J Soils Sediments (2017) 17:1224–1236

DOI 10.1007/s11368-015-1346-5

PHYTOREMEDIATION OF POLLUTED SOILS: RECENT PROGRESS AND DEVELOPMENTS

Phytoremediation of a multi contaminated soil: mercury

and arsenic phytoextraction assisted by mobilizing agent
and plant growth promoting bacteria
Elisabetta Franchi 1 & Eleonora Rolli 4 & Ramona Marasco 4 & Gloria Agazzi 2 &
Sara Borin 4 & Paola Cosmina 1 & Francesca Pedron 3 & Irene Rosellini 3 & Meri Barbafieri 3 &
Gianniantonio Petruzzelli 3

Received: 31 March 2015 / Accepted: 24 December 2015 / Published online: 15 January 2016
# Springer-Verlag Berlin Heidelberg 2016

Abstract agents, together with a bacterial consortium composed by

Purpose The possibility of using chemical and microbial ad- the most promising PGP isolates.
ditives to enhance the phytoextraction of mercury (Hg) and Results and discussion Thirteen indigenous metal-tolerant
arsenic (As) from a multi-contaminated soil could be very bacterial strains were isolated, and their in vitro characteriza-
effective, leading to a significant saving in terms of time and tion highlighted their great potential in assisting the
costs of the reclamation. The aim of this study was to evaluate phytoremediation process; most of them tolerated both trace
the efficacy of the addition of (i) thiosulfate and (ii) metal- elements and showed, at the same time, multiple PGP traits.
tolerant bacteria isolated from the polluted soil having plant The results were confirmed in vivo on model plants and lead
growth promotion (PGP) potential to perform As and Hg to the selection of the most promising PGP strains to be ap-
phytoextraction by Brassica juncea and Lupinus albus. plied in microcosm-scale phytoextraction experiments.
Materials and methods A collection of 13 bacterial isolates Thiosulfate addition significantly increased the mobilization
able to tolerate As and Hg was obtained from the contaminat- of both elements, promoting bioavailability and
ed soil, identified by partial 16S rRNA gene sequencing and phytoextraction. When a selected bacterial consortium was
tested in vitro for PGP activities. The most promising strains supplemented in addition to thiosulfate, the efficacy of the
were further tested in vivo for the evaluation of plant growth phytoaccumulation was increased up to 85 % for As and up
ability and rhizocompetence on model plants. Pot experiments to 45 % for Hg.
were conducted in microcosms, with polluted soil vegetated Conclusions The use of the common fertilizer thiosulfate ap-
with B. juncea and L. albus. Ammonium thiosulfate and po- peared to have great potential in phytoextraction practices
tassium dihydrogen phosphate were used as mobilizing since it was able to facilitate the uptake by plants of both Hg
and As. Moreover, the application of a consortium of indige-
nous PGP bacteria (PGPB) produced a further positive effect
Responsible editor: Jaco Vangronsveld on the plant biomass, supporting and enhancing the
phytoextraction strategy, thus demonstrating their potential
* Elisabetta Franchi in a microbe-assisted phytoremediation intervention.
[email protected]

Keywords Arsenic mercury . Phytoextraction . Plant

1
growth-promoting bacteria . Thiosulfate
Eni S.p.A, Renewable Energy and Environmental Laboratories, S.
Donato Milanese, Milan, Italy
2
Department of Biotechnologies and Biosciences, University of
Milano-Bicocca, Milan, Italy 1 Introduction
3
Institute of Ecosystem Study, National Council of Research,
Pisa, Italy Arsenic and mercury are non-essential elements for plants;
4
Department of Food, Environmental, and Nutritional Sciences, they are among the most toxic inorganic pollutants, and their
Università degli Studi di Milano, Milan, Italy contamination is a very serious global problem (Garelick et al.
J Soils Sediments (2017) 17:1224–1236 1225

2008; AMAP/UNEP 2013). Several plants were demonstrated phytoremediation can significantly facilitate the growth of
to accumulate toxic compounds in their aboveground bio- plants in the presence of high (and otherwise inhibitory) levels
mass, and the phytoextraction of metals and metalloids have of metals, even if the contribution of bacteria to increase metal
received significant attention as a non-impact environmentally bioavailability is limited (Glick 2010). A promising strategy
safe remediation strategy for polluted soils (Barbafieri et al. alternative to the mere chemical amendments could therefore
2013). In the case of contamination derived from more than be the application of microbe-mediated processes, in which
one metal, a thorough investigation of the soil properties is the microbial metabolites/processes in the rhizosphere affect
essential, together with an accurate selection of the plant spe- plant metal uptake by altering metal mobility and bioavailabil-
cies and the plant-associated growth-promoting bacteria ity (Aafi et al. 2012; Gutiérrez-Ginés et al. 2014).
(PGPB). In soil, metals could be associated with several This work relates to the possibility of using a single chem-
phases: (i) in soil solution as free metal ions and soluble metal ical agent for the simultaneous removal of mercury and arse-
complexes; (ii) adsorbed to inorganic soil constituents at ion nic from a multi-contaminated industrial soil and evaluates the
exchange sites; (iii) bound to soil organic matter; (iv) precip- contribution of a bioaugmentation step with selected autoch-
itated such as oxides, hydroxides, and carbonates; and (v) thonous metal resistance soil bacteria showing multiple PGP
embedded in structure of the silicate minerals. Contaminants features.
must be bioavailable to ensure an efficient phytoextraction,
and their bioavailability depends on metal solubility in pore
water. Only metals in soil solution as free metal ions, soluble 2 Materials and methods
metal complexes, or adsorbed to inorganic soil constituents at
ion exchange sites are potentially available for plant uptake 2.1 Site description and sampling
(Petruzzelli et al. 2013). Uptake of metals into root cells is a
step of major significance, but for phytoextraction to occur, The contaminated soil (SI29) was sampled in northern Italy
metals must also be transported from the root to the shoot, and (Piemonte region) in a disused industrial area. In this area,
this translocation is primarily controlled by root pressure and plants for acetaldehyde synthesis (obtained by acetylene oxi-
leaf transpiration. A major factor limiting metal uptake into dation with mercury catalysts) and for carbon black abatement
roots is slow transport from soil particles to root surfaces were active since early 70s. After, until recent years, the area
(Barber 1984). This transport takes place in soil solution, but was used as storage site for different waste materials from
metal solubility could be limited due to adsorption to soil neighboring industrial plants. The contamination was due to
particles. Thus, increasing metal solubility in the soil is an As (41.1 mg kg−1) and Hg (67 mg kg−1) exceeding legal limits
essential prerequisite to enhance the phytoextraction. The and a fair amount of long-chain (>C12) aliphatic hydrocar-
use of specific chemicals (assisted phytoextraction) has been bons. Soil sampling (0–1 m) was carried out by a small digger,
shown to dramatically stimulate the potential of metal accu- and samples were air dried and ground to pass through a 2-mm
mulation in plants (Meers et al. 2008). sieve before soil analysis. Samples for microbiological analy-
The other fundam ental aspect for an efficient sis were stored-refrigerated until use. Soil pH was determined
phytoextraction is the effect of soil microorganisms on metal using a glass electrode at a soil/water ratio of 1:2.5 (Thomas
uptake. Root exudates form complexes with metal ions keep- 1996); cation exchange capacity (CEC) was determined using
ing them in solution and making them available for uptake barium chloride (pH = 8.1) (Sumner and Miller 1996), texture
into roots (Romheld and Marschner 1986) and affect the prop- (sand, silt, and clay) by the pipette method (Gee and Bauder
erties of the rhizosphere soil stimulating selection and growth 1986), and organic matter (Nelson and Sommers 1996).
of rhizosphere microbial communities. Plants growing in
metal-contaminated soils harbor a diverse group of microor- 2.2 Selection of metal-tolerant bacteria
ganisms (Rajkumar et al. 2012) able to tolerate high concen-
tration of metals and providing a number of benefits to both One gram of sieved (2 × 2-mm mesh) and homogenized sam-
soil and plants. Among the microorganisms involved in heavy ple was incubated in triplicate in 50 ml of a mineral medium
metal phytoremediation, the rhizosphere bacteria deserve spe- (1.5 g l−1 KH2PO4, 0.5 g l−1 NaHPO4, 1 g l−1 NH4Cl, 0.1 g l−1
cial attention because they can directly improve the NaCl, 0.2 g l−1 MgSO4 · 7H2O, 0.0264 g l−1 CaCl2 · 2H2O,
phytoremediation process by increasing the metal bioavail- 0.01 g l−1 FeCl3,5 mg l−1 MnCl2 · H2O, 3 mg l−1 ZnCl2,
ability through altering soil pH, release of chelators (e.g., or- 0.9 mg l−1 CuCl2 · 2H2O, 1 mg l−1 CoCl2 · 6H2O, 1 mg l−1
ganic acids and siderophores), and oxidation/reduction reac- NaMoO4 · 2H2O, 0.3 mg l−1 NiCl2 · 6H2O, 3 mg l−1 H3BO3,
tions (Gadd 2000). The addition of chemical-chelating agents, and 0.2 mg l−1 Na2O3Se · 5H2O) at pH 6.8 with the addition of
a strategy that often works very well on a small scale in the a mixture of aliphatic hyrocarbons (C12, C13, C15, C16, and
laboratory, could be much less effective in the field. The use of C17) as unique carbon source. Enrichment cultures were in-
plant growth promotion (PGP) soil bacteria as helpers in metal cubated in 250-ml conical flasks, at 28 °C in an orbital shaker
1226 J Soils Sediments (2017) 17:1224–1236

(150 rpm) for 1 week. Cultures were then diluted at 5 % and (Basic Local Alignment Search Tool (BLAST) analysis) using
reincubated for an additional week, repeating this step three the National Center for Biotechnology Information (NCBI)
more times. After this first selection, in order to isolate the server at www.ncbi.nlm.nih.gov/blast/Blast.cgi. Phylogenetic
larger number of strains, 100 μl of serial tenfold dilutions of analysis showed high identity (99–100 %) to strains belonging
bacterial cultures was propagated on several solid media, to the following three phyla: Actinobacteria, beta subgroup of
Luria Bertani (LB) (10 g l−1 bacto-tryptone, 5 g l−1 yeast Proteobacteria, and Firmicutes. The 13 partial 16S rRNA
extract, and 1 g l−1 NaCl), NB (3 g l−1 beef extract and gene sequences (1200–1400 bp) from the isolates have been
5 g l−1 peptone), TSB (17 g l−1 casein peptone, 3 g l−1 soy deposited in the GeneBank database, and the accession
peptone, 2.5 g l−1 glucose, 2.5 g l−1 KH2PO4, and 5 g l−1 numbers assigned are from KP780207 to KP780219.
NaCl), TYEG (1 g l−1 bacto-tryptone, 0.3 g l−1 yeast extract,
and 0.5 g l−1 glucose), and R2A (0.5 g l−1 casein acid hydro- 2.4 In vitro estimation of PGP activities
lysate, 0.5 g l−1 yeast extract, 0.5 g l−1 proteose peptone,
0.5 g l−1 soluble starch, 0.5 g l−1 glucose, 0.3 g l−1 dipotassium Indolacetic acid (IAA) production was estimated following
phosphate, 0.024 g l−1 magnesium sulfate, and 0.3 g l−1 sodi- the protocol described by Brick et al. (1991), and quantitative
um pyruvate). To all media, 15 g l−1 of agar were added and analyses were carried out with the method of Loper and Scroth
supplemented with either sodium arsenate (100 mM (1986). The concentration of IAA produced by the cultures
Na3AsO4), sodium arsenite (10 mM NaAsO2), or mercury was determined by comparison with a calibration curve for the
chloride (40 μM HgCl2). After 10 days of incubation at IAA, in the range of 10–100 mg ml−1 IAA in 100 % EtOH.
28 °C, a total count was performed and 20 colonies per medi- The mineral P-solubilizing ability of the strains was deter-
um were randomly selected and maintained as pure cultures. mined with the aid of National Botanic Research Institute’s
The metal tolerance of the isolates was confirmed by growing Phosphate (NBRIP) growth medium, 10 g l−1 glucose, 5 g l−1
the strains in liquid cultures supplemented with As (V, III) and Ca3 (PO4), 2.5 g l−1 MgCl2 · 6H2O, 0.25 g l−1 MgSO4 · 7H2O,
Hg (II) at the same concentration used in the selection agar 0.2 g l−1 KCl, and 0.1 g l−1 (NH4)2SO4, supplemented with
plates. A collection of 13 isolates tolerating Hg and/or As was 0.025 g l−1 bromophenol blue (BPB) according to the protocol
obtained. developed by Nautiyal (1999). Siderophore release was deter-
mined as described by Milagres et al. (1999), and Chrome
2.3 Characterization based on 16S ribosomal ribonucleic Azurol S (CAS)-blue agar for the CAS-overlayed test was
acid gene sequencing and phylogenetic analyses prepared according to Schwyn and Neilands (1987).
Exopolysaccharide (EPS) production was estimated as de-
DNA from the isolates was extracted by Wizard® Genomic scribed by Santaella et al. (2008) using a modified Weaver
DNA Purification Kit (Promega) and used as template for 16S mineral media enriched with sucrose (Weaver et al. 1975),
ribosomal ribonucleic acid (rRNA) gene amplification with 0.1 g l−1 MgSO4 · 7H2O, 0.1 g l−1 CaCl2 · 2H2O, 0.022 g l−1
universal eubacterial primers (F27a FeSO 4 · 7H 2 O, 0.02 g l −1 EDTA, 0.43 mg l −1 ZnSO 4 ,
A G A G T T T G AT C C T G G C T C A G a n d R 1 4 9 2 a 1.30 mg l−1 MnSO4 · H2O, 0,75 mg l−1 NaMoO4 · 2H2O,
GGTTACCTTGTTACGACTT). Polymerase chain reaction 2.80 mg l−1 H3BO3, 26 μg l−1 CuSO4 · 5H2O, 70 μg l−1
(PCR) was performed in a final volume of 20 μl containing CoSO 4 · 7H 2 O, 7.9 g l −1 K 2 HPO 4 , 7.5 g l −1 KH 2 PO 4 ,
10 ng of DNA, 0.5 u of GoTaq® Polymerase (Promega), 0.1 g l−1 yeast extract, and 20 g l−1 sucrose. Bacterial isolates
0.2 μmol −1 of each primer, 0.2 mmol i −1 of dNTPs, were tested for the production of ammonia in peptone water,
1.5 mmol i−1 MgCl2, and 1× PCR buffer. DNA amplification 5 g l−1 peptone and 5 % NaCl, pH 7.2 according to protocol of
conditions were initial denaturation at 94 °C for 4 min, 35 cy- Islam et al. (2009). Freshly grown cultures were inoculated in
cles of 94 °C for 45 s, 55 °C for 1 m, 72 °C for 2 min, and a 5 ml of peptone water in each tube and incubated for 72 h at
final extension step at 72 °C for 10 min. PCR products were 30 °C. Nessler’s reagent (0.5 ml) was added in each tube, and
analyzed by 0.8 % agarose gel electrophoresis, stained with development of yellow-brownish color indicated NH3 produc-
Gel Green (Diatech Labline), and visualized with a blue light tion (Cappuccino and Sherman 1992). Proteolytic activity (ca-
transilluminator (Dark Reader®, Claire Chemical Research). sein degradation) was determined from clearing zones in skim
The PCR-amplified DNA was sequenced using a BigDye® milk agar (7 % skimmed milk, 5 g l−1 casein, 1 g l−1 glucose,
Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems 2.5 g l−1 yeast extract, and 15 g l−1 agar) after 4 days of
Inc., USA) and the same primers used in PCR, in an automat- incubation at 30 °C as described by Nielsen and Sørensen
ed DNA sequencer (ABI model 3500 Genetic Analyzer) with (1997).
a 50-cm capillary. The nucleotide sequences recovered were The potential capacity to fix atmospheric nitrogen was
edited and assembled using SeqMan application from evaluated by growing the isolates with the specific medium
Lasergene version 11.2.1 (DNASTAR®). Nucleotide se- nitrogen-fixing bacteria (NFb), 5 g l−1 D-L malic acid,
quences were then subjected to homology comparison 0.5 g l−1 K2HPO4, 0.2 g l−1 MgSO4 · 7H2O, 0.1 g l−1 NaCl,
J Soils Sediments (2017) 17:1224–1236 1227

0.02 g l−1 CaCl2 · 2H2O, 3 g l−1 glucose, 2 ml l−1 trace element in Petri dishes were the same, except that the pots were daily
solution (0.10 mg l−1 U3BO3, 1.12 mg l−1 MnSO4 · 7H2O, watered to ensure adequate hydration. After 20 days from the
1.25 mg l−1 ZnSO4 · 7H2O, 7.8 mg l−1 CuSO4 · 5H2O, and bacterial inoculum, the length of the stem; the numbers of
0.10 mg l−1 MoO3), and 1 ml l−1 vitamin solution added after leaves, gems, and flowers; and the fresh weights of the bio-
sterilization, pH 6.8 (Liba et al. 2006). To allow the selective mass of stem and roots were evaluated. The results were ana-
growth of microorganisms capable of fixing atmospheric ni- lyzed by using the statistical Student’s t test.
trogen, any usable nitrogen source is absent in this medium.
The tubes containing the medium were inoculated with the
2.6 PGPB consortium preparation and microcosm
strains and incubated at 30 °C for 7 days. To confirm the
experiments
results and minimize false positives, strains were reinoculated
in the same conditions repeating this procedure five times. The
The isolates 185.1, 188.14, 195.23, 207.37, and 209.1 (Fig. 2)
isolated strains were also tested for their capacity to form
were grown in LB medium for 72 h, and the five cell pellets
biofilm in vitro, inoculating them in glass tubes with 7 ml of
were pooled, resuspending them in suitable protective medi-
LB medium. The tubes were incubated at 30 °C for 7 days
um (1 % sodium glutamate, 7 % sucrose, and 5 % dextrane),
without agitation. The formation of a visible layer (pellicle) at
frozen, and exposed to a vacuum in small aliquots for 48 h. In
the interface between medium and air indicated a potential
order to inoculate about 108 colony-forming units (cfu) per
capability to produce biofilms.
gram of soil, aliquots of dried bacterial consortium containing
The strains capacity to promote plant growth through the
at least 1011 cfu, suitable for each microcosm (about 400 g),
production of volatile organic compounds (VOCs) was
were prepared. After drying, bacteria were stored under vac-
established with a plate assay according to the protocol of
uum in plastic vials until use. Microcosm pots, filled with
Ryu et al. (2003).
400 g of contaminated soil, were sown with 0.5 g per pot of
Brassica juncea seeds or three seeds per pot of Lupinus albus
2.5 In vivo growth promotion assays
in five replicates for each species. Plant growing was carried
out in a growth chamber in controlled conditions, 14 h of light
Seeds of Solanum lycopersicum (tomato) were sterilized (30 s
with a temperature of 24 °C and 10 h in the dark at 19 °C.
in 70 % ethanol and 5 min in 1 % sodium hypochlorite), rinsed
Relative humidity was maintained at 70 %.
five times in sterile water, placed for germination in Petri
PGPB were suspended under stirring in distilled water and
dishes containing demineralized water and 15 g l−1 agar, and
then added to soil immediately after seed germination.
incubated in a growth chamber at 25 °C, 66 % humidity
Of thiosulfate [(NH4)2S2O3], 0.27 M was used as a single
protected from light with an aluminum foil. After 5 days, the
mobilizing agent for both contaminants and was added to the
sprouts were transferred to Petri dishes containing filter paper
soil after 15 days by splitting the total dose in 5-day applica-
(nine seeds per plate) and about 108 cells (estimated in Thoma
tions to minimize the possible toxic effects on the plant spe-
cell counter chamber) of each bacterial culture were resus-
cies (Pedron et al. 2013). Control microcosms without thio-
pended in 2 ml of sterile ddH2O and used as inoculum. The
sulfate addition and without bacterial inoculum were run si-
nine shoots for negative controls were irrigated with 2 ml of
multaneously for the two species. Microcosm experiments
H2O. After the treatment with the bacterial inoculum, the
were carried out for 30 days. After harvesting, aboveground
plantlets were allowed to grow in a greenhouse for 7 days at
parts and roots were collected, separated, and washed with
25 °C, under 16-h day/night photoperiod. At the end, the
deionized water. Soil particles were washed out from the roots
lengths of stems and roots were measured and the results were
in an ultrasound bath (Branson Sonifier 250 Ultrasonic
analyzed by using the statistical Student’s t test. Seeds of
Processor; Branson, Danbury, CT). Plant samples were dried
Arabidopsis thaliana were sterilized (30 s in 70 % ethanol
in a ventilated oven at 40 °C, and the dry biomass of shoots
and 5 min in 1 % sodium hypochlorite), rinsed five times in
and roots were gravimetrically determined.
sterile water, and stratified 48 h at 4 °C. Germination was
carried out in Petri dishes with half-strength MS medium
(2.15 g l−1 Murashige and Skoog salt, Fluka; 15 g l−1 sucrose; 2.7 Bioavailable metal extraction
and 8 g l−1 agar, pH 5.7) and incubation in a growth chamber
at 22 °C, 50 % humidity, under 16-h day/night photoperiod. Potentially bioavailable fractions of contaminants were deter-
Seven days after germination, the seedlings were transferred mined (1 g soil:25 ml solution) by extraction with 0.27 M
to pots containing 15 g of soil and inoculated with about ammonium thiosulfate [(NH4)2S2O3, from Sigma-Aldrich]
1.5 × 109 cells of the bacterial strains (about 108 cells/g of soil, for Hg (Moreno et al. 2004; Pedron et al. 2013) and with
estimated with Thoma cell counter chamber). The seedlings of 0.05 M potassium dihydrogen phosphate (KH2PO4, from
the four negative controls were irrigated with 5 ml of sterile Merck) for As (Tassi et al. 2004). Moreover, 0.27 M
H2O each. The incubation conditions used for the germination (NH4)2S2O3 was also tested as extractant for As.
1228 J Soils Sediments (2017) 17:1224–1236

2.8 Mercury and arsenic quantification 3.2 Arsenic and mercury tolerance of the isolates

Mercury concentrations in soil, plant samples, and soil ex- The soil under study (SI29) showed a contamination due to As
tracts were determined by atomic absorption spectropho- and Hg, and a selection of tolerant bacteria was thus per-
tometry with an Automatic Mercury Analyzer (AMA 254, formed. By growth on media supplemented with As (V and
FKV, Bergamo, Italy), according to the SW-846 method III) and Hg (II), 13 strains showing tolerance at least to one of
7473 (US EPA 1998). The As concentration was deter- the metals (or metalloid) were selected. Specifically, 5 strains
mined using ICP-OES with a method for the generation out of 13 showed a single tolerance, 7 strains a double toler-
of hydrides (Sparks 1998). ance, and 1 strain a triple tolerance (Fig. 2). Hg tolerance was
shown by 5 isolates, As (V) tolerance by 12 isolates, and As
2.9 Statistical analysis (III) tolerance by 5 isolates.

Treatment effects on growth promotion and Hg and As 3.3 In vitro evaluation of plant growth-promoting
phytoextraction were analyzed using one-way analysis of var- activities of the isolates
iance (ANOVA). Differences between means were compared,
and a post hoc analysis of variance was performed using the After the phylogenetic characterization, the 13 metal-tolerant
Tukey’s honest significant difference test (P < 0.05). All sta- bacteria were subjected to a series of in vitro assays to verify
tistical analyses were performed using Statistica version 6.0 their plant growth promotion potential. The production of the
(Statsoft Inc., USA). auxin 3-IAA, siderophore molecules, proteases, ammonia,
e x op o l y s a cc h a r i d es , i n v i t r o bi o f i l m f o r m at i o n ,
endopolygalacturonase and endopolyglucanase, and VOCs
3 Results besides the ability to solubilize inorganic phosphate and to
fix nitrogen were examined. IAA is a phytohormone known
3.1 Phylogenetic classification of the isolates to be involved in root initiation, cell division, and cell enlarge-
ment (Salisbury 1994). Production of IAA by microbial spe-
A total of 13 indigenous, showing As and/or Hg tolerance, cies varies greatly among different species and strains and
was isolated and selected from the multi-contaminated soil of depends on the availability of substrate(s). Different biosyn-
an industrial disused area in northern Italy (SI29). thetic pathways for IAA production exist, sometimes in par-
Phylogenetic affiliation was performed by comparison of the allel in the same organism (Shahab et al. 2009) even if for
partial 16S rRNA gene fragments with the sequences stored in many years, it was assumed that tryptophan (Trp) was the only
NCBI database. BLAST analysis revealed that the closest rel- precursor of IAA (Glick et al. 1999). In this work, we evalu-
atives of the isolates belonged to the following three classes: ated the ability of the isolates to produce in vitro IAA from
Actinomycetales (Gordonia alkanivorans, Microbacterium tryptophan. This activity was fairly widespread in the collec-
paraoxydans, and Rhodococcus equi), Betaproteobacteria tion; 46 % of the isolates showed IAA production at different
(Cupriavidus necator and Achromobacter denitrificans), and extents (Fig. 2), confirming that the ability to produce this
Bacilli (Bacillus megaterium, Lysinibacillus macroides, and molecule is a common feature among PGPB and in general
Sporosarcina luteola) (Fig. 1). According to this first assess- among soil bacteria.
ment of the isolated strains, a typical microbial outline of The potential ability to fix atmospheric nitrogen was veri-
contaminated soil bacteria clearly appears. In particular, fied by the use of a combined nitrogen-deprived, organic
Gordonia and Rhodococcus genera are very common in nat- carbon-based medium which allows for the selective growth
ural environments and especially in polluted soil since they of nitrogen-fixing chemo-organotroph bacteria. Over 30 % of
are very powerful candidates for bioremediation processes the isolates showed this ability. In particular, three of the four
due to their capacity to degrade substituted and non- strains of G. alkanivorans (185.1, 195.23, and 207.37) were
substituted hydrocarbons, widespread toxic xenobiotics, and putatively able to fix N 2 and share this ability with
natural compounds not readily biodegradable (Arenskötter Macrococcus caseolyticus (205.54). However, these results
et al. 2004). Among the collections, four isolates showed would be confirmed by specific assay for nitrogenase activity.
99 % identity with the same strain (G. alkanivoras; accession Ammonia production is involved in protein catabolism, an
no. NR026488.1), and equally, two other isolates showed important process in soil fertility which leads to the mineral-
99 % identity with A. denitrificans (accession no. ization of organic residues to ammonium molecule becoming
NR118398). Even if the collection showed phylogenetic re- an available nitrogen source for plants. The bacterial strains of
dundancy, it was nevertheless decided to carry on the PGP the collection were hence tested for their ability to produce
characterization with all 13 isolates in order to explore the ammonia in peptone water. The production of this metabolite
functional diversity harbored by strains of the same species. by microorganisms can directly influence the growth of
J Soils Sediments (2017) 17:1224–1236 1229

Fig. 1 The evolutionary history was inferred using the neighbor-joining and are in the units of the number of base substitutions per site. The
method (Saitou and Nei 1987). The optimal tree with the sum of branch analysis involved 22 nucleotide sequences. All positions with less than
length = 0.66481969 is shown. The tree is drawn to scale, with branch 95 % site coverage were eliminated. That is, fewer than 5 % alignment
lengths in the same units as those of the evolutionary distances used to gaps, missing data, and ambiguous bases were allowed at any position.
infer the phylogenetic tree. The evolutionary distances were computed There were a total of 1150 positions in the final dataset. Evolutionary
using the maximum composite likelihood method (Tamura et al. 2004) analyses were conducted in MEGA6 (Tamura et al. 2013)

plants, through the supply of nitrogen (Wani et al. 2007); thus, 185.5, 193.23, and 207.37) and M. caseolyticus (205.54)
ammonia-producing bacteria can act as biofertilizers in deplet- showed to be able to produce EPSs.
ed soils such as those contaminated. Forty percent of the tested Biofilms are assemblages of cells embedded in a matrix
isolates showed this property (Fig. 2). composed of EPSs, proteins, and sometimes DNA (Morris
Bacterial production of EPSs in rhizosphere soil enhances and Monier 2003). Biofilm production is an important feature
water retention in the microbial and root environment and in PGPB, since this matrix enables the adhesion of bacteria to
regulates the diffusion of carbon sources. Bacterial EPSs are the roots and may be necessary for their colonization
secreted outside the cells and play an important role in the root (Beauregarda et al. 2013). Four strains showed the potential
colonization and in the formation of biofilm layers during ability to form biofilm in vitro (Fig. 2). Interestingly, only one
sessile growth. The four strains of G. alkanivorans (181.1, strain (M. caseolyticus 205.54) also showed EPS production,

Fig. 2 The 13 bacterial isolates are listed. Metal tolerance and in vitro PGP properties are shown. Considering the number of these features shown by
each strains, a score was assigned and a total score was calculated
1230 J Soils Sediments (2017) 17:1224–1236

confirming the fact that biofilm formation is a very complex obtain an effective phytoremediation. The strain S. luteola
event supported not only by EPS assembly. Santaella et al. 209.1 gained the highest score showing a significant increase
(2008) demonstrated that EPSs produced by a Rhyzobium of all the six measured parameters, except the number of buds,
sp. were not essential for biofilm formation on roots but were compared to the uninoculated control plants.
critical to colonization of the basal part of the root system Despite the positive results obtained with S. lycopersicum
increasing the stability of root-adhering soil. and A. thaliana, when L. albus seedlings were inoculated with
None of the isolates showed production of siderophores, the single five isolates, after 12 days of growth in the green-
proteases, endopolygalacturonase, endopolyglucanase, house, no statistically significant increases in the parameters
VOCs, and inorganic phosphate solubilization. indicative of the plant vigor were detected (data not shown).
For each isolate, the number of metal or metalloid tolerance
and in vitro PGP activities were scored (Fig. 2) and the strains
exhibiting the best scores were selected to be applied in 3.5 Microcosm experiments
in vivo assays.
The enhancement of the phytoextraction potentials of
B. juncea and L. albus induced by the addition of thiosulfate
3.4 In vivo effect of the bacterial inoculum on model plants was tested in microcosm trials, using plants inoculated by a
bacterial consortium composed by the five strains which
The ability of five strains (G. alkanivorans 185.1, C. necator showed the best scores in in vivo and in vitro assays
188.14, G. alkanivorans 195.23, G. alkanivorans 207.37, and (G. alkanivorans 185.1, 195.23, and 207.37; C. necator
S. luteola 209.1) to promote plant growth was first evaluated 188.14; and S. luteola 209.1).
in vivo on tomato plants, S. lycopersicum. The stem and root
lengths were chosen as parameters to assess the PGP potential
of the isolates. All the strains induced a statistically significant 3.5.1 Physico-chemical properties of the soil
increase in the length of sprouts (Fig. 3a). A statistically sig-
nificant increase in the primary root length was induced only The soil was characterized by a basic pH (8.06) and a content
by G. alkanivorans strains 185.1 and 207.37 and C. necator of organic matter of around 1.5 %. The soil texture showed a
188.14 (Fig. 3b). prevalence of the sand fraction (78.9 %), with 13.1 % of silt
On the basis of the positive results obtained on and 8.0 % of clay. The CEC was estimated to be
S. lycopersicum, plant growth promotion abilities of the five 15.6 cmol kg−1.
selected strains were then evaluated in vivo using the model Since the main objective of this study was to evaluate the
plant A. thaliana during a 20-day assay in the greenhouse. The possible use of assisted phytoextraction by adding a single
beneficial contribution of bacteria to plant growth was esti- mobilizing agent for both contaminants, As and Hg extract-
mated by rating for plant vigor in plants exposed to bacteria in abilities by thiosulfate were determined before plant growth.
comparison to uninoculated control plants. The differences in The results (Table 1) confirmed the efficiency of thiosulfate in
terms of numbers of leaves, buds, and flowers; length of the Hg solubilization according to previous findings in different
stem; and the fresh weight biomass of stem and roots were soils (Moreno et al. 2004, 2005; Pedron et al. 2011, 2013;
analyzed using the statistical Student’s t test (Fig. 4a–f). As Wang et al. 2014). Moreover, it was discovered that the effec-
shown, the largest increase is recorded at the level of biomass tiveness of the thiosulfate on As extractability was comparable
and root system. These data are particularly relevant with re- to that of 0.05 M phosphate, the specific additive often used
gard to an intervention of assisted phytoextraction aimed to for As phytoextraction. The two plant species, B. juncea and

Fig. 3 a, b Effect of inocula with the bacterial isolates 185.1, 188.14, 195.23, 207.37, and 209.1 on the sprouts length (a) and on the main root length (b)
of S. lycopersicum. Each inoculated seedling was related to the control by Student’s t test. *p < 0.05, **p < 0.01, ***p < 0.005
J Soils Sediments (2017) 17:1224–1236 1231

Fig. 4 a–f Effect of inocula with the bacterial isolates 185.1, 188.14, biomass of the stem (e), and the fresh weight biomass of the roots in
195.23, 207.37, and 209.1 on the stem length (a), the number of leaves A. thaliana. Each inoculated seedling was related to the control by
(b), the number of buds (c), the number of flowers (d), the fresh weight Student’s t test. *p < 0.05, **p < 0.01, ***p < 0.005

L. albus, were compared in terms of their growth and contam- presence of As and Hg, and the effect of PGPB was particu-
inant uptake with and without PGPB addition. larly relevant in the case of thiosulfate treatment as shown by
ANOVA analysis (Table 2).
3.5.2 Biomass production
Table 2 Evaluation of B. juncea and L. albus growth promotion in soil
Biomass production, together with concentration of contami- treated with the bacterial consortium (PGPB), thiosulfate (TS), and
nants in plants, is essential in quantifying the efficiency of thiosulfate plus the bacterial consortium (TS + PGPB)
phytoextraction of As and Hg. In this soil, the seed germina-
Plant species Shoot dry weight (mg)
tion of B. juncea and L. albus were not hampered by the
Non-treated PGPB TS TS + PGPB
Table 1 Extractabilities of As and Hg
B. juncea 494 ± 20 (b) 458 ± 16 (b) 326 ± 18 (a) 480 ± 26 (b)
Extractant As (mg kg−1) Hg (mg kg−1) L. albus 632 ± 33 (b) 679 ± 32 (b) 576 ± 21 (a) 658 ± 29 (b)

0.05 M KH2PO4 7.6 – The data were compared with the Bnon-treated^ control. Data are
0.27 M (NH4)2S2O3 7.4 12.9 expressed as mean values of shoot dry weight ± standard deviations of
five replicates. The means with the same letter for the same plant are not
Data are expressed as mg kg−1 dry soil significantly different (P < 0.05) as determined by the Tukey’s test
1232 J Soils Sediments (2017) 17:1224–1236

3.5.3 Arsenic and mercury phytoextraction concentration of the metal in plant tissues for the respective
dry biomass. Data are reported in Fig. 5a, b.
The addition of thiosulfate to the soil greatly promoted the Thiosulfate addition significantly increased the amounts of
uptake of Hg. Mercury concentration in the aerial parts metals removed by the plants by promoting the metal bioavail-
reached 120 mg kg−1 for B. juncea (approximately 40 times ability. From the data obtained, it could be noted a tendency to
the value of the controls) and 25 mg kg−1 for L. albus (ap- a greater phytoextraction capacity of plants after PGPB addi-
proximately 25 times the value of the controls). Mercury is tion especially in synergy with the action of the thiosulfate.
absorbed in greater quantities than As, particularly in the With respect to controls, PGPB increased the total accumula-
roots. Always, the concentrations of As and Hg were higher tion of As of about 42 % for L. albus and 85 % for B. juncea.
in roots than in the shoots. Following thiosulfate addition, the Similar results were obtained for Hg. In this case, the increases
concentration of Hg in roots of B. juncea increased from about of total accumulation were 35.8 and 44.7 % for L. albus and
30 to more than 1000 mg kg−1, while in the roots of L. albus, B. juncea, respectively.
the concentration of Hg raise from about 25 to more than
800 mg kg−1. L. albus, a N2-fixing legume, was shown to
positively influence phytoremediation strategy, when cultivat-
4 Discussion
ed in rotation with spring-summer crop. This typical species of
the Mediterranean area can survive also in winter season, and
The polluted soil under investigation showed a double con-
it is able to uptake a certain amount of metal improving, in the
tamination characterized by two of the most toxic elements,
meantime, soil fertility (Fumagalli et al. 2014). A similar
arsenic and mercury. In this study, the development of a com-
trend, even if with lower concentration values, was found also
bined strategy acting simultaneously on both and promoting a
in the case of As concentration in roots. In the time span of the
beneficial effect on the plant growth by PGPB inoculum was
experiment (only 30 days), the plants were able to absorb the
described.
metals and partially move them to the aerial parts. As known,
Thiosulfate properties as a good ligand in soil for Hg are
roots act as a barrier against contaminants (Ghosh and Singh
well known. Formation of water-soluble complexes such as
2005). However, the high increases of Hg and As concentra-
mercury-thiosulfate [Hg(S2O3)2]2− increases Hg solubility,
tions in shoots of B. Juncea support the hypothesis that repeat-
thus increasing Hg uptake by plants (Moreno et al. 2005;
ed cropping of this species could possibly reduce the bioavail-
Wang et al. 2014). This was confirmed in this study by micro-
able contaminant fraction in soil. These considerations cannot
cosm tests with B. juncea and L.albus, for which the effect of
be extended to lupine, whose cultivation can however increase
thiosulfate was particularly evident for mercury (Fig. 5b and
soil fertility (Fumagalli et al. 2014).
Table 3). Although the additive is specific for this metal, the
The concentration of As increased significantly in the
effect of thiosulfate on As bioavailability was of great interest.
shoots of B. juncea (about 15 mg kg−1), where the value in
The increase in As bioavailability promoted by thiosulfate
the control was negligible. In L. albus, after treatment, the
addition could be ascribed to the competition between arse-
concentration of As in the shoots (about 5 mg kg−1) increased
nate and sulfate ions for the same soil surfaces.
by about five times compared to the control value. These
Thiosulfate is widely used as fertilizer because, in the soil,
outcomes point out that treatment with thiosulfate produced,
it decomposes into sulfur and sulfate. Sulfate has an immedi-
with respect to non-treated plants, statistically significant dif-
ate action on the crops, while sulfur is released over time. In
ferences in the extractions of As and Hg. The Hg and As
this contaminated soil, thiosulfate may be transformed into
concentrations in the plant tissues of B. juncea and L. albus
tetrathionate and subsequently to sulfate as shown in the fol-
are reported in Table 3.
lowing scheme:
With the aim to explain the effect of thiosulfate on As
extractability and to verify the hypothesis of a competition S2 O3 2− →S4 O6 2− →SO3 2− →SO4 2−
between arsenate and sulfate (derived from thiosulfate), a
comparison extractability test was performed (Table 4). Data The reaction could be either abiotic or biotic depending on
showed a very similar As extractability with both chemicals, the soil microbial community, and the oxidation kinetics also
and thus, the truthfulness of this assertion seemed to be con- depends on the soil characteristics. In general, the reaction can
firmed. Please refer to the discussion for further details and be completed in a few days (Barbosa-Jefferson et al. 1998) but
explanations. it can be faster in the microcosms for the presence of the plants
which further stimulate microbial activity. Thus, with an ex-
3.5.4 Total metal accumulation cess of sulfate ions, which compete for the same sites on soil
surfaces, we can suppose that arsenate ions are released in the
Phytoextraction efficiency can be estimated by total accumu- liquid phase of the soil in a potentially bioavailable form for
lation (Jarrell and Beverly 1981), defined as the product of the plant uptake.
J Soils Sediments (2017) 17:1224–1236 1233

Table 3 Measures of As and Hg

contents in shoots and roots of the Plant species Treatment Metal content
selected plants
As (mg kg−1) Hg (mg kg−1)

Shoots Roots Shoots Roots

B. juncea Non-treated nd (a) nd (a) 2.44 ± 0.7 (a) 26.9 ± 6.2 (a)
PGPB nd (a) nd (a) 2.34 ± 0.4 (a) 38.7 ± 5.1 (a)
TS 12.8 ± 3.6 (b) 58.9 ± 6.2 (b) 120 ± 7.5 (b) 1115 ± 284 (b)
TS + PGPB 16.1 ± 2.6 (b) 61.5 ± 9.9 (b) 118 ± 10 (b) 1033 ± 439 (b)
L. albus Non-treated 0.80 ± 0.09 (a) 1.8 ± 0.55 (a) 0.98 ± 0.2 (a) 9.5 ± 0.6 (a)
PGPB 0.68 ± 0.03 (a) 1.7 ± 2.7 (a) 0.97 ± 0.1 (a) 11.0 ± 1.9 (a)
TS 4.1 ± 1.9 (b) 12.6 ± 2.9 (b) 23.3 ± 3.4 (b) 806 ± 42.1 (b)
TS + PGPB 5.1 ± 1.2 (b) 14.0 ± 7.7 (b) 27.7 ± 6.0 (b) 896 ± 33.4 (c)

Data are expressed as mg kg−1 dry weight. For each metal, the means with the same letter for the same plant
tissue are not significantly different (P < 0.05) as determined by the Tukey’s test
Non-treated control; TS thiosulfate; PGPB bacterial consortium with the strains185.1, 188.14, 195.23, 207.37,
and 209.1

To test this hypothesis, As extractions with sulfate were during a phytoextraction intervention the amount of the ex-
performed in comparison with thiosulfate extraction. Data tracted metal is directly proportional to the biomass produced,
show that the As extractabilities by thiosulfate and sulfate, in it is essential to obtain a high amount of plant biomass (Wu
this soil, were the same. Competition between sulfate and et al. 2006). The use of a bacterial inoculum at the root level,
arsenate for the same adsorption sites is particularly effective able to promote plant growth by increasing root development,
in an alkaline environment, and SO42− ions are found to form may represent a useful tool for an effective phytoremediation
outer and inner sphere complexes with iron oxides (Myneni strategy (Rajkumar et al. 2012). Aimed by this rational, indig-
et al. 1998; Fukushi and Sverjensky 2007; Petruzzelli et al. enous As- and Hg-tolerant bacteria from the multi-
2014). This confirms that the sulfate-arsenate competition contaminated soil were isolated and selected based on differ-
could be the basis for the release of As in solution and thus ent phylogenetic affiliations. Members of Proteobacteria,
in forms bioavailable to plants. However, a further aspect Firmicutes, and Actinobacteria subdivisions showing, in
needs to be considered; the interactions between S and As some cases, multiple tolerance toward As(V), As(III), and
affect both the absorption of the contaminant and its transport Hg(II) were detected. All isolates showed in vitro properties
to the aerial part of the plant. The addition of sulfur to the soil known to have positive influence on plant growth. Results
promotes the absorption of As by plants, and S plays an anti- revealed that all strains, with one exception (R. equi,
stress role by reducing the toxicity of As (Duan et al. 2013). 199.33), showed, in vitro, one to three PGP traits (Table 2).
Thiosulfate can act either as nutrient or detoxifying agent, due The following step was to assay the ability of plant growth-
to the stimulation of plant-defensive systems, and influence promoting effect in vivo, and to do this, a selection of those
the As accumulation in plant tissues. strains showing the best overall rating concerning metal and
The presence in the soil of heavy metals or other contam- metalloid tolerances and PGP properties was chosen.
inants adversely affects root development leading to a lower S. lycopersicum (tomato) is a fast-growing and useful species
production of plant biomass (Glick et al. 2007). Root devel- and therefore an ideal plant model to quickly verify the PGP
opment influences the growth of the whole plant; the more the potential of the isolates. Tomato axenic seedlings inoculated
root apparatus is developed, the greater will be the ability to with the selected strains demonstrated their ability to increase
uptake nutrients and water required for the plant growth. Since the length of the stem (all strains) and of the central root (three
out of five strains). Literature data indicated that bacterial
strains isolated from contaminated sites can promote the
Table 4 Comparison of As extractability by thiosulfate and sulfate growth of tomato. Jiang et al. (2008) reported that a
Burkholderia sp. strain, isolated from rice contaminated with
Extractant concentration (NH4)2S2O3 (NH4)2SO4 heavy metals and showing metal tolerance and PGP activities,
0.14 M 3.83 ± 0.05 3.87 ± 0.08
was able to promote a significant increase in the wet weight of
0.27 M 6.98 ± 0.08 6.06 ± 0.03
the biomass of tomato and corn. Many others are the cases
where bacterial strains from heavy metal-contaminated soils
Data are expressed in mg kg−1 dry soil were shown to promote tomato growth in addition to the
1234 J Soils Sediments (2017) 17:1224–1236

Fig. 5 a, b Total accumulation of As (a) and Hg (b) in the aerial part of the plants. CT control; TS thiosulfate; B Brassica; L Lupinus; and PGPB bacterial
consortium with the strains 185.1, 188.14, 195.23, 207.37, and 209.1

ability to accumulate metals in laboratory conditions (Kumar important role in adaptation strategies improving plant growth
and Patra 2013). and Hg tolerance. The effects of PGPB addition were not
The strains which demonstrated a positive effect in enhanc- particularly remarkable on the concentration of contaminants
ing tomato plant growth were also applied in an in vivo in plants due perhaps to a dilution effect related to the increase
growth promotion assay with the model plant A. thaliana. in biomass production. The addition of PGPB produced a
After 20 days of growth in a greenhouse, the estimated param- general positive effect by considering the values of total
eters (number of leaves, buds, flowers; length of the stem; and accumulation.
fresh weights of roots and stem) showed that the better strain Sessitsch et al. (2013) reported that inoculation with mi-
in terms of growth-promoting performance, in the adopted crobes increases the shoot biomass rather than the trace ele-
experimental conditions, was S. luteola (209.1). These data ment concentration. Moreover, compared to single-strain in-
confirmed that the isolates were able, each with its own pecu- oculation, the application of a multiple-strain consortium was
liarity, to stimulate the growth of various plant species com- often more effective in inducing both increased biomass and
pared to control plants not bacterized. trace element contents in shoots.
Encouraged by these positive results with model plants, we
inoculated L. albus seedlings with the single five isolates, but
after 12 days of growth in the greenhouse, no statistically 5 Conclusions
significant increases in the parameters indicative of the plant
vigor were detected. This negative result could be owed to a The combination of the addition of thiosulfate as
reduced ability of selected bacteria to establish a stable inter- phytoextraction strategy for As and Hg together with the in-
action with this plant species. Possibly, unlike results with the oculum of a consortium of metal-tolerant bacteria showing
model plant, A. thaliana showed, inoculum with a single PGP properties showed a great potential in enhancing the
strain, was not able to strong colonize L. albus root system phytoremediation approach in a soil with multiple contamina-
due perhaps to a competition between these bacteria and its tion. Thiosulfate, generally used to solubilize Hg, was suc-
own rhizosphere community. Because of these last adverse cessfully applied as single chemical agent to mobilize both
results, attempting to exploit the potential synergistic effect Hg and As. Moreover, the further addition of a PGPB consor-
of the strains, we set up a bacterial consortium prepared with tium to B. juncea and L. albus plant root in microcosms led to
all the five strains which resulted positive in PGP of a meaningful increase of the biomass and a moderate increase
A. thaliana and tomato. Microcosm experiments with in the concentration of metals in shoots. Taking into account
B. juncea and L. albus were then performed with and without total accumulation, in B. juncea treated with the PGPB con-
the inoculum of PGPB consortium. The positive effect of sortium, increases up to 85 % for As and up to 45 % for Hg
PGPB was particularly noteworthy, relating to the biomass were observed. These promising laboratory results need to be
production. In the absence of PGPB, thiosulfate treatment confirmed with field applications to evaluate an eventual det-
reduced the biomass production of B. juncea, probably due rimental effect of environmental conditions on PGPB-plant
to the high increase in the concentrations of the two contam- interactions in contaminated soil.
inants, in particular Hg, which significantly stressed the me-
tabolism of the plant. In microcosms treated with PGPB, the
Acknowledgments The experimental work carried out at ISE-CNR
addition of thiosulfate did not negatively influence the growth was fully funded by Syndial s.p.a. ER was supported by Università degli
of plants and their biomass; thus, PGPB appear to play an Studi di Milano, DeFENS, the European Social Found, and Regione
J Soils Sediments (2017) 17:1224–1236 1235

Lombardia (contract BDote Ricerca^). RM was financially supported by Glick BR, Todorovic B, Czarny J, Cheng Z, Duan J, McConkey B (2007)
the project BIOGESTECA no. 15083/RCC BFondo per la Promozione di Promotion of plant growth by bacterial ACC deaminase. Crit Rev
Accordi Istituzionali^ (Regione Lombardia, Italy). We thank Francesco Plant Sci 26:227–242
Rodriguez for the bioinformatic support. Gutiérrez-Ginés MJ, Hernández AJ, Pérez-Leblic MI, Pastor J,
Vangronsveld J (2014) Phytoremediation of soils co-contaminated
by organic compounds and heavy metals: bioassays with Lupinus
luteus L. and associated endophytic bacteria. J Environ Manage 143:
References 197–207
Islam MR, Madhaiyan M, Deka Boruah HP, Yim W, Lee G, Saravanan
Aafi NE, Brhada F, Dary M, Maltouf AF, Pajuelo E (2012) VS, Fu Q, Hu H, Sa T (2009) Characterization of plant growth-
Rhizostabilization of metals in soils using Lupinus luteus inoculated promoting traits of free-living diazotrophic bacteria and their inoc-
with the metal resistant rhizobacterium Serratia sp. MSMC541. Int J ulation effects on growth and nitrogen uptake of crop plants. J
Phytorem 14(3):261–274 Microbiol Biotechnol 12:13–22
Jarrell WM, Beverly RB (1981) Dilution effect in plant nutrition studies.
AMAP/UNEP (2013) Technical background report for the global mercu-
Adv Agron 34:197–224
ry assessment 2013. Arctic monitoring and assessment
Jiang C, Sheng X, Qian M, Wang Q (2008) Isolation and characterization
Arenskötter M, Bröker D, Steinbüchel A (2004) Biology of the metabol-
of a heavy metal-resistant Burkholderia sp. from heavy metal-
ically diverse genus gordonia. Appl Environ Microb 6:3195–3204
contaminated paddy field soil and its potential in promoting plant
Barbafieri M, Japenga J, Romkens P, Petruzzelli G, Pedron F (2013) growth and heavy metal accumulation in metal-polluted soil.
Protocols for applying phytotechnologies in metal-contaminated Chemosphere 72:154–164
soils. In: Gupta DK (ed) Plant-based remediation processes. Kumar VK, Patra DD (2013) Influence of nickel and cadmium resistant
Springer, Berlin, Heidelberg, pp 19–37 PGPB on metal accumulation and growth response of Lycopersicum
Barber SA (1984) Soil nutrient bioavailability. John Wiley & Sons, NY esculentum plants grown in fly ash amended soil. Water Air Soil Poll
Barbosa-Jefferson VL, Zhao FJ, McGrath SP, Magan N (1998) 224:1357
Thiosulphate and tetrathionate oxidation in arable soils. Soil Biol Liba CM, Ferrara FIS, Manfio GP, Fantinatti-Garboggini F, Albuquerque
Biochem 30:553–559 RC, Pavan C, Ramos PL, Moreira-Filho CA, Barbosa HR (2006)
Beauregarda PB, Chaib Y, Vlamakisa H, Losickb R, Koltera R (2013) Nitrogen-fixing chemo-organotrophic bacteria isolated from
Bacillus subtilis biofilm induction by plant polysaccharides. Proc Cyanobacteria-deprived lichens and their ability to solubilize phos-
Natl Acad Sci U S A 110(17):1621–1630 phate and to release amino acids and phytohormones. J Appl
Brick JM, Bostock RM, Silverstone SE (1991) Rapid in situ assay for Microbiol 101–5:1076–1086
indoleacetic acid production by bacteria immobilized on nitrocellu- Loper JE, Scroth MN (1986) Influence of bacterial sources on indole-
lose membrane. Appl Environ Microbiol 57:535–538 3 acetic acid on root elongation of sugarbeet. Phytopathology
Cappuccino JC, Sherman N (1992) Microbiology: a laboratory manual, 76:386–389
3rd ed. Benjamin/Cummings Publishing Company, New York 125– Meers E, Tack FMG, Van Slycken S, Ruttens A, Du Laing G,
179 Vangronsveld J, Verloo MG (2008) Chemically assisted
Duan G, Liu W, Chen X, Hu Y, Zhu Y (2013) Association of arsenic with phytoextraction: a review of potential soil amendments for in-
nutrient elements in rice plants. Metallomics 5:784–792 creasing plant uptake of heavy metals. Int J Phytoremediation
Fukushi K, Sverjensky DA (2007) A surface complexation model for 10:390–414
sulfate and selenate on iron oxides consistent with spectroscopic Milagres AMF, Machuca A, Napoleão D (1999) Detection of siderophore
and theoretical molecular evidence. Geochim Cosmochim Ac 71: production from several fungi and bacteria by a modification of
1–24 chrome azurol S (CAS) agar plate assay. J Microbiol Meth 37–1:
1–6
Fumagalli P, Comolli R, Ferre C, Ghiani A, Gentili R, Citterio S (2014)
The rotation of white lupin (Lupinus albus L.) with metal- Moreno FN, Anderson CWN, Stewart RB, Robinson BH (2004)
accumulating plant crops: a strategy to increase the benefits of soil Phytoremediation of mercury-contaminated mine tailings by in-
duced plant-mercury accumulation. Environ Pract 6:165–175
phytoremediation. J Environ Manag 145:35–42
Moreno FN, Anderson CWN, Stewart RB, Robinson BH, Ghomshei M,
Gadd GM (2000) Bioremedial potential of microbial mechanisms of
Meech JA (2005) Induced plant uptake and transport of mercury in
metal mobilization and immobilization. Curr Opin Biotechnol
the presence of sulphur-containing ligands and humic acid. New
11:271–279
Phytol 166:445–454
Garelick H, Jones H, Dybowska A, Valsami-Jones E (2008) Arsenic Morris CE, Monier JM (2003) The ecological significance of biofilm
pollution sources. In: Garelick H, Jones H (eds) Arsenic pollution formation by plant-associated bacteria. Annu Rev Phytopathol 41:
and remediation: an international perspective. Springer, New York, 429–453
pp 17–60 Myneni SCB, Traina SJ, Waychunas GA, Logan TJ (1998) Vibrational
Gee GW, Bauder JW (1986) Particle-size analysis. In: Klute A (ed) spectroscopy of functional group chemistry and arsenate coordina-
Methods of soil analysis. Part 1 Physical and mineralogical methods, tion in ettringite. Geochim Cosmochim Acta 62:3499–3514
2nd edn. Agronomy Monograph No. 9, American Society of Nautiyal CS (1999) An efficient microbiological growth medium for
Agronomy/Soil Science Society of America, Madison, WI., pp screening phosphate solubilizing microorganisms. FEMS
383–411 Microbiol Lett 170:265–270
Ghosh M, Singh SP (2005) A review on phytoremediation of heavy Nelson DW, Sommers LE (1996) Total carbon, organic carbon and or-
metals and utilization of its by products. As J Energy Env 6(04): ganic matter. In: Sparks DL (ed) Methods of soil analysis. Part 3
214–231 Chemical Methods. Soil Science Society of America Book Series,
Glick BR (2010) Using soil bacteria to facilitate phytoremediation. Soil Science Society of America Inc, Madison, USA., pp 961–1010
Biotechnol Adv 28:367–374 Nielsen P, Sørensen J (1997) Multi-target and medium-independent fun-
Glick BR, Patten CL, Holguin G, Penrose DM (1999) Biochemical and gal antagonism by hydrolitic enzymes in Paenibacillus polymyxa
genetic mechanisms used by plant growth promoting bacteria. and Bacillus pumilus strains from barley rhizosphere. FEMS
Imperial College Press, London, UK Microbiol Ecol 22–3:183–192
1236 J Soils Sediments (2017) 17:1224–1236

Pedron F, Petruzzelli G, Barbafieri M, Tassi E, Ambrosini P, Patata L Shahab S, Ahmed N, Khan NS (2009) Indole acetic acid production and
(2011) Mercury mobilization in a contaminated industrial soil for enhanced plant growth promotion by indigenous PSBs. Afr J Agric
phytoremediation. Commun Soil Sci Plan 42:2767–2777 Res 4(11):1312–1316
Pedron F, Petruzzelli G, Barbafieri M, Tassi E (2013) Remediation of a Sparks DL (1998) Methods of soil analysis. Part 3. Chemical Methods,
mercury-contaminated industrial soil using bioavailable contami- Soil Science Society of America Book Series. Soil Science Society
nant stripping. Pedosphere 23(1):104–110 of America, Madison, WI
Petruzzelli G, Pedron F, Rosellini I, Barbafieri M (2013) Sumner ME, Miller WP (1996) Cation exchange capacity and ex-
Phytoremediation towards the future: focus on bioavailable contam- change coefficients. In: Sparks DL (ed) Methods of soil analy-
inants. In: Gupta DK (ed) Plant-based remediation processes. sis. Part 3 Chemical Methods. Soil Science Society of America
Springer, Berlin, Heidelberg, pp 273–289 Book Series, Soil Science Society of America Inc, Madison,
Petruzzelli G, Pedron F, Rosellini I (2014) Effects of thiosulfate on the USA., pp 1201–1230
adsorption of arsenate on hematite with a view to phytoextraction. Tamura K, Nei M, Kumar S (2004) Prospects for inferring very large
Res J Environ Earth Sci 6(6):326–332 phylogenies by using the neighbor-joining method. Proc Natl
Rajkumar M, Sandhya S, Prasad MNV, Freitas H (2012) Perspectives of Acad Sci U S A 101:11030–11035
plant-associated microbes in heavy metal phytoremediation. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013) MEGA6:
Biotechnol Adv 30:1562–1574 molecular evolutionary genetics analysis version 6.0. Mol Biol Evol
30:2725–2729
Romheld V, Marschner H (1986) Mobilization of iron in the rhizosphere
Tassi E, Pedron F, Barbafieri M, Petruzzelli G (2004) Phosphate-assisted
of different plant species. Adv Plant Nutr 2:155–204
phytoextraction in As-contaminated soil. Eng Life Sci 4:341–346
Ryu C-M, Farag MA, Hu C-H, Reddy MS, Wei H-X, Paré PW, Kloepper Thomas GW (1996) Soil pH and soil acidity. In: Sparks DL (ed) Methods
JW (2003) Bacterial volatiles promote growth in Arabidopsis. Proc of soil analysis. Part 3 Chemical Methods, Soil Science Society of
Natl Acad Sci U S A 100–8:4927–4932 America Book Series, Soil Science Society of America Inc.,
Saitou N, Nei M (1987) The neighbor-joining method: a new method for Madison, USA., pp 475–490
reconstructing phylogenetic trees. Mol Biol Evol 4:406–425 USEPA (1998) SW-846 Method 7473. Mercury in solids and solutions by
Salisbury FB (1994) The role of plant hormones. In: Wilkinson RE (ed) thermal decomposition, amalgamation and atomic absorption spec-
Plant–environment interactions., pp 39–81 trophotometry. Method 7473, U.S. EPA, Washington, DC
Santaella C, Schue M, Berge O, Heulin T, Achouak W (2008) The Wang J, Feng X, Anderson CWN, Wang H, Wang L (2014)
exopolysaccharide of Rhizobium sp. YAS34 is not necessary for Thiosulphate-induced mercury accumulation by plants: metal up-
biofilm formation on Arabidopsis thaliana and Brassica napus take and transformation of mercury fractionation in soil—results
roots but contributes to root colonization. Environ Microbiol from a field study. Plant Soil 375:21–33
10(8):2150–2163 Wani PA, Khan MS, Zaidi A (2007) Plant growth-promoting potentials,
Schwyn B, Neilands JB (1987) Universal chemical assay for the detection and metal solubilization by Bacillus sp. isolated from alluvial soil.
and determination of siderophores. Anal Biochem 160(1):47–56 Curr Microbiol 54(3):237–243
Sessitsch A, Kuffner M, Kidd P, Vangronsveld J, Wenzel WW, Fallmann Weaver PK, Wall JD, Guest H (1975) Characterization of
K, Puschenreiter M (2013) The role of plant-associated bacteria in Rhodopseudomonas capsulata. Arch Microbiol 105:207–216
the mobilization and phytoextraction of trace elements in contami- Wu SC, Cheung KC, Luo YM, Wong MH (2006) Effects of inoculation
nated soils. Soil Biol Biochem 60:182–194 of plant growth-promoting Rhizobacteria on metal uptake by
Brassica juncea. Environ Pollut 40:124–135