Calcium Increases Xylella fastidiosa Surface Attachment, BiofilmFormation, and Twitching Motility

Luisa F. Cruz,a Paul A. Cobine,b and Leonardo De La Fuentea

Department of Entomology and Plant Pathology, Auburn University, Auburn, Alabama, USA,a and Department of Biological Sciences, Auburn University, Auburn,Alabama, USAb

Xylella fastidiosa is a plant-pathogenic bacterium that forms biofilms inside xylem vessels, a process thought to be influenced bythe chemical composition of xylem sap. In this work, the effect of calcium on the production of X. fastidiosa biofilm and move-ment was analyzed under in vitro conditions. After a dose-response study with 96-well plates using eight metals, the strongestincrease of biofilm formation was observed when medium was supplemented with at least 1.0 mM CaCl2. The removal of Ca byextracellular (EGTA, 1.5 mM) and intracellular [1,2-bis(o-aminophenoxy)ethane-N,N,N=,N=-tetraacetic acid acetoxymethyl ester(BAPTA/AM), 75 �M] chelators reduced biofilm formation without compromising planktonic growth. The concentration of Cainfluenced the force of adhesion to the substrate, biofilm thickness, cell-to-cell aggregation, and twitching motility, as shown byassays with microfluidic chambers and other assays. The effect of Ca on attachment was lost when cells were treated with tetracy-cline, suggesting that Ca has a metabolic or regulatory role in cell adhesion. A double mutant (fimA pilO) lacking type I and typeIV pili did not improve biofilm formation or attachment when Ca was added to the medium, while single mutants of type I(fimA) or type IV (pilB) pili formed more biofilm under conditions of higher Ca concentrations. The concentration of Ca in themedium did not significantly influence the levels of exopolysaccharide produced. Our findings indicate that the role of Ca in bio-film formation may be related to the initial surface and cell-to-cell attachment and colonization stages of biofilm establishment,which rely on critical functions by fimbrial structures.

The xylem-limited Gram-negative bacterium Xylella fastidiosais the causal agent of many economically important plant dis-

eases, including Pierce’s disease (PD) of grapevines, citrus-variegated chlorosis (CVC) (26), and the emerging bacterial leafscorch of blueberries (2). Other crops, including alfalfa, peach,plum, almond, and coffee, and many forest trees and landscapeplants, including oak, elm, maple, and oleander, can also be in-fected by this bacterium. Characteristic disease symptoms, such asleaf scorch, have been associated with the blockage of xylem ves-sels by bacterial biofilms, which may result in a water deficit (51).Even though the exact mechanism of X. fastidiosa pathogenesis isstill unclear, disease development depends on the ability of thebacterium to multiply and become systemic in the host (3).

Cell attachment and biofilm formation are determinants forhost infection by pathogenic bacteria (45). Genes in X. fastidiosaencode various surface-associated proteins that are thought to beimportant for attachment, intraplant migration, host coloniza-tion, and biofilm formation. Proteins involved in these processesinclude fimbrial adhesins, which are components of type I andtype IV pili that are located at the same cell pole (35). The type Ipilus is responsible for cell anchorage to surfaces (15), while thetype IV pilus extends and retracts to assist in cell twitching motility(35). Afimbrial adhesin proteins have also been implicated in bac-terial cell-to-cell attachment and biofilm maturation (22).

Bacterial biofilms provide protection to the community of cellsfrom antimicrobial compounds and dehydration, promote cell-to-cell signaling interactions, and help to optimize nutrient up-take (23). Inside the host environment, factors such as the avail-ability of water and nutrients can limit the formation of biofilms(43). Different environments vary in their concentrations of sol-utes; therefore, bacteria have evolved mechanisms for regulatingcellular nutrient concentrations (50). For instance, bacterial bio-films have been thought to improve nutrient acquisition through

the formation of exopolysaccharides (EPSs), which have a highcapacity for ion retention, a strategy particularly important undernutrient-dilute conditions, such as those of xylem sap (7). For X.fastidiosa, the role of biofilm formation in bacterial infection hasbeen demonstrated (3, 36); however, the specific host factors de-termining the dynamics of biofilm formation are still under inves-tigation.

X. fastidiosa survives in the xylem, which is the conduit for thedistribution of nutrients from the root to the leaves, includingmicro- and macronutrients required for multiple functionswithin prokaryotic and eukaryotic cells (24, 33). These essentialminerals are under tight homeostatic control because, althoughrequired, they are also toxic at high concentrations. The availabil-ity of minerals and trace elements affects host-pathogen interac-tions, especially pathogen survival, the expression of virulencetraits, and host physiology. Calcium levels modulate biofilmstructures in Vibrio cholerae (27), the expression of the type IIIsecretion apparatus and effector proteins in Yersinia pestis (12)and Pseudomonas aeruginosa (9), and the production of hydrolyticenzymes (polygalacturonase and pectate lyase) in Pectobacteriumcarotovorum (20), all of them considered virulence determinants.In addition, in animal and plant hosts, calcium modulates defenseresponses, based on regulatory systems that rely on this metal as asecondary messenger (29, 52).

Calcium, magnesium, and iron have all been implicated in the

Received 9 August 2011 Accepted 12 December 2011

Published ahead of print 22 December 2011

Address correspondence to Leonardo De La Fuente, lzd0005@auburn.edu.

Copyright © 2012, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.06501-11

0099-2240/12/$12.00 Applied and Environmental Microbiology p. 1321–1331 aem.asm.org 1321

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

X. fastidiosa infection process. However, it has been suggested tobe nonspecific in some cases (e.g., Ca and Mg), with these ele-ments playing a role as bridges for the adhesion between nega-tively charged bacterial cells and xylem vessels (30). While previ-ous studies using gene expression showed that Fe regulatesvirulence factors such as the type IV pilus and bacteriocins (55),other studies showed that cells in biofilms were more resistant toCu than planktonic cells (44) and that the Cu and Zn present inxylem fluid are correlated with the in vitro growth of X. fastidiosa(1). Moreover, a Zn-protease is induced in X. fastidiosa in citrus toutilize free amino acids in the xylem as nitrogen and carbonsources (42). Those studies suggested that the modulation of themineral content of the xylem where X. fastidiosa is colonizing thehost plant will have an impact on the virulence of this bacterium.

In the present work, we studied the effects of the concentrationof calcium on X. fastidiosa biofilm formation and twitching move-ment in vitro. Different X. fastidiosa traits implicated in the for-mation of biofilms, including cell attachment to surfaces, cell-to-cell aggregation, twitching motility, and the production of EPS,were analyzed. The presence of Ca enhances biofilm formation,cell attachment, and motility under in vitro conditions. No evi-dence for an effect of Ca on EPS production was found. Theseresults suggest that the role of Ca is related to the initial steps in theformation of biofilms.

MATERIALS AND METHODSBacterial strains and culture conditions. Xylella fastidiosa strain Te-mecula (ATCC 700964) was used as the wild-type (WT) strain in thisstudy. Four mutant strains previously obtained by random mutagenesisusing an EZ::TN Tn5 transposon system were also used: fimA (PD_0062)null cells, lacking type I pili (35); pilB (PD_1927) null cells, lacking type IVpili (35); and a fimA pilO (PD_0062 PD_1693) double mutant, lackingtype I and type IV pili (31). For confocal microscopy studies, a greenfluorescent protein (GFP)-expressing mutant of the WT (strain KLN59.3)(36) was used. All X. fastidiosa strains were grown on PW agar solid me-dium (10) or in PD2 liquid medium (11) at 28°C. Stocks of X. fastidiosacultures were stored in PW broth plus 20% glycerol at �80°C.

Biofilm quantification in 96-well plates. The biofilm formation by X.fastidiosa WT and mutant strains was assessed according to methods de-scribed previously by Zaini et al. (54), with some modifications. Briefly,cells cultured for 7 days on PW plates were scraped from the plates andsuspended in PD2 medium (optical density at 600 nm [OD600] of 0.8).Sterile polystyrene 96-well plates containing 200 �l PD2 per well wereinoculated with 10 �l of cell suspension. After 5 days of incubation at28°C, the total number of cells, planktonic growth (cells in suspension),and biofilm growth (cells adhered to the substrate) were estimated bymeasuring the OD600. Planktonic growth was quantified by transferringthe broth containing cells in suspension onto a new plate. The original96-well plate was then rinsed 3 times with Milli-Q water and stained with0.1% crystal violet to quantify biofilm growth (54).

Supplementation of media with metals and chelators. The biofilmformation of cells grown in PD2 medium supplemented with metals andchelators at different concentrations was assessed as described above. PD2medium was supplemented with CaCl2, CoCl2, CuSO4, FeSO4, KCl,MgSO4, Mn2(SO4)3, or ZnCl2 at a range of concentrations from 0.1 mMto 2.5 mM. Additionally, biofilm formation was evaluated in the presenceof either the extracellular calcium chelator EGTA at a range of concentra-tions from 0.063 mM to 1.5 mM or the cell-permeant Ca chelator 1,2-bis(o-aminophenoxy)ethane-N,N,N=,N=-tetraacetic acid acetoxymethylester (BAPTA/AM) at a range of concentrations from 4.2 �M to 100 �M.For all experiments, each plate included a control row (n � 12) of cellsgrown in nonsupplemented PD2 medium. All experiments were repli-cated at least three times.

Metal quantification. X. fastidiosa or PD2 medium was analyzed byinductively coupled plasma with optical emission spectroscopy (ICP-OES) (7,100 DV; Perkin-Elmer, Waltham, MA) with simultaneous mea-surements of B, Ca, Co, Cu, Fe, K, Mg, Mn, Mo, Ni, P, S, and Zn. Cellswere collected by centrifugation and washed three times in ultrapure,metal-free water and once in 0.5 mM EDTA. Equal numbers of cells wereacid digested for 1 h in sealed tubes at 95°C in 150 �l of metal-free con-centrated nitric acid (Optima; Fisher Scientific) as previously described(53). As controls, blanks of nitric acid were digested in parallel. Digestedsamples were centrifuged, and supernatants were diluted by using ultra-pure, metal-free water. Metal concentrations were determined by com-paring intensities to a standard curve created from certified metal stan-dards LPC Standard 1 (Spex Certiprep, Metuchen, NJ). The available Cawas measured by using the fluorescent probes Fluo-5N and Fluo-5NAM(Invitrogen, Carlsbad, CA). Standard solutions of 5 mM were prepared indimethyl sulfoxide (DMSO) and then diluted into aqueous solution foruse. Relative fluorescence (Rf) was measured with an LS50B fluorescencespectrometer (Perkin-Elmer, Waltham, MA). Excitation was set at 485nm (slit width, 3 nm), and emission was scanned from 500 to 600 nm withan Rf read at a maximum intensity of 520 nm (slit width, 3 nm).

Assessment of bacterial cell-to-cell aggregation. Bacterial aggrega-tion was examined in biofilm cells formed at the air-liquid interface ofglass culture tubes containing 5 ml of liquid cultures of WT X. fastidiosa inPD2 medium incubated with shaking at 28°C for 7 days. Cells were har-vested and suspended in 200 �l of Milli-Q water to an equal startingOD600 of 0.8. Cell-to-cell aggregation was established via the “settlingrate,” using a UV-Vis 2450 spectrophotometer (Shimadzu Scientific In-struments, Columbia, MD). The cell suspension was homogenized vigor-ously by pipetting and placed into the spectrophotometer, where theOD600 was continuously measured for 1 min. The settling rate was calcu-lated as the slope of the linear portion of the decreasing change in theOD600 over time.

Strength of bacterial surface attachment in microfluidic chambers.The strength of cell attachment to surfaces (adhesion force [AF]) wasevaluated with different chemical treatments by using microfluidic cham-bers as previously described (15). Briefly, for the fabrication of microflu-idic chambers, photolithography was used to etch a pattern onto a siliconwafer for casting with polydimethylsiloxane (PDMS). Chambers con-sisted of a molded PDMS body sandwiched between a cover glass and asupporting glass microscope slide. The design used in these experiments(15) has two parallel microchannels (80 �m wide by 3.7 cm long by 50 �mdeep), which each have two inlets to allow the separate entry of media andbacteria and an outlet to allow media to flow out the other end. Forassessments of the adhesion force, the medium treatments evaluated in-cluded nonsupplemented PD2 medium (control) and PD2 medium sup-plemented with either 2 mM Ca, 1.5 mM EGTA, 75 �M BAPTA/AM, or100 �g ml�1 tetracycline. Eight-day-old bacterial cultures were scrapedfrom PW plates, suspended with each medium treatment, and homoge-nized by vortexing. The bacterial suspension was introduced into the mi-crofluidic channels (�22°C) through tubing connected to the chamber.Medium flow was controlled by an automated syringe pump (Pico Plus;Harvard Apparatus, Holliston, MA). The microfluidic chambers weremounted onto a Nikon Eclipse Ti inverted microscope (Nikon, Melville,NY) and observed at 40� phase-contrast optics to monitor cell attach-ment. After approximately 2 h at a medium flow speed of 0.25 �l min�1,when the cells had attached, the flow speed was adjusted to 1.0 �l min�1

for 1 h to remove nonattached cells. The medium flow was then sequen-tially increased every 1 min by 10 �l min�1 from 1 to 120 �l min�1 togradually remove attached cells. Time-lapse microscopy images were ac-quired every 5 s during this time with a Nikon DS-Q1 digital camera(Nikon, Melville, NY) controlled by NIS-Elements software (Nikon, Mel-ville, NY) to record the detachment of X. fastidiosa cells. The number ofattached cells remaining in each frame was scored by using NIS-Elementssoftware. Total cell counts were averaged across the 12 frames collectedeach minute to obtain the count for each flow rate. The cell adhesion force

Cruz et al.

1322 aem.asm.org Applied and Environmental Microbiology

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

was calculated according to methods described previously by De La Fu-ente et al. (15). Each treatment was evaluated at least 7 times in indepen-dent microfluidic chamber experiments.

Quantification of exopolysaccharide production by biofilm cells.The production of EPS in biofilm cells formed at the air-liquid interface ofglass culture tubes containing 5 ml of X. fastidiosa WT liquid cultures inPD2 medium incubated with shaking at 28°C for 7 days was examined.PD2 broth was supplemented with either 2 mM Ca or 1.5 mM EGTA. Therelative quantification of EPS was performed by a phenol-sulfuric acidassay, which determines the total sugar content of the cells as an indicationof EPS production (18). Glucose was used for sugar standard curves, andnonsupplemented PD2 medium was used as the control for the treat-ments. Each treatment was replicated 10 times. For the quantification ofthe number of X. fastidiosa biofilm cells in each treatment, qPCR wasperformed. DNA from one-third of the total volume of the biofilm sus-pension was extracted by using a modified cetyltrimethylammonium bro-mide (CTAB) protocol (17). qPCR was performed by using a previouslydesigned set of X. fastidiosa-specific primers and a TaqMan probe (21).qPCR mixtures (25-�l total volume) consisted of 1� ABsolute BlueQPCR ROX mix (ABgene-Thermo Fisher Scientific, Surrey, United King-dom), 0.4 �l 6-carboxyfluorescein-Black Hole quencher 1 (FAM-BHQ1)-labeled probe, 0.2 �M each primer, and 1 �l DNA. Samples were ampli-fied on an ABI 7500 real-time PCR system (Applied Biosystems, FosterCity, CA), using the following cycling parameters: 95°C for 1 min, fol-lowed by 40 cycles of 95°C for 15 s and 60°C for 1 min. All qPCR runsincluded a 4-point standard curve prepared from 10-fold serial dilutionsof DNA extracted from an X. fastidiosa cell suspension of a known con-centration. Cells in the suspension were enumerated by spread platingserial dilutions onto PW agar and counting the CFU produced after a2-week incubation at 28°C. qPCR amplification efficiencies were compa-rable to those described previously (21), and standard curves had r2 valuesof �0.97.

Assessment of biofilm structure by confocal laser scanning micros-copy. Biofilm structure assessments were performed using glass slidesprepared as previously described by O’Toole et al. (38), with a few mod-ifications. Briefly, 20 �l of a cell suspension (OD600 � 0.8) of KLN59.3 (X.fastidiosa GFP) was inoculated into 50-ml conical polystyrene tubes, eachcontaining a 22- by 22-mm microscope slide submerged in 10 ml of PD2broth or PD2 medium supplemented with either 2 mM CaCl2 or 1.5 mMEGTA. Each treatment was replicated three times. The tubes were incu-bated at 28°C for 15 days, allowing the formation of biofilms on the sur-faces of the slide.

The biofilm was observed with a Nikon Eclipse Ti confocal laserscanning inverted microscope (Nikon, Melville, NY), using a 60� oilimmersion objective with 488-nm excitation wavelengths. The three-dimensional architecture of the biofilm was assessed by z-stack imageswith 1-�m increments, and images were acquired with a CoolSNAP HQ2camera (Photometrics, Tucson, AZ). Three-dimensional reconstructionsof the images and measurements of biofilm thickness were performedwith NIS-Elements AR software, version 3.0 (Nikon, Melville, NY).

Quantification of twitching motility on agar plates and microfluidicchambers. Twitching motility was assessed on agar plates. WT cells werespotted in quadruplicate onto PW agar plates supplemented with thefollowing treatments: 2 mM Ca, 1.5 mM EGTA, and a nonsupplementedPW control. After 7 days of incubation at 28°C, the colony peripheralfringe morphology was examined under the microscope, and the fringewidth was measured (n � 3) for each of the 4 colonies by using NIS-Elements software. Each treatment was replicated on 3 plates, for a total of36 measurements per treatment.

Additionally, the twitching motility of WT cells in microfluidic cham-bers mounted onto an inverted microscope was assessed as previouslydescribed (14). Cells were introduced into the chambers and incubatedfor 2 h before time-lapse images were recorded every 30 s for 1 h. Themovement of X. fastidiosa cells was quantified by tracking their positions

frame by frame, and the cell speed was calculated by measuring the up-stream displacement with respect to time.

Statistical analyses. All experiments were carried out in a randomizedcomplete design (RCD). Data were imported, linearized, and tabulatedwith SAS 9.2 (SAS Institute Inc.) using GLIMMIX, which combines linearregression and analysis of variance models with normal errors, to checkthe residuals. Least-squared means were compared at a level of signifi-cance of a P value of �0.05.

RESULTSEffect of metal supplementation on biofilm formation. Whendifferent metals were added to PD2 medium, Ca and Fe produceda strong increase in biofilm formation in 96-well plate assays com-pared to the nonsupplemented control (Fig. 1A). A positive cor-relation between the Ca concentration and biofilm formation wasobserved for the complete range of Ca concentrations tested. At2.5 mM additional Ca, the amount of biofilm formed was 500%higher than the amount formed with the control treatment. Un-

FIG 1 Biofilm quantification and planktonic growth of the X. fastidiosa WTstrain in PD2 medium supplemented with different metals. Values are ex-pressed relative to the amount of biofilm obtained with X. fastidiosa cells innonsupplemented PD2 medium (considered 100%). (A) Biofilm growth; (B)planktonic growth. The experiment was repeated three times, and each exper-iment contained three replicates. Data from a representative experiment arepresented.

Effects of Calcium on Xylella fastidiosa

March 2012 Volume 78 Number 5 aem.asm.org 1323

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

like for Ca, the strength of the effect of Fe on biofilm formationdecreased for concentrations of Fe higher than 1.8 mM, and at 2.5mM, the increase in biofilm formation was just 150% comparedwith the formation in nonsupplemented medium (Fig. 1A). Incontrast, increases in the concentrations of Cu, Co, Mn, and Zninhibited the accumulation of biofilm to 50% compared to cellsgrown in nonsupplemented PD2 medium. No significant differ-ences in biofilm accumulation were found for additions of K andMg (Fig. 1A). Planktonic growth was considerably less affected bythe addition of these metals. Co, K, Mg, and Mn produced a re-duction in planktonic growth of �20% (Fig. 1B). None of themetals significantly increased total growth compared to the con-trol (data not shown).

Effects of calcium chelators on biofilm formation of WT X.fastidiosa. The biofilm formation in 96-well plates in the presenceof EGTA and BAPTA/AM, extra- and intracellular calcium chela-tors, respectively, was quantified. EGTA reduced the formation of

biofilm to approximately 40% at all tested concentrations (0.063mM to 1.5 mM), while the planktonic growth was similar (�90%)to that found in the cells incubated in control medium (Fig. 2A).Similarly, the addition of BAPTA/AM, which is active as a chelatoronly when the AM ester is cleaved by the intracellular esterases,caused a negative effect on biofilm formation, reducing it to ap-proximately 40% at concentrations of 30 �M or higher (Fig. 2B).Under the same conditions, the addition of BAPTA/AM reducedplanktonic growth to approximately 80% compared to that incontrol medium.

To test the efficiency and localized effects of the chelators wedetermined the amount of available Ca using Fluo-5N, a fluores-cent BAPTA derivative. By comparing the levels of Fluo-5N-reactive Ca in the media, we detected equal Ca levels in PD2 me-dium and PD2 medium with BAPTA/AM (Table 1). This was asexpected, as the AM ester derivitization prevents Ca chelation un-til it is taken into cells. However, we detected a 10-fold decrease inthe amount of available Ca in PD2 medium with EGTA based onFluo-5N fluorescence. In PD2 medium, PD2 medium with EGTA,and PD2 medium with BAPTA/AM, the concentrations of totalCa measured by ICP-OES were equal, while the addition of 2 mMCa was additive to the base medium (Table 1). For comparison,the available reactive Ca was converted to a concentration by cre-ating a standard curve using a titration of CaCl2 solution withFluo-5N.

Analysis of intracellular and extracellular Ca in X. fastidiosabiofilms was achieved by using a Fluo-5N/AM derivative andFluo-5N. The intracellular Ca concentration in cells cultured inPD2 medium plus Ca was elevated 6-fold relative to that in PD2medium alone, while BAPTA/AM treatment decreased the avail-able intracellular Ca �2-fold (Table 2). This assay assumes rela-tively equal loading in cells, and microscopic analysis confirmedthat fluorescence was a result of the intracellular signal from X.fastidiosa (data not shown). Using cell-impermeable Fluo-5N, thewashed cells were assayed for extracellular Ca levels. The amountof available extracellular Ca in EGTA medium was decreased to0.5-fold of that in PD2 medium, while the amount of availableextracellular Ca with BAPTA/AM was 1.2-fold higher than that inPD2 medium, and that with Ca-treated medium was 4-fold higherthan that in PD2 medium (Table 2).

Effect of Ca on autoaggregation and cell-to-cell attachment.Autoaggregation was previously defined for X. fastidiosa as theformation of spherical compact aggregates of cells attached toeach other that precedes biofilm formation (13). The autoaggre-

FIG 2 Effect of removal of Ca on biofilm and planktonic growth of the X.fastidiosa WT strain. Ca and the extracellular chelator EGTA (A) and the in-tracellular chelator BAPTA/AM (B) were added to the media, and biofilmgrowth as well as bacterial growth were quantified. Values are expressed rela-tive to the amount of biofilm obtained with X. fastidiosa cells in nonsupple-mented PD2 medium (considered 100%). The experiment was repeated threetimes, and each experiment contained three replicates. Data from a represen-tative experiment are presented.

TABLE 1 Comparison of amounts of available and total calcium invarious PD2 media

TreatmentMean Rf ofavailableCa � SEMa

Available Caconcn (mM)b

Mean Ca concn(mM) � SEMdetermined byICP-OESc

PD2 127 � 4 0.083 0.073 � 0.002PD2 � Ca 2,460 � 15 1.59 2.176 � 0.080PD2 � EGTA 11 � 2 0.0073 0.074 � 0.002PD2 � BAPTA/AM 136 � 4 0.088 0.074 � 0.002a Determined at an excitation wavelength of 485 nm and an emission wavelength of520 nm.b Calculated from Fluo-5N reactivity with CaCl2 standard solution in 10 mM Tris (pH 7.0).c Inductively coupled plasma-optical emission spectroscopy (ICP-OES) was used tomeasure total calcium concentrations.

Cruz et al.

1324 aem.asm.org Applied and Environmental Microbiology

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

gation process observed in microfluidic chambers was initiated at2 days after the inoculation of PD2 medium or medium supple-mented with 2 mM Ca. Cells inoculated into PD2 medium sup-plemented with 1.5 mM EGTA showed considerably less autoag-gregation than in the control medium. At 5 days postinoculation,the 2 mM Ca treatment had the highest level of autoaggregation,followed by PD2 medium, while almost no aggregation was ob-served for the treatment with 1.5 mM EGTA during this period

(Fig. 3A). We also observed that when Ca was added to the media,X. fastidiosa increased polar attachment to the substrate, as op-posed to lying flat against the glass surface. On average, approxi-mately 60% of cells grown in PD2 medium supplemented with Cashowed polar attachment, compared to control PD2 medium,where only 30% of the cells were in this position.

Analysis of cell settling rates, as a measurement of cell-to-cellaggregation, demonstrated that cells from medium containing Cashowed significantly higher sedimentation rates (0.31 OD600

min�1) (P � 0.0001) than cells grown in nonsupplemented me-dium (mean � 0.15 OD600 min�1), EGTA-containing medium(mean � 0.08 OD600 min�1), or BAPTA/AM-containing me-dium (mean � 0.04 OD600 min�1). Similarly, significant differ-ences were found between nonsupplemented medium and eitherEGTA-containing medium (P � 0.03) or BAPTA/AM-containingmedium (P � 0.003). X. fastidiosa cells grown in medium contain-ing EGTA displayed a nonsignificant difference in the settling ratecompared to that in medium containing BAPTA/AM (P � 0.212)(Fig. 3B).

Effect of calcium on cell adhesion force. Adhesion force (AF)comparisons among WT X. fastidiosa cells in media containing 2mM Ca (AF � 223 pN), 1.5 mM EGTA (AF � 190 pN), and 75�M BAPTA/AM (AF � 179 pN) and nonsupplemented PD2 me-dium (AF � 160 pN) showed that the presence of Ca significantlyincreases the adhesion force of the cells compared with those ofcells in nonsupplemented PD2 medium (P � 0.008) and cellstreated with BAPTA/AM (P � 0.05) (Fig. 4). In contrast, no sig-nificant differences were observed between the adhesion forces ofCa- and EGTA-treated cells (P � 0.14) or between the adhesionforces of cells in nonsupplemented PD2 medium, cells in EGTA-containing medium (P � 0.15), and cells treated with BAPTA/AM(P � 0.39).

Effect of calcium on tetracycline-treated cells. Tetracyclinewas added to the media as a bacteriostatic agent to test the hypoth-esis that de novo protein synthesis is needed to observe the effect of

FIG 3 Evaluation of cell-to-cell aggregation of X. fastidiosa at different Caconcentrations. (A) Time-lapse micrographs showing the formation ofspherical compact autoaggregates inside microfluidic chambers. Imageswere captured at 1, 2, 3, and 5 days postinfection. (B) Settling rates of WTX. fastidiosa as a measure of cell-to-cell aggregation. Biofilm cells wereobtained from growth cultures in PD2 medium, PD2 medium plus 2 mMCa, PD2 medium plus 1.5 mM EGTA, and 75 �m BAPTA/AM. Differentletters on the bars indicate significant differences according to the GLIM-MIX procedure (P � 0.05). Error bars represent standard errors of themeans (n � 4).

TABLE 2 Comparison of amounts of intracellular and extracellular Caand total calcium in various PD2 media

Treatment

Mean Rf � SEMa

Extracellular Ca(n � 4)

Intracellular Ca(n � 4)

PD2 86 � 8 12 � 1PD2 � Cab 301 � 6 83 � 14PD2 � EGTAc 46 � 10 9 � 2PD2 � BAPTA/AMd 104 � 7 5 � 1a Determined at an excitation wavelength of 485 nm and an emission wavelength of520 nm.b A total of 2 mM CaCl2 was added to PD2 medium.c A total of 1.5 mM the extracellular calcium chelator EGTA was added to PD2 medium.d A total of 75 �M the intracellular calcium chelator BAPTA/AM [1,2-bis(o-aminophenoxy)ethane-N,N,N=,N=-tetraacetic acid acetoxymethyl ester] was added to PD2 medium.

FIG 4 Adhesion force of the X. fastidiosa WT strain in nonsupplementedPD2 medium, PD2 medium plus 2 mM Ca, PD2 medium plus 1.5 mMEGTA, and PD2 medium plus 100 �M BAPTA/AM in microfluidic cham-bers. Different letters on the bars indicate significant differences accordingto the GLIMMIX procedure (P � 0.05). Error bars represent standarderrors of the means (n � 7).

Effects of Calcium on Xylella fastidiosa

March 2012 Volume 78 Number 5 aem.asm.org 1325

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

Ca on attachment to surfaces. The evaluation of the adhesion forceof tetracycline-treated cells indicated that in Ca-supplementedmedium, the adhesion force of tetracycline-treated cells (AF �172 pN) was significantly lower than the adhesion force of non-treated cells (AF � 223 pN) (P � 0.029) (Fig. 5). Comparisons ofthe adhesion forces of tetracycline-treated cells in Ca-supplemented medium (AF � 172 pN) and nonsupplementedmedium (AF � 198 pN) indicated no significant differences (P �0.30). These results suggest that protein synthesis is needed for Cato have an effect on the surface attachment of X. fastidiosa.

Effect of Ca on biofilm formation and adhesion force of X.fastidiosa attachment-defective mutants. Assessments of biofilmformation of X. fastidiosa mutant cells in 96-well plates indicated apositive correlation between the increase in the concentration ofCa and the formation of biofilms for the pilB (type IV pilus-defective) and fimA (type I pilus-defective) mutants (Fig. 6A). Inthe fimA mutant, biofilm production was increased by 600% at 2mM Ca, and biofilm production was increased by 350% for thepilB mutant at 2.5 mM Ca. For the double mutant defective in typeI and type IV pili (fimA pilO mutant), the addition of Ca did notsignificantly affect the formation of the biofilm. The planktonicgrowth of the X. fastidiosa mutants was not affected by the range ofconcentrations of Ca supplements tested (Fig. 6B).

The microfluidic adhesion force assay indicated that pilB mu-tant cells had a significantly increased adhesion force in the pres-ence of 2 mM Ca (AF � 211 pN) compared with that of cells incontrol (nonsupplemented) medium (AF � 178 pN) (P � 0.044).No significant differences in adhesion force were found for thefimA (AF � 130 pN) and fimA pilO (AF � 118 pN) mutants inCa-supplemented medium compared to control medium (AF �125 pN and 113 pN, respectively) (P � 0.653) (Fig. 6C).

Exopolysaccharide production in the presence of calcium.An indirect measurement of EPS production was performed byusing a phenol-sulfuric acid assay. This assay quantifies total sug-ars in glucose equivalents and is used as an indicator of EPS pro-duction. The sugar contents per cell in biofilms from PD2 me-

FIG 6 Effects of Ca on biofilm formation, growth, and adhesion force of X.fastidiosa mutant strains defective in surface structures. (A) Biofilm formation inthe presence of Ca (0.1 mM to 2.5 mM). Values are expressed relative to theamount of biofilm obtained with X. fastidiosa cells in nonsupplemented PD2 me-dium (considered 100%). The experiment was repeated three times, and data froma representative experiment are presented. (B) Corresponding planktonic growthunder the same conditions as those described above for panel A. (C) Adhesionforce in PD2 medium and in PD2 medium supplemented with 2 mM Ca mea-sured inside microfluidic chambers. Different letters on the bars indicate signifi-cant differences according to the GLIMMIX procedure (P � 0.05). Error barsrepresent standard errors of the means (n � 6).

FIG 5 Adhesion force of tetracycline-treated and nontreated X. fastidiosa WTcells in Ca-supplemented and nonsupplemented PD2 media. Different letterson the bars indicate significant differences according to the GLIMMIX proce-dure (P � 0.05). Error bars represent standard errors of the means (n � 7).

Cruz et al.

1326 aem.asm.org Applied and Environmental Microbiology

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

dium and PD2 medium supplemented with 2 mM Ca or 1.5 mMEGTA were not significantly different (P � 0.696) (Fig. 7A) re-gardless of the amount of biofilm observed among the treatments(Fig. 7B).

Effect of Ca on the biofilm three-dimensional structure. Anexamination of z-stack three-dimensional images and measure-ments of biofilm thickness indicated that X. fastidiosa biofilm cellsgrown in medium supplemented with 2 mM Ca exhibited a sig-nificantly thicker biofilm (mean � 31.04 �m) than cells in me-dium containing 1.5 mM EGTA (mean � 9.95 �m) and in controlnonsupplemented medium (mean � 12.93 �m) (P � 0.001).Similarly, significant differences were observed between EGTA-supplemented and control nonsupplemented media (P � 0.001)(Fig. 8A). Two- and three-dimensional observations of the bio-films obtained with the different treatments also showed differ-ences in cell distribution and orientation (Fig. 8B to F). The bio-film obtained from cells in the presence of Ca was continuous andcompact throughout the slide, and cells exhibited a vertical polarorientation (Fig. 8D and E). Similarly, in nonsupplemented me-dium, the biofilm was formed as a continuous structure; however,cell-to-cell spaces were larger, and cells were horizontally oriented(Fig. 8B and C). In EGTA-containing medium, the biofilm wasformed as discontinuous clusters of cells exhibiting the largestcell-to-cell separations (Fig. 8F and G).

Effect of calcium on cell twitching motility. The effect of Caon cell motility, as determined by the colony fringe width on agarplates, indicated that colonies grown on PW agar supplementedwith 2 mM Ca (mean � 123.55 �m) had a 1.5-fold increase infringe size compared to colonies grown on plain PW agar(mean � 80.65 �m) (P � 0.001) and an 11-fold increase com-

pared with colonies grown on PW agar supplemented with 1.5mM EGTA (mean � 11.40 �m) (P � 0.001) (Fig. 9A and B).

The movement of the cells quantified in the microfluidicchambers correlated with the results obtained from the colonyfringe observations. In the presence of Ca, cells moved at higherspeeds (mean � 0.53 �m min�1) than did those in PD2 medium(mean � 0.32 �m min�1) (P � 0.001) and in PD2 medium sup-plemented with 1.5 mM EGTA (mean � 0.044 �m min�1) (P �0.001) (Fig. 9C).

DISCUSSION

The availability of mineral nutrients can act as a stimulus, elicitingchanges in bacteria that can promote the establishment and devel-opment of biofilms in plant-associated (8) and animal-associated(41) bacteria. Metals are important constituents of the xylem sap,one of the natural environments of X. fastidiosa (40). Therefore,we hypothesized that a variation in metals would have an impacton X. fastidiosa colonization and biofilm formation. Calcium lev-els specifically affected biofilm production, so we focused on re-fining the role of Ca in X. fastidiosa biofilm formation.

Divalent cations, including Ca and Mg, were previously impli-cated in the accumulation of biofilms by different bacteria. Pri-marily, these cations were suspected to act as cross-bridging mol-ecules that increase the stability of the biofilm matrix (4, 46).However, they also play a role in the regulation of bacterial geneexpression related to virulence factors and biofilm formation (39,47). For example, physiological changes related to adhesion weredemonstrated previously for marine Pseudoalteromonas sp.,where Ca influences the production of extracellular matrix mate-rial associated with biofilms (47), and for Pseudomonas putida,where the presence of Ca-binding membrane proteins are in-volved in cell attachment to seeds (19). In Erwinia carotovora, highlevels of Ca repress the expression of PehA, an endopolygalactu-ronase that is one of its main virulence determinants (20). Theopposite effect, i.e., an enhancement of virulence, has been seenfor Pseudomonas aeruginosa, where the addition of Ca increasesthe production of extracellular proteases and increases the biofilmthickness by enhancing the expressions of the alginate biosyn-thetic genes, main constituents of the extracellular matrix (47).Results from this study show that Ca is involved in the regulationof X. fastidiosa biofilm formation, cell surface attachment, andtwitching motility.

Bacterial attachment to surfaces is important in the early stepsof biofilm development. Biofilm formation begins with the tran-sient attachment of cells, followed by stable attachment to sur-faces. In advanced stages, cells increase the production of EPS tocreate a matrix that holds together the mature biofilm (25, 47, 48).Experiments from the present study indicate that the addition ofCa significantly increases the strength of the attachment of thecells to the surface. Previous investigations into the mechanism ofX. fastidiosa cell adhesion using X-ray microanalyses of occludedcitrus xylem vessels led Leite et al. to propose a model where Caand Mg ions act as cross-binding molecules that create a bridgebetween the negatively charged xylem vessels and the negativelycharged bacterial EPS (30). According to this model, extracellularcalcium is responsible for cell attachment, acting as an externalphysical force. Our study shows that in addition to the “bridging”effect that can occur with Ca, a metabolic-dependent effect is alsoresponsible for the influence of Ca on biofilm and movement.

The transient exposure of X. fastidiosa to the extracellular ch-

FIG 7 EPS produced by biofilm cells grown in nonsupplemented PD2 me-dium and PD2 medium supplemented with 2 mM Ca and 1.5 mM EGTA. (A)Measurement of EPS in cells forming a biofilm, quantified as the amount ofglucose equivalents. Different letters on the bars indicate statistically signifi-cant differences according to the GLIMMIX procedure (P � 0.05). Error barsrepresent standard errors of the means (n � 10). (B) Images of biofilm formedon the glass surface of Erlenmeyer flasks with cultures of the X. fastidiosa WTstrain with different concentrations of Ca.

Effects of Calcium on Xylella fastidiosa

March 2012 Volume 78 Number 5 aem.asm.org 1327

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

elator EGTA did not produce a considerable reduction in the ad-hesion force compared to that of Ca-treated cells. Only after con-tinuous culturing in EGTA (e.g., in the crystal violet assay) did wesee strong effects. This discrepancy could be explained by differ-ences in the incubation times of these systems and the static versusdynamic natures of the experiments. In the microtiter plate assay,cells incubated for 5 days have to overcome several division cyclesunder low-Ca conditions in the same batch of medium containedin the well; thus, new cells should contain smaller amounts ofintracellular Ca. In contrast, in the microfluidic chamber assays,cells were exposed to reduced Ca concentrations just for the lengthof the experiment (a few hours), and the medium was constantlyreplenished inside the channels. However, the fact that cellstreated with BAPTA/AM showed a significant reduction in theadhesion force compared with that of Ca-treated cells, showedreduced cell-to-cell adhesion, and produced significantly less bio-film in the static assay strongly suggests that other mechanismsthat require intracellular Ca are contributing to the formation ofthe biofilm.

The measurements of adhesion force in tetracycline-treated X.fastidiosa cells further corroborate the regulatory effect of Ca oncell adhesion. Since de novo protein synthesis is compromised by

tetracycline treatment, which inhibits bacterial protein synthesisby preventing the union of tRNA with the ribosome (6), any re-sponse to Ca under these conditions would be more strongly as-sociated with physical cell adhesion. No differences were found inthe adhesion forces between translation-arrested cells in Ca-supplemented PD2 medium and those in nonsupplemented PD2medium, indicating that the effect of Ca on the increase of theadhesion force requires cells to be metabolically active. Moreover,the results obtained with the fimA pilO double mutant missingtype I and type IV pili support this hypothesis. The double mutantdid not respond to Ca in static cultures when the biofilm wasquantified in microtiter plates, and Ca did not change the force ofadhesion to a substrate in microfluidic chambers. If the effect ofCa was due purely to extracellular electrostatic interactions at thecell-substrate surface level, this effect should have been observedwith this mutant. The role of Ca in biofilm formation and specif-ically in cell attachment requires metabolically active cells and thepresence of type I and type IV pili.

Static biofilm production by the X. fastidiosa pilB and fimApilus mutants (type I- and type IV-defective mutants, respec-tively) responded to the addition of Ca. However, only the type IVpilus mutant (pilB) responded by increasing the force of adhesion

FIG 8 Biofilm architecture and confocal scanning microscopy images of biofilm formed on microscope slides. (A) Measurements of biofilm thicknesses ofbiofilms formed under the conditions mentioned below. (B and C) Two-dimensional (B) and three-dimensional (C) images of biofilm formed in nonsupple-mented PD2 medium. (D and E) Two-dimensional (D) and three-dimensional (E) images of biofilm formed in PD2 medium supplemented with 2 mM Ca. (Fand G) Two-dimensional (F) and three-dimensional (G) images of biofilm formed in PD2 medium containing 1.5 mM EGTA. Different letters on the barsindicate significant differences according to the GLIMMIX procedure (P � 0.05). Error bars represent standard errors of the means (n � 9).

Cruz et al.

1328 aem.asm.org Applied and Environmental Microbiology

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

to a substrate. These findings suggest that the strongest increase inattachment is due to an influence of Ca on type I pili. Previously,type I pili were shown to be responsible for the strongest attach-ment to surfaces (13). The role of Ca in type IV pili could berelated to the primary function of these fimbriae that assist in celltwitching motility. The speed of the cell movements and colonyfringe width demonstrate that the presence of high concentrationsof Ca significantly increases the twitching movement of the cells.Although cell movement and cell adhesion seem functionally op-

posite, in early stages of biofilm formation, twitching helps thebacteria colonize, or in later stages, twitching helps bacteria de-colonize one area and recolonize another. Additionally, eventhough type IV pili are responsible mainly for cell motility, theyare able to contribute to cell adhesion.

The function of Ca in bacterial motility via type IV pili has beendocumented for Pseudomonas aeruginosa. The binding and releaseof Ca by a PilY1 calcium-binding site similar to those seen with thecanonical EF-hand Ca-binding site, located in the C-terminal do-main of the protein, elicit the extension and retraction of the typeIV pili required for twitching motility (37). A homologue of the P.aeruginosa pilY1 gene is present in X. fastidiosa. PilY1 is a minorprotein predicted to be associated with the tip of type IV pili (31,34). In X. fastidiosa, the exact molecular mechanism(s) that regu-lates twitching motility has not been investigated, and so far, noprevious reports have found evidence for a role of Ca in type I pilusregulation in X. fastidiosa or in other bacteria.

EPS production is one of the main factors thought to influencethe rate and degree of attachment of microbial cells to differentsurfaces and to each other (16). Ca has been implicated in theregulation of bacterial gene expression eliciting the production ofEPS in Pseudoalteromonas sp. (39). Analyses of X. fastidiosa EPSproduction showed that cells in media containing Ca (PD2 me-dium and PD2 medium supplemented with additional Ca) pro-duce equal amounts of EPS compared to cells grown in mediawhere Ca is chelated by EGTA. This result suggests that Ca doesnot affect the production of EPS by X. fastidiosa cells. In X. fasti-diosa, the quorum-sensing mechanism mediated by a diffusiblesignal factor (DSF) is involved in the regulation of EPS production(49). We anticipated that the production of EPS should be depen-dent on the density of the population attached to the substrate,which is higher in the presence of Ca. However, with the techniqueused in our studies, we were unable to find differential EPS pro-duction that correlated with changing Ca levels.

Another step in the formation of biofilms is the aggregation ofcells after initial attachment. Our experiments showed that in thepresence of exogenous Ca, cells exhibit taller biofilm aggregates,formation occurs at a higher rate, and attachment to surfaces andto other cells is stronger. The adhesion and aggregation of cells areresponsible for the final architecture of a biofilm. Our confocalobservations of biofilms formed on microscope slides showed apolar orientation of the cells with Ca-supplemented treatments.Similar observations were previously made by Killiny et al. (28) forthe biofilm formed in the foregut of an insect vector. In theirexperiments, the biofilm formation process included initial lateralattachment followed by polar attachment, possibly by the actionof the type I pili to increase the surface area of nutrient absorption.The mediation of the polar attachment may also be enhanced bythe effect of Ca on this fimbrial structure.

It should be noted that previous studies have shown that thechemical compositions of the growth media and xylem sap influ-ence X. fastidiosa aggregation and biofilm formation (1, 54).Cheng et al. (5) found previously that PW medium supplementedwith the xylem sap of PD-susceptible grapevines induced in-creased bacterial aggregation and biofilm formation compared tomedium supplemented with the sap of PD-resistant varieties. An-dersen et al. (1) previously analyzed the relationship between sapchemical components and biofilm formation and compared thechemical compositions of PD-susceptible and PD-resistant culti-vars. Increasing concentrations of Ca, Mg, and amino acids from

FIG 9 Effects of Ca on X. fastidiosa cell twitching motility. (A) Colony fringewidth of the X. fastidiosa WT strain cultured on PW agar and PW agar supple-mented with 2 mM Ca and 1.5 mM EGTA. Different letters on the bars indicatestatistically significant differences according to the GLIMMIX procedure (P �0.05). Error bars represent standard errors of the means (n � 10). (B) Repre-sentative micrographs of colony fringes on agar plates with the treatmentsmentioned above for panel A. (C) Assessments of twitching speeds of cellsincubated in PD2 medium and PD2 medium supplemented with 2 mM Ca or1.5 mM EGTA in microfluidic chambers. Different letters on the bars indicatesignificant differences according to the GLIMMIX procedure (P � 0.05). Errorbars represent standard errors of the means (n � 10).

Effects of Calcium on Xylella fastidiosa

March 2012 Volume 78 Number 5 aem.asm.org 1329

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

the PD-resistant level to that present in the PD-susceptible sapcorrelated with an increase in cell aggregation and improvedplanktonic growth. However, this is the first study to focus solelyon Ca as the stimulatory factor.

The present study shows that Ca has multiple roles during invitro biofilm formation by X. fastidiosa. Ca appears to influencebiofilm formation by both extracellular ionic bridging and intra-cellular stimulation that requires protein synthesis. Physiologicalchanges in response to Ca are increased cell attachment, mostlikely via type I pili; increased twitching motility; and increasedcell-to-cell attachment, responsible for aggregation. Ca affects theinitial stages of biofilm development related mainly to cell attach-ment and has a less prominent role in biofilm maturation at laterstages. In vivo experiments that correlate the nutritional status ofthe plant with the development of disease and expression analysesof X. fastidiosa in the presence of Ca are necessary to further elu-cidate the molecular mechanisms involved in the promotion ofbiofilm formation by Ca.

ACKNOWLEDGMENTS

This project was supported by Agriculture and Food Research InitiativeCompetitive Grants Program grant no. 2010-65108-20633 from theUSDA National Institute of Food and Agriculture and the Alabama Agri-cultural Experiment Station (AAES) Hatch Grant Program.

We thank Harvey C. Hoch (Cornell University) and Tom Burr (Cor-nell University) for providing the X. fastidiosa mutant strains and StevenLindow (University of California, Berkeley) for providing the X. fastidiosaGFP-expressing strain. We acknowledge the Auburn University ConfocalFacilities for the use of their microscope. We also acknowledge the assis-tance of Melody Smith in initial experiments with Fluo-5.

REFERENCES1. Andersen PC, Brodbeck BV, Oden S, Shriner A, Leite B. 2007. Influence

of xylem fluid chemistry on planktonic growth, biofilm formation andaggregation of Xylella fastidiosa. FEMS Microbiol. Lett. 274:210 –217.

2. Chang CJ, Donaldson R, Brannen P, Krewer G, Boland R. 2009.Bacterial leaf scorch, a new blueberry disease caused by Xylella fastidiosa.Hortscience 44:413– 417.

3. Chatterjee S, Almeida RPP, Lindow S. 2008. Living in two worlds: theplant and insect lifestyles of Xylella fastidiosa. Annu. Rev. Phytopathol.46:243–271.

4. Chen X, Stewart PS. 2002. Role of electrostatic interactions in cohesion ofbacterial biofilms. Appl. Microbiol. Biotechnol. 59:718 –720.

5. Cheng DW, Lin H, Walker MA, Stenger DC, Civerolo EL. 2009. Effectsof grape xylem sap and cell wall constituents on in vitro growth, biofilmformation and cellular aggregation of Xylella fastidiosa. Eur. J. PlantPathol. 125:213–222.

6. Chopra I, Roberts M. 2001. Tetracycline antibiotics: mode of action,applications, molecular biology, and epidemiology of bacterial resistance.Microbiol. Mol. Biol. Rev. 65:232.

7. Costerton JW, Lewandowski Z, Caldwell DE, Korber DR, LappinscottHM. 1995. Microbial biofilms. Annu. Rev. Microbiol. 49:711–745.

8. Danhorn T, Fuqua C. 2007. Biofilm formation by plant-associated bac-teria. Annu. Rev. Microbiol. 61:401– 422.

9. Dasgupta N, Ashare A, Hunninghake GW, Yahr TL. 2006. Transcrip-tional induction of the Pseudomonas aeruginosa type III secretion systemby low Ca2� and host cell contact proceeds through two distinct signalingpathways. Infect. Immun. 74:3334 –3341.

10. Davis MJ, Purcell AH, Thomson SV. 1980. Isolation medium for Pierce’sdisease bacterium. Phytopathology 70:425– 429.

11. Davis MJ, French WJ, Schaad NW. 1981. Axenic culture of the bacteriaassociated with phony disease of peach and plum leaf scald. Curr. Micro-biol. 6:309 –314.

12. DeBord KL, Galanopoulos NS, Schneewind O. 2003. The ttsA gene isrequired for low-calcium-induced type III secretion of Yop proteins andvirulence of Yersinia enterocolitica W22703. J. Bacteriol. 185:3499 –3507.

13. De La Fuente L, Burr TJ, Hoch HC. 2008. Autoaggregation of Xylella

fastidiosa cells is influenced by type I and type IV pili. Appl. Environ.Microbiol. 74:5579 –5582.

14. De La Fuente L, Burr TJ, Hoch HC. 2007. Mutations in type I and typeIV pilus biosynthetic genes affect twitching motility rates in Xylella fasti-diosa. J. Bacteriol. 189:7507–7510.

15. De La Fuente L, et al. 2007. Assessing adhesion forces of type I and typeIV pili of Xylella fastidiosa bacteria by use of a microfluidic flow chamber.Appl. Environ. Microbiol. 73:2690 –2696.

16. Dharmapuri S, Sonti RV. 1999. A transposon insertion in the gumGhomologue of Xanthomonas oryzae pv. oryzae causes loss of extracel-lular polysaccharide production and virulence. FEMS Microbiol. Lett.179:53–59.

17. Doyle J, Doyle JL. 1987. Genomic plant DNA preparation from freshtissue—CTAB method. Phytochem. Bull. 19:11–15.

18. Dubois MG, et al. 1956. Colorimetric method of determination of sugarsand related substances. Anal. Chem. 28:350 –354.

19. Espinosa-Urgel M, Salido A, Ramos JL. 2000. Genetic analysis of func-tions involved in adhesion of Pseudomonas putida to seeds. J. Bacteriol.182:2363–2369.

20. Flego D, Pirhonen M, Saarilahti H, Palva TK, Palva ET. 1997. Controlof virulence gene expression by plant calcium in the phytopathogen Er-winia carotovora. Mol. Microbiol. 25:831– 838.

21. Francis M, Lin H, Cabrera-La Rosa J, Doddapaneni H, Civerolo EL.2006. Genome-based PCR primers for specific and sensitive detection andquantification of Xylella fastidiosa. Eur. J. Plant Pathol. 115:203–213.

22. Guilhabert MR, Kirkpatrick BC. 2005. Identification of Xylella fastidiosaantivirulence genes: hemagglutinin adhesins contribute to Xylella fastid-iosa biofilm maturation and colonization and attenuate virulence. Mol.Plant Microbe Interact. 18:856 – 868.

23. Hall-Stoodley L, Costerton JW, Stoodley P. 2004. Bacterial biofilms:from the natural environment to infectious diseases. Nat. Rev. Microbiol.2:95–108.

24. Hansch R, Mendel RR. 2009. Physiological functions of mineral mi-cronutrients (Cu, Zn, Mn, Fe, Ni, Mo, B, Cl). Curr. Opin. Plant Biol.12:259 –266.

25. Hinsa SM, Espinosa-Urgel M, Ramos JL, O’Toole GA. 2003. Transitionfrom reversible to irreversible attachment during biofilm formation byPseudomonas fluorescens WCS365 requires an ABC transporter and a largesecreted protein. Mol. Microbiol. 49:905–918.

26. Hopkins DL, Purcell AH. 2002. Xylella fastidiosa: cause of Pierce’s diseaseof grapevine and other emergent diseases. Plant Dis. 86:1056 –1066.

27. Kierek K, Watnick PI. 2003. The Vibrio cholerae O139O-antigen polysac-charide is essential for Ca2�-dependent biofilm development in sea water.Proc. Natl. Acad. Sci. U. S. A. 100:14357–14362.

28. Killiny N, Prado SS, Almeida RP. 2010. Chitin utilization by the insect-transmitted bacterium Xylella fastidiosa. Appl. Environ. Microbiol. 76:6134 – 6140.

29. Lecourieux D, Ranjeva R, Pugin A. 2006. Calcium in plant defence-signalling pathways. New Phytol. 171:249 –269.

30. Leite B, et al. 2002. Genomics and X-ray microanalysis indicate that Ca2�

and thiols mediate the aggregation and adhesion of Xylella fastidiosa. Braz.J. Med. Biol. Res. 35:645– 650.

31. Li YX, et al. 2007. Type I and type IV pili of Xylella fastidiosa affecttwitching motility, biofilm formation and cell-cell aggregation. Microbi-ology 153:719 –726.

32. Reference deleted.33. Maathuis FJM. 2009. Physiological functions of mineral macronutrients.

Curr. Opin. Plant Biol. 12:250 –258.34. Mattick JS. 2002. Type IV pili and twitching motility. Annu. Rev. Micro-

biol. 56:289 –314.35. Meng YZ, et al. 2005. Upstream migration of Xylella fastidiosa via pilus-

driven twitching motility. J. Bacteriol. 187:5560 –5567.36. Newman KL, Almeida RP, Purcell AH, Lindow SE. 2003. Use of a green

fluorescent strain for analysis of Xylella fastidiosa colonization of Vitisvinifera. Appl. Environ. Microbiol. 69:7319 –7327.

37. Orans J, et al. 2010. Crystal structure analysis reveals Pseudomonas PilY1as an essential calcium-dependent regulator of bacterial surface motility.Proc. Natl. Acad. Sci. U. S. A. 107:1065–1070.

38. O’Toole GA, et al. 1999. Genetic approaches to study of biofilms. Biofilms310:91–109.

39. Patrauchan MA, Sarkisova S, Sauer K, Franklin MJ. 2005. Calciuminfluences cellular and extracellular product formation during biofilm-

Cruz et al.

1330 aem.asm.org Applied and Environmental Microbiology

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from

associated growth of a marine Pseudoalteromonas sp. Microbiology 151:2885–2897.

40. Peuke AD. 2000. The chemical composition of xylem sap in Vitis viniferaL. cv. Riesling during vegetative growth on three different Franconianvineyard soils and as influenced by nitrogen fertilizer. Am. J. Enol. Viti-culture 51:329 –339.

41. Prakash B, Veeregowda BM, Krishnappa G. 2003. Biofilms: a survivalstrategy of bacteria. Curr. Sci. 85:1299 –1307.

42. Purcino RP, et al. 2007. Xylella fastidiosa disturbs nitrogen metabolismand causes a stress response in sweet orange Citrus sinensis cv. Pera. J. Exp.Bot. 58:2733–2744.

43. Ramey BE, Koutsoudis M, von Bodman SB, Fuqua C. 2004. Biofilmformation in plant-microbe associations. Curr. Opin. Microbiol.7:602– 609.

44. Rodrigues CM, et al. 2008. Copper resistance of biofilm cells of the plantpathogen Xylella fastidiosa. Appl. Microbiol. Biotechnol. 77:1145–1157.

45. Romantschuk M. 1992. Attachment of plant pathogenic bacteria to plant-surfaces. Annu. Rev. Phytopathol. 30:225–243.

46. Rose RK. 2000. The role of calcium in oral streptococcal aggregation andthe implications for biofilm formation and retention. Biochim. Biophys.Acta 1475:76 – 82.

47. Sarkisova S, Patrauchan MA, Berglund D, Nivens DE, Franklin MJ. 2005.Calcium-inducedvirulencefactorsassociatedwiththeextracellularmatrixofmu-coid Pseudomonas aeruginosa biofilms. J. Bacteriol. 187:4327–4337.

48. Sauer K, Camper AK, Ehrlich GD, Costerton JW, Davies DG. 2002.Pseudomonas aeruginosa displays multiple phenotypes during develop-ment as a biofilm. J. Bacteriol. 184:1140 –1154.

49. Scarpari LM, Lambais MR, Silva DS, Carraro DM, Carrer H. 2003.Expression of putative pathogenicity-related genes in Xylella fastidiosagrown at low and high cell density conditions in vitro. FEMS Microbiol.Lett. 222:83–92.

50. Snyder L, Champness W. 2003. Molecular genetics of bacteria, 2nd ed.ASM Press, Washington, DC.

51. Stevenson JF, Matthews MA, Ros TL. 2004. Grapevine susceptibility toPierce’s disease. I. Relevance to hydraulic architecture. Am. J. Enol. Viti-culture 55:228 –237.

52. TranVan Nhieu G, Clair C, Grompone G, Sansonetti P. 2004. Calciumsignalling during cell interactions with bacterial pathogens. Biol. Cell 96:93–101.

53. Yang M, et al. 2006. The effects of mitochondrial iron homeostasis oncofactor specificity of superoxide dismutase 2. EMBO J. 25:1775–1783.

54. Zaini PA, De La Fuente L, Hoch HC, Burr TJ. 2009. Grapevine xylemsap enhances biofilm development by Xylella fastidiosa. FEMS Microbiol.Lett. 295:129 –134.

55. Zaini PA, et al. 2008. The iron stimulon of Xylella fastidiosa includesgenes for type IV pilus and colicin V-like bacteriocins. J. Bacteriol. 190:2368 –2378.

Effects of Calcium on Xylella fastidiosa

March 2012 Volume 78 Number 5 aem.asm.org 1331

on April 11, 2019 by guest

http://aem.asm

.org/D

ownloaded from