Spatial distributions of Pseudomonas fluorescens colony variants in mixed-culture biofilms
© Workentine et al.; licensee BioMed Central Ltd. 2013
Received: 15 April 2013
Accepted: 24 July 2013
Published: 28 July 2013
Skip to main content
© Workentine et al.; licensee BioMed Central Ltd. 2013
Received: 15 April 2013
Accepted: 24 July 2013
Published: 28 July 2013
The emergence of colony morphology variants in structured environments is being recognized as important to both niche specialization and stress tolerance. Pseudomonas fluorescens demonstrates diversity in both its natural environment, the rhizosphere, and in laboratory grown biofilms. Sub-populations of these variants within a biofilm have been suggested as important contributors to antimicrobial stress tolerance given their altered susceptibility to various agents. As such it is of interest to determine how these variants might be distributed in the biofilm environment.
Here we present an analysis of the spatial distribution of Pseudomonas fluorescens colony morphology variants in mixed-culture biofilms with the wildtype phenotype. These findings reveal that two variant colony morphotypes demonstrate a significant growth advantage over the wildtype morphotype in the biofilm environment. The two variant morphotypes out-grew the wildtype across the entire biofilm and this occurred within 24 h and was maintained through to 96 h. This competitive advantage was not observed in homogeneous broth culture.
The significant advantage that the variants demonstrate in biofilm colonization over the wildtype denotes the importance of this phenotype in structured environments.
When grown in spatially structured environments several Pseudomonas species are known to produce variants with altered phenotypic properties. Such variants are often isolated from laboratory biofilms [1–5], cystic fibrosis airways [6, 7], and the plant rhizosphere . Two variant types have been characterized in some detail; the wrinkly spreader (WS, also called rugose small colony variants) and the small colony variant (SCV), of which the primary phenotypic characteristic is the overproduction of exopolyscharides [1, 2, 6, 9].
Given that these variants arise in structurally heterogeneous environments, presumably still populated with the ancestral strain, one could expect the variants to have an advantage in specific niches within these environments. Indeed, the WS morphotype isolated from static microcosms has a competitive advantage at the air-liquid interface where it can form self-supporting mats generated by the cellulose-like polymer that it overproduces [1, 10–12]. However, besides competition studies with this morphotype very little work has been done to examine spatial interaction between colony variants and the ancestral phenotype, within the environment where the variant evolved. To the best of our knowledge only one other study has specifically examined the spatial distributions of variant and wildtype populations in a biofilm on a microscopic level. This was done with a laboratory derived P. aeruginosa colony variant and the authors concluded that the variant only had a selective advantage in certain niches within the biofilm .
We have previously isolated SCV and WS variants from biofilms of P. fluorescens. To examine spatial interactions between colony variants and the wildtype ancestral strains, strains were labeled with 4 different coloured auto-fluorescent proteins (AFPs). In order to determine if these variants had any spatial preference or advantage in the environment where they evolved we examined co-culture biofilms and planktonic populations of SCV and WS with the ancestral strains.
The emergence of phenotypic diversity in biofilms or other structurally heterogeneous environments is generally associated with selection for that phenotype in that particular environment. Such is the case for the previously studied WS from P. fluorescens SBW25, which has adaptations that allow it to out-compete wildtype genotypes from the air-liquid interface of the static microcosm where it evolved . Previously we isolated an SCV and WS variant from a ΔgacS strain of P. fluorescens biofilms and here we sought to determine if these variants might have an advantage in the biofilm environment. The hypothesis was that the variants would have a distinct advantage over the wildtype, when colonizing a surface, due to the fact that they evolved in the biofilm. In addition, the fact that the WS is over-producing a cellulose-like polymer  suggests it might be better at colonizing a surface.
To test this hypothesis, different coloured auto-fluorescent proteins (AFPs) were introduced into the four different strains of P. fluorescens; CHA0 (wildtype), CHA19 (ΔgacS), SCV, and WS. The two variants were isolated from the ΔgacS strain, which produces a higher frequency of colony morphology variants  and so both the wildtype and ΔgacS strains were included. Green fluorescent protein (GFP), yellow fluorescent protein (YFP), cyan fluorescent protein (CFP), and dsRed (referred to from here on in as red fluorescent protein, RFP) were introduced on a plasmid that is stable in P. fluorescens without antibiotic selection . Biofilms of the individual strains or mixed co-cultures were grown and imaged using confocal laser scanning microscopy (CLSM). Imaging the individual strains with each of the 4 colours of AFP revealed that expressing the different fluorescent proteins did not significantly alter the biofilm structure when compared to the biofilms stained with acridine orange . Although some variation in biofilm structure was observed between replicates, this was independent of which AFP was being expressed, indicating that no one particular AFP was affecting biofilm formation or structure.
Niche specialization is an important aspect of colony morphotypes and this is certainly the case for the variants described in this study. Here we have shown that the SCV and WS colony variants out-grow the ancestral populations in the environment from which they were isolated, that is, the peg surface in the CBD. Microscopic evaluation of spatial distributions of variant and ancestral strains in biofilms is virtually non-existent, hence, these findings represent the first detailed microscopic examination of multiple variant types within a biofilm. One previous study examined a variant and wildtype co-culture of P. aeruginosa in a tube biofilm . Here they observed that although the variant seemed to dominate initially, upon prolonged growth the wildtype eventually took over and the variant never made up more than 40% of the biofilm. The conclusion was the variant was only able to grow within certain microniches in the tube biofilm. Given the microscale heterogeneity assumed to be present in the biofilm environment  such microniche specialization could certainly be expected. However, the work here suggests that, at least for P. fluorescens, the two morphotypes are macroniche specialists, that is, they have adaptations that allow them to better colonize the entire surface, rather than small niches within the biofilm. The extensive work done with the WS morphotype from P. fluorescens SBW25 supports this concept in that this morphotype is adapted to colonize the air-liquid interface of static microcosms, a niche that cannot be colonized by the wildtype phenotype .
It is interesting to note that in the present study, the wildtype can colonize the peg surface efficiently suggesting that the emergence of diversity is not solely associated with ecological opportunity but may have other function such as resistance to stress, as is suggested by the enhanced metal tolerance these variants have over the ancestral Δ gacS strain . In addition to having properties suggestive of adaptation to surface growth variants of P. aeruginosa isolated from the lungs of infected cystic fibrosis patients also have markedly increased antibiotic resistance . This has lead to the general conclusion that these variants have more than just surface-attachment adaptations but may actually have a host of adaptations specific to the environment from which they were isolated .
In summary, we have presented a microscopic examination of variant-wildtype distributions in biofilms, which has revealed that the variants rapidly out-grow the wildtype and dominate the biofilm environment. Furthermore, we demonstrate that this is phenomenon is specific to surface associated growth and is not observed in planktonic culture.
Strains and plasmids used in this study
Strain or plasmid
P. fluorescens CHA0
P. fluorescens CHA19
Contains a marker-less deletion of the gacS coding region
P. fluorescens SCV
Small Colony Variant derived from the CHA19 strain
P. fluorescens WS
Wrinkly Spreader derived from the CHA19 strain
Rhizosphere stable plasmid, does not require antibiotic selection in P. fluorescens
pME6010 containing the coding sequence of enhanced GFP with the lac promoter
pME6010 containing the coding sequence of enhanced CFP with the lac promoter
pME6010 containing the coding sequence of enhanced YFP with the lac promoter
pME6010 containing the coding sequence of dsRed with the lac promoter
Microscopy was performed according the protocols outlined previously . The pegs were examined using a Leica DM IRE2 spectral confocal and multiphoton microscope with a Leica TCS SP2 acoustic optical beam splitter (AOBS) (Leica Microsystems). A 63 × water immersion objective used for all the imaging and the image capture was performed using Leica Confocal Software Lite (LCS Lite, Leica Microsystems). Imaging of the biofilms expressing the AFPs were obtained by breaking off a peg of the CBD and placing it on a coverslip with a drop of saline. Excitation/emission parameters for each of the AFPs were 488/500−600 for GFP, 514/525−600 for YFP, 458/465−600 for CFP, and 543/55−700 for dsRed. To reduce cross-talk between the different AFPs, images with more than one AFP were acquired sequentially by frame so only one AFP was being imaged at a time. Furthermore any AFPs that were imaged together were checked to ensure minimal cross-talk was occurring. Laser intensity and photomultiplier tube gain were kept consistent across all experiments. Image stacks were processed using Imaris 6.3.1 (Bitplane) to generate images for publication. Biovolumes for each image stack were computed using the ‘Surfaces’ feature of the Imaris software with the ‘Absolute Intensity’ setting for background removal. For each co-culutre, 4 replicates comprised of different strain-AFP combinations (to remove any fluorescent intensity bias in the quantification) were used to calculate the mean biovolume. The relative proportion of each strain was calculated compared to the total biovolume. Student’s t-test was used to compare the means of the relative volumes for each strain pair.
To determine if the WS or SCV had any growth advantage in broth culture competitions were performed with each pair combination. Equal volumes of 16 h cultures of each strain were add to a total of 150 μL LB media in 96 well plates (30-fold dilution). The plate was incubated at 30℃ with shaking (175 rpm) for 24 h. Prior to incubation samples were removed for determination of initial cell numbers. The cultures were serially diluted on LB agar and the number of each colony type were recorded. The SCV and WS could easily be distinguished from the wildtype CHA0 and CHA19 colony types. To control for any phenotypic variation occurring the broth culture the competitions were performed with the strains expressing the fluorescent proteins. Representative plates from each pair combination were imaged with a fluorescent imager (IVIS Imaging System, Caliper LifeSciences) to distinguish the two strains and the numbers were compared to the values obtained when counting based on colony morphology. No phenotypic variation occurred in broth cultures during the time period tested. Fluorescent imaging of the plates was also used to distinguish the CHA0 and CHA19 colonies as well as CHA0 and CHA19 competed with themselves. The relative fitness  of the variant (SCV or WS) compared to wildtype (CHA0 or CHA19) was calculated for each pairwise combination. A relative fitness of 1 indicates that neither strain has a competitive advantage, whereas values higher than 1 indicate that the variant is more fit in the broth culture. A one-tailed Student’s t-test was used to determine if the values were significantly greater than 1. P values were adjusted with the Holm-Bonfferoni correction to control for the family-wise error rate .
This work was supported through discovery grants from the Natural Sciences and Engineering Research Council (NSERC) of Canada to RJT and HC. NSERC has also provided a Postgraduate Scholarship (Doctoral) to MLW who was additionally supported by a PhD Studentship from the Alberta Heritage Foundation for Medical Research (AHFMR). CLSM was made possible through a Canadian Foundation for Innovation (CFI) Bone and Joint Disease Network grant to HC. We would also like to thank Mike Surette for providing materials and laboratory space.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.