Skip to main content

Garvicin Q: characterization of biosynthesis and mode of action

This article has been updated

Abstract

Bacteriocins are ribosomally synthesized antimicrobial peptides, that either kill target bacteria or inhibit their growth. Bacteriocins are used in food preservation and are of increasing interest as potential alternatives to conventional antibiotics. In the present study, we show that Lactococcus petauri B1726, a strain isolated from fermented balsam pear, produces a heat-stable and protease-sensitive compound. Following genome sequencing, a gene cluster for production of a class IId bacteriocin was identified consisting of garQ (encoding for the bacteriocin garvicin Q), garI (for a putative immunity protein), garC, and garD (putative transporter proteins). Growth conditions were optimized for increased bacteriocin activity in supernatants of L. petauri B1726 and purification and mass spectrometry identified the compound as garvicin Q. Further experiments suggest that garvicin Q adsorbs to biomass of various susceptible and insusceptible bacteria and support the hypothesis that garvicin Q requires a mannose-family phosphotransferase system (PTSMan) as receptor to kill target bacteria by disruption of membrane integrity. Heterologous expression of a synthetic garQICD operon was established in Corynebacterium glutamicum demonstrating that genes garQICD are responsible for biosynthesis and secretion of garvicin Q. Moreover, production of garvicin Q by the recombinant C. glutamicum strain was improved by using a defined medium yet product levels were still considerably lower than with the natural L. petauri B1726 producer strain.

Collectively, our data identifies the genetic basis for production of the bacteriocin garvicin Q by L. petauri B1726 and provides insights into the receptor and mode of action of garvicin Q. Moreover, we successfully performed first attempts towards biotechnological production of this interesting bacteriocin using natural and heterologous hosts.

Introduction

Hyperacute haemorrhagic septicaemia, also termed lactococcosis, is a disease of fish species that is characterized by high mortality rates, and thus leads to economic losses in aquaculture [1, 2]. The closely related species Lactococcus garvieae and Lactococcus petauri were identified as the etiological agents of this disease in rainbow trout (Oncorhynchus mykiss) and grey mullet (Mugil cephalus) [36]. L. garvieae infections have also been reported for the crustacean giant freshwater prawn Macrobrachium rosenbergii [7]. Moreover, there is an increasing number of reports on L. garvieae as an emerging zoonotic pathogen infecting ruminants and humans [1, 8]. Infections of humans by L. garvieae are often associated with contact to contaminated raw fish [9, 10]. Interestingly, L. garvieae and L. petauri strains are widely distributed in the environment and were isolated not only from fish but also from various food products including raw milk, artisan cheese and vegetables [8, 1114].

Bacterial infections of fish are traditionally treated or prevented by antibiotics and thus it is not surprising that antibiotic resistances are widespread amongst pathogens associated with seafood [15, 16]. There are numerous reports on L. garvieae isolates with (multiple) resistances towards clinically relevant antibiotics like erythromycin and tetracycline [1]. As alternatives to conventional antibiotics, vaccination of fish, bacteriophages or probiotic bacteria have been investigated for their potential to combat L. garvieae infection [1719]. Another interesting approach is the application of antimicrobial peptides produced by bacteria, so called bacteriocins, in aquaculture systems [20].

Bacteriocins are ribosomally synthesized antimicrobial peptides that directly kill target bacteria or inhibit their growth [21] and provide the producing organisms with competitive advantages over other bacteria in their ecological niche [22]. L. garvieae strains have also been described to produce various bacteriocins including garvicin Q [23], garvicin KS [24], garvicin A, B and C [25, 26], garvicin ML [28], garviecin LG34 [29] and garviecin L1-5 [30]. Garvicins KS, Q, and ML and garviecin L1-5 have a broad antimicrobial spectrum, whereas garvicin A, B and C mainly kill L. garvieae strains. Hence, these bacteriocins are interesting candidates to treat lactococcosis. Recently, it has been demonstrated that garvicin KS is a promising candidate for protection of fish against L. garvieae infection [31]. Low concentrations of garvicin KS exhibited no cytotoxicity to fish cell lines and significantly increased survival of zebrafish larvae upon L. garvieae challenge, highlighting the potential of bacteriocins for infection control in aquacultures [31]. However, application of bacteriocins is limited by a lack of knowledge about the characteristics of the peptides and by expensive production of pure peptide [21].

In the present study, we identified L. petauri B1726 as garvicin Q producer, analysed the genetic locus for garvicin Q, characterised its mode of action, and investigated conditions for its production by L. petauri B1726 and C. glutamicum.

Material and methods

Strains and growth conditions

L. petauri B1726 was routinely cultivated in MRS medium unless stated otherwise. For optimization of garvicin Q production, L. petauri B1726 was cultivated statically in M17, 2xTY or BHI medium with 2% (w/v) glucose or lactose at 30 or 37 °C in glass tubes or Schott glass bottles as indicated in figure legends. Anaerobic cultivations were performed in a Millipore anaerobic jar with Oxoid AnaeroGen bags. Listeria spp. were cultivated in BHI medium at 37 °C under shaking conditions (130 rpm). Lactococcus lactis IL1403 was grown in M17 medium with 0.5% (w/v) glucose (GM17) overnight at 30 °C under shaking conditions (130 rpm). C. glutamicum strains were cultivated overnight in 2xTY at 30 °C under shaking conditions (130 rpm). For recombinant garvicin Q synthesis, C. glutamicum CR099/pPBEx2-garQICDCgl was cultivated in 2xTY with 2% (w/v) glucose or CGXII minimal medium [32] with 2% (w/v) glucose. Media were supplemented with 25 µg kanamycin ml−1 (C. glutamicum), 10 µg chloramphenicol ml−1 (Listeria spp. and L. lactis) and/or 0.2 mM IPTG (C. glutamicum) as appropriate. All strains are listed in Table 1.

Table 1 Bacterial strains used in the present study

Genome sequencing and annotation

Whole genome sequencing of L. petauri B1726 was performed using Illumina and Oxford Nanopore Technology (ONT) sequencing as follows. High molecular weight gDNA was prepared from a MRS overnight culture of L. petauri B1726 using the MagAttract HMW DNA kit (Qiagen) according to the manufacturer’s instructions for Gram-positive bacteria. Library preparation was done using the rapid barcoding kit SQK-RBK004 (ONT). ONT sequencing was carried out for 24 h using a FLO-FLG001 flow cell on a MinION platform with Flongle adapter controlled by MinKNOW v. 21.10.4 (all ONT). Base calling of raw data was done using the super-accurate base calling model of the guppy algorithm (v. 6.0.1, ONT) on a CUDA-capable RTX 3070 GPU (Nvidia). L. petauri B1726 reads were then filtered using filtlong v. 0.2.1 with recommended settings [40]. The library for Illumina sequencing was prepared using the Nextera Flex Library Preparation kit (Illumina). Sequencing was performed on an Illumina MiSeq system. Illumina short reads were filtered before further processing using fastp (Chen et al. 2018). Coverages were 21 ×for ONT and 74 ×for Illumina data. Both, the ONT and the Illumina data sets were then passed to the hybrid assembly pipeline of Unicycler v. 0.5.0 [41], which resulted in a single circular contig of 2.106 Mega base pairs. Subsequent short read polishing was done using Polypolish v. 0.5.0 [42] and POLCA (part of MaSuRCA v. 4.0.8, [43]). Automated annotation and identification of rRNA and tRNA genes of the polished consensus sequence was done using prokka v. 1.14.6 [44]. The annotated version of the L. petauri B1726 genome has been submitted to GenBank (Accession Number: CP094882.1). Raw sequencing data is available on the NCBI Sequencing Read Archive with accession numbers SRR18554891 (ONT data) and SRR18554892 (Illumina data).

Nucleotide and amino acid sequence analysis

The online tools BAGEL4 [45] and AntiSMASH version 6.0.1 [46] were used for prediction of bacteriocin gene clusters. Terminator and promoter prediction was done with ARNold [47], BPROM and FGENESB [48] online tools. JSpeciesWS version 3.9.0 [49] was used for whole genome sequence alignment. BLASTp [50] and Clustal Omega [51] was used for protein sequence alignment. Protter version 1.0 [52] and PPM 3.0 web server [53] were applied for prediction of membrane topology of selected proteins. To gain structural insights, deep learning modelling method RoseTTAFold of Robetta [54] was used for tertiary structure prediction. The complex of group I PTSMan subunits IIC and IID and garvicin Q was predicted using AlphaFold [55]. Visualization of protein structures was done with ChimeraX [56].

Molecular biology procedures

Cloning procedures were performed using standard reagents according to protocols of the manufacturers. Primers and codon-optimized genes were purchased from a commercial service provider (Eurofins Genomics) and sequences are listed in Additional file 1: Table S1. For construction of pPBEx2-garQICDCgl, sequences of the L. petauri B1726 genes garQ, garI, garC, and garD were codon-optimized for C. glutamicum. The codon-optimized genes were amplified by PCR using Q5 high fidelity polymerase (New England Biolabs) and specific primers. The empty plasmid pPBEx2 [57] was linearized with PstI and KpnI and ligated to PCR products in an isothermal reaction [58]. E. coli DH5α was transformed with the assembled plasmid pPBEx2-garQICDCgl and the plasmid sequence was confirmed by Sanger sequencing by a commercial service provider (Microsynth Seqlab). Transformation of C. glutamicum CR099 with pPBEx2-garQICDCgl was performed according to a standard protocol [59]. Construction of pNZ-pHin2Lm was described elsewhere [34] and transformation of L. lactis IL1403 was achieved using a standard protocol [60].

Overlay agar plate assay

The overlay agar plates were prepared as described previously [61] with slight modifications. An appropriate dilution of a L. petauri B1726 overnight culture was spread on GM17 agar plates and immediately overlayed with fresh GM17 agar. After two days of incubation at 30 °C single colonies of L. petauri B1726 were observed. The sandwiched colonies were then overlayed with BHI agar containing Listeria innocua LMG2785 diluted to an optical density at 600 nm (OD600) of ~ 0.002. The agar plates were then incubated for one day at 30 °C and plates were imaged in an iBright FL1000 System.

Growth inhibition assay

To determine the antimicrobial activity of garvicin Q-containing samples a standard microtiter plate assay [62] was performed with slight modifications. Two-fold serial dilutions of the sample were prepared with BHI or GM17. The indicator strains L. innocua LMG2785 or L. lactis IL1403 were cultivated overnight and diluted in fresh medium to an OD600 of 0.2 prior to the assay. Diluted samples (100 µl) were mixed with the indicator strain suspension (100 µl) in a sterile 96-well plate corresponding to a start OD600 of 0.1. The microtiter plate was incubated at 37 °C for 4–5 h with shaking and growth of the indicator strain was determined by measuring the OD600 in an Infinite M200 plate reader (Tecan). Bacteriocin activity was determined in a semi-quantitative manner as described previously [62]. Briefly, bacteriocin units per ml (BU ml−1) were calculated as the reciprocal of the highest dilution showing at least 50% inhibition of the indicator strain.

pHluorin2 assay

The pHluorin2 assay was performed according to a previously described protocol using reporter strains L. innocua LMG2785/pNZ-pHin2Lm [34] and L. lactis IL1403/pNZ-pHin2Lm (this study). These reporter strains constitutively express the pH-sensitive fluorescent protein pHluorin2, which shows a bimodal excitation spectrum with maxima at 400 and 480 nm. If pH homoeostasis is disrupted by membrane-damaging compounds, intracellular pH drops to the level of the assay buffer (pH 6.2) and this leads to a ratiometric change in the fluorescence intensity of pHLuorin2 at the two excitation peaks. For assays, reporter strains were cultivated overnight, washed once with phosphate-buffered saline, and then resuspended in LMB buffer (pH 6.2) to an OD600 of 3. Two-fold serial dilutions of samples were prepared in a similar manner as described for the growth inhibition assay and 100 µl of sensor strain suspension was mixed with 100 µl of sample dilution. The mixture was incubated for 30 min in the dark at room temperature. Fluorescence intensities (emission at 520 nm) were determined with excitation at 400 and 480 nm. The ratios of emission intensities after excitation at 400 and 480 nm were calculated.

Purification and identification of garvicin Q

First purifications of garvicin Q were performed following an established protocol [63]. The peptide was purified by ammonium sulphate (AS) precipitation, cation exchange chromatography (CIEX) and reversed-phase chromatography (RPC) as described previously [63]. Garvicin Q eluted at ~ 49% elution buffer during RPC.

For later purifications an optimized purification protocol including hydrophobic interaction chromatography (HIC) was applied. L. petauri B1726 was cultivated overnight at 30 °C in GM17 (2% glucose) and the culture supernatant (SN) was collected by centrifugation. Then, 0.64 M ammonium sulphate was slowly added to the SN and the pH was set to 6 to match conditions during HIC. The sample was then again centrifuged to remove insoluble components. The 1 ml HiTrap Octyl FF column (Cytiva) was equilibrated with 5 column volumes of 50 mM sodium phosphate buffer at pH 6 with 0.64 M ammonium sulphate. Then ~ 100 ml of the SN retained after AS precipitation was loaded onto the column and washed with 20 column volumes of equilibration buffer. Garvicin Q was eluted with 10 column volumes of HPLC water. Fractions containing bacteriocin activity (in total 4 ml) were subjected to RPC as described earlier [63] using acetonitrile as a mobile phase instead of 2-propanol. Garvicin Q eluted at ~ 30% elution buffer. The garvicin Q preparation was either stored at 4 °C or dried under vacuum for long-term storage.

Mass spectrometry analysis was performed based on the method described in [64]. Protein fractions were concentrated to 1 mg ml−1 for tryptic digestion with 1 μg trypsin in a total volume of 100 μl for 5 h at 42 °C as recommended by the supplier. Peptide solutions were diluted 1:2 with LC–MS-grade H2O prior LC–MS measurements.

Liquid chromatography mass spectrometry (LC–MS) was conducted with an Agilent 1260 Infinity system (Agilent Technologies, Waldbronn, Germany) coupled to a quadrupole time-of-flight mass spectrometer (TripleTOF6600, AB Sciex, Darmstadt, Germany). LC was performed with an Ascentis® Express Peptide ES-C18, 2.7 μm HPLC column (53,307-U, Merck, Darmstadt, Germany) with a flow rate of 200 µL min−1 and the mobile phases (A) 0.1% formic acid in water and (B) acetonitrile. The elution gradient was as follows: 0 min, 3% B; 70 min, 40% B; 78 min 40% B, 79 min 60% B, 89 min 60% B, 90 min 3% B followed by a 12 min equilibration time between injections. Column temperature was set to 21 °C and injection volume to 10 µL. MS was conducted with a TurboV ion source operated in positive ionization mode. Ion spray voltage was set to 5.5 kV, source temperature to 450 °C, curtain gas to 35 psi, and the support gases GS1/GS2 to 50 psi/50psi. All gases were nitrogen. The quadrupole time-of-flight (QToF) mass spectrometer was operated in data-dependent acquisition (DDA) mode. Based on a ToF survey scan with a dwell time of 250 ms, product ion scans with a dwell-time of 100 ms were automatically performed for a maximum of 40 ions with a mass-to-charge ratio > 300 m/z, a charge of 2–4 passing an intensity threshold of 150 cps. After acquisition, mass-to-charge ratios were excluded from the potential candidate-ion list for 12 s. Declustering potential was set to 120 V, collision energy spread to 5 V and mass tolerance to 25 ppm. Acquired mass spectra were analyzed with PeakView 2.1 (AB Sciex, Darmstadt, Germany). Protein identification was performed with the ProteinPilot 5.1 software (AB Sciex, Darmstadt, Germany). Complete results of the MS analyses are provided in Additional files 2, 3.

SDS-PAGE, silver staining and zymogram analysis

SDS-PAGE and silver staining of garvicin Q samples were performed according to a previously published protocol [63]. For zymogram analysis the silver-stained gel was washed for 3 h in HPLC-grade water and then placed on a BHI agar plate containing embedded L. innocua LMG2785 at an OD600 of ~ 0.002. The agar plate was incubated overnight at 37 °C and imaged with a digital camera.

Adsorption assay

Adsorption of garvicin Q to L. innocua LMG2785, L. lactis IL1403, L. monocytogenes EGDe, L. monocytogenes EGY2, L. petauri B1726 and C. glutamicum ATCC13032 cells was analysed as described previously [65] with modifications. Overnight cultures were washed three times with phosphate-buffered saline. Bacteria were then resuspended at an OD600 of 3 in 5 mM sodium phosphate buffer (pH 6.5) containing HIC purified garvicin Q. Where indicated, a cocktail of protease inhibitors (cOmplete™ Protease Inhibitor Cocktail, Sigma) with 1 mM EDTA was added to assays to exclude proteolytic degradation of garvicin Q. The peptide-cell mixture was incubated for 1 h at 30 (Lactococcus spp. and C. glutamicum) or 37 °C (Listeria spp.). Then, bacteria were precipitated by centrifugation and bacteriocin activity was measured in the adsorption supernatants (ASN) using a growth inhibition assay with L. lactis IL1403 as described above. The percentage of adsorbed garvicin Q was calculated according to following formula:

$$\mathrm{adsorption} \left[\%\right]=100-\left(\frac{\mathrm{bacteriocin\, activity\, in\, ASN}}{\mathrm{original\, bacteriocin \,activity}}\times 100\right)$$

Time-kill assay

Time-kill assays were performed with L. innocua LMG2785 and L. lactis IL1403 [66]. BHI or GM17 were inoculated with the respective strains to an initial OD600 of 0.5. HIC purified garvicin Q (16-fold diluted), chloramphenicol (10 µg ml−1), nisin (100 µg ml−1) or H2O was added to the cultures. After 0, 2, 4, and 6 h tenfold serial dilutions of the cultures were plated on BHI or GM17 agar plates. The agar plates were incubated overnight at 30 or 37 °C and colony forming units (cfu) were determined by counting single colonies in appropriate dilutions.

Results

Lactococcus sp. B1726 produces a protease sensitive, amphiphilic bacteriocin

Lactococcus sp. B1726 was initially isolated from fermented balsam pear during a screening for potential bacteriocin producers with activity against pathogenic L. garvieae [67]. To corroborate these results, we tested growth inhibition of Listeria innocua LMG2785 by Lactococcus sp. B1726 colonies on overlay agar plates (Fig. 1A). A clear zone of inhibition surrounding single colonies of Lactococcus sp. B1726 was observed indicating the production of a compound with activity against L. innocua LMG2785. This result was confirmed in growth inhibition assays in a microtiter plate format using serial dilutions of culture SNs of Lactococcus sp. B1726 (Fig. 1B). While untreated Lactococcus sp. B1726 SNs showed potent inhibition of growth of L. innocua LMG2785, treatment of SNs with trypsin efficiently abolished this activity. By contrast, heat treatment (80 °C, 15 min) of SNs had no effect.

Fig. 1
figure 1

Lactococcus sp. B1726 produces a heat-stable and protease sensitive bacteriocin. A Overlay agar plates of Lactococcus sp. B1726 colonies with L. innocua LMG2785 as indicator strain. The red arrow marks a single colony of Lactococcus sp. B1726 and the white arrow highlights a zone of inhibition surrounding Lactococcus sp. B1726. B Growth inhibition assay of serial twofold dilutions of supernatants (SN) of Lactococcus sp. B1726. Where indicated SN were treated with trypsin (Trp) or incubated at 80 °C for 15 min (heat). MRS medium served as negative control. L. innocua LMG2785 was used as indicator strain. Values are mean ± standard deviation of n = 3 independent cultures. C Silver stained SDS-gel (left) and zymogram (right) of RPC fractions (C7-C9) of Lactococcus sp. B1726 SN. The black arrow indicates the antimicrobial active peptide. M = Spektra Low range ladder

To test the cationic and hydrophobic character of the Lactococcus sp. B1726 antimicrobial protein or peptide, the compound was concentrated by ammonium sulphate (AS) precipitation, cation exchange chromatography (CIEX) and reverse-phase chromatography (RPC; Additional file 1: Figure S2). Zymogram analysis of RPC elution fraction revealed a single band of a molecular mass of approx. 5 kDa with antimicrobial activity against L. innocua LMG2785 (Fig. 1C). These findings suggest that Lactococcus sp. B1726 secretes a heat-stable and protease-sensitive antimicrobial protein or peptide and are consistent with characteristics of low molecular weight bacteriocins.

Genome analysis and identification of the putative gar operon of Lactococcus sp. B1726

The genome of Lactococcus sp. B1726 was sequenced using Illumina and ONT and assembled to a single circular contig. The genome has a size of 2,106,259 bp and a GC content of 38%, which is in line with previously sequenced L. garvieae and L. petauri genomes [14, 68, 69]. Initial whole genome sequence alignment proposed that the isolate belongs to the species L. garvieae as the closest relative was L. garvieae PAQ102015-99 with an average nucleotide identity (ANI) of 99.94% and a coverage of 97.39% (Additional file 1: Table S3). However, more recent ANI analyses suggest that L. garvieae PAQ102015-99 should be reclassified to the species Lactococcus petauri [6, 69], which was first isolated from an abscess of a sugar glider (Petaurus breviceps) and more recently was also described to cause lactocococcis in fish [6, 69]. According to ANI, the most closely related type strain of our isolate was L. petauri 159,469 (98.55%) whereas the L. garvieae type-strain ATCC 49,156 showed lower ANI (93.30%) (Additional file 1: Table S3). Consequently, we classified our isolate B1726 the species Lactococcus petauri.

We searched the annotated L. petauri B1726 genome for putative virulence factors and genes described to be essential for survival of other pathogenic L. garvieae and L. petauri strains in the host [6, 8, 7072]. We identified various genes coding for adhesins, stress response proteins, and proteins involved in host-adapted metabolism (Additional file 1: Table S4). Moreover, L. petauri B1726 harbours genes involved in modification of the cell envelope including phosphoglucomutase and D-alanine-D-alanyl carrier protein ligase which may play a role in immune evasion. Of note, a gene coding for a putative hemolysin A was identified. Indeed L. petauri B1726 showed clear α-haemolytic activity on blood agar plates after incubation at 37 °C (Additional file 1: Figure S5). Interestingly, no lacG homologue for phospho-β-galactosidase was identified in the genome of L. petauri B1726. This indicates that the strain is not able to utilize lactose as a carbon source.

To identify potential bacteriocin gene clusters, the genome of L. petauri B1726 was searched with the online tools BAGEL4 and antiSMASH. Both tools identified one region of interest consisting of four genes that were designated as garQICD (Fig. 2A, Additional file 1: Figure S6). The first gene, garQ (lgb_01492), is predicted to encode a peptide precursor identical to the previously described garvicin Q [23]. Similarly, garI (lgb_01493), encodes a putative bacteriocin immunity protein with similarity to an enterocin A immunity protein (PF08951). The genes garCD (lgb_01494 and lbg_01495) encode for an ABC-transporter with a C-terminal C39 peptidase domain (PF00005, PF00664 and PF03412), and an accessory secretion protein with a HlyD typical motif (PF13437). Of these four genes, garQ, garI and garC are almost identical to the genes garQ, garI and garT of L. garvieae BCC 43,578 (GenBank Accession Number JN605800) previously shown to be associated with garvicin Q production [23]. In silico analysis using BPROM predicted promoters upstream of garQ and garC and putative rho-independent transcription terminators were identified downstream of garI and garD. This suggests that the four genes are organised in two bicistronic operons with transcripts for garQI and garCD. A similar prediction was made for the garvicin A gene cluster of L. garvieae 21,881 [25].

Fig. 2
figure 2

Putative bacteriocin gene cluster of L. petauri B1726 for production of garvicin Q. The cluster consists of four genes: garQ for the pre-peptide, garI for an immunity protein, garC and garD for the transporter. The cluster contains two predicted promoters upstream of garQ and garC (indicated by arrows) and two Rho-independent transcription terminators downstream of gari and garD (indicated by lollies). Pre-garvicin Q consists of a N-terminal signal sequence (bold) with a conserved double glycine cleavage site (underlined). Pre-garvicin Q is likely processed by GarC during secretion. GarI shows similarity to an enterocin A immunity protein (PF08951). GarC contains an intracellular C39 peptidase domain (red, PF03412), an ABC-transporter transmembrane part (blue, PF00664) and an intracellular ATP-binding domain (green, PF00005). GarD contains an extracellular HlyD -typical motif (light blue, PF13437)

The deduced amino acid sequences of GarC and GarD have a high sequence similarity to LgnC (97.62%) and LgnD (96.21%), i.e., the garvicin A transporter of L. garvieae (UniProtKB H2B2W2 and H2B2W2). GarC and GarD also have a similar amino acid sequence to LcnC (73.01%) and LcnD (56.96%), i.e. the lactococcin A transporter of L. lactis subsp. lactis (UniProtKB Q00564 and P0A3G5). Moreover, the N-terminal signal sequences of garvicin Q, garvicin A and lactococcin A share a conserved motif including a double glycine required for cleavage of the signal sequence by the transporter [26] (Fig. 2). In silico prediction of GarD membrane topology showed that it consists of a short N-terminal intracellular part and a long α-helical extracellular part (Fig. 2) and shares similarity with HlyD and LcnD 27. Altogether, in silico analyses indicated that L. petauri B1726 possesses all components required for the synthesis, immunity, and secretion of the bacteriocin garvicin Q.

Optimization of cultivation conditions for garvicin Q production

To increase the yield of garvicin Q for further analyses, the impact of different cultivation conditions and media on the production of garvicin Q by L. petauri B1726 was tested (Additional file 1: Table S7) and antimicrobial activity (BU/ml) was determined after o/N cultivation. In MRS medium, SNs obtained after static cultivation showed higher activities compared to SN of shaken L. petauri B1726 cultures (Fig. 3A), which may be associated to reduced growth and more alkaline pH in shaken cultures (Additional file 1: Table S7). To exclude a detrimental effect of oxygen on production and stability of garvicin Q, L. petauri B1726 was cultivated statically under aerobic and anaerobic conditions. No differences in final optical densities or activity in SNs were observed between aerobic and anaerobic conditions (Fig. 3B), indicating that strictly anaerobic conditions are not required for garvicin Q production but shaking leads to lower biomass yield and reduced activity in SN. The cultivation temperature affected the garvicin Q production by L. petauri B1726 (Fig. 3C).

Fig. 3
figure 3

Antimicrobial activity in SN of L. petauri B1726 cultivated under different conditions. L. petauri B1726 was grown in MRS medium overnight at 30 °C under static conditions or with shaking A, statically under aerobic and anaerobic conditions B, statically at 25, 30 or 37 °C C or in the indicated media statically at 30 °C D. Activity was tested by incubating the indicator strain L. innocua LMG2785 with serial twofold dilutions of supernatants (SN). Values are mean ± standard deviation of SN of n = 3 independent cultures

Interestingly, growth of L. petauri B1726 and pH of the cultures were comparable at 25, 30, and 37 °C (Additional file 1: Table S7) but highest activities were obtained with cultures grown at 30 °C (Fig. 3C). Also, the impact of different media on antimicrobial activity in SNs of L. petauri B1726 was investigated in static cultures (Fig. 3D, Additional file 1: Table S7). Very poor growth and activity was observed in M17 + 2% lactose, indicating that L. petauri B1726 is unable to efficiently consume lactose. This fits nicely to the lack of a lacG homologue in the genome of L. petauri B1726. Final optical densities (OD600) of L. petauri B1726 in 2xTY + 2% glucose, BHI + 2% glucose, and MRS medium were comparable. However, activities in TY and BHI were marginal compared to MRS. Cultivation of L. petauri B1726 in M17 + 2% glucose (30 °C, static) resulted in the highest final OD600 and activities (OD600 2.8, 1280 BU/ml).

Since M17 + 2% glucose performed best under the tested conditions, this medium and static cultivation at 30 °C were used for further experiments. To determine the best timepoint for collection of SNs, OD600, antimicrobial activity and pH were measured during cultivation of L. petauri B1726 (Fig. 4A, B). Under these conditions, the growth rate of L. petauri B1726 in exponential phase was 0.67 h−1 and cultures entered stationary phase after 8 h at an OD600 of 2.8. The pH of the cultures continuously decreased during growth and reached a value of 4.4 after 24 h of cultivation. Antimicrobial activity was detected first after 4 h of cultivation (OD600 approx. 0.8, pH 6.6, 160 BU/ml), reached a maximum after 8 h (OD600 of 3.5, pH 5.1, 1280 BU/ml), and remained stable until the end of the experiment. After 24 h, SNs were collected and subjected to HIC and subsequent RPC (Fig. 4C). Here, AS precipitation and CIEX were replaced by HIC, since it was observed that AS precipitation results in problems with solubility of the hydrophobic garvicin Q (Additional file 1: Table S8). Combined RPC elution fractions containing antimicrobial activity were analysed by mass spectrometry (Fig. 4D, Additional file 1: Figure S9). Three distinct peaks with a mass/charge ratio (m/z) of 1069.1500, 891.1142 and 763.9628 were observed corresponding to the five-, six and seven-fold charged ions of garvicin Q, respectively (Fig. 4C and Additional file 1: Figure S9). The experimentally determined monoisotopic mass of 5346.72 Da fits to the calculated monoisotopic mass of garvicin Q (5340.63 Da). MS of the samples after tryptic digestion confirmed the identity of garvicin Q (Additional File 2).

Fig. 4
figure 4

Kinetics of production, purification, and mass spectrometry analysis of garvicin Q of L. petauri B1726. A Optical density (OD600, solid line) and pH (dashed line) of L. petauri B1726 cultures grown in M17 medium supplemented with 2% glucose. The cells were incubated statically in 50 ml Schott glass bottles at 30 °C. B Antimicrobial activity in SN from (A) harvested at indicated time points. Activity was tested by incubating the indicator strain L. innocua LMG2785 with serial two-fold dilutions of SN. Values in (A) and (B) are mean ± standard deviation of n = 3 independent cultures. C Hydrophobic interaction chromatography (upper panel) and subsequent reverse-phase chromatography (lower panel) to purify garvicin Q from L. petauri B1726 SN after cultivation for 24 h in M17 medium supplemented with 2% glucose A. D LC–MS analysis of combined RPC elution fraction harbouring antimicrobial activity. The peaks with mass to charge ratios (m/z) of 1069.1500, 891.1142 and 763.9628 correspond to garvicin Q with five, six or seven positive charges, respectively

Previous reports have shown that garvicin Q from L. garvieae BCC 43,578 exhibits a broad antimicrobial spectrum and requires the IIC and IID subunits of the PTSMan as receptors at target cells for activity [23, 73]. To confirm the PTSMan-dependent antimicrobial activity of garvicin Q, we tested susceptibility of different bacteria either harbouring a group I PTSMan (L. lactis IL1403, L. innocua LMG2785, L. monocytogenes EGDe) or not (L. monocytogenes EGY2 and C. glutamicum ATCC13032) to purified garvicin Q (Fig. 5A). Growth of L. monocytogenes EGDe, L. innocua LMG2785, and Lactococcus lactis IL1403 were inhibited by purified garvicin Q. By contrasts L. monocytogenes EGY2, which harbours a deletion in the extracellular γ-region of the PTSMan IID subunit rendering the strain resistant to the class II bacteriocins mesentericin and pediocin [36, 63], and C. glutamicum ATCC13032 that does not possess a PTSMan were completely resistant. Interestingly, a potential interaction of garvicin Q with the IIC and IID subunits of PTSMan could be predicted in silico (Additional file 1: Figure S10). The predicted garvicin Q–PTSMan complex is similar compared to the Cryo-EM structure of the pediocin–PTSMan complex ([74] and Additional file 1: Fig. 11). PTSMan residues previously identified to be involved in interaction with garvicin Q [73] and pediocin [74] were located in close vicinity to garvicin Q. This suggests a similar mode of action of pediocin and garvicin Q. In summary, these findings support the hypothesis of Tymoszewska et al. [73] that the PTSMan serves as receptor for garvicin Q.

Fig. 5
figure 5

Mode of action of garvicin Q. A Sensitivity of L. monocytogenes EGDe, L. monocytogenes EGY2, L. innocua LMG2785, L. lactis IL1403 and C. glutamicum ATCC13032 towards purified garvicin Q. B and C Viable cell count (log cfu/ml) of L. lactis IL1403 B and L. innocua LMG2785 C after treatment with garvicin Q (GarQ). As controls, bacteria were treated with nisin (Nis; bacteriolytic), chloramphenicol (Cm; bacteriostatic) or water (H2O; negative control). D Fluorescence intensity ration with emission at 520 nm after excitation at 400 and 480 nm (ratio RFU 400/480) biosensors L. innocua LMG2785/pNZ-pHin2Lm (Lin) and L. lactis IL1403/pNZ-pHin2Lm (Lla) following treatment with a garvicin Q RPC preparation (GarQ), and nisin (6.25 µg/ml; Nis) compared to unreated controls (LMB). All values are mean ± standard deviation of n = 3 independent cultures

We then investigated the binding of garvicin Q to the same set of bacteria plus the native producer L. petauri B1726 by adsorption assays with purified garvicin Q (Table 2). Activity in garvicin Q preparations was strongly reduced (> 80%) after incubation with cells of all strains except L. petauri B1726, i.e., the natural producer. This indicates that garvicin Q binds to a wider range of bacteria, but antimicrobial activity requires the presence of a functional PTSMan. To rule out proteolytic degradation as the mechanism of reduced garvicin Q activity following incubation with bacterial biomass, we performed additional adsorption assays in the presence and absence of protease inhibitors with representatives of susceptible (L. innocua LMG2785) and resistant (C. glutamicum ATCC13032) bacteria. In both cases, garvicin activity was largely abolished following incubation with bacteria and addition of the protease inhibitor cocktail had no effect, suggesting that proteolytic degradation of garvicin Q is not responsible for the loss in activity (Fig. 6).

Table 2 Adsorption of garvicin Q to different Gram-positive bacteria with or without a group I PTSMan
Fig. 6
figure 6

Garvicin Q adsorption assay in the presence of protease inhibitors. Activity of garvicin Q (GarQ, red lines) in following incubation with C. glutamicum ATCC13032 (A; Cgl, blue lines) or L. innocua LMG2785 (B; Lin, blue lines) was tested with (solid lines) or without (broken lines) presence of a protease inhibitor cocktail with 1 mM EDTA (PI). Activity was tested by measuring optical density (OD600) after incubation of the indicator strain L. lactis IL1403 with serial twofold dilutions of supernatants of adsorption assays (ASN) with the indicated treatment combinations. Values are mean ± standard deviation of n = 3 independent assays

To further investigate the mode of action of garvicin Q, viable cell counts of two bacteria targeted by garvicin Q (L. innocua LMG2785 and L. lactis IL1403) were determined after treatment with this bacteriocin (Fig. 5B and C). Both target bacteria showed similar behaviour. Treatment with chloramphenicol had a clear bacteriostatic effect on both indicator bacteria and growth was observed in samples treated with water. Similar to the bactericidal bacteriocin nisin, treatment with garvicin Q led to a reduction in viable cell count over time suggesting a bactericidal mode of action of garvicin Q. This effect was more pronounced for L. lactis IL1403.

To identify the underlying mechanism of the bactericidal effect of garvicin Q, we conducted assays with previously described L. innocua [34] and newly generated L. lactis pHluorin2 biosensor bacteria (Additional file 1: Figure S12). These sensors allow detection of intracellular pH-shift due to disruption of membrane integrity by e.g. bacteriocins [12, 63, 75, 76]. Both sensor strains clearly showed perturbed intracellular pH following treatment with nisin or garvicin Q as indicated by a drop in fluorescence intensity ratios compared to the untreated controls (Fig. 5D). This points towards a membrane-damaging mechanism as described for e.g. pediocin [74, 77].

Heterologous expression of garQICD results in synthesis of garvicin Q

To conclusively demonstrate that the genes of the gar locus are responsible for biosynthesis of garvicin Q, they were expressed in C. glutamicum. This organism was selected as a host as it was shown to be resistant to garvicin Q, produces no own bacteriocin and was recently successfully engineered for production of the bacteriocins pediocin PA1 and (pre)nisin [63, 76]. The sequences of garQ, garI, garC, and garD were optimized for codon usage of C. glutamicum, each equipped with the strong ribosome binding site of the tuf gene of C. glutamicum and cloned into the vector pPBEx2 under control of IPTG-inducible tac promoter. The obtained plasmid pPBEx2-garQICDCgl (Additional file 1: Figure S13A) was introduced into C. glutamicum CR099. First experiments with a standard protocol for induction only in the main culture did not yield detectable activities in SNs of C. glutamicum CR099/pPBEx2-garQICDCgl (Additional file 1: Figure S13B). We thus turned to a protocol previously successfully used for production of prenisin [76] that includes induction of precultures using 0.05 mM IPTG and main cultures using 0.2 mM IPTG (Additional file 1: S13C). Activity assays on these SNs indicated presence of an antimicrobial (Additional file 1: figure S12B). This activity was observed mainly in the early phases of cultivation, i.e. after 4 h, and largely reduced at the end (24 h).

To further improve heterologous production of garvicin Q, further cultivations of C. glutamicum CR099/pPBEx2-garQICDCgl were performed in defined CGXII medium (Fig. 7). Surprisingly, no activity was detected in SNs under these conditions despite growth to high optical densities (Fig. 7B). The pH steadily increased over the course of the fermentation (Fig. 7A). As highest garvicin Q levels in SNs of the natural producer L. petauri B1726 are observed in late exponential to stationary phase when cultures had a pH of 4–5 [78], we hypothesized that acidic pH during cultivation might increase garvicin Q production. Thus, we performed additional experiments in CGXII medium without urea to allow acidification. Indeed, lack of urea resulted in a continuous acidification of culture during growth (Fig. 7A). Moreover, significant activity was detected in C. glutamicum CR099/pPBEx2-garQICDCgl SNs at all timepoints during the cultivation on CGXII medium without urea (Fig. 7B) despite significantly lower final OD600. Moreover, activity steadily increased over the course of the experiment and reached levels significantly higher than in the initial experiments on 2TY medium with glucose (Additional file 1: Figure S13). To confirm the identity and correct processing of the heterologous produced garvicin Q, the recombinant peptide was purified by HIC and RPC (Fig. 7C) as described for GarQ of the natural producer L. petauri B1726. LC–MS analysis of the RPC fraction containing antimicrobial activity yielded a signal with m/z = 1069.2010 corresponding to the five-fold charged ion of garvicin Q (Fig. 7D and Additional file 1: Figure S14). MS of the samples after tryptic digestion confirmed the identity of recombinant garvicin Q (Additional File 3).

Fig. 7
figure 7

Heterologous expression of garQICDCgl in C. glutamicum. A Optical density (OD600; solid lines) and pH (broken lines) in supernatants of C. glutamicum CR099/pPBEx2-garQICDCgl cultivated in CGXII + 2% glucose with (red lines) or without (blue lines) urea. B antimicrobial activity in SNs of these cultivations sampled at the indicated timepoints. Activity was assayed using L. lactis IL1403/pNZ-pHin2Lm as indicator. Values are ratios of fluorescence intensity at 520 nm after excitation at 400 and 480 nm (ratio RFU 400/480) and are mean of n = 3 independent experiments (supernatants of independent cultivations). C Purification of garvicin Q from SN of C. glutamicum CR099/pPBEx2-garQICDCgl after cultivation for 24 h in CGXII medium + 2% glucose without urea by hydrophobic interaction chromatography (HIC; upper panel) and subsequent reverse-phase chromatography (RPC; lower panel). D LC–MS analysis of pooled RPC elution fractions harbouring antimicrobial activity. The peak with a mass to charge ratio (m/z) of 1069.2010 corresponds to garvicin Q with five positive charges

Discussion

The class IId bacteriocin garvicin Q was previously shown to be produced by L. garvieae [23]. It is active against a wide spectrum of target bacteria including Lactococcus spp., Lactobacillus spp., Leuconostoc spp., Pediococcus spp. Enterococcus spp. and Listeria spp. [23, 73]. Moreover, it does not show cytotoxicity towards Vero and HepG2 cell lines, implicating the potential of garvicin Q for therapeutic applications [23].

Our data shows that L. petauri B1726 produces a heat-stable, protease sensitive, amphiphilic bacteriocin with a molecular mass of ~ 5 kDa. To unravel the genetic traits associated with production of this antimicrobial protein/peptide, we sequenced, annotated, and analysed the genome of L. petauri B1726. It contains several genes that encode for virulence factors and proteins for survival in the host described for pathogenic L. garvieae and L. petauri strains [6, 8, 7072]. Also, no lacG homologues required for lactose utilization were found in the L. petauri B1726 genome and the strain showed poor growth on M17 medium containing lactose. Inability to utilize lactose is another feature associated with pathogenic L. garvieae isolates [79]. Collectively, these findings suggest that L. petauri B1726 is a strain with pathogenic potential for fish and humans. This poses a serious problem for production of garvicin Q using its native producer and the European Food Safety Agency recently raised safety concerns over use of L. garvieae in food and feed [80].

Bioinformatic analyses revealed a complete operon for garvicin Q biosynthesis, which was only partially described before [23], and the peptide was identified in precipitated and purified SN proteins by mass spectrometry. The gar operon consists of four genes garQ (coding for the pre-peptide), garI (for an immunity protein), garC (ABC-transporter) and garD (additional protein presumably involved in secretion). The predicted immunity protein GarI contains an enterocin A immunity domain, and the predicted structure is similar to the antiparallel four alpha-helical bundle of enterocin A immunity protein [81]. The pediocin immunity protein is known to interact with the C-terminal domain of pediocin [82] and it is hypothesized that this somehow blocks the pediocin-induced formation of a channel in the PTSMan from the cytoplasmic side [74]. Since garvicin Q also targets the PTSMan [73], GarI might fulfil a similar role. This is supported by the fact that no N-terminal signal peptide could be identified in the GarI sequence suggesting it is localized intracellularly.

Similar to pediocin, the ABC-transporter GarC may not only be involved in secretion but also in processing of the garvicin Q pre-peptide after the double glycine motif at the end of the garvicin Q signal peptide, which is common among class II bacteriocins [83]. The role of the additional secretion protein GarD for garvicin Q production remains unclear. In L. lactis subsp. lactis, the GarD homolog LcnD was demonstrated to be essential for the secretion of lactococcin A and may stabilize the associated ABC-transporter LcnC [84]. Moreover, lcnD encodes two in frame proteins (a transmembrane protein and a cytosolic variant) in L. lactis. The truncated, cytosolic LcnD version is thought to act like a chaperone guiding lactococcin A to the ABC-transporter. Similarly, we identified a potential RBS and start codon within garD. Hence, it is possible that garD also encodes two proteins. It is noteworthy that GarD contains a HlyD-like motif. HlyD is a periplasmic adaptor protein and part of the type I secretion system of hemolysin in E. coli. HlyD monomers assemble in an α-helical nanotube structure that connect the inner and outer membrane transport proteins [27]. LcnD and GarD have a similar membrane topology as HlyD [85]. Based on these predictions it is tempting to speculate, that GarD also stabilizes GarC and builds an α-helical nanotube structure on top of GarC. Crystallization or further modelling approaches may help to prove this hypothesis.

We observed optimal garvicin Q production by L. petauri B1726 in static cultures. Shaking of L. petauri B1726 during cultivation decreased garvicin Q production and growth, presumably by increasing oxidative stress. The genome of L. petauri B1726 contains genes for enzymes involved in molecular oxygen consumption (NADH oxidase, lgb_01366) and detoxification of reactive oxygen species (superoxide dismutase, lgb_00290), however we could not identify a catalase gene. A lack of catalase may lead to H2O2 accumulation which would explain the reduced growth at shaking conditions. For L. lactis it has been demonstrated that accumulation of H2O2 reduces growth rate and recombinant expression of catalase alleviates oxidative stress [86, 87]. Temperature also affected garvicin Q production. Although the final OD600 and pH was comparable between L. petauri B1726 cultures incubated at 25, 30 and 37 °C, garvicin Q production was lowest at 37 °C. A similar temperature-dependent biosynthesis was also reported for the bacteriocin sakacin A of Lactobacillus sakei Lb706 with good production at 25 and 30 °C but not at 35 °C [88]. A peptide pheromone and a two-component system were shown to be involved in temperature-dependent regulation of biosynthesis of sakacin A and other class II bacteriocins [88, 89]. However, we could not find a similar regulatory system in close proximity to the gar operon of L. petauri B1726.

The PTSMan of L. lactis IL1404 was shown to serve as a receptor for lactococcin A and its expression in a resistant Lactobacillus sakei strain installed sensitivity [90]. The subunits IIC and IID of group I PTSMan were shown to bind garvicins A, B, C and Q and mutational analysis of Lactococcus lactis IL1403 and L. garvieae IBB3403 PTSMan identified amino acids that might be involved in this interaction [73]. These amino acid residues are conserved in PTSMan IIC and IID subunits of garvicin Q-sensitive bacteria and are localized in a transmembrane helix of the IIC subunit and in extracellular loops of the IID subunit [73]. It was proposed that the N-terminus of garvicin Q interacts with the residues located in the IID extracellular loops and that the α-helical C-terminus of garvicin Q interacts with the transmembrane residues of the IIC subunit. These interactions could then lead to structural rearrangements of the PTSMan that creates a channel in the IIC permease resulting in disruption of membrane integrity [73]. Recently, the peptide-receptor complex could be resolved for pediocin and the L. monocytogenes PTSMan [74] supporting this hypothesis. Our own structural modelling also predicts interaction of garvicin Q with the extracellular domain of the PTSMan at the same site as pediocin [74] and in the vicinity of all the conserved amino acids required for activity of garvicin Q [73]. This further corroborates the hypothesis that the N-terminal β-sheet region of garvicin Q interacts with the extracellular part of the PTSMan and its C-terminal α-helix cracks the PTSMan to form a small hydrophilic pore with the PTSMan subunits [73, 74].

Another interesting observation is that adsorption of garvicin Q to bacterial cells is not dependent on the presence of its receptor. Adsorption of garvicin Q to C. glutamicum ATCC13032 (no PTSMan) and to L. monocytogenes EGY2 (truncated IID of PTSMan) was in a similar range like adsorption to group I PTSMan harbouring bacteria. Like other bacteriocins, garvicin Q is an amphiphilic peptide. Hence, unspecific interaction with negatively charged components of the Gram-positive envelope like phospholipids or teichoic acids could occur. Interestingly, adsorption to the native producer L. petauri B1726 was weaker compared to the other tested bacteria. This might point towards a modification of the L. petauri B1726 cell envelope to reduce binding of garvicin Q. This could be achieved by modifications of the cell envelope e.g. a positive surface charge and protection from cationic antimicrobials. In Staphylococcus aureus, such modifications are accomplished by sensing of cationic antimicrobials by the two-component system GraRS, lysinylation of phosphatidylglycerol by MprF and D-alanylation of lipoteichoic acids by DltABCD [65]. For L. petauri B1726, we identified homologous genes for GraRS (lgb_00457 and lgb_00458), MprF (lgb_01471) and DltABCD (lgb_01206-lgb_01209). Hence, these factors may help to reduce adsorption of the cationic bacteriocin garvicin Q. Based on our results, we propose that the antimicrobial action of garvicin Q involves three steps: (I) adsorption of the hydrophobic/positively charged peptide to the negatively charged bacterial cell envelope, (II) interaction of garvicin Q with IIC and IID subunits of group I PTSMan membrane components, and (III) structural changes resulting in formation of a pore in the transmembrane region of the IIC subunits leading to the inability to maintain membrane potential.

Finally, heterologous expression of the predicted garQICD operon in C. glutamicum resulted in production of an antimicrobial compound with identical features as garvicin Q produced by L. petauri B1726. These results strongly suggest that the genes garQICD are responsible for production of garvicin Q and indicates that recombinant production of garvicin Q is feasible. Also, our data does not provide information in the requirements of individual genes of the operon for garvicin Q production. This could be achieved by generating and analysing knockout mutants in the native host or by expression of different combinations of the gar genes in a heterologous host. Recently, similar experiments were performed for pediocin PA-1, another class II bacteriocin with a similar operon structure. The native pediocin PA-1 operon comprises the genes pedABCD with pedA encoding for the bacteriocin, pedB for an immunity protein and pedCD for two proteins involved in transport and processing. In C. glutamicum as heterologous host, pedACD are required for production of pediocin PA-1 but pedB is dispensable, possibly because C. glutamicum does not have a PTSMan and is thus resistant [63].

A significant loss in garvicin Q activity was observed after 24 h of cultivation of C. glutamicum CR099/pPBEx2-garQICDCgl. Cultivation of the recombinant strain in CGXII without urea markedly improved the production of the peptide and this is associated with acidification of the SN. Garvicin Q activity in SNs of the natural producer L. petauri B1726 is highest in late exponential to stationary phase when pH is highly acidic (pH 4–5; Fig. 4). Similar observation were made for other bacteriocins [78]. It is thus possible that acidic pH favourably impacts on its activity, stability, and/or adsorption to the producing cells.

Regarding efficiency of garvicin Q production using natural and recombinant hosts, it is difficult to compare the levels of activity and thus of the product garvicin Q obtained with L. petauri (natural producer) and C. glutamicum (heterologous host). There are differences in media (and potential interference with media components) as well as the sensor strains and the assays used for activity measurements. For heterologous production, an antibiotic (kanamycin) is required to maintain the plasmid. Thus, unlike for the natural producer strain, the sensor used to quantify activity has to carry a corresponding resistance to avoid measuring activity of the antibiotic instead of the bacteriocin. When simply comparing the dilutions required to detect garvicin Q activity it can be estimated that garvicin Q levels are at least fourfold higher in SNs of the natural producer L. petauri B1726 compared to the recombinant C. glutamicum producer despite lower final optical densities. This leaves ample room for improvement of recombinant garvicin Q production. Adsorption assays revealed that > 90% of the activity of purified garvicin Q was lost upon incubation with C. glutamicum despite the lack of a specific receptor, i.e. a PTSMan. This indicates that actual levels of garvicin Q are much higher, but the bulk of the peptide is not released into the SN and removed with the bacterial cells before activity assays. Further studies to understand (and prevent) the adsorption to bacterial cells may help to improve recombinant production of garvicin Q and other bacteriocins with C. glutamicum and provide an alternative to production using pathogenic, natural producers L. garvieae and L. petauri.

Conclusion

We could demonstrate that L. petauri B1726 produces garvicin Q and optimized cultivation conditions for production of this bacteriocin. The genes responsible for biosynthesis of garvicin Q were identified in silico and their functionality was proven by heterologous expression in C. glutamicum. The presented data confirm the hypothesis that garvicin Q utilizes a PTSMan as a receptor and provide evidence that it kills target bacteria by disruption of membrane integrity potentially by locking the PTSMan into a conformation that leads to formation of a constitutively open pore. Moreover, we show that garvicin Q adsorbs to different types of bacteria independent of the presence of a group I PTSMan. Finally, we explore first avenues for the recombinant biotechnological production of garvicin Q.

Data availibility statement

All data generated or analysed during this study are included in this published article. Genome sequence of L. petauri B1726 is publically available on GenBank (Accession Number: CP094882.1). Raw sequencing data is available on the NCBI Sequencing Read Archive with accession numbers SRR18554891 (ONT data) and SRR18554892 (Illumina data).

Change history

  • 02 January 2023

    The incorrect citation of the additional files was corrected and the missing caption of additional file 2 was updated.

References

  1. Meyburgh CM, Bragg RR, Boucher CE. Lactococcus garvieae: an emerging bacterial pathogen of fish. Dis Aquat Organ. 2017;123:67–79.

    Article  CAS  Google Scholar 

  2. Vendrell D, Balcázar JL, Ruiz-Zarzuela I, de Blas I, Gironés O, Múzquiz JL. Lactococcus garvieae in fish: a review. Comp Immunol Microbiol Infect Dis. 2006;29:177–98.

    Article  Google Scholar 

  3. Chen S-C, Liaw L-L, Su H-Y, Ko S-C, Wu C-Y, Chaung H-C, Tsai Y-H, Yang K-L, Chen Y-C, Chen T-H, et al. Lactococcus garvieae, a cause of disease in grey mullet, Mugil cephalus Lin Taiwan. J Fish Diseases. 2002;25:727–32.

    Article  CAS  Google Scholar 

  4. Shahi N, Mallik SK, Sahoo M, Chandra S, Singh AK. First report on characterization and pathogenicity study of emerging Lactococcus garvieae infection in farmed rainbow trout, Oncorhynchus mykiss (Walbaum), from India. Transbound Emerg Dis. 2018;65:1039–48.

    Article  CAS  Google Scholar 

  5. Shahin K, Veek T, Heckman TI, Littman E, Mukkatira K, Adkison M, Welch TJ, Imai DM, Pastenkos G, Camus A, Soto E. Isolation and characterization of Lactococcus garvieae from rainbow trout, onchorhyncus mykiss, from California. California: Transbound Emerg Dis; 2021.

    Google Scholar 

  6. Kotzamanidis C, Malousi A, Bitchava K, Vafeas G, Chatzidimitriou D, Skoura L, Papadimitriou E, Chatzopoulou F, Zdragas A. First report of isolation and genome sequence of L petauri strain from a rainbow trout Lactococcosis outbreak. Curr Microbiol. 2020;77:1089–96.

    Article  CAS  Google Scholar 

  7. Chen SC, Lin YD, Liaw LL, Wang PC. Lactococcus garvieae infection in the giant freshwater prawn Macrobranchium rosenbergii confirmed by polymerase chain reaction and 16S rDNA sequencing. Dis Aquat Organ. 2001;45:45–52.

    Article  CAS  Google Scholar 

  8. Gibello A, Galán-Sánchez F, Blanco MM, Rodríguez-Iglesias M, Domínguez L, Fernández-Garayzábal JF. The zoonotic potential of Lactococcus garvieae: an overview on microbiology, epidemiology, virulence factors and relationship with its presence in foods. Res Vet Sci. 2016;109:59–70.

    Article  CAS  Google Scholar 

  9. Kim JH, Go J, Cho CR, Kim JI, Lee MS, Park SC. First report of human acute acalculous cholecystitis caused by the fish pathogen Lactococcus garvieae. J Clin Microbiol. 2013;51:712–4.

    Article  Google Scholar 

  10. Wang CY, Shie HS, Chen SC, Huang JP, Hsieh IC, Wen MS, Lin FC, Wu D. Lactococcus garvieae infections in humans: possible association with aquaculture outbreaks. Int J Clin Pract. 2007;61:68–73.

    Article  CAS  Google Scholar 

  11. Fortina MG, Ricci G, Foschino R, Picozzi C, Dolci P, Zeppa G, Cocolin L, Manachini PL. Phenotypic typing, technological properties and safety aspects of Lactococcus garvieae strains from dairy environments. J Appl Microbiol. 2007;103:445–53.

    Article  CAS  Google Scholar 

  12. Desiderato CK, Sachsenmaier S, Ovchinnikov KV, Stohr J, Jacksch S, Desef DN, Crauwels P, Egert M, Diep DB, Goldbeck O, Riedel CU. Identification of potential probiotics producing bacteriocins active against listeria monocytogenes by a combination of screening tools. Int JMol Sci. 2021;22(16):8615.

    Article  CAS  Google Scholar 

  13. Kawanishi M, Yoshida T, Kijima M, Yagyu K, Nakai T, Okada S, Endo A, Murakami M, Suzuki S, Morita H. Characterization of Lactococcus garvieae isolated from radish and broccoli sprouts that exhibited a KG+ phenotype, lack of virulence and absence of a capsule. Lett Appl Microbiol. 2007;44:481–7.

    Article  CAS  Google Scholar 

  14. Martinovic A, Cabal A, Nisic A, Sucher J, Stöger A, Allerberger F, Ruppitsch W. Genome sequences of Lactococcus garvieae and Lactococcus petauri strains isolated from traditional montenegrin brine cheeses. Microbiol Resour Announc. 2021;10: e0054621.

    Article  Google Scholar 

  15. Defoirdt T, Sorgeloos P, Bossier P. Alternatives to antibiotics for the control of bacterial disease in aquaculture. Curr Opin Microbiol. 2011;14:251–8.

    Article  Google Scholar 

  16. Done HY, Venkatesan AK, Halden RU. Does the recent growth of aquaculture create antibiotic resistance threats different from those associated with land animal production in agriculture? Aaps j. 2015;17:513–24.

    Article  CAS  Google Scholar 

  17. Ghasemi SM, Bouzari M, Emtiazi G. Preliminary characterization of Lactococcus garvieae bacteriophage isolated from wastewater as a potential agent for biological control of lactococcosis in aquaculture. Aquacult Int. 2014;22:1469–80.

    Article  Google Scholar 

  18. Sequeiros C, Garcés ME, Vallejo M, Marguet ER, Olivera NL. Potential aquaculture probiont Lactococcus lactis TW34 produces nisin Z and inhibits the fish pathogen Lactococcus garvieae. Arch Microbiol. 2015;197:449–58.

    Article  CAS  Google Scholar 

  19. Su FJ, Chen MM. Protective efficacy of novel oral biofilm vaccines against Lactococcus garvieae infection in mullet, mugil cephalus. Vaccines. 2021;9(8):844.

    Article  Google Scholar 

  20. Nayak A, Karunasagar I, Chakraborty A, Maiti B. Potential application of bacteriocins for sustainable aquaculture. Rev Aquacult. 2022. https://doi.org/10.1111/raq.12647.

    Article  Google Scholar 

  21. Cotter PD, Ross RP, Hill C. Bacteriocins—a viable alternative to antibiotics? Nat Rev Microbiol. 2013;11:95–105.

    Article  CAS  Google Scholar 

  22. Riley MA, Wertz JE. Bacteriocins: evolution, ecology, and application. Annu Rev Microbiol. 2002;56:117–37.

    Article  CAS  Google Scholar 

  23. Tosukhowong A, Zendo T, Visessanguan W, Roytrakul S, Pumpuang L, Jaresitthikunchai J, Sonomoto K. Garvieacin Q, a novel class II bacteriocin from Lactococcus garvieae BCC 43578. Appl Environ Microbiol. 2012;78:1619–23.

    Article  CAS  Google Scholar 

  24. Ovchinnikov KV, Chi H, Mehmeti I, Holo H, Nes IF, Diep DB. Novel group of leaderless multipeptide bacteriocins from gram-positive bacteria. Appl Environ Microbiol. 2016;82:5216–24.

    Article  CAS  Google Scholar 

  25. Maldonado-Barragán A, Cárdenas N, Martínez B, Ruiz-Barba JL, Fernández-Garayzábal JF, Rodríguez JM, Gibello A. Garvicin A, a novel class IId bacteriocin from Lactococcus garvieae that inhibits septum formation in L garvieae strains. Appl Environ Microbiol. 2013;79:4336–46.

    Article  Google Scholar 

  26. Tymoszewska A, Diep DB, Aleksandrzak-Piekarczyk T. The extracellular loop of Man-PTS subunit IID is responsible for the sensitivity of Lactococcus garvieae to garvicins A B and C. Sci Rep. 2018;8:15790.

    Article  Google Scholar 

  27. Alav I, Kobylka J, Kuth MS, Pos KM, Picard M, Blair JMA, Bavro VN. Structure, assembly, and function of tripartite efflux and type 1 secretion systems in gram-negative bacteria. Chem Rev. 2021;121:5479–596.

    Article  CAS  Google Scholar 

  28. Borrero J, Brede DA, Skaugen M, Diep DB, Herranz C, Nes IF, Cintas LM, Hernández PE. Characterization of garvicin ML, a novel circular bacteriocin produced by Lactococcus garvieae DCC43, isolated from mallard ducks (Anas platyrhynchos). Appl Environ Microbiol. 2011;77:369–73.

    Article  CAS  Google Scholar 

  29. Gao Y, Li D, Liu S, Zhang L. Garviecin LG34, a novel bacteriocin produced by Lactococcus garvieae isolated from traditional Chinese fermented cucumber. Food Control. 2015;50:896–900.

    Article  CAS  Google Scholar 

  30. Villani F, Aponte M, Blaiotta G, Mauriello G, Pepe O, Moschetti G. Detection and characterization of a bacteriocin, garviecin L1–5, produced by Lactococcus garvieae isolated from raw cow’s milk. J Appl Microbiol. 2001;90:430–9.

    Article  CAS  Google Scholar 

  31. Dubey S, Diep DB, Evensen Ø, Munang’andu HM, Garvicin KS. a Broad-spectrum bacteriocin protects zebrafish larvae against Lactococcus garvieae Infection. Int J Mol Sci. 2022;23(5):2833.

    Article  CAS  Google Scholar 

  32. Eikmanns BJ, Metzger M, Reinscheid D, Kircher M, Sahm H. Amplification of three threonine biosynthesis genes in Corynebacterium glutamicum and its influence on carbon flux in different strains. Appl Microbiol Biotechnol. 1991;34:617–22.

    Article  CAS  Google Scholar 

  33. Nilsen T, Nes IF, Holo H, Enterolysin A. a cell wall-degrading bacteriocin from Enterococcus faecalis LMG 2333. Appl Environ Microbiol. 2003;69:2975–84.

    Article  CAS  Google Scholar 

  34. Reich SJ, Stohr J, Goldbeck O, Fendrich B, Crauwels P, Riedel CU. Improved fluorescent Listeria spp biosensors for analysis of antimicrobials by flow cytometry. bioRxiv. 2022;2030:486417.

    Google Scholar 

  35. Bécavin C, Bouchier C, Lechat P, Archambaud C, Creno S, Gouin E, Wu Z, Kühbacher A, Brisse S, Pucciarelli MG, et al. Comparison of widely used Listeria monocytogenes strains EGD, 10403S, and EGD-e highlights genomic variations underlying differences in pathogenicity. MBio. 2014;5:e00969-e914.

    Article  Google Scholar 

  36. Dalet K, Cenatiempo Y, Cossart P, Héchard Y, Consortium ELG. A sigma(54)-dependent PTS permease of the mannose family is responsible for sensitivity of Listeria monocytogenes to mesentericin Y105. Microbiology. 2001;147:3263–9.

    Article  CAS  Google Scholar 

  37. Bolotin A, Wincker P, Mauger S, Jaillon O, Malarme K, Weissenbach J, Ehrlich SD, Sorokin A. The complete genome sequence of the lactic acid bacterium lactococcus lactis ssp. lactis IL1403. Genom Res. 2001;11:731–53.

    Article  CAS  Google Scholar 

  38. Hanahan D. Studies on transformation of Escherichia coli with plasmids. J Mol Biol. 1983;166:557–80.

    Article  CAS  Google Scholar 

  39. Baumgart M, Unthan S, Rückert C, Sivalingam J, Grünberger A, Kalinowski J, Bott M, Noack S, Frunzke J. Construction of a prophage-free variant of Corynebacterium glutamicum ATCC 13032 for use as a platform strain for basic research and industrial biotechnology. Appl Environ Microbiol. 2013;79:6006–15.

    Article  CAS  Google Scholar 

  40. Wick R, Menzel P: Filtlong. v0.2.1 edition. github2021.

  41. Wick RR, Judd LM, Gorrie CL, Holt KE. Unicycler: resolving bacterial genome assemblies from short and long sequencing reads. PLoS Comput Biol. 2017;13: e1005595.

    Article  Google Scholar 

  42. Wick RR, Holt KE. Polypolish: short-read polishing of long-read bacterial genome assemblies. PLoS Comput Biol. 2022;18: e1009802.

    Article  CAS  Google Scholar 

  43. Zimin AV, Marçais G, Puiu D, Roberts M, Salzberg SL, Yorke JA. The MaSuRCA genome assembler. Bioinformatics. 2013;29:2669–77.

    Article  CAS  Google Scholar 

  44. Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30:2068–9.

    Article  CAS  Google Scholar 

  45. Van Heel AJ, De Jong A, Song C, Viel JH, Kok J, Kuipers OP. BAGEL4: a user-friendly web server to thoroughly mine RiPPs and bacteriocins. Nucleic Acids Res. 2018;46:W278–81.

    Article  Google Scholar 

  46. Blin K, Shaw S, Kloosterman AM, Charlop-Powers Z, van Wezel GP, Medema MH, Weber T. AntiSMASH 6.0: improving cluster detection and comparison capabilities. Nucleic Acids Res. 2021;49:W29–35.

    Article  CAS  Google Scholar 

  47. Naville M, Ghuillot-Gaudeffroy A, Marchais A, Gautheret D. ARNold: a web tool for the prediction of Rho-independent transcription terminators. RNA Biol. 2011;8:11–3.

    Article  CAS  Google Scholar 

  48. Solovyev V, Salamov A. Automatic annotation of microbial genomes and metagenomic sequences. In: Li RW, editor. Metagenomics and its applications in agriculture biomedicine and environmental studies. Newyork: Nova Science Publishers; 2011.

    Google Scholar 

  49. Richter M, Rosselló-Móra R, Oliver Glöckner F, Peplies J. JSpeciesWS: a web server for prokaryotic species circumscription based on pairwise genome comparison. Bioinformatics. 2016;32:929–31.

    Article  CAS  Google Scholar 

  50. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–402.

    Article  CAS  Google Scholar 

  51. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using clustal omega. Mol Syst Biol. 2011;7:539.

    Article  Google Scholar 

  52. Omasits U, Ahrens CH, Müller S, Wollscheid B. Protter: interactive protein feature visualization and integration with experimental proteomic data. Bioinformatics. 2014;30:884–6.

    Article  CAS  Google Scholar 

  53. Lomize MA, Pogozheva ID, Joo H, Mosberg HI, Lomize AL. OPM database and PPM web server: resources for positioning of proteins in membranes. Nucleic Acids Res. 2012;40:D370-376.

    Article  CAS  Google Scholar 

  54. Baek M, DiMaio F, Anishchenko I, Dauparas J, Ovchinnikov S, Lee GR, Wang J, Cong Q, Kinch LN, Schaeffer RD, et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science. 2021;373:871–6.

    Article  CAS  Google Scholar 

  55. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, et al. Highly accurate protein structure prediction with alphafold. Nature. 2021;596:583–9.

    Article  CAS  Google Scholar 

  56. Pettersen EF, Goddard TD, Huang CC, Meng EC, Couch GS, Croll TI, Morris JH, Ferrin TE. UCSF chimerax: structure visualization for researchers, educators, and developers. Protein Sci. 2021;30:70–82.

    Article  CAS  Google Scholar 

  57. Bakkes PJ, Ramp P, Bida A, Dohmen-Olma D, Bott M, Freudl R. Improved pEKEx2-derived expression vectors for tightly controlled production of recombinant proteins in Corynebacterium glutamicum. Plasmid. 2020;112:102540.

    Article  Google Scholar 

  58. Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat Methods. 2009;6:343–5.

    Article  CAS  Google Scholar 

  59. Tauch A, Kirchner O, Löffler B, Götker S, Pühler A, Kalinowski J. Efficient electrotransformation of Corynebacterium diphtheriae with a mini-replicon derived from the Corynebacterium glutamicum plasmid pGA1. Curr Microbiol. 2002;45:362–7.

    Article  CAS  Google Scholar 

  60. Holo H, Nes IF. High-frequency transformation, by electroporation, of Lactococcus lactis subsp. cremoris grown with glycine in osmotically stabilized media. Appl Environ Microbiol. 1989;55:3119–23.

    Article  CAS  Google Scholar 

  61. Henning C, Vijayakumar P, Adhikari R, Jagannathan B, Gautam D, Muriana PM. Isolation and taxonomic identity of bacteriocin-producing lactic acid bacteria from retail foods and animal sources. Microorganisms. 2015;3:80–93.

    Article  CAS  Google Scholar 

  62. Holo H, Nilssen O, Nes IF. Lactococcin A, a new bacteriocin from Lactococcus lactis subsp. cremoris: Isolation and characterization of the protein and its gene. J Bacteriol. 1991;173:3879–87.

    Article  CAS  Google Scholar 

  63. Goldbeck O, Desef DN, Ovchinnikov KV, Perez-Garcia F, Christmann J, Sinner P, Crauwels P, Weixler D, Cao P, Becker J, et al. Establishing recombinant production of pediocin PA-1 in Corynebacterium glutamicum. Metab Eng. 2021;68:34–45.

    Article  CAS  Google Scholar 

  64. Voges R, Noack S. Quantification of proteome dynamics in Corynebacterium glutamicum by (15)N-labeling and selected reaction monitoring. J Proteomics. 2012;75:2660–9.

    Article  CAS  Google Scholar 

  65. Yang R, Johnson MC, Ray B. Novel method to extract large amounts of bacteriocins from lactic acid bacteria. Appl Environ Microbiol. 1992;58:3355–9.

    Article  CAS  Google Scholar 

  66. Chi H, Holo H. Synergistic antimicrobial activity between the broad spectrum bacteriocin Garvicin KS and Nisin, Farnesol and POLYMYXIN B against gram-positive and gram-negative bacteria. Curr Microbiol. 2018;75:272–7.

    Article  CAS  Google Scholar 

  67. Holivololona L. Screening and characterization of bacteriocins produced by lactic acid bacteria against Lactococcus garvieae. Nor Univ Life Sci. 2013;77(3):893.

    Google Scholar 

  68. Ferrario C, Ricci G, Milani C, Lugli GA, Ventura M, Eraclio G, Borgo F, Fortina MG. Lactococcus garvieae: where is it from? A first approach to explore the evolutionary history of this emerging pathogen. PLoS ONE. 2013;8: e84796.

    Article  Google Scholar 

  69. Goodman LB, Lawton MR, Franklin-Guild RJ, Anderson RR, Schaan L, Thachil AJ, Wiedmann M, Miller CB, Alcaine SD, Kovac J. Lactococcus petauri sp. nov., isolated from an abscess of a sugar glider. Int J Syst Evol Microbiol. 2017;67:4397–404.

    Article  CAS  Google Scholar 

  70. Aguado-Urda M, Gibello A, Blanco Mdel M, Fernández-Garayzábal JF, López-Alonso V, López-Campos GH. Global transcriptome analysis of Lactococcus garvieae strains in response to temperature. PLoS ONE. 2013;8: e79692.

    Article  CAS  Google Scholar 

  71. Menéndez A, Fernández L, Reimundo P, Guijarro JA. Genes required for Lactococcus garvieae survival in a fish host. Microbiology. 2007;153:3286–94.

    Article  Google Scholar 

  72. Morita H, Toh H, Oshima K, Yoshizaki M, Kawanishi M, Nakaya K, Suzuki T, Miyauchi E, Ishii Y, Tanabe S, et al. Complete genome sequence and comparative analysis of the fish pathogen Lactococcus garvieae. PLoS ONE. 2011;6: e23184.

    Article  CAS  Google Scholar 

  73. Tymoszewska A, Diep DB, Wirtek P, Aleksandrzak-Piekarczyk T. The non-lantibiotic bacteriocin garvicin Q targets man-pts in a broad spectrum of sensitive bacterial genera. Sci Rep. 2017;7:8359.

    Article  Google Scholar 

  74. Zhu L, Zeng J, Wang C, Wang J. structural basis of pore formation in the mannose phosphotransferase system by pediocin PA-1. Appl Environ Microbiol. 2022;88: e0199221.

    Article  Google Scholar 

  75. Crauwels P, Schäfer L, Weixler D, Bar NS, Diep DB, Riedel CU, Seibold GM. Intracellular pHluorin as sensor for easy assessment of bacteriocin-induced membrane-damage in Listeria monocytogenes. Front Microbiol. 2018;9:3038.

    Article  Google Scholar 

  76. Weixler D, Berghoff M, Ovchinnikov KV, Reich S, Goldbeck O, Seibold GM, Wittmann C, Bar NS, Eikmanns BJ, Diep DB, Riedel CU. Recombinant production of the lantibiotic nisin using Corynebacterium glutamicum in a two-step process. Microb Cell Fact. 2022;21:11.

    Article  CAS  Google Scholar 

  77. Ríos Colombo NS, Chalón MC, Navarro SA, Bellomio A. Pediocin-like bacteriocins: new perspectives on mechanism of action and immunity. Curr Genet. 2018;64:345–51.

    Article  Google Scholar 

  78. Abbasiliasi S, Tan JSS, Tengku Ibrahim TAA, Bashokouh F, Ramakrishnan NRR, Mustafa S, Ariff ABB. Fermentation factors influencing the production of bacteriocins by lactic acid bacteria: a review. RSC Adv. 2017;7:29395–420.

    Article  CAS  Google Scholar 

  79. Aguado-Urda M, Cutuli MT, Blanco MM, Aspiroz C, Tejedor JL, Fernández-Garayzábal JF, Gibello A. Utilization of lactose and presence of the phospho-β-galactosidase (lacG) gene in Lactococcus garvieae isolates from different sources. Int Microbiol. 2010;13:189–93.

    CAS  Google Scholar 

  80. Koutsoumanis K, Allende A, Alvarez-Ordóñez A, Bolton D, Bover-Cid S, Chemaly M, Davies R, De Cesare A, Hilbert F, Lindqvist R, et al. Update of the list of QPS-recommended biological agents intentionally added to food or feed as notified to EFSA 14: suitability of taxonomic units notified to EFSA until March 2021. Efsa J. 2021;19: e06689.

    CAS  Google Scholar 

  81. Johnsen L, Dalhus B, Leiros I, Nissen-Meyer J. 1.6-Angstroms crystal structure of EntA-im. a bacterial immunity protein conferring immunity to the antimicrobial activity of the pediocin-like bacteriocin enterocin A. J Biol Chem. 2005;280:19045–50.

    Article  CAS  Google Scholar 

  82. Johnsen L, Fimland G, Nissen-Meyer J. The C-terminal domain of pediocin-like antimicrobial peptides (class IIa bacteriocins) is involved in specific recognition of the C-terminal part of cognate immunity proteins and in determining the antimicrobial spectrum. J Biol Chem. 2005;280:9243–50.

    Article  CAS  Google Scholar 

  83. Håvarstein LS, Diep DB, Nes IF. A family of bacteriocin ABC transporters carry out proteolytic processing of their substrates concomitant with export. Mol Microbiol. 1995;16:229–40.

    Article  Google Scholar 

  84. Varcamonti M, Nicastro G, Venema G, Kok J. Proteins of the lactococcin a secretion system: lcnd encodes two in-frame proteins. FEMS Microbiol Lett. 2001;204:259–63.

    Article  CAS  Google Scholar 

  85. Franke CM, Leenhouts KJ, Haandrikman AJ, Kok J, Venema G, Venema K. Topology of LcnD, a protein implicated in the transport of bacteriocins from Lactococcus lactis. J Bacteriol. 1996;178:1766–9.

    Article  CAS  Google Scholar 

  86. Rochat T, Miyoshi A, Gratadoux JJ, Duwat P, Sourice S, Azevedo V, Langella P. High-level resistance to oxidative stress in Lactococcus lactis conferred by Bacillus subtilis catalase KatE. Microbiology. 2005;151:3011–8.

    Article  CAS  Google Scholar 

  87. van Niel EW, Hofvendahl K, Hahn-Hägerdal B. Formation and conversion of oxygen metabolites by Lactococcus lactis subsp lactis ATCC 19435 under different growth conditions. Appl Environ Microbiol. 2002;68:4350–6.

    Article  Google Scholar 

  88. Diep DB, Axelsson L, Grefsli C, Nes IF. The synthesis of the bacteriocin sakacin a is a temperature-sensitive process regulated by a pheromone peptide through a three-component regulatory system. Microbiology. 2000;146(Pt 9):2155–60.

    Article  CAS  Google Scholar 

  89. Kleerebezem M, Quadri LE, Kuipers OP, de Vos WM. Quorum sensing by peptide pheromones and two-component signal-transduction systems in gram-positive bacteria. Mol Microbiol. 1997;24:895–904.

    Article  CAS  Google Scholar 

  90. Diep DB, Skaugen M, Salehian Z, Holo H, Nes IF. Common mechanisms of target cell recognition and immunity for class II bacteriocins. Proc Natl Acad Sci USA. 2007;104:2384–9.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors gratefully acknowledge Dr. Cyril Frantzen for performing Illumina sequencing.

Funding

Open Access funding enabled and organized by Projekt DEAL. This study was funded by grants of the BMBF consortium AMPLIFY (Grant No. 031B0826A and 031B0826C). The funding bodies had no role in the design of the study, analysis of the data, or writing of the manuscript.

Author information

Authors and Affiliations

Authors

Contributions

Conceptualization: CKD, OG, BJE, and CUR; Methodology: KVO, SJR, OG, and MO Formal analysis: CKD, KMH, SJR, LH,KVO, and AR; Investigation: CKD, KMH, SJR and KVO; Resources: BJE, DBD, MO and CUR; Writing-original draft preparation: CKD; Writing—review and editing: CKD, BJE, A, MO, and CUR; Supervision: CKD, and CUR; Project administration: CUR; Funding acquisition: BJE, MO, and CUR All authors have read and agreed to the published version of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Christian U. Riedel.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

Primer and codon optimized gene sequences used in present study. Table S3. Whole genome sequence alignment of Lactococcus sp. B1726 to other Lactococcus species. JSpeciesWS was used for performing the alignment. A tetra correlation search for Isolate B1726 was done to find similar genomes. Identified genomes were compared by ANIm (average nucleotide index based on MUMer). The percentage of aligned nucleotides is also indicated as well as the source, safety level and potential bacteriocins. Table S4. Predicted virulence factors and proteins important for survival in fish encoded on the genome of L. petauri B1726. The annotated genome of L. petauri B1726 was searched for virulence factors or proteins importat for survival in the host described for pathogenic strains of L. garvieae. Table S7. Impact of different growth conditions on garvicin Q production by L. petauri B1726. L. petauri B1726 was cultivated in 5 ml medium in glass tubes overnight. The antimicrobial activity was determined by growth inhibition assay with L. innocua LMG2785 as sensor strain Table S8. Ammonium sulphate (AS) precipitation of garvicin Q. Cell-free culture supernatant of an overnight L. petauri B1726 culture was subjected to AS precipitation applying different AS concentrations. The precipitate was dissolved in demin H2O (dissolved precipitate). The dissolved precipitate was clarified by centrifugation to remove insoluble components and the clarification supernatant was used in a growth inhibition assay (clarification). The percentage of recovered activity after clarification compared to the dissolved precipitate was then calculated (recovered activity). The antimicrobial activity was determined by growth inhibition assay with L. innocua LMG2785 as sensor strain. Although the recovered activity was highest for 20–30 % of AS, the highest overall activity was determined in the dissolved precipitate using 70 % AS. These results indicate that >50 % of AS reduces solubility of garvicin Q. Figure S2. Concentration of Lactococcus sp. B1726 peptide by cation exchange (CIEX) and reverse-phase chromatography (RPC). CIEX of ammonium sulphate precipitation of cell-free supernatants of Lactococcus sp. B1726 A and RPC of the active CIEX elution fraction B. The black arrow indicates the active fraction. Antimicrobial activity was tested by spot on lawn assay with L. innocua LMG2785 as sensor strain. Figure S5. α-hemolysis by L. petauri B1726. L. petauri B1726 and L. lactis IL1403 were streaked on tryptone soya agar supplemented with 5 % sheep blood (Thermo ScientificTM). The agar plates were incubated at 37 °C. Incubation of L. petauri B1726 for ~4 days led to a clear discoloration of the blood agar indicating α-hemolysis. Figure S6. Garvicin Q operon of L. petauri B1726. Nucleotide sequence of garQ (red), garI (blue), garC (orange) and garD (green) are given. Predicted promoter sequences (− 10 and − 35 boxes) are displayed in italic and bold letters. Rho-independent terminator sequences are underlined. A putative ribosome binding site and start codon within garD is highlighted by capital and bold letters. FGENESB and BPROM were used for prediction of promoter and transcription units. ARNold was used for finding terminators. Figure S9. QTOF LC-MS analysis of natural garvicin Q in pooled RPC elution fractions harbouring antimicrobial activity. Graphs are derived from the spectrum shown in Fig. 4D zoomed in on the peaks with a mass to charge ratio (m/z) of 1069.1500 (left), 891.1142 (middle) and 763.9628 (right), which correspond to the 5-fold ([M+5H]5+; m/z spacing of 0.2), 6-fold ([M+6H]6+; m/z spacing of 0.17), and 7-fold ([M+7H]7+; m/z spacing of 0.14) positively charged ion of garvicin Q. Figure S10. Predicted interaction of garvicin Q with the IIC and IID subunit of the L. monocytogenes EGDe group I PTSMan. Garvicin Q (GarQ) is indicated in blue. The PTSMan subunits IIC and IID are indicated in pink and green. The surface of garvicin Q and PTSMan residues in close vicinity to garvicin Q (within 5 angstrom) is transparent. A and C Amino acid residues that were previously described to be involved in the interaction of garvicin Q and PTSMan by Tymoszewska et al. [5] are indicated in red. B and D Amino acid residues that were previously described to be involved in the interaction of pediocin and PTSMan by Zhu et al. 2022 [6] are indicated in cyan. AlphaFold was used for structure prediction and ChimeraX for visualization. Figure S11. interaction of pediocin PA-1 with the IIC and IID subunit of the L. monocytogenes EGDe group I PTSMan (PDB: 7VLY). Pediocin PA-1 (Ped) is indicated in blue. The PTSMan subunits IIC and IID are indicated in pink and green. In B, amino acid residues that were previously described to be involved in the interaction of pediocin and PTSMan by Zhu et al. [6] are indicated in cyan. ChimeraX was used for visualization. Figure S12. Characterisation of L. lactis IL1403/pNZ-pHin2Lm. A Fluorescence imaging of the recombinant strain L. lactis IL1403 pNZ-pHin2Lm and the wild type strain. The recombinant strain showed fluorescence compared to the wild type strain. This indicates synthesis of pHluorin2 protein. Imaging was performed in an iBrightTM FL 1000 imaging system with filter settings for green fluorescence. B Excitation spectrum of L. lactis IL1403 pNZ-pHin2Lm. Addition of the pore-forming bacteriocin nisin (green) and garvicin Q (blue) is shifting the ratio of the excitation maxima in comparison to the LMB buffer negative control (red). Figure S13. Establishment of recombinant garvicin Q production by C. glutamicum. A Schematic map of expression vector pPBEx2-garQICDCgl. Upstream of each gene the sequence of the ribosomal binding site (RBS) of the elongation factor TU was inserted. B Influence of induction of gene expression during pre-cultivation of C. glutamicum CR099/pPBEx2-garQICDCgl. Garvicin Q activity was determined by pHluorin2 assay using L. lactis IL1403/pNZ-pHin2Lm as a sensor. Samples were taken 4 h after inoculation of the main culture. Addition of 0.05 mM IPTG during pre-cultivation led to a more pronounced decrease in the fluorescence ratio indicating pore formation/garvicin Q activity. Trypsin treatment completely abolished antimicrobial activity indicated that the antimicrobial compound is a protein/peptide. C High-cell density cultivation of C. glutamicum CR099/pPBEx2-garQICDCgl for garvicin Q production. D Activity in supernatants of C. glutamicum CR099/pPBEx2-garQICDCgl or the control strain harbouring the empty plasmid pPBEx2 at 4 and 24 h of cultivation in a high-cell density approach. Assay were performed with L. lactis IL1403/pNZ-pHin2Lm as indicator. Assay buffer (LMB) served as negative control and HIC purified garvicin Q (GarQ) was used as a positive control. Values in B and D are ratios of fluorescence intensity at 520 nm after excitation at 400 and 480 nm (ratio RFU 400/480) and are mean of n = 2–3 independent experiments (supernatants of independent cultivations). Figure S14. QTOF LC-MS analysis of recombinant garvicin Q in pooled RPC elution fractions harbouring antimicrobial activity. The graph is derived from the spectrum shown in Fig. 7D zoomed in on the peak with a mass to charge ratio (m/z) of 1069.2010, which corresponds to the 5-fold positively charged ion of garvicin Q ([M+5H]5+; m/z spacing of 0.2).

Additional file 2.

Complete data set of QTOF LC-MS analysis of garvicin Q purified from supernatants the natural producer L. petauri B1726.

Additional file 3.

Complete data set of QTOF LC-MS analysis of garvicin Q purified from supernatants the recombinant producer C. glutamicum CR099/pPBEx2-garQICDCgl.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Desiderato, C.K., Hasenauer, K.M., Reich, S.J. et al. Garvicin Q: characterization of biosynthesis and mode of action. Microb Cell Fact 21, 236 (2022). https://doi.org/10.1186/s12934-022-01952-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12934-022-01952-9