We encountered limited previous research pertaining to bacterial diversity in the roots of marigold plants. These unexplored environments present promising prospects for the discovery of rare actinomycetes that are believed to be rich sources of novel bioactive compounds. Therefore, to investigate the analysis of beneficial and culturable bacterial community in the roots of these plants, we collected samples from the garden of Dongguk University, Ilsan, Republic of Korea (37° 40′ 26.4” N 126° 48′ 20.88″ E) (Supplementary Figure S1). To prepare the samples, we removed the roots from the soil. Subsequently, a thorough washing with sterile water was conducted to remove any external soil from the surface. The root surface was disinfected with 70% ethanol for 1 min and washed several times with sterile distilled water again. The surface-sterilized roots were then pulverized using a ceramic mortar. Serial dilutions using 0.85% NaCl saline were made, and aliquots of 0.1 mL from decimal dilution (10⁻³–10⁻⁷) were plated on Reasoner’s 2 agar (R2A) (Difco). After 1 week of incubation of the inoculated plates at 30°C, isolates were obtained. Fourty-eight different colonies were selected according to their different morphology and color. Each individual colony was purified by transferring to them onto new R2A, nutrient agar (NA), and tryptic soy agar (TSA) plates. Subsequently, these purified colonies were sent to Bionics (Daejeon, Republic of Korea) for 16S rRNA gene analysis. The isolates were cultivated on R2A agar at 30°C for 5 days and simultaneously maintained in 50% glycerol at −80°C.

Bacterial DNA was extracted using the TaKaRa MiniBEST Bacteria Genomic DNA extraction Kit version 3.0 (TaKaRa), following the manufacturer’s instructions. Small subunit 16S rRNA gene fragments were amplified with universal bacterial primers 27F, 518F, 805R, and 1492R; the resulting PCR products were commercially sequenced (Solgent, Korea). The 16S rRNA gene sequence identity similarities between strain RG38^T and other type strains were determined, and phylogenetic relationships were analyzed using the EzBioCloud server (Kim et al., 2012). Multiple sequence alignments were performed with the ClustalW program (Thompson et al., 1997). Phylogenetic and molecular evolutionary analyses were conducted using MEGA version 7.0 (Kumar et al., 2016). Clustering was determined using the neighbor-joining (NJ), maximum-likelihood (ML), and maximum-parsimony (MP) algorithms. The distances for the NJ, ML, and MP trees were calculated according to the two-parameter model of Kimura (1980), and the bootstrap values were established based on 1,000 replicates (Felsenstein, 1985).

The strain RG38^T was subjected to whole-genome shotgun sequencing at Macrogen (Republic of Korea) using the Illumina HiSeq 2,500 platform with a 150-bp × 2 paired-end kit. Subsequently, the data were assembled using the SPAdes analysis v.3.10.1 at Macrogen (Seoul, Republic of Korea). After assembly, the locations of protein genes were predicted, and their functions were annotated. Prokka was employed for location prediction, while BLAST and evolutionary genealogy of genes: Nonsupervised Orthologus Groups (eggNOG) 4.5 database (Huerta-Cepas et al., 2017) were used for functional determination and identification of assembled sequences against the nucleotide and protein sequence databases. The genomes of the reference strains were retrieved from the NCBI database and used as references for digital DDH and ANI analysis of the query genome of strain RG38^T. The estimated DDH values were calculated by the recommended formula 2 (identities/HSP length) using the Genome-to-Genome Distance Calculator (GGDC)¹ (Meier-Kolthoff et al., 2013), and ANI values were from EZ Biocloud platform ² (Yoon et al., 2017a). Average amino acid identity (AAI) values were calculated from protein sequences by using an online AAI calculator ³. Two-way AAI analysis was used. Genomic circular feature map was constructed using CGView server (Grant and Stothard, 2008). To evaluate the intergenomic distances between genome sequences of the strain RG38^T and its close strains belonging to the phylogenetically closest Streptomyces species, the FastANI values were also determined (Jain et al., 2018). The DNA G + C content of the strain RG38^T was calculated based on whole-genome sequence. CheckM bioinformatics tool was used to assess genome contamination and completeness ⁴ of the strain RG38^T (Parks et al., 2015). To enhance the phylogenetic classification and gain a better understanding of the relationships between the novel isolate and closely related species, phylogenomic trees were constructed based on an up-to-date bacterial 92 core gene set (UBCG) (Na et al., 2018). The draft genome of the strain RG38^T was then analyzed for the presence of BGCs using antiSMASH 5.0 online software (Naughton et al., 2017).

Comparative studies were conducted for all strains in triplicate on International Streptomyces Project (ISP) 2 media (ISP2) at 30°C. For morphology, cells were grown on ISP2 medium for 7 days at 30°C. Subsequently, cells were fixed in 2.5% (v/v) glutaraldehyde and stored for 4 h at 4°C, and fixing agent was removed by rinsing with PBS (phosphate buffer saline). Dehydration was achieved through successive washes with 30, 50, 60, 70, 80, and 90% (v/v) and absolute ethanol. Then, the cells were subjected to critical point dehydration at room temperature followed by mounting on a stub with a carbon disk. After overnight drying, a platinum coating was applied using a scanning electron microscopy (SEM) coating unit (15 nm; EM ACE200, Leica, Wetzlar, Germay). Images were analyzed using field emission-scanning electron microscopy (FE-SEM) analysis.

Cultural characteristics for the strain were determined after incubation at 30°C for 2 weeks using ISP2 media, Bennett’s agar (BA), Nutrient agar (NA), Reasoner’s 2 agar (R2A), Marine agar (MA), Luria bertani agar (LBA), and Tryptic soy agar (TSA; Difco). Growth was tested at different temperatures (4, 10, 15, 20, 25, 30, 35, 40, 45, and 50°C). NaCl tolerance was determined on ISP2 medium supplemented with 1–16% NaCl (w/v, with an interval of 1% w/v) at 30°C for 14 days. The pH range for growth was determined by cultivation at 30°C in ISP2 broth adjusted to pH 4–10 (at pH 1 unit intervals) before sterilization with citrate/NaH₂PO4 buffer (pH 4.0–5.0), phosphate buffer (pH 6.0–8.0), and Tris buffer (pH 9.0–10.0) as described previously (Kim et al., 2020; Chhetri et al., 2021a). The pH values were verified after autoclaving. Cellular fatty acid profiles and quinones were not analyzed in this study, owing to the revised guidelines of the International Code of Nomenclature of Prokaryotes or IJSEM for taxonomic descriptions. According to these updated guidelines, chemotaxonomic analyses, such as polar lipid and fatty acid analysis, are no longer obligatory for the characterization of novel bacterial species. They are only mandated for novel genera ⁵.

Biochemical tests were conducted using the API 20NE and API ZYM strips (bioMérieux), following the manufacturer’s instructions. Additionally, Tween (20, 40, and 80), casein, starch, CM cellulose, and chitin were performed as previously described (Kim et al., 2019; Chhetri et al., 2021b, 2023).

The strain RG38^T was cultured in R2A, NA, TSA, BA, MA, LBA, and ISP2 media at 30°C to determine the best optimal medium for antibiotic production. Pathogenic bacteria Staphylococcus epidermidis KACC 13234, Staphylococcus aureus ATCC 6538, Xanthomonas campestris pv. campestris KACC 10377, Bacillus subtilis KACC 16747, Micrococcus luteus KACC 13377, Botrytis cinerea KACC 40573, and Pantoea agglomerans KACC 10054 were grown in Tryptic soya broth (TSB) at 37°C. Subsequently, after a 12-h incubation period, 80 μL aliquots of each pathogenic bacterium were individually spread onto NA plates. Following this, 4-day-old agar blocks of the strain RG38^T were transferred onto plates that had been pre-inoculated with the pathogenic bacteria. The plates were then incubated at 30°C for 24 h, and the activity was quantified by measuring the inhibition zone in millimeters surrounding the agar block (Maiti et al., 2020).

The agar extraction method, employed in this study, was adapted from a previous study (Carr et al., 2010). In brief, 30 large Petri dishes (150 mm x 15 mm) were used. Strain RG38^T was spread on NA plates and then incubated at 30°C. After 4 days of incubation, the agar in the plates was chopped into small pieces. These agar pieces were collected in two separate 2 L bottle, into which ethyl acetate (1 L) was added to facilitate the absorption of agar pieces. The bottles were rotated at 150 rpm in room temperature and left for 1 day. The resultant ethyl acetate solution was filtered using Whatman™ filter paper 42 (150 mm; GE healthcare UK). The filtered solution was centrifuged to remove the remnants at 5,000 rpm for 20 min at 4°C. The organic layer was collected and completely concentrated using a rotary evaporator at 40°C as described previously (Chhetri et al., 2022). The remaining residue of the crude product was dissolved in 500 μL methanol, and subsequently, 15 μL was applied to a sterile paper disk (6 mm, Whatman). The antimicrobial activity against pathogenic bacteria was assessed, and the zone of inhibition was measured. The extract in methanol was concentrated using a vacuum evaporator and was analyzed by reverse-phase high-performance liquid chromatography (HPLC) using solvents A (H₂O) and B (acetonitrile) (Agilent 1,260, C18, 3.5 μm, 4.6 × 100 mm, the detection wavelengths were 254 and 280 nm). Major peaks were isolated and purified by HPLC on a gradient comprised of solvents A (H₂O) and B (acetonitrile): 40–75% B (0–12 min), 75–100% B (12–16 min), 100% B (16–30 min), flow rate: 3 mL/min (Agilent 1,260, SB-C18, 5 μm, 9.4 × 250 mm). Chromomycins A2 (1) and A3 (2) were isolated and purified by HPLC on a gradient comprised of solvents A (H₂O) and B (acetonitrile): 40–75% B (0–12 min), 75–100% B (12–16 min), 100% B (16–30 min), flow rate: 3 mL/min (Agilent 1,260, SB-C18, 5 μm, 9.4 × 250 mm). Solvents were evaporated under a vacuum using Scanvac Concentrator. Dried pure products were stored at −20°C. Next day, each fraction was tested for antimicrobial activity using 6 mm disk (Whatman). Active fraction obtained was further chromatographed using methanol as a mobile phase and obtained single peak, which indicated the purity of compound. At last, lyophilization of purified compound was performed. NMR spectra of isolated compounds were acquired on a Varian 400 (Varian, Palo Alto, CA, USA) at 25°C.

The lyophilized crude extract from the strain RG38^T was dissolved in LC–MS grade methanol through sonication and filtered with a 0.2 μm regenerated cellulose (RC) syringe filter (GVS filter technology, IN, USA) with a concentration of 2 mg/mL. The UPLC-QTOF-MS/MS data were obtained using the Waters Acquity UPLC system (Milford, MA, USA), comprising a binary solvent delivery system, an autosampler, and a photo diode array (PDA) detector. The UPLC column utilized was the Waters Acquity UPLC BEH C18 (150 mm × 2.1 mm, 1.7 μm). The mobile phase was 20 mM formic acid in water (A) and acetonitrile (B), with the following gradient: 10–90% B (0–14 min, v/v). The flow rate was set at 300 μL/min, and the injection volume was 2.0 μL. The autosampler and column oven were kept at 15°C and 40°C, respectively. Mass spectrometry experiments were conducted using a Waters Xevo G2 QTOF mass spectrometer (Waters MS Technologies, Manchester, UK), which was connected to the UPLC system through an electrospray ionization (ESI) interface. The ESI condition was set as follows: negative ion mode, capillary voltage 2.5 kV, cone voltage 40 V, source temperature 120°C, desolvation gas temperature 350°C, cone gas flow 50 L/h, and desolvation gas flow 800 L/h. The ion acquisition rate was set at 0.2 s. Data were centroided during acquisition using an independent reference lock-mass ion via the LockSpray™ interface to ensure accuracy and precision. Leucine enkephalin (m/z 554.2615 in negative mode) was used at a concentration of 200 pg/μl with an infusion flow rate of 5 μL/min. The acquired data were analyzed using a feature-based molecular networking analysis, which is available on the GNPS web platform ⁶ with a spectral preprocessing by MZmine3 software. All the results and parameters can be accessed with the GNPS job id for molecular network analysis ⁷ and enhanced analysis ⁸.

In the EzBioCloud analysis, 16S rRNA gene sequences indicated that the closest relatives of the strain RG38^T were Streptomyces collinus NBRC 12759^T (99.3%), S. violaceochromogenes NBRC 13100^T (99.2%), S. viridiochromogenes NBRC 3113^T (99.1%), S. griseoflavus LMG 19344^T (99%), and S. paradouxus NBRC 14887^T (99%). The 16S rRNA sequence similarity values between strain RG38^T and other species ranged from 98.9 to 98.3%. The NJ tree showed the grouping of the strain RG38^T within the genus Streptomyces and clusters with S. collinus NBRC 12759^T, S. iakyrus NRRL ISP-5482^T, and S. glaucescens NBRC 12774T (Figure 1A). Similar results were also obtained in the ML and MP trees (Supplementary Figures S2, S3). Based on the phylogenetic trees, S. collinus NBRC 12759^T, S. iakyrus NRRL ISP-5482^T, and S. glaucescens NBRC 12774^T were selected to compare the ANI, dDDH, and AAI and characteristics of the strain RG38^T with those of the type strains in the genus Streptomyces. The selected reference strains were purchased from the KCTC (Korean Collection for Type Cultures) and JCM (Japan Collection of Microorganisms). The draft genome of the strain RG38^T contained 19 contigs, with a length of 7.15 Mb and an N50 length of 781,243 bp. A total of 6,289 genes were predicted, which included 6,053 protein-coding genes and 68 tRNA genes. The ANI threshold for species demarcation is recommended to be 95–96%, and the ANI values between the strain RG38^T and its phylogenetically closest neighbors were ≤ 84.7% (Table 1). The dDDH values between the strain RG38^T and other members of Streptomyces genus were ≤ 28.8%, which was lower than the 70% species threshold recommended for species delineation (Meier-Kolthoff et al., 2013; Yoon et al., 2017b). In addition, the tight association of strain RG38^T with its closely related strains was supported by the results of phylogenomic analyses. The strain RG38^T has AAI values ranging from 68.7 to 81.6% with the all reference genomes, which exceeds the AAI criteria for genus delineation (65%). AAI values between strain RG38^T and other species of Streptomyces are shown in Table 1. These results are all below the accepted cutoff values for species delineation, indicating that the strain RG38^T should be considered and representing a new species of the genus Streptomyces. The phylogenomic tree revealed that the strain RG38^T formed an independent cluster with S. collinus NBRC 12759^T, S. iakyrus NRRL ISP-5482^T, S. glaucescens NBRC 12774^T, and S. griseoflavus NRRL B-1830^T with high topology (Figure 1B), demonstrating that the strain RG38^T represents a novel species within the genus Streptomyces. The CheckM result showed that the genome completeness was 99%, and the contamination level was 0.85%. The DNA G + C content was 73.3%. Genomic circular feature map was constructed using CGView server is shown in Figure 2A. The FastANI values between the strain RG38^T and its three reference strains S. collinus NBRC 12759^T, S. iakyrus NRRL ISP-5482^T, and S. glaucescens NBRC 12774^T were 81.4, 82.6, and 82.7%, which are lower than the 95–96% cutoff values previously proposed for species delimitation (Jain et al., 2018). Each red line segment denotes a reciprocal mapping between the strain RG38^T and reference genomes, indicating their evolutionary conserved regions (Figures 2B–D).

Predicted proteins were annotated by blasting the eggNOG database. In total, 6,115 out of 6,289 protein coding genes were classified into 24 functional categories based on the eggnog–Mapper using precomputed cluster and phylogenies from the eggnog database as previously described. The following four top categories (unknown functions were skipped) were classified: transcription (456), carbohydrate transport and metabolism (349), and inorganic ion transport and metabolism (300) (Figure 3). Thirty-five BGCs were detected in the genome of the strain RG38^T. These biosynthetic gene clusters belonged to several cluster categories, noticeably T2PKS, oligosaccharide, phosphonate, linear azol(in)e-containing peptides (LAP), other unspecified ribosomally synthesized and post-translationally modified peptide product (RiPP)-like cluster, terpene, non-ribosomal peptide synthases (NRPS), siderophore, lanthipeptide-i, lanthipeptide-ii, transAT-PKS, betalactone, indole, butyrolactone, linaridin, melanin, ectoine, and lanthipeptide-iv. Interestingly, the genome of the strain RG38^T comprised the gene clusters revealing a low similarity percentage (<20%) to Versipelostatin (5%), Herboxidiene (4%), Vazabitide A (19%), and Caniferolide A/ B/C/ D (4%). Thus, they might be involved in the biosynthesis of new compounds. Therefore, it is implied that the strain RG38^T is a potential species for producing novel competent compounds. Among them, herboxidiene is a compound isolated from Streptomyces chromofuscus strain A7847 and exhibits antitumor activity by suppressing the growth of tumor cells through interfering with the splicing of pre-mRNA coding for cell cycle regulation proteins in our body, and this activity makes herboxitriene a valuable starting point for the development of anticancer drug (Maiti et al., 2020). Moreover, six secondary metabolite biosynthesis gene clusters showed 100% similarities to known biosynthetic gene clusters: T2PKS, oligosaccharide, terpene, transAT-PKS, ectoine, and lantipeptide class iv biosynthetic gene cluster (100%). These gene clusters may be involved in the production of the secondary metabolites in RG38^T. All bio-clusters found in the genomes of the strain RG38^T is presented in Table 2. Gene cluster for geosmin, which is responsible for earthy smell of the strain RG38^T, was also found.

The phenotypic morphology of the culture was monitored after 1 week of incubation at 30°C (Figures 4A,B). Cells of the strain RG38^T are gram-positive, forming black melanin only on ISP2 medium or broth (Figures 4C,D). Strain RG38^T showed sporulation under solid culture conditions similar to those used for other Streptomyces and formed straight hyphae that were 0.45–0.75 μm wide with occasional single ramification on their side. Spores of the strain RG38^T were short cylinders of length 0.5–2.0 μm (Figure 4E). The pigmentation of both aerial and substrate mycelium was influenced by culture medium used (Figure 5). Cells exhibited robust growth across all media utilized in this study (ISP2, BA, NA, TSA, R2A, and LB). Cells grow at 10–45°C (optimum 28–30°C), at pH 5.0–10.0 (optimum, 7.0), and at 0–15% NaCl (w/v) tolerance (optimum, 0%). Strain RG38^T showed growth at 45°C and can tolerate upto 15% NaCl, however its close strains have not shown such characteristics. Cells are positive for catalase and oxidase. Hydrolysis of Tween 20, 40, 80, starch, casein, urease, and esculin was occurred but CM cellulose, chitin, and gelatin were not hydrolyzed. Additional differentiating biochemical characteristics of the strain RG38^T, as compared with its close relatives, are presented in Table 3.

Strain RG38^T showed antimicrobial activity against Staphylococcus aureus ATCC 6538, Bacillus subtilis KACC 16747, Micrococcus luteus KACC 13377, and S. epidermidis KACC 13234. However, antimicrobial activity against X. campestris pv. campestris KACC 10377, Botrytis cinerea KACC 40573, and Pantoea agglomerans KACC 10054 was not found. Test was performed a minimum of three times. The observed antibiotic activity varied depending on the specific growth medium employed. No antifungal activity was detected in the strain RG38^T. Before the extraction of compounds, it is imperative to optimize the growth medium to enhance antibiotic production. Consequently, optimization tests were conducted. Antimicrobial activity was high when the strain was cultivated in NA, LB, TSA, and MA medium (Supplementary Figure S4). Antibiotic production was not found when the strain was cultured in R2A medium. To ascertain the presence of antimicrobial activity in the fermentation broth of the strain RG38^T, we conducted a separate analysis with the resulting supernatant, subsequently employed for inhibition studies against the aforementioned pathogenic bacteria. The fermentation broth of the strain RG38^T showed weak inhibition of S. epidermidis KACC 13234. Furthermore, inhibition of S. aureus ATCC, B. subtilis KACC 16747, and M. luteus KACC 13377 was almost completely absent. This disparity in inhibition profiles between solid and liquid cultures suggests that the strain RG38^T may produce multiple distinct antibiotics that are capable of targeting various microorganisms. Furthermore, the alteration in culture methodology appears to have affected the expression patterns of secondary metabolic pathways in the strain RG38^T. Various strategies were employed in attempts to optimize the fermentation broth for bioactive compound production, such as adjustments to the bacteria-to-medium ratio, modifications in culture duration, temperature, and pH. Unfortunately, none of the culture conditions tested were able to yield bioactive compounds with strong inhibitory effects against the pathogenic bacteria. Therefore, for this study, we opted to utilize the agar extraction method for compound extraction while considering further investigations into optimizing broth culture conditions in future research.

In addition to confirm the expression of same antibiotic production, HPLC profile was created by using different extracts from both liquid and solid cultures, where the crude extract from a solid culture condition showed the antibiotic activity only (Figure 6). Two profiles were significantly different, and the chromomycin derivatives were detected exclusively from the solid culture condition only. Additionally, we compared the HPLC profiles of two crude extracts from different time (September 2023 and January 2024). Both profiles showed that RG38^T consistently produced the chromomycin derivatives.

To investigate the bioactive metabolites produced by RG38^T, fractionation of the major components was performed using semi-preparative HPLC system. As shown in Figure 7, two prominent fractions, F-1 and F-2, were observed from the HPLC chromatogram of RG38^T extract at 254 and 280 nm, respectively. The antimicrobial activities of two obtained yellow color fractions were evaluated against Staphylococcus epidermis KACC 13234 (Figure 8), Staphylococcus aureus ATCC 6538 (not shown), Micrococcus luteus KACC 13377 (not shown), and Bacillus subtilis KACC 16747 (not shown), and both fractions exhibited strong antimicrobial activities for all of them. By comparing the NMR spectra of F-1 and F-2 with the literature, the obtained fractions were identified as chromomycin A2 (1) and A3 (2), respectively (Toume et al., 2014). In this study, the NMR spectra obtained from the fractions exactly matched and aligned with the NMR spectra that were reported in previous articles (Supplementary Figures S5–S8). Detail of chemical shift is presented in Supplementary Table S1. Typically, members of the aureolic acid family, including chromomycin derivatives, show antimicrobial activity against gram-positive bacteria (Cho et al., 2020). Therefore, this study suggested that the antimicrobial activities of the RG38^T extract originated from the chromomycin A2 (1) and A3 (2), which are produced by RG38^T.

In addition, we analyze the secondary metabolites derived from the strain RG38^T to understand its chemical diversity. Thus, the MS/MS molecular network analysis was performed with the crude extract (Figure 8A). The result showed one molecular family in the range of m/z 1,100–1,200. Three red nodes in the family were annotated as aureolic acid derivatives by matching MS/MS fragmentation patterns with the GNPS spectral library. In detail, the metabolites involved in the molecular family shared a same fragment ion of m/z 269 derived from a tricyclic core of aureolic acid family, which supported the annotation results. For example, the molecule ion with m/z of 1167.5 shows a loss of 42 Da from chromomycin derivatives through deacetylation or demethylation process.

Those results were identical to the genomic analysis result of RG38^T. Combining those results, major chromatographic peaks and several peaks with similar UV absorbance patterns and MS/MS fragmentation patterns were tentatively annotated in a scaffold level to describe the chemo diversity of RG38^T-derived metabolites (Figure 8B and Table 4).

Streptomyces tagetis sp. nov. (ta.ge’tis. N.L. gen. n. tagetis, of Tagetes, the plant from which the type strain was isolated).

Strain RG38^T is gram-positive, aerobic, non-motile actinobacterium that forms branched substrate and aerial mycelium which differentiates into short cylindrical shaped spore chains at maturity. The colors of the aerial and substrate mycelium are diverse and depend on the growth medium. Cells grow well on all media that used in this study (ISP2, BA, NA, TSA, R2A, and LB). Strain RG38^T found to grow at 10–45°C (optimum 28–30°C), at pH 4.0–10.0 (optimum, 7.0), and at 0–15% NaCl (w/v) tolerance (optimum, 0%). Cells are positive for catalase and oxidase. The strain hydrolyzes Tween 20, 40, 80, casein, urease, starch, and esculin but does not exhibit hydrolysis of CM cellulose, chitin, and gelatin. When grown in ISP2 solid agar and broth, strain RG38^T produces melanin pigment. This strain also has the ability to reduce nitrate and is positive for indole production, while it is negative for glucose fermentation and arginine dihydrolase activities. Although it hydrolyzes urease and esculin, it does not hydrolyze gelatin. In 20 NE, the assimilation of only β-galactosidase and potassium gluconate was found. The assimilation of rest of the components such as D-mannitol, D-maltose, capric acid, adipic acid, malic acid, trisodium citrate, and phenylacetic acid was not found. In API ZYM, the strain RG38^T was positive for alkaline phosphatase, lipase (C14), trypsin, α-chymotrypsin, acid phosphatase, α-galactosidase, and β-galactosidase but negative for leucine arylamidase, valline arylamidase, cystine arylamidase, β-glucuronidase, α-glucosidase, and β-glucosidase.

The type strain is RG38^T (=KCTC 49624^T = TBRC 15113^T) which was isolated from the roots of marigold plant roots collected in the Ilsan region, South Korea. The DNA G + C content of the type strain is 73.3%.