Μελέτη της συμπεριφοράς βιολογικών και συνθετικών γεωργικών φαρμάκων στο έδαφος και των επιπτώσεων τους στη μικροβιακή κοινότητα

Abstract

Synthetic pesticides play a major role in crop protection as a fast, easy and cheapmethod for pest control. However, their indiscriminate use has resulted in (1) efficacyloss due to the development of resistance by pests (2) toxicity to non-target organismsand (3) hazardous effects on the environment endangering the sustainability ofecosystems. Due to the environmental side effects and health concerns, manysynthetic pesticides have been banned or are under re-evaluation. Alternativeapproaches to pest control are urgently needed. In recent years, pesticides ofbiological origin have been developed as alternatives for pest control, while modernsynthetic pesticides have become safer and more target specific in their activity.Despite of general belief that their biological origin guarantee their environmentallysafety the effects of such pesticides on non-target organisms in general and on soilmicroorganisms in particular have not been studied extensively. Soil microorganismsplay a pivotal role in soil quality and plant productivity. Nowadays numerous newadvanced methods have allowed the in depth study of soil microbial communities.The use of high-throughput molecular techniques for investigating microbial diversityin soil communities provides new understanding of the distribution and diversity ofmicroorganisms in soil habitats. However, those methodological advances in soilmicrobial ecology have been hardly taken into account in the implementation of newlegislation and regulations regarding pesticide authorization. In the present thesis anumber of different pesticides both synthetic and of biological origin, were studiedwith regards to their fate in the environment and their impact on soil microbialcommunities.The first aim of the study was the optimization of Phospholipid Fatty Acid (PLFA)analysis in order to attain a precise picture of the size and the structure of the soilmicrobial community. Σhe effects of extraction solvent (chloroform vsdichloromethane) and buffer (phosphate vs citrate) on the yield and the profile ofPLFAs extracted from an acidic (pH 5.5) and an alkaline (pH 8.6) soil followingtwo2-h sequential extractions were determined. The yield and the profile of thePLFAs obtained separately by the first and the second 2-h extraction were comparedto identify the impact of sequential extractions on the PLFA extraction efficiency.Chloroform–citrate and a 2×2-h extraction maximized PLFA yields in both soils.Multivariate analysis of the data showed that the choice of the extraction mixture did not significantly influence the profile of the PLFAs obtained by the first 2-hextraction, whereas it had a profound effect on the profile of the PLFAs obtained bythe second 2-h extraction. Most PLFAs were extracted during the first extractionexcept from 18:2ς6,9 and 22:0 which were almost equally extracted by the twosequential extractions. The choice of organic solvent significantly influenced theprofile of the PLFAs extracted; their yield increased with chloroform with theexception of 18:2ς6,9 and 22:0 which were favored by dichloromethane. Overall, a2×2-h extraction with chloroform/methanol/citrate is expected to provide maximumPLFA yields. Subsequently, the optimized PLFA protocol was used for investigatingthe effects of Quillaja saponaria plant extract (QLAgri®) and Paecilomyces lilacinusstrain 251 (PL251, BIOACT®) on the microbial community comparatively to thesynthetic nematicides oxamyl and fluensulfone. No effects of QLAgri®, oxamyl andfluensulfone were observed. In contrast BIOACT® stimulated the growth ofcopiotrophic Gram negative bacteria and fungi which was attributed to individualcomponents of the formulation like glucose and semi skimmed milk which upon theirrelease in soil stimulated those microorganisms.Ammonia-oxidizing microorganisms (AOM) have been identified as a key groupregulating N cycle through their involvement in the nitrification process in soil. Theirpivotal role in N cycling, the availability of well-standardized tools to determine theiractivity and abundance in soil make them potential microbial indicators for use in theassessment of pesticide soil microbial toxicity. Despite all the above, little is knownabout the effects of pesticides on AOM. The impact of the low dose, high potencyherbicide nicosulfuron on the community structure and abundance of AOM wasinvestigated following a tiered pot-to-field experimental approach. This approachallowed the assessment of nicosulfuron toxicity i) at extreme exposure schemes (potexperiment, Tier I) invoked by the repeated application of a range of dose rates (x0,x10, x100 and x1000 the recommended dose) and ii) under realistic exposurescenarios (x0, x1, x2 and x5 the recommended dose) in the field (Tier II). Structuralchanges on AOM community were assessed via denaturating gradient gelelectrophoresis (DGGE) at the DNA level, while changes in abundance of AOM wereassessed via qPCR of the amoA gene. At extreme exposure schemes, nicosulfuroninduced changes only in AOB community structure, while AOA community structureand AOM abundance were not significantly affected. On the other hand under realisticfield exposure scenarios no specific effect on AOM was observed. The key role of arbuscular mycorrhizal (AM) fungi in maintaining soil fertility andecosystem functioning as well as their general responsiveness to pesticides make themgood candidates as microbial indicators in pesticide soil microbial ecotoxicityassessment. In a similar to the above pot-to-field experimental approach, the impact ofnicosulfuron on community structure of AM fungi was assessed i) at extremeexposure schemes (pot experiment, Tier I) and ii) under realistic exposure scenarios inthe field (Tier II). In the pot experiment, the highest dose rates (x100 and x1000)significantly reduced AM fungal richness as determined by DGGE. This coincidedwith the progressive accumulation of herbicide concentrations in soil. In contrast, noeffects on AM fungi were observed at the nicosulfuron dose rates tested in the field.Clone libraries developed for the pot experiment samples showed that the majority ofAM fungi belonging to the previously known Glomus group and were sensitive to thehigh levels of nicosulfuron accumulated in soil at the latter culture cycles. In contrast,a Paraglomeraceae and a Glomus etunicatum ribotype were present in maize roots inall cycles and dose rates implying a tolerance to nicosulfuron induced stress. Overall,the deleterious effects of nicosulfuron on AM fungi induced by the highest dose ratesin the pot experiment could be attributed either to fungal driven toxicity or to plantdriven effects which have subsequent implications for mycorrhizal symbiosis.Thiabendazole (TBZ), imazalil (IMZ), ethoxyquin (EQ), diphenylamine (DPA),and ortho-phenyl phenol (OPP) are pesticides used in fruit packaging plants for thecontrol of post-harvest fungal infestations and physiological disorders. Theirapplication result in the production of large wastewater volumes with high pesticideloads (50-200 mg L-1). These compounds were given authorization for use at the EUlevel under the clause that an efficient treatment of the produced wastewaters shouldbe operative at a local scale. Despite that, no established depuration method isoperative to date and their direct disposal into soil via land spreading is a commonpractice. Taking into account the key role of soil microbes in ecosystem functioningthe effects of these pesticides on the diversity and activity of soil microbes wereassessed in a microcosm study. In parallel, the dissipation of those pesticides wasdetermined in order to identify possible links between persistence and soil microbialtoxicity. The functional endpoints studied were microbial respiration, potentialnitrification, acidic and alkaline phosphatases, α-glucosidase, and fluoresceindiacetate (FDA), while the effect on the size and diversity of soil microbial biomasswas studied via phospholipid fatty acid (PLFAs) analysis. Pesticides dose rates (50 mg Kg-1) were selected based on a direct disposal scenario for wastewaters. Reliableanalytical methods were developed for the extraction and determination of pesticidesand their metabolites in soil. The dissipation of all pesticides was biphasic and wasadequately described by the hockey-stick model (π2< 15, r2> 0.75). IMZ and TBZshowed hight to moderate persistence (DT50 150.1 and 47.0 days, respectively) whileOPP and DPA rapidly dissipated (DT50 0.6 and 1.30 days, respectively). EQ itself wasnot persistent (DT50 2.2 days) and was rapidly transformed in soil to its mainoxidation derivatives quinone imine (QI) and 2,4-dimethyl-6-ethoxyquinoline(EQNL). Microbial respiration was not affected by pesticide application. DPA (44.6nmol/g) and OPP (45.9 nmol/g) induced significant decreases on total PLFA yieldscompared to the control (51.9 nmol/g). However, no effects on the PLFAs profile inthe different treatments were observed. Enzymatic activities were significantlyaffected by pesticides, however the effects did not show a clear temporal pattern. Incontrast, EQ and OPP significantly inhibited potential nitrification with the effect ofthe former being more prominent and persistent. Complementary experiments wereconducted in order to further investigate the effects of EQ and its oxidationderivatives on potential nitrification in soil in relation to their dissipation in soil. Themajor metabolite of EQ in soil was QI and appeared to be responsible for the drasticinhibition of potential nitrification observed in soil. In parallel the effects of thesepesticides on the structure of basidiomycetes and AOM in soil (DGGE-DNA) and onthe abundance of ammonia oxidizers both at DNA and RNA level were investigated,using DGGE-DNA, qPCR, and RTqPCR, respectively. Pesticides induced structuralchanges to basidiomycetes and AOM communities. The abundance of amoA genecopies was not significantly affected, while the abundance of amoA gene transcriptssignificantly decreased at the first three days after pesticide application suggesting aclear toxicity effect of EQ and its derivatives on AOM activity. Our results providefirst evidence that soil spreading of pesticide-containing wastewaters from the fruitpackaging plants could influence the size of soil microbial biomass and selectedmicrobial activities depending on their pesticide content.The findings of this thesis could be the basis for the establishment of a newcomprehensive assessment of the toxicity of pesticides on soil microbes. Such anassessment approach should include a) a tiered evaluation, including experimentationunder laboratory conditions and highly conservative exposure schemes (Tier I) andunder field conditions simulating a more realistic exposure step (Tier II), b) the use of new standardized molecular and biochemical methods which could provide a moreaccurate and in-depth picture of potential effects of pesticides on soil microbes at bothdiversity and functional level and c) the evaluation of the toxicity of both parentcompounds and main metabolites but also evaluation of the toxicity of the marketedformulations since our findings showed that pesticide metabolic products andcomponents of the formulation could have deleterious effects on soil microorganisms.