Genome Sequencing and Pan-Genome Analysis of 23 Corallococcus spp. Strains Reveal Unexpected Diversity, With Particular Plasticity of Predatory Gene Sets

Abstract

Corallococcus is an abundant genus of predatory soil myxobacteria, containing two species, C. coralloides (for which a genome sequence is available) and C. exiguus. To investigate the genomic basis of predation, we genome-sequenced 23 Corallococcus strains. Genomic similarity metrics grouped the sequenced strains into at least nine distinct genomospecies, divided between two major sub-divisions of the genus, encompassing previously described diversity. The Corallococcus pan-genome was found to be open, with strains exhibiting highly individual gene sets. On average, only 30.5% of each strain's gene set belonged to the core pan-genome, while more than 75% of the accessory pan-genome genes were present in less than four of the 24 genomes. The Corallococcus accessory pan-proteome was enriched for the COG functional category "Secondary metabolism," with each genome containing on average 55 biosynthetic gene clusters (BGCs), of which only 20 belonged to the core pan-genome. Predatory activity was assayed against ten prey microbes and found to be mostly incongruent with phylogeny or BGC complement. Thus, predation seems multifactorial, depending partially on BGC complement, but also on the accessory pan-genome - genes most likely acquired horizontally. These observations encourage further exploration of Corallococcus as a source for novel bioactive secondary metabolites and predatory proteins.

Keywords: comparative genomics; evolution; myxobacteria; predator; prey; taxonomy.

FIGURE 1

Phylogenetic trees of Corallococcus isolates and other genome-sequenced myxobacteria. (A) AMPHORA2 tree based on 31 conserved genes. (B) Tree based on relative distance calculated from the genomic ANI (average nucleotide identity).

FIGURE 2

The Corallococcus core genome (A) and pan-genome (B) as a function of the number of genomes included (V1–V24). Boxes represent +/- one standard deviation around the median number of genes while whiskers indicate +/- two standard deviations.

FIGURE 3

Congruence of taxonomy, BGC, and predation trees. Tanglegrams show pairs of trees reordered for maximal alignment. Trees were variously constructed from the matrix of predatory assay data (Predation), 16S rRNA gene sequences (16S Sequence), genome-wide average nucleotide identity (Genomic ANI), and the average nucleotide identity of biosynthetic gene clusters (BGCs ANI).