ABSTRACT
Biofilm formation is an important problem for many industries. Desulfovibrio vulgaris is the representative sulfate-reducing bacterium (SRB) which causes metal corrosion in oil wells and drilling equipment, and the corrosion is related to its biofilm formation. Biofilms are extremely difficult to remove since the cells are cemented in a polymer matrix. In an effort to eliminate SRB biofilms, we examined the ability of supernatants from Pseudomonas aeruginosa PA14 to disperse SRB biofilms. We found that the P. aeruginosa supernatants dispersed more than 98% of the biofilm. To determine the genetic basis of this SRB biofilm dispersal, we examined a series of P. aeruginosa mutants and found that mutants rhlA, rhlB, rhlI, and rhlR, defective in rhamnolipids production, had significantly reduced levels of SRB biofilm dispersal. Corroborating these results, purified rhamnolipids dispersed SRB biofilms, and rhamnolipids were detected in the P. aeruginosa supernatants. Hence, P. aeruginosa supernatants disperse SRB biofilms via rhamnolipids. In addition, the supernatants of P. aeruginosa dispersed the SRB biofilms more readily than protease in M9 glucose minimum medium and were also effective against biofilms of Escherichia coli and Bacillus subtilis.
INTRODUCTION
Sulfate-reducing bacteria (SRB) are an important type of microorganisms causing iron corrosion on metal surfaces under both anaerobic and aerobic conditions1-3. Desulfovibrio vulgaris Hildenborough is a sequenced4 Gram-negative SRB that has been used as an SRB model organism to study biocorrosion and bioremediation of toxic metal ions4 as well as biofilm formation5, 6 and bioimmobilization in superfund sites7. It is also called the “petroleum pest” because it is commonly found in oil fields and causes “souring” of petroleum and damage to topside equipment and pipelines8.
Biofilms are groups of bacteria that are held together in a self-produced extracellular matrix9 and are difficult to remove with antimicrobial agents due to their antibiotic or biocide resistance relative to planktonic cells10. Desulfovibrio spp. populations in biofilms have a significant role for microbial induced corrosion because of their sulfide production and electron transfer mechanism5, and biofilms of D. vulgaris have been extensively shown to cause corrosion in many types of steels and other alloys11. The biofilms of D. vulgaris consists primarily of protein5, mannose6, fucose6, and N-acetylgalactosamine6. The biofilms of Escherichia coli consist primarily of proteinaceous curli fibres, flagella, and the polysaccharide cellulose12, and the biofilms of Staphylococcus aureus are largely composed of cytoplasmic proteins13 and extracellular genomic DNA14.
Many Gram-negative bacteria use quorum sensing (QS) molecules or autoinducers to communicate with each other15 and to form biofilms15. The QS mechanism can control particular processes related to cell density16, and QS inhibition targeting autoinducers has been used as a method to control biofilm formation16. The opportunistic pathogen Pseudomonas aeruginosa has four QS systems (Las, Rhl, Pqs, and Iqs)17. Each system has its own signal and regulatory protein. For the Las system, LasI synthesizes N-(3-oxododecanoyl)-homoserine lactone (3oxoC12HSL), and LasR is the protein receptor17. For the Rhl system, RhlI synthesizes N-butyrylhomoserine lactone, and RhlR is the protein receptor. The third QS system is quinolone-based intercellular signaling; the PQS signal is 2-heptyl-3-hydroxy-4-quinolone, and it is synthesized by the products of the PQS synthesis cluster consisting of pqsABCD, phnAB, and pqsH. PqsR is the protein receptor17. The fourth system is called Iqs, and the QS molecule is 2-(2-hydroxyphenyl)-thiazole-4-carbaldehyde17.
One aspect of biofilm formation controlled by the Rhl QS system is regulation of the synthesis of rhamnolipids, which are glycolipid biosurfactants composed of rhamnose and 3-(hydroxyalkanoyloxy)alkanoic acid (HAA)18. P. aeruginosa rhamnolipids affect biofilm architecture by participating in the maintenance of biofilm channels19 and reduce adhesion between cells20.
There are many reasons for forming bacterial biofilms such as a defense against stress (e.g., nutrient deprivation, antibiotics, or pH changes)21 and a mechanism for staying stationary in a nutrient-rich area21. However, bacteria also must have a means to leave the biofilm (dispersal) due to environmental changes (e.g., fluctuations in oxygen levels, alterations in nutrients, or increasing of toxic products)22; this process involves breaking the matrix to uncement the cells23.
Here, we show that supernatants from P. aeruginosa PA14 (henceforth PA14) disperse the biofilms of SRB, E. coli, and S. aureus. Using QS-related mutants, we determined that the biochemical basis for this dispersal is the presence of rhamnolipids in the supernatants.
RESULTS
PA14 wild-type supernatant disperses SRB biofilm
In an effort to investigate whether there are QS compounds utilized by the representative SRB D. vulgaris, we tested whether its supernatants would disperse its own biofilm formed in rich medium. D. vulgaris supernatant did not disperse its own biofilm within 2 h.
Since there was no negative effect of SRB supernatants on its own biofilm, we investigated the effect of the supernatant of other strains (e.g., P. aeruginosa, P. fluorescens, Bacillus subtilis, E. coli). Among these strains, P. aeruginosa PA14 and P. aeruginosa PAO1 dispersed SRB biofilm the most (Fig. 1). The supernatants were obtained from planktonic cultures, and the SRB biofilm grown was in 96-well plates for 24 to 48 h in modified Baar’s medium. Critically, the PA14 wild-type supernatant dispersed D. vulgaris biofilm more than 92% after 1 to 2 h of incubation. We utilized short periods of contact of the supernatants with the SRB biofilm to avoid artifacts related with growth of the bacterium.
Rhamnolipids in the supernatants disperse SRB biofilms
To determine mechanisms behind these strong dispersal results, we hypothesized that the compounds in the P. aeruginosa supernatant may be related to QS since PA14 is the best known strain for QS, and QS controls extracellular compounds like protease24 and rhamnolipids24. Hence, we tested the supernatant of mutants related to QS as well as those related to virulence factors (e.g., lasI25, lasB26, pelA27, phzM28, phzS28, pvdF29, pchE30, lecA31, toxA32, sodM33, rhlI34, rhlR34). Similar to the supernatant of the PA14 wild-type, the supernatants of the mutants lasI (autoinducer), lasB (elastase), pelA (oligogalacturonide lyase), phzM (pyocyanin), phzS (pyocyanin), pvdF (pyoverdin), pchE (pyochelin), lecA (LecA lectin), toxA (exotoxinA), and sodM (superoxide dismutase) dispersed the D. vulgaris biofilm like the supernatant of PA14 wild-type (Fig. 2), indicating the Las QS system, Pel polysaccharide, pyocyanin, pyoverdine, pyocheline, LecA lectin, exotoxinA, superoxide dismutase had no role in the biofilm dispersal.
In contrast to the other mutants, mutations in the rhamnolipids pathway, specifically in the gene encoding the autoinducer, rhlI, and the gene encoding the response regulator, rhlR, had a pronounced decrease in dispersal (Fig. 2). Since the rhlI and rhlR mutants were found to affect biofilm dispersal, additional mutations in the rhamnolipid pathway were investigated: rhlA, which encodes rhamnosyltransferse 1 subunit A and rhlB, which encodes the catalytic subunit of the rhamnosyltransferase34,35. The supernatants of both the rhlA and rhlB mutant did not disperse SRB biofilms (Fig. 2). Therefore, the dispersal compounds are related to rhamnolipids.
To corroborate the 96-well biofilm dispersal results with the PA14 wild-type and rhlB mutant, confocal microscopy was used to visualize the remaining biofilm after treatment with supernatants from these two strains. As shown in Fig. 3, the supernatant from the wild-strain nearly completely dispersed the SRB biofilm whereas the supernatant from the rhlB mutant had no effect, just like the buffer negative control.
From the rhamnolipid production pathway (Fig. 4), RhlA synthesizes 3-(3-hydroxyakanoyloxy)alkanoic acids (HAA)36,and RhlB uses HAA to make mono-rhamnolipids37. Therefore, the compounds required for SRB biofilm dispersal are HAA or mono/di-rhamnolipids, respectively. However, the results from mass spectrometry showed that there was no HAA detected in the wild-type supernatant sample. Therefore we conclude that rhamnolipids are the compounds (and not HAA) that are important for biofilm dispersal.
Rhamnolipids are present in the PA14 supernatants and disperse SRB biofilms
The genetic results clearly indicate the importance of rhamnolipids for SRB biofilm dispersal via PA14 supernatants. To corroborate these results, the supernatants of PA14, rhlA, rhlB, rhlI, and rhlR mutants were tested for the presence of rhamnolipids by mass spectrometry (Fig. 6). As expected, rhamnolipids were detected only in the PA14 supernatant samples and not from any of the mutant samples. There were four predominant peaks of rhamnolipids shown (Fig. 6) which are Rha-C10-C10, Rha-C10-C12/Rha-C12-C10, Rha-Rha-C10-C10, and Rha-Rha-C10-C12/ Rha-Rha-C12-C10. By comparing with the 10 mM rhamnolipids standard from P. aeruginosa, we determined the concentration of rhamnolipids in the PA14 supernatant using the total peak areas is over 100 fold less than the standard (0.1 mM, Table 1).
To show conclusively that rhamnolipids in the supernatants are the biochemical means by which the D. vulgaris biofilms are dispersed, purified P. aeruginosa rhamnolipids (10 mM) were tested. We found that the purified rhamnolipids disperse D. vulgaris biofilm with approximately 67% dispersal after 2 h (Fig. 2). The rhamnolipids standard did not disperse the biofilm as well as the PA14 supernatant though there were more rhamnolipids in the 10 mM standard sample than the amount of rhamnolipids in the supernatant (Table 1). This implies that the ratio of the rhamnolipids or other compounds in the supernatant are important for SRB biofilm dispersal.
Addition of the supernatant of the rhlB mutant to the rhamnolipids standard did not increase the biofilm dispersal compared to the the rhamnolipids standard alone. This suggests the dispersal is due to the rhamnolipids alone and not some other product in the supernatants.
PA14 supernatants disperse D. desulfuricans, E. coli MG1655, and Staphylococcus aureus biofilms
To test the ability of the PA14 supernatants to disperse other bacteria, biofilms of D. desulfuricans, E. coli MG1655, S. aureus, and PA14 were grown in 96-well plates, then treated with supernatants from PA14. The PA14 supernatant dispersed the biofilms of D. desulfuricans, E. coli MG1655, and S. aureus well (over 90%). Although, there are some strain specific differences, these results show that the PA14 supernatants have a great deal of utility for dispersing a wide range of biofilms.
As expected, the supernatant from the rhlR mutant did not disperse biofilms well (Fig. 5), providing further evidence of the importance of rhamnolipids. In addition, the supernatant of PA14 did not disperse its own biofilm (data not shown).
PA14 supernatant disperses SRB biofilm grown in M9G
Since biofilm matrices can change when the growth medium is altered38, we tested the ability of the PA14 supernatant to disperse SRB biofilms grown in minimal glucose medium (M9G). The PA14 supernatant dispersed 90-100% of the SRB biofilm when formed in modified’s Baar’s medium and dispersed about 80% of the biofilm formed in M9G. In contrast, the positive control protease, which dispersed 80% of the SRB biofilm formed in modified’s Baar’s medium, dispersed only about 40% of the biofilm grown in M9G.
DISCUSSION
We have demonstrated that the supernatants of PA14 disperse D. vulgaris biofilms effectively and that the biochemical basis of this dispersal is the presence rhamnolipids. The PA14 supernatants dispersed the biofilm better than protease, and they are versatile because they also dispersed D. desulfuricans, E. coli MG1655, and S. aureus biofilm. This is significant since the matrix of D. desulfuricans, E. coli, and S. aureus are primarily proteins12,13, carbohydrates39, and some extracellular genomic DNA14, respectively, which indicates that rhamnolipids are a powerful and general approach for removing biofilms.
The rhamnolipid standard did not disperse the biofilm as well as the rhamnolipids in the PA14 WT supernatant, possibly because the ratios of the rhamnolipids are critical for biofilm dispersal activity. For example, the ratio between rhamnolipids and mono rhamnolipids affects the mixture properties (e.g., foam thickness, surface electric parameters)40 and changes the emulsification index and antimicrobial properties41. In addition, the rhamnolipid standard was not pure (90%), and the growth condition used to produce the rhamnolipids is unknown. We also surmise that purifying the rhamnolipids alters their biofilm dispersal properties. Furthermore, previously it was reported that rhamnolipids from P. aeruginosa together with the QS signal 3oxoC12HSL work cooperatively to disperse E. coli biofilms42; however, we found that there was no need for 3oxoC12HSL for supernatants to disperse SRB since the supernatants from the lasI mutant were just as effective as those from the wild-type strain (Fig. 2).
METHODS
Bacterial strains and culture conditions
Desulfovibrio vulgaris and Desulfovibrio desulfuricans were grown anaerobically in modified Baar’s medium at 30°C. PA14 wild-type and its mutants were grown in lysogeny broth (LB)43 or minimal medium with 0.4% glucose (M9G)44 at 37°C. Gentamycin (Gm) 15 μg/mL was used for culturing the PA14 mutants. Escherichia coli TG1, Escherichia coli MG1655, and Pseudomonas fluorescens were grown in LB at 37°C. All the strains are shown in Table 2. The PA14 rhlA, rhlB, rhlI, rhlR, and lasI mutants were verified via PCR using the primers in Table 3.
PA14 supernatant preparation
P. aeruginosa PA14 wild-type and its isogenic mutants were grown in LB or LB with Gm15 µg/mL for the mutants. The overnight cultures (200 µL) were centrifuged, the cell pellets were resuspended in 50 mL of M9G, and cultured for 3 to 4 days. The supernatants were collected and filtered using a 0.22 µm filtration system. The samples were concentrated from 2x to 4x using a speed vacuum.
Biofilm dispersal assay using crystal violet
Biofilm formation was assayed in 96-well polystyrene plates using 0.1% crystal violet staining as described previously45 with some modifications. Diluted overnight cultures (150 µL) at an initial turbidity at 600 nm of 0.05 were inoculated into 96-well-plates with modified Baar’s medium, and the bacteria were cultured for 24 to 48 hours at 30°C without shaking. The planktonic cells were removed from each well, and the biofilm was washed gently using phosphate buffered saline (PBS). The supernatant or chemicals were added to the wells, and the plate was incubated at 30°C for 1 to 2 h. Then, the supernatants were removed from each well, and the remaining biofilm was washed gently using water. After the crystal violet was added to each well, the wells were rinsed and dried, and ethanol was added to dissolve the crystal violet. The total biofilm formation samples were measured at 540 nm. At least three independent cultures were used for each strain.
Confocal Microscopy
Diluted overnight cultures (300 µL) at an initial turbidity at 600 nm of 0.1 were inoculated into an 8 chamber cell culture slide (Dot Scientific, Inc, Burton, MI) with modified Baar’s medium, and the bacteria were cultured for 48 hours at 30°C without shaking. The planktonic cells were removed from each well, and the biofilm was washed gently using PBS. Supernatants or PBS was added to the chambers, and the slide was incubated at 30°C for 20 minutes. Confocal microscopy images were taken using a PlanApo60×/1.4 oil objective lens with Olympus Fluoview 1000 confocal laser microscope. The biofilm structure images were performed using IMARIS software (Bitplane, Zurich, Switzerland), and average biomass and thickness were obtained using COMSTAT image-processing software46. At least three different areas were observed, and two independent cultures were tested.
Mass spectrometric analysis of supernatants
Mass spectrometric analysis was performed on the supernatants by using a Waters Q-TOF Premier quadrupole/time-of-flight (TOF) mass spectrometer (Waters Corporation (Micromass Ltd.), Manchester, UK). Operation of the mass spectrometer was performed using MassLynx™ software Version 4.1 (http://www.waters.com). Samples were introduced into the mass spectrometer using a Waters 2695 high performance liquid chromatograph. The samples were analyzed using flow injection analysis (FIA), in which the sample is injected into the mobile phase flow and passes directly into the mass spectrometer, where the analytes are ionized and detected. The mobile phase used was 90% acetonitrile (LC-MS grade), and 10% aqueous 10 mM ammonium acetate. The flow rate was 0.15 mL/min. The nitrogen drying gas temperature was set to 300 °C at a flow of 7 L/min. The capillary voltage was 2.8 kV. The mass spectrometer was set to scan from 100-1000 m/z in negative ion mode, using electrospray ionization (ESI). Rhamnolipids standard (90%) was purchased from Sigma-Aldrich (St. Louis, MO).
DATA AVAILABILITY
The authors declare that data supporting the findings of this study are available within the paper.
AUTHOR CONTRIBUTIONS
Experiments were designed by TLW, LZ, DSM, BY, and TKW. The experiments were performed by TLW except for mass spectrometry which was performed by JM. The manuscript was written by TLW and TKW.
COMPETING INTERESTS
The authors declare there are no competing interests.
ACKNOWLEDGMENTS
This work was supported by the Dow Chemical Company and funds derived from the Biotechnology Endowed Professorship at the Pennsylvania State University.