Differential replication dynamics for large and small Vibrio chromosomes affect gene dosage, expression and location
© Dryselius et al; licensee BioMed Central Ltd. 2008
Received: 10 October 2008
Accepted: 26 November 2008
Published: 26 November 2008
Replication of bacterial chromosomes increases copy numbers of genes located near origins of replication relative to genes located near termini. Such differential gene dosage depends on replication rate, doubling time and chromosome size. Although little explored, differential gene dosage may influence both gene expression and location. For vibrios, a diverse family of fast growing gammaproteobacteria, gene dosage may be particularly important as they harbor two chromosomes of different size.
Here we examined replication dynamics and gene dosage effects for the separate chromosomes of three Vibrio species. We also investigated locations for specific gene types within the genome. The results showed consistently larger gene dosage differences for the large chromosome which also initiated replication long before the small. Accordingly, large chromosome gene expression levels were generally higher and showed an influence from gene dosage. This was reflected by a higher abundance of growth essential and growth contributing genes of which many locate near the origin of replication. In contrast, small chromosome gene expression levels were low and appeared independent of gene dosage. Also, species specific genes are highly abundant and an over-representation of genes involved in transcription could explain its gene dosage independent expression.
Here we establish a link between replication dynamics and differential gene dosage on one hand and gene expression levels and the location of specific gene types on the other. For vibrios, this relationship appears connected to a polarisation of genetic content between its chromosomes, which may both contribute to and be enhanced by an improved adaptive capacity.
Vibrios constitute a broad family of gammaproteobacteria with over 100 members classified (NCBI taxonomy browser). They are ubiquitous within marine and estuarine environments and the ecological roles for individual species are diverse. A common characteristic, however, is their ability to adapt and survive within various niches either as free-swimmers or in symbiotic or pathogenic association with diverse aquatic organisms such as plankton, coral, fish and shellfish. Moreover, several Vibrio species are capable of infecting humans with Vibrio cholerae, V. parahaemolyticus and V. vulnificus as the most common causes of disease [1, 2].
A shared trait among vibrios is the presence of two unequally sized chromosomes . The larger shows a more constant size, a lower interspecies sequence variability and harbour many of the genes involved in essential biosynthetic pathways while the smaller Vibrio chromosome is highly variable in size and contains relatively more species specific and unclassified genes [3–6]. This unusual structure and the distinct distribution of genetic content between the replicons has prompted studies on how the system is maintained and has also initiated discussion about fitness benefits with a divided genome [4, 6, 7].
Regarding the maintenance issue, most knowledge about chromosomal replication and partitioning has been gained from studies on V. cholerae. For example, it has been shown that the two chromosomes display different segregation patterns [8–10] and utilise separate sets of partition proteins [11, 12]. It has further been demonstrated that while the large chromosome origin of replication is similar to oriC of Escherichia coli, the small bears resemblance to those of certain plasmids . Nevertheless, the number of initiations for the two replicons remains equal and strictly follows the cell cycle . Therefore, as the difference in size between the replicons theoretically results in differing replication times it was suggested that initiation of replication is synchronised to maintain an equal number of small and large chromosomes . However, more recent studies indicate that inter-chromosomal synchrony between the V. cholerae chromosomes likely occurs at the level of termination [15, 16].
Bacteria with divided genomes must overcome additional obstacles to accurately distribute genetic material to daughter cells, yet the evolutionary success of the broad and diverse Vibrio family implies that split genomes may be beneficial. A possible advantage is that multiple replicons allow faster replication which in turn could lead to faster growth rates . This view is supported by the fact that several Vibrio species display unusually short multiplication times [18, 19]. Another potential benefit may be that multiple chromosomes provide the means to regulate gene expression in a replicon-wide manner by alterations in the 1:1 balance between copy numbers . Such regulation could facilitate large scale adaptations in response to changes in growth conditions , for example when the bacterium associates with or dissociates from a host organism. Considering that the genetic content differs for the small and large Vibrio chromosomes and that the two chromosomes utilise partly different mechanisms for initiation of replication  and partitioning , this is not an unlikely assumption. Consistent with this idea, over-expression of the distinct large or small chromosome replication initiator protein of V. cholerae results in over-initiation of the respective chromosome they control . Also, V. cholerae cells harboring unequal numbers of small and large chromosomes were recently obtained by deleting the small chromosome specific partitioning genes . Although these studies reveal technical possibilities for alterations in chromosome balance, no wild type vibrio with differing numbers of large and small chromosomes has yet been detected.
Interestingly, vibrios may possess an intrinsic mechanism to differentiate gene copy numbers between the chromosomes through gene dosage associated with replication [15, 21]. Such gene dosage occurs as replication always initiates at an origin and proceeds in a bidirectional manner towards the terminus of replication. This mode of replication means that genes located near the origin are duplicated earlier than other genes, which enables higher expression. Therefore, with a given replication speed, there are two main factors that influence gene dosage; initiation rate and replicon size. A higher initiation rate results in an increased average difference in copy numbers between origin proximate and terminus proximate genes as replication takes up a larger proportion of the cell cycle. For the same reason, gene dosage differences are more pronounced for a larger over a smaller replicon assuming equal initiation frequencies .
Although an influence from gene dosage on bacterial gene expression levels has only occasionally been reported [22–24], it is assumed to affect gene positioning . In line with this, altered gene dosage has been used as an explanation for decreased fitness and sometimes deleterious effects that can follow chromosomal rearrangements [26–28] as such events simultaneously change the distance to the origin of replication and thereby average copy numbers for a large number of genes. The impact from gene dosage has also been examined at the scale of genomic conservation, and highly expressed genes tend to locate near the origin of replication, especially for bacteria with fast cell division rates [21, 29]. Furthermore, it has been reported that a high level of gene dosage correlates with a higher degree of genomic stability as fast multiplying bacteria display a stronger conservation of gene positioning . Therefore, despite the fact that direct experimental evidence is scarce, these reports give support for gene dosage as an important factor in the evolution of bacterial genomes, especially for fast growing species.
In the case of vibrios, which have their genomes distributed between two unequally sized chromosomes and also display very short multiplication times, differential gene dosage could have a strong impact. The short multiplication times would result in large gene dosage differences within each chromosome. Also, assuming an equal replication speed for the two replicons, size differences would lead to different gene dosage between the chromosomes. Indeed, a study employing flow cytometry in combination with computer modelling revealed convincing experimental evidence for increased gene dosage differences within the large than within the small chromosome of actively growing V. cholerae . In addition, this examination showed growth rate dependent variations in relative gene copy numbers both within and between the chromosomes and that differing timings of replication initiation creates overall highest copy numbers for genes located near the origin of replication of the large chromosome. A later study employing fluorescence microscopy to detect relative abundances of large and small chromosome origins of replication confirmed these enhanced and growth rate dependent gene dosage differences for the large chromosome relative to the small . However, this report also indicated some question marks regarding an earlier replication start for the large chromosome, at least under certain growth conditions.
In an attempt to extend current knowledge about vibrio replication dynamics and gain insight into how this affect expression and genetic distribution, we here employed real-time PCR to quantify relative abundances of origin and terminus proximate DNA for both chromosomes of actively growing V. parahaemolyticus, V. cholerae and V. vulnificu s. For V. parahaemolyticus, microarray analyses at both the genomic and transcription levels were also conducted. We further examined the location of distinct gene types within five sequenced and annotated Vibrionaceae genomes and related this to replication and expression patterns.
Estimates of large and small chromosome origin/terminus ratios based on doubling times
Strain designations, purpose of use and chromosome sizes for bacteria employed in this study
Size large chr (Mb)
Size small chr (Mb)
Vibrio parahaemolyticus RIMD 2210633
Vibrio cholerae El Tor Inaba RIMD 2203577
Vibrio vulnificus ATCC27562
Vibrio cholerae O1 El Tor N16961
Vibrio vulnificus YJ016
Vibrio fischeri ES114
Photobacterium profundum SS9
Doubling times, replication times and theoretical origin/terminus ratios for V. parahaemolyticus, V. cholerae and V. vulnificus
Replication time large chr*
Replication time small chr*
Ori/ter ratio large chr#
Ori/ter ratio small chr#
Based on the established maximum replication fork movement of 1000 nt/s for E. coli , chromosome size can be used to estimate replication time. With an estimated replication time, knowledge about the doubling time enables predictions about origin/terminus (ori/ter) ratios for a replicon . For this, we used the chromosome sizes given for the genomic strain of V. parahaemolyticus RIMD2210633, and employed pulsed-field gel electrophoresis to determine approximate sizes for the chromosomes of V. cholerae RIMD2203577 and V. vulnificus ATCC27562 (Table 1) before calculating ori/ter ratios for each separate chromosome (Table 2). The highest ratio (4.31) is estimated for the large chromosome of the quickly multiplying V. parahaemolyticus. These cells are also supposed to show relatively large inter-chromosomal difference in ori/ter ratios (4.31/2.30 = 1.87). For the large chromosomes of V. cholerae and V. vulnificus, the approximately equal ratios (2.50 and 2.67, respectively) reflect that the faster doubling time for the former is compensated by the larger size for the latter. Similarly, a relatively low ori/ter ratio for the small V. cholerae chromosome (1.47) is explained by its small size. Furthermore, the large size difference between the two V. cholerae chromosomes gives an inter-chromosomal ori/ter ratio (2.50/1.47 = 1.70) that is only slightly lower than for the much faster multiplying V. parahaemolyticus cells. Finally, a comparison between V. parahaemolyticus grown in rich and minimal media suggests that large variations in ori/ter ratios can be expected, especially for the large chromosome.
Quantification of large and small chromosome origins and termini with RT-qPCR
To better visualise gene dosage differences, ori/ter ratios were determined for each chromosome (Figure 1Ab, Bb and 1Cb), and in all instances the large chromosomes show a significantly higher ori/ter ratio than the small (P < 0.05). A comparison between V. parahaemolyticus grown in minimal and rich media at 37°C shows a much larger increase in gene dosage for the large than for the small chromosome (Figure 1Ab). This is consistent with doubling time based calculations (Table 2) and agrees with previous results from V. cholerae demonstrating that growth rate has a much larger impact on gene dosage for the large than for the small chromosome [15, 16]. For low temperature cultures in rich media, however, there appear to be only minor differences in ori/ter ratios compared to rich media cultures grown at 37°C (Figure 1Ab, Bb and 1Cb). With consideration to the much faster doubling times at the higher temperature, this could suggest that replication is temporarily blocked or slowed in low temperature grown cells.
Microarray based visualisation of replication dynamics
Previous analyses have indicated a relatively fast replication progress for V. cholerae with replication fork movements around  or just below  1000 nt/s for cells grown at 37°C. The slightly lower gene dosage differences for the RT-qPCR and microarray analyses compared to the doubling time based estimates (cf. Figure 1 and 2 with Table 2) indicate an average replication speed that may exceed 1000 nt/s. However, it must be considered that determination of a correct replication speed relies on a number of factors and inconsistencies may be due to (i) variations in doubling times estimates, (ii) differences in strains and growth media, (iii) errors in DNA target quantifications, (iv) errors caused by sampling handling and (v) the possibility that reference samples display a certain degree of replicating activity. Nevertheless, our results confirm a fast replication speed for vibrios. In addition, the replication speed seems slowed down to approximately one third for cells grown at 20°C as gene dosage differences were maintained (Figure 1 and 2) while doubling times increased threefold (Table 2).
Microarray based examination of large and small chromosome expression
Although gene dosage and growth environment appear to have an influence on expression levels, the expression patterns also suggest other impacts. Several examinations have pointed out the presence of regularities in bacterial gene expression data where local expression maxima are found at periodicities of around  or slightly above [33, 34] 100 kb. These patterns have been explained by higher order nucleoid structuring where the DNA is compacted into one or two large helices containing loops of approximately 100–120 kb lengths [32–34]. To examine whether periodicities are present in our expression data, grids were fitted to match local peaks. For both chromosomes periodic patterns of approximately 100 kb in length were detected (Additional file 3). Therefore, it appears like, in addition to gene dosage, also higher order chromosomal structuring has an influence on vibrio expression.
Examination of genetic distribution within the Vibrio genome
The above experiments show that differing size and initiation timing creates higher average gene dosage and gene copy numbers on the large Vibrio chromosome. The results also show that expression levels from the large but not the small chromosome tend to follow gene dosage in a growth rate dependent manner. To gain a better understanding of these replication and expression patterns we next examined the distribution of different gene types within the Vibrio genome.
We next analysed gene distribution between the early and late replicated parts of the large V. cholerae chromosome. The early replicated part was defined as being replicated before initiation of the small chromosome, assuming an equal bi-directional replication speed for both chromosomes and a synchronous termination. The results show that orthologs to both growth essential and the most growth contributing genes are over-represented within the early replicated part, while orthologs to the least growth contributing genes do not show a significantly biased distribution (Figure 4B). This distribution is in agreement with a previous notion that essential genes, and especially highly expressed such genes, show a clear tendency to locate near the origin of replication in E. coli . The explanation given for this distribution was that the highly expressed essential genes benefit from a high gene dosage. With consideration to our expression results, this argument also seems suitable for vibrios. An additional benefit with an early replication for such genes in vibrios could be that this gives the cells more time to build up enough supplies of gene products required later in the cell cycle, including the additional metabolic burden of replicating the small chromosome . It is also possible that an early replication of genes important for growth allow the cells to quickly adopt their expression to changes in growth conditions.
We also compared the distribution of orthologs to growth related genes between the late replicating part of the large chromosome and the small chromosome. Essential and most growth contributing genes were clearly under-represented on the small chromosome while genes contributing the least to growth appeared more randomly distributed (Figure 4C). This distribution contrasts with the observation that these two genome parts show a similar gene dosage (Figure 2). However, the disproportionally low number of growth important genes on the small chromosome agrees with the gene dosage and growth rate independent expression observed for this part of the genome (Figure 3). In summary, the results displayed in Figure 4 show that the distribution of genes central for growth is connected to replication timing and gene dosage effects.
Until recently it was believed that the two chromosomes of V. cholerae always initiated replication in a synchronous manner . However, Rasmussen and co-workers introduced a new paradigm for V. cholerae replication dynamics when suggesting and confirming a model of termination synchrony . Furthermore, they offered an explanation for why previous results had indicated initiation synchrony in that only large chromosome initiation is inhibited by treatment with the antibiotic rifampicin, which is commonly used prior to chromosome copy number determinations. The data shown in Figure 1 and 2 was obtained without antibiotics and confirm a model of termination synchrony. In addition, the results show that this model is applicable to additional Vibrio species and also to a differing rate of replication.
The generally high, partly gene dosage and growth rate dependent expression levels from the large chromosome is in agreement with vibrio replication dynamics. The high abundance of genes important for growth near the origin of replication of the large chromosome also makes sense as this provides both a fast and a powerful control of their expression in response to changed growth conditions. However, the gene dosage independent expression pattern for the small chromosome is more difficult to, at least directly, explain by replication dynamics. Instead, a high abundance of transcription related genes could be responsible for the more stringent expression regulation. Such tighter expression regulation is also in agreement with a disproportionally low abundance of growth important genes, even in comparison to the part of the large chromosome that is replicated at the same time. In addition, it is possible that the high abundance of species specific genes on the small chromosome further contributes to its low and gene dosage/growth rate independent expression levels. A support for this is that more newly acquired genes tend to be much more stringently regulated than genes with a long history within the genome [45, 46].
Although the differences in gene dosage dependency between the two chromosomes may be explained by genetic distribution, a remaining question is why this genetic arrangement has evolved. One explanation could be that the lack/very low abundance of gene dosage depending genes on the small chromosome makes it more genetically flexible. Such flexibility could benefit the bacteria as a whole in that new genetic traits that improve the adaptive capacity are more easily gained . Translocation of growth essential and growth contributing genes from the large to the small chromosome could decrease this flexibility which may explain why such genes are maintained on the large chromosome. An additional factor that likely restricts movement of growth important genes is the earlier replication start for the large chromosome. As we show here, a vast majority of growth essential and growth contributing genes locate on the origin-proximate part of the large chromosome. Translocation to any other part of the genome would delay their replication timing and decrease their dosage which could affect overall bacterial fitness. Indeed, support for a differential genetic flexibility is found in pair-wise comparisons of genetic content of different Vibrio species, which show a much higher variability between the small chromosomes [4, 5]. In addition, these comparisons reveal a larger variability within the terminus-proximate than within the origin-proximate part of the large chromosome, which has also been noted by others [38, 40]. Therefore, flexibility issues and, hence, adaptive advantages may, together with differing initiation timings, be responsible for the distinct genetic distribution and differential gene dosage dependency that appears to be maintained throughout the vibrio family.
With the above discussion in mind, it is interesting that genetic arrangements similar to that of vibrios are also found within other bacterial families. Inter-strain and inter-species comparisons of the alpha-3 subgroup proteobacteria Rhodobacter sphaeroides  and the alpha-2 subgroup Brucella family [48, 49], respectively, display the presence of a large evolutionary conserved and a small fast evolving chromosome. Similarly, whole genome sequences of the beta-proteobacteria Ralstonia solanacearum  and Burkholderia pseudomallei  reveal the presence of a larger replicon that harbors most genes related to growth and survival and a smaller replicon that contain disproportionately high numbers of unclassified genes and genes related to transcription. Although replication dynamics and gene dosage effects have not been examined for these species, the strong genomic similarities with vibrios suggest that both of these factors could be influential. Moreover, the fact that a similar genomic structure appears to have arose more than once support the idea that it provides an important fitness advantage.
Bacterial strains and growth conditions
Bacterial strains used for experimental and computational analyses in this study are listed in Table 1. V. parahaemolyticus, V. cholerae and V. vulnificus from frozen glycerol stocks were grown in 2 ml cultures overnight at 20 or 37°C in either M9 media supplemented with 3% NaCl (w/v) and 0.4% glucose (3%M9), Luria-Bertani (LB) broth or LB containing 3% NaCl (3%LB). Cultures were diluted 1:1000 in 2 ml fresh media and incubated at 37 or 20°C with constant shaking at 210 rpm. At harvest, cultures were immediately transferred to – 30°C ice/NaCl/ethanol slurry followed by centrifugation (8000 g, 3 min, 4°C). To obtain non-replicating cells, overnight cultures were either plated onto LB agar followed by a 24 h incubation at 20°C and an additional 24 h incubation at 4°C or incubated in 2 ml liquid media (3%M9, LB or 3%LB) for 24 h at 20 or 37°C. Stationary phase liquid cultures were treated for 2 h with 20 μl rifampicin (50 mg/ml dissolved in DMSO) before harvest to finish ongoing replication rounds without initiating new ones.
Pulsed-field gel electrophoresis
Estimates of chromosomal sizes were performed with pulsed-field gel electrophoresis (PFGE) on a CHEF Mapper XA system using reagents provided in the GenePath Univeral Module (Bio-Rad). Samples of V. cholerae, V. vulnificus and V. parahaemolyticus were prepared according to the manufacturers' instructions and loaded along with a molecular marker onto a 0.8% agarose gel prepared with and run in 1xTAE (40 mM Tris-acetate, 1 mM EDTA, pH 8.0) containing 500 μM thiourea . The electrophoresis was run at 14°C for 48 h with a switch time of 500 s at 3 V/cm and an included angle of 106°.
Determination of doubling times and estimations of replication times and origin/terminus ratios
Given that initiation of replication occurs simultaneously for all origins on a replicon  it can be assumed that every new required round of replication results in a doubling of the origin/terminus ratio (rO/T). The following formula, taken from , was used to calculate (rO/T):
rO/T = 2C/τ
Calculations are summarised in Table 2.
Extraction of nucleic acids
Genomic DNA (gDNA) was extracted with DNeasy® Tissue Kit (Qiagen) following the manufacturer's instructions including a prolonged RNase treatment step (15 minutes) and was finally eluted in 2 × 100 μl AE buffer. DNA samples were further purified by addition of equal volumes of phenol and chloroform with subsequent phase separation by centrifugation before precipitation with 1/10 volume 3 M CH3COONa (pH 5.2) and 2.6 volumes ethanol. DNA was pelleted by centrifugation and further washed with 1 ml 70% ethanol before being dissolved in H2O.
Total RNA was extracted from bacterial pellets with 1 ml TRIzol® Reagent (Invitrogen) following the manufacturer's protocol. RNA pellets were re-suspended and treated with DNase I according to instructions that accompany the RNase-free DNaseI set (Qiagen) before purification, precipitation and wash with phenol/chloroform, 3 M CH3COONa (pH 5.2)/ethanol and 70% ethanol, respectively. RNA was re-dissolved in 100 μl H2O and further purified using RNeasy® Mini Kit (Qiagen) before elution in H2O. RNA was subjected to a further precipitation (3 M CH3COONa (pH 5.2)/ethanol) and wash (70% ethanol) before finally being dissolved in H2O.
Relative quantification of origins and termini for both Vibrio chromosomes was carried out with either 10 ng (for V. parahaemolyticus) or 2.5 ng (for V. cholerae and V. vulnificus) gDNA with real time quantitative polymerase chain reaction (RT-qPCR). Reaction volumes were 20 μl and also included 1× Power SYBR® Green PCR Master Mix (Applied Biosystems) and either of the primer pairs listed in Additional file 1. Importantly, primer pairs were selected to target stably integrated regions not thought capable to excise from their respective places on the genomes. An exception was a primer pair targeting a phage like element present at both V. parahaemolyticus termini that was selected to provide an additional reference for this species. As we aimed to quantify relative numbers of large and small chromosome origins and termini in exponentially growing cultures, reference samples with known relative quantities of these regions were required. To provide such references, gDNA was extracted from non-replicating cells that were pre-grown on LB-plates and incubated for 24 h at 4°C. To verify an equal relationship between large and small chromosome origin and termini, the samples were compared against gDNA from exponentially grown cells along with gDNA from non-replicating cells from liquid media cultures grown under two (V. vulnificus) or three (V. parahaemolyticus and V. cholerae) different conditions (see "Bacterial strains and growth conditions"). Similar ratios between origins and termini were obtained for all non-replicating samples, which confirmed their reliability as reference samples (Additional file 2). In addition, a pyrosequencing based analysis with a 20 fold coverage of the whole genome of V. parahaemolyticus confirmed approximately equal numbers of small and large chromosomes for stationary phase cells in 3%LB (unpublished data from the lab). Amplification efficiencies were validated with gDNA from both non-replicating (three separate experiments with five replicates) and exponentially grown cells (one experiment with five replicates) over a range of template concentrations (2.5–40 ng for V. parahaemolyticus and 0.625–10 ng for V. cholerae and V. vulnificus). Near linear dose responses and similar amplification efficiencies (89–100% for V. parahaemolyticus, 97–104% for V. cholerae, and 98–103% for V. vulnificus) were obtained by analysis with the 2-ΔΔCT method  and indicated that reliable comparisons between the targets could be performed. Experimental analyses were performed in five double samples on gDNA extracted at three or five different occasions.
Preparation of aminoallyl-labelled nucleic acids derived from gDNA and RNA
For generation of aminoallyl-labelled product from gDNA, the procedures described in  were followed. For generation of aminoallyl-labelled product from total RNA we used reagents included in SuperScript™ III Reverse Transcriptase Kit (Invitrogen). In brief, 20 μg RNA was mixed with 10 μg random hexamers in a 22 μl reaction that was incubated at 70°C for 5 min before cooling on ice. Addition of 5× First-Strand Buffer (8 μl), 0.1 M DTT (2 μl), SuperScript™ III RT (4 μl) and 10× dNTP-aminoallyl dUTP (4 μl) was followed by a 3 h incubation at 46°C. Samples were precipitated and washed as described earlier.
Coupling with Cy3 and Cy5 dyes, hybridisation onto microarray slides and scanning
Coupling of aminoallyl-labelled products with Cy3 or Cy5 monofunctional reactive dyes was performed as described previously . Microarray slides and hybridisation procedures are also described in , except that human CotI DNA (Invitrogen) replaced yeast tRNA and incubation/hybridisation temperatures were 55°C instead of 60°C. Fluorescence signals were measured and analysed according to previous descriptions . Microarray data was submitted to Gene Expression Omnibus (GEO)  and has the serial number GSE9968.
To examine the distribution of essential and growth related genes in the Vibrio genome, we used growth and essentiality data from a genome-wide single gene knock-out study performed in E. coli  and deduced the nearest orthologs in V. cholerae. E. coli genes were classified into three groups; one containing all essential genes, a second (most growth contributing genes) containing the genes for which knock-outs displayed the slowest growth rates in LB media (OD600 < 0.604, see ), and a third (least growth contributing genes) comprising genes for which knock-outs displays the highest growth rates in LB media (OD600 > 0.823, see ). Next, protein sequences for the essential, most growth contributing and least growth contributing E. coli genes were searched against the NCBI COG database in BLASTO  using the default settings (E = 0.001, BLOSUM62). Best hits were sorted into three parts of the V. cholerae genome; an early replicated part of the large chromosome (defined as the part being replicated before initiation of the small chromosome assuming an equal bidirectional replication speed and a simultaneous replication termination), a late replicated part of the large chromosome (the part of the large chromosome that is not defined as being early replicated) and the small chromosome. Deviations from an average distribution were determined by comparison to the total number of genes within the different parts of the genome and significance levels were determined by chi-square tests for comparison of two proportions.
To examine the distribution of different gene categories, the classification system from Clusters of Orthologous Groups of proteins (COGs)  was employed. Genes belonging to 21 functional categories were counted within the early and late replicated parts of the large and the small chromosome of five Vibrionaceae species (see Table 1). Relative over- and under-representation was determined by comparing the size of each group against an average distribution of COGs within each genome part. Significance levels for deviations from average distributions were determined by chi-square tests for comparison of two proportions. Categories showing a similar deviation from an average in a specific genome part for all Vibrionaceae species were considered over- or under-represented and categories with a similar and significant deviation for each species were considered highly over- or under-represented.
To examine the relative abundance of growth essential or highly growth contributing genes within the separate COG categories, we first determined the number of essential and highly growth contributing genes for each COG. Proportions of essential genes within each category were compared to the proportion of essential genes in the whole genome while proportions of highly growth contributing genes among the non-essential genes in each category were compared to the proportion highly growth contributing among all non-essential genes. Significance levels for deviations were determined by chi-square tests for comparison of two proportions.
We thank Liam Good and Richard Culleton for suggestions and comments on the manuscript. We are grateful to Satoru Kuhara, Kosuke Tashiro, Ken Kurokawa and Tetsuya Hayashi for their assistance in microarray preparation, and Toru Tobe, Hiroyuki Abe and Yoshitoshi Ogura for their valuable advice. We also thank Daisuke Okuzaki at the DNA-chip Development Center for Infectious Diseases RIMD, Osaka University, for technical support for analysing the microarray data. A special thanks to Alistair Chalk for time saving advice on a computational analysis and to Tatsuya Takagi for answering questions on statistical issues. RD was financially supported through a "Postdoctoral Fellowship for Foreign Researchers" provided by the Japan Society for the Promotion of Science. Further support was provided by Grants-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
- Thompson FL, Iida T, Swings J: Biodiversity of Vibrios. Microbiol Mol Biol Rev. 2004, 68: 403-431. 10.1128/MMBR.68.3.403-431.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Thompson FL, Austin B, Swings J: The Biology of Vibrios. 2006, Washington DC: ASM PressView ArticleGoogle Scholar
- Okada K, Iida T, Kita-Tsukamoto K, Honda T: Vibrios commonly possess two chromosomes. J Bacteriol. 2005, 187: 752-757. 10.1128/JB.187.2.752-757.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Iida T, Kurokawa K: Comparative genomics: genome configuration and the driving forces of evolution of Vibrios. The Biology of Vibrios. Edited by: Thompson FL, Austin B, Swings J. 2006, Washington DC: ASM Press, 67-75.View ArticleGoogle Scholar
- Reen FJ, Almagro-Moreno S, Ussery D, Boyd EF: The genomic code: inferring Vibrionaceae niche specialization. Nat Rev Microbiol. 2006, 4: 697-704. 10.1038/nrmicro1476.PubMedView ArticleGoogle Scholar
- Dryselius R, Kurokawa K, Iida T: Vibrionaceae a versatile bacterial family with evolutionary conserved variability. Res Microbiol. 2007, 158: 479-486. 10.1016/j.resmic.2007.04.007.PubMedView ArticleGoogle Scholar
- Egan ES, Fogel MA, Waldor MK: Divided genomes: negotiating the cell cycle in prokaryotes with multiple chromosomes. Mol Microbiol. 2005, 56: 1129-1138. 10.1111/j.1365-2958.2005.04622.x.PubMedView ArticleGoogle Scholar
- Fogel MA, Waldor MK: Distinct segregation dynamics of the two Vibrio cholerae chromosomes. Mol Microbiol. 2005, 55: 125-136. 10.1111/j.1365-2958.2004.04379.x.PubMedView ArticleGoogle Scholar
- Srivastava P, Fekete RA, Chattoraj DK: Segregation of the replication terminus of the two Vibrio cholerae chromosomes. J Bacteriol. 2006, 188: 1060-1070. 10.1128/JB.188.3.1060-1070.2006.PubMedPubMed CentralView ArticleGoogle Scholar
- Fiebig A, Keren K, Theriot JA: Fine-scale time-lapse analysis of the biphasic dynamic behaviour of the two Vibrio cholerae chromosomes. Mol Microbiol. 2006, 60: 1164-1178. 10.1111/j.1365-2958.2006.05175.x.PubMedPubMed CentralView ArticleGoogle Scholar
- Heidelberg JF, Eisen JA, Nelson WC, Clayton RA, Gwinn ML, Dodson RJ, Haft DH, Hickey EK, Peterson JD, Umayam L: DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature. 2000, 406: 477-483. 10.1038/35020000.PubMedView ArticleGoogle Scholar
- Yamaichi Y, Fogel MA, Waldor MK: par genes and the pathology of chromosome loss in Vibrio cholerae. Proc Natl Acad Sci USA. 2007, 104: 630-635. 10.1073/pnas.0608341104.PubMedPubMed CentralView ArticleGoogle Scholar
- Egan ES, Waldor MK: Distinct replication requirements for the two Vibrio cholerae chromosomes. Cell. 2003, 114: 521-530. 10.1016/S0092-8674(03)00611-1.PubMedView ArticleGoogle Scholar
- Egan ES, Lobner-Olesen A, Waldor MK: Synchronous replication initiation of the two Vibrio cholerae chromosomes. Curr Biol. 2004, 13: R501-R502. 10.1016/j.cub.2004.06.036.View ArticleGoogle Scholar
- Rasmussen T, Jensen RB, Skovgaard O: The two chromosomes of Vibrio cholerae are initiated at different time points in the cell cycle. EMBO J. 2007, 26: 3124-3131. 10.1038/sj.emboj.7601747.PubMedPubMed CentralView ArticleGoogle Scholar
- Srivastava P, Chattoraj DK: Selective chromosome amplification in Vibrio cholerae. Mol Microbiol. 2007, 66: 1016-1028. 10.1111/j.1365-2958.2007.05973.x.PubMedView ArticleGoogle Scholar
- Yamaichi Y, Iida T, Park KS, Yamamoto K, Honda T: Physical and genetic map of the genome of Vibrio parahaemolyticus: presence of two chromosomes in Vibrio species. Mol Microbiol. 1999, 31: 1513-1521. 10.1046/j.1365-2958.1999.01296.x.PubMedView ArticleGoogle Scholar
- Aiyar SE, Gaal T, Gourse RL: rRNA promoter activity in the fast-growing bacterium Vibrio natriegens. J Bacteriol. 2002, 184: 1349-1358.PubMedPubMed CentralView ArticleGoogle Scholar
- Joseph SW, Colwell RR, Kaper JB: Vibrio parahaemolyticus and related halophilic Vibrios. Crit Rev Microbiol. 1982, 10: 77-124. 10.3109/10408418209113506.PubMedView ArticleGoogle Scholar
- Duigou S, Knudsen KG, Skovgaard O, Egan ES, Lobner-Olesen A, Waldor MK: Independent control of replication initiation of the two Vibrio cholerae chromosomes by DnaA and RctB. J Bacteriol. 2006, 188: 6419-6424. 10.1128/JB.00565-06.PubMedPubMed CentralView ArticleGoogle Scholar
- Couturier E, Rocha EP: Replication-associated gene dosage effects shape the genomes of fast-growing bacteria but only for transcription and translation genes. Mol Microbiol. 2006, 59: 1506-1518. 10.1111/j.1365-2958.2006.05046.x.PubMedView ArticleGoogle Scholar
- Schmid MB, Roth JR: Gene location affects expression level in Salmonella typhimurium. J Bacteriol. 1987, 169: 2872-2875.PubMedPubMed CentralGoogle Scholar
- Sousa C, de Lorenzo V, Cebolla A: Modulation of gene expression through chromosomal positioning in Escherichia coli. Microbiology. 1997, 143: 2071-2078.PubMedView ArticleGoogle Scholar
- Lobner-Olesen A, Marinus MG, Hansen FG: Role of SeqA and Dam in Escherichia coli gene expression: a global/microarray analysis. Proc Natl Acad Sci USA. 2003, 100: 4672-4677. 10.1073/pnas.0538053100.PubMedPubMed CentralView ArticleGoogle Scholar
- Roth JR, Benson N, Galitski T, Haack K, Lawrence JG, Miesel L: Rearrangements of the bacterial chromosome: formation and applications. Escherichia coli and Salmonella: Cellular and Molecular Biology. Edited by: Neidhardt FC. 1996, Washington DC: American Society for Microbiology, 2: 2256-2276. 2Google Scholar
- Hill CW, Gray JA: Effects of chromosomal inversion on cell fitness in Escherichia coli K-12. Genetics. 1988, 119: 771-778.PubMedPubMed CentralGoogle Scholar
- Liu SL, Sanderson KE: Highly plastic chromosomal organization in Salmonella typhi. Proc Natl Acad Sci USA. 1996, 93: 10303-10308. 10.1073/pnas.93.19.10303.PubMedPubMed CentralView ArticleGoogle Scholar
- Kothapalli S, Nair S, Alokam S, Pang T, Khakhria R, Woodward D, Johnson W, Stocker BA, Sanderson KE, Liu SL: Diversity of genome structure in Salmonella enterica serovar Typhi populations. J Bacteriol. 2005, 187: 2638-2650. 10.1128/JB.187.8.2638-2650.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Rocha EP: The replication related organization of bacterial genomes. Microbiology. 2004, 150: 1609-1627. 10.1099/mic.0.26974-0.PubMedView ArticleGoogle Scholar
- Marians KJ: Replication fork propagation. Escherichia coli and Salmonella: Cellular and Molecular Biology. Edited by: Neidhardt FC. 1996, Washington DC: American Society for Microbiology, 1: 749-763. 2Google Scholar
- Atlung T, Hansen FG: Low-temperature-induced DnaA protein synthesis does not change initiation mass in Escherichia coli K12. J Bacteriol. 1999, 181: 5557-5562.PubMedPubMed CentralGoogle Scholar
- Carpentier AS, Torresani B, Grossmann A, Henaut A: Decoding the nucleoid organisation of Bacillus subtilis and Escherichia coli through gene expression data. BMC Genomics. 2005, 6: 84-10.1186/1471-2164-6-84.PubMedPubMed CentralView ArticleGoogle Scholar
- Jeong KS, Ahn J, Khodursky AB: Spatial patterns of transcriptional activity in the chromosome of Escherichia coli. Genome Biol. 2004, 5: R86-10.1186/gb-2004-5-11-r86.PubMedPubMed CentralView ArticleGoogle Scholar
- Wright MA, Kharchenko P, Church GM, Segre D: Chromosomal periodicity of evolutionary conserved gene pairs. Proc Natl Acad Sci USA. 2007, 104: 10559-10564. 10.1073/pnas.0610776104.PubMedPubMed CentralView ArticleGoogle Scholar
- Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M, Datsenko KA, Tomita M, Wanner BL, Mori H: Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol. 2006, 2: 2006.0008.-10.1038/msb4100050.PubMedPubMed CentralView ArticleGoogle Scholar
- Zhou Y, Landweber LF: BLASTO: a tool for searching orthologous groups. Nucleic Acids Res. 2007, 35: W678-82. 10.1093/nar/gkm278.PubMedPubMed CentralView ArticleGoogle Scholar
- Makino K, Oshima K, Kurokawa K, Yokoyama K, Uda T, Tagomori K, Iijima Y, Najima M, Nakano M, Yamashita A: Genome sequence of Vibrio parahaemolyticus: a pathogenic mechanism distinct from that of V cholerae. Lancet. 2003, 361: 743-749. 10.1016/S0140-6736(03)12659-1.PubMedView ArticleGoogle Scholar
- Chen CY, Wu KM, Chang YC, Chang CH, Tsai HC, Liao TL, Liu YM, Chen HJ, Shen AB, Li JC: Comparative genome analysis of Vibrio vulnificus a marine pathogen. Genome Res. 2003, 13: 2577-2587. 10.1101/gr.1295503.PubMedPubMed CentralView ArticleGoogle Scholar
- Ruby EG, Urbanowski M, Campbell J, Dunn A, Faini M, Gunsalus R, Lostroh P, Lupp C, McCann J, Millikan D: Complete genome sequence of Vibrio fischeri: a symbiotic bacterium with pathogenic congeners. Proc Natl Acad Sci USA. 2005, 102: 3004-3009. 10.1073/pnas.0409900102.PubMedPubMed CentralView ArticleGoogle Scholar
- Vezzi A, Campanaro S, D'Angelo M, Simonato F, Vitulo N, Lauro FM, Cestaro A, Malacrida G, Simionati B, Cannata N: Life at depth: Photobacterium profundum genome sequence and expression analysis. Science. 2005, 307: 1459-1461. 10.1126/science.1103341.PubMedView ArticleGoogle Scholar
- Clusters of Orthologous Groups of proteins. [http://www.ncbi.nlm.nih.gov/COG/]
- Xu Q, Dziejman M, Mekalanos JJ: Determination of the transcriptome of Vibrio cholerae during intraintestinal growth and midexponential phase in vitro. Proc Natl Acad Sci USA. 2003, 100: 1286-1291. 10.1073/pnas.0337479100.PubMedPubMed CentralView ArticleGoogle Scholar
- Kolesov G, Wunderlich Z, Laikova ON, Gelfand MS, Mirny LA: How gene order is influenced by the biophysics of transcription regulation. Proc Natl Acad Sci USA. 2007, 104: 13948-13953. 10.1073/pnas.0700672104.PubMedPubMed CentralView ArticleGoogle Scholar
- Kepes F: Periodic transcriptional organisation of the E. coli genome. J Mol Biol. 2004, 340: 957-964. 10.1016/j.jmb.2004.05.039.PubMedView ArticleGoogle Scholar
- Price MN, Dehal PS, Arkin AP: Horizontal gene transfer and the evolution of transcriptional regulation in Escherichia coli. Genome Biol. 2008, 9: R4-10.1186/gb-2008-9-1-r4.PubMedPubMed CentralView ArticleGoogle Scholar
- Lercher MJ, Pal C: Integration of horizontally transferred genes into regulatory interaction networks takes many million years. Mol Biol Evol. 2008, 25: 559-567. 10.1093/molbev/msm283.PubMedView ArticleGoogle Scholar
- Choudhary M, Zanhua X, Fu YX, Kaplan S: Genome analyses of three strains of Rhodobacter sphaeroides: evidence of rapid evolution of chromosome II. J Bacteriol. 2007, 189: 1914-1921. 10.1128/JB.01498-06.PubMedPubMed CentralView ArticleGoogle Scholar
- Paulsen IT, Seshadri R, Nelson KE, Eisen JA, Heidelberg JF, Read TD, Dodson RJ, Umayam L, Brinkac LM, Beanan MJ: The Brucella suis genome reveals fundamental similarities between animal and plant pathogens and symbionts. Proc Natl Acad Sci USA. 2002, 99: 13148-13153. 10.1073/pnas.192319099.PubMedPubMed CentralView ArticleGoogle Scholar
- Halling SM, Peterson-Burch BD, Bricker BJ, Zuerner RL, Qing Z, Li LL, Kapur V, Alt DP, Olsen SC: Completion of the genome sequence of Brucella abortus and comparison to the highly similar genomes of Brucella melitensis and Brucella suis. J Bacteriol. 2005, 187: 2715-2726. 10.1128/JB.187.8.2715-2726.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Salanoubat M, Genin S, Artiguenave F, Gouzy J, Mangenot S, Arlat M, Billault A, Brottier P, Camus JC, Cattolico L: Genome sequence of the plant pathogen Ralstonia solanacearum. Nature. 2002, 415: 497-502. 10.1038/415497a.PubMedView ArticleGoogle Scholar
- Holden MT, Titball RW, Peacock SJ, Cerdeño-Tárraga AM, Atkins T, Crossman LC, Pitt T, Churcher C, Mungall K, Bentley SD: Genomic plasticity the causative agent of melioidosis Burkholderia pseudomallei. Proc Natl Acad Sci USA. 2004, 101: 14240-14245. 10.1073/pnas.0403302101.PubMedPubMed CentralView ArticleGoogle Scholar
- Romling U, Tummler B: Achieving 100% typeability of Pseudomonas aeruginosa by pulsed-field gel electrophoresis. J Clin Microbiol. 2000, 38: 464-465.PubMedPubMed CentralGoogle Scholar
- Cooper S, Helmstetter CE: Chromosome replication and the division cycle of Escherichia coli B/r. J Mol Biol. 1968, 31: 519-540. 10.1016/0022-2836(68)90425-7.PubMedView ArticleGoogle Scholar
- Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods. 2001, 25: 402-408. 10.1006/meth.2001.1262.PubMedView ArticleGoogle Scholar
- Izutsu K, Kurokawa K, Tashiro K, Kuhara S, Hayashi T, Honda T, Iida T: Comparative genomic analysis using microarray demonstrates strong correlation between presence of Vp-PAI and pathogenicity in Kanagawa phenomenon-positive Vibrio parahaemolyticus. Infect Immun. 2008, 76: 1016-1023. 10.1128/IAI.01535-07.PubMedPubMed CentralView ArticleGoogle Scholar
- Gene Expression Omnibus. [http://www.ncbi.nlm.nih.gov/projects/geo/]