Eikenella exigua sp. nov., isolated from brain abscess and blood

Abstract

We herein describe the first novel species within the genus Eikenella since it was established in 1972 by the reclassification of ‘Bacteroides corrodens’ to Eikenella corrodens . From a polymicrobial brain abscess, we encountered an Eikenella isolate, PXXT, that could not validly be named E. corrodens . The isolate grew on blood agar with small, translucent, pitting colonies after 3 days of anaerobic incubation. By reviewing previously collected invasive isolates, we found an additional Eikenella strain, EI-02, from a blood culture exhibiting the same properties as PXXT. Phylogenetic analyses based on both whole genome and individual house-keeping genes confirmed that the two strains allocate in a phylogenetic cluster separate from E. corrodens . Using specific amplification and sequencing of the Eikenella nusG gene, we further detected the novel Eikenella species in six historic brain abscesses previously reported to contain E. corrodens based on 16S metagenomics. Out of 24 Eikenella whole-genome projects available in GenBank, eight cluster together with PXXT and EI-02. These isolates were recovered from brain abscess (n=2), blood (n=1), bone/soft tissue (n=3), parotid gland (n=1) and unknown (n=1). It remains to be investigated whether the new species can cause endocarditis. The average nucleotide identity value between strain PXXT and the E. corrodens type strain ATCC 23834T was 92.1 % and the corresponding genome-to-genome distance value was 47.1 %, both supporting the classification of PXXT as a novel species. For this species we propose the name Eikenella exigua. The type strain of E. exigua is PXXT (DSM 109756T, NCTC 14318T).

Description of Eikenella halliae sp. nov. and Eikenella longinqua sp. nov., derived from human clinical materials, emendation of Eikenella exigua Stormo et al. 2019 and emendation of the genus Eikenella to include species which are strict anaerobes

“…E. exigua-like (except for NML 99-0057, below) and E. longinqua strains grew best under strictly anaerobic conditions on pre-reduced Brucella agar, similar to E. exigua strain EI-02 described by Stormo et al [9]. A number of these were characterized using test substrates under anaerobic conditions as outlined previously for Gram-stain-negative anaerobes, including hydrolysis of aesculin, nitrate reduction, urease production and gelatin hydrolysis.…”

“…These strains had all been characterized by 16S rRNA gene sequencing, whole-genome sequencing and matrix-assisted laser desorption/ionization time-of-flight mass spectrophometry (MALDI-TOF) analysis, as well as biochemically, chemotaxonomically and for antimicrobial susceptibilities. On the basis of the results of these analyses, we propose emendation of the genus Eikenella to accommodate species which are strictly anaerobic as well as emendation to Eikenella exigua [9] to include some features observed from among the Canadian isolates. Lastly, we propose two novel species, Eikenella halliae sp.…”

“…Fourteen strains had 97.6 to 98.8% sequence similarities by 16S rRNA gene sequencing to E. corrodens and formed what appeared to be three distinct taxonomic groups. Concurrently, Stormo and colleagues in Norway characterized two anaerobic, Eikenellalike strains recovered from a brain abscess and blood culture which were discernible from E. corrodens, for which the name E. exigua was proposed; some information derived from the genomes described in this investigation were also included in that study [9]. The largest taxon group from among the Canadian referrals consisted of ten anaerobic isolates from eight patients recovered from blood, brain, bone and other abscesses; upon re-evaluation here, this group was found to be most consistent with E. exigua and so were referred to here as E. exigua-like isolates.…”

“…Genomic analyses had clearly separated the strains into four taxonomic groups using the draft genome of Eikenella corrodens ATCC 23834 T (GenBank accession number ACEA01) as a reference; some of these observations had been previously inferred by Stormo et al [9]. Draft whole genomes were compared using digital DNA-DNA hybridization analysis via the Genome-to-Genome Distance Calculator 2 (GGDC) using GGDC 2.0 blast+ [20].…”

“…Some were tested using the rapID ANA panel (Remel) or the API 32A panel (Biomérieux). [9]). Strains for all species were catalasenegative and did not ferment, oxidize or assimilate sugars.…”