Content uploaded by Xiaochen Yuan
Author content
All content in this area was uploaded by Xiaochen Yuan on Sep 19, 2022
Content may be subject to copyright.
RESEARCH ARTICLE
Exopolysaccharides amylovoran and levan contribute to
sliding motility in the fire blight pathogen Erwinia
amylovora
Xiaochen Yuan | Lauren I. Eldred | George W. Sundin
Department of Plant, Soil, and Microbial
Sciences, Michigan State University, East
Lansing, Michigan, USA
Correspondence
George W. Sundin, Department of Plant, Soil,
and Microbial Sciences, Michigan State
University, East Lansing, Michigan 48824,
USA.
Email: sundin@msu.edu
Funding information
National Institute for Food and Agriculture,
Grant/Award Numbers: 2015-67013-23068,
2020-51181-32158; Michigan State University
Summary
Erwinia amylovora, the causative agent of fire blight, uses flagella-based
motilities to translocate to host plant natural openings; however, little is
known about how this bacterium migrates systemically in the apoplast.
Here, we reveal a novel surface motility mechanism, defined as sliding, in
E. amylovora. Deletion of flagella assembly genes did not affect this move-
ment, whereas deletion of biosynthesis genes for the exopolysaccharides
(EPSs) amylovoran and levan resulted in non-sliding phenotypes. Since
EPS production generates osmotic pressure that potentially powers sliding,
we validated this mechanism by demonstrating that water potential posi-
tively contributes to sliding. In addition, no sliding was observed when the
water potential of the surface was lower than 0.5 MPa. Sliding is a passive
motility mechanism. We further show that the force of gravity plays a critical
role in directing E. amylovora sliding on unconfined surfaces but has a negli-
gible effect when cells are sliding in confined microcapillaries, in which
EPS-dependent osmotic pressure acts as the main force. Although amylo-
voran and levan are both required for sliding, we demonstrate that they
exhibit different roles in bacterial communities. In summary, our study pro-
vides fundamental knowledge for a better understanding of mechanisms
that drive bacterial sliding motility.
INTRODUCTION
Bacterial cells have evolved a range of active and pas-
sive motility mechanisms that enable migration under
diverse environmental conditions including in liquids and
on semisolid and solid surfaces (Henrichsen, 1972).
Active motility mechanisms include swimming, swarm-
ing, twitching, and gliding motilities. Swimming occurs
inside the medium, whereas the other forms of motility
occur on the surface. In general, swimming and
swarming are dependent on flagella, twitching is pow-
ered by type IV pili, and gliding motility is supported by
several different types of motors depending on bacterial
species (Harshey, 2003; Kearns, 2010; Mattick, 2002;
McBride, 2001). Unlike active motility mechanisms, pas-
sive motility mechanisms such as sliding motility are not
well characterized but have attracted attention in recent
years (Henrichsen, 1972; Hölscher & Kov
acs, 2017;
Kearns, 2010; Murray & Kazmierczak, 2008).
Exopolysaccharides (EPSs) are carbohydrate poly-
mers secreted by bacteria that can be found either as
capsules attached to the cell wall or as secreted slime
in the surrounding environment (Leigh & Coplin, 1992;
Sutherland, 1982; Whitfield et al., 2020). Mostly studied
as the main components of the extracellular matrix of
bacterial biofilms (Limoli et al., 2015; Sutherland, 2001),
EPSs are also involved in the passive spreading of bac-
teria (Hölscher & Kov
acs, 2017; Nogales et al., 2012).
Received: 27 May 2022 Accepted: 31 August 2022
DOI: 10.1111/1462-2920.16193
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2022 The Authors. Environmental Microbiology published by Society for Applied Microbiology and John Wiley & Sons Ltd.
Environ Microbiol. 2022;1–17. wileyonlinelibrary.com/journal/emi 1
In Bacillus subtilis, Seminara et al. (2012) proposed a
model in which osmotic pressure caused by the secre-
tion of EPS drives sliding. EPS production causes an
increase in osmotic pressure between the bacterial bio-
film and the external environment, which promotes bio-
film expansion likely through uptake of water from the
external environment. Yan et al. (2017) reported a simi-
lar motility mechanism in Vibrio cholerae, the causal
agent of pandemic disease cholera, but it is much less
understood how pathogenic bacteria use EPS-
mediated passive motility for infection. Besides EPS,
other factors reported to affect sliding are surfactants
(Hölscher & Kov
acs, 2017; Van Gestel et al., 2015), a
B. subtilis hydrophobin protein BslA (Grau et al., 2015),
the production of siderophore in Sinorhizobium meliloti
(Grau et al., 2015; Nogales et al., 2012; Van Gestel
et al., 2015), and glycopeptidolipids produced by Myco-
bacterium smegmatis (Recht & Kolter, 2001).
Erwinia amylovora is a Gram-negative enterobacte-
rium that causes the disease fire blight in many eco-
nomically important Rosaceous plants including apple,
pear, blackberry, and raspberry (Griffith et al., 2003;
Malnoy et al., 2012). Migration of this bacterium to the
host plant flowers or leaves is thought to be facilitated
by rain, wind, and insects (Holtappels et al., 2015;
Puławska et al., 2017; Pusey, 2000; Thomson, 1986).
On flowers, E. amylovora populations develop on stig-
mas, following which a flagella-dependent motility
mechanism and free moisture are needed to facilitate
the movement of bacterial cells into flowers.
E. amylovora enters into the flower via natural openings
in flower nectaries (Bayot & Ries, 1986). Infection of
flowers, leaves at shoot tips, and stems of the apple
host by E. amylovora is mediated by the type III secre-
tion system (T3SS) (Malnoy et al., 2012). Infection
occurs in cortical parenchyma cells layers of the host,
and E. amylovora cells are present in the intercellular
region or apoplast (Kharadi et al., 2021). Flagella are
not required for cell spreading in the apoplast, and sev-
eral studies have demonstrated that E. amylovora loses
flagella once inside the plant and flagella-deficient
strains exhibit similar spreading rates in young apple
leaves compared to the wild type (Bayot & Ries, 1986;
Cesbron et al., 2006; Holtappels et al., 2018;
Raymundo & Ries, 1981). Systemic downward spread-
ing of E. amylovora in trees via the apoplast is rapid,
but the underlying mechanism remains largely elusive.
For example, E. amylovora cells were detected in tis-
sues >50 cm below the inoculated shoot tips of apples
11 days after inoculation and in the susceptible root-
stock 3 weeks after inoculation (Momol et al., 1998).
Erwinia amylovora forms biofilms in planta,a
process that relies on the production of three
EPSs, amylovorna, levan, and cellulose (Castiblanco &
Sundin, 2016; Kharadi et al., 2021; Malnoy
et al., 2012). Amylovoran is a high-molecular-weight
acidic heteropolysaccharide, consisting of a pentasac-
charide repeating subunit containing galactose and glu-
curonic acid in a ratio of 4:1 (Nimtz et al., 1996). Levan
is a simple homopolymer of fructose synthesized extra-
cellularly from sucrose by levansucrase (Seemüller &
Beer, 1977) and cellulose is a homopolymer of glucose
residues (Ross et al., 1991). E. amylovora mutants defi-
cient in amylovoran production are nonpathogenic, and
levan and cellulose production mutants are reduced in
virulence (Bellemann & Geider, 1992; Castiblanco &
Sundin, 2018; Geier & Geider, 1993; Koczan
et al., 2009), illustrating the importance of these EPSs
to E. amylovora pathogenesis. We are interested in
motility mechanisms driving the systemic movement of
E. amylovora through the apple host, and we hypothe-
sized that sliding motility represented an important
mechanism for host invasion. In the present work, we
studied how E. amylovora cells migrate on surfaces
without the presence of flagella. We characterized an
EPS-mediated sliding motility mechanism and identified
two EPSs, amylovoran and levan, required for this
movement. Water is important for plant–microbe inter-
actions (Aung et al., 2018; Beattie, 2011). We provided
further evidence showing that water potential positively
regulates sliding, suggesting that osmotic-driven water
flow is the main force for E. amylovora sliding motility.
EXPERIMENTAL PROCEDURES
Bacterial strains, plasmids, primers, and
media
The bacterial strains and plasmids used in this study
are listed in Table 1.E. amylovora and P. syringae
strains were grown in LB medium at 28C. Escherichia
coli strains were grown in LB broth medium at 37C.
Swimming and swarming assays were conducted in
MM medium (Na
2
HPO
4
at 6 g/L, KH
2
PO
4
at 3 g/L, NaCl
at 0.5 g/L, NH
4
Cl at 1 g/L, MgSO
4
7H
2
O at 0.2 g/L,
CaCl
2
at 0.1 g/L, nicotinic acid at 0.2 g/L, thiamine
hydrochloride at 0.2 g/L, and sucrose at 20 g/L;
Falkenstein et al., 1989) solidified with 0.2% w/v or
0.4% of agar. The sliding motility of E. amylovora was
determined on surfaces of 1.0, 1.5, or 2.0 w/v % agar
MM, MM-sorbitol (modified MM containing 20 g/L of
sorbitol instead of sucrose), or MM supplemented with
various concentrations of NaCl. Amylovoran production
assays were conducted in modified basal medium A
(MBMA) (KH
2
PO
4
at 3 g/L, K
2
HPO
4
at 7 g/L,
(NH
4
)
2
SO
4
at 1 g/L, citric acid at 0.5 g/L, MgSO
4
at
0.03 g/L; Edmunds et al., 2013) containing 1% w/v sor-
bitol. Antibiotics were added as needed to media at the
following concentrations: ampicillin (Ap; 100 μg/ml),
chloramphenicol (Cm; 10 μg/ml), gentamicin (Gm;
15 μg/ml), or kanamycin (Km; 30 μg/ml). Genome
2YUAN ET AL.
TABLE 1 Strains and plasmids used in this study
Strains and plasmids Relevant characteristics References or source
Erwinia amylovora
Ea110 Wild type Ritchie and Klos (1977)
Ea1189 Wild type Zhao, Sundin, and Wang (2009)
ΔflhDC1 ΔflhDC1::Km;Km
r
, EAM_2034 and EAM_2033 deletion mutant in Ea1189 This study
ΔmotA1 ΔmotA1::Km;Km
r
, EAM_2032 deletion mutant in Ea1189 This study
ΔfliC ΔfliC::Cm;Cm
r
, EAM_2067 deletion mutant in Ea1189 This study
Δams Δams, clean mutant, deletion of 12-gene ams operon in Ea1189 Zhao, Sundin, and Wang (2009)
Δlsc Δlsc::Cm;Cm
r
, EAM_3468 deletion mutant in Ea1189 This study
ΔbcsA ΔbcsA, clean mutant, EAM_3387 deletion mutant in Ea1189 Castiblanco and Sundin (2018)
ΔamsΔlsc ΔamsΔlsc::Km;Km
r
, deletion mutant of 12-gene ams operon and EAM_3468 in Ea1189 This study
ΔamsG ΔamsG::Km;Km
r
, EAM_2174 deletion mutant in Ea1189 This study
ΔflhDC1Δams ΔamsΔflhDC1::Km;Km
r
, deletion mutant of 12-gene ams operon, EAM_2034, and EAM_2033 in Ea1189 This study
ΔflhDC1Δlsc Δlsc::CmΔflhDC1::Km;Km
r
and Cm
r
, EAM_3468, EAM_2034, and EAM_2033 deletion mutant in Ea1189 This study
ΔflhDC1ΔbcsA ΔbcsAΔflhDC1::Km;Km
r
, EAM_3387, EAM_2034, and EAM_2033 deletion mutant in Ea1189 This study
ΔedcC ΔedcC, clean mutant, EAM_1504 deletion mutant in Ea1189 Edmunds et al. (2013)
ΔedcE ΔedcE, clean mutant, EAM_2435 deletion mutant in Ea1189 Edmunds et al. (2013)
ΔedcCΔedcE ΔedcCΔedcE, clean mutant, EAM_1504 and EAM_2435 deletion mutant in Ea1189 Edmunds et al. (2013)
ΔamyR ΔamyR::Cm;Cm
r
, EAM_1300 deletion mutant in Ea1189 This study
Escherichia coli
DH5αsupE44 ΔlacU169 (ϕ80lacZΔM15) hsdR17 recA1endA1gyrA96 thi-1 relA1 Lab stock
Pseudomonas syringae
DC3000 Wild type Lab stock
Plasmids
pKD4 Template plasmid for kanamycin cassette, Km
r
Datsenko and Wanner (2000)
pKD3 Template plasmid for chloramphenicol cassette, Cm
r
Datsenko and Wanner (2000)
pKD46 Arabinose-inducible lambda red recombinase, Ap
r
Datsenko and Wanner (2000)
pBBR1-MCS5 Broad-host-range plasmid, Gm
r
Kovach et al. (1995)
pMP2444 gfp expressed from lac promoter, pBBR1-MCS5, Gm
r
Stuurman et al. (2000)
pBBR1-P
nptII
-mCherry mCherry expressed from nptII promoter, pBBR1-MCS5, Gm
r
This study
pBBR1-amyR amyR cloned in pBBR1-MCS5, Gm
r
This study
pBBR1-amsG amsG cloned in pBBR1-MCS5, Gm
r
This study
pBBR1-lsc lsc cloned in pBBR1-MCS5, Gm
r
This study
pBBR1-edcC edcC cloned in pBBR1-MCS5, Gm
r
This study
pBBR1-edcE edcE cloned in pBBR1-MCS5, Gm
r
This study
pmCherry_NAT Template plasmid for mcherry,Km
r
Roth and Chilvers (2019)
pPNptGreen gfp expressed from nptII promoter, pPROBE-KT, Km
r
Axtell and Beattie (2002)
pPProGreen gfp expressed from proU promoter, pPROBE-KT, Km
r
Axtell and Beattie (2002)
Abbreviations: Ap
r
, ampicillin resistance; Cm
r
, chloramphenicol resistance; Gm
r
, gentamicin resistance; Km
r
, kanamycin resistance.
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 3
(CP055227) and plasmid (CP055228) sequences of
E. amylovora strain Ea1189 were retrieved from the
National Center for Biotechnology Information (Yu
et al., 2020). Oligonucleotide primers used for cloning
are listed in Table S1.
Construction of mutant and
complementation
Deletion mutants of flhDC1,motA1,fliC,amsG,amyR,
and lsc were constructed using the red recombinase
method (Datsenko & Wanner, 2000), and the method
was recently described elsewhere (Yuan et al., 2022).
Complementation strains were constructed by cloning
the putative promoter and open reading frame (ORF)
regions of target genes into the plasmid pBBR1-MCS5
(Table 1), followed by electroporation of the resulting
plasmids into E. amylovora.
Swimming and swarming motility assays
Swimming and swarming motility assays were per-
formed in MM containing 0.2% or 0.4% of agar. In brief,
bacterial cells grown in LB were incubated at 28C until
the optical density at 600 nm (OD
600
) reached 1.0,
corresponding to the mid- to late-exponential growth
phase. Values of OD
600
were measured using a Tecan
Spark plate reader (Tecan, Männedorf, Switzerland). A
total of 5 μl (approximately 2.5 10
7
cells) of bacterial
cultures were spotted in the centre of MM plates, and
the inoculated plates were incubated at 28C for 48 h
and photographed. Diameters of the radial area repre-
senting the motility of swimming and swarming were
measured and quantified using ImageJ (Abràmoff
et al., 2004).
Apple shoot virulence, levansucrase
activity, and amylovoran production
assays
Apple shoot virulence and levansucrase activity assays
were performed as recently described (Yuan
et al., 2022). Amylovoran production was determined in
supernatants of bacterial cultures using a turbidity
assay with cetylpyrimidinium chloride (CPC)
(Bellemann et al., 1994). Briefly, cells grown in MBMA
medium supplemented with 1% sorbitol were incubated
at 28C with shaking (220 rpm) for 48 h. The 200 μlof
the supernatant were then mixed with 20 μl of CPC
(50 mg/ml) for 10 min, and amylovoran production was
calculated by measuring the resulting turbidity of mix-
tures at OD
600
and normalized to the OD
600
values of
the culture.
Sliding motility, biomass, and bacterial
population assays
The sliding motility of E. amylovora was assessed on
surfaces of MM plates containing various concentrations
of agar. A total of 5 μl of overnight bacterial cultures in
LB (OD
600
=1.0) was spotted on MM plates. The inocu-
lated plates were then incubated at 28C on a flat sur-
face and imaged daily. Sliding areas were quantified
from photographs of inoculated plates using ImageJ.
For bacterial sliding on inclined or declined surfaces,
cells were first inoculated on one side of square MM
plates and incubated at 28C for 48 h on a flat surface.
Plates were then placed on tilted surfaces, incubated for
16 h, and the length of sliding was measured. Inclined
(+) or declined () angles were measured using a pro-
tractor and adjusted based on the horizontal line.
To create confined microcapillaries, warm 1.5%
agar MM were filled into 2 ml microcentrifuge tubes,
each containing a 22 G 3.8 cm hypodermic needle
(BD, Franklin Lakes, NJ, U.S.A.), and solidified at room
temperature for 2 h. Needles were then removed, and
tubes were baked with the cap open at 28C for 12 h to
evaporate condensed water in the microcapillary. For
sliding motility, bacterial cells were transferred from LB
plates to the open end of the microcapillary using a
pipette tip.
For measuring biomass and bacterial population
size of E. amylovora during sliding, bacterial biomass
produced on 1.5% agar plates was harvested with a
loop and suspended in a pre-weighed 0.5PBS buffer.
The weight gain was determined which represents the
amount of biomass produced by E. amylovora. Bacte-
rial cells were resuspended in 0.5PBS buffer follow-
ing which the cell number was determined by serial
dilution and plate counting. To simultaneously quantify
bacterial populations around the inoculated spot and
the sliding area, bacterial biomass around the inocu-
lated spot was collected using a 6 mm standard biopsy
punch, and the remaining biomass around the sliding
area was collected with a loop.
GFP transcriptional activity assay
To measure the transcriptional activities of P
proU
-gfp
and P
nptII
-gfp,E. amylovora cells carrying reporter plas-
mids pPProGreen or pPNptGreen slid on 1.5% agar
MM containing various concentrations of NaCl were
harvested, respectively. Cells were resuspended and
washed with 0.5PBS buffer. The GFP intensity and
OD
600
values were measured using the Tecan Spark
plate reader (Tecan, Männedorf, Switzerland) with
excitement at 488 nm and emission detection at
435 nm. The reported GFP fluorescence was normal-
ized to the OD
600
value of bacterial suspensions.
4YUAN ET AL.
Biosurfactant detection and cell surface
hydrophobicity assays
Biosurfactant detection assay was performed using an
atomized oil assay as previously described, with a few
modifications (Burch et al., 2010). Bacteria grown over-
night were spotted onto LB or MM agar plates and incu-
bated at 28C for 48 h. A fine stream of light paraffin oil
(ThermoFisher Scientific, Waltham, MA, USA) was
applied onto the plate using an airbrush with an air
pressure of 1055 g/cm
2
(master airbrush model G22;
TCP Global Co., San Diego, CA, USA). Halos of the
biosurfactant were visualized and imaged immediately
after spray. Oil droplets were observed using the Leica
Zoom 2000 stereomicroscope (Leica microsystems,
Wetzlar, Germany).
Cell surface hydrophobicity was measured using
the hexadecane partitioning method (van Loosdrecht
et al., 1987). Bacterial cells from overnight LB cultures
were harvested and washed three times with 0.5PBS
buffer. Cells were then resuspended in 1 ml 0.5PBS
buffer and the OD
540
values were determined. The
250 μlofn-hexadecane (Sigma-Aldrich, St. Louis, MO,
USA) was then added to each cell suspension, and the
suspensions were vortexed for 10 min. The resulting
mixtures were incubated at 37C for 30 min. The OD
540
of the aqueous layer was measured against a blank of
hexadecane-extracted PBS. Cell surface hydrophobic-
ity was calculated as follows: cell surface hydrophobic-
ity (in %) =100(final OD/initial OD).
Construction of the pBBR1-P
nptII
-mCherry
plasmid and confocal microscopy analysis
The nptII promoter was amplified from the plasmid
pKD4 with primers nptII-F and nptII::mCherry-Rc
(Table S1), resulting in a 279-bp DNA fragment. The
709-bp mCherry fragment was amplified from the plas-
mid pmCherry_NAT with a set of primers, nptII::
mCherry-F and mCherry-Rc (Table S1). As the primer
nptII::mCherry-Rc was designed to include the reverse
complement sequence of nptII::mCherry-F, a crossover
PCR was performed using the above PCR products as
templates with primers nptII-F and mCherry-Rc to gen-
erate the nptII::mCherry fragment, followed by purifica-
tion, digestion with XbaI and HindIII restriction
enzymes, and ligation with the plasmid pBBR1-MCS5
digested by the same enzymes. The resulting plasmid
was confirmed by sequencing and transformed into
E. amylovora strain Ea1189 or various EPS-deficient
mutants by electroporation.
E. amylovora cells expressing gfp from the plasmid
pMP2444 or mCherry from the plasmid pBBR1-P
nptII
-
mCherry were observed for their sliding behaviour
using confocal microscopy. Settings for the Nikon
A1-Rsi confocal laser scanning microscope (Nikon
Instruments, Inc., Tokyo, Japan) were recently
described (Zhong et al., 2021). For image acquisition of
the entire bacterial sliding area, multiple images were
automatically collected across the bacterial film, with a
15% overlap between each image, and stitched
together to form one large-area image using the Nikon
NIS-Elements AR software. Briefly, for each field of
view within the large-area scan, confocal images were
collected in 20 μm increments through an average
thickness of 100 μm. For each confocal series, a maxi-
mum intensity projection (MIP) image was generated,
representing the brightest intensity pixels through the
Z-depth. MIP images across the bacterial film were
then stitched together to form a single large-area MIP
image.
Statistical analysis
Means and standard deviations of experimental results
were calculated using Excel, and mean comparisons
were performed using a two-tailed student’st-test
(Microsoft, Redmond, WA) or Fisher’s least significant
difference test using R software (https://www.R-
project.org).
RESULTS
Flagella are required for in vitro swimming
and swarming but not for apoplastic
spreading of E. amylovora
Bacteria rotate flagella to swim in planktonic environ-
ments or swarm over surfaces (Berg, 2003;
Kearns, 2010; Minamino et al., 2008). We confirmed
the function of flagella in E. amylovora by comparing
the swimming and swarming behaviours of the wild-
type (WT) E. amylovora strain Ea1189 with those of
three independent flagellar-deficient mutants,
Ea1189ΔflhDC1 (flagellar master regulator encoding
genes) (Wang et al., 2006), Ea1189ΔmotA1 (flagellar
motor complex encoding gene) (Blair & Berg, 1990),
and Ea1189ΔfliC (flagellin encoding gene) (Fitzgerald
et al., 2014). We found that cells of Ea1189 swam or
swarmed symmetrically from the inoculated spots with
diameters of 4.0 cm in minimal medium
(MM) solidified with 0.2% of agar and of 2.5 cm in
0.4% agar MM at 2 days post-inoculation (dpi), respec-
tively (Figure 1A,B). In contrast, flagella-deficient
mutants were non-motile in both media (Figure 1A,B),
which appeared consistent with the results from other
studies (Schachterle et al., 2019; Zhao et al., 2011;
Zhao, Wang, et al., 2009).
We also compared the disease progression
between Ea1189 and flagella-deficient mutants when
cells were inoculated at the tip of apple shoots through
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 5
wounding. At 10 dpi, necrosis was observed in
Ea1189-inoculated shoots with a percentage of lesion/
shoot length of approximately 65%. Meanwhile, shoots
inoculated with flagella-deficient mutants exhibited simi-
lar lesion lengths to those inoculated with Ea1189,
whereas no visible lesions were observed in shoots
inoculated with the amylovoran deficient mutant
Ea1189Δams (Figure 1C). Deletion of ams did not
affect swimming or swarming in vitro (data not shown).
Taken together, our data strongly suggest that flagella
and flagella-dependent motility mechanisms are not
required for the migration of E. amylovora in the
apoplast.
E. amylovora exhibits surface motility
independent of flagella
To explore the motility mechanisms independent of fla-
gella, we examined the movement of E. amylovora on
MM solidified with 1.5% agar, generating a relatively
dry and solid surface. Notably, cells of Ea1189 formed
slimy colonies at the inoculated spot and gradually
expanded outward in an asymmetric manner
(Figure S1). This movement pattern differed from those
of the symmetrical movements observed for swimming
or swarming in vitro (Figure 1A,B) and was further vali-
dated by visualizing cells constitutively expressing the
red fluorescent protein-encoding gene mCherry from
the plasmid pBBR1-MCS5 (Figure 1D). In addition, our
data confirmed that flagella are not required for this sur-
face movement, as three flagella-deficient mutants
(Ea1189ΔflhDC1, Ea1189ΔmotA1, and Ea1189ΔfliC)
exhibited motility areas that were not significantly differ-
ent from WT Ea1189 over a 4-day growth period
(Figure 1E,F). Taken together, our data support a
model that a previously uncharacterized motility mech-
anism enables surface migration of E. amylovora in a
flagella-independent manner. We named this sliding
motility as it is consistent with the description of the
FIGURE 1 Flagella are required for swimming and swarming but not for sliding or apoplastic spreading in E. amylovora. (A) Swimming and
(B) swarming motilities were determined in wild-type (WT) E. amylovora strain Ea1189 and flagella-deficient mutants, Ea1189ΔflhDC1,
Ea1189ΔmotA1, and Ea1189ΔfliC, in 0.2% and 0.4% agar minimal medium (MM), respectively. Images were taken 2 days post inoculation (dpi).
Mean and standard deviation (n=3) are shown. (C) Apple shoots inoculation assay was performed using the above-mentioned strains by a
scissor-dip method described in Experimental Procedures section. At 10 dpi, the percentage of lesion/shoot length for bacterial spreading in the
apoplast was calculated. Mean and standard deviation (n=5) are shown. (D) the surface sliding of Ea1189 harbouring the plasmid
pBBR1-P
nptII
-mCherry was determined on 1.5% agar MM. Microphotographs were captured using light microscopy with incident lighting (upper
graph) and confocal microscopy detecting mCherry fluorescence (lower graph), respectively, at 4 dpi. The scale bar represents 0.2 cm. Sliding
motilities (E) and overall sliding areas (F) were compared between WT Ea1189 and Ea1189 flagella-deficient mutants on 1.5% agar MM at 0, 1,
2, 3, and 4 dpi. The image was captured at 3 dpi. Mean and standard deviation (n=3) are shown. One representative experiment was chosen,
and three (two for the apple shoots assay) independent experiments were conducted. Asterisks indicate statistically significant differences of the
means (ns =not significant, p> 0.05; **p< 0.01 by Student’st-test).
6YUAN ET AL.
passive biofilm expansion and sliding motility from stud-
ies of several other bacterial species such as B. subtilis
and V. cholerae (Henrichsen, 1972; Hölscher &
Kov
acs, 2017; Seminara et al., 2012; Yan et al., 2017).
Interestingly, we found that unlike B. subtilis slides on
rich media such as lysogeny broth (LB) by a flagella-
independent mechanism (Fall et al., 2006),
E. amylovora did not slide on LB medium (data not
shown), suggesting that sliding motility might be
species-specific in bacteria.
Two EPSs, amylovoran and levan, are
required for the sliding motility of
E. amylovora
E. amylovora produces three EPSs, amylovoran, levan,
and cellulose (Castiblanco & Sundin, 2018; Malnoy
et al., 2012), which led to a question that whether the
above-described sliding motility is associated with the
secretion of EPSs. For this purpose, we tested sliding
phenotypes of mutant bacteria defective in the biosyn-
thesis of individual EPSs and compared them with that
of the WT Ea1189. Deletion of the ams operon, encod-
ing 12 amylovoran biosynthetic enzymes (Bugert &
Geider, 1995), the first ams gene amsG, or the
levansucrase encoding gene lsc (Seemüller &
Beer, 1977) resulted in cells that were unable to slide
on 1.5% agar MM (Figures 2A,B and S2). Deletion of
both ams and lsc (Ea1189ΔamsΔlsc) significantly hin-
dered sliding motility to a level the same as the single
deletion mutants (Figure 2A,B). In contrast, deletion of
the cellulose biosynthesis gene bcsA (Castiblanco &
Sundin, 2018) had a negligible impact on sliding motility
(Figure 2A,B). To further ensure that sliding motility is
driven by the production of EPSs, in trans expression of
amylovoran or levan biosynthesis genes was conducted
in EPS-deficient mutants and found WT levels of the
sliding phenotype (Figure S2). Since we did not observe
strong correlations between the production of these two
EPSs (Figure S3) and Geier and Geider (1993) previ-
ously found normal amounts of amylovoran in several
transposon mutants of lsc, these data indicate that both
amylovoran and levan play a key role for sliding.
Next, we speculated that the absence of flagella
would not affect EPS-mediated sliding motility. Indeed,
no changes were found between the sliding motility of
Ea1189ΔflhDC1 and that of the double mutant
Ea1189ΔflhDC1ΔbcsA (Figure S4), confirming that cel-
lulose is not involved in sliding. In contrast, double
mutants of Ea1189ΔflhDC1Δams and Ea1189ΔflhD
C1Δlsc were non-motile (Figure S4).
FIGURE 2 EPSs amylovoran and levan are required for sliding. (A) Microphotographs of bacterial sliding (A) and overall sliding areas
(B) were determined in wild-type (WT) Ea1189 and several EPS-deficient mutant strains of Ea1189, including Ea1189Δams, Ea1189Δlsc,
Ea1189ΔbcsA, and Ea1189ΔamsΔlsc, on 1.5% agar minimal medium (MM) at 2, 3, and 4 days post inoculation (dpi). The scale bar represents
0.5 cm. (C) Bacterial populations of WT Ea1189, Ea1189Δams, Ea1189Δlsc, and Ea1189ΔamsΔlsc were determined while cells were sliding on
1.5% agar MM at 0, 2, 3, and 4 dpi. (D) Weights of the total biomass produced by the sliding cells of Ea1189, Ea1189Δams, Ea1189Δlsc, and
Ea1189ΔamsΔlsc were measured at 2, 3, and 4 dpi. Assays were performed as described in Experimental Procedures section. Mean and
standard deviation (n=3) are shown. One representative experiment was chosen, and three independent experiments were performed.
Asterisks indicate statistically significant differences of the means (*p< 0.05, **p< 0.01 by Student’st-test). Ns, not significant
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 7
Sliding motility positively contributes to
biomass without affecting bacterial
population size
Bacteria migrate via various motility mechanisms to
acquire nutrients (Ni et al., 2020; Yan et al., 2017). To
investigate the role of sliding motility in E. amylovora,
we measured the bacterial population and biomass of
WT Ea1189 and various EPS-deficient mutants follow-
ing cell sliding on 1.5% agar MM. Interestingly, no sig-
nificant differences were observed in the number of
cells between WT and EPS-deficient mutants at 2, 3,
and 4 dpi (Figure 2C), implying that EPS-mediated slid-
ing motility does not benefit E. amylovora for growth.
This data also suggests that the impaired sliding motil-
ity of EPS-deficient mutants (Figure 2A,B) is not due to
a growth defect.
We observed a consistent increase in the weight of
biomass, consisting of extracellular materials and cells,
of WT Ea1189 sliding on 1.5% agar MM (Figure 2D).
Ea1189Δlsc, whose amylovoran production was nor-
mal in a liquid environment (Figure S3), generated
detectable biomass that was significantly reduced in
weight relative to that of Ea1189 (Figure 2D). Moreover,
the weight of biomass produced by two non-sliding
amylovoran-deficient mutants, Ea1189Δams and
Ea1189ΔamsΔlsc, was barely detected (Figure 2D),
which agrees with a previous study showing that MM
broth-grown E. amylovora produces amylovoran as the
main EPS (Yuan, McGhee, et al., 2021).
EPS-dependent osmotic pressure and
water availability positively affect sliding
motility
To understand how EPSs affect motility mechanisms,
we sought to test the established model that osmotic
pressure generated by EPS production drives sliding
(Seminara et al., 2012; Yan et al., 2017). To this end,
we compared the sliding motility of Ea1189 on MM con-
taining 1%, 1.5%, and 2% agar, respectively. Given that
lower agar concentrations generate higher osmotic
pressure differences between EPS and agar plate and
enhance water uptake (Yan et al., 2017), we validated
the above hypothesis by showing that an increase in
the sliding area of Ea1189 was detected corresponding
to a reduced agar concentration (Figure 3A). Interest-
ingly, we also found that Ea1189Δlsc slid on 1% agar
MM, whereas Ea1189Δams did not (Figure 3B), and
FIGURE 3 EPS-generated osmotic force drives sliding. (A) Sliding areas were determined on 1.0, 1.5, or 2.0% agar minimal medium
(MM) for E. amylovora strain Ea1189 at 0, 1, 2, 3, and 4 days post inoculation (dpi) and (B) for Ea1189, Ea1189Δams, and Ea1189Δlsc at 4 dpi.
Sliding areas and induction of P
proU
-gfp in Ea1189 harbouring the plasmid pPProGreen (C) or P
nptII
-gfp in Ea1189 harbouring the plasmid
pPNptGreen (D) were determined on 1.5% agar MM containing an increasing concentration of sodium chloride (NaCl) at 4 dpi, respectively. The
induction ratio was calculated relative to the fluorescence of cells grown with 0 mM NaCl. Three independent experiments were performed in
each experiment. Values are from one representative experiment. Error bars indicate standard deviations of the means (n=3). Asterisks
indicate statistically significant differences of the means (p< 0.05 by Student’st-test). Ns, not significant
8YUAN ET AL.
neither of these mutants were motile on MM containing
≥1.5% agar (Figure 3B), indicating that both EPSs are
needed for generating the maximum osmotic force.
Next, inspired by the importance of EPS-mediated
osmotic pressure in sliding (Figure 3A) and the notion
that EPSs are hydrated polymers mainly comprised of
water (Sutherland, 1972), we hypothesized that water
availability and water potential affect sliding. To
address this question, we conducted sliding experi-
ments on 1.5% agar MM supplemented with an
increasing amount of sodium chloride (NaCl). Several
studies demonstrated that NaCl is a permeating solute
that reduces the water potential of the growth medium
(Csonka, 1989; Halverson & Firestone, 2000). Indeed,
the expression of a green fluorescence protein (GFP)-
based transcriptional fusion P
proU
-gfp that quantita-
tively responds to water deprivation (Axtell &
Beattie, 2002) was induced by the addition of NaCl in a
dose-dependent manner in Ea1189 (Figure 3C), con-
firming a lower water potential. More importantly, we
found that the size of the sliding area was gradually
decreased, and that no sliding phenotype was
observed when 100 mM NaCl was added to the
medium conferring a water potential of approximately
0.5 MPa (Figure 3C) (Axtell & Beattie, 2002). Bacte-
rial populations were comparable between 0 and
0.5 MPa water potential on MM (Figure S5), which is
consistent with a previous study showing that the popu-
lation of E. amylovora was decreased only when the
water potential was lower than 2 MPa in flower nec-
taries (Pusey, 2000). As a negative control, P
nptII
-gfp,a
water potential-independent reporter (Axtell &
Beattie, 2002), expression was not altered in Ea1189
with or without the presence of NaCl (Figure 3D).
Lastly, surface wetting materials such as surfac-
tants are known to facilitate bacterial movement
(Harshey, 2003; Murray & Kazmierczak, 2008). In
E. amylovora, we did not detect any surfactants pro-
duced (Figure S6), suggesting that a surfactant is likely
not controlling or contributing to the sliding motility.
Since one of the major functions of surfactants is to
reduce the surface tension and increase the surface
wettability (Shekhar et al., 2015), we measured the cell
surface hydrophobicity of Ea1189, Ea1189Δams, and
Ea1189Δlsc and found that the cell surface of
E. amylovora was highly hydrophilic with a value of
hydrophobicity below 10%, and deletion of either ams
or lsc did not significantly affect this value (Figure S7).
Impact of gravity and osmotic pressure on
E. amylovora sliding
Gravity is a fundamental force; however, its impact on
bacterial motility remains largely unknown (Acres
et al., 2021). We observed a gravitational sliding motion
for E. amylovora, as revealed when cells of Ea1189
were sliding on surfaces of 1.5% agar MM positioned at
declined angles. From 10to 20, an increased slope
assisted the downward sliding phenotype of Ea1189,
and no sliding was observed for various EPS-deficient
mutants (Figure 4A). Meanwhile, the osmotic pressure
contrast created by EPS production and the availability
of water were found to mediate sliding, a process for
which flagella were not required (Figure 4B). We also
observed that Ea1189 failed to climb on a 5inclined
surface of 1.5% agar MM (Figure 4A). Although
E. amylovora formed a large bulk of slime consisting of
EPSs (Figures 1, 2, and S1), our data indicate that the
dynamic caused by EPS-generated osmotic force is
unable to overcome gravity in directing the sliding
motion of bacterial biomass on surfaces.
To further investigate how physical forces influence
sliding, we assessed the movement of E. amylovora in
confined environments, in which cells could migrate
through a vertically placed 0.8-mm-wide microcapillary
with walls comprised of 1.5% agar MM. Unlike uncon-
fined environments where E. amylovora biomass did
not propel upward (Figure 4A), we found that both
Ea1189 and Ea1189ΔflhDC1 exhibited upward sliding
motility within the microcapillary (Figure 4C), and simi-
lar sliding lengths were observed for Ea1189 sliding up
or down through the microcapillary (Figure 4C), indicat-
ing a negligible influence of gravity. On the other hand,
EPS production is needed for bacterial sliding in a con-
fined space, as Ea1189Δams was not motile within the
microcapillary environment (Figure 4C).
Sliding motility is controlled by several
EPS regulators and might act as a
virulence factor
Bis-(30-50)-cyclic dimeric guanosine monophosphate (c-
di-GMP) is a bacterial second messenger that posi-
tively regulates amylovoran production in E. amylovora
(Edmunds et al., 2013; Kharadi et al., 2021). Deletion
of genes edcC or edcE, encoding diguanylate cyclase
enzymes necessary for c-di-GMP biosynthesis,
resulted in reduced sliding motilities of Ea1189 on 1.5%
agar MM (Figure 5A), suggesting that c-di-GMP posi-
tively contributes to sliding presumably via promoting
amylovoran production. Complementation analysis by
in trans expression of edcC or edcE restored the
mutant phenotype to WT levels (Figure 5A). AmyR is a
putative sensory transduction regulator protein and a
major repressor of amylovoran (Zhao, Wang,
et al., 2009). Interestingly, AmyR also upregulates the
production of levan via an unknown mechanism (Wang
et al., 2012). We found that the sliding areas produced
by Ea1189ΔamyR were similar in size to those pro-
duced by Ea1189 at 2 and 3 dpi and were slightly
reduced in size at 4 dpi (Figure 5B), and the expression
of amyR from a plasmid (pBBR1-MCS5-amyR) strongly
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 9
inhibited the sliding motility of Ea1189ΔamyR on 1.5%
agar MM (Figure 5B). On 1.5% agar MM-sorbitol, a
modified MM containing sorbitol, the primary storage
carbohydrate in apples and other Rosaceae plants
(Loescher, 1987), instead of sucrose to prevent the pro-
duction of levan, our data showed that Ea1189ΔamyR
FIGURE 5 C-di-GMP, AmyR, and carbon sources affect sliding. (A) Sliding areas were measured on 1.5% agar minimal medium (MM) for
wild-type (WT) E. amylovora strain Ea1189, mutant strains of Ea1189, including Ea1189ΔedcC, Ea1189ΔedcE, and Ea1189ΔedcCΔedcE,
Ea1189 harbouring the empty vector pBBR1-MCS5, Ea1189ΔedcC harbouring pBBR1-edcC, and Ea1189ΔedcE harbouring pBBR1-edcE at
4 days post inoculation (dpi). (B) Sliding motilities were compared between Ea1189 harbouring pBBR1-MCS5, Ea1189ΔamyR harbouring
pBBR1-MCS5, and Ea1189ΔamyR harbouring pBBR1-amyR on 1.5% agar MM (MM-sucrose) at 2, 3, and 4 dpi, respectively. Sliding motilities
were also compared between Ea1189 and Ea1189ΔamyR when cells were sliding on 1.5% agar MM-sorbitol at 3 dpi. (C) Overall sliding areas
were measured for WT E. amylovora strain Ea110, WT Ea1189, Ea1189Δams, and Ea1189Δlsc on 1.5% agar MM at 2, 3, and 4 dpi and on
1.5% agar MM-sorbitol at 5, 6, and 7 dpi, respectively. One representative experiment was chosen, and three independent experiments with
three replicates were performed. Error bars indicate standard deviations of the means (n=3). Asterisks indicate statistically significant
differences of the means (p< 0.05 by Student’st-test). Ns, not significant
FIGURE 4 Gravity and EPS-based osmotic pressure affect sliding. (A) Sliding motility was determined on +5inclined, 10declined, or
20declined 1.5% agar minimal medium (MM) surfaces for wild-type (WT) E. amylovora strain Ea1189, Ea1189Δams, Ea1189ΔamsG, and
Ea1189Δlsc, respectively. Filled red circles indicate the bacterial inoculation spot, following which the degree of inclined or declined surfaces is
indicated by a filled white triangle. (B) Sliding lengths were measured on 10-declined 1.0 or 1.5% agar MM for WT Ea1189, Ea1189ΔflhDC1,
and Ea1189Δams with or without the presence of sodium chloride (NaCl). Mean and standard deviation (n=3) are shown. (C) Arepresentative
image showing E. amylovora slid up in a confined microcapillary. White and black arrows indicate the beginning and end of sliding over a 2-day
period. Microphotograph was taken using light microscopy with transmitted lighting. (D) Sliding lengths were measured in vertically placed
microcapillaries for WT Ea1189 (slid up or down) and Ea1189ΔflhDC1 (slid up) and Ea1189Δams (slid up), respectively. Mean and standard
deviation (n=12) are shown. Assays were performed as described in experimental procedures. Three independent experiments were
performed, and one representative experiment was chosen. Asterisks indicate statistically significant differences of the means (p< 0.05 by
Student’st-test). Ns, not significant. Different lowercase letters above the bars indicate statistically significant differences between treatments
(Fisher’s least significant difference, p< 0.05).
10 YUAN ET AL.
produced a sliding area that was 4-fold larger than that
produced by Ea1189 at 3 dpi (Figure 5B). Taken
together, these data are in line with our conclusion that
both EPSs amylovoran and levan are required for the
full sliding motility of E. amylovora on MM, as in a con-
dition that only amylovoran but not levan is synthesized
(MM-sorbitol), bacteria could still slide, but at a much
slower pace. In addition, our observation agrees with a
previous study showing that the rate of fire blight pro-
gression in apple shoots is negatively correlated with
sorbitol concentration (Suleman & Steiner, 1994).
Different strains of E. amylovora have been reported
to be genetically similar yet different in their levels of
virulence (Cabrefiga & Montesinos, 2005;Puławska &
Sobiczewski, 2012). A highly virulent E. amylovora
strain Ea110 produces more amylovoran than the
lower-virulent strain Ea1189, while the levansucrase
activity was comparable (McGhee & Jones, 2000;
Wang et al., 2010). We found that the enhanced amylo-
voran production indeed led to a significant increase in
sliding areas for Ea110 relative to Ea1189 on either
MM or MM-sorbitol (Figure 5C). As controls,
Ea1189Δams failed to slide whereas Ea1189Δlsc slid
similar to Ea1189 on MM-sorbitol (Figure 5C). Collec-
tively, we conclude that EPS-mediated sliding motility
could act as a virulence factor of E. amylovora. When
the sorbitol concentration is high, E. amylovora cells
could migrate in host plants via producing amylovoran;
however, when sucrose becomes the main carbon
source, cells require the production of both amylovoran
and levan for the maximum sliding motility.
Amylovoran and levan play different roles
in sliding
To gain further insights into EPS-mediated sliding motil-
ity, we assessed the sliding behaviour of E. amylovora
in co-inoculated mixtures of two strains (1:1 ratio), com-
prised of Ea1189 harbouring a plasmid constitutively
expressing GFP and various EPS-deficient mutants
harbouring a plasmid constitutively expressing
mCherry. Microscopic images revealed that both
Ea1189Δams and Ea1189Δlsc slid in the presence of
Ea1189 on 1.5% agar MM (Figure 6A,B), and we did
not observe a rescued sliding motility in a mixture of
Ea1189Δams and Ea1189Δlsc (data not shown),
highlighting an involvement of both EPSs in modulating
the sliding motility of E. amylovora. Fluorescence pro-
teins were not altering the sliding phenotype because
similar sliding patterns were observed for Ea1189 and
EPS-deficient mutants expressing either GFP or
mCherry (Figures 1and S8).
Interestingly, unlike Ea1189Δams, which slid simi-
larly to Ea1189 in a mixture (Figure 6B), distinct differ-
ences were observed between Ea1189 and
Ea1189Δlsc, as cells of Ea1189Δlsc were visualized
mostly at the inoculated spot rather than at the margin
of the sliding zone (Figure 6A). Quantitative analysis
confirmed this observation showing that the number of
cells of Ea1189Δlsc was 3.7 times more than that of
Ea1189 around the inoculated spot and was 3.2 times
less in the sliding area (Figure S9). In contrast, cell
numbers of Ea1189Δams were slightly less (1.4
times) than Ea1189 only in the sliding area (Figure S9).
These results collectively imply that extracellular com-
plementation of levan from WT Ea1189 could not res-
cue the sliding motility of Ea1189Δlsc, which is
different from the amylovoran-mediated sliding motility.
We further validated this hypothesis by comparing
the overall sliding area in conditions where WT Ea1189
cells were mixed with either live or dead cells of
EPS-deficient mutants for sliding. As expected, no sig-
nificant differences were observed between Ea1189 +-
Ea1189Δlsc and Ea1189 +Ea1189Δlsc
H
(
H
represents cells were being heat treated at 95C for
10 min) (Figure 6C), but the sliding area of Ea1189 +-
Ea1189Δams was greatly reduced in size when com-
pared with that of Ea1189 +Ea1189Δams
H
(Figure 6C). Ea1189ΔamsΔlsc was unable to produce
either amylovoran or levan (Figure S3). Our data
showed that cells of Ea1189ΔamsΔlsc were less
motile than Ea1189 during sliding similar to the single
deletion mutant Ea1189Δlsc (Figure 6A,D)
and produced a much smaller sliding zone in
Ea1189 +Ea1189ΔamsΔlsc than that of Ea1189 +-
Ea1189ΔamsΔlsc
H
(Figure 6C), a phenotype the same
as that of Ea1189Δams (Figure 6C). Taken together,
these data indicate different roles of amylovoran and
levan in controlling the sliding motility of E. amylovora.
The association and possible physical linkage of levan
with levan-producing cells are important for sliding,
whereas the presence of exogenous amylovoran alone
could power the sliding motility of amylovoran-deficient
cells of E. amylovora (Figure 7).
DISCUSSION
In this work, we demonstrated that E. amylovora
exhibits a unique surface motility mechanism driven by
the production of two EPSs, amylovoran and levan. We
further defined this as sliding motility, a mechanism ini-
tially described as a passive bacterial translocation
powered by expansive forces created by cell division
(Henrichsen, 1972; Hölscher & Kov
acs, 2017). It has
been shown that E. amylovora produces mucoid colo-
nies on minimal agar medium due to its ability to
secrete EPS; in fact, early researchers used this growth
morphology to distinguish E. amylovora from other
Erwinia and Pseudomonas species and to rate its abil-
ity for EPS production in various naturally occurring
strains (Falkenstein et al., 1988; Falkenstein
et al., 1989). However, little is known regarding the
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 11
consequence of EPS production on solid surfaces and,
more importantly, how this affects the physiological
behaviour of E. amylovora in vitro and in planta.
A unique feature of bacterial sliding is that this
movement occurs independently of flagella (Hölscher &
Kov
acs, 2017). The sliding motility phenotype reported
in several bacterial systems, such as P. aeruginosa
(Murray & Kazmierczak, 2008), Salmonella enterica
serovar Typhimurium (Park et al., 2015), and Serratia
marcescens (Matsuyama et al., 1992), is flagella-
FIGURE 6 Sliding behaviours of E. amylovora in bacterial communities. Microphotographs were captured using light microscopy with
incident lighting (bright field) and confocal laser scanning microscope detecting GFP (green) or mCherry (red). Ea1189 cells harbouring the
plasmid pMP2444 were co-inoculated with Ea1189Δams (a) or Ea1189Δlsc (B) harbouring the plasmid pBBR1-P
nptII
-mCherry on 1.5% agar
minimal medium at a ratio of 1:1. Cells were sliding for 4 days at 28C. Microscopic overviews and closeups, showing cells from the inoculated
spot (centre) and the edge of sliding (edge), were shown. (C) Overall sliding areas were measured in bacterial communities including
Ea1189 +Ea1189Δams
H
, Ea1189 +Ea1189Δams, Ea1189
H
+Ea1189Δams, Ea1189 +Ea1189Δlsc
H
, Ea1189 +Ea1189Δlsc,
Ea1189
H
+Ea1189Δlsc, Ea1189 +Ea1189ΔamsΔlsc
H
, Ea1189 +Ea1189ΔamsΔlsc, and Ea1189
H
+Ea1189ΔamsΔlsc
H
. Represents cells
were being heat-treated at 95C for 10 min. Mean and standard deviation (n=3) are shown. Asterisks indicate statistically significant differences
of the means (p< 0.05 by Student’st-test). Ns, not significant. (D) Microphotographs showing sliding cells of Ea1189 harbouring pMP2444 and
Ea1189ΔamsΔlsc harbouring pBBR1-P
nptII
-mCherry. Overviews and closeups, including centres, midpoints, and edges, were shown. Bars
represent 0.2 cm in overviews, 0.1 mm in closeups of (a) and (B), and 0.2 mm in closeups of (D). Data are representative of three independent
experiments.
12 YUAN ET AL.
independent in all cases. This is also true in
E. amylovora as we found that the bacterial cells slide
similarly in the presence or absence of flagella on
surfaces of 1.5% agar plates or in confined microcapil-
laries (Figures 1and 4). However, in soft agar (0.2%–
0.4%), E. amylovora primarily uses flagella-based
swimming and swarming for translocation (Figure 1A,
B). These observations could be explained by the
mechanics of flagella since flagellar filaments could
rotate and generate torque more easily in liquid or
semi-solid environments than on solid surfaces
(Mandadapu et al., 2015), but it is more likely that
E. amylovora has employed a complex signalling
pathway, by which cells could switch between flagella-
based and EPS-mediated motilities depending on dif-
ferent environmental conditions. Indeed, c-di-GMP, a
key signalling component that represses swimming but
induces biofilm formation in E. amylovora (Edmunds
et al., 2013), was also involved in sliding (Figure 5A). c-
di-GMP plays an important role in organismal sensing
of environmental cues (Jenal et al., 2017; Sondermann
et al., 2012). For example, several recent studies con-
ducted in Escherichia coli and P. aeruginosa sug-
gested that a higher surface stiffness could stimulate c-
di-GMP signalling causing a reduction in flagella-based
motility and enhanced biofilm development (Peng
et al., 2019; Vrabioiu & Berg, 2022). Nevertheless, fur-
ther studies are needed to fully understand the underly-
ing mechanism that coordinates different motility
mechanisms in E. amylovora.
Bacterial cells produce surfactants to reduce sur-
face tension (Ron & Rosenberg, 2001), a process that
has been reported to facilitate sliding behaviour
(Hölscher & Kov
acs, 2017). For example, in
P. syringae pv. tomato DC3000, Nogales and
colleagues reported a surface sliding mechanism that
relies heavily on the production of the biosurfactant syr-
ingafactin (Nogales et al., 2015). Our data indicated
that E. amylovora does not produce surfactants
(Figure S6), which is likely not a surprise because
E. amylovora is best adapted to growth in the flower
environment and in the interior of plants (apoplast and
xylem), whereas P. syringae is particularly ubiquitous
on leaves (Kharadi et al., 2021; Lindow &
Brandl, 2003). Hydrophobins are also known to contrib-
ute to bacterial sliding. Grau et al. (2015) reported that
BslA, a hydrophobin-like protein, acts as a water repel-
lent for the sliding cells of B. subtilis. Since we found
that the cell surface of E. amylovora was highly hydro-
philic (Figure S7), whether and how hydrophobins
affect E. amylovora sliding motility requires further
investigation. We demonstrated here that the sliding
motility of E. amylovora is mainly powered by the pro-
duction of EPS in vitro similar to those characterized in
Sinorhizobium meliloti, the soil-dwelling bacterium
B. subtilis, and the plant pathogen Xanthomonas citri
subsp. citri (Grau et al., 2015; Nogales et al., 2012;
Seminara et al., 2012). Seminara and colleagues pro-
posed osmotic pressure-driven sliding motility via the
secretion of EPS in B. subtilis (Seminara et al., 2012),
which has also been discovered in V. cholerae (Yan
et al., 2017). Although specific evidence demonstrating
that osmotic pressure physically spreads E. amylovora
cells on surfaces is lacking, our data showed that water
potential indeed positively contributes to sliding, and
Yan et al. (2017) reported that an osmotic pressure dif-
ferential could draw water into the secreted EPS matrix.
Furthermore, T3SS and the type III effector DspA/E are
pathogenicity factors of E. amylovora (Oh et al., 2005;
Yuan, Hulin, & Sundin, 2021). A study conducted in
P. syringae demonstrated that AvrE1, an E. amylovora
DspA/E homologue, promotes stomatal closure and the
occurrence of water-soaking lesions in host plants (Xin
et al., 2016), and a similar phenotype has been
reported in X.gardneri causing bacterial spot in tomato
(Schwartz et al., 2017). The T3SS-mediated water-
soaking lesions create an aqueous living space in the
apoplast, which is essential for bacterial multiplication
and disease development in the apoplast (Aung
et al., 2018; El Kasmi et al., 2018; Gentzel et al., 2022;
Roussin-Léveillée et al., 2022). A similar water-soaking
lesion symptom caused by E. amylovora occurs during
infection of apple flowers, infection of leaves at shoot
tips, and in cankers (Van der Zwet and Keil, 1979). We
hypothesize that the induction of an aqueous apoplast
by E. amylovora facilitates the use of sliding motility to
enable systemic spreading through host tissue without
flagella.
Interestingly, unlike V. cholerae cells that acquire
nutrients for growth by sliding (Yan et al., 2017), the
sliding cells of E. amylovora showed comparable cell
populations and enhanced production of biomass
FIGURE 7 A graphical representation of EPS-mediated sliding
motility in E. amylovora.E. Amylovora cells secrete levansucrase to
catalyse the fructosyl transfer from sucrose to levan (blue hexagons)
in the extracellular space. Levan is likely linked physically to the cell.
Amylovoran is thought to be synthesized in the bacterial cytoplasm,
polymerized in cell membranes, and secreted to the extracellular
space. Amylovoran could attach to the cell as capsules (yellow circles
around the bacterial cell) or secrete as a slime (a bigger yellow cloud
around the biomass). Osmotic pressure gradients generated due to
the EPS production could draw water from surrounding surfaces to
the EPS matrix, resulting in sliding motility. Arrows indicate a sliding
direction. The figure was created with BioRender.com.
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 13
consisting of EPS than those of the non-sliding mutants
in vitro (Figure 2). These data signify the notion that
EPS production is an energy-consuming process, but it
remains to be determined how E. amylovora balances
between virulence and fitness at the molecular level.
Recently, we reported that a functional tricarboxylic
acid (TCA) cycle is required for amylovoran production;
when cells encounter low oxygen environments,
E. amylovora could synthesize thiamine pyrophos-
phate, a cofactor of several TCA metabolic enzymes, to
stimulate the TCA cycle thus providing energetic
requirements to produce amylovoran (Yuan, McGhee,
et al., 2021). In addition, unlike the in vitro medium
environment where we conducted sliding experiments,
the hydrated apoplast region of leaves accompanying
water-soaking lesions would provide a ready source of
water and nutrients that are leaking from plant host
cells killed by the pathogen. This modified environment
is likely conducive to both cell proliferation and
increased EPS synthesis that furthers sliding motility
and systemic spread through the plant (Figure 8). Fur-
thermore, the high density of E. amylovora cells and
EPS in parenchymal cell layers of the plant also con-
tributes to the exudation of ooze droplets on the plant
surface that function in pathogen dispersal (Slack
et al., 2017; Thomson, 2000).
EPSs produced by E. amylovora have multiple func-
tions ranging from modulating bacterial lifestyles to
affecting the overall pathogenicity. Early studies dem-
onstrated that amylovoran and levan are important
structural components of biofilms formed by
E. amylovora in vitro and in planta (Bellemann
et al., 1994; Gross et al., 1992; Koczan et al., 2009).
Castiblanco and Sundin (2018) showed that cellulose
plays a role in shaping the biofilm matrix in vitro and
contributes to the virulence of E. amylovora in an apple
shoot model. More recently, amylovoran and cellulose
have been reported to be involved in autoaggregation,
a newly defined behaviour of E. amylovora in liquid
environments (Kharadi & Sundin, 2019). Our data sug-
gested that amylovoran and levan are required for slid-
ing, which poses an intriguing question: why would
E. amylovora use two different EPSs for its surface
movement? This could be explained by the finding that
amylovoran and levan exhibit different physiological
functions for sliding, as we showed that, unlike
amylovoran-deficient cells, mixing levan-deficient cells
with WT bacteria did not rescue their sliding motility
(Figure 6). Since amylovoran has been visualized as
capsules attached to the bacterial cell wall that could
also be released into the environment (Bellemann
et al., 1994), we conclude that amylovoran plays a
larger role in forming the bulk of the biomass for gener-
ating osmotic pressure (Figure 7). Water could then be
drawn into the EPS matrix due to osmotic pressure dif-
ferences, expanding the volume of biomass, and result-
ing in sliding motility (Figure 7). The localization of
levan is poorly characterized. Our data indicate that
levan, produced by levansucrase in the extracellular
space, is likely associated with the cell membrane.
Levan draws water from surrounding surfaces, and the
resulting force together with the physical linkage
between levan and cells could then push cells for slid-
ing (Figure 7).
Lastly, how do E. amylovora cells migrate in the
apoplast? This question has been a long-standing puz-
zle in the fire blight community. Despite flagella-
dependent motilities being the only characterized trans-
location methods in E. amylovora, evidence from a
large number of early studies and our data indicate that
flagella are not required for the apoplastic migration of
E. amylovora through infected trees (Bayot &
Ries, 1986; Cesbron et al., 2006; Raymundo &
Ries, 1981). Furthermore, a recent proteomic analysis
showed that the presence of flagellar proteins might
negatively affect the virulence of E. amylovora in the
apoplast (Holtappels et al., 2018), possibly due to
PAMP (pathogen-associated molecular pattern)-
triggered plant immunity (Zhang & Zhou, 2010) caused
by flagellar proteins such as flagellin. In this study, we
discovered a passive surface translocation that is inde-
pendent of flagella and flagella-based motilities in
E. amylovora. In line with previous studies implying that
EPS is likely the key player in powering the apoplastic
migration of E. amylovora (Geider, 2000; Koczan
et al., 2009; Slack et al., 2017), we showed that sliding
is driven by two EPSs amylovoran and levan. This
EPS-mediated motility was required for the in vitro
translocation of bacteria in both unconfined and con-
fined spaces and is likely to play a major role in the
migration of E. amylovora cells in planta. In summary,
we propose a model showing that the production of
FIGURE 8 Model of the apoplast region of parenchymal cell
layers of an apple leaf upon initiation of E. amylovora infection
showing (A) a small number of invading cells establishing an
infection, and (B) type three secretion mediated host cell death
resulting in leakage of water and nutrients that result in hydration of
the apoplast stimulating pathogen cell growth and increased EPS
production that furthers sliding motility and systemic spread. The
figure was created with BioRender.com.
14 YUAN ET AL.
EPSs by bacterial cells could draw the surrounding
water to the polymers, thereby expanding the volume
of biomass, and in the process, pushing the cell for slid-
ing. Such a mechanism could be facilitated by the
release of water and nutrients from plant cells caused
by the T3SS-mediated cell death and are extremely
efficient for spreading in confined spaces in the apo-
plast, as bacterial cells that are intimately bordered
could expand simultaneously for sliding (Figure 8).
ACKNOWLEDGEMENTS
The authors thank Melinda K. Frame at the Centre for
Advanced Microscopy at Michigan State University
for the assistance with confocal microscopy, Janette
L. Jacobs and Martin I. Chilvers for kindly providing
the plasmid pmCherry_NAT, and Gwyn A. Beattie at
Iowa State University for kindly providing the plasmids
pPNptGreen and pPProGreen. The authors also
thank Lindsay Brown for the assistance with figure
creation. This project was supported by funds from
the Agriculture and Food Research Initiative Competi-
tive Grants Program Grants 2015-67013-23068 and
2020-51181-32158 from the USDA National Institute
of Food and Agriculture and by Michigan State
University AgBioResearch.
CONFLICT OF INTEREST
The author declares that there is no conflict of interest
that could be perceived as prejudicing the impartiality
of the research reported.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are
available from the corresponding author upon reason-
able request.
REFERENCES
Abràmoff, M.D., Magalh˜
aes, P.J. & Ram, S.J. (2004) Image proces-
sing with ImageJ. Biophotonics International, 11, 36–42.
Acres, J.M., Youngapelian, M.J. & Nadeau, J. (2021) The influence of
spaceflight and simulated microgravity on bacterial motility and
chemotaxis. npj Microgravity,7,1–11.
Aung, K., Jiang, Y. & He, S.Y. (2018) The role of water in plant–
microbe interactions. The Plant Journal, 93, 771–780.
Axtell, C.A. & Beattie, G.A. (2002) Construction and characterization
of a proU-gfp transcriptional fusion that measures water avail-
ability in a microbial habitat. Applied and Environmental Microbi-
ology, 68, 4604–4612.
Bayot, R.G. & Ries, S.M. (1986) Role of motility in apple blossom
infection by Erwinia amylovora and studies of fire blight control
with attractant and repellent compounds. Phytopathology, 76,
441–445.
Beattie, G.A. (2011) Water relations in the interaction of foliar bacte-
rial pathogens with plants. Annual Review of Phytopathology,
49, 533–555.
Bellemann, P., Bereswill, S., Berger, S. & Geider, K. (1994) Visualiza-
tion of capsule formation by Erwinia amylovora and assays to
determine amylovoran synthesis. International Journal of Biolog-
ical Macromolecules, 16, 290–296.
Bellemann, P. & Geider, K. (1992) Localization of transposon inser-
tions in pathogenicity mutants of Erwinia amylovora and their
biochemical characterization. Microbiology, 138, 931–940.
Berg, H.C. (2003) The rotary motor of bacterial flagella. Annual
Review of Biochemistry, 72, 19–54.
Blair, D.F. & Berg, H.C. (1990) The MotA protein of E. coli is a proton-
conducting component of the flagellar motor. Cell, 60, 439–449.
Bugert, P. & Geider, K. (1995) Molecular analysis of the ams operon
required for exopolysaccharide synthesis of Erwinia amylovora.
Molecular Microbiology, 15, 917–933.
Burch, A.Y., Shimada, B.K., Browne, P.J. & Lindow, S.E. (2010)
Novel high-throughput detection method to assess bacterial sur-
factant production. Applied and Environmental Microbiology,
76, 5363–5372.
Cabrefiga, J. & Montesinos, E. (2005) Analysis of aggressiveness of
Erwinia amylovora using disease-dose and time relationships.
Phytopathology, 95, 1430–1437.
Castiblanco, L.F. & Sundin, G.W. (2016) New insights on molecular
regulation of biofilm formation in plant-associated bacteria. Jour-
nal of Integrative Plant Biology, 58, 362–372.
Castiblanco, L.F. & Sundin, G.W. (2018) Cellulose production, acti-
vated by cyclic di-GMP through BcsA and BcsZ, is a virulence
factor and an essential determinant of the three-dimensional
architectures of biofilms formed by Erwinia amylovora Ea1189.
Molecular Plant Pathology, 19, 90–103.
Cesbron, S., Paulin, J.-P., Tharaud, M., Barny, M.-A. & Brisset, M.-N.
(2006) The alternative σfactor HrpL negatively modulates the
flagellar system in the phytopathogenic bacterium Erwinia amy-
lovora under hrp-inducing conditions. FEMS Microbiology Let-
ters, 257, 221–227.
Csonka, L.N. (1989) Physiological and genetic responses of bacteria
to osmotic stress. Microbiological Reviews, 53, 121–147.
Datsenko, K.A. & Wanner, B.L. (2000) One-step inactivation of chro-
mosomal genes in Escherichia coli K-12 using PCR products.
Proceedings of the National Academy of Sciences of the
United States of America, 97, 6640–6645.
Edmunds, A.C., Castiblanco, L.F., Sundin, G.W. & Waters, C.M.
(2013) Cyclic Di-GMP modulates the disease progression of
Erwinia amylovora.Journal of Bacteriology, 195, 2155–2165.
El Kasmi, F., Horvath, D. & Lahaye, T. (2018) Microbial effectors and
the role of water and sugar in the infection battle ground. Current
Opinion in Plant Biology, 44, 98–107.
Falkenstein, H., Bellemann, P., Walter, S., Zeller, W. & Geider, K.
(1988) Identification of Erwinia amylovora, the fireblight patho-
gen, by colony hybridization with DNA from plasmid pEA29.
Applied and Environmental Microbiology, 54, 2798–2802.
Falkenstein, H., Zeller, W. & Geider, K. (1989) The 29 kb plasmid,
common in strains of Erwinia amylovora, modulates develop-
ment of fireblight symptoms. Microbiology, 135, 2643–2650.
Fall, R., Kearns, D.B. & Nguyen, T. (2006) A defined medium to
investigate sliding motility in a Bacillus subtilis flagella-less
mutant. BMC Microbiology,6,1–11.
Fitzgerald, D.M., Bonocora, R.P. & Wade, J.T. (2014) Comprehensive
mapping of the Escherichia coli flagellar regulatory network.
PLoS Genetics, 10, e1004649.
Geider, K. (2000) Exopolysaccharides of Erwinia amylovora: struc-
ture, biosynthesis, regulation, role in pathogenicity of amylo-
voran and levan. In: Vanneste, J.L. (Ed.) Fire blight: the disease
its causative agent, Erwinia amylovora. Wallingford, UK: CAB
International, pp. 117–140.
Geier, G. & Geider, K. (1993) Characterization and influence on viru-
lence of the levansucrase gene from the fireblight pathogen
Erwinia amylovora.Physiological and Molecular Plant Pathol-
ogy, 42, 387–404.
Gentzel, I., Giese, L., Ekanayake, G., Mikhail, K., Zhao, W.,
Cocuron, J.-C. et al. (2022) Dynamic nutrient acquisition from a
hydrated apoplast supports biotrophic proliferation of a bacterial
pathogen of maize. Cell Host & Microbe, 30, 502–517.e4.
Grau, R.R., de Oña, P., Kunert, M., Leñini, C., Gallegos-
Monterrosa, R., Mhatre, E. et al. (2015) A duo of potassium-
responsive histidine kinases govern the multicellular destiny of
Bacillus subtilis.mBio, 6, e00581–e00515.
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 15
Griffith, C.S., Sutton, T.B. & Peterson, P.D. (2003) Fire blight: the
foundation of phytobacteriology. St. Paul, MN, USA: APS press.
Gross, M., Geier, G., Rudolph, K. & Geider, K. (1992) Levan and
levansucrase synthesized by the fireblight pathogen Erwinia
amylovora.Physiological and Molecular Plant Pathology, 40,
371–381.
Halverson, L.J. & Firestone, M.K. (2000) Differential effects of perme-
ating and nonpermeating solutes on the fatty acid composition of
Pseudomonas putida.Applied and Environmental Microbiol-
ogy, 66, 2414–2421.
Harshey, R.M. (2003) Bacterial motility on a surface: many ways to a
common goal. Annual Review of Microbiology, 57, 249–273.
Henrichsen, J. (1972) Bacterial surface translocation: a survey and a
classification. Bacteriological Reviews, 36, 478–503.
Hölscher, T. & Kov
acs,
´
A.T. (2017) Sliding on the surface: bacterial
spreading without an active motor. Environmental Microbiology,
19, 2537–2545.
Holtappels, M., Noben, J.-P., Van Dijck, P. & Valcke, R. (2018) Fire
blight host-pathogen interaction: proteome profiles of Erwinia
amylovora infecting apple rootstocks. Scientific Reports,8,1–9.
Holtappels, M., Vrancken, K., Schoofs, H., Deckers, T., Remans, T.,
Noben, J.-P. et al. (2015) A comparative proteome analysis
reveals flagellin, chemotaxis regulated proteins and amylovoran
to be involved in virulence differences between Erwinia amylo-
vora strains. Journal of Proteomics, 123, 54–69.
Jenal, U., Reinders, A. & Lori, C. (2017) Cyclic di-GMP: second messen-
ger extraordinaire. Nature Reviews. Microbiology, 15, 271–284.
Kearns, D.B. (2010) A field guide to bacterial swarming motility.
Nature Reviews. Microbiology, 8, 634–644.
Kharadi, R.R., Schachterle, J.K., Yuan, X., Castiblanco, L.F.,
Peng, J., Slack, S.M. et al. (2021) Genetic dissection of the Erwi-
nia amylovora disease cycle. Annual Review of Phytopathology,
59, 191–212.
Kharadi, R.R. & Sundin, G.W. (2019) Physiological and microscopic
characterization of cyclic-di-GMP-mediated autoaggregation in
Erwinia amylovora.Frontiers in Microbiology, 10, 468.
Koczan, J.M., McGrath, M.J., Zhao, Y. & Sundin, G.W. (2009) Contri-
bution of Erwinia amylovora exopolysaccharides amylovoran
and Levan to biofilm formation: implications in pathogenicity.
Phytopathology, 99, 1237–1244.
Kovach, M.E., Elzer, P.H., Hill, D.S., Robertson, G.T., Farris, M.A.,
Roop, R.M. et al. (1995) Four new derivatives of the broad-host-
range cloning vector pBBR1MCS, carrying different antibiotic-
resistance cassettes. Gene, 166, 175–176.
Leigh, J.A. & Coplin, D.L. (1992) Exopolysaccharides in plant-
bacterial interactions. Annual Review of Microbiology, 46,
307–346.
Limoli, D.H., Jones, C.J. & Wozniak, D.J. (2015) Bacterial extracellu-
lar polysaccharides in biofilm formation and function. Microbiol
Spectrum, 3, 29.
Lindow, S.E. & Brandl, M.T. (2003) Microbiology of the phyllosphere.
Applied and Environmental Microbiology, 69, 1875–1883.
Loescher, W.H. (1987) Physiology and metabolism of sugar alcohols
in higher plants. Physiologia Plantarum, 70, 553–557.
Malnoy, M., Martens, S., Norelli, J.L., Barny, M.-A., Sundin, G.W.,
Smits, T.H. et al. (2012) Fire blight: applied genomic insights of
the pathogen and host. Annual Review of Phytopathology, 50,
475–494.
Mandadapu, K.K., Nirody, J.A., Berry, R.M. & Oster, G. (2015)
Mechanics of torque generation in the bacterial flagellar motor.
Proceedings of the National Academy of Sciences of the
United States of America, 112, E4381–E4389.
Matsuyama, T., Kaneda, K., Nakagawa, Y., Isa, K., Hara-Hotta, H. &
Yano, I. (1992) A novel extracellular cyclic lipopeptide which pro-
motes flagellum-dependent and -independent spreading growth
of Serratia marcescens.Journal of Bacteriology, 174, 1769–
1776.
Mattick, J.S. (2002) Type IV pili and twitching motility. Annual Review
of Microbiology, 56, 289–314.
McBride, M.J. (2001) Bacterial gliding motility: multiple mechanisms
for cell movement over surfaces. Annual Review of Microbiol-
ogy, 55, 49–75.
McGhee, G.C. & Jones, A.L. (2000) Complete nucleotide sequence
of ubiquitous plasmid pEA29 from Erwinia amylovora strain
Ea88: gene organization and intraspecies variation. Applied and
Environmental Microbiology, 66, 4897–4907.
Minamino, T., Imada, K. & Namba, K. (2008) Molecular motors of the
bacterial flagella. Current Opinion in Structural Biology, 18,
693–701.
Momol, M., Norelli, J., Piccioni, D., Momol, E., Gustafson, H.,
Cummins, J. et al. (1998) Internal movement of Erwinia amylo-
vora through symptomless apple scion tissues into the rootstock.
Plant Disease, 82, 646–650.
Murray, T.S. & Kazmierczak, B.I. (2008) Pseudomonas aeruginosa
exhibits sliding motility in the absence of type IV pili and flagella.
Journal of Bacteriology, 190, 2700–2708.
Ni, B., Colin, R., Link, H., Endres, R.G. & Sourjik, V. (2020) Growth-
rate dependent resource investment in bacterial motile behavior
quantitatively follows potential benefit of chemotaxis. Proceed-
ings of the National Academy of Sciences of the United States
of America, 117, 595–601.
Nimtz, M., Mort, A., Domke, T., Wray, V., Zhang, Y., Qiu, F. et al.
(1996) Structure of amylovoran, the capsular exopolysaccharide
from the fire blight pathogen Erwinia amylovora.Carbohydrate
Research, 287, 59–76.
Nogales, J., Bernabéu-Roda, L., Cuéllar, V. & Soto, M.J. (2012) ExpR
is not required for swarming but promotes sliding in Sinorhizo-
bium meliloti.Journal of Bacteriology, 194, 2027–2035.
Nogales, J., Vargas, P., Farias, G.A., Olmedilla, A., Sanju
an, J. &
Gallegos, M.-T. (2015) FleQ coordinates flagellum-dependent
and-independent motilities in Pseudomonas syringae
pv. Tomato DC3000. Applied and Environmental Microbiology,
81, 7533–7545.
Oh, C.-S., Kim, J.F. & Beer, S.V. (2005) The Hrp pathogenicity Island of
Erwinia amylovora and identification of three novel genes required
for systemic infection. Molecular Plant Pathology,6,125–138.
Park, S.-Y., Pontes, M.H. & Groisman, E.A. (2015) Flagella-
independent surface motility in Salmonella enterica serovar
Typhimurium. Proceedings of the National Academy of Sci-
ences of the United States of America, 112, 1850–1855.
Peng, Q., Zhou, X., Wang, Z., Xie, Q., Ma, C., Zhang, G. et al. (2019)
Three-dimensional bacterial motions near a surface investigated
by digital holographic microscopy: effect of surface stiffness.
Langmuir, 35, 12257–12263.
Puławska, J., Kałużna, M., Warabieda, W. & Mikici
nski, A. (2017)
Comparative transcriptome analysis of a lowly virulent strain of
Erwinia amylovora in shoots of two apple cultivars-susceptible
and resistant to fire blight. BMC Genomics, 18, 868.
Puławska, J. & Sobiczewski, P. (2012) Phenotypic and genetic diver-
sity of Erwinia amylovora: the causal agent of fire blight. Trees,
26, 3–12.
Pusey, P. (2000) The role of water in epiphytic colonization and infec-
tion of pomaceous flowers by Erwinia amylovora.Phytopathol-
ogy, 90, 1352–1357.
Raymundo, A. & Ries, S. (1981) Motility of Erwinia amylovora.Phyto-
pathology, 70, 1062–1065.
Recht, J. & Kolter, R. (2001) Glycopeptidolipid acetylation affects slid-
ing motility and biofilm formation in Mycobacterium smegmatis.
Journal of Bacteriology, 183, 5718–5724.
Ritchie, D. & Klos, E. (1977) Isolation of Erwinia amylovora bacterio-
phage from aerial parts of apple trees. Phytopathology, 67,
101–104.
Ron, E.Z. & Rosenberg, E. (2001) Natural roles of biosurfactants:
minireview. Environmental Microbiology, 3, 229–236.
Ross, P., Mayer, R. & Benziman, M. (1991) Cellulose biosynthesis
and function in bacteria. Microbiological Reviews, 55, 35–58.
Roth, M.G. & Chilvers, M.I. (2019) A protoplast generation and trans-
formation method for soybean sudden death syndrome causal
16 YUAN ET AL.
agents Fusarium virguliforme and F. brasiliense.Fungal Biolo-
gyBiotechnology,6,1–8.
Roussin-Léveillée, C., Lajeunesse, G., St-Amand, M., Veerapen, V.
P., Silva-Martins, G., Nomura, K. et al. (2022) Evolutionarily con-
served bacterial effectors hijack abscisic acid signaling to induce
an aqueous environment in the apoplast. Cell Host & Microbe,
30, 489–501.e4.
Schachterle, J.K., Zeng, Q. & Sundin, G.W. (2019) Three Hfq-dependent
small RNAs regulate flagellar motility in the fire blight pathogen
Erwinia amylovora.Molecular Microbiology, 111, 1476–1492.
Schwartz, A.R., Morbitzer, R., Lahaye, T. & Staskawicz, B. (2017)
TALE-induced bHLH transcription factors that activate a pectate
lyase contribute to water soaking in bacterial spot of tomato.
Proceedings of the National Academy of Sciences of the
United States of America, 114, E897–E903.
Seemüller, E. & Beer, S. (1977) Isolation and partial characterization
of two neutral proteases of Erwinia amylovora.Journal of Phyto-
pathology, 90, 12–21.
Seminara, A., Angelini, T.E., Wilking, J.N., Vlamakis, H., Ebrahim, S.,
Kolter, R. et al. (2012) Osmotic spreading of Bacillus subtilis bio-
films driven by an extracellular matrix. Proceedings of the
National Academy of Sciences of the United States of America,
109, 1116–1121.
Shekhar, S., Sundaramanickam, A. & Balasubramanian, T. (2015)
Biosurfactant producing microbes and their potential applica-
tions: a review. Critical Reviews in Environmental Science and
Technology, 45, 1522–1554.
Slack, S.M., Zeng, Q., Outwater, C.A. & Sundin, G.W. (2017) Microbi-
ological examination of Erwinia amylovora exopolysaccharide
ooze. Phytopathology, 107, 403–411.
Sondermann, H., Shikuma, N.J. & Yildiz, F.H. (2012) You’ve come a
long way: c-di-GMP signaling. Current Opinion in Microbiology,
15, 140–146.
Stuurman, N., Bras, C.P., Schlaman, H.R., Wijfjes, A.H.,
Bloemberg, G. & Spaink, H.P. (2000) Use of green fluorescent
protein color variants expressed on stable broad-host-range vec-
tors to visualize rhizobia interacting with plants. Molecular Plant-
Microbe Interactions, 13, 1163–1169.
Suleman, P. & Steiner, P.W. (1994) Relationship between sorbitol
and solute potential in apple shoots relative to fire blight symp-
tom development after infection by Erwinia amylovora.Phytopa-
thology, 84, 1244.
Sutherland, I. (1972) Bacterial exopolysaccharides. Advances in
Microbial Physiology, 8, 143–213.
Sutherland, I. (1982) Biosynthesis of microbial exopolysaccharides.
Advances in Microbial Physiology, 23, 79–150.
Sutherland, I.W. (2001) Biofilm exopolysaccharides: a strong and
sticky framework. Microbiology, 147, 3–9.
Thomson, S. (1986) The role of the stigma in fire blight infections.
Phytopathology, 76, 476–482.
Thomson, S.V. (2000) Epidemiology of fire blight. In: Fire blight: the
disease and its causative agent, Erwinia amylovora. Wallingfort
UK: CABI Publishing, pp. 9–37.
Van Der Zwet, T. & H.L. Keil. (1979) Fire blight: A bacterial disease of
rosaceous plants. No. 510. US Department of Agriculture.
Van Gestel, J., Vlamakis, H. & Kolter, R. (2015) From cell differentia-
tion to cell collectives: Bacillus subtilis uses division of labor to
migrate. PLoS Biology, 13, e1002141.
van Loosdrecht, M., Lyklema, J., Norde, W., Schraa, G. &
Zehnder, A. (1987) The role of bacterial cell wall hydrophobicity
in adhesion. Applied and Environmental Microbiology, 53,
1893–1897.
Vrabioiu, A.M. & Berg, H.C. (2022) Signaling events that occur when
cells of Escherichia coli encounter a glass surface. Proceedings
of the National Academy of Sciences of the United States of
America, 119, e2116830119.
Wang, D., Korban, S.S., Pusey, P.L. & Zhao, Y. (2012) AmyR is a
novel negative regulator of amylovoran production in Erwinia
amylovora.PLoS One, 7, e45038.
Wang, D., Korban, S.S. & Zhao, Y. (2010) Molecular signature of dif-
ferential virulence in natural isolates of Erwinia amylovora.Phy-
topathology, 100, 192–198.
Wang, S., Fleming, R.T., Westbrook, E.M., Matsumura, P. &
McKay, D.B. (2006) Structure of the Escherichia coli FlhDC
complex, a prokaryotic heteromeric regulator of transcription.
Journal of Molecular Biology, 355, 798–808.
Whitfield, C., Wear, S.S. & Sande, C. (2020) Assembly of bacterial
capsular polysaccharides and exopolysaccharides. Annual
Review of Microbiology, 74, 521–543.
Xin, X.-F., Nomura, K., Aung, K., Vel
asquez, A.C., Yao, J., Boutrot, F.
et al. (2016) Bacteria establish an aqueous living space in plants
crucial for virulence. Nature, 539, 524–529.
Yan, J., Nadell, C.D., Stone, H.A., Wingreen, N.S. & Bassler, B.L.
(2017) Extracellular-matrix-mediated osmotic pressure drives
vibrio cholerae biofilm expansion and cheater exclusion. Nature
Communications,8,1–11.
Yu, M., Singh, J., Khan, A., Sundin, G.W. & Zhao, Y.F. (2020) Com-
plete genome sequence of the fire blight pathogen Erwinia amy-
lovora strain Ea1189. Molecular Plant-Microbe Interactions, 33,
1277–1279.
Yuan, X., Eldred, L.I., Kharadi, R.R., Slack, S.M. & Sundin, G.W.
(2022) The RNA-binding protein ProQ impacts exopolysacchar-
ide biosynthesis and second messenger cyclic di-GMP signaling
in the fire blight pathogen Erwinia amylovora.Applied and Envi-
ronmental Microbiology, 88, e0023922.
Yuan, X., Hulin, M.T. & Sundin, G.W. (2021) Effectors, chaperones,
and harpins of the Type III secretion system in the fire blight
pathogen Erwinia amylovora: a review. Journal of Plant Pathol-
ogy, 103, S25–S39.
Yuan, X., McGhee, G., Slack, S. & Sundin, G.W. (2021) A novel sig-
naling pathway that connects thiamine biosynthesis, bacterial
respiration, and production of the exopolysaccharide amylo-
voran in Erwinia amylovora.Molecular Plant-Microbe Interac-
tions, 34, 1193–1208.
Zhang, J. & Zhou, J.-M. (2010) Plant immunity triggered by microbial
molecular signatures. Molecular Plant, 3, 783–793.
Zhao, Y., Qi, M. & Wang, D. (2011) Evolution and function of flagellar
and non-flagellar type III secretion systems in Erwinia amylo-
vora.Acta Horticulturae, 896, 177–184.
Zhao, Y., Sundin, G.W. & Wang, D. (2009) Construction and analysis
of pathogenicity Island deletion mutants of Erwinia amylovora.
Canadian Journal of Microbiology, 55, 457–464.
Zhao, Y., Wang, D., Nakka, S., Sundin, G.W. & Korban, S.S. (2009)
Systems level analysis of two-component signal transduction
systems in Erwinia amylovora: role in virulence, regulation of
amylovoran biosynthesis and swarming motility. BMC Geno-
mics, 10:245, 1–16.
Zhong, M., Cywiak, C., Metto, A.C., Liu, X., Qian, C. & Pelled, G.
(2021) Multi-session delivery of synchronous rTMS and sensory
stimulation induces long-term plasticity. Brain Stimulation, 14,
884–894.
SUPPORTING INFORMATION
Additional supporting information can be found online
in the Supporting Information section at the end of this
article.
How to cite this article: Yuan, X., Eldred, L.I. &
Sundin, G.W. (2022) Exopolysaccharides
amylovoran and levan contribute to sliding
motility in the fire blight pathogen Erwinia
amylovora.Environmental Microbiology,1–17.
Available from: https://doi.org/10.1111/1462-
2920.16193
EPS-MEDIATED SLIDING MOTILITY OF E. AMYLOVORA 17
