the changes in FT survival (see also supplementary information in
Appendix 1, Figure A1 .1, for changes in FT survival between individual FTG
evolved lines and their progenitors). Both group A and group B lines survived
freezing and thawing better than their progenitors (one-tailed paired t-tests: group

A, 2 df, p = 0.0014; group B, 11 df, p = 0.0023). The improvement in sunrival

38

was significantly greater in group B than in group A (one-tailed t-test with unequal
variances, 13 df, p = 0.0432), as expected given the higher initial mortality in B
progenitors. The resulting final survival values did not differ significantly between
groups A and B (two-tailed t-test with unequal variances, 13 cit, p = 0.3691).

As an aside, repeated FT survival to 10 FT cycles was also measured
between all FTG evolved lines and their progenitors (see Appendix 1, Figure
A1.2). Note that repeated FT survival is shown on a different scale as that in
Figure 10A, which is only survival after one FT cycle measured in competition
with the ancestor, to compare against previous repeated FT survival
measurements (Figures 2 and 4). Figure A13 in Appendix 1 shows the mean
repeated FT survival differences between the evolved groups and their
progenitors. On average, Evolved Group A experienced 6% less mortality per FT
cycle (one-tailed t-test p = 0.0652) and Evolved Group B improved by 14% in this
respect (one-tailed t-test p > 0.0001). Repeated FT survival of the ancestor is
about the same as Evolved Group B on average. Thus, the 12 long-term lines
that evolved for 20,000 generations in a benign environment overcame their
repeated FT sensitivity relative to the ancestor (Sleight et al. 2006) by evolution
in a FTG regime.

Figure 108 shows the corresponding changes in growth performance after
a FT cycle and dilution into fresh medium (see Appendix 1, Figure A14, for
changes in growth after FT measurements between individual FTG evolved lines
and their progenitors). Both groups undenrvent significant improvement (one-

tailed paired t-tests: group A, 2 (If, p = 0.0042; group B, 11 df, p < 0.0001). The

39

proportional gain in growth performance was greater for group A than for group B
(one-tailed t-test with unequal variances, 13 df, p = 0.0056), as expected given
the higher initial level for the group B progenitors. However, the final growth
performance was still higher in B than in A (two-tailed t-test with unequal
variances, 13 df, p = 0.0001).

The fact that group B lines showed such large improvements in their
growth performance in the second day of the FTG cycle, despite their progenitors
having evolved at 37°C for 20,000 generations, may be surprising at first glance.
However, when we performed one-day competitions in the same medium,
without a FT cycle, the evolved group B lines did not show any improvement in
growth performance (S. C. Sleight and R. E. Lenski, unpublished data). As we
will show in the next section, their improvement comes not from faster rates of
exponential growth, but instead they evolved faster recovery of their growth

capacity following the FT treatment.

Changes in growth dynamics including the duration of lag phase

Figure 11 shows the growth dynamics, based on optical density, for both
FTG-evolved groups and their progenitors under two slightly different conditions.
In Figure 11A, the growth data correspond precisely to day 2 of the FTG regime,
with cells having been frozen and thawed prior to their dilution into fresh medium
at 37°C. In Figure 118, the populations were started from stationary-phase
cultures grown at 37°C without the intervening FT cycle.

Several interesting features are apparent from the growth dynamics

following the FT cycle (Fig. 11A). First, notice that both evolved groups (grey

40

curves) increase much sooner than their respective progenitors (black curves).
Second, the group B evolved lines (solid grey curve) perform better, on average,
than the group A lines (dashed grey curve), although the difference is much less
than between their progenitors (solid black and dashed black curves,
respectively). Third, the improvement in group A (dashed grey curve) relative to
its ancestor (dashed black curve) appears to involve both earlier and faster
growth.

Now compare these dynamics with those observed without the intervening
FT treatment (Fig. 11B). First, notice that all four growth trajectories now rise
earlier than before. These differences imply that some portion of the lags after
the FT treatment reflect demographic recovery, physiological recovery, or both
from that stress. Second, the trajectories for the group B lines and their
progenitors (solid grey and solid black curves, respectively) are almost perfectly
superimposed. This similarity, in contrast to the difference seen after the FT
treatment, implies that the group B lines have adapted by accelerating their
recovery from the stress. Third, the trajectories for group A and the ancestor
(dashed grey and dashed black curves, respectively) are much closer to than
they were following the FT treatment. The remaining difference between them
appears to reflect faster growth by the evolved A lines, rather than a shorter lag
phase. After quantifying these differences, we will turn in the next section to
more intensive experiments to disentangle the roles of demographic and

physiological recovery from the FT treatment.

41

For each FTG-evolved line and its progenitor, we calculated its doubling
time during exponential-phase growth and the “apparent” lag duration (without
adjusting for FT mortality), as described in the Materials and Methods section.
Considering first the growth trajectories after the FT treatment (Fig. 11A), group
A lines evolved much faster exponential growth rates, with their mean doubling
time almost 13 minutes shorter than the ancestor (one-tailed p = 0.0075 using a
t-test for comparing a single specimen, the ancestor, with the three evolved
group A lines). The group A lines also dramatically shortened the mean duration
of the apparent lag phase by 153 minutes relative to their ancestor (one-tailed p
= 0.0066, same test as above). By contrast, the group B lines, on average, did
not evolve faster exponential growth than their progenitors, with the trend slightly
in the opposite direction (p > 0.5). However, the B lines also evolved a much
faster transition into exponential growth, with the average duration of the
apparent lag phased reduced by about 105 minutes (one-tailed paired t-test, 11
df, p = 0.0001). 1

Without the preceding FT cycle (Fig. 118), group A lines again showed
faster exponential growth rates relative to their progenitors, with the average
reduction in doubling time estimated to be about 9 minutes from these data (one-
tailed p = 0.0058 using a t-test for comparing a single specimen with a group).
However, the A lines did not evolve any shorter lag phase under this treatment,
with the observed trend in the opposite direction (p > 0.5). Relative to their
immediate progenitors, the group B lines did not improve, on average, in either

their doubling time (one-tailed paired t-test, 11 df, p = 0.3007) or the duration of

42

the lag phase prior to growth, which trended very slightly in the opposing

direction (p > 0.5).

Demographic and physiological contributions to shorter lag phases

The data in the previous section clearly show that the FTG-evolved
bacteria evolved much shorter apparent lag phases, and that these changes
were specific to the FT treatment. However, these reductions in apparent lag
phase do not distinguish between reduced cell death, such that population
growth recommenced from a higher level after the FT treatment, and faster
physiological recovery of surviving cells. To distinguish between these two
effects, one must adjust the apparent lag phase by taking into account mortality
during the FT treatment. Recall Figure 7, which shows schematically how the
actual duration of the physiological lag phase is corrected for this FT mortality.
Whereas the apparent lag is calculated by extrapolating exponential growth back
to the initial optical density, which includes both living and dead cells, the actual
lag is calculated by extrapolating back to the density of surviving cells only.

To address this issue, we performed more intensive experiments,
including growth trajectories and FT survival assays, for single representatives of
each evolved group and their progenitors. For group A, we used the middle of
the three evolved lines and the common ancestor; for group B, we used the
evolved line designated as A+1 and its progenitor (Fig. 9). Figure 12A shows the
estimation of the duration of the physiological lag for that group A evolved line
and its ancestor. The ancestor (black symbols) had an apparent lag of 324

minutes, while the apparent lag for the evolved line (grey symbols) was only 155

43

minutes, with the difference being 169 minutes. The physiological lags, adjusted
for cell mortality, were 282 minutes for the ancestor and 135 minutes for the
evolved line. Thus, the physiological lag was reduced by 147 minutes during the
evolution of this FTG-adapted line. The other 22 minutes of the difference in the
duration of the apparent lag reflects improved survival during the FT treatment.

Figure 12B shows the same analysis for the group B evolved line and its
own immediate progenitor, which had evolved previously at 37°C for 20,000
generations. The apparent lag for this evolved line was 76 minutes, while that of
its progenitor was 192 minutes, with the duration reduced during FTG-evolution
by 116 minutes. After adjusting for mortality during freezing and thawing, the
physiological lags estimated for the evolved line and its progenitor were 56 and
148 minutes, respectively, indicating that evolution had reduced the physiological
lag by about 92 minutes in this group B line.

Figure 13 shows the calculation of lag durations for the same evolved
lines and their progenitors, except in this case without a FT treatment prior to
growth. There were no adjustments for death, which does not occur at any
measurable rate under these conditions (Vasi et al. 1994). For both evolved
lines and their progenitors, lag-phase durations were shorter than those seen
after a FT treatment. Moreover, the differences between the FTG lines and their
progenitors were much smaller and, in the group A case, the order was reversed.
Without a FT treatment, the group A evolved line had a lag duration of 84
minutes, whereas that of the ancestor was only 52 minutes (Fig. 13A). Notice,

too, that this group A line evolved faster exponential-phase growth, consistent

44

with the shorter doubling times reported earlier for its group. The group B line
showed no improvement in exponential growth rate, as expected given its
progenitor’s prior evolution in the same medium and again consistent with earlier
results (Fig. 138). This group B progenitor experienced a lag of 34 minutes
without a FT treatment, while its FTG-evolved derivative had a lag of 6 minutes.
The resulting difference of 28 minutes is much less than the 92-minute difference
in lag duration observed for the same B line and its progenitor after the FT
treatment (Fig. 128).

These experiments demonstrate that greater survival during the FT
treatment and faster physiological recovery of growth capacity both contributed to
the improved fitness of lines that evolved under the FTG regime. We obtained
additional evidence for the evolution of faster physiological recovery in two
subsidiary experiments. First, we obtained some population-growth trajectories
by plating and counting cells, rather than by combining measurements of optical
densities and cell survival (see Appendix 1, Figures A1 .5-6). These experiments
showed the faster recovery of growth in the same two evolved lines relative to
their progenitors. Second, we performed fluorescence microscopy at multiple
times after a FT cycle in order to visualize actively dividing cells (see Appendix 1,
Figures A1 .7-8). These observations revealed that cell division began much
earlier in the FTG-evolved lines than in their progenitors, consistent with the
evolution of faster physiological recovery of growth capacity following this

stressful treatment.

45

DISCUSSION
We performed an experiment to investigate the evolutionary adaptation of 15 E.
coli populations to 150 two—day cycles of freezing, thawing, and growth. No
glycerol or other cryoprotective compound was added prior to freezing, unlike the
standard method used to store bacteria. The entire experiment encompassed
over 1,000 generations. Twelve of these freeze-thaw-growth (FTG) lines were
founded by progenitors that had previously evolved for 20,000 generations in the
same medium and at the same growth temperature of 37°C, except without the
freeze-thaw (FT) treatment between reaching stationary phase and dilution in
fresh medium (Fig. 6). The 37°C-evolved progenitors had higher growth rates
than did the ancestor of the other three FTG lines, but these 37°C-evolved
progenitors also suffered greater mortality during the FT treatment.

Almost all of the evolved lines showed large fitness gains in the FTG
regime (Figs. 8 and 9). We identified three distinct fitness components that
contributed to the improvement in overall performance. First, both groups of the
evolved lines exhibited substantial improvement in FT survival, although the
magnitude of improvement was greater for the group B lines, whose progenitors
had evolved in the same medium at constant 37°C (Fig. 10A). During evolution
in that benign environment, the progenitors became more sensitive to FT
mortality (Sleight et al. 2006); hence, there was greater opportunity for better
survival, which indeed occurred. However, these larger proportional gains in FT
survival in Group B were not inevitable. The same mutations giving a fitness

improvement in minimal glucose at 37°C could very well have not allowed for

46

compensatory mutations to increased FT survival. Second, the group A lines,
whose progenitors had not experienced the same medium, evolved faster
exponential-phase growth (Figs. 11-13). By contrast, the group B lines showed
no measurable gains in their exponential growth (Figs. 11-13). Evidently, the B
progenitors had largely exhausted the potential for faster exponential growth rate
during their 20,000-generation history in the same culture medium.

Third, and perhaps most interesting, both groups showed striking
improvement in the speed of their physiological recovery from FT stress. To
demonstrate and quantify this third component of adaptation, we had to
disentangle two confounding effects. One such factor is mortality during the FT
cycle, which contributes to delayed recovery at the population level, but does not
bear directly on the speed of recovery by surviving cells. We addressed this
issue by independently measuring FT mortality, and using the data to calculate
the physiological lag (Figs. 7, 12, and 13). The second complication arises
because physiological lags occur following starvation, even without the FT
treatment, upon dilution into fresh medium. Thus, one must distinguish between
recovery from starvation and from freezing and thawing, which we did by
measuring the duration of physiological lags following starvation with and without
the intervening FT treatment (Figs. 11-13). Both groups of FTG-evolved lines
showed much faster recovery of their surviving cells, with growth commencing an
hour or more sooner than it did in their progenitors. This accelerated recovery
provides a substantial head-start during the subsequent exponential growth. We

also performed fluorescence microscopy on populations while they were

47

recovering from the FT treatment; these observations confirmed that many more
viable cells were in the process of doubling for the FTG-evolved lines than for
their progenitors during the early recovery period.

In addition to the differences in adaptation between the two groups of
evolved lines, substantial variation also exists among lines in the same group,
especially in the B group. The differences in adaptation between groups A and B
reflect differences in their selective histories, but this explanation cannot explain
the striking diversity among the B lines (Fig. 9), which shared the same original
ancestor and the same two-stage history (Fig. 6). Evidently, the 12 lines in the B
group have, by chance, accumulated different sets of mutations that cause the
differences in their performance. At present, we do not know the identity of all
those mutations. However, we can ask whether the variation among them
reflects, at least in part, heterogeneity that was already present in their
progenitors that diverged for 20,000 generations at constant 37°C. In fact, there
is a significant positive correlation between the FTG fitness levels of the evolved
B lines and their progenitors (r = 0.6769, 10 df, one-tailed p = 0.0079), indicating
that some of the differential success among the diverse B lines reflects variation
in the extent of pre-adaptation among their progenitors. However, it is also clear
that most or all the B lines undenrvent substantial evolutionary adaptation during
the FTG experiment (Figs. 8 and 9). With only one derived line from each
progenitor, we cannot explicitly test whether the different B progenitors
predisposed different final states (cf. Travisano et al. 1995). However, we might

still examine this issue, in the future, by finding the mutations that contribute to

48

 

differences among the progenitors, finding other mutations responsible for
adaptation during the FTG evolution experiment, and then systematically
recombining these mutations from different lineages to ask whether the beneficial
effects of the later mutations are conditional on or, alternatively, independent of
the earlier mutations.

Many more experiments remain to be done, of course, and so we will
close by discussing briefly our on-going research and future directions. In this
study, we have demonstrated evolutionary adaptation to the FT G regime and we
have partitioned the improvement into three distinct fitness components -
survival, recovery, and growth — but we have not yet identified the physiological,
biochemical, and genetic bases of the improvements in organismal performance.
One can imagine approaching this general problem by beginning at a
physiological scale and working down toward the genetic level, or vice versa.
The approach that we are undertaking begins at the genetic level, which reflects
our own expertise as well as the power of molecular genetic analyses in bacteria.
Once we have found some of the mutations responsible for the improved
performance of the FTG-evolved lines, we can then explore the resulting
biochemical and physiological mechanisms. By having gene identities in hand,
we can then target future studies to particular pathways based on the wealth of
information that links the genetics, biochemistry and physiology of E. coli (Bock
et al. 2006).

There are several different genetic approaches that one could pursue to

find mutations substituted in experimental populations. These approaches

49

include DNA fingerprinting methods to find genes with new insertions or deletions
(Papadopoulos et al. 1999; Schneider et al. 2000; Riehle et al. 2001);
sequencing candidate genes based on specific phenotypic changes or other prior
information (Notley-McRobb and Ferenci 2000; Cooper et al. 2001b; Crozat et al.
2005; Maharjan et al. 2006; Woods et al. 2006), analyzing changes in gene-
expression profiles to suggest additional candidates (Cooper et al. 2003; Riehle
et al. 2003; Pelosi et al. 2006); and even whole-genome sequencing (Shendure
et al. 2005; Velicer et al. 2006; Herring et al. 2006).

Finding mutations, although certainly not trivial, is becoming much easier
and less costly. But finding mutations is also only a first step in the genetic
analysis. An extremely important subsequent step is to manipulate the affected
gene, for example by constructing isogenic strains that differ by a single mutation
(Cooper et al. 2001b, 2003; Crozat et al. 2005; Pelosi et al. 2006). Such work is
possible using powerful molecular genetic approaches available in this system,
but it is also painstaking and challenging work. Once such strains have been
constructed, they can be competed to test whether a particular mutation did, in
fact, contribute to the observed evolutionary adaptation or, alternatively, merely
hitchhiked and was inconsequential to any gains in performance. Of course, the
same strains can also be used to examine changes in biochemical and
physiological traits, and the connections between genotype, phenotype,
performance, and fitness can thus be elucidated.

There are many candidate biochemical and physiological pathways by

which the FTG-evolved lines might have improved their FT survival as well as

50

their subsequent recovery and transition to growth. Some possibilities are as
follows: altered induction of stress responses, including molecular chaperones;
better repair of damage to DNA or other cell components; and regulatory or
structural changes affecting membrane fluidity, osmosis, DNA supercoiling, and
ribosomes or other components of the translational machinery. We are
proceeding by a two-pronged approach. First, we have chosen a set of six
candidate genes of interest given their known involvement in some of the
relevant pathways, and these genes are being sequenced in the FTG-evolved
lines and their progenitors. Second, we are performing a genomic-fingerprinting
approach to look for insertions and deletions caused by the movement of IS
insertion-sequence elements that are native to E. coli genomes (Papadopoulos
et al. 1999; Schneider et al. 2000, 2002; Cooper et al. 2001b).

Without giving away the results of future papers, we have found two genes
in which many FTG-evolved lines have substituted parallel mutations. Such
parallelism is a hallmark of adaptive evolution (Wichman et al. 1999; Cooper et
al. 2003; Woods et al. 2006). These genes and their products are in two different
physiological pathways, one probably affecting membrane fluidity and the other
encoding a stress response. We are presently attempting to construct isogenic
strains that differ only by mutations in these genes. We will then use these
strains to analyze the effects of the mutations on the relevant pathways, as well
as on the demographic and physiological components of survival and recovery

that produce the evolutionary adaptation to the FTG regime.

51

FIGURES

37°C Evolved

Ancestor Progenitors

   

37°C Growth for 20,000 generations

FT G for FTG for
1,000 generations 1,000 generations

II

   

Figure 6. Evolutionary histories of freeze-thaw-growth (FTG) populations.
Three lines in FTG-evolved group A were founded by the ancestor of another
evolution experiment. Twelve lines in FTG-evolved group B were founded by
clones sampled from populations that previously evolved for 20,000 generations
at 37°C. All 15 of the FTG lines evolved for 1,000 generations under the FTG
regime, with alternating days of a FT cycle and growth at 37°C. The culture
medium used for growth days was the same one used during the evolution of the
progenitors of FTG group B.

52

  
 

log2 (OD/Initial OD)
N

1 Apparent lag

Actual lag

  

240 300 360 420

Time (minutes)

Figure 7. Schematic illustration showing how the apparent duration of lag
phase is corrected for FT survival to calculate the actual duration of the
physiological lag prior to growth. The apparent duration is calculated by
extrapolating the regression line based on exponential growth back to the initial
optical density. In this hypothetical example, 50% of the cells survived the FT
treatment; therefore, the initial density of viable cells is half that implied by the
initial optical density @092 0.5 = —1). The actual duration of the physiological lag
is then obtained by extending the extrapolated growth trajectory until it intersects
the initial density of viable cells.

53

4.0

3.5

 

 

 

Ancestor Group A Progenitors Group B

 

0.5

Figure 8. Relative fitness of FTG-evolved groups A and B and their
progenitors over the two-day FTG cycle. For group A, each of the three
evolved lines competed against the ancestor with the opposite Ara-marker state.
For group B, each of the twelve evolved lines and their progenitors competed
separately against the original ancestor with the opposite marker state. The
mean value for each evolved group (grey) is adjacent to the corresponding mean
for its progenitors (white). Means for each group were calculated from the
average values for each member in the group, except for the single ancestor of
group A where the mean is based on six independent measurements. Error bars
are 95% confidence intervals based on the number of independently evolved
lines, except for the ancestor, for which the interval is based on the six replicate
assays.

54

Relative FTG Fitness

N

 

 

‘9 £9"$v‘v”é’¢v°’v9

K!

Figure 9. Relative fitness of the 15 FTG-evolved lines and their progenitors.
Fitness values were measured in competition with the common ancestor over the
two-day FTG cycle. Each evolved line (grey) is paired with its progenitor (white).
The three group A lines are shown with different ancestral replicates (designated
A, B, and C). The twelve group B lines were derived from long-term lines,
designated A+1 to A-6, that previously evolved for 20,000 generations at 37°C.
Error bars show 95% confidence intervals based on six replicate fitness assays
for each line or progenitor, except for the group A ancestor, where two of the
replicate assays were paired with each of the three evolved lines in that group.

JJJI
f ,. 2. .u

55

Figure 10.
(A)

2.2

2.0 ‘

 

0.0

 

0.5

(B)

2.8

Ancestor Group A

2.0

MGMAMFT
t:

 

I:

 

Progenitors

 

 

Ancestor Group A

 

56

Progenitor:

 

Group B

 

Group B

Figure 10. Decomposition of overall fitness gains under the FT G regime
into changes in (A) FT survival and (3) growth performance. All values were
measured with, and are shown relative to, a marked variant of the common
ancestor. FT survival reflects changes in viable cell density during the first day of
FTG competitions. Growth performance reflects the net rate of increase on the
second day after dilution into fresh medium. The mean for each evolved group
(grey) is adjacent to the corresponding mean for its progenitors (white). Means
for each group were calculated from average values for each member in the
group, except for the single ancestor of group A where the mean is based on six
independent measurements. Error bars are 95% confidence intervals based on
the number of independently evolved lines, except for the ancestor, for which the
interval is based on the six replicate assays.

57

Figure 11.

 

 
   
    
   

(A)
0.035 7
- - Ancestor
0.03 - - Evolved GroupA
— Progenitors
-— Evolved Group 8
0.025
0.02

E...
a

0.01

0.005
0
0 60 120 180 240 300 360 420 480 540 600 660 720
-0.005
Time (minutes)
(8)
0.035
- - Ancestor
- - Evolved GroupA
0.03

 

-— Progenitors
Evolved Group 8

0.025- ' o-._..—.
I I
l I
' I
0.02 . '
I I
I
3 .' '
$0.015 - . I
I
’ I

’ I

0.01 ’ I
I
I
I
0
0.005
o .’ .77. g n I
0 60 1 20 1 80 240 300 360 420 480 540 600 660 720
Time (minutes)

58

Figure 11. Growth dynamics of the FTG-evolved lines and their progenitors
following (A) FT treatment and (B) stationary phase at 37°C. Evolved group
A and the ancestor are shown by dashed-grey and dashed-black curves,
respectively. Evolved group B and progenitors are shown as solid-grey and
solid-black curves, respectively. Each trajectory shows the mean calculated over
all of the evolved lines or progenitors in a group, except for the single group A
ancestor. In each case, the underlying trajectories were replicated nine-fold for
each evolved line or progenitor, including the group A ancestor.

59

Figure 12.

 

(A)
7 ,
6 e Ancestor
e Evolved GroupA
5
4

I092 (CD/Initial 00)

0 __________________
.1 ...............................
-2
o 60 120 180 240 300 360 420 480 540 600 660 720
Time (minutes)
(B)
7
e Progenitor
0

e Evolved Group B

by; (00m 00)

 

0 60 120 180 240 300 360 420 480 540 600 660 720
Time (minutes)

60

Figure 12. Lag phases following the FT treatment for two FTG-evolved
lines and their progenitors. (A) Group A evolved line and the ancestor; (B)
group B evolved line and its immediate progenitor. The evolved lines are shown
as grey symbols, and their progenitors as black symbols. The apparent duration
of the lag phase is where the log-linear regression crosses the initial optical
density. The actual physiological duration occurs when the regression crosses
the density adjusted for the proportion of surviving cells in the initial population,
shown by the dashed horizontal lines. Growth trajectories and survival
calculations are means of 27 replicate assays for each evolved line and
progenitor.

61

Figure 13.

(A)

7

e Ancestor
e Evolved Group A

 

0 60

e Progenitor
e Evolved Group B

120

120

180
Time (minutes)

180
Time (minutes)

62

240

240

300

300

Figure 13. Lag phases following stationary phase at 37°C for two FTG-
evolved lines and their progenitors. (A) Group A evolved line and the
ancestor; (B) group B evolved line and its immediate progenitor. The evolved
lines are shown as grey symbols, and their progenitors as black symbols. The
duration of the lag phase extends to where the log-linear regression crosses the
initial density; there is no appreciable mortality under this regime. Growth
trajectories are means of 27 replicate assays for each evolved line and

progenitor.

63

CHAPTER 3
ADAPTATION TO CYCLES OF FREEZING, THAWING, AND GROWTH:
GENETIC EVOLUTION
ABSTRACT
Microbial evolution experiments offer a powerful approach for coupling

changes in complex phenotypes, including fitness and its components, with
specific mutations. Here we investigate mutations substituted in 15 lines of E.
coli that evolved for 1,000 generations under freeze-thaw-growth (FTG)
conditions. To investigate the genetic basis of their improvements, we first
sequenced six candidate genes in two lines, but found no changes. We then
screened many of the lines for mutations involving insertion sequence (IS)
elements, and identified two genes where multiple lines had similar mutations.
Three lines had IS150 insertions in cls, which encodes cardiolipin synthase, and
eight lines had IS150 insertions in the uspA-uspB intergenic region, encoding two
universal stress proteins. Another line had an 11-bp deletion mutation in the cls
gene. Strain reconstructions and competitions demonstrated that this deletion is
beneficial under the FTG regime in its evolved genetic background. Further
experiments showed that this cls mutation helps maintain membrane fluidity, and
enhances survival, during freeze-thaw cycles. Reconstruction of isogenic strains
also showed that the IS150 insertions in uspA/B are beneficial under the FT G
regime. The evolved insertions reduce uspB transcription and enhance recovery
of growth after thawing, although the physiological mechanism for faster recovery

remains unknown.

64

BACKGROUND

Evolutionary biologists have long been interested in elucidating the
genetic bases of adaptation to particular environments, including especially those
environments that are novel or stressful to the organism. Evolution experiments
using bacteria and other microorganisms (Elena and Lenski 2003; Poon and
Chao 2005; Riehle et al. 2005; Herring et al. 2006; Schoustra et al. 2006, Velicer
et al. 2006) offer a powerful context for studying the genetics of evolutionary
adaptation, because one can couple changes in phenotypic traits, including
fitness and its components, with specific mutations. Previous studies on
evolutionary adaptation to stressful environments, for example, have examined
how known stress-responsive genes evolve, and can also provide new insights
into the physiology of organisms when there are unexpected changes in other
genes (Riehle et al. 2001; de Visser et al. 2004). More generally, it is of
considerable interest to know whether independent populations, when confronted
with the same environmental challenges, will evolve along parallel or divergent
paths. The convergence of multiple evolving lines on similar phenotypes, and
even the same underlying genetic changes, provides a strong indication that the
changes are adaptive as opposed to the product of random genetic drift (Bull et
al. 1997; Ferea et al. 1999; Wichman et al. 1999; Cooper et al. 2001b, 2003;
Colosimo et al. 2005; Wood et al. 2005; Woods et al. 2006; Pelosi et al. 2006).

Advantages to using bacteria for experimental evolution include the ability
to establish and propagate replicate populations from a single clone, their rapid

generations, and the ability to store and revive the ancestral clone and evolved

65

lines from different evolutionary time-points. Also, asexual reproduction allows
the use of a stably inherited neutral genetic marker that can distinguish different
competitors in assays designed to measure their relative fitness. This is
particularly relevant for competition experiments between isogenic constructs
that differ by a single mutation, to measure its fitness effect in either the exact
same selection environment or another environment (Cooper et al. 2001, 2003;
Crozat et al. 2005). This approach allows one to distinguish rigorously those
mutations that cause fitness improvement from others that may have merely
hitchhiked without producing any adaptation. The physiological effects of
adaptive mutations can then be elucidated by appropriate phenotypic
comparisons between the same isogenic constructs.

Here, we examine the genetic basis of evolutionary adaptation that
occurred in 15 populations that evolved under stressful freeze-thaw-growth
(FTG) conditions. This FTG regime consisted of a two-day cycle including 22.5 h
spent frozen at -80°C and 1.5 h thawing at room temperature, followed by 1:100
dilution into fresh medium and incubation at 37°C for 24 h. During this last step,
cells underwent a complete growth cycle including lag, growth, and stationary
phases. The evolved lines achieved large increases in fitness relative to their
progenitors when competed under the FTG regime, and these gains resulted
from both improved survival after the freeze-thaw (FT) cycle and faster recovery
to initiate exponential growth after thawing (Sleight and Lenski 2007). This
shorter lag phase is specific to recovery after freezing and thawing, and not some

more general improvement in recovery of growth following stationary phase per

66

se. Thus, it is of interest to identify and characterize some of the genetic
changes responsible for these adaptations to the FT G regime.

Various approaches have been used to find beneficial mutations
substituted in bacterial evolution experiments: sequencing candidate genes
suggested by phenotypic changes (Treves et al. 1998; Cooper et al. 2001b)
including global gene expression profiles (Cooper et al. 2003; Pelosi et al. 2006),
genomic fingerprinting including the use of native insertion sequence (IS)
elements as probes (Papadopoulos et al. 1999; Schneider et al. 2000; Zinser et
al. 2003; Schneider and Lenski 2004), and most recently even whole-genome
sequencing (Velicer et al. 2006; Herring et al. 2006). In this study, we have
pursued a two-prong strategy. First, we sequenced six candidate genes whose
diverse functions seemed particularly relevant to the stresses that cells would
experience during freezing and thawing. When these candidate genes were
sequenced in two FTG-evolved lines, however, we found no mutations in any of
them relative to their progenitors. Second, we screened the entire genomes of
many of the FTG-evolved lines and their progenitors by using a fingerprinting
approach with IS elements as probes. This approach led us to the discovery of
multiple lS-associated mutations at each of two loci, and these genes and
mutations then became the focus of this study. In the paragraphs below, we
describe some of the salient features of IS elements for our work.

IS elements are generally small (< 2.5 kb) mobile genetic elements, found
in bacteria, that only carry information related to their transposition and its

regulation (Mahillon and Chandler 1998). The number and locations of IS

67

elements vary among bacterial genomes (Deoiner 1987; Schneider et al. 2002),
and these elements are a significant source of new mutations (Naas et al. 1994;
Hall 1999; Papadopoulos et al. 1999; Schneider et al. 2000; Cooper et al. 2001b;
de Visser et al. 2004; Schneider and Lenski 2004). IS insertions can have
diverse effects on gene expression, including gene inactivation and polar effects
(Jordan et al. 1968), activation of cryptic genes (Reynolds et al. 1981; Hall 1999),
inactivation of adjacent genes (Blasband et al. 1986), and altered gene
expression caused by lS-encoded promoters (Charlier et al. 1982; Ciampi et al.
1982; Jaurin and Nonnark 1983; Bongers et al. 2003).

IS and other mobile genetic elements have been viewed as genomic
parasites, which have mostly deleterious effects on their hosts but are
maintained in genomes by transposition and horizontal transfer (Charlesworth et
al. 1994). However, these elements also sometimes have beneficial
consequences by producing mutations that increase the genetic variation
available for selection (Blot 1994; Kidwell and Lisch 2001; Schneider and Lenski
2004). Some authors have suggested that IS transposition may occur at higher
rates in bacteria under stressful conditions, including low (Reif and Saedler 1975)
or high temperature (Ohtsubo et al. 2005), starvation (Naas et al. 1995; Hall
1999; Twiss et al. 2005), oxygen stress (de Visser et al. 2004), and UV radiation
(Eichenbaum and Livneh 1998), but direct tests of the effects are few (Hall 1999)
and IS-mediated mutations certainly occur in non-stressful environments as well.
For example, the average number of lS-mediated substitutions in E. coli

genomes after 30 years of starvation (Naas et al. 1995) was roughly the same as

68

after only 4 years of growth (Papadopoulos et al. 1999). A study directly
comparing the rates of lS-associated mutations substituted under growth and
starvation conditions found no significant differences between these two
treatments (de Visser et al. 2004). In this study, we do not test whether the
stresses of freezing and thawing contributed to IS activity per se; instead, we
seek to determine whether IS mutations that were substituted in the evolving
lines contributed to their adaptation to the FTG regime.
MATERIALS AND METHODS

Derivation of bacterial strains

The bacteria used in this study derive from a common ancestor via two
successive evolution experiments, as summarized in Figure 14. In the first
experiment, 12 populations independently evolved for 20,000 generations (3,000
days) starting from two variants of the same ancestral strain of E. coli B (Lenski
et al. 1991; Lenski 2004). One ancestral variant (REL606) cannot grow on
arabinose, while the other (REL607) is a spontaneous Ara+ mutant. The Ara
marker is selectively neutral under the conditions of that long-term experiment
(Lenski et al. 1991) as well as in the freeze-thaw-growth environment used in this
study (Sleight and Lenski 2007). For the purposes of this paper, we will refer to
both variants as the “ancestor” unless we need to discriminate between them for
methodological reasons, or as the “original ancestor” to minimize confusion with
clones derived from the first evolution experiment that served as progenitors for
the second evolution experiment. The long-term experiment involves daily

transfers in DM25 (Davis Minimal medium supplemented with glucose at 25

69

pg/mL) with incubation at 37°C. The 100-fold dilution and re-growth allow about
6.6 (= log2 100) generations per day. One clone was sampled from each of the
12 populations after 20,000 generations, and these twelve clones are designated

as A-1 to A-6 and A+1 to A+6 for this study.

The second evolution experiment involves 15 populations that were
propagated under the freeze-thaw-growth (FTG) regime. Three populations were
founded by the Ara+ variant of the original ancestor, and a single population was
founded by each clone sampled from each of the 12 long-term populations.
These 15 populations evolved for 150 two-day FTG cycles, which equals at least
1,000 generations based on the 100-fold dilution and growth in alternating days.
Somewhat more generations occurred because growth also offset death during
the freeze-thaw cycle in alternating days. To start the FTG evolution experiment,
1 mL of DM25 stationary-phase culture was transferred into a freezer tube and
put in a -80°C freezer for 22.5 h. The tube then thawed at room temperature
(~22°C) for 1.5 h, after which time the culture was diluted 100-fold into fresh
DM25 and incubated at 37°C without shaking for 24 h. Thus, the evolving
populations experienced cycles of a day of freezing and thawing that alternated
with a day of growth in the same medium and at the same temperature as during
the long-term evolution experiment. Populations and clones were sampled and
stored every 100 generations (30 days) in freezer vials with glycerol for future
studies. Note that glycerol was added as a cryoprotectant for long-term storage
of samples, but no cryoprotectant was present during the actual FTG evolution

experiment itself. The three populations derived from the original ancestor are

70

designated as AncA/FTG, AncB/FTG, and AncC/FTG, while the other twelve are

designated as A—1/FTG to A-6/FTG and A+1/FTG to A+6/FTG (Fig. 14).

During the first evolution experiment, unique mutations were substituted in
each of the 12 populations after 20,000 generations, including different alleles in
nadR (Woods et al. 2006). To check for the possibility of inadvertent cross-
contamination during the second evolution experiment, the nadR gene was re-
sequenced in all 15 of the FTG lines. In all cases, the nadR alleles precisely
matched that of their intended progenitors, thereby excluding any cross-

contamination.

DNA extractions, PCR, and sequencing

DNA was extracted from cultures grown overnight in LB medium using the
Qiagen genomic-tip 100/G kit and quantified using a spectrophotometer.
Polymerase Chain Reaction (PCR) was performed using the Promega High
Fidelity Taq Polymerase kit in a PTO-200 thermocycler (MJ). PCR products were
purified using GFX PCR DNA and Gel Band Purification Kit (Amersham
Biosciences). Purified PCR products were sequenced on both strands by the
dideoxy chain termination method (Sanger et al. 1977). DNA sequences were
aligned in the SeqMan program (DNA Star). All mutations were verified by
repeating this entire process at least twice.
Candidate genes and sequencing strategy

As our first approach to identify the genetic bases of adaptation to the
FTG regime, we sequenced six candidate genes. These candidates were

chosen a priori because their described functions involve coping with the multi-

71

faceted stresses of freezing and thawing, which include low temperature,
dehydration, osmotic shock, oxygen radicals, and intracellular ice formation (Cox
and Heckly 1973; Grecz et al. 1980; Calcott and Gargett 1981; Mazur1984; Gao
and Critser 2000). Moreover, previous studies have shown that cold-shock
proteins (Willimsky et al. 1992), membrane fluidity (Sinensky 1974; Calcott
1985), oxygen radical scavenging (Stead and Park 2000), and production of
compatible solutes (Kandror et al. 2002) contribute to survival and recovery from
freezing and thawing. Therefore, the six candidate genes were: cspA, the major
cold-shock gene; fabF, involved in fatty acid unsaturation; otsA, involved in
synthesizing the cryoprotectant trehalose; rpoS, encoding the alternative sigma
factor involved in general stress responses (including transcription of otsA); rrsA,
which encodes the 16S rRNA portion of the MA operon; and sodA, encoding a
superoxide dismutase. Moreover, some of these genes appear to have unusual
sequence signatures in cold-adapted microorganisms (Hebraud and Potier1999;
Pruss et al. 1999). We sequenced all six candidate genes, including their
adjacent regulatory regions, in the original ancestor and in two FTG-evolved
clones (A-1/FTG and A+1/FTG) using the primer pairs shown in Appendix 2,
Table A2.1.
Southern hybridizations with IS probes

Extracted genomic DNA from each clone was digested with EcoRV (for
IS1, ISZ, IS3, IS4, IS30, and IS186) or Hincll (for IS150), separated on a 0.8%
agarose gels with a 1 kb ladder (New England Biolabs), and transferred to nylon

membranes (Southern 1975). Hybridizations were performed at high stringency

72

(68°C) using internal fragments of each IS probe (Appendix 2, Table A2.2) that
were PCR amplified, purified, and labeled using the DIG DNA Labeling and
Detection Kit (Roche).
Characterization of sequences adjacent to IS elements by inverse PCR

Genomic DNA from each clone was digested with EcoRV or Hincll, and
separated on agarose gels. Gel fragments containing IS elements were cut and
purified using the GFX PCR DNA and the Gel Band Purification Kit (Amersham
Biosciences). The fragments were self-ligated (end-to-end) with T4 DNA Ligase
(New England Biolabs) at 10 pg/mL, and the ligated mixtures were used as
templates in PCR reactions using primers directed outward from the appropriate
IS element (Appendix 2, Table A2.3). PCR products were purified, sequenced,
and the sequences adjacent to the IS element were compared to the whole
genome sequence for E. coli K-12 (Riley et al. 2006; Genbank accession number
NC__000913). Once the location of these adjacent sequences was determined,
primers were designed to PCR amplify the IS-insertion allele on the
chromosome. The PCR products were sequenced with the same primers
(Appendix 2, Table A23) to determine the exact position, orientation, and target-
site duplication caused by the IS insertion.
Isogenic strain construction

Isogenic constructs and reconstructs were created using the “gene
gorging” replacement technique (Herring et al. 2003). In the case of the A-2/FTG
evolved clone, which has an 11-bp deletion in the cls gene, we performed two

allelic replacements. First, the ancestral cls allele was moved into this evolved

73

clone. Next, the evolved clone with the introduced ancestral cls allele was
reconstructed to the evolved cls deletion allele to verify that phenotypic
differences were caused by the mutation and not a side-effect of the gene
gorging process. Second, the evolved cls deletion allele was introduced into the
A-2 progenitor and then reconstructed back to the ancestral cls allele. We used
the A-2/FTG evolved line, rather than other evolved lines that had IS insertions in
the cls gene, because this small mutation (11-bp deletion) is easier to
manipulate. In the case of the AncB/FTG evolved clone, which acquired an
IS150 insertion in the uspA/B intergenic region, we independently made two
isogenic constructs from which this IS 150 insertion was removed, thereby
restoring the ancestral allele of uspA/B. In this case, we did not attempt to move
the uspA/B::IS150 allele back into this construct, nor into the ancestor, because
of recombination events, during the strain construction process, with the various
IS150 copies present in the chromosome. The two independently constructed
isogenic strains from which the IS150 insertion was removed did not differ
significantly from one another in any respect, supporting their isogenicity, and the
data obtained for them is combined in the competition, growth curve, and real-
time PCR experiments reported in this paper. See Appendix 2, Figure A21, for a
schematic representation of these isogenic constructs and reconstructs.

For all constructs, we used one primer with a l-Sce-I restriction site on its
5’ end. Fragments that included about 500 bp on each side of the mutation site
or its ancestral equivalent were PCR-amplified and purified. Primers used to

make the cls constructs were RL594 (5’ GGA CCT GCG CCA ACA TC 3’) and

74

RL595 (5’ TAG GGA TAA CAG GGT AAT CCG CCG GAT CGA ATA ACT C 3’).
The primers for the uspA/B constructs were RL567 (5’ TAG GGA TAA CAG GGT
AAT ATG TAG GCC TGA TAA GCG TAG C 3’) and RL568 (5’ CCC GGT GTA
TAG GTC AGA GTA GTT 3’). The resulting ~1 kb fragment was cloned into
pCRIl-Topo and transformed into E. coli cells using the Topo TA Cloning kit
(Invitrogen). Transformants were selected and grown in LB supplemented with
ampicillin, and the plasmids were purified using the Promega Wizard Plus SV
Miniprep DNA Purification System. All cloned fragments were checked by DNA
sequencing and no additional mutations were detected.

The donor plasmid (encoding kanamycin resistance and the ~1 kb cloned
fragment) and the mutagenesis plasmid (encoding the l—Sce-I restriction enzyme,
A red recombination enzymes, and chloramphenicol resistance) were 00-
transformed into competent cells of the desired clone. Transformants were
selected using LB agar with supplemental kanamycin and chloramphenicol, and
then grown in a medium containing arabinose; this medium induces the enzymes
produced by the mutagenesis plasmid that promote homologous recombination
between the fragment on the donor plasmid and the chromosome (Herring et al.
2003). Putative constructs were then screened for the desired genotype using
PCR with primers specific to the corresponding gene on the chromosome, and
also checked for kanamycin and chloramphenicol sensitivity to ensure loss of
both plasmids. All constructs and reconstructs were sequenced twice to verify
the presence or absence, as intended, of the evolved mutation, and to ensure

that no other mutations were introduced in the vicinity of the replacement allele.

75

Quantifying FTG fitness and its components

We performed competition experiments to measure the relative fitness
levels of various ancestral, evolved, and constructed genotypes. The neutrally
marked variants of the ancestral strain allowed competitors to be distinguished
on the basis of colony color on tetrazolium-arabinose (TA) indicator agar (Lenski
et al. 1991; Sleight and Lenski 2007). For each replicate competition assay,
each competitor’s realized (net) growth rate, r, was measured over the two-day
FTG cycle, as follows:

r = |n((N2 x 100) / No)
where N2 is that competitor’s final cell density, No is its initial cell density, and the
factor of 100 takes into account the 100-fold dilution between each FT G cycle.
Thus, the overall FTG fitness of one competitor relative to another is simply the
ratio of their respective realized growth rates over the complete cycle.

We also calculated each competitor’s FT survival, 3, and subsequent
growth rate, 9, over the two separate days of the FT G cycle as follows:

3 = N1 / No

9 = |n((N2 x 100) / M)
where N1 is the viable cell density measured after the first day of the cycle prior
to the 100-fold dilution. Note that survival is a proportion, whereas growth is a
rate. In any case, the relative survival and growth of two competitors are
expressed as the ratio of the relevant parameters, so that both become
dimensionless quantities, as is the overall FTG fitness. Note also that all of these

quantities are calculated separately for each replicate assay, thus preserving

76

their statistical independence. The quantities were averaged across replicates,
and analyzed statistically as described below.
Statistical methods

We performed paired t-tests to compare the properties of two clones, with
pairing based on temporal and spatial proximity in the structure of the
experiment. In those cases where we had a priori expectations about the
direction of change, significance was computed using one-tailed t-tests;
otherwise two-tailed tests were used. We expected the FTG-evolved lines to
have improved performance relative to their progenitors in the FT G regime,
including both FT survival and subsequent growth. We also expected that
evolved alleles would confer an advantage under the FTG conditions, and when
present in the evolved genetic background. However, we had no expectations
for other phenotypic aspects of isogenic clones differing only by a specific allele.
Two-way ANOVAs were performed to compare phenotypic measures between
clones with the four combinations of ancestral and evolved backgrounds and

alleles.
Growth curves

Growth curves, based on optical density (OD) measurements, were
obtained for sets of isogenic constructs in order to compare the durations of their
lag phase prior to growth as well as their doubling times during the growth phase.
The OD measurements were performed at 420 nm with 0.2-mL cultures in a 96-
well microtiter plate that was incubated at 37°C. Data were collected for cultures

following either a FT cycle or stationary phase at 37°C, such that a comparison

77

between the two curves allows one to estimate the effect of the FT cycle on the
time required to achieve exponential growth following dilution into fresh DM25

medium (Sleight and Lenski 2007).

Measuring membrane fluidity

Fluorescence polarization (anisotropy) measures fluidity in the cytoplasmic
membrane using various probes, each one specific to a particular region of the
membrane (Harris et al. 2002; Vanounou et al. 2002). The sample incubated
with the probe is excited by polarized light, and the emitted light from the probe is
polarized differentially depending on how much rotation of the probe occurs.
When the emitted light is simultaneously measured both parallel and
perpendicular to the axis of excitation, a probe that can rotate more freely (i.e. in
a more fluid membrane) will generate a larger emission signal in the
perpendicular relative to the parallel direction, when compared to a probe that
rotates less freely (i.e. in a less fluid membrane).

Fluorescence anisotropy was measured with a Spectramax M5
fluorometer (Molecular Devices) using two probes, 6-dodecanoyl-2-
dimethylaminoaphthalene (Iaurdan) and 1,3-diphenyl-1,3,5-hexatriene (DPH).
Laurdan is an amphipathic molecule that localizes near the lipid polar head
groups in the cytoplasmic membrane, and is therefore sensitive to changes at the
water-lipid interface (Harris et al. 2002; Vanounou et al. 2002). DPH is useful for
detecting changes in saturation of fatty acyl chains in the cytoplasmic membrane
(Beney et al. 2004; Aricha et al. 2004). Either Iaurdan (5 x 10'5 M added from a

stock solution in methanol) or DPH (5 x 10'5 M added from a stock solution in

78

tetrahydrofuran) was incubated for one hour at room temperature in the dark with
0.2 mL of cultures sampled either at stationary phase or after a FT cycle.
Cultures without any probe were used as a scattering control, and were
incubated under the same conditions for each individually measured sample.

The optimal ratio of probe to culture concentration was determined by
using the minimal concentration giving an appropriate signal-to-noise ratio
(Vanounou et al. 2002). Laurdan anisotropy was measured at the 355 nm
excitation wavelength and 440 nm emission wavelength. DPH anisotropy was
measured at the 360 nm excitation wavelength and 430 nm emission
wavelength. Anisotropy (r) is calculated in the traditional manner (Harris et al.
2002) as follows:

r = (IV - GIH) / (IV + 2GIH),
where Iv and IH are the fluorescence intensities determined at vertical and
horizontal orientations of the emission polarizer when the excitation polarizer is
set in the vertical position. G is a correction factor for dissymmetry associated
with the horizontal and vertical positions of the polarizers. Lower anisotropy
values indicate higher probe rotation and thus imply a more fluid membrane
compared to higher anisotropy values.
RNA extractions and real-time (RT) PCR

RNA was extracted from cultures at stationary phase (24 h after dilution
into fresh media), after a FT cycle (frozen at -80°C for 22.5 h and then thawed for
1.5 h at room temperature), and in transitional growth conditions (2 h after the

thawed culture was diluted 1:10 Into DM25 and incubated at 37°C) using the

79

RNeasy Mini Kit (Qiagen). For that third treatment, two hours was chosen
because it corresponds to a time period when the ancestor is still deep within its
long FT-associated lag phase, whereas the evolved clone is approaching the
time when it starts growth (Sleight and Lenski 2007); hence, that timepoint may
offer the best opportunity to detect physiologically important differences in gene
expression between the evolved clone and ancestor. A 1:10 dilution was used,
instead of the 1:100 dilution used in the evolution experiment, in order to extract
enough RNA for the RT-PCR procedure; control experiments confirmed that the
difference between the ancestor and evolved clone in the duration of their lags
after a FT cycle is comparable for 1:10 and 1:100 dilutions. The same culture
volume was used for each clone to preserve the relevant cell density, while an
endogenous control served to measure the total RNA present (see below). After
extraction, RNA was treated with Rnase-free DNase (Ambion), and PCR
experiments were run using the extracted RNA as a template to ensure the
absence of DNA in the RNA samples. RT-PCR was performed using the
TaqMan One-Step RT-PCR kit (Applied Biosystems) with primers and MCB
probes specific to the uspA, uspB, and 168 rRNA genes in the ABI Prism
7900HT Sequence Detection System. Negative controls without any RNA were
also used in RT-PCR experiments, and they produced no significant background
noise. An RNA standard was obtained by mixing RNA from different samples,
and a logo dilution series through 10'5 was measured with the primers and probe
for each gene. Regression was then performed on CT (threshold cycle) values

against the dilution factor to determine the amount of RNA in unknown samples.

80

The mRNA levels for uspA and uspB levels were individually divided by the 16S
rRNA levels for each clone and sample timepoint to ensure that any differences
were not an artifact of the total amount of RNA extracted.
Ethanol sensitivity experiments

Samples of an evolved clone in stationary phase or after 3 FT cycle were
mixed with the reciprocally marked ancestor from the same condition, as in a
competition experiment. The mixed culture was then incubated in 10% ethanol
at 37°C for one hour. Cell densities were measured before and after ethanol
exposure, and used to quantify relative sensitivity using the same equation used
to quantify relative FT survival.
Novobiocin sensitivity experiments

Stationary-phase cultures of ancestral and evolved clones were separately
plated on both TA agar and TA agar supplemented with novobiocin (100 pg/mL).
The plates without novobiocin were incubated at 37°C for one day, while those
with antibiotic were incubated for two days. Novobiocin sensitivity was then
calculated from the ratio of the resulting cell densities.
Supporting data and supplementary analyses

Appendix 2, which can be found at the back of this dissertation, provides
some supporting data and supplementary analyses relevant to this Chapter. All
information pertaining to the Materials and Methods is included in the figure
legend if not already described.

RESULTS

Absence of mutations in candidate genes

81

We began our search for the beneficial mutations responsible for
adaptation to the FTG regime by sequencing six candidate genes in two of the
FTG-evolved lines. These candidates were chosen because their encoded
functions affect a wide range of physiological processes important for FT survival
and subsequent recovery. These processes (and candidate genes) include: the
cold-shock response (cspA); membrane fluidity (fabF); synthesis of
cryoprotectants (otsA); global regulation of stress responses (rpoS); protein
translation (nsA); and scavenging oxygen radicals (sodA). A total of ~10 kb was
sequenced in two focal FTG-evolved lines, A-1/FTG and A+1/FTG, but no
mutations were found relative to their progenitors. Therefore, we next sought to
pursue a more open-ended approach that would find mutations anywhere in the
genome, albeit only those mutations causing observable changes in DNA
fingerprints.

Discovery of several IS-associated mutations in FTG-evolved lines

To characterize genetic changes and genomic rearrangements associated
with IS elements in FTG-evolved lines relative to their progenitors, a Restriction
Fragment Length Polymorphism (RFLP) analysis was performed using internal
fragments from each of the seven IS elements in E. coli B as probes. One clone
sampled from 10 of the 15 independently evolved FTG populations was
compared against its progenitor for each of the seven IS elements (Table 3). An
lS-associated change is defined as either a gain or loss of a band; however, one
IS-mediated mutation can produce more than one band change, and therefore

this method may sometimes overestimate the number of underlying mutational

82

events (Papadopoulos et al. 1999; Schneider et al. 2000). Of the seven IS
elements, three showed no changes (I82, IS4, and IS30) in any of the 10 evolved
lineages. The absence of any changes associated with these three elements
does not necessarily mean that they had absolutely no transposition activity
under FTG conditions, but rather that either no mutations produced by such
activity were highly beneficial and thus substituted in any of the ten FTG-evolved
lines chosen for analysis, or that no transposition event hitchhiked with other
beneficial mutations. By contrast, there were 30 total changes in the other four
IS elements (IS1, ISB, IS 150, and IS 186), with IS 150 alone contributing to more
than half of these changes (see Figure 15 and Appendix 2, Figures A2.2-5, for
the RFLP images). This high level of activity is not specific to FTG conditions,
since IS150 also shows high activity under benign growth conditions
(Papadopoulos et al. 1999; Schneider et al. 2000; Cooper et al. 2001). See also
the exact band differences between FTG-evolved lines and their progenitors in
Appendix 2, Table A24.

The number of IS changes in FTG-evolved lines ranges from zero
(A+3/FTG, A-1/FT G) to seven (A+1/FTG) among the ten lines tested. Four of
them (A+3/FTG, A+6/FTG, A-2/FTG, A-4/FTG) had evolved mutator phenotypes
during their prior experimental evolution (Sniegowski et al. 1997; Cooper and
Lenski 2000). On average, there were more lS-associated changes in the non-
mutator than the mutator populations, although that difference was not
statistically significant (two-tailed t-test, p = 0.1946).

Localization of IS-associated mutations in FTG-evolved lines

83

Given the numerous lS-associated mutations that we found, and the fact
that parallel changes are a strong indication of adaptive evolution, we focused
our attention on characterizing those changes that arose independently in
multiple FTG-evolved lines. Figure 15 shows the Southern hybridizations using
the IS150 sequence to probe the genomes of the FTG-evolved lines and their
progenitors. Bands present in more than one FTG-evolved line, but not in their
progenitors, are circled and the chromosomal location is indicated. Two of the 10
FTG-evolved clones that were tested have an insertion in the cls gene, and four
evolved clones have an insertion in the uspA/B intergenic region. Another IS150-
associated change, which is not circled in this figure, is an insertion within the
uspD gene in A+4/FTG. However, this insertion was found in only one of ten
clones tested from this evolved population and, moreover, it was absent from all
of the other 14 FT G-evolved populations for which we screened 10 clones from
each. Therefore, this uspD mutation was not investigated further.

Systematic screening of cls in FTG-evolved lines

After the cls and uspA/B insertion mutations were discovered, we
screened 10 clones from all 15 FTG-evolved populations for the presence of
insertion mutations in these loci using PCR with primers designed to detect
mutations in regulatory as well as coding regions. For the cls gene, five FTG-
evolved lines were found to have IS insertions in at least one clone, including the
two lines in which IS150 insertions were originally found (A+4/FTG and A-
3/FTG). In a third line (A-1/FTG), an IS150 insertion was found in a single clone

that was not tested in the previous Southern hybridizations. In two other lines (A-

84

5/FTG and A-6/FTG), all clones tested in these populations had insertions of a
different IS element, IS186, either in or immediately upstream of cls.
Surprisingly, the progenitors of these two lines (A-5 and A-6), which had evolved
for 20,000 generations at constant 37°C, already harbored the cls::lS186
insertions. Thus, while these last two mutations might have contributed to
adaptation to the previous environment, they cannot be responsible for the
fitness gains specific to the FTG regime. In all of the other FTG-evolved lines
without any insertion mutations in cls, that gene and its adjacent regulatory
region were sequenced. One additional substitution was found, in which the A-
2/FTG evolved line had an 11-bp deletion that generates a premature stop codon
downstream. The physical locations of all these cls mutations are listed in Table
4, while Figure 16 provides a schematic representation. Table 5 shows the
substitution dynamics of the cls mutations in the corresponding FTG populations,
based on testing clones isolated from freezer stocks that we saved every 100
generations during the evolution experiment. We presume that all of the cls
mutations that evolved during the FTG experiment, including the deletion in line
A-2/FTG, disrupt the function of the encoded enzyme, cardiolipin (CL) synthase.
This enzyme is widely distributed across bacterial species and converts two
phosphatidylglycerol molecules into CL and glycerol in the membrane
(Hirschberg and Kennedy 1972) during stationary phase (Shibuya and Hiraoka
1992). As noted previously, a deletion mutation can be more precisely
manipulated than a new insertion of an IS element, as recombination between

multi-copy IS elements may produce unintended changes. Therefore, our

85

subsequent analyses of the effects of cls evolution on FTG performance will
focus on the cls deletion allele.
Systematic screening of uspA/B in FTG-evolved lines

By screening the same 10 clones from all 15 FTG-evolved lines, we found
mutations in uspA/B and its associated regulatory region in eight populations,
with one or more clones presenting an insertion. Remarkably, all of them were
ISI50 insertions in the same exact position and orientation in the uspA/B
intergenic region (Table 4). Cross-contamination can be ruled out because these
clones have different IS banding patterns in RFLP experiments, different cls
mutations (as described above), and unique nadR alleles derived from their
progenitors (see Materials and Methods). No mutations were found in the open-
reading frames of uspA or uspB in any of the FTG-evolved lines. Figure 17
shows that these identical IS150 insertions are located 34 bp upstream of the
uspB start codon and 95 bp downstream of the putative uspB oS promoter. The
uspA and uspB genes encode Universal Stress Proteins (Usp) A and B,
respectively. UspA is an autophosphorylating serine and threonine
phosphoprotein that is induced under a wide variety of stress conditions; it is
thought to play a role in protecting cells from DNA damage, although its exact
function is unknown (Kvint et al. 2003). Even less is known about UspB,
although it is differentially transcribed from UspA and may not be a truly universal
stress protein (Farewell et al. 1998; Kvint et al. 2003). The effects of the IS150
insertions on transcription of uspA and uspB will be discussed later. In the seven

other FTG lines without uspA/uspB::lS150 mutations, we sequenced uspA, uspB,

86

and the uspA/B intergenic region for individual clones, but no other mutations
were found. Table 6 summarizes the temporal dynamics of the
uspA/uspB::lSi50 mutations in the eight FT G-evolved populations where they
were found, based on testing clones from the freezer stocks saved every 100
generations during the evolution experiment.
Isogenic strain constructions

We constructed strains that were isogenic except for the cls deletion and
uspA/B::IS150 insertion in order to test the fitness and other phenotypic effects
of these evolved mutations. For cls, we first restored the ancestral allele in an
evolved clone from the A-2/FTG line. In order to verify that any phenotypic
differences were specifically caused by the cls alleles (and not by some
hypothetical mutation inadvertently introduced during genetic manipulations), we
then “reconstructed” the evolved state by re-introducing the cls deletion mutation
back into the same evolved clone. Similarly, in order to test the effect of the
evolved allele in the ancestral genetic background, the cls deletion mutation was
moved into the A-2 progenitor genome, and then that progenitor was
reconstructed by reintroducing the ancestral cls allele. Thus, we produced a set
of four isogenic strains comprising each of the ancestral and evolved cls alleles
in each of the progenitor and FTG-evolved backgrounds.

For uspA/B, we made two independent constructs from one evolved clone
in the AncB/FTG line in which the ancestral allele replaced the
uspA/usszzlS150. The clones with the evolved insertion allele in the same

background served as the controls for the experiments reported below; the

87

evolved allele was not introduced into the ancestral background for this locus.
We detected no significant differences between the two independent constructs;
therefore, we pooled the data from these constructs in the analyses below.
Details of the strain-construction procedures are provided in the Materials and
Methods section.
Phenotypic consequences of evolved cls mutation

We performed several experiments to investigate the physiological effects
of the cls deletion mutation in the evolved and progenitor genetic backgrounds.
First, we measured the relative fitness levels of all four isogenic strains, including
the ancestral and evolved cls alleles in the progenitor and FT G-evolved genetic
backgrounds, under four conditions (Figure 18): the entire two-day FTG regime,
FT survival, growth performance after the FT treatment, and growth performance
after stationary phase (without a FT cycle). Over the entire FTG cycle, the
evolved clone with the cls’ mutation has a significantly higher fitness than its
progenitor with the ancestral cls+ allele (Figure 18A; paired one-tailed t-test, p <
0.0001). This higher overall FTG fitness reflects both improved FT survival and
subsequent growth after FT treatment (Figures 188 and 18C; paired one-tailed t-
tests, p = 0.0001 and p = 0.0006, respectively). This overall improvement is
specific to FTG conditions, because the evolved clone has a small, but
significant, decrease in its performance without the FT treatment (Figure 18D;
paired two-tailed t-test, p = 0.0038). Note the scale differences between panels
in Figure 18; thus the error bars on the fitness values that are not significantly

different from 1.0 in panel D are, in fact, very tight. These comparisons

88

demonstrate that the evolved A-2/FTG line improved its FTG fitness relative to its
progenitor, and that the adaptation is specific to the FTG regime.

The following comparisons allow us to test whether the evolved cls”
mutation is responsible, at least in part, for this adaptation. Restoring the
ancestral cls+ mutation to the evolved strain reduces overall FTG fitness by about
31% (Figure 18A; paired two-tailed t-test, p = 0.0002). The evolved allele
significantly improves both FT survival and subsequent growth (Figures 188 and
18C; paired two-tailed t-tests, p = 0.0054 and 0.0024, respectively). The
improvement associated with the evolved cls allele is specific to the FTG regime,
as there is no significant difference between the ancestral cls+ and evolved cls‘
alleles in the same evolved A-2/FTG background during competition after
stationary phase without a FT cycle (Figure 18D; paired two-tailed t-test, p =
0.5077). These comparisons demonstrate that the cls‘ allele contributes
significantly to the FT G-specific adaptation.

When moved into the A-2 progenitor, the same evolved cls' allele shows
however only a small (roughly 5%) and marginally non-significant advantage in
overall FTG fitness (Figure 18A; paired two-tailed t-test, p = 0.0957). Also,
neither FT survival nor subsequent growth after FT treatment show any
significant effect of the cls allele in the progenitor’s background (Figures 18B and
18C; paired two-tailed t-tests, p = 0.4686 and 0.2736, respectively). The
ancestral allele has a small, but again non-significant, advantage during growth
after stationary phase without the FT cycle (Figure 18D; paired two-tailed t-test, p

= 0.1250). Evidently, the fitness effects of the evolved cls‘ allele are contingent

89

 

on interactions with one or more other mutations in the evolved A-2/FTG line.
This conclusion is further supported by a two-way ANOVA using the overall FTG
fitness data, which indicates a highly significant interaction between the genetic
background and the cls allele (p = 0.0001).

The beneficial effect of the evolved cls‘ deletion allele for survival and
recovery, and the specificity of that effect with respect to the FT treatment, was
also evident from the growth dynamics of each isogenic strain (Figure 19).
Following a FT cycle, the evolved A-2/FTG strain with the evolved cls‘ allele
recovers much more quickly than any of the other three constructs (Fig. 19A).
When we compare the recovery of the same four isogenic strains following
stationary phase at 37°C with no FT treatment, the growth curves are almost
identical (Fig. 198). Comparing the dynamics with and without the FT treatment,
we also see that the recovery of the A-2 progenitor with its ancestral cls+ allele is
delayed (i.e., shifted to the right) by almost 3 h by imposing a FT treatment. In
striking contrast, recovery of the evolved A-2/FTG strain with its cls' allele is
delayed by less than 1 h when the FT treatment is imposed. As an aside, these
four strains following a 2 h cold shock at 4°C recover at about the same rate at
37°C (data not shown), indicating that the adaptation is specific to FT recovery
and is not related to recovery from low temperatures in general.

Previous research has shown that cls mutants have altered membrane
phase-transitions during or after a temperature downshift (Pluschke and Overath
1981). To test whether the evolved cls' allele altered membrane fluidity, we

measured fluorescence anisotropy of the same four strains after either FT

90

treatment or stationary phase at 37°C (see Materials and Methods for details).
Laurdan is a fluorescent probe that localizes to the cytoplasmic membrane, and
which is sensitive to changes in the water-lipid interface where the phospholipid
head groups reside (Harris et al. 2002; Vanounou et al. 2002). The results of the
experiments are summarized in Figure 20 (note that lower values for Iaurdan
anisotropy indicate more fluid membranes).

Figure 20A shows that the evolved cls‘ allele significantly increases
membrane fluidity (reduces Iaurdan anisotropy) relative to the ancestral cls+
allele following FT treatment, and does so in both the progenitor and FTG-
evolved backgrounds. One or more other evolved alleles must also contribute to
greater membrane fluidity, because the evolved construct has greater fluidity
than the progenitor even when both strains have the same ancestral cls+ allele.
A two-way ANOVA indicates the effects of both genetic background and cls allele
are highly significant, while there is no significant interaction with respect to this
phenotype (Table 7). The evolved cls' allele therefore generally increases
membrane fluidity after freezing and thawing regardless of genetic background.
However, recall from the fitness experiments that the evolved cls' allele was
beneficial under the FTG regime only in the evolved genetic background. Taken
together, these findings indicate that the beneficial effect of increased membrane
fluidity after the FT cycle must be dependent on other mutations in the A-2/FTG
evolved line and that the beneficial effect conferred by cls must be linked to

another phenotypic effect.

91

We see no significant differences in membrane fluidity when Iaurdan
anisotropy is measured for the same four strains after stationary phase, without
the FT treatment (Figure 203, Table 8). Therefore, the effect of the evolved cls'
allele on membrane fluidity is specific to freezing and thawing. Notice also that
the anisotropy value of the A-2/FT G evolved strain, with its evolved cls allele, is
almost the same when measured after either FT treatment or stationary phase at
37°C. This similarity indicates that this FTG-evolved line is able to maintain its
membrane fluidity at the water-lipid interface after 3 FT cycle, whereas the
progenitor’s membrane evidently becomes much more rigid. This effect is
independent of the cls allele.

We also used the same approach to measure changes in membrane
fluidity in two of the FTG-evolved lines with IS150 insertions in the cls gene
(A+4/FTG and A-3/FTG) relative to their respective progenitors. Both of these
evolved lines also have significantly increased membrane fluidity relative to their
progenitors after a FT cycle, but not during stationary phase (Appendix 2, Table
A2.5), paralleling the case of the A-2/FTG evolved clone with the cls deletion
relative to its progenitor (Figure 20). By contrast, the two FT G-evolved lines
whose progenitors (A-5 and A-6) had IS186 insertions in or near the cls gene
(Figure 16) evolved no further changes in their membrane fluidity following the
FT treatment (Appendix 2, Table A2.5). Taken together, these results indicate
that the cls mutations are a major, but not sole, determinant of the evolved

changes in membrane fluidity measured after the FT treatment.

92

To this point, our results on membrane fluidity are based on Iaurdan as a
probe, which is sensitive to changes at the water-lipid interface. We also
performed fluorescence polarization measurements using the probe DPH, which
is used to detect differences in fatty acyl chains in the membrane. However, we
found no differences between multiple FTG-evolved lines and their progenitors,
either during stationary phase or after the FT treatment.

We also examined the effect of the evolved cls deletion mutation on
sensitivity to the hydrophobic antibiotic novobiocin. The cls gene is known to
affect resistance to novobiocin (lvanisevic et al. 1995), although other genes are
also involved. As we expected, the A-2 progenitor with its cls+ allele is much
more resistant to novobiocin than its isogenic counterpart with the evolved cls
deletion, with the former producing >500 times as many colonies on plates
supplemented with that antibiotic at 100 pg/mL (data not shown). Unexpectedly,
the evolved strain A-2/FTG with the cls deletion allele is slightly less sensitive to
novobiocin than its counterpart with the ancestral cls+ allele, with the difference
being about 3-fold. These results show that the evolution of the cls gene has a
measurable effect on novobiocin resistance, while also supporting our other
results that indicate epistatic interactions between cls and one or more other loci
that evolved under the FT G regime.

Phenotypic consequences of evolved uspA/B mutation

Recall that some or all clones tested in 8 of the 15 FTG evolved

populations have identical IS150 insertions in the uspA/B intergenic region

(Figure 17, Table 4). We constructed isogenic strains, in an evolved genetic

93

background only, by replacing the uspA/uspB::lS150 allele in clones from the
AncB/FT G evolved line with the uspA/B ancestral allele. Figure 21 shows the
relative fitness values measured for the ancestor (without the IS element), the
AncB/FT G evolved line with its evolved uspA/B::IS150 allele, and the same
evolved strain except with that IS150 insertion replaced by the ancestral allele.
As expected, the evolved clone has a large fitness advantage relative to
its ancestor under the full two-day FTG regime (Figure 21A; paired one-tailed t-
test, p < 0.0001). Both FT survival and subsequent growth performance
contribute significantly to the overall improvement (Figures 21 B and 21 C; paired
one-tailed t-tests, both p < 0.0001). Without the FT cycle, fitness increased
slightly, but significantly, in this evolved line (Figure 21 D; paired one-tailed t-test,
p = 0.0003), indicating that its adaptation was not entirely specific to the FTG
regime. Most of the benefit was however linked to the FT regime. Hence, this
evolved line’s fitness gain after freezing and thawing was about 60% (Fig. 21 C)
as compared to its improvement of less than 5% when measured without the FT
treatment (Fig. 21D). Again, note the difference in scales between panels C and
D of Figure 21; the confidence intervals on the values in panel D are very tight.
The fact that this AncB/FTG evolved line shows some non-specific adaptation to
the growth conditions, whereas the A-2/FTG line examined previously did not, is
probably because AncB/FTG derives from the original ancestor, whereas A-
2/FTG derives from a progenitor that had already adapted to the same growth

conditions for 20,000 generations.

94

Eliminating the IS150 insertion from the uspA/B intergenic region in the
evolved clone (AncB/FTG -IS150) causes a significant reduction in FTG fitness
(Figure 21A; paired two-tailed t-test, p = 0.0022). Both FT survival and growth
performance after the FT cycle benefit from the presence of this insertion
(Figures 21 B and 21C; paired two-tailed t-tests, p = 0.0358 and 0.0023,
respectively). The effect of this mutation is specific to FTG conditions, as there is
no significant benefit to the evolved clone of having the IS150 insertion in growth
after stationary phase without an intervening FT treatment; in fact, there is a
marginally significant advantage when the evolved uspA/uspB::lS150 allele is
replaced by the insertion-free ancestral allele without the FT treatment (Figure
21 D; paired two-tailed t-test, p = 0.0685).

Figure 22 shows the growth dynamics of these same three strains after
the FT treatment and following stationary phase at 37°C. After 3 FT cycle, the
evolved line with its IS150 insertion in uspA/B has a much shorter lag than the
ancestor, while reverting the insertion to its ancestral state reduces the difference
(Fig. 22A). Following stationary phase without a FT cycle, the IS150 insertion
has no discernible effect in the FTG-evolved background, and the difference in
growth trajectories between the evolved line and its ancestor appears to reflect
solely a difference in their exponential growth rates (Fig. 228). Following a 2 h
c0ld shock at 4°C, the evolved line with or without the IS150 insertion recovers at
about the same rate at 37°C (data not shown), indicating again that the
uspA/uspB::IS150 allele is specific to FTG adaptation and is not related to cold

temperatures in general.

95

The physiological basis for the advantage of the evolved uspA/B::IS150
allele is unknown, although the precise same allele evolved independently in 8 of
the 15 FT G populations. As a step towards understanding the effect of the IS150
insertion, we measured the transcription levels of both uspA and uspB by
performing real-time PCR (RT-PCR) in stationary phase, after a FT cycle, and in
growth-permissive conditions 2 h after thawed cultures were diluted into fresh
media (Figure 23). Under all three conditions, transcription of uspB is reduced by
more than 10-fold in the evolved uspA/B::IS150 allele relative to both the evolved
strain with the ancestral uspA/B allele restored and the ancestor itself. By
contrast, the same evolved allele has only a small effect on uspA transcription
under the same test conditions (Appendix 2, Figure A2.6). This difference is
consistent with the physical location of the IS150 insertion between the putative
cs promoter and start codon for uspB, whereas the upstream regulatory region
for uspA is not interrupted by the insertion (Figure 17).

Although UspB is called a universal stress protein, it is evidently beneficial
to reduce or even eliminate its expression for freeze-thaw survival and recovery.
In fact, UspB is a predicted membrane protein, and uspB mutants are
hypersensitive to ethanol during stationary phase (Farewell et al. 1998). Ethanol
fluidizes membranes, and the physiological response to ethanol is to increase
membrane rigidity (Dombek and Ingram 1984). We hypothesized, therefore, that
the disruption of uspB transcription increases membrane fluidity, which would be
beneficial in the FTG regime, but which should reduce survival during ethanol

exposure. To test this hypothesis, we examined both predicted effects by

96

comparing the isogenic strains with and without the IS 150 insertion that disrupts
uspB transcription. Possible differences in membrane fluidity were measured by
performing fluorescence anisotropy after a FT cycle, as well as during stationary
phase, using both Iaurdan and DPH as probes. However, we saw no significant
differences in any of these cases (Appendix 2, Table A25 for Iaurdan
measurements). The strains were tested for ethanol sensitivity through co-
cultures with the marked ancestor, as in competition assays, in medium
containing 10% ethanol for 1 h either during stationary phase or after FT
treatment. Again, however, we saw no significant differences between these
strains under either treatment (data not shown). In summary, the IS150 insertion
reduces uspB transcription, which is beneficial under the FTG regime, but the
physiological basis for that benefit remains unknown.
DISCUSSION

Fifteen populations of E. coli evolved for 150 two-day cycles of a freeze-
thaw—growth (FTG) regime. Every other day, each population was diluted 1:100
in glucose minimal medium, during which time the cells grew to stationary phase
and exhausted the available glucose; in alternating days, the population was
frozen at -80°C, without cryoprotectant, for 22.5 h and then thawed for 1.5 h at
room temperature; and the cycle was then repeated by diluting the cells in fresh
medium. Our previous research found that this regime was highly stressful, with
substantial mortality during each freeze-thaw (FT) cycle and with much longer
lags prior to re-commencing growth following the FT treatment than after

stationary phase at 37°C (Sleight and Lenski 2007). Our earlier work also

97

established that these FTG-evolved lines adapted to this regime by improving
both their FT survival and the speed with which surviving cells began to grow
following dilution in fresh medium and incubation at 37°C (Sleight and Lenski
2007). In this study, we sought to identify the genetic differences between the
FTG-evolved lines and their progenitors, and to characterize the resulting
mutations for evidence concerning their improved survival and recovery.

We pursued two approaches to find relevant mutational differences
between the FT G-evolved lines and their progenitors. First, we identified six
candidate genes whose described functions impact a number of physiological
processes that are important for FT survival and subsequent recovery. These
processes include cold-shock response, membrane fluidity, cryoprotectant
synthesis, regulation of stress responses, protein translation, and scavenging of
oxygen radicals (Cox and Heckly 1973; Sinensky 1974; Grecz et al. 1980;
Calcott and Gargett 1981; Mazur 1984; Calcott 1985; Kandror et al. 2002;
Willimsky et al. 1992; Gao and Crister 2000; Stead and Park 2000). We
sequenced six key genes involved in these processes (cspA, fabF, otsA, rpoS,
nsA, and sodA) in each of two FTG-evolved lines and the ancestor, but we found
no mutations in any of them.

We then pursued a global approach in which we screened the entire
genomes of clones from many of the FTG-evolved lines and their progenitors by
an RFLP approach using native IS elements as probes (Naas et al. 1994;
Papadopoulos et al. 1999; Schneider et al. 2000; Cooper et al. 2001b; de Visser

et al. 2004; Schneider and Lenski 2004). This approach led to the discovery of

98

new IS insertions in multiple FTG-evolved lines at each of two loci, cls and
uspA/B, and we therefore systematically screened those loci for mutations in ten
clones sampled from each of 15 FTG-evolved lines. In all, we found four FTG-
evolved lines with new cls mutations, including three different IS1 50 insertions
and one 11-bp deletion (Figure 16). We also found IS186 insertions in two other
FTG-evolved lines but, in both cases, these insertions were already present in
their progenitors, which had evolved at constant 37°C for 20,000 generations as
part of an earlier experiment; hence, these IS186 insertions do not contribute to
adaptation specific to the FTG regime, and they were not considered further.
The screening also revealed that 8 of the 15 FTG-evolved lines had I8150
insertions in the uspA/B intergenic region; remarkably, all these insertions had
the same location and orientation (Figure 17). The possibility that inadvertent
cross-contamination of lines might account for the identical mutations in uspA/B
was excluded by showing that the FTG-evolved lines had different alleles at
another locus, one in which their progenitors had evolved distinguishing alleles
as part of the previous experiment at 37°C (nadR, Woods et al. 2006). Parallel
genetic changes, as observed in cls and uspA/B, are widely regarded as strong
evidence that the resulting phenotypic changes are adaptive (Bull et al. 1997;
Ferea et al. 1999; Wichman et al. 1999; Cooper et al. 2001 b, 2003; Colosimo et
al. 2005; Wood et al. 2005; Pelosi et al. 2006; Woods et al. 2006). Therefore, we
sought to examine the effects of the evolved cls and uspA/B alleles on fitness
under the FTG regimes, as well as the physiological bases thereof, by making

isogenic strains that differed only at those loci.

99

Physiological significance of the evolved mutations in cls

The cls gene encodes the cardiolipin (CL) synthase enzyme, which
catalyzes condensation of two phosphatidylglycerol (PG) molecules to form CL
and glycerol (Hirschberg and Kennedy 1972). CL is one of the three main
phospholipids in E. coli, the other two being phosphatidylethanolamine (PE) and
PG (Ames 1968; Cronan 1968; Cronan and Vagelos 1972). The relative
amounts of these phospholipids depend on the physiological condition of cells;
PG is most abundant in exponentially growing cells, while CL is dominant in
stationary-phase cells and also under other conditions when cellular energy
levels are lowered (Ames 1968; Cronan 1968; Cronan and Vagelos 1972;
Shibuya and Hiraoka 1992; Heber and Tropp 1991). Thus, loss-of-function
mutations in the cls gene lead to excess PG that cannot be converted to CL
during stationary phase (Cronan and Vagelos 1972; Pluschke et al. 1978;
Pluschke and Overath 1981).

We used the cls deletion mutation that was substituted in population A-
2/FTG to test for effects on fitness and cell physiology because it was more
readily manipulable than the IS150 insertions that arose in three other FTG
populations. By replacing this evolved cls allele with the ancestral allele, we
found that the evolved allele contributed substantially to the fitness gain of this
line under the FTG regime (Figure 18A). It did so primarily by improving survival
during the FT treatment (Figure 18B), although recovery of subsequent growth
also improved somewhat in with the evolved cls" allele relative to the ancestral

cls+ allele (Figure 180). The evolved cls" allele conferred no advantage in

100

competition under the same conditions except without a FT cycle (Figure 18D),
showing the specificity of its beneficial effect with respect to the FTG regime.
When this same cls" allele was moved into the genetic background of the
progenitor, it provided only a slight advantage under the FTG regime (Figure
18A). Thus, while the cls‘ allele is highly beneficial to the FTG line in which it
arose, much of its advantage depends on epistatic interactions with one or more
other mutations that were substituted in that lineage. The same overall
conclusions about the beneficial effect of the evolved cls" allele, and the
dependence on genetic background, are supported by comparing growth
trajectories of these strains following either the FT treatment or stationary phase
at 37°C (Figure 19).

A temperature downshift causes a phase transition in a cell membrane
from a fluid to a more rigid, gel-like state (Hazel 1995). Prior research has shown
that the increased PG content in cls mutants increases membrane fluidity,
leading to a 6°C decrease in the midpoint of this phase transition (Pluschke and
Overath 1981). This increased fluidity probably occurs because PG has a
smaller head group than CL, which is the condensation of two PG molecules. In
general, phospholipids with smaller head groups are more cone-shaped than
cylindrical, which increases hydrogen-bound interstitial water and allows the
molecules to change shape more easily (Slater et al. 1994).

The effect of this evolved cls deletion mutation on membrane fluidity was
tested using fluorescence anisotropy with the amphipathic probe Iaurdan. No

single technique for characterizing membrane fluidity is sensitive to the entire

101

range of lipid motions, and estimates of fluidity therefore depend on the motions
that can be detected by particular methods (Hazel 1995). Laurdan localizes near
the phospholipid head groups in the cytoplasmic membrane, and hence is useful
for examining differences in membrane fluidity at the water-lipid interface. This
cls mutation significantly increases fluidity (i.e., reduces anisotropy) after a FT
cycle in both the evolved and progenitor backgrounds (Figure 20A), but it has no
significant effect on fluidity during stationary phase at 37°C (Figure 208). The
anisotropy value for this evolved line after freezing and thawing was nearly the
same as when measured in stationary phase at 37°C, indicating that it was better
able to maintain the same membrane fluidity in the face of the FT stress than
was its progenitor, whose membrane became much more rigid (higher
anisotropy) after the FT treatment. Similar changes in membrane fluidity after
freezing and thawing were also seen in FTG-evolved lines that had new IS150
insertions in the cls gene; again we saw no differences in fluidity between these
evolved lines and their progenitors during stationary phase at 37°C. Freezing
and thawing cause changes in hydration, and so the fact that anisotropy values
differ among strains only after the FT treatment might indicate a hydration-
dependent molecular rearrangement in the head-group region, as other head
groups are known to respond to changes in hydration (Hsieh et al. 1997).

Our findings support previous research showing that membrane fluidity
plays an important role in allowing organisms to tolerate various stresses over
physiological and evolutionary timescales (Ramos et al. 1997; Beney and

Gervais 2001), including especially stresses related to temperature (Marr and

102

lngraham 1962; Sinensky 1974; Behan-Martin et al. 1993; Herman et al. 1994;
Nedwell 1999). To compensate for the increased membrane rigidity caused by a
temperature downshift, adaptive responses often increase membrane fluidity, a
phenomenon known as “homeoviscous adaptation” (Sinensky 1974; Hazel
1995). This physiological response is typically associated with an increase in the
unsaturation of fatty acids. At the outset of our study, we therefore chose fabF
as a candidate gene for evolutionary adaptation to the FTG regime. That gene
encodes a long-chain acyl-ACP elongation enzyme that contributes to fatty-acid
unsaturation. However, no mutations were found when we sequenced fabF in
two of the FTG-evolved lines. Furthermore, fluorescence polarization
measurements using the probe DPH, which detects differences in fatty acyl
chains in the membrane, found no differences between multiple FTG-evolved
lines and their progenitors either after a FT treatment or during stationary phase.
Evidently, our choice of membrane fluidity as a candidate evolutionary response
was appropriate, but the particular gene target was different from the one that we
first chose to investigate.

In any case, the increased membrane fluidity associated with the evolved
cls alleles, and other mutations not yet found, may promote FT survival in several
ways. Disruption of the plasmid membrane is the primary cause of FT injury,
which leads to changes in osmotic behavior and, potentially, mechanical failure
leading to cell death (Steponkus 1984). Membrane fluidity reduces FT injury and
promotes survival during freezing and thawing in bacteria and other organisms

(Kruuv et al. 1978; Beney and Gervais 2001). We speculate that increased

103

membrane fluidity in FTG conditions may be beneficial for a number of reasons,
including: improved membrane protein function (Letellier et al. 1977; Cronan
1978; Hazel 1995), greater membrane integrity during contractions and
expansions caused by dehydration and rehydration, respectively (Steponkus
1984), reduced ice nucleation on the membrane surface (Mindock et al. 2001),
and faster recovery of growth since DNA replication requires a fluid membrane
(Castuma, et al. 1993).
Physiological significance of the evolved uspA/B mutations

Eight of the 15 FTG-evolved lines acquired IS150 insertions in the exact
same position and orientation in the uspA/B intergenic region (Figure 17). As
noted earlier, the possibility of accidental cross-contamination was excluded
because the FT G-evolved lines had the appropriate alleles at other loci that
distinguished them. Removal of this insertion element (i.e. restoration of the
ancestral uspA/B gene) from a FTG-evolved line caused a significant reduction in
fitness under the FT G regime, indicating that the insertion is beneficial (Figure
21A). This effect was precisely duplicated for two independent strain
constructions, and the fitness advantage of the uspA/B::IS150 accrues during
both FT survival and subsequent recovery of growth (Figures 21 B and 21 C). The
evolved uspA/B allele confers no significant advantage when fitness assays are
performed under the same conditions except without the FT treatment (Figure
21 D). Therefore, as we also saw for the cls, the benefit of the evolved uspA/B

state is specific to the FTG regime.

104

The location of these uspA/B::IS150 insertions in relation to the known
regulatory regions suggests that they should disrupt expression of uspB more
strongly than they would interfere with uspA (Figure 17). This expectation was
confirmed by performing RT-PCR, which showed that uspB transcription was
severely reduced when the insertion was present under all conditions examined
including during stationary phase, after FT treatment, and during subsequent
recovery of growth (Figure 23), whereas the insertion had little effect on uspA
transcription under these same conditions.

The uspB gene has a putative os promoter (Figure 17), and it should
therefore be up-regulated under various stressful conditions (Farewell et al.
1998). Yet, although UspB is named a universal stress protein, it is
demonstrably beneficial for freeze-thaw survival and recovery to dramatically
reduce its expression. Based on its sequence, UspB is probably a membrane-
associated protein, and uspB mutants have been shown to be hypersensitive to
ethanol in stationary phase (Farewell et al. 1998). Ethanol fluidizes membranes,
and therefore the physiological response to ethanol is to increase membrane
’ rigidity (Dombek and Ingram 1984). Based on these reported effects, we
hypothesized that the lS-mediated disruption of uspB transcription increases
membrane fluidity, which would benefit cells during freezing and thawing, but
which should reduce their survival during exposure to ethanol. To test our
hypothesis, we compared isogenic strains with and without the IS150 insertion
that disrupts uspB transcription with respect to both membrane fluidity and

ethanol sensitivity. We saw no differences in membrane fluidity, either during

105

stationary phase or after the FT treatment, using both Iaurdan and DPH as
probes for fluorescence anisotropy. There were also no significant differences
between these strains in their ethanol-induced mortality during stationary phase
or after FT treatment. The physiological basis for the advantage of the
uspA/B::IS150 mutant under the FTG regime therefore remains unknown. We
are tempted to speculate that this mutation is somehow involved with changes in
membrane fluidity, despite our failure to find any such effect, given the facts that
membrane fluidity is sensitive to the entire range of lipid motions and that no
single method can detect all the potentially relevant motions (Hazel 1995).

Finally, the fact that the uspA/B::IS150 insertion is demonstrably beneficial
under the FTG regime does not exclude the possibility that the genomic position
where it inserted is a “hotspot” for this class of mutation. For example, Cooper et
al. (2001b) found that lS-mediated deletions of the ribose-catabolic operon
evolved quickly and repeatedly in replicate populations of the long-term
experiment in glucose at 37°C for two reasons: the deletion mutants had a
competitive advantage and, moreover, an IS element adjacent to the operon in
the ancestral strain evidently easily recombined with new insertions to produce
deletions at a very high rate at this ribose locus. In the present context, to search
for possibly relevant sequence motifs near IS150 insertion target sites in the
FTG-evolved lines (Table 4), we employed the MAST program (Bailey and
Gribskov 1998) to look for a specific sequence motif using 100-bp regions on
either side of each target site. No obvious motifs were found among these

sequences, although we found a 6-bp DNA stretch (‘GGGGCT’) 4-bp

106

downstream from the uspA/B target site that exactly matches an IS150 insertion
into an existing IS1 element (Hall et al. 1989). A few other IS150 insertions in
our study have some similarity to this 6-bp motif, although with mismatches.
However, given the short length of the motif and the small sample of unique
IS150 insertions in our study, its biological relevance remains unclear.

In conclusion, we found two loci, cls and uspA/B, in which several lines
that had evolved under a FTG regime independently acquired mutations.
Construction of, and competition between, isogenic strains that differ only at
those loci demonstrate that the evolved alleles in both genes contribute
significantly to the fitness gains under the FTG regime, including improved FT
survival and faster recovery of subsequent growth. In the case of cls, which
encodes an enzyme involved in the conversion of phospholipids, this mutation
helps maintain membrane fluidity following FT treatment. The mutations in the
uspA/B intergenic region severely disrupt transcription of uspB, but the
physiological basis for the resulting benefit remains unknown. In addition, other
mutations in unknown genes also contribute to adaptation to the FTG regime,

some of which interact epistatically with the cls and uspA/B mutations.

107

FIGURES

 

 

37°C Growth
Ancestor ’
20,000 generations
FTG for FTG for
1,000 generations 1,000 generations

 

Figure 14. Evolutionary histories of individual freeze-thaw-growth (FTG)
populations. The ancestor of a long-term evolution experiment (Lenski et al.
1991) was used to found three FTG-evolved populations. Twelve other FTG-
evolved populations were founded by clones sampled from each of 12
populations that previously evolved for 20,000 generations at 37°C. All 15 of the
FTG lines evolved for 1.000 generations under the FT G regime, with alternating
days of a FT cycle and growth at 37°C. The culture medium used for growth was
the same as the one used in the long-term evolution experiment.

 

108

12 3f. 5 0 7 8 01011121314151017181920

W
,.
kb Q
3
10' ‘ '-
s b nun-- ‘ I" ‘9':
s > .5 _ i"
5 p ad” . “I -
U”
4> on uuouvvv'll"' '
3 b . .....:‘.. .':.:' ’-
o. 'C

2 b ’ .... 6 get:
1.5» 9 ® @“PA’B

Figure 15. Southern hybridization of Hincll-digested genomic DNA with the
IS150 probe. Sizes of bands are given in kb (kilobases) on the left. Insertion
bands that occur in multiple FTG-evolved clones are circled and labeled either
cls or uspA/B for the affected genes. The clones are numbered for each lane at
top, where each FTG-evolved clone follows its direct progenitor (odd numbers
are progenitors and even numbers are FTG-evolved clones). Lanes 1-20 are as
follows: 1) Ancestor, 2) AncB/FTG, 3) A+1, 4) A+1/FTG, 5) A+2, 6) A+2/FTG, 7)
A+3, 8) A+3/FTG, 9) A+4, 10) A+4/FTG, 11)A+6, 12) A+6/FTG, 13) A-1, 14) A-
1/FTG, 15) A-2, 16) A-2/FTG, 17) A-3, 18) A-3/FTG, 19) A—4, 20) A-4/FTG.

109

 

 

 

 

M cls (1461 bp) 17T< kch I'—
J I I Io

———>

 

 

A-2/ A+4/ A-1/ A'5
F TG FTG FTG
A-5 A-3/
FTG

Figure 16. Schematic representation of mutations in cls. Block arrows
indicate the direction of transcription for each gene and are drawn to scale.
Upward pointing arrows indicate the location of IS150 (A+4/FTG, A-1/FTG, A-
3/FTG) and IS186 (A-5, A-6) insertions, and an 11-bp deletion (A-2/FTG) in FTG-
evolved clones or progenitors already harboring insertions. The A+4/FTG and A-
3/FTG insertions were originally discovered by the IS150 Southern hybridization
(Figure 15), while all other mutations were found after screening clones in all
other FT G-evolved lines for mutations within cls. The IS150 insertion in the A-
1/FTG population occurred in a clone not tested in the original IS150 Southern
hybridization. The A-2/FTG mutation causes a premature stop codon 43 bp
downstream. All mutations occurred in the open reading frame (ORF) except in
the case of A-6, which is upstream of the cls transcription start site. The bent
arrow indicates the putative transcription start site of cls, and the sigma factor
(07°) responsible for its transcription is also shown (lvanisevic et al. 1995). The
yciU gene (330 bp) encodes a hypothetical protein and is predicted to be
transcribed in the same direction as cls. In this case, the cls mutations may have
polar effects on yciU transcription. The kch gene (1254 bp) is a potassium VIC
(Voltage-gated Ion Channel).

110

I"

[ 15150 (1443 bp)>

 

 

 

 

 

 

 

——<uspA (435 bp) I +11 I: I uspB (336 bp)>—

261 bp

—
——

356 bp

390 bp

Figure 17. Schematic representation of IS150 insertions in uspA/B
intergenic region. Block arrows indicate the direction of transcription for each
gene and are not to scale. Eight of the 15 FTG populations had at least one
tested clone with an IS150 insertion in the exact same position and orientation,
34 bp upstream of the uspB start codon. This insertion generated a 3- b-p
duplication at the target site and is 95 bp downstream of the putative uspB
promoter (oS ). The IS150 element has at least one possible promoter-like
sequence directed outward (Schwartz et al. 1988) towards uspB.

111

Figure 18.

(A)
2.0

1.8

Relative FTG Fltnees
I 5‘.

a
N

1.0

0.8

(B)

3.0

2.5

1.5

WFT Survlvd

1.0 '

0.5 ~'

 

 

 

 

 

 

A-2

cls .

 

 

 

 

 

A-2

cls-

A-2

+
cls

 

112

 

 

 

I—IH

 

 

 

A-2/F‘I'G
cls +

 

AoZIFTG

cls

 

 

 

 

 

 

 

 

A-2/FT G
cls +

A-ZIFTG
cls '

 

 

Figure 18 (continued).

(C)
1.40

1.35

1.30

1.00
0.95

0.90

(D)

1.04

1 .02

.0 .0 ."
8 8 8

Relative Growth Fitness

12

0.92

0.90

 

 

 

 

 

 

cls '

 

 

 

A-2
cls

 

P—(r—I

 

 

 

A-2
4.
cls

 

A-2
cls +

113

 

 

 

I—TF—T'I

 

 

 

A-2/FT G
cls +

 

 

 

 

A-2IFT G
cls

 

 

 

 

A-2/FT G
cls +

 

 

 

A-2/FT G

 

 

Figure 18. Relative fitness components of the following clones: A-2 cls'
construct, A-2 cls+ progenitor, A-2/FTG cls+ construct, and A-2/FTG cls‘ evolved
clone. All four types competed against an Ara+ mutant of the A-2 progenitor.
Fitness components are as follows: (A) overall FTG fitness, (B) FT survival, (C)
growth after FT, and (D) growth after stationary phase. Error bars are 95%
confidence intervals based on eight replicate assays.

114

Figure 19.

(A)
0.035 .
— A°2 d3
'0-
- - A-2 cls
0.03
0.025
0.02

0.015

OpticeIDeneIty

0.01

0.005

 

0

°Assessesseeseessseesseee

-0.005
Time (minutes)
(8)
0.035 A 2 cls-
- - A-2 cls *
0.03
0.025
0.02

0.015

OpticelDeneIty

0.01

0.005

 

0

°sseesesseseeeesesessesee
0.005
Time (minutes)

115

Figure 19. Growth dynamics following (A) FT treatment and (B) stationary
phase at 37°C. Clones shown are as follows: A-2 cls' construct (solid black), A-2
cls+ progenitor (dashed black), A-2/FTG cls+ construct (dashed grey), and A-
2/FTG cls‘ evolved clone (solid grey). Each curve is the average of 18 replicate
trajectories.

116

Figure 20.

(A)
0.050

0.045
0.040

0.035

0.020

Laurdan Anisotropy
2
8’.

0.015
0.010
0.005
0.000

(3)

0.030

0.025

0.020

0.015

LemdenAnisotropy

0.010

0.000

 

 

 

A-2

cls '

 

 

 

A—2

+
cls

 

 

 

 

 

 

 

A-2
cls '

A-2/FTG
cls +

 

 

 

 

 

 

 

 

 

A-2
cls

I—F—vl

A-2/FTG
cls

 

 

 

 

 

 

 

A-2/FTG
cls +

 

 

 

 

A-ZIFT G
cls '

 

 

 

Figure 20. Laurdan anisotropy differences following (A) FT treatment and
(B) stationary phase. Each clone was incubated with the fluorescent probe
Iaurdan, and fluorescence anisotropy was measured individually in a fluorometer
with 18-fold replication. Lower anisotropy values are indicative of a more fluid
membrane in the context of the water-lipid interface. See Materials and Methods

for details.

118

Figure 21.
(A)

2.0

1.3 w i

 

 

I——1

1.6

Relative FTG Fitness
52

1.2

 

 

 

1.0 , -—-§-=—=—

 

 

 

Ancestor AncB/FTG AncB/FTG
uspA/B 48150 uspA/B +IS150 um 48150

0.8 ~
(3)

1.8

1.7

1.6 ‘

 

1.5

 

‘
a.
——1

 

Relative FT Survival
3; .
.—_

 

 

 

 

 

 

 

 

1.2 -
1f
1.1
. Ifl___l
Ancestor AncB/FTG AncB/FT G
0.9 - uspA/B 45150 uspA/B +IS150 uspA/B 48150
0.0 ~

119

Figure 21 (continued).

(C)

1.8
1.7

1.6

.e
(7’

:"
§

.5
N

Relative Growth Afbr FT
a»

d
—I

.A
O

0.9

0.8 -

(D)

1.08

1.06

RelativeGrowthFltness

1.00

0.98

0.96

1.02 '

 

 

 

 

 

 

 

1::

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ancestor AncBIFT G AncBIFT G
uspA/B -IS150 uspA/B +IS150 uspA/B 48150
Ancestor AncB/FT G AncBIFT G
uspA/B -IS150 uspA/B +IS15O uspA/B 48150

120

Figure 21 . Relative fitness components of the following clones: ancestor
(uspA/B -IS 150), evolved clone AncB/FTG uspA/B +IS150, and the
constructed clone AncB/FTG uspA/B -IS 150. All three types competed
against the ancestor with the opposite arabinose marker. Each relative fitness
component is as follows: (A) overall FTG fitness, (B) FT survival, (C) growth after
FT, and (D) growth after stationary phase. Error bars are 95% confidence
intervals of 12 replicate assays.

121

Figure 22.

 

(A)
0.03
- - AncestoruspA/B-IS150
—— AncB/FTG uspA/B +13150
0-025 - - AncB/FTG uspA/3-18150
0.02

0.015

Optical Density
9
9.

0.005

0 -"

-"
--------- .-
-
.gUII-UIIII-eC."-

°ssseessseeeesessseeeeeee
-0.005

C
e
ou-‘T'-

Time (minutes)

(B)
0.03
- - AncestoruspA/B 48150
-— AncB/FTG uspA/B +IS15O
0.025 - - AncB/FTG uspA/B -IS150
0.02

0.015

OpticeIDensrty

0.005

 

O
°ssecesseseeeeesssesessee

-0.005
Time (minutes)

122

Figure 22. Growth dynamics following (A) FT treatment and (B) stationary
phase at 37°C. Clones shown are as follows: Ancestor uspA/B -IS150 (dashed
black), evolved clone AncB/FT G uspA/B +IS150 (solid grey), and constructed
clone AncB/FTG uspA/B —IS150 (dashed grey). Each curve is the average of 18
replicate trajectories.

123

CI Ancestor uspA/B 48150
I AncB/FTG uspA/B +IS150
I AncB/FT G uspA/B 48150

 

 

 

 

 

 

I———-I

 

 

 

 

   

 

 

 

Stationery Phase After FT Growth

Figure 23. Relative uspB transcription differences using Real-Time PCR.
uspB transcription was measured relative to 168 rRNA transcription as an
endogenous control at stationary phase, after a FT cycle, and in growth
conditions 2 h after thawed cultures were diluted into fresh media. Clones are as
follows: Ancestor uspA/B —IS150 (white), evolved clone AncB/FTG uspA/B
+IS150 (grey), and constructed clone AncB/FTG uspA/B -IS150 from which the
insertion element was removed (black). The height of each bar represents the
mean of two independent RNA extractions, each set with 3-fold replication. Error
bars are 95% confidence intervals based on the six-fold replication.

124

TABLES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Clone IS1 I82 I83 I84 I830 IS150 IS 186 Total
AncB/FTG 0 0 0 0 0 3 0 3
A+1/FTG 0 0 3 0 0 3 1 7
A+2/FTG 1 0 3 0 0 0 0 4
A+3/FTG 0 0 0 0 0 0 0 0
A+4/FTG 0 0 0 0 0 4 0 4
A+6/FTG 2 0 0 0 0 2 1 5
A-1/FTG 0 0 0 0 0 0 0 0
A-2/FTG 0 0 0 0 0 0 1 1
A-3/FTG 1 0 0 0 0 4 0 5
A-4/FTG 0 0 0 0 0 1 0 1
Total 4 0 6 0 0 1 7 3 30

 

 

 

 

 

 

 

 

 

Table 3. IS changes in FTG-evolved lines relative to progenitors. IS
changes are defined as appearance or disappearance of a hybridizing band. A
single lS-associated mutation may cause more than one IS change.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Target site Location in
Population Mutation duplication nc_000913 ‘
A+4/FTG IS150 insertion in cls 3 bp (ATC) 1305917-19
A-1/FTG IS150 insertion in cls 3 bp (GCG) 1306467-69
A-2/FTG 11-bp deletion in cls N/A 1305473-83
A-3/FTG IS150 insertion in cls 3 bp (CGA) 1305865-67
A-5 Progenitor IS186 insertion in cls 6 bp (TGGATG) 1305580-85
A-6 Progenitor IS186 insertion in cls 6 bp (GTATAA) 1306774-79
AncA/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
AncB/FTG IS150 insertion In uspA/B 3 bp (CCC) 3637778-80
A+1/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
A+2/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
A+4/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
A-3/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
A-5/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80
A-6/FTG IS150 insertion in uspA/B 3 bp (CCC) 3637778-80

 

 

 

 

Table 4. Mutations in cls and uspA/B genes.

" Pubmed Accession Number for the E. coli K-12 genome (Riley et al. 2006).

125

 

 

 

 

 

 

 

Timepoint 100 200 300 400 500 600 700 800 900 1000
A+4IFTG 1/8 1/9 2/8 6/9 6/8 4/5 0/3 5/6 5/5 3/10
A-1IFTG 0/5 0/5 0/5 0/5 0/1 0/4 n.d. 2/3 4/4 1/ 10
A-3/FTG 0/5 0/5 1 I5 2/5 4/4 4/4 3/ 5 2/3 n .d. 10l1 0

 

 

 

 

 

 

 

 

 

 

 

Table 5. cls mutations in FTG-evolved lines through time. FTG-evolved
populations having IS150 mutations in cls at generation 1000 were tracked every
100 generations during FTG evolution. The numerator is the number of clones
harboring an IS element as determined by PCR. The denominator is the number
of clones tested (n.d., not determined).

 

 

 

 

 

 

 

 

 

 

Timepoint 1 00 200 300 400 500 600 700 800 900 1 000
AncAI 0/9 0/8 0/9 0/9 0/17 0/ 10 1/6 0/10 6/12 1/10
:23] 0/7 0/10 3/10 8/10 7/8 3/8 15/32 5/9 6/8 10/10
:TI-SFTG 0/10 9/10 9/10 8/10 8/9 10/10 8/8 4/5 7/11 6/10
A+2lFTG 0/10 0/10 0/8 0/9 1/8 0/8 2/9 0/12 0/12 1/10
A+4IFT G 0/8 8/9 10/10 6/6 6/8 9/10 9/10 6/7 9/1 1 9/22
A-3IFT G 7/10 6/7 9/10 7/9 n.d. 9/9 8/8 12/ 12 10/ 10 10/10
A-5IFTG 0/10 9/12 0/9 9/9 10/10 12/12 10/10 8/8 9/9 10/10
A-SIFT G 0/10 1/7 0/8 0/8 0/10 0/5 2/8 0/12 5/6 10/10

 

 

 

 

 

 

 

 

 

 

 

Table 6. uspA/B mutations in FTG-evolved lines through time. FTG-evolved
populations having IS150 mutations in the uspA/B intergenic region at generation
1000 were tracked every 100 generations during FTG evolution. The numerator
is the number of clones harboring an IS element as determined by PCR. The

denominator is the number of clones tested (n.d., not determined).

126

 

 

 

 

 

 

 

 

 

 

Source of variation DF 88 MS F p

Allele 1 0.00109 0.00109 23.457 < 0.0001
Background 1 0.00245 0.00245 52.779 < 0.0001
Allele x Background 1 0.00002 0.00002 0.528 0.4700
Error 68 0.00316 0.00005

Total 71 0.00672

 

 

 

 

 

 

Table 7. 2-way ANOVA testing for differences in the cls allele, progenitor or
evolved background, and interaction effects in Iaurdan anisotropy after

freeze-thaw treatment.

 

 

 

 

 

 

 

 

 

 

 

 

Source of variation DF SS MS F p
Allele 1 0.00005 0.00005 0.825 0.367
Background 1 0.00001 0.00001 0.141 0.708
Allele x Background 1 0.00004 0.00004 0.769 0.384
Error 68 0.00385 0.00006

Total 71 0.00395

 

Table 8. 2-way ANOVA testing for differences in the cls allele, progenitor or
evolved background, and interaction effects in Iaurdan anisotropy after

stationary phase.

127

 

 

REFERENCES

Ames, GP 1968. Lipids of Salmonella typhimurium and Escherichia coli:
structure and metabolism. J Bacteriol 95:833-43.

Aricha, B., l. Fishov, Z. Cohen, N. Sikron, S. Pesakhov, l. Khozin-Goldberg, R.
Dagan, and N. Porat. 2004. Differences in membrane fluidity and fatty acid
composition between phenotypic variants of Streptococcus pneumoniae. J
Bacteriol 186:4638-44.

Bailey, T.L., and M. Gribskov. 1998. Combining evidence using p-values:
application to sequence homology searches. Bioinformatics 14:48-54.

Bakermans C., A.l. Tsapin, V. Souza-Eqipsy, D.A. Gilichinsky, and K.H. Nealson.
2003. Reproduction and metabolism at -10°C of bacteria isolated from
Siberian permafrost. Environ Microbiol 42321-326.

Behan-Martin, M.K., G.R. Jones, K. Bowler, and AR. Cossins. 1993. A near
perfect temperature adaptation of bilayer order in vertebrate brain
membranes. Biochim Biophys Acta 1151 :216-22.

Bell G. 1997. Selection: The Mechanism of Evolution. New York: Chapman and
Hall.

Beney, L., and P. Gervais. 2001. Influence of the fluidity of the membrane on the
response of microorganisms to environmental stresses. Appl Microbiol
Biotechnol 57:34-42.

Beney, L., Y. Mille, and P. Gervais. 2004. Death of Escherichia coli during rapid
and severe dehydration is related to lipid phase transition. Appl Microbiol
Biotechnol 65:457-64.

Bennett A.F., R.E. Lenski, and J.E. Mittler. 1992. Evolutionary adaptation to
temperature. I. Fitness responses of Escherichia coli to changes in its
thermal environment. Evolution 46:16-30.

Bennett A.F., and RE. Lenski. 1993. Evolutionary adaptation to temperature. II.
Thermal niches of experimental lines of Escherichia coli. Evolution 47:1-
12.

Bennett A.F., and RE. Lenski. 1997. Phenotypic and evolutionary adaptation of a
model bacterial system to stressful thermal environments. Pp. 135-154 in
R. Bijlsma and V. Loeschcke, eds. Environmental Stress, Adaptation and
Evolution. Birkhauser, Basel, Switzerland.

128

Bennett A.F., and RE. Lenski. 1999. Experimental evolution and its role in
evolutionary physiology. Am Zool 39:346-362.

Blasband, A.J., W.R. Marcotte, Jr., and CA. Schnaitman. 1986. Structure of the
lc and nmpC outer membrane porin protein genes of lambdoid
bacteriophage. J Biol Chem 261:12723-32.

Blot, M. 1994. Transposable elements and adaptation of host bacteria. Genetica
9325-12.

B6ck A., R. Curtiss Ill, J.B. Kaper, F.C. Neidhardt, T. Nystrbm, K. E. Rudd, and
CL. Squires, eds. 2006. Escherichia coli and Salmonella: Cellular and
Molecular Biology. ASM Press, Washington, DC. [Published online at
www.ecosal.org]

 

Bongers, R.S., M.H. Hoefnagel, M.J. Starrenburg, M.A. Siemerink, J.G. Arends,
J. Hugenholtz, and M. Kleerebezem. 2003. lSQ81-mediated adaptive
evolution recovers lactate production by lth transcription activation in a
lactate dehydrogenase—deficient strain of Lactococcus lactis. J Bacteriol
185:4499-507.

Bull, J.J., M.R. Badgett, H.A. Wichman, J.P. Huelsenbeck, D.M. Hillis, A. Gulati,
C. Ho, and l.J. Molineux. 1997. Exceptional convergent evolution in a
virus. Genetics 147:1497-507.

Burch C.L, and L. Chao. 2000. Evolvability of an RNA virus is determined by its
mutational neighbourhood. Nature 406:625-628.

Calcott PH, and RA MacLeod. 1975. The survival of Escherichia coli from
freeze-thaw damage: the relative importance of wall and membrane
damage. Can J Microbiol, 12:1960-1968.

Calcott PH, and AM. Gargett. 1981. Mutagenicity of freezing and thawing.
FEMS Microbiol Lett 10:151-155.

Calcott P.H., D. Wood, and L. Anderson. 1983. Freezing and thawing induced
curing of drug-resistance plasmids from bacteria. Cryo-Lett 4:99.

Calcott PH. 1985. Cryopreservation of microorganisms. CRC Crit Rev Biotech
4:279-297.

Castuma, C.E., E. Crooke, and A. Kornberg. 1993. Fluid membranes with acidic
domains activate DnaA, the initiator protein of replication in Escherichia
coli. J Biol Chem 268:24665—8.

129

Charlesworth, B., P. Sniegowski, and W. Stephan. 1994. The evolutionary
dynamics of repetitive DNA in eukaryotes. Nature 371 :215-20.

Charlier, D., J. Piette, and N. Glansdorff. 1982. IS3 can function as a mobile
promoter in E. coli. Nucleic Acids Res 10:5935—48.

Chippindale A. 2006. Experimental evolution. Pp. 482-501 in C.W. Fox and J.B.
Wolf, eds. Evolutionary Genetics: Concepts and Case Studies. Oxford
Univ Press, Oxford, UK.

Ciampi, M.S., M.B. Schmid, and JR. Roth. 1982. Transposon Tn10 provides a
promoter for transcription of adjacent sequences. Proc Natl Acad Sci U S
A 79:5016-20.

Cloutier J., D. Prevost, P. Nadeau, and H. Antoun. 1992. Heat and cold shock
protein synthesis in arctic and temperate strains of rhizobia. Appl Environ
Microbiol 58:2846-53.

Colosimo, P.F., K.E. Hosemann, S. Balabhadra, G. Villarreal, Jr., M. Dickson, J.
Grimwood, J. Schmutz, R.M. Myers, D. Schluter, and D.M. Kingsley. 2005.
Widespread parallel evolution in sticklebacks by repeated fixation of
Ectodysplasin alleles. Science 307:1928-33.

Cooper T.F., D.E. Rozen, and RE. Lenski. 2003. Parallel changes in gene
expression after 20,000 generations of evolution in Escherichia coli. Proc
Natl Acad Sci USA 100:1072-1077.

Cooper V.S., and RE. Lenski. 2000. The population genetics of ecological
specialization in evolving E. coli populations. Nature 407:736-739.

Cooper V.S., A.F. Bennett, and RE. Lenski. 2001a. Evolution of thermal
dependence of growth rate of Escherichia coli populations during 20,000
generations in a constant environment. Evolution 55:889-896.

Cooper V.S., D. Schneider, M. Blot, and RE. Lenski. 2001b. Mechanisms
causing rapid and parallel losses of ribose catabolism in evolving
populations of E. coli B. J Bacteriol 183:2834-2841.

Cox 0.8., and R.J. Heckly. 1973. Effect of oxygen upon freeze-dried and freeze-
thawed bacteria: viability and free radical production. Canadian J Microbiol
19:189-194.

Cronan, J.E., Jr. 1968. Phospholipid alterations during growth of Escherichia coli.
J Bacteriol 95:2054-61.

130

Cronan, J.E., and RR. Vagelos. 1972. Metabolism and function of the membrane
phospholipids of Escherichia coli. Biochim Biophys Acta 265:25-60.

Cronan, J.E., Jr. 1978. Molecular biology of bacterial membrane lipids. Annu Rev
Biochem 47: 1 63-89.

Crozat E., N. Philippe, R.E. Lenski, J. Geiselmann, and D. Schneider. 2005.
Long-term experimental evolution in Escherichia coli. XII. DNA topology as
a key target of selection. Genetics 169:523-532.

Cullum A.J., A.F. Bennett, and RE. Lenski. 2001. Evolutionary adaptation to
temperature. IX. Preadaptation to novel stressful environments by E. coli
adapted to high temperature. Evolution 55:2194-2202.

Deonier, RC. 1996. Native insertion sequence elements: locations, distributions,
and sequence relationships, pp. 2000—2011 in Escherichia coli and
Salmonella: Cellular and Molecular Biology, edited by F .C. Neidhardt.
ASM Press, Washington, DC.

de Visser J.A., and RE. Lenski. 2002. Long-term experimental evolution in
Escherichia coli. XI. Rejection of non-transitive interactions as cause of
declining rate of adaptation. BMC Evo Bio 2:19.

de Visser, J.A., A.D. Akkermans, R.F. Hoekstra, and WM. de Vos. 2004.
Insertion-sequence-mediated mutations isolated during adaptation to
growth and starvation in Lactococcus lactis. Genetics 168:1145-57.

Dombek, KM, and LO. Ingram. 1984. Effects of ethanol on the Escherichia coli
plasma membrane. J Bacteriol 157:233-9.

Drouin P., D. Prevost, and H. Antoun. 2000. Physiological adaptation to low
temperatures of strains of Rhizobium leguminosarum bv. viciae associated
with Lathyrus spp. FEMS Microb Ecol 32:111-120.

Eichenbaum, Z., and Z. Livneh. 1998. UV light induces lSiO transposition in
Escherichia coli. Genetics 149:1173-81.

Elena SP, and RE. Lenski. 2003. Evolution experiments with microorganisms:
the dynamics and genetic bases of adaptation. Nature Rev Gen 4:457-
469.

Eriksson M., J.O. Ka, and WW. Mohn. 2001. Effects of low temperature and

freeze-thaw cycles on hydrocarbon biodegradation in Arctic tundra soil.
Appl Environ Microbiol 67:5107-5112.

131

Farewell, A., K. Kvint, and T. Nystrom. 1998. uspB, a new sigmaS-regulated
gene in Escherichia coli which is required for stationary-phase resistance
to ethanol. J Bacteriol 180:6140-7.

Ferea, T.L., D. Botstein, P.O. Brown, and RF. Rosenzweig. 1999. Systematic
changes in gene expression patterns following adaptive evolution in yeast.
Proc Natl Acad Sci U S A 96:9721-6.

Funchain P., A. Yeung, J.L. Stewart, R. Lin, M.M. Slupska, and J.H. Miller. 2000.
The consequences of growth of a mutator strain of Escherichia coli as
measured by loss of function among multiple gene targets and loss of
fitness. Genetics 154:959-970.

Gao D., and J.K. Critser. 2000. Mechanisms of cryoinjury in living cells. ILAR
41:187-196.

Gould SJ. 1989. Wonderful Life: The Burgess Shale and the Nature of History.
New York: Norton.

Grant PR. 1999. Ecology and Evolution of Darwin '3 Finches. Princeton:
Princeton University Press.

Grecz N., T.L. Hammer, C.J. Robnett, and MD. Long. 1980. Freeze-thaw injury:
evidence for double strand breaks in Escherichia coli DNA. Biochem
Biophys Res Comm 93:1110-1113.

Hall, B.G., L.L. Parker, P.W. Betts, R.F. Dubose, S.A. Sawyer etal. 1989. lS103,
a new insertion element in Escherichia coli: characterization and
distribution in natural populations. Genetics 121:423-431.

Hall, B.G. 1999. Spectra of spontaneous growth-dependent and adaptive
mutations at ebgR. J Bacteriol 181:1149-55.

Harris, F.M., K.B. Best, and JD. Bell. 2002. Use of Iaurdan fluorescence intensity
and polarization to distinguish between changes in membrane fluidity and
phospholipid order. Biochim Biophys Acta 1565:123-8.

Harvey PH, and MD. Pagel. 1991. The Comparative Method in Evolutionary
Biology. Oxford: Oxford University Press.

Hazel, JR, 1995 Thermal adaptation in biological membranes: ls homeoviscous
adaptation the explanation? Annu. Rev. Physiol. 57: 19-42.

Heber, S., and BE. Tropp. 1991. Genetic regulation of cardiolipin synthase in
Escherichia coli. Biochim Biophys Acta 1129:1-12.

132

 

 

Hebraud M., and P. Potier. 1999. Cold shock response and low temperature
adaptation in psychrotrophic bacteria. J Mol Microbiol Biotechnol 1:211-9.

Herman, P., I. Konopasek, J. Plasek, and J. Svobodova. 1994. Time-resolved
polarized fluorescence studies of the temperature adaptation in Bacillus
subtilis using DPH and TMA—DPH fluorescent probes. Biochim Biophys
Acta 1190:1-8.

Herring, C.D., J.D. Glasner, and FR. Blattner. 2003. Gene replacement without
selection: regulated suppression of amber mutations in Escherichia coli.
Gene 311:153-63.

Herring, CD, A. Raghunathan, C. Honisch, T. Patel, M.K. Applebee, A.R. Joyce,
T.J. Albert, F.R. Blattner, D. van den Boom, C.R. Cantor, and B.O.
Palsson. 2006. Comparative genome sequencing of Escherichia coli
allows observation of bacterial evolution on a laboratory timescale. Nat
Genet 38:1406-12.

Hirschberg, CB, and E.P. Kennedy. 1972. Mechanism of the enzymatic
synthesis of cardiolipin in Escherichia coli. Proc Natl Acad Sci U S A
692648-51.

Hoffmann AA, and PA. Parsons. 1993. Evolutionary Genetics and
Environmental Stress. Oxford Univ Press, New York.

Hsieh, C.H., S.C. Sue, P.C. Lyu, and W.G. Wu. 1997. Membrane packing
geometry of diphytanoylphosphatidylcholine is highly sensitive to
hydration: phospholipid polymorphism induced by molecular
rearrangement in the headgroup region. Biophys J 73:870-7.

lngraham J. 1987. Effect of temperature, pH, water activity, and pressure on
growth. In: Escherichia coli and Salmonella: Cellular and Molecular
Biology (Edited by: Neidhardt FC). Washington, DC, ASM Press,1543-
1553.

lvanisevic, R., M. Milic, D. Ajdic, J. Rakonjac, and DJ. Savic. 1995. Nucleotide
sequence, mutational analysis, transcriptional start site, and product
analysis of nov, the gene which affects Escherichia coli K-12 resistance to
the gyrase inhibitor novobiocin. J Bacteriol 177:1766-71.

Jaurin, B., and S. Nonnark. 1983. Insertion of I82 creates a novel ampC
promoter in Escherichia coli. Cell 32:809-16.

Jeffreys A.G., K.M. Hak, R.J. Steffan, J.W. Foster, A.K. Be]. 1998. Growth,

survival and characterization of cspA in Salmonella enteritidis following
cold shock. Curr Microbiol 36:29-35.

133

Jordan, E., H. Saedler, and P. Starlinger. 1968. 00 and strong polar mutations in
the gal operon are insertions. Mol. Gen. Genet. 102: 353-363.

Kandror, O., A. DeLeon, and AL. Goldberg. 2002. Trehalose synthesis is
induced upon exposure of Escherichia coli to cold and is essential for
viability at low temperatures. Proc Natl Acad Sci U S A 99:9727-32.

Kidwell, MG, and DR. Lisch. 2001. Perspective: transposable elements,
parasitic DNA, and genome evolution. Evolution Int J Org Evolution 55:1-
24.

Kruuv, J., JR. Lepock, and AD. Keith. 1978. The effect of fluidity of membrane
lipids on freeze-thaw survival of yeast. Cryobiology 15:73-9.

Kvint, K., L. Nachin, A. Diez, and T. Nystrom. 2003. The bacterial universal
stress protein: function and regulation. Curr Opin Microbiol 62140-5.

Leebanon B., M.A. Drake. 2001. Acid stress, starvation, and cold stress affect
post-stress behavior of Escherichia coli O157:H7 and non-pathogenic
Escherichia coli. J Food Protect 64:970-974.

Lenski R.E., M.R. Rose, S.C. Simpson, and SC. Tadler. 1991. Long-term
experimental evolution in Escherichia coli. |. Adaptation and divergence
during 2,000 generations. Am Nat 138:1315-1341.

Lenski RE, and AF. Bennett. 1993. Evolutionary response of Escherichia coli to
thermal stress. Am Nat 142:S47-S64.

Lenski R.E., V. Souza, L.P. Duong, Q.G. Phan, T.N.M. Nguyen, and KP.
Bertrand. 1994. Epistatic effects of promoter and repressor functions of
the Tn10 tetracycline-resistance operon on the fitness of Escherichia coli.
Molec Ecol 3:127-135.

Lenski RE, and M. Travisano. 1994. Dynamics of adaptation and diversification:
a 10,000-generation experiment with bacterial populations. Proc Natl Acad
Sci USA 91 :6808-6814.

Lenski R.E., J.A. Mongold, P.D. Sniegowski, M. Travisano, F. Vasi, P.J. Gerrish,
T. Schmidt. 1998. Evolution of competitive fitness in experimental
populations of E. coli: what makes one genotype a better competitor than
another? Antonie van Leeuwenhoek 73:35-47.

Lenski R.E., C.L. Winkworth, and MA. Riley. 2003. Rates of DNA sequence

evolution in experimental populations of Escherichia coli during 20,000
generations. J Mol Evol 56:498-508.

134

Lenski RE. 2004. Phenotypic and genomic evolution during a 20,000-generation
experiment with the bacterium Escherichia coli. Plant Breed Rev 24:225-
265.

Leroi AM, RE. Lenski, and AF. Bennett. 1994. Evolutionary adaptation to
temperature. Ill. Adaptation of Escherichia coli to a temporally varying
environment. Evolution 48:1222-1229.

Letellier, L., H. Moudden, and E. Shechter. 1977. Lipid and protein segregation in
Escherichia coli membrane: morphological and structural study of different
cytoplasmic membrane fractions. Proc Natl Acad Sci U S A 74:452-456.

Leu J.Y., and AW. Murray. 2006. Experimental evolution of mating discrimination
in budding yeast. Curr Biol 16:280-286.

MacLean R.C., G. Bell, P.B. Rainey. 2004. The evolution of a pleiotropic fitness
tradeoff in Pseudomonas fluorescens. Proc Natl Acad Sci USA 101:8072-
8077.

Maharjan R., S. Seeto, L. Notley—McRobb, and T. Ferenci. 2006. Clonal adaptive
radiation in a constant environment. Science 313:514-517.

Mahillon, J., and M. Chandler. 1998. Insertion sequences. Microbiol Mol Biol Rev
62:725-74.

Marr, A.G., and J.L. lngraham. 1962. Effect of Temperature on the Composition
of Fatty Acids in Escherichia coli. J Bacteriol 84:1260-7.

Mazur, P. 1984. Freezing of living cells: mechanisms and implications. Am J
Physiol 247:C125-42.

Mindock, C.A., M.A. Petrova, and RI. Hollingswort. 2001. Re-evaluation of
osmotic effects as a general adaptative strategy for bacteria in sub-
freezing conditions. Biophys Chem 89:13-24.

Mongold J.A., A.F. Bennett, and RE. Lenski. 1996. Evolutionary adaptation to
temperature. IV. Adaptation of Escherichia coli at a niche boundary.
Evolution 50:35-43.

Mongold J.A., A.F. Bennett, and RE. Lenski. 1999. Evolutionary adaptation to

temperature. Vll. Extension of the upper thermal limit of Escherichia coli.
Evolution 53:386-394.

135

 

Monnet C., C. Beal, G. Corrieu. 2003. Improvement of the resistance of
Lactobacillus delbrueckii ssp bulgaricus to freezing by natural selection. J
Dairy Sci 86:3048-3053.

Mountfort D.O., H.F. Kaspar, R.A. Asher, and D. Sutherland. 2003. Influences of
pond geochemistry, temperature, and freeze-thaw on terminal anaerobic
processes occurring in sediments of six ponds of the McMurdo Ice Shelf,
near Bratina Island, Antarctica. Appl Environ Microbiol 69:583-592.

Muryoi N., M. Sato, S. Kaneko, H. Kawahara, H. Obata, M.W. Yaish, M. Griffith,
and B.R. Glick. 2004. Cloning and expression of apr, a gene encoding an
antifreeze protein from the arctic plant growth-promoting rhizobacterium
Pseudomonas putida GR12-2. J Bacteriol 17:5661-5671.

Naas, T., M. Blot, W. M. Fitch, and W. Arber. 1994. Insertion sequence-related
genetic variation in resting Escherichia coli K-12. Genetics 136:721-30.

Naas, T., M. Blot, W. M. Fitch, and W. Arber. 1995. Dynamics of IS-related
genetic rearrangements in resting Escherichia coli K-12. Mol Biol Evol

12:198-207.

Nedwell, D. B. 1999. Effect of low temperature on microbial growth: lowered
affinity for substrates limits growth at low temperature. FEMS Microbiol

Ecol 30:101-111.

Notley—McRobb L., and T. Ferenci. 2000. Experimental analysis of molecular
events during mutational periodic selections in bacterial evolution.
Genetics 156:1493-1501 .

Novak M., T. Pfeiffer, R.E. Lenski, U. Sauer, S. Bonhoeffer. 2006. Experimental
tests for an evolutionary trade-off between growth rate and yield in E. coli.
Am Nat 168:242-251.

Ohtsubo, Y., H. Genka, H. Komatsu, Y. Nagata, and M. Tsuda. 2005. High-
temperature-induced transposition of insertion elements in bunkholderia
multivorans ATCC 17616. Appl Environ Microbiol 71:1822-8.

Packer E.L., J.L. lngraham, and S. Scher. 1965. Factors affecting the rate of
killing of Escherichia coli by repeated freezing and thawing. J Bacteriol
89:718-24.

Panoff J.M., B. Thammavongs, and M. Gueguen. 2000. Cryoprotectants lead to

phenotypic adaptation to freeze-thaw stress in Lactobacillus delbrueckii
spp. bulgaricus CIP 101027T. Cryobiology 40:264-269.

136

Papadopoulos, D., D. Schneider, J. Meier-Eiss, W. Arber, R.E. Lenski, and M.
Blot. 1999. Genomic evolution during a 10,000-generation experiment with
bacteria. Proc Natl Acad Sci USA 96:3807-3812.

Parsons PA. 1987. Evolutionary rates under environmental stress. Pp. 311-347
in M.K. Hecht, B. Wallace, and GT. Prance, eds. Evolutionary Biology,
Vol 21. Plenum, New York.

Pelosi L., L. Ktihn, D. Guetta, J. Garin, J. Geiselmann, R.E. Lenski, and D.
Schneider. 2006. Parallel changes in global protein profiles during long-
term experimental evolution in Escherichia coli. Genetics 173:1851-1869.

Pluschke, G., Y. Hirota, and P. Overath. 1978. Function of phospholipids in
Escherichia coli. Characterization of a mutant deficient in cardiolipin
synthesis. J Biol Chem 253:5048-55.

Pluschke, G., and P. Overath. 1981. Function of phospholipids in Escherichia
coli. Influence of changes in polar head group composition on the lipid
phase transition and characterization of a mutant containing only
saturated phospholipid acyl chains. J Biol Chem 256:3207-12.

Ponder M.A., S.J. Gilmour, P.W. Bergholz, C.A. Mindock, R. Hollingsworth, M.F.
Thomashow, and J.M. Tiedje. 2005. Characterization of potential stress

responses in ancient Siberian permafrost psychroactive bacteria. FEMS
Microbiol Ecol 1 :103-1 15.

Poon, A., and L. Chao. 2005. The rate of compensatory mutation in the DNA
bacteriophage phiX174. Genetics 170:989-99.

Pruss, B.M., K.P. Francis, F. von Stetten, and S. Scherer. 1999. Correlation of
16S ribosomal DNA signature sequences with temperature-dependent
growth rates of mesophilic and psychrotolerant strains of the Bacillus
cereus group. J Bacteriol 181:2624-30.

Queitsch 0., TA. Sangster, and S. Lindquist. 2002. Hsp90 as a capacitor of
phenotypic variation. Nature 417:618-24.

Ramos, J.L., E. Duque, J.J. Rodriguez-Herve, P. Godoy, A. Haidour, F. Reyes,
and A. Femandez—Barrero. 1997. Mechanisms for solvent tolerance in
bacteria. J Biol Chem 272:3887-90.

Reif, H.J., and H. Saedler. 1975. IS1 is involved in deletion formation in the gal
region of E. coli K12. Mol Gen Genet 137:17-28.

Reynolds, A.E., J. Felton, and A. Wright. 1981. Insertion of DNA activates the
cryptic bgl operon in E. coli K12. Nature 293:625—9.

137

Riehle M.M., A.F. Bennett, and AD. Long. 2001. Genetic architecture of thermal
adaptation in Escherichia coli. Proc Natl Acad Sci USA 98:525-530.

Riehle M.M., A.F. Bennett, R.E. Lenski, and AD. Long. 2003. Evolutionary
changes in heat-inducible gene expression in lines of Escherichia coli
adapted to high temperature. Physiol Genom 14:47-58.

Riehle M.M., A.F. Bennett, and AD. Long. 2005. Changes in gene expression
following high-temperature adaptation in experimentally evolved
populations of E. coli. Physiol Biochem Zool 78:299-315.

Riley, M., T. Abe, M. B. Amaud, M. K. Berlyn, F. R. Blattner, R. R. Chaudhuri, J.
D. Glasner, T. Horiuchi, I. M. Keseler, T. Kosuge, H. Mori, N. T. Pema, G.
Plunkett, 3rd, K. E. Rudd, M. H. Serres, G. H. Thomas, N. R. Thomson, D.
Wishart, and B. L. Wanner. 2006. Escherichia coli K-12: a cooperatively
developed annotation snapshot--2005. Nucleic Acids Res 34:1-9.

Rivkina E.M., E.|. Friedmann, C.P. McKay, and DA. Gilichinksy. 2000. Metabolic
activity of permafrost bacteria below the freezing point. Appl Environ
Microbiol 66:3230-3233.

Rose MR. 1984. Laboratory evolution of postponed senescence in Drosophila
melanogaster. Evolution 38: 1 004-1 010.

Rozen DE, and RE. Lenski. 2000. Long-term experimental evolution in
Escherichia coli. VIIl. Dynamics of a balanced polymorphism. Am Nat
155:24-35.

Rozen D.E., D. Schneider, R.E. Lenski. 2005. Long-term experimental evolution
in Escherichia coli. Xlll. Phylogenetic history of a balanced polymorphism.
J Mol Evol 61:171-180.

Russell NJ. 1990. Cold adaptation of microorganisms. Phil Trans R Soc Lond B
Biol Sci 326:598-608.

Rutherford S.L., and S. Lindquist. 1998. Hsp90 as a capacitor for morphological
evolution. Nature 396:336-342.

Sanger, F., S. Nicklen, and AR. Coulson. 1977. DNA sequencing with chain-
terminating inhibitors. Proc Natl Acad Sci U S A 74:5463-7.

Scher S., E. Packer , and C. Sagan. 1964. Biological contamination of Mars. |.

Survival of terrestrial microorganisms in simulated Martian environments.
Life Sci Space Res 2:352-356.

138

Schneider D., E. Duperchy, E. Coursange, R.E. Lenski, and M. Blot. 2000. Long-
term experimental evolution in Escherichia coli. IX. Characterization of
insertion sequence-mediated mutations and rearrangements. Genetics
156:477-488.

Schneider D., E. Duperchy, J. Depeyrot, E. Coursange, R.E. Lenski, and M. Blot.
2002. Genomic comparisons among Escherichia coli strains B, K-12, and
O157:H7 using IS elements as molecular markers. BMC Microbiol 2:18.

Schneider, D., and RE. Lenski. 2004. Dynamics of insertion sequence elements
during experimental evolution of bacteria. Res Microbiol 155:319-27.

Schoustra S.E., A.J. Debets, M. Slakhorst, and RF. Hoekstra. 2006. Reducing
the cost of resistance: experimental evolution in the filamentous fungus
Aspergillus nidulans. J Evol Biol 19:1115-1127.

Schwartz, E., M. Kroger, and B. Rak. 1988. IS150: distribution, nucleotide
sequence and phylogenetic relationships of a new E. coli insertion
element. Nucleic Acids Res 16:6789-802.

Shendure J., G.J. Porreca, N.B. Reppas, X. Lin, J.P. McCutcheon, A.M.
Rosenbaum, M.D. Wang, K. Zhang, R.D. Mitra, and GM. Church. 2005.
Accurate multiplex polony sequencing of an evolved bacterial genome.
Science 309:1728-1732.

Shibuya, |., and S. Hiraoka. 1992. Cardiolipin synthase from Escherichia coli.
Methods Enzymol 209:321-30.

Siddiqui KS, and R. Cavicchioli. 2006. Cold-adapted enzymes. Annu Rev
Biochem 752403-433.

Sinensky, M. 1974. Homeoviscous adaptation--a homeostatic process that
regulates the viscosity of membrane lipids in Escherichia coli. Proc Natl
Acad Sci U S A 71:522-5.

Slater, S.J., M.B. Kelly, F.J. Taddeo, C. Ho, E. Rubin, and CD. Stubbs. 1994.
The modulation of protein kinase C activity by membrane lipid bilayer
structure. J Biol Chem 269:4866-71.

Sleight S.C., N.S. Wigginton, and RE. Lenski. 2006. Increased susceptibility to
repeated freeze-thaw cycles in Escherichia coli following long-term
evolution in a benign environment. BMC Evol Biol 6:104.

Sleight, 8.0., and RE. Lenski. 2007. Evolutionary adaptation to freeze-thaw-
growth cycles in Escherichia coli. Phys. Biochem. Zool., in press.

139

 

Sniegowski P.D., P.J. Gerrish, and RE. Lenski. 1997. Evolution of high mutation
rates in experimental populations of E. coli. Nature 387:703-705.

Sokal RR, and F.J. Rohlf. 1981. Biometry. W. H. Freeman, New York.

Southern, EM. 1975. Detection of specific sequences among DNA fragments
separated by gel electrophoresis. J Mol Biol 982503-17.

Souzu H., M. Sato, and T. Kojima. 1989. Changes in chemical structure and
function in Escherichia coli cell membranes caused by freezing and
thawing. lI. Membrane lipid state and response of cells to dehydration.
Biochim Biophys Acta 978:112-118.

Stead, D., and SF. Park. 2000. Roles of Fe superoxide dismutase and catalase
in resistance of Campy/obacter coli to freeze-thaw stress. Appl Environ
Microbiol 66:31 10-2.

Steponkus, P. L. 1984. Role of the plasma membrane in freezing injury and cold
acclimation. Ann. Rev. Plant Physiol. 35:543-584.

Thammavongs B., D. Corroler, J.M. Panoff, Y. Auffray, and P. Boutibonnes.
1996. Physiological response of Enterococcus faecalis JH2-2 to cold
shock: growth at low temperatures and freezing/thawing challenge. Lett
Appl Microbiol 23: 398-402.

Travisano M., J.A. Mongold, A.F. Bennett, and RE. Lenski. 1995. Experimental
tests of the roles of adaptation, chance, and history in evolution. Science
267287-90.

Travisano, M. 1997. Long-term experimental evolution in Escherichia coli. VI.
Environmental constraints on adaptation and divergence. Genetics 146:
471-479.

Treves, D.S., S. Manning, and J. Adams. 1998. Repeated evolution of an
acetate-crossfeeding polymorphism in long-term populations of
Escherichia coli. Mol Biol Evol 152789-97.

Twiss, E., A.M. Coros, N.P. Tavakoli, and KM. Derbyshire. 2005. Transposition
is modulated by a diverse set of host factors in Escherichia coli and is
stimulated by nutritional stress. Mol Microbiol 57:1593-607.

Vanounou, S., D. Pines, E. Pines, A.H. Parola, and I. Fishov. 2002. Coexistence

of domains with distinct order and polarity in fluid bacterial membranes.
Photochem Photobiol 7621-1 1.

140

 

Vasi F ., M. Travisano, and RE. Lenski. 1994. Long-term experimental evolution
in Escherichia coli. ll. Changes in life-history traits during adaptation to a
seasonal environment. Am Nat 144:432-456.

Velicer G.J., G. Raddatz, H. Keller, S. Deiss, C. Lanz, l. Dinkelacker, and SC.
Schuster. 2006. Comprehensive mutation identification in an evolved

bacterial cooperator and its cheating ancestor. Proc Natl Acad Sci USA
103:8107-8112.

Walker V.K., G.R. Palmer, and G. Voordouw. 2006. Freeze-thaw tolerance and
clues to the winter survival of a soil community. Appl Environ Microbiol
72:1784-1792.

Wichman H.A., M.R. Badgett, L.A. Scott, C.M. Boulianne, and J.J. Bull. 1999.
Different trajectories of parallel evolution during viral adaptation. Science
285:422-424.

Wichman H.A., L.A. Scott, C.D. Yarber, and J.J. Bull. 2000. Experimental
evolution recapitulates natural evolution. Phil Trans Roy Soc London B
355:1677-1684.

Wichman H.A., J. Millstein, J.J. Bull. 2005. Adaptive molecular evolution for
13,000 phage generations: 3 possible arms race. Genetics 170219-31.

Willimsky, G., H. Bang, G. Fischer, and M. A. Marahiel. 1992. Characterization of
cspB, a Bacillus subtilis inducible cold shock gene affecting cell viability at
low temperatures. J Bacteriol 174:6326-35.

Wood, T.E., J.M. Burke, and L.H. Rieseberg. 2005. Parallel genotypic adaptation:
when evolution repeats itself. Genetica 123:157-70.

Woods R., D. Schneider, C.L. Winkworth, M.A. Riley, and RE. Lenski. 2006.
Tests of parallel molecular evolution in a long-term experiment with
Escherichia coli. Proc Natl Acad Sci USA 103:9107-9112.

Wouters J.A., B. Jeynov , F.M. Rombouts , W.M. de Vos , O.P. Kuipers , T.
Abee. 1999. Analysis of the role of 7 kDa cold-shock proteins of
Lactococcus lactis MG1363 in cryoprotection. Microbiol 145:3185-3194.

Wouters J.A., F.M. Rombouts , W.M. de Vos , O.P. Kuipers, and T. Abee. 1999.
Cold shock proteins and low-temperature response of Streptococcus
thennophilus CNRZ302. Appl Environ Microbiol 65:4436-4442.

Young R.S., P.H. Deal, J. Bell, and J.L. Allen. 1964. Bacteria under simulated
Martian conditions. Life Sci Space Res 2:105-111.

141

Zeyl C. 2006. Experimental evolution with yeast. FEMS Yeast Res 6:685-691.
Zinser, E.R., D. Schneider, M. Blot, and R. Kolter. 2003. Bacterial evolution

through the selective loss of beneficial genes. Trade-offs in expression
involving two loci. Genetics 164:1271-7.

142

APPENDIX 1
This appendix provides some supporting data and supplementary
analyses for Chapter 2 on “Adaptation to cycles of freezing, thawing, and growth:
phenotypic evolution.” It includes the following items:

Figure A1 .1. Relative freeze-thaw survival of the 15 FTG-evolved lines and their
progenitors.

Figure A1 .2. Comparison of mortality rates between individual FTG-evolved lines
and their progenitors under the repeated freeze-thaw regime.

Figure A1 .3. Comparison of mortality rates between the FTG-evolved groups
and their progenitors under the repeated freeze-thaw regime.

Figure A1 .4. Relative growth fitness after freezing and thawing of the 15 FTG-
evolved lines and their progenitors.

Figure A1 .5. FTG timecourse for two FTG-evolved lines and their progenitors.

Figure A1 .6. Growth timecourse after stationary phase for two FTG-evolved lines
and their progenitors.

Figure A1 .7. Fluorescence microscopy of the Evolved Group A (AncB/FTG)
clone and the ancestor at different timepoints in the FTG cycle.

Figure A1 .8. Fluorescence microscopy of the Evolved Group B (A+1/FTG) clone
and its progenitor (A+1) at different timepoints in the FTG cycle.

Table A1 .1. Phenotypic measurements between individual FTG-evolved lines
relative to their progenitors.

Table A1 .2. Statistical tests for FTG-evolved groups and their progenitors.

143

 

4.0

3.5

3.0

2.5

2.0

1

'0:

Relative FT Survivd

1

"o

#1111 it 3

0.5

Figure A1 .1. Relative freeze-thaw survival of the 15 FTG-evolved lines and
their progenitors. FT survival values were measured in competition with the
common ancestor during the first half of the two-day FTG cycle. Each evolved
line (grey) is paired with its progenitor (white). The three group A lines are
shown with different ancestral replicates (designated A, B, and C). The twelve
group B lines were derived from long-term lines, designated A+1 to A-6, that
previously evolved for 20,000 generations at 37°C. Error bars show 95%
confidence intervals based on six replicate FT survival assays for each line or
progenitor, except for the group A ancestor, where two of the replicate assays
were paired with each of the three evolved lines in that group.

144

 

0.9

 

 

 

“E i I
g 0.5

cf

E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

y!" 5’ s“ ‘3’ e” v" vm'v-‘b v“ v? v”

 

Figure A1 .2. Comparison of mortality rates between individual FTG-
evolved lines and their progenitors under the repeated freeze-thaw regime.
The height of each bar shows the mean mortality rate per day for each of the
evolved lines and their progenitors, measured individually over ten daily freeze-
thaw cycles, calculated from three assays for each line. Each evolved line (grey)
is shown paired with its progenitor (white). The three group A lines are shown
with different ancestral replicates (designated A, B, and C). The twelve group B
lines were derived from long-term lines, designated A+1 to A-6, that previously
evolved for 20,000 generations at 37°C. See the Materials and Methods section
in Chapter 1 for the mortality rate calculation. Error bars show 95% confidence
intervals based on three replicate repeated FT survival assays for each evolved
line or progenitor.

145

 

 

 

 

   

 

0.5
0 0.4
E J;

J.

|—
g 0.3
LL
g 0.2
2

0.1

0.0

Ancestor Group A Progenitors Group 8

Figure A1 .3. Comparison of mortality rates between the FTG-evolved
groups and their progenitors under the repeated freeze-thaw regime. The
height of each bar shows the mean mortality rate per day for each FT G-evolved
group and their progenitor, measured over ten daily freeze-thaw cycles. The
mean for each evolved group (grey) is adjacent to the corresponding mean for its
progenitors (white). Means for each group were calculated from average values
for each member in the group, except for the single ancestor of group A where
the mean is based on nine independent measurements. See the Materials and
Methods section in Chapter 1 for the mortality rate calculation. Error bars are
95% confidence intervals based on the number of independently evolved lines,
except for the ancestor, for which the interval is based on the nine replicate
assays.

146

4.0

3.0

Relative Growth After FT

 

1.0 " ,

x ‘5 to N '5 9:
‘3’wa .m- 3,. vi": it it ‘2”
.5 3

Figure A1 .4. Relative growth fitness after freezing and thawing of the 15
FTG-evolved lines and their progenitors. Growth after FT values were
measured in competition with the common ancestor during the second half of the
two-day FTG cycle. Each evolved line (grey) is paired with its progenitor (white).
The three group A lines are shown with different ancestral replicates (designated
A, B, and C). The twelve group B lines were derived from long-term lines,
designated A+1 to A-6, that previously evolved for 20,000 generations at 37°C.
Error bars show 95% confidence intervals based on six replicate growth after FT
assays for each line or progenitor, except for the group A ancestor, where two of
the replicate assays were paired with each of the three evolved lines in that

group.

147

Figure A1 .5.

(A)

o
2

5 0

Aficiopxiovgaoo-oo

5

Time (Hours)

(3)

40

w a m w
Estexnuovazcoo

 

 

Time (Hours)

148

Figure A1 .5. FTG timecourse for two FTG-evolved lines and their
progenitors. (A) Group A evolved line (AncB/FTG) and the ancestor; (B) group
B evolved line (A+1/FTG) and its immediate progenitor (A+1). Colony forming
units (CFU) measured initially and every hour for six hours after the FT cycle in
minimal glucose media. The evolved lines are shown as grey symbols, and their
progenitors as black symbols. Each point is the average of each clone replicated
sixfold. Error bars are 95% confidence intervals.

149

Figure A1 .6.

(A)

35

u o 5 o

2 4| 1

Jarexauovzacoog

 

Time (Hours)

(3)

35

w

 

a m w o

Attopxzuovgooto

 

Time (Hours)

150

Figure A1 .6. Growth timecourse after stationary phase for two FTG-
evolved lines and their progenitors. (A) Group A evolved line (AncB/FTG) and
the ancestor; (B) group B evolved line (A+1/FTG) and its immediate progenitor
(A+1). Colony forming units (CFU) measured initially and every hour for four
hours after stationary phase in minimal glucose media. The evolved lines are
shown as grey symbols, and their progenitors as black symbols. Each point is
the average of each clone replicated sixfold. Error bars are 95% confidence
intervals.

151

Figure A1 .7.

Ancestor AncB/FTG
A

 

152

-
-

Figure A1 .7. Fluorescence microscopy of the Evolved Group A (AncB/FTG)
clone and the ancestor at different timepoints in the FTG cycle. (A)
Stationary phase, (B) After a FT cycle, (C) 1.5 hours after growth, (D) 3 hours
after growth, (E) 4.5 hours after growth. Cells at each timepoint were stained
using the Live/Dead count kit (lnvitrogen) according to protocol and a sample
was put on a microscope slide for visualization with a fluorescent microscope.
Live cells are stained green and dead cells are stained red. Images in this
dissertation are presented in color.

 

153

 

Figure A1 .8.

A+1 Progenitor A+1/FTG
A

 

154

 

Figure A1 .8. Fluorescence microscopy of the Evolved Group B (A+1/FTG)
clone and its progenitor (A+1) at different timepoints in the FTG cycle. (A)
Stationary phase, (B) After a FT cycle, (C) 1.5 hours after growth, (D) 3 hours
after growth, (E) 4.5 hours after growth. Cells at each timepoint were stained
using the Live/Dead count kit (Invitrogen) according to protocol and a sample
was put on a microscope slide for visualization with a fluorescent microscope.
Live cells are stained green and dead cells are stained red. Images in this
dissertation are presented in color.

155

 

3:58...
mu. 2 025.2 ms. co>_o>o-0...n. .3223. some .8 5.65 05 3589388 0539.93 EoEtB 290 Go cmoE
9: 38.2395 :65 8 3.3.9. «2... oozes-6...... .3222: c3253 8:25.532: cabocosn. ...._.< 2%...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8o... 33 £3. 33- 48.3. 83 33 33 :82
83 83- 2mg 33- £32 83 83 33 eta-4.
3.3 83 03.8- 2m.- 13 83 82.. mm: eta-.-
33 83.- at? 831 SE? a: 8: N3 eta}
$3 83- 83: 9:3- 5.3% 83 $3 $3 eta:
Bod v2.2 .8.-o- 8nd wine-F 83 $3 83 eta-4.
mm; 83- mam-.8 53 v8.2 83 R3 8: 33%
83 83 $3- 83 84.88 :3 83 85 eta}
83 33 gm. P P- 8::- 38. z: 8: 33 E: eta}
BE Boa v8.31 23 23- $3 83 83 33+...
83. 93 83.8- 33.- fimdm 9.3 53- 53 eta-é
8nd 988. moosm 83- 93: 23 t3 8: eta:-
53 83- ~55 :3. SN. 5 $3 $3 $3 33:
m3... 8; 83-N- 33. 392: 83 2.3 33 52...
83 83 48.3- 32: 58.3 83 ~91 mi... 3&0 .9805
83 a}: Bowm- 39 88.83 23 83 83 win .8385
22. Eva moos- mmmé 832 8: ~91 tow eta 3335
.23. .32.“. .82“. .E E .22. E 322.8 E 285E or. 96.0
£3.22. 55.3% 55:5 .32. 2:: 32a 93 .53 5.35
.325 .22. we: .32. 9:325
.E 033%”. ufifisoo wound on.

 

 

 

 

 

 

 

 

 

 

156

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Statistical test One or two p-value Greater of
tailed t-test mups

FTG Fitness: A > Ancestor One 0.0034

FTG Fitness: B > Prggenitors One < 0.0001

FT Survival: A > Ancestor One 0.0014

FT Survival: B > Progenitors One 0.0023

Growth After FT: A > Ancestor One 0.0042

Growth After FT: B > Progenitors One < 0.0001

Proportional FTG Fitness: A ?=? B Two 0.0417 A

Proportional FT Survival: A < B One 0.0432

Proportional Growth After FT: A > B One 0.0056

Final FTG Fitness: A ?=? B Two 0.0003 B

Final FT Survival: A ?=? B Two 0.3691

Final Growth After FT: A ?=? B Two 0.0001 B

Lag After FT: A > Ancestor One 0.0066

Lag After FT: B > Progenitors One 0.0001

Growth Rate After FT: A > Ancestor One 0.0075

Growth Rate After FT: B ?=?

Progenitors Two > 0.5

Lag After Stationary Phase: A ?=?

Ancestor Two > 0.5

Lag After SP: B ?=? Progenitors Two > 0.5

Growth Rate After SP: A > Ancestor One 0.0058

Growth Rate After SP: B ?=?

Progenitors Two 0.3007

 

 

Table A1 .2. Statistical tests for FTG-evolved groups and their progenitors.
Each statistical test between evolved groups and/or their progenitors is shown as
a one or two-tailed t-test with the p-value. In the case of a one-tailed t-test, there
was an a priori hypothesis about the direction of change (< or >) between two
groups. In the case of a two-tailed t-test there is no a priori hypothesis about the
direction of change. If the p-value is less than 0.05, then the far right column
shows which of the groups being tested has a significantly greater value.

157

APPENDIX 2
This appendix provides some supporting data and supplementary
analyses for Chapter 3 on “Adaptation to cycles of freezing, thawing, and growth:
genetic evolution.” It includes the following items:

Figure A2.1. cls and uspA/B isogenic constructs and reconstructs.

Figure A2.2. Southern hybridization of EcoRV-digested genomic DNA to the IS1
probe.

Figure A2.3. Southern hybridization of EcoRV-digested genomic DNA to the IS3
probe.

Figure A2.4. Southern hybridization of EcoRV-digested genomic DNA to the
IS186 probe.

Figure A2.5. Southern hybridization of EcoRV-digested genomic DNA to the
IS186 probe (second blot).

Figure A2.6. Relative uspA transcription differences using Real-Time PCR.
Table A2.1. Primer pairs used to PCR amplify candidate genes.

Table A2.2. Primer pairs used to PCR amplify IS element probes for RFLP
analysis.

Table A2.3. Primer pairs used to PCR amplify sequences adjacent to IS
elements.

Table A2.4. IS band differences between individual FTG-evolved lines relative to
their progenitors.

Table A2.5. Laurdan anistropy differences between individual FTG-evolved
clones, their progenitors, and isogenic constructs.

158

 

Original Clone Isogenic construct Isogenic reconstruct
A.

 

   

I

   

Figure A2.1. cls and uspA/B isogenic constructs and reconstructs. (A) The
A-2/FTG population has an 11-bp deletion in the cls ORF in all clones tested
within this population. An isogenic construct of the original A-2/FTG clone was
made to reverse this mutation to the ancestral state. This isogenic construct was
then reconstructed to possess the evolved mutation in order to verify that any
FTG fitness differences are caused by the mutation itself (not some hypothetical
mutation inadvertently introduced during genetic manipulations). Likewise, to
test the effect of this mutation in the reciprocal case, the A-2 progenitor was
constructed to have this evolved mutation and reconstructed back to the
ancestral state. All constructs and reconstructs were sequenced twice to verify
the presence or absence of the evolved mutation and that no other mutations
were present within approximately 500 bp on either side. (B) All clones tested in
the AncB/FTG population have an IS150 insertion in the uspA/B intergenic
region. Two independent isogenic constructs were made to remove this IS150
element and control clones that retained the l8150 insertion were isolated for
FTG fitness competition and growth curve experiments to ensure that any
differences are due to the IS150 insertion. In this case, no attempt was made to
reconstruct the isogenic constructs back to have the IS150 insertion because in
the process this IS150 may recombine with other IS150 elements and cause
unwanted genomic rearrangements. All constructs and control clones were
sequenced twice to verify the presence or absence of the IS1 50 insertion and
that no other mutations were present 100 bp on either side of the insertion.

159

 

 

‘30333. It. It

O-Dibou-I:
.--- gnu:

.go.::.-.-_
“2::— ;:n::---e:

lull: l

 

Figure A2.2. Southern hybridization of EcoRV-digested genomic DNA to
the IS1 probe. The clones are numbered for each lane at top where each FI'G-
evolved clone follows its direct progenitor (odd numbers are progenitors and
even numbers are FTG-evolved clones). Lanes 1-20 are as follows: 1)Ancestor,
2) AncB/FTG, 3) A+1, 4) A+1/FTG, 5) A+2, 6) A+2/FTG, 7) A+3, 8) A+3/FTG, 9)
A+4, 10) A+4/FTG, 11)A+6, 12) A+6/FTG, 13) A-1, 14)A-1/FTG, 15) A-2, 16)A-
2/FTG, 17) A-3, 18) A-3/FTG, 19) A4, 20) A-4/FTG.

160

12 3 4 5 6 7 8 91011121314151617181920

oven. jjddddu'.-."-

.
U

H
II
II
IE
IR .0
"r
O M
O u
H
H
I I"
II
t
I II
It
Cl
0'
0|
0 H
H

.l
I
3

 

Figure A2.3. Southern hybridization of EcoRV-digested genomic DNA to
the IS3 probe. The clones are numbered for each lane at top where each FT G-
evolved clone follows its direct progenitor (odd numbers are progenitors and
even numbers are FTG-evolved clones). Lanes 1-20 are as follows: 1) Ancestor,
2) AncB/FT G, 3) A+1, 4) A+1/FTG, 5) A+2, 6) A+2/FTG, 7) A+3, 8) A+3/FTG, 9)
A+4, 10)A+4/FTG, 11)A+6, 12) A+6/FTG, 13) A-1, 14) A-1/FTG, 15) A-2, 16)A-
2/FTG, 17) A-3, 18) A-3/FTG, 19)A-4, 20) A-4/FTG.

161

‘ .- .. a. u.

13 37 575 i e 5151’} ”14151617181920

v"l'.06uuuui"

 

0683 l I. “f
o - : .....M"M6'
i a

 

 

Figure A2.4. Southern hybridization of EcoRV-digested genomic DNA to
the IS 186 probe. The clones are numbered for each lane at top where each
FTG-evolved clone follows its direct progenitor (odd numbers are progenitors and
even numbers are FTG-evolved clones). Lanes 1-20 are as follows: 1) Ancestor,
2) AncB/FTG, 3) A+1, 4) A+1/FTG, 5) A+2, 6) A+2/FTG, 7) A+3, 8) A+3/FTG, 9)
A+4, 10) A+4/FTG, 11)A+6, 12) A+6/FTG, 13) A-1, 14)A-1/FTG, 15)A-2, 16)A-
2/FTG, 17) A-3, 18) A-3/FTG, 19) A4, 20) A-4/FTG.

162

251 2 3 4 5 6 7 8 910 11121314151617181920

9~:::::I==-UI5I"I"

in: i....-......,,gp

“2.:

 

Figure A2.5. Southern hybridization of EcoRV-digested genomic DNA to
the IS186 probe (second blot). This blot was performed to repeat four clones
from the previous IS186 blot and determine whether other clones in the A+6/FTG
population at the 500 generation and 1000 generation timepoints have deletion of
a particular band as seen in the A+6/FTG clone. The clones are numbered for
each lane at top. Lanes 1-20 are as follows: 1)A+1, 2) A+1/FTG, 3) A-2, 4) A-
2/FTG, 5) A+6, 6-15) 10 different A+6/FTG clones from the 1000 generation
population, 16-20) Five different A+6/FTG clones from the 500 generation
population. The A+1/FTG and A-2/FTG clones have similar band changes as the
previous IS186 blot and the reason for there being faint bands is unknown. All
ten clones from the 1000 generation timepoint in the A+6/FTG population have a
deleted band relative to the progenitor, but all five clones from the 500 generation
timepoint have the same banding patterns as the progenitor. This deleted band
in the 1000 generation clones was located near the hokE locus and is either a
large deletion or complex genomic rearrangement. The exact mutation was not
determined.

163

p
8

D AnoestoruspA/B 48150
I AncB/FTG uspA/B +l3150
I AncB/FTG uspA/B 48150

'8

.0
o
‘1

 

p
8

 

p
8

 

p
9

p
8

Relative uspA Transcription

 

I'M

 

 

 

 

 

Stationary Phase After FT Growth

Figure A2.6. Relative uspA transcription differences using Real-Time PCR.
uspA transcription was measured relative to 168 rRNA transcription as an
endogenous control at stationary phase, after a FT cycle, and in growth
conditions 2 h after thawed cultures were diluted into fresh media. Clones are as
follows: Ancestor uspA/B -lS150 (white), evolved clone AncB/FTG uspA/B
+IS150 (grey), and constructed clone AncB/FTG uspA/B -|S150 from which the
insertion element was removed (black). The height of each bar represents the
mean of two independent RNA extractions, each set with 3-fold replication. Error
bars are 95% confidence intervals based on the six-fold replication.

164

 

 

Gene

Gene Product

Primers

 

Cold Shock Protein A

5' ATGGTAAAGGAATGGGAATC 3'

 

5' CTGGCGGCGI | l | IAGC 3'

 

fabF

Beta-ketoacyI-ACP synthase II

5' TCCTGGCATCCGACGAA 3'

 

5' CTTATCAGAGAATTGTTAGTGTGG 3'

 

5' CAGGCTGCCATTGATTACAT 3'

 

5' TCATACCAAAGCCGACGAGTI’ 3'

 

nadR

Nicotinamide-Nucleotide
adenylyltransferase

5' GTCTGGCAATCCTGCTGGCT 3'

 

5' GGTCACCCGTGCAGGTTT 3'

 

5' CGCATTCATGCTTTCAACGAAG 3'

 

5' GCGCACCGCTGATACTCATA 3'

 

otsA

Trehalose synthase

5' TGCGGTAATCACAATCACAG 3'

 

5' CACAAATGGCGTTACAGC 3'

 

5' AGCCCACGTCAGAGTAGCGGAATA 3'

 

5' GCGAACAGGACCTTGACGAATACT 3'

 

(p08

08 subunit

5' TGGGG'ITGTCGGTAGCAGA 3'

 

5' TGATATCGCAGGCAGCAAAGGAC 3'

 

5' TATCGTCATCTTGCGTGGTATC 3'

 

5' CGAAAAGGCCTI'AGTAGAACA 3'

 

rrsA

16$ rRNA

5' CGCCCATTTAACCGACAA 3'

 

5' I | | l IATTTAATCACTACAGAGA 3'

 

5' GTGGGGAGCAAACAGGATTAGATA 3'

 

5' CTCCGGAAGCCACGCCTCAAG 3'

 

sodA

Manganese superoxide dismutase

5' CTGCCCGGCGCGAGTI'ACAATC 3'

 

5' GCCAAGGAATAGCGGGACGAGCAT 3'

 

5' TTCCTGCAAAACCATACC 3'

 

 

 

 

5' ACCCATCAGCGGAGAAT 3'

 

Table A2.1. Primer pairs used to PCR amplify candidate genes.

165

 

 

 

 

 

 

 

 

 

 

IS Primer name and seguence Size of PCR product

IS 1 IS1 F: 5’ CTCCAGTGGCTI'CTGTTTCT 3’ 668 b p
lS 1R: 5’ TGACTTTGTCATGCAGCTCC 3’

ISZ ISZF: 5’ GATGTCTTAGGGCCGGAGAA 3’ 1 181 b p
ISZR: 5’ AGCCCGCTGCCGCAGATATT 3’ ’

l S 3 IS3F: 5’ TCAGAGGTGACTCACATGAC 3’ 1 158 b p
IS3R: 5’ CAAATTG‘ITCCGGACTGAGG 3’ ’

IS 4 IS4F: 5’ ATCCGGAAACGCTCACTI'GT 3’ 1 325 b p
IS4R: 5’ ATTI'CCAGGGCCTCTCCTTT 3’ ’
IS30F:

1330 ?ngTGAGACGAACTATTACAGCAG 3 1,111 b p
5’ G‘I'TTTGAACTTCAGTGTCTTI'CTC 3’
IS150F: 5’ GCGCCTTGAAGTCGTGAATC 3’

IS150 lS150R: 5’ GGCTAATTCTTCTGCTGTI’GTAG 1,279 bp
3.

IS 186 IS186F. 5 GCCGATGAATTACTCTCACG 3 1,235 b p

 

|S186R: 5’ GCCTCG'ITAAGACGATGCCT 3’

 

 

Table A2.2. Primer pairs used to PCR amplify lS element probes for RFLP
analysis.

166

 

 

 

IS element

Primer sequence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IS 1 5' GTCATCGGGCATI'ATCTGAAC 3'
5' AGAAGCCACTGGAGCACC 3'
ISZ 5' CATAGTGCGCTGGGTTATCG 3'
5' GTCTGCGTI'TCTCCGGCCTA 3'
IS3 5' AGTAACACCGATGCGTTCAG 3'
5' CTGCTACGATAATGCCTGCG 3'
|S4 5' ACCAGATCAAGAGCCTGTCC 3'
5' CCTTCCCGAGAGTGGTAAAG 3'
|S30 5' CCTCTGCTGTAAATGTI'CGTCTCA 3'
5' GGTI'GCTGCTCAGCTAAACAACAG 3'
IS 150 5' GATCCTGTAACCATCATCAG 3'
5' CTGAAGGATGCTGTTACGG 3'
IS 186 5' CGGCATTACGTGCCGAAG 3'

 

 

 

5' GGTGGCCATTCGTGGGAC 3'

 

Table A2.3. Primer pairs used to PCR amplify sequences adjacent to IS

elements.

167

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Clone IS1 IS3 IS150 IS186
2.75 kb
AncB/FTG insertion
2.5 kb deletion
1.5 kb insertion
7 kb
A+1/FTG 4.75 kb deletion 4 kb insertion insertion
4 kb insertion 3.5 kb deletion
1.8 kb deletion 1.5 kb insertion
A+2/FTG 11 kb deletion 6 kb insertion
5.5 kb band
fainter
1.8 kb band
fainter
A+3/FTG
A+4/FTG 5 kb insertion
2.75 insertion
2 kb insertion
1.5 kb insertion
A+6/FTG 6 kb deletion, 3.75 kb deletion 6 kb deletion
1.75 kb deletion 2.75 kb deletion
A-1/FTG
A-2/FTG 8 kb deletion
A-3/FTG 3.25 kb deletion 7 kb insertion
5 kb deletion
2 kb insertion
1.5 kb insertion
A-4/FTG 2.5 kb insertion

 

Table A2.4. IS band differences between individual FTG-evolved lines
relative to their progenitors. For each FTG-evolved clone, the IS band
differences are shown for the four IS elements with changes.

168

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Clone Genotype Laurdan Anisotropy Laurdan Anisotropy
At Stationary Phase After FT

A-2 cls' uspA/B -IS150 0.0231 :l: 0.0037 0.0353 i 0.0034
A-2 cls+ uspA/B -IS 150 0.0231 :t 0.0036 0.0443 1 0.0042
A-2/FTG cls+ uspA/B -|S150 0.0222 i 0.0037 0.0314 :l: 0.0035
A-2/FTG cls' uspA/B -IS150 0.0253 :1: 0.0040 0.0248 :1: 0.0021
A+4 cls+ uspA/B -lS150 0.0193 :1: 0.0059 0.0408 1 0.0078
A+4/FTG cls‘ uspA/B +IS150 0.0200 i 0.0051 0.0248 1: 0.0033
A-3 cls+ uspA/B -|S150 0.0202 i 0.0103 0.0330 1: 0.0053
A-3/FTG cls' uspA/B +IS150 0.0197 :1: 0.0023 0.0240 1 0.0044
Ancestor cls+ uspA/B -IS150 0.0263 1 0.0033 0.0424 :1: 0.0032
3?:TG cls+ uspA/B +IS150 0.0244 :t 0.0021 0.0375 :l: 0.0027
A

B7FCTG cls+ uspA/B -|S150 0.0274 1 0.0012 0.0362 1: 0.0032
A-5 cIs' uspA/B -IS150 0.0328 i 0.0038 0.0343 i 0.0040
A-5/FTG cls' uspA/B +IS150 0.0339 :l: 0.0032 0.0376 :I: 0.0044
A-6 cIs' uspA/B -lS150 0.0249 :l: 0.0018 0.0363 1 0.0050
A-6/FTG cls' uspA/B -IS150 0.0275 1- 0.0021 0.0395 1 0.0052

 

 

 

 

 

Table A2.5. Laurdan anistropy differences between individual FTG-evolved
clones, their progenitors, and isogenic strains. For each clone, the genotype
is shown with relation to the cls and uspA/B loci, and Iaurdan anisotropy values
are shown at stationary phase or after FT. The :l: value indicates the 95%
confidence interval from at least eight independent assays.

169