|Home | About | Journals | Submit | Contact Us | Français|
Quorum sensing molecules (QSMs) are involved in the regulation of complicated processes helping bacterial populations respond to changes in their cell-density. Although the QS gene cluster (comQXPA) has been identified in the genome sequence of some bacilli, the QS system B. licheniformis has not been investigated in detail, and its QSM (ComX pheromone) has not been identified. Given the importance of this antagonistic bacterium as an industrial workhorse, this study was aimed to elucidate B. licheniformis NCIMB-8874 QS. The results obtained from bioinformatics studies on the whole genome sequence of this strain confirmed the presence of essential quorum sensing-related genes. Although polymorphism was verified in three proteins of this cluster, ComQ, precursor-ComX and ComP, the transcription factor ComA was confirmed as the most conserved protein. The cell–cell communication of B. licheniformis NCIMB-8874 was investigated through further elucidation of the ComX pheromone as 13-amino acid peptide. The peptide sequence of the pheromone has been described through biochemical characterisation.
The online version of this article (doi:10.1186/s13568-017-0381-6) contains supplementary material, which is available to authorized users.
The first evidence of microbial cell–cell communication was reported by Tomasz and Beiser in 1965, when they suggested that a hormone-like extracellular product regulated competence in Streptococcus pneumoniae (Tomasz and Beiser 1965). Later on researchers found that the product was a peptide acting as a common signal in cell–cell communication amongst Gram-positive bacteria (Dunny and Leonard 1997). However, organized responses in a microbial colony were officially reported in the luminous marine bacterium.
Aliivibrio fischeri in its symbiotic relationship with the Hawaiian squid, Euprymna scolopes. Bioluminescence was triggered and controlled by one or more signalling molecules accumulating in the extracellular environment of A. fischeri as their cell density increased and reached a critical number (quorum). Signal molecules implicated in cell–cell communication are known as auto-inducers and/or QSM and their function is to regulate gene expression in other cells of the community to control bacterial responses (Nealson et al. 1970).
In Gram negative bacteria the QSMs are diffusible, they use low molecular weight hydrophobic signal molecules. Gram-positive bacteria employ unmodified or post-translationally modified peptides as well as γ-butyrolactone. In the cytoplasm, these peptides are produced as precursors of the QSM and then cleaved, modified and exported. Once in the extracellular environment, the peptides are detected via two-component systems (Kleerebezem et al. 1997). In Gram-positive bacteria, the QSMs are secreted to the extracellular milieu and then recognised by receptors which transport the signal across the cell membrane to initiate the target gene transcription (Waters and Bassler 2005). Studies have shown that QS in Bacillus species is mediated by small peptides that control competence (for DNA uptake), sporulation and the production of certain secondary metabolites in a cell-density dependent fashion (Bassler and Miller 2013). A competence pheromone in B. subtilis was first described genetically in B. subtilis subsp. subtilis 168 (Magnuson et al. 1994). In this bacterium, cell–cell communication is regulated through the comQXPA locus. The products of this system are the ComX pheromone and the two-component transduction system ComP and ComA which regulate the occurrence of natural competence in this bacterium (Weinrauch et al. 1990; Dubnau et al. 1994). The system is activated by accumulation of the ComX pheromone in the extracellular milieu (Magnuson et al. 1994).
Studies by Ansaldi and co-workers confirmed that the comQXPA gene cluster plays an essential role in the regulation of competence development in the B. subtilis QS mechanism (Ansaldi et al. 2002). This gene cluster is present in bacilli with close genomic relationship to B. subtilis, a group within which B. licheniformis is reported to belong (Magnuson et al. 1994). De Vizio identified that a B. licheniformis NCIMB 8874 cell–cell communication operates analogously to the comQXPA-controlled pathway of B. subtilis (De Vizio 2011). The products of this system are the ComX pheromone and the two-component transduction system ComP and ComA. ComQ is the only dedicated protein required for the processing of active pheromone (Magnuson et al. 1994).
Although QS is well established in B. subtilis, further investigations of the cell–cell communication and signalling molecules in B. licheniformis were required as the biochemistry of the relevant QSMs remained unexplored; importantly, such work would help explore potential bio-inhibitory activities relevant to industrial applications such as production of proteases, amylases and specialty chemicals (Schallmey et al. 2004) as well as several antimicrobial compounds, such as bacitracin (Johnson et al. 1945) and the surfactin-resembling lichenysin (Yakimov et al. 1995).
Some researchers have focused on the production of the pheromone as a post-translationally modified peptide which requires processing of the precursor to 10 amino acids, modification of the tryptophan residue and export from the cell by ComQ (Lazazzera et al. 1999). It has also been confirmed that the pheromone was formed by isoprenylation of an inactive precursor peptide (Schneider et al. 2002). Okada and colleagues identified the pheromone structure of B. subtilis for the first time and the structure of the resulting 6-amino-acid peptide product as a QSM (Okada et al. 2005). With the B. licheniformis genome sequence in hand cloning, expression and purification methods developed for B. subtilis (Okada et al. 2005) were adopted for the study of the B. licheniformis QS system with special emphasis on its signalling molecule, the ComX pheromone.
In the present work, the genomic studies have focused on the extent of polymorphism presented in the amino acid sequences of proteins involved in the B. licheniformis QS system. Besides, the QS study on the ComX pheromone carried out by investigating the putative QS genes (comQX) of B. licheniformis NCIMB 8874 and over-expressing comQX genes using gene cloning techniques. It led to detect and identify the novel pheromone peptide using available genomic information (using next generation sequencing platform) on B. licheniformis QS genes. The ComX pheromone was purified using biochemical techniques on a recombinant E. coli culture, constructed for over-production of the pheromone in the supernatant.
The genome sequences of B. licheniformis NCIMB 8874 were sequenced and determined for the first time on the Ion Torrent Personal Genome Machine (PGM) (Life Technologies, Thermo Fisher Scientific, UK) at Genomic Services, University of Westminster. Following the first stage of sequencing procedure, library construction, the template was prepared through emulsion PCR automated system and then run on the PGM to accomplish the sequencing process (no. of reads was 1,624,672; no. of generated contigs was 168 and achieved overall depth of coverage was 59×). These sequence data have been submitted to the GenBank data bases under the accession number MBGK01000000. Details of data submission can be found at GenBank: http://www.ncbi.nlm.nih.gov.
Assembled DNA sequences data in FASTA format was obtained from the Ion Reporter 5.0 software. The assembled sequence was annotated through IonGap Annotation Service (http://iongap.hpc.iter.es/), an integrated Genome Analysis Platform for Ion Torrent sequence data. The phylogenetic analysis was carried out by aligning amino acid sequences of comQXPA cluster from the strain with homologous proteins from other Bacilli which obtained from NCBI nucleotide/protein database using “Clustal Omega” as a multiple sequence alignment program (for details of clustering method please refer to http://www.ebi.ac.uk/Tools/msa/clustalo/).
The QSM studies were performed on B. licheniformis NCIMB 8874. The reporter strain B. subtilis JRL293 [amyE: (srfA-lacZ, cat), trp, phe] was used for pheromone bioassay. Both strains were available in the Culture Collection of the University of Westminster, London, UK. Lysogeny broth (LB) and LB agar (LBA) (Sigma) were used for the maintenance of B. licheniformis NCIMB 8874. Maintenance medium for B. subtilis JRL293 was supplemented with chloramphenicol (Sigma) (5 μgml−1).
The expression strain [E. coli BL21 (DE3)] and E. coli TOP10 were used for cloning/transformations and were selected on LBA supplemented with ampicilin (100 µgml−1). E. coli BL21 ComX producer strain was cultivated in M9 minimal salts solution (sigma). The medium was supplemented with a mix of filter-sterilised amino acids (leucine, phenylalanine, histidine, serine, 40 µgml−1 each; glutamine, 400 µgml−1), and ampicillin (100 µgml−1). According to the manufacturer instruction, additional supplementation of filter-sterilised 20% (w/v) glucose, 1 M magnesium sulfate and 1 M calcium chloride was required in order to complete M9 minimal medium preparation. Filter sterilization was carried out through a 0.22 µm filter (Millipore).
Plasmid allowing the overproduction of ComQ and ComX proteins in E. coli was derived from the pET-22b(+) vector. comQ and comX were PCR amplified from chromosomal DNA with the custom primer set (comQ-Forward/comX-Reverse). Forward primer (ACGTCATATGAATCATTTTATAGACGTTGAGATTCC) hybridized to a sequence upstream of comQ contained a NdeI site while downstream comX was amplified by reverse primer (ACGTGGATCCTTATTTGAACCATAAATTAGGGTAAG) containing a BamHI site. The annealing temperature was 53 °C and the expected PCR product fragment was 1070 bp. Primers were custom prepared by Invitrogen (Thermo Fisher Scientific).
After cleavage with NdeI and BamHI, DNA fragments were cloned into the pET-22b(+) vector cut with the same enzymes. The recombinant plasmid were transformed into E. coli TOP10 and then into E. coli BL21 (DE3) as a host to express the ComX pheromone. All cloned fragments in both transformation steps were sent for sequencing using the T7 primers (Novagen) to determine the accuracy of their sequence.
Escherichia coli BL21 ComX producer strain was grown overnight in the completed M9 minimal salts medium described earlier. At stationary phase, this pre-culture (20 ml) was added to 1980 ml of the supplemented M9 medium to make 2 l bacterial culture (5 flasks in total to prepare 10 l culture) and then incubated at 37 °C and 110 rpm for 8 h. comQX gene expression was induced with 0.5 mM Isopropyl β-d-1-thiogalactopyranoside (IPTG) at 37 °C and 110 rpm overnight. The culture broth (10 l) was centrifuged for 10 min at 8000g. The supernatant was filtered through a 0.22 µm vacuum filtration unit Corning (Sigma). Reverse-phase Chromatography method (RP + C18) was performed for the initial purification and concentration of the filtered supernatant using Diaion HP-20 resin. The eluted solution was collected using absolute acetonitrile. It was then concentrated and dried through rotary evaporator and freeze dryer respectively.
The dried extract from reverse-phase chromatography was analysed through HPLC for the presence of the ComX pheromone. The column was C18 (Thermo Fisher Scientific, 5 μm × 4.6 × 150 mm) with Acclaim 120, C18 5 μm Guard Cartridges (4.6 × 10 mm) on a Dionex ICS-5000 HPLC instrument (Thermo Fisher Scientific). The dried extract was re-dissolved in 200 µl acetic acid, 600 µl acetonitrile and 1200 µl deionised water (1:3:6) to prepare a solution of 25 mgml−1. Two different sequences of amino acids were synthesised as standard (standard 1 and standard 2) which obtained from Pepceuticals Ltd. (Leicestershire, UK). These standard samples were used for pheromone quantification and also to confirm the retention time. To find the standard pheromone sequences for the HPLC run, the whole genome sequence of B. licheniformis NCIMB 8874 and the sequence of the recombinant plasmid (carrying comQX) were studied. The potential pheromone sequence as standard 1 was obtained from the whole genome sequence of B. licheniformis NCIMB 8874 and the corresponding amino acid sequences were identified through IonGap Annotation Service. This sequence was compared also through BLAST to B. licheniformis 9945A ComX sequence with 100% identity. The amino acid sequence of standard 2 was based on the sequence of the recombinant plasmid.
The mobile phase of 20% acetonitrile in 0.1% aqueous ammonium acetate (w/v) at the flow rate of 1.0 ml min−1 washed the system for 5 min equilibration and continued to another 5 min after injection the sample into the column. The run was continued with a linear gradient of 20–55% acetonitrile in 0.1% aqueous ammonium acetate for 20 min. According to the retention time of the two standard samples, the associated fractions (standard 1 retention time: 14.5–15.5 min, and standard 2 retention time: 12.5–13.5 min) were collected and purified using Automated Fraction Collector Dionex UltiMate 3000 (Thermo Fisher Scientific). The pheromone molecule was detected at 210 nm. Detection of peptides and proteins in RP-HPLC, generally involves detection between 210 and 220 nm, which is specific for the peptide bond.
MS/MS and MALDI–MS used to determine the mass spectrometry and the amino acid sequences of pheromone peptide presented in the collected samples from HPLC. This work was performed by Proteomics Services at department of Biology, York University.
β-Galactosidase assay, using a srfA-lacZ reporter strain (B. subtilis JRL293), was performed according to the standard protocol (Tortosa et al. 2001). Different samples were tested to verify the bioactivity of pheromone presented in them. These samples are including; the supernatants which were obtained from the transformed E. coli BL21 cultures before and after the addition of IPTG as well as the supernatant from B. licheniformis NCIMB 8874 culture in the late exponential phase. Besides, the filtered extract from concentrated supernatant of induced recombinant E. coli BL21 (DE3) was also tested.
To evaluate the presence of the comQXPA locus in the draft assembly of B. licheniformis protein, homologues of other bacilli were investigated after annotating the B. licheniformis NCIMB 8874 assembled sequence data through “IonGap Annotation Service” and aligning amino acid sequences of comQXPA cluster from this strain with homologous proteins from other Bacilli using “Clustal Omega” (Stark et al. 2010).
QS-related genes (comQXPA) previously identified and annotated in other Bacilli (Table 1) were compared to the B. licheniformis NCIMB 8874 genome (Accession number MBGK01000000) as homologues at protein level. The percentage identities for four annotated proteins (ComQ, ComX, ComP and ComA) are presented in Table 2.
In B. licheniformis NCIMB 8874, ComQ was identified as a 303-amino-acid protein. The alignment of B. licheniformis NCIMB 8874 ComQ with other homologues showed that the highest degree of identity appeared in other B. licheniformis strains such as 9945A, F11 and ATCC 14580. ComQ from B. amyloliquefaciens FZB42 appears to be the most divergent, with only 40% identity (Table 2).
The precursor of ComX pheromone is a 56-amino-acid protein encoded within the comX locus in B. licheniformis NCIMB 8874 and according to these results, is highly polymorphic. Among Bacillus species, B. licheniformis 9945A and B. sp. BT1B CT2 share 100 and 57% identity, respectively. The other percentage identities range from 50 to 22 (Table 2). In pre-ComX, conservation appears restricted to the N-terminal protein ends. In contrast, high diversity in the C-terminus marks divergent within the pheromone-forming region (Fig. 1). Although the alignment of ComQ and pre-ComX sequences highlights the polymorphism of these proteins at the amino acid level, they could be classified into three main phylogenetic groups (Figs. 1, ,22).
The comP nucleotide sequence of B. licheniformis NCIMB 8874 translated to a 771-amino-acid protein which performed as the sensor histidine kinase of the ComPA-two component system (Parkinson 1995; Kleerebezem et al. 1997; Bassler 1999, 2002). The results obtained from the amino acid sequence alignment revealed a variable distribution of identities ranging from 100% (B. licheniformis 9945A) to 62% (B. mojavensis) within ComP. Interestingly, polymorphism is restricted only to the N-terminal portion of the protein, whereas the C-terminus appears to be conserved (Additional file 1). Phylogenetic analysis further suggests, ComP homologues may be grouped in three distinct clusters (Fig. 2).
Conservation level for the 212-amino-acid protein ComA in Bacilli, confirmed this as the most conserved component of the QS-regulating cluster. Thus, the results showed 99% identity between B. licheniformis NCIMB 8874 and homologues from other B. licheniformis strains and Bacillus sp. BT1B_CT2. The lowest identity was observed with B. amyloliquefaciens FZB42 (Table 2). Phylogenetic analysis of ComA homologues at the protein sequence level again separated these into three main groups (Fig. 2). However, this phylogenetic tree shows the closest distance between B. licheniformis NCIMB 8874 and B. subtilis 168. It could be therefore postulated that the ComA functional implications across these two species might be similar.
comQ and comX were cloned under the control of a T7 promoter in the pET-22b(+) vector [pET-22b(+) comQX] and transformed into E. coli BL21 (DE3) which encodes T7 polymerase under the control of an IPTG-inducible promoter. To avoid contamination by medium components, defined media were used and an amino acid mix added to promote growth. Overproduced pheromone was recovered from the culture supernatant by reverse-phase chromatography and partial purification by gradient reverse-phase HPLC. To determine the optimal retention window for detecting the ComX pheromone, two peptides synthesised based on the B. licheniformis WGS data and the pET-22b(+) comQX plasmid sequence were used as pheromone standards (Table 3). Thus, two fractions were automatically collected by monitoring detection at 210 nm. Based on the retention time of the two standard samples, the associated fractions were collected and purified using the HPLC’s Automated Fraction Collector. The pheromone molecule was detected at 210 nm as this wavelength is used to measure active fractions (Fig. 3).
The fractions were next analysed by Tandem mass spectrometry (MS/MS) and Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry (MALDI–MS), confidently (by the intensity of more than 1.5 a.u.) identifying two dominant ions (Fig. 4) as EAGWGPYPNLWFK (Mass 1) and FSLIEGFKRI (Mass 2). Mass 1 strongly matched the sequence of standard 1 (Table 3), whereas Mass 2 proved identical to the sequence of ComX precursor in B. licheniformis 9945A, as archived in UniProt, under the accession no. D9YRL0 (MQEIVSFLVEHPEVLEQVIAGKASLIGVDKDQVFSLIEGFKRIEAGWGPYPNLWFK). In the present study, this newly identified 10-amino-acid peptide has been reported for the first time, though its role has not yet been investigated in any QS system. Figure 5 shows a schematic model of competence regulation in B. licheniformis by presenting the role of the pheromone in QS system. The mature pheromone is generated by the processing and secretion of the precursor protein ComX by ComQ. Once exported in the extracellular environment the signalling molecule interacts with the histidine kinase sensor protein (ComP), triggering phosphorylation of the response regulator ComA. As a transcription factor, ComA binds to DNA and activates the transcription of the srf operon, thus inducing competence.
To explore ComX pheromone bioactivity a β-galactosidase reporter assay was used based on a reporter, B. subtilis JRL293 strain, carrying a srfA-lacZ fusion enabling pheromone activity quantification through β-galactosidase activity induction (Fig. 6). These studies demonstrated that both B.licheniformis NCIMB 8874 in the late exponential growth phase and E. coli BL21 (DE3) carrying the IPTG-induced ComQX expression cassette in pET-22b(+) resulted in comparable levels of supernatant ComX bioactivity, and therefore comparable pheromone yields. Moreover, filtration extraction resulted in minor ComX activity loss.
The proteins encoded in comQXPA have been investigated and compared with homologues in related species. These studies have shown that the competence regulating locus is highly polymorphic across ComQ, ComX and the region of ComP encoding the N-terminal part of the protein, whilst the C-termini of ComP and ComA are highly conserved (Tran et al. 2000; Tortosa et al. 2001). Our own analysis performed on comQXPA locus of B. licheniformis NCIMB 8874 as sequenced in the present study confirmed further evidences for this pattern of molecular evolution since the ComQ and ComX coding regions of strains WX-02 and NCIMB 8874 were found to share only 85% identity at the nucleotide level. However these regions in ATCC 14580 strain showed 93% identity with NCIMB 8874 strain (Lapidus et al. 2002).
Previous research has revealed that the Bacillus pheromones can be classified in four pherotypes depending on their amino acid sequences and the nature of the post-translational modifications on their tryptophan residues. Accordingly, the pheromones belonging to the same group are able to generate a cross-induction phenomenon (Ansaldi et al. 2002). In this context, the confirmed polymorphism in the comQXP locus suggests a striking pattern of specificity in pheromone interactions with the receptor protein ComP (Tran et al. 2000; Tortosa et al. 2001). However, the present study demonstrated that the ComX pheromone generated by B. licheniformis NCIMB 8874 was able to activate a QS response in B. subtilis. Therefore, comparative analysis between the comQXPA loci of NCIMB 8874 strain and other selected Bacilli supported the further investigation of the relationships between the polymorphisms in this cluster and the specificity exerted in the QS system across different Bacillus spp. and strains.
The utility of reporter systems in the study of bacterial inter-species communication has been demonstrated with respect to the B. subtilis AI-2 signal against low cell-density Vibrio harveyi, a Gram-negative bacterium (Lombardia et al. 2006). In the current study the reporter strain was used to monitor the expression of the srfA operon which is a known requirement for competence development. Thus, B. subtilis srfA-lacZ cultures at low cell densities show srfA expression at basal level. Accordingly, the level of β-galactosidase activity indicates srfA expression levels induced by signalling molecules accumulated in the extracellular medium (Magnuson et al. 1994). The bioassay showed that the supernatant of B. licheniformis NCIMB-8874 collected at the late exponential growth phase as well as IPTG-induced supernatant from E. coli cells transformed with the comQX cassette resulted in the highest expression lacZ pointing towards a comparably high pheromone bioactivity in both samples. The small activity reduction observed after filtration could be attributed to the partial inactivation or loss of the pheromone during purification. In contrast, in the absence of IPTG the transformed E. coli cells show 39 Miller Unit (MU) less pheromone activity, compared to the induced supernatant as a result of srfA expression at the basal level. Future studies should explore the use of B. licheniformis reporter strains to improve the accuracy of data acquisition. Such constructs would contribute to better elucidate QS-regulated secondary metabolite production in the less well known system of B. licheniformis.
To this end, the products of the comQXP locus were individually aligned with homologues, B. licheniformis strains were usually found in the same cluster, alongside the less well-described Bacillus sp. BT1B_CT2. Interestingly, the cluster components ComQ, ComX precursor and ComP across two strains of B. licheniformis, ATCC 14580 and F11, were classified under a distinct group. Previously, it has been reported that these two strains of B. licheniformis harboured non-functional QS systems (Hoffmann et al. 2010). ComA congruence indicated that, whilst this protein is conserved in the same bacterial species, the conservation does not extend to the genus. These findings bear interesting implications on the putative conservation of ComA.
Although B. subtilis subsp. subtilis and B. licheniformis NCIMB 8874 were classified under different phylogenetic groups for ComX and ComP, their positions on pre-ComX evolutionary tree are not too distant, thus confirming the possibility of cross induction between the two species. This is experimentally supported by the pheromone bioactivity outcomes of our reporter studies. The amino acid sequence alignments between the two ComX precursor proteins, however, showed that their conservation is only restricted to the N-terminal ends. Moreover, high diversity in the amino acid sequence in C-terminus marked the pheromone-forming region, where, interestingly, the tryptophan residue is located. Classification of this pheromone under a particular pherotype based on amino acid sequence is not possible since little is known about the mechanism of its modification by ComQ. As our experimental evidence suggests that the ComX pheromone of B. licheniformis NCIMB 8874 may induce a QS response in a B. subtilis reporter strain derived from B. subtilis subsp. subtilis, despite the obvious amino acid sequence divergence. We postulate a common tryptophan modification might account for the observed functional overlap. Interestingly, the percentage identities of pre-ComX and ComP proteins of B. licheniformis NCIMB 8874 and other Bacilli showed that the similarity of pre-ComX of B. licheniformis NCIMB 8874 to associated protein in B. mojavensis and B. subtilis strains is 20–50% while the identity in ComP is 60–70%. This could be an evidence for the existence of amino acid sequence simillarity in ComP than pre-ComX.
Furthermore, null mutants of comQ fail to mature the ComX precursor into a bioactive pheromone (Magnuson et al. 1994). Indeed Tortosa and colleagues reported that co-expression of comQ and comX in E. coli leads to the production of active pheromone in the medium, demonstrating that ComQ is the only dedicated protein required for processing, modification, and release of active ComX pheromone (Tortosa et al. 2001). Our present studies are therefore focusing on the nature of tryptophan modifications present, if any.
In the current study, the finding about the importance of the co-expression of comQ and comX for pheromone production in B. licheniformis NCIMB 8874, helped to design a suitable primer set and subsequently conduct successful comQX gene cloning to produce ComX pheromone of B. licheniformis NCIMB 8874. Applying the plasmid pET-22b(+) as the vector, and the restriction enzymes (BamHI and NdeI) introduced in B. subtilis QS studies (see the methods in Schneider et al. 2002; Ansaldi et al. 2002), led to successful cloning of comQX genes of B. licheniformis NCIMB-8874 for the first time in the present research. The cloned comQX genes was also sequenced and verified in the BLAST algorithm.
The recent investigation of comQXPA-like genes in 2620 complete/6970 draft prokaryotic genomes shows that in addition to B. subtilis and its close relatives, 20 comQXPA-like loci are identified outside the B. subtilis clade, all in the phylum Firmicutes. The sequence variability in the ComX peptide is evident in both B. subtilis and non-B. subtilis clade which suggests grossly similar evolutionary constraints in the underlying quorum sensing system (Dogsa et al. 2014). The pre-ComX protein sequence comparison analysis between B. licheniformis and other Bacilli lead to similar evolutionary conclusion as Dogsa et al. studies since the percentage identity varied from 20 to 100%.
On the basis of these functional observations, we further studied the B. licheniformis QS system components produced recombinantly in E. coli. Mass spectrometry matched one of the two main peaks (Mass 1, 13 amino acid peptide) to the B. licheniformis NCIMB-8874 ComX precursor C-terminus, a synthetic analogue of which we had used as an assay standard (standard 1). Therefore, this sequence most likely corresponds to the ComX pheromone sequence as it compares well with the historically described B. subtilis ComX (Schneider et al. 2002). Curiously, the last reported QS peptide from B. subtilis Ro-E-2 features a distinctly different, 6-amino-acid residue sequence: GIFWEQ (Okada et al. 2005). Given the nucleotide level identity between B. licheniformis NCIMB-8874 pre-comX and B. subtilis Ro-E-2 is only 31%, we postulate that the ComX pheromone amino acid residue in these two strains varies substantially both in size and sequence. This is supported by the degree of phylogenetic divergence across bacilli as reported herein. Moreover, the B. licheniformis NCIMB-8874 strain pheromone shows a N-terminal cleavage site substantially different to that characterised by Ansaldi and co-workers in B. subtilis 168 (2002); thus mature peptides in the Ansaldi et al. studies exhibited diverse lengths ranging 5–10 amino acids.
Interestingly, the second sequence (Mass 2, 10-amino-acid peptide) aligns well with the element of the precursor ComX molecule found in B. licheniformis. This is the first report suggesting that elements of the ComX precursor are also secreted from these bacteria along with the bioactive pheromone peptide. It is presently unclear if this fragment has any biological significance within host QS or the communication systems of other bacteria (Fig. 4).
In conclusion, this project has utilised whole genome sequencing to identify the QS locus in B. licheniformis, enabling its sub-cloning and the biochemical production and characterisation of the previously undescribed B. licheniformis ComX pheromone. Further studies on the chemical structure of this cell communication compound are warranted in the future work alongside the elucidation of any antimicrobial role of the pheromone itself, or the newly described, co-produced precursor fragment (ComE coming from Com-Elham).
EE participated in the design of the study and carried out the molecular studies, genome sequencing, and sequencing data analysis as well as drafted manuscript. SM and TK participated in the design of the study, supervised the work and helped to draft the manuscript. DD helped to draft manuscript and gave some interpretation on data. All authors read and approved the final manuscript.
EE is a Ph.D. graduate in the field of biotechnology and molecular biology. DD is a Ph.D. biotechnologist who works as a development scientist at GlaxoSmithKline. SM is an associate professor in cellular and molecular sciences at Northumbria University. TK is a professor in biotechnology and biochemistry. He works as a graduate centre coordinator at University of Westminster.
We thank the laboratory staff member of the Faculty of Science and Technology, University of Westminster for all the supports. We also thank Dr. Miriam Dwek, Dr. Anatoliy Markiv and Dr. Mark Odell for valuable discussions.
The authors declare that they have no competing interests.
The datasets supporting the conclusions of this article are included within the article (and its supporting materials).
This article does not contain any studies with human participants or animals performed by any of the authors.
This work was part of a PhD project and supported by University of Westminster and the expenses were covered by the department of life sciences.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Elham Esmaeilishirazifard, Email: firstname.lastname@example.org.
Daniela De Vizio, Email: email@example.com.
Sterghios A. Moschos, Email: moc.liamg@soihgrets.
Tajalli Keshavarz, Email: firstname.lastname@example.org.