Niche specific amino acid features within the core genes of the genus Shewanella.

Shewanella species are found to dwell in various ecological niches. The widespread habitation where they live requires specific adaptations. Recent advances in genomic approaches, such as in sequencing technologies, generate huge amount of genomic data that lend support towards understanding the microbial evolution and diversity through comparative study. In this manuscript, we discuss a comparative analysis of core genes of phylogenetically related twelve members from the genus Shewanella. Phylogenetic analysis based on the core genes, differentiated two subgroups of the genus, one group comprises of species characterized as highpressure cold-adapted while the other group is characterized as mesophilic pressure-sensitive species. By analyzing the differences of amino acid composition of these two groups, we have identified the specific trend of amino acid usage that has been adopted by the psychro-peizo-tolerant Shewanella species. The functional categories have also been recognized which are responsible for rendering the particular amino acid compositional pattern in psychropeizophilic Shewanella species facilitating their niche adaptation.

Background

The increasing number of complete genomic sequences has unwrapped loads of ways to interpret the adaptation of genomes in their respective niche in relation to their structure, function etc. [1]. An estimation of amino acid preferences of different organisms is now possible through comparative genomic study [2]. Comparative genomics helps to determine the genes that are present across bacterial genomes of the same species (or genus) known as conserved (core) genes and disparity within these core genes is an indication that can be used to unravel highly variable and conserved genes within the genomes of interest. The conserved genes usually evolve more slowly; therefore, they will play a role to infer molecular evolution of the genomes [3]. Moreover, molecular phylogenetic trees built using the sequences of core genes can also determine the relationship among the genomes.

Shewanella is a genus from which adequate number of species had already been sequenced facilitating detection of core genes. Different species belonging to the genus Shewanella are capable of inhabiting many aquatic (from fresh water to deep sea) and sedimentary ecosystems under aerobic as well as anaerobic conditions [4]. Some of the Shewanella species, for example Shewanella violacea, had been isolated from cold environments, such as seawater in Antarctica or in the North Sea, implying that they are not only peizophilic (can breed better under high hydrostatic pressure conditions), but also psychrophilic (needs low temperatures, ranging from −15°C to +10°C for growth and breeding). They can be defined as psychro-peizo-tolerant Shewanella species.

It is known that pressure influences protein structure and folding [5]. Functional properties of proteins might get affected by pressure range that is experienced by organism [6]. In this study, we discuss the comparative analysis of the core genes of phylogenetically related twelve members from the genus Shewanella and focus on the recognition of a common trend towards a certain pattern of amino acid usage of the psychropeizo-tolerant Shewanella species.

Methodology

We have downloaded the fully annotated complete nucleotide sequences of all the twelve Shewanella genomes, considered in the present study, from NCBI FTP site (www.ncbi.nlm.nih.gov/Ftp/). Details of the 12 bacteria are listed in Table 2(see supplementary material).

Identification of core genes:

Highly similar paralogous sequences have been removed from each of the twelve set of genome sequences, based on a comparison of the gene list against itself (with identity >=90%) using Blastclust program (www.ncbi.nlm.nih.gov/Web/Newsltr/Spring04/blastlab.html ). From the rest of the genes, core genes were detected through putative one-to-one orthologous gene identification, using BLASTP (>= 85% identity and 0% gap). This search identified a total of 121 orthologous groups that had all twelve species represented, which were used for further analysis.

Phylogenetic analysis:

All core genes from 12 Shewanella genomes were used for generating a neighbor joining tree. For each twelve species, the amino acid sequences of 121 core gene sets were concatenated in order to produce twelve continuous sequences of average length 34,746 amino acids. Multiple Alignments of these 12 sets of concatenated core amino acid sequences were performed using CLUSTALW (http://www.ebi.ac.uk/clustalw/). A Neighbor-joining Phylogenetic tree based on these concatenated core amino acid sequences from the Shewanella species of interest, was generated using MEGA 5 [7].The bootstrap values are shown in the Phylogenetic tree.

Estimation of amino acid composition:

We have computed the frequencies of amino acid residues in the protein sequences of two groups of Shewanella sp., as depicted in the Phylogenetic tree. To compare the means of two groups of data, t-test is essentially a good tool. The t-values are calculated as follows:

t = (FGroup1 - FGroup2) / √ [(VarGroup1/ nGroup1) + (VarGroup2/ nGroup2)]

where, VarGroup1 and VarGroup2 are the variance of amino acid residues. FGroup1 and FGroup2 are mean frequencies of Group1 and Group2 proteomes respectively. The nGroup1 and nGroup2 are the total number of Group1 and Group2 proteins investigated in this study. Based on student's t-distribution table of significance, critical values for such t-test at various probabilities are as follows:

Probability 10% 5% 2.5% 2% 1% 0.5%

Critical Value 1.372 1.812 2.228 2.359 2.764 3.169

If t-value is positive and greater than critical value at 10% probability (1.372), then the mean frequency (FGroup1) of Group1 core proteins are significantly greater than that of the Group2 core proteins (FGroup2) at 90% or higher confidence level. If the frequency of residue or property group t-value is negative and less than -1.372, then the mean frequency of Group1 core proteins are (FGroup1) is significantly less than that of Group2 core proteins (FGroup2) at 90% or higher confidence level [8-10].

Functional classification:

The core proteins were functionally classified according to Clusters of Orthologous Groups (COGs) of proteins categories. BLAST homology study had been carried out against the COG database. Proteins that were classified in two COG categories were registered in both categories.

Results and Discussion

A final list containing a total of 121 core gene sets from all the twelve Shewanella species, were used for further analyses. Alignment of these 12 sets of concatenated core amino acid sequences using CLUSTALW (http://www.ebi.ac.uk/clustalw/) reveals 17% non-conserved amino acid sites. Neighbor-joining phylogenetic tree based on the concatenated core amino acid sequences from the Shewanella species of interest, was constructed using MEGA 5, which separates 12 Shewanella species into two groups i.e., Group 1 and Group 2 (Figure 1). Literature search describes Group 1 members of Shewanella as cold adapted species that grow at high pressure, while Group2 members are mesophilic, pressure sensitive species [11].

Figure 1

Phylogenetic tree constructed using core genes of 12 Shewanella species.Group1 consists of psychro-peizo-tolerant Shewanella species and Group2 contains mesophilic pressuresensitive Shewanella species. Bootstrap values for all the branches are mentioned in the figure.

Amino acid composition preferences:

We have analyzed the amino acid composition of 121 core gene sequence sets for each of the twelve Shewanella species. The frequencies of individual amino acids were further analyzed using student t-test, which shows that a few of the amino acid differed significantly in the core gene sets of Shewanella species present in Group1 when compared to the core gene sets of Shewanella species present in Group 2 Table 1 (see supplementary material). As specified by the t-value, amino acid residues A, D, S, N and C, are significantly preferred (marked in bold red in Table 1), while G, T, V, M, I are moderately preferred and residues R, Y, E, P and L are significantly avoided (marked in blue in Table 1) by Group 1 Shewanella species compared to Group 2. Residues S and D are helix breakers and residue E favors the formation of helical structure. On the other hand, presence of residue L stabilizes the helical conformations [12]. It is known that amino acid substitutions that diminish protein flexibility and compressibility results in an increase of stability of the protein at high pressure [13]. In addition, helix destabilizing beta branch residues I, T and V are preferred by peizophilic proteins [14]. All the above results signify that increased structural homogeneity is perhaps favored by the high-pressure environments of the deep-sea, which is attained by favoring a decreased number of helix breaking and helix destabilizing residues. Considering the amino acid properties, Group1 Shewanella members most significantly, favor the amino acid residues S, D, and N, all of which are strongly polar as well as have very low molecular weight. Interestingly, the hydrostatic pressure asymmetry index is positively correlated with the polarity of amino acids and inversely correlated with molecular weight of amino acids [15].

Thus, the amino acid composition of the core genes of the twelve species of Shewanella considered in our study shows a particular trend, which points towards a strong favor of polar and small amino acids with adequate propensity of breaking and destabilizing the helical structure, for the Group 1 members of Shewanella (Figure 1) sustaining their psychropeizophilic adaptation for residing in deep sea environment, while an opposite amino acid compositional trend is featured by the Group 2 members of Shewanella (Figure 1) supporting their mesophilic, pressure sensitive characteristics.

Functional Classification:

Core genes considered in our study can be divided into two categories depending on the average percent identity of the blast score. (a) Core gene showing low variation (54% of the total core genes with identity >95%); (b) Core gene that are highly variable (46% of the total core genes with identity <95%). Functional profiles of the core genes of the twelve Shewanella species considered in our study have been determined based on COG categories. It has been found that higher proportion of the core genes (44%) are involved in information storage and processing, 62% of which are involved in translation, ribosomal structure and biogenesis (Figure 2).

Figure 2

Distribution of COG classification of 121 core genes in twelve Shewanella species considered in this study.

Variable core genes as well as conserved core genes are classified according to COG functional categories and compared. The comparison points to the fact that functional categories like “Translation, ribosomal structure and biogenesis”, “Carbohydrate transport and metabolism” and “Amino acid transport and metabolism” are more common in conserved core genes (Figure 3). On the other hand, functional categories like “Energy production and conversion”, “Cell envelope biogenesis” and “Cell motility and secretion” are more common in variable core genes.

Figure 3

COG classification terms for conserved core genes (red bars) and highly variable core genes (blue bars). Functional classes like Translation, ribosomal structure and biogenesis (J), Carbohydrate transport and metabolism (G), Amino acid transport and metabolism (E), are more common in conserved core genes. Variable core genes have a preference over functional categories like Energy production and conversion (C), Cell envelope biogenesis (M) and Cell motility and secretion (N).

Core genes with low variation experience selection against mutations that leads to amino acid changes. But in case of highly variable core genes, positive selection for amino acid changes takes place [16]. Consequently, three functional categories which are more common in variable core genes are accountable for the specific amino acid compositional trend in psychropeizophilic Shewanella species (members of Group 1), favoring polar and small amino acids with helix breaking and destabilizing property.

Conclusion

Phylogenetic study based on the concatenated core amino acid sequences of twelve Shewanella species separated two distinct groups of Shewanella. Group 1 comprises of psychropeizophilic Shewanella species, whereas Group2 members are mesophilic, pressure sensitive species. Our studies on the composition of individual amino acid residues within these two groups of bacteria revealed that the psychropeizophilic Shewanella species show a specific trend of amino acid usage that favors the increase in frequency of strongly polar, small and tiny amino acids having the potential of avoiding helical structures. Amino acid residues S, D and N are mostly preferred by them. Functional profiles of the core genes of the twelve Shewanella species show that information storage and processing represents higher proportion of the core genes, with 62% of them involved in translation, ribosomal structure and biogenesis, signifying the importance of these two functional categories in maintaining the most important cellular processes of the genus Shewanella. We have also divided the core genes into two types: conserved core genes and variable core genes, which are evaluated on the basis of COG functional classification. Three functional categories (Cell envelope biogenesis, Cell motility and secretion, Energy production and conversion) which are quite more common in variable core genes are responsible for displaying the specific amino acid compositional trend in psychropeizophilic Shewanella species (members of Group 1). Thus, they seem to have played an important role for niche adaptation of the twelve Shewanella species considered in our study.