Understanding the adaptation of Halobacterium species NRC-1 to its extreme environment through computational analysis of its genome sequence

Abstract

The genome of the halophilic archaeon Halobacterium sp. NRC-1 and predicted proteome have been analyzed by computational methods and reveal characteristics relevant to life in an extreme environment distinguished by hypersalinity and high solar radiation: (1) The proteome is highly acidic, with a median pI of 4.9 and mostly lacking basic proteins. This characteristic correlates with high surface negative charge, determined through homology modeling, as the major adaptive mechanism of halophilic proteins to function in nearly saturating salinity. (2) Codon usage displays the expected GC bias in the wobble position and is consistent with a highly acidic proteome. (3) Distinct genomic domains of NRC-1 with bacterial character are apparent by whole proteome BLAST analysis, including two gene clusters coding for a bacterial-type aerobic respiratory chain. This result indicates that the capacity of halophiles for aerobic respiration may have been acquired through lateral gene transfer. (4) Two regions of the large chromosome were found with relatively lower GC composition and overrepresentation of IS elements, similar to the minichromosomes. These IS-element-rich regions of the genome may serve to exchange DNA between the three replicons and promote genome evolution. (5) GC-skew analysis showed evidence for the existence of two replication origins in the large chromosome. This finding and the occurrence of multiple chromosomes indicate a dynamic genome organization with eukaryotic character.

Figure 1

Predicted percentage of total protein versus isoelectric point distribution for six complete genomes including Methanobacterium thermoautotrophicum (filled triangles), Methanococcus jannaschii (X symbols ), Escherichia coli (filled diamonds), Bacillus subtilis (filled squares), Saccharomyces cerevisiae (filled circles), and Halobacterium NRC-1 (unmarked line).

Figure 2

Surface charge comparisons for halophilic and nonhalophilic proteins. Acidic character is indicated by red, basic character is indicated by blue, and neutral areas are indicated by white. (A) NRC-1 TFBe (left) and Human TFIIB (right) are shown with sites for BRE and TBP contacts indicated. (B) NRC-1 (left) and Escherichia coli GyrA (right) are shown with the binding site for the helix that is cleaved as well as the site for strand passage indicated for both molecules.

Figure 3

Codon bias plot showing differences in expected triplet frequencies. Each of the 64 triples is plotted along the X axis with one-letter amino acid designations identifying which amino acid a particular group of triplet represents. Results are normalized according to GC composition. The deviation from the expected percentage is plotted on the Y axis.

Figure 4

Graphical representation of the relative amount of homology between predicted proteins in the large chromosome of NRC-1 to those encoded by Deinococcus radiodurans and Bacillus subtilis. NRC-1 chromosomal ORFs are plotted on the X axis. Triangles represent NRC-1 ORFs with BLAST P values <e⁻²⁹, with more significant scores higher on the Y axis. Average bacterial character is shown by the percentage of predicted ORFs in a 40-ORF window that showed BLAST P values of <1 × 10⁻¹⁰. The labeled peaks correspond to regions harboring genes encoding NADH dehydrogenase (I) and menaquinone biosynthesis (II), components of the electron transport chain.

Figure 5

(A) Genetic maps of aerobic respiration genes. The first row contains the nuo and cox genes in the region from 485,000 to 506,000 bp. The second row contains the men and cyd genes in a chromosomal region from 821,000 to 829,000 bp and pNRC100, respectively. The genetic nomenclature for some nuo and men genes was changed from our earlier work (Ng et al. 2000); changes are indicated parenthetically. (B) Phylogenetic trees of nuoB and nuoI gene products. Branches are labeled for each gene with the corresponding three-letter organism designations except for Ssp 6803 (Synechocystis sp.) and NRC-1 (Halobacterium NRC-1).

Figure 6

Circular representation of the Halobacterium NRC-1 chromosome showing the GC composition of ORFs, χ-squared analysis, and location of IS elements. The outer scale refers to ORF identification numbers. Bars associated with the outmost circle denote the position of the chromosomal IS elements. χ-squared analysis is plotted as a thin black line, and average GC content of ORF is the innermost plot in darker black, each against their relative position on the circular map. The Roman numerals I and II inside the plot indicate relatively AT-rich islands.

Figure 7

Plot of the GC skew of the Halobacterium NRC-1 large chromosome. The GC ratio (Y axis) is plotted over the length of the chromosome (X axis). The numbered arrows show inflection points, where the plot crosses the zero mark. The positions of the three chromosomal orc1/cdc6 genes are indicated.