Directed evolution of cell size in Escherichia coli
© Yoshida et al.; licensee BioMed Central. 2014
Received: 31 July 2014
Accepted: 27 November 2014
Published: 17 December 2014
In bacteria, cell size affects chromosome replication, the assembly of division machinery, cell wall synthesis, membrane synthesis and ultimately growth rate. In addition, cell size can also be a target for Darwinian evolution for protection from predators. This strong coupling of cell size and growth, however, could lead to the introduction of growth defects after size evolution. An important question remains: can bacterial cell size change and/or evolve without imposing a growth burden?
The directed evolution of particular cell sizes, without a growth burden, was tested with a laboratory Escherichia coli strain. Cells of defined size ranges were collected by a cell sorter and were subsequently cultured. This selection-propagation cycle was repeated, and significant changes in cell size were detected within 400 generations. In addition, the width of the size distribution was altered. The changes in cell size were unaccompanied by a growth burden. Whole genome sequencing revealed that only a few mutations in genes related to membrane synthesis conferred the size evolution.
In conclusion, bacterial cell size could evolve, through a few mutations, without growth reduction. The size evolution without growth reduction suggests a rapid evolutionary change to diverse cell sizes in bacterial survival strategies.
Cell size is a key feature for all living things, from bacteria to mammals. In bacteria, cell size plays an important role in fitness, both directly and indirectly. For example, a bacterium’s vulnerability to predation by protists and host immune cells, such as neutrophils, depends on its cell size ,. In addition, the cell size is relevant to mechanisms of antibiotic resistance  and protection from bacterial phages . Thus, the bacterial cell size itself could be a target of selective pressure in the natural environment in addition to other targets, such as growth rate.
In bacteria, as well as eukaryotes, cell size is also closely related to cell proliferation. Cell size affects the uptake of nutrients from outside of the cell, the concentrations of cellular components and the progress of intracellular biochemical reactions . Importantly, the initiation of chromosomal replication  and the assembly of the division machinery  also depend on cell size through underlying molecular mechanisms, resulting in homeostatic and recursive reproduction of an optimal cellular state . These facts show that the bacterial cell size is strongly coupled with its growth rate.
Such a strong coupling between the cell size and the growth rate could play two contradicting roles in evolution. Such coupling may facilitate adaptation during evolution. For example, a rapid growth state could be achieved as a consequence of the selection in size and vice versa. However, coupling that is too strong may introduce conflicts during evolution in cell size. For instance, if large cell size is favorable in the environment, for example to avoid grazing pressure from predators, but not for the fast-growing states, then the cells will face fitness conflicts, called evolutionary trade-offs . Similar concerns regarding the coupling between other traits, such as gene expression responsiveness across conditions (plasticity) and cell-to-cell variation (noise), have been discussed previously .
Consistent with the possible conflicts between cell size and growth rate, most cell size mutants exhibit defective growth rates -. Previous studies have identified several mutations that introduce abnormal morphology and volume. For example, when the genes related to cell division, cell wall synthesis and membrane synthesis are inactivated, the mutant cells exhibit different shapes and sizes from the wild-type ,. In contrast, mutants with defective lipid biosynthesis cannot synthesize the cell body rapidly and become smaller . Although these kinds of mutations might contribute to size evolution, such induced mutants face the conflict of growth rate instead.
Is bacterial cell size evolvable in several directions, from small to large, without an added growth burden? If so, how many and what kind of mutations are required? Unlike the analysis of artificially constructed mutations for a limited number of target genes, experimental evolution of microbial populations could be useful for identifying such genetic paths across the whole genome. The previous experimental evolution studies suggest that cell size could evolve along with growth adaptation. Lenski’s group performed the serial transfer of Escherichia coli over thousands of generations . As an evolutionary consequence, the cells obtained not only faster growth speed than the ancestors but also larger size ,, even in the absence of explicit directional selection on the cell size.
To explore the directed evolution toward different cell sizes, directed evolution experiments to finite cell sizes were required. Because the previous long-term experimental evolution with serial passages lacked explicit size selections, the selection target and its pressure were uncontrolled, allowing for the accumulation of mutations unrelated to size changes. It is unclear how rapidly cell size can evolve in the presence of explicit size selections. Thus, directed evolution experiments with a tunable selection for cell size within fewer generations are desirable.
Here, using E. coli, we performed evolution experiments on cell size for a short period to address whether cell size evolved in response to the size selections without growth conflict. We employed a cell sorter to directly select specific cell sizes that were smaller than the ancestor. Two target sizes were repeatedly selected, along with the size distributions, mild and severe. The former target size selected cells that were slightly smaller than the ancestors (1% of the cells around the peak of the size distributions). The latter target size selected cells with far smaller sizes (the smallest 1% of the cells). The sorted cells were cultured overnight until the next size selections. Within 400 generations, smaller mutants were selected in response to each size selection. The growth rates of these mutants did not decrease. Whole genome sequencing revealed a few genomic mutations, as expected. We found that only a few mutations in the genes related to membrane synthesis could confer size evolution without growth conflict. We also tested the directed evolution toward larger cell sizes. The bacterial size evolution without growth reduction suggests that the rapid evolutionary change to diverse cell sizes represents a survival strategy.
Results and discussion
Bacterial cell size distributed broadly in a clonal population
Repeated cycles of size selection was examined with population propagation
Bacterial cell size decreased, and homogeneity increased
During the selection cycles, we observed changes in the size distributions (Figure 3B and C). In the Svr-lineage, the mean cell size decreased after 15 days (190 generations), achieving a 2.2-fold reduction compared with the T-lineage at the final day (288 generations) (Figure 3B). In contrast, the mean cell size in the Mld-lineage showed only a slight reduction. These results are consistent with the strength of the size selection. We also found that variation in the Mld-lineage started to decrease after 4 days (57 generations), relative to the Svr- and T-lineages (Figure 3C).
Cell IDs of the experiments
Ancestral clonal population of BSKY
12 clones isolated from AC
Population propagated for 5 days without size selection from an AC
Population obtained at the 22nd round in the T-lineage
12 clones isolated from T22P
Population propagated for 5 days without size selection from an T22C
Population obtained at the 22nd round in the Svr-lineage
12 clones isolated from Svr22P
Population propagated for 5 days without size selection from an Svr22C
Population obtained at the 22nd round in the Mld-lineage
12 clones isolated from Mld22P
Population propagated for 5 days without size selection from an Mld22C
Population obtained at the 22nd round in the L-lineage
12 clones isolated from L22P
Population propagated for 5 days without size selection from an L22C
Population obtained at the 8th round in the Ls-lineage
12 clones isolated from Ls8P
Population propagated for 5 days without size selection from an Ls8C
We also found that the variation in both the Svr- and the Mld-lineages decreased but to different extents (Figure 3G). The variation in Mld22Cs greatly decreased, while that in Svr22Cs decreased slightly. The selection around the peak may have stabilized the size distribution, reducing the variation in size. These evolved properties were confirmed by direct microscopic observation (Figure 3E, Additional file 1: Figure S1 and Additional file 2: Figure S2). These results indicate that the bacterial cell size could evolve through the simple selection process, not only in average but interestingly also in cell-to-cell variation, even in the clonal population.
The evolved cells maintained small size, independent of cell concentrations
In contrast, all evolved clones in the Svr- and Mld-lineages showed their evolved properties below 108 cells/ml and had different dependencies on cell concentration. The cell size of Svr22Cs was small in mean value, as were both the top 1% and the bottom 1% lines (2nd from the right in Figure 4C) and became slightly insensitive to cellular concentrations, relative to the ancestor clones. Mld22Cs notably became dense around the mean and maintained their cell size over various cell concentrations (far right in Figure 4C). Thus, size selection at a particular cell concentration promotes size evolution at the different cell concentrations.
We further tested whether size selections are necessary for maintaining these evolved traits. To answer this question, we randomly chose 12 isolates from each lineage and propagated them for 5 days in the absence of size selections. All isolates maintained the corresponding traits in size over different cell concentrations (Additional file 3: Figure S3).
Whole genome resequencing revealed a few mutations in growth-related genes
Amino acid substitutions
BPS (G to A)
Insertion (AG to AAG)
Frameshift from Lys658
Cell size evolution is not associated with a growth disadvantage
The rapid evolution of cell sizes implies survival strategies in the natural environments
Bacterial cell size plays important roles in survival under varying environmental conditions, such as nutrient availability and predation . For example, some bacteria of different phyla employ the strategy of size reduction or filamentation to protect from predation by protists ,. The observed short-term evolutionary change to diverse sizes may contribute to sustainability through a fluctuating grazing pressure or nutrient availability in the natural environment. Because trophodynamics, based on the grazing and nutrient availability, are quite complex in the wild, size evolution might be important or even advantageous. Previous field surveys explored size polymorphism within some genotypes ,. Microbes may accomplish total fitness via tuning their size distribution to solve complex trophodynamics.
Directed evolution toward larger size can occur without growth defects
Is it possible to evolve toward larger size without growth defects, just like the evolution to smaller size? Some bacterial strains exhibit large filamentous morphotypes that contribute to antipredation strategies. By applying the sorting gate for large-sized cells, the directed evolution toward larger size can be tested. In principle, filamentous morphotypes could be achieved in two extreme ways: a chained form of several cells without increasing each cell’s volume and an elongated form of single cells with a resulting increase in cell volume. Both types were observed in several bacteria, in certain conditions. Due to the technical limitations of the cell sorter, these two types were indistinguishable through light scattering patterns. Therefore, the outcome of the evolution might depend on growth competition between the two types. Nevertheless, we tried to introduce evolution to large-sized, filamentous cells without growth defects (L-lineage, Additional file 4: Figure S4 and Additional file 5: Figure S5A–C). After 22 days, the size distributions were unchanged from that of the ancestor (Additional file 5: Figure S5D). Microscopic observation, however, revealed that the chained form accounted for most of the longer cells (Additional file 5: Figure S5E). FM4-64 membrane staining showed several membrane septa between each cellular compartment in a single filament (Additional file 6: Figure S6). Moreover, the individual cell sizes within a filament remained unchanged. Thus, the simplest selection regime introduced the filamentous morphotype, but failed to introduce an evolution to larger volume between the septa, within a short time frame.
In addition to the instrumental limitation in selecting large cell sizes in a filament, one possible reason for the outcome is that a chained form with many dividing septa provides a growth advantage immediately after cell sorting versus the singular, elongated form. To avoid such putative growth competitions induced by the sorting inaccuracy and to eliminate the delusive morphotypes, we examined an alternative selection regime (Ls-cycle, Additional file 7: Figure S7). In addition to the growth selection during propagation, the revised method consisted of single-cell sorting for the populations with the largest cell size. The single-cell sorting could isolate a possible mutant from the chained forms, which are prone to divide immediately. A possible limitation of this method, however, is that the single-cell bottleneck might allow incidental fixation of growth-defective mutants relative to ancestors by genetic drift. After 8 days, however, the cells exhibited larger size distributions without growth defects during the exponential phase, before accumulating deleterious mutations (Additional file 8: Figure S8 and Additional file 9: Figure S9). Microscopic observations revealed a long, filamentous shape, composed of elongated cells in most cases (Additional files 1, 2 and Additional file 8: Figure S8F). Some filaments contained membrane septa, but the intervals between the septa were wider than that of the cells in the L-lineage, which is consistent with the elongation of each cell. Thus, the revised selection regime in the Ls-lineage drove the evolution to larger size within 139 generations without growth defects.
The clonal isolates from the evolved population in the Ls-lineage also exhibited large size (Additional file 8: Figure S8E). In addition, these isolates kept their large size after 5 days (80–90 generations) of propagation in the absence of size selections (Additional file 3 and Additional file 10: Figure S10). These results indicate that the evolved large size was inheritable. Unfortunately, whole genome resequencing detected no significant base pair substitutions nor small insertions and deletions. Thus, further analysis or superior techniques would be needed to detect possible genetic signatures.
Starting from a clonal E. coli population, the directed evolution of cell sizes without inducing a growth burden was tested empirically. Cell size evolved to small within 400 generations in response to directed size selections. Severe selection led rapidly to a small cell size, more so than mild selection. In addition to the mean cell size, the width of the size distribution also evolutionally changed. Importantly, the cell size evolution were unaccompanied by disadvantages to the cells’ growth rate in the absence of the size selections. In conclusion, these data indicate that bacterial cell size could evolve, by means of a few mutations, in response to size selection, without strong constraints due to trade-offs of growth rate, suggesting that the rapid evolutionary change to diverse cell sizes is important for bacterial survival strategies.
We constructed the E. coli derivative strain from DH1, DH1ΔgalK:: Plac-gfpuv5-Pkan-kan, called BSKY. To construct BSKY, we used the plasmid pLacGK (pBR322 derivative) containing the lac promoter (Plac), its operator, the derivative gene of gfp (gfpuv5), kanamycin resistance gene (kan) and its promoter (Pkan). The corresponding fragment, Plac-gfpuv5-Pkan-kan, was flanked by an rrnBT1T2 terminator (upstream of Plac) and a t7 terminator (downstream of kan). The construction of pLacGK was reported previously . The fragment of Plac-gfpuv5-Pkan-kan, including the terminators of both ends, was amplified from pLacGK with the primers T2-f (5′-aagcagaaggccatcctgacgga-3′) and T7-R2 (5′-atccggatatagttcctcctttga-3′) and was inserted into the galK gene on the plasmid pT0 for subsequent genomic recombination, as described previously . Using the primers chgalKl (5′-aagcccacgttttacggatc-3′) and chgalKr (5′-ggcccgccgtgcagctggtt-3′), this plasmid was employed as a template for PCR of the target sequence Plac-gfpuv5-Pkan-kan, flanked by the homologous sequences of galK. The amplified fragment was used for genome replacement at the chromosomal location of galK in DH1 . The final recombinant, DH1ΔgalK::Plac-gfpuv5-Pkan-kan, was called BSKY.
Bacterial cells were grown in a minimal medium, modified M63 (mM63) that contained 62 mM K2HPO4, 39 mM KH2PO4, 15 mM (NH4)2SO4, 2 μM FeSO4 · 7H2O, 15 μM thiamine hydrochloride, 203 μM MgSO4 · 7 H2O and 22 mM glucose  and was supplemented with 100 μM kanamycin (Km) and 100 μM isopropyl β-d-1-thiogalactopyranoside (IPTG). Ancestral cells were cultured at 37°C for several passages until the growth rate stabilized and were cloned before use for experimental evolution. Other culture conditions are detailed elsewhere, where relevant to the other experimental parameters.
Relative cell size, GFP fluorescence and cell concentration was evaluated using a flow cytometer (FACSAria cell sorter; Becton Dickinson) equipped with a 488-nm argon laser. Relative cell size was measured by the detector for the forward-scattering light, while GFP fluorescence was collected through a 515–545 nm emission filter (GFP). The GFP fluorescence was used to distinguish the cells from debris. The flow data were analyzed by scripts written in R . Systematic errors, resulting from events that occurred at the bottom or top of the instrument’s range, were eliminated. Cell samples, mixed with known concentrations of fluorescent beads (3 μm Fluoresbrite YG Microspheres; Polysciences), were loaded to calculate the cell concentrations. For cell sorting, the cells of particular sizes were sorted according to their forward-scattered light intensity. To correct the FSC measurements from daily variation in instrumental condition, we calibrated the measured data by daily measurements of four different beads with known diameters: 0.75 μm for Fluoresbrite Plain YG 0.75 micron Microspheres (Polysciences), 1.0 μm for Fluoresbrite calibration grade 1.0 micron YG Microspheres (Polysciences), 2.0 μm for Latex Microsphere Suspensions (Duke Scientific Corporation) and 3.0 μm for Fluoresbrite calibration grade 3.0 micron Microspheres (Polysciences).
Starting from a genetically identical cell population of BSKY, we conducted experimental rounds consisting of two simple selections: size selection via FACS and growth selection in culture. We prepared the cells for the evolutionary experiments by overnight culture. Through the evolutionary experiment, we sampled the overnight culture, and the particular fractions of the population exhibiting the target sizes were sorted to the fresh medium using FACS. The size selections were examined in the smallest 1% of the population (severe selection) and around the peak (mild selection), (Svr- and Mld-lineage, respectively). The numbers of the sorted cells were calculated from the growth rate of the previous round, so they would reach approximately 107 cells/ml in the next overnight culture. The typical values were 20 to 2000 cells in 1 ml of fresh medium. These cycles were repeated daily, along with the general serial transfer line without the size selection. We stored each sampled population at −80°C for later experiments. Every day, we calculated the number of generations per day (g) as the following equation: g = log2(Nt+Δt/Nt), where Nt is the initial cell concentration and Nt+Δt is the final cell concentration in the cell culture. We used the number of sorted cells as Nt, while Nt+Δt was measured by flow cytometry as described above. The total number of generations was calculated by summing g.
Single clone assay experiment
Cells from freezer stocks were plated on mM63 agar and incubated at 37°C for 4 days. Twelve colonies were picked from each strain and suspended in mM63 medium. Then, they were stored at −80°C with glycerol. Clonal isolated cells were inoculated into mM63 medium from freezer stocks. The initial cell concentration was 103 cells/ml, and the cells were incubated for 20 hours. These cultures were then diluted, and 102 cells were transferred to 1 ml of fresh medium. The cells were sampled every 2–3 hours. The growth rates (Malthusian parameter) during the exponential growth phase were calculated from the slopes of the growth curves according to the standard Malthusian growth model.
Genomic DNA preparation
Glycerol-stock cells were inoculated into mM63 medium and grown until OD600 = 0.5 at 37°C. The cell cultures were subsequently diluted to OD600 = 0.05 with fresh medium, and grown to stationary phase. Rifampicin (final concentration 300 μg/ml) was subsequently added, and the culture was continued for another 3 hours to block the initiation of DNA replication. The cells were collected by centrifugation at 25°C at 5000 × g for 5 min, and the pelleted cells were stored at −80°C prior to use. Genomic DNA was extracted using standard procedures, following instructions from the Aqua Pure Genomic DNA Isolation kit (Bio-Rad) and Wizard Genomic DNA Purification kit (Promega). Then, we stored the genomic DNA at −20°C.
The genomic DNA library was prepared for Roche 454 sequencing using emulsion PCR kits (GS Junior Titanium emPCR Kit (Lib-L), Roche). Whole-genome sequencing was performed on the Roche 454 GS Junior with the genomic DNA library (average read length of 386 bp; average per-site coverage of 5.2; average coverage of 98% of the genome per strain) with the appropriate kits (GS Junior Titanium Sequencing Kit and GS Junior Titanium PicoTiterPlate Kit, Roche). These preparation and analyses were examined using the Genome Information Research Center (GIRC) at Osaka University (Japan). Using the GS Reference Mapper software (ver. 2.6; Roche), these reads were then aligned onto the E. coli DH1 reference chromosome (Accession: CP001637, Version: CP001637.1, GI: 260447279, Size: 4,630,707 bp) to identify putative mutations. Candidate mutations were detected as ’HC (High Confidence) Differences,” “HC Structural Rearrangements” and “HC Structural Variants” by the software with recommended parameter settings (system default: Seed step: 12; Seed length: 16; Seed count: 1; Hit-per-seed limit: 70; Minimum overlap length: 40; Minimum overlap identity: 90; Alignment identity score: 2, Alignment difference score: −3; Repeat score threshold: 12). As detailed in the commercial manual, confidence was determined by the built-in algorithm in the software, where high-confidence was determined along with the following three rules: There must be at least 3 non-duplicate reads with the difference (1); There must be both forward and reverse reads showing the difference (2); If the difference is a single-base overcall or undercall, then the reads with the difference must form the consensus of the sequenced reads and the signal distribution of the differing reads must vary from the matching reads (3). All high-confidence SNP sites and candidates for DIP variations (deletions, insertions and inversions) were checked by capillary Sanger sequencing of PCR products amplified directly from the genome.
Cells in the exponential growth phase (106 cells/ml, 10 μl of the culture) were placed on a thin agarose pad (1.5%), and the pad was subsequently placed down on a glass dish, resulting in a monolayer of cells between the agarose pad and glass dish. The culture condition was the same as the single colony assay experiment. Images were acquired at 60× magnification using a fluorescence microscope (TE2000; Nikon) and a cooled CCD camera (DV887; Andor). The gain of the camera was 100, and exposure time was 50 ms. The images were analyzed using ImageJ software (NIH) to measure cell length.
Availability of supporting data
The data sets supporting the results of this article are included within the article and its additional files.
We thank Natsuko Tsuru, Yoshihiro Shimizu and Yoichiro Ito for technical assistance in constructing the ancestral strain. We thank Yusuke Takahashi for whole genome analysis. We also thank Shota Nakamura at the Department of Infection Metagenomics at Osaka University for whole genome resequencing.
- Justice SS, Hunstad DA, Cegelski L, Hultgren SJ: Morphological plasticity as a bacterial survival strategy. Nat Rev Microbiol. 2008, 6 (2): 162-168. 10.1038/nrmicro1820.PubMedView ArticleGoogle Scholar
- Pernthaler J: Predation on prokaryotes in the water column and its ecological implications. Nat Rev Microbiol. 2005, 3 (7): 537-546. 10.1038/nrmicro1180.PubMedView ArticleGoogle Scholar
- Miller C, Thomsen LE, Gaggero C, Mosseri R, Ingmer H, Cohen SN: SOS response induction by beta-lactams and bacterial defense against antibiotic lethality. Science. 2004, 305 (5690): 1629-1631. 10.1126/science.1101630.PubMedView ArticleGoogle Scholar
- St-Pierre F, Endy D: Determination of cell fate selection during phage lambda infection. Proc Natl Acad Sci U S A. 2008, 105 (52): 20705-20710. 10.1073/pnas.0808831105.PubMedPubMed CentralView ArticleGoogle Scholar
- Young KD: The selective value of bacterial shape. Microbiol Mol Biol Rev. 2006, 70 (3): 660-703. 10.1128/MMBR.00001-06.PubMedPubMed CentralView ArticleGoogle Scholar
- Donachie WD: Relationship between cell size and time of initiation of DNA replication. Nature. 1968, 219 (5158): 1077-1079. 10.1038/2191077a0.PubMedView ArticleGoogle Scholar
- Weart RB, Lee AH, Chien AC, Haeusser DP, Hill NS, Levin PA: A metabolic sensor governing cell size in bacteria. Cell. 2007, 130 (2): 335-347. 10.1016/j.cell.2007.05.043.PubMedPubMed CentralView ArticleGoogle Scholar
- Chien AC, Hill NS, Levin PA: Cell size control in bacteria. Curr Biol. 2012, 22 (9): R340-R349. 10.1016/j.cub.2012.02.032.PubMedPubMed CentralView ArticleGoogle Scholar
- Stearns SC: Trade-offs in life-history evolution. Funct Ecol. 1989, 3 (3): 259-268. 10.2307/2389364.View ArticleGoogle Scholar
- Lehner B: Conflict between noise and plasticity in yeast. PLoS Genet. 2010, 6 (11): e1001185-10.1371/journal.pgen.1001185.PubMedPubMed CentralView ArticleGoogle Scholar
- Hill NS, Kadoya R, Chattoraj DK, Levin PA: Cell size and the initiation of DNA replication in bacteria. PLoS Genet. 2012, 8 (3): e1002549-10.1371/journal.pgen.1002549.PubMedPubMed CentralView ArticleGoogle Scholar
- Hill NS, Buske PJ, Shi Y, Levin PA: A moonlighting enzyme links Escherichia coli cell size with central metabolism. PLoS Genet. 2013, 9 (7): e1003663-10.1371/journal.pgen.1003663.PubMedPubMed CentralView ArticleGoogle Scholar
- Yao Z, Davis RM, Kishony R, Kahne D, Ruiz N: Regulation of cell size in response to nutrient availability by fatty acid biosynthesis in Escherichia coli. Proc Natl Acad Sci U S A. 2012, 109 (38): E2561-E2568. 10.1073/pnas.1209742109.PubMedPubMed CentralView ArticleGoogle Scholar
- Typas A, Banzhaf M, Gross CA, Vollmer W: From the regulation of peptidoglycan synthesis to bacterial growth and morphology. Nat Rev Microbiol. 2012, 10 (2): 123-136.Google Scholar
- Kruse T, Bork-Jensen J, Gerdes K: The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex. Mol Microbiol. 2005, 55 (1): 78-89. 10.1111/j.1365-2958.2004.04367.x.PubMedView ArticleGoogle Scholar
- Lenski RE, Travisano M: Dynamics of adaptation and diversification: a 10,000-generation experiment with bacterial populations. Proc Natl Acad Sci U S A. 1994, 91 (15): 6808-6814. 10.1073/pnas.91.15.6808.PubMedPubMed CentralView ArticleGoogle Scholar
- Stanek MT, Cooper TF, Lenski RE: Identification and dynamics of a beneficial mutation in a long-term evolution experiment with Escherichia coli. BMC Evol Biol. 2009, 9: 302-10.1186/1471-2148-9-302.PubMedPubMed CentralView ArticleGoogle Scholar
- Chou HH, Chiu HC, Delaney NF, Segre D, Marx CJ: Diminishing returns epistasis among beneficial mutations decelerates adaptation. Science. 2011, 332 (6034): 1190-1192. 10.1126/science.1203799.PubMedPubMed CentralView ArticleGoogle Scholar
- Stewart EJ, Madden R, Paul G, Taddei F: Aging and death in an organism that reproduces by morphologically symmetric division. PLoS Biol. 2005, 3 (2): e45-10.1371/journal.pbio.0030045.PubMedPubMed CentralView ArticleGoogle Scholar
- Akerlund T, Nordstrom K, Bernander R: Analysis of cell size and DNA content in exponentially growing and stationary-phase batch cultures of Escherichia coli. J Bacteriol. 1995, 177 (23): 6791-6797.PubMedPubMed CentralGoogle Scholar
- Drake JW, Charlesworth B, Charlesworth D, Crow JF: Rates of spontaneous mutation. Genetics. 1998, 148 (4): 1667-1686.PubMedPubMed CentralGoogle Scholar
- Kishimoto T, Iijima L, Tatsumi M, Ono N, Oyake A, Hashimoto T, Matsuo M, Okubo M, Suzuki S, Mori K, Kashiwagi A, Furusawa C, Ying BW, Yomo T: Transition from positive to neutral in mutation fixation along with continuing rising fitness in thermal adaptive evolution. PLoS Genet. 2010, 6 (10): e1001164-10.1371/journal.pgen.1001164.PubMedPubMed CentralView ArticleGoogle Scholar
- Jung YM, Lee JN, Shin HD, Lee YH: Role of tktA gene in pentose phosphate pathway on odd-ball biosynthesis of poly-beta-hydroxybutyrate in transformant Escherichia coli harboring phbCAB operon. J Biosci Bioeng. 2004, 98 (3): 224-227. 10.1016/S1389-1723(04)00272-5.PubMedView ArticleGoogle Scholar
- Jung IL, Phyo KH, Kim IG: RpoS-mediated growth-dependent expression of the Escherichia coli tkt genes encoding transketolases isoenzymes. Curr Microbiol. 2005, 50 (6): 314-318. 10.1007/s00284-005-4501-1.PubMedView ArticleGoogle Scholar
- Hahn MW, Moore ER, Hofle MG: Bacterial filament formation, a defense mechanism against flagellate grazing, is growth rate controlled in bacteria of different phyla. Appl Environ Microbiol. 1999, 65 (1): 25-35.PubMedPubMed CentralGoogle Scholar
- Bernard L, Courties C, Servais P, Troussellier M, Petit M, Lebaron P: Relationships among bacterial cell size, productivity, and genetic diversity in aquatic environments using cell sorting and flow cytometry. Microb Ecol. 2000, 40 (2): 148-158.PubMedGoogle Scholar
- Servais P, Courties C, Lebaron P, Troussellier M: Coupling bacterial activity measurements with cell sorting by flow cytometry. Microb Ecol. 1999, 38 (2): 180-189. 10.1007/s002489900160.PubMedView ArticleGoogle Scholar
- Ying BW, Ito Y, Shimizu Y, Yomo T: Refined method for the genomic integration of complex synthetic circuits. J Biosci Bioeng. 2010, 110 (5): 529-536. 10.1016/j.jbiosc.2010.05.014.PubMedView ArticleGoogle Scholar
- Kashiwagi A, Sakurai T, Tsuru S, Ying BW, Mori K, Yomo T: Construction of Escherichia coli gene expression level perturbation collection. Metab Eng. 2009, 11 (1): 56-63. 10.1016/j.ymben.2008.08.002.PubMedView ArticleGoogle Scholar
- Ihaka R, Gentleman R: R: A language for data analysis and graphics. J Comput Graph Stat. 1996, 5 (3): 299-314.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.