Holomycin, an Antibiotic Secondary Metabolite, Is Required for Biofilm Formation by the Native Producer Photobacterium galatheae S2753

Analysis of the biosynthetic gene cluster of holomycin of P. galatheae S2753.

antiSMASH, version 5.0, identified 11 BGCs in P. galatheae S2753, eight of which were on the large chromosome and three on the small one. Based on a previous prediction (17), BGC11 on the small chromosome potentially encodes the enzymes involved in the biosynthesis of holomycin. Sequence analysis and functional annotation showed that BGC11 contains 10 genes, eight of which are homologous to those in the holomycin BGCs of Y. ruckeri and S. clavuligerus (Fig. 1A). Therefore, the holomycin biosynthesis in P. galatheae likely follows the same paths as those in S. clavuligerus and Y. ruckeri (31, 32). As a core protein, the nonribosomal peptide synthetase (NRPS) HlmE contains a characteristic arrangement of cyclization (Cy), adenylation (A), and thiolation (T) domains (Fig. 1A and Table 1). According to the proposed mechanism (31, 32), HlmE initiates the synthesis pathway by covalently loading an l-cysteine and forming a dipeptide bond with a second l-cysteine. An acyl-coenzyme A (CoA) dehydrogenase, HlmB, then oxidizes the thiol group to allow the cyclization of the aminopyrrolinone ring of holomycin. Subsequently, a thioesterase, HlmC, a PPC-DC decarboxylase, HlmF, and a flavin mononucleotide (FMN)-dependent oxidoreductase, HlmD, work together to generate the second thiol group, remove one molecule of carbon dioxide, and then form a dihydroholothin molecule. Finally, an N-acyltransferase, HlmA, adds an acetyl group to form a holomycin.

Besides the eight conserved genes, BGC11 harbors a gene, hlmX, encoding a homolog to Hom9 of Y. ruckeri (Table 1), and a unique gene, hlmY (Fig. 1A), which encodes a putative metallophosphoesterase (Table 1). Both HlmX and Hom9 were predicted as putative ArsR/SmtB-family transcriptional regulators with a C-terminal rhodanese-like domain.

The other two holomycin-producing bacteria, S. clavuligerus and Y. ruckeri, use different self-protection strategies during holomycin production. S. clavuligerus encodes a disulfide-forming dithiol oxidase, HlmI, to control the formation of the intramolecular disulfide bridge in holomycin (35), while Y. ruckeri deploys a RNA methyltransferase, Hom12, to modify the potential holomycin antibiotic targets (32). However, a homolog of hlmI (35) or hom12 (32) was not found in S2753 BGC11 or in other places on the genome.

BGC11 is responsible for holomycin production in S2753.

To experimentally test if BGC11 indeed is responsible for holomycin production, an in-frame scarless deletion mutant of the core gene hlmE was generated by using homologous recombination and SacB-mediated counterselection (see details in Materials and Methods). Diagnostic PCRs were performed and confirmed that hlmE was deleted from S2753 (Fig. 1B). Wild-type (WT) S2753 and the ΔhlmE strains were then grown to stationary phase in different media, i.e., the marine-enriched APY medium and a marine minimal medium supplemented with chitin, glucose, or mannose. The produced secondary metabolites were extracted and analyzed by high-performance liquid chromatography coupled to diode array detection and high-resolution mass spectrometry (HPLC-DAD-HRMS). While holomycin was detected in WT cultures, it was not produced by the ΔhlmE mutant (Fig. 1C and D), demonstrating that hlmE and, thus, BGC11 are responsible for the biosynthesis of holomycin in P. galatheae S2753. Consistent with this, the extracted metabolites of ΔhlmE cultures failed to inhibit the growth of either the Gram-negative bacterium Vibrio anguillarum (Fig. 1E) or the Gram-positive Staphylococcus aureus (Fig. 1F), both of which were inhibited by the extracts from the WT strain. The antimicrobial activity of the crude extracts was complemented by ectopically expressing hlmE under its native promoter on the plasmid vector pBBR1-MCS2 but not by a vector control (Fig. 1G).

Biofilm formation is reduced in the ΔhlmE mutant.

Deletion of hlmE did not affect the growth rate, as the doubling time during exponential growth and the maximum yield were 0.71 ± 0.04 h and 6.21 ± 0.56 optical density (OD) units for the WT and 0.70 ± 0.03 h and 6.21 ± 0.34 OD units for the ΔhlmE mutant in marine media. Statistically, the growth of the ΔhlmE mutant was not significantly different from that of the WT (P > 0.05 by Student's t test). Despite the similar growth dynamics, the ΔhlmE mutant formed less biofilm after a 2-day incubation than the WT (Fig. 2). The biofilm formation was partially restored in the ΔhlmE::pBBR1-MCS2-hlmE complemented strain that was used in Fig. 1G but not restored in the ΔhlmE::pBBR1-MCS2 control strain with the empty vector (Fig. 2). The partial complementation may be due to the varied expression time and level of hlmE.

Biofilm formation by the P. galahtheae ΔhlmE strain is restored by exogenously applied holomycin but not by S,S′-dimethyl–red-holomycin.

Since holomycin is secreted extracellularly, the effect of holomycin production on biofilm formation could be mediated by its role in intercellular interaction, and addition of exogenous holomycin should restore biofilm formation of the ΔhlmE mutant. To test this hypothesis, the production of endogenous holomycin was first determined as 3.59 ± 0.05 μM in the WT biofilm formation cultures. Holomycin with final concentrations of 0, 0.9, 1.8, 3.6, and 7.2 μM was then added to the ΔhlmE cultures after 17 h of incubation at 25°C, at the time when cells entered the exponential-stationary transition phase and were expected to produce holomycin. Holomycin supplied in concentrations higher than 1.8 μM restored the biofilm formation of the ΔhlmE mutant to the same level as the WT cultures (Fig. 3, P > 0.05 by Student's t test). To further explore the functional group of holomycin in triggering biofilm formation, S,S′-dimethyl holomycin (Fig. 3), a methylated, less antibacterial chemical analogue of holomycin (36), was also added to the ΔhlmE cultures after 17 h of incubation. Biofilm formation of ΔhlmE cultures was not restored even by S,S′-dimethyl holomycin concentrations up to 7.2 μM (Fig. 3), suggesting that the disulfide group is essential for the role of holomycin in triggering biofilm formation or that the two methyl groups in S,S′-dimethyl holomycin diminish the biological function of holomycin in triggering the biofilm formation of P. galatheae S2753.

The zinc ion concentration in media negatively correlates with holomycin production in P. galatheae S2753.

The disulfide group of holomycin chelates zinc ions in reducing environments (37). To test if zinc was involved in holomycin production and biofilm formation, wild-type P. galatheae was grown in mannose marine minimal medium with and without zinc addition. A calibration curve of holomycin was constructed and used for the quantitative analysis of holomycin production in cultures. The growth of P. galatheae was not influenced by zinc at the tested concentration; however, the increasing zinc concentration gradually inhibited holomycin production in the WT (Fig. 4). Holomycin production in WT cultures was almost abolished when zinc was added up to 2 mM in the wild-type cultures. Meanwhile, biofilm formation of wild-type P. galatheae was also gradually inhibited by increasing zinc concentrations (Fig. 4). To sum up, zinc negatively influenced biofilm formation and holomycin production in P. galatheae.

Exogenous holomycin triggers biofilm formation of other marine isolates.

To test whether holomycin at subinhibitory concentrations would affect the biofilm formation of other marine bacteria, 16 isolates from six genera (i.e., Phaeobacter, Ruegeria, Vibrio, Photobacterium, Pseudoalteromonas, and Cobetia) from the Galatheae Collection (Tables 2 and 3) were grown in marine broth. We first determined the MICs of holomycin for these strains. Five strains were resistant to holomycin (MIC, >93 μM) (Table 2), including S2754 and S2755, which were isolated from the same stone where the S2753-bearing mussel was located. The MICs of other bacteria were in the range of 2.9 to 93 μM (Table 2). Holomycin was then added to the cultures at subinhibitory concentrations at either 0 h or 17 h (Fig. 5 and 6). Biofilm formation of three Vibrio strains and one Photobacterium strain was significantly enhanced by holomycin added after the 17-h incubation (Fig. 6), while the effect was not significant when added at 0 h (Fig. 6). Biofilm formation of the other strains was not affected by holomycin at either time point (Fig. 5). These data showed that holomycin produced by S2753 could affect the biofilm formation of other marine bacterial species.

Bioinformatics analyses.

The genome sequences used in this study were extracted from NCBI using the relevant accession numbers and uploaded to the MaGe Genoscope for holomycin BGC synteny analysis and gene annotation (55, 56). Genomes were submitted to antiSMASH, version 5.0 (33), for the prediction of gene clusters involved in the production of holomycin. Protein domain prediction was done in InterPro Domains and Conserved Domain Search Service (CD Search) (57). The DNA sequencing data were analyzed using BioEdit. The sequence alignment was done in ClustalX.

Microorganisms and growth conditions.

Escherichia coli strains PIR1 (C101010; Invitrogen, Denmark) and TOP10 (404010; Invitrogen, Denmark) were used for cloning. E. coli WM3064 (58) was used as the donor strain in bacterial conjugations and grown in the presence of 300 μM 2,6-diaminopimelic acid (DAP). A concentration of 10 μg/ml kanamycin (Kan) or chloramphenicol (Cm) was used in the E. coli liquid cultures, and 30 μg/ml both antibiotics was used in E. coli agar cultures (59). Cultures of wild-type and mutant strains of P. galatheae S2753 (16, 25) were grown in marine minimal medium (52) supplemented with sole carbon sources (0.2% mannose, 0.2% colloidal chitin, or 0.2% glucose), marine broth 2216, marine agar 2216, and modified enriched growth medium (APY) (56) containing, per liter, 5 g of peptone, 3 g of yeast extract, and artificial seawater. The pH was adjusted to 7.0; 12 g of agar was added per liter of APY to prepare the agar plates. Generally, colonies appeared 15 h after plating. In the P. galatheae cultures, kanamycin was added at 200 μg/ml to agar plates and 150 μg/ml to liquid cultures; chloramphenicol was used at 30 μg/ml. Zinc ions were added to the medium to the working concentration from a 1 M zinc chloride stock solution (pH 6.5) in water. Sixteen marine bacteria were selected to investigate the influence of holomycin in their biofilm, of which eight strains, i.e., Vibrio sp. strains (16) S0703, S1396, S1399, S2757, S3027, and S3030, Pseudoalteromonas ruthenica S2756 (16), and Cobitia sp. strain S3029 (16), were isolated from mussel surfaces. Both Pseudoalteromonas piscicida S2755 (16) and Photobacterium sp. S2754 (16) were sampled from a stone located in the same place as S2753. Bacteria S2541 and S2545 were also included because they belong to the genus Photobacterium (16). Additionally, Ruegeria sp. strain S2684 (16), Vibrio coralliilyticus S2052 (16), Phaeobacter piscinae S26 (19), and Pseudoalteromonas galatheae S4498 (16) were selected, as they were studied in several previous or ongoing projects. These selected marine strains were cultured on marine agar (MA) plates or in marine broth (MB). All marine cultures were inoculated at 25°C and all E. coli strains at 37°C. The strains used in this study are listed in Table 3 with genotype description.

DNA manipulation and plasmid construction.

The restriction enzymes and the quick ligase for DNA modification were purchased from New England Biolabs (NEB; Bionordika, Denmark). DNA polymerase (TaKaRa Biomedical Technology Europe [France]) and Q5 high-fidelity polymerase (NEB) were used for PCR amplification, except for colony PCRs, which were performed using TEMPase (Ampliqon; VWR, Denmark). All PCR products and plasmids were purified using GFX PCR DNA and gel band purification kit (28–9034-70; GE Healthcare) and Monarch plasmid miniprep kit, respectively. All plasmids and primers were designed in ApE–A plasmid Editor (v2.0) (a program designed by M. Wayne Davis). Integrated DNA Technologies (IDT; Belgium) synthesized all the primers in this study. All plasmids and primers used in the study are listed in Tables 4 and 5.

The suicide plasmid pDM4-del-hlmE was constructed by the restriction cloning method. Approximately 1.0-kb upstream and downstream regions flanking hlmE were amplified using primer pairs DhlmE-P1/2 and DhlmE-P3/4 (Table 5). Amplified DNA fragments were ligated into the pJET1.2/blunt cloning vector with the CloneJET PCR cloning kit (K1231; Thermo Scientific, Denmark); subsequently, they were subcloned into the suicide vector pDM4 (60) by using the XbaI and XhoI digestion sites to form pDM4-del-hlmE. Gene hlmE and its native promoter region were amplified by primer pair P_hlm-hlmE/hlmE-6×His (Table 5) and cloned into the expression vector pBBR1-MCS2 via KpnI and XbaI to generate the complementation plasmid pBBR1-MCS2-hlmE. Correct plasmid assembly was confirmed by PCR (Table 5), restriction digestion, and sequencing (Macrogen Europe, Netherlands).

Bacterial conjugation.

The electroporation of E. coli WM3064 and the conjugation experiments were performed as described previously, with some modifications of the culture conditions (58, 59, 61). WM3064 cells carrying each plasmid were grown at 37°C in LB-DAP medium with antibiotics until an optical density at 600 nm (OD₆₀₀) of 0.4 to 0.6. As a donor, 1 ml of WM3064 culture was harvested by centrifugation (6,000 × g for 1 min). Cells were washed twice with LB medium and resuspended in 50 μl LB medium with DAP. It was then mixed with 500 μl of P. galatheae culture with an OD₆₀₀ between 0.4 and 0.5. The donor-recipient cell suspension was concentrated by centrifugation (6,000 × g for 1 min), resuspended with 20 μl APY medium with 300 μM DAP, and mixed briefly by pipetting three times. All mixtures were spotted onto a 0.22-μm-pore-size mixed-cellulose ester (MCE) membrane (GSWP02500; MF-Millipore, Merck, Germany) placed on an MA 2216 plate with 300 μM DAP. The plate was incubated for 3 to 4 h at 37°C or for 3 to 15 h at 25°C. The conjugations were suspended in 1 ml of APY medium and incubated for 20 min at 25°C. Each 100-μl conjugation was plated onto antibiotic-containing APY plates. The plates were incubated at 25°C for 16 to 24 h. Following conjugation, single colonies were grown in 2 ml APY medium supplied with chloramphenicol at 25°C with shaking overnight. Resulting antibiotic-resistant strains were screened by PCR to determine the transconjugants.

Confirmation of the transconjugants and first cross event.

Genomic DNA for PCR analyses was isolated using a NucleoSpin tissue kit according to the protocol of the manufacturer (Macherey-Nagel, Fisher Scientific, Denmark). PCR primers are listed in Table 5. Primer pairs Cm-Fw/Rv, Km-Fw-Rv, Pc0/Pc4, and Pc1/Pc2 were designed to amplify the replication region for detecting the plasmids in donor strains and transconjugants. The PCR steps were 94°C for 30 s; 30 cycles of 94°C for 10 s, 58°C for 5 s, and 72°C for 30 s; and then 72°C for an additional 10 min. Routine DNA manipulations were carried out by following standard methods as described above.

Construction of the hlmE in-frame deletion mutants in S2753.

The suicide plasmid for knocking out the hlmE gene was constructed by restriction cloning and was transferred into S2753 by intergeneric conjugation as described. The PCR-verified mutants, in which the suicide plasmid had integrated into the anticipated place in the S2753 genome, were grown at 25°C in APY medium with shaking to an OD₆₀₀ of 0.5. The cells were then diluted and spread on a half-APY medium (500 ml/liter APY medium, 500 ml/liter distilled H₂O, 15 g agar, pH 7.0) supplied with 10% (wt/vol; final concentration) sucrose (autoclaved at 100°C for 1 h or filtered) and incubated at 16°C for 48 h. All primers used in this study are listed in Table 3.

Purification of the deletion mutants.

P. galatheae S2753 swarms on agar plates when cultured below 42°C (unpublished data). Therefore, several purification steps following the second crossover event are required to get a genetically homogenous clone. Cells from the edge of a swarming colony were inoculated in 1 ml APY medium without antibiotics with shaking at 25°C for 8 h. The cells were then diluted and transferred onto a new half-APY–10% sucrose medium plate and incubated at 16 to 18°C for 48 h. Single colonies on the new plate were transferred onto APY-agar and APY-agar plates containing 30 μg/ml chloramphenicol. Colonies sensitive to chloramphenicol were collected and confirmed by PCR and DNA sequencing using the same protocol of verifying mutants with the first crossover. If the PCR result showed a mosaic genetic feature of the selected colonies, the purification step was repeated at least once.

Extraction of liquid cultures for chemical analysis.

Chemical extraction was prepared as described by Giubergia et al. (52). Cultures were incubated at 25°C for 48 h with shaking and then transferred to 50 ml falcon tubes. An equal volume of ethyl acetate was added to the culture and mixed by inversion. The mixture was incubated for 10 min with occasional inversion until a clear division of layers was present. The organic phase (top layer) was transferred to a glass tube. These tubes were placed in a 35°C heating block and evaporated with nitrogen until dry. Extracts were resuspended in methanol (1/20 volume of the initial culture) and stored at −20°C.

UHPLC-HRMS profiling of holomycin from the wild-type and mutant strains.

Chemical analysis was performed as described by Giubergia et al. (52). Ultra-high-performance liquid chromatography–high-resolution mass spectrometry (UHPLC-HRMS) was performed on an Agilent Infinity 1290 UHPLC system (Agilent Technologies, Santa Clara, CA) equipped with a diode array detector. The separation was obtained on an Agilent Poroshell 120 phenyl-hexyl column (2.1 by 150 mm; particle size, 1.9 μm) with a linear gradient consisting of water and acetonitrile, both buffered with 20 mM formic acid, starting at 10% acetonitrile and increasing to 100% in 10 min, at which point the concentration was held for 2 min, returned to 10% acetonitrile in 0.1 min, and left for 3 min (0.35 ml/min, 60°C). An injection volume of 1 μl was used. MS detection was performed in the positive detection mode on an Agilent 6545 quadrupole time-of-flight (QTOF) MS equipped with an Agilent dual-jet-stream electrospray ion source with a drying gas temperature of 250°C, a gas flow of 8 liters/min, a sheath gas temperature of 300°C, and a flow rate of 12 liters/min. The capillary voltage was set to 4,000 V and nozzle voltage to 500 V. Mass spectra were recorded at 10, 20, and 40 eV as centroid data for m/z 75 to 1,700 in MS mode and m/z 30 to 1,700 in MS/MS mode, with an acquisition rate of 10 spectra/s. Lock mass solution in 70:30 methanol-water was infused into the second sprayer using an extra LC pump at a flow rate of 15 μl/min and a 1:100 splitter. The solution contained 1 μM tributylamine (Sigma-Aldrich) and 10 μM hexakis(2,2,3,3-tetrafluoropropoxy)phosphazene (Apollo Scientific Ltd., Cheshire, United Kingdom) as lock masses. The [M+H]⁺ ions (m/z 186.2216 and 922.0098, respectively) of both compounds were used. The secondary metabolite profile was analyzed in Agilent Qualitative Analysis B.07.00. Five series of calibration solutions of pure holomycin (H458490; Toronto Research Chemicals, Canada) were used to create the HPLC standard calibration curve of holomycin. The peak area of holomycin in the biofilm samples and zinc cultures was recorded and used to calculate the holomycin concentration in cultures.

Well diffusion inhibition assay.

This experiment was performed with a modified protocol from Wietz et al. (24). Vibrio anguillarum 90-11-287 and Staphylococcus aureus 8325 were cultured at 25°C for 24 h in MB and LB media, respectively. To test the susceptibility of the two pathogenic strains to the extracts from P. galatheae S2753 cultures, the strains were homogeneously added into warm (44.5°C) IO agar (3% instant ocean, 0.3% Bacto Casamino Acids [BD-223050; Denmark] supplemented with 0.4% glucose, 1% agar) (for V. anguillarum) or IO agar with 1% peptone (for Staphylococcus aureus 8325). The plates were solidified and dried in a flow bench. Wells with 6-mm diameter were punched with home-made tips, and 45 μl of culture extract was added to each well. The agar plates containing the V. anguillarum 90-11-287 and S. aureus 8325 cultures were incubated at 25 and 37°C for 48 and 24 h, respectively. The inhibition assay was then evaluated by analyzing the formation of clearing zones around the well.

Growth experiments.

Precultures of the P. galatheae WT and mutant were prepared by inoculating a single colony into the proper liquid medium. After 24 h of incubation at 25°C, the preculture was diluted to an OD₆₀₀ of 0.01 in 30 ml medium in a 250-ml flask and incubated at 25°C with 160-rpm shaking. The OD₆₀₀ value was measured in a 1-ml cuvette every 0.5 to 6 h for 72 h using a Novaspec III spectrophotometer (Amersham Biosciences) and plotted using Origin, version 2019 (OriginLab Corporation, Northampton, MA, USA). After the final OD₆₀₀ measurement, cultures diluted to 10⁻⁶ and 10⁻⁷ were plated on marine agar and incubated overnight for colony counting.

Biofilm formation.

A modified protocol based on previous works (62–64) was used. Precultures were diluted to an OD₆₀₀ of 0.01 in 1 ml MB. Border wells were filled with 100 μl Milli-Q water to prevent desiccation. The 96-well microtiter plate was incubated in a humidity chamber with a wet paper towel at the bottom for 48 h at 25°C. In the complementation experiments, growth kinetics were tracked and holomycin or S,S-dimethyl-holomycin synthesized by the method of Buijs et al. (65) were added to the cultures at the exponential-stationary transition phase (17 h). In the biofilm assay of selected Galatheae Collection strains, holomycin was added to cultures in a 2-fold dilution from 93 μM to 1.5 μM, either at the initial inoculation time or after 17 h of incubation. After incubation, the OD₆₀₀ was measured in a SpectraMax i3 (Molecular Devices). Culture media and nonadhering bacteria were removed, and the wells were washed with 150 μl Milli-Q water and dried for 15 min in a flow bench. To each well was added 125 μl of 1% crystal violet, and staining proceeded for 15 min. After removing the crystal violet, wells were washed three times with 200 μl Milli-Q water and dried for another 15 min. An amount of 200 μl 96% ethanol was added to each well and incubated for 30 min to dissolve the staining color. Thereafter, 100 μl of the ethanol-crystal violet mixture in each well was transferred to a new microtiter plate. The crystal violet intensity was measured at OD₅₉₀ in a SpectraMax i3. Data were analyzed in Microsoft Excel. One-way analysis of variance (ANOVA) test and the statistical plot graphs were analyzed in Origin, version 2019 (OriginLab Corporation, Northampton, MA, USA).