Insights into Genomic Features and Potential Biotechnological Applications of Bacillus halotolerans Strain HGR5

Six water samples were collected aseptically from two hydrothermal springs in Hammam Ouled Tebben and Hammam Guergour (GPS: 35°47΄40.32˝N, 5°07΄30.18˝E; 36°19΄17.19˝N, 5°03΄19.36˝E respectively) (Wilaya of Sétif) for the isolation of thermotolerant bacteria. Water temperature was measured on-site with an electronic thermometer (Electronic temperature controller; Ahlborn Mess- und Regelungstechnik GmbH, Germany), and pH measurements (pH 211 Microprocessor pH Meter; HANNA Instruments Deutschland GmbH, Germany) were performed in the laboratory.

Dilutions from 10^–1 to 10^–7 were prepared from each sample with physiological water and used for bacterial isolation on three different media: GYM, R2A, and King’s B, by streaking 100 μl of each dilution on the Petri dishes. The plates were incubated at 30°C for 96 hours. The colony-forming units (CFU) per plate were measured for all dilutions in the three different media. The bacterial colonies were selected based on their macroscopic characteristics (i.e., color, shape), and successive streaks on TSA medium obtained pure cultures. Finally, pure colonies were suspended in a TSB medium containing 20% glycerol and stored at –20°C.

The isolates (n = 72) were cultivated at 30°C on TSA medium for 48 h. The bacterial colonies were harvested and suspended in saline solution, and the concentration was adjusted to 10⁸ CFU/ml, corresponding to an optical density (OD₆₀₀) of 0.1.

All isolates were screened in triplicate for lipase, protease, amylase, chitinase, cellulase, and xylanase activity on agar media containing specific substrates for each enzyme. 2 μl of bacterial suspension was spread over the surface of the agar. The dishes were incubated at 30°C for 48 hours. The enzymatic index was measured for all these activities.

Lipolytic activity was screened using Tween 80 as substrate. The degradation was performed in media containing 1.0% peptone, 0.5% NaCl, 0.01% CaCl₂·H₂O, 1.5% agar, and 1.0% substrate (Tween 80) (Ramnath et al. 2017). After incubation, precipitations around colonies show Tween 80 degradation (Ramnath et al. 2017).

The detection of proteolytic activity was screened on media supplemented with 3% casein, a zone of hydrolysis around the colonies indicates caseinolytic activity (Zhang et al. 2015).

Starch degradation was measured on agar supplemented with 1% starch. After two days of incubation, the starch degradation was detected by flooding the dishes with Lugol iodine solution; clear zones around colonies indicate positive activity (Benammar et al. 2020). Colloidal chitin agar was used to screen chitinolytic activity, according to Gasmi et al. (2019). Positive isolates were identified based on the presence of hydrolysis zones. Whereas assays for cellulase and xylanase were carried out on agar mediums enriched with CMC and xylan, respectively (Liu et al. 2011; Liang et al. 2014). The plates were flooded in 1% Congo red solution for 15 minutes after 48 hours of incubation, then washed with 1 M NaCl solution to visualize hydrolysis clearance zones (Shanthi and Roymon 2018).

Nineteen of the isolates were examined through a confrontation test on agar for activity against the three phytopathogenic fungus Fusarium graminearum (FG), Phytophthora infestans (PI), and Alternaria alternata (ALT). Co-cultivation was performed in PDA medium by placing an 8 mm mycelial agar disc in the center of the dishes. Then, two bacterial spots were inoculated 2.5 cm away from the fungi, toward the periphery. Controls were prepared by growing only the fungus under the same conditions. Dishes were incubated at 28°C and monitored after 5 and 14 days (Alenezi et al. 2016). The inhibition rate (IR), which is used to assess antifungal activity, was calculated using the following formula: IR(%)=(Dc−Dt)Dc×100%$${\rm{IR}}({\rm{\% }}) = {{({\rm{Dc}} - {\rm{Dt}})} \over {{\rm{Dc}}}} \times 100{\rm{\% }}$$ where:

Dc – control fungus diameter (average of 3 replicates),

The extraction of the genomic DNA was performed using a Wizard^® Genomic DNA Purification Kit (Promega, USA) following the manufacturer’s instructions. The two primers, 27F (5’AGAGTTTGATCCTGGCTCAG3’) and 1492R (5’GGTTACCTTGTTACGACTT-3’) were used for the amplification of the 16S rRNA genes using HOT FIREPol^® DNA polymerase (Solis BioDyne, Estonia). The PCR amplification mixture was composed of a 1 × concentration of reaction buffer, 2 mM magnesium MgCl₂, 200 μM of each dNTP, 0.3 μM of both the forward and reverse primers, 1 unit (U) of HOT FIREPol^® DNA polymerase enzyme, and 2 μl of isolated template DNA. The total reaction volume was adjusted to 20 μl using nuclease-free water. Cycling conditions were: initial activation at 95°C for 10 min, followed by 35 cycles of denaturation at 95°C for 30 sec, primers annealing at 58°C for 30 sec, and extension at 72°C for 1 min, then the final extension at 72°C for 10 min. Sequencing was performed by Eurofins Genomics (France), and based on the fluorescence-adapted Sanger method. Fragments were separated by capillary electrophoresis on an Applied Biosystems™ 3730xl Genetic Analyzer (Thermo Fisher Scientific, Inc., USA), and the resulting gene sequences were analyzed using the 16S Based-ID tool of EzBioCloud (https://www.ezbiocloud.net; Yoon et al. 2017). 16S rDNA-based phylogenetic tree was constructed using a neighbor-joining method in MEGA X (Kumar et al. 2018) using the 16S gene sequences from the closest Bacillus species.

The genomic DNA of B. halotolerans HGR5 was obtained, as mentioned above. The quantity and purity of the DNA were checked using the NanoDrop™ 2000/2000c Spectrophotometer (Thermo Fisher Scientific, Inc., USA). The sequencing library was prepared using the Illumina^® TruSeq^® DNA PCR-Free kit (Illumina, Inc., USA) according to the manufacturer’s recommendations. Whole genome sequencing was performed on the NovaSeq™ 6000 (Illumina, Inc., USA) platform using the S4 reagent kit v1.5 with a 2 × 151-bp paired-end read protocol.

Default parameters were used for all bioinformatic tools unless otherwise noted. TrimGalore v0.6.2 (Krueger 2015) was used for read trimming, and read quality was checked with FastQC v0.11.9 (Andrews 2010). Raw paired-end reads were de novo assembled with Unicycler v0.5.0 (Wick et al. 2017), and assembly metrics were calculated using QUAST v0.5.2 (Mikheenko et al. 2018). Whole genome annotation was carried out using the Bakta pipeline v1.8.1 (Schwengers et al. 2021). ResFinder (Bortolaia et al. 2020) and VFDB (Liu et al. 2019) were used to screen the genome sequence for antibiotic resistance and virulence genes, respectively, in addition to Pathfinder v1.1 (Cosentino et al. 2013) used to assess the pathogenicity of the strain.

The whole-genome shotgun project of B. halotolerans HGR5 has been deposited at GenBank under the accession number JASXRX000000000. The associated BioSample and BioProject accessions are SAMN35725142 and PRJNA983391, respectively.

Nucleotide sequences for the four housekeeping genes 16S rDNA, cdaA, gyrA, and gyrB were either extracted from the assembled genome of HGR5 isolate or retrieved from GenBank for the sixty-seven Bacillus type strains. For each gene, nucleotide sequences were aligned using muscle (default parameters) (Edgar 2004a; 2004b), and a maximum likelihood tree was constructed using mega X (Kumar et al. 2018) with Hasegawa-Kishino-Yano model, Gamma distribution with invariant sites G + I, and bootstrapping (n = 1,000) parameters.

Functional sequence annotation to establish clusters of orthologous groups (COG) categories was achieved using EggNOG-mapper (Cantalapiedra et al. 2021), where gene sequences were assigned gene ontology (GO) terms. Then, the distribution analysis of GO terms was performed and displayed using TB tools (Chen et al. 2020).

The biosynthetic gene clusters (BGCs) involved in the synthesis of secondary metabolites (SMs) were identified using the online tools antiSMASH v7.0.0 (Blin et al. 2023) and BAGEL4 (van Heel et al. 2018). Also, the ClusterBlast and KnownClusterBlast modules were employed for the comparative analysis of the obtained gene clusters as described by (Tsalgatidou et al. 2022). The detection of CRISPR-Cas systems was carried out online by the CRISPRCasFinder (Couvin et al. 2018).

Annotation of genes involved in antifungal activities was achieved using the carbohydrate-active enzymes (CAZymes) (http://www.cazy.org; Drula et al. 2022) database through the dbCAN3 annotation tool Run-dbcan v4.0.0 (https://github.com/linnabrown/run_dbcan; Zhang et al. 2018). Then, CAZymes were grouped based on the substrates and the corresponding Merops families (Rawlings et al. 2016).

The biodegradation of coconut oil, tributyrin, and polycaprolactone diol (PLCD; M_n ~ 2,000) by B. halotolerans HGR5 was performed on LB agar according to the protocol developed by (Molitor et al. 2020). A 50% (v/v) substrate emulsion was prepared in sterile distilled water containing 0.5% gum arabic. A volume of 30 ml of this emulsion was then added to 1 liter of LB agar and mixed using a homogenizer (POLYTRON^® PT 1200 E; Kinematica AG, Switzerland), then poured into Petri dishes. Clear zones around the colonies reveal positive activities against tributyrin and PCLD (Molitor et al. 2020). At the same time, coconut oil degradation was detected under UV radiation at 350 nm by observing fluorescent halos around positive colonies (Ramnath et al. 2017).

To investigate the thermotolerance of strain HGR5, tubes containing 5 ml of TSB broth were inoculated with 50 μl of the bacterial suspension and then incubated with shaking at 160 rpm at 30, 35, 40, 45, 50, 55 and 60°C respectively. After 48 h, the growth was recorded by measuring the optical density at 600 nm (Baltaci et al. 2020). For halotolerance, 50 μl of the bacterial suspension was inoculated into 5 ml of TSB broth with NaCl concentrations of 0; 2,5; 5; 7,5; 10; 12,5; 15%. The tubes were then incubated at 35°C (growth optimal temperature) under shaking at 160 rpm for 96 hours. Every 24 h, a volume of 300 μl was collected, and the OD₆₀₀ was measured using a microplate reader (VEG 500;

ESSE 3, Italy). Finally, the optimal pH was also determined by inoculating 50 μl of bacterial suspension into 5 ml of TSB broth with a pH range adjusted from 5 to 11 (Baltaci et al. 2020). The optical density of the bacteria was recorded at 24, 48, 72, and 96 h using a microplate reader (Wu et al. 2021). All experiments described above were performed with three repetitions.

The water temperature and pH of Hammam Ouled Tebben and Hammam Guergour were at 52°C; 44.5°C and 6; 6.9, respectively.

Three media (i.e., R2A, King B, and GYM) were employed for bacterial isolation, and enumeration was performed by membrane filtration of 100 ml of water from each source. The highest densities were observed in the R2A medium, around 39 CFU/100 ml for Hammam Ouled Tebben and 49 CFU/100 ml for Hammam Guergour. However, the other two media also showed significant densities, with the GYM medium exhibiting 33 CFU/100 ml for the two samples. For the King’s B medium, an average density of 35 CFU/100 ml and 30 CFU/100 ml was obtained for Hammam Ouled Tebben and Hammam Guergour, respectively. Thus, 72 isolates were obtained, 35 from Hammam Ouled Tebben and 37 from Hammam Guergour.

Screening of enzymatic activities revealed that 68% (n = 49/72) of the isolates produced protease, whereas 66,6% (n = 48/72) exhibited lipolytic activity against Tween 80. On the other hand, for glycosyl hydrolases, 62.5%, 73.6%, 50%, and 5.55% of the isolates were able to degrade starch, chitin, cellulose, and xylan, respectively (Table SI). The enzymatic profiles of bacteria and actinomycetes isolated from Algerian hot springs have been described in several studies (Gomri et al. 2018; Benammar et al. 2020; Medjemadj et al. 2020; Bouacem et al. 2022). These bacteria displayed a variety of typical extracellular enzymatic activities, suggesting their potential applications in various biotechnological applications.

Nineteen of the isolates with the best enzymatic profiles (Fig. 1, highlighted in bold in Table SI) were selected for the antifungal activity against F. graminearum, P. infestans, and A. alternata. Sixteen of them showed significant antifungal activity in the confrontation test (Table SII). Our selection criteria, which included a < 50% inhibition rate on ALT, PI, and FG, enabled us to select three strains with the highest antifungal activities, which are HGG15, HGK11, and HGR5. Their inhibition rates were 60, 64, and 60% for ALT, 53.13, 79.69, and 53.13% for PI, 56.90, 51,72, and 58.62% for FG, respectively.

Fig. 1.

Numerous studies have highlighted the promising antifungal activity of B. halotolerans. For instance, B. halotolerans MS50-18A had an antagonistic efficacy towards Phytophthora capsici, Fusarium solani, Rhizoctonia solani, and Fusarium oxysporum, which was higher than 60% (Sagredo-Beltrán et al. 2018). Also, four strains (named BFOA1 to BFOA4) belonging to this species have been isolated from the roots of plants grown in semi-arid areas of North Africa (Algeria and Tunisia). They had a considerable antagonistic potential toward F. oxysporum f. sp. albedinis, and 16 other phytopathogenic fungi of the genus Fusarium (Slama et al. 2019). They also exhibited antifungal activity against A. alternata, Rhizoctonia bataticola, P. infestans, and Botrytis cinerea. In a different investigation, B. halotolerans strain KLBC XJ-5 was employed as a biocontrol agent against B. cinerea mold on harvested strawberries. It was able to maintain the nutrient quality of the fruit while also controlling mycelial development (Wang et al. 2021).

On the other hand, the B. halotolerans KKD1 strain showed a potent antagonistic effect against F. graminearum, with an average size of the zone of inhibition above 11 mm. Additionally, the lipopeptide within the extracts of this strain presented a suppressive effect on this phytopathogen (Wu et al. 2021). The mining of its genome revealed that this antifungal activity resulted from the production of several bioactive molecules. The production of fengycin/plipastatin, surfactin, bacillibactin, and polyketides was mostly responsible for the activity attained. Therefore, the expression of surfactin and fengycin lipopeptides, which have been recognized as potent antifungals of KKD1, was confirmed by HPLC and MALDI-TOF-MS (Wu et al. 2021). In addition, the B. halotolerans KDD1 strain displayed positive protease and amylase activities, which aligns with our findings (Wu et al. 2022).

The two B. halotolerans strains Hil4 and Cal.l.30, isolated from medicinal plants, demonstrated antifungal activity against B. cinerea. They successfully controlled gray mold on harvested grapes and cherry tomatoes. Moreover, the supernatants of these strains also showed high activity against this plant pathogen. On the other hand, UHPLC-HRMS analysis of the extracts showed that B. halotolerans Hil4 and Cal.l.30 synthesized and secreted various secondary metabolites with antibacterial activity, such as fengycin, surfactin, mojavensin A, and others (Thomloudi et al. 2021; Tsalgatidou et al. 2022). Similarly, the culture supernatant of B. halotolerans strain QTH8 exhibited antifungal activity by significantly reducing the conidial germination of Fusarium pseudograminearum. Fengycin, iturin, and surfactin were identified by MALDI-TOF-MS in the QTH8 extract. Therefore, PCR was used to confirm the presence of the required genes to produce these compounds (Li et al. 2022).

The generated16S DNA genes sequences (i.e., MN420494.2, MN420495.1, MN420495.1 GenBank accession numbers for HGR5, HGG15, and HGK11 strains, respectively) were submitted to the EzBioCloud database through the 16S-Based ID tool. Unfortunately, the obtained sequences did not cover the whole 16S rDNA gene sequence (i.e., completeness of about 80%), making partial taxonomic identification with a validation at least of the genus. Thus, we obtained a 98.01% and 97.46% similarity sequence with B. halotolerans ATCC^® 25096™ for HGG15 and HGK11 strains, respectively. Nevertheless, the highest percentage of identity for HGR5 was 94.57%, which is below the 97% threshold needed to suggest a potential species affiliation. Thus, identifying this strain required further molecular analyses, such as whole genome sequencing.