Predataxis behavior in Myxococcus xanthus
James E. Berleman, Jodie Scott, Tatiana Chumley, and John R. Kirby1
Department of Microbiology, University of Iowa, 51 Newton Road, Iowa City, IA 52242
Edited by Howard C. Berg, Harvard University, Cambridge, MA, and approved September 18, 2008 (received for review May 6, 2008)
Spatial organization of cells is important for both multicellular
development and tactic responses to a changing environment. We
find that the social bacterium, Myxococcus xanthus utilizes a
chemotaxis (Che)-like pathway to regulate multicellular rippling
during predation of other microbial species. Tracking of GFP-
labeled cells indicates directed movement of M. xanthus cells
during the formation of rippling wave structures. Quantitative
analysis of rippling indicates that ripple wavelength is adaptable
and dependent on prey cell availability. Methylation of the recep-
tor, FrzCD is required for this adaptation: a frzF methyltransferase
mutant is unable to construct ripples, whereas a frzG methyles-
terase mutant forms numerous, tightly packed ripples. Both the
frzF and frzG mutant strains are defective in directing cell move-
ment through prey colonies. These data indicate that the transition
to an organized multicellular state during predation in M. xanthus
relies on the tactic behavior of individual cells, mediated by a
Che-like signal transduction pathway.
chemotaxis ? morphogenesis ? predation ? rippling ? chemosensory
multicellular species (1). In most cases the origins of multicel-
lularity have been obscured by time, but in general it is thought
between sister cells of the same organism. Many bacterial species
also show multicellular organization in colony architecture or
biofilm ultrastructure (2, 3). In bacteria, multicellularity is often
induced by specific environmental conditions rather than as a
function of programmed growth. In this report, we examine how
the predatory bacterium, Myxococcus xanthus reorganizes its
multicellular structure after invading a colony of suitable Esch-
erichia coli prey.
M. xanthus is capable of forming several distinct multicellular
structures (4). M. xanthus cells are motile on solid surfaces
through gliding as a combination of individuals and groups
movement is driven by a combination of polar type IV pili and
distributed focal adhesion sites (5–7). Under conditions of high
cell density, M. xanthus cells form fruiting body aggregates of
?105-106cells. Within fruiting bodies, low-nutrient availability
induces vegetative rods to differentiate into nonmotile spores
(8). During predation, M. xanthus cells organize into rippling
wave structures. Rippling behavior coordinates cell movement
across areas of several square millimeters and can involve ?108
cells working as a unit.
Multicellular rippling in M. xanthus is a phenomenon that is
superficially similar to many patterns displayed by other biolog-
ical and chemical systems (9). Yet, it is unique in that counter
migrating cells reflect off each other producing a convective-
rather than diffusion-based wave pattern (10). One model is that
rippling constitutes an intermediate phase of cell organization
before a more permanent fruiting body structure (11). However,
rippling is not observed under all conditions that promote
fruiting body formation (12). Additionally, rippling is induced by
M. xanthus during predation on a variety of microbial species or
degradation of macromolecular growth substrates such as pep-
tidoglycan, protein, and chromosomal DNA (13, 14). An alter-
ulticellularity has emerged several times in separate lin-
eages of life, giving rise to plants, animals, and many other
movement to enhance contact with prey macromolecules to
In swimming E. coli cells, individuals find optimal positions
within spatial gradients of chemoeffectors by altering their
behavior (smooth swimming and tumbling), a process controlled
by the chemotaxis (Che) signal transduction pathway (15).
Nonchemotactic mutants are less competent than chemotactic
strains in both capillary assays and infection models (16, 17).
Each individual cell makes decisions on the basis of its own
perception of the local chemical environment, yet complex
multicellular patterns can emerge, including expanding swarm
rings and focal aggregates (18, 19). Chemical signals are sensed
through an array of chemoreceptors and the signal is transduced
to the flagellar motors to control the swimming behavior of cells
(20). The sensitivity of the receptors in Che-like signal trans-
duction is subject to methylation via the opposing biochemical
activities of the methyltransferase (CheR) and methylesterase
(CheB). Altering the sensitivity of the receptors allows cells to
adapt their behavior to subtle changes across a wide range of
chemical concentrations. E. coli cheR and cheB mutants are
ineffective at directing cell movement through chemical gradi-
ents as cells are locked in either a constitutively smooth swim-
ming or a constitutively tumbly behavioral state (15).
In M. xanthus, the Che-like Frz pathway has previously been
pole during surface gliding, similar to regulation of flagellar
rotation in E. coli (21). Therefore, if predatory rippling in M.
xanthus depends on the behavior of individuals similar to that
seen during E. coli chemotaxis, then we predict that the Frz
pathway will be required for controlling this behavior. If this
hypothesis is correct, M. xanthus cells should display the two
hallmarks of bacterial chemotactic behavior during predatory
rippling: directed movement and adaptation to a stimulus.
Directed Movement During Cellular Reorganization to Rippling. To
analyze predatory behavior we make use of M. xanthus strain
DZ2, which has a low intrinsic level of autolysis and E. coli strain
?2155, which is a diaminopimelic acid (DAP) auxotroph whose
M. xanthus behavior in relation to an unchanging source of prey.
During predation assays, M. xanthus cells organize into dramat-
ically different structures depending on whether they are inside
or outside of the prey colony (Fig. 1A). In the absence of prey,
M. xanthus cells glide in amorphous groups such that no stable
pattern of organization can be observed after 1 h of incubation
(Fig. 1B). In contrast, cells within the prey colony organize into
dynamic, parallel rippling waves and maintain this organization
over time (Fig. 1C). In both the presence and absence of prey,
cells are actively motile and glide at similar velocities: 2.7 ?/?
Author contributions: J.E.B. and J.R.K. designed research; J.E.B., J.S., and T.C. performed
research; J.E.B., J.S., and J.R.K. analyzed data; and J.E.B. and J.R.K. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1To whom correspondence should be addressed. E-mail: email@example.com.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2008 by The National Academy of Sciences of the USA
November 4, 2008 ?
vol. 105 ?
no. 44 ?
0.9 ?m/min in the absence of prey and 2.8 ?/? 0.7 ?m/min in the
presence of prey.
To determine whether M. xanthus cells display directed move-
ment during predatory rippling, M. xanthus strain DZ2 was
mixed with a strain carrying GFP fused to the highly expressed
pilA promoter and fluorescence microscopy was performed in
the presence and the absence of prey [Fig. 1 B and C and
supporting information (SI) Movies 1 and 2]. Tracking of cells
was performed by mixing DZ2 cells with GFP-labeled cells on an
agar-covered microscope slide and indicates that in the absence
of prey, M. xanthus cell movement occurs in curved trajectories,
with changes of direction occurring gradually through cell bend-
ing and collisions with other cells, rather than through oscillation
of the leading cell pole (Fig. 2A). Net movement of cells was
analyzed by plotting the total displacement of 100 individual cell
displacement vectors after 30 min of movement (Fig. 2A). Net
displacement occurs in all directions, with various magnitudes,
and the multicellular makeup of the group consists of an
amorphous pattern with no relationship to the initial pattern in
the absence of prey stimuli (Fig. 1).
Previous studies on cell behavior during rippling led to the
conclusion that there is no net cell displacement, i.e., no directed
movement (22). However, those experiments were performed in
closed systems where the ripple-inducing substrate was distrib-
uted throughout the sample. To examine rippling behavior in an
open system, M. xanthus strain DZ2 cells were mixed with
GFP-labeled cells and pipetted adjacent to an E. coli strain
?2155 prey colony on an agar-covered microscope slide. M.
xanthus ripples can be observed as dense bands of autofluores-
cence that aggregate and dissipate in a synchronized, oscillating
manner with a period of ?8 min (Figs. 1C and 2C). New ripple
aggregates that form are shifted by ?1⁄2 wavelength relative to
the previous ripple crests (Fig. 2C). Analysis of individual cell
behavior indicates that movement occurs along primarily linear
paths that are perpendicular to the orientation of the multicel-
lular ripple structure (compare Figs. 1C and 2B). Thus, there is
little movement in the y-axis and cells are constrained to moving
along a nearly one-dimensional path. Both the direction and
magnitude of the displacement vectors are biased to the positive
x-axis, the direction of swarm migration through the E. coli prey
colony. Thus, although rippling behavior can last for days, the
individual structures survive only for a few minutes and seem to
be indicative of a shifting equilibrium of M. xanthus cell density
rather than the intermediate construction of a larger multicel-
Adaptation to Prey Stimuli. Although rippling is a stable multicel-
lular behavior, it takes time for the cells to coalesce into waves
remaining E. coli colony is visible as a dark crescent on the Right edge. The
open box denotes the area of the phase contrast image (Right) which is
centered on the near edge of the prey colony where the transition from
nonrippling to rippling behavior can be observed. (White scale bar, 1 mm;
(Right) of DZ2 mixed in a 50:1 ratio with GFP-labeled DZ10547, showing the
characteristic mesh pattern of M. xanthus gliding behavior in the absence of
prey. (C) Within the prey colony, the rippling pattern is stably maintained,
in (A) the absence of prey and (B) the presence of E. coli prey. Experimental
setup is the same as in Fig. 1. Thirty cell tracks are shown for each condition.
Cell movement over time is represented by the change in color of each track,
with the beginning of the track labeled in dark blue and the end in dark red.
Vector displacements of 100 cells for each condition are shown on the Right.
(C) Distribution of cells was monitored at 2-min intervals by measuring fluo-
of cell movement in the given area. The wave structure dissipates as M.
xanthus cells leave the initial wave aggregates, but a similar pattern of
aggregation emerges, shifted by1⁄2 wavelength, at 8 min.
Tracking movement of M. xanthus cells. Tracking of GFP-labeled cells
www.pnas.org?cgi?doi?10.1073?pnas.0804387105 Berleman et al.
and rippling terminates soon after the available prey macromol-
ecules have been consumed, such that rippling behavior lasts for
a finite interval. To determine whether changes occur through-
out the duration of rippling behavior, M. xanthus cells were
pipetted adjacent to E. coli cells on a cloned fruiting light (CFL)
low-nutrient agar plate and the distance between wave crests was
measured from the beginning to the end of predation (Fig. 3A).
There is entry of M. xanthus cells into the prey colony, but no
measurable rippling during the first 18 h of the assay. After
initiating, the ripples display a short wavelength of ?60 ?m.
There is a gradual increase from 18 to 66 h in the wavelength of
the ripples with a maximum wavelength of 140 ?m. After 66 h,
rippling behavior was still occasionally observed in scattered
areas of the colony, but was not quantifiable.
To determine whether changes in rippling behavior occur over
time as a result of adaptation by M. xanthus cells to the gradual
loss of prey, predation assays were performed in which various
levels of prey were provided, while the initial M. xanthus cell
density was held constant (For quantification see Fig. 3B, for
images see Fig. 4C). High densities of prey stimulate formation
of a large quantity of ripples with a short wavelength between
crests. Lower cell densities of prey result in longer wavelengths
until a threshold is reached at which rippling is no longer
observed. The output wavelength observed during rippling
ranges from 70 to 135 ?m, consistent with the changes observed
over time (Fig. 3A). It is expected that the presence of prey will
also cause a gradual increase in the M. xanthus cell density as the
prey are consumed and the M. xanthus cells grow and divide.
Thus the changes in wavelength observed could be due, in part,
to increasing predator cell density. To determine whether the M.
xanthus cell density also impacts the rippling wavelength, we
performed similar assays in which the predator cell density was
varied, while the prey cell density was held constant (Fig. 3C).
Changes in the initial M. xanthus cell density have little effect on
the output wavelength of rippling. At the cell densities tested, no
significant change in ripple wavelength was observed.
Although the predator cell densities tested had no effect on
ripple spacing, analysis of a wider range of initial cell densities
indicates that the cell density of M. xanthus is critical for the
timing of rippling induction. Fig. 3D shows the average time
required for 10 parallel waves to appear. The resulting curve is
biphasic, with a steep slope below 5 ? 105cells and a shallow
curve at higher cell densities. This indicates that at predator
densities at and above 5 ? 105cells, rippling is able to initiate
rapidly, with the only delay being the result of factors such as
biosynthesis of motility organelles, sensing of prey, and the time
required to reorganize into ripples. At lower predator densities
there is a longer delay, which is likely the result of insufficient M.
xanthus cell density to facilitate rippling. Taken together these
results indicate that rippling wavelength is adaptable and in
proportion to prey cell availability.
Molecular Controls of Adaptation. In E. coli, adaptation of cell
motility to chemical stimuli occurs through the reversible meth-
ylation of the methyl-accepting chemotaxis proteins (MCPs)
catalyzed by the methyltransferase, CheR, and the methylester-
ase, CheB (23). The homologs in the M. xanthus Frz pathway that
modify the FrzCD receptor are FrzF and FrzG, respectively (20).
While cells lacking FrzF reverse direction of movement very
to any obvious defect in the reversal frequency (24). To deter-
mine the impact of the Frz pathway on multicellular structures,
we examined the wild-type, ?frzF, and ?frzG strains during
gliding, fruiting body formation, and rippling behaviors (Fig. 4).
During colony expansion, wild-type and ?frzG cells glide in a
meshed pattern of individuals and groups (Fig. 4A). In contrast,
?frzF displays a dispersed pattern of individuals and very small
groups. Under low-nutrient, high-cell-density conditions, dis-
tinct fruiting body aggregates can be observed forming in the
total M. xanthus strain DZ2 cells pipetted onto CFL media adjacent to a 3-?l colony containing 5 ? 107cells of E. coli ?2155 prey. The dashed line indicates times
when quantifiable rippling was not observed. (B) Rippling wavelength at 40 h changes as a function of E. coli ?2155 prey cell density (2 ? 106to 6 ? 107total
cells) in M. xanthus strain DZ2 (diamonds), the wavelength is constitutively short in ?frzG (open sqaures), and no rippling is observed in ?frzF (?). (C) Rippling
wavelength shows no change as a function of M. xanthus strain DZ2 cell density at 40 h (2 ? 105to 4 ? 107total cells) incubated with 5 ? 107cells of E. coli ?2155
prey. (D) Rippling induction time changes as a function of M. xanthus strain DZ2 cell density (6 ? 102to 1 ? 107total cells) incubated with 5 ? 107cells of E. coli
Quantitative analysis of rippling pattern dynamics. (A) Rippling wavelength as a function of incubation time during predation. A 1-?l suspension of 106
Berleman et al.
November 4, 2008 ?
vol. 105 ?
no. 44 ?
wild-type and ?frzG strains, but the ?frzF mutant displays
disorganized, frizzy aggregates of cells (Fig. 4B) as described
previously (20). In the presence of prey, wild-type M. xanthus
ripples in accordance with the abundance of prey (Fig. 4C). The
?frzF mutant does not show rippling behavior at any of the prey
hyperrippling phenotype, in which ripple structures are tightly
packed resulting in a constitutively short output wavelength
see Fig. 3B). This hyperrippling phenotype is consistent with an
inability to adapt to decreasing prey availability. Rippling was
not observed in either the wild-type strain or the ?frzG mutant
below 106total prey cells.
To determine whether the rippling defects observed in the
?frzF and ?frzG mutants are the result of a biochemical role in
which the FrzCD MCP homolog is incorrectly methylated during
predatory rippling, we isolated cells from rippling and nonrip-
pling conditions and probed for the presence of the methylated
and demethylated forms of the FrzCD receptor using ?-FrzCD
antibody. Previous work has demonstrated that altering meth-
ylation of MCPs affects mobility under standard SDS/PAGE
conditions and can be detected using immunoblot analysis (25).
The apparent lower molecular weight form of FrzCD corre-
sponds to a more highly methylated version of the receptor while
demethylation of FrzCD produces an apparent increase in
molecular weight (26).
Immunoblot analysis of strain DZ2 after growth in nutrient-
rich liquid medium shows that both methylated and demethyl-
ated forms of FrzCD are detectable (Fig. 5). Both forms of
FrzCD were also detected when cells were incubated on a solid
surface for 24 h in the absence of prey. However, in the presence
of prey, FrzCD can only be detected in the methylated state. In
a ?frzF methyltransferase mutant, FrzCD was detected consti-
tutively in the demethylated state under the three conditions
tested (Fig. 5, lanes 4–6). In a ?frzG methylesterase mutant, the
FrzCD protein was detected primarily in the methylated state
(Fig. 5, lanes 7–9). Together, these data indicate that the
presence of prey can alter the methylation state of the FrzCD
receptor and that both the FrzF and FrzG proteins are required
in response to prey.
Adaptation Components Are Essential in a Predatory Taxis Assay. E.
coli cheB and cheR mutants have opposite behavioral pheno-
types; cheB is tumbly and cheR is smooth swimming. Yet, both
mutants are defective in their ability to direct movement in
various chemotaxis assays (15, 27, 28). Similarly, the ?frzF and
?frzG mutants show opposite phenotypes with respect to rip-
pling behavior; ?frzF does not ripple, whereas ?frzG ripples at
a constitutively short wavelength. To determine the effect of
these mutations on general predation ability, we devised a
predatory taxis assay. In this assay, M. xanthus cells are pipetted
into the center of a long strip of E. coli prey cells and the swarm
expansion is measured in the dimension containing prey (x-axis)
and the dimension lacking prey (y-axis) (Fig. 6A). In the absence
of prey, M. xanthus strain DZ2 expands uniformly at a rate of
?100 ?m/h (Fig. 6B). This rate is constant across a wide range
of basal nutrient levels (0.01–10 g/liter casitone). In the presence
of prey, the rate of swarm expansion in wild type is unchanged
during the first 24 h when rippling is not observed. Soon after
rippling induction, swarm expansion increases to a rate of ?190
?m/h. In a ?frzG mutant the rate of swarm expansion is ?100
?m/h both in the presence and absence of prey, even after the
induction of rippling behavior (Fig. 6C). In the ?frzF mutant,
swarm expansion is maintained at ?100 ?m/h in the presence of
begins at 100 ?m/h during the first 24 h but decreases over time
to a rate of 50 ?m/h during the last 24 h of the assay (Fig. 6D).
Together, these data indicate that adaptation via methylation of
FrzCD is required for directed movement during predation.
During bacterial chemotaxis, the methylation state of the re-
ceptors is regulated by the adaptation proteins CheB and CheR
ing 107M. xanthus strain DZ2 cells was pipetted on CFL agar in the absence of
prey. Stereo and phase microscopy images were captured at 72 h at (A) the
colony edge at which group gliding behavior can be observed and (B) the
colony center where fruiting body aggregation is observed. A ?frzF mutant is
defective in forming gliding groups and also forms unorganized ‘‘frizzy’’
aggregates. ?frzG displays a phenotype with only subtle differences from the
wild type in the absence of prey. (C) After 40 h in coculture with E. coli prey,
wild-type cells modulate their rippling wavelength in response to the avail-
ability of prey. The ?frzF mutant does not form ripple structures. The ?frzG
mutant forms numerous ripples at all prey cell densities in which rippling is
induced. The direction of migration through E. coli prey is from Left to Right.
(Black bar, 100 ?m in a and c; White bar, 1 mm (b).)
Cooperation analysis in adaptation mutants. A 10-?l aliquot contain-
Cells of DZ2 (wild type, lanes 1–3), ?frzF (cheR, lanes 4–6), and ?frzG (cheB,
lanes 7–9) were harvested after 24-h growth in liquid CYE (lanes 1, 4, and 7)
and plated on CFL media (108total cells) either in the absence (lanes 2, 5, and
8) or presence of 109E. coli prey cells (lanes 3, 6, and 9). Cells were harvested
after 24 h at 32 °C and all samples were analyzed using SDS/PAGE to separate
the methylated and demethylated forms of FrzCD, as described previously
(26). frzCDC, a strain that expresses a truncated form of FrzCD, was used as a
negative control (lane 10).
Immunoblot analysis of the methylation state of the FrzCD receptor.
www.pnas.org?cgi?doi?10.1073?pnas.0804387105 Berleman et al.
and provides a molecular memory that allows cells to compare
the current signal to the recent past and adjust their behavior
accordingly (23). Recently, Che-like pathways have been shown
to control processes other than chemotaxis including biofilm
formation, spore cell differentiation, and flagella biosynthesis
(29–31). Here, we show that a Che-like mechanism is used by M.
xanthus cells to navigate through prey colonies and causes the
formation of transient multicellular rippling structures during
predation. We refer to this behavior as predataxis, as the
behavior of M. xanthus cells during predation is a novel example
of how bacteria can regulate cell movement in response to a
There are three main points to consider when comparing
directed movement during predation by M. xanthus to the results
from more conventional chemotactic assays on M. xanthus
gliding. One, chemotaxis assays use soft 0.3% agar, on which
cells glide more rapidly relative to the 1.5% agar used in the
predation assays described here (32, 33). Two, no small molecule
has yet been identified that can induce rippling behavior, al-
though several nondiffusing macromolecules such as peptidogly-
effectors on M. xanthus movement have been identified. Che-
morepellents, such as isoamyl alcohol, cause individual M.
xanthus cells to hyperreverse (34, 35). Chemoattractants such as
phosphatidyl ethanolamine (PE) derivatives with specific fatty
acid side chains can elicit a positive chemotaxis response and
cause individual cells to inhibit reversals (32, 36). It is surprising
then, that during predation a positive tactic response is observed
in combination with an increase in cellular reversals. Taken
together, these data indicate that the ability of M. xanthus cells
to navigate through a nondiffusing colony of suitable prey is
fundamentally different from the way cells respond to small
On the basis of the observed individual M. xanthus cell velocity
average of 2.8 ? 0.7 ?m/min, the maximum rate of swarm
expansion should be in the range of 60–105 ?m/h if cell
movement is random. We observed that swarm expansion occurs
at a rate of ?100 ?m/h in the absence of prey, but increases to
as much as 190 ?m/h in the presence of prey. Because no
significant increase in the velocity of individual cells was ob-
served after rippling induction, the increased rate of swarm
expansion in the direction of prey must be the result of the
inhibition of random cell movement, allowing cells to dedicate
nearly all of their potential velocity (120–205 ?m/h if there are
no reversals) into vectors oriented toward more prey.
Individual M. xanthus cells move more slowly than M. xanthus
entire prey colony on its own. Thus, we explain the emergence of
rippling behavior as a balance between two constraints—
maintaining contact with neighboring M. xanthus cells to achieve
maximum gliding velocity, balanced with an impetus to maintain
contact with prey cells for maximum feeding potential. Cells that
move ‘‘forward’’ through a prey colony will inevitably distance
themselves from their peers, in which case these cells would benefit
through a prey colony will eventually lose contact with prey as they
move toward a higher density of M. xanthus cells, again leading to
a benefit from a cell reversal. The combination of these two
structures during predation.
M. xanthus demonstrates a number of cooperative traits
including cell density-dependent macromolecule degradation,
cell–cell exchange of motility components, and a higher effi-
ciency of prey cell lysis when rippling is observed (11, 14, 37, 38).
Yet, the formation of ripples alone is insufficient for maximum
predation, because the ?frzG mutant is competent for rippling,
yet defective at increasing swarm expansion through a prey
colony. Thus, cooperative group behavior in M. xanthus occurs
against a backdrop of the tactic behavior of individuals.
Strain Handling. M. xanthus strains DZ2 (wild type), DZ4483 (?frzF), DZ4482
(?frzG), DZ4486 (frzCDC), and DZ10547 (pilA::gfp/pilA?) were used in this study
(24, 39). E. coli strain ?2155 was used as prey (40). For routine culturing, M.
xanthus was grown in CYE broth and E. coli in LB broth (24, 41). Kanamycin was
for analysis of M. xanthus cells under low-nutrient conditions (8). Ten millimolar
MOPS, pH 7.6 buffer was used for harvesting and washing of cells.
Cell Tracking. Microscope slides were prepared with rubber gaskets loaded
with a 750-?l volume of CFL media solidified with 1.0% agarose. One micro-
liter of M. xanthus (106total cells) and 3 ?l of E. coli (5 ? 107total cells) were
added to the agarose pad and incubated at 32 °C. M. xanthus strain DZ2 was
mixed with DZ10547 50:1 in the absence of prey and 200:1 in the presence of
prey. For microscopy, cover slips were added to the slides to prevent evapo-
Fluorescence images were captured using a Leica microscope with the appro-
for a phase contrast image. The images were captured by a Hamamatsu Orca
C9495 camera and analyzed with QCapture Pro and Microsoft Excel software.
calculating net cell displacement, the coordinate positions of cells were mea-
sured after 30 min and were collected for cells that do not leave the field of
view. Data were pooled from four samples from independent cultures. Cell
velocities were calculated on the basis of cell displacement changes in 1-min
intervals. In total ?100 cells from three independent samples were analyzed
for each condition.
Quantitative Analysis of Rippling Behavior. Cultures of M. xanthus and E. coli
(pH 7.6) buffer. Serial dilutions of E. coli cultures were prepared in twofold
steps ranging from 2 ? 106to 6 ? 107. Serial dilutions of M. xanthus cultures
were prepared in fourfold steps ranging from 6 ? 102to 4 ? 107. Three
incubated at 32 °C for 40 h and rippling behavior was observed using a Nikon
SMZ1500 stereomicroscope. The rippling wavelength for each sample was
determined by measuring eight consecutive waves in three separate areas of
using a QImaging MicroPublisher 5.0 RTV CCD camera and QCapture Pro
software. Images were compiled into movies with Microsoft Moviemaker.
Immunoblotting. Overnight cultures of M. xanthus were harvested and an ali-
quot was pelleted and saved at ?20 °C as a control. The remaining cells were
xanthus cells was pipetted in the center of a 40 ? 2.5 mm strip of E. coli ?2155
The arrows indicate when rippling behavior begins. (A) Assay diagram, (B) M.
xanthus strain DZ2, (C) ?frzG, and (D) ?frzF.
Tactic assay of predatory behavior. A 1-?l aliquot containing 106M.
Berleman et al.
November 4, 2008 ?
vol. 105 ?
no. 44 ?
washed twice and 108M. xanthus cells were spread on CFL plates either with or
without 109E. coli strain ?2155 cells. Plates were incubated for 24 h at 32 °C.
resuspended in TE buffer and sonicated. The soluble protein fraction was nor-
and detected using anti-FrzCD antibody (primary) used at 1/10,000 and anti-IgG
(rabbit)-HRP conjugate (secondary) at 1/15,000, similar to previously described
pipetted in 2 ?l adjacent aliquots to generate a thin strip of prey with
dimensions of 20 mm ? 2.5 mm on a CFL agar plate. A 1-?l aliquot
containing ?106M. xanthus cells was pipetted in the center of the prey
long axis of the E. coli strip is treated as the x-axis and the short axis as the
y-axis. Measurements were collected from three independent samples for
ACKNOWLEDGMENTS. We thank D. Zusman for providing us with strains
and antibodies, and S.J. Arends for helpful advice regarding microscopy
and Excel. We also thank E. Mauriello and Kirby laboratory members for
discussion and manuscript comments. This work was supported by National
Institutes of Health Grants T32 AI007511 (to J.B.) and AI59682 (to J.K.).
1. King N (2004) The unicellular ancestry of animal development. Dev Cell 7:313–325.
2. Aguilar C, Vlamakis H, Losick R, Kolter R (2007) Thinking about Bacillus subtilis as a
multicellular organism. Curr Opin Microbiol 10:638–643.
3. Camilli A, Bassler BL (2006) Bacterial small-molecule signaling pathways. Science
4. Zusman DR, Scott AE, Yang Z, Kirby JR (2007) Chemosensory pathways, motility and
development in Myxococcus xanthus. Nat Rev Microbiol 5:862–872.
5. Mignot T, Merlie JP, Jr, Zusman DR (2005) Regulated pole-to-pole oscillations of a
bacterial gliding motility protein. Science 310:855–857.
6. Mignot T, Shaevitz JW, Hartzell PL, Zusman DR (2007) Evidence that focal adhesion
complexes power bacterial gliding motility. Science 315:853–856.
7. Sliusarenko O, Zusman DR, Oster G (2007) The motors powering A-motility in Myxo-
coccus xanthus are distributed along the cell body. J Bacteriol 189:7920–7921.
8. Berleman JE, Kirby JR (2007) Multicellular development in Myxococcus xanthus is
stimulated by predator-prey interactions. J Bacteriol 189:5675–5682.
9. Igoshin OA, Mogilner A, Welch RD, Kaiser D, Oster G (2001) Pattern formation and
traveling waves in myxobacteria: Theory and modeling. Proc Natl Acad Sci USA
10. Sliusarenko O, Neu J, Zusman DR, Oster G (2006) Accordion waves in Myxococcus
xanthus. Proc Natl Acad Sci USA 103:1534–1539.
11. Stevens A, Sogaard-Andersen L (2005) Making waves: Pattern formation by a cell-
surface-associated signal. Trends Microbiol 13:249–252.
12. Kuner JM, Kaiser D (1982) Fruiting body morphogenesis in submerged cultures of
Myxococcus xanthus. J Bacteriol 151:458–461.
13. Shimkets LJ, Kaiser D (1982) Induction of coordinated movement of Myxococcus
xanthus cells. J Bacteriol 152:451–461.
14. Berleman JE, Chumley T, Cheung P, Kirby JR (2006) Rippling is a predatory behavior in
Myxococcus xanthus. J Bacteriol 188:5888–5895.
15. Baker MD, Wolanin PM, Stock JB (2006) Signal transduction in bacterial chemotaxis.
17. Stecher B, et al. (2004) Flagella and chemotaxis are required for efficient induction of
Salmonella enterica serovar Typhimurium colitis in streptomycin-pretreated mice.
Infect Immun 72:4138–4150.
18. Budrene EO, Berg HC (1991) Complex patterns formed by motile cells of Escherichia
coli. Nature 349:630–633.
19. Budrene EO, Berg HC (1995) Dynamics of formation of symmetrical patterns by
chemotactic bacteria. Nature 376:49–53.
20. Francis NR, Wolanin PM, Stock JB, Derosier DJ, Thomas DR (2004) Three-dimensional
structure and organization of a receptor/signaling complex. Proc Natl Acad Sci USA
21. McBride MJ, Weinberg RA, Zusman DR (1989) ‘‘Frizzy’’ aggregation genes of the
gliding bacterium Myxococcus xanthus show sequence similarities to the chemotaxis
genes of enteric bacteria. Proc Natl Acad Sci USA 86:424–428.
22. Sager B, Kaiser D (1994) Intercellular C-signaling and the traveling waves of Myxococ-
cus. Genes Dev 8:2793–2804.
23. Hazelbauer GL, Falke JJ, Parkinson JS (2008) Bacterial chemoreceptors: High-
performance signaling in networked arrays. Trends Biochem Sci 33:9–19.
24. Bustamante VH, Martinez-Flores I, Vlamakis HC, Zusman DR (2004) Analysis of the Frz
signal transduction system of Myxococcus xanthus shows the importance of the
conserved C-terminal region of the cytoplasmic chemoreceptor FrzCD in sensing
signals. Mol Microbiol 53:1501–1513.
25. DeFranco AL, Koshland DE, Jr (1980) Multiple methylation in processing of sensory
signals during bacterial chemotaxis. Proc Natl Acad Sci USA 77:2429–2433.
26. McCleary WR, McBride MJ, Zusman DR (1990) Developmental sensory transduction in
Myxococcus xanthus involves methylation and demethylation of FrzCD. J Bacteriol
27. Parkinson JS (1976) cheA, cheB, and cheC genes of Escherichia coli and their role in
chemotaxis. J Bacteriol 126:758–770.
28. Yonekawa H, Hayashi H, Parkinson JS (1983) Requirement of the cheB function for
sensory adaptation in Escherichia coli. J Bacteriol 156:1228–1235.
29. Hickman JW, Tifrea DF, Harwood CS (2005) A chemosensory system that regulates
biofilm formation through modulation of cyclic diguanylate levels. Proc Natl Acad Sci
30. Berleman JE, Bauer CE (2005) A che-like signal transduction cascade involved in
controlling flagella biosynthesis in Rhodospirillum centenum. Mol Microbiol 55:1390–
31. Kirby JR, Zusman DR (2003) Chemosensory regulation of developmental gene expres-
sion in Myxococcus xanthus. Proc Natl Acad Sci USA 100:2008–2013.
32. Kearns DB, Shimkets LJ (2001) Lipid chemotaxis and signal transduction in Myxococcus
xanthus. Trends Microbiol 9:126–129.
33. Shi W, Zusman DR (1994) Sensory adaptation during negative chemotaxis in Myxococ-
cus xanthus. J Bacteriol 176:1517–1520.
34. Shi W, Zusman DR (1994) Sensor/response in Myxococcus xanthus to attractants and
repellents requires the frz signal transduction system. Res Microbiol 145:431–435.
35. Xu Q, Black WP, Mauriello EM, Zusman DR, Yang Z (2007) Chemotaxis mediated by
NarX-FrzCD chimeras and nonadapting repellent responses in Myxococcus xanthus.
Mol Microbiol 66:1370–1381.
36. Curtis PD, Geyer R, White DC, Shimkets LJ (2006) Novel lipids in Myxococcus xanthus
and their role in chemotaxis. Environ Microbiol 8:1935–1949.
37. Rosenberg E, Keller KH, Dworkin M (1977) Cell density-dependent growth of Myxo-
coccus xanthus on casein. J Bacteriol 129:770–777.
38. Nudleman E, Wall D, Kaiser D (2005) Cell-to-cell transfer of bacterial outer membrane
lipoproteins. Science 309:125–127.
Natl Acad Sci USA 98:14907–14912.
40. Dehio C, Meyer M (1997) Maintenance of broad-host-range incompatibility group P
and group Q plasmids and transposition of Tn5 in Bartonella henselae following
conjugal plasmid transfer from Escherichia coli. J Bacteriol 179:538–540.
41. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual
(Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY).
www.pnas.org?cgi?doi?10.1073?pnas.0804387105 Berleman et al.