The Xanthomonas axonopodis pv. citri flagellum is required for mature biofilm and canker development.

TLDR: Confocal laser scanning microscopy of biofilms produced in static culture demonstrated that the flagellum is also involved in the formation of mushroom-shaped structures and water channels, and in the dispersion ofBiofilms, indicating a regulatory role for DSF in both types of motility.

Abstract: Xanthomonas axonopodis pv. citri (Xac) is the causative agent of citrus canker. This bacterium develops a characteristic biofilm on both biotic and abiotic surfaces. To evaluate the participation of the single flagellum of Xac in biofilm formation, mutants in the fliC (flagellin) and the flgE (hook) genes were generated. Swimming motility, assessed on 0.25 % agar plates, was markedly reduced in fliC and flgE mutants. However, the fliC and flgE mutants exhibited a flagellar-independent surface translocation on 0.5 % agar plates. Mutation of either the rpfF or the rpfC gene, which both encode proteins involved in cell-cell signalling mediated by diffusible signal factor (DSF), led to a reduction in both flagellar-dependent and flagellar-independent surface translocation, indicating a regulatory role for DSF in both types of motility. Confocal laser scanning microscopy of biofilms produced in static culture demonstrated that the flagellum is also involved in the formation of mushroom-shaped structures and water channels, and in the dispersion of biofilms. The presence of the flagellum was required for mature biofilm development on lemon leaf surfaces. The absence of flagellin produced a slight reduction in Xac pathogenicity and this reduction was more severe when the complete flagellum structure was absent.

Figures

Fig. 3. Adhesion and biofilm development of Xac strains in vitro. (a) Adhesion in a 96-well polystyrene plate. Aliquots (150 ml) of overnight cultures were used to inoculate the plates. OD600 measurements and crystal violet (CV) assays were performed at 1, 3, 6 and 24 h after inoculation. (b) Biofilm development on wells of a 96-well polystyrene plate. Overnight cultures in nutrient medium were diluted in YMM to final OD600 of 0.1. The OD600 measurements and CV assays were performed at 24, 48 and 72 h after inoculation. Three independent experiments with six replicates for each strain were performed. Error bars represent SEM.

Table 1. Strains, plasmids and primers used in this study

Fig. 5. Canker development, bacterial growth and biofilm formation on the surface of leaves infected with Xac strains by the spray method. (a) Lemon leaves at 22 days p.i. with 1 105 or 1 104 c.f.u. ml”1. (b) Quantification of canker lesions in lemon leaves at 22 days p.i. Twenty leaves were assayed and the results are the means of four independent experiments±SD. The OD600 of the suspensions used for inoculation is indicated above the bars. Asterisks indicate a statistically significant difference between the flgE mutant and the wild-type at both concentrations of bacterial suspension. (c) Growth in vivo of Xac strains on C. limon leaves infected with 1 106 c.f.u. ml”1. Values are means of three independent experiments±SD. Datasets marked with an asterisk are significantly different according to the two-tailed Student’s t-test (P,0.05). (d) Biofilm structure analysis at 6 days p.i. using CLSM and GFP-labelled cells at a concentration of 1 106 c.f.u. ml”1. The panels show projections in the x–y and x–z planes with a magnification of 100. Red indicates chlorophyll autofluorescence. Three independent experiments were assayed with essentially the same outcome.

Fig. 1. Effects of mutation of the fliC, flgE and gumB genes on motility in Xac. (a) Flagellum-dependent swimming motility on 0.25 % NYGB agar plates. (b) Motility on 0.5 % PYM agar plates with 1 % glucose.

Fig. 2. Motility and gene expression in Xac strains with disruptions in the Rpf/DSF signalling system. (a) Flagellum-dependent swimming motility on 0.25 % agar NYGB plates. (b) Expression of the fliC gene assessed by semi-quantitative RT-PCR using cDNA from wild-type Xac, and rpfC and rpfF mutants grown in YMM medium to stationary phase (n53). (c) Motility on 0.5 % PYM agar plates with 1 % glucose. (d) Expression of the gumB gene assessed by semi-quantitative RT-PCR in wild-type Xac, and rpfC and rpfF mutants grown in YMM medium to stationary phase (n53).

Full text

The Xanthomonas axonopodis pv. citri flagellum is
required for mature biofilm and canker development
Florencia Malamud,
1
Pablo S. Torres,
1
Roxana Roeschlin,
2
Luciano A. Rigano,
1
Ramo
´
n Enrique,
2
Herna
´
n R. Bonomi,
3
Atilio P. Castagnaro,
4
Marı´a Rosa Marano
2
and Adria
´
n A. Vojnov
1
Correspondence
Adria
´
n A. Vojnov
avojnov@fundacioncassara.org.ar
Received 3 August 2010
Revised 17 November 2010
Accepted 19 November 2010
1
Instituto de Ciencia y Tecnologı´a Dr. Cesar Milstein, Fundacio
´
n Pablo Cassara
´
, CONICET, Saladillo
2468 C1440FFX, Ciudad de Buenos Aires, Argentina
2
IBR – Depto. Microbiologı´a, Facultad de Ciencias, Bioquı´micas y Farmace
´
uticas, U.N.R. Suipacha
531, S2002LRK, Rosario, Argentina
3
Fundacio
´
n Instituto Leloir-CONICET, Av. Patricias Argentinas 435 C1405BWE, Buenos Aires,
Argentina
4
Estacio
´
n Experimental Agroindustrial Obispo Colombres, Av. William Cross 3150, Las Talitas,
Tucuma
´
n, Argentina
Xanthomonas axonopodis pv. citri (Xac) is the causative agent of citrus canker. This bacterium
develops a characteristic biofilm on both biotic and abiotic surfaces. To evaluate the participation
of the single flagellum of Xac in biofilm formation, mutants in the fliC (flagellin) and the flgE (hook)
genes were generated. Swimming motility, assessed on 0.25 % agar plates, was markedly
reduced in fliC and flgE mutants. However, the fliC and flgE mutants exhibited a flagellar-
independent surface translocation on 0.5 % agar plates. Mutation of either the rpfF or the rpfC
gene, which both encode proteins involved in cell–cell signalling mediated by diffusible signal
factor (DSF), led to a reduction in both flagellar-dependent and flagellar-independent surface
translocation, indicating a regulatory role for DSF in both types of motility. Confocal laser scanning
microscopy of biofilms produced in static culture demonstrated that the flagellum is also involved
in the formation of mushroom-shaped structures and water channels, and in the dispersion of
biofilms. The presence of the flagellum was required for mature biofilm development on lemon leaf
surfaces. The absence of flagellin produced a slight reduction in Xac pathogenicity and this
reduction was more severe when the complete flagellum structure was absent.
INTRODUCTION
Biofilms are bacterial communities in which cells are
embedded in an extracellular polysaccharide (EPS) matrix
that can be attached to a surface (Branda et al., 2005;
Southey-Pillig et al., 2005). Life in these communities
provides protection to the organisms from deleterious
conditions (Davey & O’Toole, 2000), and biofilm forma-
tion is considered to be important for the disease cycle of
bacterial pathogens of both animals and plants. We are
interested in understanding the role of biofilm formation
in the development of canker disease by Xanthomonas
axonopodis pv. citri (Xac), one of the most devastating
diseases of citrus species. Xac is a foliar pathogen that
enters the plant leaves through stomata and can infect both
fruits and leaves (Brunings & Gabriel, 2003). Confocal laser
scanning microscopy (CLSM) of citrus cankers, using
Xac bacteria expressing green fluorescent protein (GFP),
showed the occurrence of structured arrangements of cells
(Rigano et al., 2007b). Our aim here was to identify
bacterial factors involved in the development of such three-
dimensional structures.
The development of bacterial biofilms is generally a
multistep process, which is initiated when the bacteria
reach a surface. First, the bacteria attach to a surface
reversibly, where they can move freely across it until they
become immobilized (Stoodley et al., 2002). To achieve a
mature biofilm, new individuals appear in the community
either by recruitment from planktonic bacteria or through
replication of cells already present in the biofilm. Finally,
some grouped cells disperse to develop new structures
elsewhere (Heydorn et al., 2000; Tolker-Nielsen et al.,
Abbreviations: CLSM, confocal laser scanning microscopy; CV, crystal
violet; DSF, diffusible signal factor; EPS, extracellular polysaccharide;
Xac, Xanthomonas axonopodis pv. citri; Xcc, Xanthomonas campestris pv.
campestris.
Two supplementary figures are available with the online version of this
paper.
Microbiology (2011), 157, 819–829 DOI 10.1099/mic.0.044255-0
044255
G
2011 SGM Printed in Great Britain 819

2000). When the bacterial population reaches a certain
density, the biofilm begins to mature through the pro-
duction of an extracellular matrix which contributes greatly
to the final architecture of the community (Branda et al.,
2005).
Flagella have been implicated in surface attachment in
several bacteria (Lemon et al., 2007; O’Toole & Kolter,
1998a, b). For instance, flagella are necessary for swarming
motility in Pseudomonas aeruginosa, which in turn is
important in determining the final structure of the biofilm
(Merritt et al., 2007). Similarly, Escherichia coli is not able
to form an organized structure when flagella are lost. It has
been proposed that E. coli flagella serve to overcome
surface repulsion (Van Houdt & Michiels, 2005) and also
allow attached cells to migrate along the abiotic surface
to facilitate biofilm expansion (Pratt & Kolter, 1998).
However, flagella are important but not essential for
surface attachment in Vibrio cholerae El Tor (Watnick &
Kolter, 1999).
Previous work from our laboratory has established that
synthesis of the EPS xanthan contributes to the formation of
structured biofilms in Xac and that both synthesis of
xanthan and cell–cell signalling involving diffusible signal
factor (DSF) contribute to structured biofilm formation in
the related Xanthomonas campestris pv. campestris (Xcc)
(Rigano et al., 2007a; Torres et al., 2007). Synthesis and
perception of DSF require proteins coded in the rpf cluster
(for
regulation of pathogenicity factors). RpfF directs DSF
synthesis whereas RpfC is involved in DSF perception
(Barber et al., 1997; Slater et al., 2000). Mutants in the rpfF
(DSF-minus) and rpfC (DSF overproducer) genes in Xcc can
only form unstructured arrangements of bacteria (Torres
et al., 2007). Synthesis of xanthan is directed by genes from
the gum operon (da Silva et al., 2002), which is highly
conserved in Xanthomonas spp. Disruption of the first gene,
gumB, leads to complete loss of xanthan production
(Vojnov et al., 1998); strains of Xac or Xcc with mutations
in gumB cannot form structured biofilms (Rigano et al.,
2007b; Torres et al., 2007). Beyond DSF signalling and
xanthan production, relatively little is known about factors
influencing biofilm formation in Xanthomonas spp. and the
role that the flagellum plays in this process.
Both Xcc and Xac bear a single polar flagellum. Flagellar
gene clusters encode all the structural proteins of the
flagellum in Xac (da Silva et al., 2002). In addition, it has
been shown that several genes encoding auxiliary protein
subunits are involved in regulation of the flagellar assembly
(Khater et al., 2007). As in many other bacterial species,
FliC and FlgE proteins are the flagellin (Vonderviszt et al.,
1998) and the hook components (Aizawa, 1996), respect-
ively, of the flagellar structure. This structure is formed by
a thin filament that protrudes from the cell body bound to
a basal body that includes a rotor and a stator (Chevance &
Hughes, 2008).
Here we address the role of the single flagellum in biofilm
formation in Xac. Our initial approach was to test the
effects of mutations in fliC and flgE on motility and biofilm
structure. Our results indicate that the flagellum and
flagellum-dependent motility are important for initial
adherence to surfaces, for the development of a mature
structured biofilm and for biofilm dispersion. Unexpectedly,
our observations reveal a second type of motility that is
flagellum-independent and EPS-dependent. Both types of
motility are regulated by the DSF signalling system. Thus,
these findings indicate possible additional roles for cell–cell
signalling and EPS in the biofilm formation process.
METHODS
Bacterial strains. Xanthomonas strains were cultured at 28 uC with
shaking in PYM (Cadmus et al., 1976) or in Y minimal medium
(YMM) (Sherwood, 1970). To examine biofilm development, bacteria
were grown in YMM containing 1 % (w/v) glucose as the carbon source
(Rigano et al., 2007b). E. coli was grown at 37 uC in Luria–Bertani
medium (Sambrook et al., 1989). Bacterial growth was measured in a
Spectronic 20 Genesys spectrophotometer (Thermo Electron) at
600 nm. When required, the antibiotics ampicillin (100
mgml
21
),
kanamycin (50
mgml
21
), spectinomycin (100 mgml
21
), tetracycline
(10
mgml
21
) or gentamicin (20 mgml
21
) were added to the growth
media.
Construction of fliC and flgE mutants. Molecular techniques and
extraction of genomic DNA from Xac were performed according to
Rigano et al. (2007b). To obtain the Xac
DfliC mutant strain (fliC), the
fliC gene was partially deleted. First, two fragments of 408 and 484 bp
were amplified from the region encoding the wild-type Xac fliD and
fliC genes by using primers flic1/flic2 and flic3/flic4, respectively
(Table 1). The fragments were digested with restriction enzyme XbaI,
purified and ligated to one another. A PCR was then performed using
this ligation as template and primers flic1/flic4. The PCR product was
resolved by 0.8 % agarose gel electrophoresis and a fragment of
892 bp was excised and purified from the gel using the Wizard SV Gel
and PCR Clean-Up System (Promega). The product was cloned into
pGEM-T easy (Promega) and subcloned into the suicide vector
pk18mobsacB (Katzen et al., 1999), producing the pK-
DfliC construct.
The Xac flgE ::
V strain (flgE) was constructed by allelic exchange.
Two regions, upstream and downstream from flgE, were amplified
using flge1/flge2 (546 bp) and flge3/flge4 (547 bp) primers, respec-
tively (Table 1). The two fragments were digested with HindIII,
purified and ligated to one another, and a PCR was then performed
using this ligation as template and primers flge1/flge4. The PCR
product of 1093 bp was resolved and purified as described above. The
product was cloned into pGEM-T easy and the 2 kb Sm
r
/Spc
r
cassette
(
V) was inserted in the HindIII restriction site. This construct was
subcloned into pk18mobsacB, giving pK-flgE ::
V.
The two constructs, pK-
DfliC and pK-flgE ::V, were used to transform
the wild-type strain by electroporation (do Amaral et al., 2005).
Transformed bacteria were selected on PYM agar medium supple-
mented with kanamycin (Mu
¨
ller et al., 1993). Sucrose-sensitive clones
were grown in the absence of antibiotics and double recombination
events were selected on PYM agar plates supplemented with 5 %
sucrose and spectinomycin in the case of the flgE mutant. Km
s
/
sucrose
r
candidates were screened by PCR followed by sequencing of
the amplified fragments. Isolated mutants were confirmed by
flagellum staining as described by Kearns & Losick (2003) and
motility was further analysed.
For genetic complementation of the fliC mutant, the entire gene and
245 bp from the upstream region were amplified by PCR using
F. Malamud and others
820 Microbiology 157

primers c-Flic-sense and Flic4 (Table 1). The amplified fragment was
cloned into pLAFR3 (Staskawicz et al., 1987) to obtain pLAFR-fliC.
Similarly, the flgE gene was amplified using primers c-Flge-sense and
c-Flge-antisense (Table 1); the amplicon was cloned into pLAFR3 to
obtain pLAFR-flgE. Both constructs were confirmed by DNA
sequencing and electroporated into fliC and flgE mutant strains.
Bacterial adhesion and biofilm quantification. The crystal violet
(CV) technique was used to quantify biofilm development of the
different strains on an abiotic surface (O’Toole & Kolter, 1998b).
Bacterial strains were grown overnight in PYM nutrient medium and
then inoculated into YMM to a final OD
600
of 0.1. Aliquots of 150 ml
were used to fill the wells of a 96-well polystyrene plate and incubated
at 28 uC for 24, 48 or 72 h. To confirm similar bacterial growth, the
OD
600
was measured before the adhesion assay was performed.
To analyse the capacity of the strains to adhere to the abiotic surface,
bacteria from an overnight culture growth in YMM at 28 uC and
200 r.p.m. were used. Aliquots of 150
ml were dispensed into the wells
of a 96-well plate and bacteria were incubated as previously described
for 1, 3, 6 and 24 h. The OD
600
and cell attachment were measured as
described above.
For quantification of biofilm development and adhesion of cells to an
abiotic surface, the medium was gently removed using a pipette, and
the 96-well plate was washed using 0.9 % NaCl and stained with 0.1 %
CV solution. After 30 min incubation the unbound CV stain was
Table 1. Strains, plasmids and primers used in this study
Strain, plasmid or primer Relevant genotype or sequence (5§–3§) Source or reference
Strains
Xac Wild-type, Ap
r
Rigano et al. (2007b)
Xac GFP Xac,Km
r
Ap
r
This study
fliC Xac
Dflic,Ap
r
This study
flgE Xac flge ::V,Ap
r
Sp
r
This study
fliC-GFP Xac
Dflic,Ap
r
Km
r
This study
flgE-GFP Xac flge ::
V,Ap
r
Sp
r
This study
c-fliC Xac
Dflic,Ap
r
Tc
r
This study
c-flgE Xac flge ::
V,Ap
r
Sp
r
Tc
r
This study
rpfF Xac
DrpfF,Km
r
Ap
r
Siciliano et al. (2006)
rpfC Xac
DrpfC,Km
r
Ap
r
Siciliano et al. (2006)
gumB Xac
Dgumb,Km
r
Ap
r
Rigano et al. (2007b)
E. coli DH5
a hsdR recA lacZYA w80 lacZDM15 Gibco-BRL
Plasmids
pGem-Teasy Ap
r
, lacZ Promega
Pk mobsacB Km
r
, sacB mob lacZ Katzen et al. (1999)
pBBR2-GFP Km
r
Rigano et al. (2007b)
pLAFR 3 Tc
r
Staskawicz et al. (1987)
pK-
DfliC Km
r
This study
pK-flgE ::
V Km
r
Sp
r
This study
pLAFR-fliC Tc
r
This study
pLAFR-flgE Tc
r
This study
Primers
Flic1 GC
GAATTCGCCTTGTTGATGCGTGCCTGD This study
Flic2 GC
TCTAGATTCCGCAGAACGTGCTGAGCd This study
Flic3 GC
TCTAGAGAGATACCGTCGTTGGCGTTd This study
Flic4 GC
GGATCCCAGGCGGACGGAGTTTATTT§ This study
Flge1 GGGAATCGCAAAAGCGGGAT This study
Flge2
AAGCTTGCAGCTCAACGTCACTGGCT|| This study
Flge3
AAGCTTCGTAGACATTGATGCCGCCG|| This study
Flge4 CGCAGCCAATGCCGATCTGA This study
c-Flic-sense GC
GAATTCCATTGCTGCGGCAGGTAACTD This study
c-Flge-sense GC
GAATTCGATGATCGTTTGAGTGACCTGGTCCD This study
c-Flge-antisense GC
GGATCCCCTCACCAACGGAACACTCCACAT§ This study
RT-flic sense CAGCTTGGTGCCGTTGAAGTT This study
RT-flic antisense TGAACGCTCAGCGGAACCTCA This study
RT-gumb sense AAACACGATGACATTGCCGC This study
RT-gumb antisense GCCATATTTCGTTGCCGCTC This study
RT-16S sense TGGTAGTCCACGCCCTAAACG This study
RT-16S antisense CTGGAAAGTTCCGTGGATGTC This study
*Km
r
,Ap
r
,Gm
r
,Sp
r
and Tc
r
indicate kanamycin, ampicillin, gentamicin, spectinomycin and tetracycline resistance, respectively. Sites for restriction
enzymes are underlined: DEcoRI, dXbaI, §BamHI, ||HindIII.
Flagellum and biofilm formation in Xanthomonas
http://mic.sgmjournals.org 821

removed and the wells were washed with distilled water. The CV in
each well was solubilized by adding 150
ml 70 % ethanol and was
quantified by absorbance at 570 nm.
The adhesion value was normalized according to the number of cells.
This value was termed relative adhesion (A
570
/OD
600
).
In vitro and in vivo analysis of biofilm by CLSM. For in vitro
experiments each strain expressing the autofluorescent protein GFP
was grown at 28 uC on PYM supplemented with kanamycin. The
OD
600
was adjusted to 0.004 in YMM. Aliquots of 500 ml were
transferred to chambered coverglass slides containing a 1
mm thick
borosilicate glass (no. 155411; Lab-Tek, Nunc), as described by Russo
et al. (2006).
Biofilm formation was monitored with a Zeiss LSM 510/Axiovert
100M confocal laser scanning microscope at 1, 2, 3, 4 and 5 days post-
inoculation (days p.i.) of the bacteria on the chambered coverglass
slides. The assay was performed in triplicate and three-dimensional
images were generated by the program Zeiss LSM Image Browser,
version 3.2.0.
For in vivo studies, leaves of lemon (Citrus limon) were infected by
spraying with a 1610
6
c.f.u. ml
21
suspension of GFP-labelled wild-
type Xac or the fliC or flgE mutants. After 6 days of incubation, areas
of approximately 1 cm
2
were cut from the leaves and mounted on the
adaxial leaf surface under glass coverslips. The samples were observed
with a Nikon C1 confocal laser scanning microscope. The simulated
images obtained were analysed with the program Nikon EZ-C1.
Assays were performed in triplicate.
Quantification of biofilm structures. Quantifications of biofilm
volume and thickness of three-dimensional CLSM biofilm image
stacks were performed by using the
COMSTAT program (Heydorn et al.,
2000). For comparative analysis of the wild-type, fliC and flgE mutant
strains, three independent biofilm experiments were performed. In
each round, seven image stacks were acquired at the following time
points: 1, 2, 3, 4 and 5 days p.i.
RNA purification and RT-PCR. Total RNA of Xac cells was isolated
by using an RBC kit, according to the manufacturer’s instructions
(RBCBioscience). The RNA was treated with RNase-free DNase
(Promega) and the integrity of the nucleic acid was checked by
agarose gel electrophoresis. Total RNA (1
mg) was used to synthesize
the cDNA using MMLV-RT (Promega) and the oligonucleotide dN6.
Expression of the gumB, fliC and 16S genes was determined by RT-
PCR using the primers listed in Table 1.
The density corresponding to the RT-PCR bands was quantified by
using ImageJ 1.41 software from the National Institutes of Health
(http://rsbweb.nih.gov/ij/download.html). Each gene was normalized
using 16S rRNA as the housekeeping gene. The ratio between
normalized fliC or gumB in rpfC and rpfF mutants and in wild-type
Xac was determined. Data are expressed as the mean±
SEM (n53).
Motility assays. Motility assays were performed as described by
Rashid & Kornberg (2000). Briefly, bacteria were grown overnight in
PYM medium, then 3
ml of bacterial cultures with normalized OD
600
were used to inoculate plates of 0.25 % agar NYGB medium (Barber
et al., 1997) (swimming) and 0.5 % agar PYM. After 72 h, motility
was assessed qualitatively by examining the circular halo formed by
the growing bacterial cells. The assays were performed in triplicate.
Flagellar staining with CV. Flagellar staining was done based on a
modification of the method described by Mayfield & Inniss (1977).
The staining solution was prepared by mixing 10 vols solution A (2 g
tannic acid, 10 ml 5 % phenol, 10 ml AlKO
8
S
2
.12H
2
O) with 1 vol.
12 % (w/v) CV in ethanol. Then 3
ml of bacterial culture grown
overnight in PYM (OD
600
2) was placed over a coverglass with a
coverslip and 10
ml of the solution was used. Observations were made
with an optical microscope (Carl Zeiss).
Fig. 1. Effects of mutation of the fliC, flgE and gumB genes on motility in Xac. (a) Flagellum-dependent swimming motility on
0.25 % NYGB agar plates. (b) Motility on 0.5 % PYM agar plates with 1 % glucose.
F. Malamud and others
822 Microbiology 157

Plant material and inoculations. Citrus limon ‘Eureka’ was used as
the host plant for Xac. Plants were kept in a controlled temperature
room at 20–25 uC and with a 16 h photoperiod. Bacteria were grown
in PYM with the appropriate antibiotics. Bacterial suspensions were
diluted in 10 mM MgCl
2
to a final concentration of 1610
4
,1610
5
or
1610
6
c.f.u. ml
21
and were inoculated by spraying as previously
described (Rigano et al., 2007b). Bacterial growth in the host plant
was quantified as previously described (Rigano et al., 2007b), using
1610
6
c.f.u. ml
21
to spray the leaves. Statistical significance of
differences between bacterial populations was determined by
Student’s unpaired two-tailed t-test. P-values ,0.05 were considered
statistically relevant. Cankers from 20 inoculated leaves were
quantified and their areas were calculated using the program
ImageJ version 1.41.
RESULTS
FliC and FlgE are required for swimming motility
The genomic arrangement near the fliC and flgE genes in
Xac is shown in Supplementary Fig. S1(a), available with
the online version of this paper. Xac strains with either a
fliC deletion or a disruption in the flgE gene with a
spectinomycin resistance cassette were created as described
in Methods (Supplementary Fig. S1b). The mutations were
confirmed by PCR (Supplementary Fig. S1c) and DNA
sequence analysis (data not shown). The absence of the
flagellar structure in these mutants was confirmed by CV
staining (Kearns & Losick, 2003). A typical single polar
flagellum was observed in wild-type Xac, but was absent in
the two mutants (Supplementary Fig. S1d). This obser-
vation confirms the expected and fundamental role for FliC
and FlgE in the morphogenesis of the flagellum in Xac.
The ability of the different Xac strains to swim in low-
percentage agar was assayed to evaluate the role of the
flagellum in motility. NYGB 0.25 % agar plates were
inoculated with the wild-type strain and the fliC and flgE
mutants. After incubation at 28
u
C for 72 h, both mutants
showed a considerable reduction in their motility com-
pared with the wild-type (Fig. 1a). This was expected
because the same phenotype has been reported for these
kinds of mutations in other bacteria (Murray &
Kazmierczak, 2006; Young et al., 1999). The complemented
mutant strains showed a partial restoration of their
motility, confirming that no other genes had been affected
in the mutants (Supplementary Fig. S2).
Fig. 2. Motility and gene expression in Xac strains with disruptions in the Rpf/DSF signalling system. (a) Flagellum-dependent
swimming motility on 0.25 % agar NYGB plates. (b) Expression of the fliC gene assessed by semi-quantitative RT-PCR using
cDNA from wild-type Xac, and rpfC and rpfF mutants grown in YMM medium to stationary phase ( n53). (c) Motility on 0.5 %
PYM agar plates with 1 % glucose. (d) Expression of the gumB gene assessed by semi-quantitative RT-PCR in wild-type Xac,
and rpfC and rpfF mutants grown in YMM medium to stationary phase (n53).
Flagellum and biofilm formation in Xanthomonas
http://mic.sgmjournals.org 823

Frequently asked questions

Q1. What is the role of EPS in the formation of microcolonies?

Bacterial attachment, the formation of aggregates or microcolonies, and finally biofilm maturation have been recognized as the three stages of biofilm development that are responsive to diverse environmental signals (Stanley & Lazazzera, 2004). 

Q2. What is the role of rpfF in Xcc?

Mutants in the rpfF (DSF-minus) and rpfC (DSF overproducer) genes in Xcc can only form unstructured arrangements of bacteria (Torres et al., 2007). 

Q3. What is the role of flagella in the development of biofilms?

For instance, flagella are necessary for swarming motility in Pseudomonas aeruginosa, which in turn is important in determining the final structure of the biofilm (Merritt et al., 2007). 

Q4. What is the role of flagella in Xanthomonas spp?

It has been proposed that E. coli flagella serve to overcome surface repulsion (Van Houdt & Michiels, 2005) and also allow attached cells to migrate along the abiotic surface to facilitate biofilm expansion (Pratt & Kolter, 1998). 

Q5. What was the technique used to quantify biofilm development?

The crystal violet (CV) technique was used to quantify biofilm development of the different strains on an abiotic surface (O’Toole & Kolter, 1998b). 

Q6. What is the role of EPS in the sliding motility?

It is not known how DSF signalling and EPS impinge on the proposed sliding motility, although it is possible that EPS plays a role in modifying the properties of the agar surface to promote motility. 

Q7. What was the effect of the fliC mutant on the growth of lemon leaves?

biofilms from the fliC and flgE mutants developed in a more dispersed fashion with either few or no aggregated structures. 

Q8. What was the effect of the complemented mutants on the morphogenesis of the flag?

The complemented mutant strains showed a partial restoration of their motility, confirming that no other genes had been affected in the mutants (Supplementary Fig. S2). 

Q9. What is the conserved gene in Xanthomonas spp?

Synthesis of xanthan is directed by genes from the gum operon (da Silva et al., 2002), which is highly conserved in Xanthomonas spp. 

Q10. What is the role of xanthan in the formation of biofilms?

Previous work from their laboratory has established that synthesis of the EPS xanthan contributes to the formation of structured biofilms in Xac and that both synthesis of xanthan and cell–cell signalling involving diffusible signal factor (DSF) contribute to structured biofilm formation in the related Xanthomonas campestris pv. campestris (Xcc) (Rigano et al., 2007a; Torres et al., 2007). 

Q11. What was the flgE gene amplified using?

the flgE gene was amplified using primers c-Flge-sense and c-Flge-antisense (Table 1); the amplicon was cloned into pLAFR3 to obtain pLAFR-flgE.