Research in Microbiology 161 (2010) 101e108

www.elsevier.com/locate/resmic

Exopolysaccharide production by nitrogen-fixing bacteria within nodules

of Medicago plants exposed to chronic radiation in the Chernobyl
exclusion zone
Nathalie Pawlicki-Jullian a,*, Bernard Courtois a, Michelle Pillon a, David Lesur b,
Anne Le Flèche-Mateos c, Jean-Claude Laberche d, Nadia Goncharova e, Josiane Courtois a
a
Laboratoire des Polysaccharides Microbiens et végétaux, IUT Dpt GB, 80025 Amiens, France
b
Laboratoire des Glucides UMR CNRS 6219, 80039 Amiens, France
c
Urgent Response to Biological Threats, Institut Pasteur Paris, 75724 Paris, France
d
Laboratoire de Biologie des Plantes et Contrôle des Ravageurs EA 3900, 80039 Amiens, France
e
UNESCO Chair International Sakharov Environmental University, 220009 Minsk, Belarus
Received 26 May 2009; accepted 23 December 2009
Available online 18 January 2010

Abstract

Nitrogen-fixing bacteria isolated from root nodules of Medicago plants growing in the 10 km zone around the Chernobyl nuclear power plant
were screened for the production of new water-soluble acidic exopolysaccharides (EPSs). The different strains belonged to the Enter-
iobacteriaceae family (Enterobacter ludwigii, Raoultella terrigena, Klebsiella oxytoca), except for one which belonged to the Rhizobiaceae
family (Sinorhizobium meliloti). All of the bacteria produced highly viscous EPS with an average molecular weight comprised between 1  106
and 3  106 Da. Five different compositions of EPS were characterized by physico-chemical analyses and 1H NMR spectroscopy: galactose/
mannose (2/1), galactose/glucose (1/1), galactose/glucose/mannose (1/2/1), fucose/galactose/glucose (2/1/1) and fucose/galactose/glucose/
mannose (2/2/1/1 or 1/1/2/4). Glucuronic acid, a charged monosaccharide, was also recovered in most of the different EPSs.
Ó 2010 Elsevier Masson SAS. All rights reserved.

Keywords: Chernobyl exclusion zone; Exopolysaccharides; Nitrogen-fixing bacteria; Enterobacteriaceae; Medicago; Diversity

1. Introduction nitrifying and sulfate-reducing bacteria in soil around the

The Chernobyl nuclear reactor accident in April 1986 had
great impact upon the radio-ecological situation in both the
Ukraine and Belarus, but also across all of Eastern Europe [4].
During the past twenty years, a great deal of information
concerning the consequences of the influence of chronic
ionizing radiation on humans, animals and plants has been
collected [35,37,13], but very few specific studies have been
carried out on soil bacteria. In 1998, Romanovskaya et al. [31]
showed a loss of diversity in the population of cellulolytic,
* Corresponding author. Tel.: þ33 322 534 093; fax: þ33 322 957 117.
E-mail address: nathalie.jullian@u-picardie.fr (N. Pawlicki-Jullian).
Chernobyl nuclear plant. However, some soil bacteria
(ammonifying, amylolytic, nitrogen-fixing, and spore-form-
ing) revealed themselves [11] to be more resistant to g-irra-
diation than microscopic fungi (eukaryotic organisms).
Likewise, soil bacteria from the genus Methylobacterium, that
are resistant to g-radiation, were also resistant to desiccation,
UV radiation and hydrogen peroxide [32]. Similar results were
also observed when X-rays were used to evaluate bacterial
resistance [42]. The survival of bacteria in the Chernobyl
accident zone is probably due to repairing of radiation-
damaged DNA. Bacteria that form nitrogen-fixing symbiotic
associations with legumes are usually distributed in the classes
of Alphaproteobacteria, with the well-known group of Rhi-
zobiaceae, and Betaproteobacteria [36,38,27]. Recently, new

0923-2508/$ - see front matter Ó 2010 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.resmic.2009.12.009
102 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108

types of legume symbionts have been isolated and identified as

belonging to Gammaproteobacteria [5]. Most often, these
bacteria secrete into the environment exopolysaccharides
(EPSs), which play a role in root nodule development [17] but
also in soil aggregation [2] and heavy metal entrapment [23].
In contaminated soils of the Chernobyl zone, bacterial endo-
symbionts of leguminous plants, may have been affected in
several ways, including that of damage to DNA repair systems
resulting in chronic genetic instability. Thus, genetic modifi-
cations involving plasmid content, nodulation, nitrogen fixa-
tion efficiency and quantitative and qualitative production of
exopolysaccharides have occurred.
The aim of this work was to utilize the Chernobyl exclusion
zone as an experimental laboratory in which Fabacea
nitrogen-fixing strains were in contact with strong mutagenic
agents in relation to chronic irradiation since the accident at
the Chernobyl nuclear power plant in April 1986. It would be
extremely advantageous to transform this very negative cat-
astrophy into a tool for isolating new metabolites of interest,
synthesized by isolated bacterial strains in Belarussian soil.
Forty-five different bacterial strains isolated from root nodules
of Medicago sativa and Medicago lupulina plants growing in
various contaminated zones of the Chernobyl area in Belarus Fig. 1. Exclusion zone of the Polesky State Radioecological Reserve (PSRR)
were compared for their genetic pool, their symbiotic prop- in the Chernobyl area (National report [35]) with the three defined zones (zone
1: 185 kBq/m2, zone 2: 555 kBq/m2, zone 3: 9250 kBq/m2). +: Nuclear power
erties towards alfalfa and their capacity to produce EPSs. plant, : border.
Particular attention was paid to isolating different original
polysaccharides with potential industrial applications in
different fields (cosmetics, medicine, environment, agronomy, family (M. sativa and M. lupulina) were harvested. Roots were
etc.). In order to determine whether component diversity in carefully cleaned and washed free of soil.
EPSs is derived from the mutation of bacteria caused by
radiation or originated from species variation, we decided to
2.2. Isolation and morphological characterization of the
use as the control plant in vitro alfalfa plants (seeds from
bacterial strains
Laboulet, France) with well-established culture conditions and
no radioactivity, and as control bacteria the Sinorhizobium
For isolation of bacteria, one to five nodules were separated
meliloti M5N1 strain [9], a reference used in our laboratory as
from the roots of the different plants, surface-sterilized and
the symbiont of Medicago plants. Because of the anthropic
crushed in sterile 20% glucose. The grinding was then cultured
status of the Chernobyl exclusion zone, it was impossible to
in tryptone yeast extract (TY) medium [6] at 30  C under
find identical plant ecotypes in another region. Likewise, the
shaking at 120 rpm for 24e48 h. Colonies were then isolated
International Sakharov Environmental University of Minsk, to
on Rhizobium complete (RC) medium [9]. Morphological
which one of our co-author belongs, did not have in its
features of colony size, growth rate, pigmentation, consistency
seedbank Medicago seeds prior to the Chernobyl catastrophy
and mucosity were evaluated on RC medium supplemented or
(Goncharova N., Pers. Comm.).
not with sucrose 1% (w/v). Gram analysis, the oxidase reac-
tion as well as the API 20E system were used to identify the
different bacterial strains. The production of EPSs was qual-
2. Materials and methods itatively assessed on RC medium, supplemented with aniline
blue that specifically reacts with b-(1,3)-glucoside linkages
2.1. Plant harvest [29], and on Luria broth (LB), supplemented with calcofluor
white (0.02%, w/v), a fluorescent dye known to bind a-(1e4)
The present study was conducted in Southern Belarus
linkages in glycans and EPSs [25]. All strains were maintained
(Gomel region, 52 200 N, 30 590 E) in the 30 km exclusion
in liquid RC at 4  C or in 20% (v/v) glycerol at 80  C.
zone of the Polesky State Radioecological Reserve (PSRR).
This site was established in June 1986 following the accident
at the Chernobyl nuclear power plant, and covers a total area 2.3. Plasmid analysis
of about 170,000 ha. From this site, we defined three zones
(Fig. 1) according to the levels of radioactive exposure doses The isolated strains were subjected to profile analyses using
(185, 555 and 9250 kBq/m2), and after official authorization to vertical 0.7% agarose gel electrophoresis according to the
enter the exclusion area, a total of 32 plants from the Fabacea method of Eckhardt [12].
 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108
2.4. DNA extraction and PCR amplification
Cells were lysed by suspending a loopful of plate-grown
isolated colony in 50 mL of sterile water in a 1.5 mL poly-
propylene tube. Then, they were stirred for 60 s on a vortex
and heating at 95  C for 10 min. Primers nodbox1 and nod-
box3 were used simultaneously with primers mucRf and
mucRr to amplify a 646 bp of the nodbox4 promoter [3] and
a 431 bp of the mucR gene [8]. PCR reactions were performed
according to the method described by Sanchez-Contreras [34].
2.5. Sequencing of the rpoB gene
Chromosomal DNAs were prepared by using the Promega
Genomic DNA purification kit (Promega, Madison, WI, USA).
The rpoB gene, coding for the beta subunit of RNA poly-
merase, was sequenced at Platform 8 of the Institut Pasteur’s
Genome Center, and the complete sequences were edited
using the Lasergene package (DNASTAR, Madison, WI,
USA). The different rpoB sequences, determined in this study,
were submitted to GenBank databases under accession number
GU199599 to GU199603.
2.6. Transmission electron microscopy
Nodules, separated from the roots of the different M. sativa
and M. lupulina plants harvested in various contaminated soils
were fixed for 3 h at 4  C in 3% (v/v) glutaraldehyde in
a 0.1 M sodium cacodylate buffer (pH 7.4), postfixed for 3 h at
4  C in 1% (v/v) osmium tetraoxide in the same buffer,
dehydrated in a graded ethanol series and propylene oxide, and
then embedded in araldite resin (Sigma). Ultrathin (0.05 mm)
sections prepared with an LKB ultratome III were laid on
copper grids, stained with uranyl acetate at room temperature
and then examined with a Jeol 1200EX transmission electron
microscope operating at 80 kV.
2.7. Plant cultivation and nodulation tests
Plant assays were performed with alfalfa seeds (Laboulet
Semences, France). Seeds were surface-sterilized and germi-
nated on RC medium at 30  C in darkness. Germinated
seedlings were aseptically transferred to tubes containing
a nitrogen-free liquid medium [40]. Forty-eight seedlings were
tested for each strain. Seedlings were grown under a 16/8 h
photoperiod at 30  C. Upon the appearance of secondary roots,
seedlings were inoculated with exponential phase liquid
bacteria (108 bact/mL). Four weeks after inoculation, effec-
tiveness was estimated by visual observation of plant vigor,
foliage color and by counting the number of formed nodules.
Acetylene reduction assays were carried out [18] with whole
plants in order to estimate nitrogenase activity.
2.8. Production and purification of EPS
For production of EPS, the glutamate-D-mannitol salt
(GMS) medium (pH 7.0) supplemented with biotin, thiamine
103
and trace elements as described by Zevenhuizen and van
Neerven [44] was used to prevent biochemical contamination
of EPS by glycomannans from yeast extract. After 72 h of
culture, the bacteria were separated by centrifugation
(14,000 g for 30 min), and the supernatant was precipitated by
using 3 vol. cold isopropanol (overnight, 4  C). The stringy
formed white mass was recovered by centrifugation (10,000 g
for 15 min), re-suspended in distilled water, and dialyzed
against distilled water (24 h at 4  C). The EPS solution was
subsequently freeze-dried and weighed.
2.9. Chemical analysis of EPS
The colorimetric m-hydroxydiphenyl [41] and resorcinol
methods [26] were used to determine the ratio between neutral
sugars and uronic acids. Proteins were estimated by the
Bradford method [7] using bovine serum albumin (Sigma) as
a standard. Mw of the different EPS was estimated by HPLC on
a linked column (Shodex OHpak KB-805, OHpak KB-806)
eluted with 0.1 M sodium nitrate with a flow rate of 0.7 mL/min
and coupled to a differential refractometer (RI, Shimadzu)
and a Mini-Dawn multi-angle laser light scattering detector
(MALLS, Wyatt Technology). A dextran standard was used to
calibrate the column. The different EPSs were hydrolyzed in
4 M trifluoroacetic acid (TFA) at 100  C for 3 h. The mono-
saccharide composition was performed using a Dionex ICS
3000 system and a pulsed amperometric detector. Aliquots of
the extract were passed through a 4  50 mm Propac PA1
pre-column (Dionex, France) before separation of anionic
compounds on a 4  250 mm Propac PA1 column (Dionex,
France) at 35  C. The column was equilibrated with eluant A
(90% H2O, 10% NaOH 100 mM) for 10 min before each sample
run. The solutions were gassed with helium. Gradient elution
was performed with a multi-step gradient as follows: 0e20 min,
90% H2O and 10% NaOH 100 mM, 20e25 min, 100% NaOH
100 mM, 25e60 min, 100% NaOH 100 mM, sodium acetate
500 mM, at a flow rate of 1 mL/min. Peak analysis was per-
formed using Chromeleon software, version 7.0. The results
were confirmed by analysis of the alditol acetate derivatives on
a HewlettePackard 6890 gas chromatograph coupled to a mass
selective detector with an HP1 column (HewlettePackard,
50 m), using He as the carrier gas at a flow of 0.9 mL/min. The
injection temperature was 250  C. The monosaccharides were
identified by comparison of their retention times with that of
pure standards analyzed under the same conditions. The relative
area of anomeric protons, and the presence of substituents
within the different EPS were performed by 1H nuclear
resonance (NMR) using a Bruker AM300-WB spectrometer in
D2O (99.9 atom 2H, Aldrich) solutions at 80  C.
2.10. Statistical analyses
Data from nodule occupancy were analyzed with non-
parametric tests [21] because of the too small number
of observations for each level of treatment. The different
means were compared by the Kruskal and Wallis test and the
104 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108
ManneWhitney test. The mean difference in the treatments
was considered to be significant at the level of P ¼ 0.05.
3. Results
3.1. Plant sampling and nodule occupancy
In order to isolate a collection of bacterial strains with
a variety of EPS production capacities and variabilities,
Medicago plants were collected from different contaminated
zones of the Chernobyl exclusion zone (Fig. 1). The root
system of all harvested plants bears nodules having various
numbers and shapes. In the least contaminated zone (zone 1,
185 kBq/m2), numerous nodules (20e25 per plant) were
observed over all the parts of a normal root system (Fig. 2A1).
They were 1e2 mm long, with a typical shape (Fig. 2A2), and
sometimes a pink coloring certainly due to the presence of
leghemoglobin. On the roots of the plants from the second
defined zone (zone 2, 555 kBq/m2) we noted the presence of
nodules either with a typical shape (2e5 mm long), or
sometimes with bifurcate or fan shapes (Fig. 2B). These
nodules were mainly observed on the lateral roots, but in
a reduced number (less than 10 per plant) compared with the
first zone. In the third and most contaminated zone (zone 3,
9250 kBq/m2), we observed plants with thick primary roots
(Fig. 2C1). From some part of these primary roots, we noted
an important system of lateral roots where many (>50) small
round nodules (1 mm diameter) were present (Fig. 2C2). A
statistical analysis of the nodule occupancy using non-
parametric tests showed differences according to each level
of treatment. Nodule occupancy from plants from zone 1 and
zone 2 did not show differences at the level of P ¼ 0.05,
whereas it showed significant differences from plants of
zone 3.
Ultrathin sections and microscopic analyses were carried
out on different nodules. We noted the presence of infected
bacteroid-containing cells in the different observations
(Fig. 2D1). The observed bacteroids were pleomorphic. The
intercellular spaces were not infected. We also noted cyto-
plasms with numerous ribosomes (Fig. 2D2), and some
bacteroids seemed to divide. We did not observe internal or
parallel membranes in the bacteroids. Intracellular inclusions
that were either electron-opaque or electron-transparent,
resembling polyphosphate or polyhydroxybutyrate (PHB)
storage granules, respectively, were present. Only 1e2
bacteroids per symbiosom were observed in transmission
electron micrographs of the small and round nodules isolated
from Medicago plants of zone 3.
3.2. Isolation, selection and identification of
exopolysaccharide-producing bacteria
Bacterial strains were isolated from different nodules har-
vested on roots from 32 different plants of the different zones
(Table 1). A total of 47 isolates (22 from zone 1, 11 from zone
2 and 14 from zone 3) were able to produce exopoly-
saccharides on RC medium supplemented with 1%
saccharose. White and circular colonies (2 mm) with regular
borders were observed after 20 h, but quickly an important
production of mucus covered the agar plates, except for two
strains (Ez-185-4 and Ez-185-5). Because of the weak
production of mucus of these two strains, they were not further
investigated. The observed mucosity varied according to the
isolates, but no diversity was found between the different
sample zones. After 2 days of growth in RC liquid medium,
we noted acidification of the medium culture (from 6.0 to 4.2)
for all strains except one (Ez-555-11). This decrease in pH was
no doubt due to the metabolism of sugar and ammonium
sulfate. All bacteria strains except Ez-555-11 were considered
as medium-growing-strains. Indeed, the generation time varied
according to the strain, but remained at around 4e5 h, which is
long compared to control strain M5N1 (2 h). The different
isolates were Gram-negative bacteria, rod-shaped with
different sizes, and all of them except for Ez-555-11 were
facultative anaerobic, oxidase-negative, and recognized as
belonging to the Enterobacteriaceae family. The Ez-555-11
strain was identified as Gram-negative but oxidase-positive.
All isolated bacteria were able to stain blue with aniline blue.
However, differences in the intensity of staining were
observed. A few bacterial strains (15) showed staining similar
to that of control strain S. meliloti M5N1, suggesting the
presence of curdlan-type polysaccharides, while the others
strains presented heterogenous staining: deep-blue in the
center of the colony and light-blue around the colony. Cal-
cofluor white, a fluorescent dye that binds to glycans, has been
used to determine the kind of produced polysaccharides. We
noted that only one bacterial strain (Ez-555-11) fluoresced on
LB/calcofluor plates like our control, suggesting a similar
composition for the produced EPS. Altered fluorescence of
colonies on plates containing the fluorescent dye calcofluor
was noted for the other strains.
Visualization of the plasmid content of the different
bacterial strains by the Eckhardt method did not highlight any
large plasmid except for one strain (Ez-555-11) in which 2
megaplasmids (1400 and 1700 kb) were recovered, as was the
case for the control strain. Six different bacterial strains (Ez-
185-2, Ez-185-3, Ez-185-6, Ez-555-4, Ez-555-5, Ez-555-6)
showed the presence of only 1 cryptic plasmid (136 kb).
DNA amplification of two regions of the S. meliloti
genome, the nodbox4 promoter and the mucR gene, was per-
formed. Only the Ez-555-11 strain showed a pattern similar to
that of the control, with the presence of two bands at 646 and
431 bp, corresponding to the size of the two genes respec-
tively. For the others strains, we did not observe amplification.
The sequencing of the rpoB gene [1,39], isolated from the
chromosomal DNA of 5 different strains (Ez-185-8, Ez-185-
17, Ez-555-6, Ez-9250-2 and Ez-9250-7), as well as a phylo-
genetic study based on the rpoB gene sequences led to
identification of these different strains (Table 1). The closest
relative of isolates Ez-185-8 and Ez-185-17 (935 nucleotides)
was Enterobacter ludwigii type strain CIP108481, with the
same percent similarity, 99.5. The closest relatives of isolates
Ez-555-6 (929 nucleotides) and Ez-9250-2 (936 nucleotides)
is Raoultella terrigena type strain ATCC33257 with the same
 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108 105

Fig. 2. (A1) Medicago plant from Chernobyl exclusion zone 1, (A2) nodules on the root system from plant harvested in zone 1, (B) atypical nodule from Medicago
plants harvested in Chernobyl exclusion zone 2, (C1) Medicago plant from Chernobyl exclusion zone 3, (C2) lateral root system with small and round nodules from
zone 3, (D1) transmission electron micrograph of plant cells containing bacteroids from A2 (G  6000), (D2) transmission electron micrograph of plant cells with
bacteroids and numerous ribosomes from C (G  15,000). N, nucleus of the cell plant; c, cytoplasm; CW, cell wall; IS, intercellular space; r, ribosomes.

percent similarity 100. The closest relatives of isolates Ez-
9250-7 (983 nucleotides) is Klebsiella oxytoca type strain
ATCC13182 with 97.4% similarity.
3.3. Nodulation tests
We next checked whether the various bacterial isolates
could induce nodule formation on in vitro alfalfa plants. A
series of nodulation tests was performed in tubes as described
in Materials and methods. Nodules were produced in all
inoculated seedlings; however, the number of nodulated plants
varied from 4 to 100%, and the number of nodules per plant
varied from 2 to 10. When in vitro alfalfa plants were inocu-
lated with strains from Chernobyl exclusion zone 3, only
small, round translucent nodules like those in the harvested
plants were observed. No atypical nodule (bifurcate or with
a fan shape) was noted. Four weeks after inoculation, all
inoculated plants were dark green and well-developed,
whereas non-inoculated in vitro alfalfa plants did not show
nodules and were chlorotic and stunted. Likewise, these
control plants were not able to reduce acetylene, whereas the
inoculated alfalfa plantlets were positive for this characteristic.
All bacterial strains were thus efficient, but this efficiency
varied from 6.2 to 98.7 nmol C2H4 mg1 plantlet/h. The
weakest efficiency was recovered for plantlets inoculated with
bacterial strains from zone 3.
3.4. Production and chemical analysis of the EPSs
After 72 h of culture on GMS medium, an EPS yield of
from 0.2 to 3.0 g/L was obtained according to the bacterial
strain (Table 1). Among screened strains, most were able to
106 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108

Table 1
EPS production and composition from bacterial strains isolated from M. sativa or M. lupulina root nodules in 185 kBq/m2 zone 1 (corresponding to 0.25 mGy/h of g
rays in air), 555 kBq/m2 zone 2 (corresponding to 0.90 mGy/h of g rays in air) and 9250 kBq/m2 zone 3 (corresponding to 2.5 mGy/h of g rays in air).
Bacterial strain number Related species Plant EPS yield g/l/3 NS/UA (%)a %
days growth Fuc Gal Glu Man
Ez-185-1 Enterobacter ludwigii M. lupulina 1 90/10 40 20 30
Ez-185-2 Enterobacter ludwigii M. lupulina 1.3 90/10 40 20 30
Ez-185-3 Enterobacter ludwigii M. lupulina 1.4 90/10 40 20 30
Ez-185-6 Enterobacter ludwigii M. lupulina 1.2 90/10 40 20 30
Ez-185-7 Enterobacter ludwigii M. lupulina 1.5 80/20 35 20 25
Ez-185-8 Enterobacter ludwigii M. lupulina 1.8 80/20 35 20 25
Ez-185-9 Enterobacter ludwigii M. lupulina 0.9 80/20 35 20 25
Ez-185-10 Enterobacter ludwigii M. lupulina 1.5 80/20 40 20 20
Ez-185-12 Klebsiella oxytoca M. lupulina 0.4 90/10 55 35
Ez-185-13 Enterobacter ludwigii M. lupulina 0.7 80/20 35 20 25
Ez-185-14 Raoultella terrigena M. lupulina 0.95 90/10 35 25 15 15
Ez-185-15 Raoultella terrigena M. lupulina 1.05 90/10 35 25 15 15
Ez-185-16 Klebsiella oxytoca M. lupulina 1.2 90/10 55 35
Ez-185-17 Enterobacter ludwigii M. lupulina 0.7 90/10 40 20 20
Ez-185-18 Enterobacter ludwigii M. lupulina 0.6 90/10 40 20 20
Ez-185-19 Enterobacter ludwigii M. lupulina 0.8 90/10 40 20 20
Ez-185-20 Klebsiella oxytoca M. sativa 0.84 90/10 55 35
Ez-185-21 Klebsiella oxytoca M. sativa 1.7 90/10 55 35
Ez-185-22 Klebsiella oxytoca M. sativa 2.96 90/10 55 35
Ez-185-23 Klebsiella oxytoca M. sativa 1.98 90/10 55 35
Ez-555-1 Klebsiella oxytoca M. lupulina 1.05 90/10 55 35
Ez-555-2 Klebsiella oxytoca M. lupulina 0.8 90/10 55 35
Ez-555-3 Klebsiella oxytoca M. lupulina 2.25 90/10 55 35
Ez-555-4 Raoultella terrigena M. lupulina 1.4 85/15 25 35 25
Ez-555-5 Raoultella terrigena M. lupulina 1.1 80/20 20 40 20
Ez-555-6 Raoultella terrigena M. lupulina 2.15 85/15 25 35 25
Ez-555-7 Raoultella terrigena M. lupulina 1.8 85/15 25 35 25
Ez-555-8 Raoultella terrigena M. sativa 1.15 85/15 25 35 25
Ez-555-9 Raoultella terrigena M. sativa 0.95 85/15 25 35 25
Ez-555-10 Raoultella terrigena M. sativa 0.85 85/15 25 35 25
Ez-555-11 Sinorhizobium meliloti M. sativa 1.0 100 50 50
Ez-9250-1 Raoultella terrigena M. sativa 1.2 90/10 10 15 25 40
Ez-9250-2 Raoultella terrigena M. sativa 0.95 90/10 10 15 25 40
Ez-9250-3 Raoultella terrigena M. sativa 2.64 90/10 25 35 25
Ez-9250-4 Raoultella terrigena M. sativa 1.1 90/10 10 15 25 40
Ez-9250-5 Raoultella terrigena M. sativa 0.8 90/10 10 15 25 40
Ez-9250-6 Klebsiella oxytoca M. sativa 0.9 90/10 55 35
Ez-9250-7 Klebsiella oxytoca M. sativa 1.1 90/10 55 35
Ez-9250-8 Klebsiella oxytoca M. sativa 1 90/10 55 35
Ez-9250-9 Klebsiella oxytoca M. sativa 1.23 90/10 55 35
Ez-9250-10 Klebsiella oxytoca M. sativa 1.2 90/10 55 35
Ez-9250-11 Klebsiella oxytoca M. sativa 1.3 90/10 55 35
Ez-9250-12 Klebsiella oxytoca M. sativa 1.05 90/10 50 40
Ez-9250-13 Raoultella terrigena M. sativa 0.9 90/10 30 30 15 15
Ez-9250-14 Raoultella terrigena M. sativa 0.85 90/10 30 30 15 15
a
NS: neutral sugar, UA: uronic acid.

produce water-soluble polysaccharides under the conditions
used. All the EPSs were heteropolysaccharides with a high Mw
(1  106 to 3  106 Da). No protein (less than 1%) was
recovered. The percentage of neutral sugars varied from 80 to
100%. As described by Curie and Perry [10], high pH anion
exchange chromatography coupled with pulsed amperometric
detection (HPAED-PAD) was used for detection and quanti-
fication of both neutral and charged monosaccharides from our
different produced EPS. Under the conditions described in
Materials and methods, we were able to separate and quantify
the different constitutive monosaccharides. The results,
detailed in Table 1, were confirmed by GLC-MS analysis. All
strains but one produced uronic acids. That one strain, Ez-555-
11, produced an EPS composed of galactose and glucose with
a molar ratio of 1/1, probably galactoglucan as already
described in mutants of S. meliloti Rm1021 [16]. The charged
monosaccharides corresponded to uronic acids, mainly glu-
curonic acid. For some strains (Ez-185-7, Ez-185-8, Ez-185-9,
Ez-185-10, Ez-185-13, Ez-555-4, Ez-555-5, Ez-555-6), we
noted, on the chromatograms, the presence of another charged
monosaccharide, but which has not yet been determined.
Exopolysaccharides from E. ludwigii were constituted of
fucose, galactose and glucose at a molar ratio 2/1/1. These
bacteria were mainly recovered in the least contaminated zone
 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108
(185 kBq/m2). Bacteria identified as K. oxytoca produced an
EPS composed of galactose and mannose at a ratio of 2/1 and
were present in the three contaminated zones. In the least
contaminated zone and in the most contaminated zone,
R. terrigena produced an EPS composed of fucose, galactose,
glucose and mannose (2/2/1/1 or 1/1/2/4), whereas in the
intermediate contaminated zone, the EPS was composed of
galactose, glucose and mannose (1/2/1).
The NMR spectra showed that the different EPSs produced
were substituted by acetate esters and pyruvate ketals.
4. Discussion
In the present study, we report on the isolation, production
and characterization of different acidic water-soluble EPSs
produced by various nitrogen-fixing bacterial strains isolated
from Medicago plants harvested in the Chernobyl exclusion
zone. As some strains were trapped by M. sativa and others by
M. lupulina, they could belong to S. meliloti and Sino-
rhizobium medicae, respectively. However, all the bacterial
strains except one have been identified by biochemical tests
and molecular analyses (rpoB gene sequences) as belonging to
the family Enterobacteriaceae, and in particular to the genera
Enterobacter, Klebsiella and Raoultella. The isolation of these
unexpected taxa from Medicago nodules can be disregarded,
as they resulted from an inefficient sterilization procedure.
Bacteria that form nitrogen-fixing symbiotic associations with
legumes are usually distributed in the Alphaproteobacteria
[36,38] and Betaproteobacteria [27]. Recently, Gammapro-
teobacteria within the genus Enterobacter have been described
as being plant-growth-promoting [33], but also as being able to
nodulate legumes [5]. The acquisition of endophytic inter-
activity with leguminous plants must be either due to lateral
gene transfer from Rhizobium strains or to coexistence with
non-culturable rhizobia [28]. Rhizobia symbionts can be
overwhelmed and masked by endophytes. In the soil of the
Chernobyl exclusion zone, the level of radioactivity as well as
the level of heavy metals is quite high, and bacteria must be
well adapted [15,22]. Recently, E. ludwigii, a bacterial strain
from a clinical environment, was characterized as being
associated with biofilms of corroded oil pipelines with
immediate contact with toxic substances and heavy metals
[30]. In the same way, K. oxytoca, an opportunistic pathogen,
was isolated from iron-rich sediments and was capable of
growing on high concentrations of heavy metals [24]. Under
our culture conditions, isolated nitrogen-fixing enterobacteria
were able to produce highly viscous exopolysaccharides of
different composition with uronic acid residues and with
various proportions of pyruvate and acetate. Several exopo-
lysaccharide-producing enterobacteria have already been
identified [19], but there is no specific information about EPS
production and composition in E. ludwigii, R. terrigena or
K. oxytoca. Recently, the structure of an EPS secreted by
K. oxytoca BAS-10 containing two glucuronic acids at the
non-reducing terminus has been reported [24]. In our study,
only one EPS-producing nitrogen-fixing rhizobacteria screened
(Ez-555-11) produced galactoglucan. This EPS corresponds to
107
the EPS II produced under specific conditions by S. meliloti
wild type [43]. Studies on UV radiation on Rhizobium trifolii
11B [14] showed induction of spontaneous mutations in the
bacteria in relation to changes in the production and compo-
sition of EPS. Chemical mutagenesis (NTG) performed on
S. meliloti M5N1-producing succinoglycan led to a mutant
producing a partially acetylated (1e4)-b-D-glucuronan [20],
and able to induce nodule formation on alfalfa roots.
The objective of the study, i.e. screening of nitrogen-fixing
bacteria from Medicago plants exposed to chronic radiation in
the Chernobyl exclusion zones so as to select those producing
copious amounts of EPS with a specific composition, was
achieved. Some of the isolated EPSs are attractive for indus-
trial applications due to their composition (fucosylated EPSs,
presence of mannose, etc.) and their viscosity. Further works
are now necessary to reveal the exact and complete structure
of the different EPSs and detailed biological and environ-
mental activities such as the uptake of heavy metals in soils.
Acknowledgments
We thank Gerard Vasseur for his technical assistance in
ultrathin sections, and Bernard Baudoin for his assistance in
transmission electron microscopy.
References
[1] Adékambi, T., Drancourt, M., Raoult, D. (2009) The rpoB gene as a tool
for clinical microbiologists. Trends Microbiol. 17, 37e45.
[2] Alami, Y., Achouak, W., Marol, C., Heulin, T. (2000) Rhizosphere soil
aggregation and plant growth promotion of sunflower by an exopoly-
saccharide-producing Rhizobium sp. strain isolated from sunflower roots.
Appl. Environ. Microbiol. 66, 3393e3398.
[3] Baev, N., Endre, G., Petrovics, G., Banfalvi, Z., Kondorosi, A. (1991) Six
nodulation genes of nod box locus 4 in Rhizobium meliloti are involved in
nodulation signal production: nodM codes for d-glucosamine synthetase.
Mol. Gen. Genet. 228, 113e124.
[4] Bell, J.N., Shaw, G. (2005) Ecological lessons from the Chernobyl
accident. Environ. Int. 31, 771e777.
[5] Benhizia, Y., Benhizia, H., Benguedouar, A., Muresu, R., Giacomini, A.,
squartini, A. (2004) Gamma Proteobacteria can nodulate legumes of the
genus Hedysarum. Syst. Appl. Microbiol. 27, 462e468.
[6] Beringer, J.E. (1974) R-factor transfer in Rhizobium leguminosarum. Soil
Biol. Biochem. 84, 188e198.
[7] Bradford, M.M. (1976) A rapid and sensitive method for quantitation of
microgram quantities of protein utilizing the principle of protein-dye-
binding. Anal. Biochem. 72, 248e254.
[8] Bromfield, E.S.P., Wheatcroft, R., Barran, L.R. (1994) Medium for direct
isolation of Rhizobium meliloti from soil. Soil Biol. Biochem. 26, 423e428.
[9] Courtois, J., Courtois, B., Guillaume, J.B. (1988) High frequency
transformation of Rhizobium meliloti. J. Bacteriol. 170, 5925e5927.
[10] Curie, H.A., Perry, C.C. (2006) Resolution of complex monosaccharide
mixtures from plant cell wall isolates by high pH anion exchange
chromatography. J. Chromatogr. A 1128, 90e96.
[11] Denisova, T.V., Kaseev, K.S.H., Kolesnikov, S.I., Val’kov, V.F. (2005)
The influence of gamma radiation on the biological properties of soil.
Euras. Soil Sci. 38, 776e779.
[12] Eckhardt, T. (1978) A rapid method for the identification of plasmid
deoxyribonucleic acid in bacteria. Plasmid 1, 584e588.
[13] Geras’kin, S.A., Fesenko, S.V., Alexakhin, R.M. (2008) Effects of non-
human species irradiation after the Chernobyl NPP accident. Environ.
Int. 34, 880e887.
108 N. Pawlicki-Jullian et al. / Research in Microbiology 161 (2010) 101e108
[14] Ghai, J., Ghai, S.K., Kalra, M.S. (1984) Ultraviolet-irradiation induced
spontaneous mutation of Rhizobium trifolii 11B in relation to water-
soluble and water-insoluble polysaccharide production ability. Enzyme
Microb. Technol. 7, 83e86.
[15] Giller, K.E., McGrath, S.P., Hirsch, P.R. (1989) Absence of nitrogen
fixation in clover grown on soil subject to long-term contamination with
heavy metals is due to survival of only ineffective Rhizobium. Soil Biol.
Biochem. 21, 841e848.
[16] Glazerbrook, J., Walker, G.C. (1989) A novel exopolysaccharide can
function in place of the Calcofluor-binding exopolysaccharide in nodu-
lation of alfalfa by Rhizobium meliloti. Cell 56, 661e672.
[17] Gray, J.X., Zhan, H., Levery, S.B., Battisti, L., Rolfe, B.G., Leigh, J.A.
(1991) Heterologous exopolysaccharide production in Rhizobium sp.
strain NGR234 and consequences for nodule development. J. Bacteriol.
173, 3066e3077.
[18] Hardy, R.W., Burns, R.C., Holsten, R.D. (1973) Application of the
acetylene reduction assay for the measurement of nitrogen fixation. Soil
Biol. Biochem. 5, 47e81.
[19] Hebbar, K.P., Gueniot, B., Heyraud, A., Colin-Morel, P., Heulin, T.,
Balandreau, J., et al. (1992) Characterization of exopolysaccharides
produced by rhizobacteria. Appl. Microbiol. Biotechnol. 38, 248e253.
[20] Heyraud, A., Courtois, J., Dantas, L., Colin-Morel, P., Courtois, B.
(1993) Structural characterization and rheological properties of an
extracellular glucuronan produced by a Rhizobium meliloti M5N1 mutant
strain. Carbohydr. Res. 240, 71e78.
[21] Laberche, J.Cl. (2008) Statistiques Et Expérimentation En Biologie.
Paris: Ellipses Marketing.
[22] Lakzian, A., Murphy, P., Giller, K.E. (2007) Transfer and loss of naturally-
occurring plasmids among isolates of Rhizobium leguminosarum bv. viciae
in heavy metal contaminated soils. Soil Biol. Biochem. 39, 1066e1077.
[23] Lamelas, C., Benedetti, M., Wilkinson, K.J., Slaveykova, V.I. (2006)
Characterization of Hþ and Cd2þ binding properties of the bacterial
exopolysaccharides. Chemosphere 65, 1362e1370.
[24] Leone, S., De Castro, C., Parrilli, M., Baldi, F., Lanzetta, R. (2007)
Structure of the iron-binding exopolysaccharide by the Gram-Negative
Bacterium Klebsiella oxytoca BAS-10. Eur. J. Org. Chem. 31, 5183e5189.
[25] Long, S., Reed, J.W., Himawan, J., Walker, G.C. (1988) Genetic analysis
of a cluster of genes required for synthesis of the calcofluor-binding
exopolysaccharide of Rhizobium meliloti. J. Bacteriol. 170, 4239e4248.
[26] Monsigny, M., Petit, C. (1988) Colorimetric determination of neutral
sugars by resorcinol acid micromethod. Anal. Biochem. 175, 525e530.
[27] Moulin, L., Munive, A., Dreyfus, B., Boivin-Masson, C. (2001) Nodu-
lation of legumes by members of the b-subclass of Proteabacteria. Nature
411, 948e950.
[28] Muresu, R., Polone, E., Sulas, L., Baldan, B., Tondello, A., Delogu, G.,
et al. (2008) Coexistence of predominantly nonculturable rhizobia with
diverse, endophytic bacterial taxa within nodules of wild legumes. FEMS
Microbiol. Ecol. 63, 383e400.
[29] Nakanishi, I., Kimura, K., Suzuki, T., Ishikawa, M., Banno, M.,
Sakane, T., et al. (1976) Exopolymers from curdlan production:
incorporation of glucose-related sugars by Agrobacterium sp. strain
ATCC31749. J. Gen. Appl. Microbiol. 22, 1e11.
[30] Neria-Gonzalez, I., Wang, E.T., Ramirez, F., Romero, J.M., Hernandez-
Rodriguez, C. (2006) Characterization of bacterial community associated
to biofilms of corroded oil pipelines from the southeast of Mexico.
Anaerobe 12, 122e133.
[31] Romanovskaya, V.A., Solokolov, G., Rokitko, P., Chernaya, N.A. (1998)
Effect of radioactive contamination on soil bacteria in the 10-km zone
around the Chernobyl nuclear power plant. Microbiology 67, 226e231.
[32] Romanovskaya, V.A., Rokitko, P.V., Mikheev, A.N., Gushcha, N.I.,
Malashenko, Y.R., Chernaya, N.A. (2002) The effect of g-radiation and
dessication on the viability of the soil bacteria isolated from the alienated
zone around the Chernobyl nuclear power plant. Microbiology 71, 608e613.
[33] Ruppel, S. (2000) Role of rhizosphere and endophytic bacteria in plant
nutrition. Arch. Acker- Pfl. Boden. 45, 329e341.
[34] Sanchez-Contreras, M., Lloret, J., Martin, M., Villacieros, M., Bonilla, I.,
Rivilla, R. (2000) PCR use of highly conserved DNA regions for identifi-
cation of Sinorhizobium meliloti. Appl. Environ. Microbiol. 66, 3621e3623.
[35] Savchenko, V.K. (1996) The Ecology of the Chernobyl Catastrophe.
Scientific Outlines of an International Programme of the Collaborative
Research. Carnforth: Unesco, Paris and the Parthenon Publishing
Company.
[36] Sawada, H., Kuykendall, D., Young, J.M. (2003) Changing concepts in
the systematics of bacterial nitrogen-fixing legume symbionts. J. Gen.
Appl. Microbiol. 49, 155e179.
[37] Shevchuk, V.E., Gurachevsky, V.L. (2006) 20 years after the Chernobyl
Catastrophe: the Consequences in the Republic of Belarus and their
Overcoming. National report, Minsk.
[38] Sy, A., Giraud, E., Jourand, P., Garcia, N., Willens, A., de Lajudie, P.,
et al. (2001) Methylotrophic Methylobacterium bacteria nodulate and fix
nitrogen in symbiosis with legumes. J. Bacteriol. 183, 214e220.
[39] Tayeb, L.A., Lefevre, M., Passet, V., Diancourt, L., Brisse, S.,
Grimont, P.A. (2008) Comparative phylogenies of Burkholderia, Ral-
stonia, Comamonas, Brevumdimonas and related organisms derived from
rpoB, gyrB and rrs gene sequences. Res. Microbiol. 159, 169e177.
[40] Thornton, H.G. (1952) Introduction. The symbiosis between Rhizobium
and leguminous plants and the influence on this of the bacterial strain.
Proc. R. Soc. Lond. Series B: Biol. Sci. 139, 170e176.
[41] van den Hoogen, B.M., van Weeren, P.R., Lopes-Cardozo, M., van
Golde, L.M.G., Barneveld, A., van de Lest, C.H.A. (1998) A microtiter
plate assay for the determination of uronic acids. Anal. Biochem. 257,
107e111.
[42] Zavilgelsy, G.B., Abilev, S.K., Sukhodolets, V.V., Ahmad, S.I. (1998)
Isolation and analysis of UV and radio-resistant bacteria from Chernobyl.
J. Phytochem. Phytobiol. Biol. 43, 152e157.
[43] Zevenhuizen, L.P.T. (1997) Succinoglycan and galactoglucan. Carbo-
hydr. Polym. 33, 139e144.
[44] Zevenhuizen, L.P.T., van Neerven, A.R.W. (1983) (1,2)-b-[D]-glucan and
acidic oligosaccharides produced by Rhizobium meliloti. Carbohydr. Res.
118, 127e134.