Association of purine asymmetry, strand-biased gene distribution and PolC within Firmicutes and beyond: a new appraisal

Abstract

Background: The Firmicutes often possess three conspicuous genome features: marked Purine Asymmetry (PAS) across two strands of replication, Strand-biased Gene Distribution (SGD) and presence of two isoforms of DNA polymerase III alpha subunit, PolC and DnaE. Despite considerable research efforts, it is not clear whether the co-existence of PAS, PolC and/or SGD is an essential and exclusive characteristic of the Firmicutes. The nature of correlations, if any, between these three features within and beyond the lineages of Firmicutes has also remained elusive. The present study has been designed to address these issues.

Results: A large-scale analysis of diverse bacterial genomes indicates that PAS, PolC and SGD are neither essential nor exclusive features of the Firmicutes. PolC prevails in four bacterial phyla: Firmicutes, Fusobacteria, Tenericutes and Thermotogae, while PAS occurs only in subsets of Firmicutes, Fusobacteria and Tenericutes. There are five major compositional trends in Firmicutes: (I) an explicit PAS or G + A-dominance along the entire leading strand (II) only G-dominance in the leading strand, (III) alternate stretches of purine-rich and pyrimidine-rich sequences, (IV) G + T dominance along the leading strand, and (V) no identifiable patterns in base usage. Presence of strong SGD has been observed not only in genomes having PAS, but also in genomes with G-dominance along their leading strands - an observation that defies the notion of co-occurrence of PAS and SGD in Firmicutes. The PolC-containing non-Firmicutes organisms often have alternate stretches of R-dominant and Y-dominant sequences along their genomes and most of them show relatively weak, but significant SGD. Firmicutes having G + A-dominance or G-dominance along LeS usually show distinct base usage patterns in three codon sites of genes. Probable molecular mechanisms that might have incurred such usage patterns have been proposed.

Conclusion: Co-occurrence of PAS, strong SGD and PolC should not be regarded as a genome signature of the Firmicutes. Presence of PAS in a species may warrant PolC and strong SGD, but PolC and/or SGD not necessarily implies PAS.

Figure 1

Instantaneous GC-skew (blue lines) and AT-skew (red lines) trajectories in model representatives of Trend I. (A) Bacillus anthracis str.Ames, (B) Listeria monocytogenes 07PF0776, (C) Staphylococcus aureus 04–02981, (D) Enterococcus faecalis V583, (E) Clostridium difficile CD196, (F) Thermoanaerobacter tengcongensis MB4, (G) Streptobacillus moniliformis DSM 12112, (H) Illyobacter polytropus DSM 2926.

Figure 2

Instantaneous GC-skew (blue lines) and AT-skew (red lines) trajectories in model representatives of Trend II. (A) Streptococcus agalactiae NEM 316, (B) Acidaminococcus intestini RyC-MR95, (C) Geobacillus kaustophilus HTA426, (D) Veillonella parvula DSM 2008, (E) Thermodesulfobium narugense DSM 14796, (F) Clostridiales genomosp BVAB3 UPII9 5, (G) Acinetobacter sp. ADP1, (H) Candidatus Protochlamydia amoebophila UWE25.

Figure 3

Instantaneous GC-skew (blue lines) and AT-skew (red lines) trajectories in model representatives of Trend III. (A) Ruminocococcus albus 7, (B) Fusobacterium nucleatum subsp. nucleatum ATCC 25586, (C) Leptotrichia buccalis C-1013-b, (D) Mycoplasma mycoides SC PG1, (E) Thermotoga maritima MSB8, (F) Aquifex aeolicus VF5.

Figure 4

Instantaneous GC-skew (blue lines) and AT-skew (red lines) trajectories in model representatives of Trend IV & V. Trend IV - (A) Oenococcus oeni PSU1, (B) Sulfobacillus acidophilus DSM 10332, (C) Mycobacterium tuberculosis CDC 1511, (D) Bartonella henselae str. Houston-1. Trend V - (E) Mycoplasma synoviae 53, (F) Acidobacterium capsulatum ATCC 51196.

Figure 5

Scatter plots of Local GC-skew and AT-skew values in model representatives of organisms following different trends in purine usages. (I) Trend I - Bacillus anthracis str.Ames (L) and Streptobacillus moniliformis DSM 12112 (R); (II) Trend II - Geobacillus kaustophilus HTA 426 (L) and Acinetobacter sp. ADP1 (R); (III) Trend III - Ruminocococcus albus 7 (L) and Fusobacterium nucleatum subsp.nucleatum ATCC 25586 (R); (IV) Trend IV - Oenococcus oeni PSU1 (L) and Bacteroides fragilis 638R (R); (V) Trend V - Mycoplasma synoviae 53 (L) and Acidobacterium capsulatum ATCC 51196 (R).

Figure 6

SGD and Genomic GC-content distribution profiles in organisms showing different trends. (A) Distribution of SGD in Trend I (red solid line), PolC-Trend II (blue solid line), non-PolC-Trend II (blue dotted line), PolC-Trend III (green solid line) and non-PolC Trend IV (violet dotted line) organisms; (B) Distribution of Genomic GC-content for Trend I and Trend II organisms.

Figure 7

Trends in individual base usages in Staphylococcus aureus 04–02981 for genes encoded by both LeS and LaS. Subscripts 1, 2, 3 indicate the percentage of occurrences of the respective base at 1st ( A , B ), 2nd ( C , D ) and 3rd ( E , F ) codon sites, intergenic indicate the percentage of intergenic regions ( G , H ) and the subscript T stands for the total percentage ( I , J ) of occurrence of the base in individual genes of the organism.

Figure 8

Trends in individual base usages in Streptococcus agalactiae NEM316 for genes encoded by both LeS and LaS. Subscripts are same as in Figure  7.

Figure 9

Trends in individual base usages in Geobacillus kaustophilus HTA426 for genes encoded by both LeS and LaS. Subscripts are same as in Figure  7.

Figure 10

Trends in individual base usages in Oenococcus oeni PSU 1 for genes encoded by both LeS and LaS. Subscripts are same as in Figure  7.