Ac

DMUF

a

ARRA

KPRPMRI

1

taaoe(ainome

ht

EY

0h

Applied Soil Ecology 62 (2012) 42– 49

Contents lists available at SciVerse ScienceDirect

Applied Soil Ecology

journa l h o me page: www.elsev ier .com/ locate /apsoi l

ssessment of Miscanthus × giganteus for rhizoremediation of long term PAHontaminated soils

idier Techer ∗, Claudia Martinez-Chois, Philippe Laval-Gilly, Sonia Henry, Amar Bennasroune,arielle D’Innocenzo, Jairo Falla

niversité de Lorraine, Laboratoire des Interactions Ecotoxicologie, Biodiversité, Ecosystèmes (LIEBE), CNRS UMR 7146, IUT Thionville-Yutz, Espace Cormontaigne, Yutz, F-57970,rance

r t i c l e i n f o

rticle history:eceived 11 October 2011eceived in revised form 8 June 2012ccepted 24 July 2012

eywords:AHhizoremediation

a b s t r a c t

The purpose of this study was to investigate the potential of Miscanthus × giganteus (M × G) for rhizore-mediation of long term PAH-polluted soils. To evaluate its growth ability on contaminated substrates, apot experiment was a pot experiment was conducted. Plant development was compared to that obtainedin a reference soil. Pollutant dissipation with several other physico-chemical and microbiological param-eters (including bacterial community diversity molecular analysis) were investigated in order to bettercharacterize the rhizosphere effects of M × G. Field trials were also conducted to confirm the feasibilityof crop installation in situ. Plants demonstrated a physiological adaptation to soils from various PAH con-

−1

olluted soilsiscanthus × giganteus

ehabilitationndustrial crop

tamination levels (from 26 to 364 mg PAH kg dry soil) both in laboratory and in field scale conditions.Changes in rhizosphere bacterial community were observed via specific bacterial phylotype selectionin the root vicinity. Despite the lack of conclusive trends for phytoremediation, slight decreases in total4-ring PAH concentration would suggest a positive influence of growing plants in the long term. Further-more, significant organic carbon inputs and nitrate losses were measured after 17 weeks of laboratorycultivation, indicating a global improvement of soil agronomic quality.

. Introduction

Polycyclic aromatic hydrocarbons (PAHs) are composed ofwo or more fused benzene-rings in linear, angular or clusterrrangements (Blumer, 1976; Cerniglia, 1992). These pollutantsre commonly found in air, water, sediment and soil subsurfacesf industrialized countries, since the most prominent source ofmission is related to anthropogenic processing of fossil fuelsBlumer, 1976; Wilcke, 2007). Once introduced into soils, PAH show

strong affinity for organic phases due to their high hydrophobic-ty (Blumer, 1976; Bogan and Sullivan, 2003). They may dissolve inon-aqueous liquid phase (NAPL), partition onto coal tar particles

r in organic matter and become progressively sequestered in soilicropores (Bogan and Sullivan, 2003). Such “aging” or “weath-

ring” processes may conduce to PAH accumulation for many

Abbreviations: M × G, Miscanthus × giganteus; PAH, polycyclic aromaticydrocarbon; PCR–TGGE, polymerase chain reaction–temperature gradient gel elec-rophoresis; ch., chlorophyll.∗ Corresponding author at: Université de Lorraine, Laboratoire des Interactions

cotoxicologie, Biodiversité, Ecosystèmes (LIEBE), CNRS UMR 7146, IUT Thionville-utz, Espace Cormontaigne, Yutz, F-57970, France. Tel.: +33 6 13 07 92 41.

E-mail address: erwann9@hotmail.com (D. Techer).

929-1393/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.apsoil.2012.07.009

© 2012 Elsevier B.V. All rights reserved.

years (Bogan and Sullivan, 2003). Therefore, soils can act as “sinks”for these contaminants (Blumer, 1976; Bogan and Sullivan, 2003;Wilcke, 2007). Because of their hazard risk to human health and toliving organisms, with potential carcinogenic and mutagenic prop-erties, PAH removal from the environment is an important concern(Bouchez et al., 1995).

Among the several technologies that have been tested forPAH dissipation in soils during the last decades, bioremediationthrough bacterial degradation has been shown to be one of themost environmentally friendly and promising technique (Haritashand Kaushik, 2009). Indeed, the chemically stable PAH moleculesbecome more reactive following the enzymatic action of microbialdioxygenases that catalyze successive ring oxidations and fissions,allowing for progressive mineralization of the mother compound(Cerniglia, 1992).

Biodegradation of organic contaminants in soils has beendemonstrated to be stimulated in the vicinity of plant roots, mainlydue to the enhancement of microbial growth and activity follow-ing root exudation (Aprill and Sims, 1990; Gerhardt et al., 2009).The interrelationships between plants and soil microorganisms

involved in the breakdown of pollutants lead to the formulation ofthe “rhizoremediation technology” (Aprill and Sims, 1990; Nicholset al., 1997). Finally, it appears that such a plant-assisted remedia-tion technology has the advantages of being: (i) less expensive than

d Soil E

edamssa

rifmcmdt2o

pmsitchmsomt(

ociiplawpacc

2

2

wiasc(P1fo(7

a

D. Techer et al. / Applie

ngineered physicochemical techniques due to its solar-energyriven approach (root exudation is dependent on initial CO2 fix-tion by plants), (ii) particularly effect-efficient as a “polishing”ethod with generally no additional soil contamination, and (iii)

ubstantial facilities to be implemented in situ when treating largeurface areas of contaminated soils (Gerhardt et al., 2009; Haritashnd Kaushik, 2009; Schwitzguébel et al., 2002).

In spite of its mentioned advantages, the efficiency of PAHhizodegradation may be mitigated due to the presence of othernorganic and organic pollutants. Indeed, polluted soils located inormer industrial sites involved in petroleum and coke transfor-

ation are often characterized by the presence of heavy metals,yanides and other co-contaminants (Roy et al., 2005). These latteray induce harmful effects toward plant growth as well as root

evelopment and impede subsequent plant-assisted bioremedia-ion in rhizosphere soils (Haritash and Kaushik, 2009; Roy et al.,005). Therefore, special attention must be devoted to the choicef plants before implementing phytoremediation programs in situ.

Miscanthus × giganteus (M × G) is a sterile perennial grassroducing a high-yield biomass under temperate climates. Further-ore, this Gramineae is capable of growing on a large variety of

oils (Lewandowski et al., 2003). Previous works showed the abil-ty of this crop to develop on domestic sludge amended soils (upo 200 t of material/ha), without any significant heavy metal bio-oncentration in aerial part (Fernando et al., 2004). Arduini andis collaborators also reported its tolerance to chromium and cad-ium (two metals often found in hydrocarbon industrially polluted

oils) following hydroponic studies (Arduini et al., 2006a,b). More-ver, the potential of M × G root exudates to selectively enhanceicrobial growth and degradation activity of PAH-degrading bac-

erial consortia was recently demonstrated through in vitro assaysTécher et al., 2011).

The aim of this study was to investigate the growth performancef Miscanthus × giganteus (biomass production and chlorophyllontents) associated with subsequent PAH dissipation and changesn bacterial communities in two industrially polluted soils differentn their hydrocarbon and heavy metal contents. For this purpose,ot experiments were conducted under controlled conditions in the

aboratory. Moreover, to address the paucity of information avail-ble in the literature concerning simultaneous rhizoremediationith crops and soil quality improvements, several physicochemicalarameters (pH, soluble organic carbon, nitrite, nitrate, phosphatend potassium contents) were investigated. Field trials were alsoonducted in order to test the feasibility of Miscanthus × giganteusrop cultivation at a regional scale.

. Materials and methods

.1. Soil collection and preparation

Two industrially polluted soils, respectively named M and H,ere collected at a former coke plant in Northern France. Accord-

ng to the FAO (World Reference Base for Soil Resources, 2006), Mnd H were classified as Technosol soils. Soil type M came from aite with the following characteristics (BASOL, 2008): a total PAHoncentration of 26 mg kg−1 dry soil, a heavy metal concentrationsmg kg−1 dry soil) in the range of 48–140 for Cr, 530–10,000 forb, 880–7700 for Zn and the presence of cyanides varying from1 to 163 mg kg−1 dry soil. Soil type H came from a site with theollowing characteristics (BASOL, 2008): a total PAH concentrationf 364 mg kg−1 dry soil, but a lower heavy metal contamination

mg kg−1 dry soil) in the range of 58–67 for Cr, 220–290 for Pb,40–830 for Zn and no detectable cyanide contamination.

Soil preparation before initial (t0) physicochemical analysesnd pot experiments included air drying, sieving (<2 mm) and

cology 62 (2012) 42– 49 43

mixing. Moreover, an artificial soil was used as a referential forMiscanthus × giganteus growth comparison. This reference soil wasreconstituted according to ISO standard protocols (InternationalStandardization Organisation, ISO 11269-2, 1995) and was com-posed (w/w) of 55% of a commercial potting A3 soil (LUFA Speyer)(He et al., 1995), 42.5% acid-washed sand and 2.5% sphagnum peat.Physicochemical characteristics of the three soils were determinedaccording to standard techniques (Cunningham et al., 1996). Thereference soil was classified as a sandy loam soil with a texture ofsand (62%), silt (18.5%), clay (19.5%) and a pHwater of (7.16). The soiltype M was a sandy loam soil and had the following composition:sand (66%), silt (20%) and clay (14%), total organic C (11.5%) and C/N(21). The soil type H was a loamy sand soil and presented the fol-lowing composition: sand (79%), silt (15%), clay (6%), total organicC (8.6%) and C/N (16).

2.2. Experimental design and plant analyses

The greenhouse experiment had a completely randomized blockdesign with five replications that had the following treatments: ref-erence soil planted with M × G, soil types M and H planted withM × G and, soil types M and H unplanted. M × G growth in thetwo polluted soils M and H was monitored during 17 weeks andcompared to growth measured in the reference soil. The unplantedcontrol pots of the two polluted soils were set up in parallel fora better assessment of the effects of planting on soil quality (PAHdissipation, soil physicochemical characteristics and bacterial com-munity changes).

Before planting, rhizomes of M × G from 7 to 10 cm wereplaced in quartz sand for shoot emergence during two weeks.Then plastic pots were filled with 700 g dry soil, receiving onegerminated rhizome (for those planted) and brought to 70% ofwater-holding capacity using deionised water. Growth conditionsincluded the emission of photosynthetically active radiation at 450and 650 nm (OSRAM Fluora L36W/77) and a day/night cycle of16/8 h at 20–23 ◦C. Stem length at the first internodes was mea-sured weekly.

Plants were harvested after 4.5 months of growth. The lastemerged leaf of each plant was weighed and immediately storedinto liquid nitrogen for further determination of chlorophyll pig-ment contents. For this purpose, 80 mg of frozen plant materialwas mixed with 500 mg glass beads (0.1 mm diameter) and 750 �LN,N-dimethylformamide (DMF) during 5 min. Additional steps ofwashing (with 1 mL of DMF), centrifugation (12,000 × g; 90 s) andmixing were repeated five times to recover all photosyntheticpigments. Absorbances of chlorophyll a and b were read at 664and 647 nm into a glass microtiter plate and chlorophyll contentscalculated according to Inskeep and Bloom (1985) formula. Theremaining harvested plant, i.e. rhizomes, roots, stems and leaveswere separated to estimate their fresh weight and dry weight tothe nearest 0.1 g after 48 h in an oven at 105 ◦C.

2.3. Soil chemical analyses

The content of the 16 PAH congeners listed as priority pollutantsby the US Environmental Protection Agency (US-EPA) was analyzedaccording to the French standard methods using Soxhlet extractionand HPLC quantification (AFNOR XP X33-012, 2000).

In order to perform the analyses of water extractable soil nutri-ents, a soil water extraction procedure was set up, comprising2 h mixing on an orbital shaker with ultrapure water (soil:waterratio (v:v) of 1:5) at ambient temperature. A settling period was

maintained for 10 min prior to filtration (0.45 �m). Then, pH andelectrical conductivity were directly measured, followed by dis-solved organic carbon, nitrite, nitrate, phosphate and potassiumcontents.

4 d Soil

sa2bIr

(sin

m1Talb1b

pnabopip

q(

2

6pcowrrt

a1gm

rmLwwroHL

fmsmd

4 D. Techer et al. / Applie

Dissolved organic carbon was indirectly estimated by mea-uring total reducing sugars and amino-acids. This was achievedccording to a microplate assay developed by (Kenealy and Jeffries,003): 35 �L of soil water suspension was mixed with 150 �L 2,2-icinchoninic acid (BCA) reagent (BC Assay protein quantitation kit;nterchim) and incubated at 80 ◦C for 45 min. Absorbances wereead at 562 nm against appropriate blanks and glucose standards.

For nitrite determination, 40 �L of Griess-Ilosvay reagentAlexander and Clark, 1965) was added to 200 �L of soil wateruspension. Absorbance was measured at 540 nm after 15 min ofncubation at ambient temperature against appropriate blanks anditrite standards.

Nitrate concentration in soil water suspensions was also deter-ined. 10 �L of sodium salicylate (0.5%, w/v) were mixed with

00 �L of soil water suspension and were evaporated 2 h at 80 ◦C.hen 20 �L sulphuric acid (98%) were added and incubated 10 mint ambient temperature. 150 �L distilled water was added, fol-owed by 150 �L of a sodium hydroxide (40%, w/v) and potassiumitartrate (6%, w/v) solution. Absorbance was read at 415 nm after5 min of incubation at ambient temperature against appropriatelanks and nitrate standards.

For soluble phosphate determination, a molybdic reagent wasrepared by mixing (per liter) 74 mL H2SO4 (98%) with 6 g ammo-ium molybdate in 125 mL of ultrapure water and 0.1455 gntimony potassium tartrate in 50 mL H2SO4 (4 N). Then, 1 g ascor-ic acid was dissolved into 50 mL of the molybdic solution. 40 �Lf this combined reagent was added to 200 �L of soil water sus-ension. The absorbance was measured at 720 nm after 15 min of

ncubation at ambient temperature against appropriate blanks andhosphate standards.

Finally, potassium content in the form of oxides (K2O) wasuantified in soil water extracts by measuring atomic absorptionSPECTRAA 55 – Varian spectrophotometer).

.4. Soil microbial analyses

Considering the sandy texture of H and M (respectively 79% and6% of sand) and the lack of soil particles aggregation, all soil sam-les collected from planted pots after the cultivation period wereonsidered as “rhizosphere” soils contrary to “non rhizosphere”nes in unplanted pots. Subsamples (5 g) of soils per pot were mixedith 45 mL of 0.85% NaCl. Serial tenfold dilutions were performed,

anging from 10−2 to 10−5 for the enumeration of N2-fixing bacte-ia, from 10−2 to 10−7 for total culturable microflora and from 10−1

o 10−7 for nitrite oxidizing bacteria (NOB).For N2-fixing bacteria, 100 �L of each dilution was sprayed on

modified Rennie medium (Muratova et al., 2003) solidified with.5% noble agar (SIGMA). Numbers of colony forming unit (CFU) perram of dry soil was calculated according to the NF ISO 7218/A1ethod after 5 days of incubation (25 ◦C).Densities of total culturable microflora, PAH degrading bacte-

ia and NOB were estimated by the most probable number (MPN)ethod respectively described in detail by Binet et al. (2000) and

ipponen et al. (2004), using 96-well microtiter plates. Each wellas inoculated with 25 �L of soil water suspensions, including 24ells per dilution for total microflora and PAH degrading bacte-

ia and 12 wells per dilution for NOB. MPNs of bacteria per gramf dry soil were calculated according to the computer program ofugues and Plantat (1982) after appropriate incubation time andog-transformed.

Molecular analyses of bacterial community structure were per-ormed in complement of the aforementioned cultivation based

ethods. DNA was extracted from rhizosphere and non rhizo-phere soil samples and also from rhizoplane by a bead-beatingethod including a CTAB-vitamin based purification procedure as

escribed in (Técher et al., 2010). For rhizoplane analysis, 80 mg of

Ecology 62 (2012) 42– 49

fresh roots were carefully rinsed into distilled water and added toEppendorf tube containing DNA extraction buffer instead of 500 mgof soil sample. Universal primers targeting the 16S rDNA v3 vari-able region (Muyzer et al., 1993) were used to amplify fragmentsof about 200 bp which correspond to positions 341–534 in theEscherichia coli sequence (with a GC-clamp on the forward primer341F). PCR reactions followed by TGGE runs and polyacrylamidegel silver-staining were carried out as described in (Técher et al.,2010). TGGE runs were performed to compare the bacterial com-munity structures of unplanted soils, planted ones and rhizoplane.The banding patterns were analyzed for the presence/absence ofbands using the clustering method of the unweighted pair groupwith mathematical averages, i.e. UPGMA (Quantityone softwarev.4, BioRad). Bands of interest were excised, suspended into 100 �Lof sterile ultrapure water, reamplified using the same primer set(but without GC-clamp) and sequenced (GENOSCREEN). Sequenceswere compared with those deposited in the public domain of theRibosomal Database Project (Cole et al., 2006). Bacterial identifica-tions are provided into supplementary data.

2.5. Field trials

Field trials were conducted to evaluate the feasibility of M × Gcultivation in situ without any agricultural maintenance. For thispurpose, two areas of 16 m2 (4 m × 4 m) were delimited in eachof the two sites where the soil types M and H were respectivelycollected. 16 rhizomes were planted just before winter in eachof the delimited area, with a density of 1 rhizome m2. Those rhi-zomes emerged after winter but half of them had to be replacedbecause of freezing. Plant growth (stem length at the first intern-odes) was monitored during 18 weeks and distinction was madebetween “initially planted” and “replaced” rhizomes. Numbers ofnew sheets were also counted.

2.6. Statistical analyses

All data were expressed as mean ± standard deviation. Thedifferences between groups were determined by ANOVA forparametric distribution or Kruskal–Wallis for non parametric dis-tribution, followed by Student Newman–Keuls post hoc tests(p < 0.05).

3. Results and discussion

3.1. Plant growth

After 60 days of growth, a significantly higher shoot length wasreported in the reference soil compared to H and M (Fig. 1) reflect-ing the absence of pollutant. An increase in plant growth rate wasnoticed in H and M after two months, leading to similar shootlengths with control plants until the 80th day of experiment (Fig. 1).Then, plant development in H and in the reference soil involvedsignificantly taller stem lengths than that of individuals in M untilthe end of the experiment (Fig. 1). Taking into account the lowerconcentration of PAH in M (Table 2) but its higher heavy metaland cyanide contents (Section 2.1), the presence of these last con-taminants may have been detrimental to plant development. Asfor instance, excessive root absorption of trace element is knownto cause cellular disorders (production of reactive oxygen species(ROS), blocking of functional groups in biomolecules) and tissueinjuries leading to a decrease in plant growth (Roy et al., 2005;

Schützendübel and Polle, 2002). However, plants kept growing dur-ing all the cultivation period in the three soil types (Fig. 1) andneither necrosis nor chlorosis symptom could be observed on aerialparts (data not shown). According to the work of Scebba et al.

D. Techer et al. / Applied Soil Ecology 62 (2012) 42– 49 45

Table 1Evaluation of biomass (dry weight), hydration rate and chlorophyll content of plants grown in the reference soil and in polluted soils H and M; ch.a: chlorophyll a content,ch.b: chlorophyll b content; data represent the mean of five replicate pots per treatment ± standard deviation of the mean.

Physiological parameters Soil types Rhizomes Roots Stems Leaves

Dry weight (g) Reference soil 2.7 ± 0.6 3.5 ± 2.7 1.3 ± 0.1 2.3 ± 0.3H soil 1.2 ± 0.7b 1.7 ± 1.0 1.2 ± 0.3 1.6 ± 0.3b

M soil 1.5 ± 0.4b 1.8 ± 0.4 0.7 ± 0.1b 1.3 ± 0.5b

Hydration rate (%) Reference soil 62 ± 8 43 ± 7 67 ± 4 41 ± 14H soil 65 ± 3 55 ± 4b 68 ± 3 61 ± 11M soil 66 ± 3 60 ± 4b 68 ± 7 57 ± 7

Chlorophyll content(mg g−1 fresh leaves)

Reference soil ch.a: 1.9 ± 0.4; ch.b: 0.5 ± 0.1H soil ch.a: 1.6 ± 0.2b; ch.b: 0.4 ± 0.1b

M soil ch.a: 1.4 ± 0.4b,a; ch.b: 0.4 ± 0.1b,a

(ogwc

3

otlaePaaiicrcb

TCr

d

a Significant difference compared to plants from H.b Significant difference compared to plants from the reference soil.

2006), plants may have developed resistance mechanisms to abi-tic stressors (induction of ROS-scavenging enzymes like catalases,uaiacol and ascorbate peroxidases and superoxide dismutase) thatould have allowed for a growth adaptation to the various soil

onditions.

.2. Biomass

Below-ground biomass production showed a significant devel-pment of rhizomes for plants in the reference soil compared tohose in H and in M (Table 1), whereas root production was simi-ar (Table 1). This last parameter is of great importance to ensuren efficient soil volume exploration which could favor plant nutri-nt acquisition and provide elevated root surface exchanges withAH-degrading microbes (and subsequent biostimulation of pollut-nt degradation) (Gerhardt et al., 2009). The highest leaf biomassnd chlorophyll contents (ch.a and ch.b) were also measured fromndividuals originated from the reference soil (Table 1). Consider-ng that chlorophyll systems are key components in energy and

arbon fixation for photosynthetic plants (Peter et al., 1999), thoseesults may explain the higher plant growth in the reference soilompared to H and M. However, hydration rates of both aerial andelow-ground parts of plants grown in H and in M were similar to

able 2oncentrations of individual polycyclic aromatic hydrocarbons (PAHs) in polluted soils Heplicate pots per treatment ± standard error of the mean.

Rings PAH H (mg kg−1 dry soil)

t0 Unplanted

3 Acenaphthylene 1 ± 1 d.l

3 Acenaphthene 8 ± 7 11 ± 11

3 Fluorene 3 ± 1 1 ± 0.6a

3 Phenanthrene 23 ± 2 26 ± 3

3 Anthracene 23 ± 9 34 ± 3

Total 3 ring PAH 58 ± 20 72 ± 18

4 Pyrene 134 ± 11 115 ± 12

4 Benzo(a)anthracene 61 ± 6 48 ± 6

4 Chrysene 29 ± 2 17 ± 3b

Total 4 ring PAH 224 ± 19 180 ± 21

5 Benzo(b)fluoranthene 22 ± 9 23 ± 4

5 Benzo(k)fluoranthene 15 ± 4 9 ± 4

5 Benzo(a)pyrene 31 ± 2 20 ± 3b

Total 5 ring PAH 68 ± 15 52 ± 11

6 Indeno(1,2,3,cd)pyrene 14 ± 5 19 ± 4

Total PAH 364 ± 59 323 ± 54

.l, below detection limit.a Significant difference compared to other treatment.b Significant difference compared to t0.

that of individuals grown in the reference soil (with root hydrationbeing even higher in H and M, Table 1), indicating thus the phys-iological adaptation of Miscanthus × giganteus to the two pollutedsoils.

3.3. PAH dissipation

Initial pollutant concentrations varied from 364 mg totalPAH kg−1 dry soil for H to 26 mg total PAH kg−1 dry soil for M(Table 2). Soil contamination was dominated by the occurrence ofheavy molecular weight compounds of 4 aromatic rings and more(Table 2). Pyrene was the major pollutant in each type of soil:respectively 134 mg kg−1 dry soil for H and 12 mg kg−1 dry soil forM (Table 2). High molecular weight PAH are well-known for theirrecalcitrance to biodegradation because of their low bioavailability,particularly in soils like H and M characterized by a total organiccarbon content above 5% (Bossert and Bartha, 1986; Kästner, 2008;Shuttleworth and Cerniglia, 1995). Nevertheless, unexpected butsignificant dissipation of PAH was noticed in unplanted soils

compared to t0 and concerned anthracene (3 rings) for M, chrysene(4 rings) and benzo(a)pyrene (5 rings) for H (Table 2). A markeddissipation was even reported for fluorene (3 rings) in H com-pared to the initial soil concentration and to planted pots (Table 2).

and M, at t0 and after greenhouse experiments. Data represent the mean of five

M (mg kg−1 dry soil)

Planted t0 Unplanted Planted

d.l d.l d.l d.ld.l 4 ± 1 6 ± 4 1 ± 0.73 ± 0.2 d.l d.l d.l20 ± 2 2 ± 0.3 1± 0.3 1 ± 0.120 ± 8 3 ± 0.4 1 ± 0.3b 1 ± 0.2b

43 ± 10 9 ± 2 8 ± 5 3 ± 1

107 ± 3 12 ± 2 6 ± 2 6 ± 1b

42 ± 6 4 ± 2 4 ± 2 2 ± 114 ± 2b 1 ± 0.4 0.2 ± 0.2 d.l

163 ± 11b 17 ± 4 10 ± 4 8 ± 2b

29 ± 4 d.l d.l d.l8 ± 5 d.l d.l d.l28 ± 5 d.l d.l d.l

65 ± 14 – – –

16 ± 1 d.l d.l d.l

287 ± 36 26 ± 6 18 ± 9 11 ± 3

46 D. Techer et al. / Applied Soil Ecology 62 (2012) 42– 49

Fig. 1. Evolution of Miscanthus × giganteus growth depending on the soil type (ref-eci

Aoillpo2ooastlewrdoripdapomr

3

ntMdSezdratpa(

Fig. 2. TGGE patterns of 16S rDNA fragments after nucleic acid extractions follow-ing pot experiments in soil types H and M (t0: initially collected soil samples, NP:

rence soil and PAH-contaminated soils H and M), * indicates significant differenceompared to plants grown in the reference soil and “a” compared to plants grownn H.

ccording to the litterature, these decreases regardless the numberf aromatic rings could be attributed to “priming effects” followingnitial soil preparation. Indeed, soil collection and aeration fol-owed by sieving, mixing and watering during the preparation ofaboratory experiments may have enhanced the availability (withotential leaching) of some PAH and also the concomitant increasef degrading activity of competent soil microflora (Joner et al.,004). Taking these into account, such priming effects presumablyccurred in planted pots and took part in the significant dissipationf some PAH compared to t0 (acenaphthene and chrysene in H andnthracene and pyrene in M planted pots, Table 2). Besides, in theseoils, root exploration associated with the exudation of severalypes of organic acids may have prevented from any contaminanteaching while progressively enhancing PAH availability (Gaot al., 2010). Thus, priming effects and initial root developmentould have contributed to hide any conclusive rhizosphere effect

egarding PAH dissipation on the time scale of this study. However,espite the fact that there was no clear phytoremediation effect inur experimental conditions, the slight but significant dissipationeported only in planted soils compared to t0 when consider-ng the sum of 4-ring PAH (in bold in the Table 2) suggested aositive impact of growing Miscanthus × giganteus. The completeisappearance of some PAH in planted soils (acenaphthene in Hnd chrysene in M, Table 2) and the slight reduction observed inlanted pots for the sum of 3-ring PAH in contrast with unplantednes tended to confirm this assumption, suggesting that rhizore-ediation of PAH may be more marked and noticeable for a wider

ange of contaminant on a longer time scale of study.

.4. Microbiological analyses

Enumeration of total culturable microflora, PAH degraders, free-itrogen fixing and nitrite oxidizing bacteria, were respectively inhe range of 107, 104, 106 and 105 bacteria g−1 dry soil for soil types

and H (data not shown). No significant difference in bacterialensities could be noticed between unplanted and planted pots.uch results pointed out that the culture period of the greenhousexperiment was presumably not long enough to demonstrate a rhi-osphere effect of M × G on bacterial population densities. Theseata also suggested that the different screened culturable bacte-ial communities were initially widespread in soil types M and H,nd consequently well-adapted to the varying levels of pollution in

hese soils. Such a hypothesis could be explained by the selectiveressure that diffuse (but chronic) anthropogenic contaminationt industrial sites exerts upon indigene microbial communitiesJohnsen and Karlson, 2007).

unplanted pots, P: planted pots and R: rhizoplane; arrows indicate TGGE bandsspecific to rhizoplane).

To thoroughly investigate the impact of Miscanthus × giganteuson rhizosphere bacterial communities, molecular analysis of bacte-rial diversity were performed in soils and at the soil–root interface(rhizoplane) (Fig. 2). TGGE profiles of 16S rDNA amplified from soilsamples collected at t0 and after experiments (in unplanted andplanted pots) were similar for each type of soil (Fig. 2). In contrast,community profiles from rhizoplane were characterized by the pre-dominance of some specific bacterial phylotypes (corresponding toTGGE bands indicated by an arrow) that were not identified in soilsamples (Fig. 2). These observations indicated a shift in bacterialcommunity in the vicinity of roots, confirming that planting withM × G impacted the microbial community.

3.5. Soil leachates

To get insights into the potential of Miscanthus × giganteus culti-vation for industrial wasteland site restoration, several agronomicparameters were measured in soil leachates after plant growth

(Table 3). In H soil, no pH variation was observed after experimentwhereas a slight but significant acidification was noted in M, bothin unplanted and planted pots (Table 3). This pH decrease in M mayhave resulted in substantial trace element remobilization (Qureshi

D. Techer et al. / Applied Soil Ecology 62 (2012) 42– 49 47

Tab

le

3A

nal

yses

of

wat

er

soil

extr

acts

in

pol

lute

d

soil

s

H

and

M, a

t

t 0an

d

afte

r

gree

nh

ouse

exp

erim

ents

. Dat

a

rep

rese

nt

the

mea

n

of

five

rep

lica

te

pot

s

per

trea

tmen

t ±

stan

dar

d

dev

iati

on

of

the

mea

n.

Trea

tmen

tsp

H

(ult

rap

ure

wat

er)

Elec

tric

al

con

du

ctiv

ity

(�S

cm−1

)D

isso

lved

orga

nic

carb

on(m

g

g−1d

ry

soil

)N

O2

−(�

g

g−1d

ry

soil

)N

O3

−(�

g

g−1d

ry

soil

)K

2O

(�g

g−1d

ry

soil

)PO

4−

(�g

g−1d

ry

soil

)

HM

HM

HM

HM

HM

H

M

H

M

t 08.

06

±

0.03

8.04

±

0.03

a18

9

±

24a

407

±

3

4.5

±

0.1a

4.3

±

0.1b

6.6

±

1.4a

7.5

±

0.3

335.

1

±

21.1

a36

7.9

±

18.7

a4.

1

±

0.1a

4.7

± 0.

1a10

.4

±

2.9

1.0

±

0.3

Un

pla

nte

d

soil

8.20

±

0.10

7.89

±

0.09

101

±

6

432

±

31

4.0

±

0.1

3.8

±

0.1

0.9

±

0.2

6.0

±

1.8

93.5

±

30.5

144.

6

±

15.9

2.0

±

0.1

4.1

±

0.3

19.9

±

2.2c

1.3

±

0.5

Plan

ted

soil

8.14

±

0.13

7.86

±

0.05

94

±

11

501

±

50c

4.1

±

0.1b

4.1

±

0.5

1.1

±

0.2

5.6

±

1.3c

27.0

±

17.9

b11

.4

±

4.4b

1.5

±

0.2b

3.

8

±

0.3

19.9

±

1.4c

0.9

±

0.5

aSi

gnifi

can

t

dif

fere

nce

com

par

ed

to

the

oth

er

non

-mar

ked

trea

tmen

ts.

bSi

gnifi

can

t

dif

fere

nce

com

par

ed

to

un

pla

nte

d

trea

tmen

t.c

Sign

ifica

nt

dif

fere

nce

com

par

ed

to

t 0.

Fig. 3. Evolution of Miscanthus × giganteus plantation on the industrial wasteland

site corresponding to H area and M area; M i/H i: initially planted rhizomes (beforewinter), M r/H r: replaced rhizomes (after winter).

et al., 2004), enhancing soil electrical conductivity (Table 3). Withthis in mind, the observed shift in plant growth after 80 days ofplanting in M (Fig. 1) could be explained by a subsequent increaseof the soil solution toxicity.

For both type of soils, noticeable decreases in levels of dissolvedorganic carbon, nitrite, nitrate and potassium after experiments(Table 3) were presumably due to nutrient uptake by microflora andplants. However, carbon contents in H and M planted pots remainedsignificantly higher than in unplanted ones (Table 3), likely affectedby root deposits (Nguyen, 2003).

Dissolved inorganic phosphate concentrations increased signifi-cantly in H planted soil (Table 3). In rhizosphere soils, increased rootexudation of organic acids may enhance phosphate solubilization(Neumann and Mheld, 1999). Significant increases in the phos-phate soluble fraction were also noticed in H and M unplanted soils(Table 3). Additional investigations about the activity of other spe-cific functional groups of soil microorganisms involved in phosphorcycle may be required to get better insights into such observations(Velázquez et al., 2007).

3.6. Field trials

For the first series of plantation, an average shoot length of 30 cmwas measured for plants in H against 47 cm for plants in M soil(Fig. 3). These observations seemed contradictory to those obtainedduring laboratory experiments where plants grown in H were tallerthan those grown in M (Fig. 1). On the one hand, better agronomicproperties of the loamy M soil in comparison to the sandy H soilmay have favored plant development in situ. On the other hand,higher contents of soluble heavy metals and cyanides in M soil incomparison to H soil (mainly characterized by hydrophobic hydro-carbon contamination) may have lead to enhanced soil toxicitydue to pollutant remobilization during soil preparation (grinding,sieving) for laboratory experiments. Finally, the growth differenceinitially noticed between the two planted areas (M i/H i, Fig. 3)could be attributed to environmental conditions. Indeed, compar-isons between rhizomes that emerged the following spring seasonindicated a similar stem length of 45 cm for all plants from the twoplanted areas (Fig. 3). Plants in H area were more “dried” with frayedleaves (personal observations), thereby reflecting a more drasticlightening and wind exposition. Stem density also varied, with anaverage distribution of 3.06 sheets m−2 in M against 1.8 m−2 in H.This parameter, previously mentioned by Price et al. (2004) for Mis-

canthus × giganteus growth monitoring on agricultural soils, couldbe also a valuable tool for evaluation of growth performance inpolluted soils.

4 d Soil

4

atpettbgasnisinit

A

mMdF–

A

fj

R

A

A

A

A

A

B

B

BB

B

B

C

C

8 D. Techer et al. / Applie

. Conclusion and perspectives

The establishment of Miscanthus × giganteus rhizosphere wasssociated with the emergence of certain bacterial phylotypes athe soil–root interface. On the time scale of this study, no clearhytoremediation effect could be found in H and M soils. How-ver, when comparing the total 4-ring PAH contamination levels tohat determined at the beginning of the experiments, there was lit-le evidence that rhizoremediation would effectively take place inoth soils. Yet, that would require a longer time to be clearly distin-uished from priming effects. Concerning the agronomic impacts,

marked decrease in nitrate content was observed in rhizosphereoil whereas an increase in dissolved organic molecules indicatedon-negligible inputs of plant organic carbon that may take part

n the formation of the pool of soil organic matter. Finally, the firsteries of field trials confirmed the ability of M × G to develop onndustrially polluted soils, without any specific agricultural mainte-ance. Future monitoring of plant cultivation and soil remediation

n situ may pave the way for sustainable phytoremediation withhe production of a non-food biomass energy crop.

cknowledgements

This study was partially supported by the French govern-ental agency ADEME (Agence de l’Environnement et de laaîtrise de l’Energie), EDF (Electricite De France) and Communauté

’agglomérations Porte de France, Thionville. We gratefully thankranc ois Griffaton and Emilie Charrue from Lycée Agricole de Metz

Courcelles Chaussy for their technical assistance.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.apsoil.2012.07.009.

eferences

FNOR XP X33-012, 2000. Association Franc aise de Normalisation. Characterisationof Sludges—Determination of Polynuclear Aromatic Hydrocarbons (PAH) andPolychlorinated Biphenyls (PCB).

lexander, M., Clark, F.E., 1965. Nitrifying bacteria. In: Black, C.A. (Ed.), Meth-ods of Soil Analysis, Part 2. American Society of Agronomy, Madison, WI, pp.1477–1483.

prill, W., Sims, R.C., 1990. Evaluation of the use of prairie grasses for stimu-lating polycyclic aromatic hydrocarbon treatment in soil. Chemosphere 20,253–265.

rduini, I., Ercoli, L., Mariotti, M., Masoni, A., 2006a. Response of miscanthus to toxiccadmium applications during the period of maximum growth. Environ. Exp. Bot.55, 29–40.

rduini, I., Masoni, A., Ercoli, L., 2006b. Effects of high chromium applicationson Miscanthus during the period of maximum growth. Environ. Exp. Bot. 58,234–243.

ASOL, 2008. Base de données sur les sites et sols pollués (ou potentiellement pol-lués), http://basol.environnement.gouv.fr, Ministère de l’Ecologie, de l’Energie,du Développement durable et du Territoire, mis à jour 01.09.08.

inet, P., Portal, J.M., Leyval, C., 2000. Dissipation of 3-6-ring polycyclic aromatichydrocarbons in the rhizosphere of ryegrass. Soil Biol. Biochem. 32, 2011–2017.

lumer, M., 1976. Polycyclic aromatic compounds in nature. Sci. Am. 234, 35–45.ogan, B.W., Sullivan, W.R., 2003. Physicochemical soil parameters affecting seques-

tration and mycobacterial biodegradation of polycyclic aromatic hydrocarbonsin soil. Chemosphere 52, 1717–1726.

ossert, I.D., Bartha, R., 1986. Structure–biodegradability relationships of polycyclicaromatic hydrocarbons in soil. Bull. Environ. Contam. Toxicol. 37, 490–495.

ouchez, M., Blanchet, D., Vandecasteele, J.P., 1995. Degradation of polycyclicaromatic hydrocarbons by pure strains and by defined strain associations: inhi-bition phenomena and cometabolism. Appl. Microbiol. Biotechnol. 43, 156–164.

erniglia, C.E., 1992. Biodegradation of polycyclic aromatic hydrocarbons. Biodegra-

dation 3, 351–368.

ole, J.R., Chai, B., Farris, R.J., Wang, Q., Kulam-Syed-Mohideen, A.S., McGarrell,D.M., Bandela, A.M., Cardenas, E., Garrity, G.M., Tiedje, J.M., 2006. The ribosomaldatabase project (RDP-II): introducing myRDP space and quality controlled pub-lic data. Nucl. Acids Res. 35, D169–D172.

Ecology 62 (2012) 42– 49

Cunningham, S.D., Anderson, T.A., Schwab, A.P., Hsu, F.C., Donald, L.S., 1996. Phy-toremediation of soils contaminated with organic pollutants. Adv. Agron.,55–114.

Fernando, A.L., Godovikova, V., Oliveira, J.F.S., 2004. Miscanthus × giganteus: contri-bution to a sustainable agriculture of a future/present – oriented biomaterial.Materials Science Forum 455–456 (Advanced Materials Forum II, 455–456).

Gao, Y., Ren, L., Ling, W., Gong, S., Sun, B., Zhang, Y., 2010. Desorption of phenanthreneand pyrene in soils by root exudates. Bioresour. Technol. 101, 1159–1165.

Gerhardt, K.E., Huang, X.-D., Glick, B.R., Greenberg, B.M., 2009. Phytoremediationand rhizoremediation of organic soil contaminants: potential and challenges.Plant Sci. 176, 20–30.

Haritash, A.K., Kaushik, C.P., 2009. Biodegradation aspects of polycyclic aromatichydrocarbons (PAHs): a review. J. Hazard. Mater. 169, 1–15.

He, Y., Yediler, A., Sun, T., Kettrup, A., 1995. Adsorption of fluoranthene on soil andlava: effects of the organic carbon contents of adsorbents and temperature.Chemosphere 30, 141–150.

Hugues, B., Plantat, J.L., 1982. Calcul du nombre le plus probable et de sonintervalle de confiance dans le cas ou le nombre d’inoculums par dilutionest important (Calcul of the most probable number and its confidence limitwhen the number of inoculum per dilution is considerable). Chemosphere 11,1135–1140.

Inskeep, W.P., Bloom, P.R., 1985. Extinction Coefficients of Chlorophyll a and b inN,N-Dimethylformamide and 80% Acetone. American Society of Plant Biologists,Rockville, MD, ETATS-UNIS.

ISO 11269-2, 1995 International Standard Organization. Soil Quality—Determination of the Effects of Pollutants on Soil Flora—Part 2: Effects ofChemicals on the Emergence and Growth of Higher Plant.

Johnsen, A.R., Karlson, U., 2007. Diffuse PAH contamination of surface soils: environ-mental occurrence, bioavailability, and microbial degradation. Appl. Microbiol.Biotechnol. 76, 533–543.

Joner, E.J., Hirmann, D., Szolar, O.H.J., Todorovic, D., Leyval, C., Loibner, A.P., 2004.Priming effects on PAH degradation and ecotoxicity during a phytoremediationexperiment. Environ. Pollut. 128, 429–435.

Kästner, M., 2008. Degradation of aromatic and polyaromatic compounds. In: Rehm,H.-J., Reed, D.G. (Eds.), Biotechnology, second ed., pp. 211–239.

Kenealy, W.R., Jeffries, T.W., 2003. Rapid 2,2′-bicinchoninic-based xylanaseassay compatible with high throughput screening. Biotechnol. Lett. 25,1619–1623.

Lewandowski, I., Scurlock, J.M.O., Lindvall, E., Christou, M., 2003. The developmentand current status of perennial rhizomatous grasses as energy crops in the USand Europe. Biomass Bioenergy 25, 335–361.

Lipponen, M.T.T., Martikainen, P.J., Vasara, R.E., Servomaa, K., Zacheus, O., Kontro,M.H., 2004. Occurrence of nitrifiers and diversity of ammonia-oxidizing bacteriain developing drinking water biofilms. Water Res. 38, 4424–4434.

Muratova, A., Bner, T., Narula, N., Wand, H., Turkovskaya, O., Kuschk, P., Jahn, R.,Merbach, W., 2003. Rhizosphere Microflora of Plants Used for the Phytoreme-diation of Bitumen-contaminated Soil. Nature Publishing Group, Basingstoke,ROYAUME-UNI.

Muyzer, G., de Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbialpopulations by denaturing gradient gel electrophoresis analysis of polymerasechain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol.59, 695–700.

Neumann, G., Mheld, V., 1999. Root Excretion of Carboxylic Acids and Protons inPhosphorus-deficient Plants. Springer, Dordrecht, PAYS-BAS.

Nguyen, C., 2003. Rhizodeposition of organic C by plants: mechanisms and controls.Agronomie 23, 375–396.

Nichols, T., Wolf, D., Rogers, H., Beyrouty, C., Reynolds, C., 1997. Rhizosphere micro-bial populations in contaminated soils. Water Air Soil Pollut. 95, 165–178.

Peter, R., Evert, R.F., Eichhorn, S.E., 1999. Biology of Plants, 6th ed. W.H. Freeman andCompany, New York.

Price, L., Bullard, M., Lyons, H., Anthony, S., Nixon, P., 2004. Identifying the yieldpotential of Miscanthus × giganteus: an assessment of the spatial and tempo-ral variability of M. × giganteus biomass productivity across England and Wales.Biomass and Bioenergy 26, 3–13.

Qureshi, S., Richards, B.K., Steenhuis, T.S., McBride, M.B., Baveye, P., Dousset, S., 2004.Microbial acidification and pH effects on trace element release from sewagesludge. Environ. Pollut. 132, 61–71.

Roy, S., Labelle, S., Mehta, P., Mihoc, A., Fortin, N., Masson, C., Leblanc, R., Châteauneuf,G., Sura, C., Gallipeau, C., Olsen, C., Delisle, S., Labrecque, M., Greer, C.W., 2005.Phytoremediation of heavy metal and PAH-contaminated Brownfield sites. PlantSoil 272, 277–290.

Scebba, Arduini, F., Ercoli, I., Sebastiani, L., 2006. Cadmium Effects on Growth andAntioxidant Enzymes Activities in Miscanthus sinensis. Academia, Praha, TchequeRepublique.

Schützendübel, A., Polle, A., 2002. Plant responses to abiotic stresses: heavy metalinduced oxidative stress and protection by mycorrhization. J. Exp. Bot. 53,1351–1365.

Schwitzguébel, J.-P., van der Lelie, D., Baker, A., Glass, D., Vangronsveld, J., 2002. Phy-toremediation: European and American trends successes, obstacles and needs.J. Soils Sediments 2, 91–99.

Shuttleworth, K.L., Cerniglia, C.E., 1995. Environmental aspects of PAH biodegrada-

tion. Appl. Biochem. Biotechnol. 54, 291–302.

Técher, D., Martinez-Chois, C., D’Innocenzo, M., Laval-Gilly, P., Bennasroune, A.,Foucaud, L., Falla, J., 2010. Novel perspectives to purify genomic DNA fromhigh humic acid content and contaminated soils. Sep. Purif. Technol. 75,81–86.

d Soil E

T

V

rhizosphere soil and its growth promotion on black pepper (Piper nigrum L.)

D. Techer et al. / Applie

écher, D., Laval-Gilly, P., Henry, S., Bennasroune, A., Formanek, P., Martinez-Chois,

C., D’Innocenzo, M., Muanda, F., Dicko, A., Rejsek, K., Falla, J., 2011. Contribution ofMiscanthus × giganteus root exudates to the biostimulation of PAH degradation:an in vitro study. Sci. Total Environ. 409, 4489–4495.

elázquez, E., Rodríguez-Barrueco, C., Ramachandran, K., Srinivasan, V., Hamza,S., Anandaraj, M., 2007. Phosphate solubilizing bacteria isolated from the

cology 62 (2012) 42– 49 49

cuttings. In: First International Meeting on Microbial Phosphate Solubilization,Springer Netherlands, pp. 325–331.

Wilcke, W., 2007. Global patterns of polycyclic aromatic hydrocarbons (PAHs) insoil. Geoderma 141, 157–166.