FEMS Microbiology Ecology 46 (2003) 73^80
www.fems-microbiology.org
Soil antimony pollution and plant growth stage a¡ect
the biodiversity of auxin-producing bacteria isolated from
the rhizosphere of Achillea ageratum L
Christine Picard
1;
, Marco Bosco
Dipartimento di Scienze e Tecnologie Agroambientali, Area di Microbiologia, Alma Mater Studiorum, Universita' di Bologna, Viale Fanin 42,
40127 Bologna, Italy
Received 27 April 2003 ; received in revised form 18 June 2003; accepted 8 July 2003
First published online 28 August 2003
Abstract
A total of 4512 rhizobacteria were isolated at three stages of plant growth from Achillea ageratum colonizing a polluted site with an
antimony concentration gradient. For 222 of these isolates auxin production (auxþ ) was verified in vitro. The percentage of auxþ isolates
increased with soil antimony concentration, as well as with plant growth stage. An amplified rDNA restriction analysis clustered the auxþ
isolates into 51 clusters, one of which was numerically predominant and present throughout plant development and at all antimony
concentrations. The auxþ population was genetically very diverse, and this diversity was related to both antimony concentration and plant
growth stage.
4 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.
Keywords : Plant growth-promoting rhizobacterium; Auxin production; Heavy metal; Achillea ageratum L. ; Antimony pollution; Biodiversity
1. Introduction
Pollution of the biosphere by toxic metals has accelerated dramatically since the beginning of the industrial revolution. The sources of this pollution include mining and
smelting of metalliferous ores, burning of fossil fuels, municipal wastes, fertilizers, pesticides and sewage. Toxic
metal contamination of soil and groundwater poses a major environmental and human health problem [1]. Antimony is a toxic trace element of growing interest due to
the increased anthropogenic input into the environment
* Corresponding author. Tel. : +39 (051) 20 96 270;
Fax : +39 (051) 20 96 274.
E-mail address : [email protected] (C. Picard).
1
Collaborator, via a fellowship under the OECD Co-operative
Research Programme : Biological Resource Management for Sustainable
Agriculture Systems.
[2]. It is known to provoke DNA damage [3], to disturb
the hematic and gastrointestinal systems and to cause other toxic e¡ects [4].
Unlike organic pollutants, heavy metals cannot be degraded, but instead persist inde¢nitely in the environment,
complicating their remediation. The techniques presently
used for the clean-up of soils contaminated by heavy metals are mainly based on ex situ decontamination, using
physico-chemical methods of extraction, which are very
expensive. In addition, they destroy the soil structure
and leave the soil biologically inactive [5].
Certain plants have the ability to grow on heavy-metalcontaminated soils, and some of them can accumulate
heavy metals, which have no known biological function
[6]. Such plants are investigated for their potential use in
in situ phytoremediation, i.e. the use of plants to remove,
contain or transform environmental contaminants, which
is considered one of the most promising new technologies
for heavy metal remediation [7]. Metals such as antimony,
arsenic, cadmium, copper, mercury, nickel, selenium and
zinc can be extracted from the soil by hyperaccumulator
plants and concentrated into their roots and shoots. This
process is called phytoextraction, as the metal-enriched
plant biomass can be harvested using standard agricultural
0168-6496 / 03 / $22.00 4 2003 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/S0168-6496(03)00206-X
FEMSEC 1567 17-9-03
74
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
methods and smelted to recover the metal [8]. However,
most of the plants known to be metal hyperaccumulators
emerged spontaneously, so their physiology, ecology and
agronomic performances are often scarcely known.
In nature, the roots of plants interact with a large number of di¡erent microorganisms, and these interactions,
together with soil and climatic conditions, are major determinants of the extent to which plants grow and spread
[9]. Among soil microorganisms, several bacterial genera
stimulate root proliferation and protect surrounding
plants by producing auxins and/or other compounds
[10^15]. They are called plant growth-promoting rhizobacteria (PGPR) [12]. Recent studies have shown that hyperaccumulator plants grown in heavy-metal-contaminated
soils presented a more proliferated root system than in
non-contaminated soils [16,17]. The phenomenon of
PGPR has been investigated to clarify their potential
role in the phytoextraction process. Indian mustard plants
(Brassica juncea L.), when germinated in vitro on selenium-containing media [18] or in pot experiments with
cadmium-contaminated soils [19], produced more root
hairs when inoculated with PGPR strains. Furthermore,
these inoculated plants accumulated more selenium than
the non-inoculated ones [18]. Thus, PGPR are promising
bacterial partners for the improvement of plant growth on
heavy-metal-contaminated soil, and thus for phytoremediation.
In the case of antimony-polluted soils, plants such as
Achillea ageratum L., Plantago lanceolata L. and Silene
vulgaris (Moench) Garcke were recently investigated for
phytoextraction purposes [2]. In fact, these plants have
been shown to be the best antimony hyperaccumulators
among 59 plant species tested, and this regardless of the
antimony concentration in the soil [20]. However, no data
exist on their root-associated micro£ora.
The aim of this research was to characterize the diversity of a PGPR population in antimony-contaminated
soils, and how this is in£uenced by metal and plant. In
particular, we focused on PGPR genotypes present in antimony-contaminated soils regardless of the concentration
of the pollutant, and the overall development of hyperaccumulator plants. For this purpose, thousands of rhizobacteria were isolated at three stages of plant growth from
the rhizospheres of A. ageratum L. (Compositae), naturally colonizing a site contaminated by antimony 25 years
ago.
producing reference strains were used: P. £uorescens Pf5
[11] and Pseudomonas corrugata 3-1 [22].
2.2. Characterization of the site
The model site was a 25-year-old abandoned open-sky
antimony mine in Tuscany (Italy). This site is characterized by the existence of a gradient in the concentration of
antimony. The description of the site contamination status
was already available thanks to previous studies ([2,20],
Boscagli, personal communication). Based on this knowledge, four soils were chosen for plant collection: soils with
low (I), moderate (II), high (III) and very high (IV) antimony concentrations (Table 1).
2.3. Isolation and selection of bacteria able to produce
auxins in vitro
The rhizosphere of four individual A. ageratum L.
plants for each of the four soils was collected at three
stages of plant growth, corresponding to (1) the emergence
of new roots and shoots from the rhizome, (2) the beginning of £owering and (3) the death of the £ower shoots
and basal leaves. In this way, a total of 48 rhizospheric
samples were collected (four plantsUfour soil antimony
concentrationsUthree stages of plant growth). Immediately after each collection, plants with their rhizospheres
were taken to the laboratory, and isolation of rhizobacteria was performed within 24 h.
For each sample, 10 g of roots were washed in 100 ml of
sterile water, and then blended in a Waring blender with
100 ml of sterile water. Serial dilutions were plated onto
solid tryptone soya medium (Oxoid, Basingstoke, Hampshire, UK). The plates were incubated at 27‡C, and numbers of cfu were determined after 7 days.
For each sample, 94 randomly selected colonies were
picked and each colony was separately transferred into a
single well of a 96-well microplate containing 100 Wl of
liquid tryptone soya medium. In total, 4512 bacterial colonies were collected in this way (94 isolatesU48 rhizospheres). After 3 days of growth at 27‡C in the dark,
Table 1
Average antimony contents in the soils, as reported by Baroni et al. [20]
Soil
I (n = 4)
II (n = 4)
III (n = 7)
IV (n = 5)
2. Materials and methods
2.1. Bacterial strains
Four auxin-producing reference strains were used: Pseudomonas £uorescens M114 [13] and CHA0 [15], Pseudomonas putida M.3.1 (Digat, INRA, Angers, France) and
Pseudomonas aureofaciens PGS12 [21]. Two non-auxin-
Antimony concentration (mg kg31 of dried soil)
Totala
Solubleb
Extractablec
27.74
114.80
1 314.83
16 388.75
0.08
0.10
0.48
15.76
0.18
1.59
3.12
798.38
n = number of samples analyzed.
a
Total antimony content was measured by digestion of 0.5 g of dried
soil with a 1:1 (v:v) HCl+HNO3 mixture.
b
The estimated soluble antimony content was obtained by water extraction (100 ml of bidistilled water for 50 g of dried soil).
c
The estimated extractable antimony content was obtained by acetic
acid extraction (200 ml of 0.43 mol acetic acid for 5 g of dried soil).
FEMSEC 1567 17-9-03
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
75
auxin-producing isolates (auxþ ) were detected by a colorimetric technique derived from that of Salkowski [23^25],
known to be speci¢c for the detection of indolic compounds (IAA, IpyA and IAM) [24]. The Salkowski technique was adapted to the 96-well microplate format by
adding 200 Wl of a 35% perchloric acid^0.01 M FeCl3 solution to each 100 Wl culture. The speci¢city of the test was
systematically veri¢ed by including the four auxþ reference
strains, for which the addition of the Salkowski reagent
should result in a pink coloration when auxin compounds
are produced in the medium, and the two non-auxin-producing (aux3 ) reference strains, for which no visually detectable color reaction is produced. Among the 4512 isolates, only those presenting the diagnostic pink coloration
were puri¢ed and multiplied on new solid tryptone soya
medium. The test was repeated on pure cultures for con¢rmation of auxin production. Then, each puri¢ed auxþ
isolate was preserved at 380‡C in 40% glycerol.
2.4. Cell preparation prior to polymerase chain reaction
(PCR) ampli¢cation
Bacterial isolates were grown overnight at 27‡C on
YPGA, which contained (per liter) 5 g of yeast extract,
5 g of peptone, 10 g of glucose and 15 g of agar. A single
colony of each strain was suspended in 50 Wl of distilled
water and heated at 95‡C for 10 min to lyse the cells. The
lysate was then cooled in ice, brie£y centrifuged in a microcentrifuge, and used for PCR ampli¢cation.
2.5. Ampli¢ed rDNA restriction analysis (ARDRA)
The DNA encoding the 16S rRNA of each isolate was
ampli¢ed using the universal primers fD1 and rD1 [26],
corresponding to positions 8^27 and 1524^1540 on the
Escherichia coli rrs gene sequence. This primer pair is capable of amplifying nearly full-length 16S ribosomal DNA
from a wide variety of bacterial taxa [26]. Ampli¢cation
was performed as previously described [22] ; each mixture
contained 2 Wl of lysed cell suspension in 25 Wl of reaction
bu¡er containing 1.5 mM MgCl2 , 150 ng of each primer,
each deoxynucleoside triphosphate at a concentration of
250 mM, and 0.5 U of Taq DNA polymerase. The reaction mixtures were incubated in a thermocycler at 95‡C for
1.5 min and then subjected to 35 cycles consisting of 95‡C
for 30 s, the annealing temperature for 30 s, and 72‡C for
4 min. The annealing temperature was 60‡C for the ¢rst
¢ve cycles, 55‡C for the next ¢ve cycles, and 50‡C for the
last 25 cycles. Finally, the mixtures were incubated at 72‡C
for 10 min and then at 60‡C for 10 min. Five Wl of each
ampli¢cation mixture was analyzed by agarose (1.2% w/v)
gel electrophoresis in Tris-borate-EDTA (TBE) bu¡er.
Restriction analysis was performed with 5 Wl of ampli¢ed product and 15 Wl of restriction bu¡er containing 3 U
of either MspI, Sau96I, HinfI, RsaI, HhaI, TaqI or HindIII (Roche Molecular Biochemicals). After a 3-h diges-
Fig. 1. Percentages of auxin-producing isolates along the antimony concentration gradient (A) and at the di¡erent stages of plant growth (B).
Each column represents the mean of four replicates and bars indicate
the S.E.M. White, emerging stage; gray, £owering stage; black, death
stage. Soils are described in Table 1.
tion at the appropriate temperature, the enzyme was inactivated by heating the preparations at 70‡C for 15 min,
and the restriction fragments were separated by gel electrophoresis (2% agarose) in TBE bu¡er. A 1-kb DNA
ladder (Invitrogen) was used as molecular size marker.
For each isolate, PCR ampli¢cation and restriction analysis were performed at least three times.
2.6. Statistical analysis
The AMOVA procedure was used to estimate the variance components for ARDRA patterns by partitioning
the variations among plant growth stages and/or antimony
FEMSEC 1567 17-9-03
76
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
Table 2
Number of auxþ isolates from the rhizosphere of Achillea and their distribution among the six main ARDRA clusters
Plant growth stagea
Em
Fl
De
Total isolates
a
Soil
I
II
III
IV
I
II
III
IV
I
II
III
IV
using the Biolog Microlog software. E⁄ciency of Biolog
tests was checked by including reference strains 3-1 and
M.3.1
ARDRA cluster
1
2
3
4
5
6
3
8
3
8
2
1
3
1
2
2
4
4
42
5
5
5
0
0
0
1
1
1
0
1
2
21
2
1
0
1
0
1
1
0
5
4
1
2
18
0
0
2
0
2
0
1
4
0
0
0
1
10
2
0
0
0
0
2
0
1
4
1
0
0
10
0
2
0
0
0
0
2
0
4
0
2
0
10
Em, emerging stage; Fl, £owering stage; De, death stage.
concentrations. The AMOVA technique is a method for
analyzing molecular variance that produces estimates of
variance components re£ecting the correlation of haplotypic diversity at di¡erent levels of a hierarchical subdivision. The signi¢cance of the variance components is tested
by a permutational approach [27]. The vectors for the
presence of ARDRA markers (1 for the presence of each
band on a gel; 0 for the absence of each band on a gel) for
each strain were used to compute the genetic distance for
each pair of strains. The parameter used was the Euclidean metric measurement (E) of Exco⁄er et al. [27], as
de¢ned by Hu¡ et al. [28] as follows : E = O2xy = n(132nxy /
2n), where 2nxy is the number of markers shared by two
strains and n is the total number of polymorphic sites. All
analyses were performed with the Arlequin program [29],
which is used in several scienti¢c ¢elds (microbiology,
medicine, population genetics) and is available at the following URL: http://anthropologie.unige.ch/arlequin/.
For each collection, the number of auxþ isolates is the
mean of four replicates. Standard error between replicates
was determined by analysis of variance followed by Tukey’s multiple comparison test (GraphPad PRISM Software, version 2.0).
3. Results
3.1. Isolation of auxin-producing bacterial strains
Bacteria isolated on solid tryptone soya medium were
enumerated 7 days after plating for each of the four soils
examined. Stages of plant growth and antimony concentration in the soil had no e¡ect on the densities of the total
cultivable bacterial populations, since they ranged independently from 4U107 to 2U108 cfu g31 of dry roots.
A total of 4512 colonies (94 isolatesU48 rhizospheres)
were randomly picked and tested for auxin production.
Auxin production was veri¢ed for 222 isolates, by comparison of their pink culture coloration with that of the four
auxin-producing reference strains (below, these isolates are
referred to as auxþ isolates). No color change was observed for the culture medium of the auxin-negative reference strains 3-1 and Pf5. The percentage of auxþ isolates
increased with increasing concentration of antimony in the
soil (Fig. 1A). This increase was quite low when bacteria
were isolated from plants at the emerging stage, since the
frequency of auxþ isolates varied from 4% of the total
bacterial population isolated from soil I, to 5.7% from
soil IV. For the other plant growth stages, the increase
of the frequency of auxþ isolates with the increase of
soil antimony concentration was more evident, i.e. from
6.5% and 4% (soil I, £owering and plant death stage, respectively) to 20% and 15% (soil IV, £owering and plant
death stage, respectively). On the other hand, some £uctuation of the total number of auxþ isolates was observed
with the stages of plant growth, since the frequency of
auxþ isolates increased from the emerging stage (5%) to
the £owering stage (11.73%), and decreased a little with
the death of the £ower shoots and basal leaves (8.25%)
(Fig. 1B).
Table 3
Gel-detectable restriction fragment sizes (in bp) of ampli¢ed 16S rDNAs
from main ARDRA clusters and from two reference strains
2.7. Characterization of the isolates
Gram characteristics of the isolates were determined by
the KOH test of Suslow et al. [30]. Isolates were phenotypically characterized using the Biolog system (Biolog,
Hayward, CA, USA), which is based on the di¡erential
utilization of a large number of organic compounds. Biolog GN microplates were inoculated as recommended by
the manufacturer and were incubated at 27‡C for 24 h.
Formazan accumulation in bacterial cells was measured by
determining the optical density at 550 nm with an automatic microplate reader. The isolates were identi¢ed on
the basis of their patterns of utilization of 95 substrates
Restriction endonuclease
HinfI
ARDRA cluster
1
1000, 200
2
1000, 200
3
510, 360, 330
4
1200
5
1100
6
1100
Reference strains
3-1
1000, 200
CHA0
1000, 200
FEMSEC 1567 17-9-03
MspI
RsaI
550,
550,
510,
700,
510,
700,
850,
800,
510,
820,
820,
820,
460, 150
350, 150
290
290
350, 250
290
550, 460, 150
550, 350, 150
350, 150
350, 150
450
450
450
450
850, 350, 150
800, 350, 150
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
Table 4
Distribution of ARDRA clusters as a¡ected by soil and plant parameters
Plant growth stagea
Soil
Total number of
ARDRA clusters
a
I
II
III
IV
Em
Fl
De
18
20
21
21
21
28
26
Em, emerging stage; Fl, £owering stage; De, death stage.
77
by only one isolate. The remaining clusters (ARDRA-7^
25) included two to nine isolates that never originated
from more than three soils and two stages of plant growth.
The number of diverse clusters isolated from each soil
increased with its antimony concentration, from 18 for soil
I to 21 for soils III and IV. Considering the stages of plant
growth, the number of diverse clusters was lower at the
emerging stage (21 cluster) than at the £owering and the
plant death stages (28 and 26, respectively) (Table 4).
3.2. ARDRA analysis
3.3. AMOVA analysis
The 16S rDNAs of 214 out of 222 auxþ isolates were
successfully ampli¢ed (data not shown). The remaining
isolates were not investigated since they did not survive
the cryoconservation in glycerol. The results of ARDRA
with seven enzymes were subjected to cluster analysis. The
di¡erent patterns obtained with each of the seven endonucleases were readily distinguishable from one another and
highly reproducible from one experiment to the next. ARDRA with HinfI, MspI, and RsaI revealed a signi¢cant
level of genetic diversity between the isolates, since 51
clusters were found (designated ARDRA-1 to ARDRA51). Isolates exhibiting the same HinfI-MspI-RsaI cluster
were indistinguishable when analysis was carried out with
any of the four other endonucleases (data not shown).
In particular, one cluster (ARDRA-1) was predominantly represented, corresponding to 42 (19.6%) isolates.
Furthermore, it was present in the Achillea rhizosphere for
all the soil antimony concentrations and throughout the
stages of plant growth (Table 2). Five other clusters (ARDRA-2^6) included 21 isolates (10%), 18 isolates (8.4%)
(respectively for ARDRA-2 and -3) and 10 isolates (4.7%)
(for ARDRA cluster 4^6). They were present throughout
the stages of plant growth of Achillea, but in not more
than three of the four soils (Table 2). Restriction patterns
of these six main ARDRA clusters are presented in Table
3. Twenty-six clusters (ARDRA-26^51) were represented
Each ARDRA cluster was compared with the other
clusters, and an Euclidean distance matrix (E) was calculated (data not shown). To analyze the ARDRA variation
in the 214 isolates, we performed an AMOVA with the
Euclidean distance matrix. Antimony concentrations in
soil and stages of plant growth were considered two di¡erent groups, to evaluate independently their respective effect on the biodiversity of auxþ isolates. The AMOVA
data revealed that 4.01% of the genetic diversity observed
between the auxþ isolates was related to the soil antimony
concentration, at a signi¢cant P value (P = 0.04). We also
analyzed all of the possible combinations of two soils. The
data obtained (Table 5) clearly show that a greater di¡erence in antimony concentration between two soils (I vs.
IV) corresponds to a higher genetic diversity between the
auxþ isolates (7.95%, P = 0.04).
Similar analyses carried out on the stage of plant
growth (Table 5) show that a part of the genetic variability
between the auxþ isolates is due to this factor (7.74%,
P = 0.04, and up to 10.76% between £owering and death
stages, P = 0.07).
3.4. Identi¢cation of the most representative
aux+ isolates
The 42 isolates belonging to the main ARDRA cluster
Table 5
Genetic diversity between the auxþ isolates determined by the AMOVA analysis on the ARDRA pattern variation
Soil
Stage of plant growthb
Variance component
% of variance between sampling
% of variance within sampling
Pa
I vs. II vs. III vs. IV
(I+II) vs. (III+IV)
I vs. II
II vs. III
III vs. IV
I vs. III
I vs. IV
II vs. IV
Em vs. Fl vs. De
Em vs. Fl
Em vs. De
Fl vs. De
4.01
3.31
1.60
1.57
1.73
4.99
7.95
2.67
7.74
10.76
7.82
5.77
95.99
96.69
98.40
98.43
98.27
95.01
92.05
97.33
92.26
89.24
92.18
94.23
0.04
0.03
0.03
0.01
0.01
0.04
0.04
0.02
0.04
0.07
0.05
0.02
Values in boldface type are not signi¢cant.
a
Probability of a more extreme variance distribution.
b
Em, emerging stage; Fl, £owering stage; De, death stage.
FEMSEC 1567 17-9-03
78
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
(ARDRA-1) were all Gram-negative. Biological tests performed on 16 isolates belonging to ARDRA-1 and the two
reference strains 3-1 and M.3.1 showed that nine of the
isolates should be assigned to P. corrugata, with similarity
levels of at least 95%, as well as strain 3-1. The seven other
isolates were assigned to P. putida biotype B (similarity
levels of 94^98%), as well as reference strain M.3.1.
4. Discussion
The e¡ect of heavy metal contamination on soil microbial populations has often been analyzed by biochemical
techniques, such as measurements of respiration and biomass. Less numerous are the studies that have analyzed
the e¡ect of heavy metals on soil microbial community
structure or on bacterial diversity, and, to our knowledge,
no data exist on the in£uence of long-term soil antimony
contamination on the bacterial auxin-producing community present in the rhizosphere of hyperaccumulator
plants. The design of heavy metal phytoextraction applications needs detailed data on both plant and plant
growth-promoting bacterial genotypes for plant inoculation. Bearing that in mind, we showed in the present research that the auxin-producing population present in the
Achillea rhizosphere was genetically very diverse, and that
this diversity was in part due to the antimony concentration in soil, and in part to the plant physiological stage.
However, only one ARDRA cluster of auxþ was isolated
throughout plant growth development, regardless of the
antimony concentration in the soil. This highlights the
importance of considering not only the heavy metal concentration in the soil but also the plant development
stage, when screening bacterial strains for such applications.
Concerning the antimony concentration parameter, the
total rhizospheric bacterial population did not change
along the gradient, while the auxþ bacterial population
£uctuated with the concentration of this pollutant. This
is in apparent contradiction with numerous studies on
other heavy metal soil contamination, showing a decrease
of the total microbial population with increasing metal
concentration [6,31,32]. Furthermore, we observed that a
part of the genetic variability between auxþ isolates is
associated with the antimony concentration in the soil,
and that the number of diverse clusters isolated from
each soil increases with the antimony concentration. These
data are also in contradiction with numerous other studies
that noted a lower diversity index between bacteria isolated from high metal-contaminated soils, compared to
low metal-contaminated soils [33^36]. However, it is not
possible to directly compare the results of the present
study with those of the above cited works, as many of
them investigated short-term e¡ects of heavy metals in
experimentally contaminated soils [34,35], or long-term
contaminated soils, but without plants [33,36]. Fewer in-
vestigations address the e¡ects of heavy metals on microorganisms in the rhizosphere of plants naturally colonizing
long-term polluted soils. For nickel-rich soils, it has been
shown that bacteria isolated from the rhizosphere of the
hyperaccumulator plant Alyssum bertolonii were genetically greatly diverse from those isolated from the non-rhizospheric soil [37]. Another study showed that the zinc
concentration has no or only weak e¡ects on the total
number of bacteria isolated from the rhizosphere of Thlaspi caerulescens or Trifolium pratense [38], which is completely in accord with our data. Concerning speci¢c bacterial populations such as PGPR, an investigation on
Rhizobium leguminosarum biovar trifolii in a range of
heavy-metal-contaminated soils is partly in agreement
with our data, since it observed that heavy metals have
a quantitative e¡ect on the population of rhizobia, but not
on the genetic diversity [39].
Concerning the plant physiological stage parameter, the
numerical £uctuation of the auxþ bacterial population
could be explained by assuming that there was a selection
for the auxin producers resulting from the composition of
the root exudates, which are known to evolve with plant
development [9]. Furthermore, our results showed that the
plant stage not only a¡ects the number of auxþ isolates,
but also the diversity of their populations. The highest
level of genetic diversity between auxþ isolates was found
when the Achillea roots were highly colonized by this type
of bacteria (£owering and shoot death stages), suggesting
that the selection of particular auxþ strains occurs when
the roots are old. These results are consistent with published data for other plant species, which showed that the
stage of plant growth signi¢cantly a¡ects the level of genetic diversity of PGPR populations, and that this phenomenon is related to the quantity and quality of the root
exudates [22].
The results of this research have practical importance in
the context of using plant growth-promoting agents in
heavy metal phytoextraction experiments. Indeed, we
found that only the ARDRA-1 auxþ genotype is present
in the Achillea rhizosphere throughout plant growth development, regardless of the antimony concentration present
in the soil. Thus, we hypothesize that if it were included in
an inoculum, it would e⁄ciently colonize the Achillea rhizosphere. Furthermore, these strains were identi¢ed as
P. corrugata or P. putida biotype B. Strains belonging to
these species are widely distributed in nature and are often
associated with a variety of plant species [19,22,40]. Some
of them, isolated from seeds, hypocotyl tissue or the rhizosphere, are e¡ective PGPR strains used as an inoculum to
promote the root development of plants in agricultural
systems [41], as well as in cadmium-contaminated environments [19]. Releasing such auxþ strains in the Achillea
rhizosphere could be of particular help in the phytoextraction of the antimony contaminant, especially in highly
contaminated soils that naturally support a very poor
£ora.
FEMSEC 1567 17-9-03
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
Acknowledgements
The initial part of this work was done at the DiBA of
the Universita' degli Studi di Firenze (Italy) and at the
ENSAIA of Nancy (France). We received grants from
the OECD (C.P.), and from the Italian Ministry of University and Scienti¢c and Technological Research (PRIN
9905212224_003 to M.B.). We are grateful to Aldemaro
Boscagli (Universita' degli Studi di Siena, Italy) for help in
antimony mine site samplings and soil analysis, and to
Gianni Simonti for technical assistance. We thank M.
Schroth (University of California, USA), F. O’Gara (University College Cork, Ireland), G. De¤fago (ETH Zurich,
Switzerland), M. Digat (INRA Angers, France), and J.
Loper (Horticultural Crops Research Laboratory, Corvallis, USA) for providing reference bacterial strains.
[14]
[15]
[16]
[17]
[18]
[19]
References
[1] Giller, K., Witter, E. and McGrath, S. (1998) Toxicity of heavy
metals to microorganisms and microbial process in agricultural soils,
A review. Soil Biol. Biochem. 30, 1389^1414.
[2] Baroni, F., Boscagli, A., Protano, G. and Riccobono, F. (2000) Antimony accumulation in Achillea ageratum, Plantago lanceolata and
Silene vulgaris growing in an old Sb-mining area. Environ. Pollut.
109, 347^352.
[3] Schaumlo¡el, N. and Gebel, T. (1998) Heterogeneity of the DNA
damage provoked by antimony and arsenic. Mutagenesis 1, 3281^
3286.
[4] Hammel, W., Steubing, L. and Debus, R. (1998) Assessment of the
ecotoxic potential of soil contaminants by using a soil-algae test.
Ecotox. Environ. Safe 40, 173^176.
[5] McGrath, S. (1998) Phytoextraction for soil remediation. In: Plants
that Hyperaccumulate Heavy Metals (Brooks, R., Ed.), pp. 261^287.
CAB International, Wallingford.
[6] Brooks, R., Lee, J., Reeves, R. and Ja¡re, T. (1977) Detection of
nickeliferous rocks by analysis of herbarium species of indicator
plants. J. Geochem. Explor. 7, 49^57.
[7] McGrath, S., Zhao, F. and Lombi, E. (2001) Plant and rhizosphere
processes involved in phytoremediation of metal-contaminated soils.
Plant Soil 232, 207^214.
[8] Cunningham, S. and Lee, C. (1995) Phytoremediation : Plant-based
remediation of contaminated soils and sediments. In: Special Publication No. 43: Bioremediation : Science and Applications (Skipper,
H. and Turco, R., Eds.), pp. 145^156. Soil Science Society of America, Madison, WI.
[9] Lynch, J. and Whipps, J. (1990) Substrate £ow in rhizosphere. Plant
Soil 129, 1^10.
[10] Andersen, J.B., Koch, B., Nielsen, T.H., SYrensen, D., Hansen, M.,
Nybroe, O., Christophersen, C., SYrensen, J., Molin, S. and Givskov,
M. (2003) Surface motility in Pseudomonas sp. DSS73 is required for
e⁄cient biological containment of the root-pathogenic microfungi
Rhizoctonia solani and Pythium ultimum. Microbiology 149, 37^
46.
[11] Howell, C. and Stipanovic, R. (1979) Control of Rhizoctonia solani in
cotton seedlings with Pseudomonas £uorescens and with an antibiotic
produced by the bacterium. Phytopathology 69, 480^482.
[12] Kloepper, J. and Schrot, M. (1978) Plant growth-promoting rhizobacteria on radishes. Proc. Int. Conf. Plant Pathog. Bact. 2, 879^882.
[13] O’Gara, F., Treacy, P., O’Sullivan, D., O’Sullivan, M. and Higgins,
P. (1986) Biological control of phytopathogens by Pseudomonas sp.:
genetic aspects of siderophore production and root colonisation. In:
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
79
Iron, Siderophores and Plant Diseases (Swinburne, T., Ed.), pp. 331^
339. Plenum Press, New York.
Picard, C. and Bosco, M. (2003) Genetic diversity of phlD gene from
2, 4-diacetylphloroglucinol-producing Pseudomonas spp. strains from
the maize rhizosphere. FEMS Microbiol. Lett. 219, 167^172.
Stutz, E., De¤fago, G. and Kern, H. (1986) Naturally occurring £uorescent pseudomonads involved in suppression of black root rot of
tobacco. Phytopathology 76, 181^185.
Schwartz, C., Morel, J-L., Saumier, S., Whiting, S. and Baker, A.
(1999) Root development of the zinc-hyperaccumulator plant Thlaspi
caerulescens as a¡ected by metal origin, content and location in soil.
Plant Soil 208, 103^115.
Whiting, S., Leake, J., McGrath, S. and Baker, A. (2000) Positive
responses to Zn and Cd by roots of the Zn and Cd hyperaccumulator
Thlaspi caerulescens. New Phytol. 145, 199^210.
De Souza, M., Chu, D., Zhao, M., Zayed, A., Ruzin, S., Schichnes,
D. and Terry, N. (1999) Rhizosphere bacteria enhance selenium accumulation and volatization by Indian mustard. Plant Physiol. 119,
565^573.
Belimov, A., Safronova, V., Sergeyeva, T., Egorova, T., Matveyeva,
V., Tsyganov, V., Borisov, A., Tikhonovich, I., Kluge, C., Preisfeld,
A., Dietz, K. and Stepanok, V. (2001) Characterization of plant
growth promoting rhizobacteria isolated from polluted soils and containing 1-aminocyclopropane-1-carboxylate deaminase. Can. J. Microbiol. 47, 642^652.
Baroni, F., Boscagli, A., Protano, G. and Riccobono, F. (2000) Antimony contents in plant species growing in an Sb-mining district (Tuscany, Italy). In: Trace Elements ^ Their Distribution and E¡ects in
the Environment (Market, B. and Friese, K., Eds.), pp. 341^361.
Elsevier, Amsterdam.
Georgakopoulos, D., Hendson, M., Panopoulos, N. and Schroth, M.
(1994) Cloning of a phenazine biosynthetic locus of Pseudomonas
aureofaciens PGS12 and analysis of its expression in vitro with the
ice nucleation reporter gene. Appl. Environ. Microbiol. 60, 2931^
2938.
Picard, C., Di Cello, F., Ventura, M., Fani, R. and Guckert, A.
(2000) Frequency and biodiversity of 2,4-diacetylphloroglucinol-producing bacteria isolated from the maize rhizosphere at di¡erent stages
of plant growth. Appl. Environ. Microbiol. 66, 948^955.
Benizri, E., Courtade, A., Picard, C. and Guckert, A. (1998) Role of
maize root exudates in the production of auxins by Pseudomonas
£uorescens M3.1. Soil Biol. Biochem. 30, 1481^1484.
Glickmann, E. and Dessaux, Y. (1995) A critical examination of the
speci¢city of the Salkowski reagent for indolic compounds produced
by phytopathogenic bacteria. Appl. Environ. Microbiol. 61, 793^796.
Salkowski, E. (1885) Ueber das Verhalten der Skatolcarbosaure im
Organismus. Z. Physiol. Chem. 923^933.
Weisburg, W., Barns, S., Pelletier, D. and Lane, D. (1991) 16S ribosomal DNA ampli¢cation for phylogenetic study. J. Bacteriol. 173,
697^703.
Exco⁄er, L., Smouse, P. and Quattro, J. (1992) Analysis of molecular variance inferred from metric distances among DNA genotypes:
application to human mitochondrial DNA restriction data. Genetics
131, 479^491.
Hu¡, D., Peakall, R. and Smouse, P. (1993) RAPD variation among
natural populations of outcrossing bu¡alograss Buchloe dactyloides
(Nutt.) Engelm. Theor. Appl. Genet. 86, 927^934.
Schneider, S., Kue¡er, J.-M., Roessli, D. and Exco⁄er, L. (1997)
Arlequin ver. 1.1: a sofware for population genetic data analysis.
Genetics and Biometry Laboratory, University of Geneva, Geneva.
Suslow, T., Schroth, M. and Isaka, M. (1982) Application of a rapid
method for Gram di¡erentiation of plant pathogenic and saprophytic
bacteria without staining. Phytopathology 72, 917^918.
Bafiafith, E., Diaz-Ravina, M., Frostegard, A. and Campbell, C. (1998)
E¡ect of metal-rich sludge amendments on the soil microbial community. Appl. Environ. Microbiol. 64, 238^245.
Roane, T. and Kellogg, S. (1996) Characterization of bacterial com-
FEMSEC 1567 17-9-03
80
[33]
[34]
[35]
[36]
C. Picard, M. Bosco / FEMS Microbiology Ecology 46 (2003) 73^80
munities in heavy metal contaminated soils. Can. J. Microbiol. 42,
593^603.
Ellis, R., Neish, B., Trett, M., Best, G., Weightman, A., Morgan, P.
and Fry, J. (2001) Comparison of microbial and meiofaunal community analyses for determining impact of heavy metal contamination.
J. Microbiol. Methods 45, 171^185.
Fritze, H., Perkiomaki, J., Saarela, U., Katainen, R., Tikka, P., Yrjala, K., Karp, M., Haimi, J. and Romantschuk, M. (2000) E¡ect of
Cd-containing wood ash on the micro£ora of coniferous forest humus. FEMS Microbiol. Ecol. 32, 43^51.
Kelly, J., Haggblom, M. and Tate, R. (1999) Changes in soil microbial communities over time resulting from one time application of
zinc: a laboratory microcosm study. Soil Biol. Biochem. 31, 1455^
1465.
Sandaa, R., Torsvik, V. and Enger, O. (2001) In£uence of long-term
heavy-metal contamination on microbial communities in soil. Soil
Biol. Biochem. 33, 287^295.
[37] Mengoni, A., Barzanti, R., Gonnelli, C., Gabbrielli, R. and Bazzicalupo, M. (2001) Characterization of nickel-resistant bacteria isolated
from serpentine soil. Environ. Microbiol. 3, 691^698.
[38] Delorme, T., Gagliardi, J., Angle, J. and Chaney, R. (2001) In£uence
of zinc hyperaccumulator Thlaspi caerulescens J. and C. Presl. and
the non metal accumulator Trifolium pratense L. on soil microbial
populations. Can. J. Microbiol. 47, 773^776.
[39] Obbard, J., Sauerbeck, D. and Jones, K. (1993) Rhizobium leguminosarum biovar. Trifolii in soils amended with heavy metal contaminated sewage sludges. Soil Biol. Biochem. 25, 227^231.
[40] SYrensen, J., Skouv, J., JYrgensen, A. and Nybroe, O. (1992) Rapid
identi¢cation of environmental isolates of Pseudomonas aeruginosa,
P. £uorescens and P. putida by SDS-PAGE analysis of whole-cell
protein-patterns. FEMS Microbiol. Ecol. 101, 41^50.
[41] Patten, C. and Glick, B. (2002) Role of Pseudomonas putida indoleacetic acid in development of the host plant root system. Appl. Environ. Microbiol. 68, 3795^3801.
FEMSEC 1567 17-9-03