Shared Secondary Metabolites Between Salicornia brachiata and Its Bacillus Endophytes Reveal a Cooperative Chemical Defense

ABSTRACT

This study explores the chemical ecology of the halophytic plant Salicornia brachiata and its associated Bacillus endophytes using LC‐MS‐based metabolomics. A total of 43 Bacillus strains were isolated and screened for the production of bioactive lipopeptides, with 26 strains showing both lipopeptide production and antimicrobial activity, particularly against Mycobacterium smegmatis MTCC6. Fermentation of a potent endophyte, Bacillus subtilis NPRoot‐3, yielded purified surfactins and fengycins, which displayed significant inhibitory activity with low IC₅₀ values against Mycobacterium smegmatis. LC‐MS analysis of aqueous and methanolic extracts of S. brachiata revealed molecular features putatively corresponding to surfactins, fengycins, and bacillibactins, indicating the presence of similar bioactive compounds in both the host plant and its endophytic microbiome. These findings suggest that the antimycobacterial properties of S. brachiata may be partly attributed to metabolites produced by its endophytes. Overall, this study highlights cooperative defense mechanisms between the plant and its endophytes, confirms the role of endophyte‐derived secondary metabolites in shaping plant bioactivity and offers promising leads for the discovery of novel antimycobacterial agents.

LC–MS metabolomics reveals that Salicornia brachiata and its Bacillus endophytes share bioactive lipopeptides, with several strains producing potent antimycobacterial compounds. Purified surfactins and fengycins exhibit strong inhibitory activity, suggesting a cooperative chemical defense. These findings highlight endophyte‐driven plant bioactivity and uncover promising leads for novel antimycobacterial agents.

1. Introduction

Exploring natural sources, particularly plants and their associated micro‐organisms, is an appealing strategy due to their rich chemical diversity, structural complexity, and significant bioactive potential [1]. Among these, halophytic plants, which thrive in saline environments, have garnered attention for their unique chemical profiles and associated broad biological activities, including antimicrobial properties. Salicornia brachiata, a halophyte found in marshy coastal areas in Gujarat and the western coast of India, is well known for its diverse nutraceutical and pharmaceutical applications [2]. Although extracts of S. brachiata have demonstrated antimycobacterial activity, the specific bioactive metabolites responsible for this effect remain elusive till date [3]. Endophytic micro‐organisms associated with halophytes represent an underexplored reservoir of bioactive secondary metabolites. In our previous work, we reported the isolation of antimycobacterial compounds, such as bacillibactins and surfactins from endophytes inhabiting S. brachiata [4]. Notably, Bacillus sp. emerged as a dominant genus among these endophytes, many of which exhibited significant antimycobacterial activity. Cyclic lipopeptides, such as surfactins, iturins, and fengycins, commonly produced by Bacillus sp. are widely reported to display a diverse range of biological activities, including antimicrobial, cytotoxic, mosquito larvicidal effects. [5]. Given this background, we hypothesized that the antimycobacterial potential of S. brachiata may be, at least in part, attributed to the production of bioactive metabolites by its endophytic microbiota. To explore this hypothesis, we compared the metabolic profiles of S. brachiata and its associated Bacillus endophytes using HPLC and LC‐MS. This study aims to establish a metabolic link between the plant and its endophytes, with a particular focus on cyclic lipopeptides, to elucidate their potential role in the antimycobacterial activity of the plant‐endophyte system. Our findings contribute to a deeper understanding of the chemical ecology of halophyte‐endophyte interactions, particularly in the context of microbial defense mediated by shared secondary metabolites.

2. Results and Discussion

2.1. Culturing of Bacillus sp. Isolates

A total of 43 Bacillus strains were cultured in modified Landy's medium, which is reported to enhance the production of cyclic lipopeptides in significantly high quantities [6]. The cultures were incubated at 30°C with shaking at 150 rpm for 5 days to obtain a lipopeptide‐rich broth, which was further processed for lipopeptide extraction.

2.2. Extraction and Detection of Lipopeptides

To assess the lipopeptide‐producing potential of the Bacillus endophytes, fermentation broths were subjected to acid precipitation to obtain a crude lipopeptide mixture. The presence of surfactins and fengycins was confirmed by HPLC chromatograms, which showed characteristic peaks in two distinct retention‐time (RT) regions, consistent with previous reports [7]. The peaks eluting between RT 8−9 min corresponded to fengycins, while peaks at RT 17−18 min were indicative of surfactins (Figure 1). Of the 43 Bacillus strains tested, 26 produced detectable levels of lipopeptides. The remaining strains either did not produce lipopeptides or produced them in quantities below the detection threshold (Table 1). The extracted lipopeptides were further analyzed using HPLC and MALDI‐TOF‐MS, revealing two characteristic ranges of molecular masses: 1400 to 1600 Da for fengycins and 900 to 1100 Da for surfactins. These findings are in line with previous reports indicating that Bacillus sp. produce diverse families of lipopeptides, including surfactins, fengycins, and iturins [6, 7].

FIGURE 1.

HPLC chromatogram showing the detection of surfactins and fengycins in the acid precipitate of Bacillus subtilis NPRoot‐3 and its comparison with the separated compounds.

TABLE 1.

Primary screening of Bacillus endophytes against Mycobacterium smegmatis MTCC6.

S. No.	Isolates	Strain	Lipopeptide	Zone of inhibition (Diameter‐mm)
1	NPRoot‐3	B. subtilis	+	8
2	NPA‐4	B. subtilis	+	10
3	4SPT8‐6A	B. subtilis	+	4
4	4VPT1‐9	B. subtilis	+	10
5	5VPT7‐10	B. subtilis	+	10
6	5VPT1‐13	B. subtilis	+	14
7	6SPT4‐14	B. subtilis	+	5
8	7NPSHOOT‐4	B. subtilis	+	10
9	4NPBL‐2	B. subtilis	+	10
10	NPB‐5	B. subtilis	+	12
11	NPA‐10	B. tequilensis	+	12
12	4VPT4‐5	B. tequilensis	+	8
13	5SPT4‐2	B. tequilensis	+	10
14	5SPT7‐7	B. tequilensis	+	8
15	5VPT1‐11	B. tequilensis	−	10
16	5SPT8‐5A	B. tequilensis	−	−
17	1VPT5‐7	B. infantis	−	−
18	1VPT5‐10	B. infantis	−	−
19	4NPA‐2	B. safensis	+	6
20	4SPT4‐10	B. safensis	+	10
21	4SPT4‐16	B. safensis	+	−
22	4SPT6‐1	B. safensis	+	4
23	6SPT7‐1	B. safensis	+	10
24	6VPT1‐2	B. safensis		−
25	7NPA‐13	B. stratosphericus	+	6
26	7NPB‐3B	B. velezensis	+	2
27	4SPT8‐2	B. filamentosus	−	4
28	4SPT12‐17	B. aerius	−	−
29	4VPT5‐1	B. pumilus	−	3
30	4VPT5‐1A	B. zhangzhouensis		−
31	5NPA‐1	B. amyloliquefaciens	+	3
32	5SPROOT‐1	B. paralicheniformis	−	−
33	5NPB‐5	B. aerius	+	8
34	5SPT7‐2	B. paralicheniformis	+	−
35	5SPROOT 3	B. stratosphericus	+	2
36	5SPT4‐2A	B. paralicheniformis	−	−
37	5VPT7‐8	B. paralicheniformis	+	−
38	5VPT11‐1	B. endophyticus	−	−
39	6VPT5‐9	B. zhangzhouensis	−	−
40	6NPA‐7	B. aerius	−	−
41	6VPT1‐8	B. pumilus	−	−
42	7NPB‐1	B. stratosphericus	−	−
43	7NPRoot‐4	B. cereus	−	−

Among the positive strains, Bacillus subtilis NPRoot‐3 exhibited the most promising lipopeptide profile and was therefore selected for further purification of surfactins analogs. Fractionation of its crude lipopeptide extract yielded five subfractions: F‐1, F‐2, F3‐3, F‐E, and F‐M. The F‐M fraction was further purified using an open column‐chromatography followed by semi‐preparative HPLC, leading to the isolation of six distinct compounds (1–6). MALDI‐TOF‐MS analysis of these compounds revealed ion signals corresponding to [M+H]⁺, [M+Na]⁺, and [M+K]⁺ adducts. Comparison with literature data indicated that all six compounds belonged to the surfactin family and were identified as C₁₃‐surfactin (1), C₁₄‐surfactin (2 and 3), C₁₅‐surfactin (4), and C₁₆‐surfactin (5 and 6) (Figure 2, Table 2). To determine the amino acid sequence of the peptide moiety, tandem mass spectrometry (MS/MS) was performed on the [M + Na]⁺ ions. Fragmentation patterns in the MS/MS spectra consistently confirmed the peptide sequence as Leu‐Leu‐Asp‐Val‐Leu‐Leu‐Glu‐(fatty acid) across all six surfactin homologs [8].

FIGURE 2.

Structures of six surfactins identified by MS/MS fragmentation pattern.

TABLE 2.

Identification of surfactins on the basis of MS/MS fragmentation pattern.

Compound	Major fragments in the MS/MS spectra of [M + Na]+ ion					[M + Na]+	Surfactin homologues
1	1031.18	917.48	804.44	688.43	590.34	1031.18	C13‐surfactin
	773.9	594.3	481.2	382.28	264.17
2	1044.95	931.52	818.48	702.46	604.4	1044.95	C14‐surfactin
	707.42	594.35	481.26	382.27	267.18
3	1044.98	931.55	818.52	702.50	604.4	1044.98	C14‐surfactin
	707.4	594.34	481.27	382.37	267.18
4	1059.28	945.5	832.4	716.4	618.39	1059.28	C15‐surfactin
	707.40	594.33	481.28	382.18	267.14
5	1073.36	959.5	846.5	717.49	618.40	1073.36	C16‐surfactin
	707.4	594.34	481.28	382.14	267.17
6	1073.21	959.5	846.4	717.49	618.40	1073.21	C16‐surfactin
	707.4	594.33	481.25	382.18	267.16

2.3. Bioassay

2.3.1. Antimycobacterial Activity of Crude Extracts From Bacillus Endophytes

Lipopeptide extracts from all 43 Bacillus isolates were screened for antimycobacterial activity against Mycobacterium smegmatis MTCC6. Of these, 26 isolates exhibited inhibitory activity, while 17 showed no detectable inhibition (Table 1). Among the active isolates, inhibition zone diameters ranged from 2 to 14 mm. Twelve isolates, B. subtilis NPA‐4, B. subtilis NPB‐5, Bacillus safensis 6SPT7‐1, B. subtilis 4NPB‐L, B. subtilis 4VPT1‐9, B. subtilis 5VPT7‐10, B. subtilis 5VPT1‐13, B. subtilis 7NPSHOOT‐4, Bacillus tequilensis NPA‐10, B. tequilensis 5SPT4‐2, B. tequilensis 5VPT1‐11, and Bacillus safensis 4SPT4‐10 displayed zones of inhibition ≥ 10 mm. DMSO was used as the vehicle control, as it served as the solvent for sample preparation (Figure 3). Among these, only four isolates (B. subtilis NPA‐4, B. tequilensis 5SPT4‐2, B. safensis, 4SPT4‐10, and B. subtilis 4VPT1‐9) demonstrated significant potency, with IC₅₀ values ranging from 100 to 300 µg/mL compared to the standard drug rifampicin (Table 3). The most active strains were B. tequilensis 5SPT4‐2 and B. safensis 4SPT4‐10, with IC₅₀ values as low as 132.37 and 164.72 µg/mL, respectively. These results are consistent with previous findings about antimycobacterial activity of bacillibactin siderophores isolated from the B. subtilis NPRoot‐3 endophyte against Mycobacterium tuberculosis H37Rv strain [9].

FIGURE 3.

Disk diffusion assay of 10 isolates exhibiting significant zone of inhibition ≥10 mm diameter against Mycobacterium smegmatis MTCC6.

TABLE 3.

IC₅₀ determination of potent Bacillus isolates against M. smegmatis MTCC6 with positive control Rifampicin.

S. No.	Isolates	IC50
1	B. subtilis NPRoot‐3	> 2 mg/mL
2	B. subtilis NPB‐5	> 2 mg/mL
3	B. subtilis 7NPSHOOT‐4	1.9 mg/mL
4	B. tequilensis NPA‐10	> 2 mg/mL
5	B. subtilis NPA‐4	296.1 µg/mL
6	B. tequilensis 5SPT4‐2	132.37 µg/mL
7	B. safensis 4SPT4‐10	164.72 µg/mL
8	B. subtilis 4VPT1‐9	210.746 µg/mL
9	B. velezensis 7NPB‐3B	> 2 mg/mL
10	B. subtilis 5VPT1‐13	1.4 mg/mL
11	B. subtilis 4NPBL‐2	1 mg/mL
12	B. safensis 6SPT7‐1	> 1 mg/mL
13	Rifampicin	2.8 µg/mL

2.3.2. Bioactivity of Surfactins and Fengycins

Crude mixtures of surfactins and fengycins isolated from B. subtilis NPRoot‐3 also displayed antimicrobial activity against M. smegmatis MTCC6, as well as other bacterial and fungal pathogens. Surfactins exhibited more pronounced activity than fengycins. Specifically, surfactins showed inhibitory effects against Staphylococcus aureus MCC2043, M. smegmatis MTCC6, Enterococcus faecalis MCC2409, and Candida albicans NCIM1152, with IC₅₀ values of 20.3, 251.2, 10.4, and 186.1 µg/mL, respectively (Table 4). Fengycins also inhibited M. smegmatis, although with lesser potency.

TABLE 4.

Antimicrobial capacities of surfactins and fengycins isolated from B. subtilis NPRoot‐3 isolate.

Pathogen	Fengycins		Surfactins
	Zone of inhibition	IC50 (µg/mL)	Zone of inhibition	IC50 (µg/mL)
A. baumanii MCC2076	+ (1 mm)		+ (3 mm)
C. albicans NCIM1152	—		+ (13 mm)	186.11
X. campestris NCIM5028	+ (3 mm)		+ (4 mm)
S. aureus MCC2043	+ (3 mm)		+ (5 mm)	20.3
M. smegmatis MTCC6	+ (5 mm)	1000	+ (5 mm)	251.2
C. herbarum NCIM1112	—		+ (10 mm)
E. aerogenes NCIM3092	—		+ (1 mm)
E. coli MCC2412	—		—
P. aeruginosa MTCC3541	—		—
E. faecalis MCC2409	+ (8 mm)	193.93	+ (8 mm)	10.41
K. pneumoniae MCC2451	—		+ (1 mm)
Vehicle control DMSO	—	—	—	—

2.4. Correlation of Metabolites Using LC‐MS

Following the identification of lipopeptide metabolites in Bacillus isolates, a comparative LC‐MS analysis was conducted to examine whether similar compounds were also present in the crude extract of the host halophytic plant S. brachiata. The obtained MS spectra of both aqueous and organic extracts of S. brachiata revealed molecular ions corresponding to surfactins, fengycins, and bacillibactins. For surfactins, the observed masses (m/z) were 1008.72 (M+H)⁺ and 1029.73 (M+Na)⁺ for C₁₃ homolog; 1022.74 (M+H)⁺ and 1043.75 (M+Na)⁺ for C₁₄; 1036.75 (M+H)⁺ and 1057.75 (M+Na)⁺ for C₁₅; and 1087.78 (M+Na)⁺ for C₁₈ homolog (Figure 2). Similarly, in case of fengycins, C₁₄A, C₁₅A, C₁₆A, C₁₅B, C₁₆B, and C₁₇B homologs were detected, with (M+H)⁺ masses of 1437.00, 1450.01, 1464.00, 1477.85, 1491.88, and 1508.04, respectively [10]. Additionally, bacillibactins belonging to the siderophore class were detected in the aqueous plant extract, with m/z values of 923.35 (M+Na)⁺, 905.26 (M+Na) ⁺, 900.05 (M+H)⁺, and 882.68 (M+Na)⁺ (Figure 4). The presence of structurally and functionally similar metabolites in both S. brachiata and its endophytic Bacillus strains suggests a potential chemical relationship between the host and its microbiome. This correlation may be attributed to intergeneric genetic exchange or metabolic interactions, supporting the hypothesis that endophytes significantly contribute to the antimycobacterial properties of S. brachiata.

FIGURE 4.

3D LC‐MS chromatogram plot of Salicornia aqueous extracts and organic extracts showing the presence of masses of surfactins, fengycin, and bacillibactins visualized using MZmine 3.0.

3. Conclusion

To conclude, this is the first comprehensive screening of a large number of Bacillus endophytes for their bioactive potential against M. smegmatis MTCC6. The significant inhibitory effects observed, particularly for cyclic lipopeptides such as surfactins and fengycins, highlight the potential of the genus Bacillus as a promising source of naturally occurring antimycobacterial compounds. These findings further suggest their potential applicability in the management of tuberculosis as well as Mycobacterium‐related infections. Furthermore, the detection of structurally similar metabolites in both S. brachiata and its associated endophytes, coupled with comparable bioactivities, provides new insights into the chemical ecology of this halophytic plant‐microbe association. The data suggest that S. brachiata may have evolved a symbiotic relationship with Bacillus endophytes, benefiting from their production of bioactive secondary metabolites with antimicrobial properties that contribute to plant defense against pathogens and insect pests. Overall, this work strengthens the case for exploring endophytic Bacillus species as natural reservoirs of therapeutically valuable compounds and sheds light on the ecological and evolutionary dynamics underlying plant‐endophyte interactions.

4. Methodology

4.1. Cultivation of Bacillus Isolates

To evaluate the lipopeptide‐producing potential and antimicrobial activity, Bacillus strains previously isolated from the halophyte Salicornia brachiata collected along the Gujarat coast, India [4], were cultivated in modified Landy's medium [6, 7]. The medium contained (per litre:) 35 g glucose, 4 g yeast extract, 8 g glutamic acid, 0.6 g MgSO₄.7H₂O, 1.1 g KH₂PO_4, 0.6 g KCl, 0.2 mg CuSO₄, 10 mg MnSO₄.4H₂O, and 0.2 mg FeSO_4. The pH was adjusted to 7.0 before sterilization by autoclaving at 115°C for 30 min. Each isolate was inoculated into 50 mL medium in 150 mL Erlenmeyer flasks and incubated for 5 days at 30°C with shaking at 150 rpm.

4.2. Extraction and Detection of Lipopeptides

4.2.1. Detection of Lipopeptides in Bacillus Endophytes

Lipopeptides were extracted using the classical acid precipitation method. The pH of the fermented broth was adjusted to 2 with 6N HCl and incubated overnight at 4°C to allow precipitation. The resulting pellets were extracted thrice with methanol, pooled, and concentrated to obtain crude lipopeptide mixtures. Surfactins and fengycins were detected using analytical reversed‐phase ultrahigh‐pressure liquid chromatography (RP‐UHPLC; Thermo Scientific Dionex UltiMate 3000) system comprising LPG‐3400SD pump, WPS‐3000 autosampler, and DAD‐3000 detector. Separation was achieved on a Luna C8 column (Phenomenex, 5 µm, 250 × 4.6 mm) using water (solvent A) and acetonitrile (solvent B) as the mobile phase. A 50 µL sample was injected and eluted using a gradient at 2 mL/min with UV detection at 214 nm: 25%−45% B (0−5 min), 45%−60% B (5−10 min), 60%−95% B (10−15 min) and 95% B (15−20 min).

4.2.2. Isolation of Surfactins and Fengycins

One strain, B. subtilis NPRoot‐3, was cultivated on a larger scale for purification of individual lipopeptides. The detected surfactin and fengycin peaks were separated using a semipreparative RP‐UHPLC (Thermo Scientific Dionex UltiMate 3000 system) equipped with a YMC‐Triart C18 column (5 µm, 250 × 10 mm ID). Chromatographic conditions were identical to the analytical method, except for a flow rate of 4 mL/min. The surfactin‐rich extract (0.929 g) was further separated by silica gel column chromatography using a stepwise petroleum ether−ethyl acetate gradient (1:0−0:1), followed by 100% methanol, generating sub‐fractions F‐1, F‐2, F3‐3, F‐E, and F‐M. Fraction F‐M (100% MeOH, 232.9 mg) was further separated into five major fractions (FM‐1 − FM‐5) on a RP‐C18 silica gel column using ACN−H₂O gradient (80%−90% ACN over 30 min). Fraction FM‐2 (16 mg) was applied to semi‐preparative HPLC using gradient of ACN−H₂O (80%−90% ACN over 20 min; 5.0 mL/min) to yield compound 1 (FM‐2‐2, 7.84 mg). From fraction FM‐3, compounds 2 (FM‐3‐1, 11.17 mg) and 3 (FM‐3‐2, 3.66 mg) were purified using RP‐HPLC over C8 column with an isocratic solvent system (65% ACN in H₂O). Separation of FM‐5 on HPLC (C8 column; ACN−H₂O, 75:25, v/v) led to isolation of compounds 4 (FM5‐1, 3.14 mg), 5 (FM5‐5, 0.82 mg), and 6 (FM5‐7, 6.61 mg). These compounds were characterized using MALDI‐MS/MS spectral data (Applied Biosystems 4800 MALDI‐TOF‐TOF analyzer).

4.3. Bioassays

4.3.1. Screening of Bacillus Endophytes Against Mycobacterium smegmatis MTCC6

All 43 isolates were screened for inhibitory property against M. smegmatis MTCC6 employing the disk diffusion assay [4]. Briefly, 100 µL log‐phase culture was spread onto Muller–Hinton agar plates. Sterile 6 mm disks were loaded with 1 mg of each crude extract and incubated at 30°C for 24 h, and thereafter, zones of inhibition were observed [11].

4.3.2. IC₅₀ Determination for Antimycobacterial Endophytes

Fourteen isolates showing significant inhibition were selected for IC₅₀ determination using an MTT cell viability assay following CLSI guidelines. Crude extracts were prepared as 100 mg/mL stock solutions in dimethyl sulphoxide (DMSO) to perform concentration‐dependent MTT assay in 96‐well microplate in triplicate. DMSO as the vehicle control, M. smegmatis MTCC6 culture without extract as the negative control, and rifampicin as the positive control with similar concentrations as test, were also included in the protocol. Briefly, each well contained 100 µL Muller–Hinton broth and serial dilutions of crude extracts (2 mg/mL to 100 µg/mL) were tested against log‐phase culture of M. smegmatis MTCC6 (10⁵ cells), and plates were incubated overnight at 30°C with shaking at 150 rpm. After 24 h, 10 µL of MTT dye (final concentration 0.5 mg/mL) was added under dark conditions and incubated for 1−2 h.The resulting formazan crystals were dissolved in 100% DMSO, and absorbance was measured at 570 nm using a microplate reader. Differences in absorbance were calculated relative to the controls.

4.3.3. Antimicrobial Potential of Surfactins and Fengycins

Purified compounds were screened against a panel of bacterial and fungal pathogens using the disk diffusion method, as described above. Test organisms included Acinetobacter baumanii MCC2076, Candida albicans NCIM1152, Xanthomonas campestris NCIM5028, S. aureus MCC2043, Cladosporium herbarum NCIM1112, Enterobacter aerogenes NCIM3092, Escherichia coli MCC2412, Pseudomonas aeruginosa MTCC3541, Enterococcus faecalis MCC2409, and Klebsiella pneumoniae MCC2451. IC₅₀ values were determined using the MTT assay for the most susceptible pathogens.

4.4. Correlating the Presence of Lipopeptides and Bacillibactins in Salicornia brachiata Extracts Using LC‐MS

Rathod and team reported antimycobacterial activity in S. brachiata extracts against Mycobacterium sp. [3]. To assess whether endophyte‐derived lipopeptides and bacillibactins were also present in plant tissues, aqueous and methanolic extracts were prepared. Briefly, 25 g of powdered plant material was extracted three times in 200 mL of methanol or water to obtain crude extracts. Metabolite profiling was performed using an Agilent 1290 Infinity UHPLC system coupled with an Agilent 6545 Q‐TOF mass spectrometer (comprising of Agilent G7104A‐1290 flexible pump and Agilent G7129B‐1290 autosampler). Samples were injected through Betasil C18 column (150 mm × 3 mm, 3 µm; Thermofisher Scientific) at a flow rate of 300 µL/min employing a gradient of 0.1% formic acid in water (solution A) and 0.1% formic acid in acetonitrile (solution B). The obtained data was processed and visualized as a 3D plot using MZmine 3.0 software.

Author Contributions

Sanju Singh, Yedukondalu Nalli, and Pramod B. Shinde conceived and designed experiments. Pramod B. Shinde obtained funding. Sanju Singh, Yedukondalu Nalli, Krishan, and Apexa Gajjar performed experiments. Sanju Singh and Yedukondalu Nalli analyzed the data. Sanju Singh, Usha P. Shinde, and Pramod B. Shinde drafted the manuscript. All authors read and approved the final manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.