Bacterial motility was first described by Leeuwenhoek (1, 2), and many types of bacterial movement have since been described and characterised in detail. In liquid media, bacteria move constantly by either active use of their flagella (including axial filaments) or passive use of Brownian motion in a fluid (3, 4). Even in a structure attached to a solid surface – known as a biofilm – their cells are not as immobile as one might think.
Several types of bacterial movement across solid surfaces have been characterised in detail. These mostly fall into three categories:
Here, we describe an additional type of bacterial motility, in which bacterial cells migrate from biofilm up across non-nutritive solid surface carried by capillary forces. As soon as we noticed this phenomenon and browsed available literature, we found that this type of motility had not been described before.
The aim of this study was to prove the concept, that is, that bacteria can also move on solid non-nutritive surfaces through capillary action.
To that end and to further investigate the phenomenon, we ran experiments on glass slides partly immersed in liquid and partly exposed to air (Figure 1), on which we grew an array of bacterial species that differ significantly in physiology, morphology, and motility. One group uses the flagella to move, namely
Experimental setup for growing bacterial biofilms on an air/liquid interface
Pure cultures were kept in a MicrobankTM system (Pro-Lab Diagnostics, Richmond Hill, Canada) and re-cultivated on Luria-Bertani (LB) agar (tryptone 10 g, yeast extract 5 g, NaCl 5 g, Agar 20 g, deionised water 1000 mL, pH=7±0.2) at 37 °C for 16 h before the experiments started.
Followed suspension of two 10 μL loops of biomass in 10 mL of sterile saline (0.3% NaCl) in 15 mL Falcon tubes. To obtain homogenous suspension, the tubes were vortexed at 45 Hz for 1 min (Kartell Labware, Milano, Italy). Suspension concentrations were ~108 colony forming units per one mL (CFU/mL) of suspension. The CFUs were counted on plates, as follows: 1 mL of suspension was serially diluted in 9 mL of sterile 0.3% NaCl, 0.1 mL dilutions were then inoculated on LB agar, spread with a sterile L-shaped cell spreader, incubated at 37 °C for 24 h, and counted.
Biofilms were grown on glass microscope slides (75×25 mm, VWR International, Leuven, Belgium) as follows: 1 mL of bacterial suspension (~108 CFU/mL) was inoculated in 10 mL of LB medium in 50 mL Falcon tubes. Sterilised slides (autoclaved at 121 °C for 20 min) were then vertically inserted in the tubes so that they were partly immersed in the inoculated medium, while the rest was exposed to air (Figure 1). The tubes were lightly capped to let in air and incubated at 37 °C for seven days. Two incubation settings were applied: i) tubes were gently shaken at 50 rpm on an orbital shaker in a vertical position, or ii) tubes were not shaken to exclude the possibility of bacterial cell migration via aerosols.
As soon as we obtained initial results, we ran a new set of experiments described above with
Experimental combinations performed in order to explain the migration of bacterial cells up the glass slide
Biofilm growth (days) | |||||||
---|---|---|---|---|---|---|---|
103 CFU/mL | 107 CFU/mL | 103 CFU/mL | 107 CFU/mL | Carbol-fuchsin | |||
Exp 1 | Exp 2 | Exp 17 | Exp 18 | Exp 33 | |||
Exp 3 | Exp 4 | Exp 19 | Exp 20 | Exp 34 | |||
Exp 5 | Exp 6 | Exp 21 | Exp 22 | Exp 35 | |||
Exp 7 | Exp 8 | Exp 23 | Exp 24 | Exp 36 | |||
Exp 9 | Exp 10 | Exp 25 | Exp 26 | Exp 37 | |||
Exp 11 | Exp 12 | Exp 27 | Exp 28 | Exp 38 | |||
Exp 13 | Exp 14 | Exp 29 | Exp 30 | Exp 39 | |||
Exp 15 | Exp 16 | Exp 31 | Exp 32 | Exp 40 |
Additionally, we immersed glass slides in LB medium enriched with 1 mL of carbol-fuchsin solution (Biognost, Croatia) to assess capillary forces by monitoring if and to what extent the nutrient medium would move up the glass slide over the same period of incubation (biofilm growth). To do that, we prepared a different slide for every time point.
After incubation at specified time points, the slides were removed from the tubes and gently washed with sterile saline to remove unattached or loosely attached cells. The bottom of each slide (resting on the microscope stage) was wiped off with a paper cloth soaked in 70 % ethanol, and the top side was prepared for microscopy. The slides were heat-fixed, treated with Alcian blue solution [1 g of Alcian blue (Fluka Analytical, Munich, Germany) dissolved in 10 mL of ethanol and 90 mL of deionised water] for 2 min to visualise biofilm extracellular polymeric substances (EPS), washed with tap water, and then treated with carbol-fuchsin solution for 1 min to visualise bacterial cells, washed with tap water, and viewed with an Olympus CX21 (Tokyo, Japan) light microscope under 1000× magnification. Images were taken with a 5-megapixel mobile phone camera (Samsung Galaxy J1, Seoul, South Korea). Biofilm growth was monitored on the liquid section (that was immersed in the nutrient medium), at the liquid/air “interface” (border), and on the air-exposed (non-immersed) section. Each experiment was done in triplicate. Since the aim of this “proof-of-concept” study was to prove a new type of bacterial motility, the results are described qualitatively, without quantifying biofilm biomass.
It was challenging to obtain good images of
Swarming and twitching surface motility was assessed to check for other types of motility in our bacterial species. We followed the method described by Antunes et al. (8) using Luria-Bertani medium containing 0.5 % agarose. An overnight bacterial culture was suspended in 1mL PBS. With a pipette tip, 10 μL of the bacterial suspension was inoculated to the bottom of the polystyrene Petri dish, tightly wrapped in parafilm to prevent the loss of moisture, and incubated in a humid atmosphere at 37 °C/24 h. Swarming motility was determined at the air-agarose layer while twitching motility was determined after the removal of the agarose layer and staining the Petri dish with 0.5% crystal violet for 10 min. The longest diameter of the motility was measured. Isolates were grouped into categories based on the average values of motility: < 25 mm poor; 25–50 mm intermediate; > 50 mm highly motile isolates.
Our experiments showed that biofilm formation of all tested bacteria was the strongest at the air/liquid interface and characterised by massive biomass (Figures 2 and 3). All the bacteria also migrated from the biofilm at the air/liquid interface to the clean glass surface. Shaking did not contribute to bacterial presence in the air-exposed zone, as both shaking and non-shaking tests produced practically the same results.
Biofilms grown on glass microscopy slides at the air/liquid interface after 7 days of incubation. Far right: macroscopic view of biofilm formed at air/liquid interface. NS – experiments without shaking; S – experiments with shaking. Scale bar=50 μm
Biofilm of
Cells of all the tested bacteria were observed all over the air-exposed section, reaching distances of ~35 mm above the interface in experiments with
Migration of
Migration of
Migration of
Microcolonies of
The air-exposed section of
The upward movement of cells was the clearest in experiments with
Typical microcolony of
Migration of
Figures 10 and 11 summarise in drawing the results of tests with carbol-fuchsin, designed to reveal bacterial movement with the capillary movement of nutrient medium. The dots, smudges, and spots of coloured nutrient reached as far as 30–35 mm above the air/liquid interface after ten days of incubation. With
Migration of carbol-fuchsin (left) and
Migration of carbol-fuchsin (left) and
Even when we replaced the nutrient medium with saline,
Routine swarming and twitching assays (Table 2) confirmed swarming of
Swarming and twitching motility determined by standard assays for
82±14 mm | N/A | N/A | 16±7 mm | |
N/A | 35±12 mm | 45±14 mm | 51±9 mm |
N/A – not applicable
Our results demonstrate that various types of bacteria are capable of moving across an inert, non-nutritive solid surface. However, it is important to point out that the air-exposed section of the glass slide surface was moist from nutrient medium or saline evaporation, and surface moistness is one of the requirements for bacterial movement. We did not measure the exact amount of condensed water on the air-exposed surface, but it was clearly visible.
At this point, it is intriguing to see what motility mechanism(s) enabled the bacteria to climb vertically up the glass slide.
In that respect, the standard “twitching assay” is more similar to our conditions, as it monitors colony spreading over inert surfaces like a Petri dish (19), glass slides (13), or even cellophane (20). However, in all those experiments, inert surfaces were covered by a layer of nutrient-rich agar, and our air-exposed glass slide sections were not, yet the tested strains did twitch across them. This type of movement was the most prominent with
Angelini et al. (11) reported the type of movement most similar to the ones presented in this study. In their experiment,
The bacterial migration up the glass slide was most likely initiated by capillary forces, since the bacteria migrated even when immersed in saline, in which no multiplication is expected. This migration was probably facilitated by the production of EPS. Similar observations were reported by Be’er et al. (25) on agar plates and Hennes et al. (26), who argued that bacteria inside a droplet overcame the pinning capillary forces of a water drop on an agar surface, and collectively “surfed” across agar in speeds well above that of mass swarming. They proposed that surfactin lowered surface tension and created inward osmotic flow. In our experiments this may indicate that biofilm formation (and the production of EPS) above the air/liquid interface created a suction force for nutrient media, enabling further cell division, more suction, and step by step upward migration of bacterial cells.
Another question raised here is, do the bacteria just migrate upward from biofilm at the interface, or do they also actively multiply along the way? The presence of microcolonies on the air-exposed section seems to evidence active multiplication (proliferation). This, in turn, suggests that nutrients are transferred from the medium all the way up the glass slide, possibly facilitated by the production of EPS, as argued above. Another explanation could be the “bust and boom” survival strategy, where weak bacterial cells die in unfavourable conditions, and the remaining cells live on their expense (27).
Even though the tested bacteria differ in morphology, physiology, and dominant types of motility, all of them migrated vertically from biofilm at the air/liquid interface over non-nutritive surface. It is important to note that our experimental conditions differ significantly from standard assays, which enabled us to describe a novel type of motility that seems to be a common property of various bacterial species. We propose to call it “capillary movement of biofilm” to describe this phenomenon. It is further facilitated by biofilm production of EPS.
Further experiments that would involve quantification of cell migration (with imaging software), comparison between motile and non-motile mutants, mixed microbial communities, and determination of physicochemical properties such as capillary forces, transport of nutrients, and the importance of water vapour in the gas phase, could shed more light on this newly described phenomenon.