wheat seedlings. These seedlings had been raised
from seeds inoculated with phenazine-producing
pseudomonads and grown in pots of natural soil
containing the take-all fungus. After 4 weeks the
seedlings were harvested and the rhizosphere soil was
found to have antibiotic levels of 30 – 40 nanograms per
gram of root with adhering soil. The levels of PCA were
about 10 times higher when the inoculated seedlings
were grown in sterile soil, but this would be expected
because of the lack of competition from other rhizo-
sphere bacteria. Compared with controls (untreated
seeds) the PCA-treated seeds caused a significant
reduction of disease. In contrast to the natural
phenazine-producing strains, phenazine-minus strains
obtained by transposon mutagenesis gave no reduction
of disease and had no detectable antibiotic levels,
but back-mutation to a phenazine-plus phenotype
restored the ability to reduce disease. Identical results
were obtained with a diacetyl-phloroglucinol (DAPG)-
producing strain. So, there is clear evidence for a role
of antibiotic-producing fluorescent pseudomonads in
take-all suppressive soils.
An interesting feature revealed by these studies is that
the population of antibiotic-producing pseudomonads
builds up progressively on wheat crops, but only (or
largely) when the take-all fungus is present. Wheat can
be grown repeatedly in the absence of the take-all fun-
gus in glasshouse conditions, and this does not lead to
a build-up of the antagonistic pseudomonads. So it
seems that the crop has to go through a build up of
disease before the disease suppression sets in.
The likely reason for this is shown by the
experimental results in Fig. 12.3. Wheat seeds were
inoculated with different inoculum densities of
DAPG-producing pseudomonads and then sown in a
glasshouse in pots of natural soil containing take-all
inoculum. The seedlings were sampled after 4 weeks and
assessed for disease severity and plant height. There was
no disease control until the population level of DAPG
strains on the roots exceeded 10^4 colony-forming
units per gram of root. But, at colony levels of 10^5 and
above, there was a marked and significant decrease in
disease severity and a corresponding increase in plant
height. This “all or nothing” effect is characteristic of
a bacterial signalling system called quorum sensing–
a term derived from the meetings of committees,
where a certain number of people (a quorum) has to
be present before a decision can be taken. Quorum-
sensing by a population of Gram-negative bacteria
involves the continued release of molecules called
N-acyl homoserine lactones (Fig. 12.4). When the
concentration of these molecules reaches a certain level
(indicating that the population is large enough) the
relevant genes are switched on. In this case, it is the
genes controlling antibiotic production and therefore
the control of take-all disease. The production of
phenazine antibiotics is known to be under the
control of a quorum-sensing system (Chin-A-Woeng
et al. 2003), but there is no evidence as yet that DAPG
production is regulated in a similar way.FUNGAL INTERACTIONS 239
O
O
N
H
O
H
Fig. 12.4Structure of N-hexanoyl-L-homoserine lactone,
a quorum-sensing molecule that regulates the synthesis
of phenazine antibiotics by fluorescent pseudomonads.
(From Wood et al. 1997.)5
2
1
Pseudomonas fluorescens (cfu g–1 root)3
101
4
160
100
80
120
102 103 104 105 106 107
140
Disease severity
Shoot height (mm)
Fig. 12.3The effects of different population levels
of a diacetyl-phloroglucinol producing fluorescent
pseudomonad on disease level (or shoot height)
of wheat seedlings grown in take-all infested soil.
(Reproduced from Raaijmakers & Weller 1998.)