Assessing insecticide and fungicide effects on the culturable soil bacterial community by analyses of variance of their DGGE fingerprinting data

European Journal of Soil Biology 45 (2009) 466–472

Contents lists available at ScienceDirect

European Journal of Soil Biology journal homepage: http://www.elsevier.com/locate/ejsobi

Original article

Assessing insecticide and fungicide effects on the culturable soil bacterial community by analyses of variance of their DGGE fingerprinting data Enderson P. de B. Ferreira a, *, Andre´ N. Dusi b, Janaı´na R. Costa c, Gustavo R. Xavier c, Norma G. Rumjanek c a

´s, 74375-000 Goia ´ s, Brazil ˆ nio de Goia National Rice and Beans Research Center, BR GO-462, km 12, PO Box 179, SantoAnto National Vegetables Research Center, BR 060, km 09, Brası´lia, 70359-970 DistritoFederal, Brazil c National Agrobiology Research Center, BR 465, km 07, PO Box 74.505, Serope´dica, 23890-000 Rio de Janeiro, Brazil b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 September 2008 Received in revised form 7 July 2009 Accepted 8 July 2009 Available online 18 July 2009 Handling editor: Christoph Tebbe

To assess the effects of three insecticides (aldicarb, chlorpyrifos, deltamethrin) and two fungicides (tebuconazole and metalaxyl þ mancozeb) on the PCR-DGGE fingerprints of culturable soil bacterial communities (CSBC), a greenhouse experiment was carried out with soil samples from an Integrated System for Agroecological Production (ISAP), a Conventional Potato Production Area (CPPA) and a Secondary Forest Area (SFA) close to the CPPA. Samples were obtained at 15 day intervals starting at 32 until 77 days after sowing (DAS) to perform the PCR-DGGE analysis of the CSBC cultured on media amended with soil suspension. Analysis of variance from PCR-DGGE data indicated significant differences among treatments. Regardless the type of pesticide applied, CSBC was disturbed and similarity values varied from 5% to 90% in comparison to the control. Significant shifts on CSBC were only detected among treatments in the first two harvests, while CSBC tended to be more akin to each other at the last two harvest dates. The most significant responses observed were due to different soil sample origins, where values of 5% of similarity to the control were observed on CPPA soil. The use of analysis of variance on PCR-DGGE data was useful to a better understanding of the changes on CSBC induced by pesticides applications. Crown Copyright Ó 2009 Published by Elsevier Masson SAS. All rights reserved.

Keywords: Solanum tuberosum 16S rDNA Culture-dependent Bacterial community structure

1. Introduction A large number of pesticides is used to control potato pests in Brazil [6]. From those registered in the Ministry of Agriculture, 41 are for pest and 58 for disease control [2]. Although the use of pesticides is intended to provide satisfactory crop yields by controlling commonly occurring pests and disease in production fields, some may be toxic to the environment, as well as to humans. Reports on 320 chemicals registered for agricultural use in Brazil revealed that 18 insecticides and 5 fungicides, had potential risks to humans [10]. Adverse effects caused by pesticides are related to the central and peripheral nervous systems, in addition to elicitation of immunosuppressive or carcinogenic responses [14]. Some pesticides may accumulate throughout the food chain, affecting several trophic levels. The understanding of the impact a pesticide may cause to the environment is a complex issue, being necessary to observe the overall hierarchical chain, from a single molecule to the entire ecosystem, passing by the cell, the organism and the community [30]. * Corresponding author. Tel.: þ55 62 3533 2265; fax: þ55 62 3533 2100. E-mail address: [email protected] (E.P.deB. Ferreira).

The study of pesticide effects on non-target populations is an accepted strategy to evaluate its associated potential environmental risks. Among non-target populations, soil microorganisms are extremely important, since they play an essential role in nutrient turnover [3], maintaining generative capacity in agroecosystems [7]. The processes of ecological succession are, among other factors, mediated by microorganisms and depend on a fine balance of their population dynamics [23]. Under these circumstances, the impact inflicted on soil microbial populations caused by a specific pesticide is a potential indicator of the toxicity level of this product, and may represent a component of a broad study aiming to evaluate its potential impact on the environment [24]. Recently, microbial diversity has been studied through molecular methods, mainly by the analysis of the ribosomal genes, which are amplified by Polymerase Chain Reaction (PCR) and sequenced after cloning, or by the study of microbial community profiles obtained using molecular tools such as Random Amplified Polymorphic DNA – RAPD [16], Amplified Ribosomal DNA Restriction Analysis – ARDRA [42], Terminal Restriction Fragment Length Polymorphism – T-RFLP, RISA [19], Denaturing/Temperature Gradient Gel Electrophoresis – DGGE/TGGE [31] and Single-Strand Conformation Polymorphism – SSCP [38].

1164-5563/$ – see front matter Crown Copyright Ó 2009 Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.ejsobi.2009.07.003

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

DGGE fingerprints are commonly analyzed by cluster analysis (dendrogram) and/or non-parametric analysis, which pose great difficulty to establish whether clusters are significantly different. Non-parametric statistics shows some advantages to parametric approaches, such as reduction or even avoidance of bias caused by outliers. Besides, no assumptions are needed about the distribution of the analyzed values and homogeneity of variances as well as additivity of effects [21]. The disadvantage of non-parametric statistics is its complexity, power analysis, and time consuming. In contrast, parametric statistics are simple and easy to compute but rely upon the assumption of a ‘‘Gaussian’’ distribution. Parametric statistics are known to be generally robust even when the assumption of ‘‘Gaussian’’ distribution is violated [29]. The objectives of this work were to evaluate the impacts of different insecticides and fungicides on the PCR-DGGE profiles of culturable soil bacterial communities (CSBC) as compared to forest area, and to determine the potential use of the analysis of variance and mean tests on the interpretation of PCR-DGGE data. 2. Material and methods

Soil samples were collected from an Integrated System for Agroecological Production (ISAP) located in the county of Serope´dica, Rio de Janeiro, Brazil. The ISAP is being subjected to agroecological management since 1993 and plant diversity is being stimulated by intercropping and the use of green manure. Soil samples were also collected from a Conventional Potato Production Area (CPPA), and in a neighboring Secondary Forest Area (SFA), both located in Brası´lia, DF, Brazil. Soil from SFA was used for potato cultivation because despite differences in the structure and function of heterotrophic microbial communities in forest soils have been linked primarily to the quantities and qualities of soil organic materials [40] these soils are frequently used as a reference of non disturbed environment. In the CPPA, potato has been intensively cultivated for several years, and the use of chemical fertilizers and pesticides is the strategy to obtain high yields. The SFA is a typical area of the ‘‘Cerrado’’ ecosystem common in the Brazilian central plateau. In each area, soil sampling was performed at 0–20 cm depth. Soil analyses were performed according to EMBRAPA [15] and results are shown on Table 1. The experiment was carried out in a greenhouse at Embrapa Agrobiologia, Serope´dica, Rio de Janeiro, Brazil, from July to September 2003. This period comprises the winter season with climatic conditions relatively constant over the entire experimental period (Table 2). The experimental design was a completely randomized block with 3 replicates, in a factorial arrangement. The factorial arrangement was composed of: 3 soils (ISAP, CPPA and SFA); 5 pesticides (3 insecticides: deltamethrin, aldicarb and chlorpyrifos; and 2 fungicides: tebuconazole and metalaxyl þ mancozeb); and 4 harvest periods (32, 47, 62 and 77 days after sowing – DAS). For each soil type a control treatment without insecticide or fungicide Table 1 Fertility analysis of the soils used to evaluate the effects of insecticides and fungicides on the culturable soil bacterial communities. Al

Ca þ Mg

application was assigned. This factorial arrangement resulted in a total of 216 pots, each one representing an independent experimental unit. Potato cv. Achat was cultivated in 1.5 kg pots containing the same soil amount for all treatments. Pesticides were applied according to Table 3. Two spray applications of pesticides were performed at 30 and 45 DAS, to simulate field conditions, except for aldicarb, incorporated during pot preparation. 2.2. Preparation of medium for soil microorganism cultivation A culture medium (Meio Solo – MS) was used where the nutrient source for microbial growth was soil (Zilli, 2004) [44], collected on each site described on Section 2.1. Initially, soil samples were sieved  through a 2 mm wire mesh, dried at 65 C and ground in a rolling mill [41]. Afterwards, different amounts of ground soil (10, 20, 40 and 80 g) and agar (1.5, 2 and 4 g) on a final volume of 100 mL, were tested to reach the best condition for gelling of the culturable medium. The MS medium was sterilized and poured into Petri dishes to solidify. 2.3. Soil microbial community cultivation

2.1. Soil sampling and experimental design

pH

467

Ca

Mg

Cmolc dm3

P

K

C

mg dm3

%

OM

N

AF

6.7

0

4–0.3

3.0

1.3

117.3

142.1

1.7

2.9

0.19

CPPA

5.4

0.3

2.7

2.1

0.6

17.7

150.3

0.4

0.6

0.13

SFA

5.5

1.5

1.2

0.9

0.3

0.7

62.1

1.5

2.6

0.15

Al, aluminum; Ca, Calcium; Mg, magnesium; P, phosphorus; K, potassium; C, carbon; OM, organic matter and N, nitrogen.

After removing the plants from each pot, including the root system, the soil was homogenized and a 10 g sample of bulk soil was taken and placed in Erlenmeyer flasks (250 mL) containing sterilized water (90 mL) to compose the soil suspensions used for the CSBC cultivation. The flasks were placed on a shaker at 200 rpm for 30 min. A total of 100 mL of the soil suspension from each treatment was sampled and inoculated in Petri dishes containing the corresponding MS medium. Plates were incubated until abun dant microbial growth could be observed (28 C; 5 d). After this period, sterilized water (2 mL) was added to the surface of the culturable medium, cells were mixed with a platinum loop and the suspension was collected with an automatic pipette. Approximately 1 mL of this material was transferred to a microtube (1.5 mL) and centrifuged (9300  g; 15 min). The supernatant was discarded and  the pellet stored at 20 C overnight. 2.4. DNA extraction DNA extraction was performed following the protocol described by Schwieger and Tebbe [37] and modified by Xavier et al. [43]. The stored pellets were suspended in 0.6 mL of TES buffer (0.05 M NaCl; 0.01 M EDTA; 0.05 M Tris HCl pH 8.0; 1% SDS) and vortexed. Samples were subjected to five freeze and thaw cycles, consisting of freezing  in liquid nitrogen (5 min) and heating under agitation (65 C; 180 RPM; 5 min). After each freeze and thaw cycle, samples were mixed by vortexing, 0.168 mg of proteinase K were added to each  sample, followed by incubation under agitation (65 C; 180 RPM; 1 h). At the end of the incubation period, 0.6 mL of phenol:chloroform:isoamyl alcohol (25:24:1) was added to each sample and centrifuged (7500  g; 6 min). The supernatant was transferred to a 1.5 mL microtube and 0.6 mL of chloroform:isoamyl alcohol (24:1) was added followed by centrifugation (7500  g; 6 min). A 0.5 mL of the supernatant was transferred to a 1.5 mL microtube and 0.5 mL of  cold isopropanol was added. Samples were incubated at 20 C for 60 min and centrifuged (16,100  g; 20 min). The supernatant was removed and the pellet was dried in a cold trap centrifuge and raised in 50 mL of TE buffer (10 mM Tris; 1 mM Na-EDTA; pH 8.0). 2.5. PCR-DGGE conditions PCR was performed with three different dilutions of the DNA samples: 1:20, 1:40 and 1:80 in ultrapure water. After amplification each replicate was mixed together in a single microtube. For each

468

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

Table 2 Values of some climatic components during entire experimental period. 

Temperature ( C)

July August September

Mean of the maximum

Mean of the minimum

27.7 25.7 26.5

16.3 15.7 18.2

Humidity (%)

Precipitation (mm)

Evaporation (mm

Solar Irradiation (h)

63.7 66 69

10.2 81.1 58.4

130.5 92.6 68.2

221.4 170.7 143.1

Source: Meteorological farming station/PESAGRO-RIO, km 47, County of Serope´dica, Rio de Janeiro, Brazil.

DNA dilution, PCR amplification was carried out on a final volume of 35 mL, containing 1 mL of DNA dilution, PCR buffer (10 mM), MgCl2 (3.5 mM), dNTP (0.2 mM of each), Bovine Serum Albumin (0.17 mg mL1), Taq DNA polymerase (Invitrogen) (0.7 U) and the bacterial primers 1401r and 968f (0.2 mM of each) described by Heuer and Smalla [20], with a GC-clamp attached on the 968f primer [32], spanning the region roughly between nucleotides 968 and 1401 of the 16S subunit ribosomal DNA (16S rDNA), which includes the variable regions V6–V8 [9], resulting in amplicons of about 500 bp. To determine the best denaturing conditions for the samples, an assay was performed with a 6% polyacrylamide gel containing from zero to 100% 7 M Urea and 40% formamide. The best denaturing conditions occurred on a gradient varying from 50 to 65% of the denaturing agents. For the assays, the polyacrylamide (6%) gels with gradients of 50–65% of denaturants (urea/formamide) were prepared by mixing two solutions: one with 45% denaturant contained 3.5 M (21% w/v) urea plus 20% (v/v) formamide; and the other with 65% denaturant 4.55 M (27.3% w/v) urea plus 26% (v/v) formamide. Depending on the efficiency of PCR amplification, 12– 20 mL of the mixed amplified products were loaded in a denaturing polyacrylamide gel 6% (N-acrylamide, N0 -methylbisacrylamide, 37:1) in 0.5 TAE buffer (Tris-base, 20 mM pH 7.8; sodium acetate 10 mM and Na-EDTA, 0.5 mM). Electrophoresis was carried out in a DcodeTM Detection Mutation System (Bio-Rad) under constant  voltage (120 V; 60 C; 16 h). The gels were stained using SYBR Gold (Molecular Probe) (20) and visualized under UV light in an IMAGO (B&L) photo-documentation system. 2.6. Statistical analysis For each DGGE gel the central sample was loaded at each side of the gel in order to reduce distortion effects during the electrophoresis. The presence or absence of bands in the PCR-DGGE profiles were computed as a binary matrix according to Kozdro´j and Elsas [26]. Each block consisted of one replicate of the complete set of treatments with either insecticide or fungicide plus control treatment and it was subjected to electrophoresis in a single gel. The band profile data obtained for each gel were used for constructing a dendrogram, using Unweighted Pair Group Method with Arithmetic mean (UPGMA) and the Jaccard similarity index. Dendrograms were performed by NTSYSpc version 2.10t [35].

The distance similarity between each treatment and the control was determined using the dendrogram. Initially, the distance residue data were tested to verify the premises of parametric analysis. Cochran’s [5,11] and Lilliefors’ [4,27] tests were used to determine the residue variance homogeneity and normality, respectively. Accepting the assumptions for these tests, data were submitted to the analysis of variance, according to the experimental design described in Section 2.1. Data were submitted to the analysis of variance and mean comparisons were performed at the 5% level of probability by the Scott-Knott test [39]. This test was chosen because, compared to Tukey, Duncan, and Student ‘‘t’’ tests, ScottKnott avoids ambiguous results, in which two statistically different treatments are equal to another treatment [8]. 3. Results 3.1. Cultivation of soil microorganisms on MS For the twelve possible combinations of ground soil and agar described on Section 2.2, the best result for medium gelling for the 3 types of soils studied was achieved using 40 g of soil and 2 g of agar on a final volume of 100 mL, resulting in a gelling condition resembling the conventional culturable medium prepared with 1.5 g of agar on a final volume of 100 mL. The soil/agar proportion above was used to prepare the MS used for CSBC cultivation of all studied soils. Each aliquot of 100 mL of the soil suspension used resulted in a cultured bacterial community from 10 mg of soil. The cultivation on MS resulted in an abundant microbial growth after  incubation (28 C; 5 d), indicating the efficiency of the MS in promoting microbial population growth. 3.2. Culturable soil bacterial community structure The PCR-DGGE fingerprinting showed clear and distinguishable profiles, representing the CSBC under different treatments for ISAP soil (Fig. 1). Some bands were preferentially associated with a specific pesticide treatment when compared to the control, as indicated by the dashed arrows, or they may have been characteristic of a harvest period, as indicated by solid arrows (Fig. 1A and B). In contrast, some bands were common to all treatments, independent of either the applied pesticide or the harvest, as shown into the

Table 3 Chemical group, dosage, method of application and application schedule of five pesticides applied to potato grown in pot culture. Active ingredient

Chemical group

Dosage of active ingredient (kg ha1)

Application method

Application schedule

Insecticides Aldicarb Chlorpyrifos Deltamethrin

Carbamate Organophosphate Pyrethroid

1.95 0.72 0.008

Soil mixing Spray Spray

Sowing 30 and 45 DAS 30 and 45 DAS

Fungicides Metalaxyl þ mancozeb

Phenylamide þ Ethylenebis(dithiocarbamate)

Spray

30 and 45 DAS

Tebuconazole

Triazole

0.2 1.6 0.2

Spray

30 and 45 DAS

DAS, Days After Sowing.

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

469

At the first harvest, the CSBC from CPPA distinguished itself from the others through the lowest similarity values observed when compared to the control. The general shift of the CSBC from CPPA calculated as the mean value for all pesticide treatments was 16.2%, while calculated mean values were as great as 50.8% and 42.2% for ISAP and SFA soils, respectively. This observation suggests that a lower resilience is a characteristic of the CSBC from CPPA when compared to the others, perhaps as a consequence of a low biodiversity, which is generally found under conventional agricultural management [22], resulting in a strong impact caused by pesticide application. On the other hand, a significant difference was found among pesticides on the CSBC of SFA at 32 DAS, in which the lowest effect was for metalaxyl þ mancozeb application (Table 4). However, considering just insecticides, significant differences were only observed at 77 DAS when marked effects occurred for aldicarb, chlorpyrifos and tebuconazole applications (Table 4). This data suggest a substantial shift on the CSBC due to these insecticides but it does not allow any assumption on the magnitude and kind of impact caused by each compound on the CSBC. The main effect seems to have been influenced by the specific CSBC that colonized each soil used for potato cultivation. 3.4. Interpretation of the similarity index as revealed by statistical approach

Fig. 1. PCR-DGGE fingerprinting of culturable soil bacterial communities of ISAP soil cultivated with potato plants and sprayed with insecticide (A) and fungicide (B).

rectangle (Fig. 1B), suggesting that these bands may represent a well established group or groups of bacteria and, therefore, not disturbed by the pesticide treatment. The same tendency was observed for the CSBC from CPPA and SFA soil origins (DGGE gels not shown). 3.3. Similarity dendrogram analysis of CSBC Fig. 2 shows the similarity dendrograms of CSBC of different soils after insecticide spraying on potato plants, constructed using Jaccard similarity index and UPGMA grouping. The dendrograms did not show a clear effect of the insecticides sprayed on the CSBC, but indicated a general tendency of some insecticides to be more effective than others when compared to the control (Fig. 2). Similar results were observed for fungicides spraying (data not shown). Dendrograms similarities values were determined between each treatment and the control for the four harvests and the three replicates, all from one DGGE running for each set of insecticides or fungicides plus control treatment. The data obtained were submitted to analysis of variance and compared at the 5% level of probability by the Scott-Knott test. The results from the PCR-DGGE analysis showed that regardless the type of pesticide applied, a disturbance on the CSBC was observed during the entire period of the study, in comparison to the control (Table 4). Significant differences were observed among soils within harvests (Table 4). The pesticides applied showed significant differences for the similarity of CSBC among treated and control plants as far as the three different soils and the three first harvests (32, 47 and 62 DAS) were considered. In general, the lowest similarity coefficients were observed at the first harvest (32 DAS), where aldicarb, chlorpyrifos and tebuconazole had a higher impact on the CSBC than deltamethrin and metalaxyl þ mancozeb when compared to the control community (Table 4).

The similarity values between treatments and controls were used to generate bar graphs for each pesticide and each soil sample (Fig. 3). The similarity values between insecticide-treated and control culturable soil bacterial profiles showed a distinctive behavior for each soil regardless the type of insecticide applied. The CSBC present in soil samples collected at ISAP showed similarity values varying between 50% and 60%, without major changes during the test period. On the other hand, CSBC from soil samples collected at CPPA were strongly disturbed at the beginning of the experiment as indicated by the similarity values between treated and control communities which were reduced to values of 10–15%. However, these communities showed a tendency to recover as indicated by a gradual increase in the similarity values between treated and control samples. CSBC present in soil samples collected at the SFA were not initially as heavily influenced by insecticide application as the communities from CPPA, but a marked tendency of recovering to a community composition closer to the control was observed from the second and third harvests, which nevertheless was not maintained at the last harvest (Fig. 3). The effect of metalaxyl þ mancozeb and tebuconazole (Fig. 3D and E), on the CSBC had a tendency to show an erratic behavior when compared to the insecticide effects (Fig. 3A–C). The results shown on Table 4, corroborate that the main effects observed in this experiment were due to the soil sample used for growing potato, but within each soil, there were distinct behaviors according to the pesticides applied (Fig. 3). 4. Discussion In this work, the cultivation of CSBC on MS medium was the strategy chosen to obtain representative 16S rDNA profiles of live and active bacterial communities from different soils cultivated with potato and treated with pesticides. The MS medium was used as an alternative to conventional culture medium, which has serious limitations to microbial diversity studies, since they cannot supply specific nutrients for the growth of different microbial species [34]. MS medium is not suitable to isolate soil microorganisms neither to replace kits used for direct DNA soil extraction, but it is suitable to promote the growth of bacteria previously described as unculturable on the conventional culture medium, as reported by

470

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

Fig. 2. UPGMA dendrograms of culturable soil bacterial communities of the soil cultivated with potato and sprayed with insecticides, constructed using Jaccard similarity index. Harvests performed at 32, 47, 62 e 77 days after sowing (DAS). ISAP ¼ Integrated System for Agroecological Production; CPPA ¼ Conventional Potato Production Area; SFA ¼ Secondary Forest Area. Aldic ¼ Aldicarb; Chlorp ¼ Chlorpyrifos; Deltam ¼ Deltamethrin e Contr ¼ Control.

Zilli (2004 – Doctorate thesis). Although these bacterial types have not yet been isolated, the sequence recoveries are evidence of their capability to survive and to compete for nutrients when inoculated on the MS medium, suggesting that its use allows the assessment of the live and active fraction of the soil microbial community. The use of statistical analysis requires some assumptions such as normality and homogeneity of variance of the residue, which were accepted in this study, considering 3 replicates and 45 plots. On the other hand, some mean tests, such as Tukey and SNK tests show ambiguities among treatments, in which two treatments are statistically different but at the same time are equal to another treatment [8]. These authors evaluating the power and type I error rates for Scott-Knott, Tukey and SNK tests under extensive experimental situations and conditions of normality and non-normality of the residue recommend the use the Scott-Knott test because it shows high discriminative power, low rates of type I error for all distributions considered and to be robust when violations of normality are considered. The insecticides used in this work inhibit the acetyl-cholinesterase enzyme (aldicarb and chlorpyrifos) and the potassium channels (deltamethrin). Considering this type of mechanism, there is not a predictable direct effect of these compounds on the soil microbial cells. However, some works reported an increase in soil microbial carbon and fungal population [18,33] and decrease of the bacterial

population [33] as a result of insecticide application. Therefore, the observed effects of insecticides on the CSBC could result from nutrient release due to the death of soil arthropod populations, which might have favored the development of saprophytic microbial groups, among others. Regarding fungicide treatments, their mechanisms of action also suggest that the effect on the heterogeneity of PCR-DGGE profiles of CSBC should also be indirect. However in this case, the observed results might be based on the direct effects of these compounds on the fungal population, which could stimulate the CSBC by the release of nutrients and by a decrease in the competition for nutrients. When the insecticides were compared with metalaxyl þ mancozeb and tebuconazole it was observed distinctive pattern of behavior (Fig. 3). Soil management adopted on ISAP and soil characteristics of SFA favored an increase in microbial heterogeneity, which might had been associated with an increase in the redundant functional groups as well as in the resilience of these soils to the impact of pesticide applications. Sanginga et al. [36] suggested that special conditions that lead to the stimulation of biological activity might exist in soils under forests and agroecological systems that are not present in soils under conventional production systems. Therefore, it is possible that the established management conditions at the ISAP had stimulated the development of soil bacterial populations and the growth of redundant functional

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

471

Table 4 Percentage of similarity of culturable soil bacterial communities between pesticide and the control treatments at four harvests of potato (32, 47, 62 e 77 DAS). Averages of three replicates. Pesticides

ISAP

CPPA

SFA

Deltamethrin Aldicarb Chlorpyrifos Metalaxyl þ Mancozeb Tebuconazole Mean of areas

32 DAS 55.0 Aa 52.3 Aa 55.0 Aa 49.0 Aa 42.7 Aa 50.8 a

Mean of pesticides

17.2 Ab 5.3 Ab 14.0 Ab 27.0 Ab 17.0 Ab 16.1 c

35.8 37.3 37.3 62.3 37.3 42.0

Bb Ba Ba Aa Ba b

36.0 31.7 35.4 46.1 32.3

A A A A A

Deltamethrin Aldicarb Chlorpyrifos Metalaxyl þ Mancozeb Tebuconazole Mean of areas

47 DAS 52.0 Ab 47.7 Aa 47.7 Ab 39.7 Ab 44.0 Ab 46.2 b

34.7 26.7 32.2 28.7 26.7 29.8

Ab Ab Ab Ab Ab c

91.0 67.3 87.3 77.7 77.3 80.1

Aa Aa Aa Aa Aa a

59.2 47.2 55.7 48.7 49.3

A A A A A

Deltamethrin Aldicarb Chlorpyrifos Metalaxyl þ Mancozeb Tebuconazole Mean of areas

62 DAS 53.5 Aa 53.5 Aa 46.0 Aa 52.7 Aa 49.3 Aa 51.0 a

51.0 34.0 43.0 45.3 42.7 43.2

Aa Ab Aa Aa Aa b

58.7 61.8 57.2 46.0 52.0 55.1

Aa Aa Aa Aa Aa a

54.4 49.8 48.7 48.0 48.0

A A A A A

Deltamethrin Aldicarb Chlorpyrifos Metalaxyl þ Mancozeb Tebuconazole Mean of areas CV Plot (%)

77 DAS 53.8 Aa 46.3 Aa 43.7 Aa 51.7 Aa 31.3 Aa 45.4 a 26.48

50.0 41.7 27.3 44.7 40.0 40.7

Aa Aa Bb Aa Aa a

40.0 37.7 40.0 56.3 44.3 43.7

Aa Aa Aa Aa Aa a

47.9 41.9 37.0 50.9 38.6

A B B A B

DAS, Days After Sowing; ISAP, Integrated System for Agroecological Production; CPPA, Conventional Potato Production area; SFA, Secondary Forest Area; CV, Coefficient of variation; Similarity values in the same column within harvest followed by the same upper case letter are not different by the Scott-Knott test (p > 0.05); Similarity values in the same line followed by the same lower case letter are not different by the Scott-Knott test (p > 0.05).

groups. Even if part of these redundant functional groups were affected by pesticide application, the remaining groups with some tolerance to the active ingredients would continue to function normally, minimizing the treatment effect and contributing to the resilience of the bacterial community in this soil. On the CPPA soil, the strong impact observed initially, represented by the low similarity index in relation to the control (Fig. 3) could have resulted from the low microbial diversity commonly associated with this kind of management, with the predominance of a few groups or species due to an increase in soil pH, and a decrease in soil organic matter, microbial biomass and soil microbial activity [13]. As suggested by Lupwayi et al. (1998) [28], under these conditions a decrease in the diversity of bacteria might be a consequence of the reduction of both substrate richness and evenness. The initial impact in the SFA soil was weaker than the effect observed in the CPPA soil. Forest soils normally present greater contents of organic matter and microbial diversity than conventionally cultivated soils (Haynes et al., 2003) [17], which may have contributed to the reduction of the initial impact due to pesticide application. However, the recovery observed in this case might be a consequence of opportunist bacterial groups, which after depletion of the nutrient concentration have rapidly reduced their activities. Although some studies have shown the effect of pesticides on specific soil microbial groups [1,25], studies focusing on the effect of these compounds on the dynamics of microbial communities are still relatively scarce. A better understanding of the behavior of microbial communities aiming at the characterization of ecological

Fig. 3. Similarity among culturable soil bacterial communities under potato cultivation after pesticides spraying as compared to the control on the different harvests (32, 47, 62 and, 77 DAS – Days after sowing). ISAP ¼ Integrated System for Agroecological Production; CPPA ¼ Conventional Potato Production Area; SFA ¼ Secondary Forest Area. A – Deltamethrin, B – Aldicarb, C – Chlorpyrifos, D – Metalaxyl þ mancozeb, E – Tebuconazole.

472

E.P.deB. Ferreira et al. / European Journal of Soil Biology 45 (2009) 466–472

parameters may indicate the effect of the continued use of pesticides on ecosystem health [12]. 5. Conclusions Results obtained in this study indicate significant effects of pesticide application on the PCR-DGGE profiles of CSBC. Although the impacts had not been clearly observed on both DGGE gels and dendrograms, the use of the analysis of variance, followed by mean comparison tests allowed to observe shifts on the similarity of the CSBC, in which significant differences among treatments and control were shown on the different soils within harvesting times. These differences could be best observed at the first and second harvests, when the main effects were linked to the different soils. These results also indicate the potential of the analysis of variance as an useful tool for the interpretation of PCR-DGGE data that could be widely used in soil microbial ecology and diversity studies. Acknowledgements The authors would like to thank Embrapa Agrobiologia, Embrapa Hortaliças, CAPES and CNPq for their financial and technical support, Dr. Phillip Chalk for his relevant suggestions, and Dra. Aline de Holanda Nunes Maia for the useful considerations on the statistics analysis. References [1] M.H. Abd-Alla, S.A. Omar, S. Karanxha, The impact of pesticides on arbuscular and nitrogen-fixing symbioses in legumes, Appl. Soil Ecol. 14 (2000) 191–200. [2] Agrofit, Sistema de Agroto´xicos Fitossanita´rios (2005).www.agricultura.gov.br. [3] P. Aneja, M. Daı´, D.A. Lacorre, B. Pillon, T.C. Charles, Heterologous complementation of the exopolysaccharide synthesis and carbon utilization phenotypes of Sinorhizobium meliloti Rm1021 polyhydroxyalkanoate synthesis mutants, FEMS Microbiol Lett. 239 (2004) 277–283. [4] M.A. Arcones, Y. Wang, Some new tests for normality based on U-processes, Stat. Probab. Lett. 76 (2006) 69–82. [5] D. Argaç, Testing for homogeneity in a general one-way classification with fixed effects: power simulations and comparative study, Comput. Stat. Data Anal. 44 (2004) 603–612. [6] A.C. A´vila, C.A. Lopes, F.H. França, F.J.B. Reifschneider, G.P. Henz, J.A. Buso, O. Furumoto, A cultura da batata, Embrapa Commun. Transf. Tecnol. (1999) 184. [7] P.J. Bohlen, C.A. Edwards, Q. Zhang, R.W. Parmelee, M. Allen, Indirect effects of earthworms on microbial assimilation of labile carbon, Appl. Soil Ecol. 20 (2002) 255–261. [8] L.C. Borges, D.F. Ferreira, Poder e taxas de erro tipo I dos testes Scott-Knott, Tukey e Student-Newman-Keuls sob distribuiço˜es normal e na˜o normais dos resı´duos, Rev. Mat. Estat. 21 (2003) 67–83. [9] J. Brosius, M.L. Palmer, P.L. Kennedy, H.H. Noller, Complete nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli, Proc. Natl. Acad. Sci. 75 (1978) 4801–4805. [10] E.D. Caldas, L.C.K.R. Souza, Chronic dietary risk assessment for pesticides residues in Brazilian food, J. Public Health 34 (2000) 529–537. [11] W.G. Cochran, Problems arising in the analysis of a series of similar experiments, J. R. Stat. Soc. 4 (1937) 102–118. [12] H.A.M. De Kruijf, D.P. van Vuuren, Following sustainable development in relation to the north–south dialogue: ecosystem health and sustainability indicators, Ecotoxicol. Environ. Saf. 40 (1998) 4–14. ˜ a, J. Bueno, S.J. Gonza´lez-Prieto, T. Carballas, Cultivation effects on [13] M. Dı´az-Ravin biochemical properties, C storage and 15N natural abundance in the 0–5 cm layer of an acidic soil from temperate humid zone, Soil Tillage Res. 84 (2005) 216–221. [14] D.J. Ecobichon, Toxic effects of pesticides, in: M.O. Amdur, J. Doull, C.D. Klaassen (Eds.), Asarett and Doll’s Toxicology: the Basic Science of Poisons, fourth ed. Mc Graw Hill, New York, 1993, pp. 565–622. [15] EMBRAPA (Empresa Brasileira de Pesquisa Agropecua´ria), Manual for Methods of Soil Analysis, second ed. National Soil Research Center, 1997, 212 pp. [16] R.B. Franklin, D.R. Taylor, A.L. Mills, Characterization of microbial communities using randomly amplified polymorphism (RAPD), J. Microbiol. Methods 35 (1999) 225–235. [17] R.J. Haynes, C.S. Dominy, M.H. Graham, Effect of agricultural land use on soil organic matter status and the composition of earthworm communities in KwaZulu-Natal, S. Afr. Agric. Ecosyst. Environ. 95 (2003) 453–464.

[18] M.R. Hart, P.C. Brookes, Soil microbial biomass and mineralisation of soil organic matter after 19 years of cumulative field applications of pesticides, Soil Biol. Biochem. 28 (1996) 1641–1649. [19] M. Hartmann, B. Frey, R. Ko¨lliker, F. Widmer, Semi-automated analyses of soil microbial communities: comparison of T-RFLP and RISA based on descriptive and discriminative statistical approaches, J. Microbiol. Methods 61 (2005) 349–360. [20] H. Heuer, K. Smalla, Application of denaturing gradient gel electrophoresis and temperature gradient gel electrophoresis for studying soil microbial communities, in: J.D. van Elsas, J.T. Trevors, E.M.H. Wellington (Eds.), Modern Soil Microbiology, Marcel Dekker, New York, 1997, pp. 353–373. [21] M. Huehn, Nonparametric measures of phenotypic stability. Part. 1: theory, Euphytica 47 (1990) 189–194. [22] K. Jangid, M.A. Williams, A.J. Franzluebbers, J.S. Sanderlin, J.H. Reeves, M.B. Jenkins, D.M. Endale, D.C. Coleman, W.B. Whitman, Relative impacts of land-use, management intensity and fertilization upon soil microbial community structure in agricultural systems, Soil Biol. Biochem. 40 (2008) 2843–2853. [23] A.C. Kennedy, Bacterial diversity in agroecosystems, Agric. Ecosyst. Environ. 74 (1999) 65–76. [24] A.D. Kent, E.W. Triplett, Microbial communities and their interactions in soil and rhizosphere ecosystems, Annu. Rev. Microbiol. 56 (2002) 211–236. [25] B.D. Kishinevsky, R. Lobel, D. Gurfel, C. Nemas, Soil fumigation with methyl bromide as a means of increasing the occurrence of the inoculum strain in peanut nodules, Soil Biol. Biochem. 24 (1992) 845–848. [26] J. Kozdro´j, J.D. van Elsas, Structural diversity of microbial communities in arable soils of a heavily industrialised area determined by PCR-DGGE. Fingerprinting and FAME profiling, Appl. Soil Ecol. 17 (2001) 31–42. [27] H. Lilliefors, On the Kolmogorov–Smirnov test for normality with mean and variance unknown, J. Am. Stat. Assoc. 62 (1967) 399–402. [28] N.Z. Lupwayi, W.A. Rice, G.W. Clayton, Soil microbial diversity and community structure under wheat as influenced by tillage and crop rotation, Soil Biol. Biochem. 30 (1998) 1733–1741. [29] G. Marrelec, H. Benali, P. Ciuciu, M. Pelegrini-Issac, J.B. Poline, Robust Bayesian estimation of the hemodynamic response function in eventrelated BOLD fMRI using basic physiological information, Hum. Brain Mapp. 19 (2003) 1–17. [30] M.N. Moore, Biocomplexity: the post-genome challenge in ecotoxicology, Aquat. Toxicol. 59 (2001) 1–15. [31] G. Muyzer, DGGE/TGGE a method for identifying genes from natural ecosystems, Curr. Opin. Microbiol. 2 (1999) 317–322. [32] G. Muyzer, E.C. de Waal, A.G. Uitterlinden, Profiling of complex microbial populations by denaturing gel electrophoresis analysis of polymerase chain reaction amplified genes coding for 16S rRNA, Appl. Environ. Microbiol. 59 (1993) 695–700. [33] S. Pandey, D.K. Singh, Total bacterial and fungal population after chlorpyrifos and quinalphos treatments in groundnut (Arachis hipogea L.) soil, Chemosphere 55 (2004) 197–205. [34] L. Ranjard, F. Poly, S. Nazaret, Monitoring complex bacterial communities using culture-independent molecular techniques: application to soil environment, Res. Microbiol. 151 (2000) 167–177. [35] F.J. Rohlf, NTSYS-pc. Numerical Taxonomy and Multivariate Analysis System, Version 2.10, Exeter Software, New York, 2002. [36] N. Sanginga, K. Mulongoy, M.J. Swift, Contribution of soil organisms to the sustainability and productivity of cropping systems in the tropics, Agric. Ecosyst. Environ. 41 (1992) 135–152. [37] F. Schwieger, C.C. Tebbe, A new approach to utilize PCR-single strand conformation polymorphism for 16S rDNA gene-based microbial community analysis, Appl. Environ. Microbiol. 64 (1998) 4870–4876. [38] A. Schmalenberger, C.C. Tebbe, M.A. Kertesz, H.L. Drake, K. Ku¨sel, Twodimensional single strand conformation polymorphism (SSCP) of 16S rRNA gene fragments reveals highly dissimilar bacterial communities in an acidic fen, Eur. J. Soil Biol. 44 (2008) 495–500. [39] A.J. Scott, M. Knott, A cluster analysis method for grouping means in the analysis of variance, Biometrics 30 (1974) 507–512. [40] P. Seatre, E. Baath, Spatial variation and patterns of soil microbial community structure in a mixed spruce-birch stand, Soil Biol. Biochem. 32 (2000) 907–917. [41] J.L. Smith, H.U. Myung, Rapid procedures for preparing soil and KCl extracts for 15 N analysis, Commun. Soil Sci. Plant Anal. 21 (1990) 2273–2279. [42] C. Viti, L. Giovannetti, Characterization of cultivable heterotrophic bacterial communities in Cr-polluted and unpolluted soils using Biolog and ARDRA approaches, Appl. Soil Ecol. 28 (2005) 101–112. [43] G.R. Xavier, F.V. Silva, J.E. Zilli, N.G. Rumjanek, Adaptaça˜o de me´todo para extraça˜o de DNA microbiano, Embrapa Agrobiologia, Serope´dica, 2004, Documento 171, 24 pp. [44] J.E. Zilli, Avaliaça˜o do impacto dos herbicidas glifosato e imazaquin na comunidade bacteriana do solo e associada a´s raı´zes de plantas de soja. 2004. 138 p. Tese Doutorado- Universidade Federal Rural do Rio de Janeiro, Serope´dica, RJ.