ecological interaction of a biocontrol pseudomonas...

Ecological interaction of a biocontrol Pseudomonas £uorescens

strain producing 2,4-diacetylphloroglucinol with the soft rot

potato pathogen Erwinia carotovora subsp. atroseptica

Don Cronin, Yvan Moeënne-Loccoz1, Anne Fenton, Colum Dunne,

David N. Dowling2, Fergal O'Gara *

Microbiology Department, University College Cork, Cork, Ireland

Received 23 July 1996; revised 27 January 1997; accepted 10 February 1997

Abstract

Erwinia carotovora subspecies atroseptica is the agent of soft rot of potato and causes important crop damage in Europe.

Synthetic 2,4-diacetylphloroglucinol (DAPG) inhibited the growth of E. carotovora subsp. atroseptica under in vitro conditions

and Pseudomonas fluorescens F113, which produces DAPG, was studied for biocontrol of E. carotovora subsp. atroseptica.

Wild-type F113 (or the spontaneous rifampicin-resistant mutant F113Rif) inhibited growth of E. carotovora subsp. atroseptica

on solid medium, displayed bactericidal activity towards the pathogen in liquid medium, and prevented Erwinia-mediated

rotting of wounded potato tuber under in vitro conditions. F113Rif reduced the population size of E. carotovora subsp.

atroseptica in soil and on potato tuber dice in competition experiments carried out with unplanted soil and soil planted with

diced potato tubers, respectively. Co-inoculation of potato tuber seeds with F113Rif and E. carotovora subsp. atroseptica

reduced Erwinia contamination of the seed tubers compared with single inoculation with the pathogen. F113G22 is a

Tn5 : :lacZY-induced DAPG-negative biosynthetic derivative of F113 and showed no antibiosis towards E. carotovora subsp.

atroseptica in vitro. In contrast to F113Rif, F113G22 did not inhibit Erwinia-mediated rotting of wounded potato tuber in

vitro, did not influence survival of E. carotovora subsp. atroseptica in unplanted soil or soil planted with potato tuber dice and

did not reduce Erwinia contamination of potato seed tubers. F113G22(pCU203) is a complemented derivative with restored

DAPG-producing ability. F113G22(pCU203) had similar effects against E. carotovora subsp. atroseptica as F113 (or F113Rif)

under in vitro conditions and in soil microcosms. The results indicate that P. fluorescens F113 is a promising biocontrol agent

against the potato soft rot agent E. carotovora subsp. atroseptica and suggest that the pseudomonad's ability to produce DAPG

is a key factor in its inhibition of the pathogen.

Keywords: Biocontrol; Pseudomonas ; DAPG; Erwinia

0168-6496 / 97 / $17.00 ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V.

PII S 0 1 6 8 - 6 4 9 6 ( 9 7 ) 0 0 0 1 0 - X

FEMSEC 795 27-8-97

* Corresponding author. Tel: +353 (21) 27 20 97; fax: +353 (21) 27 59 34; e-mail: [email protected]

1Current address: Institute of Plant Sciences/Phytopathology, ETH Zentrum, Zuërich, Switzerland.

2Current address: Department of Applied Biology and Chemistry, Carlow Regional Technical College, Carlow, Ireland.

FEMS Microbiology Ecology 23 (1997) 95^106

1. Introduction

Erwinia carotovora is a broad host-range pathogen

causing seed piece decay, blackleg and aerial soft rot,

especially in potatoes [1]. One of the four subspecies

of E. carotovora, i.e. subsp. atroseptica is considered

the most important soft rot bacterium in Europe due

to its ability to be pathogenic at low temperature in

moist soil [2]. E. carotovora subsp. atroseptica sur-

vives better than the other three subspecies of E.

carotovora in cool soils and causes blackleg earlier

in the season. The pathogenicity of E. carotovora

subsp. atroseptica (hereafter referred to as E. caroto-

vora) is almost entirely limited to potato crops. Cur-

rently, control of this pathogen relies on (i) seed

technology; i.e. tuber pasteurisation, sanitation and

use of certi¢ed seed, (ii) application of chemical pes-

ticides in the ¢eld and (iii) the use of potato cultivars

resistant to E. carotovora. However, the recent devel-

opment of more virulent strains of E. carotovora, the

toxicity of chemical control towards humans and the

environment, and the high cost of the control meth-

ods currently available have stimulated the search

for alternative methods of control such as biological

control.

The predominant group of bacteria that has been

studied for biocontrol of Erwinia-mediated soft rot

corresponds to the £uorescent pseudomonads since

they are aggressive rhizosphere colonisers and pro-

duce a wide range of antimicrobial compounds [3,4].

Certain £uorescent pseudomonads can protect plants

from diseases caused by root pathogens and often

this biocontrol e¡ect involves antimicrobial com-

pounds such as siderophores [5], hydrogen cyanide

[6], and antibiotics like phenazine-1-carboxylate [7],

pyoluteorin [8] and 2,4-diacetylphloroglucinol

(DAPG; [9^11]). Biological control of E. carotovora

has been achieved by inoculating potato seed pieces

with bacteria antagonistic to E. carotovora prior to

planting [12,13].

A number of DAPG-producing £uorescent pseu-

domonads have been studied for biocontrol of

pathogenic fungi [9,14^16] and bacteria [15,16].

While the importance of DAPG has been shown

for the biocontrol of certain pathogenic fungi, only

circumstantial evidence is available regarding the

role of DAPG in the biocontrol of E. carotovora

by £uorescent pseudomonads. DAPG producer P.

£uorescens F113 suppresses damping-o¡ of sugarbeet

in soil microcosms prepared with soil naturally in-

fested by Pythium [14]. The objective of this study

was to evaluate strain F113 for biocontrol of E. ca-

rotovora subsp. atroseptica and to investigate the role

of DAPG in the interaction between the biocontrol

pseudomonad and the potato soft rot pathogen.

2. Materials and methods

2.1. Strains and culture conditions

Pseudomonas £uorescens F113 produces the sec-

ondary metabolite DAPG [10]. A spontaneous rifam-

picin-resistant mutant of F113 (i.e. F113Rif) was

used in some of the experiments [17]. Strain

F113Rif produces DAPG and displays a similar

growth rate in liquid laboratory media compared to

the wildtype [17]. F113G22 is a Tn5 : :lacZY-induced

DAPG-negative biosynthetic mutant of F113 and

F113G22-(pCU203) is a complemented derivative of

F113-G22 [14]. E. carotovora subsp. atroseptica strain

SCRI47 [12] was kindly provided by Rhodes and

Logan (Queen's University of Belfast, Northern Ire-

land). A spontaneous streptomycin-resistant mutant

of the strain (hereafter referred to as Eca), which

causes potato soft rot in vitro and grows like the

wildtype in liquid culture, was used in this study.

All strains were maintained on LB agar [18] at 4³C

and were grown at 28³C with shaking. Growth me-

dia included LB, sucrose asparagine (SA; [19]), SA-

Fe (i.e. SA supplemented with 100 WM FeCl3) and

two iron-rich minimal media derived from Shana-

han's minimal medium [10]. The minimal media

were composed of succinate (27 g l

31; i.e. succi-

nate/KNO3 medium) or sucrose (34.2 g l

31; i.e. su-

crose/KNO3 medium), KNO3 (4.25 g l

31), Na2HPO4

(4.375 g l

31) and KH2PO4 (1.875 g l

31) and iron (27

mg l

31). Cells from overnight cultures (approxi-

mately 10

9CFU ml

31) were washed twice in quar-

ter-strength Ringer's solution (Oxoid) prior to inoc-

ulation. Antibiotics were used at the following

concentrations (Wg ml

31) : rifampicin, 100 (Rif100);

kanamycin, 50 (Km50); chloramphenicol, 200

(Cm200); and streptomycin, 100 (Sm100).

FEMSEC 795 27-8-97

D. Cronin et al. / FEMS Microbiology Ecology 23 (1997) 95^10696

2.2. E. carotovora subsp. atroseptica growth inhibition

test

The e¡ect of F113, F113G22 and F113G22-

(pCU203) on growth of Eca was investigated using

a growth inhibition bioassay similar to that of Fen-

ton et al. [14]. Media used included solid LB, SA,

SA-Fe, sucrose/KNO3 and succinate/KNO3. Three

5-Wl samples of cell suspension from each pseudomo-

nad (2U10

8CFU per 5-Wl sample) were spotted onto

solid media and the plates were incubated overnight

at 28³C. An Eca cell suspension (10

9CFU ml

31) was

sprayed onto the plates using an aerosol sprayer

(Sigma) and the plates were further incubated at

28³C for 72 h. Three plates of each medium were

used per pseudomonad studied.

The e¡ect of DAPG on growth of Eca was inves-

tigated using a similar in vitro assay, as follows.

Synthetic DAPG produced through acetylation of

phloroglucinol and assayed for impurities by NMR

and HPLC analysis (Chemistry Department, UCC,

Cork, Ireland) was dissolved in 99% methanol (1

mM DAPG) and diluted in sterile distilled water to

a concentration of 1 WM DAPG. A 15 Wl aliquot was

spotted onto the centre of each LB plate used. The

plates were dried in a laminar-£ow hood and were

kept at 4³C until inoculation with Eca was per-

formed. Eca was sprayed and the plates were incu-

bated, as described above. The experiment was per-

formed in triplicate.

2.3. In vitro competition between E. carotovora subsp.

atroseptica and P. £uorescens

Competition experiments were conducted by co-

inoculating SA-Fe broth with Eca and F113Rif,

F113G22, or F113G22(pCU203) in a 1:1 ratio.

Each strain was inoculated at the level of about

10

6CFU ml

31. The experiment was performed in

250 ml Erlenmeyer £asks containing 100 ml of me-

dium and the £asks were incubated at 28³C with

shaking (rotary shaker; 280 rpm). Flasks inoculated

with single strains were used as controls. Colony

counts were carried out periodically using 1 ml sam-

ples from the £asks. Three 20 Wl volumes were spot-

ted onto SA+Rif100 (for F113Rif), SA+Km50 (for

F113G22), SA+Km50+Cm200 (for F113G22

(pCU203)), and SA+Sm100 (for Eca). The plates

were incubated at room temperature (approximately

20³C) for 72 h and the colonies were counted. The

experiment was performed in duplicate.

2.4. In vitro pathogenicity bioassay

The e¡ect of P. £uorescens on the pathogenicity of

Eca was investigated using an in vitro bioassay in

which pseudomonads were each co-inoculated with

Eca into wounded potato tuber slices. Tubers (about

5 cm in diameter) of potato cv. `Kerr's Pink' were

washed free of soil in sterile distilled water and were

soaked for three min in 2% sodium hypochlorite so-

lution to disinfect the tuber surface. The tubers were

rinsed in sterile distilled water and cut in slices

(about 1 cm thick). Three wells were aseptically

punched in each slice using a 5 mm diameter metal

rod. The tuber slices were dried with sterile ¢lter

paper (Whatman No. 541). Each well was co-inocu-

lated with 10 Wl cell suspensions (i.e. 10

7CFU of

each strain) of Eca and of either F113Rif,

F113G22, or F113G22(pCU203). The experiment

was also carried out with each strain inoculated sin-

gly. The tuber slices were placed into Petri dishes

perforated for aeration and the latter were put into

anaerobic jars (BBL Gas Pak Systems) lined with

moist tissue paper. The jars were incubated at 12³C

for 96 h and the surface area of rotted potato tissue

was measured. The experiment was performed in du-

plicate.

2.5. Survival of Eca in unplanted soil microcosms

Soil microcosms consisted of 100 g of natural soil

sampled from the top horizon (sandy loam texture;

pHH2O6.9) of a brown podzolic soil. Sampling took

place in a ¢eld located near Ovens (County Cork,

Ireland) that was under sugarbeet cultivation. The

soil was sieved through a 5-mm mesh screen and

soil microcosms were prepared as described else-

where [14,20]. Cell suspensions of Eca were added

to the soil (about 10

6CFU g

31of soil) and the

soil was mixed throughout. Pseudomonads were

then inoculated into the soil at cell concentrations

of approximately 10

4, 10

6or 10

8CFU g

31of soil

and the soil was mixed uniformly. The water content

of the soil was adjusted to 25% w/w (i.e. 60% satu-

ration of the soil porosity) using distilled water and

FEMSEC 795 27-8-97

D. Cronin et al. / FEMS Microbiology Ecology 23 (1997) 95^106 97

the soil was watered, by spraying, every three days.

The pots were placed in a growth chamber (12³C;

16-h photoperiod) in a randomized block design.

Pseudomonads tested included F113Rif, F113G22

and F113G22(pCU203). Pseudomonads and Eca

were also inoculated singly, as controls. At weekly

intervals, a 1-g soil sample was taken from each pot

and extracted in quarter-strength Ringer's solution

for three min. A serial dilution was prepared in the

same solution and 100 Wl volumes were spread-plated

in duplicate onto SA+Rif100, SA+Km50,

SA+Km50+Cm200 and SA+Sm100 to enumerate

F113Rif, F113G22, F113G22(pCU203) and Eca, re-

spectively. The experiment was performed in dupli-

cate.

2.6. Survival of Eca in soil microcosms planted with

potato tuber pieces

Soil microcosms were prepared using natural soil

sampled from the sandy-loam surface horizon of a

brown podzolic soil located near Bandon (County

Cork, Ireland) and cropped with sugarbeet. The

soil for this experiment was not collected at the

Ovens site because sugarbeet was not grown at the

time of sampling. Both ¢eld sites correspond to the

same pedon and the soil materials sampled displayed

essentially similar physico-chemical characteristics

(data not shown). Soil microcosms and incubation

conditions were as described above, except that 150

g of soil was used per pot. Eca was inoculated into

soil before planting, at the concentration of about

2U10

6CFU g

31of soil. Tuber pieces from potato

cv. `Kerr's Pink' were soaked into a cell suspension

of F113Rif, F113G22, or F113G22(pCU203) for 5

min. The experiment was also run with single inoc-

ulation of Eca or pseudomonad. The tuber pieces

were planted at a depth of 3.5 cm (one tuber piece

per pot) and the soil water content was adjusted to

26% w/w (about 60% saturation of the soil porosity)

using distilled water. The pots were placed in the

growth chamber for six weeks, as described above.

At the end of the experiment, the tuber pieces were

removed from the pots and colonisation of the tuber

surface was determined. The tuber pieces were ex-

tracted in quarter-strength Ringer's solution for 3

min. Dilution, plating, and colony counts were

done as described above. The experiment was ¢rst

performed using one-cm

3tuber dice (pseudomonads:

8U10

8CFU tuber dice

31) and was run in triplicate.

The same experiment was then carried out using pre-

sprouted potato seed tubers (pseudomonads: 8U10

8

CFU seed tuber

31) and four replications were

studied per treatment.

2.7. Statistical analyses

Two to four replications were used per treatment.

Colony counts were log10-transformed before anal-

ysis. All results were analysed by analysis of var-

iance, using procedures of the Statistical Anal-

ysis System (SAS; The SAS Institute, Cary, NC,

USA). When appropriate, treatments were com-

pared using Fisher's least signi¢cant di¡erence

(LSD) test. All analyses were conducted at P=0.05

level.

3. Results

3.1. E¡ect of F113, F113G22 and F113G22(pCU203)

on growth of E. carotovora subsp. atroseptica

in vitro

The e¡ect of DAPG producer P. £uorescens F113

on growth of Eca in vitro was studied by growing

the pseudomonad as a single colony in the center of

plates and monitoring the growth of Eca subse-

quently inoculated onto the plates by spraying of a

cell suspension. Strain F113 inhibited growth of Eca

on LB, SA, SA-Fe, and sucrose/KNO3 plates (Table

1) but had no e¡ect on succinate/KNO3 plates,

where F113 does not produce DAPG (data not

shown). Similar results were obtained when plates

were incubated at 12³C instead of 28³C prior to in-

oculation with Eca. The experiment was also carried

out by spraying plates with the DAPG-sensitive in-

dicator strain Bacillus subtilis A1 [14] and similar

results were obtained regarding the ability of F113

to inhibit the growth of strain A1 on the ¢ve media

studied.

DAPG-negative biosynthetic mutant F113G22 did

not inhibit growth of Eca on LB, SA-Fe, sucrose/

KNO3 and succinate/KNO3 plates (Table 1). How-

ever, F113G22 inhibited growth of Eca on SA plates,

a medium that is low in iron and consequently fa-

FEMSEC 795 27-8-97


vours siderophore production and in which F113

produces low amounts of DAPG [10]. Similar results

were obtained when the pseudomonad was grown at

12³C instead of 28³C. F113G22 failed to inhibit

DAPG-sensitive B. subtilis strain A1 on all media

with the exception of solid SA. F113G22(pCU203),

a complemented mutant of F113G22 with restored

DAPG-producing ability inhibited growth of Eca

like F113 on SA and sucrose/KNO3 plates (Table

1). F113G22(pCU203) also inhibited growth of Eca

on LB and SA-Fe plates, albeit to a lesser extent

compared to F113. F113G22(pCU203), like F113,

had no e¡ect on Eca on succinate/KNO3 plates.

Similar results were obtained when F113G22-

(pCU203) was grown at 12³C instead of 28³C.

F113G22(pCU203) also inhibited DAPG-sensitive

B. subtilis strain A1 on all media tested except on

solid succinate/KNO3.

3.2. E¡ect of synthetic DAPG on growth of

E. carotovora subsp. atroseptica in vitro

The addition of synthetic DAPG as a 15 Wl spot of

DAPG solution onto LB agar inhibited the growth

of Eca on the plates when DAPG was used at con-

centrations greater than the threshold value of 23

WM DAPG (Fig. 1). Growth of Eca was not inhib-

ited on plates receiving a 15 Wl spot of methanol in

which no DAPG had been dissolved. For concentra-

tions higher than 23 WM DAPG, the increase in di-

ameter of the zone of inhibition was linear with in-

creasing log values of DAPG concentration.

FEMSEC 795 27-8-97

Table 1

Inhibition of growth of E. carotovora subsp. atroseptica by P. £uorescens F113, F113G22 and F113G22(pCU203) under in vitro condi-

tions at 28³C

Media conducive to DAPG production Media not conducive to DAPG production

LB SA-Fe Sucrose/KNO3 Succinate/KNO3 SA

No pseudomonad none a1

none a none a none a none a

F113 +++2b +++ b ++ b none a +++ b

F113G22 none a none a none a none a ++ c

F113G22(pCU203) ++ c ++ c ++ b none a +++ b

Pseudomonads were spotted and grown on the plates prior to inoculation with Eca.

1For each medium, letters in bold are used to indicate statistical di¡erences between treatments assessed by LSD.

2Diameter of inhibition zone between 2.4 and 3.0 cm are indicated by +++ and those between 1.5 and 2.3 cm are indicated by ++.

Fig. 1. Inhibition of Eca by synthetic 2,4-diacetylphloroglucinol on LB agar.


3.3. In vitro competition between E. carotovora subsp.

atroseptica and P. £uorescens

Competition experiments were conducted by co-in-

oculating SA-Fe broth with Eca and F113Rif,

F113G22, or F113G22(pCU203) in a 1:1 ratio. Co-

inoculation with Eca and F113Rif resulted in lower

cell numbers of Eca compared with those obtained

with single inoculation of Eca (Fig. 2A,B). In con-

trast, F113G22 had no e¡ect on Eca cell numbers

and cell numbers of F113G22 were actually lower

when the strain was co-inoculated with Eca compared

with F113G22 inoculated singly (Fig. 2C,D). DAPG

producer F113G22(pCU203) reduced cell numbers of

Eca in competition experiments (Fig. 2E,F) and the

e¡ect of the pseudomonad on Eca was similar to that

exerted by F113Rif (Fig. 2B,F), although occurring

30 instead of 12 h after co-inoculation.

3.4. E¡ect of P. £uorescens on Eca pathogenicity

in vitro

P. £uorescens F113Rif reduced the extent of Eca-

mediated soft rot of the potato tuber slices in vitro

when both bacteria were co-inoculated into wells

punched into the tuber slices (Fig. 3). Strain

F113Rif failed to a¡ect Eca-mediated soft-rotting

of potato tissue when co-inoculation was performed

using 10 Wl cell suspensions containing 105instead of

107

CFU of F113Rif. In contrast to F113Rif,

DAPG-negative derivative F113G22 did not prevent

Eca-mediated soft rot compared to situations where

only Eca was inoculated. The complemented strain,

F113G22(pCU203) also reduced Eca-mediated soft

rotting of the tuber slice and the e¡ect of the

pseudomonad was similar to that of F113Rif.

No soft rotting took place when any of the

FEMSEC 795 27-8-97

Fig. 2. In£uence of P. £uorescens F113Rif (A,B), F113G22 (C,D) and F113G22(pCU203) (E,F) on the survival of Eca in SA-Fe broth.

Strains were inoculated singly in A, C and E. F113Rif (B), F113G22 (D) and F113G22(pCU203) (F) were co-inoculated each with Eca.

F113Rif and F113G22(pCU203) signi¢cantly reduced cell numbers of Eca in the co-inoculation experiments. Survival of F113G22 was sig-

ni¢cantly lower when the pseudomonad was in the presence of Eca.


three pseudomonads were inoculated in the absence

of Eca.

3.5. E¡ect of P. £uorescens on survival of Eca in

unplanted soil microcosms

Co-inoculation of soil with F113Rif (at 104, 10

6or

108CFU g

31soil) and 10

6CFU Eca g

31soil re-

sulted in lower cell numbers of Eca compared to

those in soil inoculated with Eca alone (Fig. 4A).

The e¡ect of F113Rif, F113G22 and F113G22-

(pCU203) introduced in soil at 108CFU g

31soil

on cell numbers of Eca co-inoculated at 106CFU

g31

soil was then investigated. Unlike F113Rif,

F113G22 had no e¡ect on Eca cell numbers in soil

during the course of the six-week experiment (Fig.

4B). The complemented derivative F113G22-

(pCU203) reduced the survival of Eca in soil,

although to a lesser extent compared with F113Rif.

3.6. E¡ect of P. £uorescens on colonisation of tuber

dice by Eca in soil microcosms

There was no di¡erence in survival on tuber dice

between strains F113Rif, F113G22 and F113G22

(pCU203) in soil microcosms in the absence of Eca

(Fig. 5A,C,E). Inoculation of one cm3tuber dice

with F113Rif prior to planting in soil previously in-

oculated with Eca (2U106CFU g

31soil) reduced the

extent of colonisation of tuber dice by Eca in the six-

week experiment (Fig. 5A,B). Inoculation of tuber

dice with the DAPG-negative mutant F113G22 did

not a¡ect their colonisation by Eca compared to that

of uninoculated tuber dice (Fig. 5C,D). However,

inoculation of the tuber dice with the complemented

FEMSEC 795 27-8-97

Fig. 4. In£uence of P. £uorescens F113Rif, F113G22 and F113G22(pCU203) on the survival of Eca in unplanted soil microcosms. Eca

was introduced into soil at about 106CFU g

31soil. The in£uence of the inoculum level of F113Rif on Eca survival is shown in A.

F113Rif signi¢cantly reduced the survival of Eca in soil at all three inoculum levels. The e¡ect of F113Rif, F113G22 or

F113G22(pCU203) introduced at about 108CFU g

31soil on Eca survival is shown in B. F113Rif and F113G22(pCU203) signi¢cantly re-

duced the survival of Eca in soil, whilst F113G22 did not. Vertical bars represent standard deviations.

Fig. 3. In£uence of P. £uorescens F113Rif, F113G22 and

F113G22(pCU203) on soft rotting of wounded potato tissue by

Eca in vitro. Vertical bars represent standard deviations. Letters

a and b are used to indicate means that are not signi¢cantly dif-

ferent by LSD on log10 transformed values.


pseudomonad F113G22(pCU203), with restored

ability to produce DAPG, reduced Eca colonisation

of the tuber dice, in a similar fashion as when

F113Rif was used (Fig. 5E,F). In this experiment,

no soft rotting of the potato tissue was found, even

in the treatment where uninoculated tuber dice were

planted in Eca-inoculated soil.

3.7. E¡ect of P. £uorescens on colonisation of seed

tubers by Eca in soil microcosms

Colonisation of pre-sprouted potato seed tubers

by Eca was reduced when F113Rif was inoculated

onto the seed tubers prior to planting them in Eca-

enriched soil. F113G22 colonised the potato tubers

to a lesser extent than F113Rif in uninoculated soil

(Fig. 6A) and in soil inoculated with Eca (Fig. 6B).

F113G22 was unable to in£uence colonisation of the

seed tubers by Eca, whereas F113G22(pCU203) re-

duced seed tuber colonisation by Eca to the same

extent as F113Rif did. No soft rotting of the seed

tubers was found in any of the treatments.

4. Discussion

Certain £uorescent pseudomonads can improve

plant growth through the control of harmful micro-

organisms and this often involves the production of

enzymes and/or secondary metabolites such as anti-

biotics and siderophores [7,9,10,21,22]. In this work,

the e¡ect of DAPG-producing P. £uorescens F113

on the potato soft rot pathogen E. carotovora

subsp. atroseptica was investigated.

F113 inhibited growth of Eca on a number of

solid media that are conducive to DAPG production

FEMSEC 795 27-8-97

Fig. 5. Colonisation of potato tuber dice by Eca in soil in the presence of P. £uorescens F113Rif, F113G22 or F113G22(pCU203). Eca

was introduced into unplanted soil and the pseudomonads were inoculated each onto the tuber dice prior to planting. All bacteria were

inoculated singly in A, C and E, and results of co-inoculation experiments are shown in B, D and F. Only F113Rif and

F113G22(pCU203) signi¢cantly reduced colonisation of the tuber dice by Eca.


but had no e¡ect on succinate/KNO3 plates, where

F113 does not produce DAPG (Table 1). These re-

sults suggest that DAPG production may be in-

volved in Eca growth inhibition. The e¡ect of

DAPG on Eca was con¢rmed by the results of ex-

periments in which F113G22, a transposon-induced

DAPG-negative biosynthetic mutant of F113 and

F113G22(pCU203), a DAPG-producing comple-

mented derivative of F113G22 were compared with

F113. Interestingly, F113G22 inhibited growth of

Eca on SA plates, a low-iron medium that favours

siderophore production in F113 and F113G22, sug-

gesting that siderophore production also can contrib-

ute to biological control of Eca under these labora-

FEMSEC 795 27-8-97

Fig. 6. In£uence of P. £uorescens F113Rif, F113G22 and F113G22(pCU203) on colonisation of potato seed tubers by Eca in soil micro-

cosms at 6 weeks after inoculation. Eca was introduced into unplanted soil and the pseudomonads were inoculated onto the seed tubers

prior to planting. Results from single inoculations are shown in A and those from co-inoculation experiments in B. Vertical bars represent

standard deviations. Only F113Rif and F113G22(pCU203) signi¢cantly reduced colonisation of the seed tubers by Eca.


tory conditions. Strains F113 and F113G22 produce

minute amounts of siderophore on iron-rich media

SA-Fe, LB and sucrose/KNO3, which indicate that

Eca growth inhibition on these media resulted from

the strain's ability to produce DAPG. These results

show that siderophore production by F113 may play

a role in biocontrol of Eca under iron limiting con-

ditions, but that DAPG is the metabolite primarily

responsible for the control of Eca by F113 in vitro.

The involvement of several secondary metabolites in

biocontrol of black root rot of tobacco and damp-

ing-o¡ of cress by pseudomonads has been demon-

strated by Laville et al. [23] and Maurhofer et al.

[24], respectively. On all three media where F113

produces DAPG (i.e. LB, SA-Fe and sucrose/

KNO3), F113G22(pCU203) inhibited growth of

Eca to a lesser extent compared to F113. F113G22

(pCU203) produces less DAPG than F113 under cer-

tain laboratory conditions [14], which further indi-

cates the role of DAPG in Eca growth inhibition.

Vincent et al. [11] also failed to fully complement a

Tn5-induced DAPG-negative mutant of Pseudomo-

nas for DAPG production. Previous work has shown

the prevalence of DAPG production in strains of P.

£uorescens inhibitory to E. carotovora subsp. atrosep-

tica [15] but the current study is the ¢rst one to

demonstrate speci¢cally the role of DAPG produc-

tion by pseudomonads in their inhibition of E. caro-

tovora subsp. atroseptica in vitro. Supplementation

of LB plates with synthetic DAPG indicated that a

concentration threshold of 23 WM DAPG was re-

quired to inhibit Eca growth at 28³C (Fig. 1). The

level of Eca inhibition implemented by F113 and

F113G22(pCU203) on LB plates was also achieved

using synthetic DAPG at concentrations of 230 WM

and 63 WM, respectively. Previous work has shown

that E. carotovora subsp. atroseptica was inhibited by

DAPG used at the concentration of 50 Wg (i.e. 20

Wmoles) of DAPG per disc [16].

The antagonistic e¡ect of F113 on Eca may be

bacteriostatic 9 preventing the growth of Eca, or

bactericidal 9 a¡ecting Eca survival. This point

was investigated in competition experiments in vitro,

where F113Rif, F113G22 and F113G22(pCU203)

were co-inoculated each with Eca. Strain F113Rif

was bactericidal to Eca as cell numbers of the latter

decreased sharply before the time where Eca usually

reaches the highest cell density in pure culture (Fig.

2). The e¡ects of F113G22 and F113G22(pCU203)

on Eca indicate that the bactericidal e¡ect of

F113Rif on Eca resulted from the strain's ability to

produce DAPG. In this work, no inhibition of Eca

growth on plates took place at DAPG concentra-

tions below 23 WM, which implies that DAPG pro-

duced by F113 becomes toxic to Eca only when the

concentration of DAPG exceeds a threshold level.

This is in accordance with the facts that F113Rif

had no e¡ect on Eca in liquid LB until 12 h after

inoculation and that F113G22(pCU203), which pro-

duces DAPG to a lesser extent compared to F113,

required 30 h to a¡ect Eca cell numbers. Interest-

ingly, F113G22 had no e¡ect on cell numbers of

Eca, but then Eca appeared to be detrimental to

the pseudomonad. This experiment indicated that

(i) F113Rif exhibits a bactericidal mode of control

against Eca, (ii) a build-up of DAPG may be neces-

sary in the medium for biocontrol of Eca by F113,

and (iii) the partially complemented strain

F113G22(pCU203) takes longer to exert the same

e¡ect on Eca as F113Rif, possibly due to a slower

accumulation of DAPG or the inability to produce

DAPG in concentrations as high as those achieved

by F113Rif.

DAPG producer F113Rif also inhibited Eca-medi-

ated soft rotting of potato tuber slices when both

bacteria were co-inoculated into wounded tissue in

vitro (Fig. 3). Results obtained with F113G22 and

F113G22(pCU203) showed that biocontrol of Eca

on potato tuber slices by F113Rif involved the

strain's ability to produce DAPG.

Further investigation was aimed at determining

how soil factors would interfere with the ability of

F113Rif to control Eca in soil and in the rhizosphere

of potato. F113Rif exhibited a bactericidal e¡ect to-

wards Eca (introduced at 10

6CFU g

31soil) in un-

planted soil when F113Rif was co-inoculated with

Eca at cell densities of between 10

4and 10

8CFU

g

31soil (Fig. 4A). The time period required for

F113Rif to reduce Eca cell numbers decreased

when higher inoculum sizes of F113Rif were used

in the soil. Potato tissue colonisation by Eca was

reduced when potato tuber dice or seed tubers were

inoculated with F113Rif at the time of plant-

ing (Figs. 5 and 6). The ability of F113Rif to

control the growth of Eca on potato tuber dice

and seed tubers in soil suggests that the environ-

FEMSEC 795 27-8-97


mental conditions at the surface of these potato tu-

ber pieces are conducive to DAPG production by

F113Rif.

Unlike F113Rif, F113G22 did not a¡ect Eca sur-

vival in unplanted soil (Fig. 4B), although Eca had

no e¡ect on cell numbers of F113G22 in soil (data

not shown). F113G22 had no e¡ect either on the

colonisation of potato tuber dice or seed tubers by

Eca (Figs. 5 and 6). Interestingly, F113G22 did not

colonise the seed tubers to the same extent as

F113Rif in the absence of Eca. This is the ¢rst report

where the loss of DAPG-producing ability a¡ects

colonisation of the pseudomonad in soil microcosm.

In contrast to F113G22, F113G22(pCU203) reduced

the cell numbers of Eca in unplanted soil micro-

cosms, con¢rming that DAPG was involved in the

control of Eca in soil. The pseudomonad also re-

duced colonisation of potato tuber dice and seed

tubers by Eca. Collectively, these results demonstrate

the importance of DAPG production in the e¡ect of

F113 and F113Rif on Eca.

In the current work, the soil microcosm experi-

ments were designed with the objective to assess

the nature of the interactions between DAPG-pro-

ducing pseudomonads and E. carotovora subsp. atro-

septica. Eca did not cause soft rotting of the potato

tuber pieces under the experimental conditions chos-

en despite colonising the tuber dice and seed tubers.

E. carotovora subsp. atroseptica can survive in soil

for up to 10 weeks when the soil is cool and moist

and enters the potato via the roots or wounds in the

tuber [25]. E. carotovora subsp. atroseptica is an

opportunistic pathogen and can reside in the po-

tato tuber at low levels for some time before start-

ing to proliferate and becoming pathogenic [2].

This suggests that the current soil microcosm ex-

periments were too short to observe soft rot symp-

toms.

The results of the study indicate that DAPG-pro-

ducing P. £uorescens F113 is a promising biocontrol

agent against the potato soft rot pathogen E. caro-

tovora subsp. atroseptica and suggest that the strain's

biocontrol ability results from its capacity to pro-

duce the secondary metabolite DAPG. Further

work will evaluate the ability of F113 to suppress

Eca-mediated soft rot of potato in soil microcosms,

under experimental conditions that favour develop-

ment of the disease.

Acknowledgments

We thank Pat Higgins for technical assistance.

This work was supported in part by grants from

the European Commission; Biotech BIO2-CT93-

0196, BIO2-CT96-0053, BIO2-CT94-3001 and BIO-

CT92-0084.

References

[1] Peèrombelon, M.C.M. and Kelman, A. (1987) Blackleg and

other potato diseases caused by soft rot Erwinias : proposal

for revision of terminology. Plant Dis. 71: 283^285.

[2] Aleck, J.R. and Harrison, M.D. (1978) The in£uence of in-

oculum density and environment on the development of po-

tato blackleg. Am. Potato J. 55: 479^494.

[3] Kloepper, J.W. (1983) E¡ect of seed pieces inoculation with

plant growth-promoting rhizobacteria on populations of Er-

winia carotovora on potato roots and in daughter tubers. Phy-

topathology 73: 217^219.

[4] Gross, D.C. (1988) Maximising rhizosphere populations of

£uorescent pseudomonads on potatoes and their e¡ects on

Erwinia carotovora. Am. Potato J. 65: 697^710.

[5] Bakker, P.A.H.M., Bakker, A.W., Marugg, J.D., Weisbeek,

P.J. and Schippers, B. (1987) Bioassays for studying the role

of siderophores in potato growth stimulation by Pseudomonas

spp. in short potato rotations. Soil Biol. Biochem. 19: 443^

449.

[6] Voisard, C., Keel, C., Haas, D. and Deèfago, G. (1989) Cya-

nide production by Pseudomonas £uorescens helps suppress

black root rot of tobacco under gnotobiotic conditions.

EMBO J. 8: 351^358.

[7] Thomashow, L.S. and Weller, D.M. (1988) Role of a phena-

zine antibiotic from Pseudomonas £uorescens in biological

control of Gaeumannomyces graminis var. triciti. J. Bacteriol.

170: 3499^3508.

[8] Howell, C.R. and Stipanovic, R.D. (1979) Control of Rhizoc-

tonia solani on cotton seedlings with Pseudomonas £uorescens

and with an antibiotic produced by the bacterium. Phytopa-

thology 69: 480^482.

[9] Keel, C., Schnider, U., Maurhofer, M., Voisard, C., Laville,

J., Burger, U., Wirthner, P., Haas, D. and Deèfago, G. (1992)

Suppression of root diseases by Pseudomonas £uorescens

CHA0: Importance of the bacterial secondary metabolite

2,4-diacetylphloroglucinol. Mol. Plant^Microbe Interact. 5:

4^13.

[10] Shanahan, P., O'Sullivan, D.J., Simpson, P., Glennon, J.D.

and O'Gara, F. (1992) Isolation and characterisation of an

antibiotic-like compound from a £uorescent pseudomonad

and investigation of physiological parameters in£uencing its

production. Appl. Environ. Microbiol. 58: 353^358.

[11] Vincent, M.N., Harrison, L.A., Bracken, J.M., Kovacevich,

P.A., Mukerji, P., Weller, D.M. and Pierson, E.A. (1991) Ge-

netic analysis of the antifungal activity of a soilborne Pseudo-

FEMSEC 795 27-8-97


monas aureofaciens strain. Appl. Environ. Microbiol. 57:

2928^2934.

[12] Rhodes, D.J. and Logan, C. (1986) E¡ects of £uorescent pseu-

domonads on the potato blackleg syndrome. Ann. Appl. Biol.

108: 511^518.

[13] Xu, G.W. and Gross, D.C. (1986) Selection of £uorescent

pseudomonads antagonistic to Erwinia carotovora and sup-

pressive of potato seed piece decay. Phytopathology 76:

414^422.

[14] Fenton, A.M., Stephens, P.M., Crowley, J., O'Callaghan, M.

and O'Gara, F. (1992) Exploitation of gene(s) involved in 2,4-

diacetylphloroglucinol biosynthesis to confer a new biocontrol

capability to a Pseudomonas strain. Appl. Environ. Microbiol.

58: 3873^3878.

[15] Nowak-Thompson, B., Gould, S.J., Kraus, J. and Loper, J.E.

(1994) Production of 2,4-diacetylphloroglucinol by the biocon-

trol agent Pseudomonas £uorescens Pf-5. Can. J. Microbiol.

40: 1064^1066.

[16] Levy, E., Gough, F.J., Berlin, K.D., Guiana, P.W. and Smith,

J.T. (1992) Inhibition of Septoria tritici and other phytopatho-

genic fungi and bacteria by Pseudomonas £uorescens and its

antibiotics. Plant Pathol. 41: 335^341.

[17] Carroll, H., Moeënne-Loccoz, Y., Dowling, D.N. and O'Gara,

F. (1995) Mutational disruption of the biosynthesis genes cod-

ing for the antifungal metabolite 2,4-diacetylphloroglucinol

does not in£uence the ecological ¢tness of Pseudomonas £uo-

rescens F113 in the rhizosphere of sugarbeets. Appl. Environ.

Microbiol. 61: 3002^3007.

[18] Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular

Cloning: a Laboratory Manual. Cold Spring Harbour Labo-

ratory, Cold Spring Harbour, N.Y.

[19] Scher, F.M. and Baker, R. (1982) E¡ect of Pseudomonas pu-

tida and a synthetic iron chelator on induction of soil suppres-

siveness to Fusarium wilt pathogens. Phytopathology 72:

1567^1573.

[20] Fedi, S., Moeënne-Loccoz, Y., Dowling, D.N. and O'Gara, F.

(1995) A versatile most probable number system to quantify

lacZY-marked pseudomonads present at low cell numbers in

the rhizosphere. Lett. Appl. Microbiol. 20: 220^224.

[21] Dowling, D.N. and O'Gara, F. (1994) Metabolites of Pseudo-

monas involved in the biocontrol of plant disease. TIBTECH

12: 133^141.

[22] Lindburg, G.D. (1981) An antibiotic lethal to fungi. Plant Dis.

65: 680^683.

[23] Laville, J., Voisard, C., Keel, C., Maurhofer, M., Deèfago, G.

and Haas, D. (1992) Global control in Pseudomonas £uores-

cens mediating antibiotic synthesis and suppression of black

root rot of tobacco. Proc. Natl. Acad. Sci. USA 89: 1562^

1566.

[24] Maurhofer, M., Keel, C., Voisard, D., Haas, D. and Deèfago,

G. (1992) In£uence of enhanced antibiotic production in Pseu-

domonas £uorescens strain CHA0 on its disease suppressive

capacity. Phytopathology 82: 190^195.

[25] Peèrombelon, M.C.M. and Kelman, A. (1980) Ecology of the

soft rot erwinias. Ann. Rev. Phytopathol. 18: 361^387.

FEMSEC 795 27-8-97


ecological interaction of a biocontrol pseudomonas...

Documents