Two Corynebacterium species were proposed decades ago, isolated from clinical samples and divided into biovars: "Corynebacterium genitalium" biovars I-V and "Corynebacterium pseudogenitalium" biovars C1 -C6. Several biovars have been re-classified as new species. Nevertheless, biovar I and C5, together with their respective specific epithets "Corynebacterium genitalium" and "Corynebacterium pseudogenitalium", remained not validly published after more than 40 years. Several more strains, temptatively classified as "C. genitalium" biovar I and "Corynebacterium pseudogenitalium" C5, have been isolated from clinical and environmental samples. Both species presented Gram-positive, non-spore forming rod-shaped cells, able to grow aerobically with CO2. Core-genome analysis identified "C. genitalium" to be most closely related to Corynebacterium tuscaniense, Corynebacterium urinipleomorphum, Corynebacterium aquatimens and C appendicis, and Corynebacterium gottingense as the most closely related species to "C. pseudogenitalium". Comprehensive genomic, genotypic, phenotypic analyses, as well as chemotaxonomic, support the proposal for "C. genitalium" and "C. pseudogenitalium" as distinct species within the genus Corynebacterium. The designated type strains of the two species are Furness 392-1T = ATCC 33030T = CCUG 38989T = CCM 9178T = DSM 113155T for C. genitalium sp. nov., nom. rev., and Furness 162-C2T = ATCC 33039T = CCUG 27540T = CCM 9177T = DSM 113154T for C. pseudogenitalium sp. nov., nom. rev. (c) 2022 The Author(s). Published by Elsevier Masson SAS on behalf of Institut Pasteur. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Although many clinically significant strains belonging to the family Enterobacteriaceae fall into a restricted number of genera and species, there is still a substantial number of isolates that elude this classification and for which proper identification remains challenging. With the current improvements in the field of genomics, it is not only possible to generate high-quality data to accurately identify individual nosocomial isolates at the species level and understand their pathogenic potential but also to analyse retrospectively the genome sequence databases to identify past recurrences of a specific organism, particularly those originally published under an incorrect or outdated taxonomy. We propose a general use of this approach to classify further clinically relevant taxa, i.e., Phytobacter spp., that have so far gone unrecognised due to unsatisfactory identification procedures in clinical diagnostics. Here, we present a genomics and literature-based approach to establish the importance of the genus Phytobacter as a clinically relevant member of the Enterobacteriaceae family.

Clinical and environmental-associated strains (n=17), genotypically related to Corynebacterium spp., yet distinct from any species of the genus Corynebacterium with validly published names, have been isolated during the last 20 years and tentatively identified as Corynebacterium sanguinis, although the combination, “Corynebacterium sanguinis” was never validly published. The comprehensive genotypic and phenotypic characterisations and genomic analyses in this study support the proposal for recognizing the species within the genus Corynebacterium, for which the name, Corynebacterium sanguinis sp. nov., is reaffirmed and proposed. Strains of Corynebacterium sanguinis are Gram-positive, non-motile, non-spore-forming, short, pleomorphic and coryneform bacilli, growing aerobically, with CO2. They contain mycolic acids, major respiratory menaquinones, MK-8 (II-H2) and MK-9 (II-H2), and polar lipids, phosphatidylglycerol, diphosphatidylglycerol, phosphatidylinositol, phosphoglycolipid, glycolipids and a novel lipid that remains to be characterized and identified. Strains of Corynebacterium sanguinis are genotypically most similar to Corynebacterium lipophiliflavum, with 16S rRNA gene sequence similarities of 98.3% and rpoB sequence similarities of 94.9–95.2%. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis were able to clearly differentiate Corynebacterium sanguinis from the most closely related species. The genome size of Corynebacterium sanguinis is 2.28–2.37Mbp with 65.1–65.5mol% G+C content. A total of 2202–2318 ORFs were predicted, comprising 2141–2251 protein-encoding genes. The type strain is CCUG 58655T (=CCM 8873T=NCTC 14287T).

Two strains included in a whole-genome sequencing project for methicillin-resistant Staphylococcus aureus (MRSA) were identified as non-Staphylococcus aureus when the sequences were analysed using the bioinformatics software ALEX (www.1928diagnostics.com, Gothenburg, Sweden). Sequencing of the sodA gene of these strains identified them as Staphylococcus argenteus. The collection of MRSA in western Sweden was checked for additional strains of this species. A total of 18 strains of S. argenteus isolated between 2011 and December 2017 were identified.

High quality 16S ribosomal RNA (rRNA) gene sequences from the type strains of all species with validly published names, as defined by the International Code of Nomenclature of Bacteria, are a prerequisite for their accurate affiliations within the global genealogical classification and for the recognition of potential new taxa. During the last few years, the Living Tree Project (LTP) has taken care to create a high quality, aligned 16S and 23S rRNA gene sequence database of all type strains. However, the manual curation of the sequence dataset and type strain information revealed that a total of 552 "orphan" species (about 5.7% of the currently classified species) had to be excluded from the reference trees. Among them, 322 type strains were not represented by an SSU entry in the public sequence repositories. The remaining 230 type strains had to be discarded due to bad sequence quality. Since 2010, the LTP team has coordinated a network of researchers and culture collections in order to improve the situation by (re)-sequencing the type strains of these "orphan" species. As a result, we can now report 351 16S rRNA gene sequences of type strains. Nevertheless, 201 species could not be sequenced because cultivable type strains were not available (121), the cultures had either been lost or were never deposited in the first place (66), or it was not possible due to other constraints (14). The International Code of Nomenclature of Bacteria provides a number of mechanisms to deal with the problem of missing type strains and we recommend that due consideration be given to the appropriate mechanisms in order to help solve some of these issues.

Three strains of a Gram-positive, catalase-positive, fermentative, non-lipophilic, previously unknown bacterium were isolated from urogenital samples taken from mares in Scotland (M401624/00/1) and Sweden (VM 2074 and VM 2298(T)). All were deposited with the CCUG with tentative identifications as Corynebacterium spp. The strains were characterized using a polyphasic taxonomic approach. Biochemically, the strains were very similar to each other, but phylogenetically distinct from Corynebacterium species with validly published names (≤95% sequence similarity). rpoB gene sequence data confirmed the strains belonged to the same species (>99% sequence similarity) and were distinct from species with validly published names (>13% sequence divergence). On the basis of phenotypic and sequence data, the strains represent a novel species within the genus Corynebacterium, for which the name Corynebacterium uterequi is proposed. The type strain is VM 2298(T) (=CCUG 61235(T)=DSM 45634(T)), isolated from equine uterus.