Engulfment requires the coordinated, targeted synthesis and degradation of peptidoglycan at the leading edge of the engulfing membrane to allow the mother cell to completely engulf the forespore. Proteins such as the DMP and Q:AH complexes in Bacillus subtilis are essential for engulfment, as are a set of accessory proteins including GerM and SpoIIB, among others. Experimental and bioinformatic studies of these proteins in bacteria distinct from Bacillus subtilis indicate that fundamental differences exist regarding the organization and mechanisms used to successfully perform engulfment. As a consequence, the distribution and prevalence of the proteins involved in engulfment and other proteins that participate in different sporulation stages have been studied using bioinformatic approaches. These works are based on the prediction of orthologs in the genomes of representative Firmicutes and have been helpful in tracing hypotheses about the origin and evolution of sporulation genes, some of which have been postulated as sporulation signatures. To date, an extensive study of these signatures outside of the representative Firmicutes is not available. Here, we asked whether phyletic profiles of proteins involved in engulfment can be used as signatures able to describe the sporulation phenotype. We tested this hypothesis in a set of 954 Firmicutes, finding preserved phyletic profiles defining signatures at the genus level. Finally, a phylogenetic reconstruction based on non-redundant phyletic profiles at the family level shows the non-monophyletic origin of these proteins due to gain/loss events along the phylum Firmicutes.

One of the main characteristics of prokaryotic genomes is the ratio in which guanine-cytosine bases are used in their DNA sequences. This is known as the genomic GC content and varies widely, from values below 20% to values greater than 74%. It has been demonstrated that the genomic GC content varies in accordance with the phylogenetic distribution of organisms and influences the amino acid composition of their corresponding proteomes. This bias is particularly important for amino acids that are coded by GC content-rich codons such as alanine, glycine, and proline, as well as amino acids that are coded by AT-rich codons, such as lysine, asparagine, and isoleucine. In our study, we extend these results by considering the effect of the genomic GC content on the secondary structure of proteins. On a set of 192 representative prokaryotic genomes and proteome sequences, we identified through a bioinformatic study that the composition of the secondary structures of the proteomes varies in relation to the genomic GC content; random coils increase as the genomic GC content increases, while alpha-helices and beta-sheets present an inverse relationship. In addition, we found that the tendency of an amino acid to form part of a secondary structure of proteins is not ubiquitous, as previously expected, but varies according to the genomic GC content. Finally, we discovered that for some specific groups of orthologous proteins, the GC content of genes biases the composition of secondary structures of the proteins for which they code.