Science of the Total Environment 421–422 (2012) 184–196

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Characterization of the bacterial archaeal diversity in hydrocarbon-contaminated soil

De-Chao Zhang, Christoph Mörtelmaier, Rosa Margesin ⁎Institute of Microbiology, University of Innsbruck, Technikerstrasse 25, A-6020 Innsbruck, Austria

⁎ Corresponding author. Tel.: +43 512 5076021; fax:E-mail address: Rosa.Margesin@uibk.ac.at (R. Marge

0048-9697/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.scitotenv.2012.01.043

a b s t r a c t

a r t i c l e i n f o

Article history:Received 4 October 2011Received in revised form 16 January 2012Accepted 17 January 2012Available online 3 March 2012

Keywords:BacteriaArchaeaSoilHydrocarbonsBiodegradationMicrobial diversity

A polyphasic approach combining culture-based methods with molecular methods is useful to expandknowledge on microbial diversity in contaminated soil.Microbial diversitywas examined in soil samples from a former industrial site in the EuropeanAlps (mainly usedfor aluminum production and heavily contaminated with petroleum hydrocarbons) by culture-dependent andculture-independent methods. The physiologically active eubacterial community, as revealed by fluorescence-in-situ-hybridization (FISH), accounted for 6.7% of the total (DAPI-stained) bacterial community. 4.4% and 2.0%of the DAPI-stained cells could be attributed to culturable, heterotrophic bacteria able to grow at 20 °C and10 °C, respectively. The majority of culturable bacterial isolates (34/48) belonged to the Proteobacteria (with apredominance of Alphaproteobacteria and Gammaproteobacteria), while the remaining isolates were affiliatedwith the Actinobacteria, Cytophaga–Flavobacterium–Bacteroides and Firmicutes. A high fraction of the culturable,heterotrophic bacterial population was able to utilize hydrocarbons. Actinobacteria were the most versatileand efficient degraders of diesel oil, n-alkanes, phenol and PAHs. The bacterial 16S rRNA gene clone librarycontained 390 clones that grouped into 68 phylotypes related to the Proteobacteria, Bacteroidetes, Actinobacteriaand Spirochaetes. The archaeal 16S rRNA gene library contained 202 clones and 15 phylotypes belonging tothe phylum Euryarchaeota; sequences were closely related to those of methanogenic archaea of the ordersMethanomicrobiales, Methanosarcinales, Methanobacteriales and Thermoplasmatales. A number of bacterial andarchaeal phylotypes in the clone libraries shared high similarities with strains previously described to beinvolved in hydrocarbon biodegradation.Knowledge of the bacterial and archaeal diversity in the studied soil is important in order to get a better insightinto the microbial structure of contaminated environments and to better exploit the bioremediation potential byidentifying potential hydrocarbon degraders and consequently developing appropriate bioremediation strategies.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Petroleum hydrocarbons are the most widespread contaminants inthe environment. The contamination of soil with high levels of hydro-carbons results in an increased soil organic carbon content, which –

depending on composition and concentration – may be utilized formicrobial growth or may be toxic to microorganisms (Bossert andBartha, 1984; Alexander, 1999; Maier et al., 2000). The impact of lowand high doses of environmental pollutants such as hydrocarbonscan range from stimulation to total inhibition of microorganisms(Ramakrishnan et al., 2010). The capacity of a broad spectrum ofmicro-organisms to utilize hydrocarbons as the sole source of carbon andenergy was the basis for the development of biological remediationmethods. The ability to degrade hydrocarbons is widespread amongsoil microorganisms. They may adapt rapidly to the contamination,as demonstrated by significantly increased numbers of hydrocarbondegraders after a pollution event (Margesin and Schinner, 2001; Greeret al., 2010).

+43 512 5072929.sin).

rights reserved.

Microbial community structures in hydrocarbon-contaminatedsoils are influenced by a number of factors, such as soil type, concen-tration and bioavailability of the contaminants, nutrient contents,temperature, oxygen content and pH (Margesin and Schinner, 2001;Greer et al., 2010). To evaluate soil microbial community compositionin contaminated soils, culture-dependent and culture-independentmethods have been used (Margesin and Schinner, 2005; Alonso-Gutierrez et al., 2009; Fabiani et al., 2009). Microbial abundance isoften based on culture-dependent methods. However, culturablecells may only represent less than 1% of the total microbial communityin an environment (Amann et al., 1995; Rappé and Giovannoni, 2003)and numerous bacteria enter a viable but non-culturable (VBNC) statein response to environmental stress (McDougald et al., 2009). There-fore, culture-independent, molecular assays, such as profiling soilDNA, rRNA, or phospholipid fatty acids, are increasingly used in envi-ronmental microbiology. Culture-independent approaches have beenclaimed to be more reliable for diversity analyses given the estab-lished cultivation methods favor the isolation of fast-growing micro-organisms (Felske et al., 1999). Direct recovery of bacterial 16S rDNAfrom soil theoretically represents the entire microbial populationfrom environmental samples (Spiegelman et al., 2005). However,

185D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

molecular methods also have their limitations, such as variable effi-ciency of lysis and DNA extraction and differential amplification oftarget genes (Kirk et al., 2004).

Studies on microbial community composition in contaminated al-pine soils have focused so far on the bacterial population (Margesinet al., 2003b; Labbé et al., 2007; Margesin et al., 2007), whereas infor-mation on the impact of Archaea is missing. In this study, we used acombination of culture-dependent and culture-independent methods(analysis of Bacteria and Archaea 16S rRNA gene clone libraries andfluorescence-in-situ-hybridization (FISH)) to investigate the microbi-al diversity in soil samples from an Alpine hydrocarbon-contaminatedindustrial site. Knowledge of the bacterial and archaeal diversity inthe studied soil is important in order to get a better insight into themicrobial structure of contaminated environments and to betterexploit the bioremediation potential by identifying potential hydro-carbon degraders and consequently developing appropriate bioreme-diation strategies. Since both traditional, culture-based, andmolecularmethods have their limitations (Kirk et al., 2004), a multi-technique(polyphasic) approach combining these methods is advantageous.

2. Materials and methods

2.1. Sampling site and soil analysis

Soil samples were collected from an industrial site in March 2008.The study site was located in the European Alpine region in Bozen/Bolzano, South Tyrol, Italy. It used to be a former industrial district,built in 1930 and mainly used for aluminum production. In the 70sthe production was reduced, and in 1990 the area was closed downand expropriated. Currently the area is not anymore used as industrialsite and unused. Storage tanks for heavy oil (formerly used for cheapenergy supply) were located at a depth of 1–3 m below ground sur-face. In 2008, the upper surface (0–4 m) was removed in the courseof a remediation treatment in order to lay open an area of approxi-mately 1000 m2 that was contaminated with hydrocarbons due topregressive leakage of heavy oil storage tanks. Our study area withinthis area had a size of approx. 100 m2.

Five composite soil samples (15 kg each; each obtained from5 sub-samples) were collected with the help of a bucket from the study areaand immediately transported to the laboratory. All soil samples weregently crumbled and sieved through an 8-mm screen in order to elim-inate rough materials, thoroughly mixed, and stored at field humidityin polyethylene bags at 4 °C until processing.

Each sample was examined (2–3 replicates) for physico-chemicaland microbiological (see Section 2.2) parameters. Physico-chemicalparameters (soil dry mass, soil carbonate content, soil pH in CaCl2, nu-trient content) were measured as described (Schinner et al., 1996).Hydrocarbon content in the range of C10 to C40 was determined bygas chromatography after extraction with heptane (DIN ISO 16703,modified according to ÖNORM EN 14309).

2.2. Enumeration of culturable aerobic soil bacteria

Numbers of culturable soil bacteria were determined by the plate-count method for viable cells on R2A agar containing cycloheximide(100 mg L−1) (Margesin et al., 2011). Hydrocarbon-utilizers werequantified on oil-agar containing diesel oil (Margesin and Schinner,1997). Colony-forming units (CFU) were counted after 14, 21 and28 days at 10 °C and 20 °C, respectively. All enumerations wereperformed with three replicates and CFUs were calculated on anoven-dry mass (105 °C) basis.

2.3. Phylogenetic analysis of culturable bacteria

Genomic DNA of 73 culturable bacterial strains, differing in phe-notypic characteristics (colony morphology, pigmentation, growth

characteristics) was extracted using the UltraClean Microbial DNAisolation kit (Mo Bio Laboratories). The 16S rRNA genes were ampli-fied as described earlier (Zhang et al., 2010a). Strains having >98%16S rRNA gene sequence similarity and matching the same GenBanksequence were assigned to the same phylotype (see also Section 2.10.).

2.4. Characterization of culturable bacteria

The 48 strains that were identified as potentially unique (seeSection 3.5) were characterized with regard to their growth temper-ature range, degradation abilities and enzyme activities.

2.4.1. Growth temperature rangeSuspensions of bacterial cells (pre-grown on R2A agar plates at

15 °C) in 0.9% NaCl were used to inoculate R2A agar plates that wereincubated at 1, 5, 10, 15, 20, 25, 30 and 37 °C, using two replicatesper strain and temperature. Growth was monitored up to an incuba-tion time of 7–21 days.

2.4.2. Utilization of aliphatic and aromatic hydrocarbons for growthSuspensions of bacterial cells in 0.9% NaCl were used to inoculate

mineral medium (without yeast extract) agar plates (Margesin andSchinner, 1997) that contained one of the following hydrocarbonsas the sole source of carbon: diesel oil (30 μL per plate), phenol(2.5 mM), naphthalene, anthracene, pyrene (2 and 10 mg of eachcompound per plate, dissolved in acetone). Inoculated plates withouthydrocarbons as well as sterile hydrocarbon-containing mediumserved as negative controls. Two replicates were used for all experi-ments. Plates were incubated up to 14–21 days at 15 °C and growthwas monitored regularly.

Diesel-oil utilizing strains were further tested in liquid culture(mineral medium) on their ability to degrade n-alkanes (C16, C20 andC28; each 800 mg L−1) and high amounts of diesel oil (3800 mg L−1).After 14–21 days at 15 °C, the residual hydrocarbon concentration wasmeasured by gas chromatography after extraction with heptane (DINEN ISO 9377-2, modified). Strains able to utilize phenol on agar plateswere grown in liquid culture at 15 °C in the samemediumwith increas-ing phenol concentrations in order to determine the highest amount ofphenol that could be degraded by these strains (Margesin et al., 2003a).

2.4.3. Screening for enzyme activitiesAmylase, protease, cellulase and esterase–lipase activities were

tested as described (Margesin et al., 2003a; Gratia et al., 2009) onR2A agar supplemented with starch, skim milk (each compound0.4% w/v), carboxymethylcellulose and trypan blue (0.4% and 0.01%w/v, respectively) or Tween 80 and CaCl2 (0.4% v/v and 0.01% w/v,respectively). Plates were evaluated after 7 days at 15 °C.

2.5. Total bacterial counts

Total counts of bacteria were determined in the filtrates of soilsuspensions by non-selective DAPI (4′,6′-diamino-2-phenylindole;1.5 μg mL−1) staining using CitiDAPI (2.5 μg DAPI mL−1, Citifluor™AF1 antfading; Electron Microscopy Sciences) and epifluorescencemicroscopy as described below (see Section 2.6.).

2.6. Fluorescence-in-situ-hybridization (FISH)

FISH analysis was done as described by Margesin et al. (2011)according to a procedure after Bertaux et al. (2007), using Nycodenz-based cell extraction (Barra Caracciolo et al., 2005). The oligonucleotideprobe EUB338 (Eubacteria; Stahl and Amann, 1991) was used to quan-tify Eubacteria (Loy et al., 2007). The probe NONEUB (Wallner et al.,1993) was used as a negative control. The probes EUB338 andNONEUB were Cy-3 labeled. Since soil particles interfered due to auto-fluorescence with counting by automated methods (Kobabe et al.,

186 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

2004; Li et al., 2004), counting was done manually using epifluores-cence microscopy (Leica DM5000B) equipped with a digital cameraby counting at least 30 images per hybridization approach. Bacterialcell numbers were calculated on an oven-dry mass (105 °C) basis(Schinner et al., 1996).

2.7. Total community DNA extraction from soil and PCR amplification of16S rRNA genes

Total community DNA was extracted with two replicates from 10 gof the soil samples using the PowerMax soil DNA isolation kit (Mo BioLaboratories) according to the manufacturer's instructions and a com-posite extract was prepared. Extracted community DNA was purifiedusing polyvinylpolypyrrolidone spin columns (Whyte andGreer, 2005).

Bacterial 16S rRNA genes were amplified from the purified soilsample DNA extract by PCR using Bacteria-specific primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) (Dorsch and Stackebrandt, 1992) and758R (5′-CTACCAGGGTATCTAATCC-3′) (Stackebrandt and Charfreitag,1990). Archaeal 16S rRNA genes were amplified by a semi-nest PCRreaction using Archaea-specific primers as described by Steven et al.(2008). The presence of the appropriately sized PCR fragment (731 bpfor bacterial 16S rRNA and 590 bp for archaeal 16S rRNA) was con-firmed by agarose gel electrophoresis (0.8% agarose, 1X TAE buffer).

2.8. Construction of 16S rRNA gene clone libraries and restrictionfragment length polymorphism (RFLP)

PCR products were purified using the Gene JET™ PCR purification kit(Fermentas) and the cleaned PCR products were cloned in pGEM-T vec-tors (Promega) according to the manufacturer's instructions. PlasmidDNA was transformed into Escherichia coli JM109 (Promega) with astandard transformation protocol. Plasmids containing the ligated 16SrRNA gene inserts were identified using blue/white screening on LBagar containing ampicillin (100 μg mL−1). These plates were surfacespread with 4 μL of filter-sterilized 1 M isopropyl-β-D-thiogalactopyra-noside (IPTG) and 40 μL of 0.05 M 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside (X-Gal; prepared in N,N-dimethyl formamide).

Plasmid DNA from candidate-positive colonies was isolated andused as a template for reamplification using primers M13F and M13R.The amplified 16S rRNA genes were re-checked by agarose gel electro-phoresis (0.8% agarose, 1× TAE buffer).

Amplified 16S rRNA genes were restricted using the enzymes RsaIand HhaI (Invitrogen) at 37 °C overnight. Restriction digests were ana-lyzed by agarose gel electrophoresis (2% agarose, 0.5× TBE buffer).Unique restriction patternswere identified visually and representativesof each restriction pattern were used as a template for 16S rRNA genesequencing. Sequencing reactions were carried out by Eurofins MWGOperon (Ebersberg, Germany).

2.9. Phylogenetic and statistical analyses of 16S rRNA gene sequences

The 16S rRNA gene sequences were submitted for comparison andidentification to the GenBank databases using the NCBI Blastn algo-rithm, the EMBL databases using the Fasta algorithm and the RibosomalDatabase Project II (RDP) using its Sequence Match (Cole et al., 2005).Sequences that demonstrated strong homology were then aligned toreference sequences using Clustal X1.8 (Thompson et al., 1997) andphylogenetic analysis was performed using the MEGA 4.0 software(Tamura et al., 2007). Phylogenetic trees were constructed using aneighbor-joining algorithm with the Kimura's two-parameter model(Kimura, 1980); bootstrap valueswere performedwith 1000 replicates.

Sequences having >98% similarity and matching the same Gen-Bank sequence were assigned to the same phylotype.

The coverage of the clone library was calculated using the formula[1−(n/N)]×100, where n is the number of phylotypes appearingonly once in a library and N is the total number of sequences analyzed

(Good, 1953). The Shannon index (H′) of diversity, the reciprocal ofSimpson's reciprocal index of diversity (1/D), and the bias-correctedChao1 richness estimator index (Chao, 1984) were determined withEstimates S 7.5. (http://viceroy.eeb.uconn.edu/EstimateS). Evenness(the relative abundance of each phylotype) was calculated with theformula E=−eH′/N, where H′ is the Shannon index of diversity andN is the total number of phylotypes in a library (Krebs, 1989).

2.10. Nucleotide sequence accession numbers

All sequences generated in this study were deposited in theGenBank database. Accession numbers for isolates of the domain Bac-teria were GQ161991, FJ948107, GQ240227, GQ246952, GQ246953,GQ240228, GQ131578, FJ972171, GQ131577, GQ131579, GQ161990and HQ588828–HQ588864. Accession numbers for Bacteria-relatedclones were HQ603923–HQ603990, and for Archaea-related cloneswere HQ588813–HQ58827.

3. Results

3.1. Soil properties

The soil samples were a mixture of gravel, sand and clay; the C-horizon was predominantly porphyry, with low amounts of dolomite.The soil pH (CaCl2) was 6.0–6.5. Contents of soil nutrients were low(2.6 mg NH4–N kg−1 soil, 4.5 mg NO3–N kg−1 soil, b1 mg P kg−1

soil) and carbonate content was 3–4%. At the time of sampling (March2008), the mean soil temperature in the sampling area was 8–10 °C.

The soil contained 13,300 mg hydrocarbons/kg dry soil. 40% and60% of this contamination consisted of C10–C20 and C20–C40 hydrocar-bons, respectively, which points to a high content of heavy oils. Thewater-extractable fraction of the hydrocarbon contamination con-tained 0.04 mg hydrocarbons L−1 soil eluate.

3.2. Total bacterial counts and FISH

DAPI-staining of the soil sample revealed total bacterial counts of(1.2±0.1)×109 cells g−1 dry soil. The fraction of cells detectable withthe FISH-probe EUB338 (specific for members of the domain Bacteria;(7.8±1.6)×107 cells g−1 dry soil) represented (6.7±1.2)% of DAPI-stained cells.

3.3. Enumeration of culturable bacteria

The enumeration of viable heterotrophic and hydrocarbon-degrading bacteria revealed (5.2±2.1)×107 heterotrophic cfu g−1

dry soil at 20 °C. At 10 °C, still (2.4±1.2)×107 heterotrophic cfu g−1

soil were found, which corresponds to 46% of the heterotrophsfound at 20 °C. A considerable fraction of the heterotrophic population(25–28%) was able to utilize hydrocarbons at the two temperaturestested. 4.4% and 2.0% of the DAPI-stained cells could be attributed toheterotrophic bacteria able to grow at 20 °C and 10 °C, respectively.Similarly, 67% and 31% of the FISH-detected eubacterial communitywas able to grow on R2A medium at 20 and 10 °C, respectively.

3.4. Characterization of culturable bacteria

All investigated 48 strains could grow at temperatures ranging from10 to 30 °C. 31 strains (65%) showed no growth at 37 °C; all culturablestrains that belonged to the CFB group in this study (see Section 3.5.)were among these strains. 5 and 11 strains (11 and 23%) could notgrow at temperatures below 10 °C (5 or 1 °C, respectively).

13/48 (27%) and 11/48 (23%) strains produced protease or esterase–lipase, respectively, at 15 °C, and 8 strains (17%) produced amylaseor CM-cellulase activity. No relation between enzyme production andtaxonomic affiliation could be recognized.

187D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

The ability of the strains investigated to utilize a number of hydro-carbons for growth at 15 °C is shown in Table 1; 16 strains (33% of allstrains tested) did not utilize any of the compounds tested and are notincluded in this table. 9 strainswere able to utilize all compounds test-ed on agar plates at 15 °C; among these strains were 4 Actinobacteria(3 Rhodococcus), 1 Firmicutes representative (Bacillus), and 2 repre-sentatives each of the Gammaproteobacteria (2 Serratia) and the CFBgroup (both of them were representatives of recently describednovel species of the genera Dyadobacter and Hymenobacter; Zhanget al., 2010c, 2011a). In general, Actinobacteria turned out to bethe most versatile and efficient hydrocarbon utilizers.

Strains that were able to degrade diesel oil (3800 mg L−1) in liquidculture (strains BZ4, BZ17, BZ22, BZ 91; 3 Rhodococcus, 1 Pimelobacter)also degraded C16 and C20. A low amount of C28 (10–13%)was degradedby 2 Rhodococcus strains (BZ17, BZ22). Phenol could be fully degradedin liquid culture at amounts of 10 mM (strains Arthrobacter BZ73 andPimelobacter BZ91), 7.5 mM (strain Rhodoccoccus BZ4) or 5 mM (strainPseudoxanthomonas BZ60). Strain Rhodoccoccus BZ4 was additionallyable to degrade high amounts (50 mg L−1) of anthracene and pyrenein liquid culture (data not shown).

3.5. Identification of bacterial isolates

Based on phenotypic characteristics (colony morphology, pigmen-tation, growth properties) and 16S rRNA gene sequencing, 48/73strains were identified as potentially unique (b98% sequence similar-ity) and further studied. The majority of these strains (34/48; 71%)belonged to the phylum Proteobacteria (Fig. 1), with a predominanceof the two classes Alphaproteobacteria and Gammaproteobacteria(each 15/34), while Betaproteobacteria were only present in lowamounts (4/34).

Table 1Utilization of hydrocarbons for growth by culturable bacterial strains (2=good growth; 1=not shown. Strains marked with an asterisk (*) were described as novel species.

Strain Nearest phylogenetic neighbor Phylum/class Utiliz

Diese

30 μL

BZ4 Rhodococcus erythreus DSM 43066T Actinobacteria 2BZ17 Rhodococcus cercidiphyllus DSM 45141T Actinobacteria 2BZ22 Rhodococcus cercidiphyllus DSM 45141T Actinobacteria 2BZ91 Pimelobacter simplex DSM 20130T Actinobacteria 2BZ69 Bacillus idrienses SMC 4352-2T Firmicutes 2BZ56 Serratia proteamaculans DSM 4543T Gammaproteobacteria 2BZ65 Serratia quinivorans DSM 4597T Gammaproteobacteria 1BZ26 Dyadobacteria psychrophilus (sp. nov.)⁎ CFB 2BZ33r Hymenobacter psychrophillus (sp. nov.)⁎ CFB 2BZ3 Rhizobium daejeonense KCTC12121T Alphaproteobacteria 2BZ44 Roseomonas mucosa MDA5527T Alphaproteobacteria 2BZ13 Sphingobium xenophagum BN6T Alphaproteobacteria 2BZ21 Pseudomonas tremae ATCC 43299T Gammaproteobacteria 2BZ73 Arthrobacter sulfureus DSM 20167T Actinobacteria 1BZ6 Rhizobium selenireducens B1T Alphaproteobacteria 1BZ82 Rhizobium cellulosilyticus ALA10B2T Alphaproteobacteria 1BZ2 Sphingopyxis chilensis DSM 14889T Alphaproteobacteria 1BZ5 Rhizobium selenireducens B1T Alphaproteobacteria 1BZ59 Castellaniella defragrans 54PinT Betaproteobacteria 1BZ19 Rheinheimera soli BD-D46t Gammaproteobacteria 1BZ93 Pseudomonas bauzanensis (sp. nov.)⁎ Gammaproteobacteria 1BZ27 Pseudomanas congelans DSM 14939T Gammaproteobacteria 1BZ49 Serratia grimesii ATCC 14460T Gammaproteobacteria 1BZ55 Pseudomanas argentinensis CH01T Gammaproteobacteria 1BZ64 Pseudomonas fluorescens KCTC 296T Gammaproteobacteria 1BZ70 Serratia grimesii ATCC 14460T Gammaproteobacteria 1BZ43 Microbacterium deminutum KV-483T Actinobacteria 0BZ41 Agromyces bauzanensis (sp. nov.)⁎ Actinobacteria 0BZ88 Microbacterium foliorum DSM 12966T Actinobacteria 0BZ31r Roseomonas vinaceus DSM 19362T Alphaproteobacteria 0BZ78 Tistrella bauzanensis (sp. nov.)⁎ Alphaproteobacteria 0BZ60 Pseudoxanthomonas spadix DSM 18855T Gammaproteobacteria 0

The genus Rhizobium (4 strains) dominated among Alphaproteo-bacteria; other representatives of this class were related to the generaRoseomonas (2), Tistrella (1), Sphingopyxis (2), Sphingobium (1),Porphyrobacter (1), Caulobacter (1) and Brevundimonas (3). Threestrains (BZ2, BZ13 and BZ30) were closely related to aromatichydrocarbon-degrading species of the family Sphingomonadaceae.For example, strain BZ13 showed 98.8% 16S rRNA gene sequencesimilarity to Sphingomonas xenophaga DSM 6383T, which is able todegrade xenobiotic aromatic compounds (Stolz et al., 2000). StrainBZ35 had a high sequence similarity (99.1%) to Porphyrobactersanguineus IAM 12620T, which is capable of degrading biphenyl anddibenzofuran (Hiraishi et al., 2002).

Betaproteobacteria were represented by the genera Janthinobacter-ium (3) and Variovorax (1). The dominating genera among Gammapro-teobacteria were Pseudomonas (6), Serratia (4) and Luteimonas (3),while only one representative of the genera Rheinheimera and Pseudox-anthomonaswas found.

Themajority of the remaining 14/48 strains were related to bacteriafrom the phylum Actinobacteria (9/48; 19%), with representatives of thegenera Rhodococcus (4),Dietzia (1), Arthrobacter (1), Agromyces (1) andMicrobacterium (2). Only few representatives of the phyla Cytophaga–Flavobacterium–Bacteroides (CFB) group (3/48; genera: Dyadobacter,Hymenobacter and Pedobacter) and Firmicutes (2/48; genus: Bacillus)were detected (Fig. 2).

Among the 48 strains sequenced in this study, 8 strains could beidentified as novel bacterial species. The majority of them belongedto the phylum Proteobacteria with two representatives each of theclasses Alphaproteobacteria (Sphingopyxis bauzanensis: type strainBZ30T, Zhang et al., 2010a; Tistrella bauzanensis: type strain BZ78T,Zhang et al., 2011c) and Gammaproteobacteria (Luteimonas terricola:type strain BZ92rT, Zhang et al., 2010b; Pseudomonas bauzanensis:

weak growth; 0=no growth). Strains unable to utilize any of the compounds tested are

ation of hydrocarbons on agar plates at 15 °C

l oil Phenol Naphtalene Anthracene Pyrene

2.5 mM 2 mg 10 mg 2 mg 10 mg 2 mg 10 mg

2 2 2 2 2 2 12 2 2 2 2 0 02 2 2 2 2 1 12 2 2 2 2 1 02 2 2 2 2 2 02 2 2 1 1 1 02 2 2 1 1 0 02 2 2 2 2 2 01 2 2 2 2 2 22 2 2 1 1 1 12 1 1 1 1 1 01 0 0 0 0 0 02 0 0 0 0 0 02 1 1 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 00 0 0 0 0 0 00 0 0 0 0 0 02 1 1 1 1 1 02 0 0 0 0 0 02 2 1 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 02 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 01 0 0 0 0 0 02 1 1 0 0 0 0

BZ21 (HQ588832)BZ27 (HQ588846)

Pseudomonas congelans DSM 14939 (AJ492828)BZ64 (HQ588845)BZ55 (HQ588844)BZ77 (HQ588840)Pseudomonas argentinensis CH01 (AY691188)

Pseudomonas bauzanensis BZ93T (GQ161991)Pseudomonas denitrificans IAM12023 (AB021419)

BZ19 (GQ240227)Rheinheimera soli BD-d46 (EF575565)

BZ65 (HQ588852)Serratia quinivorans DSM 4597(AJ233435) BZ56 (HQ588837)BZ70 (HQ588839)BZ49 (HQ588851)

BZ60 (HQ588838)Pseudoxanthomonas spadix IMMIBAFH-5 (AM418384)

BZ39 (HQ588836)BZ29r (HQ588834)

Luteimonas terricola BZ92rT (FJ948107)Luteimonas mephitis B1953/27.1 (AJ012228)

BZ15 (HQ588854)Variovorax boronicumulans BAM-48 (AB300597)BZ31w (HQ588842) Janthinobacterium agaricidamnosum W1r3 (Y08845)

BZ45 (GQ246952) BZ59 (GQ246953)

BZ31r (HQ588841)Roseomonas vinaceus CPCC100056 (EF368368)

BZ44 (HQ588850)Tistrella bauzanensis BZ78T (GQ240228)

Tistrella mobilis IAM 14872 (AB071665)BZ2 (HQ588828)

Sphingopyxis bauzanensis BZ30T (GQ131578)Sphingopyxis witflariensis W-50 (AJ416410)BZ13 (HQ588831)

Sphingobium xenophagum BN6 (X94098)BZ35 (HQ588835)Porphyrobacter sanguineus ATCC 25661 (AB062106)

BZ3 (HQ588847)BZ6 (HQ588848)

BZ5 (HQ588853)BZ82 (HQ588849)Rhizobium galegae 59A2 (AF025853)

BZ23 (HQ588833)Caulobacter henriciiATCC15253 (AJ227758)

BZ11 (HQ588830)BZ38 (HQ588843)BZ10 (HQ588829)Brevundimonas alba DSM4736 (AJ227785)

Bacillus thuringiensis 1 (DQ105966)

100

100

100

100

5458

64

100

100

63100

100

100

99

96

100

84

100

90

100

78100

100

46100

99

100

100

99

100

95

100

60

53

98

76

100

100

53

100

99 100

10094100

59100

5860

0.02

Alphaproteobacteria

Betaproteobacteria

Gam

maproteobacteria

Fig. 1. Phylogenetic relationships of 34 culturable bacterial strains isolated in this study (BZ) and related to the Proteobacteria. The tree was inferred by neighbor-joining analysis.Bootstrap values (%) are based on 1000 replicates and are shown for branches with more than 50% support. GenBank accession numbers of 16S rRNA sequences are given inparentheses. Bar, 0.2% sequence divergence. Strains isolated in this study are indicated in bold and novel species (already described) are highlighted.

188 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

BZ4 (HQ588862)Rhodococcus erythreus DSM 43066 (X79289)

BZ91 (HQ588857)Rhodococcus yunnanensis YIM70056 (AY602219)

BZ17 (HQ588858)

BZ22 (HQ588861)BZ84 (HQ588860)Dietzia schimae YIM65001 (EU375845)

BZ73 (HQ588859)Arthrobacter sulfurous DSM 20167 (X83409)

Agromyces bauzanensis BZ41T (FJ972171)Agromyces humatus CD5 (AY618216)

BZ88 (HQ588856)BZ43 (HQ588855)

Microbacterium deminutum KV-483 (AB234026)

Bacillus arsenicus Cona/3 (AJ606700)BZ85 (HQ588864)

BZ69 (HQ588863)Bacillus idriensis SMC4352-2 (AY904033)

Dyadobacter psychrophilusBZ26T (GQ131577)Dyadobacter koreensis KCTC 12534 (EF017660)

Runella slithyformis ATCC 29530 (M62786)

Hymenobacter psychrophilusBZ33rT (GQ131579)Hymenobacter aerophilus DSM 13606 (EU155008)

Pontibacter actiniarum KMM6156 (AY989908)

Pedobacter bauzanensis BZ42T (GQ161990)

Pedobacter terricola DS-45 (EF446147)Pedobacter terrae DS-57 (DQ889723)

Sulfolobus acidocaldarius DSM 639 (D14053)

100

100

88100

100

100

100

99

86

100

100

69

100100

100

100

100

99

100 100

100

100100

100

10064

0.05

Actinobacteria

Firm

icutesC

FB

Fig. 2. Phylogenetic relationships of 14 culturable bacterial strains isolated in this study (BZ) and related to the CFB group, Actinobacteria and Firmicutes. The tree was inferred byneighbor-joining analysis. Bootstrap values (%) are based on 1000 replicates and are shown for branches with more than 50% support. GenBank accession numbers of 16S rRNAsequences are given in parentheses. Bar, 0.5% sequence divergence. Strains isolated in this study are indicated in bold and novel species (already described) are highlighted.

189D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

type strain BZ93T, Zhang et al., 2011b). Three of the novel speciesbelonged to the CFB-group (Dyadobacter psychrophilus: type strainBZ26T, Zhang et al., 2010c; Hymenobacter psychrophilus: type strainBZ33rT, Zhang et al., 2011a; Pedobacter bauzanensis: type strainBZ42T, Zhang et al., 2010d). Only one novel species could be assignedto Actinobacteria (Agromyces bauzanensis: type strain BZ41T, Zhanget al., 2010e).

3.6. Diversity of bacterial 16S rRNA gene sequences in the clone library

The bacterial clone library was composed of 390 clones thatgrouped into 68 phylotypes (Figs. 3–4), representing an estimatedcoverage of 93%. Five phylotypes (14 clones) could not be classifiedinto known phylogenetic groups by the RDP Classifier or by phyloge-netic tree branching. The remaining 63 of the 68 bacterial phylotypescould be divided into the following four phyla: Proteobacteria (330clones/390; 85% of the total clones), Bacteroidetes (35/390; 9%),Actinobacteria (7/390; 2%), Spirochaetes (4/390; 1%).

Twenty-one of the 68 Bacteria phylotypes showed >97% sequencesimilarity to 16S rRNA gene sequences in the public database, whichsuggests that these phylotypes have been previously described. Re-markably, 44 of the phylotypes had less than 95% sequence similarityto their closest relatives in the public databases, thus suggesting theymay represent novel bacterial genera. Three phylotypes may repre-sent novel bacterial species (b97% sequence similarity).

3.6.1. ProteobacteriaProteobacteria were highly dominant in the bacterial clone library

and comprised 49 different phylotypes with representatives from allfive classes (Alpha-, Beta-, Gamma-, Delta-, and Epsilon-Proteobacteria).Betaproteobacteria showed the highest diversity among the Proteobac-teria and consisted 20 phylotypes. The Betaproteobacteria dominance(210/390; 54% of the total clones) was largely due to the high propor-tions of phylotype Bac5 (69/390; 18%), phylotype Bac3 (48/390; 12%)and of phylotype Bac10 (41/390; 10%). The phylotype Bac5, represent-ing 17.6% of the universal bacterial library, was 98% identical to Acido-vorax sp. strain NA3 isolated from polycyclic aromatic hydrocarbon(PAH)-contaminated soil (Singleton et al., 2009). The phylotype Bac10had a very similar sequence to that of clone SBPostOx65 from PAH-contaminated soil (Richardson et al., 2011). The phylotype Bac3 was98% identical to obligately chemolithoautotrophic and sulfur-oxidizingbacteria (Thiobacillus sp. strain FTL9) isolated from Antarctic LakeFryxell (Sattley and Madigan, 2006).

The Deltaproteobacteria and Gammaproteobacteria phylotypesrepresented two other large groups among the Proteobacteria andhad the same proportion in the bacterial clone library (each 12% ofthe total clones). The dominating Deltaproteobacteria-related phylo-type Bac91 shared 99% 16S rRNA gene sequence similarity to clone5C54 from petroleum-contaminated sediments (Allen et al., 2007).The predominant phylotype within the Gammaproteobacteria wasphylotype Bac124, which showed 98% similarity to a bacterium

Fig. 3. Phylogenetic relationships of bacterial clone sequences and environmental 16S rRNA sequences related to the Gammaproteobacteria (Gamma), Betaproteobacteria (Beta),Alphaproteobacteria (Alpha), Deltaproteobacteria (Delta), Epsilonproteobacteria (Epsilon) and unclassified bacteria (Un). The tree was inferred by neighbor-joining analysis.Bootstrap values (%) are based on 1000 replicates and are shown for branches with more than 50% support. GenBank accession numbers of 16S rRNA sequences are given inparentheses. Bar, 0.5% sequence divergence. Clone sequences from this study with their prevalence (in brackets) in the clone library are indicated in bold.

190 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

Bac14(7) (HQ603975)Petroleum affected groundwater clone D88 (DQ423675)

Bac182(1) (HQ603976)Wastewater sludge digester clone QEDQ1CG08 (CU923071)

Bac79(2) (HQ603979)Bac-403(1) (HQ603988)

Bac44(9) (HQ603978)PenicillinG production wastewater clone B16 (EU234211)

Bac151(4) (HQ603977)Marathonas Reservoir clone A07-70-BAC (GQ340095)

Bac300(2) (HQ603986)Arctic meltwater clone SSIM-E2v (FJ946537)

Bac180(4) (HQ603981)Oil field clone G-9 (FJ900971)

Bac324(1) (HQ603987)Hydrocarbon-contaminated soil clone CM3D11 (AM936244)

Bac164(4) (HQ603984)Sulfate-reducing bioreactor clone ET10-25 (DQ443981)Bac320(1) (HQ603990)Petroleum-contaminated aquifer clone 5S32 (DQ664013)

Bac1(4) (HQ603974)Heavy oil seeps clone 91-14 (EF157111)Bac229(3) (HQ603983)Tar oil-contaminated aquifer clone LA10Ba08 (GU133260)

Bac251(1) (HQ603985)Natural Asphalt Lake clonePL-B1_5_2_D10 (GU120595)Bac227(1) (HQ603982)

Tar Ponds sediment clone STPCL-11 (HM124398)Bac286(1) (HQ603989)Spirochaeta Africana Z-7692 (X93928)Bac178(3) (HQ603980)

Chemical industrial anaerobic digestor clone1 (FJ462035)

Sulfolobus acidocaldarius DSM 639 (D14053)

100

100

94100

100

100

100

100

69100

100

100

100

10099

55

100

9996

10097

93100

99100

86

100

0.05

CF

BSpi

Un

Un

Act

Fig. 4. Phylogenetic relationships of bacterial clone sequences and environmental 16S rRNA sequences related to the CFB group, Actinobacteria (Act), Spirochaetes (Spi) and unclas-sified bacteria (Un). The tree was inferred by neighbor-joining analysis. Bootstrap values (%) are based on 1000 replicates and are shown for branches with more than 50% support.GenBank accession numbers of 16S rRNA sequences are given in parentheses. Bar, 0.5% sequence divergence. Clone sequences from this study with their prevalence (in brackets) inthe clone library are indicated in bold.

191D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

(Thermomonas sp. strain S47) found in sewage sludge (Phuong et al.,2009).

The Alphaproteobacteria represented 5% of the bacterial clones(20/390) and consisted of 8 phylotypes; their sequences were similarto those of culturable bacteria and were affiliated to the generaPelagibius, Tetracoccus, Porphyrobacter, Caulobacter, Sandaracinobacterand Brevundimonas.

The Epsilonproteobacteria represented 1% of the bacterial clones(4/390) and the two phylotypes (bac179 and bac336) were relatedto the sulfur-oxidizing bacterium Sulfuricurvum kujiense and to clone1068 (unculturable) from petroleum-contaminated groundwater(Watanabe et al., 2000), respectively.

3.6.2. Cytophaga–Flavobacterium–Bacteroides (CFB)The CFB group represented 9% of the clones and consisted of 10 phy-

lotypes. The two dominating CFB group-related phylotypes (phylotypes44 and 14)were closely related to clone B16 found inwastewater and toclone H2-2D.88 from petroleum-affected groundwater, while phylo-type Bac324 was very closely related (99% sequence identity) to cloneCM3D11 from a pilot-scale bioremediation process of a hydrocarbon-contaminated soil (Militon et al., 2010).

3.6.3. SpirochaetesThe lowest fraction of the bacterial clone library was affiliated to

the phylum Spirochaetes (1%). Both Spirochaetes phylotypes (Bac268and Bac178) were classified into the family Spirochaetaceae.

3.6.4. ActinobacteriaA low fraction of the bacterial clone library was affiliated to the phy-

lum Actinobacteria (2%). The two phylotypes (Bac1 and Bac229) wererelated to sequences obtained from heavy oil seeps (Kim and Crowley,2007) and from tar-oil-contaminated aquifer sediment (Winderl et al.,2010), respectively.

3.7. Diversity of archaeal 16S rRNA gene sequences in the clone library

202 Archaeal 16S rRNA gene clones were grouped into 15 phylo-types, representing an estimated coverage of 96%. Phylogenetic analy-sis revealed that all 15 phylotypes fell into the phylum Euryarchaeota(Fig. 5), while representatives of the phylum Crenarchaeota werenot found. Members of the Euryarchaeota were represented bysequences closely related to those of methanogenic archaea of theorders Methanomicrobiales (Methanoculleus spp. and Methanocorpus-culum spp.), Methanosarcinales (Methanosaeta spp. and Methanosar-cina spp.), Methanobacteriales and Thermoplasmatales. The 16S rRNAgene sequences of phylotypes related to the orders Methanomicro-biales and Methanosarcinales were highly similar to those of culturedmethanogenic archaea (e.g., Methanospirillum hungatei, Methanocul-leus palmolei, Methanosaeta thermophila and Methanosaeta concilii).The Methanobacteriales and Thermoplasmatales phylotypes, however,were related to uncultured archaea.

The dominating phylotypes Arch-3 and Arch-1 (29% and 28% of thetotal clones, respectively) shared 99% sequence similaritywith the acet-oclastic methanogenM. concilii and 97% similarity withMethanoculleus

Methanospirillum hungatei (M60880)Acetate-degrading methanogenic clone BA03 (AB092917)

Arch186(2) (HQ588823)Anaerobic butyrate-degrading reactor clone BHA (AB248620)

Arch163(6) (HQ588822)Arch153(1) (HQ588827)

Anaerobic digestion of sludge clone (CU916902)Arch12(14) (HQ588816)

Rich minerotrophic fen clone MHLsu47-1G (EU155925)Arch72(5) (HQ588818)

Wastewater sludge clone 57-1(AF424773)Methanoculleus palmolei DSM 4273T (Y16382)Methanoculleus chikugoensis JCM 10825T (AB038795)Arch1(57) (HQ588813)Biodegraded oil reservoir clone PL-9A5 (AY570665)

Arch205(1) (HQ588824)Methanosarcina sp. FR (AF020341)Biodegraded oil reservoir clone PL-37A1 (AY570658)

Arch40(1) (HQ588826)Methanosaeta thermophila DSM 6194T (AB071701)

Arch125(1) (HQ588821)Arch3(59) (HQ588815)Methanosaeta concilii DSM 3671T (X16932)

Arch88(1) (HQ588819)Oil contaminated groundwater clone KuA13 (AB077223)

Thermopiles hot springs clone ThpA 13 (EF444620)Arch104(1) (HQ588820)Arch2(51) (HQ588814)

Manzallah Lake clone Hados.Sedi.Arch.3 (AB355124)Psychrophilic anaerobic digesters clone AG10 (AY161261)

Arch57(1) (HQ588817)Hydrocarbon contaminated aquiferclone WCHD3-33 (AF050619)

Arch144(1) (HQ588825)Subsurface groundwater clone WA16 (AB237749)Landfill leachate clone GZK14 (AJ576214)

Aquifex pyrophilus DSM 6858 (M83548)54

99

73

100

79

54

100

100

100

59100

94

100

86100

83 98

9491

100

95

64

94

99

9299

88

55

7387

0.05

Methanom

icrobialesM

ethanosarcinalesM

ethanobacterialesT

hermoplasm

atales

Fig. 5. Phylogenetic relationships of archaeal clone sequences and environmental 16S rRNA sequences related to the Euryarchaeota (Phylum) including Methanomicrobiales,Methanosarcinales, Thermoplasmatales and Methanobacteriales. The tree was inferred by neighbor-joining analysis. Bootstrap values (%) are based on 1000 replicates and areshown for branches with more than 50% support. GenBank accession numbers of 16S rRNA sequences are given in parentheses. Bar, 0.5% sequence divergence. Clone sequencesfrom this study with their prevalence (in brackets) in the clone library are indicated in bold.

192 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

chikugoensis. Several other phylotypes (Arch-57, Arch-104, Arch-2) inthe archaeal clone library were closely related to archaeal clones re-trieved fromoil-contaminated aquifer or oil contaminated groundwater(Dojka et al., 1998; Watanabe et al., 2002).

Six of the 15 Archaea phylotypes showed>97% sequence similarityto 16S rRNA gene sequences in the public database, which suggeststhat these phylotypes have been previously described, while 9 of thephylotypes had less than 95% sequence similarity to their closest rela-tives in the public databases, thus suggesting they may representnovel archaeal genera.

3.8. Statistical analysis of the clone libraries

The number of clones, phylotypes, and biodiversity indices calcu-lated for the bacterial and archaeal clone libraries are shown inTable 1. Both bacterial and archaeal clone libraries had high librarycoverages (93% and 96%, respectively), which demonstrates that themajor part of the microbial diversity was investigated in this study.The bacterial library was characterized by a higher diversity thanthe archaeal library in terms of numbers of phylotypes and speciesrichness (Shannon index H′, Chao1). Simpson's reciprocal index ofdiversity (1/D) also pointed to a higher diversity among Bacteriacompared to Archaea. Evenness (E), however, estimated comparable

bacterial and archaeal diversity with regard to the relative abundanceof each phylotype in the two libraries.

4. Discussion

In this study we characterized the microbial community composi-tion in soil samples from a former Alpine industrial site contaminatedwith petroleum hydrocarbons. The lack of uncontaminated (pristine)soil corresponding to soil samples from the site investigated in thisstudy makes it impossible to compare microbial communities beforeand after contamination. The contamination included heavy oilwhich is characterized by a high ratio of aromatics and naphthenesto linear alkanes and contains high amounts of nitrogen, sulfur, oxy-gen and heavy metals. Since the contamination was aged, only lowamounts of hydrocarbons were water-extractable. The numbers ofculturable, heterotrophic bacteria were comparable to those reportedearlier from oil-contaminated soils in alpine regions (Margesin andSchinner, 1997; Margesin et al., 2003b; Margesin, 2007). The highfraction of culturable microorganisms able to utilize petroleum hy-drocarbons reflects the adaptation of the indigenous soil populationto the contamination. The temperature conditions prevailing at thestudied site (8–10 °C) were reflected by the ability of the culturable,

193D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

heterotrophic population to grow and utilize a number of aliphatic oraromatic hydrocarbons at 10–15 °C.

The enumeration of microbial populations is typically performedto gain information on the biodegradation potential of the hydrocar-bons present in a studied sample, however, it is associated with anumber of difficulties both from the methodological point of viewand from the interpretation of the results. Only a small fraction of mi-croorganisms can be isolated and cultured on laboratory media sincethe growth requirements for many strains are unknown and it is notknown if the culturable fraction is representative of the bacterial soilpopulation (Torsvik et al., 1998; Kirk et al., 2004). Thus, plate countsunderestimate the true viable population density (Amann et al.,1995; Atlas and Bartha, 1998). This was demonstrated in our studyby the fact that the culturable bacterial community accounted for4.4% (20 °C) and 2.0% (10 °C) of the total (DAPI-stained) bacterialcommunity. It is possible that 1% of the entire population.

The physiologically active eubacterial community, as revealed byFISH, accounted for 6.7% of the total (DAPI-stained) bacterial commu-nity. A high amount of the FISH-detected eubacterial communitycould be cultured under the applied culture conditions (R2A mediumat 20 and 10 °C), which indicates that large part of the studied soilpopulation was active. FISH is based on the detection of rRNA. Sincethe rRNA content is associated with the metabolic state of microbialcells, FISH is a valuable tool to describe the composition of the moreactive, ecologically relevant part of the microbial community(Amann et al., 1995; Wagner et al., 2003) even in complex matrices,such as soil (Barra Caracciolo et al., 2005).

The 16S rRNA gene sequences of 48 potentially unique, culturablestrains revealed that these strains were predominantly Proteobacteria(71%; mainly Alphaproteobacteria and Gammaproteobacteria); 19% ofthe studied population belonged to high G+C content Gram-positiveActinobacteria. Representatives of the genera Pseudomonas, Pedobacter,Brevundimonas, Rhodococcus, Arthrobacter and Bacillus, such as foundin our study, have been previously found in contaminated soil samples(Saul et al., 2005; Popp et al., 2006; Greer et al., 2010), including alpinesoils (Margesin et al., 2002, 2003a, 2003b). Our findings corroboratethat these genera are well adapted to hydrocarbon contaminants andthat they are relatively easy to culture. Our biodegradation studiesshowed that Actinobacteria strains (especially rhodococci) were themost versatile and efficient degraders among the investigated cultur-able bacteria. Interestingly, the genus Rhizobium dominated amongAlphaproteobacteria; representatives of this genus with the ability forN-fixation have been found in the rhizosphere of oil-contaminatedsoils and play a role in phytoremediation (Wang et al., 2007;Jurelevicius et al., 2010; Xu et al., 2010). Beside of pseudomonads, thegenus Janthinobacterium, often found in cold environments (Schlosset al., 2010; Lee et al., 2011) and able to degrade PAHs (Bodour et al.,2003) was frequently found among Betaproteobacteria.

The culture-independent characterization of the studied soil pop-ulation (16S rRNA gene clone library of Bacteria) also revealed a clearpredominance of Proteobacteria (85%) with representatives from allfive classes. These results agree with previous findings that all proteo-bacterial classes respond to the input of hydrocarbons (Greer et al.,2010). Contrary to culturable bacteria, Proteobacteria were highlydominated by Betaproteobacteria. Several phylotypes of this classshared high similarities with bacterial strains involved in the degra-dation of PAHs (Eriksson et al., 2003; Chang et al., 2007; Singletonet al., 2009).

In our study, both the culture-based and culture-independentanalysis of bacterial diversity demonstrated a clear dominance ofProteobacteria. This phylum is the most abundant in soil libraries(Janssen, 2006). The predominance of Gram-negative bacteria incontaminated alpine soils has been observed before and is seen asthe result of a shift in bacterial community composition towards anenrichment of Proteobacteria following hydrocarbon contamination(Labbé et al., 2007). Comparable results have been described in arctic

soils (Juck et al., 2000). Similarly, catabolic genotypes derived fromGram-negative bacteria (Pseudomonas, Acinetobacter) and involvedin the degradation of aliphatic and aromatic petroleum hydrocarbonswere significantly higher in contaminated alpine than in the corre-sponding pristine soils. In addition, there was a significant positivecorrelation between the level of contamination and the number of ge-notypes of Gram-negative bacteria, while genotypes of Gram-positivebacteria (Rhodococcus, Mycobacterium) were found at a similar fre-quency in pristine and contaminated soils (Margesin et al., 2003b).

The Cytophaga–Flavobacterium–Bacteroides (CFB) group, alsoknown as the phylum ‘Bacteroidetes’, is considered to comprise organ-isms associated with the degradation of organic compounds (Pinhassiet al., 1999; Weon et al., 2005). Representatives of this phylum arecommonly found in soil 16S rRNA gene libraries and frequently isolat-ed from cold soils (Saul et al., 2005; Margesin et al., 2009). Interest-ingly, all three culturable bacterial strains (BZ26, BZ33r and BZ42)assigned to the CFB group in this study were identified as novel spe-cies of the genera Dyadobacter, Hymenobacter and Pedobacter (Zhanget al., 2011a, 2010c, 2010d).

The fraction of Gram-positive representatives in the studied soilmicrobial population was much lower in the bacterial clone library(2% Actinobacteria) compared to results obtained from the culture-dependent technique (19% Actinobacteria, 4% Firmucutes). This dis-crepancy has already been observed before in hydrocarbon-contaminated soil and the possibility of an artifact has been discussed(Saul et al., 2005). Although bead and chemical lysis treatments wereused in our study, DNA extraction from Gram-positive bacteria maynot have been as effective as from Gram-negative bacteria. Becauseof their cell wall structure, Gram-positive bacteria are harder to lysethan Gram-negative bacteria (Bürgmann et al., 2001). However,even after repeated (threefold) bead treatments the numbers ofsequences from Gram-positive bacteria could not be significantly in-creased in the extract prepared from a hydrocarbon-contaminatedsoil sample (Popp et al., 2006). Furthermore, the sequences fromtwo sequential extractions showed no significant difference in phylo-genetic distribution at the phylum or subphylum level (Popp et al.,2006). In addition, culture-independent analysis includes sequencesfrom microorganisms that are not yet cultured or are difficult to cul-ture. Low abundances of Gram-positive bacteria (high G+C as wellas low G+C content bacteria) in bacterial clone libraries constructedfrom oil-contaminated soils have also been described by others(Yoshida et al., 2005; Popp et al., 2006; Liu et al., 2009).

Themolecular characterization also involved a survey on the abun-dance of Archaea (16S rRNA gene clone library). Archaea have beendetected in several environments contaminated with petroleum hy-drocarbons (Kasai et al., 2005; Yoshida et al., 2005; Liu et al., 2009)and have shown to be involved in themineralization of monoaromaticand polyaromatic components of petroleum hydrocarbons, especiallybymethanogenesis under anaerobic conditions (Anderson and Lovley,2000; Chang et al., 2006). It was proposed that petroleum contamina-tion can prevent ventilation to soil, which results in the formation ofanaerobic zones (Coates et al., 2001). However, Archaea are also pre-sent in aerobic ecosystems (DeLong, 1998) and have ecological signif-icance in many cold environments (Perreault et al., 2007; Steven et al.,2007, 2008).

All 5 phylotypes detected in our study belonged to the phylumEuryarchaeota and their sequences were closely related to those ofmethanogenic Archaea. Liu et al. (2009) also observed that the archae-al population in heavily oil-contaminated soils consisted mainly ofEuryarchaeotawith abundant methanogen-like operational taxonom-ic units. These authors described a more complex archaeal diversity inthe contaminated soil compared to the pristine control soil. Petroleumcontamination of the aerobic soil zone resulted in the establishment ofa methanogenic community with substantial hydrocarbon-degradingpotential (Kasai et al., 2005). In contrast, Röling et al. (2004) observeda dramatic decrease of archaeal abundance in oil-spilled beach

194 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

sediments and concluded that archaeal communities are unlikely toplay a significant role in oil spill bioremediation.

In our study we observed a predominance of phylotypes sharinghigh similarities with the methanogens M. concilii (Arch-3, 29% of thetotal clones) andM. chikugoensis (Arch-1, 28%) belonging to the ordersMethanosarcinales and Methanomicrobiales, respectively. Methanosaetaphylotypes were abundantly obtained in a petroleum-contaminatedsoil library but not in uncontaminated soil (Kasai et al., 2005).Methanosaeta-related sequences have been cloned from hydrocarbon-contaminated groundwater (Dojka et al., 1998). Given the only recog-nized form of energy production in the genusMethanosaeta is aceticlas-tic methanogenesis, Dojka et al. (1998) proposed that the aceticlasticmethanogen archaeon represented by the WCHD3-03 sequence wasprimarily important in hydrocarbon degradation in cold environments(10–12 °C). Methanomicrobiales (Methanocalculus sp., Methanoculleussp.) and Methanosarcinales (Methanosarcina sp., Methanosaeta sp.)were detected in our study. Representatives of these orders were alsofound among clones retrieved from crude-oil sludge (Yoshida et al.,2005).

Microbial diversity indices have been developed to describe struc-tures and dynamics of soil microbial communities (Nannipieri et al.,2003). The determination of abundances of sequence types in a diver-sity survey is a useful first step in predicting abundances of microor-ganisms in the studied environment (Dojka et al., 1998). Our data(Table 2) reveal a higher diversity among Bacteria compared toArchaea. Generally, archaeal diversities have been reported to belower than bacterial diversities (Roh et al., 2010). However, the calcu-lation of Evenness (E) showed a similar relative abundance of eachphylotype in the bacterial and archaeal clone libraries. Shannon di-versity index (H′) of the bacterial clone library in this study was3.30 (diversity in terms of species richness). Slightly higher Shannondiversity indices of 3.93 and 3.78 were reported in oil hydrocarbon-contaminated soils (Popp et al., 2006); Kumar and Khanna (2010)calculated at H′ value of 3.92 in PAH-contaminated soil. Shannondiversity indices of 2.76–2.93 were interpreted as a result of the lossof bacterial diversity in contaminated Antarctic soils compared tothe pristine controls (Saul et al., 2005). In general, there is a trendtowards lower Shannon indices in soils with aged hydrocarbon con-tamination (Andreoni et al., 2004; Popp et al., 2006). Due to the lackof data for comparison in literature, we unfortunately cannot com-pare the diversity indices obtained from the archaeal library withthose of others obtained in contaminated soils.

In our study, a number of bacterial and archaeal phylotypes sharedhigh similarities with strains previously described to be involved inhydrocarbon biodegradation (see also Section 3.), implying that micro-organism with the potential to degrade petroleum hydrocarbons arewell-established in the studied soil. In addition, the number of se-quences potentially representing novel species (8 culturable novelspecies have been recently described) or genera (44/68 bacterial phylo-types and 9/15 archaeal phylotypes in the clone libraries) was remark-ably high. This demonstrates that the investigated site represents alarge pool of microorganisms well adapted to the contamination andpotentially contributing to bioremediation processes, which will bethe focus of further studies. Special focus will be given to Actinobacteria,

Table 2Numbers of clones and phylotypes and biodiversity indices for the bacterial and ar-chaeal 16S rRNA clone libraries.

Clone library Bacteria Archaea

Total no. of clones 390 202Total no. of phylotypes 68 15Coverage (%) 93 96Shannon index (H′) 3.30 1.70Simpson's index (1/D) 14.76 4.32Chao1 95 29Evenness (E) 0.40 0.37

that turned out to be the most versatile and efficient degraders of abroad range of hydrocarbons (alkanes, aromatic and polyaromatichydrocarbons) among culturable soil bacteria. However, the potentialof Proteobacteria and members of the CFB group also needs to be evalu-ated since representatives of both phyla are known to be involved inhydrocarbon biodegradation processes. It will further be relevant tostudy the contribution of the archaeal community to bioremediation.

Acknowledgments

This research work was supported by a grant from the “AutonomeProvinz Bozen, Südtirol”, Amt für Geologie und Baustoffprüfung.We thank F. Schinner for helpful discussions and P. Thurnbichler,J. Mair and S. Kasenbacher for technical assistance.

References

Alexander M. Biodegradation and bioremediation. 2nd ed. San Diego: Academic Press;1999.

Allen JP, Atekwana EA, Atekwana EA, Duris JW,Werkema DD, Rossbach S. The microbialcommunity structure in petroleum-contaminated sediments corresponds to geo-physical signatures. Appl Environ Microbiol 2007;73:2860–70.

Alonso-Gutierrez J, Figueras A, Albaiges J, Jimenez N, Vinas M, Solanas AM, et al. Bacte-rial communities from shoreline environments (Costa da Morte, NorthwesternSpain) affected by the Prestige oil spill. Appl Environ Microbiol 2009;75:3407–18.

Amann R, Ludwig W, Schleifer KH. Phylogenetic identification and in situ detection ofindividual microbial cells without cultivation. Microbiol Rev 1995;59:143–69.

Anderson RT, Lovley DR. Hexadecane decay by methanogenesis. Nature 2000;404:722–3.

Andreoni V, Cavalca L, Rao MA, Nocerino G, Bernasconi S, Dell'Amico E, et al. Bacterialcommunities and enzyme activities of PAHs polluted soils. Chemosphere 2004;57:401–12.

Atlas RM, Bartha R. Microbial ecology: fundamentals and applications. 4th ed. CA:Benjamin/Cummings Publishing Company, Inc; 1998.

Barra Caracciolo A, Grenni P, Cupo C, Rosetti S. In situ analysis of native microbial com-munities in complex samples with high particulate loads. FEMS Microbiol Lett2005;253:55–8.

Bertaux J, Gloger U, Schmid M, Hartmann A, Scheu S. Routine fluorescence in situhybridization in soil. J Microbiol Methods 2007;69(4):51–60.

Bodour AA, Wang JM, Brusseau ML, Maier RM. Temporal change in culturable phenan-threne degraders in response to long-term exposure to phenanthrene in a soilcolumn system. Environ Microbiol 2003;5:888–95.

Bossert I, Bartha R. The fate of fuel spills in soil ecosystems. In: Atlas RM, editor. Petro-leum Microbiology. New York: Macmillan; 1984. p. 435–73.

Bürgmann H, Pesaro M, Widmer F, Zeyer J. A strategy for optimizing quality and quan-tity of DNA extracted from soil. J Microbiol Methods 2001;45:7-20.

Chang W, Um Y, Holoman TR. Polycyclic aromatic hydrocarbon (PAH) degradationcoupled to methanogenesis. Biotechnol Lett 2006;28:425–30.

Chang YT, Lee JF, Chao HP. Variability of communities and physiological characteristicsbetween free-living bacteria and attached bacteria during the PAH biodegradationin a soil/water system. Eur J Soil Biol 2007;43:283–96.

Chao A. Nonparametric estimation of the number of classes in a population. Scand J Stat1984;11:265–70.

Coates JD, Chakraborty R, Lack JG, O'Connor SM, Cole KA, Bender KS, et al. Anaerobicbenzene oxidation coupled to nitrate reduction in pure culture by two strains ofDechloromonas. Nature 2001;411:1039–43.

Cole JR, Chai B, Ferris RJ, Wang Q, Kulam SA, McGarrell DM, et al. The Ribosomal data-base project (RDP-II): sequences and tools for high-throughput rRNA analysis.Nucleic Acids Res 2005;33:D294–6.

DeLong EF. Everything in moderation: Archaea as ‘non-extremophiles’. Curr OpinGenes Dev 1998;8:649–54.

DIN EN ISO 9377-2. Water quality — determination of hydrocarbon oil index. Part 2:Method using solvent extraction and gas chromatography. Berlin: Beuth VerlagGmbH; 2000.

DIN ISO 16703. Soil quality — determination of content of hydrocarbons in the rangeC10 to C40 by gas chromatography. Berlin: Beuth Verlag GmbH; 2006.

Dojka MA, Hugenholtz P, Haack SK, Pace NR. Microbial diversity in a hydrocarbon- andchlorinated-solvent-contaminated aquifer undergoing intrinsic bioremediation.Appl Environ Microbiol 1998;64:3869–77.

Dorsch M, Stackebrandt E. Some modifications in the procedure of direct sequencing ofPCR amplified 16S rDNA. J Microbiol Methods 1992;16:271–9.

Eriksson M, Sodersten E, Yu Z, Dalhammar G, Mohn WW. Degradation of polycyclicaromatic hydrocarbons at low temperature under aerobic and nitrate-reducingconditions in enrichment cultures from northern soils. Appl Environ Microbiol2003;69:275–84.

Fabiani A, Gamalero E, Castaldini M, Coss GP, Musso C, Pagliai M, et al. Microbiologicalpolyphasic approach for soil health evaluation in an Italian polluted site. Sci TotalEnviron 2009;407:4954–64.

Felske A, Wolterink A, van Lis R, de Vos WM, Akkermans ADL. Searching for predomi-nant soil bacteria: 16S rDNA cloning versus strain cultivation. FEMS Microbiol Ecol1999;30:137–45.

195D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

Good IL. The population frequencies of species and the estimation of populationparameters. Biometrika 1953;40:237–64.

Gratia E, Weekers F, Margesin R, D'Amico S, Thonart P, Feller G. Selection of a cold-adapted bacterium for bioremediation of wastewater at low temperatures. Extre-mophiles 2009;13:763–8.

Greer CW, Whyte LG, Niederberger TD. Microbial communities in hydrocarbon-contaminated temperate, tropical, alpine, and polar soils. In: Timmis KN, editor.Handbook of Hydrocarbon and Lipid Microbiology. Berlin: Springer-Verlag; 2010.p. 2313–28.

Hiraishi A, Yonemitsu Y, Matsushita M, Shin YK, Kuraishi H, Kawahara K. Characteriza-tion of Porphyrobacter sanguineus sp. nov., an aerobic bacteriochlorophyll-containing bacterium capable of degrading biphenyl and dibenzofuran. ArchMicrobiol 2002;178:45–52.

Janssen PH. Identifying the dominant soil bacterial taxa in libraries of 16S rRNA and16S rRNA genes. Appl Environ Microbiol 2006;72:1719–28.

Juck D, Charles T, Whyte LG, Greer CW. Polyphasic microbial community analysis ofpetroleum hydrocarbon-contaminated soils from two northern Canadian commu-nities. FEMS Microbiol Ecol 2000;33:241–9.

Jurelevicius D, Korenblum E, Casella R, Vital RL, Seldin L. Polyphasic analysis of the bac-terial community in the rhizosphere and roots of Cyperus rotundus L. grown in apetroleum-contaminated soil. J Microbiol Biotechnol 2010;20:862–70.

Kasai Y, Takahata Y, Hoaki T, Watanabe K. Physiological and molecular characterizationof a microbial community established in unsaturated, petroleum-contaminatedsoil. Environ Microbiol 2005;7:806–18.

Kim JS, Crowley DE. Microbial diversity in natural asphalts of the Rancho La Brea TarPits. Appl Environ Microbiol 2007;73:4579–91.

Kimura M. A simple method for estimating evolutionary rates of base substitutionsthrough comparative studies of nucleotide sequences. J Mol Evol 1980;16:111–20.

Kirk LJ, Beaudette LA, Hart M, Moutoglis P, Klironomas JN, Lee H, et al. Methods ofstudying soil microbiol diversity. J Microbiol Methods 2004;58:169–88.

Kobabe S, Wagner D, Pfeiffer EM. Characterisation of microbial community composi-tion of a Siberian tundra soil by fluorescence in situ hybridisation. FEMS MicrobiolEcol 2004;50:13–23.

Krebs C. Ecological methodology. New York: Harper and Row; 1989.Kumar M, Khanna S. Diversity of 16S rRNA and dioxygenase genes detected in coal-tar-

contaminated site undergoing active bioremediation. J Appl Microbiol 2010;108:1252–62.

Labbé D, Margesin R, Schinner F, Whyte LG, Greer CW. Comparative phylogenetic anal-ysis of microbial communities in pristine and hydrocarbon-contaminated alpinesoils. FEMS Microbiol Ecol 2007;59:466–75.

Lee Y, Kim SY, Jung J, Kim EH, Cho KH, Schinner F, et al. Cultured bacterial diversity andhuman impact on alpine glacier cryoconite. J Microbiol 2011;49:355–62.

Li Y, Dick WA, Tuovinen OH. Fluorescence microscopy for visualization of soil microor-ganisms — a review. Biol Fertil Soils 2004;39:301–11.

Liu R, Zhang Y, Ding R, Li D, Gao Y, Yang M. Comparison of archaeal and bacterial com-munity structures in heavily oil-contaminated and pristine soils. J Biosci Bioeng2009;5:400–7.

Loy A, Maixner F, Wagner M, Horn M. probeBase-an online resource for rRNA-targetedoligonucleotide probes: new features 2007. Nucleic Acids Res 2007;35:D800–4.

Maier RM, Pepper IL, Gerba CP. Environmental microbiology. London: Academic Press; 2000.Margesin R. Alpine microorganisms: useful tools for low-temperature bioremediation.

J Microbiol 2007;45:281–5.Margesin R, Schinner F. Bioremediation of diesel-oil contaminated alpine soils at low

temperatures. Appl Microbiol Biotechnol 1997;47:462–8.Margesin R, Schinner F. Biodegradation and bioremediation of hydrocarbons in ex-

treme environments (mini-review). Appl Microbiol Biotechnol 2001;56:650–63.Margesin R, Schinner F. Manual for Soil Analysis—Monitoring and Assessing Soil Biore-

mediation (Soil Biology, Vol. 5). ed. Berlin, Heidelberg, New York, Tokyo: SpringerVerlag; 2005.

Margesin R, Zacke G, Schinner F. Characterization of heterotrophic microorganisms inalpine glacier cryoconite. Arct Antarct Alp Res 2002;34:88–93.

Margesin R, Gander S, Zacke G, Gounot AM, Schinner F. Hydrocarbon degradation and en-zyme activities of cold-adapted bacteria and yeasts. Extremophiles 2003a;7:451–8.

Margesin R, Labbé D, Schinner F, Greer CW,Whyte LG. Characterization of hydrocarbon-degrading microbial populations in contaminated and pristine alpine soils. ApplEnviron Microbiol 2003b;69:3085–92.

Margesin R, Hämmerle M, Tscherko D. Microbial activity and community compositionduring bioremediation of diesel-oil-contaminated soil: effects of hydrocarbonconcentration, fertilizers and incubation time. Microb Ecol 2007;53:259–69.

Margesin R, Jud M, Tscherko D, Schinner F. Microbial communities and activities inalpine and subalpine soils. FEMS Microbiol Ecol 2009;67:208–18.

Margesin R, Plaza GA, Kasenbacher S. Characterization of bacterial communities atheavy-metal-contaminated sites. Chemosphere 2011;82:1583–8.

McDougald D, Rice SA, Kjelleberg S. New perspectives on the viable but nonculturableresponse. Biologia 2009;54:617–23.

Militon C, Boucher D, Vachelard C, Perchet G, Barra V, Troquet J, et al. Bacterial commu-nity changes during bioremediation of aliphatic hydrocarbon-contaminated soil.FEMS Microbiol Ecol 2010;74:669–81.

Nannipieri P, Ascher J, Ceccherini MT, Landi L, Pietramellara G, Renella G. Microbialdiversity and soil functions. Eur J Soil Biol 2003;54:655–70.

ÖNORM EN 14309. Characterization of waste — determination of hydrocarbon contentin the range of C10 to C40 by gas chromatography. Wien: ÖsterreichischesNormungsinstitut; 2005.

Perreault NN, Andersen DT, Pollard WH, Greer CW, Whyte LG. Characterization of theprokaryotic diversity in cold saline perennial springs of the Canadian High Arctic.Appl Environ Microbiol 2007;73:1532–43.

Phuong K, Kakii K, Nikata T. Intergeneric coaggregation of non-flocculating Acinetobacter spp.isolates with other sludge-constituting bacteria. J Biosci Bioeng 2009;107:394–400.

Pinhassi J, Azam F, Hemphälä J, Long RA, Martinez J, Zweifel UL, et al. Coupling betweenbacterioplankton species composition, population dynamics, and organic matterdegradation. Aquat Microb Ecol 1999;17:13–26.

Popp N, Schlömann M, Mau M. Bacterial diversity in the active stage of a bioremedia-tion system for mineral oil hydrocarbon-contaminated soils. Microbiology2006;152:3291–304.

Ramakrishnan B, Megharai M, Venkateswarlu K, Naidu R, Sethunathan N. The impactsof environmental pollutants on microalgae and cyanobacteria. Crit Rev Env SciTechnol 2010;40:699–821.

Rappé M, Giovannoni SJ. The uncultured microbial majority. Annu Rev Microbiol2003;57:369–94.

Richardson SD, Lebron BL, Miller CT, Aitken MD. Recovery of phenanthrene-degradingbacteria after simulated in situ persulfate oxidation in contaminated soil. EnvironSci Technol 2011;45:719–25.

Roh S, Kim K, Nam Y, Chang H, Park E, Bae J. Investigation of archaeal and bacterial di-versity in fermented seafood using barcoded pyrosequencing. ISME J 2010;4:1-16.

Röling WF, de Brito Couto IR, Swannell RP, Head IM. Response of Archaeal communitiesin beach sediments to spilled oil and bioremediation. Appl Environ Microbiol2004;70:2614–20.

Sattley WM, Madigan MT. Isolation, characterization, and ecology of cold-active, che-molithotrophic, sulfur-oxidizing bacteria from perennially ice-covered Lake Fryxell,Antarctica. Appl Environ Microbiol 2006;72:5562–8.

Saul DJ, Aislabie JM, Brown CE, Harris L, Foght JM. Hydrocarbon contamination changesthe bacterial diversity of soil from around Scott Base, Antarctica. FEMS MicrobiolEcol 2005;53:141–55.

Schinner F, Öhlinger R, Kandeler E, Margesin R. Methods in Soil Biology. ed. BerlinHeidelberg: Springer Verlag; 1996.

Schloss PD, Allen HK, Klimowicz AK, Mlot C, Gross JA, Savengsuksa S, et al. Psychro-trophic strain of Janthinobacterium lividum from a cold Alaskan soil produces pro-digiosin. DNA Cell Biol 2010;29:533–41.

Singleton DR, Ramirez LG, Aitken MD. Characterization of a polycyclic aromatic hydro-carbon degradation gene cluster in a phenanthrene-degrading Acidovorax strain.Appl Environ Microbiol 2009;75:2613–20.

Spiegelman D, Whissell G, Greer CW. A survey of the methods for the characterizationof microbial consortia and communities. Can J Microbiol 2005;51:355–86.

Stackebrandt E, Charfreitag O. Partial 16S rRNA primary structure of five Actinomycesspecies: phylogenetic implications and development of an Actinomyces israelii-specific oligonucleotide probe. J Gen Microbiol 1990;136:37–43.

Stahl DA, Amann R. Development and application of nucleic acid probes. In:Stackebrandt E, Goodfellow M, editors. Nucleic Acid Techniques in Bacterial Sys-tematics. New York: Wiley; 1991. p. 205–48.

Steven B, Briggs G, McKay CP, Pollard WH, Greer CW, Whyte LG. Characterization of themicrobial diversity in a permafrost sample from the Canadian high Arctic usingculture-dependent and culture-independent methods. FEMS Microbiol Ecol2007;59:513–23.

Steven B, Pollard WH, Greer CW, Whyte LG. Microbial diversity and activity through apermafrost/ground ice core profile from the Canadian High Arctic. Environ Micro-biol 2008;10:3388–403.

Stolz A, Schmidt-Maag C, Denner EBM, Busse HJ, Egli T, Kämpfer P. Description ofSphingomonas xenophaga sp. nov. for strains BN6T and N, N which degrade xenobi-otic aromatic compounds. Int J Syst Evol Microbiol 2000;50:35–41.

Tamura K, Dudley J, Nei M, Kumar S. MEGA4: molecular evolutionary genetics analysis(MEGA) software version 4.0. Mol Biol Evol 2007;24:1596–9.

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The CLUSTAL_X win-dows interface: flexible strategies for multiple sequence alignment aided by qual-ity analysis tools. Nucleic Acids Res 1997;25:4876–82.

Torsvik V, Daae FL, Sandaa RA, Ovreas L. Review article: novel techniques for analysing mi-crobial diversity in natural and perturbed environments. J Biotechnol 1998;64:53–62.

Wagner M, Horn M, Daims H. Fluorescence in situ hybridisation for the identificationand characterisation of prokaryotes. Curr Opin Microbiol 2003;6:302–9.

Wallner G, Amann R, Beisker W. Optimizing fluorescent in situ hybridisation with rRNAtargeted oligonucleotide probes for flow cytometric identification of microorgan-isms. Cytometry 1993;14:136–43.

Wang J, Xu H, Guo S. Isolation and characteristics of a microbial consortium for effec-tively degrading phenanthrene. Pet Sci 2007;4:68–75.

Watanabe K,Watanabe K, Kodama Y, Syutsubo K, Harayama S.Molecular characterizationof bacterial populations in petroleum-contaminated groundwater discharged fromunderground crude oil storage cavities. Appl Environ Microbiol 2000;66:4803–9.

Watanabe K, Kodama Y, Hamamura N, Kaku N. Diversity, abundance, and activity of ar-chaeal populations in oil-contaminated groundwater accumulated at the bottom ofan underground crude oil storage cavity. Appl Environ Microbiol 2002;68:3899–907.

Weon HY, Kim BY, Kwon SW, Park IC, Cha IB, Tindall BJ, et al. Leadbetterella byssophilagen. nov., sp. nov., isolated from cotton-waste composts for the cultivation of oys-ter mushroom. Int J Syst Evol Microbiol 2005;55:2297–302.

Whyte LG, Greer CW. Molecular techniques for monitoring and assessing soil bioreme-diation. In: Margesin R, Schinner F, editors. Manual for Soil Analysis — Monitoringand Assessing Soil Bioremediation (Soil Biology, Vol.5). Berlin, Heidelberg, NewYork, Tokyo: Springer-Verlag; 2005. p. 201–31.

Winderl C, Penning H, Netzer F, Meckenstock RU, Lueders T. DNA-SIP identifies sulfate-reducing Clostridia as important toluene degraders in tar-oil-contaminated aquifersediment. ISME J 2010;4:1314–25.

Xu L, Teng Y, Li ZG, Norton JM, Luo YM. Enhanced removal of polychlorinated biphenylsfrom alfalfa rhizosphere soil in a field study: the impact of a rhizobial inoculum. SciTotal Environ 2010;408:1007–13.

196 D.-C. Zhang et al. / Science of the Total Environment 421–422 (2012) 184–196

Yoshida N, Yagi K, Sato D, Watanabe N, Kuroishi T, Nishimoto K, et al. Bacterial commu-nities in petroleum oil in stockpiles. J Biosci Bioeng 2005;99:143–9.

Zhang DC, Liu HC, Xin YH, Zhou YG, Schinner F, Margesin R. Sphingopyxis bauzanensissp. nov., a novel psychrophilic bacterium isolated from soil. Int J Syst Evol Micro-biol 2010a;60:2618–22.

Zhang DC, Liu HC, Xin YH, Zhou YG, Schinner F, Margesin R. Luteimonas terricola sp.nov., a psychrophilic bacterium isolated from soil. Int J Syst Evol Microbiol2010b;60:1581–4.

Zhang DC, Liu HC, Xin YH, Zhou YG, Schinner F, Margesin R. Dyadobacter psychrophilussp. nov., a novel psychrophilic bacterium isolated from soil. Int J Syst Evol Micro-biol 2010c;60:1640–3.

Zhang DC, Schinner F, Margesin R. Pedobacter bauzanensis sp. nov., isolated from soil.Int J Syst Evol Microbiol 2010d;60:2590–3.

Zhang DC, Schumann P, Liu HC, Xin YH, Zhou YG, Schinner F, et al. Agromyces bauzanensissp. nov., isolated from soil. Int J Syst Evol Microbiol 2010e;60:2341–5.

Zhang DC, Busse HJ, Liu HC, Zhou YG, Schinner F, Margesin R.Hymenobacter psychrophilussp. nov., a psychrophilic bacterium isolated from soil. Int J Syst Evol Microbiol2011a;61:859–63.

Zhang DC, Liu HC, Zhou YG, Schinner F, Margesin R. Pseudomonas bauzanensis sp. nov., apsychrophilic bacterium isolated from soil. Int J Syst Evol Microbiol 2011b;61:2333–7.

Zhang DC, Liu HC, Zhou YG, Schinner F, Margesin R. Tistrella bauzanensis sp. nov., anovel bacterium isolated from soil. Int J Syst Evol Microbiol 2011c;61:2227–30.