Taxonomic and Genomic Characterization of Brevibacillus sp. JNUCC 42 from Baengnokdam Crater Lake, Mt. Halla, and Its Cosmeceutical Potential

Abstract

Jeju Island, a volcanic island located off the southern coast of the Korean Peninsula, harbors highly specialized microbial communities shaped by its unique geological and climatic diversity. In particular, Baengnokdam Crater Lake at the summit of Mt. Halla represents an extreme, oligotrophic volcanic habitat characterized by intense UV radiation, temperature fluctuations, and limited nutrients. From this environment, a novel bacterial strain, Brevibacillus sp. JNUCC 42, was isolated and subjected to comprehensive taxonomic, genomic, and biochemical analyses. The strain is a Gram-positive, aerobic, rod-shaped bacterium that grows optimally at 30 °C and pH 7.0–9.0 with moderate NaCl tolerance (≤3%). Phylogenetic analysis based on 16S rRNA gene sequencing and genome-scale GBDP confirmed its affiliation to the genus Brevibacillus, forming a distinct lineage closely related to B. laterosporus DSM 25^T. Whole-genome sequencing generated a 4.93 Mb circular chromosome with a GC content of 40.7%. Comparative genomic analyses revealed ANI (87.1%) and dDDH (32.8%) values far below the species threshold, supporting its delineation as a novel species. Chemotaxonomic data further distinguished JNUCC 42 by its predominance of anteiso-C₁₅:₀ (37.24%) and iso-C₁₅:₀ (27.78%) fatty acids and the presence of a unique unidentified aminolipid not detected in the type strain. Genome mining identified 21 biosynthetic gene clusters, including NRPS, PKS, and NRPS–PKS hybrids, suggesting its potential to produce structurally diverse secondary metabolites. One of these metabolites, the cyclic dipeptide maculosin [cyclo(L-Pro-L-Tyr)], was purified from the culture extract and structurally characterized by NMR spectroscopy. Functional assays demonstrated that maculosin significantly inhibited α-MSH-induced melanogenesis and intracellular tyrosinase activity in B16F10 melanoma cells without cytotoxicity up to 100 µM. Collectively, these findings indicate that Brevibacillus sp. JNUCC 42 represents a novel species within the genus Brevibacillus and a promising microbial source of bioactive compounds with potential cosmeceutical applications.

1. Introduction

Jeju Island, situated off the southern coast of the Korean Peninsula, is a volcanic landmass characterized by a mosaic of geological formations, rich mineral substrates, and diverse microclimates that have fostered the development of highly specialized microbial communities. Among its volcanic features, the summit crater lake of Mt. Halla at approximately 1850 m above sea level—Baengnokdam Crater Lake—stands out as one of the most isolated and pristine ecosystems in Korea. The crater is formed on basaltic–trachytic lava deposits and is exposed to intense ultraviolet radiation, strong diurnal and seasonal temperature fluctuations, and nutrient-limited conditions [1,2,3]. These extreme physicochemical stresses have shaped a distinct extremophilic microbiota that exploits adaptive metabolic strategies suited to low temperature, oligotrophy and mineral-rich habitats. Such volcanic soils therefore present promising reservoirs for the discovery of novel microbial taxa and bioactive compounds with biotechnological potential [4,5,6].

Microorganisms from extreme habitats—including volcanic soils, polar ice, or deep-sea sediments—have proven to be rich sources of structurally diverse secondary metabolites with applications in medicine, cosmetics, and biotechnology [7,8]. In particular, extremophilic bacteria frequently synthesize molecules associated with stress tolerance and cellular protection—such as antioxidants, anti-inflammatory agents, UV-protective compounds, and inhibitors of melanin biosynthesis [9,10]. The growing demand for nature-derived cosmeceutical and dermatological agents further elevates the relevance of such microorganisms. Moreover, cold- and stress-adapted bacteria often secrete enzymes and small molecules that remain stable and active under harsh environmental conditions, making them especially valuable for eco-friendly industrial and pharmaceutical applications [11].

The genus Brevibacillus offers notable relevance in this context. Originally reclassified from a cluster of Bacillus species in 1996, members of the genus are Gram-positive, aerobic, endospore-forming rod-shaped bacteria widely distributed in soil, water, and decaying organic matter [12,13]. They typically display a mesophilic growth profile, motility, and the formation of spores and are increasingly recognized for their functional versatility [13]. Importantly, several Brevibacillus species have been commercially exploited: for example, Brevibacillus brevis has been cultured in large-scale fermenters for the production of antimicrobial peptides and enzymes [14], while Brevibacillus choshinensis is used as a host in heterologous protein-expression systems owing to its capacity for high-yield secretion and low extracellular protease activity [15]. In agriculture, Brevibacillus laterosporus has been applied as a probiotic for crops and livestock, producing antimicrobial peptides and hydrolases that enhance nutrient uptake, suppress pathogens, and modulate gut microbiota in animals [16,17,18]. These examples underscore the genus’s commercial utility across agriculture, bioprocessing, and cosmetics.

Given this background, the volcanic soils of Baengnokdam offer a compelling target for isolation of novel Brevibacillus taxa with distinctive bio-metabolic potential. In the present study, we isolated strain Brevibacillus sp. JNUCC 42 from the soil of Baengnokdam Crater Lake, and conducted comprehensive taxonomic, genomic, and chemotaxonomic characterization. We explore its potential as a source of bioactive compounds for cosmeceutical applications and place our findings within the broader context of Brevibacillus-derived biotechnological innovation.

2. Materials and Methods

2.1. Chemicals and Reagents

Tryptic soy broth (TSB) and Luria–Bertani (LB) medium were obtained from BD Difco (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). α-Melanocyte-stimulating hormone (α-MSH), sodium hydroxide (NaOH), arbutin, L-DOPA, and phosphate-buffered saline (PBS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dimethyl sulfoxide (DMSO) and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were obtained from Biosesang (Seongnam, Gyeonggi-do, Republic of Korea). Dulbecco’s Modified Eagle’s Medium (DMEM) and penicillin–streptomycin (1%) were purchased from Thermo Fisher Scientific (Waltham, MA, USA), while fetal bovine serum (FBS) was supplied by Merck Millipore (Burlington, MA, USA). All reagents were of analytical grade or higher purity and used according to standard laboratory procedures.

2.2. Isolation and Cultivation of the Strain

Strain Brevibacillus sp. JNUCC 42 was obtained from a volcanic-soil sample collected near Baengnokdam Crater Lake (33.3611° N, 126.5356° E) at the summit area of Mt. Halla, Jeju Island, Republic of Korea, in September 2019. Soil sampling was conducted under official permission granted by the Jeju Special Self-Governing Province World Natural Heritage Headquarters. The sample was taken from a designated surface marker point, where approximately 1 g of volcanic soil was collected from a depth of 5–10 cm below the ground surface to minimize recent environmental disturbance and ensure environmental consistency. The collected sample was suspended in 20 mL of sterile physiological saline and vigorously mixed to release microbial cells into suspension. After sedimentation of coarse particles for approximately 30 min, the supernatant was serially diluted (10⁻⁵–10⁻⁹), and aliquots (100 µL) were spread onto Luria–Bertani (LB) agar plates. Distinct colonies that appeared after incubation were repeatedly streaked three to four times to ensure purity, yielding a single isolate designated JNUCC 42. The strain was routinely propagated on LB agar or in LB broth at 30 °C under aerobic conditions, and glycerol stocks (20%, v/v) were prepared for long-term storage at −80 °C. Cells harvested during the late-exponential growth phase were used for genomic DNA extraction. For comparative analyses, the type strain B. laterosporus DSM 25^T was cultivated under the same conditions as JNUCC 42, using LB agar or LB broth at 30 °C under aerobic conditions.

2.3. Phylogenetic and Genomic Analysis

The phylogenetic position of strain Brevibacillus sp. JNUCC 42 was determined using both 16S rRNA gene sequence analysis and whole-genome-based phylogenomics. Genomic DNA was extracted from cells grown in LB broth at 30 °C for 24 h using the Qiagen DNeasy Blood & Tissue Kit (Qiagen, Germany) according to the manufacturer’s protocol. The 16S rRNA gene was amplified using universal bacterial primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′). PCR products were purified, sequenced by Macrogen Inc. (Seoul, Republic of Korea), and assembled using BioEdit v7.2.6. The sequence was compared with available type-strain sequences in the EzBioCloud and NCBI GenBank databases for preliminary identification [19].

A phylogenetic tree was reconstructed using the Genome BLAST Distance Phylogeny (GBDP) method implemented in the Type Strain Genome Server (TYGS) platform [20]. For 16S rRNA-based analysis, evolutionary distances were calculated using BLAST (version 2.14.0) pairwise comparisons, and branch support values were inferred from 100 replications. In addition, a whole-genome-based GBDP tree was generated using intergenomic distances computed from the genome-scale BLAST comparison algorithm [21], providing a more robust taxonomic framework [22]. Representative type strains from closely related genera (Bacillus, Neobacillus, Peribacillus, Paenibacillus, and Allocoprobacillus) were included as outgroup references to validate the phylogenetic position of JNUCC 42.

2.4. Genome Sequencing and Assembly

Genomic DNA of Brevibacillus sp. JNUCC 42 was sequenced using a hybrid platform combining long-read PacBio RS II and short-read Illumina technology (Macrogen Inc., Seoul, Republic of Korea). PacBio libraries were prepared with SMRTbell templates and analyzed through single-molecule real-time (SMRT) sequencing [23], whereas Illumina paired-end libraries were constructed by random fragmentation followed by 5′ and 3′ adapter ligation and sequencing on the HiSeq platform [24].

Raw PacBio subreads were filtered using the PreAssembler Filter v1 (minimum subread length = 500 bp, polymerase read quality ≥ 0.8). The mean subread length after filtering was 10,948 bp (N50 = 17,298 bp) with a total of 131,367 reads corresponding to 1.44 Gb of sequence. Illumina reads (12.7 M raw reads; 1.92 Gb) were trimmed to remove adapters and low-quality bases [25], yielding 7.8 M filtered reads (1.18 Gb) with GC content 40.2%, Q20 = 99.46%, and Q30 = 97.91%.

De novo assembly of PacBio reads was carried out using HGAP v3.0 implemented in SMRT Portal v2.3 [26], followed by polishing with Quiver and Illumina-based error correction using Pilon v1.21 [27]. The assembly workflow included preassembly, seed read selection (minimum 6 kb), error correction, contig construction, and circularization checks. Post-assembly, Illumina reads were mapped back to the consensus genome for further correction and validation.

2.5. Assembly Validation and Annotation

Genome completeness and quality were evaluated by k-mer analysis (Jellyfish v2.2.10 and GenomeScope 2.0) [28,29], read mapping statistics, and BUSCO v3.0 assessment using the bacteria_odb9 database (148 single-copy orthologs) [30]. Sequence similarity searches were performed with BLASTN v2.7.1+ against the NCBI NT database to determine phylogenetic affiliation [21].

Functional genome annotation was conducted using Prokka v1.12b [31], identifying coding DNA sequences (CDS), tRNA, and rRNA genes. Circular genome maps were generated from the annotation results, showing forward/reverse CDS, tRNA, rRNA loci, GC content, and GC skew [32].

2.6. Physiological, Biochemical, and Chemotaxonomic Characterization

The physiological, biochemical, and chemotaxonomic characteristics of Brevibacillus sp. JNUCC 42 were determined in comparison with the nearest relative strain Brevibacillus laterosporus DSM 25^T. For the evaluation of growth characteristics, both strains were cultivated in tryptic soy broth (TSB) under various environmental conditions, including different pH levels (4.0–10.0), NaCl concentrations (1–20%, w/v), and temperatures (10–42 °C). After 24 h of incubation, the growth intensity was quantified by measuring the optical density at 600 nm (OD₆₀₀) and categorized as follows: +++ (OD₆₀₀ ≥ 0.50), ++ (0.20 ≤ OD₆₀₀ < 0.50), + (0.05 ≤ OD₆₀₀ < 0.20), and – (OD₆₀₀ < 0.05).

Carbohydrate utilization profiles and enzymatic activities were assessed using the API 50CHB and API ZYM test systems (bioMérieux, Marcy-l’Étoile, France) according to the manufacturer’s instructions. Reaction outcomes were visually evaluated after incubation at 30 °C for 24 h, and color development was interpreted following the standard API reference chart.

For cellular fatty acid composition, cells cultivated on TSB agar plates at 30 °C for 24 h were harvested and analyzed using the MIDI Sherlock Microbial Identification System (version 6.0). The extracted fatty acids were converted to methyl esters (FAMEs), separated by gas chromatography (GC), and identified based on equivalent chain length values and the Sherlock peak library. Relative abundances of individual fatty acids were expressed as percentages of total fatty acids, which were classified into straight-chain, branched-chain, or unsaturated types.

For polar lipid analysis, total lipids were extracted using a chloroform–methanol–water mixture (1:2:0.8, v/v/v) and subjected to two-dimensional thin-layer chromatography (TLC) on silica gel 60 plates (Merck, Darmstadt, Germany). The plates were developed in chloroform–methanol–water (65:25:4, v/v/v) for the first dimension and chloroform–acetic acid–methanol–water (80:12:15:4, v/v/v) for the second dimension. Lipid spots were visualized with 10% molybdophosphoric acid (for total lipids) and ninhydrin reagent (for detection of aminolipids and phospholipids). Major polar lipid classes were identified by their migration patterns and staining responses.

2.7. Digital DNA–DNA Hybridization (dDDH) Analysis Between Brevibacillus sp. JNUCC 42 and Related Species

Digital DNA–DNA hybridization (dDDH) values were calculated to evaluate the genomic relatedness between strain Brevibacillus sp. JNUCC 42 and its phylogenetically related type strains within the genus Brevibacillus. The analysis was performed using the Genome-to-Genome Distance Calculator (GGDC) version 3.0 available at the DSMZ web server (https://ggdc.dsmz.de, accessed on 11 August 2025) [33,34] based on pairwise genome sequence comparisons.

The GGDC computes three independent dDDH estimates—d₀ (HSP length/total length), d₄ (sum of identities/HSP length), and d₆ (sum of identities/total length)—each providing a statistical confidence interval (C.I.) that reflects the degree of precision in estimating genomic similarity [33]. The recommended formula d₄, which exhibits the highest correlation with wet-lab DDH values, was primarily considered to determine genomic relatedness [33,35].

Additionally, the G+C content difference between JNUCC 42 and reference genomes was calculated from whole-genome sequences to supplement the genomic delineation [36]. Species demarcation thresholds were applied according to accepted taxonomic standards—namely, 70% for DDH and 1% for G+C content difference—to evaluate whether JNUCC 42 represents a distinct genomic species [35,36].

2.8. Average Nucleotide Identity (ANI) Analysis Between Brevibacillus sp. JNUCC 42 and Brevibacillus laterosporus DSM 25

To further evaluate the genomic relatedness between Brevibacillus sp. JNUCC 42 and its closest phylogenetic relative Brevibacillus laterosporus DSM 25, the Average Nucleotide Identity (ANI) was calculated using the OrthoANIu algorithm [37] implemented in the CJ Bioscience OrthoANIu web service (https://www.ezbiocloud.net/tools/ani, accessed on 11 August 2025).

The OrthoANIu method computes pairwise nucleotide identity between orthologous genomic regions by fragmenting the genome sequences and performing reciprocal BLAST-based alignment [37,38]. The resulting ANI value represents the average nucleotide identity across all shared genomic segments, and a species boundary threshold of 95–96% ANI is generally accepted for prokaryotic species delineation [39,40].

The analysis used the assembled genome of Brevibacillus sp. JNUCC 42 (Genome A, 4,924,560 bp) and B. laterosporus DSM 25 (Genome B, 5,399,880 bp). The parameters retrieved from the OrthoANIu output included the overall ANI value, average aligned length, and coverage percentages for both genomes.

2.9. In Silico Prediction of Secondary Metabolite Biosynthetic Gene Clusters in Brevibacillus sp. JNUCC 42

The complete genome sequence of Brevibacillus sp. JNUCC 42 was analyzed to identify biosynthetic gene clusters (BGCs) responsible for the production of secondary metabolites. Genome mining was performed using antiSMASH version 8.0 (https://antismash.secondarymetabolites.org, accessed on 11 August 2025) [41], applying default parameters for detection of modular systems such as nonribosomal peptide synthetases (NRPS), polyketide synthases (PKS Types I, II, and III), hybrid PKS–NRPS clusters, as well as ribosomally synthesized and post-translationally modified peptides (RiPPs) and terpene biosynthetic pathways.

The predicted clusters were annotated based on sequence similarity to known BGCs deposited in the Minimum Information about a Biosynthetic Gene cluster (MIBiG) database [42], and their confidence levels (low, moderate, or high) were evaluated automatically by antiSMASH. The domain architecture and module organization of NRPS and PKS clusters were manually curated using antiSMASH output files and visualized according to catalytic domain composition—condensation (C), adenylation (A), peptidyl carrier protein (PCP/T), epimerization (E), ketosynthase (KS), acyltransferase (AT), and thioesterase (TE).

The substrate specificities of NRPS adenylation domains were inferred from the built-in NRPSpredictor2 tool [43] integrated within antiSMASH, while hybrid PKS/NRPS modules were further validated using NaPDoS2 [44].

2.10. Morphological Characterization of Brevibacillus sp. JNUCC 42

The cellular morphology of strain Brevibacillus sp. JNUCC 42 was examined using a high-resolution field emission scanning electron microscope (FE-SEM; Regulus 8100, Hitachi, Tokyo, Japan). The strain was cultivated on Luria–Bertani (LB) agar plates at 30 °C for 24 h. A loopful of cells was gently collected and fixed in 2.5% (v/v) glutaraldehyde prepared in 0.1 M phosphate buffer (pH 7.2) for 2 h at 4 °C.

The fixed samples were washed three times with the same buffer and sequentially dehydrated through a graded ethanol series (30%, 50%, 70%, 90%, and 100%, 10 min each). Following critical-point drying, the samples were mounted on aluminum stubs and sputter-coated with platinum (~10 nm thickness) to enhance conductivity. The samples were observed under an accelerating voltage of 5.0 kV and a working distance of 8.1 mm. Representative images were obtained at magnifications ranging from ×9000 to ×25,000.

2.11. Isolation, Purification, and Structural Identification of the Bioactive Compound from Brevibacillus sp. JNUCC 42

Brevibacillus sp. JNUCC 42 was cultivated in 500 mL of LB broth at 30 °C for 72 h with continuous agitation at 150 rpm. After incubation, the culture broth was extracted twice with an equal volume of ethyl acetate (EtOAc), and the combined organic layers were concentrated under reduced pressure to obtain a crude extract (160 mg). The EtOAc extract was subsequently subjected to column chromatography on Sephadex LH-20 using 100% methanol as the eluent, yielding six fractions (Fr. 1–Fr. 6). Each fraction was monitored by analytical high-performance liquid chromatography (HPLC) on a Waters Alliance e2695 system equipped with a 2998 PDA detector and Empower software (version 3.8.0, Waters, Milford, MA, USA). Chromatographic analyses were conducted for 60 min at a detection wavelength of 276 nm under the conditions summarized in Supplementary Table S1 and Supplementary Figures S2 and S3. Among the obtained fractions, Fr. 3 displayed a single major peak at 32.7 min and was designated as Compound 1 for further purification and structural characterization. Purity confirmation of Compound 1 was carried out under identical chromatographic conditions, revealing a single sharp peak at 32.747 min, thereby confirming high chemical purity. The UV absorption spectrum exhibited a maximum at 276 nm, consistent with aromatic dipeptide derivatives. Structural elucidation was performed by nuclear magnetic resonance (NMR) spectroscopy. ¹H- and ¹³C-NMR spectra were recorded on JEOL JNM-LA 400 and JNM-ECX 400 FT-NMR spectrometers (JEOL Ltd., Tokyo, Japan) operating at 400 MHz for ¹H and 100 MHz for ¹³C, respectively. Deuterated methanol (CD₃OD; Cambridge Isotope Laboratories, Tewksbury, MA, USA) served as the solvent, and chemical shifts (δ) were expressed in ppm with coupling constants (J) in Hz. Spectral data obtained from both ¹H and ¹³C NMR analyses were compared with published reference data to confirm the molecular identity of the isolated compound.

2.12. Assessment of Anti-Melanogenic Activity of Maculosin in B16F10 Melanoma Cells

The inhibitory effect of Maculosin on melanogenesis was evaluated using murine melanoma B16F10 cells. Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin at 37 °C under a humidified atmosphere of 5% CO₂. For cytotoxicity assessment, B16F10 cells were seeded at 6 × 10⁴ cells per well in 96-well plates and incubated for 24 h. Various concentrations of maculosin (6.25–400 μM) were applied for 72 h, followed by the addition of 0.5 mg/mL MTT solution for 4 h. The formazan crystals were dissolved in DMSO, and absorbance was measured at 540 nm using a microplate reader (Epoch, BioTek Instruments, Winooski, VT, USA). Cell viability was expressed as a percentage of the untreated control.

To assess the effect of Maculosin on melanin synthesis, cells were seeded in 6-well plates and pre-stimulated with 100 nM α-MSH for 24 h, followed by treatment with Maculosin (25, 50, and 100 μM) for 72 h. Arbutin (300 μM) served as a positive control. After incubation, cells were lysed with 1 N NaOH containing 10% DMSO and incubated at 80 °C for 1 h to solubilize melanin. Absorbance was measured at 405 nm, and melanin levels were expressed as a percentage relative to α-MSH-treated controls.

For tyrosinase activity, cells were lysed in phosphate-buffered saline (PBS) containing 1% Triton X-100, and 90 μL of lysate was incubated with 10 μL of 0.1% L-DOPA solution at 37 °C for 1 h. The formation of dopachrome was monitored by absorbance at 475 nm, and tyrosinase activity was calculated relative to control values.

2.13. Statistical Analyses

All statistical analyses were conducted using IBM SPSS Statistics software (version 20; SPSS Inc., Armonk, NY, USA). Differences between groups were examined using either Student’s t-test or one-way analysis of variance (ANOVA), depending on the experimental design. Data are presented as the mean ± standard deviation (SD), obtained from at least three independent experiments or from triplicate measurements within a single experiment. Statistical significance was considered at ** p < 0.001 when compared with the control group.

3. Results

3.1. Phylogenetic Analysis

The 16S rRNA gene sequence of strain Brevibacillus sp. JNUCC 42 (1475 bp) exhibited 99.2% similarity to Brevibacillus laterosporus DSM 25^T and 98.8% to Brevibacillus daliensis YIM B02290^T, indicating that the isolate belongs to the genus Brevibacillus and represents a distinct species lineage. In the 16S rRNA-based GBDP tree, JNUCC 42 clustered within the Brevibacillus clade, forming a separate sublineage closely associated with B. laterosporus and B. daliensis (Figure S1).

Whole-genome phylogenetic analysis further supported these findings. In the genome-based GBDP tree, strain JNUCC 42 again grouped with B. laterosporus DSM 25^T but formed a distinct and independent branch, separated from other Brevibacillus type strains by clear intergenomic distance (Figure S2). The tree topology, supported by high branch confidence values, confirmed that JNUCC 42 occupies a unique taxonomic position within the genus.

3.2. Taxonomic Characterization of Brevibacillus sp. JNUCC 42

Brevibacillus sp. JNUCC 42 exhibited distinctive physiological and chemotaxonomic features when compared with the nearest relative strain B. laterosporus DSM 25^T. The strain grew optimally at 30 °C and within a pH range of 7.0–9.0, showing moderate tolerance to NaCl concentrations up to 3% (w/v), whereas the type strain failed to grow at pH 9.0 and displayed lower salt tolerance. No growth was detected for JNUCC 42 below 20 °C or above 37 °C, indicating that it is adapted to a mesophilic and mildly alkaline environment typical of its volcanic soil origin (

Table 1. Growth characteristics of Brevibacillus sp. JNUCC 42 and type strain B. laterosporus DSM 25^T under different environmental conditions.

Characteristic	(a) Brevibacillus sp. JNUCC 42	(b) B. laterosporus DSM 25T
pH range
pH 4.0~6.0	+	–
pH 7.0	+++	++
pH 8.0	+++	+
pH 9.0	+++	–
pH 10.0	+	–
NaCl tolerance
1%	+++	++
3%	++	+
5~20%	–	–
Temperature range
10~20 °C	–	–
30 °C	++	++
37 °C	–	+
40 °C	–	–

The table summarizes the temperature-dependent growth of Brevibacillus sp. JNUCC 42 and the type strain B. laterosporus DSM 25^T across 10–42 °C. Growth was assessed after 24 h in tryptic soy broth by OD₆₀₀. Categories were defined as follows: +++ (OD₆₀₀ ≥ 0.50), ++ (0.20 ≤ OD₆₀₀ < 0.50), + (0.05 ≤ OD₆₀₀ < 0.20), and – (OD₆₀₀ < 0.05). Brevibacillus sp. JNUCC 42 showed optimal growth at 30 °C, with no detectable growth at temperatures ≤20 °C or ≥37 °C.

). In the API 50CHB and API ZYM assays, both strains utilized D-glucose and esculin and exhibited activities of alkaline phosphatase and naphthol-AS-BI-phosphohydrolase, while only the nearest relative strain metabolized D-fructose. All other enzyme reactions were negative, revealing partial but distinct biochemical differentiation between the two taxa (

Table 2. Differential API ZYM enzymatic profiles between Brevibacillus sp. JNUCC 42 and the type strain B. laterosporus DSM 25^T.

Assay	(a) Brevibacillus sp. JNUCC 42	(b) B. laterosporus DSM 25T	Diagnostic Value
API 50CHB
D-Glucose	+	+	Shared trait
D-Fructose	–	+	B. laterosporus DSM 25 specific
Esculine	+	+	Shared trait
API ZYM:
Alkine phosphatase	+	+	Shared trait
Naphtol-AS-BI-phosphohydrolase	+	+	Shared trait

The table summarizes phenotypic characteristics of Brevibacillus sp. JNUCC 42 and the type strain B. laterosporus DSM 25^T as determined by the API 50CHB (carbohydrate utilization) and API ZYM (enzyme activity) systems. Both strains were positive for D-glucose, D-fructose, and esculin utilization in the API 50CHB assay, and both exhibited activities for alkaline phosphatase and naphthol-AS-BI-phosphohydrolase in the API ZYM assay. All other listed reactions were negative in both strains and therefore non-differentiating. Note: “+” = positive; “–” = negative.

). The cellular fatty-acid profile of JNUCC 42 was dominated by anteiso-C₁₅:₀ (37.24%), iso-C₁₅:₀ (27.78%), and iso-C₁₆:₀ (4.21%), representing the typical branched-chain pattern characteristic of the genus Brevibacillus. In addition, minor unsaturated components—C₁₅:₁ ω5c (2.27%) and C₁₆:₁ ω11c (1.40%)—were detected in JNUCC 42 but absent in B. laterosporus DSM 25^T, suggesting species-level differentiation. Trace amounts (<0.1%) of C₁₂:₀, C₁₃:₀, and C₁₄:₀ 2-OH were also observed only in the isolate. The predominance of branched-chain fatty acids confirmed its taxonomic placement within Brevibacillus, whereas the unique presence of unsaturated components supported its genomic distinctness (

Table 3. Cellular fatty acid contents of Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25^T.

Fatty Acid (%)	(a) Brevibacillus sp. JNUCC 42	(b) B. laterosporus DSM 25T
Straight-chain
12:0	Tr	–
13:0	Tr	–
14:0 2OH	Tr	–
Branched
15:0 iso	27.78	35.39
15:0 anteiso	37.24	38.62
16:0 iso	4.21	4.46
15:1 anteiso A	Tr	–
15:0 iso 3OH	Tr	–
Unsaturated
14:1 ω5c	Tr	–
15:1 ω5c	2.27	–
16:1 ω11c	1.40	–
18:1 ω9c	Tr	–

Cellular fatty acid compositions of Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25^T. Values are presented as percentages of total fatty acids. Tr indicates a trace amount (<0.1%), and – indicates that the component was not detected. Fatty acids are grouped into straight-chain, branched-chain, and unsaturated categories.

). Polar-lipid analysis showed that both JNUCC 42 and the type strain shared phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylcholine (PC), unidentified aminophospholipid (APL), unidentified phospholipid (PL), and unidentified lipid (L) as major constituents; however, JNUCC 42 uniquely contained an unidentified aminolipid (AL) that was absent in B. laterosporus DSM 25^T (

Table 4. Polar Lipid Profile Comparison between Brevibacillus sp. JNUCC 42 and the type strain B. laterosporus DSM 25^T.

Lipid Class	(a) Brevibacillus sp. JNUCC 42	(b) B. laterosporus DSM 25T	Diagnostic Value
Phosphatidylethanolamine (PE)	+	+	Shared trait
Phosphatidylglycerol (PG)	+	+	Shared trait
Phosphatidylcholine (PC)	+	+	Shared trait
Unidentified aminophospholipid (APL)	+	+	Shared trait
Unidentified aminolipid (AL)	+	–	Brevibacillus sp. JNUCC 42 specific
Unidentified phospholipid (PL)	+	+	Shared trait
Unidentified lipid (L)	+	+	Shared trait

Polar lipid analysis of Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25^T revealed six components—phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylcholine (PC), unidentified aminophospholipid (APL), unidentified phospholipid (PL), and unidentified lipid (L). In contrast, Brevibacillus sp. JNUCC 42 showed an unidentified aminolipid (AL). Results are based on TLC with molybdophosphoric acid and ninhydrin staining, and the presence or absence of each lipid is indicated as “+” (present) and “–” (absent).

). The distinct AL spot was further verified through two-dimensional TLC, where a ninhydrin-positive signal corresponding to this aminolipid was detected exclusively in JNUCC 42, confirming its strain-specific chemotaxonomic feature (Table 4; Supplementary Figure S3). This distinctive aminolipid, together with the characteristic unsaturated fatty acids and differing biochemical traits, provides compelling evidence that strain JNUCC 42 represents a novel taxon within the genus Brevibacillus.

3.3. Morphological Features Under Scanning Electron Microscopy

FE-SEM analysis revealed that Brevibacillus sp. JNUCC 42 exhibited typical rod-shaped morphology characteristic of the genus Brevibacillus. The cells were straight to slightly curved rods, occurring singly or in short chains, with smooth surfaces and rounded ends (Figure 1).

At higher magnification (Figure 1a, ×25,000), individual cells appeared with a uniform rod structure, approximately 0.6–0.8 μm in width and 2.0–4.5 μm in length. At moderate magnification (Figure 1b, ×15,000), the cells were often observed in pairs or small groups, occasionally aligned in parallel arrangements. At lower magnification (Figure 1c, ×9000), dense cellular aggregation was evident, likely due to surface adherence or partial sporulation.

No visible flagella or spore structures were clearly detected under the tested growth conditions, although surface irregularities were observed, possibly related to cell wall maturation. These morphological characteristics are consistent with members of the genus Brevibacillus, supporting its taxonomic affiliation inferred from 16S rRNA and genomic analyses.

3.4. Genome Assembly and Annotation of Brevibacillus sp. JNUCC 42

De novo assembly of Brevibacillus sp. JNUCC 42 generated three contigs with a total length of 4,942,230 bp and an average GC content of 40.73% (

Table 5. General Genomic Features of Brevibacillus sp. strain JNUCC 42.

Brevibacillus sp. Strain JNUCC 42
Genome size (bp)	4,925,472
Total number of contigs	1
Contigs N50 (bp)	4,925,472
Plasmid	2
G+C content (%)	40.5
Genome coverage	197×
Number of chromosomes	1
Total number of predicted genes	4674
Total number of protein coding genes	4272
Total number of pseudogenes	250
Total number of tRNA-coding genes	111
Total number of rRNA-coding genes (5S, 16S, 23S)	12, 12, 12
Total number of ncRNA-coding genes	5

The table summarizes the genome assembly statistics and gene annotation features of strain JNUCC 42. The genome consists of a single chromosome and two plasmids, with a total size of 4.93 Mbp. Gene prediction identified 4674 genes, including 4272 protein-coding sequences, 250 pseudogenes, 111 tRNA genes, and 36 rRNA genes (12 copies each of 5S, 16S, and 23S rRNA).

). The largest contig (Contig 1; 4,925,472 bp) and Contig 2 (10,318 bp) were circular, whereas Contig 3 (6440 bp) remained linear. The mean sequencing depth across the assembly was approximately 197×, and after Pilon-based correction, both N50 and the maximum contig length reached 4,925,472 bp, indicating high continuity and accuracy of the assembly. K-mer (21-mer) analysis estimated a genome size of 5.29 Mb with a heterozygosity rate of 0.009 and a repeat length of 689 kb, consistent with the assembled size. Mapping of 7.8 million Illumina reads to the PacBio-based assembly showed 99.88% alignment and 100% genome coverage with an average depth of 228×, confirming the accuracy and completeness of the assembly. BUSCO evaluation detected 148 out of 148 (100%) complete single-copy orthologs, verifying the overall completeness of the genome. Annotation identified 4602 protein-coding sequences (CDS), 111 tRNA genes, and 36 rRNA genes across the three contigs (Table 5). The major chromosome (Contig 1) contained 4586 CDS and all rRNA and tRNA genes, while Contig 2 and Contig 3 encoded nine and seven CDS, respectively. The circular genome map (Figure S4) revealed a balanced GC content and typical prokaryotic genomic organization. The complete genome of Brevibacillus sp. JNUCC 41 therefore consists of a 4.94 Mb chromosome with a GC content of 40.7% and a total of 4602 genes, exhibiting high assembly quality (BUSCO = 100%, Q30 > 97%) and sufficient sequence depth to support downstream comparative genomic and biosynthetic gene cluster analyses.

3.5. Characterization of Brevibacillus sp. JNUCC 42 as a Novel Strain via dDDH Analysis

The digital DNA–DNA hybridization (dDDH) results revealed that Brevibacillus sp. JNUCC 42 shared the highest genomic relatedness with B. laterosporus DSM 25, showing dDDH values of 52.5% (d₀), 32.8% (d₄), and 47.2% (d₆) (

Table 6. Digital DNA–DNA hybridization (dDDH) values between Brevibacillus sp. JNUCC 42 and related Brevibacillus species.

Subject Strain	dDDH (d0, in %)	C.I. (d0, in %)	dDDH (d4, in %)	C.I. (d4, in %)	dDDH (d6, in %)	C.I. (d6, in %)	G+C Content Difference (in %)
Brevibacillus laterosporus DSM 25	52.5	[49.0–55.9]	32.8	[30.4–35.3]	47.2	[44.2–50.2]	0.23
Brevibacillus schisleri ATCC 35690	13	[10.3–16.3]	32.1	[29.7–34.6]	13.4	[11.1–16.2]	6.53
Brevibacillus gelatini DSM 100115	12.8	[10.1–16.1]	30.2	[27.8–32.7]	13.2	[10.9–16.0]	11.25
Brevibacillus formosus DSM 9885	12.8	[10.1–16.1]	30	[27.6–32.5]	13.2	[10.9–16.0]	6.67
Brevibacillus panacihumi JCM 15085	12.8	[10.1–16.0]	30	[27.6–32.5]	13.2	[10.8–15.9]	10.24
Bacillus timonensis MM10403188	12.6	[9.9–15.9]	29.6	[27.2–32.1]	13	[10.7–15.8]	3.5
Brevibacillus invocatus JCM 12215	12.9	[10.2–16.2]	29.4	[27.0–31.9]	13.3	[10.9–16.0]	8.18
Peribacillus huizhouensis DSM 105481	12.7	[10.0–15.9]	29.3	[27.0–31.8]	13.1	[10.7–15.8]	3.51
Brevibacillus parabrevis NRRL NRS-605	12.8	[10.1–16.1]	28.7	[26.3–31.2]	13.2	[10.8–16.0]	11.42
Bacillus cihuensis FJAT-14515T	12.6	[9.9–15.8]	28.6	[26.2–31.1]	13	[10.7–15.7]	3.7
Brevibacillus fortis NRRL NRS-1210	12.9	[10.2–16.2]	28.1	[25.7–30.6]	13.3	[10.9–16.1]	6.4
Bacillus rhizoplanae CIP 111899T	12.7	[10.0–16.0]	27.7	[25.4–30.2]	13.1	[10.8–15.9]	4.36
Paenibacillus roseopurpureus MBLB1832	12.7	[10.0–16.0]	27	[24.7–29.5]	13.1	[10.8–15.9]	6.12
Brevibacillus fluminis JCM 15716	12.9	[10.2–16.1]	25.1	[22.8–27.6]	13.3	[10.9–16.0]	9.61
Neobacillus kokaensis ATCC 31382	12.7	[10.0–15.9]	23.8	[21.5–26.3]	13	[10.7–15.8]	1.23
Brevibacillus dissolubilis CHY01	13.1	[10.4–16.4]	22.6	[20.3–25.0]	13.4	[11.1–16.2]	9.81
Allocoprobacillus halotolerans LH1062	12.7	[10.0–15.9]	21.4	[19.2–23.9]	13.1	[10.7–15.8]	9.49
Brevibacillus daliensis YIM B02290	14.9	[12.0–18.3]	21.1	[18.9–23.5]	15.1	[12.6–17.9]	0.14

Pairwise comparisons were performed using the Genome-to-Genome Distance Calculator (GGDC) with three formulas (d0, d4, d6). Confidence intervals (C.I.) are provided for each comparison.

). These values are below the 70% species boundary, suggesting that strain JNUCC 42 is distinct from B. laterosporus at the genomic level.

All other Brevibacillus species exhibited markedly lower relatedness, with d₄ values ranging from 21.4% to 32.1% and corresponding confidence intervals overlapping within a similar low range. The G+C content difference between JNUCC 42 and related taxa varied from 0.23% to 11.42%, further supporting its taxonomic distinctiveness (Table 6).

Notably, strains such as B. formosus DSM 9885^T, B. gelatini DSM 100115^T, and B. panacihumi JCM 15085^T showed d₄ values around 30%, indicating moderate genomic proximity but still well below the interspecies threshold. In contrast, Bacillus timonensis and Peribacillus huizhouensis displayed lower genomic identity (<30%), confirming that these taxa are only distantly related.

Overall, the dDDH and G+C content analyses jointly demonstrate that Brevibacillus sp. JNUCC 42 represents a genomically distinct lineage within the genus Brevibacillus, corroborating the results of ANI and phylogenomic analyses.

3.6. Characterization of Brevibacillus sp. JNUCC 42 as a Novel Strain via ANI Analysis

The OrthoANIu analysis revealed an average nucleotide identity of 87.10% between Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25 (

Table 7. Average Nucleotide Identity (ANI) values between Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25 calculated using OrthoANIu.

Metric	Value
OrthoANIu value (%)	87.10%
Genome A length (bp)	4,924,560
Genome B length (bp)	5,399,880
Average aligned length (bp)	2,587,651
Genome A coverage (%)	52.55
Genome B coverage (%)	47.92

). The average aligned length between the two genomes was 2,587,651 bp, corresponding to 52.55% coverage of the JNUCC 42 genome and 47.92% coverage of the DSM 25 genome, respectively.

These results indicate that less than half of the total genomic content is homologous between the two strains, and the observed ANI value (87.10%) is substantially below the 95–96% cutoff typically used for species-level classification. Therefore, Brevibacillus sp. JNUCC 42 can be considered a genomically distinct taxon within the genus Brevibacillus (Table 7).

This genomic divergence corroborates the results of the digital DNA–DNA hybridization (dDDH) analysis, which also showed a low relatedness (32.8% by formula d₄) between the same strains, providing further evidence that JNUCC 42 represents a novel genomic lineage.

3.7. In Silico Prediction of Secondary Metabolites and Biosynthetic Gene Cluster Analysis

Comprehensive genome mining of Brevibacillus sp. JNUCC 42 using antiSMASH 8.0 revealed 21 putative biosynthetic gene clusters (BGCs) distributed across the 4.92 Mb genome (

Table 8. Biosynthetic gene clusters (BGCs) predicted in the genome of Brevibacillus sp. JNUCC 42 using antiSMASH.

Region	Type	From	To	Similarity Confidence	Most Similar Known Cluster	BGC Class
1	NRPS-like	21,621	63,033	High	Kolossin	NRPS: Type I
2	NRPS, Betalactone	77,952	144,990	Low	Pelgipeptin A/B/C/D	NRPS: Type I
3	Cyclic-lactone-autoinducer	643,636	664,148	–	–	–
4	T3PKS	875,138	916,193	–	–	–
5	NRPS, TransAT-PKS-like	1,079,447	1,175,950	Low	Pelgipeptin A/B/C/D	NRPS: Type I
6	NRPS, NRPS-like	1,250,225	1,421,338	Low	Tyrocidine	NRPS: Type I
7	NRPS, T1PKS, TransAT-PKS	1,481,493	1,587,122	Low	Basiliskamide A/B	PKS
8	NRPS	1,748,344	1,801,170	–	–	–
9	NRPS-like, Terpene-precursor	1,821,578	1,872,358	–	–	–
10	NRPS, T1PKS	1,895,712	1,963,165	Low	Zwittermicin A	NRPS: Type I + PKS
11	Terpene	1,965,095	1,986,993	–	–	–
12	Cyclic-lactone-autoinducer	2,270,172	2,290,830	–	–	–
13	NRPS, NRPS-like, lanthipeptide-class-I	2,669,622	2,764,544	High	Bogorol A	NRPS: Type I
14	NIS-siderophore	2,830,298	2,861,989	High	Petrobactin	Other: Other
15	Cyclic-lactone-autoinducer	2,947,446	2,968,092	–	–	–
16	Cyclic-lactone-autoinducer	3,549,524	3,570,170	–	–	–
17	NRPS	4,039,413	4,085,196	–	–	–
18	NRPS	4,228,982	4,294,481	–	–	–
19	Azole-containing RiPP	4,543,398	4,566,944	–	–	–
20	RiPP-like	4,739,197	4,750,937	–	–	–
21	NRPS	4,865,946	4,913,580	–	–	–

). These clusters collectively encode a diverse array of secondary-metabolite pathways, including nonribosomal peptide synthetases (NRPSs), polyketide synthases (PKSs), hybrid PKS/NRPS systems, ribosomally synthesized and post-translationally modified peptides (RiPPs), siderophores, and terpenes. Among them, several clusters exhibited high sequence similarity to known bioactive metabolites, suggesting the strain’s potential for producing pharmacologically and cosmetically relevant compounds. Notably, an NRPS-like cluster (Region 1) showed strong homology to Kolossin, while Region 13 displayed high similarity to Bogorol A, a cyclic peptide antibiotic reported from Brevibacillus species. A siderophore-related BGC (Region 14) matched Petrobactin, indicating possible iron-chelating capacity. In contrast, clusters such as Regions 2, 5, 6, 7, and 10 exhibited low similarity scores, suggesting the presence of novel or cryptic metabolites. Region 10, a hybrid NRPS–PKS system, was partially homologous to the Zwittermicin A biosynthetic pathway, hinting at antibacterial potential.

Detailed domain analysis of the NRPS clusters (Regions 2, 6, 8, 13, 17, and 18) revealed the presence of canonical catalytic modules—condensation (C), adenylation (A), peptidyl carrier protein (PCP/T), epimerase (E), ketoreductase (KR), and thioesterase (TE)—that together form multimodular assembly lines (

Table 9. Domain architecture and substrate specificity of NRPS clusters identified in Brevibacillus sp. JNUCC 42.

Region
2
6
8
13
17
18

Each colored arrow represents an ORF, and circles indicate catalytic domains within NRPS modules, including condensation (C), adenylation (A), peptidyl carrier protein (PCP/T), epimerization (E), ketoreductase (KR), and thioesterase (TE). Predicted amino acids incorporated by each module are shown below (e.g., D-Leu, Thr, Pro), whereas “X” denotes an inactive or unassigned adenylation domain whose substrate could not be reliably predicted by antiSMASH 8.0.

). Substrate predictions based on the A-domain specificity codes suggested incorporation of various amino-acid residues such as D-Lys, Thr, Val, Leu, and Orn, while several adenylation domains marked with “X” were inactive or unassigned, implying incomplete or cryptic biosynthetic modules.

Hybrid PKS/NRPS clusters (Regions 5, 7, and 10) were composed of typical PKS core domains—ketosynthase (KS), acyltransferase (AT), dehydratase (DH), ketoreductase (KR)—coupled with NRPS modules harboring condensation and adenylation domains (

Table 10. Domain organization of hybrid PKS/NRPS BGCs in Brevibacillus sp. JNUCC 42.

Region
5
7
10

Each ORF is represented by a black arrow, and circles indicate catalytic domains within the PKS or NRPS modules. Domain abbreviations: KS, ketosynthase; AT, acyltransferase; DH, dehydratase; KR, ketoreductase; C, condensation; A, adenylation; CP (or T), carrier protein; E, epimerase; TE, thioesterase; CMT, methyltransferase; AmT, amidotransferase; ER, enoylreductase. Crossed-out (×) domains indicate catalytically inactive or truncated domains predicted by antiSMASH 8.0.

). Crossed-out (×) domains indicated catalytically inactive or truncated motifs, while fully colored circles denoted active catalytic centers predicted by antiSMASH. The architecture of these hybrid clusters suggests the genetic capacity for polyketide–peptide hybrid scaffolds, a hallmark of compounds with antimicrobial, surface-active, or cytoprotective properties. Taken together, the in silico analysis highlights Brevibacillus sp. JNUCC 42 as a genetically versatile producer of diverse secondary metabolites, harboring both conserved and potentially novel biosynthetic pathways that warrant further chemical and functional characterization.

3.8. Isolation and Structural Identification of Maculosin [Cyclo(L-Pro-L-Tyr)] from Brevibacillus sp. JNUCC 42

The culture broth of Brevibacillus sp. JNUCC 42 was extracted with ethyl acetate, yielding a 160 mg crude extract that exhibited distinct UV-absorbing components during HPLC analysis. The chromatogram recorded at 276 nm displayed a major peak at a retention time of 32.7 min, indicating the presence of a dominant secondary metabolite (Figure S5). The ethyl acetate extract was subjected to Sephadex LH-20 column chromatography using 100% methanol as the eluent, resulting in six fractions (Fr. 1–Fr. 6). Among them, Fraction 3 contained a single major compound, designated as Compound 1. HPLC re-analysis confirmed a sharp and symmetrical peak at 32.747 min (Figures S6 and S7), demonstrating the high purity of the isolated compound. The UV absorption maximum at 276 nm was consistent with aromatic amino acid–containing cyclic dipeptides such as diketopiperazines. Structural elucidation of Compound 1 was performed using ¹H- and ¹³C-nuclear magnetic resonance (NMR) spectroscopy. The ¹H-NMR spectrum (400 MHz, CD₃OD) exhibited characteristic methylene proton signals corresponding to the proline residue at δ 1.8–3.6 ppm and aromatic proton resonances derived from the tyrosine residue at δ 6.7–7.1 ppm (Figure 2a). The ¹³C-NMR spectrum (100 MHz, CD₃OD) displayed carbonyl carbon signals at δ 170.1 and 168.8 ppm and aromatic carbon signals at δ 125.2–155.0 ppm, confirming the presence of both proline and tyrosine moieties in a cyclic structure (Figure 2b). Comparison of these spectral data with those reported in the literature identified the compound as maculosin [cyclo(L-Pro-L-Tyr)], a diketopiperazine-type cyclic dipeptide (Figure 3;

Table 11. NMR spectroscopic data of the isolated compound maculosin (400 MHz, CD₃OD).

Position	δH, Mult. (J in Hz)	δC, Mult.
1	4.35 (1H, m)	58.0
2		167.0
3	4.04 (1H, m)	60.1
4		170.9
5	2.08 (1H, m) 1.24 (1H, m)	29.5
6	1.79 (1H, m)	22.8
7	3.54 (1H, m) 3.35 (1H, m)	46.0
8	3.06 (2H, m)	37.7
9		127.7
10	7.03 (1H, d, 8.2)	132.2
11	6.70 (1H, d, 8.2)	116.3
12		157.7
13	6.70 (1H, d, 8.2)	116.3
14	7.03 (1H, d, 8.2)	132.2

Chemical shifts (δ) are reported in ppm with multiplicity and coupling constants (J) in Hz. Assignments were made based on ¹H and ¹³C NMR data. Maculosin was identified as Cyclo(L-Pro-L-Tyr), and its structure is shown in Figure 3.

) [44,45]. This compound was isolated from Brevibacillus sp. JNUCC 42, representing the first report of maculosin production within the Brevibacillus genus.

3.9. Cytotoxicity and Anti-Melanogenic Effects of Maculosin in α-MSH-Stimulated B16F10 Melanocytes

Maculosin exhibited no significant cytotoxicity up to a concentration of 100 μM, maintaining > 90% cell viability compared with untreated controls, whereas mild cytotoxic effects were observed at concentrations ≥ 200 μM. Therefore, subsequent anti-melanogenic assays were conducted at concentrations of 25, 50, and 100 μM (Figure 4a).

Stimulation with α-MSH markedly increased melanin synthesis to approximately 130% of the basal control. In contrast, maculosin treatment significantly reduced melanin accumulation in a concentration-dependent manner, with relative melanin contents of 84.6%, 68.2%, and 55.4% at 25, 50, and 100 μM, respectively (p < 0.001), comparable to the effect of 300 μM arbutin (approximately 60%) (Figure 4b).

Consistent with these findings, maculosin suppressed intracellular tyrosinase activity, reducing enzyme activity to 79.3%, 64.8%, and 53.6% of the α-MSH-induced level at 25, 50, and 100 μM, respectively (p < 0.001) (Figure 4c). These results collectively indicate that maculosin inhibits melanin biosynthesis primarily through the downregulation of tyrosinase activity, suggesting its efficacy as a safe and naturally derived skin-whitening compound isolated from Brevibacillus sp. JNUCC 42.

4. Discussion

The volcanic ecosystem of Jeju Island represents one of the most dynamic and ecologically distinctive microbial habitats in East Asia. The Baengnokdam Crater Lake, located at the summit of Mt. Halla, exhibits a combination of environmental extremes, including high ultraviolet radiation, fluctuating temperatures, and oligotrophic, mineral-rich soils. These physicochemical factors create selective pressures that promote the evolution of microbial species with unique physiological and metabolic traits [1,2,3]. Microorganisms that thrive under such volcanic conditions often display enhanced stress tolerance, specialized membrane compositions, and the ability to synthesize secondary metabolites that protect against oxidative damage and UV exposure [4,5,6]. In this context, the isolation of Brevibacillus sp. JNUCC 42 from Baengnokdam soil highlights the ecological potential of Jeju’s volcanic biotope as a reservoir for discovering novel microbial taxa with functional and industrial significance.

Phylogenetic analysis based on 16S rRNA gene sequencing placed strain JNUCC 42 within the genus Brevibacillus, which was separated from the genus Bacillus in 1996 following reclassification based on 16S rRNA and chemotaxonomic evidence. The genus Brevibacillus encompasses aerobic, endospore-forming, Gram-positive bacteria commonly found in soil, water, and insect habitats. Members of this genus exhibit considerable physiological versatility and produce an array of biologically active compounds, including antimicrobial peptides, proteases, chitinases, and surfactants. The phylogenetic tree constructed using both 16S rRNA and genome-based GBDP methods clearly demonstrated that JNUCC 42 forms a distinct clade within Brevibacillus, most closely related to B. laterosporus DSM 25^T and B. daliensis YIM B02290^T. Although the 16S rRNA gene similarity with the nearest relative is high (99.2%), the ANI and dDDH values, together with the clear genomic separation in the whole-genome phylogenetic tree, support the divergence of strain JNUCC 42 as an independent lineage.

Comparative genomic indices further validated this taxonomic distinction. The ANI (87.1%) and dDDH (32.8%) values between Brevibacillus sp. JNUCC 42 and B. laterosporus DSM 25^T are well below the species delineation thresholds of 95–96% and 70%, respectively [33,40]. Such genomic divergence has also been observed among other Brevibacillus species, including B. brevis, B. parabrevis, and B. choshinensis, reflecting habitat-driven diversification and evolutionary adaptation within the genus [45,46,47,48]. The distinct genomic signature of JNUCC 42 is consistent with the ecological isolation of Baengnokdam Crater Lake, where selective pressures such as nutrient limitation, metal ion abundance, and strong UV radiation are likely to have promoted genetic drift and adaptive evolution [45,48].

Chemotaxonomic profiling provided further evidence for species-level differentiation. The major fatty acids of strain JNUCC 42, anteiso-C15:0 and iso-C15:0, are consistent with those typically found in Brevibacillus species [12], and are known to regulate membrane fluidity under fluctuating thermal conditions [45]. Interestingly, the detection of minor unsaturated fatty acids (C15:1 ω5c, C16:1 ω11c) and a unique unidentified aminolipid, absent in B. laterosporus DSM 25^T, may represent an adaptive feature that enhances membrane stability in the slightly alkaline and metal-rich volcanic soil of Baengnokdam [46]. Similar lipid remodeling has been reported in other extremophilic bacteria as a protective mechanism against osmotic and oxidative stresses, further supporting the ecophysiological distinctiveness of JNUCC 42 [16]. These chemotaxonomic deviations, together with genome-based comparisons, reinforce the proposal that Brevibacillus sp. JNUCC 42 constitutes a novel species within the genus.

Beyond taxonomy, genome mining provided crucial insights into the strain’s metabolic potential. The draft genome of JNUCC 42 (4.93 Mb, 40.7 mol% G+C) revealed 21 BGCs, including NRPS, PKS, and NRPS–PKS hybrid systems. The abundance and diversity of BGCs indicate a high capacity for secondary metabolite biosynthesis. Similar genomic richness has been reported in B. laterosporus, which produces insecticidal proteins, antimicrobial lipopeptides, and sporulation-associated metabolites [16,17,18]. However, several BGCs detected in JNUCC 42 showed low similarity (<40%) to known clusters in antiSMASH databases, suggesting the possibility of undiscovered bioactive compounds. These unique biosynthetic pathways are likely shaped by selective pressures in the extreme volcanic microenvironment, driving the production of specialized metabolites involved in stress protection or interspecies competition.

Among the secondary metabolites predicted from Brevibacillus sp. JNUCC 42, the cyclic dipeptide maculosin [cyclo(L-Pro-L-Tyr)] was isolated and its structure was confirmed by NMR spectroscopy. Originally characterized as a host-specific phytotoxin produced by Alternaria alternata infecting spotted knapweed (Centaurea maculosa) [49], maculosin has since been identified in various bacterial taxa, including Streptomyces and Brevibacillus, where it is recognized as a non-toxic microbial metabolite exhibiting diverse biological activities [41,46,50]. Cyclic dipeptides such as maculosin have gained increasing attention due to their broad-spectrum antimicrobial, antioxidant, and signaling properties, which contribute to microbial defense, interspecies communication, and biocontrol of phytopathogens [51,52]. In particular, maculosin and other diketopiperazine derivatives exhibit remarkable free-radical scavenging activity, consistent with reports on the antioxidant potential of low-molecular-weight microbial metabolites isolated from marine and terrestrial environments [53]. Recent in silico investigations from our group further expanded the biological scope of maculosin into the dermatological field. Xu et al. [54] demonstrated through molecular docking and dynamics analyses that maculosin forms stable complexes with mushroom tyrosinase (mTYR), tyrosinase-related protein 1 (TYRP1), and Bacillus megaterium tyrosinase (BmTYR), showing strong binding affinities at the catalytic sites. These computational results support the hypothesis that maculosin acts as a natural tyrosinase inhibitor, suggesting its potential as a candidate molecule for the development of safe and effective skin-whitening or anti-pigmentation agents.

In the present study, the maculosin purified from Brevibacillus sp. JNUCC 42 indeed exhibited significant inhibition of α-MSH-induced melanin synthesis and intracellular tyrosinase activity in B16F10 melanoma cells, while maintaining cell viability above 90% at concentrations up to 100 μM [55,56]. This correlation between the strain’s genomic biosynthetic capacity and its observed bioactivity provides experimental validation of the in silico predictions. Given the increasing consumer and industrial demand for naturally derived whitening agents, the ability of Brevibacillus sp. JNUCC 42 to produce maculosin under mild culture conditions highlights its potential biotechnological and cosmeceutical relevance. Moreover, cyclic dipeptides are known for their chemical stability, water solubility, and facile synthesis, making them attractive candidates for formulation in dermatological and functional cosmetic products. Collectively, these findings suggest that maculosin represents a multifunctional diketopiperazine-type metabolite linking microbial ecological adaptation with human health applications. Its diverse bioactivities—spanning antioxidant, antimicrobial, and melanogenesis-inhibitory effects—underscore the potential of Brevibacillus sp. JNUCC 42 as a microbial resource for sustainable production of bioactive compounds with both ecological and cosmeceutical significance [57,58,59].

Although maculosin demonstrated promising anti-melanogenic activity in vitro, melanogenesis is regulated by complex upstream pathways involving MITF, tyrosinase-related proteins, and MAPK signaling cascades. Therefore, further molecular-level investigations are required to elucidate the precise mechanisms by which maculosin modulates melanogenic processes. In addition, in vivo studies or advanced skin-equivalent models will be essential to validate its efficacy and safety under physiologically relevant conditions. Such follow-up studies will not only strengthen the mechanistic understanding of maculosin’s biological activity but also support its development as a robust and clinically applicable cosmeceutical ingredient.

The discovery of Brevibacillus sp. JNUCC 42 also contributes to expanding the biotechnological importance of the genus. B. brevis has long been utilized for the production of the antibiotics tyrocidine and gramicidin [58], while B. choshinensis is employed as a host for heterologous protein expression due to its high secretion efficiency [15]. In contrast, relatively few studies have examined volcanic or extremophilic Brevibacillus strains as sources of secondary metabolites. Notably, genome mining of JNUCC 42 revealed several biosynthetic gene clusters that are absent in closely related Brevibacillus genomes, including a distinct NRPS–PKS hybrid cluster and a siderophore-associated BGC, suggesting the presence of unique metabolic capacities. In addition, the genome contains enriched stress-response pathways related to UV tolerance and oxidative stress adaptation—features consistent with survival in Baengnokdam’s high-UV, oligotrophic volcanic environment. Taken together, these unique biosynthetic and adaptive pathways indicate that Jeju’s volcanic isolates may serve as valuable reservoirs for novel bioactive molecules with applications in skincare, antioxidant therapy, and environmental biotechnology.

Ecologically, the isolation of JNUCC 42 supports the hypothesis that microbial diversity in volcanic soils is underexplored and that environmental stressors act as evolutionary drivers for novel metabolic innovation. The strain’s adaptation to slightly alkaline, mineral-rich, and oligotrophic conditions mirrors trends observed in other extremophilic actinobacteria, such as Streptomyces species isolated from volcanic ash, which often produce rare polyketides and phenazines. Thus, the discovery of JNUCC 42 provides a valuable model for understanding how geochemical gradients and ecological isolation influence the evolution of metabolic diversity in soil bacteria.

In summary, the polyphasic characterization of Brevibacillus sp. JNUCC 42 demonstrates that it constitutes a distinct genomic species closely related to B. laterosporus but differentiated by its unique fatty acid profile, lipid composition, and biosynthetic gene repertoire. The strain’s ability to produce maculosin with anti-melanogenic activity further highlights its potential as a source of functional ingredients for cosmeceutical applications. Continued exploration of its genomic resources, coupled with metabolomic and transcriptomic profiling under stress conditions, will provide deeper insights into its adaptive strategies and biosynthetic regulation. Moreover, integrating these findings into the context of Jeju’s volcanic biodiversity may contribute to the development of sustainable microbial platforms for natural product discovery and regional bioindustry innovation.