Pumilacidin-Like Lipopeptides Derived from Marine Bacterium Bacillus sp. Strain 176 Suppress the Motility of Vibrio alginolyticus

ABSTRACT

Bacterial motility is a crucial factor during the invasion and colonization processes of pathogens, which makes it an attractive therapeutic drug target. Here, we isolated a marine bacterium (Vibrio alginolyticus strain 178) from a seamount in the tropical West Pacific that exhibits vigorous motility on agar plates and severe pathogenicity to zebrafish. We found that V. alginolyticus 178 motility was significantly suppressed by another marine bacterium, Bacillus sp. strain 176, isolated from the same niche. We isolated, purified, and characterized two different cyclic lipopeptides (CLPs) from Bacillus sp. 176 using high-performance liquid chromatography, mass spectrometry, and nuclear magnetic resonance spectroscopy. The two related CLPs have a pumilacidin-like structure and were both effective inhibitors of V. alginolyticus 178 motility. The CLPs differ by only one methylene group in their fatty acid chains. In addition to motility suppression, the CLPs also induced cell aggregation in the medium and reduced adherence of V. alginolyticus 178 to glass substrates. Notably, upon CLP treatment, the expression levels of two V. alginolyticus flagellar assembly genes (flgA and flgP) dropped dramatically. Moreover, the CLPs inhibited biofilm formation in several other strains of pathogenic bacteria without inducing cell death. This study indicates that CLPs from Bacillus sp. 176 show promise as antimicrobial lead compounds targeting bacterial motility and biofilm formation with a low potential for eliciting antibiotic resistance.

IMPORTANCE Pathogenic bacteria often require motility to establish infections and subsequently spread within host organisms. Thus, motility is an attractive therapeutic target for the development of novel antibiotics. We found that cyclic lipopeptides (CLPs) produced by marine bacterium Bacillus sp. strain 176 dramatically suppress the motility of the pathogenic bacterium Vibrio alginolyticus strain 178, reduce biofilm formation, and promote cellular aggregation without inducing cell death. These findings suggest that CLPs hold great promise as potential drug candidates targeting bacterial motility and biofilm formation with a low overall potential for triggering antibiotic resistance.

INTRODUCTION

Bacterial motility is one of several crucial factors during the initial infection and colonization processes of pathogens. In particular, it helps pathogens overcome the repulsive forces between the bacterial cell wall and the host tissues and facilitates host attachment (1–3). The flagellum is one of the main locomotive organelles responsible for bacterial motility and is thus considered an important virulence factor in many pathogens (2, 4, 5). Flagellar motility is a complex biological process requiring the synthesis and assembly of numerous flagellar components, energetic coupling, and chemotactic control over flagellar rotation to afford directed motion (3, 4).

Vibrio alginolyticus is a common Gram-negative halophilic bacterium. It is an opportunistic pathogen and a major cause of vibriosis in aquatic animals, resulting in severe economic losses worldwide (6, 7). Recent studies suggested that V. alginolyticus is also a potential human pathogen (8). Unlike other Vibrio species, V. alginolyticus usually infects tissues via superficial routes (i.e., wounds, ears, and eyes) rather than the gastrointestinal tract (8). In V. alginolyticus, there are two types of flagella, polar and lateral. Polar flagella are constitutively expressed, while lateral flagellar expression is induced when polar flagellar function has been disabled (9–11). Furthermore, the flagellar assembly pathway of V. alginolyticus has been implicated in cell adhesion processes (12, 13).

Overuse of broad-spectrum antibiotics and the accompanying proliferation of drug-resistant bacteria have stimulated efforts to develop environment-friendly biocontrol measures to reduce health hazards and environmental pollution (14, 15). In recent years, the antimicrobial properties of biological surfactants have been increasingly recognized and harnessed for antibacterial, antifungal, and antiviral applications (16–18). Lipopeptides are the most widely reported class of biosurfactants having antimicrobial and antiadhesive activity against pathogenic bacteria, due to the amphipathic nature of their peptide and fatty acid components (19, 20). Many antimicrobial lipopeptides identified and characterized to date are derived from bacterial species in the Bacillus genus. These include surfactin, fengycin, iturin, bacillomycin, mycosubtilin, lichenysin, and pumilacidin (21–26). To date, none of these lipopeptides have been reported to possess any antimotility activity.

In this study, we demonstrate that the marine bacterium V. alginolyticus strain 178 exhibits strong motility on agar plates and causes high levels of mortality in zebrafish. More excitingly, we show that this motility is dramatically inhibited by two cyclic lipopeptides (CLPs) derived from a competing bacterium (Bacillus sp. strain 176). We purified and characterized the active antimotility compounds and determined their structural and functional properties. In order to explore the mechanism of action of the CLPs, we also investigated their impact on cell aggregation, adherence, and the expression of flagellar assembly components in V. alginolyticus.

RESULTS

The discovery and pathogenicity assays of the vigorously swarming strain V. alginolyticus 178.

The initial purpose of this study was to investigate the marine bacterial resource in the seamount area of the tropical Western Pacific. During the course of bacterial isolation, V. alginolyticus strain 178 attracted our attention, because it displayed vigorous swarming on the 2216E medium with 1% agar (Fig. 1A). To further analyze the swarming of strain 178, we investigated its motility on plates with different agar concentrations varying from 0.5% to 2%. Surprisingly, strain 178 could cover the entire 9-cm plate surface within 12 h with an agar concentration up to 1% (Fig. 1A). With the increase of agar concentration in the medium, the motility of this bacterium gradually decreased. However, it could still occupy the full plate surface within 24 or 36 h even with agar concentrations up to 1.5% or 2%, respectively (Fig. 1A). In view of the high homology (99% identity) with marine bacterium Vibrio alginolyticus by 16S rRNA gene sequencing and phylogenetic analysis (see Fig. S1A in the supplemental material), the vigorously swarming strain was designated V. alginolyticus 178. To further validate the strong motility ability of V. alginolyticus 178, we compared its motility ability with that of seven other V. alginolyticus strains (C1, C11, C38, C48, CJ11, CJ26, and CT30). The results showed that most V. alginolyticus strains were motile; however, V. alginolyticus 178 exhibited the greatest swarming motility of all eight strains analyzed (see Fig. S2).

FIG 1.

Characterization of marine bacterium strain V. alginolyticus 178. (A) Motility assays of V. alginolyticus 178 in the plates containing 0.5%, 1%, 1.5%, or 2% agar at different time points. (B) The pathogenicity assay of V. alginolyticus 178 to zebrafish. (C) Cumulative mortality rates of zebrafish caused by V. alginolyticus 178.

V. alginolyticus is a leading cause of vibriosis, causing opportunistic infections in humans in association with raw seafood contamination (8). To identify the possible pathogenicity of V. alginolyticus 178, we performed injection experiments in zebrafish. The results showed that dead zebrafishes were observed on the first day after infection, and mortalities continued increasing until the sixth day of treatment with high (3 × 10⁴ cells) and low (3 × 10³ cells) concentrations of V. alginolyticus 178 cells. Compared with zebrafish in the control group, obvious hemorrhage and abdominal dropsy were observed in zebrafish after injection with V. alginolyticus 178 (Fig. 1B). The cumulative mortality rates in the groups challenged with high and low bacterial concentrations reached 100% and 50%, respectively, and were significantly higher (P < 0.01) than that of the control group (Fig. 1C). For Koch's postulates, 87.33% (262/300) of the reisolated bacteria were identified as sharing the same 16S rRNA gene fragment with strain 178. Thus, V. alginolyticus 178 is a marine bacterium with strong motility and severe pathogenicity.

Screening and identification of marine bacteria with antimotility activity against V. alginolyticus 178.

Bacterial motility is proposed to be closely related to virulence (4). Therefore, motility may be an attractive target for drugs that can prevent the infective process. To find potential natural products inhibiting the motility of V. alginolyticus 178, we performed the screening described in the “Materials and Methods” section. Notably, bacterium strain 176 was demonstrated to significantly suppress the motility of V. alginolyticus 178 based on the plate assays (Fig. 2A). Also, the colony margins of V. alginolyticus 178 showed clear differences in morphology with or without coculturing with strain 176 (Fig. 2B). Furthermore, the expansion of the colony margins of V. alginolyticus 178 was inhibited even without touching strain 176, which led to the hypothesis that bacterium strain 176 produced an extracellular diffusible inhibitor of V. alginolyticus 178 motility.

FIG 2.

Antimotility assay of Bacillus sp. 176 (b) against V. alginolyticus 178 (a). (A) Plate-based motility assay shows a zone of inhibition around the colony of Bacillus sp. 176. (B) Morphology changes in the margins of V. alginolyticus 178 in the absence (left) or presence (right) of Bacillus sp. 176.

It is noteworthy that strain 176 was isolated from the same niche as that of V. alginolyticus 178, and its 16S rRNA gene sequence shared high similarities (greater than 98%) with Bacillus aerophilus, Bacillus aerius, and Bacillus pumilus (see Fig. S1B in the supplemental material). Additionally, strain 176 also clustered with these strains according to phylogenetic analysis (see Fig. S1B). Strain 176 was thus defined as Bacillus sp. 176.

Purification, structural elucidation, and activity assays of lipopeptides derived from Bacillus sp. 176.

To obtain the active compound inhibiting the motility of V. alginolyticus 178, purification was performed as described in “Materials and Methods.” The filter paper disc assay was carried out to follow the trail of the bioactive compound(s) using V. alginolyticus 178 as the indicator bacterium. Notably, two compounds were found to have antimotility activity based on high-pressure liquid chromatography (HPLC) analyses (see Fig. S3A in the supplemental material). Electrospray ionization mass spectrometry (ESI-MS) results showed that the mass-to-charge ratios (m/z) of compounds 1 and 2 displayed the predicted molecular formulas C₅₇H₁₀₁N₇O₁₃ and C₅₈H₁₀₃N₇O₁₃, respectively (see Fig. S3B). Based on our purification scheme, we deduced that these two compounds were lipopeptides; therefore, the amino acid sequences of these two compounds were further analyzed by tandem mass spectrometry (MS/MS). In the MS/MS spectra, fragment ion sets were observed in compounds 1 and 2, and similar molecular weights of [M + Na]⁺ were detected at m/z 1114.7336 and 1128.7505 (Fig. 3A). For compound 1, the peptide chain sequence was methoxy glutamic acid (mGlu)-leucine (Leu)-Leu-Leu-methoxy aspartic acid (mAsp)-Leu-isoleucine (Ile) (Fig. 3B and C). The cleavage mode of amino acid residues indicated the existence of a cyclic structure (27). MS/MS data also showed that compound 2 contained the same amino acid sequence as did compound 1 (Fig. 3B and C).

FIG 3.

Structural elucidation of cyclic lipopeptides (CLP1, left; CLP2, right) produced by Bacillus sp. 176. (A) MS/MS spectra of CLP1 and CLP2. (B) Fragment ions of CLP1 and CLP2. (C) Structures of CLP1 and CLP2. The arrows indicate the positions of CLP1 and CLP2 that are different.

Moreover, ¹H nuclear magnetic resonance (NMR) and ¹³C NMR spectroscopy analyses were performed to clarify the exact structure of these active compounds. In the ¹H NMR spectrum, a long alkyl chain (δ 1.4-1.1) and a peptide backbone (7 amine protons, δ 8.22 to 7.76 and 7 α-amino acid protons, δ 4.6 to 4.1) were observed in compound 1. A further proton next to oxygen was detected with a resonance at δ 5.02, which might be a part of lactone ring. Also, two intense singlets at δ 3.60 and 3.58 were similar to those in the ¹H NMR spectrum of the lipopeptide esters (28, 29), suggesting the existence of two methoxy groups in compound 1 (see Fig. S4 in the supplemental material). In the ¹³C NMR spectrum, signals of 10 carbonyl groups (δ 173.36 to 169.98) were detected in compound 1, 9 of which were attributed to amino acids. The resonances at δ 50.11 and 51.10 in the ¹³C NMR spectrum also demonstrated the methoxy group attached on the Glu and Asp residues, respectively. In addition, a signal at δ 72.18 (corresponding to δ_H 5.02) was characteristic for β-hydroxy fatty acids (30). The similar structure of compound 2 was identified by NMR analysis, in which an excess methylene group was located on its alkyl chain (see Fig. S5). According to the results of structural analyses, compounds 1 and 2 were defined as pumilacidin-like lipopeptides (CLP1 and CLP2) by comparison with spectroscopic data in previous studies (Fig. 3C) (26).

Next, we checked the antimotility activities of these two cyclic lipopeptides (CLP1 and CLP2). Notably, CLP1 and CLP2 (100 μg/ml) showed similar antimotility activity against V. alginolyticus 178, but there was no significant synergistic antimotility effect observed (Fig. 4A). Hence, the cyclic lipopeptide used in the present study was a mixture with equal proportions of CLP1 and CLP2 and defined as CLPs in the following study. Furthermore, the bactericidal effect of CLPs against V. alginolyticus 178 was investigated. The results showed that the growth of V. alginolyticus 178 in the presence of CLPs with a final concentration of 100 μg/ml was similar to that of the negative control group (Fig. 4B). This indicated that CLPs from strain 176 affected only the motility of V. alginolyticus 178 but did not kill it. Motility is a crucial virulence factor of Vibrio spp.; thus, we checked the pathogenicity of V. alginolyticus 178 after treatment with CLPs. The results showed that CLP-treated V. alginolyticus 178 caused a significantly lower (P < 0.01) cumulative mortality rate than that caused by control V. alginolyticus 178 and killed only 26.7% of zebrafishes after injection for 7 days (Fig. 4C), which strongly suggested that CLPs could effectively reduce the pathogenicity of V. alginolyticus 178.

FIG 4.

Effects of CLP1 and CLP2 on the motility and growth of V. alginolyticus 178. (A) Plate-based antimotility activity assays of CLP1 (a), CLP2 (b), CLP1 + CLP2 (c), and DMSO (d). Sterile filter paper disks, each containing 200 μg of purified CLPs or a DMSO control, are indicated. (B) Growth assay of V. alginolyticus 178 in the presence of 100 μg/ml of CLPs or DMSO. (C) Pathogenicity assays of V. alginolyticus 178 to zebrafish in the absence or presence of CLPs.

Cell aggregation of V. alginolyticus 178 promoted by CLPs.

Next, we sought to investigate the effects of CLPs on V. alginolyticus 178 cells. Surprisingly, the obvious aggregation and sedimentation of V. alginolyticus 178 cells were observed at the bottom of the test tube after treatment with 300 μg/ml CLPs (Fig. 5A). The aggregation activity of CLPs was further verified by crystal violet staining. After treatment with 300 μg/ml CLPs for 24 h, most of the V. alginolyticus 178 cells were aggregated at the bottom of 96-well microtiter plates and were then dyed by the crystal violet (Fig. 5B). There were fewer cells from strain 178 treated with dimethyl sulfoxide (DMSO) observed at the bottom of the well (Fig. 5B). Cell aggregation was quantified through measuring the crystal violet staining solution with optical density at 595 nm (OD₅₉₅). The results showed that the aggregation was dose dependent, and the minimum threshold concentration of CLPs was 25 μg/ml (Fig. 5C).

FIG 5.

V. alginolyticus 178 cell aggregation assay in the presence of CLPs. (A) Cell aggregation of V. alginolyticus 178 observed in culture tubes after treatment with either DMSO or 300 μg/ml CLPs. (B) Cell aggregation of V. alginolyticus 178 observed in microtiter plates stained with crystal violet. The cells were treated with DMSO or 300 μg/ml CLPs. (C) Quantitative cell aggregation assay at different concentrations of CLPs. **, P < 0.01.

Moreover, the morphological features of V. alginolyticus 178 after treatment with 100 μg/ml CLPs were checked by scanning electron microscopy (SEM). For the control, or in the presence of DMSO, cells of strain 178 were distributed evenly with regular morphology (Fig. 6A and B). However, cells of V. alginolyticus 178 were found to gather around each other in 50 μg/ml CLP-treated groups (Fig. 6C). More cell aggregation was detected in the 100 μg/ml CLP-treated group (Fig. 6D), and the cause seemed to be the conglutination of flagella. The motility ability of Vibrio spp. is closely related to the existence of flagella (31); thus, the effects of CLPs on the flagellar biosynthesis of V. alginolyticus 178 were checked by transmission electron microscopy (TEM). The TEM results showed that one polar flagellum and numerous lateral flagella existed around each cell of V. alginolyticus 178 (Fig. 7A and B), which might explain why this bacterium possesses vigorous swarming ability. Consistent with the SEM results, cells of V. alginolyticus 178 were found to gradually aggregate with the increase in CLP concentration (Fig. 7C and D). Notably, flagella of V. alginolyticus 178 could be clearly observed when treated by CLPs (Fig. 7C and D), and there were no significant differences in the number of flagella for cells with or without CLP treatment (Fig. 7C and D). Thus, we propose that CLPs might mainly affect the functions related to the motility of flagella rather than their quantity.

FIG 6.

SEM images of V. alginolyticus 178 cells following treatment with CLPs. V. alginolyticus 178 cell morphology without any treatment (A), with DMSO treatment (B), with 50 μg/ml CLPs (C), and with 100 μg/ml CLP treatment (D). The scale bar applies to all panels.

FIG 7.

TEM images of V. alginolyticus 178 cells following treatment with CLPs. V. alginolyticus 178 cell morphology without any treatment (A), with DMSO treatment (B), with 50 μg/ml CLPs (C), and with 100 μg/ml CLP treatment (D). The scale bar applies to all panels. The white and black arrows indicate the polar flagellum and periflagella, respectively.

Cell aggregation is correlated to biofilm formation (32). Next, we investigated cell aggregation in the bacteria with or without biofilm formation capacity. Of the eight bacterial species selected, four possessed biofilm formation ability, including Vibrio anguillarum, Pseudomonas aeruginosa, Bacillus subtilis, and Bacillus sp. 176, while Vibrio splendidus, Vibrio vulnificus, Staphylococcus aureus, and Pseudomonas stutzeri could not form the biofilm (see Fig. S6 in the supplemental material). Compared with biofilm formation by nontreated, or DMSO-treated, strains, biofilm formation by Vibrio anguillarum, P. aeruginosa, B. subtilis, and Bacillus sp. 176 was suppressed dramatically after treatment with 300 μg/ml CLPs (see Fig. S6). For bacteria that cannot form biofilm, the aggregation and sedimentation of cells were significantly promoted by 300 μg/ml CLPs and had an effect on V. alginolyticus 178 (see Fig. S6).

Effect of CLPs on cell adherence to glass slides of V. alginolyticus 178.

Lipopeptide is the most widely reported class of biosurfactants having antiadhesive action against pathogenic bacteria (19, 20). Therefore, we checked the effect of CLPs on the cell adherence of V. alginolyticus 178. Cells treated with DMSO or CLP (100 μg/ml) were incubated with submerged glass slides for 5, 15, or 30 min in 2216E liquid medium. Thereafter, the cells attached to the glass slides were stained with crystal violet and counted under a light microscope. Compared with the DMSO-treated group, the capacity of cells to adhere to slides was dramatically decreased in the CLP-treated group (see Fig. S7A and B in the supplemental material). The cell counts of strain 178 adherence to the slides in the DMSO-treated group were 4.52-, 3.42-, or 5.37-fold (P < 0.01) more than those in the CLP-treated group at 5, 15, or 30 min, respectively (see Fig. S7C). These results indicated that CLPs from Bacillus sp. 176 could significantly impact the capacity of V. alginolyticus 178 to adhere to the glass slide surface.

The influence of CLPs on the flagellar assembly of V. alginolyticus 178.

Both the motility and adherence of bacteria have been mentioned in previous studies as correlating with the flagellum (2, 12). Since CLPs could inhibit the motility and adherence of V. alginolyticus 178, the influence of CLPs on the flagellar assembly was further tested on the transcription level by quantitative real-time PCR (qRT-PCR). Seven important genes (σ⁵⁴, flhF, fliG, pomA, pomB, flgA, and flgP) responsible for the flagellar assembly of V. alginolyticus 178 were selected for the qRT-PCR assay. Among the proteins encoded by these genes, σ⁵⁴ and FlhF were located in the cytoplasm, FliG was in the cytoplasmic membrane, PomA and PomB formed a complex connecting the cytoplasmic membrane and the cell membrane, FlgA was seated in the peptidoglycan layer of the cell membrane, and FlgP was outside the cell membrane (Fig. 8A). The results showed that relative expression levels of the pomA, pomB, σ⁵⁴, flhF, and fliG genes were all significantly upregulated after treatment with 100 μg/ml CLPs, namely, they were 2.83-, 6.32-, 4.46-, 2.11-, and 3.28-fold, respectively, greater than those of the control (Fig. 8B). It is noteworthy that all of the upregulated proteins were located in the intracellular portion of the flagellar structure (Fig. 8A). However, the mRNA expression of both flgA and flgP was decreased dramatically after CLP treatment and was only 0.23- and 0.46-fold, respectively, less than that of the control (Fig. 8B). Notably, the downregulated proteins were extracellular molecules of flagella from V. alginolyticus 178.

FIG 8.

Effects of CLPs on the expression of flagellar assembly-related genes in V. alginolyticus 178. (A) Proposed overall flagellar structure of V. alginolyticus. (B) Quantitative RT-PCR assays of flagellar gene expression in V. alginolyticus 178. Cells were treated with either DMSO or 100 μg/ml CLPs. All data are relative to the expression levels found in the DMSO control ± the standard error (n = 4). *, P < 0.05; **, P < 0.01.

DISCUSSION

Bacterial swarming motility, a frequently observed behavior, is an impressive form of community motility in which founding colonies begin to concentrically expand at some rate (33). Swarmer cells display increased tolerance for antibiotics and increased virulence in infection models compared with nonswarming cells (33). In this study, the marine bacterium V. alginolyticus 178 showed vigorous swarming capability even on 2% agar plates (Fig. 1A; Fig. S2A). Importantly, this bacterium could severely infect zebrafish and lead the fishes to die in a short time (Fig. 1B and C), which makes it a potential pathogen in marine aquaculture.

To obtain potential compounds against V. alginolyticus 178, we performed screening via targeting bacterial motility. Bacillus sp. 176 was found to significantly inhibit the motility of V. alginolyticus 178 (Fig. 2). Correspondingly, two cyclic lipopeptides (CLPs) (including CLP1 and CLP2) produced by Bacillus sp. 176 were identified with the typical amino acid sequence of pumilacidin (Glu-Leu-Leu-Leu-Asp-Leu-Ile) (Fig. 3C), differing in only the fatty acid chains (C₁₃H₂₇- and C₁₄H₂₉-) (Fig. 3C) with reported pumilacidins (34). Interestingly, the fatty acid moieties of the two lipopeptides were different in only one methylene group (CH₂-) (Fig. 3C). Although numerous reports have demonstrated the mechanism of lipopeptides as biosurfactants, only a few relationships between their chemical structures and functions have been well defined. Notably, both Glu and Asp in CLP1 and CLP2 are acidic amino acids and are further modified by methoxy (CH₃O-), an electron-donating group (35). Thus, some amino acids of CLP1 and CLP2 could be negatively charged at a physiological pH (7.35 to 7.45), which might be beneficial for the interaction with cations in the environment. It is well known that lipopeptides are produced mainly by Bacillus species and have been found to possess various biological activities on biological controls, including the surfactin, iturin, and fengycin families (26). Among these lipopeptides, pumilacidin was a novel lipopeptide similar to surfactin and was first characterized from Bacillus pumilus (26). Although there have been only a few studies of pumilacidin after its discovery, it has been reported to possess different functions, such as antiviral (26), antifungal (34), and antibacterial (36) applications. To date, none of the lipopeptides have been reported to possess antimotility activity.

CLPs from Bacillus sp. 176 were further found to significantly increase the cell aggregation of V. alginolyticus 178 (Fig. 5). As biosurfactants, CLPs are also amphipathic molecules with hydrophilic and hydrophobic moieties (18) that influence the polarity of the bacterial cell membrane (37). Thus, the CLPs sequentially induced cell aggregation. This function of CLPs seemed nonspecific, because similar effects also occurred in other bacteria that did not form a biofilm, including V. splendidus, V. vulnificus, S. aureus, and P. stutzeri (see Fig. S6). However, biofilm-forming bacteria, such as V. anguillarum, P. aeruginosa, B. subtilis, and even Bacillus sp. 176, displayed an effective decrease in biofilm formation upon treatment with CLPs (see Fig. S6). Therefore, it would be of great interest to investigate whether the quorum-sensing system of V. alginolyticus 178 is affected by CLPs.

It is known that polar and lateral flagella are usually important locomotive organelles and virulence factors for bacterial motility and colonization (2, 4, 5). However, based on the TEM results (Fig. 7), the flagellar numbers of V. alginolyticus 178 did not change obviously, which indicates that CLPs might affect the function rather than the quantity of flagella. In Vibrio species, flagellar genes are hierarchically expressed under strict control (5, 38). The FlhF protein determines the polar location and production of flagella (5). The pomA and pomB genes are positively regulated by σ²⁸, expressing the inner membrane sodium ion channel complex PomA/PomB of V. alginolyticus (39). Also, FliG is the essential rotor component of flagella and has been reported to interact with the cytoplasmic region of PomA by electrostatic interactions (40–42). FlgA is a periplasmic chaperone essential for P ring formation (43), and FlgP is an outer membrane lipoprotein functioning in motility to mediate flagellar stability and influence attachment and colonization (31). These two molecules are important for motility of the flagellar complex by keeping the energy transfer and the structural stabilization (1). The expression levels of inner membrane genes were significantly upregulated after CLP treatment (Fig. 8); however, two other genes, flgA and flgP, were downregulated (Fig. 8).

Altogether, this study illustrates how Bacillus sp. 176 produces pumilacidin-like lipopeptides to defeat piracy by V. alginolyticus 178. Going further, this study tells us that some bacteria have evolved clever strategies for competing with their opponents for better survival. Moreover, CLPs from Bacillus sp. 176 could effectively inhibit the biofilm formation of several other pathogenic bacteria but not kill them. Thus, this demonstrates the potential for CLPs from Bacillus sp. 176 to become promising drug candidates for targeting bacterial motility or biofilm formation with a low potential for antibiotic resistance.

MATERIALS AND METHODS

Bacterial strain isolation, identification, and culture conditions.

Sediments and seawater were collected near the Yap Trench during the seamount cruise of the R/V Kexue in the tropical Western Pacific in March 2016 (139°3802′E, 11°44162′N). The marine bacterial strains used in this study were isolated from the above-described samples via the dilution method as described previously (44) and cultured in modified Zobell 2216E broth (5 g/liter tryptone, 1 g/liter yeast extract, 1 liter filtered seawater [pH adjusted to 7.4 to 7.6]) at 28°C. The single colonies were further purified in 2216E plates with 1% agar for several rounds before downstream applications. To phylogenetically classify the bacterial strains, genomic DNA was extracted from different isolates, and PCR was performed to amplify the 16S rRNA gene sequence using the universal primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1541R (5′-AAGGAGGTGATCCACCC-3′). The 16S rRNA gene sequence was then compared with related sequences in public databases using NCBI-BLAST (see http://www.ncbi.nlm.nih.gov/BLAST) and the phylogenetic analysis program MEGA6 (45).

Motility and pathogenicity assays of V. alginolyticus 178.

Isolates of V. alginolyticus 178 were cultured overnight in Zobell 2216E liquid medium at 28°C with shaking at 170 rpm. Thereafter, cells were diluted with fresh 2216E liquid medium, and the OD₆₀₀ was adjusted to 0.2. One microliter of the cell suspension was deposited in the center of 2216E plates containing 0.5%, 1%, 1.5%, or 2% agar and incubated at 28°C. The diameter of the resulting colony was measured at different time points (6 h, 12 h, 24 h, and 36 h after incubation) to evaluate the motility of V. alginolyticus 178 at different agar concentrations. The experiment was repeated three times with three replicates per data point in each experiment. The comparison of motility activity between V. alginolyticus 178 and seven other V. alginolyticus strains (C1, C11, C38, C48, CJ11, CJ26, and CT30) was performed as described above.

The pathogenicity of V. alginolyticus 178 was assayed using the cumulative mortality rates of the zebrafish. Briefly, 90 zebrafishes were randomly divided into three groups (30 fish per group, and 10 fish per tank) with each group distributed among three tanks. Two experimental groups were challenged by abdominal injection of 50 μl of V. alginolyticus 178 cells at concentrations of 6 × 10⁴ and 6 × 10⁵ cells/ml, respectively. In the control group, 50 μl of sterile normal saline was injected. The mortality rates of the zebrafishes after injection were determined by directly counting the number of zebrafish that survived for 7 days postchallenge. The bacteria were reisolated and confirmed as the causative agents of mortality following Koch's postulates as described previously (46). All zebrafish experiments were carried out in accordance with the ethical guidelines of the Chinese Academy of Sciences.

Screening of bacteria inhibiting motility of V. alginolyticus 178.

To screen for bacteria inhibiting the motility of V. alginolyticus 178, cell suspensions of different isolates were incubated overnight and adjusted to 0.2 at OD₆₀₀ as described above. One microliter of V. alginolyticus 178 was then seeded in the center of 2216E plates with 1% agar, and the same volume of another bacterial isolate was seeded around the colony of V. alginolyticus 178. The plates of V. alginolyticus were placed in a 28°C incubator and checked for colony expansion after a 24-h incubation period.

Purification of antimotility compound from Bacillus sp. 176.

To get the active compound inhibiting V. alginolyticus 178 motility, Bacillus sp. 176 was cultured in a 250-ml glass flask filled with 2216E medium overnight at 28°C with shaking at 160 rpm. Then, 5 ml of seed culture was transferred into a 3,000-ml flask containing 500 ml of fermentation medium (20 g/liter peptone, 10 g/liter yeast extract, 5 g/liter glucose [pH adjusted to 7.5 in seawater]). Fermentation was carried out for 2 days at 28°C on a rotary shaker at 160 rpm. Thereafter, the culture was centrifuged at 8,000 × g for 10 min to obtain the supernatant. The supernatant was adjusted to a pH of 2 to 3 using a 6 N HCl solution and kept at 4°C overnight until precipitation was complete. The precipitate was then collected by centrifugation at 8,000 × g at a temperature of 4°C for 10 min, washed with 0.1 N HCl, and frozen under vacuum conditions. Next, the bioactive substance in the solid was extracted with methanol, concentrated under reduced pressure, filtered through a 0.22-μm membrane, and applied to a Sephadex LH-20 column for fractionation using a methanol mobile phase for elution (47). Each eluted fraction was concentrated and assessed for antimotility activity using the paper disc method as described previously (48). Active fractions were then concentrated and dried by evaporation before being redissolved in methanol. Further purification of the bioactive substances was achieved via high-performance liquid chromatography (HPLC) (Agilent 1260 Infinity, USA) on an Eclipse XDB-C18 column (5 μm, 9.4 by 250 mm) (Agilent, USA). The column was eluted with methanol at a flow rate of 2.0 ml/min, and UV detection at 214 nm was used to identify product peaks. Fractions of each eluted peak were tested for antimotility activity as described above. The active fractions were passed through the HPLC column two times to assess stability (27).

Structural elucidation of antimotility compounds derived from Bacillus sp. 176.

The molecular weight of the purified bioactive compound was determined by electrospray ionization mass spectrometry (ESI-MS) with a Bruker maXis mass spectrometer (Bruker, Germany). Data from ESI-MS were acquired in positive ion mode under the following conditions: 3,200-V capillary voltage, 4.0 liters/min dry gas, and 200°C dry gas temperature. Tandem MS (MS/MS) was also performed on this instrument using the multiple reaction monitoring (MRM) mode described by the Anderson and Hunter method (49). The chemical composition of the purified substance was characterized using ¹H and ¹³C nuclear magnetic resonance (NMR). The NMR spectra of the bioactive compounds were recorded on a Bruker Advance III 600 spectrometer (Bruker, Germany) with DMSO-d6 solvent at 600 and 151 MHz. All experiments were performed at 25°C.

Antimotility assay with purified CLPs.

A plate-based assay was used to analyze the antimotility activity of the cyclic lipopeptides purified from Bacillus sp. 176. Briefly, 1 μl of an overnight culture of V. alginolyticus 178 (OD₆₀₀, ∼0.2) was seeded in the center of 2216E plates (containing 1% agar) and incubated at 28°C for 6 h. Next, a sterile paper disk filter containing 20 μl of 10 mg/ml purified CLPs dissolved in DMSO (200 μg total) or DMSO only was placed around the bacterial strain. Then, the inhibitory activity of CLPs against V. alginolyticus 178 was observed on plates after incubating them for an additional 4 h at 28°C.

The growth curve of V. alginolyticus 178 in the presence of CLPs was tested according to previous methods with minor modifications (32). Briefly, a single colony of V. alginolyticus 178 was transferred into 5 ml of Zobell 2216E liquid medium and grown overnight. Then, 0.5 ml of the overnight seed culture was transferred into a flask containing 50 ml fresh 2216E medium in the presence of either purified CLPs at a final concentration of 100 μg/ml or DMSO with same volume as specified above. Different cultures were incubated in a rotary shaker at 28°C with shaking, and the OD₆₀₀ was monitored at different time intervals.

The effect of CLPs on the pathogenicity of V. alginolyticus 178 was further checked with the method described above. Briefly, 120 zebrafishes were divided randomly into four groups. In the challenged groups, 3 × 10⁴ cells of V. alginolyticus 178 treated with 100 μg/ml CLPs or DMSO was injected into the abdomen of zebrafishes. Zebrafishes injected with untreated V. alginolyticus 178 or sterile saline were employed as the positive and negative controls, respectively. The cumulative mortality rate of zebrafishes with different treatments were calculated after different time points.

Aggregation assay of V. alginolyticus 178 in the presence of CLPs.

Cell aggregation assays were performed as described previously (50). Briefly, V. alginolyticus 178 was grown in saline LB broth (10 g/liter peptone, 5 g/liter yeast extract, 1 liter filtered seawater [pH adjusted to 7.4 to 7.6]) at 28°C overnight. The cell suspension of V. alginolyticus 178 was diluted 1:100 with saline LB broth and put into culture tubes (3 ml each) or the wells of a flat-bottom 96-well microtiter plate (200 μl in each well) (Corning, USA). The cells were treated with DMSO or 300 μg/ml CLPs and incubated at 28°C for 24 h. For a quantitative assay, different concentrations (25, 50, 100, 200, or 300 μg/ml) of CLPs were added to the well. Wells containing V. alginolyticus 178 cell suspension without treatment, or with DMSO treatment, were employed as controls. Five replicate wells were used for each treatment, and all plates were incubated at 28°C for 24 h. At the end of incubation, the planktonic cells were discarded with a pipettor, and the aggregated cells in each well were washed three times with sterile saline. The aggregated cells were then fixed with 200 μl of methanol (99% purity) per well for 15 min, and the plates were emptied and left to dry. Then, the wells were stained with 200 μl of a 1% (wt/vol) solution of crystal violet in water for 10 min at room temperature. Excess stain was rinsed out, the plates were air dried, and the dye bound to the aggregated cells was resolubilized with 200 μl of 30% (vol/vol) acetic acid in water. The absorbance of each well was measured in a SYNERGY-H1 microplate reader (BioTek, USA) at 595 nm using 30% acetic acid as the blank.

To check the aggregation and morphology changes after CLP treatment, an overnight cell culture of V. alginolyticus 178 was diluted 1:100 into fresh 2216E medium and cultured for another 3 h to an OD₆₀₀ of 0.2 to 0.3. The cell suspension was then treated with 50 or 100 μg/ml CLPs or DMSO for an additional 3 h. Samples were fixed with a 5% glutaraldehyde solution for downstream scanning electron microscope (SEM) or transmission electron microscope (TEM) imaging.

To check what other bacteria could be aggregated by the CLPs from Bacillus sp. 176, eight species of important marine bacteria from different genera, including V. anguillarum, V. splendidus, V. vulnificus, Pseudomonas aeruginosa, P. stutzeri, Staphylococcus aureus, Bacillus sp. 176, and B. subtilis, were selected to further analyze the spectrum of action of the CLPs. The aggregation assays were performed as described above in the presence of DMSO or 300 μg/ml CLPs.

Adherence assays of V. alginolyticus 178 in the presence of the CLPs.

To compare the ability of V. alginolyticus 178 to adhere to the glass slides, an experiment was performed essentially as described by Lee et al. (32). Briefly, cell suspensions (OD₆₀₀, ∼0.3) of V. alginolyticus 178 incubated with or without 100 μg/ml CLPs were dispersed onto glass slides at a final cell density of 1 × 10⁶ cells/ml. After 5, 15, and 30 min of incubation at 28°C without agitation, the slides were removed, briefly washed, and stained with crystal violet. The number of bacterial cells adhering to one side of the slide was determined under a light microscope. The average number of adherent cells per field was obtained from at least 20 observations.

RNA extraction and quantitative real-time PCR (qRT-PCR) analysis.

Cells were incubated in the presence of DMSO or 100 μg/ml CLPs, as in the aggregation assay. The cells were harvested with sterile saline, and the cell suspensions were centrifuged at 12,000 × g at 4°C for 10 min. Total RNA was extracted with TRIzol reagent (Invitrogen, USA). First-strand cDNA synthesis was carried out with DNase I (Promega, USA)-treated total RNA and random primers (TaKaRa, Japan) based on the Promega M-MLV RT manufacturer's specifications. Synthesis was performed at 42°C for 1 h and then terminated by heating at 95°C for 5 min. The mRNA levels of flagellar assembly-related genes were determined using qRT-PCR on mRNA/cDNA samples from cultures coincubated with either DMSO or 100 μg/ml CLP-treated cells. Specific primers were designed according to the corresponding sequences in the genome of V. alginolyticus (Table 1). The comparative threshold cycle (C_T) (2^−ΔΔCT) method was used to analyze the expression level (51). Two 16S rRNA gene primers for V. alginolyticus, P15 and P16 (Table 1), were used as internal controls to verify successful transcription and to calibrate the cDNA template for corresponding samples. qRT-PCR was performed using an Applied Biosystems 7500 PCR system (Life Technologies, USA), and the collected data were analyzed with the system's accompanying SDS software. The assay was conducted in a volume of 20 μl consisting of 10 μl of 2× SYBR premix Ex Taq II (TaKaRa, Japan), 0.8 μl of each forward and reverse primer (10 μmol/liter), 0.4 μl of 50× ROX reference dye, 2 μl of DNA extract (10 ng/μl), and 6 μl of nuclease-free water. The reaction was performed at 95°C for 30 s and then subjected to 40 cycles of primer annealing at 95°C for 5 s and primer extension at 60°C for 31 s. Dissociation curve analysis of amplification products was performed at the end of each PCR to confirm that only one PCR product was amplified and detected. All data were given in terms of relative mRNA expressed as means ± standard error (n = 4).

TABLE 1.

Primers used for quantitative real-time PCR

Primer name	Sequence (5′–3′)
PomA
P1	CGGATGAACCCGAAGACCT
P2	ATCGCAACAAGACCAACCAA
PomB
P3	ACAATGAATCTGCCGAAGCG
P4	GCAATCTGACGAACTAAAGGACG
Sigma factor 54
P5	TAGGCAACCGAGACTACAAGC
P6	GAGTGATGCGACTACCTGGAC
FlhF
P7	GGTGGCGGACCAAATCAATA
P8	ATCTTTAGCAACTCTTACGGGACA
FliG
P9	GGTTCAGCGTGTTGGTAGTGC
P10	GCGTGGGTCCATCCATTTC
FlgA
P11	ATTGCCCTGTTCCGTTAGAT
P12	ACCGCTTTGATAATGACCTTT
FlgP
P13	CCAGTGGTGCTTTCAATACCC
P14	ATTGCCCTGTTCCGTTAGAT
16S rRNA
P15	TCGTGTYGTGARATGTTGGGT
P16	CCACCTTCCTCCRGTTTRTCA

Statistical analysis.

The significant differences among groups were subjected to one-way analysis of variance (ANOVA) and multiple comparisons by using the SPSS 18.0 program. Statistical significance was defined in our study as a P value of <0.05 or <0.01.

Accession number(s).

The GenBank accession number for the 16S rRNA gene of V. alginolyticus 178 is KY203854. The GenBank accession number for the 16S rRNA gene of Bacillus sp. 176 is KY203855.