Introduction

Pseudomonas diminuta and Pseudomonas vesicularis were reclassified into the new genus Brevundimonas based on their genotypic and phenotypic characteristics (Segers et al. 1994). The genus Brevundimonas (B. diminuta, type species) belongs to the family Caulobacteraceae, class Alphaproteobacteria, and phylum Pseudomonadota. Currently, the genus includes 37 validly published species in the List of Prokaryotic names with Standing in Nomenclature (https://lpsn.dsmz.de/genus/brevundimonas) (Parte et al. 2020). The majority of organisms belonging to the genus Brevundimonas have been isolated from diverse environments, such as the soil (Pham et al. 2016), black sand (Choi et al. 2010), aquatic habitats (Abraham et al. 2010), seawater (Fritz et al. 2005), and lake sediment (Qu et al. 2019). These species have also been found under specific conditions, e.g., in alkaline soil (Yoon et al. 2006), an oligotrophic pond (Friedrich et al. 2021), oil-contaminated soil (Chaudhary and Kim 2018), or the sewage water of a hospital (Scotta et al. 2011). These species have also been isolated from certain patients with weak immunity. In particular, B. diminuta and B. vesicularis have been isolated from patients with diseases such as pneumonia, bacteremia, meningitis, and lymphangitis (Ryan and Pembroke 2018; Altamirano-Beltrán et al. 2023). In addition, B. vancanneytii and B. huaxiensis were isolated from blood samples (Estrela and Abraham 2010; Liu et al 2021), ‘B. huaxiensis’ (not a validly published strain) from faeces (Huang et al. 2022), and ‘B.brasiliensis’ (not valid) from cerebrospinal fluid sample (Soares et al. 2023) of patients.

The genus Brevundimonas are Gram-negative and rod-shaped bacteria, containing ubiquinone-10 (Q-10) as well as C18:1 ω7c and C16:0 as the major isoprenoid quinone and cellular fatty acids, respectively, and displaying relatively high DNA G+C content (Dahal and Kim 2018; Lee et al. 2020).

In this study, we isolated a Gram-stain-negative, aerobic, short rod-shaped bacterial strain NCCP 15609T from a blood sample. Based on its genomic, biochemical, and chemotaxonomic characteristics, we established the taxonomic position of strain NCCP 15609T in the genus Brevundimonas, suggesting that it represents a novel species, and proposing to name it Brevundimonas sanguinis, meaning isolated from the blood.

Materials and methods

Bacterial strain isolation and culture conditions

Strain NCCP 15609T was isolated from the blood sample of a 3-year-old male patient from a hospital located at Jeonju in the Republic of Korea (GPS position: 35°84′61′′N, 127°13′95′′E). The isolate was cultured on a 5% sheep blood agar plate (BAP) at 37 °C and stored in 20% glycerol (w/v) at − 80 °C. The strain was registered at the National Culture Collection for Pathogens (NCCP, Republic of Korea) as NCCP 15609T and the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ, Germany) as DSM 116005T, respectively. The reference strain B. naejangsanensis DSM 23858T was purchased from DSMZ.

Identification and phylogenetic analysis

The strain was cultured on BAP at 37 °C for 1 day and the resulting colonies were identified by matrix-assisted laser desorption/ionisation–time-of-flight mass spectrometry (MALDI–TOF MS) using MALDI Biotyper software (Bruker Daltonik, Germany). The genomic DNA of strain NCCP 15609T was extracted using a MagAttract HMW DNA Kit (Qiagen) following the manufacturer’s instructions. PCR amplification of the 16S rRNA gene was carried out using the universal primers 27F (5′- AGAGTTTGATCMTGGCTCAG-3′) and 1492R (5′- TACGGYTACCTTGTTACGACTT-3′) (Chun and Goodfellow 1995). The 16S rRNA gene sequence similarities were calculated using the EzBioCloud server (www.ezbiocloud.net). The 16S rRNA gene sequences were aligned using CLUSTAL_W (Thompson et al. 1994) supplied with BioEdit version 7.2 software (Hall 1999). Phylogenetic trees were constructed using the software package MEGA version 11 (Tamura et al. 2021) with three algorithms (i.e., neighbour-joining (NJ) (Saitou and Nei 1987), maximum-likelihood (ML) (Felsenstein 1981), and maximum-parsimony (MP) (Fitch 1971). The phylogenetic tree topology was evaluated using bootstrap analysis based on 1,000 resamplings (Felsenstein 1985). In addition, the whole genomes were aligned and the phylogenetic tree was created using CLC genomics Workbench 23.0.5 (QIAGEN). The phylogenetic tree, based on whole genomes, was constructed using strains with 16S rRNA sequence similarity exceeding 97.0%.

Phenotypic and biochemical analysis

Growth at 4, 10, 15, 18, 25, 30, 37, 42, 45, and 50 °C was tested on nutrient agar (NA). The tolerance of NaCl (final concentration: 0%–5% [w/v], 1% intervals, and 0.5%) and pH values (pH 4–11, 1 pH intervals) were incubated on nutrient broth (NB) at 30 °C. The pH range was adjusted using HCl and NaOH. The strain was incubated on a fluid thioglycollate medium (MBcell) at 30 °C for the aerobe growth test. Cell morphology was observed using a transmission electron microscope (Tecnai 12, FEI). Cell motility was assessed using motility test medium (BD Biosciences) and examined using phase-contrast microscopy (ICC50, Leica). Gram staining was performed using a Gram staining kit (MBcell). Catalase and oxidase activities were determined using 3% (v/v) hydrogen peroxide and 1% (w/v) tetramethyl-p-phenylenediamine, respectively. Casein [2% (w/v) skimmed milk], cellulose (0.5%, w/v), gelatin (1.0%, w/v), starch (0.2%, w/v), and Tween 80 (1%, v/v) hydrolyses were tested using NA according to the protocols of Smibert et al. (1994). The DNase activity was determined using DNase test agar (Difco). API 20E, API ZYM, and API 50CH kits (bioMérieux) were used for other biochemical tests, enzyme activities, and carbohydrate utilisation following the manufacturer’s instructions.

Chemotaxonomic analysis

The isolated and reference strains NCCP 15609T and B. naejangsanensis DSM 23858T were grown on NA for 3 days at 30 °C. The cellular fatty acids were extracted using the fatty acid methyl esters method, and separated by gas chromatography. The fatty acid composition was identified using the Microbial Identification System (MIDI) version 6.1 and the TSBA6 database (Sasser 1990). For the extraction of polar lipids and isoprenoid quinone, the cells of strain NCCP 15609T grown in NB for 3 days at 30 °C were harvested and freeze-dried. Polar lipids were analysed using standard procedures (Minnikin et al. 1984). Extracted polar lipids were separated in the first and second dimensions with a mixture of chloroform:methanol:water (65:25:4 by volume) and chloroform:acetic acid:methanol:water (80:18:12:5 by volume), respectively, by thin-layer chromatography (TLC) using TLC silica gel 60F254 (Merck) (Da Costa et al. 2011). Separated polar lipids were identified by spraying with detection reagents. Isoprenoid quinones were extracted from 50 mg of lyophilised cells with chloroform:methanol (2:1 by volume) and purified by TLC. The purified quinones were analysed by high-performance liquid chromatography (HPLC) using an isocratic solvent system of methanol:2-butanol (2:1 by volume) (Collins 1985).

Genomic analysis

The genomic DNA of strain NCCP 15609T was extracted using the MagAttract HMW DNA kit (Qiagen). Whole-genome sequencing of the isolate was performed using the PacBio sequel single-molecule real-time (SMRT) sequencing technology (Pacific Biosciences) at Cosmogenetech Co., Ltd. (Republic of Korea). De novo assembly was performed using the Hierarchical Genome Assembly Process version 3 (HGAP3) (Chin et al. 2013). The assembled genome was annotated using the Prokaryotic Genome Annotation Pipeline (PGAP) version 6.5 of the National Center for Biotechnology Information (NCBI) (Tatusova et al. 2016). The functional annotation and classification of Clusters of Orthologous Groups (COG) of proteins were performed using Blast2GO version 6.0 (Conesa et al. 2005) and eggNOG-mapper version 2.12 (Cantalapiedra et al. 2021), respectively. Virulence Factor Database (VFDB) (https://www.mgc.ac.cn/VFs/) was used to predict the virulence genes. Antimicrobial resistance gene search was performed using ABRicate version 1.0.1 (https://github.com/tseemann/abricate).

The DNA G + C content was directly calculated from the genome sequence. The average nucleotide identity (ANI) was used to compare strain NCCP 15609T and related type species genome sequences. The ANI value was calculated by an ANI calculator using the OrthoANIu algorithm (https://www.ezbiocloud.net/tools/ani) (Yoon et al. 2017). The digital DNA–DNA hybridisation (dDDH) analysis was estimated using the Genome-to-Genome Distance Calculator 3.0 (GGDC) (https://ggdc.dsmz.de/ggdc.php#) (Meier-Kolthoff et al. 2022). Additionally, the average amino acid identity (AAI) analysis and AAI tree were performed, which provided useful measurements for species delineation using EzAAI (http://leb.snu.ac.kr/ezaai) (Kim et al. 2021).

Results and discussion

Identification and phylogenetic analysis

Strain NCCP 15609T was identified as B. diminuta using MALDI–TOF. A score of less than 2.0 was obtained, indicating low-confidence identification. The 16S rRNA gene sequence (1,425 bp) was obtained for strain NCCP 15609T (GenBank accession no. OQ271317). The sequence comparison using the EzBioCloud server indicated that the isolate was closely related to members of the genus Brevundimonas. Strain NCCP 15609T showed the highest 16S rRNA gene sequence similarity to B. naejangsanensis DSM 23858T (99.8%), followed by that to B. guildfordensis Sa3CVA3T(99.8%), B. diminuta ATCC 11568T(98.6%), B. vancanneytii LMG 2337T (98.5%), and B. faecalis CS20.3T (98.3%). Sequence similarities with other members of the genus Brevundimonas remained below 98.0%. Phylogenetic trees based on whole genomes (Fig. 1) and 16S rRNA sequences (Fig. S1) revealed that strain NCCP 15609T formed a distinct branch within the genus Brevundimonas. In tree based on whole genomes (Fig. 1) and neighbour-joining tree (Fig. S1), strain NCCP 15609T and B. naejangsanensis DSM 23858T were grouped with high bootstrap values of 100% and 99%, respectively, followed by B. guildfordensis (with bootstrap support of 100% and 66%, respectively). These three strains also formed a clade in the ML and MP algorithm trees. Based on the phylogenetic tree using the whole genome, B. naejangsanensis was selected as a reference strain.

Fig. 1
figure 1

Phylogenetic tree based on whole genomes showing the phylogenetic position of strain NCCP 15609T and related strains of the genus Brevundimonas. The numbers indicate the bootstrap values as a percentage of 1000 replicates. Bar, 0.05 substitutions per nucleotide position

Full size image

Phenotypic and biochemical analysis

The cells were observed to be Gram-stain-negative, non-motile, non-spore-forming, aerobic, and short rod to coccus shaped (2.0–3.5 × 4.0–5.5 μm in size; Fig. 2). Strain NCCP 15609T grew on NA and BAP5. The strain could grow with 0%–2.0% (w/v) NaCl (optimally at 0% NaCl), at pH 6.0–9.5 (optimally at pH 7.5–8.0), and 10–42 °C (optimally at 30 °C). Colonies were observed to be circular, smooth, whitish yellow colour, and 2–3 mm in diameter after incubation on NA for 3 days. The strain was found to be positive for catalase, oxidase, and tryptophan deaminase, but negative for urease and DNase activities. Casein was hydrolysed, but cellulose, gelatin, starch, and Tween 80 were not hydrolysed. There were several phenotypic characteristics, such as the absence of Tween 80 hydrolysis, acetoin production, and valine arylamidase enzyme activity that differentiated strain NCCP 15609T from the phylogenetically related strain B. naejangsanensis DSM 23858T. The species description and Table 1 present the detailed phenotypic characteristics of NCCP 15609 T.

Fig. 2
figure 2

Transmission electron microscopy of cell of NCCP 15609T. The cell were grown on nutrient broth at 30 °C for 3 days. Bar, 0.5 μm

Full size image
Table 1 Phenotypic characteristics that differentiate strain NCCP 15609T from phylogenetically related B. naejangsanensis DSM 23858T
Full size table

Chemotaxonomic analysis

The fatty acid profile of strain NCCP 15609T (> 10.0% of total fatty acids) comprised summed feature 8 (C18:1 ω7c and/or C18:1 ω6c) and C16:0, which were the same as those of related species B. naejangsanensis DSM 23858T (Table S1). These common major fatty acid profiles were similar to those of the genus Brevundimonas (Dahal and Kim 2018). However, strain NCCP 15609T contained a higher proportion of C17:0, C17:1 ω8c, and a lower proportion of C16:0 and summed feature 8 compared with the reference type strain. The predominant polar lipids of strain NCCP 15609T were a phosphatidylglycerol (PG), two 1,2-di-O-acyl-3-O-[D-glycopyranosyl (1 → 4)-α-D-glucopyranuronosyl] glycerol (DGL1–2), and an 1,2-di-O-acyl-3-O- α-D -glycopyranuronosyl glycerol (MGDOx). A smaller amount of two unidentified phospholipids (PL1–2), two unidentified lipids (UL1–2), an unidentified glycolipid (GL), an unidentified phosphoglycolipid (PGL), and one DGL were also detected (Fig. S2). The main respiratory quinone of strain NCCP 15609T was Q-10, which is also the main quinone of the genus Brevundimonas (Lee et al. 2020).

Genomic analysis

Whole-genome sequencing of strain NCCP 15609T yielded a single chromosomal contig of 3,063,090 bp, resulting in a total of 159,941 subreads (N50 = 9,210 bp) (GenBank accession no. NZ_CP116770). The genome size of NCCP 15609T was smaller than that of B. naejangsanensis DSM 23858T (GenBank accession no. NZ_ATXN00000000; 3,164,519 bp), although the genome size was similar to that of closely related Brevundimonas species (Table S2) and was within the range of those in the genus Brevundimonas (2.84–3.42 Mb) (Huang et al. 2022).

The DNA G + C content of strain NCCP 15609T was 67.4%, similar to the 66.6% G + C content of the related species B. naejangsanensis DSM 23858T. In general, the members of the genus Brevundimonas have high DNA G + C contents (Liu et al. 2021). The ANI, dDDH, and AAI values between NCCP 15609T and B. naejangsanensis DSM 23858T were 92.5%, 49.5% ± 2.6%, and 94.7%, respectively. NCCP 15609 T could thus represent a distinct species since ANI, dDDH, and AAI cutoffs of 95%–96%, 70%, and 95% respectively, are generally used for microbial species definition (Meier-Kolthoff et al. 2013; Yoon et al. 2017; Konstantinidis and Tiedje 2005). Species with 16S rRNA sequence similarities exceeding 97.0% were analysed for ANI, dDDH, and AAI, and each result fell below the respective cutoff levels (Table S2). Furthermore, in the AAI tree (Fig. S3), as in previous trees, strain NCCP 15609 T was confirmed to be grouped with B. naejangsanensis and followed by B. guildfordensis. This tree analysis also confirmed that NCCP 15609 T was distinct from other species within the genus Brevundimonas.

The genome of strain NCCP 15609T contained 3,029 genes, including 2,949 protein-coding sequences (CDSs), 52 tRNA genes, and 6 rRNA genes (two copies each of 5S, 16S, and 23S). Most of the identified coding sequences were classified as COG gene categories of unknown function (S, 21.2%) and amino acid transport and metabolism (E, 7.5%), followed by translation, and ribosomal structure and biogenesis (J, 6.6%). In addition, NCCP 15609T contained 130 and 34 genes related to DNA replication and repair (L, 4.9%) and defence mechanisms (V, 1.3%), respectively (Table S3). Compared with B. naejangsanensis, NCCP 15609T contained higher genes in the COG category L. In particular, it contained more genes related to the Type IV secretion system that transports DNA and proteins, which is associated with pathogenicity. Strain NCCP 15609T also encodes DNA recombinational repair proteins (RecA, RecF, RecO, RecR, RecJ, and RecN), nucleotide excision repair protein (Mfd), and mismatch repair proteins (MutL and MutS). Reactive nitrogen species (NO) produced by the bacterial-infected host cell causes bacterial DNA damage. However, it was reported that these DNA repair proteins enable the growth of the pathogen in the host by protecting against NO (Guillemet et al. 2016). Moreover, genes encoding the toxin-antitoxin system (ParD, YefM) were also identified in strain NCCP 15609T. On the other hand, the motility of strain NCCP 15609T could not be confirmed through microscopy, and genome analysis failed to detect motility-related genes.

We also analysed virulence genes based on VFDB, and the results showed that 27 and 28 virulence genes were identified in strains NCCP 15609 T and B. naejangsanensis, respectively (Table S4). In a previous study, virulence genes (icl, tufA, kdsA, htpB, and acpXL) were found in most of members of Brevundimonas (Huang et al. 2022). In this study, only four virulence genes were identified in strain NCCP 15609T and B. naejangsanensis, except the tufA encoded translation elongation factor. In addition, ahpC, which plays a role in protecting cells from oxidative stress, and genes related to the chemotaxis response regulator (cheY1) and chemotaxis signal system (cheB) were identified in both strains. Genes related to immune modulation, gmd1 and wzt2, were only detected in strain NCCP 15609T, whereas genes related to flagella biosynthesis protein flhA, and two-component response regulator, pilR and fleR, were only detected in strain B. naejangsanensis. Additionally, the ABRicate results showed that strain NCCP 15609T contained the antibiotic florfenicol resistance gene floR, whereas B. naejangsanensis contained a sulfonamide resistance dihydropteroate synthase gene sulII and a tetracycline resistance gene tet(G).

Taxonomic conclusion

The genotypic, phenotypic, and chemotaxonomic analyses presented in this study clearly indicate that the strain differs from the reference type strain. The physiological characteristics of strains NCCP 15609T and B. naejangsanensis are summarised in Table 1. In conclusion, we suggest that strain NCCP 15609T represents a novel species of the genus Brevundimonas, which we propose to name Brevundimonas sanguinis sp. nov.

Description of Brevundimonas sanguinis sp. nov.

Brevundimonas sanguinis (san’gui.nis. L. gen. masc. n. sanguinis: of the blood).

Cells are observed to be Gram-stain-negative, non-motile, non-spore-forming, aerobic, and short rod to coccus shaped, of diameters and lengths of approximately 2.0–3.5 and 4.0–5.5 μm, respectively. Colonies are observed to be circular, smooth, whitish yellow colour, and of a diameter of 2–3 mm upon incubation on NA for 3 days. The cells grow on NA and BAP with 0%–2.0% (w/v) NaCl (optimally 0%), 6.0–9.5 (optimally at pH 7.5–8.0) and at 10–42 °C (optimally at 30 °C). The strain is found to be positive for catalase, oxidase, and tryptophan deaminase activities, but negative for urease and DNase activities. Cells are confirmed to be positive for casein hydrolysis and citrate utilisation and negative for hydrolysis of arginine, cellulose, gelatin, starch, and Tween 80, nitrate reduction, and production of H2S, and indole. The predominant fatty acids (> 10%) are summed feature 8 (C18:1 ω7c and/or C18:1 ω6c) and C16:0. The major polar lipids are phosphatidylglycerol, two 1,2-di-O-acyl-3-O-[D-glycopyranosyl (1 → 4)-α-D-glucopyranuronosyl] glycerol, and 1,2-di-O-acyl-3-O- α-D -glycopyranuronosyl glycerol. The main respiratory quinone is ubiquinone-10. The type strain is NCCP 15609T (= DSM 116005T), isolated from a patient blood sample in the Republic of Korea, with a genome of 3.06 Mb and a DNA G + C content of 67.4%, containing 2,949 CDSs. The GenBank accession numbers for the 16S rRNA gene and genome sequence of strain NCCP 15609T are OQ271317 and NZ_CP116770, respectively.