The genome sequences of P. thermoglucosidasius DSM 2542 (GenBank: LAKX00000000) were derived from the NCBI database at.¹ Bacteriocin gene mining in the genome of P. thermoglucosidasius DSM 2542 was performed by BAGEL4 web-server at.² Prediction of signal sequence in bacteriocin was performed by SignalP 5.0 Server at.³ BLAST protein sequence analysis was performed using the NCBI database at⁴ and the UniProt database at.⁵ Protein sequence alignment was performed using the UniProt database at.⁶ Protein structure prediction was performed by the I-TASSER web server at.⁷ To evaluate the secondary structure of the peptide a DichroWeb online analysis was performed on a web-server at.⁸

For routine cultivation of thermophilic bacteria CASO broth (Roth) or NB medium was used and supplemented with 15 g/L agar (Oxoid) when needed. For NB medium preparation 10 g/L of tryptone from casein (Roth), 5 g/L of meat extract, and 5 g/L of NaCl (Roth) were used. For mesophilic bacteria cultivation LB or LB-agar (Biolab) media were used. Yeasts were grown in YPD medium, which contained 2% peptone from casein (Roth), 1% yeast extract (Roth), and 2% glucose (Roth) and supplemented with 15 g/L agar (Fisher Scientific) when needed. Medium selection for each strain in experiments is listed in Supplementary File 1.

Growth media supplemented with glycerol [20% (w/w)] were used as stock culture stored at −75°C temperature. Mesophilic bacteria strains were cultivated in a thermostat or thermoshaker with aeration at 37°C, thermophilic bacteria at – 55°C, and yeasts at – 30°C.

For the spot on a lawn assay using mesophiles, an overnight culture of the indicator strain was prepared by cultivating it at 30/37°C. Then, 1% of the culture was inoculated into melted agar medium and poured into Petri plates. Once the medium solidified, 10 μl of the bacteriocin sample was spotted on the plate. The plate was then incubated overnight in a thermostat at 30/37°C. The next day, the growth inhibition zone of the indicator strain was evaluated on the plate.

For the antimicrobial activity assay using thermophiles, a fresh colony of the indicator strain was inoculated into a liquid medium and cultivated overnight at 55°C in a thermoshaker. The following morning, the culture was inoculated (2–3%) into fresh medium and cultivated until the optical density (OD) reached 1–1.5. Then, the culture was mixed with melted agar medium at a ratio of 1:10 and poured into Petri plates.

For the spot on a lawn assay, 10 μl of the sample was spotted on the plate. To compare the activity between different samples, serial two-fold dilutions of the samples were made and dispersed on the plate. The arbitrary unit of antibacterial activity per milliliter (AU/ml) was defined using a formula 2ⁿ × 1 ml/V, where V represents the sample volume (ml) and n represents the titer of the reciprocal highest dilution that results in growth inhibition of the indicator strain (Daba et al., 1991). Next, the plate was incubated overnight at 55° in a thermostat. The following day, the growth inhibition zone of the indicator strain was evaluated on the plate. A list of bacteria and yeast strains with their respective cultivation conditions can be found in Supplementary File 1.

DNA amplification was conducted using Phusion HF DNA polymerase, while genomic DNA extraction was performed using the GeneJET Genomic DNA Purification Kit. For plasmidic DNA extraction, the GeneJET Plasmid Miniprep Kit was used. Amplified or digested DNA were cleaned with the GeneJET PCR Purification Kit. For DNA cloning T4 DNA Ligase enzyme was used. For DNA digestion restriction endonucleases FastDigest NcoI and FastDigest BamHI were used. DNA extraction or amplification was confirmed by 1% agarose gel electrophoretic analyses. All reagents and kits mentioned in this section were purchased from Thermo Fisher Scientific and used according to manufacturers’ recommendations.

A synthetic DNA coding Geo6A precursor peptide with optimized codon sequences for E. coli was ordered using GeneArt Strings DNA Fragments (Thermo Fisher Scientific) biosynthesis service. The DNA sequence was designed (Supplementary File 2) to code His6-tag and TEV peptidase recognition sites at the N-terminus of the peptide. DNA fragment His-TEV-geo6A was ligated into vector pET15b between NcoI and BamHI restriction sites using T4 DNA ligase and restriction endonucleases. The ligated DNA was transformed into E. coli DH5α by electroporation. Positive transformants were selected on LB-agar plates containing 50 μg/mL ampicillin and by colony PCR using Duet-up1 (5′-GGATCTCGACGCTCTCCCT-3′) and T7 Terminator (5′-GCTAGTTATTGCTCAGCGG-3′) primers. The constructed vector containing cloned gene (pET-His-TEV-geo6A) was extracted from bacteria culture and the DNA sequence of the insert was confirmed by Sanger sequencing.

E. coli BL21 (DE3) cells were transformed with pET-His-TEV-geo6A vector and inoculated to liquid LB medium containing 50 μg/ml ampicillin and cultivated in a thermoshaker. The overnight culture was inoculated (1%) to fresh medium with antibiotics and grown until OD reached 0.9–1.2. Then gene expression was induced using IPTG 1 mM end concentration. After 4 h of cultivation in thermoshaker bacterial culture was collected by centrifugation at 5,000 ×g for 20 min at 4°C. Harvested cells were resuspended in buffer A (25 mM sodium phosphate, 500 mM NaCl, pH 7.4) and kept at −75°C temperature until further use.

After induction harvested cells were sonicated on ice for 20 min using ultrasound sonicator VCX 130 (Sonics and Materials) with cycle 5 s ON and 10 s OFF and using 35% amplitude. Cell debris was separated by centrifugation at 15,000 ×g for 20 min at 4°C. The supernatant containing soluble fraction was diluted with binding buffer A up to 100 ml and filtered using a 0.45 μm filter. Further protein purification was performed using the BioLogic DuoFlow chromatography system (Bio-Rad) and the sample was loaded on Ni²⁺ affinity chromatography column HisTrap HP 1 ml (GE) previously equilibrated in the buffer A (section 2.6). After sample application, the column was washed with buffer B (25 mM sodium phosphate, 500 mM NaCl, 20 mM imidazole, pH 7.4). Protein was eluted using a step gradient with buffer C (25 mM sodium phosphate, 500 mM NaCl, 500 mM imidazole, pH 7.4). Elution fractions containing peptides were collected and further loaded on HiTrap HP 5 ml (GE) desalting column using buffer D (50 mM sodium phosphate, 100 mM NaCl, pH 7.4). Next, elution fractions of desalted protein were mixed with TEV protease, which was made in our laboratory according to Kapust et al. (2001). The reaction was performed in buffer D containing 5 mM DTT and incubated in a water bath for 4 h at 37°C temperature. After treatment with protease sample was diluted in buffer E (20 mM sodium phosphate, pH 7.4) and loaded on cation exchange column UNO S-1 (Bio-Rad) previously equilibrated in the same buffer D. Further, the column was washed with buffer D and protein was eluted using linear grading with buffer E (20 mM sodium phosphate, 1 M NaCl, pH 7.4). Elution fractions were collected and analyzed using tricine-SDS-PAGE (section 2.8).

Cell samples or elution samples were analyzed by tricine-SDS-PAGE (Schägger, 2006). The electrophoresis was carried out using 4% stacking and 16% separating polyacrylamide gels. Following protein electrophoresis, the gel was stained using PageBlue Protein Staining Solution (Thermo Fisher Scientific) as per the manufacturer’s instructions.

For thermostability assay purified recombinant Geo6 was incubated at different temperatures: 60°C, 80°C and 100°C for 2 h and at 95°C up to 6 h. After incubation, serial two-fold dilutions of samples were made using medium, and further antibacterial activity of bacteriocin was assessed by spot on a lawn assay.

For evaluation of stability at different pH ranges, purified recombinant Geo6 was mixed with 25 mM Britton-Robinson buffer solutions (Ebihara et al., 2016) with different pH values in a ratio of 1:20. After incubation at room temperature for 2 h samples were neutralized with NaOH or HCl solutions to pH 7. Further, samples were diluted two-fold using a medium and assessed for antibacterial activity using spot on a lawn assay.

MICs of Geo6 were determined according to the method of Wiegand et al. (2008) with some modifications. An indicator strain sensitive to bacteriocin was grown overnight at 55°C in a thermoshaker. The next day it was inoculated (2–3%) into fresh medium and cultivated until OD reached 1–1.5. The culture was further diluted and adjusted to 1 × 10⁶ CFU/ml. A 95 μl of the medium was mixed with 5 μl of Geo6 and serial two-fold dilutions of suspension were made with the medium. Next, diluted mixtures were dispersed (100 μl) in a 96-well plate. A volume of 100 μl of previously prepared cell suspension was dispersed and mixed with the diluted bacteriocin in every well on the plate. As a result, the final volume of suspension in the well containing diluted bacteriocin and bacteria cells was 200 μl and the final concentration of the bacteria culture in the well was 5 × 10⁵ CFU/ml. For each experiment, three replicates were made in the plate. Wells with positive controls contained a mixture of medium (95 μl) and Geo6 (5 μl), and negative controls contained bacteria suspension (5 × 10⁵ CFU/ml) in the medium (100 μl). The plate was placed into a bag to prevent medium evaporation and put into a box, which was incubated overnight at 55°C in a thermoshaker. The next day, the growth inhibition was evaluated in the wells of the plate.

An indicator strain G. kaustophilus HTA 426^T sensitive to bacteriocin was grown in BHI medium at 55°C till OD 0.5 was reached. Next, the bacteria culture was treated with 600 nM of Geo6 and further cultivated for 30 min. Then, the cell culture was centrifuged at 3000 ×g for 5 min and further washed two times with 0.9% NaCl solution (centrifuged at 3000 ×g for 5 min). After the last wash, cells were suspended in 0.9% NaCl solution and stained using LIVE/DEAD BacLight Bacterial Viability Kit (Invitrogen) containing SYTO9 and fluorescent dyes according to manufacturers’ recommendations. Cells were incubated in the dark at room temperature for 15 min. Next, cells were centrifuged at 3000 ×g for 5 min and washed two times with 0.9% NaCl solution (centrifuged at 3000 ×g for 5 min). Further, 10 μl of Geo6 treated cells and nontreated control cells suspensions were spotted onto objective slides and examined using a fluorescence microscope using filters FITC (480/500 nm) and Texas Red (490/635 nm).

Peptide solution was prepared in 10 mM sodium phosphate buffer (pH 7.4) and at a final concentration of 10 μM. The circular dichroism (CD) data was collected using a J-810 CD Spectrophotometer (JASCO) in a Cuvette of 0.125 cm path length and in a range of 190–280 nm at 0.5 nm intervals. The spectrum of the scan was recorded with a 2 nm optical bandwidth and presented as the average of 4 scans. Baseline measurements were recorded using a buffer solution and then subtracted from the sample scan.

As an indicator of thermal denaturation in varying pH or other conditions a fluorescent SYPRO Orange dye (Invitrogen) that binds to hydrophobic parts of protein during its unfolding was used in this work. SYPRO Orange 50× stock solution was prepared by diluting 2.5 μl of 5,000× concentrate into 250 μl of deionized (DI) water. A volume of 2.5 μl of 50× SYPRO Orange was aliquoted into a white HardShell 96-well PCR plate (Bio-Rad). Different pH buffer solutions (25 mM Briton-Robinson, pH 3–11; Bendikienė et al., 2004) were added to aliquoted SYPRO Orange followed by adding an appropriately diluted solution of purified Geo6 into the wells at a final concentration of 0.25 nmol/ml (1.3 μg/ml). The final volume of the mixture was 25 μl. The negative control was 25 mM of Briton-Robinson buffer solutions in a range of pH 3 to 11, with 2.5 μl of SYPRO Orange. Lysozyme (final concentration of 100 μg/ml) diluted in 25 mM Briton-Robinson buffer (pH 5 and 6) and mixed with 2.5 μl of SYPRO Orange was used as a positive control. A curve experiment for DNA with the elimination of the initial denaturation step at 95°C was used. The melt curve temperature range was set from 20°C to 95°C in increments of 0.5°C for 5 s. Each experimental condition was prepared in triplicate. Thermal shift analysis or differential scanning fluorimetry (DSF) of Geo6 was performed in CFX Touch Real TIME PCR System (Bio-Rad) using CFX Manager (Bio-Rad) software.

To reveal new antimicrobial peptides encoded in thermophilic bacteria, we screened a series of NCBI-available genomes from various Geobacillus and Parageobacillus spp. bacteria stored in the microorganism collection of the Department of Microbiology and Biotechnology at Vilnius University’s Life Sciences Center. Genome analysis using the BAGEL4 web tool revealed that the genomic DNA of the thermophilic bacteria P. thermoglucosidasius DSM 2542 encoded a gene cluster of 5,709 bp in length (Figure 1) and presumably is responsible for the biosynthesis of a novel bacteriocin. This strain was confirmed to produce antimicrobial substances using a spot on a lawn assay (Figure 2). The gene cluster contained 6 genes, which were named here geo6ABCDEF, could be related to the biosynthesis of the novel bacteriocin, that we have named geobacillin 6 (Geo6).

BLAST analysis and alignment of the geo6A gene product (Figures 3, 4) revealed that the encoded 48 amino acid length peptide, Geo6, shares sequence similarity with a group of leaderless bacteriocins: lacticins Q and Z, aureocin A54, epidermicin NI01, and BhtB (Perez et al., 2018). The highest sequence similarity was observed with lacticin Q (52%) and lacticin Z (54%).

Further analysis of the gene cluster observed that the geo6D, geo6E, and geo6F genes encode proteins with 78, 78, and 84% sequence similarity, respectively, to proteins annotated as ABC transporters. While the geo6B and geo6C genes encode proteins with 51 and 34% sequence similarity, respectively, to proteins annotated as Pleckstrin homology (PH) domain-containing proteins. Additionally, they exhibit 46 and 29% similarity, respectively, to proteins annotated as transmembrane YdbS and YdbT proteins, respectively.

The PH domains are small modular domains, which occur in a large variety of signaling proteins and serve as simple targeting domains that bind lipids in eukaryotic cells. However, these proteins are also present in bacteria (Cozier et al., 2004; Xu et al., 2010). Based on Butcher and Helmann (2006) research, the YdbST proteins can be responsible for resistance to antimicrobial compound(s) produced by Bacillus amyloliquefaciens. A recent study suggests that proteins containing PH domains and proteins similar to YbdST constitute transporters dedicated to leaderless bacteriocins (Pérez-Ramos et al., 2021). In addition, studies of gene clusters coding for bacteriocin lacticin Q (Iwatani et al., 2013) and aureocin A54 (Nascimento et al., 2012) showed that their biosynthetic loci encode proteins similar to ABC transporters. The secretion of the lacticin Q in L. lactis QU5 is strictly controlled by the presence of LnqBCDEF, which confers the secretion and self-immunity of lacticin Q. It was revealed that the LnqEF proteins are the minimal unit required for sufficient immunity. Whereas LnqBCD, showing sequence similarity to ABC transporters, are considered accessory proteins, which support the activity of the LnqEF (Iwatani et al., 2012, 2013).

However, studies have demonstrated that the production of leaderless bacteriocins can occur independently of the biosynthesis machinery. It has been shown that the synthesis of active bacteriocin is possible with the production of only the precursor peptide (Netz et al., 2002b).

Based on previous studies and analysis of the geo6ABCDEF genes, it can be suggested that these genes are responsible for the biosynthesis of Geo6 (Table 1). The geo6A gene encodes the bacteriocin precursor peptide Geo6, while the geo6BCDEF genes are dedicated to transport and immunity, as observed in other biosynthetic gene clusters of leaderless bacteriocins. Some gene clusters of leaderless bacteriocins also contain genes responsible for biosynthesis regulation (Perez et al., 2018). However, in the identified gene cluster, which was encoded in a relatively short DNA contig (6,966 bp, GenBank: LAKX01000053.1), only one gene outside of geo6ABCDEF was found. No other genes associated with bacteriocin biosynthesis or regulation were identified.

Further, gene geo6A was heterologously expressed in E. coli for biosynthesis and purification of recombinant bacteriocin. For this purpose, we used a synthetic gene with optimized codons for E. coli bacteria. In addition, the gene was designed to encode His6-tag and TEV peptidase recognition site in the N-terminus of the peptide (Figure 5). The new gene of bacteriocin precursor, His-TEV-geo6A, was cloned into pET-15b vector downstream T7 promoter and induced for expression in E. coli BL21 (DE3) cells. The synthesized peptide His-TEV-Geo6 was purified from soluble cell lysate fraction using Ni⁺² affinity chromatography, desalted by gel chromatography, then treated with TEV protease to cleave His-tag sequence and finally purified to homogeneity by cation exchange chromatography (Figure 6A). The recombinant bacteriocin Geo6 had additional Gly upstream of the Met at the N-terminus of the peptide.

Purified Geo6 peptide was evaluated for the antibacterial activity using spot on a lawn assay against 36 bacteria and 4 yeast strains, including Gram-positive, Gram-negative, and pathogenic strains (Table 2). Results demonstrated that bacteriocin was active against 12 out of 19 thermophilic bacteria strains tested, including Aeribacillus pallidus DSM 3670^T, Anoxybacillus tepidamans DSM 16325^T, Parageobacillus toebii DSM 14590^T, and various Geobacillus spp. bacteria strains (Figure 6B). Additionally, we observed small antibacterial activity against one mesophilic bacterium, Bacillus zanthoxyli CH07. However, the yeasts Saccharomyces cerevisiae or Candida spp., as well as the mesophilic bacteria Bacillus spp., Staphylococcus spp., Pseudomonas aeruginosa, Enterococcus faecalis, Salmonella enterica, Escherichia coli, Acinetobacter baumannii, and Stenotrophomonas maltophilia strains used in the assay were not susceptible to the bacteriocin.

In addition, the MIC values of Geo6 were evaluated against some susceptible thermophilic bacteria strains (Table 2). It was demonstrated that the MIC value for G. thermodenitrificans DSM 465^T was 2.34 μM, for G. lituanicus DSM 15325^T – 585 nM, for G. thermoleovorans DSM 5366^T – 585 nM, for G. kaustophilus HTA 426–293 nM, and for P. toebii DSM 14590^T it was 4.45 μM.

For comparison, previous studies have shown that pallidocin, a glycocin-type bacteriocin produced by thermophilic bacteria A. pallidus strain 8, was active in the pM range against other thermophiles, and a 2.4 pM concentration of pallidocin inhibited the growth of thermophilic bacterium P. genomospecies 1 strain NUB36187 (Kaunietis et al., 2019). Other leaderless bacteriocins are active in the μM and nM ranges (Fujita et al., 2007; Iwatani et al., 2007) and have a broad spectrum of activity. Studies showed that lacticin Q was active against a variety of tested Gram-positive bacteria strains, including Bacillus spp., Lactobacillus spp., Enterococcus spp., Pediococcus spp., Lactococcus spp., Streptococcus spp., and some other species. The highest activity was detected against B. coagulans JCM 2257^T (MIC – 1.6 nM), L. sakei subsp. sakei JCM 1157^T (MIC – 3.2 nM), P. dextrinicus JCM 5887^T (MIC – 7.3 nM). No activity was detected against the Gram-positive bacteria Escherichia coli JM109 (Iwatani et al., 2007). Bacteriocin aureocin A54 had a narrower antibacterial activity spectrum, it was active against some tested Gram-positive indicator bacteria strains, including Brochothrix sp., Carnobacterium spp., Lactococcus spp., and Staphylococcus spp. The lowest MIC values were detected against B. campestris ATCC 43754–2 μM, and L. lactis ATCC 19257–250 nM (Acedo et al., 2016).

Our results show that Geo6, as well as other leaderless bacteriocins lacticin Q or aureocin A54, are active in nM concentrations, in contrast to the bacteriocin pallidocin from another thermophilic bacterium, which can kill bacteria in pM concentrations. However, Geo6, similar to pallidocin, is active against Gram-positive and mostly against closely related thermophilic bacteria, including Geobacillus spp., Aeribacillus sp., Anoxybacillus sp., and Parageobacillus sp. strains. Other leaderless bacteriocins, such as lacticin Q and aureocin A54, inhibit Gram-positive bacteria as well. However, they have a wider antibacterial spectrum and kill more than just closely related bacteria.

Cells of G. kaustophilus HTA 426^T were treated with Geo6, and membrane integrity was assayed using the fluorescent dyes SYTO9 and PI (Figure 7). The experiment showed that bacteria cells treated with the bacteriocin emitted red fluorescence, since propidium iodide can enter only cells with damaged membranes. In contrast, control cells not exposed to Geo6 emitted green fluorescence, indicating an intact membrane. These results suggest that Geo6 destabilizes the membrane of indicator cells. This can be supported by the fact that the peptide has a high charge, its theoretical isoelectric point (pI) value is 9.3. Cationic peptides tend to be attracted to an anionic cell membrane. Moreover, our suggestion that Geo6 targets the cell membrane is in agreement with a recent study on leaderless bacteriocins, epidermicin NI01 (Staphylococcus epidermidis) and aureocin A53, which revealed the unique mode of action of these peptides. It comprises several distinctive methods of membrane disruption, and it was found that a four-helix bundle structure is necessary to complete the mechanism of membrane disruption (Hammond et al., 2020). Research on aureocin A53 and lacticin Q showed permeabilization of the bacterial cell membrane, followed by membrane potential disruption, efflux of various vital compounds, and termination of macromolecular synthesis (Netz et al., 2002a; Fujita et al., 2007). It was also shown that lacticin Q forms at least 4.6 nm pores (Yoneyama et al., 2009).

The purified recombinant Geo6 was exposed to different temperatures, and the thermal stability of the bacteriocin was evaluated after treatment by using a spot-on-lawn assay. The results showed that after treatment at 60°C, 70°C, and even 100°C for 2 h, the peptide retained 100% of its antibacterial activity. Furthermore, incubation of the peptide for up to 6 h at 95°C did not abolish its activity, and after incubation at room temperature for 24 h, the peptide remained 100% active. Additionally, Geo6 stability was evaluated in different pH ranges. Incubation in the pH range of 3–11 showed that Geo6 retained 100% of its activity in acidic and alkaline environments. These results are consistent with observations from experiments with other leaderless bacteriocins, such as lacticin Q or Bht-B, which remained active after treatment at high temperatures and in a wide pH range (Fujita et al., 2007).

Analysis of the Geo6 peptide sequence using the SignalP bioinformatics tool confirmed that it has no signal or leader peptide, consistent with other leaderless bacteriocins. Three-dimensional structure prediction using I-TASSER suggested that Geo6 is predominantly formed from α-helixes. Furthermore, structure assembly simulation showed that Geo6 forms four distinct helixes, which fold into a compact, globular assembly (Figure 8), similar to other leaderless bacteriocins such as epidermicin NI01 (PDB: 6SIF), aureocin A54 (PDB: 2N8O) and lacticin Q (PDB: 2N8P). Previous studies using structure prediction tools suggested a similar assembly for other leaderless bacteriocins such as lacticin Z, weissellicins Y and M, and enterocins L50A and L50B (Acedo et al., 2016). The secondary structure of Geo6 was evaluated using circular dichroism (CD) spectroscopy (Figure 9). The CD spectra revealed the presence of α-helical structure of the peptide in an aqueous solution. These findings are consistent with the predictions made by I-TASSER and are in agreement with previous studies on epidermicin NI01 and aureocin A54 (Hammond et al., 2020). Computational analysis of the spectra further estimated that α-helices may make up 75% of the peptide. It is well-known that the presence of α-helices and their enhanced conformational stability contribute significantly to the thermal stability of proteins (Yakimov et al., 2016). This observation provides a plausible explanation for the remarkable stability of Geo6 at high temperatures.

A recent study using the CD spectroscopy method confirmed the formation of helical structures in epidermicin NI01 and aureocin A53 in aqueous buffers. It also revealed that epidermicin NI01 undergoes denaturation at 60°C and exhibits a reversible two-state conversion between helical and unfolded forms (Hammond et al., 2020). Due to these interesting findings, we decided to employ a thermal shift assay, also known as differential scanning fluorimetry (DSF), to analyze the thermal denaturation of Geo6 under different pH ranges and temperatures. Additionally, to the best of our knowledge, this method has never been employed to study the stability and structural changes of bacteriocins.

The fluorescent dye (SYPRO Orange) used in this assay is highly fluorescent in non-polar environments, such as hydrophobic sites on unfolded proteins, compared to an aqueous solution where the fluorescence is quenched. By monitoring the increase in fluorescence, we can observe the binding of the dye to the hydrophobic amino acids of the unfolded peptide, which are normally hidden within the helical structure. Additionally, the DSF method enables the use of small sample volumes and low concentrations (Niesen et al., 2007).

The analysis showed that Geo6 remained in its unfolded state over a range of pH 3–7 during the increase of temperatures from 20°C to 95°C, and the formation of typical melt curves did not occur during the DSF assay (Figure 10). Such results imply that Geo6 remained extremely stable under the DSF conditions and stand in line with the results obtained in the spot on a lawn assay. Partial unfolding or thermal denaturation of Geo6 was detected in the range of pH 9–11, implying that in these conditions the peptide starts to denature (melting temperatures (T_m) of 76.5–80.5°C were determined). However, the observed melting curves did not indicate full denaturation or aggregation. A slight increase in relative fluorescence units (RFU) was also detected at pH 8, and the appropriate negative controls within the pH range of 9–11 also exhibited increased RFU. As a result, the findings of the DSF assay for Geo6 within the pH range of 8–11 are inconclusive and represent a major limitation of this study in accurately determining the stability of Geo6 within this pH range. Given that Geo6 has a pI value of 9.3, it can become negatively charged at pH levels higher than its pI, potentially leading to the destabilization of the bacteriocin. Despite the inconclusive results at pH 8–11, the denaturation or unfolding of Geo6 within the pH range 3–7 was not observed, which contradicts the findings of the previous study on epidermicin NI01 and aureocin A53 using CD spectroscopy analysis (Hammond et al., 2020).