• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Feb 23, 2010; 107(8): 3776–3781.
Published online Feb 2, 2010. doi:  10.1073/pnas.0910934107
PMCID: PMC2840483

Cell density and mobility protect swarming bacteria against antibiotics


Swarming bacteria move in multicellular groups and exhibit adaptive resistance to multiple antibiotics. Analysis of this phenomenon has revealed the protective power of high cell densities to withstand exposure to otherwise lethal antibiotic concentrations. We find that high densities promote bacterial survival, even in a nonswarming state, but that the ability to move, as well as the speed of movement, confers an added advantage, making swarming an effective strategy for prevailing against antimicrobials. We find no evidence of induced resistance pathways or quorum-sensing mechanisms controlling this group resistance, which occurs at a cost to cells directly exposed to the antibiotic. This work has relevance to the adaptive antibiotic resistance of bacterial biofilms.

Keywords: swarming motility, antibiotic resistance, group trait, surfactants, biofilms

Swarming is defined as flagella-driven bacterial group motility over a surface, which is observed in the laboratory on media solidified with agar (1 4). The percentage of agar is critical for enabling swarming. Some bacteria like Vibrio parahaemolyticus and Proteus mirabilis can swarm readily on higher percentage agar (1.5–3%; referred to here as hard agar), whereas others like Salmonella, Escherichia coli, Serratia, Pseudomonas, and Bacillus swarm only on lower percentage agar (0.5–0.8%; referred to here as medium agar to distinguish it from even lower percentage soft agar in which the bacteria swim individually within water-filled channels inside the agar). Hard-agar swarmers differentiate into specialized swarm cells that are elongated and have increased flagella. Medium-agar swarmers generally do not display a similar differentiated morphology (5, 6). In many of the latter class of swarmers (e.g., Serratia, Pseudomonas, Bacillus), movement is enabled by powerful extracellular surfactants whose synthesis is under quorum-sensing control (7, 8). Surfactants lower surface tension and allow rapid colony expansion (9 11). Salmonella and E. coli do not appear to make such surfactants (12).

An elevated resistance to multiple antibiotics has been reported for swarming populations of Salmonella enterica (13, 14), Pseudomonas aeruginosa (15), and a variety of other medium-agar swarmers, including Serratia marcescens and Bacillus subtilis (16). This resistance was reported to be linked specifically to swarming and was not observed in the same bacteria growing on hard agar, where they cannot move, or in soft agar, where they swim inside the agar. Therefore, the resistance was attributed to a physiology specific to swarmer cells. The resistance was not attributable to selection for antibiotic-resistant mutants, because the swarmer cells were killed with lethal doses of antibiotic when inoculated in fresh liquid media (13, 15), reminiscent of the nongenetic or “adaptive resistance” seen in bacterial biofilms (17).

The present study was initiated to reexamine the data showing that the adaptive resistance of Salmonella swarmers is attributable to their special physiology, a conclusion at odds with a microarray study that found essentially similar genome-wide expression profiles (a proxy for physiology) for Salmonella growing on medium vs. hard agar (i.e., swarming vs. nonswarming conditions, respectively) (6). We show here that adaptive resistance is a property of high cell densities within the swarming colony and not of swarming-specific physiology as concluded earlier. We test predictions of the Salmonella results in two other swarming bacteria—Bacillus and Serratia—and show that cell density and mobility are common protective features for survival against antimicrobials.

Results and Discussion

Antibiotic Resistance Is a Property of High Cell Density and Is Favored by Mobility.

In soft agar (0.3%), cells swim individually inside the agar and are referred to as swimmers. In medium agar (0.6%), cells move as a group on the surface and are referred to as swarmers. The original E-test strip assay showing differential antibiotic resistance of swimmer and swarmer cells of Salmonella (13, 14) is shown in Fig. 1A. The strips have a predefined gradient of antibiotic concentrations (highest at the top end), and the antibiotic diffuses into the surrounding medium when the strip is placed on the surface of the agar. The three antibiotics tested target different processes in the cell, namely, DNA replication (ciprofloxacin), protein synthesis (kanamycin), and membrane integrity (polymyxin). Bacteria are inoculated at points indicated by the asterisks, from which they migrate outward (the antibiotics do not elicit a chemotactic response). When they encounter the antibiotic, both swimmers and swarmers are expected to stop as a result of cell killing. A pear-shaped clear zone was visible for the swimmers, delineating the area into which the antibiotic had diffused, and arrested their migration. The lower end of this zone marks the minimum inhibitory concentration (MIC) for swimmers. Swarmers displayed no inhibition zones on these plates, consistent with earlier results showing that swarmers are resistant to higher antibiotic concentrations than swimmers. However, this test might potentially overestimate MIC values for swarmers because they achieve higher cell densities than swimmers. Sampling of local cell density showed that the advancing edge of a swarm colony has ~30-fold higher cell density compared with that of a swimming edge and ~75-fold higher density than an exponentially growing broth culture (Fig. 1B and Fig. S1). The high density is a property of surface growth, and similar densities are achieved on both medium and hard agar (see below).

Fig. 1.
Antibiotic response and cell densities of bacteria moving within soft agar (swim) or over the surface of medium agar (swarm). (A) E-test strips containing a gradient of indicated antibiotics (decreasing from top to bottom) were placed in the center of ...

To test if the higher resistance exhibited by swarmer cells is attributable to their higher cell density rather than their swarmer cell status, the surface of medium- and hard-agar plates was uniformly inoculated with a low density of cells (Fig. 2A; 0-h time point). E-test strips containing ciprofloxacin were applied on the surface of these plates at various times from 0 to 3 h of growth (i.e., at increasing cell density), and the plates were photographed 3 h after the application (Fig. 2B). On hard agar, where cells cannot move, the inhibition zone around the E-test strips was maximal when cell density was low (0–1.5 h), diminished significantly after 2 h of growth, and was erased by 3 h, showing a clear dependence on cell density. [Cells are still in the exponential phase of growth at these time points (6)]. On medium agar, where cells swarm, the antibiotic zone was colonized even earlier, at 1.5 h. If observed at 4 h after E-test strip application, no inhibition zones remained on swarm agar even on the 0-h plate (also see swarm plates in Fig. 1A). This experiment demonstrates the importance of high cell density in promoting growth in the antibiotic zones, irrespective of swarmer cell status. However, the swarmer population apparently has the additional advantage of mobility, promoting earlier migration into these zones after a buildup of cell density. We note that the previous conclusion that nonswarmers do not exhibit resistance was arrived at by placing E-strips on hard agar seeded with low cell densities of bacteria, similar to the 0-h sample in Fig. 2A (13, 14).

Fig. 2.
Nonswarmers show cell density-dependent antibiotic resistance. (A) Phase-contrast images (100× magnification) showing the density distribution of cells during 0–2 h of growth on the surface of swarm and nonswarm hard-agar plates. The pour-and-drain ...

Cell density dependence of antibiotic resistance could also be observed in concentrated broth-grown cells (Fig. S2).

Border-Crossing Assay to Measure Adaptive Resistance of Swimming vs. Swarming Bacteria.

The antibiotic zone around the E-test strips is narrow and readily infiltrated by swarmers. To test how far swarmers would travel on a wider antibiotic surface, we set up a test we call the “border-crossing test,” where the border is a plastic barrier dividing a Petri plate into two chambers. Media was poured into both chambers, but antibiotic was added only to the right chamber (Fig. 3A). A thin (~1-mm tall) agar bridge was constructed over the barrier (Methods), which allowed the bacteria inoculated in the left chamber to cross over and migrate to the right. The narrow bridge minimized antibiotic diffusion, as seen by lack of significant growth inhibition left of the border. The cross-border plates gave swarmers more time to declare their MICs, because it takes several generations for Salmonella at the border to colonize the entire right chamber on control plates (5–6 h at 37 °C). Both swimmers and swarmers had to navigate the border crossing in a similar space. Swarmers were clearly able to move on higher antibiotic media than swimmers, although the rate of advancement of the swarm front decreased with increasing antibiotic until the swarmers were eventually arrested at the border. It took 10-fold higher kanamycin and 200-fold higher ciprofloxacin to arrest swarmers compared with that required to arrest swimmers at the border (Fig. 3A).

Fig. 3.
Border-crossing assay, adaptive resistance, and cell death in Salmonella. (A) Cells were inoculated in the left no-antibiotic chamber and allowed to migrate to the right antibiotic-containing chamber (Methods). Numbers refer to μg/mL of indicated ...

If cells surviving on the antibiotic surface had turned on pathways to enable resistance, they should exhibit a growth advantage when transferred to fresh swarm media containing similar antibiotic concentrations. Transferred cells were picked from the moving edge to ascertain that they were still in the exponential phase of growth. Such a transfer maintains the physiological status of the cells but does not deliver the original high cell densities. These cells were killed on transfer, ascertaining that their antibiotic resistance was not induced (Fig. 3B). When plated clonally, similar results were obtained (i.e., the resistant population was unable to grow from single cells on transfer to either medium- or hard-agar antibiotic plates) (Fig. S3). We conclude that some feature of the swarming colony other than long-lived induced resistance must contribute to its survival; the ability to swarm acts as a preadaptation for survival in an antibiotic-containing solid environment.

Swarming Bacteria Sustain Cell Death While Navigating the Antibiotic Surface.

If antibiotic resistance is not induced but is somehow afforded by high cell density and an ability to move, a subpopulation of cells that is in direct or prolonged contact with the antibiotic likely gets killed. The advancing edge of a swarming Salmonella colony has an ~2–3 cell-wide zone consisting of a monolayer of cells but is generally multilayered behind this edge, with cells moving continuously through these layers (Movie S1). Survivors are likely those that are in the upper layers or those that minimize their exposure to the antibiotic by circulation through the multilayers. To test if cells from the antibiotic region are killed, they were treated with live/dead stain that stains live cells green and dead cells red. Cell death was clearly apparent in cells taken from the antibiotic regions (Fig. 3C), consistent with the visibly lower growth resulting from cells transferred from the antibiotic region compared with those from the control region (Fig. 3B, Left). Thus, the swarming colony endures death of a subpopulation while continuing to move.

Quorum-Sensing Regulators Are Not Involved in Tolerance to Antibiotics.

In a process referred to as quorum-sensing, bacteria can produce and detect signaling molecules to control their behavior in response to variation in cell density (18). To determine if the high cell densities of swarming bacteria turn on quorum-sensing pathways that aid migration on the antibiotic surface, we tested a luxS mutant defective in synthesis of the only known quorum-sensing signaling molecule in Salmonella, an N-acylhomoserine lactone derivative called AI-2 (19). AI-2 synthesis has been reported to be up-regulated in Salmonella swarmers (20), and there is a report that AI-2 affects antibiotic susceptibility of Streptococcus anginosus (21). A luxS mutant showed antibiotic resistance similar to the WT control (Fig. S4). Thus, AI-2 is not necessary for the ability of Salmonella swarmers to migrate into antibiotic zones.

We conclude from the experiments in Figs. 13 and Figs. S2S4 that adaptive resistance or tolerance to antibiotics is not a property of a gene expression program specific to Salmonella swarmers, as concluded earlier (13, 14), but is rather afforded by high cell densities. The known quorum-sensing pathway in Salmonella does not play a role in adaptive resistance. Although all cells—swarmers, nonswarmers, and broth-grown swimmers—can tolerate higher antibiotic concentrations at high cell densities, the ability to move gives swarmers an added advantage in overriding the antibiotic. Individuals within the dense-moving group, likely those directly exposed to the antibiotic, undergo cell death, protecting cells that are likely not directly exposed.

Faster Migration Enables Higher Adaptive Resistance.

Because cell density and mobility are apparently the only protective features of the swarm and a subset of cells is being killed by the antibiotic, slower swarmers would be expected to suffer higher casualties because of longer exposure to the antibiotic, and would therefore invade less territory compared with faster swarmers. To test this prediction, we compared resistance to kanamycin and ciprofloxacin at two different temperatures, which promote different rates of movement. At 30 °C and 37 °C, the Salmonella swarming fronts move at the average rate of 1.5 mm/h and 5 mm/h, respectively. A temperature of 37 °C enabled migration over higher antibiotic concentrations than one of 30°C (Fig. 4A, Left), even though sensitivities of broth-grown cells to these antibiotics are similar at both temperatures (Fig. 4A, Right). Similarly, a Salmonella mutant that swarms at a slower rate was unable to move over antibiotic concentrations easily colonized by the WT at the same temperature (Fig. S5). These data show a direct relation between adaptive resistance and swarming speed, and they support the notion that the exposure time of the group to the antibiotic is a critical factor affecting resistance.

Fig. 4.
Faster migration enables higher adaptive resistance. (A) Cross-border swarm plates were inoculated with Salmonella as described in Fig. 3A. The 37 °C and 30 °C plates were incubated for 16 and 26 h, respectively, the time at which control ...

Testing Predictions of Salmonella Results in Other Swarming Bacteria.

The swarming behavior of many medium-agar swarmers, particularly those with peritrichous flagella, is very similar. Swarming initiates only after a buildup of cell density; cells at the edge of the colony are not as motile as those immediately behind; multilayered bacterial rafts swirl in different directions in highly motile regions; and cells are motile only in groups, becoming immotile if accidentally isolated. This shared behavior likely comes from a common cell shape, common flagellar mechanics, and common challenge of moving against surface friction. Because adaptive resistance to antimicrobials is exhibited by many medium-agar swarmers, it is reasonable to assume that the survival strategy used by Salmonella will be shared by these other swarmers. We therefore tested two predictions from the Salmonella results in other swarming bacteria that show adaptive resistance—S. marcescens and B. subtilis.

The prediction that adaptive resistance should be governed by swarming speed was satisfied in Salmonella in two experimental setups: at two different temperatures with WT (Fig. 4A) and at the same temperature with WT vs. a slow-swarming mutant (Fig. S5). To extend these results further, we compared two different bacteria—S. enterica and S. marcescens—that swarm at different speeds at the same temperature. The swarming dynamics and group morphology of these bacteria are otherwise indistinguishable (Movie S2, compare with Movie S1). The faster speed of Serratia is attributable to a secreted lipopeptide surfactant (9, 22). Under conditions optimal for Serratia motility (30°C), the swarming front moves at the average rate of 7 mm/h in Serratia and 1.5 mm/h in Salmonella. Cross-border experiments comparing migration of the bacteria on kanamycin and ciprofloxacin plates are shown in Fig. 4B. Broth-grown cells of both bacteria are sensitive to low levels of these antibiotics (Fig. 4B, Right). Serratia is more sensitive to kanamycin than Salmonella, yet it efficiently colonized kanamycin 20 as well as ciprofloxacin 0.5 plates, whereas Salmonella was stopped near the border. Thus, the relation between adaptive resistance and swarming speed observed in Salmonella could be extended to Serratia.

Another prediction of the Salmonella experiments is that the cells that get killed are those directly exposed to the antibiotic, whereas those that are protected are in the interior of the multilayered swarming colony. To test this, we turned to B. subtilis, which initially sends forth a monolayer of cells, exposing them directly to the antibiotic (10). The monolayer migrates rapidly, aided by a lipopeptide surfactant (23) (Movies S3 and S4). The colony later becomes multilayered, attributable both to subsequent waves of bacteria that travel over the monolayer and growth within the monolayer. The behavior of Bacillus with increasing polymyxin concentrations is shown in Fig. 5A. The initial monolayer was seen traversing the right chamber at all antibiotic concentrations (likely aided by rapid spreading of the surfactant), although its rate of advancement decreased with increasing polymyxin. Live/dead staining revealed increasing cell death in this monolayer with increasing antibiotic (Fig. 5B). Cells from the left continued to move in and swarm over the monolayer, but they were stalled at the border at higher antibiotic concentrations. To test if a multilayered swarm would offer more protection, we altered the experimental setup, allowing swarmer cells to build up density in the left chamber before pouring antibiotic media into the right chamber. As a dense colony, cells were able to cross over and survive on higher antibiotic zones compared with a single-layered colony (Fig. 5C; compare polymyxin 20 and polymyxin 50 in Fig. 5C with similar plates in Fig. 5A). These experiments demonstrate both that cells directly in contact with the antibiotic get killed and that a multilayered colony is at an advantage compared with a monolayered colony while navigating antimicrobial territory, satisfying the expectations from the Salmonella results.

Fig. 5.
Behavior of Bacillus in the cross-border assay. (A) Cross-border migration of B. subtilis over increasing polymyxin concentrations. Numbers refer to μg/mL. Plates were incubated at 37 °C for 16 h. (B) Increasing cell death in the monolayer ...

That observations with swarming Salmonella can be extended to different bacterial genera speaks to a commonly conserved behavior of their swarms.

Adaptive Resistance: Self-Sacrifice or Selfish Behavior?

Our study has shown that high bacterial densities promote survival of swarming bacteria in certain types of harsh environments. The survival occurs without apparently altering gene expression but at a cost to some individuals. Swarms typically move after reaching a density threshold. Movement to a different location involves risk, and in some cases, part of or all the moving swarm could get wiped out. This cost of movement might appear to represent a form of self-sacrifice or “altruism,” a trait particularly observed in species with complex social structures (24). However, migration can be favored as a “selfish” trait, even when the death rate during movement is high (25), such that the death that occurs in a swarm does not, by itself, point to altruism. From another perspective, there is a clear group benefit of high density in a swarm. Group benefits are sometimes associated with (or identified with) altruistic behavior. At face value, however, our observations could sit well with more than one model in which group benefit follows from phenotypic heterogeneity and intercellular interactions with selection at the level of the individual (26), combined with a Poisson-type distribution of group sizes in each generation (27), or “safety in numbers” leading to “byproduct benefit” (28). None of these models require altruism. To all appearances, our observations would favor the selfish model, in which all cells are actively trying to stay alive but some get caught in the swarm in a way that leads to their death. For example, survival may be highest on top (furthest from the antibiotic), but some bacteria just get “piled on” and die because they cannot get off the bottom in time. In general, there may be positions within a swarm that are better for survival than others, and if bacteria are capable of sensing where those best locations lie, there could be considerable selfishness to individual bacterial attempts to reach those positions, leading to classic “selfish herd” dynamics (29). We point to a parallel behavior of red fire ants, which, when floods arrive, survive by binding together the entire colony into a dense ball that floats on the flood waters until the ants drift to higher ground (Fig. S6). The ants constantly reposition themselves to minimize their exposure to water (30).

Whatever the underlying mechanism and evolutionary basis, a group-level trait, namely, swarming behavior, confers a fitness advantage to individual members of the group when the environment contains something harmful. The mobility of the swarm allows it to “outrun” harsh conditions to reach safer ground. The dead bacteria in immediate contact with the antibiotic might provide a physical barrier that protects those on top. They might also feed the group with nutrients released on their death, as seen during the cannibalistic behavior of B. subtilis bacteria, which feed on their siblings to delay committing to spore formation (31).

In summary, three different swarming bacteria exhibit a common survival strategy against antibiotics. This strategy involves maintaining high cell density, circulating within the multilayered colony to minimize exposure to the antibiotic, and the death of individuals that are directly exposed.


Strains and Growth Conditions.

WT Salmonella enterica serovar Typhimurium (strain 14028) and its mutant luxS strain have been described (12, 32), as has WT Serratia marcescens (strain 274) (22). The flhE mutant of S. enterica is a complete gene deletion, constructed by Jaemin Lee (University of Texas, Austin). WT Bacillus subtilis strain 3610 was obtained from Daniel Kearns (Indiana University, IN). All strains were grown in LB (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl) broth. LB plates (25 mL) were solidified with 0.3%, 0.6%, or 1.5% Bacto agar (Difco). For S. enterica, 0.5% glucose was added to swarm plates. Plates were allowed to dry on the bench top overnight and were used the next day. For S. marcescens, optimal motility is observed at 30°C and is inhibited at 37°C (22). The optimal temperature for S. enterica and B. subtilis motility is 37°C. Border-crossing plates were prepared by pouring 30 mL of LB swarm media into each chamber in a two-step process: The antibiotic media were poured first and allowed to harden before the nonantibiotic side was filled. Before the latter media hardened, a sterile culture stick was introduced into the molten media and the meniscus was dragged over the plastic border, connecting the two sides with an ~1-mm tall agar bridge. Motility plates were inoculated with 5 μL of an exponentially growing broth culture at an OD600 of ~0.7. The drop was allowed to dry for ~15 min with the lid off and was then transferred to the incubator. Plates were photographed using either a BioRad Geldoc system or a “bucket of light” device (33) and a Canon Rebel XSI digital camera.

Cell Density Measurements.

Cell density measurements were determined by controlled sampling using the flat end of a round culture stick (2-mm diameter). The stick was held upright and gently touched to the culture to be sampled (swimming, swarming, or broth culture). The stick was then rinsed in 1 mL of LB. The sampling procedure was repeated two more times, using a different stick each time, and the bacteria from all three samplings were collected in the same 1 mL of LB. Serial dilutions of this tube were plated on LB agar plates to determine the cfus. The cell numbers were normalized to cells per sampling stick. Three independent experimental repeats of this method yielded similar cell numbers. Broth-grown cells were concentrated as follows: 15 mL of an exponentially growing LB culture at an OD600 of ~0.7 was pelleted by centrifugation at 5,000 × g, diluted to various extents with LB, and sampled with the end of a culture stick as described previously.

E-Test Assay.

The E-test strip comprises a predefined gradient of antibiotic concentrations on a plastic strip. The strips were purchased from AB Biodisk and applied to the surface of swim or swarm plates before spot inoculation of cells on either side. On hard-agar plates, broth-grown cells were uniformly inoculated by the pour-and-drain method described by Wang et al. (6) before application of the strips. All assays were repeated at least three times.

Live/Dead Staining.

The stain was purchased from Invitrogen, and cells were stained according to the manufacturer’s specifications. The kit includes two nucleic acid stains—green-fluorescent STYO 9 and red-fluorescent propidium iodide (PI). STYO 9 labels both live and dead bacteria alike, whereas PI reduces STYO 9 stain intensity only after crossing damaged cellular membranes. To determine red/green cell numbers, at least 500 cells were counted in each sample analyzed.


Phase-contrast images were obtained with a DP-12 digital camera (Olympus) attached to an Olympus BH2 microscope. Red and green fluorescence of cells stained with live/dead stain was monitored using an Olympus BX60 microscope equipped with a Photometrics Quantix camera system. Optimal excitation wavelengths for STYO and PI are 480 nm and 490 nm, respectively. Red and green images were captured and overlaid using MetaMorph software (Molecular Devices Corporation) or MDC. Movies of swarming bacteria were recorded using an Olympus IX50 microscope, maintained in a temperature- and humidity-controlled environment, and equipped with LD ×20 and ×60 phase-contrast objective lenses. Motion was captured with a digital camera at 30 frames per second and a spatial resolution of 640 × 480 pixels. To improve clarity and minimize vibration attributable to bacterial motion, 11 mL of LB swarm medium was used per plate.

Supplementary Material

Supporting Information:


We thank Avraham Be’er for use of his microscope facility and help with recording the movies. We are grateful to James Bull, Raghavendra Gadagkar, Richard Meyer, and Vidyanand Nanjundiah for their comments and ideas on evolutionary mechanisms and to David Moynahan for sharing his fire ant image. This work was supported by National Institutes of Health Grant GM 57400.


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0910934107/DCSupplemental.


1. Harshey RM. Bacterial motility on a surface: Many ways to a common goal. Annu Rev Microbiol. 2003;57:249–273. [PubMed]
2. McCarter LL. Dual flagellar systems enable motility under different circumstances. J Mol Microbiol Biotechnol. 2004;7:18–29. [PubMed]
3. Rather PN. Swarmer cell differentiation in Proteus mirabilis . Environ Microbiol. 2005;7:1065–1073. [PubMed]
4. Verstraeten N, et al. Living on a surface: Swarming and biofilm formation. Trends Microbiol. 2008;16:496–506. [PubMed]
5. Tolker-Nielsen T, et al. Assessment of flhDC mRNA levels in Serratia liquefaciens swarm cells. J Bacteriol. 2000;182:2680–2686. [PMC free article] [PubMed]
6. Wang Q, Frye JG, McClelland M, Harshey RM. Gene expression patterns during swarming in Salmonella typhimurium: Genes specific to surface growth and putative new motility and pathogenicity genes. Mol Microbiol. 2004;52:169–187. [PubMed]
7. Ochsner UA, Reiser J. Autoinducer-mediated regulation of rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa . Proc Natl Acad Sci USA. 1995;92:6424–6428. [PMC free article] [PubMed]
8. Eberl L, et al. Involvement of N-acyl-L-hormoserine lactone autoinducers in controlling the multicellular behaviour of Serratia liquefaciens . Mol Microbiol. 1996;20:127–136. [PubMed]
9. Matsuyama T, et al. A novel extracellular cyclic lipopeptide which promotes flagellum-dependent and -independent spreading growth of Serratia marcescens . J Bacteriol. 1992;174:1769–1776. [PMC free article] [PubMed]
10. Kearns DB, Losick R. Swarming motility in undomesticated Bacillus subtilis . Mol Microbiol. 2003;49:581–590. [PubMed]
11. Caiazza NC, Shanks RM, O’Toole GA. Rhamnolipids modulate swarming motility patterns of Pseudomonas aeruginosa . J Bacteriol. 2005;187:7351–7361. [PMC free article] [PubMed]
12. Toguchi A, Siano M, Burkart M, Harshey RM. Genetics of swarming motility in Salmonella enterica serovar typhimurium: Critical role for lipopolysaccharide. J Bacteriol. 2000;182:6308–6321. [PMC free article] [PubMed]
13. Kim W, Killam T, Sood V, Surette MG. Swarm-cell differentiation in Salmonella enterica serovar typhimurium results in elevated resistance to multiple antibiotics. J Bacteriol. 2003;185:3111–3117. [PMC free article] [PubMed]
14. Kim W, Surette MG. Swarming populations of Salmonella represent a unique physiological state coupled to multiple mechanisms of antibiotic resistance. Biological Procedures Online. 2003;5:189–196. [PMC free article] [PubMed]
15. Overhage J, Bains M, Brazas MD, Hancock RE. Swarming of Pseudomonas aeruginosa is a complex adaptation leading to increased production of virulence factors and antibiotic resistance. J Bacteriol. 2008;190:2671–2679. [PMC free article] [PubMed]
16. Lai S, Tremblay J, Déziel E. Swarming motility: A multicellular behaviour conferring antimicrobial resistance. Environ Microbiol. 2009;11:126–136. [PubMed]
17. Stewart PS. Mechanisms of antibiotic resistance in bacterial biofilms. Int J Med Microbiol. 2002;292:107–113. [PubMed]
18. Bassler BL, Losick R. Bacterially speaking. Cell. 2006;125:237–246. [PubMed]
19. Schauder S, Shokat K, Surette MG, Bassler BL. The LuxS family of bacterial autoinducers: Biosynthesis of a novel quorum-sensing signal molecule. Mol Microbiol. 2001;41:463–476. [PubMed]
20. Kim W, Surette MG. Coordinated regulation of two independent cell-cell signaling systems and swarmer differentiation in Salmonella enterica serovar Typhimurium. J Bacteriol. 2006;188:431–440. [PMC free article] [PubMed]
21. Ahmed NA, Petersen FC, Scheie AA. AI-2 quorum sensing affects antibiotic susceptibility in Streptococcus anginosus . J Antimicrob Chemother. 2007;60:49–53. [PubMed]
22. Alberti L, Harshey RM. Differentiation of Serratia marcescens 274 into swimmer and swarmer cells. J Bacteriol. 1990;172:4322–4328. [PMC free article] [PubMed]
23. Tsan P, Volpon L, Besson F, Lancelin JM. Structure and dynamics of surfactin studied by NMR in micellar media. J Am Chem Soc. 2007;129:1968–1977. [PubMed]
24. Gadagkar R. Survival Strategies: Cooperation and Conflict in Animal Societies. Cambridge, MA: Harvard Univ Press; 1997.
25. Hamilton WD, May RM. Dispersal in stable habitats. Nature. 1977;269:578–581.
26. Atzmony D, Zahavi A, Nanjundiah V. Altruistic behaviour in Dictyostelium discoideum explained on the basis of individual selection. Curr Sci. 1997;72:142–145.
27. Chuang JS, Rivoire O, Leibler S. Simpson's paradox in a synthetic microbial system. Science. 2009;323:272–275. [PubMed]
28. Sachs JL, Mueller UG, Wilcox TP, Bull JJ. The evolution of cooperation. Q Rev Biol. 2004;79:135–160. [PubMed]
29. Hamilton WD. Geometry for the selfish herd. J Theor Biol. 1971;31:295–311. [PubMed]
30. Taber S. Fire Ants. College Station, TX: Texas A&M Univ Press; 2000.
31. González-Pastor JE, Hobbs EC, Losick R. Cannibalism by sporulating bacteria. Science. 2003;301:510–513. [PubMed]
32. Wang Q, Mariconda S, Suzuki A, McClelland M, Harshey RM. Uncovering a large set of genes that affect surface motility in Salmonella enterica serovar Typhimurium. J Bacteriol. 2006;188:7981–7984. [PMC free article] [PubMed]
33. Parkinson JS. A “bucket of light” for viewing bacterial colonies in soft agar. Methods Enzymol. 2007;423:432–435. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...