Systematic review of descriptions of novel bacterial species: evaluation of the twenty-first century taxonomy through text mining.

Although described bacterial species increased in the twenty-first century, they correspond to a tiny fraction of the actual number of species living on our planet. The volume of textual data of these descriptions constitutes valuable information for revealing trends that in turn could support strategies for improvement of bacterial taxonomy. In this study, a text mining approach was used to generate bibliometric data to verify the state-of-art of bacterial taxonomy. Around 9700 abstracts of bacterial classification containing the expression 'sp. nov.' and published between 2001 and 2018 were downloaded from PubMed and analysed. Most articles were from PR China and the Republic of Korea, and published in the International Journal of Systematic and Evolutionary Microbiology. From about 10 800 species names detected, 93.33 % were considered valid according to the rules of the Bacterial Code, and they corresponded to 82.98 % of the total number of species validated between 2001 and 2018. Streptomyces, Bacillus and Paenibacillus each had more than 200 species described in the period. However, almost 40 % of all species were from the phylum Proteobacteria. Most bacteria were Gram-stain-negative, bacilli and isolated from soil. Thirteen species and one genus homonyms were found. With respect to methodologies of bacterial characterization, the use of terms related to 16S rRNA and polar lipids increased along these years, and terms related to genome metrics only began to appear from 2009 onward, although at a relatively lower frequency. Bacterial taxonomy is known as a conservative discipline, but it gradually changed in terms of players and practices. With the advent of the mandatory use of genomic analyses for species description, we are probably witnessing a turning point in the evolution of bacterial taxonomy.

Genome-based reclassification of Azospirillum brasilense Sp245 as the type strain of Azospirillum baldaniorum sp. nov.

Azospirillum sp. strain Sp245T, originally identified as belonging to Azospirillum brasilense, is recognized as a plant-growth-promoting rhizobacterium due to its ability to fix atmospheric nitrogen and to produce plant-beneficial compounds. Azospirillum sp. Sp245T and other related strains were isolated from the root surfaces of different plants in Brazil. Cells are Gram-negative, curved or slightly curved rods, and motile with polar and lateral flagella. Their growth temperature varies between 20 to 38 °C and their carbon source utilization is similar to other Azospirillum species. A preliminary 16S rRNA sequence analysis showed that the new species is closely related to A. brasilense Sp7T and A. formosense CC-Nfb-7T. Housekeeping genes revealed that Azospirillum sp. Sp245T, BR 12001 and Vi22 form a separate cluster from strain A. formosense CC-Nfb-7T, and a group of strains closely related to A. brasilense Sp7T. Overall genome relatedness index (OGRI) analyses estimated based on average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) between Azospirillum sp. Sp245T and its close relatives to other Azospirillum species type strains, such as A. brasilense Sp7T and A. formosense CC-Nfb-7T , revealed values lower than the limit of species circumscription. Moreover, core-proteome phylogeny including 1079 common shared proteins showed the independent clusterization of A. brasilense Sp7T, A. formosense CC-Nfb-7T and Azospirillum sp. Sp245T, a finding that was corroborated by the genome clustering of OGRI values and housekeeping phylogenies. The DNA G+C content of the cluster of Sp245T was 68.4-68.6 %. Based on the phylogenetic, genomic, phenotypical and physiological analysis, we propose that strain Sp245T together with the strains Vi22 and BR12001 represent a novel species of the genus Azospirillum, for which the name Azospirillum baldaniorum sp. nov. is proposed. The type strain is Sp245T (=BR 11005T=IBPPM 219T) (GCF_007827915.1, GCF_000237365.1, and GCF_003119195.2).

Glutamine synthetase evolutionary history revisited: Tracing back beyond the Last Universal Common Ancestor.

Glutamine synthetase (GS; EC 6.3.1.2, L-glutamate: ammonia ligase) is an essential enzyme in nitrogen assimilation. It catalyzes glutamine synthesis using glutamate and ammonium with ATP hydrolysis. Four forms of GSs have been described in literature. These enzyme types are discriminated based on their primary and quaternary structures. GS-encoding genes are believed to be of the oldest functioning genes studied, and its evolutionary history was explored in classic studies in the 90s. Here, we evaluated GS-homologous sequences from the three life domains to revisit their origins and evolutionary history. There are clear examples of ancient duplications and interdomain horizontal gene transfers. We present GS-encoding genes as one multigenic family that comprises three distinct groups. Our findings are presented in light of two main hypotheses for GS origins and evolutions, and we argue in favor of gene duplications giving rise to the three genes in the Last Universal Common Ancestral. Type I family is the most diverse one, presenting a subgroup of polyamine metabolizing enzymes, besides many examples of noncatalytic GS homologs. Many instances of gene loss, duplication, and transfer have occurred after life diversification, contributing to GS complex evolutionary history.