S. thermophilus strains B59671 (NRRL, Agricultural Research Service Culture Collection, NCAUR, USDA, Peoria, IL), ST109, ST113, ST128 (in-house collection), LMG18311, LMD-9, and Streptococcus salivarius K12 (ATCC BAA1024) were grown in tryptone-yeast extract-lactose (TYL) broth at 37°C. Pediococcus acidilactici strain F (gift from B. Ray, University of Wyoming) and Lactobacillus helveticus 1942 (in-house collection) were grown in deMan, Rogosa, and Sharpe (MRS) medium. Escherichia coli strains DH5α and BL21, Enterococcus faecalis VA797, Enterococcus faecium H3, Bacillus cereus ATCC1457, and Listeria innocua GV9-LS were grown in Brain Heart Infusion (BHI) medium, and Streptococcus mutans strains GS-5 and UA159 were grown in Todd Hewitt Broth (THB) at 37°C. All strains were grown aerobically with E. coli, B. cereus and L. innocua broth cultures agitated at 200 rpm. Gene expression within the LMD-9 blp locus was induced by adding a final concentration of 250 ng/mL of synthesized BlpC30 (stock solution 1 mg/mL in H₂O) to the growth medium. BlpC30 (30 mer induction peptide) was prepared by microwave-assisted solid phase peptide synthesis with Fmoc-protected amino acids, using a CEM Liberty synthesizer (CEM Corp., Matthews, NC). The purified peptide was analyzed by HPLC and the molecular mass was confirmed by MALDI-TOF MS.

Bacteriocin activity was measured using an agar diffusion method (Gilbreth and Somkuti, 2005). Cultures of S. thermophilus B59671 and LMD-9 were grown to an OD₆₆₀ of ~0.8 in TYL and pelleted at 10,000×g for 10 min at 4°C for collection of cell-free supernatants (CFS). CFS (~50 μL; ~320 AU/mL) was loaded into precast wells in an agar medium (1.5%) seeded with the target bacteria (0.5% v/ v). Bacterial targets included: S. thermophilus ST109, ST113 (positive control strain), 128, and LMG18311, S. salivarius K12 in TYL; P. acidilactici (positive control strain) and L. helveticus in MRS; B. cereus, E. faecalis and E. faecium in BHI; and S. mutans GS-5 and UA159 in THB. Plates were allowed to equilibrate at 4°C overnight, and then incubated at 37°C and examined for the appearance of zones of inhibition. ST113 was routinely run as a negative control to show that inhibition zones were not due to acid production.

P. acidilactici F cells were grown overnight at 37°C, and then diluted in fresh MRS (25 fold) and grown until mid-log phase (O.D._{600 nm} = 0.5). Cells were harvested by centrifugation (10,000×g for 5 min at 4°C) and resuspended in 0.1% peptone water at the same optical density. Cells were then incubated at 37°C in the absence or presence of 510 AU/mL thermophilin 110. Culture O.D. was monitored for up to 3 h, and cells were collected at various times with 50 μL of bacteria placed on acetone-cleaned 12 mm Micro-cover glass slides (Thermo Scientific, Portsmouth, NH, United States) and allowed to adhere for 30 min. Samples were covered with 2 mL of 2.5% glutaraldehyde, (Electron Microscopy Sciences, Hatfield, PA, United States) and allowed to fix for 30 min. Samples were rinsed twice (30 min each) with 2–3 mL 0.1 M imidazole, (Electron Microscopy Sciences, Hatfield, PA, United States), followed by washing with 50, 80, and 90% ethanol (The Warner-Graham Company, Cockeysville, MD, USA) at 30-min intervals (2–3 mL each). The samples were then washed three times with 2 mL of 100% ethanol (30-min per wash) before critical point drying. The samples were stacked in a wire basket, separated by cloth, and placed in a Critical Point Drying Apparatus (Denton Vacuum, Inc., Cherry Hill, NJ, United States), using liquid carbon dioxide (Welco Co, Allentown, PA, United States) for approximately 20 min. The samples were mounted on stubs and sputter gold coated for 1 min (EMS 150R ES, EM Sciences, Hatfield, PA). Samples were then viewed with an FEI Quanta 200 F Scanning Electron Microscope (Hillsboro, OR, United States) with an accelerating voltage of 10 KV in high vacuum mode.

Sequencing of the blp gene cluster in S. thermophilus B59671 was performed by chromosome walking using purified genomic DNA as a template. The initial PCR primers were designed based on the nucleic acid sequence of LMD-9, corresponding to the 5′ end of blpA (5′-ATGTTTCGTTTTCGTAGG-3′) and the 3′ end of blpX (5′-AATTTTTACGATTCATTT-3′). After each round of sequencing, a new primer was designed based on the obtained sequence and walking was continued until the entire gene cluster was sequenced in both directions. Nucleic acid sequencing was performed using a 3730 PRISM^® DNA analyzer (Applied Biosystems, Foster City, CA) with the ABI PRISM Big Dye terminator cycle sequencing reagent. Sequences were analyzed and contigs were created using Sequencher 4.9 (Gene Codes Corp, Ann Arbor, MI). Open Reading frame (ORF) identification and promoter analysis were performed using Clone Manager software (Sci-ed Software, Cary, NC).

S. thermophilus strains were grown in TYL broth until the culture reached an OD₆₀₀ between 0.8 and 1.0. RNA was extracted using the RiboPure™-Bacteria kit (Ambion-Life Technologies, Grand Island, NY). To eliminate any residual DNA, RNA was treated with DNase I (Ambion) for 30 min at 37°C. RT-PCR was performed using a Lightcycler^® 96 (Roche Diagnostics Corp., Indianapolis, IN). Primers for blpC, blpD, blpU and 16S rRNA (control) are listed in Table 1. Cycling conditions were selected as the following: 40 cycles of 95°C for 30 s, 45°C for 30 s, and 60°C for 30 s. The melt curve analysis was performed from 60 to 95°C with fluorescence readings taken continuously after a 1% temperature increase. cDNA synthesis was performed with Transcriptor Universal cDNA Master (Roche) with 1 μg of total RNA. RT-PCR analysis was conducted with 0.015 ng cDNA using the Fast Start Essential DNA Green Master (Roche). Lightcycler^® 96 software was used to determine C_q values and relative quantification. Results were reported from a minimum of three independent RT-PCR reactions with the 16S rRNA gene used as a reference and shown to be stably expressed.

An integrative vector for knocking out blpK in S. thermophilus B59671 was constructed as previously described (Renye and Somkuti, 2013). The primers used to amplify 600 bp fragments of blpK flanking genes were designed as follows: orf6 Fwd (SpeI): CCCACTAGTCGCTAAGGGCTTACTTGA and orf6 Rev. (PstI): CCCCTGCAGGG-AAATCTTCCTCTAATT; and ISB Fwd (XhoI): CCCCTCGAGGCTTTGTTAGGTAATATC and ISB Rev. (KpnI): CCCGGTACCGAGGACAACATTCTCAATCG. The amplified fragments were cloned into pKS1 (Shatalin and Neyfakh, 2005) at the corresponding restriction sites, resulting in the plasmid, pSTKOK, and subsequently transformed and propagated in E. coli DH5α. The integrative vector pSTKOK was then electroporated into strain B59671 using a previously described protocol (Somkuti and Steinberg, 1988), with transformants selected on TYL containing 15 μg/mL erythromycin (erm) at 30°C. Transformants were sub-cultured in TYL+ erm at 37°C to allow for chromosomal integration of pSTKOK, and then sub-cultured in TYL at 30°C to allow for the second homologous recombination event to occur. A B59671 mutant with blpK replaced by a kanamycin resistance marker was confirmed by PCR using the primers: orf5Fwd: GAAAACAAAGATATGTATATTCGCCTA and ISB2Rev: AACGTTTTCAAGAGCGCAAT. S. thermophilus transformants were maintained in TYL containing 150 μg/mL kanamycin.

The nucleic acid sequence (162 base pairs) corresponding to the mature BlpU peptide (excluding the 23 amino acid leader peptide) was amplified from S. thermophilus B59671 with the primers: BlpU Fwd (BamHI): ATAGGATCCATCGAAGGTCGTGGATGTAGCTGGGGAGGT (nucleotides encoding the Factor Xa recognition sequence (isoleucine-glutamic acid-glycine-arginine) are italicized) and BlpU Rev. (XhoI): ATCTCGAGTCACCACCAGCATGTTGCTC. PCR was performed using the high-fidelity Pfx Taq polymerase (Life Technologies, Grand Island, NY); and the resulting DNA fragment was cloned into pGEX 6P-1 (GE Healthcare Life Sciences, Piscataway, NJ) at the corresponding restriction sites, downstream of the isopropyl β-D-1-thiogalactopyranoside (IPTG) inducible promoter. Cloning was confirmed by PCR using primers specific to blpU.

The plasmid was transformed into E. coli BL21 for recombinant expression of the fusion protein. E. coli containing pGEX 6P-1/blpU was grown overnight in BHI broth containing 100 μg/mL ampicillin (amp) at 37°C under agitation at 200 rpm. The culture was then diluted 1:100 into 100 mL BHI amp and grown at 37°C until the optical density (OD₆₀₀) reached 0.8–1.0, at which point IPTG was added at a final concentration of 0.1 mM. The culture was incubated for an additional 6 h, after which cells were collected by centrifugation at 10,000×g for 10 min, washed twice with phosphate-buffered saline (PBS) prior to resuspension in 20 mL PBS, and lysed by sonication (10 pulses of 30 s) using a microtip cell disruptor (Heat Systems Ultrasonics, Farmingdale, NY, United States). The lysate was centrifuged to remove cellular debris, then mixed with 1x bed volume of a 50% slurry glutathione Sepharose 4B resin (GE Healthcare) and incubated at 25°C for 2 h and 4°C overnight. The GST-BlpU fusion protein (~31 kDa) was visualized on a 12% Bis-Tris SDS-PAGE with Simply Blue™ SafeStain (ThermoFisher Scientific, Waltham, MA), was eluted from the Sepharose 4B resin as described by the manufacturer and dialyzed (12–14 kDa molecular weight cut off) against Factor Xa digestion buffer (20 mM Tris/HCl-100 mM NaCl-2 mM CaCl, pH 8.0) at 4°C overnight. The solution within the dialysis bag was recovered and analyzed by SDS-PAGE and then digested with 50 μL (1 mg/mL) of Factor Xa protease. The digested mixture was analyzed by 12% Bis-Tris SDS-PAGE with silver nitrate staining and analyzed for antimicrobial activity against P. acidilactici using a gel overlay method (Gilbreth and Somkuti, 2005).

The mature BlpU peptide (53 aa: GCSWGGFAKQGVATGVGNGLRLGIKTRTWQGA-VAGAAGGAIVGGVGYGATCWW) was synthesized and purified by high-performance liquid chromatography (Lifetein LLC, South Plainfield, NJ) to a purity of 85.9%. The peptide was dissolved in dimethyl sulfoxide (DMSO) at a stock concentration of 4.6 mg/mL. Further dilution to a working stock of 100 μg/mL was carried out in ddH₂O. Antimicrobial activity was determined in a well-diffusion assay using 50 μL of the working stock solution.

Chromosome walking was used to sequence the blp gene cluster within strain B59671, leading to a single 10.7 kb contig shown to contain 14 open reading frames (ORFs) (Figure 1). The resulting contig was aligned with the circularized whole genome sequence (WGS) of strain B59671 (Renye et al., 2017) to identify the location of the blp locus and confirm the accuracy of the WGS sequencing results. As reported for other S. thermophilus strains, the three-gene operon containing homologs of blpA, blpB, and blpC was directly adjacent to the two-gene operon encoding for the histidine kinase (BlpH) and response regulator (BlpR) required for inducing the expression of genes within the blp locus (Bolotin et al., 2004; Fontaine et al., 2007; Sun et al., 2011; Kang et al., 2012). When compared with the blp locus in LMD-9 (Accession number: CP000419), it was determined that the predicted proteins, BlpR and BlpH, differed by 7 and 11 amino acids, respectively, in B59671.

In strain B59671, the region between the blpRH operon and blpX contained two ORFs which encode putative antimicrobial peptides based on the presence of a double glycine (GG) leader sequence. The predicted mature peptide encoded by blpU showed 100% identity with the corresponding peptides in LMD-9, LMG18311 (Accession number: CP000023) and ST106, but differed with the peptide expressed in ST109 by a single amino acid substitution, valine to glycine at the N-terminus (Renye et al., 2019). In all strains blpU was part of a two-gene operon with orf3, which was predicted to encode the immunity protein required for protection against BlpU (Fontaine and Hols, 2008). The second ORF within the B59671 blp gene cluster was predicted to encode an 81 amino acid peptide with 98% identity to BlpK in strain CNRZ1066 (Hols et al., 2005), but differed more with the BlpK homologs in ST106 and ST109 due to the presence of a four-amino acid repeat (IGGA) insertion at positions 20–23 of the mature peptide (Renye et al., 2019).

The nucleic acid sequences upstream of the blpABC, blpU-orf3, and blpK operons were compared to the previously characterized blp-inducible promoters within the LMD-9 gene cluster (Fontaine et al., 2007). The blpABC promoter in strain B59671 was identical to the characterized region in LMD-9, including the presence of two imperfect direct repeats (DR), the left repeat (LR): ACCGTTTGGGACG and right repeat (RR): ACTTTTTGGGACG, shown to be essential for BlpC30-induced gene expression (Blomqvist et al., 2006). The promoters upstream regions of blpU-orf3 and blpK matched the promoter region upstream of the blpE/orf7/blpF operon in LMD-9. This promoter sequence differed from the blpU promoter in LMD-9 by two nucleotides in the LR sequence: ACCATTCGGGACA (B59671) and ACTACTCGGGACA (LMD-9) (Fontaine et al., 2007).

Previously it was reported that thermophilin 110 was naturally produced by S. thermophilus strain B5967; however, bacteriocin production by strain LMD-9 required induction with the BlpC quorum sensing peptide (BlpC30) at 250 ng/mL (Renye et al., 2016). Semiquantitative RT-PCR was used to compare the expression of blp components known to be upregulated by BlpC30 when strains B59671 and LMD-9 reached late exponential phase (OD₆₆₀ of ~0.8). The level of blpC mRNA was substantially higher in B59671 when compared to both non-induced and induced cultures of LMD-9 (Figure 2), similar to what was observed when transcription levels were compared between B59671 and ST106 (McAnulty et al., 2024). Following induction with the BlpC30 peptide, BlpC expression increased by 6-fold in LMD-9, ~16-fold lower than the expression level in B59671. Transcription of the blpU-orf3 operon was more than 45-fold higher in B59671 when compared to both induced and non-induced cultures of LMD-9 (Figure 2). Consistent with a previous report (Fontaine et al., 2007), transcription of the blpuU-orf3 operon responded poorly to induction with BlpC30 in LMD-9 (~3-fold). However, the expression of BlpD increased by ~45 fold following induction with BlpC30 in LMD-9 (Figure 2), a level of expression similar to what was observed for BlpU in B5967. In B59671, the expression level of BlpK was comparable to that of BlpU, which was expected since their promoter regions were identical (data not shown).

Thermophilin 110 was previously reported to induce cell lysis of P. acidilactici (Gilbreth and Somkuti, 2005) and Cutibacterium acnes (Renye et al., 2023) based on a drop in culture optical density (OD_600nm). In this study, when P. acidilactici cells were exposed to 512 AU/mL thermophilin 110, the optical density of the culture dropped from 1.06 (OD_600nm) to 0.299 (OD_600nm) by 2 h; and viable counts dropped from 8.49 log CFU/mL (T₀) to 5.96 log CFU/mL (T₂). Additionally, lysed cells were observed by SEM within 0.5 h (Figure 3B, arrows), with several fully lysed cells observed after 1.5 h (Figure 3C, arrows). These results confirmed the bactericidal activity of thermophilin 110.

Additionally, the activity spectrum of thermopilin 110 was confirmed using a selection of previously reported susceptible bacteria (Gilbreth and Somkuti, 2005; Renye and Steinberg, 2021), since a recent study revealed that S. thermophilus B59671 produces a second active bacteriocin, thermophilin 13 (McAnulty et al., 2024). Cell-free supernatants from B59671C⁻ maintained intraspecies activity against four S. thermophilus strains and showed broad-spectrum activity against B. cereus (Table 2), most likely due to the production of thermophilin 13. Antimicrobial activity was not observed against enterococci, Lb. helveticus, S. mutans, and S. salivarius when CFS from B59671C⁻ was tested, suggesting that these unique activities were dependent on the production of thermophilin 110.

To determine which genes encoded active bacteriocins within the B59671 blp gene cluster, the integrative vector pKS1 was used to construct deletion mutants for both blpU and blpK. A plasmid was successfully constructed to replace blpU with a kanamycin resistance maker (kan^r), but the knock out mutant was not obtained after several attempts (data not shown). However, the vector pSTKOK was generated and successfully used to delete the chromosomal copy of blpK from B59671. The mutant, S. thermophilus B59671K⁻, was sensitive to erythromycin and resistant to kanamycin, confirming a double crossover recombination event where kan^r was inserted within the chromosome. The mutant was further confirmed by PCR analysis using primers corresponding to the genes orf5 and ISSB, which flank blpK. Insertion of kan^r increased the PCR product size from 1.3 kB to 2.8 kB. Cell-free supernatant from an overnight culture of S. thermophilus B59671K⁻ (blpK knockout) was shown to inhibit the growth of P. acidilactici PAF (Figure 4, well 3), with a zone similar in size to the parent culture (well 1). However, ST B59671C⁻ did not inhibit pediococcal growth (Figure 4, well 2), validating that the active bacteriocin was encoded within the blp locus. These results suggested that BlpU was a bacteriocin with anti-pediococcal activity.

Since the creation of a blpU knockout strain was unsuccessful, overexpression of the predicted mature peptide (without the 23 amino acid leader sequence) was demonstrated using E. coli BL21 as the host bacterium. The plasmid pGEX 6P-1/blpU was constructed to express a GST-BlpU fusion protein, which also contained a Factor Xa cleavage sequence at the N-terminus of BlpU. The recombinant gene was located downstream of an IPTG inducible promoter, and transcription was induced when the culture reached an OD₆₀₀ ~ 0.8, with the culture incubated for an additional 6 h. After lysis of the induced E. coli cells, the fusion protein was recovered using 50% slurry of glutathione Sepharose 4B resin. The eluted protein was run on an SDS-PAGE gel which showed the presence of an expected fusion protein, ~31 kDa (Figure 5A, lane 2). The band corresponding to the fusion protein was not observed in uninduced cells (Figure 5A, lane 3), or in a lysate from E. coli BL21 without the expression plasmid, attesting that the protein band was not a host cell protein. The GST-BlpU fusion protein did not show antimicrobial activity against P. acidilactici, most likely due to the presence of the GST tag on its N-terminus (data not shown). After Factor Xa cleavage, a small peptide (~5 kDa) was observed by SDS-PAGE analysis after staining with silver nitrate (Figure 5B, lane 2), and a faint inhibition zone corresponding to the released peptide was observed when a washed SDS-PAGE was overlayed onto an agar medium inoculated with P. acidilactici (Figure 5C, lane 2). This confirmed that the GST tag must be removed to render BlpU active; and the location of the active BlpU on gel overlay corresponded to the antimicrobial peptide present in CFS from S. thermophilus B59671 (Figure 5C, lane 3). The inhibition zone results strongly suggested that BlpU is a broad-spectrum bacteriocin produced by this strain.

To further establish that BlpU is an active bacteriocin in the absence of other components encoded within the blp gene cluster of S. thermophilus B59671, the mature peptide was chemically synthesized. The resulting peptide (purity ~86%) was resuspended in DMSO at a concentration of 4.6 mg/mL. The peptide at a working concentration of 100 μg/mL (in sterile water) was applied to inhibit the growth of S. thermophilus ST113, P. acidilactici, L. inoccua, and S. mutans (Figure 6, column 2). Inhibition zones were comparable to the zones observed using a partially purified thermophilin 110 preparation (~512 AU/mL) as a control (Figure 6, column 1). Inhibition zones were not observed around wells loaded with a diluted DMSO solution as a negative control, evidencing the activities arose from the presence of BlpU (Figure 6, column 3). S. thermophilus B59671 was shown to be resistant to all bacteriocin preparations.