Paenibacillus hubeiensis sp. nov., a New Selenium-Resistant Bacterium Isolated from the Rhizosphere of Galinsoga parviflora in a Selenium-Rich Area of Enshi, Hubei Province

1. Introduction

The genus Paenibacillus was originally described by Ash et al. based on the 16S rRNA gene analysis [1], and later emended by Shida et al., who proposed six new species in the genus [2]. The continuous emergence of new members increased the number of species belonging to Paenibacillus. At the time of writing (June 2025), the genus contains 323 species with validly published and correct names (https://lpsn.dsmz.de/genus/paenibacillus). Some members of this genus are rod shaped, aerobic or facultative anaerobic, motile or nonmotile, and endospore-form bacteria [3,4,5].

Paenibacillus species can be found in many ecological niches, such as fish gut [6], river otter [7], seashore land [8], hot spring [9], salt lake [10], digestive syrup [11], plants [12], and soil [13]. Among the genus, groups of species play a vital role in agricultural application through nitrogen fixation, phosphate solubilization, production of ACC deaminase, and IAA secretion [14,15,16], which are regarded as plant growth-promoting bacteria/rhizobacteria (PGPB/PGPR) [17,18,19].

Selenium (Se), an essential trace nutrient, has multiple biological functions, including roles in male reproductive health, the endocrine system, muscle function, the central nervous and cardiovascular systems, and immune syestems [20]. Research on Se has attracted extensive focus, partly because of selenocysteine, the constitutive part of the 21st amino acid, making it the only trace nutrient incorporated into proteins through genetic coding [21]. For example, certain Se functional microorganisms, called Se-respiring bacteria, play a pivotal role in the selenium cycle. Some selenate- or selenite-respiring bacteria are capable of producing Se nanoparticles [22]. Lin et al. [23] reported that Priestia sp. LWS1, a Se-resistant PGPR, could promote plant growth by 19% and Se accumulation by 75% in Oryza sativa.

Therefore, screening and isolating agricultural beneficial Se-resistant microorganisms from various habitats is imperative. In this study, a selenium-resistant PGPR, ES5-4^T was isolated from the rhizosphere of Galinsoga parviflora and characterized using polyphasic analysis; the results indicate that it is likely a novel species of the genus Paenibacillus.

2. Materials and Methods

2.1. Sample Collection and Isolation of Bacterial Strain

G. parviflora is the numerous species growing in selenium-rich ore area of Enshi, Hubei Province, China (30°8´43´´N 109°49´32´´E). Therefore, the rhizobacterial soil was collected in October 2020. The selenium (Se) content of the soil is 3.57 mg/kg, higher than 0.4 mg/kg, which was considered selenium-rich soil based on the national standard for soil selenium grade (GB/T 44971-2024). The bacterial isolation was based on the method described by Li et al. [24]. Briefly, the soil sample was diluted to 10^-3 to 10^-5 with sterile ddH₂O using a 10-fold dilution gradient and then smeared on Trypticase soy broth agar (TSA) medium supplemented with 20 mg/L of Se solution (Na₂SeO₃). After incubation for 3 days at 28 °C in an incubator, the single colony was selected, purified, and incubated on TSA plates with higher Se concentrations successively (50, 100, 200, 500, 1000, and 2000 mg/L). Finally, the strain ES5-4^T was obtained and maintained on TSA and preserved in 30% glycerol suspension at −80°C.

2.2. 16. S rRNA Gene Sequencing and Phylogenetic Analysis

The strain was cultivated in Trypticase soy broth (TSB) medium for 12 h to the exponential growth stage at 180 rpm in a rotary shaker (ZWY-2101C) at 28 °C. Next, the cell suspensions were used for the extraction of genomic DNA using a DNA isolation kit (Sangon Biotech, Shanghai, China), according to the manufacturer’s instructions. The 16S rRNA gene was amplified by PCR using the universal primers 27F and 1492R [25]. The PCR products were purified and sequenced by Sangon Biotech. The obtained 16S rRNA gene sequence was analyzed using the EzBioCloud server (https://www.ezbiocloud.net/) [26]. Multiple sequence alignments were performed using ClustalW software (http://www.clustal.org/clustal2/) [27]. Next, phylogenetic analysis of the 16S rRNA gene of strain ES5-4^T was conducted based on neighbor-joining (NJ), maximum-parsimony (MP), and maximum likelihood (ML) algorithms using MEGA 7 software [28]. Phylogenetic distances were calculated according to Kimura’s two-parameter model [29]. The bootstrap values were set as 1000 replications, as described by Felsenstein [30].

2.3. Genomic Characterization

Strain ES5-4^T was incubated to the exponential growth stage as mentioned above. The cell suspension was then centrifuged at 4 °C at 8000 × g, the supernatant was discarded, the mycelium was rinsed three times with sterile deionized water, and the mycelium was collected and used for genome sequencing. A draft genome sequence of strain ES5-4^T was conducted using the Illumina HiSeq X platform (Personalbio, Shanghai, China). Raw sequence reads were filtered and quality controlled using FastQC [31]. The high-quality reads were de novo assembled using SPAdes v3.15.3 [32] in the UGENE software package [33]. The G+C content of the genomic DNA was directly determined using the draft genome sequence. Values of average nucleotide identity (ANI) based on BLAST (OrthoANIb) between strain ES5-4^T and its close relatives (Paenibacillus illinoisensis NBRC 15959^T, Paenibacillus pabuli NBRC 13638^T, Paenibacillus taichungensis DSM 19942^T, and Paenibacillus xylanivorans A59^T) was calculated on the EzBioCloud server [26]. The Genome-to-Genome Distance Calculator (GGDC) online service (http://ggdc.dsmz.de/distcalc2.php) was used to calculate the digital DNA–DNA hybridization (dDDH) values by applying Formula 2 [34]. Annotation was performed using the NCBI Prokaryotic Genome Annotation Pipeline [35] and online eggnog-mapper v2 server (https://eggnog-mapper.embl.de/) [36].

To confirm the taxonomic status of the type strain, a genome-based ML phylogenetic tree (bac120 tree) [37] was constructed based on the protein sequences of the genus Paenibacillus, along with the outgroup. Generally, genomic information of type strains in the genus Paenibacillus was collected using scripts (https://github.com/zdf1987/Usefull_Scripts/GetTypeStrainGenomes) and the proteomes were downloaded using the NCBI software datasets (https://www.ncbi.nlm.nih.gov/datasets/docs/v2/download-and-install/). Other genomic datasets of strains that closely related to strain ES5-4^T were downloaded manually. Gene sequences of each gene cluster were aligned using MUSCLE [38], and the ML tree was constructed using FastTree 2.1 [39] based on the common genes in the bac120 set after alignment trimming and concatenation. The bac120 tree was constructed automatically using EasyCGTree v4.2 (https://github.com/zdf1987/EasyCGTree4) [40].

2.4. Phenotypic and Biochemical Characterization

The phenotypic characteristics of strain ES5-4^T were evaluated on TSA. Gram staining was conducted using a Gram-staining kit (DM0015, Leagene, Beijing, China). Cell morphology of the strain was visualized using scanning (Sirion 200) and transmission electron microscopes (SEM, Sirion 200; TEM, Tecnai G2 spirit Bio-Twin) [41]. The effect of temperature (4, 10, 20, 28, 37, and 45 °C) on bacterial growth was observed on the TSA after 3 days of incubation. The pH in the range of 4.0 to 10.0 with an increment of 1.0 for bacterial growth was also assessed. The pH was adjusted with 1 M HCl or NaOH. Salt tolerance was evaluated using various NaCl concentrations (0%, 0.5%, 1.0%, 2.0%, 3.0%, 5%, 7%, and 10% w/v). The bacterial motility experiment was conducted by observing cell growth in test tubes with semisolid TSB with 0.5% agar after 3 days of incubation in an incubator at 28 °C. Other physiological and biochemical characteristics of the strain were determined using API 50CH (bioMérieux, France) and Biology GEN III micro test systems (Biology, USA). The reference strains (Paenibacillus oceanisediminis JCM17814^T, Paenibacillus dongdonensis KCTC 33221^T) used in the study were provided by Marine Culture Collection of China (MCCC). All experiments were performed in triplicate.

2.5. Chemotaxonomic Analysis

Strain ES5-4^T and its reference type strains were cultivated in TSB until to the logarithmic phase at 180 rpm in a rotary shaker (ZWY-2101C) at 28 °C. The cells were collected, washed, and lyophilized in a freeze dryer (Labconco, Kansas City, USA) [42]. The dried cells were used for cellular fatty acids, respiratory quinones, and polar lipid analysis. Cellular fatty acids were extracted and analyzed using the Sherlock Microbial Identification System (MIDI) following Sasser’s protocol [43]. Quinones were purified and analyzed using high-performance liquid chromatography [44]. Polar lipids were first extracted based on the method of Minnikin et al. [45], and identified using two-dimensional thin-layer chromatography, according to Komagata and Suzuki [46].

2.6. Effects of Se⁴⁺ Concentration on the Growth of Strain ES5-4^T

To assess the effect of Se⁴⁺ (supplied as Na₂SeO₃) on the cell growth of strain ES5-4^T, TSA and TSB media with various Se concentrations (0, 20, 50, 100, 200, 500, 1,000, 2,000, and 5,000 mg/L) were prepared. For the TSB test, 600 μL of overnight cultured cell suspension, equivalent to 3% inoculation volume (v/v), were added to 50-mL sterile Erlenmeyer flasks containing 20 mL of TSB with aforementioned Se⁴⁺ concentration and cultured for 2 days in a rotary shaker (ZWY-2101C) at 180 rpm at 28 ℃. Next, the optical density at the 600 nm wavelength (OD₆₀₀) was measured using a Multiskan Go microplate spectrophotometer (Thermo Fisher Scientific, CA, USA). Cell growth on TSA plates containing equivalent Se⁴⁺ concentrations was also performed. Each plate was evenly divided into three compartments, and 0.5 μL of overnight cultured cell suspension was dripped into each compartment, cultured in a biochemical incubator (Xinmiao, Shanghai, China) for 2 days at 28 ℃, and the growth of strain ES5-4^T was recorded. The diameter of the colonies of strain ES5-4^T was measured using a digital vernier caliper (DWGR-2954, DELIXI, China). All experiments were performed in triplicate.

2.7. Identification of the Functional Gene for Selenocompound Metabolism

Twenty-two genes involved in selenocompound metabolism (map00450) in bacteria of the genus Paenibacillus were identified according to a strategy similar to that previously described [47]. Briefly, profile hidden Markov models (HMMs) of the 22 relevant genes were retrieved from the KOfam database (https://www.genome.jp/ftp/db/kofam/) [48]. These HMMs were merged and adjusted to meet the requirements of the HMMER v3.0 software (http://hmmer.org), which was used by the analysis pipeline based on EasyCGTree v4.2 (https://github.com/zdf1987/EasyCGTree4) [49]. Subsequently, the "Gene_Prevelence.pl" script within the EasyCGTree software package was used to search for homologous genomes using customized HMMs with an e-value cutoff of 1e-80. All candidate genes were further submitted to the online annotation tool KofamKOALA (https://www.kegg.jp/kofamkoala/) [48] for homolog verification. Finally, the online tool ImageGP2 (https://www.bic.ac.cn/BIC/#/) was employed to generate a heatmap of the gene distribution.

2.8. Statistical Analysis

Data were analyzed using a one-way ANOVA. Pairwise differences among treatments were performed using Duncan’s multiple range test. Data were visualized using SigmaPlot 14.0 (Systat Software, San Jose, CA, USA). All statistical analyses were conducted using SPSS 20.0 (IBM, Armonk, NY, USA), with a significance level set at p < 0.05.

3. Results and Discussion

3.1. Phylogenetic Analysis of the 16S rRNA Gene

The 16S rRNA gene sequence of the strain ES5-4^T was 1,403 bp in length, spanning the complete gene sequence (1552 bp; RefSeq accession NZ_JAVYAE020000044) with three mismatches, and has been deposited at GenBank/EMBL/DDBJ (accession number: OR584159). Strain ES5-4^T shared the highest 16S rRNA gene sequence similarity (98.4%) with P. pabuli NBRC 13638^T based on the results from EzBioCloud server [26], followed by P. dongdonensis KUDC 0114^T (98.3%) [50], P. taichungensis BCRC 17757^T (98.3%) [51] and P. xylanivorans A59^T (98.0%) [52]. Alignment of the full-length 16S rRNA gene on the genome against the genome database (refseq_genomes and wgs) on NCBI suggested that there were some genomes from uncharacterized bacteria sharing 16S rRNA gene identities of > 98.5% to strain ES5-4^T, including Paenibacillus sp. FSL W8-0426 (100%, NZ_CP150203.1), P. pabuli E1 (98.8%, NZ_CP073714.1), P. xylanilyticus W4 (98.7%, NZ_CP044310.1), Paenibacillus sp. 11B (98.6%, NZ JAUKNS010000005.1), P. xylanilyticus L-2 (98.6%, NZ CP182582.1), Paenibacillus sp. 1781tsa1 (98.58%, NZ JAMYWY010000001.1), Paenibacillus sp. PvP094 (98.5%, NZ JBEPOM010000001.1), and Paenibacillus sp. FSL R5-0713 (98.5%, NZ_CP150229.1).

Phylogenetic analysis using ML method indicated that strain ES5-4^T clustered with P. dongdonensis KUDC 0114^T (Figure 1). The MP method revealed that the strain clustered with P. illinoisensis NRRL NRS-1356^T and P. xylanilyticus XIL 14^T (Figure S1). The above two topologies could not be reproduced in the trees using the NJ method (Figure S2). In addition, this phylogeny was not well supported (bootstrap value < 70%) in all three trees. This meant phylogenetic analysis based on the 16S rRNA gene in the genus Paenibacillus was not reliable. P. dongdonensis KUDC 0114^T (= KCTC 33221^T) and P. oceanisediminis L 10^T (= JCM 17814^T) were used as reference strains in the following tests because they have no available genomes for genome-based distinguishing from strain ES5-4^T.

3.2. Genome-Based Characteristics

The whole Genome Shotgun project of strain ES5-4^T has been deposited at GenBank/EMBL/DDBJ with the accession number JAVYAE000000000. The whole draft genome contained 58 contigs, with a total length of 7,190,355 bp, and the genome G + C content was 50.5%. A total of 6,300 protein coding genes were predicted, of which 642, 3, 046 and 5, 370 genes were assigned to the Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Cluster of Orthologous Groups (COG) databases, respectively. Further KEGG pathway analysis revealed 34 genes involved in flagellar assembly (ko02040), 3 genes (narI, narJ, and narH) associated with nitrate reduction, 3 genes (cysI, cysJ, and nirB) with sulfite reduction, the nitrogen fixation gene nifU, and 317 ABC transporters; no photosynthesis gene cluster was detected.

According to the taxa of the 16S rRNA gene tree, 205 Paenibacillus genomes (including the seven aforementioned atypical strains together with genomes of strain ES5-4^T and Staphylococcus aureus NCTC 8325^T (GCA000013425.1, as outgroup) were used to infer a robust tree based on the bac120 gene set (Figure 2). The bac120 tree was well supported on all branches. Strain ES5-4^T was closely related to Paenibacillus sp. FSL W8-0426 (GCA_037969725.1) and was subsequently clustered with ten Paenibacillus species, including P. pabuli NBRC 13638^T, P. taichungensis DSM 19942^T, and P. xylanivorans A59^T. The results of the 16S rRNA gene tree and bac120 tree both demonstrated that the strain was within the genus Paenibacillus.

ANIb values between strain ES5-4^T and the closely related strains P. pabuli NBRC 13638^T (GCA004000925.1), P. taichungensis DSM 19942^T (GCA013359905.1), P. xylanivorans A59^T (GCA001280595.1) and P. illinoisensis NBRC 15959^T (GCA004000925.1) were 76.4%, 76.2%, 76.3% and 76.8%, respectively, whereas the corresponding dDDH values were 21.2%, 20.9%, 20.6% and 21.6%, respectively (Table 1). These values were lower than the proposed and acceptable species boundaries of 95-96% for ANI and 70% for dDDH [53,54]. In addition, strain ES5-4^T showed an ANI value of 98.8% with strain Paenibacillus sp. FSL W8-0426 (GCA_037969725.1), which had a 16S rRNA gene similarity of 100%. Taken together, these findings imply that the strain ES5-4^T represents a novel Paenibacillus species.

3.3. Phenotypic and Biochemical Characteristics

Strain ES5-4^T was found to be Gram positive, aerobic, and motile with flagella (Figure 3A), and the features were consistent with the description of Paenibacillus flagellatus [55]. SEM image of strain ES5-4^T cells showed a rod-shaped profile with 0.4–0.6 μm width and 1.0–2.5 μm length (Figure 3B). Colonies on TSA were white, circle, moist, and smooth after incubation for 72 h at 28 °C. Growth occurred at NaCl concentrations of 0–5.0% (w/v), pH of 5.0–10.0, and temperature of 4–42°C. The optimal growth conditions for strain ES5-4^T were 0.5%–1.0% (w/v) NaCl, pH of 5.0–7.0, and 28 °C; those for P. dongdonensis KCTC 33221^T were 0.5%–2.0% (w/v) NaCl, pH of 6.0–7.0, 28 °C; and those for for P. oceanisediminis JCM17814^T were 0.5–1.0% (w/v) NaCl, pH of 7.0–8.0, and 28 °C (Table 2).

In the API 50CH tests, strain ES5-4^T was only positive for esculin, and negative for L-arabinose, ribose, D-xylose, β- methyl-D-xyloside, galactose, glucose, fructose, rhamnose, mannitol, amygdalin, arbutin, salicin, cellobiose, maltose, lactose, melibiose, saccharose, trehalose, raffinose, and gluconate, unlike reference strain P. oceanisediminis JCM17814^T and P. dongdonensis KCTC 33221^T, which displayed positive reactions in these tests. Compared with the reference strain P. dongdonensis KCTC 33221^T, which was positive for H₂S production, hydrolysis of starch, cellulose, Tween 20, 40, and 60, strain ES5-4^T was positive only for H₂S production and hydrolysis of Tween 40 and 60. P. oceanisediminis JCM17814^T was positive for hydrolysis of starch, cellulose, and Tween 40 and negative for other relative tests (Table 2).

In the BIOLOG GEN III tests, strain ES5-4^T was positive for α-D-glucose, D-mannose, D-fructose, D-galactose, D-fucose, L-glutamic acid, L-serine, D-galacturonic acid, L-galactomic acid lactone, D-gluconic acid, D-glucuronic acid, glucuronamide, mucic acid, quinic acid, citric acid, and L-malic acid. In comparion, P. oceanisediminis JCM17814^T was negative for utilization of D-fucose, L-glutamic acid, L-serine, glucuronamide, mucic acid, and citric acid whereas P. dongdonensis KCTC 33221^T was negative for L-serine and glucuronamide. These differential characteristics among the three strains are summarized in Table 2.

3.4. Effects of Se⁴⁺ Concentration on the Growth of Strain ES5-4^T

The effect of Se⁴⁺ concentration on the growth of strain ES5-4^T was investigated using TSB and TSAs. As Se⁴⁺ concentration increased, the cell suspension color changed from yellow to deep orange red and then to light orange red in the TSB (Figure S3), indicating that the white Se⁴⁺ was converted to red selenium nanoparticles (SeNPs) by strain ES5-4^T [56]. The optical density values at 600 nm (OD₆₀₀) were disturbed by the red color caused by SeNPs except for strain concentration caused by Se⁴⁺ solution. Therefore, the values showed a trend of first decreasing, then increasing, and then decreasing again, reaching their maximum at a Se⁴⁺ concentration of 500 mg/L (Figure S3). The cell colors changed from white to red when strain ES5-4^T grew on TSA (Figure S4). In addition, with the increase of Se⁴⁺ concentration, the colony diameter seemed to decrease significantly (p < 0.05) (Figure 4). Bacillus sp. KW3, Corynebacterium sp. VS5, and Pseudomonas sp. VW7 have been reported to be resistant to Se⁴⁺ above 200 mM [57]. The minimal inhibitory concentrations of Lysinibacillus xylanilyticus and L. macrolides for Se⁴⁺ were 120 and 220 mM, respectively [58]. Strain ES5-4^T can be resistant to Se⁴⁺ concentration at least 5,000 mg/L, approximately responding to 30 mM.

Besides strain ES5-4^T, many other Paenibacillus species could also reduce white Se⁴⁺ to red SeNPs. For example, Paenibacillus selenitireducens ES3-24^T isolated from a selenium mineral soil was able to reduce Se to red elemental selenium [59]. Long et al. [60] reported the same function of Paenibacillus motobuensis LY5201.

SeNPs are versatile in their application, such as antioxidant [61], antibacterial [62], anticancer and immunomodulatory [63], bioremediation [64], detoxification of heavy metals, nutraceutical functions, as well as the fortification of selenium dietary especially in selenium deficiency areas [65]. Strain ES5-4^T might also be applicable in the aforementioned aspects. However, this requires further research.

3.5. Genetic Features of Paenibacillus for Selenocompound Metabolism

Given strain ES5-4^T’s resistance to Se⁴⁺, the distribution of the 22 genes involved in selenocompound metabolism was analyzed in the genus Paenibacillus and 14 genes were detected (Figure 5, Dataset S1). Genes trxB/TRR, metC, patB/malY, metE, MARS/metG, sufS, yitJ, and metH/MTR were almost universal in these strains, with metH/MTR absent in Paenibacillus shenyangensis A9. selD/SEPHS was only present in P. shenyangensis A9, which catalyzes selenophosphate synthesis using inorganic selenium and ATP as substrates [66]. metB, which can catalyze selenocysteine to selenocystathionine [67], was present only in P. xylanexedens DSM 21292, P. shenyangensis A9, and Paenibacillus sp. 1781tsa1. The gene for the reaction (E4.4.1.11), capable of producing methaneselenol from selenomethionine [68], was only found in Paenibacillus sp. FSL W8−0426, P. silvae CGMCC 1.12770, and strain ES5-4^T (Figure 5). The thioredoxin reductase gene trxB/TRR catalyzed selenite (SeO₃^2-) to hydrogen selenide (H₂Se), which was spontaneously oxidized by oxygen to become Se (0) and eventually formed SeNPs [69,70,71]. Cystathionine-β-lyase coded by metC catalyzes selenocystathionine to selenohomocysteine [72,73]. Together, the results suggest that species in the genus Paenibacillus were versatile in selenocompound metabolism. Notably, strain ES5-4^T carries two copies of metC and trxB/TRR, which likely confers its selenium resistance (Figure 5).

3.6. Chemotaxonomic Characteristics

The major cellular fatty acids (> 10%) were anteiso-C_15:0 (46.5%) and C_16:0 (21.7%) for strain ES5-4^T, iso-C_14:0 (10.1%), anteiso-C_15:0 (45.6%), and iso-C_16:0 (20.0%) for P. oceanisediminis JCM 17814^T and C_16:0 (18.3%), and anteiso-C_15:0 (50.3%) for P. dongdonensis KCTC 33221^T (Table S1). The polar lipids of strain ES5-4^T mainly comprised diphosphatidylglycerol (DPG), phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and unidentified polar lipid (PL), which is similar to that of the closely related strain P. oceanisediminis JCM17814^T (Figure S5). The predominant respiratory quinone of strain ES5-4^T was MK-7, in conformity to the genus Paenibacillus [74,75].

4. Conclusions

A strain designated ES5-4^T was isolated from rhizosphere (with Se content of 3.57 mg/kg) of G. parviflora growing in the selenium-rich ore area of Enshi, Hubei Province, China. 16S rRNA and bac120 gene trees indicated that strain ES5-4^T formed a distinct clade in the genus Paenibacillus. The ANI and dDDH values between ES5-4^T and its closely related species were below the boundaries for ANI of 95%–96% and dDDH of 70%. Furthermore, strain ES5-4^T differed from the reference strains P. oceanisediminis JCM17814^T and P. dongdonensis KCTC 33221^T in phenotype and chemotaxonomic characteristics, such as growth conditions, motile, starch and cellulose hydrolysis, utilization of some substrates, polar lipids, and cellular fatty acids. Based on these phenotypic, chemotaxonomic and genotypic features, strain ES5-4^T can be considered a novel Paenibacillus species, for which the name Paenibacillus hubeiensis sp. nov. is proposed.

Description of Paenibacillus hubeiensis sp. nov.

Paenibacillus hubeiensis (hu. bei. en’sis. N.L. masc. adj. hubeiensis, pertaining to the province of Hubei, where the selenium-rich ore is located, from which the strain was isolated)

The strain (ES5-4^T) is Gram-staining positive, motile, aerobic, and rod shaped. The colonies on TSA are milky, circle, moist and smooth after incubation for 3 days at 28°C. The cells can be grown at 4–42 °C (optimum, 28 °C), pH 5.0–10.0 (optimum, 5.0-7.0) and is tolerated to 5% (optimum, 0.5% – 1.0%) of NaCl. It can tolerate up to 5,000 mg/L Se⁴⁺ and can produce SeNPs. It is only positive for esculin in the API 50CH analysis. In the BIOLOG GEN III tests, it is positive for 16 types of organic carbon sources. The major fatty acids are anteiso-C_15:0 (46.5%) and C_16:0 (21.7%). The predominant respiratory quinone is MK-7. The polar lipids mainly comprise diphosphatidylglycerol (DPG), phosphatidylethanolamine (PE), phosphatidylglycerol (PG) and an unidentified phospholipid (PL).

The type strain, ES5-4^T (= GDMCC 1.3540^T = KCTC 43478^T), was isolated from the rhizosphere of G. parviflora, collected from a selenium-rich ore area of Enshi, Hubei Province, China (30°8´43´´N 109°49´32´´E). The genomic G + C content of the strain was 50.5%. The 16S rRNA gene sequence and genome were deposited at GenBank/EMBL/DDBJ with the accession number of OR584159 and JAVYAE000000000, respectively.