Evaluation of nanoremediation strategy in a Pb, Zn and Cd contaminated soil

Science of the Total Environment 706 (2020) 136041

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Evaluation of nanoremediation strategy in a Pb, Zn and Cd contaminated soil Carmen Fajardo a,⁎, Sebastián Sánchez-Fortún c, Gonzalo Costa c, Mar Nande c, Pedro Botías b, Jesús García-Cantalejo b, Gerardo Mengs c, Margarita Martín c a b c

Facultad de Farmacia, Universidad de Alcalá, 28805 Alcalá de Henares (Madrid), Spain Unidad de Genómica, Universidad Complutense de Madrid, 28040 Madrid, Spain Facultad de Veterinaria, Universidad Complutense, 28040 Madrid, Spain

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Multiple-heavy metal polluted soil was treated with nZVI for 4 months. • nZVI moderately reduced Pb and Zn mobility in the short-term. • HM immobilization decreased ecotoxicity in the short-term. • Decreased effectiveness of nanoremediation was found over the exposure time. • Nanoremediation induced structural and functional shifts in bacterial population.

a r t i c l e

i n f o

Article history: Received 26 September 2019 Received in revised form 6 December 2019 Accepted 8 December 2019 Available online 10 December 2019 Editor: Henner Hollert Keywords: Heavy metal Availability nZVI Ecotoxicogenomics Next Generation Sequencing Nanoremediation

a b s t r a c t We addressed the efficiency of a nanoremediation strategy using zero-valent iron nanoparticles (nZVI), in a case of co-mingled heavy metals (HM) pollution (Pb, Cd and Zn). We applied a combined set of physical-chemical, toxicological and molecular analyses to assess the effectiveness and ecosafety of nZVI (5% w/w) for environmental restoration. After 120 days, nZVI showed immobilization capacity for Pb (20%), it was scarcely effective for Zn (8%) and negligibly effective for Cd. The HMs immobilization in the nZVI treated soils (compared to control soil), reaches its maximum after 15 days (T3) as reflected in the decrease of HM toxicity towards V. fischeri. The overall abundance of the microbial community was similar in both sets of samples during all experiment, although an increase in the number of metabolically active bacteria was recorded 15 days post treatment. We studied the induced impact of nanoremediation on the soil microbial community structure by Next Generation Sequencing (NGS). Even when higher HM immobilization was recorded, no significant recovery of the microbial community structure was found in nZVI-treated soil. The most marked nZVI-induced structural shifts were observed at T3 (increase in the Firmicutes population with a decrease in Gram-negative bacteria). Predictive metagenomic analysis using PICRUSt showed differences among the predicted metagenomes of nZVI-treated and control soils. At T3 we found decrease in detoxification-related proteins or over-representation of germination-related proteins; after 120 days of nZVI exposure, higher abundance of proteins involved in regulation of cellular processes or sporulation-related proteins was detected.

⁎ Corresponding author at: Campus Universitario, Ctra. Madrid-Barcelona km. 33, 600, Edificio de Farmacia, 28805 Alcalá de Henares (Madrid), Spain.

https://doi.org/10.1016/j.scitotenv.2019.136041 0048-9697/© 2019 Elsevier B.V. All rights reserved.

2

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

This study highlights the partial effectiveness of nanoremediation in multiple-metal contaminated soil in the short term. The apparent lack of recovery of biodiversity after application of nZVI and the decreased effectiveness of nanoremediation over time must be carefully considered to validate this technology when assurance of medium- to long-term immobilization of HMs is required. © 2019 Elsevier B.V. All rights reserved.

1. Introduction In recent decades, there has been increasing demand for healthy environments that are free of contaminants, and particularly, the soil/water system is of maximum interest. Among the risks that threaten soil/ water system maintenance, heavy metal (HM) contamination is of great concern, particularly in soil, due to its numerous sources, toxicity, nonbiodegradability and accumulative behaviour. HMs interfere with many physiological processes, including alterations of enzyme specificity, disruption of cellular function, and damage to the cell membrane and structure of DNA (Bruins et al., 2000). Thus, widespread HM pollution in the soil/water system is currently one of the most serious environmental problems, and the scientific community has addressed the impact of HMs on microorganisms using different approaches (Chen et al., 2014; Xie et al., 2016; Xu et al., 2017). For instance, our group recently showed that highly polluted soil containing several HMs fully inhibited the survival and development of eukaryotic organisms, such as Caenorhabditis elegans, while prokaryotes responded to the effects of HMs by changing their structure and functionality (Fajardo et al., 2019a). Consequently, there is still a need for efficient in situ stabilization strategies to ensure a reduction of metal and metalloid bioaccessibility. Among the proposed soil remediation techniques, nanoremediation has been studied in recent years as a promising approach for the detoxification of HM contaminated soils (Xue et al., 2018). The most commonly used nanoparticles (NPs) in nanoremediation are those composed of Fe, in particular, nanoscale zero-valent iron (nZVI). The interest in the application of nZVI as a decontamination tool began after Wang and Zhang (1997), in which they described the ability of nZVI to reduce trichlorethylene and PCBs in groundwater. From that report, the effectiveness of nZVI against different recalcitrant harmful pollutants, such as chlorinated compounds, pesticides and HMs, has been documented in numerous publications (see Semerád and Cajthaml, 2016). However, most studies were performed in the context of aqueous systems and groundwater remediation (Barrera-Diaz et al., 2012; Němeček et al., 2014; Němeček et al., 2015; Yu et al., 2015). Despite the research effort carried out to assess the effectiveness of nZVI in decreasing mobility of metals in the soil environment (Shing et al., 2012; Fajardo et al., 2012; Gil-Diaz et al., 2014; Wang et al., 2014; Gil-Diaz et al., 2017; Vitkova et al., 2018; Anza et al., 2019), pollutant mixtures, which frequently occur in natural environments, have attracted less attention. The presence of different pollutants in a soil induces complicated chemical interactions that produce synergic effects that can lead to an increased toxicity. The co-occurring pollutants induce modifications in the humic content, adsorption behaviours, and biological processes. These interactions condition soil remediation strategies, since they make great influence on soil remediation efficiency. Therefore a thorough knowledge of this type of polluted scenarios is necessary, since it would allow us helpful to better understand their impacts on remediation efficiency and further study directions in this field (Ye et al., 2017, 2019). These issues are crucial to applying nanoremediation to achieve effective long-term immobilization in real polluted scenarios. In addition, the ecotoxicity of interactions between nZVI and the complex soil matrix environment is an issue of great concern and has been subjected to active investigation (see Semerád and Cajthaml, 2016; Xue et al., 2018). However, even today, there is still a lack of a suitable, comprehensive, and standardized set of tests for performing an ecotoxicological evaluation of nanomaterials, and further research in this direction is needed.

It is well known that NPs induce several negative effects on living organisms: DNA damage, disrupt membrane integrity, disrupt cellular uptake, among other effects. Oxidative stress from the formation of reactive oxygen species (ROS) is considered to be the primary toxic mechanism of NPs and impacts lipids, proteins and nucleic acids (Lefevre et al., 2016; Lei et al., 2018). To assess these impacts and the possible defence mechanisms presented by cells, the Organization to Economic Co-operation and Development (OECD) (Rasmussen et al., 2016) presented some recommendations for the use of NPs: i) obtaining results over long periods of exposure of soil to nanomaterials, ii) using new reference organisms, iii) optimizing methods and obtaining new reference parameters, and iv) including OMICs to define the effects of nanomaterials on living organisms. In response to these recommendations, our group has proposed a methodology based on molecular techniques, such as transcriptomics, Next Generation Sequencing (NGS), molecular markers, and ecotoxicity bioassays, to define biomarkers as new reference parameters to assess the impact of nZVI on living organisms (Fajardo et al., 2012; Saccà et al., 2014; Fajardo et al., 2018). In those previous studies, we reported that nZVI impacts soil organisms, to some extent, and causes changes of the structure and composition of microbial communities, although the overall functionality of the soil ecosystem was less affected. However, in spite of the uncertainty regarding the toxicity of nZVI, and even when the potential environmental hazards associated with the application of nZVI should not be underestimated, it is a promising option in highly damaged environments. The goal of this study was to address the efficiency of the nanoremediation strategy (nZVI-based), considered over long period of exposure (120 days), in a case of co-mingled HM pollution and a highly damaged soil environment. The objectives of this study included i) monitoring the capability and stability of nZVI for immobilization of Pb, Zn and Cd from a contaminated soil and ii) monitoring its impact on different soil organisms. In this context, we applied a combined set of physical-chemical, toxicological and molecular analyses to assess the effectiveness and ecosafety of nZVI as suitable tool for environmental restoration. 2. Materials and methods 2.1. Commercial nanoparticles NANOFER 25S iron nanoparticles (nZVI) were commercially synthesised and supplied by NANO IRON s.r.o. (Rajhrad, Czech Republic) as an aqueous dispersion of stabilized nZVI (coated with sodium polyacrylic acid, 3%). Additional details about the physical and chemical characteristics of nZVI are available at www.nanoiron.cz. The Zeta Potential (ZP) and size of nZVI were measured as previously described (Fajardo et al., 2019b). The zeta potential of the aqueous nZVI suspension was −44.7 mV. The size of the nZVI particles ranged from 61 to 71 nm in diameter, with an average particle size of 67 nm. 2.2. Experimental design Standard soil Lufa 2.3 supplied as field fresh with active microflora and provided by the Agricultural Investigation and Research Institute (LUFA Speyer, Germany) was used. The primary physicochemical characteristics of this sandy loam soil (USDA) were as follows: total organic carbon, 0.67%; nitrogen, 0.08%; pH 5.7; cation exchange capacity,

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

7.5 meq/100 g; and water holding capacity, 35.4 g/100 g (all values refer to dry matter). Four replicates consisting of 2 kg of soil each were artificially contaminated at 2700 mg kg−1, 4200 mg kg−1, and 310 mg kg−1 with commercial chloride salts of Pb, Zn, and Cd, respectively, purchased from Sigma-Aldrich (Sigma-Aldrich, Inc.). The HM concentrations used in the present study were equivalent to the reference levels for the protection of human health in an industrial soil according to the Spanish Regulation Orden 2770/2006, RD 9/2005 (BOCM, 2006). HM aqueous suspensions were added to the soil samples and manually mixed, adjusting the moisture content to 22% (w/w). Spiked soils were consolidated for 40 days under controlled conditions of humidity and Tª (22% w/w, 21 ± 0.5 °C). After this period, the commercial nZVI suspension was added (5% nZVI w/w) to two soil replicates (nZVI-treated samples) and carefully mixed. Additionally, two soil replicates that received only sterile water were used as controls. The nZVI concentration was selected according to the ranges determined in previous immobilization studies of heavily HM-polluted samples (Fajardo et al., 2019b). All of the samples were incubated for 120 days in a climatic chamber at 21 ± 0.5 °C. The soils were weighed periodically to maintain humidity (22% w/w). Samples were collected from the control and treated samples at 0 (T1), 7 (T2), 15 (T3), 30 (T4) and 120 (T5) days post incubation. These sampling times were chosen to obtain data after shortterm and long exposure period. At each sampling time pH and Eh were measured according to the official Spanish methodology for soil analysis (MAPA, 1994) using pH and ORP (Redox) electrodes (Metrohm Meter, Metrohm AG, Switzerland), respectively. 2.3. Determination of the Pb, Cd and Zn concentration: ICP-OES analyses Samples from the control and nZVI-treated soils were subjected to a sequential extraction procedure to determine the concentrations of Pb, Cd and Zn in the most available fractions (exchangeable and linked to carbonates) and in the sediment, according to published protocols (Gil-Diaz et al., 2014). In brief, 2.5 g of soil was mixed with 25 mL of MgCl2 (1 M, pH 7) for 1 h and then centrifuged (13,000 rpm for 30 min). The supernatant was removed and filtered (0.45 μm cellulose membrane filter) and named the exchangeable fraction (EX). The solid sediment residue left after removal of the supernatant was extracted with 25 mL of buffer CH3COONa/CH3COOH (1 M, pH 5) for 5 h and centrifuged; the supernatant was recovered, filtered and designated as carbonate fraction (CB). Finally, the resultant soil sediment residue (SED), was collected and air-dried. The total Cd, Pb and Zn levels in each fraction were determined by Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES, Perkin Elmer Optima 3300DV) following a microwave extraction according to EPA protocol 3051-A (CAI Técnicas Geológicas, Complutense University of Madrid). 2.4. Bacterial cell viability Samples from nZVI-treated and control soils (1 g) were mixed with 9 mL of PBS buffer and 45 μL of Tween 20 (0.5% v/v), vortexed for 15 min (400 rpm), and allowed to sediment for 5 h at 4 °C. Then, 1 mL of the supernatant was collected, and stained using a LIVE/DEAD BacLight™ Bacterial Viability Kit (Molecular Probes, Carlsbad, CA) according to the manufacturer's protocol. Fluorescence was measured in a Tecan Genios microplate reader (Tecan Group Ltd., Männedorf, Switzerland). A fresh culture of Pseudomonas sp. and PBS buffer were used as positive and negative controls, respectively. 2.5. Fluorescence in situ hybridization (FISH) assays The total cell abundance (cell nr g−1 soil) was determined in the nZVI-treated and control samples by direct counting using 4′, 6′-diamidino-2-phenylindole (DAPI) and published protocols

3

(Martin et al., 2008). The abundance of bacteria (bacterial nr g−1 soil) in the samples was analysed by FISH using the general Cy3labelled commercially synthesised oligonucleotide probe EUB338I-III (Bacteria domain) and published protocols (Barra-Caracciolo et al., 2005a, 2005b). The results are expressed as the percentage of DAPIpositive cells that hybridized with the fluorescent probe. 2.6. Toxicity assays Microtox® Test The Microtox® Basic Test was carried out using the manufacturer's protocol (Microbics Corporation, Carlsbad, USA). A Microtox® Model 500 Analyser (AZUR Environmental, Carlsbad, CA, USA) was used to measure the luminosity of the reconstituted bacteria after 5 and 15 min of exposure to the culture filtrate. Since there were no significant differences between the two measurements, the luminescence inhibition after a 15-min exposure was taken as the endpoint (Kaiser, 1998; Froehner et al., 2000). In the Microtox® test, inhibition of light emission was measured in relative units of luminescence. The data were used to calculate the volume required to induce full luminescence inhibition for each of the samples tested. The behaviour of the bacteria was tested with the reference toxins ZnSO4.7H2O and phenol according to normative AFNOR T90-320 (AFNOR, 1991). Toxicity is expressed as the concentration of HMs in the control and nZVI-treated samples (at T1, T2, T3, T4 and T5) that produced 50% bioluminescence inhibition (EC50). To better interpret the results, the toxicity impact index (TII50, also called Equitox/m3) was used as the measurement unit, where TII50 is directly proportional to the toxicity of the sample: TII50 = 100 / EC50. All analyses were carried out in quadruplicate (n = 4). 2.7. DNA extraction, metagenome library construction and sequencing Total bacterial community DNA was isolated from of each soil replicate (0.5 g) using the Power Soil DNA isolation kit (MoBio Laboratories, Inc., Carlsbad, CA, USA) in accordance with the manufacturer's protocols at T1, T3 and T5. DNA libraries from each sample were prepared at the Genomics Unit of the Complutense University of Madrid (Spain). In brief, the V3–V4 region of the prokaryotic 16S rRNA was amplified for each sample with primers containing the 341F and 805R sequences and Illumina-specific adapters. In a second PCR amplification, two specific 8-nucleotide index and i5/i7 Illumina adapters were added to the previous amplicons. A library pool was prepared for sequencing by mixing equal amounts of the individual sample libraries. The library pool was subjected to electrophoresis on an agarose gel, and bands containing the amplified fragments were excised from the agarose gel and purified. The library pool was sequenced (Illumina MiSeq) with 2 × 300 reads using the 600 cycle MiSeq Reagent Kit v3 in accordance with the manufacturer's recommended protocol. The FASTQ files containing the sequencing reads were analysed using the CLC Genomics Workbench version 11.0.1 (QIAGEN Aarhus A/S www.qiagenbioinformatics.com). Sequence data were trimmed using 0.05 as a limit for quality scores with 2 as the maximum number of ambiguities. The reads after trimming were analysed using the CLC Microbial Genomics Module version 4.0. The optional merge paired reads method was run with default settings (mismatch cost = 1; minimum score = 40; gap cost = 4 and maximum unaligned end mismatch = 5). Sequence reads were clustered and chimeric sequences detected using an identity of 97% as the Operational Taxonomic Unit (OTU) threshold. Reference OTU data used in the present study were downloaded from the Greengenes database v13_5 for 16S rRNA. The Shannon diversity index was calculated considering the assigned species. The raw sequencing data were deposited in the NCBI Sequence Read Archive database (BioProject ID: PRJNA593489) (http://www.ncbi.nlm. nih.gov/bioproject/593489).

4

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

2.8. Predictive metagenome analysis The PICRUSt tool, the online version available on the online Galaxy platform (https://huttenhower.sph.harvard.edu/galaxy), was used to predict the metagenome functional content based on the 16S rRNA amplicon data sets. The table of OTUs generated by the CLC Microbial Genomics Module (QIAGEN) from each microbiome was used as the input. The predicted metagenomes were functionally annotated with PICRUSt using the Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathways (hierarchical level 3). 2.9. Statistical analyses Statistical analyses of the physicochemical and microbiological parameters were performed using GraphPad Prism 5 software (San Diego, CA, USA). We used one-way or two-way analysis of variance, followed by Tukey's test or Bonferroni post-test, for multiple comparisons to determine the significance of the differences between the groups (p b 0.05). The version available on the online WebMeV (Multiple Experiment Viewer) platform (http://mev.tm4.org) was used to perform hierarchical clustering. The table of OTUs generated by the CLC Microbial Genomics Module from each microbiome classified at phylum or class levels was used as the input.

Fig. 1. Total cellular number (dashed lines) and bacterial viability (solid lines) measured in the control and nZVI-treated soil samples (mean values ± standard deviation).

3.2. Effectiveness of the nanoremediation strategy: heavy metal immobilization

3. Results 3.1. Soil chemical parameters, cellular abundance and FISH assays

To evaluate the effectiveness of nZVI on HM immobilization, the Pb, Cd and Zn concentrations were measured in the control and nZVItreated samples during the experimental period (Table 2). Application of nZVI to metal-polluted soil under the assessed experimental conditions decreased the Pb and Zn contents in the most available fraction (EX) compared to control soil, but not the Cd concentration. Fig. 2 shows the percentage of each HM concentration on the EX fraction with respect to the total concentration of each metal along the incubation time; differences between control and treated samples were maximum at T3 (13.5% and 18.1% for Pb and Zn, respectively). When compared the concentration of each HM between the nZVI-treated and control soils at each sampling time, the largest drop was also observed at T3, when the Pb and Zn concentrations in the EX fraction of nZVI-treated soil were 55.7% and 70.2%, respectively, compared to those found in the control sample. This reduction in HM concentrations was accompanied by an increase of the metal contents found in the sediment fraction (SED). At T5, only Pb immobilization remained relevant; a reduction of approximately 20% of the Pb content of the most available fraction was found (compared to control soil). The Zn concentration measured in the EX fraction was only 8% lower than that of the control soil at this sampling time. Thus, nZVI showed a moderate immobilization capacity for Pb in polluted soil, whereas it was poorly efficient for Zn removal and negligibly efficient for Cd removal.

The control and nZVI-treated soil samples were characterized considering both, physical-chemical and microbiological parameters during the experimental period. The obtained results are reported in Table 1. Previous studies reported that nZVI addition causes an increase in the pH values, due to oxidation of bare nZVI under aerobic conditions into iron oxide/hydroxides (Lefevre et al., 2016). However, in this study nZVI addition caused a slight decrease in pH values. This result can be referred to the specific physic-chemical characteristics of the soil matrix used (pH, buffer capacity, CEC, etc.), and/or to the specific coating of nZVI particles that could modify the above-mentioned behaviour. The most marked differences between the control and treated samples were observed under redox conditions; Eh (mV) was significantly lower in nZVI-added samples, particularly until 15 days post incubation (T3). nZVI application decreased E h to −90 mV at T1, and after 1–4 months (T4 and T5, respectively), Eh stabilized at approximately 130 mV. The overall abundance of the microbial community, as determined by DAPI counts, was comparable between the control and nZVItreated samples during all the experiments. However, in some cases, the bacterial cell number was higher in the nZVI-added samples than in the control soil samples (T3). Moreover, no significant differences were recorded between both groups of samples in terms of bacterial cell viability, which increased over the period between T2 and T4 for both sample groups (Fig. 1).

Table 1 pH, redox potential, total number of cells and Bacteria counted by FISH in the soil samples at each sampling time (mean values ± standard deviation). Means within a row followed by the same lowercase letter are not significantly different (p b 0.05); means within a column followed by the same capital letter are not significantly different (p b 0.05). pH Control T1 T2 T3 T4 T5

5.71 5.84 5.79 5.82 5.85

Cell nr. g−1 soil

Eh (mV)

± ± ± ± ±

nZVI 0.00 (Aa) 0.00 (Ba) 0.01 (Ca) 0.01 (BCa) 0.01 (Ba)

5.61 5.64 5.76 5.77 5.72

Control ± ± ± ± ±

0.01 (Ab) 0.01 (Ab) 0.00 (Ba) 0.01 (Bb) 0.01 (Cb)

263 342 313 284 280

± ± ± ± ±

0.0 (Aa) 0.0 (Ba) 0.0 (Ca) 0.6 (Da) 0.0 (Ea)

nZVI −90 ± 0.0 (Ab) −17 ± 0.0 (Bb) 83 ± 0.0 (Cb) 129 ± 0.6 (Db) 137 ± 0.0 (Eb)

% bacteria

Control 4.5 5.8 6.0 3.9 2.6

× × × × ×

nZVI 5

10 105 105 105 105

± ± ± ± ±

9.4 8.3 2.4 7.9 6.2

× × × × ×

4

10 (ABa) 104 (ABa) 105 (Aa) 104 (ABa) 104 (ABa)

5.3 6.0 6.2 6.7 2.5

× × × × ×

5

10 105 105 105 105

± ± ± ± ±

1.2 1.3 8.3 3.9 7.6

× × × × ×

5

10 (ABa) 105 (ABa) 104 (Aa) 105 (ACa) 104 (Ba)

Control

nZVI

81 79 72 67 64

85 83 81 71 65

± ± ± ± ±

1.1 (Aa) 5.3 (Aa) 4.2 (Ba) 0.0 (Ba) 3.3 (Ba)

± ± ± ± ±

2.8 (Aa) 1.6 (Aa) 0.1 (Ab) 0.8 (Ba) 1.3 (Ba)

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

5

Table 2 Concentration of HM (μg g−1 soil) in control and nZVI-treated samples in the different soil fractions (EX, exchangeable, CB, carbonate-bound and SED, sediment fractions). The results are expressed as the mean values ± standard deviation. μg g−1 soil

T1 Control

Cd

Pb

Zn

EX CB SED TOTAL EX CB SED TOTAL EX CB SED TOTAL

T2 nZVI

257 ± 26

2605 ± 437

4005 ± 601

T3

T4

T5

Control

nZVI

Control

nZVI

Control

nZVI

Control

nZVI

230 ± 10 28 ± 3 9.2 ± 0.9 267 640 ± 20 1020 ± 50 1046 ± 50 2706 2520 ± 250 880 ± 30 466 ± 130 3866

240 ± 10 34 ± 3 6.7 ± 0.7⁎ 281 390 ± 10⁎⁎ 940 ± 20 1390 ± 40 2720 1920 ± 130⁎ 820 ± 20 858 ± 200⁎

200 ± 10 13.8 ± 0.7 3.4 ± 0.3 217 700 ± 40 810 ± 70 810 ± 81 2320 2850 ± 140 740 ± 40 213 ± 21 3803

200 ± 10 16 ± 0.8 5.9 ± 0.6⁎ 222 390 ± 30⁎⁎⁎

227 ± 20 24 ± 2 5.2 ± 0.3 256 493 ± 30 787 ± 50 1327 ± 50 2607 2220 ± 60 580 ± 20 630 ± 50 3430

233 ± 20 21 ± 2 3.1 ± 0.3 257 362 ± 30⁎⁎ 783 ± 60 1460 ± 130 2605 1790 ± 80 610 ± 20 451 ± 20 2851

235 ± 20 17 ± 0.8 7,2 ± 0.6 259 498 ± 30 755 ± 50 1452 ± 60 2705 2159 ± 130 521 ± 20 706 ± 50 3386

227 ± 20 22 ± 1.0 6,1 ± 0.5 255 397 ± 20 732 ± 50 1589 ± 70 2718 1986 ± 100 615 ± 20 789 ± 50 3390

3598

1030 ± 80 913 ± 91 2333 2000 ± 100⁎⁎ 920 ± 50⁎⁎ 600 ± 60⁎ 3520

⁎ Significant differences (p b 0.05) between control and treatment. ⁎⁎ Significant differences (p b 0.01) between control and treatment. ⁎⁎⁎ Significant differences (p b 0.001) between control and treatment.

3.3. Ecotoxicological assays. Microtox® Test We performed ecotoxicological analyses using V. fischeri as test organism to address the HM toxicity in the treated and untreated soil over the exposure period with nZVI. The TII50 values of the control and nZVI-treated soil samples were obtained by the Microtox® Test. Toxicity indices of 100, 95.50 ± 6.66, 77.46 ± 4.20, 62.81 ± 7.46 and 55.39 ± 8.11 for the control samples and of 100, 95.04 ± 5.55, 58.67 ± 4.83, 53.18 ± 9.47 and 51.41 ± 9.42 for the nZVI treated samples (at 0, 7, 15, 30 and 120 days, respectively) were found. Comparative analysis between the relative toxicities corresponding to the control and nZVI-treated samples showed statistically significant differences (p b 0.001) between them only 15 days post-addition (T3) (Fig. 3). At T3, the presence of nZVI in treated soil significantly reduced the relative toxicity of HMs by approximately 25% compared to the control values, while the reduction in the relative toxicity values obtained at T4 and T5 days post-treatment was much lower. 3.4. Soil microbiome analyses: taxonomical and functional characterization We applied 16S rRNA gene high-throughput sequencing to unveil the impact of the nanoremediation strategy on the soil bacterial structure and biodiversity. Thus, the taxonomic analysis of the soil microbial community structure in nZVI-treated and non-treated samples was

Fig. 2. Percentage of each HMs concentration in the exchangeable fraction (EX) with respect to the total concentration of each metal during the incubation time. *p b 0.05.

performed at 0 (T1), 15 (T3) and 120 (T5) days of treatment. Table 3 shows the number of OTUs assigned to each analysed microbiome and the percentage of OTUs classified at the phylum, class, genus, and species levels (according to the CLC Microbial Genomics Module and the Greengenes database). The Shannon index suggested a similar diversity in the assessed soil samples. The total number of phyla found in the control samples ranged from 20 to 25, with Firmicutes, Proteobacteria, Actinobacteria, Bacteroidetes and Chloroflexi, the most abundant phyla during the incubation period, accounting for N94% of all the identified OTUs. Bacterial community analyses of nZVI-treated soil (at T1, T3, and T5) indicated that a range of 10–22 bacterial phyla were present, and that sample T5 contained the lowest number of identified phyla. Only seven phyla displayed a relative abundance N 0.5% in the nZVI-treated soil samples; Proteobacteria, Firmicutes, and Actinobacteria composed N90% of all the identified sequences (Fig. 4). Previously reported results highlighted that a mixture of HMs at high concentrations induced a selective pressure on the soil microbiome during the same exposure period (Fajardo et al., 2019a): the relative abundance of Proteobacteria clearly decreased, whereas the relative abundance of Firmicutes increased during the incubation period. Similarly, the results obtained in this study show that the development of the soil microbial community after nZVI treatment followed a similar

Fig. 3. Microtox® Test values for control and nZVI-treated soils at each sampling time. The points represent the mean ± standard deviation values and are expressed as the Toxicity Impact Index (TII50). ***p b 0.001.

6

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

Table 3 Data obtained after bacterial community DNA sequencing and classification of OTUs according to the CLC Microbial Genomics Module and the Greengenes database (mean data ± standard deviation). Sample

No. OTUs

%Classified to phylum

%Classified to class

%Classified to genus

%Classified to species

Shannon index

Control T1 Control T3 Control T5 nZVI T1 nZVI T3 nZVI T5

3515 2044 2495 2939 1539 1166

99.98 ± 0.03 100 ± 0.00 100 ± 0.00 100 ± 0.00 100 ± 0.00 100 ± 0.00

99.94 ± 0.01 99.99 ± 0.01 99.50 ± 0.00 99.97 ± 0.01 100 ± 0.00 99.99 ± 0.01

57.90 68.67 65.28 66.76 76.36 70.98

13.48 17.97 17.52 17.13 22.31 22.15

1.7947 1.6837 1.6623 1.7002 1.4908 1.5435

± ± ± ± ± ±

276 206 96 120 188 4

pattern (Fig. 4); the differences described below were those found as statistically significant (p b 0.05). The percentage of cells belonging to the Firmicutes group significantly increased during the exposure time and reached a maximum at T3, accounting for 96% of all the identified OTUs; afterwards, Firmicutes significantly decreased in abundance (88% at T5). By contrast, the abundance of Proteobacteria dramatically decreased during the exposure period, particularly at T3 (0.7%). However, despite both sets of soil samples evolving similarly, the taxonomical profiles of the control and nZVI-treated samples clearly differed at all sampling times (Fig. 4). At T1, T3 and T5, the Firmicutes population was increased in the nZVI-treated samples compared to the control, whereas Proteobacteria almost disappeared. At T5, a significant decrease in the abundance of OTUs affiliated with the Actinobacteria phylum was found in nZVI-added soil compared to the untreated sample. The observed differences among the phylogenetic profiles of the control and treated samples were reflected by hierarchical clustering; soil microbiomes at the phylum level grouped the samples into two distinct clusters, where the microbiomes of nZVI-added soil at T1 clustered closer to control samples, and nZVI-treated soil at T3 and T5 appeared separated from control soil samples (Fig. 4). At the class level (Fig. 5), Bacilli was remarkably overrepresented during the incubation period. This enrichment in Bacilli cells was observed for both sets of soil samples, although, once again, it was particularly evident in nZVI-treated soil. In fact, OTUs identified as Bacilli in nZVI-treated samples represented 83.9% of all the identified sequences at T3. In parallel, the α-Proteobacteria class slightly decreased during the exposure period, becoming negligible at this sampling time (0.4%). At T5, a higher abundance of Bacilli and Clostridia classes and a decrease of α-Proteobacteria were found in the nZVI-treated samples compared to the control samples. As expected, the detected increase in the number of OTUs affiliated with the Bacilli class was related to the increase of the Bacillus genus, which was the most abundant microbial group in the nZVI-treated samples at T3 and T5 (46.3% and 46.7%, respectively). Among the identified bacterial species, B. flexus predominated in all soil microbiomes (the control and nZVI-spiked samples); nevertheless, its abundance was higher in nZVI-spiked soil, and for longer exposure times, a higher percentage was found. To establish a correlation between the phylogenetic structure and the metabolic profiles of the analysed microbiomes, we used PICRUSt (Langille et al., 2013) and the KEGG pathway to identify predicted functions in the metagenomes. The data obtained showed 261 different predicted functions in the soil microbiomes. We defined dominant pathways as those with a relative abundance N 0.5% of the total observed analysis. Table S1 (Supplementary information) summarizes the most significant predicted functions (KEGG pathways) in the nZVI-treated soil samples. Environmental information processing was the dominant predicted function in all the nZVI-treated samples. The subcategories ABC transporters and transporters were highly represented, particularly at T3. Moreover, at T3-T5, increases in sporulationrelated pathways were evident. A more precise observation of the overrepresented metabolic pathways revealed an increase of key proteins in the nZVI-treated soil samples at T3 and T5 compared to those at T1 (Fig. 6). We

± ± ± ± ± ±

1.02 0.22 2.52 2.66 0.30 9.79

± ± ± ± ± ±

0.77 0.61 1.20 1.56 0.45 4.04

± ± ± ± ± ±

0.029 0.063 0.002 0.023 0.017 0.340

considered the differences to be relevant when the percentage difference of each function between two samples was higher than 0.05%; a ratio higher than 1.75-fold indicated over-represented functions, and a ratio lower than 0.5-fold indicated down-represented functions. Thus, for genetic information, there was a predominance of sequences annotated to the ribosomal-protein-alanine Nacetyltransferase [EC: 2.3.1.128], N-acetylmuramoyl-L-alanine amidase (KO0676, KO1449), and several proteins with unclassified functions in the subcategory sporulation (KO6295, KO3091). By contrast, some predicted proteins were under-represented at T3 and T5 compared to those at T1 (KO2014, KO0799). Compared to the control samples, addition of nZVI did not produce significant changes in the predicted KEGG pathways; only increases in two functional traits, environmental information processing (membrane transport, transporters) and unclassified (cellular processes and signalling, sporulation), were detected at T3 and T5. In accordance with the above-described pattern, the differences between untreated and treated samples were particularly marked at T3. Moreover, the results showed that there were no differences among the predicted metagenomes of the nZVI-treated and untreated soil samples at T1 (Table 4); at T3, a decrease in abundance of the KO2014 and KO0799 proteins was recorded, and at T5, KO1768 was also decreased. By contrast, the metagenomes of the nZVI-added samples at T3 and T5 revealed an increased relative abundance of sequences annotated to metabolism-enzymes (KO0676 and KO1449); KO6295, a germination protein, also exhibited a significant relative abundance at T3. At T5, KO 3091 and KO6407 sporulation-related proteins, as well as KO3310 and KO1534 (membrane transport) were increased in the predicted metagenome of the nZVI samples compared to control. 4. Discussion Anthropogenic contamination of soils by HMs (e.g., Cd, Pb and Zn) occurs from many sources, and HM emissions from industrial or mining activities can cause severe pollution. Inappropriate treatments of these wastes often result in extremely high concentrations of HMs, not only in mine tailings and industrial areas but also in the surrounding soils and cropping and farming lands (Rodríguez et al., 2009; Liu et al., 2013). Thus, research is needed to develop appropriate methods of managing such sites to neutralize the potentially adverse effects of HMs on these environments. In this study, we evaluated the effectiveness of a nanoremediation strategy in highly damaged environments, such as heavily HMcontaminated soil. The chemical analyses indicated that the addition of nZVI stabilized Zn and Pb, even at the high concentrations used in this study. However, nZVI treatment was unable to effectively immobilize Cd from polluted soil. After 15 days of treatment (T3), nZVI reduced the Pb and Zn contents (44.3% and 29.8%, respectively) of the most bioavailable soil fraction compared to the same fraction of the control soil. Thus, Pb stabilization was more efficient than Zn stabilization, in accordance with previous works (Gil-Diaz et al., 2014). These different effects can be related to the differences in the standard redox potential of the metals and the associated main removal mechanism, i.e., precipitation for Pb and

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

7

Fig. 4. Relative abundances of phyla in the control and nZVI-treated soil microbiomes at T1, T3 and T5 (A). Heat map and dendrogram (B) of the most abundant bacterial phyla in the microbial communities of soil samples.

adsorption for Zn (Jiang et al., 2018). However, after 120 days of exposure, the Pb concentration in nZVI-treated soil was 20% lower than that in control soil, whereas the decrease of the Zn content was only 8%. Based on these findings, the use of nZVI could moderately reduce mobility of multiple heavy metals but differences not only in the effectiveness but also in the stability of the nanoremediation process were found depending on the pollutant considered. A short reactive lifespan of nZVI has been reported since the reactivity of nZVI are significantly reduced after aging due to the oxidation that takes place upon contact

with soil (Lei et al., 2018). The results obtained in this study revealed that despite the relative effectiveness of nZVI, the extent and duration of the potential beneficial effects of the treatment seem to be lower than 4 months, particularly for Zn. Further research is needed to confirm these findings under fieldscale operation in natural scenarios, where the redox and pH conditions can be situated far from ‘natural’ soil chemistry, or where the cooccurrence of organic and inorganic aged pollutants could affect the nanoremediation approach.

8

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

Fig. 5. Most abundant identified classes in the analysed soil microbiomes of the control and nZVI-treated samples at T1, T3 and T5.

To assess nZVI as an efficient remediation strategy, we evaluated the reduction in HM toxicity associated with decreased bioavailability after nZVI treatment. Additionally, the impacts of nanoremediation on soil organisms and functionality were considered to avoid the unintended consequences and potential environmental hazards associated with this treatment. The results obtained confirm that the immobilization of Pb and Zn at T3 was reflected in the decrease of HM toxicity (compared to control sample) towards V. fischeri, although thereafter, the toxicity values found for the treated and untreated samples did not differ. Moreover, although the total number of cells was similar in both sets of samples, we detected an increase in the bacterial cell number in the nZVI-added samples compared to control soil at T3. Considering that the FISH protocol targets specific regions of the 16S rRNA gene and allows the identification of metabolically active bacteria in complex systems, the observed increase in the abundance of bacteria at this sampling time indicated increased metabolic activity of this crucial microbial group in the soil community of nZVI-treated samples. Therefore, these results suggest that a less harmful scenario was emerging in the soil microbial community at this point. However, similar to the above-described pattern, the percentage of cells in the Bacteria domain was not significantly impacted after T3. Therefore, these findings

support the suggested loss in efficacy of the nanoremediation process 15 days post-addition. Regarding the impact of nanoremediation on the structure of the soil microbial community, the NGS results revealed that Firmicutes (Bacillus) selectively displaced almost all other bacterial groups and dominated the nZVI-treated soil microbiomes during the entire experimental period. Recently, we reported that HMs caused severe disturbances in exposed bacteria and that cells belonging to the Firmicutes phylum were the most resistant and selectively displaced other bacterial phyla (Fajardo et al., 2019a). This result was likely related to their capability to form endospores under a stressed environment, which provides them an advantage in adapting to and/or resisting unfavourable environmental stimuli. After the addition of nZVI, the obtained results indicated an increase of the Firmicutes population and a decrease of Gram-negative bacteria, mainly Proteobacteria. The most marked nZVI-induced structural shifts were observed at T3, when Firmicutes accounted for 96% of all the identified OTUs. The increase of the Firmicutes (Bacillus) population was in agreement with previous results found by Němeček et al. (2014), who reported stimulation of Gram-positive bacteria in soil after nZVI

Fig. 6. Evolution over time of the major over- and down-represented predicted functions in the nZVI-treated soil samples at T1, T3 and T5.

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

injection. Therefore, at T3, the decrease in the bioavailable HM content, in addition to the potential negative impact of nZVI on Gram-negative bacteria (Xue et al., 2018), could explain the predominance of Firmicutes group. The presence of Firmicutes could represent a defence strategy against the initial environmental insult caused by HMs (through the formation of endospores), which maintains a residual population that can develop when a less harmful scenario emerges in the soil microbial community (i.e., a decrease of HM toxicity due to their lower bioavailability). By contrast, at T5, even when the Pb and Zn concentrations in the most available fractions of control and nZVI-treated soil were quite similar, the nZVI oxidation would enable Gram-negative bacteria to develop, and subsequently, the communities were slightly reestablished. The success of any nanoremediation approach should be supported by the complete recovery of the damaged environment, not only from a physical-chemical perspective but also from a microbiological perspective. However, our results indicated that even when HM immobilization was recorded (particularly at T3), no significant recovery of the microbial community structure or diversity was found in nZVI-treated soil compared to control soil. The results from the predicted metagenomes revealed that iron complex outer membrane receptor protein (KO2014) was decreased in the nZVI-treated samples at T3 and T5 compared to control (Table 4). This protein, also known as the Ton B-dependent receptor, belongs to a family of beta barrel proteins located in the outer membrane of Gram-negative bacteria, and it is associated with the uptake and transport of large substrates. Therefore, its decrease was in agreement with the decrease of Gram-negative bacteria during soil exposure to nZVI. Another decreased protein was glutathione-S-transferase (KO0799), a member of a family of eukaryotic and prokaryotic enzymes best known for their ability to catalyse the conjugation of the reduced form of glutathione to xenobiotic substrates for the purpose of detoxification. This activity detoxifies endogenous compounds, such as H2O2 and other ROS compounds produced by the presence of toxic ions. Likewise, adenylate cyclase (KO1768) was decreased in the nZVI-treated samples at T5. This enzyme catalyses the conversion of adenosine triphosphate to 3′,5′-cyclic AMP (cAMP) and pyrophosphate. cAMP is an important signalling molecule in many bacterial species that is involved in the regulation of gene expression in response to a variety of environmental stimuli (Lory et al., 2004). Regarding the over-represented functions in the predicted metagenomes of nZVI-treated samples, the dominant proteins are shown in Table 4. N-acetylmuramoyl-L-alanine amidase (KO1449) and ribosomal-protein-alanine N-acetyl-transferase (KO0676), over represented at T3 and T5, are enzymes involved in important biological processes (genetic information, formation of flagella, sporulation and regulation of pap pilin transcription) as a response to a variety of environmental stimuli (Hernday et al., 2004). Stage V sporulation protein AE (KO6407) and spore germination protein KA (KO6295) are proteins

9

involved in different stages of sporulation/germination processes of the Firmicutes phylum (Fimlaid and Shen, 2015). KO6407 was overrepresented in the T5 sample, whereas KO6295 was over-represented in the T3 sample, when Firmicutes abundance reached the maximum value (96%). Additionally, RNA polymerase sporulation-specific sigma factor (KO3091), responsible for the expression of sporulation-specific genes, was over-represented at T5. Other over-represented proteins (KO1534 and KO3310) appeared at T5. The first, a Cd2+/Zn2+exporting ATPase, is a membrane protein that couples the hydrolysis of ATP to the efflux of cytoplasmic metals, and it is involved maintaining the homeostasis of different metals, which is important for bacterial survival in oxidant environments (Argüello et al., 2011). The alanine or glycine cation symporter (AGCS) family plays important roles in the regulation of cellular processes, considering that amino acids are an important nitrogen source to support life under stressed conditions (Bualuang et al., 2015). Thus, an overall view of the predicted metagenomes support the hypothesis of a decrease in the harmful conditions in the nZVI-treated soil in the short-term (T3), in accordance to a decreased abundance in cellular detoxification proteins, blockage of the iron cellular uptake or overrepresentation of germination-related proteins. However, after longer exposure (T5) we speculate that the microbial community evolves cellular mechanisms to adapt to unfavourable conditions (i.e., higher abundance of proteins involved in regulation of cellular processes, as the maintenance of metals homeostasis, or sporulation-related proteins). According to Wuana and Okieimen (2011), any successful remediation strategies must ensure the reduction of contaminant bioavailability only if reduced bioavailability leads to risk reduction and if the bioavailability reductions are demonstrated to be long term. Previous studies have reported that the use of nZVI treatment as a remediation strategy had a good immobilization effect on several HM, including Cd, Pb and Zn, in other soil types (Jiang et al., 2018; Xue et al., 2018). However, the results obtained in our study highlighted the moderate effectiveness of nanoremediation in multiple-metal contaminated soil in the short term. Moreover, the apparent lack of recovery of biodiversity after application and the decreased effectiveness of nanoremediation over time must be carefully considered to validate this technology when assurance of medium- to long-term immobilization of HMs is required. Thus, although nanoremediation is considered a promising strategy applicable to polluted soils with high concentrations of multiple HMs, it would be interesting to consider the use of other strategies in combination with nZVI treatment. Several recent studies have reported the feasibility and efficiency of the nano-biotechnological approach to remove environmental pollutants (Němeček et al., 2016; Galdames et al., 2017), which leads us to the challenging question of whether bioremediation can be used after the first few weeks of nZVI application, when the bioavailability and toxicity of HMs are decreased, to enhance detoxification and improve the efficiency of the nanoremediation strategy for restoring contaminated sites.

Table 4 Ratio of the major predicted traits in the metagenomes of the nZVI-treated and control samples at T1, T3 and T5. The values are the nZVI-fold compared to the control samples. KEGG description

KO 2014 KO 0799 KO 1768 KO 3091 KO 0676 KO 6295 KO 1449 KO 6407 KO 1534 KO 3310

Iron complex outer membrane receptor protein Glutathione S-transferase [EC:2.5.1.18] Adenylate cyclase [EC:4.6.1.1] RNA polymerase sporulation-specific sigma factor Ribosomal-protein-alanine N-acetyltransferase [EC:2.3.1.128] Spore germination protein KA N-acetylmuramoyl-L-alanine amidase [EC:3.5.1.28] Stage V sporulation protein AE Cd2+/Z2+-exporting ATPase [EC:3.6.3.3 3.6.3.5] Alanine or glycine: cation symporter, AGCS family

“-” indicates no relevant differences between samples.

nZVI/control T1

T3

T5

-

0.12 0.12 1.76 1.77 1.79 -

0.22 0.20 0.16 2.16 1.96 2.06 2.30 1.75 2.38

10

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041

5. Conclusions Nanoremediation strategy applied to multiple HM-contaminated soil decreased Zn and Pb mobility, particularly in the short-term. However, over longer time scales of treatment (120 days), the efficiency of this strategy was low, and toxicity values were comparable between treated and untreated soil. Therefore, rationale doubts emerge about the feasibility of the large-scale application of nZVI, especially from the point of view of its efficacy and cost under field conditions, particularly when applied to an aged soil where pollutants can show reduced availability due to interaction with the soil matrix components. Moreover, the balance between its benefits and risks must be carefully weighed. Attention must be focused on the adverse impact of nanoremediation on living organisms, considering that synergistic interactions between nZVI and target contaminants can take place in the field, and that the stability of these interactions can vary over a long period. The combined application of nZVI and bio-remediation should be taken into consideration, although further research on this issue is still required. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2019.136041. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors thank the Spanish Ministry of Economy and Competitiveness for supporting the Project CTM2013-46870-C2-1-P. This work was supported by grant S2017/BMD-3691 InGEMICS-CM, funded by Comunidad de Madrid (Spain) and European Structural and Investment Funds (Genomics Unit). References AFNOR, 1991. Détermination de l’inhibition de la luminescence de V. fischeri. NF T90-320. Paris. p. 331. Anza, M., Salazar, O., Epelde, L., Alkorta, I., Garbisu, C., 2019. The application of nanoscale zero-valent iron promotes soil remediation while negatively affecting soil microbial biomass and activity. Front. Environ. Sci. 7, 19. https://doi.org/10.3389/ fenvs.2019.00019. Argüello, J.M., González-Guerrero, M., Raimunda, D., 2011. Bacterial transition metal P1BATPases: transport mechanism and roles in virulence. Biochemistry 50, 9940–9949. https://doi.org/10.1021/bi201418k. Barra-Caracciolo, A., Grenni, P., Ciccoli, R., Di Landa, G., Cremisini, C., 2005a. Simazine biodegradation in soil: analysis of bacterial community structure by in situ hybridization. Pest Manag. Sci. 61, 863–869. https://doi.org/10.1002/ps.1096. Barra-Caracciolo, A., Grenni, P., Cupo, C., Rossetti, S., 2005b. In situ analysis of native microbial communities in complex samples with high particulate loads. FEMS Microbiol. Lett. 253, 55–58. https://doi.org/10.1016/j.femsle.2005.09.018. Barrera-Diaz, C.E., Lugo-Lugo, V., Bilyeu, B., 2012. A review of chemical, electrochemical and biological methods for aqueous Cr(VI) reduction. J. Hazard. Mater. 223-224, 1–12. https://doi.org/10.1016/j.jhazmat.2012.04.054. BOCM, 2006. Orden 2770/2006 (España), de 11 de agosto, de la Consejería de Medio Ambiente y Ordenación del Territorio, por la que se procede al establecimiento de niveles genéricos de referencia de metales pesados y otros elementos traza en suelos contaminados de la Comunidad de Madrid (BOCM 28 de agosto de 2006). Bruins, M.R., Kapil, S., Oehme, F.W., 2000. Microbial resistance to metals in the environment. Ecotoxicol. Environ. Saf. 45, 198–207. https://doi.org/10.1006/eesa.1999.1860. Bualuang, A., Kageyama, H., Tanaka, Y., Incharoensakdi, A., Takabe, T., 2015. Functional characterization of a member of alanine or glycine: cation symporter family in halotolerant cyanobacterium Aphanothece halophytica. Biosci. Biotechnol. Biochem. 79, 230–235. https://doi.org/10.1080/09168451.2014.968091. Chen, J., He, F., Zhang, X., Sun, X., Zheng, J., Zheng, J., 2014. Heavy metal pollution decreases microbial abundance, diversity and activity within particle-size fractions of a paddy soil. FEMS Microbiol. Ecol. 87, 164–181. https://doi.org/10.1111/15746941.12212. Fajardo, C., Ortiz, L.T., Rodriguez-Membibre, M.L., Nande, M., Lobo, M.C., Martin, M., 2012. Assessing the impact of zero-valent iron (ZVI) nanotechnology on soil microbial structure and functionality: a molecular approach. Chemosphere 86, 802–808. https://doi.org/10.1016/j.chemosphere.2011.11.041.

Fajardo, C., García-Cantalejo, J., Botías, P., Costa, G., Nande, M., Martin, M., 2018. New insights into the impact of nZVI on soil microbial biodiversity and functionality. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 27, 1–11. https://doi.org/ 10.1080/10934529.2018.1535159. Fajardo, C., Costa, G., Nande, M., García-Cantalejo, J., Botías, P., Martin, M., 2019a. Pb, Cd, and Zn soil contamination: monitoring functional and structural impact on microbiome. Appl. Soil Ecol. 135, 56–64. https://doi.org/10.1016/j.apsoil.2018.10.022. Fajardo, C., Costa, G., Nande, M., Martín, C., Martin, M., Sanchez-Fortun, S., 2019b. Heavy metals immobilization capability of two iron-based nanoparticles (nZVI and Fe3O4): soil and freshwater bioassays to assess ecotoxicological impact. Sci. Total Environ. 656, 421–432. https://doi.org/10.1016/j.scitotenv.2018.11.323. Fimlaid, K.A., Shen, A., 2015. Diverse mechanisms regulate sporulation sigma factor activity in the Firmicutes. Curr. Opin. Microbiol. 24, 88–95. https://doi.org/10.1016/j. mib.2015.01.006. Froehner, K., Backhaus, T., Grimme, L.H., 2000. Bioassays with Vibrio fischeri for the assessment of delayed toxicity. Chemosphere 40, 821–828. https://doi.org/10.1016/ S0045-6535(99)00274-X. Galdames, A., Mendoza, A., Orueta, M., de Soto García, I.S., Sánchez, M., Virto, I., Vilas, J.L., 2017. Development of new remediation technologies for contaminated soils based on the application of zero-valent iron nanoparticles and bioremediation with compost. Resour. Effic. Technol. 3, 166–176. https://doi.org/10.1016/j.reffit.2017.03.008. Gil-Diaz, M., Alonso, J., Rodriguez-Valdes, E., Pinilla, P., Lobo, M.C., 2014. Reducing hte mobility of arsenic in brownfield soil using stabilised zero-valent iron nanoparticles. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 49, 1361–1369. https:// doi.org/10.1080/10934529.2014.928248. Gil-Diaz, M., Alonso, J., Rodriguez-Valdes, E., Gallego, J.R., Lobo, M.C., 2017. Comparing different comercial zero valent iron nanoparticles to immobilize as and Hg in browwnfield soil. Sci. Total Environ. 584-585, 1324–1332. https://doi.org/10.1016/j. scitotenv.2017.02.011. Hernday, A.D., Braaten, B.A., Broitman-Maduro, G., Engelberts, P., Low, D.A., 2004. Regulation of the pap epigenetic switch by CpxAR: phosphorylated cpxr inhibits transition to the phase ON state by competition with Lrp. Mol. Cell 16, 537–547. https://doi. org/10.1016/j.molcel.2004.10.020. Jiang, D., Zeng, G., Huang, D., Chen, M., Zhang, C., Huang, C., Wan, J., 2018. Remediation of contaminated soils by enhanced nanoscale zero valent iron. Environ. Res. 2018 (163), 217–227. https://doi.org/10.1016/j.envres.2018.01.030. Kaiser, K.L.E., 1998. Correlations of Vibrio fischeri bacteria test data with bioassay data for other organisms. Environ. Health Perspect. 106, 583–591. https://doi.org/10.1289/ ehp.98106583. Langille, M., Zaneveld, J., Caporaso, J., McDonald, D., Knights, D., Reyes, J., Clemente, J., Burkepile, D., Thurber, R., Knight, R., Beiko, R., Huttenhower, C., 2013. Predictive functional profiling of microbial communities using 16SrRNA marker gene sequences. Nat. Biotechnol. 31, 814–823. http://www.nature.com/doifinder/10.1038/nbt.2676. Lefevre, E., Bossa, N., Wiesner, M.R., Gunsch, C.K., 2016. A review of the environmental implications of in situ remediation by nanoscale zero valent iron (nZVI): behaviour, transport and impacts on microbial communities. Sci. Total Environ. 565, 889–901. https://doi.org/10.1016/j.scitotenv.2016.02.003. Lei, C., Sun, Y., Tsang, D.C.W., Lin, D., 2018. Environmental transformations and ecological effects of iron-based nanoparticles. Environ. Pollut. 232, 10–30. https://doi.org/ 10.1016/j.envpol.2017.09.052. Liu, G., Tao, L., Liu, X., Hou, J., Wang, A., Li, R., 2013. Heavy metal speciation and pollution of agricultural soils along Jishui river in non-ferrous metal mine area in Jiangxi Province, China. J. Geochem. Explor. 132, 156–163. https://doi.org/10.1016/j. gexplo.2013.06.017. Lory, S., Wolfgang, M., Lee, V., Smith, R., 2004. The multi-talented bacterial adenylate cyclases. Int. J. Med. Microbiol. 293, 479–482. https://doi.org/10.1078/1438-4221-00297. MAPA, 1994. Official Methods of Analyses. vol. III. Ministerio de Agricultura Pesca y Alimentación, Spain, pp. 205–285. Martin, M., Gibello, A., Lobo, C., Nande, M., Garbi, C., Fajardo, C., Barra-Caracciolo, A., Grenni, P., Martinez-Inigo, M.J., 2008. Application of fluorescence in situ hybridization technique to detect simazine degrading bacteria in soil samples. Chemosphere 71, 703–710. Němeček, J., Lhotský, O., Cajthaml, T., 2014. Nanoscale zero-valent iron application for in situ reduction of hexavalent chromium and its effects on indigenous microorganism populations. Sci. Total Environ. 485, 739–747. https://doi.org/10.1016/j. scitotenv.2013.11.105. Němeček, J., Pokorný, P., Lacinová, L., Černík, M., Masopustová, Z., Lhotský, O., Filipová, A., Cajthaml, T., 2015. Combined abiotic and biotic in-situ reduction of hexavalent chromium in groundwater using nZVI and whey: a remedial pilot test. J. Hazard. Mater. 300, 670–679. https://doi.org/10.1016/j.jhazmat.2015.07.056. Němeček, J., Pokorný, P., Lhotský, O., Knytl, V., Najmanová, P., Steinová, J., Černík, M., Filipová, A., Filip, J., Cajthaml, T., 2016. Combined nano-biotechnology for in-situ remediation of mixed contamination of groundwater by hexavalent chromium and chlorinated solvents. Sci. Total Environ. 563–564, 822–834. https://doi.org/10.1016/ j.scitotenv.2016.01.019. Rasmussen, K., González, M., Kearns, P., Sintes, J.R., Rossi, F., Sayre, P., 2016. Review of achievements of the OECD working party on manufactured nanomaterials testing and assessment programme from exploratory testing to test guidelines. Regul. Toxicol. Pharmacol. 74, 147–160. https://doi.org/10.1016/j.yrtph.2015.11.004. Rodríguez, L., Ruiz, E., Alonso-Azcárate, J., Rincón, J., 2009. Heavy metal distribution and chemical speciation in tailings and soils around a Pb-Zn mine in Spain. J. Environ. Manag. 90, 1106–1116. https://doi.org/10.1016/j.jenvman.2008.04.007. Saccà, M.L., Fajardo, C., Costa, G., Lobo, C., Nande, M., Martin, M., 2014. Integrating classical and molecular approaches to evaluate the impact of nanosized zero-valent iron (nZVI) on soil organisms. Chemosphere 104, 184–189. https://doi.org/10.1016/j. chemosphere.2013.11.013.

C. Fajardo et al. / Science of the Total Environment 706 (2020) 136041 Semerád, J., Cajthaml, T., 2016. Ecotoxicity and environmental safety related to nano-scale zerovalent iron remediation applications. Appl. Microbiol. Biotechnol. 100, 9809–9819. https://doi.org/10.1007/s00253-016-7901-1. Shing, R., Misra, V., Shing, R.P., 2012. Removal of Cr(VI) ny nanoscale zero-valent iron (nZVI) from soil contaminated with tannery wastes. Bull. Environ. Contam. Toxicol. 88, 210–214. https://doi.org/10.1007/s00128-011-0425-6. Vitkova, M., Puschenreiter, M., Komarek, M., 2018. Effect of nano zero-valent iron application on As, Cd, Pb and Zn availability in the rhizosphere of metal (oid) contaminated soil. Chemosphere 200, 217–226. https://doi.org/10.1016/j. chemosphere.2018.02.118. Wang, C.B., Zhang, W., 1997. Synthesizing nanoscale iron particles for rapid and complete dechlorination of TCE and PCBs. Environ. Sci. Technol. 31, 2154–2156. https://doi.org/ 10.1021/es970039c. Wang, Y., Fang, Z., Kang, Y., Tsang, E.P., 2014. Immobilization and phytotoxicity of chromium in contaminated soil remediated by CMC-stabilized nZVI. J. Hazard. Mater. 275, 230–237. https://doi.org/10.1016/j.jhazmat.2014.04.056. Wuana, R.A., Okieimen, F.E., 2011. Heavy metals in contaminated soils: a review of sources, chemistry, risks and best available strategies for remediation. ISRN Ecol. 2011, 402647. https://doi.org/10.5402/2011/402647. Xie, Y., Fan, J., Zhu, W., Amombo, E., Lou, Y., Chen, L., Fu, J., 2016. Effect of heavy metals pollution on soil microbial diversity and bermudagrass genetic variation. Front. Plant Sci. 7, 755–761. https://doi.org/10.3389/fpls.2016.00755.

11

Xu, X., Zhang, Z., Hu, S., Ruan, Z., Jian, J., Chen, C., Shen, Z., 2017. Response of soil bacterial communities to lead and zinc pollution revealed by Illumina MiSeq sequencing investigation. Environ. Sci. Pollut. Res. Int. 24, 666–675. https://doi.org/10.1007/s11356016-7826-3. Xue, W., Huang, D., Zeng, G., Wan, J., Cheng, M., Zhang, C., Hu, C., Li, J., 2018. Performance and toxicity assessment of nanoscale zero valent iron particles in the remediation of contaminated soil: a review. Chemosphere 210, 1145–1156. https://doi.org/10.1016/ j.chemosphere.2018.07.118. Ye, S., Zeng, G., Wu, H., Zhang, C., Liang, J., Dai, J., 2017. Co-occurrence and interactions of pollutants, and their impacts on soil remediation—a review. Crit. Rev. Environ. Sci. Technol. 47, 1528–1553. Ye, S., Zeng, G., Wu, H., Zhang, C., Liang, J., Dai, J., Ye, S., Xiong, W., Song, B., Wu, S., Yu, J., 2019. The effects of activated biochar addition on remediation efficiency of cocomposting with contaminated wetland soil. Resour. Conserv. Recycl. 140, 278–285. Yu, B., Jin, X., Kuang, Y., Megharaj, M., Naidu, R., Chen, Z., 2015. An integrated biodegradation and nano-oxidation used for the remediation of naphthalene from aqueous solution. Chemosphere 141, 205–211. https://doi.org/10.1016/j. chemosphere.2015.07.050.