Application of Genomic Tools in Bioremediation of Atrazine Contaminated Soil and Groundwater

Department of Food and Environmental Sciences Faculty of Agriculture and Forestry University of Helsinki

APPLICATION OF GENOMIC TOOLS IN BIOREMEDIATION OF

ATRAZINE CONTAMINATED SOIL AND GROUNDWATER

Aura Nousiainen

Academic Dissertation

To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public examination in Auditorium 1041 at Viikki Biocenter 2, on February 13th 12 o’clock noon. 1

Supervisors: Docent Kirsten Jørgensen Marine Research Centre Finnish Environment Institute

Dr Katarina Björklöf Laboratory Centre Finnish Environment Institute

Pre-examiners: Dr Françoise Binet Ecosystèmes, Biodiversité, Evolution, ECOBIO University of Rennes

Professor Martin Romantschuk Department of Environmental Sciences University of Helsinki

Opponent: Professor Carsten Suhr Jacobsen Department of Plant and Environmental Sciences University of Copenhagen and Department of Geochemistry and CENPERM Geological Survey of Denmark and Greenland

Custos: Professor Kaarina Sivonen Department of Food and Environmental Sciences University of Helsinki

Front cover photo by Kirsten Jørgensen ISBN 978-951-51-0814-2 (paperback), ISBN 978-951-51-0815-9 (PDF) ISSN 2342-5423 (paperback), ISSN 2342-5431 (PDF) Electronic version available at http://ethesis.helsinki.fi Hansaprint, Helsinki 2015

A scientist in his laboratory is not a mere technician: he is also a child confronting natural phenomena that impress him as though they were fairy tales.

Marie Curie in Madame Curie : A Biography (1937)

ABSTRACT

The use of pesticides has allowed the efficient use of agricultural soil and provided humans with greater yields and agri-food security. Unfortunately, many pesticides have also adverse effects to the environment or human health, and may end up where they were not intended: the precious groundwater reserves. The use of atrazine, a herbicide used for controlling broad-leaf weeds, was banned in the EU for this reason in 2004, but is still globally one of the most widely used herbicides today. Although atrazine can be completely mineralized by microbes, in the subsurface, slow or incomplete degradation of atrazine is often observed. The ability of atrazine degradation by microbes can be utilized in bioremediation, a technique in which contaminants are removed by microbial activity. This study was undertaken to elucidate the potential use of genetic tools, such as quantitative PCR (qPCR), radiorespirometry, microautoradiography (MAR), clone libraries and genetic fingerprinting methods, in atrazine contaminated soils, and to apply them in atrazine bioremediation. Collaboration with our Indian partner permitted comparison between atrazine treated, cropped agricultural soils and boreal subsoil contaminated two decades ago with residual atrazine from weed control in municipal areas. Four different bioremediation methods, natural attenuation, bioaugmentation, biostimulation, and their combination, were used to reduce atrazine concentration in soil. Atrazine degradation gene copy numbers often reflected the atrazine degradation potential, indicating their robustness as monitoring tools in different soils. The most efficient bioremediation treatment was bioaugmentation by atrazine degrading bacterial strains Pseudomonas citronellolis or Arthrobacter aurescens, or by an atrazine degrading bacterial consortium: in the agricultural soil, up to 90% of atrazine was degraded in less than a week, whereas in the boreal subsoil, 76% of atrazine was mineralized. In the clone library constructed from boreal soil, several clones related to taxa which include known atrazine degraders were found. In this soil, biostimulation with additional carbon was an efficient treatment at reduced temperature. In general, the efficiency of atrazine removal in different treatments was bioaugmentation and biostimulation > bioaugmentation > biostimulation > natural attenuation. Previous exposure to atrazine was the most influential factor in atrazine

4 disappearance from soil, as recent exposure always correlated with faster atrazine degradation, and greatly affected the composition of the microbial community, elucidated by LH-PCR. These results serve as an example on how soil origin, exposure history, organic content and use must be taken into account while choosing the best bioremediation method. Knowledge on the presence of genetic degradation potential can be helpful in choosing the treatment method. While bioaugmentation removed 90% of atrazine from soil, its application in field scale may be challenging. Our results show, that biostimulation with carbon alone may serve as the treatment method of choice, even in the challenging subsoil surroundings where atrazine concentrations are low.

TIIVISTELMÄ

Torjunta-aineiden käyttö on turvannut ihmisten ruoansaantia mahdollistamalla viljelymaan tehokkaan käytön ja vähentämällä satomäärien vaihtelua. Monilla torjunta- aineilla on kuitenkin ympäristön ja terveyden kannalta haitallisia vaikutuksia, ja ne saattavat myös päätyä sinne, minne niitä ei tarkoitettu: arvokkaisiin pohjavesivarantoihin. Euroopan Unioni kielsi mm. vesakontorjunnassa käytetyn torjunta-aine atratsiinin vuonna 2004, mutta se on yhä eräs maailman käytetyimmistä herbisideistä. Vaikka mikrobit kykenevät mineralisoimaan atratsiinin, syvissä maakerroksissa sen hajoaminen hidastuu merkittävästi. Biopuhdistus on tekniikka, jossa haitta-ainetta hajotetaan mikrobitoiminnan avulla. Mikrobien kykyä hajottaa atratsiinia voidaankin käyttää hyväksi biopuhdistuksessa. Tämän tutkimuksen tarkoituksena oli selvittää molekyylibiologisten menetelmien, kuten kvantitatiivisen PCR:n (qPCR), radiorespirometrian, mikroautoradiografian (MAR), kloonikirjastojen ja geneettisten sormenjälkitestien soveltuvuutta atratsiinin biopuhdistukessa ja testata eri puhdistusmenetelmiä erityyppisissä pilaantuneissa maissa. Yhteistyö intialaisen kumppanin kanssa mahdollisti vertailun aktiivisessa viljelykäytössä olevan trooppisen peltomaan ja pohjavesikontaktin välityksellä kaksi vuosikymmentä sitten pilaantuneen boreaalisen pohjamaan välillä. Maan atratsiinimääriä pyrittiin vähentämään käyttämällä neljää eri biopuhdistusmenetelmää: luontaista puhdistumista, biostimulaatiota, bioaugmentaatiota sekä biostimulaation ja –augmentaation yhdistelmää Atratsiininhajotusgeenien lukumäärät vastasivat usein biohajotuspotentiaalia eri maatyypeissä, mikä osoittaa niiden olevan luotettava parametri biohajoamisen seurannassa. Tehokkain biopuhdistusmenetelmä oli bioaugmentointi atratsiinia hajottavilla mikrobikannoilla Pseudomonas citronellolis ja Arthrobacter aurescens tai atratsiinia hajottavalla mikrobiyhteisöllä. Trooppisessa viljelymaassa jopa 90 % atratsiinista hajosi alle viikossa, kun taas boreaalisessa pohjamaassa mineralisoitui enintään 76 %. Boreaalisesta maasta tehdystä kloonikirjastosta tunnistettiin useampi atratsiininhajottajia sisältävä bakteerisuku. Tässä maassa myös biostimulaatio hiilenlähteellä oli tehokas hajotuskeino alhaisessa lämpötilassa. Yleisesti ottaen

6 atratsiinin hajotustehokkuus eri maissa noudatti järjestystä bioaugmentaatio ja biostimulaatio > bioaugmentaatio > biostimulaatio > luontainen puhdistuminen. Aiempi altistus atratsiinille vaikutti merkittävimmin atratsiinin vähenemiseen, sillä hiljattain tehty atratsiinikäsittely korreloi nopeaan atratsiinin hajoamiseen. Atratsiinikäsittely myös vaikutti suuresti mikrobiyhteisön rakenteeseen, mikä kävi ilmi geneettisestä sormenjäljestä. Nämä tulokset osoittavat, kuinka maankäyttö, maantieteellinen alkuperä, käsittelyhistoria ja orgaanisen aineen määrä tulee ottaa huomioon valittaessa sopivinta biopuhdistusmenetelmää. Lisäksi tieto maan geneettisestä hajotuspotentiaalista auttaa menetelmän valinnassa. Vaikka bioaugmentaation avulla voitiin vähentää atratsiinin määrää 90 %, sen käyttö kenttämittakaavassa on haastavaa. Tämän tutkimuksen tulokset osoittavat, että pelkkä biostimulaatio hiilenlähteellä saattaa olla riittävä puhdistusmenetelmä jopa vähän atratsiinia sisältävän pohjamaan puhdistuksessa.

TABLE OF CONTENTS

Abstract ...... 4 Tiivistelmä ...... 6 List of Original Publications ...... 9 Author’s Contribution ...... 9 Abbreviations ...... 10 1. Introduction ...... 11 1.1 Microbial Ecology ...... 11 1.2 Soil metagenomics ...... 13 1.3 Bioremediation ...... 16 1.4 Scaling up ...... 18 1.5 Atrazine use and environmental fate ...... 20 1.6 Toxicity of atrazine ...... 25 1.7 Atrazine bioremediation ...... 26 2. Objectives of this Thesis ...... 29 3. Material and methods ...... 31 3.1 Soils ...... 31 3.2 Analytical methods ...... 32 3.3 Bioremediation treatments ...... 34 4. Results ...... 36 4.1 Atrazine degradation genes in different soils and bioremediation treatments ...... 36 4.2 Bioremediation treatments in different soils ...... 36 4.3 Microbial community dynamics in atrazine treated soil ...... 37 5. Discussion ...... 41 5.1 Application of genomic tools in atrazine bioremediation ...... 41 5.2 Natural atrazine biodegdration potential ...... 42 5.3 Microbial community and species dynamics ...... 44 5.4 Atrazine bioremediation strategies...... 46 5.5 scaling up ...... 48 5.6 Summary ...... 50 6. Conclusions and Future Prospects ...... 51 7. Acknowledgements ...... 52 8. References ...... 54

LIST OF ORIGINAL PUBLICATIONS

I) Nousiainen AO, Björklöf K, Sagarkar S, Mukherjee S, Purohit HJ, Kapley A, Jørgensen KS (2014). Atrazine degradation in boreal subsoil and tropical agricultural soil. Journal of Soils and Sediments 14: 1179-1188

II) Sagarkar S, Mukherjee S, Nousiainen A, Björklöf K, Purohit HJ, Jørgensen KS, Kapley A (2013). Monitoring bioremediation of atrazine in soil microcosms using molecular tools. Environmental Pollution 172: 108-115

III) Nousiainen AO, Björklöf K, Sagarkar S, Nielsen JL, Kapley A, Jørgensen KS. Effect of biostimulation and bioaugmentation on atrazine degradation in boreal subsoil. Submitted manuscript

IV) Sagarkar S, Nousiainen AO, Shaligram S, Björklöf K, Lindström K, Jørgensen KS, Kapley A (2014). Soil mesocosm studies on atrazine bioremediation. Journal of Environmental Management 139: 208-216

AUTHOR’S CONTRIBUTION

I) Aura Nousiainen took part in planning of the sampling and collecing samples. She developed the qPCR method for different types of soils, and carried out the qPCR analyses. She carried out the qPCR and statistical analyses. She interpreted the qPCR and radiorespirometric results. She was responsible for writing the manuscript and is the corresponding author.

II) Aura Nousiainen took part in planning of the sampling and collecting samples. She participated in development of the qPCR method and took part in calculating the results and writing the manuscript.

III) Aura Nousiainen took part in planning of the sampling and collecting samples. She planned the experimental setup, carried out the analyses, and interpreted the results. She was responsible for writing the manuscript and is the corresponding author.

IV) Aura Nousiainen carried out the community analyses by LH-PCR fingerprinting, and calculated species diversities and indexes. She interpreted the microbial community results and wrote a part of the paper.

ABBREVIATIONS

DEA Desethylatrazine DEDIA Desethyl-deisopropylatrazine DGGE denaturing gradient gel electrophoresis DIA Deisopropylatrazine DNA deoxyribonucleic acid FISH Fluorescent in situ hybridization HA Hydroxyatrazine HPLC high pressure liquid chromatography LH-PCR length heterogeneity-PCR MAR microautoradiography mRNA messenger ribonucleic acid NGS next generation sequencing PAH polyaromatic hyrdocarbon PCR polymerase chain reaction qPCR quantitative polymerase chain reaction RNA ribonucleic acid rRNA ribosomal ribonucleic acid RT-qPCR reverse transcription – quantitative polymerase chain reaction SIP stable isotope probing T-RFLP terminal restriction fragment length polymorphism

1. INTRODUCTION

1.1 Microbial Ecology

Microbial ecology is the study of the presence, nature and interactions of the bacterial, archaeal, and fungal communities living in the environment; soil, sediment, ocean, freshwater or inside of the animal gut. Central questions in microbial ecology are:

Who are there? What can they do? What are they actually doing?

The microbial ecologist attempts to answer these questions by identifying the abundance and diversity, inter-species relationships, as well as biochemical turnover of compounds between the different organisms in the ecosystem. The species identities of the bacteria residing in soil (who are there) are indicative of the kinds of processes they might be capable of carrying out (what can they do). This knowledge is based on what we know about their cultivable relatives. What the bacteria are actually doing can be assessed by the identification and quantification of the RNA molecules transcribed from functional (i.e. catabolic) genes. The emerging –omics -approaches aim to pinpoint these attributes to a specific point in time.

Micro-organisms are responsible for degrading dead, organic matter. They are also key players in the global nitrogen and carbon cycles, and in this way have a pronounced impact on the entire biosphere. Bacteria are everywhere, and in fact, sterile environments are a rarity in the environment. In general, bacterial

11 diversity, determined by species richness and evenness, is low in simple, heterogeneous environments, and increases as more ecological niches become available.

Soil is a heterogeneous habitat, where the structure of soil dictates how water and gas move in the matrix (Young and Crawford, 2004). Water and gas content of the soil, in turn, affect the factors governing biological activity. The most prominent environmental parameters governing the microbial activity are temperature, water and nutrient availabilities, bioavailability of the contaminant, pH, degradation and sorption to soil particles. (Arias-Estévez et al, 2008). These parameters can change in the scale of centimeters or meters, even if the change is not evident from soil surface (Ettema and Wardle, 2002), dependent on plant root exudates (Saetre and Bååth, 2000) and nutrient hot-spots (Wachinger et al, 2000). Burrowing animals, such as earthworms, impact soil structure, and bacterial growth may be slowed down by bacteriophages or predation by protozoa. Thus, apparently heterogeneous soil may in fact contain a great diversity of microhabitats, which can sustain exceptional species richness (Ettema and Wardle, 2002). An excellent review by Giller likened soil animal communities to rainforests, where resource partitioning, habitat heterogeneity and trophic specialization allow species coexistence (Giller 1996). It is likely, that microhabitat diversity is the main reason for the number of different microbes in soil. The abundance of bacterial species in soil is large: 103 – 106 different bacterial genomes have been observed in one gram of soil (Torsvik et al 1998, Roesch et al 2007 Gans 2005).

Micro-organisms have two different survival strategies. The K-strategists are fastidious, slow-growing microbes which invest much of the available energy in slow growth and few cell divisions. The r-strategists survive on feast or famine principle, where available energy is invested in quick cell division (Andrews and

Harris 1986). Competing for limited resources make microbes resilient and quickly adaptable to changes in their surroundings. Sometimes a feast for microbes is resulted by an accidental spill of xenobiotic compounds in the environment. When the environmental parameters change, certain types of organisms benefit from the alteration. Organisms that have the capability to either withstand the toxicity, or to utilize the newly introduced compound in its metabolism, have a competitive edge. The microbial ability to degrade xenobiotics can be utilized to remediate contaminated environments. However, the success of the treatment is directed by the demands of the organisms harboring the degradation genes.

1.2 Soil metagenomics

Only a small fraction of bacteria can be studied live in laboratory conditions (Epstein 2013), and therefore DNA-based methods are used to attain improved knowledge on soil functions. The study of genetic material (DNA or RNA) recovered directly from the environment is called metagenomics.

Metagenomics is a powerful tool for answering the fundamental questions in microbial ecology. The question “who are there” is elucidated by comparing through sequencing the DNA sequences of a housekeeping gene. The gold standard for defining bacterial and archaeal species is the gene coding for the 16S subunit of the rRNA (ribosomal RNA), which is widely used for interpreting the structure of bacterial communities (Hugenholtz et al. 1998, Rondon et al. 2000). The problem with this approach is the absence of a direct link between the phylogenetic identities of the microbe and its ecophysiology, and therefore little insight is gained of ecosystem functionality by studying 16s rRNA genes (Rodrígues-Valera 2004). Answering the question of “what can they do” is slightly more complicated, but is often achieved by identification of the presence and activity of catabolic genes, which allow bacteria to utilize the chemical compounds in their surroundings for cell constituents and as an energy source. Fast adaptation to changing environmental conditions is possible due to genetic plasticity attributable to mobile, short 13

DNA material, such as plasmids and insertion sequences, which can easily be shared between same or even distantly related species (Ochman et al. 2000). Because the mobile elements may be transferred between different bacterial species, two bacteria having high 16s rRNA sequence homologies may carry out different metabolic reactions. The question “what are they doing” is answered by measuring the activity encoded by the RNA transcribed from catabolic genes or other biomolecules.

The greatest challenge of soil metagenomics lies in the nucleic acid extraction from soil matrix (van Elsas and Boersma 2011). Extraction efficiency from bacteria is dependent on the structure of the cell wall, which varies from species to species, as well as the growth form (Prosser 2002). Different soils contain variable amounts of humic substances, which co-extract with the nucleic acids. Therefore different DNA extraction methods influence the results obtained from downstream applications (Martin-Laurent et al 2001). Although the nucleic acid extraction from soil bacteria, and the subsequent purification of nucleic acids from the soil matrix is prone to biases, the most serious distortion is created when the nucleic acids are subjected to the cornerstone technique, PCR (polymerase chain reaction), where the gene of interest is exponentially multiplied in vitro (van Elsas and Boersma 2011).

Because most soil bacteria can only be studied based on their DNA molecules, several DNA-targeting methods to study the metagenome have been developed to answer the central questions in microbial ecology (Table 1). The composition of the bacterial community is elucidated by sequencing the 16s rRNA genes either by cloning, or by using the Next Generation Sequencing (NGS) methods developed just a few years ago (Solomon et al 2013). Tracking changes in the composition of microbial community can be carried out by creating “DNA-fingerprints” of the bacterial community. These methods, thoroughly reviewed by Kirk et al. (2004) and Nocker et al. (2007), are based on creating images of the community at different time points by separating the 16s (or other) genes in the microbial community, and calculating the changes over time using mathematical methods. The function of the microbial community is addressed by detection and quantification of the catabolic genes, using e.g. quantitative PCR methods or Fluorescent in situ hybridization techniques (FISH). Environmental transcriptomics, 14 the quantification of gene expression by mRNA molecules in the environment, enables measuring the activity of catabolic genes (Yergeau et al. 2012, Monard et al. 2013).

The most powerful metagenomic tools are often combinations of DNA techniques and chemical or microscopy methods. For example, by using radionuclides or stable isotopes as markers in stable isotope probing (SIP), the flow of molecules used by bacteria as nutrients to building blocks of DNA can be measured (Radajewski et al. 2000). Also fluorescent in situ hybridization techniques are combined with radionuclides for metabolic studies (Lee et al. 1999).

Table 1. Examples of genomic tools used in microbial ecology Tool Used for Reference Cloning and sequencing Identification of community members Pace et al. of ribosomal DNA and their relative abundances, diversity 1986 indexes

2nd Generation Identification of community members Margulies et al. 2005, Sequencing (NGS) and their relative abundances Bentley et al. 2008, Rothberg et al. 2011

Quantitative PCR, Quantification of genes (DNA) and their Holland et al 1991, RT-qPCR expression (mRNA) Rasmussen et al 1998

DGGE, T-RFLP, LH- Tracking changes in the bacterial Muyzer et al. 1993 PCR and other community structure over time Avaniss-Aghajani et fingerprinting methods al. 1994, Suzuki et al. 1998

Fluorescent in situ Identification and metabolic functions of Amann et al., 1995 hybridization (FISH) community members in situ by microscopy

Stable isotope probing Uptake of labelled substrates under Radajewski et al 2000 (SIP) defined conditions

1.3 Bioremediation

Bioremediation is a name for biotechnologies, listed in Table 2, that aim for environmental decontamination by microbiological processes either on site, where contamination can either be remediated without removing soil (in situ), treat the excavated soil on site, or away from the site (ex situ). It is an environmental friendly option for using the excavated, contaminated soil contrasts to landfilling or incineration. The benefits of bioremediation are low cost, permanent mineralization of the contaminant and the possibility to carry out the process in situ, although applicability screening may be required in groundwater remediation (Caliman et al. 2011).

Table 2. Bioremediation technologies Technology Process Bioaugmentation Addition of contaminant degrading microbes to the environment. Biostimulation Stimulating the indigenous microbes by providing growth supplements Phytoremediation Use of plants in contaminant elimination. Natural attenuation Monitoring the contaminant removal of the indigenous microbes. Biofilters Use of immoblilized biofilters in filtering contaminated water or air. Bioventing Enhancing aerobic degradation by providing (soil)microbes with air Land farming Tilling of contaminated topsoil. Pump and treat Pumping contaminated water to surface for treatment and reinjection.

The advantages in in situ treatments are obvious: if physical removal of soil is not required, soil transportation costs are nonexistent, and impact on landscape is moderate. In situ bioremediation treatments include land farming, bioventing, biofilters, bioaugmentation, biostimulation, natural attenuation and pump and treat method (Boopathy 2000). The greatest challenges in in situ treatments are the low mass transportation of contaminants, distribution of contaminant degrading organisms and the biostimulants, as well as long treatment times and the unpredictability of the biological systems bioremediation techniques rely on.

Even though ex situ bioremediation is more labor intensive than in situ methods, ex situ land farming, degradation in bioreactors, and composting hazardous wastes are often more economically sound option to landfilling, as shorter transportation distances are often sufficient, and they may also be more efficient, as the ex situ methods allow the optimization of treatment conditions. Albeit dredging of contaminated sediments is risky, their decontamination by ex situ bioremediation has been identified as a powerful, cost limiting tool in decontamination (Perelo 2010). Many successful bioremediation treatments have been used to decontaminate topsoil, but remediating subsoil is more demanding (Höhener and Ponsin 2014). Soil acts as a filter, creating a restricting barrier for microbial cells, and their dispersion is hindered. Fungi have been used to overcome this obstacle (Winquist et al 2014), but more often decontamination of the deeper soil layers is carried out by biostimulation. Vadose zone remediation is often accomplished by supplementing the subsoil with electron donors and acceptors, inorganic nutrients and oxygen (Salminen et al 2006, National Research Council 1993). These components can be applied in groundwater monitoring wells, where water acts as the medium conveying them to bacteria. The roadblocs in bioremediation are critically reviewed by Romantschuk et al. (2000).

Bioremediation has successfully been used in pilot– and field scale, and statistically relevant, significant contaminant removals have been shown for e.g. crude oil (Bragg et al. 1994), trichloroethylene (Eguchi et al. 2001), PAH compounds (Winquist et al. 2014) and for bioreduction of uranium (Williams et al. 2013). The preferred treatment method depends on the nature and concentration of the contaminant, as well as the environmental conditions.

1.4 Scaling up

Because multiple parameters affect the treatment outcome, and bioremediation relies on unpredictable biological systems, the direct use of results from a laboratory experiments to field scale is not feasible (Perelo 2010). Several factors affect terrestrial systems only when the scale –the mass of soil being decontaminated, is increased (Figure 1). Moreover, a number of parameters, such as temperature, rainfall and microbial competition, cannot easily be controlled in the field. Because the complexity of the system increases with size, gradual upscaling from microcosm to mesocosm to pilot scale is often used to bridge bioremediation research into biotechnological applications (Allard and Neilson 1997).

The scales that are relevant in the field conditions span from that of a microbial cell (10-6 m) to the field (10-100 m). Contaminant degradation takes place in microscale, in vicinity of the degrading bacteria, which can be tested in microcosms. In order to ascertain that biodegradation takes place in all parts of the contaminated environment, several macroscale phenomena, such as the contaminant dispersion through the bulk soil and the heterogeneity of the environment, must be taken into account (Sturman et al 1995). Increased scale affects microbial degradation, sorption kinetics, and mass transport (Rittmann et

18 al 1992), these phenomena can be tested in mesocosm scale (1-10 kg of contaminated soil).

When the monitoring tools are optimized, and contaminant degradation can be demonstrated, the system can be tested in pilot scale. In pilot- and field scales, transport may be dominated by diffusion and preferential flow. This may limit the availability of the reactants, and the decontamination time may be increased by 4- 10 times compared to laboratory scale (Sturman et al 1995). Despite this, bioremediation of many different kinds of contaminants has been successfully carried out (Megharaj et al. 2011).

Bench top scale Mesocosm scale Pilot scale Full scale

Slower mass transport

Increasing microbial competition

Water concentration fluctuations

Temperature fluctuations

Increasing heterogeneity

Figure 1: Changing parameters in upscaling. Modified from Sturman et al. 1995.

1.5 Atrazine use and environmental fate

Atrazine (1-Chloro-3-ethylamino-5-isopropylamino-2,4,6-triazine) is a pesticide used for blocking the photosystem II in broad leaf weeds mainly in the cultivation of maize, sorghum and sugarcane. It is globally one of the most widely used pesticides today, with up to 37 000 tonnes applied annually in United States alone (Thelin and Stone 2010). Atrazine is notoriously recalcitrant groundwater polluter, commonly encountered in the streams and groundwater in agricultural, urban and undeveloped areas (Gilliom et al. 2007, Vuorimaa et al. 2007, Gerecke et al. 2002). Persistence to degradation has led to banning atrazine in the European Union in 2004 (2004/248/EC), but is still heavily used in all other continents.

Atrazine persistence in groundwater was the main reason for banning its use in the European Union in 2004, and its concentration in Finnish surface- and groundwater, as well as drinking water, is regulated by European Union and national legislation. Permissible concentrations are set lower than most countries, e.g. the United States and Canada. According to the Water Framework Directive, atrazine and its degradation product concentration in surface waters should not exceed 0.6 µg L-1 for any given time, and the annual average should be lower than 0.2 µg L-1 (Water Framework Directive, 2000/60/EC). According to a recommendation by the World Health Organization, atrazine concentration in drinking water should not exceed 2 µg L-1 (WHO 1996). The Finnish legal limit, based on European Union legislation, is significantly lower than this. The permitted value of a single pesticide in groundwater 0.1 µg L-1, and the sum of different pesticides should not exceed 5.0 µg L-1 (Government Decree on Water Resources Management 2006/1040). This limit is exceeded in many sites where groundwater is used as a source for drinking water. In Finland, groundwater reserves are vulnerable since they are often restricted to shallow Quaternary deposits (Backman et al 2007). A survey of pesticides and their degradation products in 189 groundwater reserves revealed that the limit values were exceeded in 8% of sites (Figure 2) (Vuorimaa et al 2007). Atrazine, (in 26% of cases), and its degradation products DEA (20% of cases), DEDIA (9% of cases) and DIA (9% of cases), were the most commonly observed pesticides. 20

Figure 2. The amounts of pesticides, and their geographic distribution in 295 groundwater samples analyzed in a Finnish survey during 2002-2005. From Vuorimaa et al. 2007.

Even though atrazine was invented and first introduced in the environment in 1958, indication of its microbial degradation became evident already in the early 1960’s (Skipper et al 1967), but the genes involved in the two possible degradation pathways were not discovered before the mid-1990´s (Figure 3). Degradation is initiated either by a dehalogenation of the chlorine atom (pathway I in Figure 3), or by a dealkylation of one of the side chains, generating deisopropylatrazine (DIA) or desethylatrazine (DEA)

(pathway II in Figure 3), which also accumulate in the groundwater. Both pathways funnel to a central metabolite, cyanuric acid, which is finally completely mineralized.

The microbiological dealkylation of the amino side chains is governed by thcBCD genes coding for a P-450 cytochrome enzyme system (Nagy et al 1995). A similar dealkylation is carried out by the trzA gene in certain Rhodococcus species. Because the dealkylation reaction is not specific for atrazine degradation, this pathway is little studied. In contrast, a great deal is known about the presence, prevalence, genetic structure and function of the atrazine hydrolytic dechlorination pathway, mediated by genes atzABCDEF (Martinez et al. 2000, de Souza et al. 1996). This pathway is linked to a catabolic plasmid, pADP1 (de Souza et al 1998.), but some of the genes may also be chromosomal (Cai et al 2003, Devers et al. 2007). The pathway is initiated by hydrolytic removal of the chlorine atom by atrazine chlorohydrolase atzA, coded by the atzA gene. A triazine hydrolase, functional homolog to atzA, was later identified in Gram positive bacteria, and named trzN (Mulbry et al. 2002).

Figure 3. Atrazine, some of its degradation products and degradation genes. (I) Degradation pathway initiated by dechlorination, (II) degradation pathway initiated by dealkylation.

Atrazine is biologically degraded by several microbial taxa, as well as certain fungi (Kaufman and Blake 1970, Levanon 1993), either in co-metabolic reactions, or as a source of energy and nutrients (Mudhoo and Garg 2010). Since atrazine contains a fairly high amount of N, many bacteria use it as a nitrogen source (Yang et al 2010). Some known atrazine degrading bacteria are listed in Table 3. Because atrazine degradation genes are often plasmid borne, and often flanked by mobile genetic elements, they are easily spread across different bacterial taxa, as well as different geographic areas in a relatively short time (de Souza et al 1998). The fast global dissemination of the atrazine degradation genes is likely due to plasmids acting as harbors of catabolism genes, while transposons transfer the degradation genes between various broad host-range plasmids (Wackett et al. 2002). Lateral gene transfer is a beneficial treat in bioremediation treatments because of the ease by which degradation genes are transferred between potential degrader organisms.

Table 3: List of known cultivated atrazine degrading bacterial genera Genus Origin Phylogeny Degradation genes Gene location Reference Pseudomonas Herbicide spill site γ-proteobacteria atzABCDEF plasmid Mandelbaum et al. 1995 Chelatobacter Grassland soil α-proteobacteria atzABC/trzD plasmid Rousseaux et al. 2001 Aminobacter Grassland soil α-proteobacteria atzC/trzD chromosome Rousseaux et al. 2001 Stenotrophomonas Grassland soil γ-proteobacteria atzA plasmid Rousseaux et al. 2001 Arthrobacter Grassland soil Actinobacteria trzN/atzBC plasmid Rousseaux et al. 2001 Nocardia Atrazine treated soil Actinobacteria trzN Plasmid Mulbry et al. 2002 Rhodococcus EPTC-treated soil Actinobacteria thcBCD plasmid Shao and Behki 1996 Rhodococcus EPTC-treated soil Actinobacteria atrA plasmid Shao and Behki 1995 Rhodococcus EPTC-treated soil Actinobacteria trzA plasmid Shao et al 1995 Sinorhizobium Agricultural field α-proteobacteria trzN/atzBC chromosome Devers et al. 2007 Polaromonas Agricultural field β-proteobacteria trzN/atzBC chromosome Devers et al. 2007 Agrobacterium Agricultural field α-proteobacteria atzA plasmid Struthers et al. 1998 Burkholderia Agricultural field β-proteobacteria unknown unknown Ngigi et al. 2012 Microbacterium River sediment Actinobacteria dechlorinating unknown Vargha et al. 2005 Bacillus River sediment Firmicutes dechlorinating unknown Vargha et al. 2005 Micrococcus River sediment Actinobacteria dechlorinating unknown Vargha et al. 2005 Deinococcus River sediment Deinococcus-Thermus dechlorinating, dealkylating unknown Vargha et al. 2005 24

According to the pesticide manufacturer, atrazine half-life is >77 days in water, and 43 days in soil (Syngenta 2009). However, several studies have suggested, that atrazine has significantly longer half-lives in subsoil (Agertved et al. 1991, Talja et al. 2008, Willems et al. 1996). The major process limiting mass transportation and bioavailability of atrazine in soil is sorption (Sun et al 2010). Several studies support this finding by showing that reduced atrazine degradation is attributed to soil type. Especially soils with high organic carbon content and clay minerals sorb atrazine (Binet et al 2006, Park et al 2006 Kasozi et al. 2012). Sorption also directs atrazine leaching, which is typically greater in topsoil than subsoil (Bedmar et al 2011). Atrazine degrading bacteria provide enhanced bioremediation in highly sorbent soils, as they have means by which they can degrade soil-sorbed atrazine, either by degrading the sorbed pesticide or desorbing it by surfactants (Park et al. 2006). Another bottleneck in atrazine degradation in subsoil is the lack of available electron acceptors, such as oxygen. Studies carried out in constructed wetland microcosms suggest that anaerobic conditions may retard atrazine mineralization from days to years (Douglass et al. 2014).

1.6 Toxicity of atrazine

Although atrazine is not acutely toxic, it has been identified as an endocrine disrupting agent. Animal studies have shown that exposure to atrazine alters the hypothalamic control of the pituitary-ovarian function in rats (Cooper et al. 2000) and depresses plasma corticosterone levels in salamander larvae (Larson et al 1998) In vitro experiments revealed that dopamine and norepinephrine levels decrease in neuronal cells of adrenal medulla (Das et al 2000). The appearance of mammary tumors in atrazine-treated female rats is evidence of its possible carcinogenicity at high doses (Eldridge et al 1999).The toxicity of manufactured anthropogenic chemicals is tested by using standardized methods published by e.g. International Organization for Standardization and United States Environmental Protection Agency. Many of the standard tests measure the toxicity of a single compound to the test organism, even though chemicals may have a significant combined effect even at low concentrations (Kortenkamp 2007). The combined effect of atrazine and other common groundwater pollutants at environmentally relevant

25 concentrations has been shown to cause changes in the immune, endocrine and nervous parameters in mice (Jaeger et al. 1999). The European Union has taken steps in recognizing the combined effect of endocrine disrupting chemicals in the environmental protection and human health as communications to the Council (European Commission 2012), but no legislation exists today.

1.7 Atrazine bioremediation

The most commonly studied atrazine degrading species are members of Actinobacteria and Proteobacteria, and especially the γ-proteobacterium Pseudomonas has been used in bioremediation. The relative abundance of γ-proteobacteria and Actinobacteria is associated with atrazine use history, as these bacteria become more common in atrazine contaminated soil (Ghosh et al 2009). During the past decade, several successful pilot- and field scale bioremediation studies on removing atrazine from topsoil have been carried out. Atrazine removal of 90% or more is not uncommon in favourable conditions (Table 4). Up to 99% of atrazine has been removed by bioaugmentation with atrazine degrading bacteria (Chelinho et al. 2010), but if a single limiting factor, such as lack of available water, is limiting degradation, biostimulation alone can be used to remove 50% of atrazine (Moreno et al. 2007). However, atrazine concentrations used in many studies are not environmentally relevant, and simulate a chemical spill scenario (Strong et al. 2000, Struthers et al. 1998). According to the manufacturers, typical atrazine application in the field is 2.25 kg ha-1 annually. Assuming average weight of agricultural soil is 1400 kg m-3, the atrazine concentration in the top 10 cm of soil would be 1.6 mg kg-1. A comparison of atrazine concentrations used in different studies listed in Table 4 show the discrepancy between atrazine concentrations in bioremediation studies and the environment. A few studies using agriculturally relevant concentrations have been made in microcosm scale using a white-rot fungus Trametes vesicolor in dry environment (Bastos and Magan, 2009).

Table 4. Examples of atrazine bioremediation studies. Microcosm: 5-200 g soil, Mesocosm: 500 g – 10 kg, Pilot: 100 kg, Field: direct application in environment

Soil Method Treatment Treatment Max Atrazine Reference scale atrazine concentration degraded

Dry clay soil Trametes versicolor (fungus) Microcosm 98% 0.5 mg kg−1 Bastos & Magan 2009 Semiarid field soil BS2 Water Microcosm 50% 0.2-1000 mg kg−1 Moreno et al. 2007 Sandy and clay soils MNA3 MNA Field 10 kg/ha Aislabie et al. 2004 Sterile garden soil BA, BS Surfactants, Acinetobacter Microcosm 80% 250 mg kg−1 Singh & Cameotra 2014 Sterile soil BA Arthrobacter sp. Microcosm 98% 300 mg kg−1 Li et al. 2008 Unexposed garden BS Biogas slurry, compost, manure, citrate Mesocosm 34% 25 mg kg−1 Kadian et al. 2008 soil River sediment, water BA Atrazine degrading microbial community Microcosm 95% 1.0 mg L−1 Satsuma 2009 Unexposed sandy soil PR4 Forage grass Pilot 45% 5 mg kg−1 Lin et al 2008 Spill site soil BA Recombinant E. coli Field 77% 4500 mg kg−1 Strong et al 2000 Unexposed field soil BA/BS Pseudomonas ADP Mesocosm 98% 40-400 L ha −1 Lima et al 2009 Unexposed field soil BA Atrazine degrading microbial consortium Microcosm 72% 100 mg kg−1 Newcombe & Crowley 1999 Spill site soil BA/ BS Sucrose, Agrobacterium radiobacter Microcosm 94% 50-200 mg kg−1 Struthers et al 1998 Unexposed field soil BA/BS Citrate, Pseudomonas ADP Microcosm 99% 2 mg kg−1 Chelinho et al. 2010 Agricultural soil BA Pseudomonas sp., C heintzii Mesocosm 65% 1 kg ha −1 Monard et al. 2008

1 Bioaugmentation,2 Biostimulation,3 Monitored natural attenuation,4 Phytoremediation 27

In general, the level of atrazine removal observed in different experiments is dependent on the method used, experiment scale, and atrazine concentration (Table 4). Majority of the studies rely on bioaugmentation, either as a sole treatment method or in combination with biostimulation. The most common organism used in bioaugmentation is Pseudomonas citronellolis, perhaps since its genetic pathways are well known, and a lot of background knowledge regarding this organism is available. However, as atrazine is degraded by bacteria belonging to diverse phyla, many species have been applied in atrazine bioremediation. Because bacteria have various survival strategies, choosing the degrader according to the target environment may reduce the negative effects of competition. A consortium of different species may be used to ensure the presence of different degradation pathways, and to reduce the loss of bioaugmented strain (Satsuma 2009, Newcombe and Cowley 1999). Bioaugmentation is usually very effective in small experiment scale.

2. OBJECTIVES OF THIS THESIS

This work was conducted in collaboration between two research institutes, the Finnish Environment Institute in Finland, and the National Environmental Engineering Research Institute in India. This partnership made possible sampling in both countries, which allowed the comparisons of atrazine bioremediation in different soil types, climates and atrazine use histories. This work aims to elucidate how atrazine degradation genes are reflected in atrazine removal from two soil types using different bioremediation treatments. The genetic tools are used as an indication of bacterial atrazine degradation. The soils used in this study represent two scenarios; one in which atrazine is routinely applied in agricultural fields, and second where residual atrazine contaminates the subsoil via groundwater exposure. Because bioremediation requires the activity of living organisms, the treatment options must be optimized for different environments. The two different soils are likely to have dissimilar bacterial compositions, and the atrazine degrading bacteria probably have different requirements for enhanced atrazine degradation.

Specifically, I wanted to concentrate on the following issues:

 The link between degradation genes and atrazine disappearance from soil (Articles I, II and III) Hypothesis: low concentration of atrazine maintains degradation genes, even if atrazine does not degrade  The natural degradation potential in different soils. (Article I) Hypothesis: longer atrazine use history is reflected in faster degradation  What bioremediation treatments are efficient in different soils? (Article II and III) Hypothesis: bioaugmentation is most efficient in all soils, but biostimulation may enhance degradation in carbon poor soil  What is the microbial community and species dynamics in atrazine polluted soil? (Articles III and IV) Hypothesis: Atrazine application changes the shape of the microbial community, favouring certain bacterial groups after an adaptation period.  Can the experimental setups be scaled up? (Article IV) Hypothesis: atrazine removal rates are slower and not as efficient in larger scale.

3. MATERIAL AND METHODS

The material and methods used in this thesis are briefly overviewed here, and described in detail in articles as indicated in Tables 6 and 7.

3.1 Soils

The soils were collected in four atrazine-amended agricultural fields in Aurangabad region, Maharashtra, India, and in two groundwater areas, located in coniferous forests along the Salpausselkä I ice-marginal formation at sites where atrazine and its degradation products contaminate the groundwater. An example of the typical characteristics of the soils from India and Finland are shown in Table 5

Table 5: typical characteristics for Indian and Finnish soils used in this study. Examples are boreal subsoil from unsaturated layer from Lohja and agricultural soil treated for 3 years with atrazine (Article I). Indian agricultural soil Finnish subsoil Atrazine exposure history 3 years Exposure by groundwater pH value 7.3 ± 0.1 5.2 ± 0.2 Dry matter (%) 63 94 Organic matter (%) 0.05 ± 0.0 4.9 ± 0.3 -1 NO3 mg kg (dw) 120 >10 -1 NH4 mg kg (dw) 80 4.4 Atrazine mg kg-1 (dw) 0.007 >0.005

3.2 Analytical methods

The primary tools for following bioremediation in microcosms were I) quantifying the atrazine degradation genes directly from soil, and II) monitoring atrazine disappearance either by HPLC or mineralization by radiorespirometry. The combination of these tools permitted estimating how well degradation gene quantity reflects atrazine disappearance. The identities of bacterial communities in atrazine-contaminated environments were studied by cloning and sequencing the 16s rRNA genes from soil metagenome. Clone libraries were constructed from agricultural field soil (Article II), atrazine degrading cultivable bacteria (Article II) and bacteria colonizing the trapping material of a passive sampler set in two groundwater monitoring wells (Article III). The effect of different treatments to bacterial community was illustrated by creating “DNA fingerprints” from the soil metagenome by LH-PCR. This method was used to elucidate how a bacterial community shifts during different bioremediation treatments, and if any specific groups enrich depending on the treatment type. As the PCR primers we used only target bacteria carrying out the hydrolytic dechlorination of atrazine, a non-DNA based approach to visualize atrazine degrading organisms from the community was used. Microautoradiography (MAR) is based on the accumulation of radioactively labeled compound (atrazine) in (atrazine) degrading microbes (bacterial, archaeal and fungal). Using this method, the direct visualization of the proportion of atrazine degrading bacteria from total population was possible by microscopy.

Table 6. The analytical methods used in thesis. Application Method Soil origin Article reference Cultivation of atrazine Cultivation in liquid or I-IV degrading strains from culture solid medium collections for inoculation

Analytical methods Quantification of degradation qPCR (atzAB, trzN) India, I-IV genes Finland Atrazine mineralization Radiorespirometry Finland I, III, IV Atrazine disappearance HPLC1 India, II, IV Finland Microbial community analysis LH-PCR India IV Determination of active MAR Finland III degraders Identification of bacterial Cloning and sequencing India, II, III species Finland Identification of bacteria in Cloning and sequencing Finland III groundwater bacteria in passive samplers 1 Carried out by Sneha Sagarkar or Ramboll analytics Oy.

3.3 Bioremediation treatments

We used natural attenuation, bioaugmentation and biostimulation methods either as a sole treatment method or as a combination. The contaminated soils used in this study can be divided into two categories: I) boreal, non-agricultural subsoil and II) tropical agricultural soil. Bioremediation experiments were carried out in microcosm, mesocosm and pilot scales in order to see if atrazine degradation is slower, or perhaps not complete when the phenomena often observed in upscaling are affecting the process. Biostimulation was carried out using citrate or molasses (boreal soil) and citrate (agricultural soil) as additional carbon sources. Atrazine degrading bacteria Pseudomonas citronellolis and Arthrobacter aurescens were acquired from culture collections, and an atrazine degrading consortium of two Pseudomonas and one Arthrobacter strains was isolated during the study. These bacteria were used in bioaugmentation. The treatments carried out in different soils are listed in Table 7. Sampling for qPCR from microcosms was carried out destructively by periodically removing 5 g sample from 100 g parallel treatment bottles, and mixing well between each sampling. DNA samples from mesocosms were obtained directly from the 1 kg mesocosm by removing 10 g of soil, and mixing the mesocosm bulk soil after each sampling. In the pilot scale experiment, 100 g subsamples were collected from three different points, at different depths, in the mesocosms using a soil core to form a bulk sample, from which a 10 g subsample was taken for chemical and molecular analyses.

Table 7. Soils, organisms and substrates used in different bioremediation treatments. Treatment Soil Experiment Organisms used Substra Article origin scale tes reference used Natural Finland Microcosm I attenuation India Microcosm I Mesocosm II Pilot scale IV Bioaugmentation Finland Microcosm Pseudomonas citronellolis / III Arthrobacter aurescens India Mesocosm Pseudomonas citronellolis + II Arthrobacter aurescens Pilot scale Consortium IV Biostimulation Finland Microcosm Citrate, III Molasses III India Mesocosm Citrate II Pilot scale IV Bioaugmentation Finland Microcosm Pseudomonas citronellolis / Citrate / III + biostimulation Arthrobacter aurescens Molasses India N/A N/A N/A

4. RESULTS

4.1 Atrazine degradation genes in different soils and bioremediation treatments

In the boreal soil, atzA and trzN genes could be detected in low copy numbers (102 and 103 copies g-1 soil (dw), respectively) in the soil layers connected to groundwater, but atzB was absent in all soil layers. The Tropical agricultural soil contained 104 to 105 copies of degradation genes (Article I). When bioaugmentation was used as treatment method, the gene copy numbers initially increased, but then remained rather constant at 104 copies g-1 soil in the boreal soil (Article III). In contrast, the gene copy numbers fluctuated in the tropical soil, as an increase on day 6, from 105 to 108 copies g-1 soil, was followed by a gradual decrease (Article IV).

4.2 Bioremediation treatments in different soils

Bioremediation treatments removed atrazine from different soils to various degrees (Table 8). Atrazine did not naturally mineralize in the boreal soil, but the tropical agricultural soil readily mineralized up to 55% of the labelled atrazine in 7 days in microcosms (Article I). When the treatment scale was increased, atrazine removal natural attenuation was unpredictable: in mesocosms, only 16% of atrazine was removed, but in the pilot scale study, 90% of atrazine had disappeared by the end of the experiment. The length of the treatment probably influenced these results, as the mesocosm was only incubated for 6 days, whereas pilot scale experiment lasted for two months. The length of atrazine use history had little effect on mineralization rates, although the most efficient mineralization was observed in soils recently treated with atrazine. The addition of atrazine degrading bacteria increased degradation in both of the soils studied here: In the boreal soil, 65% of labelled atrazine was removed, and in the tropical soils the natural

degradation was enhanced up to 92% (Table 8). Interestingly, carbon amendment had a notable effect on atrazine mineralization in 10 °C in the boreal soil. The most efficient treatment method for the Boreal soils was the combination of bioaugmentation and biostimulation with additional carbon. The combination treatment was not tested for tropical soils, but in the pilot scale experiment, all treatments removed at least 90% of atrazine. The chosen method had greatest influence on treatment times. Only 6 days of incubation was necessary when bioaugmentation was used, but biostimulation required one month to reach the same level of atrazine removal. Natural attenuation was the slowest treatment method, requiring two months of treatment.

Table 8. Bioremediation treatments, the highest atrazine removal achieved. Boreal soil* Tropical soil**

Natural attenuation 10 °C: 100 days, <1% (Article I) Mesocosm: 6 days, 16% (Article II) 30 °C: 100 days, 1% (Article I) Pilot scale: 60 days, 90% (Article IV)

Bioaugmentation 30 °C: 15 days, 65% (Article III) Mesocosm: 6 days, 92% (Article II) with atzAB or trzN+atzB Pilot scale: 6 days, 90% (Article IV)

Biostimulation 10 °C: 30 days, 52% (Article III) Mesocosm: 6 days, 52% (Article II) with citrate or molasses 30 °C: 30 days, 1% (Article III) Pilot scale: 30 days, 90% (Article IV)

Biostimulation & 30 °C: 76% (Article III) N/A Bioaugmentation

4.3 Microbial community dynamics in atrazine treated soil

Clone libraries were constructed from clones obtained from the passive samplers in groundwater wells in the boreal soil (Figure 4). Even though the groundwater wells were close to each other, the compositions of the libraries were not identical. In groundwater well 608, the most commonly observed clone was identified as a relative of Acinetobacteria (24% of the clones), whereas in well PM, the most abundant clone belonged to the Rhodobacteria (31% of the clones). Many clones had close relatives in bacterial taxonomic groups known to contain atrazine degraders. Although there is no

37 direct indication that the clones obtained in this study are actual atrazine degraders, degradation genes atzA and atzB amplified from the same sample verifies that atrazine degraders did colonize the sampler.

Clone libraries were also constructed from the tropical agricultural soil (Article II). The bacterial diversity was high, (Shannon’s diversity index H’=3.54, H’max=3.81, D1=0.27). Five phyla, Proteobacteria, Actinobacteria, Verrucomicrobia, Firmicutes and Chloroflexi were identified in the clone libraries.

Figure 4. Closest identities of the clones obtained from the passive samplers in groundwater

Microautoradiography (MAR) was used to estimate the fraction of active atrazine degraders in different bioremediation treatments. The mean proportions of the atrazine degrading bacteria were 0.3% of the total cell count in native soil, 1.9% in bioaugmentation treatment, and 1.8% in biostimulation treatment (Article III). These results seem to fall in line with results gained from mineralization experiment (Table 8), but because the MAR experiment was conducted without parallels, no correlation can be made using statistical methods.

The species dynamics of the microbial community in pilot scale experiment of the tropical soil was monitored by LH-PCR during different bioremediation treatments. The resulting molecular fingerprints revealed, that regardless of bioremediation treatment method, the community structure remained rather constant (Figure 5, from Article IV). However, the communities diverged dependent on pretreatment with atrazine.

Figure 5. DNA-fingerprints created by LH-PCR from different bioremediation treatments (from Article IV).

5. DISCUSSION

5.1 Application of genomic tools in atrazine bioremediation

We used genetic, analytical, microscopy and isotope methods to assess bioremediation. As we could not investigate gene expression directly due to low gene copy numbers, the results from qPCR and radiorespirometry were used to close the gap between genetic potential (gene copy numbers), and the effect they had in atrazine concentration (atrazine mineralization or HPLC).

Boreal soils

The DNA extracted from boreal soils contained considerably less inhibitory substances than the tropical agricultural soils, but low template copy numbers complicated the analyses. Each sample was run in triplicate, but in many cases, only one or two of the parallel runs yielded PCR product (Article I). In order to estimate a representative copy number in soil, a value between the theoretical detection limit (100 copies/reaction) and zero can be used, instead of zero (Björklöf et al. 1995). In this way, underestimation of copy numbers that are just below detection limit can be avoided

Because the intermediates DEA and DIA are commonly encountered in the Finnish groundwater (Vuorimaa et al. 2007), it is likely that this pathway is active in boreal soil. As the genes coding for this pathway are not specific for atrazine degradation, no DNA- based methods exist for detecting this pathway. In order to avoid the underestimation of atrazine degradation potential, we used microautoradiography (MAR) for detecting atrazine degrading bacteria (Article IV) in the boreal soil. We successfully visualized the active cells in three different bioremediation treatments. As we hypothesized, bioremediation treatments increased the number of atrazine degrading bacteria. However, due to the lack of parallel treatments, this is only an exploratory experiment on the application of MAR in atrazine degradation, and therefore statistical analyses and

41 correlation with other methods cannot be conducted. Regardless, it is a promising result for carrying out larger scale MAR studies, and coupling MAR with fluorescent dyes targeting different phylogroups (MAR-FISH) would shed light on which pathways are actually active (Okabe et al 2004).

Tropical soils

Even low concentrations of humic substances inhibit the activity of the Taq-polymerase, resulting in reduced yields (Tsai and Olson 1992). This phenomenon is particularly critical in qPCR, where the quantification of templates is based on a concurrent amplification of gene standards. On the other hand, if unspecific products are amplified, the copy numbers may be overestimated. The DNA extracted from tropical agricultural soils contained a great deal of inhibitory substances. Furthermore, the primers used in this study (Devers et al 2004) also hybridized with other genes, e.g. nitrate reductases, as well as some fungal sequences. This result implicates the need for more specific primers, when atrazine degradation genes are amplified from carbon rich soils. However, primer design may be difficult, as the short amplicon length required for qPCR limits the number of suitably located priming sites.

The quality of the quantification was enhanced by dilution of the samples, as well as meticulously assessing the quality of the end products by melting curve analysis and sequencing (Article I). The optimized protocol was applied in the mesocosm experiment, which demonstrated the correlation between degradation gene copy numbers and atrazine disappearance (Article II).

5.2 Natural atrazine biodegradation potential

Boreal soils

Atrazine degradation in boreal subsoil was very low (<1%) in both 10 °C and 30°C, even though low a number of degradation genes were present in the soil (Article 1). Low microbial activity due to low temperatures, low organic carbon or the low pH might 42 explain the lack of degradation activity in boreal soil (Mueller et al 2010). Also, the low concentration of atrazine may be limiting its degradation. In order to gain energy or building blocks from compounds, the amount of energy the microbe needs to spend to synthesize the necessary enzymes must be smaller than what it gains from degrading the compound. It is possible that this limit concentration was not reached in the boreal site, as atrazine concentration in soil was below analytic detection limit (5 µg kg-1).

Tropical soils

Contrary to the findings in boreal soil, up to 55% of atrazine mineralized in tropical agricultural soil, which was reflected in the degradation gene copy numbers (log 4-5 g soil-1) (Article I), and 16% of atrazine disappeared in one week, indicative of efficient natural attenuation (Article II). Previous studies have shown that atrazine use history affects the rate of atrazine disappearance in soil (Shaner and Henry 2007). In the tropical agricultural soil, atrazine mineralization was affected by recent application, rather than the long history of atrazine use (Figure 6A). This is in line with previous work, showing that a single treatment with atrazine enhances atrazine degradation, and the degradation rates are not faster after subsequent treatments (Mueller et al 2010). The same phenomenon was apparent in the pilot scale experiment (Article IV), where part of the soil was treated with a single application of atrazine prior to the beginning of the experiment. The cluster analysis of LH-PCR profiles created from the pilot scale experiment indicated that the factor driving the microbial populations to differentiate was atrazine use history, rather than the different bioremediation treatments (Figure 6B). It is noteworthy, that microbial community is rapidly affected by atrazine, as an adaptation period of four months has been reported to enhance atrazine degradation (Rhine et al. 2003).

Figure 6. (A) Atrazine mineralization in tropical agricultural soils having different atrazine use histories (B) Clustering of different bioremediation treatments for soil having no atrazine use history (cluster I) and soil treated with atrazine previous to bioremediation (cluster II) (Articles I and IV).

5.3 Microbial community and species dynamics

Boreal soils

Due to low DNA yields (0.5 ng µl-1) and microbial numbers in the boreal soil (log 4.5 16s rRNA copies g-1 soil), the microbial community could not be investigated by a clone library constructed directly from soil, and instead passive samplers set in groundwater monitoring wells were used to obtain a DNA sample (3.5 ng µl-1) of the community present in groundwater zone. In these libraries, many clones were closely related to cultured bacteria, many identified as Proteobacteria, and a few Actinobacteria (Article III). In order to colonize the passive sampler, the bacteria must be able to grow on the activated carbon beads inside the sampler. Therefore microbial communities collected with Bio-Trap® samplers may be more indicative of the active part of the microbial community (Sublette et al. 2006). In our case, the sampler seemed to select for bacterial clones having cultivable representatives, as indicated by similarities to cultivable strains in GenBank.

The current molecular tools can only target the genes on the pathway starting by hydrolytic dechlorination, because the enzymes carrying out atrazine dealkylation are not specific for atrazine degradation. To address this problem, we used microautoradiography (MAR), which targets metabolic, rather than genetic functions, to calculate the number of atrazine degrading bacteria in the community in different bioremediation treatments (Article III). Additionally, this method also permits the estimation of the amount of active degraders in population, while DNA-based methods only allow the estimation of the degradation potential.

The number of active atrazine degraders was small in all treatments, but bioremediation treatments increased the numbers of active degraders in total population as expected; natural attenuation (0.3%) < biostimulation (1.8%) < bioaugmentation (1.9%). MAR may perhaps be used to assess the prevalence of different atrazine degradation pathways. If the gene copy numbers in hydrolytic dechlorination pathway exceed the number of MAR positive cells, it is likely that this is the dominant pathway, albeit the copy number of the plasmid harboring the hydrolytic dechlorination genes is not known. If, however, the MAR positive cells are present in greater number than hydrolytic dechlorination genes, it is likely that some other pathway is responsible for atrazine being metabolized. This was observed in the untreated soil, indicating that in native soil, both pathways are present, but biostimulation treatments direct the degradation towards the hydrolytic dechlorination pathway.

Tropical soils

In the mesocosm experiment, many representatives of the known atrazine degrading bacterial species were identified (Article II). When atrazine was used for selection of atrazine degrading cultivable bacteria, γ-proteobacteria and Actinobacteria were enriched (Article II). This is in line with previous findings (Ghosh et al 2009). The key problem in bioaugmenting atrazine contaminated soil is that Pseudomonas strains are not always able to compete with indigenous bacteria (Morán et al 2006). In our experiments, the Pseudomonas genotype atzA/atzB was always present when this strain was used in

45 bioremediation, indicating that either the strain survived competition, or the Pseudomonas genotype was present in the indigenous community.

In the pilot scale study, we used LH-PCR as a method for tracking changes in the microbial community of different bioremediation treatments. We included atrazine use history as a parameter in this experiment, because previous exposure to atrazine rapidly affects the atrazine degrading microbial community, as exemplified in Article I and several other studies (Yassir et al. 1999, Popov et al. 2005). The LH-PCR fingerprints revealed that the population composition during the course of the treatment is quite stable, as differences between peak profiles over the two month treatment time were minimal. As atrazine degraders may appear in the atrazine treated soil in a matter of days (Rhine et al. 2003), it is possible, that the shift in the microbial community had already taken place at the time of our first sampling at 1 month after the beginning of the experiment. However, clear differences were seen between the treatments that were previously exposed to atrazine, as exemplified in figure 4 (From Article IV).

5.4 Atrazine bioremediation strategies

The rate and extent of atrazine degradation is dependent on the activity of the atrazine degrading bacteria, and therefore engineering the environmental conditions favorable for these organisms is the key issue in bioremediation strategy design. Atrazine degradation genes are found in several bacterial phyla, probably because they are often associated with plasmids and mobile genetic elements, which permits lateral gene transfer between different species. We identified several atrazine degrading species in different soils using molecular cloning from boreal groundwater zone and tropical agricultural soil, and several atrazine degrading bacteria were isolated from the agricultural soil. The isolated strains were screened for degradation genes, and the most efficient degraders were used in bioaugmentation.

Bioaugmentation with atrazine degrading bacteria resulted in efficient atrazine removal in both soils, with the maximum of 90% atrazine degraded in soil either by Pseudomonas 46 citronellolis, or by a consortium of atrazine degrading bacteria (Table 7). Similar results are common in microcosm scale, but also pilot experiments with more than 90% atrazine removal have been carried out (Lima et al. 2009). However, the addition of bacteria in subsoil is not a simple task, due to competition and adverse environmental conditions (Morán et al 2006). In situ methods also suffer from other challenges. These issues are thoroughly reviewed by Rittmann (1995) and Alexander (1994). The main concerns are that the contaminant concentration might be too high or too low to be degraded by microbes, or it might be unavailable to the organism. In brief, because protein synthesis requires a great deal of energy, the bacteria will not initiate this unless certain limit concentration of the energy-yielding compound is available, ensuring that there is a payoff for the enzyme synthesis. Also, the distribution of bacteria in subsoil is a demanding task. However, it may be unnecessary, if the absence of catabolic genes on site is not the factor limiting degradation (Thompson et al 2005).

Boreal soils

Atrazine did not attenuate naturally in the boreal soil, but bioaugmentation efficiently removed it, especially in the elevated temperature. There was no clear difference between degradation efficiencies of the Pseudomonas and Arthrobacter strains used. In 10 °C, biostimulation alone was efficient, possibly because carbon, instead of nitrogen, was the growth limiting factor in situ. When carbon was added, nitrogen becomes the limiting factor, which gives the bacteria able to degrade the nitrogen rich atrazine molecule a competitive edge. Biostimulation with molasses resulted in over 50% reduction of atrazine, which would be sufficient for decreasing the atrazine concentration in the raw groundwater below the guideline value concentration (Article III). Citrate addition resulted in 23% atrazine removal, but unlike molasses, there was no lag period (Article III). Interestingly, biostimulation was not a successful treatment method for boreal soil in the elevated temperature. The results obtained from this study only allow speculation, but perhaps the indigenous bacteria were psycrophilic, and their growth was impaired at 30 °C. It is also possible that the temperature affected oxygen consumption or nitrogen fixation, which leads to unfavourable conditions for atrazine degradation. Although in situ biostimulation of subsoil might be simpler to carry out than bioaugmentation, by 47 feeding the nutrients via injection wells into the saturated zone, it too has its drawbacks, such as the excessive need of oxygen in aerobic degradation or fast-growing biomass clogging the aquifer (Alexander 1994). However, this is also the problem in bioaugmentation, if the addition of degraders is successful.

Tropical soils

Atrazine was naturally degraded to some extent in the tropical soil, probably due to optimal growing conditions of the atrazine degrading organisms, and because of the higher atrazine concentration present in the soil after continuous atrazine treatment. Previous studies have indicated, that previous atrazine exposure enhances atrazine degradation (Sinkkonen et al. 2013), and that time since last exposure increases mineralization lag time (Cheyns et al. 2012). Time since previous exposure to atrazine was the most influential factor in the natural attenuation of tropical, agricultural soils (Article I). Previous atrazine exposure was therefore selected as a parameter in the pilot scale study. In this experiment, atrazine use history influenced atrazine removal, as removal of 90% atrazine was 60% faster in bioaugmented soils previously treated with atrazine. Previous atrazine exposure also affected the structure of the microbial community, as exemplified in figure 4.

5.5 Scaling up

Application of laboratory scale bioremediation experiments on pilot or field scale are dependent on several factors. The greatest challenge is spatial segregation, and limitations in mass transportation (Sturman et al 1995). However, large atrazine concentrations can be degraded in pilot scale by biostimulation with citrate and bioaugmentation by Pseudomonas strain ADP (Lima et al. 2009).

The initial assessment of atrazine degradation potential in boreal and tropical soils was carried out in small scale, using 10 g of soil in the mineralization experiments (Article I). The same scale was used in the bioaugmentation and biostimulation tests for boreal soils (Article III). The soil mass used in the mesocosm study of tropical agricultural soil was 48

10 kg (Article II), and 100 kg soil in plastic tanks (600 mm high, Ø 700 mm) pilot scale experiment (Article IV). Due to limited mass transport, contaminant degradation in larger scales may be restricted (Sturman et al 1995). In an in situ uranium bioremediation by a geobacter species, the most influential limiting its growth was identified as grazing by protozoa (Holmes et al. 2013). In hydrocarbon bioremediation, treatment time increased threefold when experiments where moved from the laboratory into field (Diplock et al. 2009). In our experiments, larger scale did not limit atrazine removal (Table 9), as the extent of atrazine removal was approximately 90% regardless of soil mass in bioaugmented and biostimulated treatments. However, treatment times were twice as long, except when bioaugmentation was used (Table 9). Degradation in biostimulated pilot scale experiment was slower by half than in mesocosm. Natural attenuation is the most time-consuming treatment method, as treatment of two months was required for removing 94% atrazine in pilot scale. Atrazine removal in naturally attenuated mesocosms was only the half of pilot scale, probably due to the shorter experiment period. Despite the drawback of longer treatment periods, natural attenuation is the least work intensive bioremediation method, and often due to large heterogeneity of bacterial species, sometimes bioremediation is successful without the amendment of bacteria or supplements (Megharaj et al 2011).

Table 9. Atrazine removal in different bioremediation mesocosm and pilot scale treatments. Initial atrazine concentration in both experiments was 300 mg kg-1 atrazine. Statistical analyses and standard deviations in Article II (mesocosm) and Article IV (pilot scale). Treatment 10 kg 100 kg

Natural attenuation 48% (14 days) 94% (60 days) Biostimulation 97% (14 days) 90% (30 days) Bioaugmentation 92% (7 days) 90% (6 days)

5.6 Summary

 The application of genomic tools in monitoring atrazine degradation is fast, robust and often correlated to degradation. However, special care must be taken when the method is optimized for different soils (Articles I and II).  There is a need for more specific primers for atrazine degradation genes, if the method is applied to soils which contain a large amount of PCR inhibiting substances and a large bacterial diversity.  Atrazine mineralizes naturally, if frequent exposure to atrazine is maintained. Residual atrazine may remain in subsoil and is not degraded after the use is discontinued (Article I).  Bioaugmentation effectively removes atrazine even in pilot scale experiments. However, also biostimulation with additional carbon is an efficient treatment method in carbon limited environments (Articles II, III and IV).  Atrazine exposure affects the soil microbial community structure more than different treatment methods. (Articles III and IV)  Bioremediation treatments for tropical agricultural soil were successfully scaled up without a decrease in the efficiency of atrazine degradation (Article IV).

6. CONCLUSIONS AND FUTURE PROSPECTS

The use of genetic methods as monitoring tools is beneficial in bioremediation because of their sensitivity and specificity. As the knowledge on how these tools behave in different types of samples increases, focus can be diverted towards the rapid application of the tools rather than meticulous validation of methodology. In this work, the tropical agricultural soil was bioremediated in three different scale experiments, but the same remains to be achieved for the boreal subsoil. This task is challenging due to the low concentration and the location of the contaminant in subsoil, which pose limits to both the analytical methods and the treatment method. An in situ treatment of groundwater zone would require a pump-and-treat –method, or perhaps a permeable barrier constructed in the contaminated groundwater.

The function of the two degradation pathways is yet to be elucidated. The role of the dealkylation pathway could be investigated by combining the MAR method used in this work, with FISH applications targeting different phylogenetic groups. The knowledge on which phylogenetic groups are actively degrading atrazine in different environments is helpful when the decision of treatment method is needed. In general, the stimulation of the autochthonous degrader population is simpler than introducing degraders from another system. However, as atrazine degradation genes are associated with mobile elements, the strains used in bioaugmentation might merely act as a mediator in the dissemination of the degradation genes in the indigenous population. In order to monitor the occurrence of the augmented genetic material, a method for detecting the augmented plasmid using a marker gene could be developed.

The combination of robust monitoring toolbox, and the knowledge on soil microbiological function, is the beacon leading the way towards efficient, cost-effective large scale bioremediation.

7. ACKNOWLEDGEMENTS

During the years I worked on my Ph.D., I often thought of all the people I am grateful to, and things I am thankful for. Therefore it is odd that at this point, when I finally am allowed to express all my gratitude, the defined page number limits me for listing everything and everyone who have aided me on this journey. I should write another book for that.

This study was carried out at the excellent facilities of the Finnish Environment Institute, and made possible by the financial support of the Academy of Finland, the Finnish Doctoral Programme in Environmental Science and Technology (EnSTe), Maa- ja vesitekniikan tuki ry., The Finnish Concordia Fund, and the University of Helsinki. The EnSTe graduate school, as well as the Doctoral Programme in Microbiology and Biotechnology (MBDP), are acknowledged for arranging interesting courses, travel grants, and peer support. Excellent past and present staff members, especially Kaisa, Ilse and Jukka, are acknowledged for their help.

I am first and foremost indebted to my supervisor Kirsten Jørgensen for providing me the intriguing research topic, and having faith at my skills and persistence to execute the scientific work required to make this thesis a reality. I was lucky to have such an enlightened supervisor. Your door was always open for me, and yet you allowed me to work independently in order to grow to the scientist I am today. My thesis advisory board has shared feedback, guidance and ideas from the beginning to the end: Lotta Björklöf, your kindness and guidance have helped me on this rocky path more than you think. Kristina Lindström, it was you who got me excited about bioremediation way back then, and your example and support has been one of the main reasons I became a scientist. Merja Kontro, thank you for sharing your wisdom (and atrazine degrading bacteria) with me.

I sincerely thank the reviewers of my work, Prof. Martin Romantschuk and Dr. Françoise Binet, for the valuable comments for improving the quality of my thesis.

I was lucky to co-author all my papers with such wonderful scientists and people. I have gained not only wonderful coworkers, but great friends I know I get to keep forever. Atya, 52

Sneha and Shinjini, thank you for bringing more than a little Indian warmth in my life. I gained a few Indian habits and a passion for Aloo Parathas on that memorable sampling trip to Aurangabad.

I thank the NENUN network for the networking opportunities, wonderful courses and workshops, and funding my research exchange to Aalborg University to execute some of the most interesting work I carried out during my study. Jeppe, I am indebted to you for taking such good care of me in Denmark, and your expert guidance when I was struggling with my results.

A large number of coworkers and sisters in arms have kept me on track during these years. The therapy group at the lab, Noora, Päivi, Pirjo, Anna R, Anna S, Pia, I don’t think I would have stayed sane if it weren’t for you. Sharing lunches, offices and labs for so many years has knitted us close. Bioremmi: Anu, Elina, Kaisa and Lara, you have been an inspiration and brought great joy in my life during the long winters, motherhood, and glittering defense parties. Stiina, Leena, Salla, Shinjini, Windi, and other friends at the University, sharing all this with peers like you has made this road light to travel. Thank you for the good times shared in Finland and around the world.

Petri, Matti, Petsku, Mikko, Tapsa and all the Clan Pakila affiliates: thank you for keeping my feet off the ground. I needed that.

I am deeply grateful to my small but valuable family, grandma Kirsti, aunt Merja, cousin Anna and her family, but first and foremost to my mum Kevät and dad Ilkka. No matter the choices I make, you have always shown me nothing but support and encouragement, and I have been blessed to have such a strong ground to stand on.

Last, but not least, my dearest thanks go to my family. Juha, thank you for putting up with my scientific follies, extending deadlines, and never-ending attempts to save the planet. The busy years of building careers and family do not frighten me as you stand beside me. Otso, and our yet unborn child, thank you for making this planet worth saving. Rakastan teitä.

8. REFERENCES

Alexander M 1995. Biodegradation and Bioremediation. Academic press, San Diego, California.

Aislabie J, Hunter D, Ryburn J, Fraser R, Northcott GL, Di HJ 2004. Atrazine mineralisation rates in New Zealand soils are affected by time since atrazine exposure. Soil res 42:783–792

Allard A-S, Neilson AH 1997. Bioremediation of Organic Waste Sites: A Critical Review of Microbiological Aspects. Int Biodeter Biodegr 39: 253–285

Amann RI, Ludwig W, Schleifer KH 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59:143–169

Andrews JH, Harris RF 1986. r-and K-selection and microbial ecology. Adv Microb Ecol 9: 99–147

Arias-Estévez M, López-Periago E, Martínez-Carballo E, Simal-Gándara J, Mejuto JC, García-Río L 2008. The mobility and degradation of pesticides in soils and the pollution of groundwater resources. Agric Ecosys Environ 123 247–260.

Avaniss-Aghajani E, Jones K, Chapman D, Brunk C 1994. A molecular technique for identification of bacteria using small subunit ribosomal RNA sequences. Biotechniques 17: 144–146, 148–149

Backman B, Luoma S, Schmidt-Thomé P and Laitinen J 2007. Potential risks for shallow groundwater aquifers in coastal areas of the Baltic Sea, a case study in the Hanko area in South Finland. CIVPRO Working Paper 2007:2

Bastos A, Magan N 2009. Trametes versicolor: Potential for atrazine bioremediation in calcareous clay soil, under low water availability conditions. Int Biodeter Biodegr 63: 389–394

Bedmar F, Daniel PE, Costa JL, Giménez D 2011. Sorption of acetochlor, S-metolachlor, and atrazine in surface and subsurface soil horizons of Argentina. Environ Toxicol Chem 30:1990–1996

Bentley DR, Balasubramanian S, Swerdlow HP, Smith GP, Milton J, Brown CG, et al. 2008. Accurate whole human genome sequencing using reversible terminator chemistry. Nature 456: 53–59

Binet F, Kersanté A, Munier-Lamy C, Le Bayon R-C, Belgy M-J, Shipitalo MJ 2006. Lumbricid macrofauna alter atrazine mineralization and sorption in a silt loam soil. Soil Biol Biochem 38: 1255–1263

Björklöf K, Suoniemi A, Haahtela K, RomantschukM 1995. High frequency of conjugation versus plasmid segregation of RPI in epiphytic Pseudornonas syringae populations. Microbiology 141: 2719–2727

Boethling RS and Alexander M 1979. Effect of concentration of organic chemical on their biodegradation by natural microbial communities. Appl Environ Microbiol 37:1211–1216

Bragg JR, Prince RC, Harner EJ, Atlas RM 1994. Effectiveness of bioremediation for the Exxon Valdez oil spill. Nature 368: 413–418

Boopathy R 2000. Factors limiting bioremediation technologies. Biores Tech 74: 63–67

Cai B, Han Y, Liu B, Ren Y, Jiang S 2003. Isolation and characterization of an atrazine- degrading bacterium from industrial wastewater in China. Lett Appl Microbiol 36:272– 276 55

Caliman FA, Robu BM, Smaranda C, Pavel VL, Gavrilescu M 2011. Soil and groundwater cleanup: benefits and limits of emerging technologies. Clean Technol Environ Policy 13: 241–268

Chelinho S, Moreira-Santos M, Lima D, Silva C, Viana P, André S, Lopes I, Ribeiro R, Fialho AM. Viegas CA, Sousa JP 2010. Cleanup of atrazine-contaminated soils: ecotoxicological study on the efficacy of a bioremediation tool with Pseudomonas sp. ADP. J Soils Sediments 10:568–578

Cheyns K, Martin-Laurent F, Bru D, Aamand J, Vanhaecke L, Diels J, Merckx R, Smolders E, Springael D 2012. Long-term dynamics of the atrazine mineralization potential in surface and subsurface soil in an agricultural field as a response to atrazine applications. Chemosphere 86:1029–1034

Cooper RL, Stoker TE, Tyrey L, Goldman JM, McElroy KW 1999. Atrazine disrupts the hypothalamic control of pituitary-ovarian function. Toxicol sci 53: 297–307

Das PC, McElroy WK, Cooper RL 2000. Differential modulation of catecholamines by chlorotriazine herbicides in Pheochromocytoma (PC12) cells in vitro. Toxicol Sci 56: 324–331.

Devers M, Soulas G, Martin-Laurent F 2004. Real-time reverse transcription PCR analysis of expression of atrazine catabolism genes in two bacterial strains isolated from soil. J Microbiol Methods 56: 3–15

Devers M, El Azhari N, Udiković-Kolić N, Martin-Laurent F 2007. Detection and organization of atrazine-degrading genetic potential of seventeen bacterial isolates belonging to divergent taxa indicate a recent common origin of their catabolic functions. FEMS Microbiol Lett 273: 78–86

56 de Souza ML, MJ Sadowsky, LP Wackett 1996. Atrazine Chlorohydrolase from Pseudomonas sp. Strain ADP: Gene Sequence, Enzyme Purification, and Protein Characterization. J Bacteriol 178: 4894–4900 de Souza ML, J Seffernick, B Martinez, MJ Sadowsky LP Wackett 1998. The Atrazine Catabolism Genes atzABC Are Widespread and Highly Conserved. J Bacteriol 180(7): 1951–1954

Diplock EE, Mardlin DP, Killham KS, Paton GI 2009. Predicting bioremediation of hydrocarbons: Laboratory to field scale. Environ Poll 157: 1831–1840

Douglass JF, Radosevich M, Tuovinen O 2014. Mineralization of atrazine in the river water intake and sediments of a constructed flow-through wetland. Ecol Engin, in press. doi:10.1016/j.ecoleng.2014.08.016

European Commission 2012. Communication from the Commission to the Council. The combination effects of chemicals Chemical mixtures /* COM/2012/0252 final */

Gans J, Wolinsky M, Dunbar J 2005. Computational improvements reveal great bacterial diversity and high metal toxicity in soil. Science 309: 1387–1390

Gerecke A.C, Scharer M, Singer H.P, Muller SR, Schwarzenbach RP, Sagesser M, Ochsenbein U, Popow G 2002. Sources of pesticides in surface waters in Switzerland: pesticide load through waste water treatment plants–current situation and reduction potential. Chemosphere 48: 307–315

Ghosh D, Roy K, Srinivasan V, Mueller T, Tuovinen OH, Sublette K, Peacock A, Radosevich M 2009. In-situ enrichment and analysis of atrazine-degrading microbial communities using atrazine-containing porous beads. Soil Biol Biochem 41:1331–1334

Gilliom RJ, Barbash JE. Crawford CG, Hamilton PA, Martin JD, Nakagaki N, Nowell LH, Scott JC, Stackelberg PE, Thelin GP, Wolock DM 2007. Pesticides in the Nation’s Streams and Ground Water, 1992–2001. United States Geological Survey

Eldridge JC, Wetzel LT, Stevens JT, Simpkins JW 1999. The mammary tumor response in triazine-treated female rats: A threshold-mediated interaction with strain and species- specific reproductive senescence. Steroids 64: 672–678

Eguchi M, Kitagawa M, Suzuki Y, Nakamuara M, Kawai T, Okamura K, Sasaki S, Miyake Y 2001. A field evaluation of in situ biodegradation of trichloroethylene through methane injection. Water res 35:2145–2152

Epstein SS 2013. The phenomenon of microbial uncultivability. Curr Opin Microbiol 16: 636–642

Ettema CH and Wardle DA 2002. Spatial soil ecology. Trends Ecol Evol 17: 177–183

European Commission 2004. Decision concerning the non-inclusion of atrazine in annex I to the council directive 91/414/EEC and the withdrawal of authorisations for plant protection productscontaining this active substance. Off J Eur Union L78:53–55, Available: http://ec.europa.eu/food/plant/protection/evaluation/existactive/oj_atrazine.pdf [accessed 09 September 2014].

Giller PS 1996. The diversity of soil communities, the ‘poor man's tropical rainforest’. Biodivers Conserv 5: 135–168

Heid CA, Stevens J, Livak KJ, Williams PM 1996. Real Time Quantitative PCR. Genome Res 6:986–994

Holland PM, Abramson RD, Watson R, Gelfand DH 1991. Detection of specific polymerase chain reaction product by utilizing the 5′-3′ exonuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci USA 88:7276–7280 58

Holmes DE, Giloteaux L, Williams KH, WrightonKC, Wilkins MJ, Thompson CA, Roper TJ, Long PE, Lovley DR 2013. Enrichment of specific protozoan populations during in situ bioremediation of uranium-contaminated groundwater. ISME J 7: 1286– 1298

Hugenholtz P, Goebel BM, Pace NR 1998. Impact of Culture-Independent Studies on the Emerging Phylogenetic View of Bacterial Diversity. J Bacteriol 180:4765–4774

Höhener P, Ponsin V 2014. In situ vadose zone bioremediation. Curr Opin Biotechnol 27:1–7

Kadian N, Gupta A, Satya S, Mehta RK, Malik A 2008. Biodegradation of herbicide (atrazine) in contaminated soil using various bioprocessed materials. Bioresource Technol 99: 4642–4647

Kaufman DD, Blake J 1970. Degradation of atrazine by soil fungi. Soil Biol Biochem 2:73–80

Kasozi GN, Nkedi-Kizza P, Li Y, Zimmerman AR 2012. Sorption of atrazine and ametryn by carbonatic and non-carbonatic soils of varied origin Environ Pollut 169:12– 19

Kirk JL, Beaudette LA, Hart M, Moutoglis P, Klironomos JN, Lee H, Trevors JT 2004. Methods of studying soil microbial diversity. J Microbiol Methods 58:169–188

Kortenkamp A 2007. Ten years of mixing cocktails: A review of combination effects of endocrine-disrupting chemicals. Environ Heal Persp 115: 98–105

Larson DL, McDonald S, Fivizzani AJ, Newton WE, Hamilton SJ 1998. Effects of the herbicide atrazine on Ambystoma tigrinum metamorphosis: Duration, larval growth, and hormonal response. Physiol Zool 71: 671–679

Lee N, Nielsen PH Andreasen KH, Juretschko S, Nielsen JL, Schleifer K-H, Wagner M 1999. Combination of fluorescent in situ hybridization and microautoradiography—a new tool for structure-function analyses in microbial ecology Appl Environ Microbiol 65: 1289–1297

Levanon D. 1993. Roles of fungi and bacteria in the mineralization of the pesticides atrazine, alachlor, malathion and carbofuran in soil. Soil Biol Biochem 25:1097–1105

Li Q, Li Y, Zhu X, Cai B 2008. Isolation and characterization of atrazine-degrading Arthrobacter sp. AD26 J Environ Sci 20:1226–1230

Lima D, Viana P, André S, Chelinho S, Costa C, Ribeiro R, Sousa JP, Fialho AM, Viegas CA 2009. Evaluating a bioremediation tool for atrazine contaminated soils in open soil microcosms: The effectiveness of bioaugmentation and biostimulation approaches Chemosphere 74: 187–192

Lin CH, Lerch RN, Garrett HE, George FM 2006. Bioremediation of atrazine- contaminated soil by forage grasses: transformation, uptake, and fetoxification J Environ Qual 37: 196–206

Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bemben LA, et al. 2005. Genome sequencing in microfabricated high-density picolitre reactors. Nature 437:376– 380

Mandelbaum RT, Allan DL, Wackett LP 1995. Isolation and characterization of a Pseudomonas sp. that mineralizes the s-triazine herbicide atrazine. Appl Environ Microbiol 61:1451–1457

Martin-Laurent F, Philippot L, Hallet S, Chaussod R, Germon JC, Soulas G, Catroux G 2001. DNA extraction from soils: Old bias for new microbial diversity analysis methods. Appl Environ Microbiol 67:2354–2359

Martinez B, Tomkins J, Wackett LP, Wing R, Sadowsky MJ 2001. Complete nucleotide sequence and organization of the atrazine catabolic plasmid pADP-1 from Pseudomonas sp. strain ADP J Bacteriol. 183:5684–5697

Megharaj M, Ramakrishnan B, Venkateswarlu K, Sethunathan N, Naidu R 2011. Bioremediation approaches for organic pollutants: a critical perspective. Environ Int 37:1362–1375.

Monard C, Martin-Laurent F, Lima O, Devers-Lamrani M, Binet F 2013. Estimating the biodegradation of pesticide in soils by monitoring pesticide-degrading gene expression Biodegradation 24:203–213

Monard C, Martin-Laurent F, Vecchiato C, Francez AJ, Vandenkoornhuyse P, Binet F 2008. Combined effect of bioaugmentation and bioturbation on atrazine degradation in soil 40: 2253–2259

National Research Council 1993. The Current practice of bioremediation. In “In Situ bioremediation, when does it work?”. Committee on In Situ Bioremediation, Water Science and Technology Board, Commission on Engineering and Technical Systems. National Academy Press, Washington, D.C. 1993

Morán AC, Müller A, Manzano M, González B 2006. Simazine treatment history determines a significant herbicide degradation potential in soils that is not improved by bioaugmentation with Pseudomonas sp. ADP. J Appl Microbiol 101: 26–35

Moreno JL, Aliaga A, Navarro S, Hernandés T, García C 2007. Effects of atrazine on microbial activity in semiarid soil. Appl Soil Ecol 35:120–127

Mudhoo A and Garg VK 2010. Sorption, transport and transformation of atrazine in soils, minerals and composts: a review. Pedosphere 21:11–25

Mueller TC, Steckel LE, Radosevich M 2010: Effect of soil pH and previous atrazine use history on atrazine degradation in a tennessee field soil. Weed sci 58:478–483

Mulbry WW,Hong Z, Nour SM, Topp E 2002. The triazine hydrolase gene trzN from Nocardioides sp. strain C190: Cloning and construction of gene-specific primers. FEMS Microbiol Lett 206:75–79

Muyzer G, Dewaal EC, Uitterlinden AG 1993. Profiling of complex microbial- populations by denaturing gradient gelelectrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S ribosomal-RNA. Appl Environ Microbiol 59: 695–700

Nagy I, Compernolle F, Ghys K, Vanderleyden J, De Mot R 1995. A single cytochrome P-450 system is involved in degradation of the herbicides EPTC (S-ethyl dipropylthiocarbamate) and atrazine by Rhodococcus sp. strain NI86/21. Appl Environ Microbiol 61: 2056–2060

Newcombe DA and Crowley DE 1999.Bioremediation of atrazine-contaminated soil by repeated applications of atrazine-degrading bacteria Appl Microbiol Biotechnol 51: 877– 882

Ngigi AN, Getenga ZM, Boga HI, Ndalut PK 2012. Biodegradation of s-triazine herbicide atrazine by Enterobacter cloacae and Burkholderia cepacia sp. from long-term treated sugarcane-cultivated soils in Kenya. J Environ Sci Heal B 47: 769–78

Nocker A, Burr M, Camper AK 2007. Genotypic microbial community profiling: A critical technical review. Microb Ecol 54: 276–289

Ochman H, Lawrence JG, Groisman EA 2000. Lateral gene transfer and the nature of bacterial innovation. Nature 405, 299–304

Okabe S, Kindaichi T, Ito T 2004. MAR-FISH: An ecophysiological approach to link phylogenetic affiliation and in situ metabolic activity of microorganisms at a single-cell resolution. Microbes Environ 19: 83–98

Pace NR, Stahl DA, Lane DJ, Olsen GJ 1986. The analysis of natural microbial populations by ribosomal RNA sequences. Adv Microb Ecol 9: 1–55

Perelo LW 2010: Review: In situ and bioremediation of organic pollutants in aquatic sediments. J Hazard Mater 177: 81–89

Popov VH, Cornish PS, Sultana K,. Morris EC 2005. Atrazine degradation in soils: the role of microbial communities, atrazine application history, and soil carbon. Soil Res 43: 861–871

Prosser JI 2002. Molecular and functional diversity in soil micro-organisms. Plant Soil 244: 9–17,

Radajewski S, Ineson P, Parekh NR and Murrell JC 2000. Stable-isotope probing as a tool in microbial ecology. Nature 403:646–649

Rasmussen R, Morrison T, Herrmann M, Wittwer C 1998. Quantitative PCR by continuous tluorescence monitoring of a double strand DNA specific binding dye. Biochemica 2

Rhine ER, Fuhrmann JJ, Radosevich M 2003. Microbial community responses to atrazine exposure and nutrient availability: Linking degradation capacity to community structure. Microb Ecol 46: 145–160

Rittmann BE, Valocchi AJ, Seagren E, Ray C, Wrenn B 1992. A critical review of in situ bioremediation. Gas Res Inst, Top. Rep., GRI-92/0322

Rittmann BE 1995. In Situ Bioremediation. Noyes Publications, Park Ridge, NJ, USA

Rodrígues-Valera F 2004. Environmental genomics, the big picture? FEMS Microbiol Lett´231: 153–158

M Romantschuk M, Sarand I, Petänen T, Peltola R, Jonsson-Vihanne M, Koivula T, Yrjälä K, Haahtela K 2000. Means to improve the effect of in situ bioremediation of contaminated soil: an overview of novel approaches. Environ Pollut 107: 179–185

Rothberg JM, Hinz W, Rearick TM, Schultz J, Mileski W, Davey M, et al. 2011. An integrated semiconductor device enabling non-optical genome sequencing. Ecology 58: 1120–1127

Rousseaux S, Hartmann A, Soulas G 2001. Isolation and characterisation of new Gram- negative and Gram-positive atrazine degrading bacteria from different French soils. FEMS Microbiol Ecol 36: 211–222

Roesch, LFW, Fulthorpe RR, Riva A, Casella G, Hadwin AKM, Kent AD, Daroub S, Camargo FAO, Farmerie WG, Triplett EW 2007. Pyrosequencing of 16S rRNA reveal contrasts in microbial diversity of four soils. ISME J 1: 283–290

Rondon MR,August PR, Bettermann AD, Brady SF, Grossman TH, Liles MR, Loiacono KA, Lynch BA, MacNeil IA, Minor C, Lai Tiong C, Gilman M, Osburne MS, Clardy J, Handelsman J, Goodman RM 2000. Cloning the soil metagenome: a strategy for accessing the genetic and functional diversity of uncultured microorganisms. Appl Environ Microbiol 66 2541–2547

Saetre P and Bååth E 2000. Spatial variation and patterns of soil microbial community structure in a mixed spruce–birch stand. Soil Biol Biochem 32: 909–917

Salminen JM, Hänninen PJ, Leveinen J, Lintinen PTJ, Jørgensen KS 2006. Occurrence and rates of terminal electron-accepting processes and recharge processes in petroleum hydrocarbon-contaminated subsurface. J Environ Qual 35: 2273–2282

Satsuma K 2009. Complete biodegradation of atrazine by a microbial community isolated from a naturally derived river ecosystem (microcosm) Chemosphere 77: 590–596

Shaner DL and Henry WB 2007. Field history and dissipation of atrazine and metolachlor in Colorado. J Environ Qual 36:128–134

Shao ZQ and Behki R 1995. Cloning of the Genes for Degradation of the Herbicides EPTC (S-Ethyl Dipropylthiocarbamate) and Atrazine from Rhodococcus sp. Strain TE1. Appl Environ Microbiol 61: 2061–2065

Shao ZQ and Behki R 1996. Characterization of the expression of the thcB gene, coding for a pesticide-degrading cytochrome P-450 in Rhodococcus strains. Appl Environ Microbiol 62: 403–407

Shao ZQ, Steffens W, Mulbry W, Behki RM 1995. Cloning and expression of the s- triazine hydrolase gene (trzA) from Rhodococcus corallinus and development of Rhodococcus recombinant strains capable of dealkylating and dechlorinating the herbicide atrazine. J Bacteriol 177: 5748–5755

Singh AK and Cameotra SS 2014. Influence of microbial and synthetic surfactant on the biodegradation of atrazine Environ Sci Pollut Res 21: 2088–2097

Sinkkonen A, Kauppi S, Pukkila V, Nan H, Plociniczak T, Kontro M, Strömmer R, Romantschuk M 2013. Previous exposure advances the degradation of an anthropogenic s-triazine regardless of soil origin. J Soils Sediments 13: 1430–1438

Skipper HD, Gilmour CM, Furtick WR 1967. Microbial Versus Chemical Degradation of Atrazine in Soils. Soil Sci Soc Am J 31: 653–656

Strong LC, McTavish H, Sadowsky MJ, Wackett LP 2000. Field-scale remediation of atrazine-contaminated soil using recombinant Escherichia coli expressing atrazine chlorohydrolase. Environ Microbiol 2: 91–98

Struthers JK, Jayachandran K, Moorman TB 1998. Biodegradation of atrazine by Agrobacterium radiobacter J14a and use of this strain in bioremediation of contaminated soil. Appl Environ Microbiol 64: 3368–3375

Sturman PJ, Stewart PS, Cunningham AB, Bouwer EJ, Wolfram JH 1995. Engineering scale-up of in situ bioremediation processes: a review. J Contam Hydrol 19: 171–203

Sublette K, Peacock A, White D, Davis G, Ogles D, Cook D, Kolhatkar R, Beckmann D, Yang X 2006. Monitoring subsurface microbial ecology in a sulfate-amended, gasoline- contaminated aquifer. Ground Water Monitor R 26: 70–78

Sun K, Gao B, Zhang Z, Zhan G, Zhao Y, Xing B 2010. Sorption of atrazine and phenanthrene by organic matter fractions in soil and sediment. Environ Pollut 158: 3520– 3526

Suzuki M, Rappé MS, Giovannoni SJ 1998. Kinetic bias in estimates of coastal picoplankton community structure obtained by measurements of small-subunit rRNA gene PCR amplicon length heterogeneity. Appl Environ Microbiol 64: 4522–4529

Solomon KV, Haitjema CH, Thompson DA, O’Malley MA 2013. Extracting data from the muck: deriving biological insight from complex microbial communities and non- model organisms with next generation sequencing. Curr Opin Biotech 28: 103–110

Syngenta 2009. Safety data sheet, camix plus, version 1. 09.07.2009, Syngenta Crop Protection AG, PO Box CH-4002 Basel, Switzerland

Talja KM, Kaukonen S, Kilpi-Koski J, Malin I, Kairesalo T, Romantschuk M, Tuominen J, Kontro MH 2008. Atrazine and terbutryn degradation in deposits from groundwater environment within the boreal region in Lahti, Finland. J Agric Food Chem 56: 11962– 11968

Thelin GP and Stone WW 2010. Method for estimating annual atrazine use for counties in the conterminous United States, 1992–2007. U.S. Geological Survey Scientific Investigations Report 2010–5034, 129 p.

Thompson IP, van der Gast CJ, Ciric L, Singer AC 2005. Bioaugmentation for bioremediation: the challenge of strain selection. Environ Microbiol 7: 909–915

Torsvik V, Daae FL, Sandaa R, Øvreås L 1998. Novel techniques for analysing microbial diversity in natural and perturbed environments. J Biotechnol 64: 53–62

Top EM and Springael D 2003. The role of mobile genetic elements in bacterial adaptation to xenobiotic organic compounds. Curr Opin Biotech 14: 262–269

Tsai YL, Olson BH 1992. Rapid method for separation of bacterial DNA from humic substances in sediments for polymerase chain reaction. Appl Environ Microbiol 58: 2292–2295 van Elsas JD and Boersma FGH 2011. A review of molecular methods to study the microbiota of soil and the mycosphere. Eur J Soil Biol 47: 77–87

Vargha M, Takats Z, Márialigeti K 2005, Degradation of atrazine in a laboratory scale model system with Danube river sediment. Water Res 39: 1560–1568.

Vuorimaa P, Kontro M, Rapala J, Gustafsson 2007. Torjunta-aineiden esiintyminen pohjavedessä. (Occurrence of pesticides in groundwater). The Finnish Environment 42, Finnish Environment Institute, Helsinki (in Finnish)

Wachinger G, Fiedler S, Zepp K, Gattinger A, Sommer M, Roth K 2000. Variability of soil methane production on the micro-scale: spatial association with hot spots of organic material and Archaeal populations. Soil Biol Biochem 32: 1121–1130

Wackett LP, Sadowsky MJ, Martinez B and Shapir N 2002. Biodegradation of atrazine and related s-triazine compounds: from enzymes to field studies. Appl Microbiol Biotechnol 58: 39–45

WHO 1996. Atrazine in Drinking-water. Background document for development of WHO Guidelines for Drinking-water Quality WHO/SDE/WSH/03.04/32.

Williams KH, Bargar JH, Lloyd JR, Lovley DR 2013. Bioremediation of uranium- contaminated groundwater: a systems approach to subsurface biogeochemistry. Curr Opin Biotech 24: 489–497

Winquist E, Björklöf K, Schultz E, Räsänen M, Salonen K, Anasonye F Cajthaml T, Steffen KT, Jørgensen KS, Tuomela M 2014. Bioremediation of PAH-contaminated soil with fungi – From laboratory to field scale. Int Biodeter Biodegr 86: 238–247

Yang C, Li Y, Zhang K, Wang X, Ma C, Tang H, Xu P 2010. Atrazine degradation by a simple consortium of Klebsiella sp. A1 and Comamonas sp. A2 in nitrogen enriched medium. Biodegradation 21: 97–105

Yassir A, Lagacherie B, Houot S, Soulas G 1999. Microbial aspects of atrazine biodegradation in relation to history of soil treatment Pestic Sci 55: 799–809

Yergeau E, Sanschagrin S, D Beaumier, Greer CW 2012. Metagenomic analysis of the bioremediation of diesel-contaminated Canadian high arctic soils. PLOS one 7: e30058. 68

Young IM and Crawford JW 2004. Interactions and Self-Organization in the Soil- Microbe Complex. Science 304(5677): 1634–1637

J Soils Sediments DOI 10.1007/s11368-014-0868-6

SOILS, SEC 4 & ECOTOXICOLOGY & RESEARCH ARTICLE

Atrazine degradation in boreal nonagricultural subsoil and tropical agricultural soil

Aura O. Nousiainen & Katarina Björklöf & Sneha Sagarkar & Shinjini Mukherjee & Hemant J. Purohit & Atya Kapley & Kirsten S. Jørgensen

Received: 14 October 2013 /Accepted: 1 February 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract after the first treatment with atrazine. However, in boreal soil, Purpose The purpose of this study was to determine the decades after atrazine use had been discontinued, residual natural atrazine degradation activity and the genetic potential atrazine was not degraded even though a small number of in a soil profile spanning down to the groundwater zone, degradation genes could still be detected in soil. There is a collected in Finland at a site where past use of atrazine has need for more specific primers for qPCR in tropical soils. contaminated the groundwater, and in Indian agricultural topsoils having different histories of atrazine use. Keywords 14C-mineralization . Atrazine degradation . Materials and methods Atrazine degradation potential was Biodegradation . Quantitative PCR . Soil assessed by quantifying the atrazine degradation genes atzA, trzN,andatzB by quantitative PCR reaction. Atrazine mineralization was studied by radiorespirometry in order to find out 1 Introduction if these genes were expressed. Results and discussion Indian soils contained a large number Atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-tri- 4 5 −1 up to 10 –10 copies (g dry weight (dw) soil) of atrazine azine) has been widely and globally used for broad-leaf weed degradation genes after the first treatment with atrazine. These control since its introduction in 1958. In Finland, atrazine was genes were also expressed, as up to 55 % of atrazine miner- used in weed control on railroads and plant nurseries until its alized. Some unspecific binding of primers required thorough retail was discontinued in 1992 (Savela and Hynninen 2004). investigation and confirmation by sequencing of the qPCR Recently, several studies have shown atrazine to be a possible products in the agricultural soil samples. The degradation endocrine disruptor and a carcinogen (Langerveld et al. 2009; capability of the nonagricultural boreal soil profile was much Hayes et al. 2010). Atrazine and its degradation products are lower: atrazine degradation genes were present at detection persistent contaminants in groundwater, and for this reason, 2 −1 limit (10 copies g soil), but mineralization studies indicated atrazine use has been banned in the EU since 2004 (European that these genes were not transcribed, since no or very little Commission 2004). According to the EU Water Framework atrazine mineralization was observed. Directive, the limit for pesticides in groundwater is 0.1 μgl−1. Conclusions Our results indicate that when atrazine was ap- Despite the controversy regarding its ill effects on health and plied in agricultural practice, the soil atrazine degradation degradation rate, globally, atrazine is still one of the most capacity was high. The organisms responsible for the degra- widely used herbicides today (Pathak and Dikshit 2012). dation were effectively degrading atrazine already 3 months Several factors affect the rates of atrazine degradation. Previous research has shown that atrazine application history Responsible editor: Qixing Zhou hasalargeeffectonatrazinemineralization(Ostrofskyetal A. O. Nousiainen (*) : K. Björklöf : K. S. Jørgensen 1997;Aislabieetal.2004; Morán et al. 2006). Environmental Finnish Environment Institute, P.O box 140, 00251 Helsinki, Finland conditions, such as temperature, depth from soil surface e-mail: [email protected] (Roeth et al. 1969), pH (Mueller et al. 2010), and oxygen content of the surrounding matrix (Ghosh et al. 2004)have S. Sagarkar : S. Mukherjee : H. J. Purohit : A. Kapley Environmental Genomics Division, National Environmental different effects on atrazine degradation. Atrazine can disap- Engineering Research Institute, Nehru Marg, Nagpur 440020, India pear from the topsoil very fast, but both primary and J Soils Sediments secondary compounds may be transported to the deeper soil and accidentally polluted boreal subsoils soils can be layers and leached into the groundwater where conditions are used to evaluate what happens to the atrazine degrada- not optimal for degradation (Bottoni et al. 1996). Due to their tion potential in soil when atrazine use is discontinued. persistence to degradation, these compounds may be The purpose of this study was to investigate the atrazine transported long distances in the groundwater zone in soils mineralization potential, and quantity of atrazine degra- having low organic content and high hydraulic conductivity dation genes in a soil profile at a natural site where (Helling and Gish 1986). atrazine contamination had occurred decades previously, Atrazine biodegradation has been studied widely since and in agricultural soils which are continuously exposed 1980s, and atrazine degradation genes, providing complete to atrazine. From these soils, we quantified atzA, trzN, mineralization of atrazine, have been discovered. The path- and atzB genes to find out how the copy numbers relate ways are described in the online pathway resource KEGG to actual mineralization of atrazine in microcosms. We (http://www.kegg.jp) (Kanehisa et al. 2012). Mineralization chose atzA and trzN genes because they both encode for has been shown to be possible in both aerobic (e.g., Martin- the initial atrazine dechlorination step. atzB is the next Laurent et al. 2004) and anaerobic conditions (Ghosh et al. key gene in the pathway encoding for the hydroxy 2004). Known atrazine degraders include species from γ- atrazine degradation, which is a prerequisite for the proteobacteria, such as the gram-negative Pseudomonas,as degradation of atrazine to the key intermediate, cyanuric well as gram-positive bacteria, such as Arthrobacter and acid, through which atrazine is completely mineralized. Rhodococcus.InPseudomonas, the dechlorination route is well characterized and known to be coded by atzA-atzF genes found in ADP plasmid. The trzN gene has an analogous 2Materialsandmethods function to atzA (Mulbry et al. 2002) and is found mainly in plasmid TC1 or chromosomal DNA in gram-positive bacteria 2.1 Sampling and sample characterization such as Arthrobacter (Strong et al. 2002). Also, gram-negative bacteria, such as Sinorhizobium and Polaromonas are known 2.1.1 Boreal soil profiles to contain this gene (Devers et al. 2007). The genes in atz and trz pathways have been detected all around the world in Boreal soil samples were collected near the Uusiniitty agricultural soils by PCR (e. g., Smith et al. 2005; Yamazaki waterworks in Lohja, southern Finland, where atrazine et al. 2008; Sagarkar et al. 2013)andatz genes by quantitative and its degradation products had been observed in the PCR (Martin-Laurent et al. 2004), but little is known of the groundwater in concentrations up to 0.26 μgl−1 presence of atz genes and particularly trzN gene in subsurface (Vuorimaa et al. 2007). The site is on a coniferous forest soils and groundwater, where atrazine may migrate after a ridge which has not been in agricultural use. It is possible long time since exposure to topsoil. that the atrazine detected in groundwater has leached from Investigation of functional genes by quantitative PCR the nearby areas where atrazine has been used for various (qPCR) methods has provided researchers with informa- purposes. Three soil cores were collected, and soil from tion about the presence, quantity, and quality of biolog- these cores were combined as three bulk samples from ical transformations in soil of xenobiotic compounds. three layers: topsoil (surface) representing the first meter These tools can be used to elucidate how the microbial of topsoil, intermediate layer representing the deeper soil community shifts during pesticide biodegradation unsaturated with groundwater (unsaturated) at 2–5m,and (Bælum et al. 2006) and how the genetic potential is saturated zone, representing the soil saturated with ground- linked to pesticide mineralization (Martin-Laurent et al. water (saturated) at 6–9 m (Fig. 1). All soils were sieved 2004;Monardetal.2008). The qPCR method benefits through an 8 mm sieve, mixed well, and packed in plastic from greater sensitivity than conventional PCR methods. buckets. Samples were stored in +4 °C until use. Chem- However, when template DNA concentration in soil is ical analyses, mineralization experiment, and DNA extrac- low, the effects of PCR inhibition, mostly due to humic tion for the boreal soil were initiated 1–2 days after acids and other soil compounds co-extracting with sampling. The groundwater level at the sampling area nucleic acids (Braid et al. 2003) are more pronounced varied at the time of sampling between 4.8 and 6.5 m. (Frostegård et al. 1999). Furthermore, the vast diversity The soils were sandy, low in organic matter, and had a of soil microorganisms has to be taken into account low pH value (Table 1). Atrazine and its degradation when primers are designed in order to avoid unspecific products DEA, DIA, and DEDIA were extracted from primer binding. Therefore, also well-established primers the soil samples with 70 % methanol and analyzed by need to be tested in different soil types and soil layers. HPLC/MS/MS by Ramboll analytics Oy (Lahti, Finland). As atrazine is still widely used in many parts of the world, a The concentrations were below the detection limit comparison between atrazine degradation in agricultural soils 0.05 mg kg−1. J Soils Sediments

nonvegetated field with no atrazine use history as “no treatment” control (Fig. 1). The top 10 cm layer was collected from three sites on each field. Three samples were taken from each field. Each sample consisted of three subsamples, collected on the center, halfway, and near the edges of the field. Subsam- ples were mixed together in a plastic bucket, roots and stones were removed. Samples were stored in +4 °C until use. DNA was extracted 4 days, mineralization experiment started 20 days, and chemical analyses were carried out 1 month after sampling. Soils were rich in organic content and had a neutral pH (Table 1). Atrazine was detected in two of the samples (Table 1), and the degradation product DEA was detected in the amounts of 0.06 mg kg−1 in the sample with a 3-year Fig. 1 Description of material used in this research. Each square repre- history in atrazine use. DIA and DEDIA were not detected in sents a composite soil sample collected from boreal or tropical sampling any of the samples. site 2.2 Nucleic acid extraction from soil samples and control strains 2.1.2 Atrazine-treated tropical topsoils 2.2.1 Soil samples Agricultural soil samples were collected in sugarcane fields in the Maharashtra Province, India, where atrazine is used in Total DNA was extracted from the soil samples by the sugarcane cultivation. We sampled on three fields, each hav- FastDNA Spin Kit for Soil (MP Biomedicals, OH, USA) ing an area of approximately 2 ha, which had been treated with according to the manufacturer’s instructions. The samples atrazine for various periods (Table 1). The fields had a history were homogenized in a FastPrep instrument on setting 4.5 of 3 months, 3 years, and 8 years of atrazine use. The latest for 30 s. DNA yields were between 9.4 and 48 ng μl−1. atrazine treatment had been 3 months ago on the fields with 3- month and 3-year histories. These fields had not been harvested and were vegetated with sugarcane at the time of sampling. 2.2.2 Control strains The last atrazine treatment on the field with 8 years of atrazine application history had been 5 months ago, and this field had Control strains Pseudomonas citronellolis ADP (DSM 11735) been harvested. We also collected soil from an adjacent and Arthrobacter aurescens TC1 (ATTC BAA-1386) were

Ta b l e 1 Characteristics of boreal and tropical soil samples used in this study

Origin Depth Vegetation Time since the Atrazine pH value Dry Organic NO3 NH4 Atrazine (m) last application exposure matter (%) matter mg/kg mg/kg mg/kg with atrazine history (%) (dw)a (dw)a (dw)a

Boreal topsoil 0–1 + No application No exposure 4.9±0.1 96±0.1 1.1±0.0 b 17 c Boreal soil 2–5 Subsurface No application Exposure by 5.2±0.2 94±0.0 0.5±0.0 b 4.4 c unsaturated d ground water Boreal soil 6–9 Subsurface No application Exposure by 6.0±0.2 85±0.2 0.2±0.0 b 3.8 c saturated ground water Tropical 0–0.1 + 3 months 3 months 6.9±0.4 61 4.8±0.4 180 86 0.005 Tropical 0–0.1 + 3 months 3 years 7.3±0.1 63 4.9±0.3 120 80 0.007 Tropical 0–0.1 − 5 months 8 years 6.9±0.1 72 3.6±0.8 190 87 c Tropical 0–0.1 − No application No exposure 7.1±0.1 69 4.2±0.4 180 84 c a Single determination of sample pooled from three subsamples b Below detection limit 10 mg kg−1 (dw) c Below detection limit 0.005 mg kg−1 (dw) d Ground water level at 5 m + Vegetated at the time of sampling - Not vegetated at the time of sampling J Soils Sediments obtained from culture collections. Both bacteria were cultivat- were confirmed by amplifying the target gene using M13 ed on minimum medium adapted from Cai et al. 2003.Liquid primers, which target the vector, and the correct insert was −1 −1 atrazine medium contained 0.9 g l KH2PO4,3gl confirmed by sequencing with T7 primers by Macrogen Cor- −1 Na2HPO4×12H2O, 0.2 g l MgSO4×7H2O, supplemented poration (Seoul, South Korea). The plasmids containing the with 3 g l−1 sodium citrate as carbon source and 100 mg l−1 of target genes atzA, trzN,andatzB, respectively, were purified atrazine as the sole nitrogen source. Atrazine was added from by the PureLink™ Quick Plasmid Miniprep Kit (Invitrogen, a 100 mg ml−1 stock solution (in methanol). For solid media, Carlsbad, CA, USA) and further purified from genomic DNA 2 % agar was added. The bacteria were grown in +28 °C on by running the plasmid extracts on 2.5 % agarose gel, excising solid medium. Exponential growth phase cells from the correct bands, and purifying them from the gel by the P.citronellolisADP and A. aurescens TC1 were harvested. Wizard SV Gel and PCR Clean-Up System (Promega, WI, The ADP plasmid from P.citronelloliswas purified by the USA). The DNA concentrations of the plasmids were quanti- PureLink™ Quick Plasmid Miniprep Kit (Invitrogen, Carls- fied using the Qubit Fluorometer (Invitrogen Molecular bad, CA, USA). Total DNA from A. aurescens was extracted Probes, OR, USA) and the HS (High Sensitivity) assay. Each with FastDNA Spin Kit (MP Biomedicals, OH, USA). plasmid extract was quantified with Qubit using three separate standard series, and each standard series was measured three 2.3 Conventional PCR times. An average of these measurements was used to calculate the copy numbers of the specific gene-containing plasmid 1– 5 All samples were analyzed by conventional PCR. The 50 μl in solutions. Standard dilution series containing 10 10 cop- μ reaction mix contained 7–18 ng of template DNA, 10 pmol of ies of the gene were prepared and aliquoted in 10 lbatches. − forward and reverse primers (Table 2), 0.2 pmol of dNTP Plasmids were stored in 70 °C until use.

(Fermentas, Espoo, Finland), 2.5 mM MgCl2,10μgBSA (New England Biolabs, MA, USA), and 1 U Fermentas Ta q 2.5 Quantitative PCR polymerase with 1× buffer (Fermentas, Vilnius, Lithuania). The cycling conditions were 95 °C denaturation for 5 min, The qPCR reaction was optimized by testing different anneal- 30 cycles of 1 min denaturation at 95 °C, 1 min at annealing ing temperatures, MgCl2, and BSA additions. To reduce the temperature (Table 2), 1 min extension at 72 °C, and finally effect of PCR-inhibiting compounds, different sample dilu- 5 min final extension at 72 °C. As nearly all reactions were tions and purifications using 5.5 M guanidine thiocyanate and negative, quantitative PCR was used for improved sensitivity. post-DNA extraction purification with phenol/chloroform extraction and ethanol precipitation of DNA were tested (data 2.4 Preparation of qPCR standards not shown). qPCR reaction was carried out in triplicate by an Applied Biosystems 7300 Real time PCR instrument. Each Using the ADP plasmid and DNA extracted from A. aurescens 25 μl reaction constituted of 0.3 μmol of each primer, 12.5 μl cells PCR was carried out to amplify the target genes atzA, of Maxima™ SYBR Green qPCR Master Mix (Fermentas, atzB,andtrzN using primers specified in Table 2. PCR con- Vilnius, Lithuania) which includes SYBR® Green I dye, ditions were as follows: initial denaturation at 94 °C for 5 min, dNTP, KCl, (NH4)2SO4, and Maxima™ Hot Start Ta q DNA 30 cycles of 94 °C 1 min, 1 min at specific annealing temper- Polymerase and ROX as a passive reference dye. The amount ature, and 1 min at 72 °C followed by final extension of 72 °C of template added in each reaction was between 0.2 and 32 ng. 5 min. The PCR product was checked for correct size on 1 % DNA extracts from Indian soils were diluted 1:10 to reduce agarose gel. Each PCR product was cloned in pCR®4-TOPO® the effect of PCR inhibition. Calibration curves for each run vector (Invitrogen, Carlsbad, CA, USA) and transformed in were generated by running the appropriate qPCR standard chemically competent TOP10 Eschericia coli cells according dilution series. No endogenous control was used. qPCR cy- to the manufacturer’s instructions. The transformed clones cling conditions were 95 °C initial denaturation, 40 cycles of

Ta b l e 2 Primers used in PCR and qPCR reactions for atrazine degradation genes

Gene Primer sequences Target enzyme Annealing temperature Product size Reference atzA atzAf 5′-ACG GGC GTC AAT TCT ATG AC-3′ Atrazine chlorohydrolase 58 °C 204 bp Devers et al. 2004 atzAr 5′-CAC CCA CCT CAC CATAGA CC-3′ trzN trzNf 5′-GCG ACG GGA AGT TCG GTC-3′ Triazine hydrolase 61 °C 216 bp Ghosh et al. 2008 trzNr 5′-CGA GCG TCA TCG ATG ACC T-3′ atzB atzBf 5′-AGG GTG TTA GGT GGT GAA C-3′ Hydroxyatrazine hydrolase 56 °C 200 bp Devers et al. 2004 atzBr 5′-CAC CAC TGT GCT GTG GTA GA-3′ J Soils Sediments

1 min 95 °C, 1 min at specific annealing temperature (Table 2), by conventional PCR. PCR was carried out in a 50 μlreaction and 1 min 72 °C. An additional step at 30 s at 81 °C after the containing 0.05–1 ng of template. PCR was carried out as elongation step was used as fluorescence reading step to described previously. The PCR end products were purified exclude signal derived from primer dimer or unspecific prod- with the PureLink Quick Plasmid Miniprep Kit (Invitrogen, ucts having melting point below 81 °C. Melting curve analysis Carlsbad, CA, USA), cloned in the pCR®4-TOPO® vector was performed at the end of each run from 60 to 95 °C. qPCR (Invitrogen, Carlsbad, CA, USA) according to the manufac- end products were analyzed in 1.5 % agarose gel, 120 V 1 h. turer’s instructions, and transformed in chemically competent The bands were visualized by the SYBR Green dye DH5α E. coli cells (Invitrogen, Carlsbad, CA, USA). The (Invitrogen, USA). Gene copy numbers were calculated by transformed clones were verified by amplifying the target relating the cycle threshold (Ct) fluorescence of the samples to gene using M13 primers which target the vector, and correct the calibration curve. The efficiency of the qPCR reactions insert was confirmed by sequencing with T7 primers by were 100±10 %. Macrogen Corporation (Seoul, South Korea) in Amsterdam, the Netherlands. The sequencing results were compared with 2.6 Reaction analysis sequences in GenBank on 26 July 2011.

Obtained gene quantities in the boreal samples were low, often close to the assay detection limit. Especially in case of boreal 2.8 Atrazine mineralization samples, sometimes, only one or two of the three replicates produced a signal. We expect the result to be closer to the Atrazine mineralization to CO2 was studied by detection limit than to zero when no signal was detected in radiorespirometry. An amount of 2 kBq of ring-labeled 14C- only one or two of the three replicates, and for these samples, a atrazine [specific activity 160 mCi mmol−1; 99 % radiochem- value of log (assay detection limit)—0.5 log was used instead ical purity (Larodan Fine Chemicals Ab, Sweden)] was added of zero when mean values of the three parallels were calculat- to 10 g soil samples in 100 ml infusion bottles. The concen- ed (Björklöf et al. 1995). For this reason, some of the average tration of radiolabeled atrazine in the soil was 0.1 mg kg−1. gene copy numbers seem lower than detection limit. Detection This small amount was used, because it mimics the in situ limit was determined as the theoretical detection limit of one conditions as closely as possible. All samples were studied in gene copy in each reaction. For the boreal samples, it was 100 triplicate. Positive control consisted of 10 g topsoil and copies per reaction, and for tropical samples, 1,000 copies per P. citronellolis control strain. Abiotic control was topsoil reaction due to dilution of samples. In some qPCR reactions autoclaved on three consecutive days before the experiment. from tropical soil, additional PCR nonspecific products were The sealed bottles were incubated in 10 °C and 30 °C, detected. In these cases, only the reactions with a strong band representing the temperature levels in boreal and tropical of correct size and correct melting temperature were included climates, in the dark for up to 80 days. The production of 14 in further sample analysis. Nonspecific products were ran- C-labeled CO2 in the bottles was trapped 1 ml 1 M NaOH, domly selected and sequenced to determine their identity. which was analyzed in a Winspectral 1414 Liquid scintillation counter (Wallac Oy, Turku, Finland) using Ultima Gold high 2.7 Cloning and sequencing qPCR end products flash point LSC- scintillation cocktail (PerkinElmer, MA, USA). 2.7.1 Cloning correct size bands of atzB and trzN atzB and trzN qPCR end products of right size were excised 2.9 Statistical analysis from the gel and DNA extracted from the gel slice using the Wizard SV Gel and PCR Clean-Up System (Promega, WI, One-way analysis of variance (ANOVA)with Bonferroni post USA). The qPCR products were cloned in the pCR®4-TOPO® hoc tests were conducted using the SPSS v. 15 software to vector (Invitrogen, Carlsbad, CA, USA) according to the define the differences between (a) copy numbers of different manufacturer’s instructions and transformed in chemically genes in different depths of boreal soil and (b) the amounts of competent CJ236 E. coli cells (Takara Bio Inc, Otsu, Japan) atrazine mineralized after incubation in microcosms in the and sequenced. same soils; as well as (c) copy numbers of different genes in different tropical fields and (d) the amounts of atrazine min- 2.7.2 Cloning correct size bands of atzA and nonspecific eralized after incubation in microcosms in the same soils. In products of atzA, atzB, and trzN the case of replicate qPCR samples which did not produce signal, and samples where none of the triplicates produced Transformation to CJ236 cells was not always successful. In signal, a value of log (assay detection limit)—0.5 log was used these cases, the purified qPCR end products were reamplified instead of zero. J Soils Sediments

3Results 3.2 Analysis of qPCR specificity

3.1 Quantification of degradation genes In boreal soils, atzA and trzN qPCR products contained only single bands of expected size on gels. In the case of atzB The quantification of all genes in boreal samples was chal- assay, also additional products were seen. As the tropical lenging due to low copy numbers, which were close to the soils contained more organic material, which is prone to detection limit (Fig. 2, upper panel). However, when the disturb the function of the polymerase, extra care was taken reaction was successful, a product with correct size and melt- in analyzing the results. Most of the qPCR assays from ing temperature was always present. atzA and trzN genes were tropical soils produced additional products, which were de- absent in the topsoil and they were only found in the deeper, tected both on agarose gels and in melting curve profiles. unsaturated, and saturated soil layers. The copy numbers of For this reason, approximately 70 % of the qPCR results trzN were 102 gene copies g−1 dry weight (dw) and atzA up to were analyzed further and discarded if the product was not 103 gene copies g−1 (dw) in boreal soils. The ANOVA test verified to be correct. In order to confirm the identity of the confirmed that the trzN gene copy numbers in deeper soil expected and unexpected qPCR products, the sequences of layers differed from that in topsoil (Bonferroni test P= the most prominent bands seen on agarose gel were cloned 0.000), but the differences of atzA copy numbers between andidentifiedbysequencing(Table3). Because right sized the topsoil and deeper layers were less significant (Bonferroni bands corresponded to the expected products in the case of test P=0.086). The atzB gene was successfully amplified in all genes, the obtained quantification result could be consid- only one sample from topsoil in only one replicate sample. ered reliable, and the quantification results reported here are ANOVA tests confirmed that the difference of atzB copy based on confirmed correct PCR products only. The se- numbers to no template control was not statistically significant quences of qPCR product sequences having the right size (Bonferroni test P=0.105). In the tropical soil, atzA approxi- reported in this study have been deposited in EMBL Data mately 104 copies g−1 soil (dw) were detected in all samples Library, accession numbers HE579075–HE579082. Several (Fig. 2, lower panel). trzN genes were only found in one field prominent unexpected sized bands were also sequenced. (3 years atrazine treatment, 3 months since last application, These PCR products were consistently of the same size, vegetated) but in large quantity (0.9×105 copies g−1 soil). The for atzA, approximately 600, 450, 400, and 350 bp, and atzB gene was present in all soils in up to 104 gene copies g−1 for atzB, approximately 900, 600, and 300 bp. For the trzN soil. According to the ANOVA test, the differences between reaction, the unexpected bands did not show any persistent the copy numbers of atzA and atzB in different fields were not pattern, and therefore, bands from both boreal and tropical significant (Bonferroni test P=0.129 and 0.161, respectively), soils with sizes between 480 and 950 base pairs were but the trzN copy numbers in 3 years treated soil was different sequenced. Examples of the sequence identities of unexpect- from the other fields (Bonferroni test P=0.000). ed bands in Table 3 show that the amplicons contained a

Fig. 2 Atrazine degradation gene quantities in boreal and tropical soils. Dashed line represents the theoretical assay detection limit. Atrazine treatment history is indicated a by 3 months, b by 3 years, c by 8 years, and d by no treatment J Soils Sediments

Ta b l e 3 Closest sequence identities to Genbank sequences from qPCR end product clones of atrazine degradation genes

Gene targeted product size (bp) Clone origin Closest match in GenBank Match length (bp) Max ident (%) atzA 204a Tropical, 3 months FN568345.1 atzA of Pseudomonas nitroreducens 206 100 atzA 596 Tropical, 3 months No match atzA 473 Tropical, 3 months AB546581.1 uncultured Sordariomycetidae 73 100 gene for 28S rRNA atzA 396 Tropical, 3 months AJ786661.1 Limonium gibertii ribulose 1,5 75 100 bisphosphate carboxylase/oxygenase atzA 360 Tropical, 3 months AB490235.1 Ochrobactrum anthropi nitrate 247 96 reductase and FM246832.1 Uncultured marine bacterium 83 100 partial aprA gene trzN 216b Boreal topsoil GU459314.1 Arthrobacter sp. T3AB1 triazine 216 100 hydrolase (trzN)gene trzN 742 Tropical, 3 years No match trzN 664 Tropical, 3 years No match trzN 572 Tropical, no treatment FM246832.1 uncultured marine bacterium 81 100 partial aprA gene atzB 200c Tropical, 3 months, FN568346.1 Pseudomonas nitroreducens 200 99 partial atzB gene atzB 948 Boreal unsaturated soil AB546581.1 uncultured Sordariomycetidae 90 100 gene for 28S rRNA atzB 594 Boreal unsaturated soil AB546581.1 uncultured Sordariomycetidae 90 100 gene for 28S rRNA atzB 307 Boreal topsoil No match a Correct band size, sequence deposited in GenBank, accession numbers HE579075–HE579076 (two clones) b Correct band size, sequence deposited in GenBank, accession numbers HE579079–HE579082 (four clones) c Correct band size, sequence deposited in GenBank, accession numbers HE579077–HE579078 (two clones) long stretch of unspecific DNA and a short, less than 100 bp more pronounced in 30 °C: P values for ANOVA post hoc fragments of fungal 28 s genes or alkaline phosphatase aprA tests were 0.014 for 3 months treated field, 0.055 for 3 years gene, and in a single case of atzA band, also a longer treated field, and 1.000 for 8 years treated field, whereas in fragment of a nitrite reductase. 10 °C, the values were 0.229 (3 months), 0.162 (3 years), and 1.000 (8 years). In both temperatures studied, the best mineralization was achieved in the samples collected from fields 3.3 Conventional PCR where previous exposure of atrazine had been 3 months ago, and the same mineralization trend was observed in both tem- No conventional PCR reactions were positive for the boreal peratures studied. Both of these study sites were vegetated samples. In the tropical soil samples, only atzB could be with sugarcane at the time of sampling, whereas the other amplified from two samples which had been treated with study sites were unvegetated. It is probable that the effect of atrazine 3 months previous to our sampling. All other PCR plants was small, as we used bulk soil for our experiments, reactions for atzB as well as for atzA and trzN were negative. and during the mineralization experiment, the plant component was not present. 3.4 Atrazine mineralization in microcosms

Atrazine mineralization was hardly detectable in the boreal soils (Fig. 3). In 10 °C, the mineralization of the sterile control 4 Discussion differed from the mineralization of all samples (P=0.034, 0.020, and 0.025). In 30 °C, the differences between samples In the present study, the presence of atrazine degradation from different depths were not significant (P=1.000), and they genes atzA, trzN,andatzB,andtheactualatrazineminerali- did show statistically relevant differences as to the positive zation were studied in three layers of soil contaminated with (P=0.143, 0.177, or 0.154) but not to autoclaved controls (P= atrazine used in the area decades ago, as well as in soil from 1.000). In the tropical topsoils, atrazine mineralization up to three agricultural fields with varying atrazine use histories. 25 % was obtained in 10 °C, and up to 55 % in 30 °C. (Fig. 3). Our results suggest that even after a long time since last Differences between treated fields and nontreated field were exposure to atrazine, some genes responsible for its J Soils Sediments

Fig. 3 Atrazine mineralization in boreal soils in 10 °C (1A) and 30 °C (1B). Atrazine mineralization in tropical soils 10 °C (2A) and 30 °C (2B) degradation are present in soil even though atrazine does not finding could be elucidated by quantifying the transcribed mineralize. RNA directly from soil. Unfortunately, due to a very low Bottoni et al. (1996) showed that atrazine and its degrada- concentration of nucleic acids, RNA could not be reliably tion products have a tendency to leach into the groundwater. quantified in this study. However, atrazine mineralization Although atrazine mineralization in topsoil can be efficient, and the presence of degradation genes would have indicated studies concerning subsoils show that degradation conditions that degradation enzymes would have been present, and both are not optimal in deeper soil layers (Willems et al. 1996; transcription of DNA to RNA and translation of RNA to Miller et al. 1997). We found low numbers of atrazine degra- enzymes would be ongoing. dation genes from the boreal soil profile which had been Martin-Laurent et al. (2004) studied the effects of organic polluted with atrazine by vertical groundwater flow. Degrada- amendments in different agricultural soils treated with atra- tion genes atzA and trzN, coding the initial step in atrazine zine. In their study, the atzA gene copy numbers correlated degradation, were only found in the deeper layers exposed to with mineralization capacity. We found no correlation be- the groundwater flow. The atzB gene was detected in one tween any of the studied genes and the mineralization capacity sample, but due to the low concentration, its copy numbers of the soil, as relatively similar copy numbers were observed could not be reliably quantified. In our mineralization exper- in all our agricultural soils regardless of their atrazine treat- iment, atrazine was only added in micromolar concentrations ment history. The level of copy numbers in our agricultural mimicking conditions in situ. Although the degradation genes samples was similar to those reported by Martin-Laurent et al. were present in the deeper soil layers, no significant mineral- (2004). Cheyns et al (2012) discovered that lag time in atra- ization was detected in the soil profile. Studies on gene ex- zine mineralization increased with time from last atrazine pression on pure cultures of atrazine degraders have revealed application. We observed a similar trend: time since last that different bacterial species express degradation genes con- application had a significant effect on how well atrazine trarily: Pseudomonas is expressing the genes even when no mineralized, whereas the long-term history of atrazine use in atrazine is present, whereas in Chelatobacter, the genes are each field was not reflected in atrazine mineralization efficien- only expressed in high concentration of atrazine (Devers et al. cy. Atrazine degradation genes were present in high copy 2004). Our study might support the assumption that low numbers in the field which was treated with atrazine only concentration of atrazine is not high enough for degradation 3 months previously, and the same soil also readily mineral- gene transcription even if the genes are present in soil. This ized atrazine. However, the field treated 5 months previous to J Soils Sediments our sampling mineralized atrazine at a clearly reduced rate, analysis of qPCR product length and sequence. With the despite the 8-year history with atrazine use and even though acquired sequence information decisions about which results the degradation genes were present on this field in similar to use for calculating results could be made. copy numbers. The soils mineralizing atrazine most efficiently were vegetated at the time of sampling. The rhizosphere effect of pesticide degradation (Anderson et al. 1993) is another 5 Conclusions possible explanation to faster mineralization rate, although we estimated its effect to be small since roots and the soil We detected known atrazine degradation genes in deep layers attached to them were removed, and during the mineralization of boreal soil, which has been in contact with atrazine- experiment only bulk soil was used. contaminated groundwater. Atrazine had not been used in this Our data shows, that even though gene quantities remain area, but it was still present in the groundwater at low con- rather constant over time, the degradation potential decreases centration. We did not detect any atrazine mineralization in rapidly if frequent reexposure with atrazine is not maintained. these soils, but it is likely that the organisms capable of In a study conducted by Barriuso and Houot (1996), the atrazine degradation are still present. We suspect that the longest atrazine application history yielded the most efficient atrazine concentration in the area is high enough to maintain mineralization. In our study, no trend was observed corre- some of the degradation genes, but too low for their transcrip- sponding to atrazine use history, and the factor best explaining tion. To our knowledge, this is the first study describing the mineralization rate was time since last treatment with atrazine degradation genes in boreal, nonagricultural subsoils. atrazine. Also, 10 % of atrazine was mineralized in our agri- In our agricultural samples, the longer atrazine application cultural control soil which had never been treated with atra- history seemed to have smaller effect on atrazine degradation zine. Ostrofsky et al. (2002) reported similar degradation rates than time from previous atrazine exposure, as high copy in nonexposed soil, which after the first atrazine addition numbers of degradation genes could be detected only 3 months treatment mineralized over 60 %. The atz genes have been after a field was treated with atrazine for the first time. This thought to be globally widespread (de Souza et al 1998). For would indicate that atrazine degradation genes appear very this reason, it is likely that the organisms capable of atrazine fast after the first treatment with atrazine. The gene copy degradation are not uncommon in soil, but their numbers are numbers did not change according to the long application small when atrazine is not present. history, and small amounts of the genes were present even in Because qPCR reaction is more sensitive than traditional soils where atrazine selection pressure is very low. Our results PCR, it can be used for samples having low template copy suggest that atrazine is easily degraded in topsoil, as frequent numbers (Salminen et al. 2008). Due to the low template contact with atrazine is maintained. However, some atrazine number and high concentration of inhibitory substances, remains in the deeper soil layers years after its use has been qPCR reaction was very difficult to optimize. Additional stopped, and that low atrazine concentration does not stimu- purification of our DNA extracts and additional PCR-aiding late degrading activity. substances did not decrease the PCR inhibition. However, The most frequently used primers for qPCR of atrazine dilution of the samples 1:10 proved beneficial in inhibition degradation genes were found to amplify unspecific products reduction. As our qPCR reactions could not be optimized to in tropical agricultural soils, and should not be used without avoid unspecific products, we decided to verify the expected checking the quality of the amplified product. PCR products and identify the unexpected PCR products by sequencing. Our results confirm that the qPCR products hav- Acknowledgments We thank Ilse Heiskanen, Ritva Väisänen, and ing correct size and right melting temperature always contain Laura Häkkinen for the technical assistance, and the municipality of the target sequence. This is in line with a quality assurance Lohja for granting us access to the sampling site in Finland. This research performed by Monard et al (2008). The sequencing results was funded by the Academy of Finland, grant no.124822, to K. Jørgensen revealed that the primer specificity of the atzA, atzB,andtrzN and Dept. of Biotechnology, Ministry of Science and Technology, New Delhi, to A. Kapley (Indo-Finnish programme on environmental genes was not optimal for the tropical samples. This might be biotechnology). due to the number and diversity of the organisms inhabiting the agricultural soil, which had greater carbon concentration than the boreal soil (Schnüner et al. 1985). The primers References amplifying qPCR product length of 50 to 150 base pairs, which is optimal for qPCR is difficult to achieve with some DNA sequences due to secondary structures and high CG Aislabie J, Hunter D, Ryburn J, Fraser R, Northcott GL, Di HJ (2004) concentration in primers. For organic soils, new or alternative Atrazine mineralisation rates in New Zealand soils are affected by time since atrazine exposure. Aust J Soil Res 42:783–792 degradation gene primers should be designed and tested in Anderson TA, Guthrie EA, Walton BT (1993) Bioremediation in the order to obtain less unspecific binding and avoid the tedious rhizosphere. Environ Sci Technol 27:2630–2636 J Soils Sediments

Bælum J, Henriksen T, Bruun Hansen HC, Jacobsen CS (2006) regulate immune and growth-related functions in developing Degradation of 4-chloro-2-methylphenoxyacetic acid in top- and Xenopus laevis tadpoles. Environ Res 109:379–389 subsoil is quantitatively linked to the class III tfdA gene. Appl Martin-Laurent F, Cornet L, Ranjard L, López-Gutiérrez JC, Philippot L, Environ Microbiol 72:1476–1486 Schwartz C, Chaussod R, Catroux G, Soulas G (2004) Estimation of Barriuso E, Houot S (1996) Rapid mineralization of the s-triazine ring of atrazine-degrading genetic potential and activity in three French atrazine in soils in relation to soil management. Soil Biol Biochem agricultural soils. FEMS Microbiol Ecol 48:425–435 28:1341–1348 Miller JL, Wollum AG, Weber JB (1997) Degradation of carbon-14- Björklöf K, Suoniemi A, Haahtela K, Romantschuk M (1995) High atrazine and carbon-14-metolachlor in soil from four depths. J frequency of conjugation versus plasmid segregation of RP1 in Environ Qual 26:633–638 epiphytic Pseudomonas syringae populations. Microbiology 141: Monard C, Martin-Laurent F, Vecchiato C, Francez AJ, 2719–2727 Vandenkoornhuyse P, Binet F (2008) Combined effect of bioaug- Bottoni P, Keizer J, Funari E (1996) Leaching indices of some major mentation and bioturbation on atrazine degradation in soil. Soil Biol triazine metabolites. Chemosphere 32:1401–1411 Biochem 40:2253–2259 Braid MD, Daniels LM, Kitts CL (2003) Removal of PCR inhibitors from Morán AC, Müller A, Manzano M, Gonzáles B (2006) Simazine treat- soil DNA by chemical flocculation. J Microbiol Methods 52:389– ment history determines a significant herbicide degradation potential 393 in soils that is not improved by bioaugmentation with Pseudomonas Cai B, Han Y, Liu B, Ren Y, Jiang S (2003) Isolation and characterization sp. ADP. J Appl Microbiol 101:26–35 of an atrazine-degrading bacterium from industrial wastewater in Mueller TC, Steckel LE, Radosevich M (2010) Effect of soil pH and China. Lett Appl Microbiol 36:272–276 previous atrazine use history on atrazine degradation in a Tennessee Cheyns K, Martin-Laurent F, Bru D, Aamand J, Vanhaecke L, Diels J, field soil. Weed Sci 58:478–483 Merckx R, Smolders E, Springael D (2012) Long-term dynamics of Mulbry WW, Zhu H, Nour SM, Topp M (2002) The triazine hydrolase the atrazine mineralization potential in surface and subsurface soil in gene trzN from Nocardioides sp. strain C190: cloning and construc- an agricultural field as a response to atrazine applications. tion of gene-specific primers. FEMS Microbiol Lett 206:75–79 Chemosphere 86:1029–1034 Ostrofsky EB, Traina SJ, Tuovinen OH (1997) Variation in atrazine de Souza ML, Seffernick J, Martinez B, Sadowsky MJ, Wackett LP mineralization rates in relation to agricultural management practice. (1998) The atrazine catabolism genes atzABC are widespread and J Environ Qual 26:647–657 highly conserved. J Bacteriol 180:1951–1954 Ostrofsky EB, Robinson JB, Traina SJ, Tuovinen OH (2002) Analysis of Devers M, Soulas G, Martin-Laurent F (2004) Real-time reverse tran- atrazine-degrading microbial communities in soils using most- scription PCR analysis of expression of atrazine catabolism genes in probable-number enumeration, DNA hybridization, and inhibitors. two bacterial strains isolated from soil. J Microbiol Meth 56:3–15 Soil Biol Biochem 34:1449–1459 Devers M, El Azhari N, Kolic NU, Martin-Laurent F (2007) Detection and Pathak RK, Dikshit AK (2012) Atrazine and its use. Int J Res Chem organization of atrazine-degrading genetic potential of seventeen bac- Environ 2:1–6 terial isolates belonging to divergent taxa indicate a recent common Roeth FW, Lavy TL, Burnside OC (1969) Atrazine degradation in two origin of their catabolic functions. FEMS Microbiol Lett 273:78–86 soil profiles. Weed Sci 17:202–205 European Commission (2004) Decision concerning the non-inclusion of Sagarkar S, Mukherjee S, Nousiainen A, Björlöf K, Purohit HJ, atrazine in annex I to the council directive 91/414/EEC and the Jørgensen KS, Kapley A (2013) Monitoring bioremediation of withdrawal of authorisations for plant protection products contain- atrazine in soil microcosms using molecular tools. Environ Pollut ing this active substance. Off J Eur Union L78:53–55, Available: 172:108–115 http://ec.europa.eu/food/plant/protection/evaluation/existactive/oj_ Salminen J, Tuomi P, Jørgensen K (2008) Functional gene abundances atrazine.pdf [accessed 27 August 2010] (nahAc, alkB, xylE) in the assessment of the efficacy of bioremedi- Frostegård Å, Courtois S, Ramisse V, Clerc S, Bernillon D, Gall F, ation. Appl Biochem Biotechnol 151:638–652 Jeannin P, Nesme X, Simonet P (1999) Quantification of bias related Savela ML, Hynninen EL (2004) Pesticide sales in Finland 2003. Kemia- to the extraction of DNA directly from soils. Appl Environ Kemi 31:57–59 Microbiol 65:5409–5420 Schnüner J, Clarholm M, Rosswall T (1985) Microbial biomass and Ghosh P, Pranab Kumar K, Philip L (2004) Atrazine degradation in activity in an agricultural soil with different organic matter contents. anaerobic environment by a mixed microbial consortium. Water Soil Biol Biochem 17:611–618 Res 38:2277–2284 Smith D, Alvey S, Crowley DE (2005) Cooperative catabolic pathways Ghosh D, Roy K, Williamson KE, White DC, Wommack KE, Sublette within an atrazine-degrading enrichment culture isolated from soil. KL, Radosevich M (2008) Prevalence of lysogeny among soil FEMS Microbiol Ecol 53:265–273 bacteria and presence of 16S rRNA and trzN genes in viral- Strong LC, Rosendahl C, Johnson G, Sadowsky MJ, Wackett LP (2002) community DNA. Appl Environ Microbiol 74:495–502 Arthrobacter aurescens TC1 metabolizes diverse s-triazine ring Hayes TB, Khoury V, Narayan A, Nazir M, Park A, Brown T, Adame L, compounds. Appl Environ Microbiol 68:5973–5980 Chan E, Buchholz D, Stueve T, Gallipeau S (2010) Atrazine induces Vuorimaa P, Kontro M, Rapala J, Gustafsson (2007) Torjunta-aineiden complete feminization and chemical castration in male African esiintyminen pohjavedessä. (Occurrence of pesticides in groundwa- clawed frogs (Xenopus laevis). PNAS 107:4612–4617 ter). The Finnish Environment 42 (2007), Finnish Environment Helling CS, Gish TJ (1986) Soil characteristics affecting pesticide move- Institute, Helsinki (in Finnish) ment into ground water. In: Evaluation of pesticides in ground water, Willems HPL, Lewis KJ, Dyson JS, Lewis FJ (1996) Mineralization of 2, chapter 2. American Chemical Society, pp 14–38 4-D and atrazine in the unsaturated zone of a sandy loam soil. Soil Kanehisa M, Goto S, Sato Y, Furumichi M, Tanabe M (2012) KEGG for Biol Biochem 28:989–996 integration and interpretation of large-scale molecular data sets. Yamazaki K, Fuji K, Iwasaki A, Takagi K, Satsuma K, Harada N, Nucleic Acids Res 40:D109–D114 Uchimura T (2008) Different substrate specificities of two atrazine Langerveld AJ, Celestine R, Zaya R, Mihalko D, Ide CF (2009) Chronic hydrolases (TrzNs) from Nocardioides Species. FEMS Microbiol exposure to high levels of atrazine alters expression of genes that Lett 286:171–177 II

Environmental Pollution 172 (2013) 108e115

Contents lists available at SciVerse ScienceDirect

Environmental Pollution

journal homepage: www.elsevier.com/locate/envpol

Monitoring bioremediation of atrazine in soil microcosms using molecular tools

Sneha Sagarkar a, Shinjini Mukherjee a, Aura Nousiainen b, Katarina Björklöf b, Hemant J. Purohit a, Kirsten S. Jørgensen b, Atya Kapley a,* a Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nehru Marg, Nagpur 440 020, India b Finnish Environment Institute, PO Box 140, FI-00251 Helsinki, Finland article info abstract

Article history: Molecular tools in microbial community analysis give access to information on catabolic potential and Received 14 February 2012 diversity of microbes. Applied in bioremediation, they could provide a new dimension to improve Received in revised form pollution control. This concept has been demonstrated in the study using atrazine as model pollutant. 27 July 2012 Bioremediation of the herbicide, atrazine, was analyzed in microcosm studies by bioaugmentation, Accepted 29 July 2012 biostimulation and natural attenuation. Genes from the atrazine degrading pathway atzA/B/C/D/E/F, trzN, and trzD were monitored during the course of treatment and results demonstrated variation in atzC, trzD Keywords: and trzN genes with time. Change in copy number of trzN gene under different treatment processes was atzA fi trzN demonstrated by real-time PCR. The ampli ed trzN gene was cloned and sequence data showed fi Atrazine homology to genes reported in Arthrobacter and Nocardioides. Results demonstrate that speci c target Bioaugmentation genes can be monitored, quantified and correlated to degradation analysis which would help in pre- Biostimulation dicting the outcome of any bioremediation strategy. Molecular tools Ó 2012 Elsevier Ltd. All rights reserved. Natural attenuation

1. Introduction (Hayes et al., 2003). The European Union has banned atrazine, based on its persistent contamination of groundwater (Prosen and Bioremediation is the green route to remove many pollutants Zupancic-Kralj, 2005) but it is still in use in other countries. The from environment. Despite decades of research, it is always high incidence of atrazine contamination, along with an increasing a challenge to take lab-scale work into field trials. The main concern about the toxicological properties of atrazine, has prompted drawback lies in understanding the capacity of the target soil researchers to seek bioremediation options for atrazine-polluted ecosystem. This study explores the use of molecular tools to sites (Wackett et al., 2002; Lima et al., 2008). understand the catabolic capacity of the target soil and demon- Biodegradation of atrazine has been reported to proceed via strates the advantage of combining conventional bioremediation different pathways, resulting in the formation of metabolites like techniques with molecular tools to monitor bioremediation. Atra- hydroxyatrazine, deethylatrazine, or deisopropylatrazine (Zeng zine has been used as model pollutant in microcosm studies. et al., 2004). All pathways finally lead to the central metabolite, Atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5- cyanuric acid, which is further mineralized to CO2. Genes involved triazine), a broad-leaf weed control herbicide, is popularly used in in the degradative route are indicated in Fig. 1. The atzA, B, C, D, E, F agriculture of sugarcane and corn (Siripattanakul et al., 2009). and trzN, D, F genes encode for enzymes that degrade atrazine by Because of its high mobility in soil, persistence and its massive dehalogenation, wherein atzA/trzN initiate the reaction with the application, atrazine has often been detected in surface and formation of hydroxyatrazine (De Souza et al., 1998; Mulbry et al., groundwater at concentrations well above the permitted limits 2002). Microbial N-dealkylation of the ethyl and isopropyl side (Hayes et al., 2002, 2003). Atrazine can act as an endocrine disrupter, chains by nonspecific monooxygenases produces deethylatrazine affecting the endocrine system, the central nervous system and the and deisopropylatrazine, respectively (Devers et al., 2004). This immune system (Lasserre et al., 2009). Atrazine exposed male frogs oxidative-hydrolytic pathway is characterized by genes thcB and were both demasculinized and completely feminized as adults atrA, reported in Rhodococcus species. Degradation via this route is less common and mostly reported by a consortium rather than individual bacteria (Govantes et al., 2009). Atrazine degradation has * Corresponding author. been well characterized in two strains; Pseudomonas sp. ADP where E-mail addresses: [email protected], [email protected] (A. Kapley). atzA, B, C, D, E, F genes are located on a plasmid (Garcia-Gonzalez

Fig. 1. Atrazine degradation pathway. Various routes for degradation of atrazine indicating genes reported in the pathway. The route for atrazine degradation observed in this study is highlighted in dashed lines. Genes ampliﬁed using speciﬁc PCR primers in this study are indicated in bold type face and are underlined. et al., 2005), and Arthrobacter aurescens TC1 that metabolizes appropriate and limiting nutrient amendments to soils (Getenga, atrazine to cyanuric acid via trzN, atzB, C genes (Sajjaphan et al., 2003; Quietal.,2009). Furthermore, soils have also been stimu- 2004). lated by bioaugmentation, whereby non-indigenous microorgan- Enhancement of atrazine degradation by indigenous soil isms have been amended into soils, to enhance the degradation of bacteria has been demonstrated by biostimulation, viz., applying the pollutant (Jansson et al., 2000; Krutz et al., 2010). 110 S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115

This study demonstrates that molecular tools can add a new 2.4. Rarefaction analysis and diversity index dimension to bioremediation by correlating degradation pattern Rarefaction curve was obtained by applying Estimates-S 8.2 software (http:// with gene profile. The experimental design to evaluate the effi- purl.oclc.org/estimates). The Shannon diversity index (H) was used to characterize ciency of atrazine degradation in contaminated soil was as follows: species diversity in the sample (Kapley et al., 2007). The following link was used for Five microcosms were setup, three, using different methods of the calculation of Shannon diversity index http://math.hws.edu/javamath/ryan/ treatment, viz., (a) natural attenuation; (b) biostimulation; and (c) DiversityTest.html. bioaugmentation and two controls were used, viz., microcosms with (d) sterilized soil and (e) soil without the history of atrazine 2.5. Microcosms assemblement use. Degradation was estimated by monitoring atrazine levels and gene profile was monitored using molecular tools. 1 kg of contaminated composite soil was collected from the same sugarcane field, as described in Section 2.1, sieved through 2 mm sieve into plastic trays, and mixed with commercial grade atrazine at a concentration of 300 mg kg 1 of dry soil. 2. Material and methods This was then separated into three 250 g aliquots in wide mouth jars having a capacity to hold 1 kg (to give 750 mL headspace to carry out aerobic biodegra- 2.1. Soil sampling dation). Microcosms were covered with aluminum foil pierced with holes and kept at room temperature (30C 1 C). Moisture content of the soil was monitored Contaminated soil samples were obtained from a sugarcane field with a three every week and water content was maintained at 40% until the end of the experi- year history of atrazine. The sampling site was located at Yadavmala, Ahmadnagar ment adding sterile distilled water. Additional composite soil from the same field district, Maharashtra, India. Atrazine was sprayed on the fields three months prior to was used as Control 1, and 1 kg soil from a corn field in the same geographical area sample collection. The top 10 cm layer was collected from three sites in the field; was collected to be used in Control 2 microcosm. Five microcosms were set up as each site had three sub-sampling sites from which soil was mixed after removing follows; stones and roots to represent a homogenous sample. Soil samples were transported to the lab and stored at 4 C until used. Control soil was obtained from the same 1. BA: Bioaugmentation, by addition of 105 cfu/g soil, of Pseudomonas citronellolis fi geographical region, from an agricultural eld growing corn without history of ADP (DSM 11735) and Arthrobacter aurescens ATCC-BAA-1386, both strains 1 atrazine use. Soil used in microcosm setup was measured to contain 28.33 g kg of reported for atrazine degradation (Mandelbaum et al., 1995; Shapir et al., 1 total organic carbon and 140 mg kg nitrogen. 2005). Cultures were pre-grown in 0.1 Luria Broth (HiMedia, India) containing 100 mg L 1 atrazine. After 18 h of growth at 30 C, absorbance was measured at A . 2.5 107 cfu (per 250 g soil), was calculated based on 2.2. Soil microbial diversity analysis using culture dependent tools 600 absorbance values and the required amount of culture was harvested by centrifugation. The cell pellet was washed with buffer and re-suspended in 10 g of soil was suspended in 0.1 M sodium phosphate buffer at pH 7 and 5mLof20 medium described in Section 2.2 and was inoculated in the sonicated for 10 min at room temperature with low energy in a water bath, to microcosm. Before harvesting the cells, the presence of all genes listed in disperse the clumps. Various dilutions of the suspension were plated on agar plates 1 1 1 Table 1 were confirmed by PCR to rule out abortion of plasmid. that contained 0.9 g L KH2PO4, 6.5 g L Na2HPO4:12H2O, 0.2 g L MgSO4:7H2O 2. BS: Biostimulation by addition of modified media from Cai et al. (2003).20 (Cai et al., 2003). 100 mg L 1 atrazine (Sigma) was used as both nitrogen and carbon medium was prepared and 5 mL of this was added to the microcosm. source and 2% bacteriological agar was added to solidify the medium. In conditions 3. NA: Natural Attenuation, where only sterilized water was added to maintain where atrazine was used as nitrogen source, 1 g L 1 sodium citrate was used as the water content at 40% (w/w). carbon source. The plates were incubated at 30 C for 3 days and the colonies were 4. Control 1: Sterilized Yadavmala soil (by autoclaving thrice), where atrazine was picked at random. Isolates were sub-cultured three times to ensure that they added at the concentration of 300 mg kg 1 of dry soil and sterilized water was represent a pure culture. RAPD (Random amplified polymorphic DNA) was carried added to maintain water content at 40% (w/w). out to check the genetic variability of isolates. RAPD data was analyzed using Bio- 5. Control 2: 250 g of soil from a corn field in the same geographical area, without numerics software (Applied Maths, St. Martens-Latem, Belgium) (data not shown). history of atrazine use, was mixed with same amount of atrazine as for the 18 isolates were selected from plates that contained atrazine as both nitrogen and other microcosms and where only sterilized water was added to maintain carbon source and 10 from plates that contained atrazine as nitrogen source. These water content at 40% (w/w). isolates were grown in respective atrazine medium broth with 100 mg L 1 atrazine at 30 C. Total DNA was prepared by a cell lysis protocol and 16S rDNA was amplified by PCR using universal 16S rDNA primers, 27F and 1492R (Kapley et al., 2007). In 10 g portions of microcosm soil were removed periodically for chemical or brief, PCR reactions were performed under the following conditions: 30 cycles of molecular analysis for three weeks. All treatments were carried out in dupli- denaturation at 94 C for 1 min, annealing at 55 C for 1 min and extension at 72 C cates. For molecular analysis, extracted DNA samples were stored at 20 Cuntil for 1 min. 1466 bp PCR products were gel-purified using the gel extraction kit from used. Qiagen and sequenced from the 50 end. Samples were identified through BLAST analysis of the partial sequence data. 2.6. Nucleic acid extraction from soil and control strains

2.3. Soil microbial diversity analysis using culture independent tools Control strains Pseudomonas citronellolis and Arthrobacter aurescens were obtained from culture collections. Both bacteria were cultivated on minimal medium Metagenome of soil described in Section 2.1, was used as a template to amplify adapted from Cai et al. (2003) as given in Section 2.2. Bacteria were grown in liquid 16S rDNA. Amplified products were resolved on a 1% agarose gel and the amplicon media. Total genomic DNA was extracted (Kapley et al., 2007), 50 mL aliquots were was gel-purified using the gel extraction kit from Qiagen and cloned into the p-Drive stored at 20 C until required. cloning vector (Qiagen PCR Cloning Plus Kit). Recombinant clones were initially DNA was prepared from 100 mg soil in 5 replicates using a Fast Spin Kit for soil screened by blueewhite colony selection on LB ampicillin plates. Transformants DNA extraction (M.P. Biomedicals, France). DNA from all replicates was pooled to were screened for the presence of insert by EcoRI digestion. 16S rDNA gene from represent a homogeneous mixture. 100 mL aliquots were stored at 20 C until these transformants were amplified and used for digesting with HaeIII and AluI required. restriction enzymes in an ARDRA (amplified ribosomal DNA restriction analysis). Based on the digestion pattern (data not shown), 31 clones out of 100 were sequenced and identified by analysis of partial sequence data of the 16S rDNA insert. 2.7. Monitoring atrazine degradation genes in soil microcosms Sequencing was carried out using the T7 or SP6 primer. All sequences reported in the study have been deposited in NCBI GenBank. Accession numbers HM196283e Genes from the atrazine degradation pathway were monitored in soil micro- HM196305 represent 16S rDNA sequence of cultured isolates and Accession cosms by PCR. Table 1 lists the genes monitored in this study, their expected numbers HM196306eHM196335 represent 16S rDNA sequences of clones. Corre- amplified product size, annealing temperature, source of specific primers for atzA, B, sponding sequences can be viewed at http://www.ncbi.nlm.nih.gov/. C, D, E, F, trzN and trzD. Thermocycling steps and primer details are listed in A phylogenetic tree was constructed on the basis of 16S rDNA gene sequences Supplementary data, Table T1. Amplification products were analyzed by electro- using Clustal-X software version 2 (Thompson et al., 1997). First, a pair-wise align- phoresis in a 1% (w/v) agarose gel and stained with ethidium bromide. Amplicon ment was carried out using the multiple sequence alignment tool, with variables, sizes were determined by comparison with a DNA ladder (1 kb, Invitrogen, USA). PCR 100% gap penalty and 20% gap cost. Subsequently, global alignment homology was carried out in three replicates for each gene target. Pseudomonas citronellolis calculation was carried out, resulting in a distance matrix. This matrix was used for was used as positive control for atzA, B, C, D, E, F and Arthrobacter aurescens for trzN. the dendrogram construction using the neighbor joining method. Bootstrap analysis There was no positive control for trzD, the amplification product was confirmed by was carried out for 1000 iterations. analyzing its sequence data. S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115 111

Table 1 Target genes, expected ampliﬁcation product size and detection of atrazine degrading genes by PCR in microcosms with different bioremediation treatments.

Sr. no Target Gene coding for Expected size Presence of degradation Presence of degradation Presence of degradation References gene (base pairs) genes at 0 ha genes after 1st weeka genes after 2nd weeka

NA BS BA C1 C2 NA BS BA C1 C2 NA BS BA C1 C2 1 atzA Atrazine chlorohydrolase 204 nd nd þ nd nd nd nd þ nd nd nd nd þ nd nd Devers et al. (2004) 2 atzB Hydroxydechloroatrazine 523 nd nd þ nd nd nd nd þ nd nd nd nd þ nd nd De Souza et al. (1998) ethylaminohydrolase 3 atzC N-Isopropylammelide 227 nd nd þ nd nd nd þþnd nd þþþnd nd Devers et al. (2004) isopropylaminohydrolase 4 atzD Cyanuric acid 203 nd nd þ nd nd nd nd þ nd nd nd nd þ nd nd Devers et al. (2004) amidohydrolase 5 atzE Biuret amidohydrolase 202 nd nd þ nd nd nd nd þ nd nd nd nd þ nd nd Devers et al. (2004) 6 atzF Allophanate hydrolase 233 nd nd þ nd nd nd nd þ nd nd nd nd þ nd nd Devers et al. (2004) 7 trzN Triazine hydrolase 430 nd nd þ nd nd þþþnd nd þþþnd nd Mulbry et al. (2002) 8 trzD Cyanuric acid 663 nd nd nd nd nd nd nd nd nd nd nd þ nd nd nd Fruchey et al. (2003) amidohydrolase

Conventional PCR protocol used for detection of genes listed in Table 1. Thermo cycling steps and primer details are listed in supplementary data, Table T1. NA e natural attenuation; BS e biostimulation; BA e bioaugmentation; C1 e control (sterilized soil); C2 e control (soil without history of atrazine use). Positive controls used in the study: P. citronellolis e atzA, B, C, D, E, F; A. aurescens e trzN; No control for trzD, conﬁrmed by sequencing. a þ e detected; nd e not detected.

2.8. Preparation of qPCR standards and quantiﬁcation methodology 2.10. GC-FID analysis of atrazine

Quantification of trzN gene was carried out by a protocol modified from the Atrazine was extracted from soil using dichloromethane (3 replicates), evapo- Plasmid Standard Construction instructions by Applied Biosystems, available online rated to dryness and re-suspended in methanol before analyzing by GC as described at http://www.appliedbiosystems.com/support/tutorials/pdf/quant_pcr.pdf. by Wang et al. (2009). Gas chromatographic conditions were as follows; GC-14C Gas trzN gene was amplified from standard culture Arthrobacter aurescens. The PCR chromatograph, FID detector and N2000 Chromatography Workstation; using product was checked for correct size on 1% agarose gel and was cloned in pdrive- a capillary column (30 m 0.53 mm i.d.) with the interior coated with 14% OV-1701. vector (Qiagen PCR Cloning Plus Kit) and cells were transformed according to the The operating conditions used were: Temperature conditions: Injection port at manufacturer’s instructions. Random 5 clones were selected and sequenced to 250 C, and column was at 200 C, while the detector was at 250 C. Gas pressures confirm the presence of trzN gene in the insert. All clones demonstrated the were e nitrogen: 50 kPa, hydrogen: 50 kPa, air: 50 kPa. Atrazine, Deethylatrazine presence of correct insert (data not shown) and hence, one positive clone was and Deisoproply atrazine from Sigma (USA) were used as standards. Concentration randomly selected as ‘standard’. Positive clone was grown overnight in LB media of atrazine in the samples was calculated as described by Wang et al. (2009). with ampicillin at 37 C. Plasmid purification was carried out using Plasmid Min- Atrazine degradation was represented in Fig. 2 using SigmaPlot version 11.2 iprep Kit (Qiagen, Germany). Plasmid DNA was quantified (Nanodrop 1000, Thermo (Systat Software Inc.), which was also used to perform the statistical analysis and Scientific, USA) and mass of a single plasmid was calculated using the following analysis of variance (ANOVA) to determining statistical significance of the formula; treatments.

ð Þ¼ ð Þ* : * ð = Þ m g plasmid size bp 1 096 10 21 g bp 3. Results Derivation of DNA Mass Formula 3.1. Bacterial diversity in agricultural soil

Soil sample collected from the sugarcane ﬁeld, was analyzed for bacterial diversity, before setting up the microcosms. Culture-

Based on the results, a dilution series was prepared containing 101e105 copies of the gene and a standard curve was made. Real time PCR was performed with these standard plasmids and experimental samples (Peng et al., 2010). qPCR reaction was carried out in iCycler (Bio-Rad, USA) using SYBR Green chemistry. Each 25 ml reaction constituted of 50 pmol of each primer, 12.5 mlof Ò MaximaÔ SYBR Green qPCR Master Mix (biorad) which includes SYBR Green I dye, dNTP, KCl, (NH4)2SO4 and MaximaÔ Hot Start Taq DNA Polymerase. Thermocycling conditions were same as regular PCR.

2.9. Conﬁrmation of trzN gene in microcosms

trzN was amplified by PCR from the metagenome of first week microcosms BA, BS and NA. Amplified gene products were gel purified (QIAquick Gel Extraction Kit, Qiagen, Germany) and cloned (Qiagen PCR Cloning Plus Kit, Qiagen, Germany). Six to eight clones from each microcosm were selected for sequencing the insert, based on random selection. Sequence data was analyzed by BLAST (NCBI). DNA nucleotide sequences were translated to their amino acid sequences using the ExPASy Translate Tool (http://us.expasy.org/tools/dna.html) to check homology of experimental data to proteins reported. Amino acid sequences were aligned with published trzN Fig. 2. Atrazine degradation profile in soil microcosms. Microcosms have been defined sequences using the multiple sequence alignment function in Clustal W and viewed by the following abbreviations; BA, for bioaugmentation; BS, for biostimulation; NA, using Jal view (Verma et al., 2011). A phylogenetic tree was constructed using for natural attenuation; C1, for Control 1, containing sterilized soil and C2, for Control 2, experimental and reported trzN protein sequences. containing soil without any history of atrazine use. 112 S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115 dependent and culture-independent tools were used, where microcosm and 16% under natural attenuation. Both controls did culture-dependent analysis only focused on bacteria that grew on not show any change in residual atrazine after one week. Results of atrazine plates. Table 2 reports the diversity analysis, wherein, 18 the analysis at second week showed 98% removal in BA microcosm, genera represented by 7 phyla were detected. 97% removal in BS microcosm and 48% in NA. The sterilized control Majority of the cultivable isolates were classified into two phyla; (C1) did not show any utilization of atrazine, while the control Proteobacteria and Actinobacteria. Beta-proteobacteria was the using soil without the history of atrazine use demonstrated 5% dominant class represented by 46% of the isolates, while Gamma- atrazine utilization. For all the treatments, one way ANOVA was proteobacteria was represented by 12% and Alpha-proteobacteria performed, P value <0.001 showed that there is a statistically by 4%. The phylum Actinobacteria was represented by 27% and significant difference amongst the treatment groups. Firmicutes were represented by 8% of analyzed cultivable diversity. Atrazine degradation intermediates deethylatrazine and deiso- Five phyla were detected using culture-independent tools, viz., propylatrazine were also analyzed by GC-FID, however, were not Proteobacteria, Actinobacteria, Verrucomicrobia, Firmicutes and detected in any microcosm (data not shown). Chloroflexi. Supplementary Fig. S1 demonstrates the bacterial diversity and linkage between groups. 3.3. Monitoring atrazine degrading genes Shannon’s index was used to compare the diversity of the community estimated by each method. Shannon’s diversity index 0 Table 1 presents a summary of genes detected at different time for cultivable bacteria was calculated to be H ¼ 2.33 with 0 points in the course of the study. Initially, atzA, B, C, D, E, F and H max ¼ 3.25 and divergence from equiprobability, D1 ¼ 0.67, and trzN were detected only from the bioaugmented microcosm. This evenness was calculated as 0.776. For culture-independent anal- 0 0 is not unexpected since the microcosm has been seeded with ysis, diversity index H was calculated to be 3.54 with H max ¼ 3.81 105 cfu of Pseudomonas and Arthrobacter cultures reported for and D1 ¼ 0.27 and with evenness ¼ 0.928. A low D1 value of culture atrazine degradation via atzA, B, C, D, E, F and trzN pathways independent study indicates a high degree of diversity. Conversely, respectively. It is assumed that all other desired genes were below a high D1 value of culture-dependent analysis indicates that the detection limits by conventional PCR and atrazine stress would system has diverged substantially from equiprobability and is not initiate the increase in copy number of relevant microbial very diverse. These results are not unexpected since it is widely community to eventually bring about degradation of the pollutant. known that only 1e5% of bacteria are cultivable. The Rarefaction After the first week, when BA, BS and NA samples demonstrated curve indicates that a reasonable number of clones from the library 92%, 52% and 16% removal of atrazine, triazine hydrolase (trzN) were sampled in the culture independent diversity analysis gene, was detected in all three microcosms. N-isopropylammelide (Supplementary Fig. S2). isopropylaminohydrolase gene (atzC) was amplified only from BA and BS microcosms of the first week, but could be amplified even 3.2. Comparison of treatment strategies in bioremediation of in NA in the second week. In the second week of the study, the atrazine gene for cyanuric acid amidohydrolase (trzD) was observed in the BS microcosm. The control samples did not show amplification of The degradation pattern in different microcosms during two any gene. weeks of investigation (Fig. 2) indicated that removal of pollutant Since the trzN gene appeared to be the most dominant gene, was fastest in the case of bioaugmentation, where 92% atrazine was appearing in first week samples of all three treatments, real-time removed in the first week as compared to 52% in the biostimulated assay was carried out to explore its change in copy number at

Table 2 Analysis of bacteria in agricultural ﬁeld soil that was used for the microcosms. Bacterial 16S rDNA was used both for culture-dependent and culture-independent analysis.

Phylum/class Genus/family Organism Number of representatives on the basis of

Culture Cloning Alpha-proteobacteria Brucella Brucella sp. 1 Uncultured alpha-proteobacterium e 5 Beta-proteobacteria Alcaligenes Achromobacter sp. 11 Burkholderiaceae Aquabacterium sp. 1 Burkholderiaceae Burkholderiaceae sp. 1 Unidentified Burkholderia e 1 Uncultured beta-proteobacterium e 4 Gammaproteobacteria Pseudomonadaceae Pseudomonas sp. 1 3 Moraxellaceae Acinetobacter sp. 1 Xanthomonadaceae Stenotrophomonas maltophilia 2 Uncultured Xanthomonadaceae e 1 Uncultured Gammaproteobacteria e 1 Actinobacteria Nocardioidaceae Arthrobacter sp. 7 Nocardiaoides sp. 1 Uncultured Actinobacteria e 1 Verrucomicrobia Uncultured Verrucomicrobiaceae e 2 Firmicutes Bacillus Bacillus megaterium 1 Bacillus pichinotyi 1 Bacillus licheniforms 1 Paenibacillus sp. 1 Chloroflexi Uncultured Chloroflexi bacterium e 3 Uncultured acidobacteria bacterium e 4 Uncultured bacterium ee 1 Unidentified bacterium ee1 Total 26 31 S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115 113

Table 3 same gene reported in literature. Three clones (IF-BA6, IF-N1 and Quantiﬁcation of trzN gene in microcosms during bioremediation process. IF-N3) showed only 95e97% similarity. Details of homology can be Copy no. of trzN gene g 1 of soil viewed in Supplementary Table T2. All clone sequence shave been

Bioaugmentation Biostimulation Natural Control 1 Control 2 deposited at NCBI GenBank. attenuation Sequence data was translated and protein diversity of trzNwas 1st 5.7 105 7.56 104 1.26 104 NA NA also studied. The phylogenetic tree demonstrated that the protein week (0.82) (0.53) (1.4) is highly conserved (Fig. 3). Two outgroups were used in tree 2nd 3.75 105 6.88 104 2.22 104 NA 1.09 103 construction, AtzA; atrazine cholorohydrolase [P72156] to repre- week ( 0.99) ( 1.1) ( 0.51) ( 0.46) sent functional similarity to (TrzN) and Cytosine deaminase-like NA e not detected; Control 1 e microcosm with sterilized soil; Control 2 e metal dependent hydrolase http://www.ncbi.nlm.nih.gov/protein/ microcosm of soil without history of atrazine usage; Figures in the bracket represent ZP_09653040 [ZP 09653040] based on protein database of Clus- standard deviation. ters of Orthologous Groups (COGs). Fig. 3 demonstrates that the trzN clones generated in this study are highly homologous yet can different stages of degradation. Results indicated that the highest be broadly divided into three subgroups. Furthermore, Translated number of this target gene was observed in metagenome of the data, aligned using Clustal W demonstrated that, the proteins fi bioaugmented microcosm in the rst week, followed by bio- were mainly homologous. Two reference proteins for triazine stimulation and natural attenuation respectively. But in the second hydrolase were used in this alignment; TrzN translated protein week gene copy number of BA and BS microcosm decreased and from Nocardioides sp. DN36 (accession no. AB539567) and TrzN marginally increased in NA microcosm (Table 3). protein of Arthrobacter aurescens TC1 (accession no. NC_008712), protein id ¼ “YP_949961”“http://www.ncbi.nlm.nih.gov/protein/ 3.4. Confirmation of the trzN gene by sequence analysis YP_949961.1”. As can be seen in Supplementary Fig. S3,only five clones (IF-N1, IF-N2, IF-N4, IF-N7 and IF-BA6) showed the The trzN gene was amplified and cloned from all three treat- presence of Glutamic Acid (E) residue at the 55th position, like the ment microcosms, BA, BS and NA. Sequence data of clones was reference strains. Others showed mainly Glutamine (Q) at the compared to five reported trzN genes; Arthrobacter aurescens TC1 55th position. Clone IF-BA2 showed presence of glycine (G) in (NC_008712), Arthrobacter sp. AD26 (EU091479), Pseudomonas sp. place of arginine (R) at 97th residue which is a substrate binding AD39 (FJ161692), Nocardioides sp. DN36 (AB539567), and Nocar- pocket. This could suggest slight variation in function. The dioides sp. AN3 (AB427184). 99e100% similarity index indicates the alignment, visualized using Jal View program, is represented as high homology of trzN gene amplified from this soil sample to the Supplementary Fig. S3.

Fig. 3. Phylogenetic tree showing the relationship between reported TrzN proteins with clones from this study. Sequence data of clones has been translated and used in tree construction. The tree was rooted with atrazine cholorohydrolase [P72156] AtzA which has similar function to triazine hydrolase (trzN) and Cytosine deaminase-like metal dependent hydrolase http://www.ncbi.nlm.nih.gov/protein/ZP_09653040 [ZP 09653040] (based on protein database of Clusters of Orthologous groups (COGs)). Translated TrzN proteins from this study are indicated by bold type face. Control sequences downloaded from GenBank and used for tree construction are indicated by normal type face. Accession number is given in square brackets. Bootstrap values obtained after 1000 iterations are indicated at the nodes. Bar reﬂects 0.1 sequence distance. 114 S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115

4. Discussion degradation of atrazine in NA microcosm, in the first week of sampling, indicating the efficient natural catabolic capacity of the Analyzing the microbial capacity of inherent soil microbes plays soil. However, further degradation of atrazine in NA microcosm was a key role in selecting the bioremediation strategy. This study slow and no detectable change was observed in the third week (data demonstrates the importance of such an analysis using atrazine as not shown). These results indicate that even if the soil has a natural a model pollutant and soil from an agricultural field that uses atrazine, microbial population that is capable of atrazine degradation, bio- as target ecosystem. The functional capacityof the soil for degradation augmentation and biostimulation enhance this process 5e6 times. of any pollutant would be decided by the gene pool present. Hence, Both processes were aided by additional input of carbon source in key genes related to atrazine degradation were monitored in all the form of citrate at a concentration of 0.4 mg g 1 soil. The role of microcosms. Fig. 1 indicates the different routes reported in atrazine citrate as an additional carbon source in bioaugmentation studies degradation. In this study, we could not detect the degradative was demonstrated by Silva et al. (2004), who used 4e40 mg g 1 soil intermediates, deethylatrazine and deisopropylatrazine, nor could to enhance degradation. Extrapolating these concentrations into we amplify the thc genes in any of the microcosms or bacterial isolates field trials would not be practical, and hence, we reduced citrate (data not shown). Hence, we assume that the degradative route fol- concentration to 0.4 mg g 1 soil. However, many other parameters, lowed in the soil microcosms is via the hydroxyatrazine pathway, including addition of more time points for analysis, would need to which was corroborated by the detection of the target genes in the be studied to take lab-scale work into the field. Our results show pathway. The detection of trzNandatzC in all three microcosms is very little degradation in soil without any history of atrazine use. indicative of the fact that atrazine is being degraded via the hydrox- This highlights an important aspect in choosing the best bioreme- yatrazine pathway (trzN) to cyanuric acid (atzC) then to biuret and diation strategy: thorough knowledge of target site is needed and CO2 (trzD). Results in Table 1, demonstrated that atzA gene was only natural attenuation may not be the answer in all cases. detected in the case of bioaugmentation, while trzN could be detected This study highlights the importance of use of molecular tools, in BA, BS and NA microcosms, suggesting the dominant pathway in where details of expression of degradative capacity are monitored soil. Reports also suggest the prevalence of trzN gene over atzA(Singh during bioremediation. et al., 2004; Devers et al., 2007; Vaishampayan et al., 2007; Arbeli and Fuentes, 2010). trzD gene was only observed under biostimulation, 5. Conclusions after two weeks of study, while atzC could be detected in NA, BS and BA microcosms in the second week of study. Although not conclusive, Analysis of target genes indicated that the atrazine degradative the results suggest that nutritional amendments in soil enable the route followed in this soil was via the hydroxyatrazine intermediate. atrazine degraders to proliferate. As reported in literature, target trzN gene profiles were demonstrated to correlate with degradative genes detected in this study, suggest the dominance of Arthrobacter efficiencies. Bioaugmentation demonstrated the highest efficiency strain over Pseudomonas in Indian soil (Singh et al., 2004; of pollutant removal followed by biostimulation. Natural attenua- Vaishampayan et al., 2007). tion can only be considered as a bioremediation option, if the soil has Since the triazine hydrolase gene was detected in all three a history of pollutant use. This study highlights the importance of treatments, we attempted to understand if the gene profile was understanding the target soil biology before embarking on a biore- correlated to degradative efficiency. Real-time analysis of the mediation program. The uses of molecular tools make it easier to metagenome of BS, BA and NA microcosms demonstrated different understand the target soil ecosystem (Kapley and Purohit, 2009); levels of this gene that could be correlated to degradation effi- catabolic capacity can be monitored by detection of target genes, ciency. The first and second week samples showed the highest copy copy number statistics can predict treatment efficiency and a gene number in the BA microcosms, which also demonstrated the based monitoring tool can be developed to track the bioremediation highest degradation efficiency. Real time PCR data (Table 3) process. In relation to the ever-increasing load of pesticides entering demonstrated decreasing order of presence of trzN gene in the first into the environment and influencing the ecosystems, it becomes an week samples: BA > BS > NA. Results of degradation (Fig. 2) also important task for the biologist to improve clean-up processes as indicate the same order: BA > BS > NA. Real-time analysis of second well as monitoring systems. week samples, however, shows very small decrease in trzN gene in BA and BS, while a marginal increase was observed in NA micro- Acknowledgments cosm. Probably, this is so, because the BA and BS microcosms have already degraded the pollutant and atrazine levels do not instigate Funding from Department of Biotechnology, Ministry of Science the growth of target bacteria. Levels of atrazine in the NA micro- and Technology, New Delhi, to Dr. Atya Kapley for the Indo-Finland cosm ensure presence of target gene. project (G-1649) and from the Academy of Finland to Dr. Kirsten Sequence analysis of trzN gene clones from all three treatments, Jørgensen (project code 124822) are acknowledged. Sneha Sagarkar BA, BS and NA microcosms demonstrated very high homology to is grateful to the Council of Scientific and Industrial Research (CSIR), reported triazine hydrolases of cultured isolates in GenBank. India for the award of junior research fellowship. Protein BLAST demonstrated the presence of a functional segment. Protein homology was very high, but minor changes in the amino Appendix A. Supplementary data acid sequences were observed, which indicates that the enzyme activity obtained from these uncultured clones could be slightly Supplementary data related to this article can be found at http:// different (Sarkar et al., 1986). Further studies will have to be carried dx.doi.org/10.1016/j.envpol.2012.07.048. out in future to validate the role of amino acid replacement. To mimic conditions of a natural soil, the microcosm soil was References collected from a sugarcane field using atrazine for weed control. Before setting up the microcosms, diversity of the soil was analyzed. Arbeli, Z., Fuentes, C., 2010. Prevalence of the gene trzN and biogeographic patterns Taxonomic data (Table 2) indicates the presence of bacteria like among atrazine-degrading bacteria isolated from 13 Colombian agricultural Achromobacter, Arthrobacter, Pseudomonas, Nocardioides, etc., soils. FEMS Microbiology Ecology 73, 611e623. Cai, B., Han, Y., Liu, B., Ren, Y., Jiang, S., 2003. Isolation and characterization of an that have been reported to contain atrazine degradative genes atrazine-degrading bacterium from industrial wastewater in China. Letters in (Devers et al., 2007; Krutz et al., 2010). Fig. 2 demonstrates 16% Applied Microbiology 36, 272e276. S. Sagarkar et al. / Environmental Pollution 172 (2013) 108e115 115

De Souza, M.L., Wackett, L.P., Sadowski, M.J., 1998. The atzABC genes encoding Mulbry, W.W., Zhu, H., Nour, S.M., Topp, E., 2002. The triazine hydrolase gene trzN atrazine catabolism are located on a self transmissible plasmid in Pseudomonas from Nocardioides sp. strain C190: cloning and construction of gene-specific sp. strain ADP. Applied and Environmental Microbiology 64, 2323e2326. primers. FEMS Microbiology Letters 206, 75e79. Devers, M., Soulas, G., Martin-Laurent, F., 2004. Real-time reverse transcription PCR Peng, J.J., Cai, C., Qiao, M., Li, M., Zhu, Y.G., 2010. Dynamic changes in functional gene analysis of expression of atrazine catabolism genes in two bacterial strains copy numbers and microbial communities during degradation of pyrene in isolated from soil. Journal of Microbiological Methods 56, 1e3. soils. Environmental Pollution 158, 2872e2879. Devers, M., Azhari, N.El., Kolic, N.U., Martin-Laurent, F., 2007. Detection and orga- Prosen, H., Zupancic-Kralj, L., 2005. Evaluation of photolysis and hydrolysis of nization of atrazine-degrading genetic potential of seventeen bacterial isolates atrazine and its first degradation products in the presence of humic acids. belonging to divergent taxa indicate a recent common origin of their catabolic Environmental Pollution 133, 517e529. functions. FEMS Microbiology Letters 273, 78e86. Qui, Y., Pang, H., Zhou, Z., Zhang, P., Feng, Y., Sheng, D.G., 2009. Competitive Fruchey, I., Shapir, N., Sadowsky, M.J., Wackett, L.P., 2003. On the origins of cya- biodegradation of dichlobenil and atrazine coexisting in soil amended with nuric acid hydrolase: purification, substrates, and prevalence of AtzD from a char and citrate. Environmental Pollution 157, 2964e2969. Pseudomonas sp. strain ADP. Applied and Environmental Microbiology 69, Sajjaphan, K., Shapir, N., Wackett, L.P., Palmer, M., Blackmon, B., Tomkins, J., et al., 3653e3657. 2004. Arthrobacter aurescens TC1 atrazine catabolism genes trzN, atzB, and atzC Garcia-Gonzalez, V., Govantes, F., Porrua, O., Santero, E., 2005. Regulation of the are linked on a 160-kilobase region and are functional in Escherichia coli. Pseudomonas sp. strain ADP cyanuric acid degradation operon. Journal of Applied and Environmental Microbiology 70, 4402e4407. Bacteriology 187, 155e167. Sarkar, H.K., Viitanen, P.V., Poonian, M.S., Kaback, H.R., 1986. Substitution of Getenga, Z.M., 2003. Enhanced mineralization of atrazine in compost amended soil glutamine-60 with glutamic acid causes the lac permease of Escherichia coli to in laboratory studies. Bulletin of Environmental Contamination and Toxicology become temperature sensitive. Biochemistry 25, 2778e2781. 75, 937e941. Shapir, N., Rosendahl, C., Johnson, G., Andreina, M., Sadowsky, M.J., Wackett, L.P., Govantes, F., Porrúa, O., García-González, V., Santero, E., 2009. Atrazine biodegra- 2005. Substrate specificity and colorimetric assay for recombinant TrzN derived dation in the lab and in the field: enzymatic activities and gene regulation. from Arthrobacter aurescens TC1. Applied and Environmental Microbiology 71, Microbial Biotechnology 2, 178e185. 2214e2220. Hayes, T., Haston, K., Tsui, M., Hoang, A., Haeffele, C., Vonk, A., 2002. Herbicides: Silva, E., Fialho, A.M., Sa-Correia, I., Burns, R.G., Shaw, L.J., 2004. Combined bio- feminization of male frogs in the wild. Nature 419, 895e896. augmentation and biostimulation to clean up soil contaminated with high Hayes, T., Haston, K., Tsui, M., Hoang, A., Haeffele, C., Vonk, A., 2003. Atrazine- concentrations of atrazine. Environmental Science and Technology 38, 632e637. induced hermaphroditism at 0.1 PPB in American leopard frogs (Rana Singh, P., Suri, C.R., Cameotra, S.S., 2004. Isolation of a member of Acinetobacter pipiens): laboratory and field evidence. Environmental Health Perspectives species involved in atrazine degradation. Biochemical and Biophysical Research 111, 568e575. Communications 317, 697e702. Jansson, J.K., Björklöf, K., Elvang, A.M., Jørgensen, K.S., 2000. Biomarkers for moni- Siripattanakul, S., Wirojanagud, W., McEvoy, J., Limpiyakorn, T., Khan, E., 2009. toring efficacy of bioremediation by microbial inoculants. Environmental Atrazine degradation by stable mixed cultures enriched from agricultural soil Pollution 107, 217e223. and their characterization. Journal of Applied Microbiology 106, 986e992. Kapley, A., Baere, T., Purohit, H.J., 2007. Eubacterial diversity of activated biomass from Thompson, J., Gibson, T., Plewniak, F., Jeanmougin, F., Higgins, D., 1997. The CLUSTAL a common effluent treatment plant. Research in Microbiology 158, 494e500. X windows interface: flexible strategies for multiple sequence alignment aided Kapley, A., Purohit, H.J., 2009. Genomic tools in bioremediation. Indian Journal of by quality analysis tools. Nucleic Acids Research 25, 4876e4882. Microbiology 49 (2), 108e113. Vaishampayan, P.A., Kanekar, P.P., Dhakephalkar, P.K., 2007. Isolation and character- Krutz, L., Shaner, D.L., Weaver, M.A., Webb, R., Zablotowicz, R., Reddy, K., Huanga, Y., ization of Arthrobacter sp. strain MCM B-436, an atrazine-degrading bacterium, from Thomsona, S., 2010. Agronomic and environmental implications of enhanced s- rhizospheric soil. Journal of Biodeterioration and Biodegradation 60, 273e278. triazine degradation. Pesticide Management Science 66, 461e481. Verma, V., Raju, S.C., Kapley, A., Kalia, V.C., Kanade, G.S., Daginawala, H.F., Purohit, H.J., Lasserre, J., Fack, F., Revets, D., Planchon, S., Renaut, J., Hoffmann, L., Gutleb, A.C., 2011. Degradative potential of Stenotrophomonas strain HPC383 having genes Muller, C.P., Bohn, T., 2009. Effects of the endocrine disruptors atrazine and PCB 153 homologous to dmp operon. Bioresource Technology 102, 3227e3233. on the protein expression of MCF-7 human cells. Journal of Proteome Research 8, Wackett, L.P., Sadowsky, M.J., Martinez, B., Shapir, N., 2002. Biodegradation of 5485e5496. atrazine and related s-triazine compounds: from enzymes to field studies. Lima, D., Viana, P., André, S., Chelinho, S., Costa, C., Ribeiro, R., Sousa, J.P., Applied Microbiology and Biotechnology 58, 39e45. Fialho, A.M., Viegas, C.A., 2008. Evaluating a bioremediation tool for atrazine Wang, R., Jiang, Z., Ma, Y., Diao, Z., Zhang, Y., 2009. Research on how to determine contaminated soils in open soil microcosms: the effectiveness of bio- the residue of atrazine in the soil. Bioinformatics and Biomedical Engineering. augmentation and biostimulation approaches. Chemosphere 74, 187e192. In: ICBBE 2009, 3rd International Conference, http://dx.doi.org/10.1109/ICBBE. Mandelbaum, R.T., Allan, L.D., Wackett, L.P., 1995. Isolation and characterization of 2009.5163332. a Pseudomonas sp. that mineralizes the s-triazine herbicide atrazine. Applied Zeng, Y., Sweeney, C.L., Stephens, S., Kotharu, P., Wackett, L.P., 2004. Atrazine and Environmental Microbiology 61, 1451e 1457. Pathway Map, Biodegradation Database. III

Bioremediation strategies for removal of residual atrazine in the boreal groundwater zone

Aura O. Nousiainena, Katarina Björklöfa, Sneha Sagarkarb, Jeppe Lund Nielsenc, Atya Kapleyb, Kirsten S. Jørgensena a Finnish Environment Institute, P.O. Box 140, FI-00251, Helsinki, Finland b Environmental Genomics Division, National Environmental Engineering Research Institute, Nehru Marg, Nagpur 440 020, India c Department of Biotechnology, Chemistry and Environmental Engineering, Sohngaardsholmsvej 49, Aalborg University, 9000 Aalborg, Denmark

Corresponding Author: Aura Nousiainen Finnish Environment Institute P.O. Box 140 FIN-00251 Helsinki, Finland Tel. +358 400 148 734; Fax: +358 9 465 913; E-mail address: [email protected] 1

Abstract

Strategies for bioremediation of atrazine, a pesticide commonly polluting groundwater in low concentrations, were studied in two boreal nonagricultural soils. Atrazine did not mineralize in soil without bioremediation treatments. In biostimulation treatment with molasses, up to 52% of atrazine was mineralized at 10 °C, even though the degradation gene copy numbers did not increase, indicating that atrazine degradation might not solely be facilitated by atzA/trzN–atzB genes. In combined biostimulation treatment using citrate or molasses and augmentation with Pseudomonas citronellolis ADP or Arthrobacter aurescens, up to 76% of atrazine was mineralized at 30 °C, and the atrazine degradation gene numbers increased up to 107 copies g-1 soil. Clone libraries from passive samplers in groundwater monitoring wells revealed the presence of phylogenetic groups formerly shown to include atrazine degraders, and the presence of atrazine degradation genes atzA and atzB. Microautoradiography revealed that bioremediation strategies allowed increasing the proportion of active atrazine degraders from 0.3% up to 1.9% of the total bacterial count. These results show that the mineralization of low concentrations of atrazine in the groundwater zone at low temperatures is possible by bioremediation treatments.

Keywords: Atrazine degradation, bioremediation, quantitative PCR, microautoradiography, 14C-mineralization 3

Introduction

Intensive use of pesticides has increased the amount of accidental exposure of these harmful substances to the ecosystem. Although many soil bacteria have potential for herbicide biodegradation, several herbicides are frequently detected in soil and groundwater. Atrazine (1-Chloro-3-ethylamino-5-isopropylamino-2,4,6-triazine) was banned by the European Union a decade ago due to its recalcitrance, but despite the concerns of environmental impacts and possible risks to human health, it is still one of the most widely used herbicide against broad leaf weeds worldwide today.

Although atrazine is a recalcitrant compound, it is susceptible to microbial degradation in topsoil (Aislabie et al. 2004; Nousiainen et al. 2014; Sagarkar et al. 2013; Sagarkar et al. 2014). Bacteria can degrade atrazine by two different pathways, initiated by dealkylation and dechlorination. The latter route is well described, and catalyzed by the enzymes encoded by the atzABCDEF genes (De Souza 1996). Instead of atzA, the functional homolog, trzN gene, may be more commonly observed in the environment (Arbeli and Fuentes 2010). An alternative atrazine degradation pathway, initiated by a dealkylation reaction, is carried out by a P-450 cytochrome system in certain strains of Rhodococcus species (Nagy et al. 1995). This reaction is not specific for triazine breakdown, and hence the presence of this pathway cannot directly be linked with atrazine degradation by genomic tools.

Even high concentrations of atrazine can be degraded in topsoil (Wang et al. 2013), but often it partly leaches into the subsoil, where its degradation is considerably slower (Accinelli et al. 2001). Atrazine removal is economically and technically most difficult to achieve when it is present in subsoil, and at low concentration. Unfortunately, this is the most common type of atrazine pollution. A survey of pesticides in Finnish groundwater revealed atrazine concentrations up to 0.34 µg L-1 (Vuorimaa et al. 2007). Similar low values are common in groundwaters globally (Hallberg 1989; Jurado et al. 2012; Stuart et al. 2012). As the European Union has set the maximum limit for pesticides in drinking water to 0.1 µg L-1, even a trace amount of atrazine can cause the groundwater to be unfit for drinking water use. The majority of the studies related to atrazine degradation in subsoil are carried out using agricultural topsoil having high atrazine concentration (Li et al. 2008; Lima et al. 2009; Singh and Cameotra 2014), and little is known on atrazine degradation in non-agricultural subsoil where, despite the low concentrations, it potentially endangers groundwater reserves. Thus, the outcomes of research on high atrazine concentrations may not be applicable in the bioremediation of atrazine in the groundwater zone.

The environmental impacts of herbicides are frequently assessed by evaluating their half-lives under standardized conditions. However, environmental conditions seldom resemble the laboratory conditions, and degradation can be reduced when the herbicide is used in the field (Bromilow et al. 1999). Atrazine is moderately soluble in water, and it therefore leaches into the groundwater, where its half-life can increase up to 9 times (Blume et al. 2004). Pesticide degradation rates can be dramatically influenced by soil pH (Mueller et al., 2010), organic material (Cheyns et al. 2012), temperature (Kookana et al. 2010), and soil layer (Willems et al. 1996).

Finnish atrazine-contaminated subsoils are typically non-agricultural, often acidic, low in organic carbon, and ambient temperatures are low. For these reasons, residual atrazine is still found in the groundwater decades after its use was discontinued (Vuorimaa et al. 2007). In bioremediation, the environmental conditions are modified to improve degradation. Because many bacteria use atrazine as a nitrogen source (Udiković-Kolić et al. 2012), the addition of carbon may enhance atrazine degradation in soils where the amount of available carbon, rather than nitrogen, is 4 limiting growth. Several studies have shown that bioaugmentation with Pseudomonas citronellolis ADP harboring the atzABCDEF genes enhance atrazine degradation in topsoil (Chelinho et al. 2010; Lima et al. 2009). Bioaugmented Pseudomonas citronellolis is capable of atrazine degradation at 12 °C (Monard et al. 2008), but no studies explored its exploitation in subsoil.

The purpose of this study was to investigate bioremediation in boreal subsoils contaminated with low concentrations of atrazine. The natural degradation potential of one of the soils used in this study has been previously described as low (Nousiainen et al. 2014). We tested the effects of temperature, atrazine concentration, carbon source and bioaugmentation with atrazine degrading microbial strains to enhance atrazine degradation in subsoil.

Materials and methods

Sampling

Soil was collected in two different sampling campaigns from two similar sites on the Salpausselkä ice-marginal formation, in Sveitsi natural park (SNP), Southern Finland (60°38’18’’ N, 24°49’17’’E) on 15.12.2011, and in Lohjanharju (LOH), Southern Finland (60°17’39’’N, 24°12’46’’E) on 15.09.2009. Both sites were in a coniferous forest, adjacent to waterworks where atrazine and its degradation products had been observed in the raw groundwater. At both sites, the most likely source of atrazine pollution was from non-agricultural use of atrazine, e.g. on railroads. Soil from site SNP was sandy or silty, and had low organic carbon content (< 1% dry weight). The ground water level was very high, and the saturated layer was only 20 cm below surface. Saturated soil was collected from site SNP from 2-5 m for bioremediation experiments, and from 50 cm for microautoradiography (MAR). The soil from site LOH had similar characteristics than the soil from site SNP in regards to texture and carbon content. The groundwater level in the area varies depending on rainfall and season. The soil pore gas composition at the depth of 2.5 m was measured with a Dräger gas meter at site LOH. The oxygen levels were similar to those above ground, indicating low microbial activity. Material for testing atrazine degradation in increased atrazine concentration was collected from site LOH from three different depths; bulk samples representing topsoil (1- 2 m), unsaturated soil (2-5 m) and saturated soil (6-9 m). All soil samples were transported to the laboratory, sieved through 8 mm sieve, mixed well and packed in plastic buckets. Soils were stored at +4 °C until use (>7 months).

Preparation of bacterial inocula

Atrazine degrading strains Pseudomonas citronellolis strain ADP (ATTC BAA-1386) genotype atzA/atzB, and Arthrobacter aurescens strain TC1 (DSM 11735) genotype trzN/atzB, were obtained from the culture collections ATCC and DSMZ, respectively. Prior to microcosm experiments, the bacteria were grown in liquid 0.1 × Luria- Bertani (LB) medium amended with 0.5 mg L-1 atrazine at 30 °C, 200 rpm. Cells were harvested from overnight cultures by centrifugation, and washed with 0.9% NaCl. A turbid suspension in 0.9% NaCl was prepared from both strains, and approximately 2 ×105 cells g-1 soil were used for bioaugmenting the microcosms.

Construction of microcosms and experimental setup

The atrazine mineralization potential of the autochthonous microbial community was tested by amending the soil with 1.2 mg kg-1 commercial atrazine (50% active ingredient) product (SriZon, Crystal Phospates, Delhi, India) to samples from three soil layers (topsoil, unsaturated and saturated) of the LOH site, creating a higher atrazine concentration than is usually detected in groundwater in order to reach the concentration required for the detection by LC/MS/MS. Atrazine was dissolved in deionised laboratory water (Milli-Q, Merck Millipore, Darmstadt, Germany) (24 mg l-1), and the liquid was mixed thoroughly in 3.5 kg of soil. The final concentration was 0.006 mol g-1 soil. Aliquots of 65 g were taken for molecular analyses, and 10 grams was used for radiorespirometry.

The effect of additional carbon (citrate or molasses) and/or atrazine degrading bacteria on atrazine mineralization was tested in microcosms using 10 g soil samples from site SNP. The carbon sources were added in the microcosm treatments as a component of modified Cai medium (Cai et al. 2003). The carbon components added were an equal mass of carbon atoms in either sodium citrate (3 g L-1) or molasses (3.74 g L-1). Modified Cai medium was also prepared without any additional carbon amendment. The added atrazine concentration in this experiment consisted only of the label, 6.25×10-10 mol g-1.

Pseudomonas citronellolis and Arthrobacter aurescens were grown overnight in Cai medium at 30 °C. The cells were harvested by centrifugation and resuspended in 0.9% NaCl solution. The bioaugmented mineralization treatments treatments were amended by adding 0.5 ml of bacterial solution.

Atrazine mineralization by radiorespirometry

Atrazine mineralization was carried out by radiorespirometry as described elsewhere (Nousiainen et al., 2014) using 10 g of soil, and modified Cai medium with or without carbon source, labelled atrazine as tracer (6.25×10-10 mol g-1 soil (fw)), or label and additional atrazine (0.006 mol g-1 soil (fw)) and/or bacteria according to Table 1. All microcosms were carried out in triplicate. Negative controls were prepared by autoclaving the soil microcosms three times. Microcosms were incubated at 30 °C and 10 °C, 100 rpm for 115 days. The parameters of the mineralization kinetics were determined by fitting a modified Gompertz model (Zwietering et al. 1990) to the mineralization curves using Sigma Plot 4.0.

Chemical analysis of atrazine, HA DIA, DEA and DEDIA

Parallel samples containing 60 g of soil from Site LOH were set up for chemical analysis. The concentration of atrazine and its degradation products hydroxyatrazine (HA), desethylatrazine (DEA), deisopropylatrazine (DIA) and desethyl-deisopropylatrazine (DEDIA) in soil were examined at the beginning, middle (5 weeks for 30 °C incubations, 8 weeks for 10 °C incubations), and end of the experiment (8 weeks for 30 °C incubations, 16 weeks for 10 °C incubations) by Ramboll Analytics (Lahti, Finland) by gas-liquid chromatography – mass spectrometry.

Quantification of degradation genes by qPCR

Parallel samples for each mineralization bottle were prepared without the labeled atrazine for molecular analyses. Total DNA was extracted from 0.5 g of soil periodically during the course of the experiments by FastDNA Spin Kit

6 for Soil (MP Biomedicals, OH, USA) according to the manufacturer’s instructions. The number of degradation genes atzA, trzN and atzB were quantified as described elsewhere (Nousiainen et al., 2014).

Estimation of active atrazine degraders by MAR

The numbers of active atrazine degrading microorganisms in native soil from site SNP, in soil amended with 1g citrate kg-1, and in soil spiked with an overnight mixture of P. citronellolis and A. aurescens grown in mineral medium containing 100 mg L-1 atrazine, were investigated by microautoradiography as described in details elsewhere (Nielsen and Nielsen, 2005) with minor modifications. Briefly, 3.0 g of soil was weighed in three 3 mL glass vials, and 20 µCi of 14C ring-labeled atrazine (specific activity of 160 mCi mmol-1) were mixed well by vortexing and incubated under aerobic conditions for 5 d in 30 ºC at 100 rpm. Parallel samples were used for determining mineralization by radiorespirometry and the number of degradation genes.

Microbial cells were extracted from 1 g of soil by density gradient centrifugation (Burmølle et al., 2003), using OptiPrep Density Gradient Medium (Sigma-Aldrich, MO, USA). Extracted cells were homogenized and washed by glass tissue grinder (Thomas Scientific®) with 500 µL of sterile tap water three times. Dilutions of the washed cells were filtered on 0.22 µm white polycarbonate filters. The filters were placed on regular microscopy glass slides (24 × 60 mm), air dried on the bench. The slides were then dipped in liquid film emulsion (Kodak autoradiography emulsion type NTB, NY, USA), air dried and exposed for two days at +4 ºC. Exposure conditions and development procedures were performed as described elsewhere (Nielsen and Nielsen, 2005).

The exposed and developed filters were studied with a laser scanning microscope (Zeiss LSM 510 Meta). 30 random pictures were taken from each filter. The number of active cells was estimated using the threshold criterion of minimum 5 silver grains within a circle (Ø=4 μm) covering a cell. The total number of bacterial cells in the sampling site was determined by DAPI (4',6-diamidino-2-phenylindole) staining as described elsewhere (Tuomi et al., 2004).

Microbial community in groundwater

Because the DNA concentration in the sample soils was too low for constructing a clone library, unamended Bio-Trap samplers® (MicrobialInsights, TN, USA), containing only activated carbon granules as the trapping material, were used to obtain a sample of the groundwater bacterial community. The samplers were installed in two groundwater monitoring wells at approximately 5 m depth at the LOH sampling area. Monitoring well 608 was situated on the edge of a forest, at a sandy ridge next to the waterworks, whereas monitoring well PM was located approximately 100 m away on a slight slope, surrounded by small-scale industries. After four months, the samplers were retrieved, transported to the laboratory and opened in sterile conditions. The bacterial DNA was extracted directly from the sampler material using the same protocol as for extracting DNA from soil.

The number of atzA and atzB genes in the DNA was quantified as described above. The bacterial community composition was investigated by constructing a clone library of 750-100 bp fragments of the 16S rRNA genes in the extracted DNA pool (Yli-Hemminki et al., 2014). One clone library was constructed from each groundwater well, each consisting of 100 clones. Cloned inserts were sequenced using T7 and T3 sequencing primers at Macrogen, (Seoul, South Korea). The sequences were deposited in the GenBank with the accession numbers KJ670501- KJ670677. The sequences were compared with those found in the NCBI Sequence Database using the BLAST tool on

19.5. 2011. Species richness was estimated by calculating Colwell rarefactions with EstimateS software version 9.1 and plotting the rarefaction curves.

Results

Atrazine mineralization

The addition of atrazine degrading bacteria markedly increased the mineralization of atrazine (Fig. 1c, Table 1). In the bioaugmented treatments, the highest atrazine mineralization rates were observed at 30 °C. The most efficient treatment was amendment by Arthrobacter aurescens and citrate (76%), followed by Arthrobacter aurescens and citrate (62%), and Pseudomonas citronellolis and molasses (55%). Atrazine mineralization was marginal in the microcosms not amended with carbon or bioaugmented with atrazine degraders (Fig. 1b). However, biostimulation with carbon sources resulted in efficient mineralization, and maximum mineralization was at the same level as in bioaugmented samples incubated at 30 °C (Fig. 1d, Table 1). The highest atrazine mineralization rate, 27 ± 1.4% day- 1, was at 30 °C in unsaturated soil amended with atrazine and augmented with Pseudomonas citronellolis (Table 1). Higher atrazine concentration did not lead to more efficient atrazine degradation by autochthonous microbes, as less than 5% of atrazine was mineralized regardless of the soil layer (Fig. 1b). However, the addition of atrazine degrading bacteria resulted in 46% atrazine removal at 10 °C and 67% at 30 °C. In all negative controls at, >0.1% of atrazine mineralized.

Quantification of atrazine degradation genes by qPCR

The degradation genes atzA, trzN and atzB genes were quantified at the beginning, in the middle and at the end of the experiment. The gene copy levels did not show a marked change after the midpoint of the experiment (21 days). The copy numbers in situ, and at the end of the experiment (42 d), are displayed in Table 1.

The gene copy numbers in situ were low, but increased when bioaugmentation treatments were used (Table 1). One- way ANOVA revealed a statistically relevant increase in degradation gene copy numbers in the bioaugmented microcosms (p = 0.000, 0.000 and 0.000 for atzA, trzN and atzB gene copy numbers, respectively). Citrate amendment supported higher atzA gene copy numbers, irrespective of bioaugmentation with bacteria (p = 0.031), suggesting that the native bacterial population can be activated with citrate amendments. The smallest changes of gene copy numbers were observed in microcosms with a high atrazine concentration and no carbon amendments.

Concentrations of atrazine and its degradation products

The concentration of atrazine and its degradation products HA and DIA in atrazine spiked soil at the beginning, in the middle, and the end of the experiment are presented in Fig. 2. DEA was only found in the middle and at the end of the experiment at very low concentrations (>0.023 g kg-1) The concentrations of DEDIA were below detection limit (0.005 g kg-1). At 30 °C, atrazine concentration was reduced by 17-61% in all soil layers while at 10 °C, the greatest atrazine removal was 34%. The accumulation of degradation products, especially HA and DIA was observed. At the beginning of the experiment no HA was detected, but it started to accumulate at the midpoint of the experiment indicating possible chemical atrazine degradation, as though the degradation genes atzA and trzN were not detected by qPCR. The commercial atrazine used to spike the microcosms contained DIA, which was found in the beginning of 8 the experiment at approximately 0.35 mg kg-1 soil (dw). It is therefore difficult to estimate the proportion of DIA formed out of degraded atrazine. The maximum concentration of DEA in soil was 0.01 mg kg-1.

Active atrazine degrading cells

The total numbers of bacteria after five days of incubation, based on DAPI count, were 5.3×106 cells g-1 in unamended soil, 7.1×106 cells g-1 in soil amended with citrate and 2.0×106 cells g-1 in soil inoculated with Pseudomonas citronellolis and Arthrobacter aurescens. The mean numbers of active atrazine degrading cells were 1.8×104, 1.3×105, and 3.8×104 cells g-1 soil at 30 °C, respectively. The mean proportions of the atrazine degrading bacteria estimated by MAR in different treatments were thus 0.3% of the total cell count in native soil, 1.9% in bioaugmentation treatment, and 1.8% in biostimulation treatment. Atrazine mineralized poorly in this experiment: radiorespirometric measurements parallel to the MAR experiments showed that < 1% of the labelled atrazine mineralized.

In the unamended soil, the silver grain densities around MAR-positive cells were generally smaller than in amended soils, indicting less active atrazine incorporation. When citrate or atrazine degrading bacteria were added, the differences in silver grain densities varied between the MAR-positive cells, indicating that some of the atrazine degraders were stimulated by the presence of citrate and that the amended degraders were actively degrading atrazine.

Microbial community in groundwater

In groundwater monitoring well 608, the most common bacterial phylae were identified as α-, β- and γ –proteobacteria and Actinobacteria (Table 2). Sequences of strains belonging to Rhodococcus, Pseudomonas and Arthrobacter were found on this site. 77% of the clones were most closely related to members of cultured species; the most numerous of these clones were identified as Acinetobacter, Aminobacter and Rhodococcus. The bacterial community composition in groundwater well PM differed from that of well 608 (Table 2). All clones belonged to either α-, β- or γ – proteobacterial phylae, and Rhodobacter were commonly observed. Several matches to clones originating from activated carbon filters were found. As in well 608, many clones (84%) in well PM were identified as close relatives of members of cultured species.

The copy numbers of atzA and atzB in well 608 were 6.3 × 103 and 2.7 × 104 g-1 activated carbon used as sampler material, respectively, and in well PM 1.3 × 104 and 4.6 × 104 g-1 sampler material, respectively.

Rarefaction analysis revealed that the size of the clone libraries adequately represented the bacterial community inhabiting the Bio-Trap sampler® (Fig. SM-1).

Discussion

We explored the potential of two bioremediation techniques for soils with low atrazine concentration. Atrazine did not naturally mineralize in the boreal subsoils, most likely due to the low organic content, and low number of degradation genes (Nousiainen et al., 2014). However, atrazine mineralization could be enhanced significantly by bioremediation treatments.

Biostimulated and bioaugmented (BS/BA) microcosms mineralized up to 76% of labelled atrazine (Fig. 1). At elevated temperature (30 °C), atrazine mineralization was directed by the bioaugmented strains, likely because the incubation temperature was close to growth optimum of the strains used. In the combined treatments, additional carbon sources did not seem to have a pronounced effect on mineralization. At temperature similar to the ambient temperature in the boreal zone (10 °C), biostimulating the indigenous, psychrophilic bacteria with molasses was a sufficient treatment method, with up to 52% of available atrazine was mineralized. Citrate addition had a less pronounced effect, as only 23% of atrazine was mineralized. Interestingly, the effect of carbon addition was minimal at 30 °C. The data collected in this study does not provide explanations, but it is possible that the indigenous bacteria suffered from the elevated temperature.

The survival of bioaugmentation strains is a key problem, and the persistence of Pseudomonas ADP in triazine bioremediation has previously been disputed (Morán et al., 2006; Zhao et al., 2003). Other studies using Arthrobacter have shown that additional carbon decreases the density of the bioaugmented strain during bioremediation (Xie et al., 2012). In our study, Arthrobacter aurescens genotype trzN/atzB was always present when Arthrobacter was augmented. The Pseudomonas citronellolis-genes atzA/atzB were only detected when additional carbon was used, indicating that Pseudomonas citronellolis cannot compete with the autochthonous population if additional carbon is not used. This effect was pronounced when citrate was used, as the net mineralization rate of atrazine was higher in citrate amended microcosms than in treatments with molasses or controls without carbon. It is noteworthy that in all the microcosms where genotype atzA/atzB was added, also the trzN gene was observed. The prevalence of trzN over atzA gene in the soil environment reported by Arbeli and Fuentes (2010) might explain why it was often encountered in our experiment. Overall, the degradation gene numbers were high in bioaugmented treatments, but often below detection limit in the biostimulated treatments. Despite the low gene copy numbers, a significant amount of atrazine disappeared, e.g. 55% when molasses was used (Table 1). It is possible, that carbon addition directed degradation towards dealkylation, rather than dechlorination of atrazine (Ngigi et al., 2013). In fact, the most commonly observed atrazine degradation products in the Finnish groundwaters are derived from atrazine dealkylation (Vuorimaa et al. 2007), suggesting that this pathway is more active in the environment, or that the degradation products on this pathway have long half-lives.

It is not possible from the observations made in this study to estimate the significance of the different degradation pathways, and to differentiate between the degradation efficiencies of bioaugmented strains and autochthonous bacteria. Despite this, our results suggest that the gene pool provided by the autochthonous bacteria is sufficient for degrading trace amounts of atrazine, and it possibly contains multiple pathways for atrazine degradation. In trace atrazine concentrations, even a 50% reduction would be sufficient to decrease the concentration below the legal limit of pesticides in groundwater. We used microautoradiography (MAR) to estimate proportions of atrazine degrading bacteria from the total bacteria. This method measures the ability to assimilate atrazine at the cellular level, and thus the presence and degradation efficiency of microbial cells capable of degrading the compound. MAR has predominantly been used in aqueous environments, in which soil colloids provide excellent conditions for single cell resolution and easy visualization of the produced silver grains. In soils and sediments, extraction of the cells is necessary, but may lead to significantly reduced cell recovery (Pascaud et al., 2012). Despite the challenges created by the soil particles, MAR is a robust method for circumventing the need to study specific functional genes, because the activity of the cells may be observed at the cellular level. By studying the ratio of MAR-positive cells of the total DAPI count, we avoid extraction biases. In our study, the proportion of atrazine degrading bacteria was small in all treatments, but it increased when bioremediation treatments were used. The results correlated with degradation gene copy numbers, as the numbers in situ are several orders of magnitude smaller than in biostimulated and bioaugmented 10 treatments. In one gram of biostimulated soil, both the number of MAR positive cells, and the copy numbers of atzA and atzB genes were approximately104 cells and copies g-1 soil, respectively. The correlation depends on plasmid copy number. If the catabolic plasmid is present in single copy, the gene copy numbers correlate to the MAR positive cell numbers, whereas multiple copy plasmid will result in number of degradation genes exceeding the number of MAR positive cells. The first scenario was true for in situ conditions, where the number of MAR positive cells (103 cells g-1 soil) surpass the number of degradation genes (102 copies atzA, 103 copies atzB g-1 soil). It is also likely that some of the MAR-positive bacteria use a degradation pathway containing unknown genes we could not target by the applied PCR approach. The highest proportion of atrazine degraders was expected and observed in the bioaugmentation treatments. Bioaugmentation treatments also increased the number of degraders, supporting our hypothesis, that carbon addition stimulates the native atrazine degrading bacterial population in carbon limited soil. To our knowledge, this is the first attempt to visualize and enumerate indigenous and bioaugmented active atrazine degrading bacteria by MAR.

The knowledge on the composition of the autochthonous bacterial community is needed when the bioremediation strategy is chosen. Often the factor limiting degradation is not the absence of contaminant degrading bacteria or degradation genes, but rather the environmental conditions, which do not favor degradation (Thompson et al., 2005). If potential degraders are absent, bioaugmentation may be the only remediation option. Clone libraries revealed that bacteria belonging to known atrazine degrading species were residing in the groundwater zone. In well 608, the most common clones were identified as Acinetobacter (24%), which are known to be versatile degraders and efficient atrazine degraders in high concentration (Singh et al. 2004). Other clones related to potential atrazine degraders included Microbacterium species (Morohoshi et al. 2011), as well as Polaromonas (Devers et al. 2007) Aminobacter (McDonald et al. 2005), Rhodococcus (Behki et al. 1995), Pseudomonas and Arthrobacter, species extensively studied because of their atrazine degrading capabilities. The presence of degraders in the samplers was also confirmed by a positive signal of degradation genes by qPCR.

Many of the clones in our library were close matches to cultured bacteria. A likely reason is that only the actively growing fraction of the bacterial community is able to colonize the carbon beads of the Bio-Trap® sampler (Sublette et al., 2006). Despite this bias, the clone libraries identified several organisms that belong to phylogenetic groups known to contain atrazine degraders. The prerequisite of growth, and the presence of degradation genes on sampler material indicates, that the atzA and atzB containing bacteria are actively growing. Although the methods used in this study do not allow the linkage between clone identity and its atrazine degrading capability, it is possible that some of the clones identified have atrazine degradation genes.

The atrazine degrading ability of the autochthonous bacteria was tested by adding high concentrations of atrazine to soil, mimicking agricultural use. At 10 °C, atrazine disappearance was retarded, and only up to 5% of atrazine was mineralized. However, at 30 °C, atrazine was converted to hydroxyatrazine. Bioaugmenting with Pseudomonas citronellolis facilitated atrazine mineralization at both 10 °C and 30 °C. The commercial atrazine product contained DIA as an impurity, and therefore estimation of DIA accumulation from atrazine is difficult. The presence of DIA in the commercial atrazine product might also explain why it is a common groundwater contaminant.

The positive effect of carbon additions on atrazine mineralization is probably a result of cometabolic processes (Willems et al., 1996). Contrary findings on carbon promoted degradation have been observed in agricultural soils having high atrazine concentrations (Silva et al., 2004; Xie et al., 2012). These contradicting results highlight the importance of understanding the effect of environmental conditions, as they heavily impact bioremediation treatment. 11

We hypothesize that in our soil, the low carbon content was the factor limiting bacterial growth, and our results show that changing this parameter allowed efficient removal of residual atrazine in soil.

In conclusion, residual atrazine persisted in nutrient-poor subsoil because the nutritional conditions did not favor its degradation, even though potential degraders were present. The addition of atrazine degrading bacteria is a powerful tool to remediate contaminated soil. In the particular boreal subsoil used in this study, the combination of biostimulation and bioaugmentation efficiently removed atrazine. The addition of carbon, especially in the form of citrate, proved an efficient aid in atrazine degradation in combination with atrazine degrading bacteria at 30 °C. Although the biostimulation approach did only remove approximately half of the labelled atrazine, it was especially efficient at low temperature, suggesting that simple biotechnological applications may be sufficient to remediate atrazine from nutrient poor environments.

Acknowledgments

We thank Elina Saario, Jaana Keto and Ilse Heiskanen for technical assistance, and the municipalities of Lohja and Hyvinkää for granting us access to the sampling sites. This research was funded by the Academy of Finland, grant n:o 12482, the Department of Biotechnology, Ministry of Science and Technology, New Delhi, for the Indo-Finland project (G-1649), as well as the EnSTe graduate school, the NORDFORSK Nordic Environmental Nucleotide Network and the Danish Research Council for Strategic Research via the Centre “EcoDesign”.

References

Accinelli C, Dinelli G, Vicari A, Catizone P (2001) Atrazine and metolachlor degradation in subsoils. Biol Fert Soils 33:495–500

Aislabie J, Hunter D, Ryburn J, Fraser R, Northcott GL, Di HJ (2004) Atrazine mineralisation rates in New Zealand soils are affected by time since atrazine exposure. Soil Res 42:783–792

Arbeli Z and Fuentes C (2010) Prevalence of the gene trzN and biogeographic patterns among atrazine-degrading bacteria isolated from 13 Colombian agricultural soils. FEMS Microbiol Ecol 73:611–623

Behki R, Topp E, Dick W, Germon P (1995) Metabolism of the herbicide atrazine by Rhodococcus strains. Appl Environ Microbiol 59:1955–1959

Blume E, Bischoff M, Moorman TB, Turco RF (2004) Degradation and binding of atrazine in surface and subsurface soils. J Agric Food Chem 52:7382–7388

Bromilow RH, Evans AA, Nicholls PH (1999) Factors affecting degradation rates of five triazole fungicides in two soil types: 2. Field studies. Pestic Sci 55:1135–1142

Burmølle M, Hansen LH, Oregaard G, Sørensen SJ (2003) Presence of N-Acyl homoserine lactones in soil detected by a whole-cell biosensor and flow cytometry. Microb Ecol 45:226–236 12

Cai B, Han Y, Liu B, Ren Y, Jiang S (2003) Isolation and characterization of an atrazine-degrading bacterium from industrial wastewater in China. Lett Appl Microbiol 36:272–276

Chelinho S, Moreira-Santos M, Lima D, Silva C, Viana P, André S, Lopes I, Ribeiro R, Fialho AM. Viegas CA, Sousa JP (2010) Cleanup of atrazine-contaminated soils: ecotoxicological study on the efficacy of a bioremediation tool with Pseudomonas sp. ADP. J Soils Sediments 10:568–578

Cheyns K, Calcoen J, Martin-Laurent F, Bru D, Smolders E, Springael D (2012) Effects of dissolved organic matter (DOM) at environmentally relevant carbon concentrations on atrazine degradation by Chelatobacter heintzii SalB. Appl Microbiol Biotechnol 95:1333–1341

Devers M, El Azhari N, Udikovic Kolic N, Martin-Laurent F (2007) Detection and organization of atrazine- degrading genetic potential of seventeen bacterial isolates belonging to divergent taxa indicate a recent common origin of their catabolic functions. FEMS Microbiol Lett 273:78–86

Hallberg GR (1989) Pesticides pollution of groundwater in the humid United States. Agric Ecosyst Environ 26:299– 367

Jurado A, Vàzquez-Suñé E, Carrera J, López de Alda M, Pujades E, Barceló D (2012) Emerging organic contaminants in groundwater in Spain: A review of sources, recent occurrence and fate in a European context. Sci Total Environ 440:82–94

Kookana R, Holz G, Barnes C, Bubb K, Fremlin R, Boardman B (2010) Impact of climatic and soil conditions on environmental fate of atrazine used under plantation forestry in Australia. J Environ Manage 91:2649–3656

Li Q, Li Y, Zhu X, Cai B (2008) Isolation and characterization of atrazine-degrading Arthrobacter sp. AD26. J Environ Sci 20:1226–1230

Lima D, Viana P, André S, Chelinho S, Costa C, Ribeiro R, Sousa JP, Fialho AM, Viegas CA (2009) Evaluating a bioremediation tool for atrazine contaminated soils in open soil microcosms: The effectiveness of bioaugmentation and biostimulation approaches. Chemosphere 74:187–192

McDonald I, Kämpfer P, Topp E, Warner KL, Cox MJ, Connell Hancock TL, Millen LG, Larkin MJ, Ducrocq V, Coulter C, Harper DB, Murrel CJ, Oremland RS (2005) Aminobacter ciceronei sp. nov. and Aminobacter lissarensis sp. nov., isolated from various terrestrial environments. Int J Syst Evol Micro 55:1827–1832.

Monard C, Martin-Laurent F, Vecchiato C, Francez AJ, P Vandenkoornhuyse P, Binet F (2008) Combined effect of bioaugmentation and bioturbation on atrazine degradation in soil. Soil Biol Biochem 40:2253–2259

Morán AC, Müller A, Manzano M, González B (2006) Simazine treatment history determines a significant herbicide degradation potential in soils that is not improved by bioaugmentation with Pseudomonas sp. ADP. J Appl Microbiol 101:26–35

Morohoshi T, Wang W-Z, Someya N, Ikeda T (2011) Genome sequence of Microbacterium testaceum StLB037, an N-acylhomoserine lactone-degrading bacterium isolated from potato leaves. J Bacteriol 193: 2072–2073

Mueller TC, Steckel LE, Radosevich M. (2010) Effect of soil pH and previous atrazine use history on atrazine degradation in a Tennessee field soil. Weed Sci 58:478–483.

Nagy I, Compernolle F, Ghys K, Vanderleyden J, De Mot R (1995) A single cytochrome P-450 system is involved in degradation of the herbicides EPTC (S-ethyl dipropylthiocarbamate) and atrazine by Rhodococcus sp. strain NI86/21. Appl Environ Microbiol 61:2056–2060

Ngigi AH, Getenga ZM, Dörfler U, Boga HI, Kuria B, Ndalut P, Schroll R (2013) Effects of carbon amendment on in situ atrazine degradation and total microbial biomass. J Environ Sci Heal B 48:40–48.

Nielsen JL and Nielsen PH (2005) Advances in microscopy: Microautoradiography of single cells, in Methods in Enzymology. Academic Press, San Diego, USA, pp. 237–256.

Nousiainen, A. Björklöf K, Sagarkar S, Mukherjee S, Purohit HJ, Kapley A, Jørgensen KS (2014) Atrazine degradation in boreal non-agricultural subsoil and tropical agricultural soil. J Soils Sediment 14:1179–1188

Pascaud A, Soulas M-L, Amellal S, Soulas G (2012) An integrated analytical approach for assessing the biological status of the soil microbial community. Eur J Soil Biol 49:98–106

Sagarkar S, Nousiainen A, Shaligram S, Björklöf K, Lindström K, Jørgensen KS, Kapley A (2014) Efficient bioremediation of atrazine contaminated soil: Mesocosm studies. J Environ Manage 139:208–216

Sagarkar S, Mukherjee S, Nousiainen A, Björklöf K, Purohit HJ, Jørgensen KS, Kapley A (2013) Monitoring bioremediation of atrazine in soil microcosms using molecular tools. Environ Pollut 172:1–8

Silva E, Fialho AM, Sà-Correia I, Burns RG, Shaw LJ (2004) Combined bioaugmentation and biostimulation to cleanup soil contaminated with high concentrations of atrazine. Environ Sci Technol 38:632–637

Singh P, Suri CR, Cameotra SS (2004) Isolation of a member of Acinetobacter species involved in atrazine degradation. Biochem Biophys Res Comm 317:697–702

Singh P and Cameotra SS (2014) Influence of microbial and synthetic surfactant on the biodegradation of atrazine. Environ Sci Pollut Res 21:2088–2097 de Souza ML, Sadowsky MJ, Wackett LP (1996) Atrazine chlorohydrolase from Pseudomonas sp. strain ADP: gene sequence, enzyme purification, and protein characterization. J Bacteriol 178:4894–4900

Stuart M, Lapworth D, Crane E, Hart A (2012) Review of risk from potential emerging contaminants in UK groundwater. Sci Total Environ 416:1–21

Sublette K, Peacock A, White D, Davis G, Ogles D, Cook D, Kolhatkar R, Beckmann D, Yang X (2006) Monitoring subsurface microbial ecology in a sulfate-amended, gasoline-contaminated aquifer. Ground Water Monit R 26:70–78

Thompson IP, van der Gast CJ, Ciric L, Singer AC (2005) Bioaugmentation for bioremediation: the challenge of strain selection. Environ Microbiol 7:909–915

Tuomi, PM, Salminen JM, Jørgensen KS (2004) The abundance of nahAc genes correlates with the 14C-naphtalene mineralization potential in petroleum hydrocarbon-contaminated oxic soil layers. FEMS Microbiol Ecol 51:99–107

Udiković-Kolić N, Scott C, Martin-Laurent F (2012) Evolution of atrazine-degrading capabilities in the environment. Appl Microbiol Biotechnol 96:1175–1189

Vuorimaa P, Kontro M, Rapala J, Gustafsson (2007) Torjunta-aineiden esiintyminen pohjavedessä. (Occurrence of pesticides in groundwater). The Finnish Environment 42, Finnish Environment Institute, Helsinki (in Finnish)

Wang Q, Xie S, Hu R (2013) Bioaugmentation with Arthrobacter sp. strain DAT1 for remediation of heavily atrazine- contaminated soil. Int Biodeter Biodeg 77:63–67

Willems HPL, Lewis KJ, Dyson JS, Lewis FJ (1996) Mineralization of 2,4-D and atrazine in the unsaturated zone of a sandy loam soil. Soil Biol Biochem 28:989–996

Xie, S, Wan R, Wang Z, Wang Q (2012) Atrazine biodegradation by Arthrobacter strain DAT1: effect of glucose supplementation and change of the soil microbial community. Environ Sci Pollut Res 20: 4078–4084

Yli-Hemminki P, Jørgensen KS, Lehtoranta J (2014) Iron–manganese concretions sustaining microbial life in the baltic sea: The structure of the bacterial community and enrichments in metal-oxidizing conditions. Geomicrobiol J 31:263–275

Zhao S, Arthur LE, Coats JR (2003) Influence of microbial inoculation (Pseudomonas sp. strain ADP), the enzyme atrazine chlorohydrolase, and vegetation on the degradation of atrazine and metolachlor in soil. J Agric Food Chem 51: 3043–3048.

Zwietering MH, Jongenburger I, Rombouts FM, van’t Riet K (1990) Modeling of the bacterial growth curve. Appl Environ Microbiol 56:1875–1881

Figure captions

Fig. 1 Mineralization of 14C-ring labelled atrazine in atrazine spiked soil bioaugmented at 30 °C and 10 °C, site LOH (a), and mineralization in three soil layers at 30 °C and 10 °C, site LOH (b). Mineralization in bioaugmented and biostimulated (BS/BA) microcosms at 30 °C, site SNP (c). Mineralization in biostimulated (BS) microcosms at 10 °C and 30 °C, site SNP (d)

Fig. 2 Atrazine (a), hydroxyatrazine (HA) (b) and deisopropylatrazine (DIA) (c) concentrations in spiked microcosms from site LOH soil at the beginning (black), middle (5 weeks for 30 °C incubations, 8 weeks for 10 °C incubations) (grey) and end (8 weeks for 30 °C incubations, 16 weeks for 10 °C incubations) (white) of the experiment. * = below detection limit (0.005 mg kg-1)

Supplementary Fig. 1 Rarefaction curves from the clone libraries obtained from BioTrap® samplers in groundwater monitoring wells PM and 608

Figures

Fig. 1

Fig. 2

Supplementary Fig. 1

Table 1: Experimental setup of mineralization experiment, and results from mineralization and degradation gene atzA, trzN and atzB quantifications. Atrazine degradation gene copy numbers in triplicate at the beginning (in situ) and at the end of the experiment, with standard deviations. Gene copy number at 42 d Temp Soil Soil layer Added carbon, c added Atrazine Atrazine Minerali- log atzA log trzN log atzB -1 -1 -1 (°C) site genotype of atrazine, mineralized t½ (days) zation g soil g soil g soil bioaugmented strain t=0 (mol g- in 115 d (%) rate 1 soil fw) (% d-1) SNP unsaturated In situ 0.0 a a a 1.3 ±0.9 2.7 ±0.12 3.0 ±0.05

30 SNP Unsaturated Citrate, 6.25×10-10 52 ±30 7.7±2.1 20.2±18.0 7.7 ±0.3 6.2 ±0.4 7.7 ±0.7 Pseudomonas SNP Unsaturated Molasses, 6.25×10-10 55 ± 32 12.8±5.0 11.7±3.9 6.9 ±0.4 <2.4 7.3 ±0.5 Pseudomonas SNP Unsaturated No carbon, 6.25×10-10 65 ±38 13.3±2.8 11.2±2.3 <2.4 6.4 ±0.2 6.6 ±0.3 Pseudomonas SNP Unsaturated Citrate, 6.25×10-10 76 ±44 5.1±0.5 16.6±1.7 2.0 ±0.4 5.9 ±0.4 5.6 ±0.4 Arthrobacter SNP Unsaturated Molasses, 6.25×10-10 62 ±36 7.6±1.4 8.7±2.5 <2.4 5.5 ±0.3 5.1 ±0.2 Arthrobacter SNP Unsaturated No carbon, 6.25×10-10 39 ±23 b 8.7±3.1 <2.4 5.9 ±0.4 5.6 ±0.3 Arthrobacter SNP Unsaturated Citrate 6.25×10-10 1 ±1 b b 3.3 ±0.9 <2.4 4.7 ±0.3 SNP Unsaturated Molasses 6.25×10-10 1 ±1 b b <2.4 <2.4 4.4 ±1.5

SNP Unsaturated No carbon 6.25×10-10 21 ±12 17.6 5.9 <2.4 3.0 ±0.2 3.0 ±0.3 10 SNP Unsaturated Citrate 6.25×10-10 23 ±13 13.7 8.1 5.4 ±0.2 3.1 ±0.7 2.9 ±0.8 SNP Unsaturated Molasses 6.25×10-10 52 ±30 28.9±0.1 2.2±0.7 <2.4 <2.4 4.3 ±1.4 SNP Unsaturated No carbon 6.25×10-10 1 ±1 b b 3.4 ±1.1 <2.4 <2.4 LOH unsaturated In situ 0.0 a a a >1.1 2.36 ±0.6 >1.78

b b 30 LOH Topsoil High atrazine 0.006 1.7 ±1.0 <1.1 <3.58 <1.78 LOH Unsaturated High atrazine 0.006 2.1 ±1.2 b b <1.1 <3.58 <1.78 LOH Saturated High atrazine 0.006 4.8 ±2.8 b b <1.1 1.3 ±0.1 <1.78 LOH Unsaturated High atrazine, 0.006 67 3.5±0.01 27.4±1.5 4.3 ±0.2 <0.5 4.0 ±0.2 Pseudomonas 10 LOH Topsoil High atrazine 0.006 1.3 ±0.7 b b 1.5 ±1 1.0 ±0.7 2.6 ±0.2 LOH Unsaturated High atrazine 0.006 0.8 ±0.5 b b <1.1 0.9 ±0.6 3.0 ±0.3 LOH Saturated High atrazine 0.006 1.7 ±1.0 b b 1.3 ±0.5 2.4 ±0.1 2.8 ±0.3 LOH Unsaturated High atrazine, 0.006 46 b 17.1 5.0 ±0.2 1.8 ±1.0 3.7 ±0.2 Pseudomonas a NA = not applicable b ND = not determined

Table 2: Closest identities of the most abundant (n > 2) clones enriched in the BioTrap® Samplers in ground water well 608. Closest match Phylogenetic Minimum No of Query Clone origin affiliation similarity clones accession No (%) Acinetobacter sp. γ-proteobacteria 98 21 AJ551148.1 Antarctic/Deep sea

Uncultured Herminiimonas sp. β-proteobacteria 96 16 HQ132390.1 Heavy metal contaminated estuarine sediment Aminobacter aminovorans α-proteobacteria 97 10 NR025301.1 Azo dye degradomg consortium

Rhodococcus qingshengii actinobacteria 98 6 HQ439600.1 Data not available

Aminobacter sp. α-proteobacteria 98 5 FJ711220.1 Carbonate cave

Acinetobacter sp. γ-proteobacteria 98 3 AJ551155.1 Antarctic/Deep sea

Aminobacter sp. α-proteobacteria 92 3 DQ401867.1 Dichlorobenil-treated soil

Herminiimonas saxobsidens β-proteobacteria 97 3 AB512141.1 Rock colonized with lichen

Aminobacter sp. α-proteobacteria 97 2 DQ401867.1 Dichlorobenil-treated soil

Aminobacter lissarensis α-proteobacteria 99 2 NR041724.1 Terrestrial environment

Rhodococcus sp. actinobacteria 98 2 FN646613.1 Soil from Himalayan glacier

Uncultured Comamonadaceae sp. β-proteobacteria 97 2 AF523046.1 Bottled natural mineral water

Table 3: Closest identities of the most abundant (n > 2) clones enriched in the BioTrap® Samplers in ground water well PM. Closest match Phylogenetic Minimum No of Query Clone origin affiliation similarity clones accession No (%) Rhodobacter sp. α-proteobacteria 99 16 HM156123.1 Glacier forefield Rhodobacter sp. α-proteobacteria 98 14 GU441680.1 Glacier cryoconite

21 Uncultured Herminiimonas sp. β-proteobacteria 95 11 HQ132390.1 Heavy metal contaminated estuarine sediment Polaromonas sp. β-proteobacteria 96 9 AB426569.1 Freshwater bacterioplankton Methylotenera sp. β-proteobacteria 97 6 NR074693.1 Lake sediment Pseudomonas sp. γ-proteobacteria 98 5 GQ345341.1 PAH containing river sediment Methylotenera mobilis β-proteobacteria 98 4 DQ287786.1 Lake sediment Hydrogenophaga sp. β-proteobacteria 98 3 EU130968.1 Activated carbon filters in water treatment Polaromonas sp. β-proteobacteria 97 3 EU130980.1 Activated carbon filters in water treatment Polaromonas sp. β-proteobacteria 98 3 EU130985.1 Activated carbon filters in water treatment Pseudomonas putida γ-proteobacteria 97 2 DQ178233.1 Phenol-and creosol degrading species

Journal of Environmental Management 139 (2014) 208e216

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Soil mesocosm studies on atrazine bioremediation

Sneha Sagarkar a, Aura Nousiainen b, Shraddha Shaligram a, Katarina Björklöf b, Kristina Lindström c, Kirsten S. Jørgensen b, Atya Kapley a,* a Environmental Genomics Division, National Environmental Engineering Research Institute (CSIR-NEERI), Nehru Marg, Nagpur 440 020, India b Finnish Environment Institute, PO Box 140, FI-00251 Helsinki, Finland c Department of Food and Environmental Sciences, Division of Microbiology, University of Helsinki, FI-00014 Helsinki, Finland article info abstract

Article history: Accumulation of pesticides in the environment causes serious issues of contamination and toxicity. Received 22 November 2013 Bioremediation is an ecologically sound method to manage soil pollution, but the bottleneck here, is the Received in revised form successful scale-up of lab-scale experiments to field applications. This study demonstrates pilot-scale 14 February 2014 bioremediation in tropical soil using atrazine as model pollutant. Mimicking field conditions, three Accepted 17 February 2014 different bioremediation strategies for atrazine degradation were explored. 100 kg soil mesocosms were Available online set-up, with or without atrazine application history. Natural attenuation and enhanced bioremediation were tested, where augmentation with an atrazine degrading consortium demonstrated best pollutant Keywords: Atrazine removal. 90% atrazine degradation was observed in six days in soil previously exposed to atrazine, while Bioremediation soil without history of atrazine use, needed 15 days to remove the same amount of amended atrazine. Consortium The bacterial consortium comprised of 3 novel bacterial strains with different genetic atrazine degrading LH-PCR potential. The progress of bioremediation was monitored by measuring the levels of atrazine and its Mesocosm intermediate, cyanuric acid. Genes from the atrazine degradation pathway, namely, atzA, atzB, atzD, trzN and trzD were quantified in all mesocosms for 60 days. The highest abundance of all target genes was observed on the 6th day of treatment. trzD was observed in the bioaugmented mesocosms only. The bacterial community profile in all mesocosms was monitored by LH-PCR over a period of two months. Results indicate that the communities changed rapidly after inoculation, but there was no drastic change in microbial community profile after 1 month. Results indicated that efficient bioremediation of atrazine using a microbial consortium could be successfully up-scaled to pilot scale. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction reversal in male frogs (Hayes et al., 2003). Additionally, depending on the physicochemical properties of triazine herbicides, they can Herbicides help increase crop yield, but their accumulation in accumulate in sediments, severely affecting the organisms living in soil and water resources results in environmental pollution and the benthic zone (Sánchez-Sánchez et al., 2013). Because of its wide health hazards (Kanissery and Sims, 2011). Besides agricultural application and slow rate of natural degradation, it is detected land, the most affected areas are the pesticide manufacturing and in soil, sediments and groundwater at concentrations well above storage sites. The triazine herbicides are amongst the most widely the permitted limits. The European Union has banned atrazine, used pesticides in agriculture, wherein, atrazine is a broad-leaf based on its persistent contamination of groundwater (Prosen and weed control herbicide, widely used in agriculture of sugarcane Zupancic-Kralj, 2005). and corn. Atrazine (2-chloro, 4-ethylamino, 6-isopropylamino, Removal of herbicides from soil is mostly dependent on the 1,3,5-triazine), is reported to be an endocrine disrupter, affecting catabolic capacity of the soil microflora. Yet, bioremediation, a the endocrine system, the central nervous system and the immune sustainable method for clean-up, has not yet been fully exploited. system (Lasserre et al., 2009). It is also reported to cause sex Despite decades of research on bioremediation, it continues to be an area of priority research, since pollution levels are still high in many locations and success stories taking lab-scale research to the Abbreviations: BS, Basal salt; CA, Cyanuric acid; LH-PCR, Length heterogeneity field are few (Röling et al., 2004; Jørgensen et al., 2000). Especially PCR; HEX, Hexachloro fluorescein; FAM, Fluorescein amidite. fi fi * Corresponding author. Tel.: þ91 712 2249883. eld scale studies of bioaugmentation with ef cient contaminant E-mail addresses: [email protected], [email protected] (A. Kapley). degraders are rare (Jorgensen, 2007). We have earlier shown that http://dx.doi.org/10.1016/j.jenvman.2014.02.016 0301-4797/Ó 2014 Elsevier Ltd. All rights reserved. S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216 209 atrazine degradation in tropical agricultural soil can be enhanced 16S rDNA as Arthrobacter sp. AK_YN10 (Accession no. HE716859), significantly by biostimulation and by augmentation with known Pseudomonas sp. AK_AAN5 (Accession no. HE716858), Pseudomonas atrazine degrading bacterial strains (Sagarkar et al., 2013). sp. AK_CAN1 (Accession no. HE716855). Designing a successful bioremediation program involves com- Ring-labelled 14C e atrazine mineralization potential of the parison of different treatment strategies for removal of the consortium was checked by radiorespirometric analysis. The pollutant in a defined time scale. However, if degradation is not method involved incubation of 106 cfu ml 1 g 1 of each bacterial complete, it could lead to the accumulation of intermediates which strain with 2 kBq of ring-labelled 14C-atrazine [specific activity may also be toxic. Hence, it is not only important to monitor the 160 mCi/mmol; 99% radiochemical purity (Larodan Fine Chemicals disappearance of the pollutant but also to monitor the in- Ab, Sweden)] in 10 mL liquid BS media in 100 mL infusion bottles. In termediates of the degradation pathway. In the case of atrazine order to maintain a higher atrazine concentration, 100 mg L 1 of degradation, the most widely reported process is the dechlorina- pure non-labelled atrazine was also added to the bottles. Along tion route, where atrazine is converted to hydroxyatrazine. This is with this, the mineralization potential of the consortium was also initiated by the action of atzA gene (atrazine chlorohydrolase) or checked in 10 g soil. 2 kBq of 14C ring-labelled and 100 mg kg 1 of trzN gene (triazine hydrolase) (Krutz et al., 2010). Atrazine is pure non-labelled atrazine was added to this soil. All the bottles degraded to cyanuric acid by atzABC genes in gram-negative bac- were sealed and incubated at 30 C in the dark for up to 30 days at 14 teria (Iwasaki et al., 2007) and trzN-atzBC genes in gram-positive 100 rpm. The production of C labelled CO2 in the bottles was bacteria (Vibber et al., 2007). The intermediate cyanuric acid is trapped in 1 mL of 1 M NaOH, which was analyzed in a Winspectral further degraded by the atzD,E,F or trzD-atzE,F gene cluster to 1414 Liquid scintillation counter (Wallac Oy, Turku, Finland) using ammonia and carbon dioxide (Krutz et al., 2010). Ultima Gold high flash point LSC-scintillation cocktail (PerkinElmer, This study attempts to bridge this gap by scaling up bioreme- MA, USA). The counting efficiency of the instrument was deter- diation studies from microcosm level (Sagarkar et al., 2013)to mined at the start of the experiment by an external standard mesocosm levels using 100 kg soil mesocosms for bioremediation. method and determined to be 91.5% 0.5%. Bottles were aerated Atrazine was used as model pollutant and natural and enhanced after every two days with sterile air. The experiment was carried bioremediation was compared in treatments that mimicked history out in triplicate. of pollutant use versus contaminated soil without history of herbicide use. We monitored the presence of atrazine and its inter- 2.3. Soil sampling and characterization mediate, cyanuric acid in the mesocosms. The mesocosms were divided into two groups, each containing three treatment options: Soil samples were obtained from NEERI campus garden pre- natural attenuation (capacity of inherent soil microflora) and two mises, Nagpur, India. Soil was characterized as silty clay loam. enhanced bioremediation strategies, biostimulation (enhancing 700 kg soil was collected from three sites in the garden; each site concentration of nutrients in the soil) and bioaugmentation had three sub-sampling sites from which soil was mixed after (introduction of additional gene pool using bacterial inoculum in removing stones and roots to represent a homogenous sample. This the soil). The mesocosms were placed outdoors so they would be soil was then used for the mesocosm setup. Soil properties: 0.82% exposed to environmental conditions in the field. Bioaugmentation organic carbon, 0.81% total nitrogen, 0.005% total phosphate, was done with a bacterial consortium isolated from contaminated 7.67 kg hectare 1 available phosphates, pH-6.8. agricultural soil. Analytical tools were used to measure pollutant levels while molecular tools were used to study microbial popu- 2.4. Mesocosms assemblement lation dynamics and gene profile during the course of treatment. Mesocosms were setup using garden soil as described in Section 2. Materials and methods 2.3. Soil was sieved through 2 mm sieve into plastic trays. In this way a very homogenous batch of soil was obtained. The sieved soil 2.1. Chemicals was separated into two sets of 300 kg each. One set was mixed with commercial grade atrazine (Dhanuka, India) at a concentration of Analytical-grade atrazine (purity, 99%; water solubility at 20 C, 300 mg kg 1 of dry soil. From here, 100 kg soil aliquots were 33 mg L 1), cyanuric acid (purity, 99%) were purchased from Sigma, transferred into three wide mouth plastic tanks (700 mm diameter USA. [14C] ring labelled atrazine (specific activity 160 mCi/mmol; and 600 mm height) having 200 L water holding capacity (to give radiochemical purity, 99%) was purchased from Larodan Fine 100 L headspace to carry out aerobic biodegradation). These three Chemicals Ab, Sweden. mesocosms will be henceforth referred to as ‘History’ mesocosms. The other set of 300 kg soil was further divided into 100 kg aliquots 2.2. Consortium designing and mineralization of [U-ring-14C] and transferred into three plastic tanks as above. No atrazine was atrazine mixed here at this time point. These are the ‘No-History’ mesocosms. All six mesocosms were kept in the garden premises for a Isolation of bacteria was carried out from a soil sample collected period of one month, under ambient environmental conditions from an agricultural field Yadamavala, Maharashtra, India, where (about 30 C during the day and 20 C during the night; post- sugarcane was cultivated and atrazine was used for the last three monsoon season in central India). Moisture content of the soil years. The site is described in Sagarkar et al. (2013). Based on pre- was monitored every week and water content was maintained at 1 liminary growth data on BS media (Basal media) 0.9 g L KH2PO4, 40% until the end of the experiment with sterile water. 1 1 6.5 g L Na2HPO4:12H2O, 0.2 g L MgSO4:7H2O(Cai et al., 2003) After a period of one month, the bioremediation program was containing atrazine, 40 isolates were selected for screening for initiated by mixing commercial grade atrazine at a concentration of atrazine degradation. Degradation was monitored by HPLC (data 300 mg kg 1 of dry soil in each of the six mesocosms. The two not shown) and five isolates were selected for consortium design. groups, i.e. ‘History’ and ‘No-History’ consisted of three mesocosms Mineralization of atrazine using 14C ring-labelled atrazine was each, which were subjected to three different treatments as follows: tested for all five isolates and based on growth rate and degradation profile (data not shown); a three-membered consortium was 1. BA: Bioaugmentation, by addition of a consortium comprising designed. The isolates were identified by sequence analysis of their of; Arthrobacter sp. AK_YN10, Pseudomonas sp. AK_AAN5 and 210 S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216

Pseudomonas sp. AK_CAN1. Cultures were pre-grown in 0.1 X (R2 ¼ 0.996), log (16S rDNA) ¼0.29 Ct þ 13.61(R2 ¼ 0.999). The qPCR Luria Broth (HiMedia, India) containing 100 mg L 1 atrazine. reaction was carried out in an iCycler (Bio-Rad, USA). Each 25 mlre- After 18 h of growth at 30 C, absorbance was measured at A600. action constituted of 50 pmol of each primer, 12.5 mlofMaximaÔ Ò The total bacterial count inoculated was 106 cfu per gram of soil SYBR Green qPCR Master Mix (Bio-Rad) which includes SYBR Green I 5 per mesocosm (viz. 3.5 10 cells per bacterial isolate per dye, dNTP, KCl, (NH4)2SO4 and MaximaÔ Hot Start TaqDNA Poly- gram). Culture was harvested by centrifugation, cell pellet was merase. 20e100 ng template was added in each reaction. qPCR washed with buffer and re-suspended in 2 L of 20X BS media cycling conditions were; initial denaturation for 10 min at 95 Cfol- and was inoculated in the BA mesocosms by spraying. Thus, two lowed by 40 cycles of denaturation at 95 C for 45 s, annealing for 45 s mesocosms were treated by bioaugmentation, one from History at specific temperature and extension at 72 C for 45 s; and final group and one from No-History group. extension of 72 C for 10 min. Melting curve analysis was performed 2. BS: Biostimulation; by addition of modified BS media. BS media at the end of each run from 60 Cto98C. qPCR end products were described above was supplemented with 1 g L 1 sodium citrate analyzed in 1.5% agarose gel at 120 V, 1 h. The bands were visualized as additional carbon source. 20 X media was prepared and 2 L of by ethidium bromide. this was added to the mesocosm so that the final concentration of citrate would be 0.4 mg g 1 soil. Two mesocosms were 2.7. Bacterial community fingerprinting by length heterogeneity-PCR treated by biostimulation, one from History group and one from No-History group respectively. 2.7.1. LH-PCR and capillary electrophoresis 3. NA: Natural Attenuation, where only sterilized water was added Length heterogeneity PCR (LH-PCR), developed by Suzuki et al. to the mesocosm to maintain water content at 40% (w/w). Again, (1998), utilizes the natural length variance in the 16s rRNA gene two mesocosms were treated by natural attenuation, one from between different bacterial taxa. First, the VIeVIII section of the 16s History group and one from No-History group. rRNA gene, containing the most sequence variance in the 16s rRNA gene, is amplified using a primer labelled with fluorescent marker A diagrammatic illustration of mesocosm assembly is depicted FAM (6-carboxyfluorescein). The PCR products are then mixed with in Supplementary Fig. 1. a known size standard labelled with HEX (Hexachloro-fluorescein) 100 g aliquots of soil from each mesocosm were sampled peri- as internal size standard. The mixture is then separated in a odically for a period of two months. Sampling was carried out in capillary gel electrophoresis, and the fluorescent labels are detected three different points in each mesocosm, using a core sampler. Soil by laser. from all three locations within a mesocosm, was mixed and 3 sets of Soil DNA was used as template in PCR amplifying the three first 10 g aliquots were used for further analysis. The remaining soil was hyper variable regions of the 16s rRNA gene. PCR master mix con- put back into the mesocosm right away. These samples were used tained 10 pmol of universal primers fD1 (50-AGAGTTTGATCCTGGCT- for chemical and/or molecular analysis. All analysis was carried out CAG-30) and 518PRUN (50-FAM ATTACCGCGGCTGCTGG-30), 0.2 pmol in triplicate. For molecular analysis, extracted DNA samples were of dNTP (Fermentas, Vilnius, Lithuania), 2.5 mM MgCl2,10mgBSA stored at 20 C until used. (New England Biolabs, MA, USA) and 1 unit Fermentas Taq polymerase in 1 buffer (Fermentas, Vilnius, Lithuania). PCR amplifica- 2.5. Nucleic acid extraction from soil and control strains tion conditions were 10 min of initial denaturation at 95 C, followed by 25 cycles of denaturation at 95 C for 45 s, primer annealing at Control strains Pseudomonas citronellolis ADP (DSM 11,735) and 55 C for 1 s, extension at 72 Cfor2minandafinal extension step of Arthrobacter aurescens TC1 (ATCC BAA-1386) were obtained from 15 min at 72 C. The size of the PCR products was checked on 1.5% culture collections. Both bacteria were cultivated on BS medium agarose gel. The fine differences in PCR product sizes resulting from adapted from Cai et al. (2003). Total genomic DNA was extracted natural length variance in 16s rRNA gene was resolved by poly- (Kapley et al., 2007) and 50 mL aliquots were stored at 20 C until acrylamide capillary electrophoresis as described earlier (Mikkonen required. This was used as positive control in all PCR/Real-time PCR et al., 2011). The optimal signal/noise ratio was obtained by diluting reactions. The atrazine degradation genes atzA, atzB, atzD, trzN, and the PCR products 1:10 with water.1 ml of the diluted PCR product was trzD were amplified using the primers in Supplementary Table 1. mixed with Hi-Di Formamide (Applied Biosystems), and a 50HEX e Soil DNA was prepared from 100 mg soil in 3 replicates using labelled size standard, containing PCR product from three different a FastSpin Kit for soil DNA extraction (M.P. Biomedicals, France). bacterial strains, was added to each sample (Tiirola et al., 2003). The DNA from all replicates was pooled to represent a homogeneous final volume of capillary electrophoresis mixture was 15 ml. Samples mixture. 100 mL aliquots were stored at 20 C until required. were denatured by heating to 98 C for 2 min, and run through a 47 cm capillary containing POP-6 polymer on an ABI PRISM 310 Ge- 2.6. Preparation of qPCR standards and quantification methodology netic Analyzer. Sample injection time was 10 s, run voltage 15 kV, temperature 60 C and run time 70 min. Quantification of atrazine degrading genes was carried out as described in Sagarkar et al. (2013). trzN gene was amplified from 2.7.2. Community profile processing and analysis standard culture Arthrobacter aurescens TC1 (ATCC BAA-1386) and The raw data of the HEX and FAM traces from each sample was atzA, atzB, atzD and 16S rDNA was amplified from P. citronellolis imported as 12-bit densitometric curves by the Curve Converter while Pseudomonas sp. AK_CAN1 was used to amplify trzD gene. operation to BioNumerics 5.0 (Applied Maths, Sint-Martens-Latem, PCR products were cloned and standard dilution series of the genes Belgium). Sample FAM curves were normalized by the HEX labelled were prepared as described in the instructions of Plasmid Standard internal standards. The fragment lengths smaller than 450 bp and Construction by Applied Biosystems, available online at http:// greater than 550 bp were excluded in order to only include ex- www.appliedbiosystems.com/support/tutorials/pdf/quant_pcr.pdf. pected fragment lengths. qPCR reaction was optimized by testing different annealing tem- Clustering analysis was performed to see which treatments had peratures. The calibration curve equations for the primers are; log an effect on the bacterial community change. Also the similarities (atzA) ¼0.29 Ct þ 10.66 (R2 ¼ 0.994), log (trzN) ¼0.36ct þ 12.46 between two parallel DNA extractions were checked by clustering (R2 ¼ 0.994), log (atzB) ¼0.34ct þ 13.14 (R2 ¼ 0.992), log all samples together. Clustering was calculated with pairwise (atzD) ¼0.35ct þ 12.19 (R2 ¼ 0.994), log (trzD) ¼0.42ct þ 10.64 clustering method using Ward’s clustering and Pearson correlation S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216 211 coefficient. No signal detection threshold was applied, and opti- Table 1 mization of 0.30% was allowed to include 1 bp shifts between two Atrazine degrading genes present in the bacterial strains isolated in this study. profiles. Sr. no. Isolate name Atrazine degrading genes

Differences in bacterial diversities were visualized as curve- atzA atzB atzC atzD atzE atzF trzN trzD based fingerprints. The curve profiles of the two parallel DNA ex- 1 Arthrobacter aurescens e þþeeeþ e tractions were averaged using the Create averaged fingerprint (NC008711)a script in BioNumerics. The data was exported in MS Excel and 2 Arthrobacter sp. AK_YN10 e þþeeeþ e normalized by the total fluorescence. (HE716859) Shannon’s index of diversity (H0) was calculated from the 3 Bacillus sp. AK_CAN2 eeeeeeee normalized fluorescence data of the densitometric curves as in (HE716856) 4 Pseudomonas citronellolis þþþþþþee Mikkonen et al. (2011). Shortly, each rise in the curve which (AM088478)a continued at least for 0.5 bp was interpreted as a unit of diversity, 5 Pseudomonas sp. AK_AAN5 þþþþþþee and the sum of this rise was compared with the total sum of all the (HE716858) rises in calculating H0. 6 Pseudomonas sp. AK_CAN1 eeþ eeeeþ (HE716855) 7 Variovorax sp. AK_AAN2 eeeeeeee 2.8. Atrazine extraction from soil (HE716857)

þ: detected; e: not detected. Soil aliquots of 10 g were shaken with 20 mL of methanol water a Control strains. (4:1 v/v) for 1 h. The supernatant was decanted after the mixture was centrifuged at 10,000 rpm. This procedure was repeated two fi more times. Combined extract was then concentrated to 15 mL Based on the gene pro le, three isolates, Arthrobacter sp. AK_YN10, using a rotary evaporator. Concentrated extract was then acidified Pseudomonas sp. AK_AAN5 and Pseudomonas sp. AK_CAN1 were to pH 1 with HCl and partitioned three consecutive times with chosen as members of the consortium to be used in bio- chloroform (30 mL). The combined organic phases were again augmentation. As can be seen from Table 1, two isolates, Bacillus sp. evaporated to near dryness, and then re-dissolved in 10 mL of AK_CAN2 and Variovorax sp. AK_AAN2 did not show the presence of methanol. Extracted atrazine can be quantified by HPLC. any target genes and hence were not used as consortium members. Pseudomonas sp. AK_AAN5 demonstrated the presence of the 2.9. HPLC analysis complete atzA,B,C,D,E,F pathway and hence would be a good candidate for bioaugmentation. The Arthrobacter isolate contained A Perkin Elmer high-performance liquid chromatography a different (analogous) pathway for atrazine degradation viz. the (HPLC) system equipped with a variable-wavelength photodiode trzN, atzB, atzC route, which converts atrazine to cyanuric acid. array detector (PDA) (Perkin Elmer, USA) was used for detection. Strain AK_CAN1 demonstrated the presence of the atzC gene, which Separation was performed with a reversed-phase column (Lichro- is responsible for the conversion of the atrazine degradation inter- sphere RP18; size: 250 mm 4.6 mm I.D.; particle size: 5 mm). The mediate, ammelide (or isopropylammelide) to the central metabolic system consisted of two model pumps to drive the flow rate of the intermediate, cyanuric acid and the trzD gene that is responsible for mobile phase. For cyanuric acid detection, the flow rate was further degradation of cyanuric acid. To encompass diverse degra- 0.5 mL min 1 [methanol/0.1 M potassium phosphate buffer (pH dation enzymes, all three isolates were chosen for bioaugmentation. 7) ¼ 30:70, v:v] and the variable-wavelength UV detector was set to Before application of the designed consortium into the meso- fi 218 nm (Sadowsky et al., 1998). For atrazine detection, the flow rate cosms, the ef ciency of the consortium was tested in BS media, as -1 was 0.5 mL min 1 (methanol/water ¼ 70:30, v:v) and the variable- well as, in soil amended with 100 mg kg atrazine. Approximately wavelength UV detector was set to 220 nm (Wang et al., 2009). 80% ring labelled atrazine was utilized in 14 days when atrazine Spectra were referenced against a reference wavelength of 470 nm was used as the sole carbon and nitrogen source in both BS media and compared to those obtained with standard cyanuric acid and and soil. Results can viewed in Supplementary Fig. 2 that demon- atrazine. Concentration of cyanuric acid and atrazine were quan- strates the mineralization capacity of the designed consortium. tified by integrating peak areas of different concentration of the purified compound. 3.2. Comparison of treatment efficiency in mesocosms

2.10. Statistical analysis Three different bioremediation strategies were setup to study atrazine degradation in soil; bioaugmentation (BA), biostimulation The statistical analysis of the experimental data was carried out (BS) and natural attenuation (NA), in 100 kg soil mesocosms. All with Sigma Plot, version 11.2 (Systat Software). One-way analysis of three treatments were tested against two conditions, History variance (ANOVA) was used for determining statistical significance (simulated by addition of the herbicide one month prior to starting of the treatments. the treatments) and No-History (treatment started directly on garden soil). Disappearance of atrazine was the yardstick used as 3. Results the measure of treatment efficiency. However, monitoring only atrazine removal does not guarantee complete degradation, hence 3.1. Consortium development and monitoring efficiency we also monitored the central metabolic intermediate of the atrazine degradation pathway, cyanuric acid. Additionally, we 40 bacterial strains were isolated from soil of a sugarcane field quantified the atrazine degrading gene pool in each of the six where atrazine was used in weed control. Screening of isolates for mesocosms to thoroughly analyze the bioremediation process in atrazine degradation potential was analyzed by HPLC (data not each treatment. shown). Five isolates were selected and were tested for the presence of genes reported in the atrazine degradation pathway. Table 1 lists 3.2.1. Removal of atrazine the profile of genes detected by PCR amplification in all five isolates The degradation pattern in the different treatment strategies is and two control strains (P. citronellolis and Arthrobacter aurescens). demonstrated in Fig. 1. Zero h time point indicates the start of the 212 S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216

Fig. 1. Monitoring bioremediation efﬁciency in different treatments with respect to degradation. 1A) Atrazine degradation in mesocosms with respect to various time points. Atrazine concentration was represented as mg kg 1 of soil. 1B) Cyanuric acid degradation in mesocosm with respect to various time points. Cyanuric acid concentration was represented as mg kg 1 of soil.

treatment process, in both History and No-History sets of meso- cyanuric acid present in soil mesocosms. The zero h time point cosms. As shown in the figure, the atrazine levels in the History indicates the level of cyanuric acid at the start of the treatment mesocosms were higher than the amount added in soil for treat- process. In both sets, ‘History’ and ‘No-History’, up to 15 days, the ment. This is because the soil in all three History mesocosms was concentration of cyanuric acid detected was roughly similar in all treated with atrazine one month prior to the treatment process, to three treatments. Over time, the concentration levels changed ac- generate the condition of “history of pesticide use”. cording to atrazine degradation. Comparing treatment processes in Both History and No-History mesocosm groups demonstrated both groups of mesocosms, History and No-History, Fig. 1B shows, the degradation of atrazine within one month of treatment. that cyanuric acid levels increased till the 30th day of treatment However, as demonstrated in Fig. 1A, soil previously exposed to and subsequently decreased in the case of enhanced bioremedia- atrazine showed better efficiency of degradation. Comparing bio- tion (BA and BS). However, in the case of natural attenuation, even augmentation in History and No-History treatments, we observed after 60 days of treatment, the concentration of CA did not that 90% of atrazine was degraded in 6 days when soil has been decrease. This pattern is indicative of degradative efficiency in previously exposed to the herbicide, while in the case of No- enhanced bioremediation process. In the History group, bio- History, it took 15 days to remove 90% atrazine. Similarly, bioaugmentation treatment demonstrated 150 mg kg 1 CA accumu- stimulation treatment with History demonstrated 70% atrazine lation in one month, which degraded to 42 mg kg 1 in the 2nd degradation in 6 days, while only 37% atrazine was removed in the month of treatment. Under treatment of biostimulation, ‘No-History’ biostimulation treatment mesocosm in 6 days. In the 264 mg kg 1 CA accumulated in 30 days and reduced to natural attenuation treatment, 94% atrazine was degraded in 30 115 mg kg 1 in 60 days. Across the No-History group, 49 mg kg 1 days mesocosms with History condition, while with No-History CA accumulated in the case of bioaugmentation, in 30 days and treatment, 77% degradation was observed. reduced to 20 mg kg 1 in 60 days. In the case of biostimulation, For all the treatments, one way ANOVA was performed, P value 127 mg kg 1 CA was accumulated in 30 days and reduced to <0.005 indicated statistically significant differences amongst the 20 mg kg 1 in 60 days. treatment groups. Mesocosms with natural attenuation treatments in both study groups demonstrated no reduction of CA in 2 months. Removal of 3.2.2. Removal of cyanuric acid (CA) CA from the bioaugmentation and biostimulation treatments Cyanuric acid is the central intermediate in the atrazine degra- indicated efficient atrazine degradation without accumulation of dation pathway. Ring opening occurs at this stage and atrazine is major intermediates. mineralized, and hence, its levels were monitored to check for A P-value <0.005 indicated statistically significant difference in complete removal of atrazine. Fig. 1B demonstrates the levels of CA accumulation amongst the treatment groups. S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216 213

Table 2 Quantiﬁcation of atrazine degrading genes during the bioremediation process.

Target genes, copy no. g 1 of soil

Mesocosms atzAa trzNa atzBa atzDa trzDa 16Sa

Control soil sample 6.66Eþ03 ND 9.28Eþ03 2.62Eþ03 ND 1.53Eþ10 History mesocosms BA Zero h 3.22Eþ05 8.20Eþ05 5.00Eþ06 3.92Eþ03 ND 4.30Eþ10 6th day 4.64Eþ07 4.64Eþ08 1.93Eþ07 3.39Eþ03 3.39Eþ05 5.11Eþ07 15th day 1.23Eþ05 1.34Eþ07 6.74Eþ06 6.43Eþ03 6.43Eþ04 5.08Eþ10 30th day 1.43Eþ05 1.82Eþ07 2.90Eþ06 6.09Eþ03 1.09Eþ04 2.98Eþ10 60th day 4.38Eþ04 6.40Eþ05 2.93Eþ05 4.52Eþ03 4.56Eþ03 4.15Eþ10 BS Zero h 1.08Eþ04 3.68Eþ05 2.53Eþ05 2.74Eþ03 ND 6.62Eþ10 6th day 1.02Eþ05 1.15Eþ07 5.97Eþ07 2.35Eþ03 ND 1.07Eþ08 15th day 2.54Eþ05 6.82Eþ07 6.02Eþ07 2.49Eþ04 ND 5.08Eþ10 30th day 1.21Eþ04 8.50Eþ06 4.63Eþ06 2.28Eþ03 ND 2.97Eþ09 60th day 4.32Eþ03 1.88Eþ07 6.61Eþ06 2.99Eþ03 ND 4.59Eþ10 NA Zero h 1.85Eþ03 4.16Eþ04 9.84Eþ05 2.93Eþ03 ND 1.48Eþ10 6th day 3.16Eþ04 2.80Eþ07 3.17Eþ07 1.01Eþ04 ND 6.68Eþ07 15th day 1.06Eþ04 2.04Eþ07 7.20Eþ06 3.23Eþ03 ND 8.09Eþ09 30th day 2.14Eþ05 6.82Eþ04 4.81Eþ04 2.09Eþ04 ND 4.75Eþ10 60th day 5.26Eþ03 1.64Eþ04 5.11Eþ04 1.51Eþ04 ND 3.29Eþ10 No-history mesocosms BA Zero h 3.34Eþ05 ND 2.39Eþ05 3.31Eþ04 ND 1.58Eþ10 6th day 2.76Eþ05 1.79Eþ08 7.51Eþ07 8.01Eþ03 8.01Eþ05 2.45Eþ08 15th day 8.34Eþ04 4.26Eþ07 2.03Eþ07 6.87Eþ03 6.87Eþ04 1.02Eþ11 30th day 3.06Eþ04 2.20Eþ07 5.81Eþ06 2.99Eþ03 1.99Eþ04 3.40Eþ10 60th day 5.90Eþ03 4.00Eþ04 8.49Eþ05 3.02Eþ03 8.34 þ 03 6.25Eþ08 BS Zero h 1.07Eþ04 ND 6.44Eþ03 2.99Eþ03 ND 2.98Eþ10 6th day 1.61Eþ04 7.96Eþ05 2.14Eþ05 4.26Eþ03 ND 7.14Eþ07 15th day 9.90Eþ03 4.34Eþ07 1.79Eþ07 3.36Eþ03 ND 6.62Eþ09 30th day 2.38Eþ04 5.02Eþ07 1.90Eþ07 4.11Eþ03 ND 1.99Eþ10 60th day 1.66Eþ05 6.94Eþ03 3.44Eþ05 1.78Eþ04 ND 2.22Eþ08 NA Zero h 2.50Eþ04 ND 1.37Eþ04 4.01Eþ03 ND 1.02Eþ11 6th day 1.36Eþ04 9.90Eþ05 3.43Eþ05 3.49Eþ03 ND 4.04Eþ07 15th day 8.56Eþ03 3.98Eþ06 1.38Eþ07 2.15Eþ03 ND 3.40Eþ10 30th day 3.10Eþ04 3.96Eþ05 1.38Eþ07 3.83Eþ03 ND 2.13Eþ10 60th day 3.81Eþ04 4.29Eþ03 4.22Eþ05 1.51Eþ04 ND 1.44Eþ08

Control soil sample was the control sample from history mesocosm before 1 month atrazine pretreatment. BA-Bioaugmentation, BS-Biostimulation, NA-natural attenuation. ND-Not detected. a Average of triplicates.

3.2.3. Dynamics of atrazine degrading genes in the mesocosms 3.2.4. Community profiling by LH-PCR The atrazine-degrading gene copy numbers for atzA,B,D, trzN,D The community profiles, represented in Fig. 2 were relatively and 16S rRNA genes were evaluated by qPCR. Table 2 demonstrates similar between different treatments. An in silico PCR of the con- their copy number per gram of soil with respect to time in ‘History sortium members revealed that the PCR product lengths from these and No-History’ mesocosms respectively. Control soil sample in- bacteria were 511 and 521 bp, respectively. Hence, increased dicates qPCR data of soil that was used to setup the mesocosms. occurrence of these two amplification products would mean the Results demonstrate that the copy number of the first two genes of dominance of the bioaugmented bacteria. However, results did not the atrazine degradation pathway, atzA/trzN and atzB, increased in show any dramatic difference between the occurrence of these PCR the 6th day of treatment. The bioaugmented mesocosms showed a fragments in the bioaugmented mesocosms, when compared to the higher gene titre. trzN was not detected in the soil initially, but biostimulated and naturally attenuated mesocosms. This indicated could be detected in History mesocosms from zero h. The copy that with low pressure of atrazine after the 1st and 2nd month, number increased at day 6 from 105 to 108 g 1 soil. In No-History either the consortium was not dominating, or that the mesocosm mesocosms, trzN was not detected at zero h, but was detected at soils used in all treatments already contained a large number of a high concentration (108) on day 6. Analysing the copy mumbers bacteria producing this same PCR fragment. obtained for genes responsible for the first two steps of atrazine Pairwise comparison of similarities between different meso- degradation (atzA, trzN, atzB) we observed higher copy numbers in cosms revealed that the grouping factor was the previous atrazine History mesocosms on the 6th day of the study. The copy numbers exposure rather than different soil treatments (Fig. 3). When gradually decreased as the treatment progressed. In the History community profiles having the same bioremediation treatments group, on the 60th day, the copy numbers decreased 100e1000 but different atrazine exposure histories were compared as pre- fold. Two exceptions were atzA in NA mesocosm and atzB in BS sented in Fig. 2, some differences between mesocosms could be mesocosm. The decrease was less significant in the No-History seen. A consistent, sharp peak at 470 bp was observed in all sam- group of mesocosms, where there was no change in copy ples. Its intensity seemed to be more pronounced in mesocosms numbers of atzA and atzB genes in the NA mesocosm. with previous atrazine history. A collection of low peaks at around Genes responsible for the ring-opening of cyanuric acid, atzD 480 bp was inconsistent, except in the naturally attenuated mes- and trzD, demonstrated a uniform pattern throughout the study. ocosms, where a single, relatively high peak was observed. Three The copy numbers of atzD genes remained approximately constant wide peak areas were seen between 485e505 bp, 510e530 bp and up to 60 days, while trzD was detected only in bioaugmented 530e540 bp. In LH-PCR profile as the parallel samples always mesocosms in both study groups. grouped together in pairwise comparison, the bias stemming from 214 S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216

Fig. 2. Microbial community dynamics in the mesocosms monitored by LH-PCR during the bioremediation treatments. Differences in bacterial diversities were visualized as curve- based fingerprints. The curve profiles of the two parallel DNA extractions were averaged using the Create averaged fingerprint script in BioNumerics. Each profile is the average of duplicates normalized by the total fluorescence intensity. sample handling could be eliminated (data not shown). Shannon’s indicating enrichment of an atrazine degrading population. The diversity indices calculated from different treatments show that diversity value for zero hour with no atrazine history was less than there was minor difference in species richness and evenness be- the zero hour with atrazine history. tween different mesocosms after 1 and 2 months (Supplementary Table 2). However, a clear difference could be seen between zero 4. Discussion hour samples and mesocosm samples. Zero hour samples were less diverse and mesocosm samples were more diverse and species The key factors influencing any bioremediation process are the evenness at zero hour was less than that of mesocosm samples, functional capacities of the soil microflora, and the environmental

Fig. 3. A) Cluster analysis of the LH-PRC profiles of the mesocosms. Clustering was done calculating pairwise similarities based on the curves, using Pearson correlation. Comparison was optimized to 58% (1 bp). Comparison method was Ward. N-No-History mesocosm, M-History mesocosm, 1 and 2 are the time points in months, B) Three-dimensional multidimensional scaling ordination for the 14 LH-PCR profiles of White ¼ No-History, 1st and 2nd month, red ¼ History Mesocosms, 1st and 2nd month, Yellow ¼ zero points (atrazine and without atrazine). The observed groups (Roman numbers) contain: (I) No-History group (II) History group. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216 215 conditions in soil. Soil contains a large heterogeneity of species and the amount of genes required for ring opening of cyanuric acid, metabolic routes which get adapted to new situations and com- atzD/trzD. This would translate into faster degradation of atrazine munity structures can change rapidly when new molecules acting compared to cyanuric acid, leading to accumulation of CA. These as energy source appears in their vicinity. For this reason, some- results highlighted two important factors; (i) the importance of times bioremediation treatments are successful without any addi- monitoring the intermediate along with the pollutant and (ii) tion of bacteria (Megharaj et al., 2011). To assess all these factors in enhanced bioremediation worked more efficiently than natural this study on atrazine bioremediation, three bioremediation stra- bioremediation. tegies; bioaugmentation, biostimulation and natural attenuation Atrazine degrading genes have a fairly recent evolutionary his- processes were compared. We carried out bioremediation treat- tory (UdiKolvic-Kolic et al., 2012). Despite this, the genes have been ments in 100 kg mesocosms to scale up microcosm studies and found globally (Krutz et al., 2007). Therefore it was not unexpected placed them outdoors to expose them to ambient environmental to find atrazine degrading genes in the NA mesocosm, this also conditions. To simulate field conditions, we also added a parameter indicates that native soil also developed an atrazine degrading of history. community when it was amended with atrazine. An efficient bioremediation strategy should ensure that neither In order to understand the dynamics of microbial communities the pollutant, nor its intermediate should remain in soil after in different bioremediation treatments, we studied their effects on treatment. Hence, the efficiency of bioremediation across different bacterial diversity based on natural length variation in 16S rRNA treatment strategies was analyzed by measuring the amount of gene. We chose this method for fingerprinting as our target was not atrazine and cyanuric acid remaining in the mesocosms. to identify individual bacterial species, but rather to monitor which Bioaugmentation was carried out by using bacterial isolates as of the bioremediation treatments would result in greatest diversity, consortium that was designed by keeping in mind the two major and whether the PCR products originating from the bacteria added reported dechlorination mechanisms in atrazine degradation. Even in the bioaugmentation treatment would be apparent in the com- though lab isolate Pseudomonas sp. AK_AAN5 contains all genes munity profiles. According to the parsimony tree describing the from the atz operon, we used two more strains in bioaugmentation. similarities between LH-PCR profiles from different mesocosm The first dechlorinating step in atrazine degradation is regulated treatments, the results suggest that the major factor influencing the by the gene products of atzA and trzN. Hence, we included both community’s structure was whether the soil had been pre-exposed bacterial isolates containing these genes; Pseudomonas sp. strain to atrazine, rather than the bioremediation treatment. The peak AK_AAN5 and Arthrobacter sp. strain AK_YN10 in our bio- profiles showed that the fresh addition of atrazine favoured most augmentation strategy. Besides, other reports suggest that the clearly a group of bacteria having a PCR product length of superior catalytic properties of trzN might be responsible for the approximately 470 bp with the primers used. This fragment prob- proliferation of this gene in agricultural soil and become predom- ably contained a group of bacteria which got adapted to atrazine inant over its analogue, atzA (Arbeli and Fuentes, 2010) which reasonably fast. The consortium members added in bio- would be beneficial in our bioaugmentation study. Strain AK_YN10 augmentation treatments would create LH-PCR peaks at 511 bp and demonstrated the presence of genes that convert atrazine to cya- 521 bp in the LH-PCR profiles. An increase in fragments of these nuric acid (Table 1). Further degradation is carried out by the gene lengths in bioaugmented mesocosms would be an indication of products of atzD or trzD (Sagarkar et al., 2013). Amplification data successful augmentation. However, no significant increase in 1 (Table 1) demonstrated that the only tested bacterial isolate con- month was seen, suggesting that either the augmented strains did taining either of these genes was Pseudomonas sp. AK_CAN1. Hence, not survive till this time in the soil, or that the soil already con- this isolate was also included in the consortium. Earlier reports tained a large proportion of these known degrader bacteria as the have demonstrated that the addition of citrate improves degrada- experiment started. Higher gene abundances (Table 2) of bio- tion efficiency of atrazine (Udikovic-Kolic et al., 2012; Sagarkar augmented mesocosms did not support the theory of rapid decline et al., 2013). This is probably because atrazine is preferred as a ni- of bioaugmented bacterial strains, however, degradation profiles trogen source, even though some bacteria can use atrazine as both suggest that the atrazine degradation rate was highest between carbon and nitrogen source. This is especially beneficial when 0 and 15 days and that the gene abundance responses were most degradation is being carried out in soil verses a pure culture study. pronounced at 6 days. Hence, it is possible that the bioaugmented Results of biostimulation prove that addition of citrate improves population was taxonomically similar to the natural population and degradation efficiency, when compared to natural attenuation, hence the bacterial profile was resilient and not disturbed. The where no nutrients have been added. community structures in biostimulated mesocosms were quite In this study, bioremediation was monitored using analytical as similar between soils having history and no history with atrazine, well as molecular tools. Fig. 1 demonstrates the residual levels of the only clear difference being the sharp peak at 470 bp. The mi- atrazine and CA in the mesocosms. We found that all treatments crobial communities in naturally attenuated mesocosms did not showed the disappearance of atrazine within one month of treat- change over between 1 and 2 months, which indicates that the ment with different rate of removal; BA > BS > NA. Comparing communities were quite stable at that time. It also seems that between the groups, History and No-History, we found that atra- atrazine use history of one month did not influence the microbial zine was degraded faster in History mesocosms, although atrazine community in the naturally attenuated treatment. completely disappeared in 60 days of treatment in both sets. Sur- The pretreatment of soil with atrazine affected the bacterial prisingly, atrazine was not detected even in the mesocosms using diversity, indicated by the Shannon’s diversity index at the begin- natural attenuation. However, as shown in Fig. 1B, the concentra- ning of the experiment. Surprisingly, with different bioremediation tion of cyanuric acid accumulated in the mesocosm soil indicated treatments, bacterial diversities increased with time. It is possible that degradation was not complete even after 60 days. The con- that the bioaugmentation at the beginning of the experiment centration of cyanuric acid is higher in the History mesocosms. This decreased the diversity at this point, and after a decline of the was because the amount of atrazine was higher and hence greater added inoculum the community developed. This is in accordance accumulation of the intermediate. When comparing atrazine with the findings of Cheyns et al. (2012), who also found that the degradation and cyanuric acid accumulation with gene abundances atrazine degrading populations decayed when atrazine application (Table 2), we observed that the amount of genes of the upper stopped in field plots. Bacterial diversity is generally considered to pathway, atzA, trzN and atzB were 1000 or 10,000 times higher than be beneficial for the functioning of the soil, as many different 216 S. Sagarkar et al. / Journal of Environmental Management 139 (2014) 208e216 species are required to contain a multitude of metabolic pathways. Iwasaki, A., Takagi, K., Yoshioka, Y., Fujii, K., Kojima, Y., Harada, N., 2007. Isolation b Often however, a xenobiotic enriches certain species which are able and characterization of a novel simazine-degrading -proteobacterium and detection of genes encoding s-triazine degrading enzymes. Pest Manag. Sci. 63, to degrade the contaminant, and this can be seen as a decrease in 261e268. diversity (Reardon et al., 2004). The structures of microbial pop- Jorgensen, K.S., 2007. In situ bioremediation. Adv. Appl. Microbiol. 61, 285e305. ulations are largely dictated by the kind of habitat they inhabit. The Jørgensen, K.S., Puustinen, J., Suortti, A.M., 2000. Bioremediation of petroleum hydrocarbon-contaminated soil by composting in biopiles. Environ. Pollut. 107, soil used in this study was perhaps complex enough to provide 245e254. niches for complex bacterial communities, which were further Kanissery, R.G., Sims, G.K., 2011. Biostimulation for enhanced degradation of her- enhanced with our bioremediation treatments. For extrapolating bicides in soil. Appl. Environ. Soil Sci., 1e10. http://dx.doi.org/10.1155/2011/ fi 843450. Published online. these results further to eld bioremediation, plot scale studies are Kapley, A., De Baere, T., Purohit, H.J., 2007. Eubacterial diversity of activated biomass needed. from a common effluent treatment plant. Res. Microbiol. 158, 494e500. Krutz, L.J., Shaner, D.L., Weaver, M.A., Webb, R.M., Zablotowicz, R.M., Reddy, K.N., Huang, Y., Thomson, S.J., 2010. Agronomic and environmental implications of 5. Conclusion enhanced s-triazine degradation. Pest Manag. Sci. 66, 461e481. Krutz, L.J., Zablotowicz, R.M., Reddy, K.N., Koger 3rd, C.H., Weaver, M.A., 2007. A very efficient microbial consortium consisting of 3 bacterial Enhanced degradation of atrazine under field conditions correlates with a loss e strains isolated from tropical agricultural soil was designed. The of weed control in the glasshouse. Pest Manag. Sci. 63, 23 31. Lasserre, J.P., Fack, F., Revets, D., Planchon, S., Renaut, J., Hoffmann, L., Gutleb, A.C., novelty of this consortium lies in the inclusion of isolates that Muller, C.P., Bohn, T., 2009. Effects of the endocrine disruptors atrazine and PCB possess different analogue degradation genes. The bioremediation 153 on the protein expression of MCF-7 human cells. J. Proteome Res. 8, 5485e process was monitored using both genomic as well as analytical 5496. Megharaj, M., Ramakrishnan, B., Venkateswarlu, K., Sethunathan, N., Naidu, R., 2011. tools. Genes involved in degradation of atrazine were identified in Bioremediation approaches for organic pollutants: a critical perspective. Envi- the members of the consortium and quantified in the mesocosms. ron. Int. 37, 1362e1375. The bacterial community profile was mapped for two months and Mikkonen, A., Lappi, K., Wallenius, K., Lindström, K., Suominen, L., 2011. Ecological inference on bacterial succession using curve-based community fingerprint demonstrated that after one month, bioaugmentation did not data analysis, demonstrated with rhizoremediation experiment. FEMS Micro- disturb the natural microflora. The study demonstrated successful biol. Ecol. 78, 604e616. scale-up of microcosm experiments into pilot scale treatments. Prosen, H., Zupancic-Kralj, L., 2005. Evaluation of photolysis and hydrolysis of atrazine and its first degradation products in the presence of humic acids. Environ. Pollut. 133, 517e529. Acknowledgements Reardon, C.L., Cummings, D.E., Petzke, L.M., Kinsall, B.L., Watson, D.B., Peyton, B.M., Geesey, G.G., 2004. Composition and diversity of microbial communities recovered from surrogate minerals incubated in an acidic uranium- Funding from Department of Biotechnology, Ministry of Science contaminated aquifer. Appl. Environ. Microbiol. 70, 6037e6046. and Technology, New Delhi, to Dr. Atya Kapley for the Indo-Finland Röling, W.F., Milner, M.G., Jones, D.M., Fratepietro, F., Swannell, R.P., Daniel, F., project (G-1649) and from the Academy of Finland to Dr. Kirsten Head, I.M., 2004. Bacterial community dynamics and hydrocarbon degradation fi fl Jørgensen (project no. 124822) are acknowledged. Sneha Sagarkar during a eld-scale evaluation of bioremediation on a mud at beach contaminated with buried oil. Appl. Environ. Microbiol. 70, 2603e2613. is grateful to the Council of Scientific and Industrial Research (CSIR), Sadowsky, M.J., Tong, Z., de Souza, M., Wackett, L.P., 1998. AtzC is a new member of India for the award of a senior research fellowship. Aura Nousiainen the amidohydrolase protein superfamily and is homologous to other atrazine- e is grateful to the EnSte Graduate School for a fellowship. metabolizing enzymes. J. Bacteriol. 180, 152 158. Sagarkar, S., Mukherjee, S., Nousiainen, A., Björklöf, K., Purohit, H.J., Jørgensen, K.S., Kapley, A., 2013. Monitoring bioremediation of atrazine in soil microcosms Appendix A. Supplementary data using molecular tools. Environ. Pollut. 172, 108e115. Sánchez-Sánchez, R., Ahuatzi-Chacón, D., Galíndez-Mayer, J., Ruiz-Ordaz, N., Sal- merón-Alcocer, A., 2013. Removal of triazine herbicides from aqueous systems Supplementary data related to this article can be found at http:// by a biofilm reactor continuously or intermittently operated. J. Environ. dx.doi.org/10.1016/j.jenvman.2014.02.016. Manage. 128, 421e426. Suzuki, M., Rappé, M.S., Giovannoni, S.J., 1998. Kinetic bias in estimates of coastal Picoplankton community structure obtained by measurements of small-subunit References rRNA gene PCR Amplicon length heterogeneity. Appl. Environ. Microbiol. 64, 4522e4529. Arbeli, Z., Fuentes, C., 2010. Prevalence of the gene trzN and biogeographic patterns Tiirola, M.A., Suvilampi, J.E., Kulomaa, M.S., Rintala, J.A., 2003. Microbial diversity in among atrazine-degrading bacteria isolated from 13 Colombian agricultural a thermophilic aerobic biofilm process: analysis by length heterogeneity PCR soils. FEMS Microbiol. Ecol. 73, 611e623. (LH-PCR). Water Res. 37, 2259e2268. Cai, B., Han, Y., Liu, B., Ren, Y., Jiang, S., 2003. Isolation and characterization of an Udikovic-Kolic, N., Scott, C., Martin-Laurent, F., 2012. Evolution of atrazine atrazine-degrading bacterium from industrial wastewater in China. Lett. Appl. degrading capabilities in the environment. Appl. Microbiol. Biotechnol. 96, Microbiol. 36, 272e276. 1175e1189. Cheyns, K., Martin-Laurent, F., Bru, D., Aamand, J., Vanhaecke, L., Diels, J., Merckx, R., Vibber, L.L., Pressler, M.J., Colores, G.M., 2007. Isolation and characterization of novel Smolders, E., Springael, D., 2012. Long term dynamics of the atrazine mineral- atrazine-degrading microorganisms from an agricultural soil. Appl. Microbiol. ization potential in surface and subsurface soil in an agricultural filed as a Biotechnol. 75, 921e928. response to atrazine applications. Chemosphere 86, 1028e1034. Wang, R., Jiang, Z., Ma, Y., Diao, Z., Zhang, Y., 2009. Research on how to determine Hayes, T., Haston, K., Tsui, M., Hoang, A., Haeffele, C., Vonk, A., 2003. Atrazine the residue of atrazine in the soilICBBE 2009, 3rd International Conference. induced hermaphroditism at 0.1 PPB in American leopard frogs (Ranapipiens): Bioinform. Biomed. Eng., 1e4. http://dx.doi.org/10.1109/ICBBE.2009.5163332. laboratory and field evidence. Environ. Health Perspect. 111, 568e575.