Accepted Manuscript

Plant diversity enhances the natural attenuation of polycyclic aromatic compounds

(PAHs and oxygenated PAHs OPAHs) in grassland soils

Benjamin A. Musa Bandowe, Sophia Leimer, Hannah Meusel, Andre Velescu, Sigrid
Dassen, Nico Eisenhauer, Thorsten Hoffmann, Yvonne Oelmann, Wolfgang Wilcke

PII: S0038-0717(18)30371-7
DOI: https://doi.org/10.1016/j.soilbio.2018.10.017
Reference: SBB 7321

To appear in: Soil Biology and Biochemistry

Received Date: 22 May 2018

Revised Date: 21 September 2018
Accepted Date: 23 October 2018

Please cite this article as: Musa Bandowe, B.A., Leimer, S., Meusel, H., Velescu, A., Dassen, S.,
Eisenhauer, N., Hoffmann, T., Oelmann, Y., Wilcke, W., Plant diversity enhances the natural attenuation
of polycyclic aromatic compounds (PAHs and oxygenated PAHs OPAHs) in grassland soils, Soil Biology
and Biochemistry (2018), doi: https://doi.org/10.1016/j.soilbio.2018.10.017.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

1 Plant diversity enhances the natural attenuation of polycyclic aromatic compounds

2 (PAHs and oxygenated PAHs OPAHs) in grassland soils

3 Benjamin A. Musa Bandowea,b*, Sophia Leimera, Hannah Meuselb, Andre Velescua, Sigrid

PT
4 Dassenc, Nico Eisenhauerd,e, Thorsten Hoffmannf, Yvonne Oelmanng, Wolfgang Wilckea

RI
a
6 Institute of Geography and Geoecology, Karlsruhe Institute of Technology (KIT), Reinhard-Baumeister-

7 Platz 1, 76131 Karlsruhe, Germany

SC
b
8 Multiphase Chemistry Department, Max Planck Institute for Chemistry, Hahn-Meitner-Weg 1, 55128

9 Mainz, Germany

10
c

U
Department of Terrestrial Ecology, Netherlands Institute of Ecology, NIOO KNAW,
AN
11 Droevendaalsesteeg 10, 6708 PB Wageningen, The Netherlands
d
M

12 German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Deutscher Platz 5e,

13 04103 Leipzig, Germany

D

e
14 Institute of Biology, Leipzig University, Deutscher Platz 5e, 04103 Leipzig, Germany
TE

f
15 Institute of Inorganic and Analytical Chemistry, University of Mainz, Duesbergweg 10-14, 55128

16 Mainz, Germany
EP

g
17 Geoecology, University of Tübingen, Rümelinstraβe 19-23, 72070 Tübingen, Germany

18
C

19
AC

20 * Corresponding Author (B.A.M. Bandowe, Email: benjamin.bandowe@kit.edu;

21 benjamin.bandowe@mpic.de)

22

23

1
 ACCEPTED MANUSCRIPT

24 ABSTRACT

25 Increasing plant species richness stimulates microbial activity in soil, which might favor

26 biodegradation of polycyclic aromatic compounds (PACs). To explore the relationship between

PT
27 plant community composition and PACs in grassland soils (Fluvisols exposed to an urban

28 atmosphere), we determined the concentrations of 29 polycyclic aromatic hydrocarbons (PAHs)

RI
29 and 15 oxygenated PAHs (OPAHs) in topsoils of 80 plots of a grassland biodiversity experiment.

30 The plots included different levels of plant species richness (1, 2, 4, 8, 16, 60 species) and 1-4

SC
31 plant functional groups (grasses, small herbs, tall herbs, and legumes) in a randomized block

design. The concentrations (ng g-1) of ∑29PAHs and ∑15OPAHs in the soils were 271-2407 and

U
32

57-329, respectively. Concentrations of 16 (out of 44) PACs and ∑29PAHs decreased

AN
33

34 significantly with increasing plant species richness, after accounting for the effects of block and
M

35 initial soil organic C concentration (ANCOVA, p<0.05). Microbial turnover as the mechanism

36 underlying this relationship was supported by the findings that (i) the regression of the
D

37 concentrations of PAH with > 4 aromatic rings on plant species richness yielded slopes that
TE

38 were negatively correlated with their octanol-water partitioning coefficients, (ii) two OPAHs

39 accumulated in soils with higher plant species richness, and (iii) higher plant species richness
EP

40 increased four OPAH/parent-PAH ratios. Accordingly, structural equation modelling indicated

41 that the higher concentration of 1,2-acenaphthenequinone (a metabolite of acenaphthene) and the

C

42 higher 1,2-acenaphthenequinone/acenaphthene and 1-indanone/fluorene ratios with increasing

AC

43 species richness was attributable to increasing soil microbial biomass. We conclude that higher

44 plant species richness can be used to enhance biodegradation of aged PACs in soil. We however

45 caution that OPAHs (some of which are more toxic than their related PAHs) might accumulate in

46 soils during such a plant-assisted remediation process.

2
 ACCEPTED MANUSCRIPT

47 1. Introduction

48 Polycyclic aromatic compounds (PACs), such as polycyclic aromatic hydrocarbons (PAHs) and

49 oxygenated PAHs (OPAHs) are ubiquitous in soils. These compounds are mainly released from

PT
50 combustion processes, with additional sources of OPAHs mainly from transformation of PAHs

51 by biological (microbial, enzymatic) and abiotic (photochemical, thermal) reactions (Fatiadi,

RI
52 1967; Keyte et al., 2013; Lundstedt et al., 2007; Wilcke et al., 2014b). Soils at industrial and

53 urban sites and near roads are frequently contaminated with PACs (Arp et al., 2014; Bandowe et

SC
54 al., 2014a; Bandowe et al., 2011; Choi et al., 2009; Wilcke, 2000). Several representatives of the

PACs are toxic (ecotoxic, genotoxic, mutagenic, estrogenic), bioaccumulative, persistent, and

U
55

therefore cause severe damage to ecosystems, animal and human health (Arp et al., 2014; IARC,
AN
56

57 2010; Lundstedt et al., 2007). PACs in soils are dissipated by processes such as biodegradation,
M

58 plant uptake, bioaccumulation, volatilization, and leaching, which determine their overall

59 residence time and fate (Semple et al., 2003).

D

60 Plants modulate the concentration of PACs in soil by several processes, particularly by their
TE

61 effect on microbial and enzyme activity and associated turnover of PACs (Collins et al., 2006;

62 Cousins et al., 1999; Desalme et al., 2013; Pilon-Smits, 2005; Wild et al., 2005a, b). The use of
EP

63 plants to clean-up soils contaminated with PACs and other organic pollutants, termed

64 phytoremediation, is a cost-effective and environment-friendly method. The results of intensive

C

65 research on identifying processes, mechanisms, and useful plants or plant mixtures involved in
AC

66 phytoremediation and natural attenuation of organic pollutants have been summarized in several

67 reviews (Aken et al., 2009; Arthur et al., 2005; Pilon-Smits, 2005). The previous results have

68 shown that some plants such as members of the family of Fabaceae (legumes) and some

69 combinations of several plants can enhance PACs degradation. The limitations of these studies

3
 ACCEPTED MANUSCRIPT

70 include the fact that experiments were often run with a low number of plant species (partly in a

71 non-full-factorial design), for a limited study duration, and the usage of spiked soils. Moreover,

72 the degradation of primary OPAHs, and the formation and accumulation of OPAHs from PAHs

73 during natural attenuation and/or technical (bio)remediation of PACs-contaminated soils have

PT
74 hardly been considered in phytoremediation experiments (Bamforth and Singleton, 2005;

RI
75 Chibwe et al., 2015; Lundstedt et al., 2007; Wilcke et al., 2014b). This is despite the fact that

76 some OPAHs pose comparable or even more severe (eco)toxicological risks than the PAHs (Dai

SC
77 et al., 2018; Lundstedt et al., 2007; Wincent et al., 2015).

78 In the past, the influence of biodiversity on many ecosystem functions has been studied in

79
U
designed experiments, in which the species richness and other properties of the plant community
AN
80 were systematically manipulated (Spehn et al., 2005; Tilman et al., 1996; Weisser et al., 2017).

One of the largest long-term biodiversity experiments is the Jena Experiment (started in 2002),
M

81

82 which varies plant species richness from 1 to 60 in a randomized block design in a temperate
D

83 grassland. The Jena Experiment includes the full range of plant species richness from
TE

84 monocultures to the natural plant species richness of an undisturbed meadow (Roscher et al.,

85 2004; Weisser et al., 2017). Furthermore, the Jena Experiment is designed in a way that allows to
EP

86 separate plant species richness effects per se from effects of specific functional groups (legumes,

87 small and tall herbs, and grasses) or mixtures of functional groups. Many of the reported plant
C

88 diversity effects from the Jena Experiment on microclimatic conditions, water balance, nutrient
AC

89 cycles, microbial abundance and diversity, and enzyme concentrations in soil could directly

90 affect the turnover of PACs in soil (Eisenhauer et al., 2010; Eisenhauer et al., 2011; Oelmann et

91 al., 2011; Rosenkranz et al., 2012; Weisser et al., 2017). Furthermore, the Jena Experiment was

92 established on a formerly plowed and thus homogenized arable soil in the urban area of the city

4
 ACCEPTED MANUSCRIPT

93 of Jena. We expected a homogeneous PACs contamination because of atmosphere-soil

94 partitioning between a well-mixed topsoil and a homogeneous urban atmosphere only driven by

95 small-scale variations in soil organic carbon (SOC) concentrations. The Jena Experiment

96 therefore offers a unique opportunity to gain mechanistic insights in the role of plant diversity on

PT
97 PACs concentrations in soil.

RI
98 The objectives of this study were to investigate (i) if plant diversity (species richness,

99 functional group richness, and presence of specific functional groups) influences PACs

SC
100 concentrations in grassland soil, and, if so, (ii) which mechanisms drive the plant diversity-PACs

101 concentrations relationship. We hypothesize that the concentrations of most PAHs and primary

102
U
OPAHs decrease with increasing plant species richness and plant functional group richness and
AN
103 in the presence of individual plant functional groups because of enhanced dissipation, while

secondary OPAHs are formed as degradation products. We furthermore hypothesize that the
M

104

105 reason for the changes in PACs concentrations are mainly attributable to plant diversity-induced
D

106 variations in microbial degradation.

TE

107

108 2. Materials and Methods

EP

109 2.1. Study site and experimental design

110 This study was conducted as part of the Jena Experiment (www.the-jena-experiment.de), which is a
C

111 grassland plant diversity experiment studying the role of plant diversity for element cycling and
AC

112 trophic interactions (Roscher et al., 2004). The Jena Experiment is located in the city of Jena,

113 Germany (50°55′N, 11°35′E; 130 m above sea level, ca. 110.000 habitants) on the floodplain of

114 the Saale River. Therefore, the experimental soil has been exposed to an urban atmosphere,

115 which usually contains elevated concentrations of several pollutants including PACs (Baek et al.,

5
 ACCEPTED MANUSCRIPT

116 1991; Menichini, 1992). Mean annual air temperature is 9.9°C, and mean annual precipitation is

117 610 mm (1980–2010) (Hoffmann et al., 2014). The soil is an Eutric Fluvisol, a natural soil free

118 of technogenic contributions that developed from up to 2-m thick loamy fluvial sediments,

119 almost free of stones (IUSS Working Group, 2014). As a result of the fluvial dynamics, the

PT
120 texture ranges from sandy loam near the river to silty clay with increasing distance from the

RI
121 river. This systematic variation in soil texture is considered in the experimental design as the

122 plots are arranged in four blocks that are parallel to the river Saale and each has approximately

SC
123 homogeneous soil texture. The site was converted from grassland to an arable field in the 1960s

124 and thereafter fertilized and plowed for crop production until the beginning of the grassland plant

125
U
diversity experiment in 2002 (Roscher et al., 2004; Weisser et al., 2017). See also the website of
AN
126 the Jena Experiment for experimental details: www.the-jena-experiment.de.

The entire experimental design has been described previously (Roscher et al., 2004; Weisser
M

127

128 et al., 2017). Briefly, the main experiment comprises 82 plots (20 m × 20 m) grouped in four
D

129 blocks (considering the systematic variation in soil texture). The 82 plots were established from
TE

130 seeds in May 2002 with different levels of plant species richness (1, 2, 4, 8, 16, or 60 plant

131 species) and plant functional group richness (1-4 plant functional groups out of grasses, small
EP

132 herbs, tall herbs, and legumes) chosen by the random replacement method. The species were

133 selected from a pool of 60 species frequently occurring in nutrient-rich agricultural grasslands
C

134 (Arrhenatherion meadows) (Ellenberg and Leuschner, 2010). Each level of plant species richness
AC

135 is replicated four times per block, resulting in 16 plots per richness level except for the 16-

136 species plots, which were replicated 14 times, and the 60-species plots, which were replicated

137 four times (once per block) in the whole experiment. There was a strong correlation between the

138 number of sown species and the realized plant species richness (R2 > 0.9 in each year during

6
 ACCEPTED MANUSCRIPT

139 2003–2007) (Marquard et al., 2009) corroborating the successful establishment of the plant

140 species richness gradient. The management was adapted to meadows that are managed with low

141 intensity and used for hay production by mowing twice a year in June and September. The plots

142 were regularly weeded by hand to maintain the sown species composition. Mown and weeded

PT
143 biomass was removed from the experimental plots. During the experimental period, the plots

RI
144 were not fertilized. Two plots with monocultures had to be abandoned (Cynosurus cristatus L.

145 and Bellis perennis L.), because of their bad performance, resulting in 80 investigated plots in

SC
146 this study.

147

148 2.2. Sampling and chemical analysis

U
AN
149 Soil samples were taken in April 2011, i.e. 9 years after the establishment of the grassland, on

each of the 80 plots with stainless-steel corers from the 0-0.05 m soil depth layer. Samples were
M

150

151 air-dried and sieved (< 2 mm). An aliquot of each soil sample was milled and their total carbon
D

152 (TC) concentrations were determined with an elemental analyzer (vario EL cube, Elementar
TE

153 Analysensysteme GmbH, Hanau, Germany). The inorganic C (IC) concentration of each sample

154 was also determined from an aliquot of soil after combusting soil organic carbon (SOC) in a
EP

155 muffle oven (550°C, 2 h). The SOC concentrations were quantified as the difference between the

156 TC and IC.

C

157 The concentrations of 44 PACs (29 PAHs, 15 OPAHs) in the soil from Jena, European
AC

158 reference materials (ERM-CC013a), and procedural blanks (inert sorbent, Isolute HM-N,

159 Biotage, Sweden) were determined using previously published methods (Bandowe et al., 2014a;

160 Bandowe et al., 2010; Bandowe et al., 2011; Bandowe and Wilcke, 2010; Lundstedt et al., 2014).

161 In brief, soils (11-13 g) were mixed with Isolute HM-N and transferred into 33-mL ASE

7
 ACCEPTED MANUSCRIPT

162 extraction cells. Each sample was spiked with 50 µL of each of a mixture of 11 deuterated PAHs

163 (10 µg mL-1 of each) and 2 deuterated OPAHs (20 µg mL-1 each). Each sample was then

164 extracted twice by pressurized liquid extraction with an accelerated solvent extractor (ASE, 200).

165 Dichloromethane was used for the first extraction followed by acetone:

PT
166 dichloromethane:trifluoroacetic acid (1%) [250: 125: 1 v/v/v] for the second extraction. The

RI
167 instrumental conditions for the ASE were the same as in previous work (Bandowe and Wilcke,

168 2010). The two extracts from each sample were combined, passed through Na2SO4, spiked with

SC
169 hexane and rotary evaporated until < 1 mL remained. Extracts were transferred to 3 g silica gel

170 (10% deactivated) in a 6-mL glass column. Each sample was eluted with (a) 15 mL hexane:

171
U
dichloromethane (5:1 v/v) and (b) 8 mL dichloromethane followed by 5 mL acetone. Fractions a
AN
172 and b, which contain PAHs and OPAHs, respectively, were collected in separate flasks. Each

flask was spiked with about 0.75 mL of toluene, and then rotary evaporated to < 1 mL before
M

173

174 being transferred to 2-mL vials to determine the concentration of target PACs. PAHs and
D

175 OPAHs were determined in two different runs with a gas chromatograph-mass spectrometer
TE

176 (Agilent 7890 A GC coupled to Agilent 5975 C mass spectrometer). The GC-MS was operated

177 in the electron ionization mode with selected ion monitoring of target PACs. To check the
EP

178 accuracy of our analytical procedure, we simultaneously analyzed aliquots (n = 2) of the

179 European reference material (ERM-CCO13a: Polycyclic Aromatic Hydrocarbons in Soil) from
C

180 the Federal Institute of Materials Research and Testing (BAM), Berlin, Germany. Procedural
AC

181 blanks (inert bulk sorbent: Isolute HM-N, n = 3) were also extracted and analyzed with the same

182 methods as samples and reference materials. All data recording and setting up of calibration

183 functions were done with Agilent ChemStation software. Concentrations of target compounds

184 were determined by the internal standard procedure. The average mass of target compounds in

8
 ACCEPTED MANUSCRIPT

185 blanks was deducted from that in the samples before calculating the final concentrations per dry

186 mass of extracted soil. Further details of quality control procedures are specified in previous

187 papers (Bandowe et al., 2010; Bandowe et al., 2011; Bandowe and Wilcke, 2010). Table S1 lists

188 the names and abbreviations of all analyzed PACs. Results of our quality control procedures are

PT
189 reported in the Supplementary Information.

RI
190

191 2.3. Plant and soil microbial data

SC
192 The community Leaf Area Index (LAI) per plot was determined during peak standing biomass in

193 May and August of the years 2003 to 2008 with a LAI-2000 plant canopy analyzer (LI-COR).

194
U
We calculated the mean LAI per plot for the period 2003-2008 from annual means per plot for
AN
195 further analyses. Aboveground plant community biomass was measured in May and August of

the years 2007-2010 during peak standing biomass on each of the study plots. Biomass was
M

196

197 sampled by clipping the vegetation at 3 cm above ground in four rectangles of 0.2 m x 0.5 m per
D

198 plot. The samples were dried to constant weight (70°C, >= 48 h) and weighed. We averaged the
TE

199 mean biomass per plot and year over the time period 2007-2010 for our analyses. Detailed

200 description of the methods of determination (of LAI and aboveground biomass) and the original
EP

201 data set have been published elsewhere (Weigelt et al., 2010; Weigelt et al., 2016).

202 Microbial biomass C and basal respiration were determined in soils from each plot in May or
C

203 early June of the years 2007 to 2010. Soil samples were taken with a steel corer (pooled sample
AC

204 from 5 cores per plot, depth 5 cm, diameter 5 cm) and sieved (2 mm). Basal respiration [µl O2 g–1

205 dry soil h–1] and microbial biomass carbon [mg C g–1 dry soil] were determined using an O2-

206 microcompensation apparatus following the procedures described in previous papers (Eisenhauer

207 et al., 2010; Strecker et al., 2015, 2016). Mean microbial biomass C (via substrate-induced

9
 ACCEPTED MANUSCRIPT

208 respiration) and basal respiration per plot were published and further described in Eisenhauer et

209 al. (2010) and Strecker et al. (2016). Belowground microbial community structure was

210 investigated in bulk soils sampled from each plot of the Jena Experiment in September 2010. The

211 microbial community composition was analyzed by 454-pyrosequencing. Further details of the

PT
212 methods, data processing and original data on microbial community composition can be found in

RI
213 previous publications (Dassen et al., 2017a,b). From the dataset, we determined the observed

214 number of bacterial OTUs (bacterial richness), bacterial community evenness (1-D or inverse

SC
215 Simpsons index), observed number of fungal OTUs (fungal richness), and fungal community

216 evenness (1-D or inverse Simpsons index). We further determined the sum of the relative

217
U
abundance of potentially PAH-degrading microbial groups based on literature (Cerniglia and
AN
218 Sutherland, 2010; Fernández-Luqueño et al, 2011). From the microbial datasets we could

identify the following groups: Actinobacteria (Rhodococcus spp., Cellulomonas sp.,

M

219

220 Arthrobacter sp.), Bacteriodetes (Flavobacterium spp.), Firmicutes (Bacillus spp.),

D

221 Alphaproteobacteria (Paracoccus sp., Novosphingobium spp., Sphingobium spp., Sphingomonas

TE

222 spp.), Betaproteobacteria (Burkholderia sp.), Gammaproteobacteria (Aeromonas sp., Halomonas

223 sp., Pseudomonas spp.), and Trichosporon sp.

EP

224

225 2.3. Calculations and statistical analysis

C

226 The sum of the concentrations of all analyzed PAHs is referred to as Ʃ29PAHs, including (a)16
AC

227 US-EPA PAHs as ƩUS-EPA PAHs, (b) 6 non-alkylated PAHs with 2 to 3 benzene rings as

228 ƩLMW-PAHs, (c) 15 non-alkylated PAHs with 4-7 benzene rings as ƩHMW-PAHs. The sum of

229 the concentrations of all analyzed OPAHs is referred to as ∑15OPAHs. The ratios of the

230 concentration of individual OPAHs and their related parent-PAHs were also calculated, if the

10
 ACCEPTED MANUSCRIPT

231 OPAH was a known transformation product of the PAH to which it was referred (Bamforth and

232 Singleton, 2005; Casellas et al., 1997; Cerniglia, 1993; Mahajan et al., 1994; Schocken and

233 Gibson, 1984). The concentrations of individual PACs, which were below detection limits and

234 OPAH/parent-PAH concentration ratios, which could not be calculated in soils of 40 or more

PT
235 plots, were excluded from the statistical analysis (i.e., 1,4-naphthoquinone (1,4-NQ), 1,8-

RI
236 naphthalic anhydride (1,8-NAA), 1,4-naphthoquinone/naphthalene (1,4-NQ/NAPH), 1,8-

237 naphthalic anhydride/acenaphthylene (1,8-NAA/ACENY)).

SC
238 A hierarchical ANCOVA approach was adopted to test the influence of plant community

239 composition on measured PACs concentrations (individual and sums) and OPAH/parent-PAH

240
U
concentration ratios. After the establishment of the various plant mixtures, the soil organic matter
AN
241 and PACs concentrations changed depending on the plant mixture on a specific plot.

Furthermore, while the whole experimental plot was still in contact with a homogeneous urban
M

242

243 atmosphere, the receptor properties of the surface changed in relation with the plant mixture on a
D

244 specific plot and its effects on soil properties. Our analysis aimed at detecting the effects of the
TE

245 various plant mixtures on PACs concentrations nine years after establishment of the grassland,

246 while securing homogeneous start conditions by eliminating the influence of the initial variation
EP

247 in soil organic matter concentrations. In a first ANCOVA, we therefore included (after the block

248 effect and the 2002-SOC concentrations to eliminate the variation in PACs concentrations driven
C

249 by the initial soil heterogeneity), plant species richness (log-transformed; continuous variable)
AC

250 followed by functional group richness (continuous variable) (Table S2). In a second ANCOVA,

251 we analyzed the influence of the presence or absence of specific plant functional groups by

252 including the explanatory variables in the following order: block (factor), 2002-SOC (covariate),

253 plant species richness (log-transformed; continuous variable), presence/absence of legumes,

11
 ACCEPTED MANUSCRIPT

254 grasses, tall herbs, and small herbs (factors) (Table S3). The order of the explanatory variables

255 followed the statistical design of the experiment and the expected strength of the variable’s effect

256 because legumes usually have the strongest effect on ecosystem functioning, followed by

257 grasses, and small herbs usually modify ecosystem functions the least.

PT
258 To meet the requirements of the ANCOVA, the data had to be transformed with a Box-Cox

RI
259 power transformation (Equation 1) to render the residuals of the linear model as close to normal

260 distribution as possible (R functions powerTransform() and bcPower() from package car) (Fox

SC
261 and Weisberg, 2011).

( )
, ≠0
=
log( + 1) , = 0
U
262 (Equation 1)
AN
263 The hierarchical ANCOVA was calculated with the function aov(). All statistical analyses were

264 performed with the R 3.2.0 software package (R Core Team, 2015).
M

265 Before performing of Pearson’s correlations and structural equation modeling, all data was
D

266 Box-Cox transformed according to Equation 1 to obtain normal distribution of the variables.
TE

267 Pearson's correlations were calculated with the R function cor.test(). We calculated Pearson’s

268 correlations between the measured PACs and ratios and the microbial properties (basal
EP

269 respiration, microbial biomass C, bacterial richness, bacterial community evenness, fungal

270 richness, fungal community evenness, sum of relative abundance of potentially PAH-degrading
C

271 bacteria, relative abundance of potentially PAH-degrading Bacteriodetes, Alphaproteobacteria,

AC

272 Betaproteobacteria, Gammaproteobacteria, and Trichosporon sp.), LAI, and aboveground plant

273 biomass. Structural equation modelling allows for testing direct and indirect relationships

274 between variables in a multivariate approach (Grace, 2006). Structural equation modeling was

275 performed to test the hypothesis that plant species richness effects on the 1-indanone (1-

276 INDA)/fluorene (FLUO) and 1,2-acenaphthenequinone (1,2-ACQ)/acenaphthene (ACEN)

12
 ACCEPTED MANUSCRIPT

277 concentration ratios and 1,2-ACQ operate via modifications of the microbial community or via

278 modifications of compound deposition. We considered these ratios and compound as particularly

279 indicative of the assumed stimulation of microbial PAHs degradation by plant species richness

280 As potential mediating variables in the SEM, we selected those variables that showed a

PT
281 significant (p < 0.05) relationship with plant species richness and at least one PAC concentration

RI
282 or PAC concentration ratio (Table S4), i.e. microbial biomass C, basal respiration, aboveground

283 plant biomass, and relative abundance of potentially PAH-degrading Trichosporon sp. The

SC
284 residual variances of microbial biomass C and basal respiration were allowed to be correlated by

285 including a covariance structure in the SEMs. The adequacy of the models was determined via χ2

286
U
tests and root mean square error of approximation (RMSEA). Structural equation modeling was
AN
287 performed in R using the function sem() from the package lavaan (Rosseel, 2012).
M

288 Octanol water partition coefficients (KOW) for PAHs were taken from two sources (Mackay et

289 al., 2006; Neff et al., 2005), while those of the OPAHs were estimated with KOWWIN v1.67 EPI
D

290 SuiteTM version 4.11 (US EPA: http://www.epa.gov/opptintr/exposure/pubs/episuitedl.htm).

TE

291
EP

292 3. Results

293 The concentrations of Ʃ29PAHs, ƩUS-EPA PAHs and benzo[a]pyrene (B(A)P) averaged 825 ng
C

294 g-1 (range: 260 – 2400), 677 ng g-1 (211 – 2048) and 42 ng g-1 (11 – 140), respectively. The PAH
AC

295 mixtures were on average dominated by the benzo[b+j+k]fluoranthenes (B(BJK)), pyrene

296 (PYR), phenanthrene (PHEN), and fluoranthene (FLUA) (Figure S3). The dominance of HMW-

297 PAHs is revealed by the ƩHMW/ƩLMW-PAHs ratio which averaged 5.5 (3 – 8).

13
 ACCEPTED MANUSCRIPT

298 The concentrations of Ʃ15OPAHs averaged 142 ng g-1 (range: 57-405). The OPAH mixtures

299 were dominated by 9,10-anthracenedione (9,10-ANQ), benzo[a]fluorenone (B(A)FLUone), and

300 9-fluorenone (9-FLO) (Figure S3).

301 At the time when the Jena Experiment was established, SOC concentrations ranged from 14 to

PT
302 28 g kg-1 (mean: 19 g kg-1) and were unrelated to the later established plant species richness

RI
303 gradient (Fig. S1a). The SOC concentrations in 2011 (the year of our sampling campaign) were

304 higher than in 2002, ranging from 15 – 34 g kg-1 (mean: 25 g kg-1), but still correlated with those

SC
305 of 2002 (Lange et al., 2015b). The SOC concentrations in 2002 correlated more closely with

306 those of the Σ29PAHs and Σ15OPAHs than the SOC concentrations in 2011 (r = 0.44 and 0.53

307
U
vs. 0.22 and 0.32, respectively, and p < 0.05 in all cases, Fig. S2). In 2011, SOC concentrations
AN
308 increased with increasing plant species richness (Lange et al. 2015b).
M

309 The ANCOVA revealed that after consideration of the effects of block and SOC

310 concentrations in 2002 to eliminate soil heterogeneity at the start of the experiment, higher plant
D

311 species richness significantly decreased the concentrations of 16 PAHs (including the most
TE

312 abundant compounds benzo[b+j+k]fluoranthenes (B(BJK)) and pyrene (PYR), the ∑29PAHs,

313 ƩUS-EPA PAHs, ƩLMW-PAHs, and ƩHMW-PAHs (Fig. 1, Tables 1 and S2). The higher mean
EP

314 PAH concentration of the 16-species mixtures likely are due to natural variation, because two out

315 of 14 plots with 16 species had the highest and third highest initial SOC concentrations (in 2002,
C

316 Fig. S1a), respectively. Because the sum of 29 PAHs significantly increased with the SOC
AC

317 concentrations in 2002 (Fig. S2), the initial variation in SOC concentrations is likely the reason

318 for the higher PAH concentrations in the 16-species mixtures. As we account for SOC 2002 in

319 the hierarchical ANOVA before fitting the plant species richness effect, the statistical results

320 consider and remove this natural variation. Similarly, there was a negative relationship between

14
 ACCEPTED MANUSCRIPT

321 plant species richness and the concentrations of benzo[a]fluorenone (B(A)FLUone) (the second

322 most abundant OPAH, Fig. 2c), benzo[a]anthracene-7,12-dione (7,12-B(A)A) (Fig. 2d), and

323 5,12-naphthacenedione (5,12-NACQ) (Table S2). In contrast, increasing plant species richness

324 led to significant increases in the concentrations of 1-naphthaldehyde (1-NLD) and 1,2-

PT
325 acenaphthenequinone (1,2-ACQ) (Fig. 2 a,b). Furthermore, increasing plant species richness

RI
326 resulted in significant increases in the concentration ratios 1-indanone/fluorene (1-INDA/FLUO),

327 2-biphenylcarboxaldehyde/phenanthrene (2-BPCD/PHEN), 1,2-

SC
328 acenaphthenequinone/acenaphthylene (1,2-ACQ/ACENY), and 1,2-

329 acenaphthenequinone/acenaphthene (1,2-ACQ/ACEN) (Fig. 3). There was a negative

330
U
relationship between the octanol-water partitioning coefficient (KOW) and the slope of the
AN
331 regression line of the concentrations of PAHs with four or more aromatic rings on plant species

richness, while PAHs with less than four rings did not show this relationship (Fig. 4a). For the
M

332

333 OPAHs, we found partly positive and partly negative slopes of the regressions with KOW values
D

334 (Fig. 4b). The slopes tended to be steeper with increasing KOW value, albeit these relationships
TE

335 were not significant.

336 Increasing the number of plant functional groups significantly increased the concentration of
EP

337 1-naphthaldehyde (1-NLD) and the 1-naphthaldehyde/1-methylnaphthene (1-NLD/1-MNAPH)

338 ratio (Table 1). The presence of legumes resulted in significant increases in the concentration of
C

339 BP and decreases in the benzo[a]anthracene-7,12-dione/benzo[a]anthracene (7, 12-

AC

340 B(A)A/B(A)A) ratio (Fig. S4, Table S3). The presence of grasses significantly reduced the

341 concentration of naphthalene (NAPH), but increased the concentration of 1-naphthaldehyde (1-

342 NLD) and the 1-naphthaldehyde/1-methylnaphthalene (1-NLD/1-MNAPH) ratio (Fig. S5, Table

343 S3).

15
 ACCEPTED MANUSCRIPT

344 Pearson’s correlations revealed no significant relationships between LAI and PACs

345 concentrations, but aboveground plant biomass (2007-2010) correlated positively with the

346 concentrations of two OPAHs, 1-naphthaldehyde and 1,2-acenapthenequinone (Table S4). There

347 was a significant positive correlation between plant species richness and the abundance of the

PT
348 assumed PAH-degrading species Trichosporon sp. (Table S4). Moreover, plant species richness

RI
349 was marginally significantly positively correlated with fungal richness, fungal evenness, and

350 marginally negatively correlated with relative abundance of alphaproteobacteria (Table S4).

SC
351 We tested the quality of our Structural Equation Models with the help of a χ2 test and by

U
352 considering the Root Mean Square Error of Approximation (RMSEA). According to a χ2 test, the
AN
353 data on the 1-indanone (1-INDA)/fluorene (FLUO) and 1,2-acenaphthenequinone (1,2-

354 ACQ)/acenaphthene (ACEN) and 1,2-ACQ, respectively, did not significantly deviate from the
M

355 structural equation models (for each: χ2 = 4.5, p > 0.05). The RMSEA was not significantly

different from zero for each of the models. The paths from plant species richness to microbial
D

356

357 biomass C, basal respiration, plant biomass, and the potentially PAH-degrading Trichosporon sp.
TE

358 were significant in each of the structural equation models (Fig. 5a-c, Tables S5-S7). The paths
EP

359 from basal respiration, plant biomass, Trichosporon sp., and the direct paths from plant species

360 richness to the 1-INDA/FLUO and 1,2-ACQ/ACEN ratios, and the 1,2-ACQ concentrations were
C

361 not significant, while microbial biomass C showed a marginally significant (1-INDA/FLUO and
AC

362 1,2-ACQ) or significant (1,2-ACQ/ACEN) effect (Fig. 5). This indicates that the plant species

363 richness effect on these PAC ratios and concentration was explained by microbial biomass C.

364

365 4. Discussion

16
 ACCEPTED MANUSCRIPT

366 The mean concentrations of the ƩUS-EPA PAHs in soils of the Jena Experiment were higher

367 than the median concentration of 194 ng g-1 reported for rural grassland soils, but lower than the

368 median concentration of 1100 ng g-1 in urban areas (Wilcke, 2000). The proximity of the

369 experimental site to roads and the city of Jena may explain the elevated PAH concentrations

PT
370 relative to rural grasslands (Choi et al., 2009; Wilcke, 2000). The PAH mixtures, which were on

RI
371 average dominated by the benzo[b+j+k]fluoranthenes (B(BJK)), pyrene (PYR), phenanthrene

372 (PHEN), and fluoranthene (FLUA) (Fig. S3), are typical of temperate soils in industrialized

SC
373 countries (Wilcke, 2000, 2007). The OPAHs patterns were also similar to those reported in soils,

374 street dust, and air samples from other urban, traffic, and industrial sites (Bandowe et al., 2014a;

375
U
Bandowe et al., 2014b; Bandowe and Nkansah, 2016; Bandowe et al., 2011; Wei et al., 2015a;
AN
376 Wei et al., 2015b; Wilcke et al., 2014b), because the OPAHs in the studied soils are displaying

the fingerprint of traffic and other diffuse urban sources.

M

377

378 The fact that the SOC concentrations in 2002 were unrelated with the later established plant
D

379 species richness confirmed that there was no a-priori bias of the experimental design (Fig. S1a).
TE

380 The SOC concentrations increased between 2002 and 2011, because of the transformation from a

381 plowed arable soil depleted in organic matter to a permanent grassland (Lange et al., 2015b). The
EP

382 accumulation of soil organic matter was significantly accelerated by increasing plant species

383 richness (Lange et al., 2015a). Therefore, a positive correlation between plant species richness
C

384 and SOC concentrations was observed in the year 2011 (Fig. S1b). Because the SOC
AC

385 concentrations in 2002 and 2011 were still correlated (Fig. S6), the spatial distribution of SOC

386 concentrations in soil did not change fundamentally.

387 Increasing SOC concentrations with increasing plant species richness should result in

388 increasing PACs concentrations in soil with increasing plant species richness, because under

17
 ACCEPTED MANUSCRIPT

389 equilibrium conditions, there is a stronger partitioning of PACs from the atmosphere into soils

390 with a higher SOC concentration (Wilcke and Amelung, 2000; Wilcke et al., 2014a). Moreover,

391 at our study site a higher plant species richness leads to a higher LAI (Weisser et al., 2017),

392 which should also enhance PACs partitioning into the soil because of the stronger scavenging

PT
393 from the atmosphere. However, the reverse was observed at our study site (Table 1, Fig. 1): the

RI
394 SOC concentrations in 2011 correlated less strongly with the PACs concentrations (Fig. S2), and

395 there was no correlation between the LAI and the concentrations of any PAC (Table S4).

SC
396 Therefore, we assume that the potential additional atmospheric input of PACs with increasing

397 plant species richness between 2002 and 2011 and the possibly decreased biodegradation

398
U
because of stronger sorption of PACs to the higher SOC concentrations were minor.
AN
399 The consistently negative influence of higher plant species richness on PACs concentrations

(Table 1, Fig. 1) can be explained by increasing removal of PACs bound to plant surfaces and, to
M

400

401 a minor extent, for LMW-PAHs also taken up via the roots (Collins et al., 2006) by mowing or
D

402 increased dissipation of PACs with increasing plant species richness. The higher aboveground
TE

403 plant biomass on the plots of the more diverse plant mixtures (Weisser et al., 2017) can

404 increasingly trap atmospherically deposited PACs with increasing plant species richness and
EP

405 prevent them from reaching the soils. The biomass with the PACs will then be removed from the

406 plots during mowing, which would result in a negative relationship between plant species
C

407 richness and PACs concentrations, as observed (Table 1, Fig. 1). However, as for LAI, we did
AC

408 not detect any significant relationship between aboveground biomass and PACs concentrations at

409 our study site, except for a positive correlation for 1,2-acenaphthenequinone and 1-

410 naphthaldehyde (Table S1). Thus, we rule out a strong effect of mowing on the increasing

411 dissipation of PACs with increasing plant species richness. With respect to the correlation of the

18
 ACCEPTED MANUSCRIPT

412 aboveground biomass with the two OPAHs 1,2-acenaphthenequinone and 1-naphthaldehyde, we

413 speculate that the higher aboveground biomass creates a favorable microclimatic condition for

414 the formation of these two OPAHs from their parent PAHs.

415 Enhanced dissipation of PAHs can be related to enhanced volatilization, leaching and/or

PT
416 microbial degradation. Enhanced volatilization would only affect the low molecular weight

RI
417 compounds because the high molecular weight PAHs are little volatile (Cousins et al., 1999;

418 Wang et al., 2015). Volatilization from soil should be higher in plots with lower plant diversity,

SC
419 because in the Jena Experiment, the topsoils of the plots with lower plant diversity were drier

420 until 2010 (Fischer et al. 2018), and showed a slower organic matter accumulation (Lange et al,

421
U
2015a, Fig. S1b). Both, lower soil moisture and lower organic matter concentrations should favor
AN
422 to enhanced volatilization of organic compounds from soil. Consequently, PACs, particularly

those of the LMW-PAHs, should be increasingly volatilized with decreasing plant species
M

423

424 richness. Our results are contrary to that and therefore suggest that volatilization is not a major
D

425 driver of the plant diversity-PACs concentrations pattern (Table 1), which is further corroborated
TE

426 by the fact that we did also not observe different species-richness effects on LMW- and HMW-

427 PAHs (Table 1, Fig. 1)

EP

428 Enhanced leaching with increasing plant species richness is unlikely, because in the first five

429 years of the Jena Experiment, downward water fluxes were not related with plant species
C

430 richness (Leimer et al., 2014). Increased leaching in species-rich plots might, however, be
AC

431 expected in later years because Fischer et al. (2015) showed that in 2012 there was a positive

432 relationship between plant species richness and infiltration rates. Furthermore, soil moisture

433 decreased with increasing plant species richness from 2010 onward (Fischer et al., 2018). Fischer

434 et al. (2015, 2018) attributed these findings to enhanced soil aggregation by the higher plant

19
 ACCEPTED MANUSCRIPT

435 species richness in the long run, which might result in more preferential flow-related downward

436 transport of PACs. Our assumption that leaching did not play a major role from 2002 to 2011 is

437 supported by the fact that there was also no significant correlation between plant species richness

438 and the concentrations of LMW-PAHs, which would be leached preferentially (Wilcke, 2000).

PT
439 It has been reported from our experimental site that increasing plant species richness enhances

RI
440 microbial activity (microbial biomass C and basal respiration) since the fifth vegetation period

441 (i.e. since the year 2006, Eisenhauer et al., 2010; Eisenhauer et al., 2011; Strecker et al., 2016).

SC
442 The higher microbial biomass C and activity with increasing plant species richness was

443 attributed to the fact that the topsoil under species-rich mixtures was moister than under species-

444
U
poor ones until 2010 (Fischer et al., 2018; Rosenkranz et al., 2012). Increasing microbial
AN
445 biomass C and activity likely result in enhanced biodegradation of PACs (Ai et al., 2018). The

latter occurs to a small degree metabolically (only low molecular weight compounds) but mostly
M

446

447 co-metabolically (Bamforth and Singleton, 2005; Cerniglia, 1993; Peng et al., 2008).
D

448 At the field site of the Jena Experiment, increasing plant species richness leads to enhanced
TE

449 input of root-derived exudates into the soil (Lange et al., 2015a; Lange et al., 2014). The

450 enhanced input of root exudates can specifically stimulate the degradation of PAHs, beyond the
EP

451 general stimulation of microbial activity, because root exudates can improve the bioavailability

452 of aged PACs (through desorption), with the overall consequence of enhanced degradation of
C

453 PACs in soils (Aken et al., 2009; Alkorta and Garbisu, 2001; Cébron et al., 2011; Gao et al.,
AC

454 2011; Pilon-Smits, 2005; Tajedo-Agredano et al., 2013).

455 However, not all soil-borne microorganisms have the ability to degrade PAHs. Hence, various

456 studies have isolated and identified specific microorganisms that can degrade PAHs (Tajedo-

457 Agredano et al., 2013; Fernández-Luqueno et al., 2011; Peng et al., 2008; Cerniglia, 1993). The

20
 ACCEPTED MANUSCRIPT

458 relationship between plant diversity and microbial community composition (analyzed by

459 pyrosequencing) in the soils of the Jena Experiment (sampled in 2011) was recently reported

460 (Dassen et al., 2017 a, b). Hence, we examined the relationship between PACs concentrations

461 and soil microbial biomass C, basal respiration and community composition as possible

PT
462 pathways for plant species richness effects on PACs concentrations. Our assumptions were that

RI
463 (1.) higher plant species richness leads to higher microbial biomass C and activity and hence

464 stronger degradation of PAHs resulting in lower PAH concentrations, and (2.) higher plant

SC
465 species richness leads to more diverse soil microbial communities capable of degrading PAHs,

466 which may result in a lower concentration of PAHs but higher concentrations of OPAHs that are

467 PAH metabolites.

U
AN
468 The three structural equation models consistently suggest that the mechanism that links plant

species richness to increases the two concentrations ratios and the concentration of 1,2-
M

469

470 acenaphthenequinone is the plant diversity-induced increase in microbial biomass C, while basal
D

471 respiration, aboveground plant biomass and the relative abundance of Trichosporon sp. did not
TE

472 contribute significantly to the explanation of the plant species richness effect (Fig. 5a-c). Thus,

473 the structural equation modeling provides further support for the assumption that enhanced
EP

474 microbial biomass C is the main pathway that explains the positive plant species richness effect

475 on the dissipation of PAHs.

C

476 Enhanced biodegradation is further supported by our findings of increasing OPAH/parent-

AC

477 PAH ratios (Fig. 3) and by the negative relationship between the KOW values and the slopes of

478 the regression lines of individual PAH concentrations on plant species richness (Fig. 4a) for

479 PAHs with four or more rings. The fact that the 2-3 ring PAHs did not fit into the line shown in

480 Fig. 4, might be attributable to the particularly high susceptibility of these compounds to

21
 ACCEPTED MANUSCRIPT

481 leaching, plant uptake, and more dynamic air-soil exchanges (as a result of rapid changes in the

482 air-soil fugacity ratios), which might confound a degradation effect (Cabrerizo et al., 2011). The

483 positive slopes (for a number of OPAHs), which indicate the formation of these OPAHs as

484 metabolic products of PAHs, support our interpretation of microbial PACs turnover as main

PT
485 explanation of the relationship between plant species richness and PACs concentrations in soil

RI
486 (Fig. 4b, Fig. 5). The negative slope of (mainly) HMW-OPAHs (i.e. OPAHs with ≥ 4 rings)

487 instead suggest that these compounds are deposited to soil from combustion sources as primary

SC
488 OPAHs or formed in the atmosphere from post-emission transformation of PAHs as secondary

489 OPAHs. Their extent of accumulation or persistence in the soil is similar to those of the HMW-

490 PAHs.
U
AN
491 Higher plant species richness favored the accumulation of particularly LMW- OPAHs (i.e.

OPAHs with 1-3 rings) including 1-naphthaldehyde and 1, 2-acenaphthenequinone but decreased
M

492

493 the concentrations of some high HMW-OPAHs, such as benzo[a]fluorenone (B(A)FLUone) and
D

494 benzo[a]anthracene-7,12-dione (Fig. 2 and 4b). We interpret this finding as an indication of the
TE

495 production of LMW-OPAHs as metabolites from the degradation of PAHs and the dominance of

496 the degradation of HMW-OPAHs after being deposited to our study soil. 1-Naphthaldehyde has
EP

497 been isolated as a product formed from the microbial degradation of 1-methylnaphthalene

498 (Mahajan et al., 1994). 1,2-Acenaphthenequinone is a metabolic product from the microbial
C

499 degradation of acenaphthylene and acenaphthene (Schocken and Gibson, 1984) and was also
AC

500 reported to be formed from PAHs in soils after 19-weeks incubation (Wilcke et al., 2014b). Both,

501 benzo[a]fluorenone (B(A)FLUone) and benzo[a]anthracene-7,12-dione (7,12-B(A)A) belonged

502 to the compounds that were predominantly degraded in the study of Wilcke et al. (2014b). We

503 conclude that increasing plant species richness and the associated enhanced biodegradation of

22
 ACCEPTED MANUSCRIPT

504 PAHs results in the net accumulation of some LMW-OPAHs while the HMW-OPAHs are

505 increasingly degraded like PAHs.

506 Only few PACs concentrations and ratios were significantly affected by functional group

507 richness (1-naphthaldehyde (NLD) and 1-NLD/1-methylnaphthalene (1-MNAPH) ), presence of

PT
508 legumes (biphenyl (BP) and 7,12-benz(a)anthracene-7,12-dione (7,12-B(A)A)/benz(a)anthracene

RI
509 (B(A)A)) or presence of grasses (napththalene (NAPH), 1-NLD and 1-NLD/1-MNAPH; Table 1,

510 Figs. S4, S5). The observation that plant functional group richness has little effect on PACs

SC
511 concentrations and concentration ratios in soils is in line with the results of a recent study (Ai et

512 al., 2018). Functional group richness and presence of grasses seemed to favor the accumulation

513
U
of 1-naphthaldehyde (1-NLD), which might indicate its enhanced formation from the
AN
514 biodegradation of 1-methylnaphthalene (1-MNAPH). Our interpretation is further supported by

the increased 1-naphthaldehyde (1-NLD)/1-methylnaphthalene (1-MNAPH) ratios. Naphthalene

M

515

516 (NAPH) was the only PAC with lower concentrations in the presence of grasses. The weak effect
D

517 of grasses on PACs concentrations was unexpected, because several previous experiments have
TE

518 shown that grasses are capable of enhancing the removal of PAHs from soil through increased

519 microbial activity around their rhizosphere, secretion of exudates that improve bioavailability,
EP

520 and their extensive root morphology (Aprill and Sims, 1990; Binet et al., 2000; Dzantor et al.,

521 2000; Günther et al., 1996; Lee et al., 2008; Phillips, 2008). Binet et al. (2000) studied the
C

522 degradation of PAHs in the rhizosphere of ryegrass and detected the formation of the OPAHs
AC

523 (anthracene-9,10-dione, benzo[a]anthracene-7,12-dione) from the degradation of PAHs. The

524 increase in biphenyl concentrations in the presence of legumes might be explained by several

525 processes including (i) their release by legumes, (ii) their deposition in the presence of legumes,

526 and (iii) their decreased dissipation in the presence of legumes. The positive effect of legumes on

23
 ACCEPTED MANUSCRIPT

527 the benzo[a]anthracene-7,12-dione/benzo[a]anthracene concentration ratio either suggests that

528 benzo[a]anthracene-7,12-dione is also produced by legumes or that the benzo[a]anthracene

529 concentrations are specifically decreased. Legumes have also been previously reported to favor

530 phytoremediation of PAHs (Dzantor et al., 2000; Lee et al., 2008; Phillips, 2008).

PT
531

RI
532 5. Conclusions

533 Our study demonstrates that increasing plant species richness results in decreasing

SC
534 concentrations of PAHs and HMW-OPAHs in soils of a grassland. Consequently, species-rich

grassland mixtures can favor the natural attenuation of PACs-contaminated soils, and that the

U
535

main underlying mechanism is elevated soil microbial biomass. This can be applied in
AN
536

537 remediation efforts. However, we also demonstrate that some low molecular weight OPAHs
M

538 accumulate in soil during the degradation of PAHs. This might increase ecotoxicological risks

539 associated with these compounds and therefore requires thorough monitoring during
D

540 phytoremediation. We recommend further studies on the PACs in soils sampled from the Jena
TE

541 experimental sites after 2011.

542
EP

543 Acknowledgments
C

544 We thank the many people who helped with the management of the experiment and in particular
AC

545 the initiators, E.-D. Schulze, B. Schmid, and W. W. Weisser, as well as the scientific

546 coordinators C. Roscher, A. Weigelt, and A. Ebeling. Thanks also to all the helpers who assisted

547 during the weeding campaigns. We thank A. Ebeling, O. González, C. Roscher, S. Scheu, T.

548 Strecker, V.M. Temperton, A. Weigelt, George Kowalchuk, Wim van der Putten and Gerlinde

549 De Deyn for contributing data. We gratefully acknowledge financial support from the Swiss

24
 ACCEPTED MANUSCRIPT

550 National Science Foundation (SNF 200021_131938/1, SNF 200021E_131195/1) and the

551 Deutsche Forschungsgemeinschaft (DFG FOR1415, Wi1601/20-1) and additional support from

552 the Friedrich Schiller University Jena and the Max Planck Society. B.A.M. Bandowe

553 acknowledges the P.R.I.M.E. Fellowship from the German Academic Exchange Service. N.E.

PT
554 acknowledges funding by the German Centre for Integrative Biodiversity Research (iDiv) Halle-

RI
555 Jena-Leipzig funded by the German Research Foundation (DFG; FZT 118).

U SC
AN
M
D
TE
C EP
AC

25
 ACCEPTED MANUSCRIPT

556 Table 1. Overview of PACs concentrations and OPAH/parent-PAH concentration ratios that are
557 significantly influenced by plant species richness, functional group richness, presence of
558 legumes, and presence of grasses. Positively affected compound concentrations are underlined
559 and in italic letters. All other compound concentrations were negatively affected. Detailed
560 statistical results are shown in Tables S2 and S3.

PT
561
OPAH/parent-PAH
PAHs OPAHs
ratio

RI
NAPH, ACENY, FLUA,
PYR, B(A)A, CHRY,

SC
B(BJK), B(E)P, PERY, 1-INDA/FLUO, 2-
Plant species 1-NLD, 1,2-ACQ,
IND, DIBE, B(GHI), BPCD/PHEN, 1,2-
richness (log- B(A)FLUone, 7,12-
ACQ/ACENY, 1,2-
transformed) COR, Σ29PAHs, ΣUS- B(A)A, 5,12-NACQ
ACQ/ACEN
EPA PAHs, ΣLMW-

U
PAHs, ΣHMW-PAHs
AN
Functional group
- 1-NLD 1-NLD/1-MNAPH
richness

Presence of legumes BP - 7,12-B(A)A/B(A)A

M

Presence of grasses NAPH 1-NLD 1-NLD/1-MNAPH

D

562 Abbreviations: Naphthalene (NAPH), 1-methylnaphthalene (1-MNAPH), biphenyl (BP), acenaphthylene (ACENY),
TE

563 acenaphthene (ACEN), fluorene (FLUO), phenanthrene (PHEN), fluoranthene (FLUA), Pyrene (PYR),
564 benzo(a)anthracene (B(A)A), chrysene+triphenylene (CHRY), benzo(b+j+k)fluoranthenes (B(BJK)),
565 benzo(e)pyrene (B(E)P), perylene (PERY), indeno(1,2,3-cd)pyrene (IND), dibenzo(a,h)anthracene (DIBE),
566 benzo(ghi)perylene (B(GHI)), coronene (COR), 1-indanone (1-INDA), 1-naphthaldehyde (1-NLD), 1,2-
EP

567 acenaphthenequinone (1,2-ACQ), benzo(a)fluorenone (B(A)FLUone), benzo(a)anthracene-7,12-dione (7,12-

568 B(A)Dione), 5, 12-naphthacenequinone (5,12-NACQ)
C
AC

26
 ACCEPTED MANUSCRIPT

569 Figure 1. Relationship between log-transformed plant species richness and concentrations in ng
570 g-1 of a) ∑29PAHs, (black; ANCOVA: F = 5.23, p = 0.025) and ∑US-EPA PAHs (gray; F =
571 5.13, p = 0.026), b) phenanthrene (PHEN; F = 3.96, p = 0.050), the third most abundant
572 individual PAH (cf. Fig. S3), c) pyrene (PYR; F = 4.65, p = 0.034), the second most abundant
573 individual PAH (cf. Fig. S3) and d) the benzo[b+j+k]fluoranthenes (B(BJK)); F = 5.30, p =

PT
574 0.024), the most abundant PAHs. The points display means and the whiskers the standard error.

RI
575 Figure 2. Relationship between log-transformed plant species richness and concentrations in ng
576 g-1 of a) 1-naphthaldehyde (1-NLD; ANCOVA: F = 4.95, p = 0.030), b) 1,2-

SC
577 acenaphthenequinone (1,2-ACQ; F = 6.78, p = 0.011) and c) benzo[a]fluorenone (B(A)FLUone;
578 F = 4.31, p = 0.041), and d) benzo[a]anthracene-7,12-dione (7,12-B(A)A; F = 4.25, p = 0.043).
579 The points display means and the whiskers the standard error.

580
U
Figure 3. Relationship between log-transformed plant species richness and (unitless)
AN
581 OPAH/parent PAH concentration ratios of a) 1-indanone (1-INDA)/fluorene (FLUO; ANCOVA:
582 F = 4.84, p = 0.031), b) 2-biphenylcarboxxaldehyde (2-BPCD)/phenanthrene (PHEN; F = 6.02, p
M

583 = 0.016), c) 1,2-acenaphthenequinone (1,2-ACQ)/acenaphthylene (ACENY; F = 13.12, p =

584 0.001), and d) 1,2-acenaphthenequinone (1,2-ACQ)/acenaphthene (ACEN; F = 9.11, p = 0.004).
D

585 The points display means and the whiskers the standard error.
TE

586 Figure 4. Relationship between log-transformed octanol-water-partitioning coefficients (KOW)

587 and the slopes of regression lines of the concentrations of a) PAHs on log-transformed plant
588 species richness (the red triangles show the PAHs with <4 rings and blue dots the PAHs >4
EP

589 rings) and b) OPAHs. The dotted regression line in b) illustrates a non-significant tendency. KOW
590 values of PAHs were taken from two sources (Mackay et al., 2006; Neff et al., 2005), while
C

591 those of the OPAHs were estimated with KOWWINv1.67EPI SuiteTM version 4.11 (US EPA:
AC

592 http://www.epa.gov/opptintr/exposure/pubs/episuitedl.htm). Chrysene+Triphenylene and the

593 benzo[b+j+k]fluoranthenes were omitted because we were unable to separate the individual
594 compounds with our analytical method and therefore could not assign KOW values for these
595 compound mixtures. See Table S1 for an explanation of the compound abbreviations.

596 Figure 5. Structural equation model of the plant species richness effect via soil microbial
597 biomass C, soil basal respiration, aboveground plant biomass, and relative abundance of the

27
 ACCEPTED MANUSCRIPT

598 potentially PAH-degrading microorganism species Trichosporon sp. on the concentration ratios
599 of a) 1-indanone (1-INDA)/fluorene (FLUO) and b) 1,2-acenaphthenequinone (1,2-
600 ACQ)/acenaphthene (ACEN), and c) the 1,2-ACQ concentrations. Solid black arrows indicate
601 significant, dashed arrows marginally significant, and grey arrows non-significant relationships.
602 Data was Box-Cox transformed to obtain normal distribution. Numbers on arrows indicate

PT
603 standardized path coefficients (for unstandardized path coefficients see Table S5) with the
604 significance levels ., p < 0.10, *, p < 0.05, **, p < 0.01, ***, p < 0.001. Percentages indicate the

RI
605 fraction of the total data variance explained by the model.

U SC
AN
M
D
TE
C EP
AC

28
 ACCEPTED MANUSCRIPT

Figure 1

PT
RI
U SC
AN
M
D
TE
EP
C
AC

29
 ACCEPTED MANUSCRIPT

Figure 2

PT
RI
U SC
AN
M
D
TE
EP
C
AC

30
 ACCEPTED MANUSCRIPT

Figure 3

PT
RI
U SC
AN
M
D
TE
EP
C
AC

31
 ACCEPTED MANUSCRIPT

Figure 4

PT
RI
U SC
AN
M
D
TE
EP
C
AC

32
 ACCEPTED MANUSCRIPT

Figure 5

PT
RI
U SC
b
AN
M
D
TE
EP

c
C
AC

33
 ACCEPTED MANUSCRIPT

References
Ai, F., Eisenhauer, N., Jousset, A., Butenschoen, O., Ji, R., Guo, H., 2018. Elevated tropospheric
CO2 and O3 concentrations impair organic pollutant removal from grassland soil. Scientific
Reports 8, 5519.

Aken, B.V., Correa, P.A., Schnoor, J.L., 2009. Phytoremediation of polychlorinated biphenyls:

PT
new trends and promises. Environmental Science & Technology 44, 2767-2776.

Alkorta, I., Garbisu, C., 2001. Phytoremediation of organic contaminants in soils. Bioresource

RI
Technology 79, 273-276.

Aprill, W., Sims, R.C., 1990. Evaluation of the use of prairie grasses for stimulating polycyclic

SC
aromatic hydrocarbon treatment in soil. Chemosphere 20, 253-265.

Arp, H.P.H., Lundstedt, S., Josefsson, S., Cornelissen, G., Enell, A., Allard, A.-S., Kleja, D.B.,
2014. Native oxy-PAHs, N-PACs, and PAHs in historically contaminated soils from Sweden,

U
Belgium, and France: their soil-porewater partitioning behavior, bioaccumulation in Enchytraeus
crypticus, and bioavailability. Environmental Science & Technology 48, 11187-11195.
AN
Arthur, E.L., Rice, P.J., Rice, P.J., Anderson, T.A., Baladi, S.M., Henderson, K.L., Coats, J.R.,
2005. Phytoremediation—an overview. Critical Reviews in Plant Sciences 24, 109-122.
M

Baek, S.O., R.A. Field, M.E. Goldstone, P.W. Kirk, J.N. Lester, and R. Perry., 1991. A review of
atmospheric polycyclic aromatic hydrocarbons: sources, fate, and behavior. Water Air & Soil
Pollution 60, 279-300.
D

Bamforth, S.M., Singleton, I., 2005. Bioremediation of polycyclic aromatic hydrocarbons:

TE

current knowledge and future directions. Journal of Chemical Technology and Biotechnology 80,
723-736.
EP

Bandowe, B.A.M., Lueso, M.G., Wilcke, W., 2014a. Oxygenated polycyclic aromatic
hydrocarbons and azaarenes in urban soils: a comparison of a tropical city (Bangkok) with two
temperate cities (Bratislava and Gothenburg). Chemosphere 107, 407-414.
C

Bandowe, B.A.M., Meusel, H., Huang, R.-J., Ho, K., Cao, J., Hoffmann, T., Wilcke, W., 2014b.
PM 2.5-bound oxygenated PAHs, nitro-PAHs and parent-PAHs from the atmosphere of a Chinese
AC

megacity: seasonal variation, sources and cancer risk assessment. Science of the Total
Environment 473, 77-87.

Bandowe, B.A.M., Nkansah, M.A., 2016. Occurrence, distribution and health risk from
polycyclic aromatic compounds (PAHs, oxygenated-PAHs and azaarenes) in street dust from a
major West African Metropolis. Science of the Total Environment 553, 439-449.

34
 ACCEPTED MANUSCRIPT

Bandowe, B.A.M., Shukurov, N., Kersten, M., Wilcke, W., 2010. Polycyclic aromatic
hydrocarbons (PAHs) and their oxygen-containing derivatives (OPAHs) in soils from the Angren
industrial area, Uzbekistan. Environmental Pollution 158, 2888-2899.

Bandowe, B.A.M., Sobocka, J., Wilcke, W., 2011. Oxygen-containing polycyclic aromatic
hydrocarbons (OPAHs) in urban soils of Bratislava, Slovakia: patterns, relation to PAHs and
vertical distribution. Environmental Pollution 159, 539-549.

PT
Bandowe, B.A.M., Wilcke, W., 2010. Analysis of polycyclic aromatic hydrocarbons and their
oxygen-containing derivatives and metabolites in soils. Journal of Environmental Quality 39,

RI
1349-1358.

Binet, P., Portal, J., Leyval, C., 2000. Dissipation of 3–6-ring polycyclic aromatic hydrocarbons

SC
in the rhizosphere of ryegrass. Soil Biology and Biochemistry 32, 2011-2017.

Cabrerizo, A., Dachs, J., Moeckel, C., Ojeda, M.A.-J., Caballero, G., Barceló, D., Jones, K.C.,
2011. Ubiquitous net volatilization of polycyclic aromatic hydrocarbons from soils and

U
parameters influencing their soil− air partitioning. Environmental Science & Technology 45,
4740-4747.
AN
Casellas, M., Grifoll, M., Bayona, J.M., Solanas, A.M., 1997. New metabolites in the
degradation of fluorene by Arthrobacter sp. strain F101. Applied and Environmental
Microbiology 63, 819-826.
M

Cébron, A., Louvel, B., Faure, P., France‐Lanord, C., Chen, Y., Murrell, J.C., Leyval, C., 2011.
Root exudates modify bacterial diversity of phenanthrene degraders in PAH‐polluted soil but not
D

phenanthrene degradation rates. Environmental Microbiology 13, 722-736.

TE

Cerniglia, C.E., 1993. Biodegradation of polycyclic aromatic hydrocarbons. Current Opinion in

Biotechnology 4, 331-338.
EP

Cerniglia, C.E., Sutherland, J.B., 2010.Degradation of polycyclic aromatic hydrocarbons by

Fungi. In: Timmis, K.N. (Eds.), Handbook of Hydrocarbon and Lipid Microbiology. Springer-
Verlag, Berlin-Heidelberg, pp. 2080-2101
C

Chibwe, L., Geier, M.C., Nakamura, J., Tanguay, R.L., Aitken, M.D., Simonich, S.L.M., 2015.
Aerobic bioremediation of PAH contaminated soil results in increased genotoxicity and
AC

developmental toxicity. Environmental Science & Technology 49, 13889-13898.

Choi, S.-D., Shunthirasingham, C., Daly, G.L., Xiao, H., Lei, Y.D., Wania, F., 2009. Levels of
polycyclic aromatic hydrocarbons in Canadian mountain air and soil are controlled by proximity
to roads. Environmental Pollution 157, 3199-3206.

Collins, C., Fryer, M., Grosso, A., 2006. Plant uptake of non-ionic organic chemicals.
Environmental Science & Technology 40, 45-52.

35
 ACCEPTED MANUSCRIPT

Cousins, I.T., Beck, A.J., Jones, K.C., 1999. A review of the processes involved in the exchange
of semi-volatile organic compounds (SVOC) across the air–soil interface. Science of the Total
Environment 228, 5-24.

Dai, Y., Wu, Y., Ding, Q., Zeng, J., Li, X., Zheng, J., Lin, X., 2018. Oxygenated derivative is
more influential than unsubstituted polycyclic aromatic hydrocarbon on ammonia-oxidizing
archaea in an acidic soil. Journal of Soils and Sediments 18, 2573-2580.

PT
Dassen, S., Cortois, R., Martens, H., de Hollander, M., Kowalchuk, G.A., Van der Putten, W.H.,
De Deyn, G.B., 2017a. Differential responses of soil bacteria, fungi, archaea and protists to plant

RI
species richness and plant functional group identity. Microbial Ecology 26, 4085-4098.

Dassen, S., Cortois, R., Martens, H., de Hollander, M., Kowalchuk, G.A., van der Putten, W.H.,

SC
de Deyn, G.B., 2017b. Sequences of the 16S and 18S gene and pH determination of the Jena
experiment main plots measured in 2010. PANGAEA,
http://doi.org/10.1594/PANGAEA.874990

U
Desalme, D., Binet, P., Chiapusio, G.V., 2013. Challenges in tracing the fate and effects of
atmospheric polycyclic aromatic hydrocarbon deposition in vascular plants. Environmental
AN
Science & Technology 47, 3967-3981.

Dzantor, E.K., Chekol, T., Vough, L., 2000. Feasibility of using forage grasses and legumes for
phytoremediation of organic pollutants. Journal of Environmental Science & Health Part A 35,
M

1645-1661.

Eisenhauer, N., Beßler, H., Engels, C., Gleixner, G., Habekost, M., Milcu, A., Partsch, S.,
D

Sabais, A., Scherber, C., Steinbeiss, S., 2010. Plant diversity effects on soil microorganisms
support the singular hypothesis. Ecology 91, 485-496.
TE

Eisenhauer, N., Milcu, A., Sabais, A.C., Bessler, H., Brenner, J., Engels, C., Klarner, B.,
Maraun, M., Partsch, S., Roscher, C., 2011. Plant diversity surpasses plant functional groups and
EP

plant productivity as driver of soil biota in the long term. PLoS One 6, e16055.

Ellenberg, H., Leuschner, C., 2010. Vegetation Mitteleuropas mit den Alpen: in ökologischer,
dynamischer und historischer Sicht, 6th Edition ed. Ulmer, Stuttgart, Germany.
C

Fatiadi, A.J., 1967. Effects of temperature and of ultraviolet radiation on pyrene absorbed on
AC

garden soil. Environmental Science & Technology 1, 570-572.

Fernández-Luqueno, F., Valenzuela-Encinas, C., Marsch, R., Martinez-Suárez, C., Vázquez-

Nunez, E., Dendooven, L., 2011. Microbial communitities to mitigate contamination of PAHs in
soils – possibilities and challenges: a review. Environmental Science and Pollution Research 18,
12-30.

Fischer, C., Tischer, J., Roscher, C., Eisenhauer, N., Ravenek, J., Gleixner, G., Attinger, S.,
Jensen, B., de Kroon, H., Mommer, L., Scheu, S., Hildebrandt, A., 2015. Plant species diversity

36
 ACCEPTED MANUSCRIPT

affects infiltration capacity in an experimental grassland through changes in soil properties. Plant
and Soil 397, 1-16. doi: 10.1007/s11104-014-2373-5

Fischer, C., Leimer, S., Roscher, C., Ravenek, J., de Kroon, H., Kreutziger, Y., Baade, J., Beßler,
H., Eisenhauer, N., Weigelt, A., Mommer, L., Lange, M., Gleixner, G., Wilcke, W., Schröder,
B., Hildebrandt, A., 2018. Plant species richness and functional groups have different effects on
soil water content in a decade-long grassland experiment. Journal of Ecology, doi:

PT
10.1111/1365-2745.13046

Fox, J., Weisberg, S., 2011. An R companion to applied regression. R package version 2.0-10.

RI
Gao, Y., Yang, Y., Ling, W., Kong, H., Zhu, X., 2011. Gradient distribution of root exudates and
polycyclic aromatic hydrocarbons in rhizosphere soil. Soil Science Society of America Journal

SC
75, 1694-1703.

Grace, J.B., 2006. Structural equation modelling and natural systems. Cambridge, UK:
Cambridge University Press.

U
Günther, T., Dornberger, U., Fritsche, W., 1996. Effects of ryegrass on biodegradation of
AN
hydrocarbons in soil. Chemosphere 33, 203-215.

Hoffmann, K., Bivour, W., Früh, B., Koßmann, M., Voß, P.-H., 2014. Klimauntersuchungen in
Jena für die Anpassung an den Klimawandel und seine erwarteten Folgen.
M

IARC, 2010. Some non-heterocyclic polycyclic aromatic hydrocarbons and some related
exposures. IARC monographs on the evaluation of carcinogenic risks to humans 92, 1-853.
D

IUSS Working Group, World Reference Base for Soil Resource., 2014. International soil
TE

classification system for naming soils and creating legends for soil maps. World Soil Resources
Reports No, 106, FAO, Rome.
EP

Keyte, I.J., Harrison, R.M., Lammel, G., 2013. Chemical reactivity and long-range transport
potential of polycyclic aromatic hydrocarbons–a review. Chemical Society Reviews 42, 9333-
9391.
C

Lange, M., Eisenhauer, N., Sierra, C.A., Bessler, H., Engels, C., Griffiths, R.I., Mellado-
Vázquez, P.G., Malik, A.A., Roy, J., Scheu, S., 2015a. Plant diversity increases soil microbial
AC

activity and soil carbon storage. Nature Communications 6, 6707.

Lange, M., Habekost, M., Eisenhauer, N., Roscher, C., Bessler, H., Engels, C., Oelmann, Y.,
Scheu, S., Wilcke, W., Schulze, E.-D., 2014. Biotic and abiotic properties mediating plant
diversity effects on soil microbial communities in an experimental grassland. PloS one 9,
e96182.

37
 ACCEPTED MANUSCRIPT

Lange, M., Steinbeiss, S., Habekost, M., Gleixner, G., Luo, G., Guderle, M., Meyer, S.T., 2015b.
Collection of data on soil carbon (particulate and dissolved) in the Jena Experiment (Main
experiment, time series since 2002). PANGAEA, htttps://doi.org/10.1594/PANGAEA.848946.

Lee, S.-H., Lee, W.-S., Lee, C.-H., Kim, J.-G., 2008. Degradation of phenanthrene and pyrene in
rhizosphere of grasses and legumes. Journal of Hazardous Materials 153, 892-898.

PT
Lundstedt, S., Bandowe, B., Wilcke, W., Boll, E., Christensen, J.H., Vila, J., Grifoll, M., Faure,
P., Biache, C., Lorgeoux, C., Larsson, M., Frech Irgum, K., Ivarsson, P., Ricci, M., 2014. First
intercomparison study on the analysis of oxygenated polycyclic aromatic hydrocarbons (oxy-

RI
PAHs) and nitrogen heterocyclic polycyclic aromatic compounds (N-PACs) in contaminated
soil. Trends in Analytical Chemistry 57, 83-92.

SC
Lundstedt, S., White, P.A., Lemieux, C.L., Lynes, K.D., Lambert, I.B., Öberg, L., Haglund, P.,
Tysklind, M., 2007. Sources, fate, and toxic hazards of oxygenated polycyclic aromatic
hydrocarbons (PAHs) at PAH-contaminated sites. AMBIO: A Journal of the Human
Environment 36, 475-485.

U
Mackay, D., Shiu, W.-Y., Ma, K.-C., Lee, S.C., 2006. Handbook of physical-chemical properties
AN
and environmental fate for organic chemicals. CRC Press.

Mahajan, M.C., Phale, P.S., Vaidyanathan, C.S., 1994. Evidence for the involvement of multiple
pathways in the biodegradation of 1-and 2-methylnaphthalene by Pseudomonas putida CSV86.
M

Archives of Microbiology 161, 425-433.

Marquard, E., Weigelt, A., Temperton, V.M., Roscher, C., Schumacher, J., Buchmann, N.,
D

Fischer, M., Weisser, W.W., Schmid, B., 2009. Plant species richness and functional
composition drive overyielding in a six‐year grassland experiment. Ecology 90, 3290-3302.
TE

Menichini, E., 1992. Urban air pollution by polycyclic aromatic hydrocarbons: levels and
sources of variability. Sci. Tot. Environ. 116, 109-135.
EP

Neff, J.M., Stout, S.A., Gunster, D.G., 2005. Ecological risk assessment of polycyclic aromatic
hydrocarbons in sediments: identifying sources and ecological hazard. Integrated Environmental
Assessment and Management 1, 22-33.
C

Oelmann, Y., Buchmann, N., Gleixner, G., Habekost, M., Roscher, C., Rosenkranz, S., Schulze,
AC

E.D., Steinbeiss, S., Temperton, V.M., Weigelt, A., 2011. Plant diversity effects on aboveground
and belowground N pools in temperate grassland ecosystems: development in the first 5 years
after establishment. Global Biogeochemical Cycles 25, GB2014, 1-11.

Peng, R.-H., Xiong, A.-S., Xue, Y., Fu, X.-Y., Gao, F., Zhao, W., Tian, Y.-S., Yao, Q.-H., 2008.
Microbial biodegradation of polyaromatic hydrocarbons. FEMS Microbiology Reviews 32, 927-
955.

38
 ACCEPTED MANUSCRIPT

Phillips, L.L.A., 2008. The relationship between plants and their root-associated microbial
communities in hydrocarbon phytoremediation systems. University of Saskatchewan, Saskatoon,
Canada.

Pilon-Smits, E., 2005. Phytoremediation. Annual Review of Plant Biology 56, 15-39.

R Core Team, 2015. R: A language and environment for statistical computing. Vienna, Austria:

PT
R Foundation for Statistical Computing. URL: https://www.R-project.org/

Roscher, C., Schumacher, J., Baade, J., Wilcke, W., Gleixner, G., Weisser, W.W., Schmid, B.,

RI
Schulze, E.-D., 2004. The role of biodiversity for element cycling and trophic interactions: an
experimental approach in a grassland community. Basic and Applied Ecology 5, 107-121.

SC
Rosenkranz, S., Wilcke, W., Eisenhauer, N., Oelmann, Y., 2012. Net ammonification as
influenced by plant diversity in experimental grasslands. Soil Biology and Biochemistry 48, 78-
87.

U
Rosseel Y., 2012. Lavaan: An R Package for Structural Equation Modeling. Journal of Statistical
Software 48, 1-36.
AN
Schocken, M.J., Gibson, D.T., 1984. Bacterial oxidation of the polycyclic aromatic hydrocarbons
acenaphthene and acenaphthylene. Applied and Environmental Microbiology 48, 10-16.
M

Semple, K.T., Morriss, A., Paton, G.I., 2003. Bioavailability of hydrophobic organic
contaminants in soils: fundamental concepts and techniques for analysis. European Journal of
Soil Science 54, 809-818.
D

Spehn, E., Hector, A., Joshi, J., Scherer-Lorenzen, M., Schmid, B., Bazeley-White, E.,
TE

Beierkuhnlein, C., Caldeira, M., Diemer, M., Dimitrakopoulos, P., 2005. Ecosystem effects of
biodiversity manipulations in European grasslands. Ecological Monographs 75, 37-63.
EP

Strecker, T., Macé, O.G., Scheu, S., Eisenhauer, N., 2016. Functional composition of plant
communities determines the spatial and temporal stability of soil microbial properties in a long-
term plant diversity. Oikos 125, 1743-1754
C

Strecker, T., Gonzalez, O., Scheu, S., Eisenhauer, N., 2015. Analysis of temporal microbial
properties from experimental plots of the Jena experiment (2003-2014). PANGAEA,
AC

https://doi.org/10.1594/PANGAEA.854693

Tejeda-Agredano, M.C., Gallego, S., Grifoll, M., Ortega-Calvo, J.J., Cantos, M., 2013. Influence
of sunflower rhizosphere on the biodegradation of PAHs in soil. Soil Biology and Biochemistry
57, 830-840.

Tilman, D., Wedin, D., Knops, J., 1996. Productivity and sustainability influenced by
biodiversity in grassland ecosystems. Nature 379, 718.

39
 ACCEPTED MANUSCRIPT

Wang, Y., Wang, S., Luo, C., Li, J., Ming, L., Zhang, G., Li, X., 2015. The effects of rice canopy
on the air–soil exchange of polycyclic aromatic hydrocarbons and organochlorine pesticides
using paired passive air samplers. Environmental Pollution 200, 35-41.

Wei, C., Bandowe, B.A.M., Han, Y., Cao, J., Zhan, C., Wilcke, W., 2015a. Polycyclic aromatic

PT
hydrocarbons (PAHs) and their derivatives (alkyl-PAHs, oxygenated-PAHs, nitrated-PAHs and
azaarenes) in urban road dusts from Xi’an, Central China. Chemosphere 134, 512-520.

RI
Wei, C., Han, Y., Bandowe, B.A.M., Cao, J., Huang, R.-J., Ni, H., Tian, J., Wilcke, W., 2015b.
Occurrence, gas/particle partitioning and carcinogenic risk of polycyclic aromatic hydrocarbons
and their oxygen and nitrogen containing derivatives in Xi'an, central China. Science of the Total

SC
Environment 505, 814-822.

Weigelt, A., Marquard, E., Temperton, V.M., Roscher, C., Scherber, C., Mwangi, P.N., von
Felten, S., Buchmann, N., Schmid, B., Schulze, E. –D., Weisser, W.W., 2010. The Jena

U
Experiment: six years of data from a grassland biodiversity experiment (Ecological Archives
E091-066). Ecology 91, 930. URL: http://esapubs.org/archive/ecol/E091/066/
AN
Weigelt, A, De Luca, E., Roscher, C., Temperton, V., Buchmann, N., Fischer, M., Scherer-
Lorenzen, M., Schmid, B., Schulze, E,-D., Weisser, W., Luo, G., Meyer, S.T., 2016. Collection
of above ground community and species specific plant biomass from the Jena Experiment (time
M

series since 2002). PANGAEA, https://doi.org/10.1594/PANGAEA.866358

D

Weisser, W.W., Roscher, C., Meyer, S.T., Ebeling, A., Luo, G., Allan, E., Beßler, H., Barnard,
R.L., Buchmann, N., Buscot, F., 2017. Biodiversity effects on ecosystem functioning in a 15-
TE

year grassland experiment: Patterns, mechanisms, and open questions. Basic and Applied
Ecology 23, 1-73.
EP

Wilcke, W., 2000. Polycyclic aromatic hydrocarbons (PAHs) in soil—a review. Journal of Plant
Nutrition and Soil Science 163, 229-248.

Wilcke, W., 2007. Global patterns of polycyclic aromatic hydrocarbons (PAHs) in soil.
C

Geoderma 141, 157-166.

AC

Wilcke, W., Amelung, W., 2000. Persistent organic pollutants in native grassland soils along a
climosequence in North America. Soil Science Society of America Journal 64, 2140-2148.

Wilcke, W., Bandowe, B.A.M., Lueso, M.G., Ruppenthal, M., del Valle, H., Oelmann, Y.,
2014a. Polycyclic aromatic hydrocarbons (PAHs) and their polar derivatives (oxygenated PAHs,
azaarenes) in soils along a climosequence in Argentina. Science of the Total Environment 473,
317-325.

40
 ACCEPTED MANUSCRIPT

Wilcke, W., Kiesewetter, M., Bandowe, B.A.M., 2014b. Microbial formation and degradation of
oxygen-containing polycyclic aromatic hydrocarbons (OPAHs) in soil during short-term
incubation. Environmental Pollution 184, 385-390.

Wild, E., Dent, J., Thomas, G.O., Jones, K.C., 2005a. Direct observation of organic contaminant
uptake, storage, and metabolism within plant roots. Environmental Science & Technology 39,
3695-3702.

PT
Wild, E., Dent, J., Thomas, G.O., Jones, K.C., 2005b. Real-time visualization and quantification
of PAH photodegradation on and within plant leaves. Environmental Science & Technology 39,

RI
268-273.

Wincent, E., Jönsson, M.E., Bottai, M., Lundstedt, S., Dreij, K., 2015. Aryl hydrocarbon receptor

SC
activation and developmental toxicity in zebrafish in response to soil extracts containing
unsubstituted and oxygenated PAHs. Environmental Science & Technology 49, 3869-3877.

U
AN
M
D
TE
C EP
AC

41
 ACCEPTED MANUSCRIPT
Plant diversity enhances the natural attenuation of polycyclic aromatic

compounds (PAHs and oxygenated PAHs) in grassland soils

Benjamin A. Musa Bandowea,b*, Sophia Leimera, Hannah Meuselb, Andre Velescua, Sigrid

Dassenc, Nico Eisenhauerd,e, Thorsten Hoffmannf, Yvonne Oelmanng, Wolfgang Wilckea

PT
a
Institute of Geography and Geoecology, Karlsruhe Institute of Technology (KIT), Reinhard-

RI
Baumeister-Platz 1, 76131 Karlsruhe, Germany
b
Multiphase Chemistry Department, Max Planck Institute for Chemistry, Hahn-Meitner-Weg 1,

SC
55128 Mainz, Germany
c
Department of Terrestrial Ecology, Netherlands Institute of Ecology, NIOO KNAW,

U
Droevendaalsesteeg 10, 6708 PB, Wageningen, The Netherlands
AN
d
German Centre for Integrative Biodiversity Research (iDiv), Halle-Jena-Leipzig, Deutscher Platz 5e,
M

04103 Leipzig, Germany

e
Institute of Biology, Leipzig University, Deutscher Platz 5e, 04103 Leipzig, Germany
D

f
Institute of Inorganic and Analytical Chemistry, University of Mainz, Duesbergweg 10-14, 55128
TE

Mainz, Germany
g
Geoecology, University of Tübingen, Rümelinstraβe 19-23, 72070 Tübingen, Germany
EP

Highlights
●Effects of increasing plant diversity on PACs concentrations in soil are investigated
C

●Plots with higher plant species richness (SR) had lower concentrations of several PACs
AC

●Higher SR, however, caused higher OPAHs concentrations and OPAH/parent-PAH ratios

●Higher SR favors OPAHs formation by microbial transformation of PAHs.

* Corresponding Author (B.A.M. Bandowe; Email: benjamin.bandowe@kit.edu;

benjamin.bandowe@mpic.de)