, Volume 27, Issue 7, pp 871–889 | Cite as

Comparative analysis of transcriptomic responses to sub-lethal levels of six environmentally relevant pesticides in Saccharomyces cerevisiae

  • Fátima N. Gil
  • Alina C. Gonçalves
  • Jörg D. Becker
  • Cristina A. ViegasEmail author


Accidental spills and misuse of pesticides may lead to current and/or legacy environmental contamination and may pose concerns regarding possible risks towards non-target microbes and higher eukaryotes in ecosystems. The present study was aimed at comparing transcriptomic responses to effects of sub-lethal levels of six environmentally relevant pesticide active substances in the Saccharomyces cerevisiae eukaryotic model. The insecticide carbofuran, the fungicide pyrimethanil and the herbicides alachlor, S-metolachlor, diuron and methyl(4-chloro-2-methylphenoxy)acetate were studied. Some are currently used agricultural pesticides, while others are under restricted utilization or banned in Europe and/or North America albeit being used in other geographical locations. In the present work transcriptional profiles representing genome-wide responses in a standardized yeast population upon 2 h of exposure to concentrations of each compound exerting equivalent toxic effects, i.e., inhibition of growth by 20% relative to the untreated control cells, were examined. Hierarchical clustering and Venn analyses of the datasets of differentially expressed genes pointed out transcriptional patterns distinguishable between the six active substances. Functional enrichment analyses allowed predicting mechanisms of pesticide toxicity and response to pesticide stress in the yeast model. In general, variations in transcript numbers of selected genes assessed by Real-Time quantitative reverse transcription polymerase chain reaction confirmed microarray data and correlated well with growth inhibitory effects. A possible biological relevance of mechanistic predictions arising from these comparative transcriptomic analyses is discussed in the context of better understanding potential modes of action and adverse side-effects of pesticides.


Pesticide Sub-lethal Toxicogenomics Toxicity mechanisms Molecular response Saccharomyces cerevisiae 


Concerns exist over possible environmental risks of plant protection agrochemicals due to careless storage or disposal in dealership/mix-load sites (Chirnside et al. 2007), accidental spills (Vega et al. 2007) and/or misuse (Abrantes et al. 2010; Chelinho et al. 2012). The potential of many pesticides to undergo mobility via the soil-water pathway (e.g., via leaching and/or surface runoff from rainfalls and flood events) might result in contamination of aquatic environmental compartments which, in some cases, may persist long after their ban (Abrantes et al. 2010; Chelinho et al. 2012; Sousa et al. 2018). Presumably this can be the case for the six model pesticides selected to be used in the present study, namely the insecticide carbofuran (CAB), the fungicide pyrimethanil (PYR) and the herbicides alachlor (ALA), S-metolachlor (S-MET), diuron (DIU) and methyl(4-chloro-2-methylphenoxy)acetate (MCPA methyl ester, MCPA-ME). Main properties and current status of these pesticides are summarized in Supplementary Table S1. Briefly, PYR, S-MET and MCPA related products are plant protection substances currently in use in agriculture, i.e., authorized under the EC Regulation 1107/2009, while DIU and ALA are under restricted utilization and environmental surveillance in the EU and US (EC 2009; Sousa et al. 2018; USEPA 2009). On the other hand, CAB was banned from agricultural usage in the EU and US but it is still in use in several geographical locations like South America, Asia, Africa (Azizullah et al. 2011; Chelinho et al. 2012; Harabawy and Ibrahim 2014; Mansano et al. 2016; Sousa et al. 2018). Detailed information on pesticide environmental fate can be found in the PPDB – Pesticide Properties DataBase (Lewis et al. 2016). Briefly, the pesticides under study show low volatility (ALA, S-MET, MCPA-ME and PYR) or are nonvolatile (DIU, CAB). In general, all six can show some ability to sorb to organic carbon in soil particles and have variable mobility through soil layers and therefore potential to reach natural underground and surface waters (Lewis et al. 2016). For example, DIU is slightly mobile whereas the other five pesticides are moderately mobile through layers of organic agricultural soil, but this tendency can be more relevant in less organic and more mineral soils. In addition, their persistency in soil varies between moderate (DIU, PYR) or low (ALA, S-MET, MCPA-ME, CAB), but most of them (except MCPA products) are relatively resistant to aqueous photolysis and hydrolysis (Lewis et al. 2016).

These six active substances represent five different chemical-families and their primary mode of action (MoA) in the respective target-organisms is generally known (Table S1). As referred above, in worst-case situations, bioavailable levels may reach and threaten non-target microbial and higher organisms that provide important ecosystem services such as primary production, trophic relationships and nutrient recycling (Azizullah et al. 2011; Chelinho et al. 2012; DeLorenzo et al. 2013; Geret et al. 2011; Junghans et al. 2003; Liu et al. 2006; Mansano et al. 2016; Seeland et al. 2012). In particular, both ALA and DIU are priority hazardous substances in the field of water policy in the EU, and similarly ALA and CAB in the US, mainly due to potentially unacceptable ecological risks for aquatic primary producers and higher eukaryotes (EC 2013; Sousa et al. 2018; USEPA 2009). Concretely, the MoA of CAB in target insect pests is inhibition of the enzyme acetylcholinesterase, thus interfering in neurofunction. This specific effect of CAB has been also found in not only invertebrates but also vertebrates like fish, and apparently shows synergy with organophosphorus insecticides (Bocquené et al. 1995; Fulton and Key 2001, Hernández-Moreno et al. 2011). CAB can contaminate aquatic compartments in the vicinity of sources of emission like for instance treated fields or sites where accidental spills may occur (Azizullah et al. 2011; Chelinho et al. 2012). A number of studies have reported potential negative effects of CAB in the health of non-target organisms of diverse taxa and trophic levels from freshwater, marine and terrestrial ecosystems, including primary producers (Azizullah et al. 2011; Megharaj et al. 1993) as well as microbial (Mansano et al. 2016) and higher consumer organisms (Chelinho et al. 2012; Harabawy and Ibrahim 2014; Hernández-Moreno et al. 2011; Saxena et al. 2014). Regarding the MoA of ALA and S-MET on target weeds, it is mainly attributed to inhibition of the endoplasmic reticulum membrane-bound very-long-chain fatty acid elongase system (i.e., extends long-chain fatty acid CoA-esters to form C26-fatty acids), but as far as we are aware the primary molecular target of these herbicides remains not completely clear (Junghans et al. 2003; Schmalfuβ et al. 2000). These two structurally related herbicidal active substances have been used worldwide in several agricultural contexts and differ considerably with respect to their current status in the EU in that S-MET is authorized for use in agriculture while ALA is not (EC 2009; EC 2013; Lewis et al. 2016). Interestingly, they also diverge with respect to cytotoxicity towards rat and human hepatocytes (Kale et al. 2008) and to site-specificity and potency concerning tumor induction in rats (Genter et al. 2009). In what concerns the phenylurea-derived herbicide DIU, it acts by inhibiting photosynthesis in target weeds and photosynthetic microorganisms (Giacomazzi and Cochet 2004). Due to wide-spectrum uses (e.g., weed control in a variety of crops, control of mosses in non-crop areas, algicide in commercial fish production, residential ponds and aquariums, etc.), DIU can be found in different environments such as soil, sediments and water (Lewis et al. 2016; Sousa et al. 2018). It has the potential to exert negative side-effects in non-target phytoplankton, nitrogen-cycling bacteria and protozoa in aquatic ecosystems (DeLorenzo et al. 2013; Mansano et al. 2016; Tadonléké et al. 2009) or in laboratory model mammals (Federico et al. 2011; Domingues et al. 2011; Ihlaseh et al. 2011), but its MoA in non-photosynthetic organisms is not fully understood. Regarding MCPA-ME, it belongs to the chlorophenoxy chemical family of active ingredients with auxin-like phytohormonal MoA (Peixoto et al. 2009). To date, MCPA-based herbicides have been formulated as acid MCPA (4-chloro-2-methylphenoxy acetic acid), aqueous salts (e.g., sodium or dimethylamine salt forms) or esters (e.g., thioethyl-, ethylhexyl-ester forms), being registered for control of annual and perennial broad-leaved weeds in cereals, grasses, orchards and non-crop areas (Lewis et al. 2016; Peixoto et al. 2009). Even though we found no information in the literature whether or not the active substance MCPA-ME is actually used in herbicidal formulations in the EU, it was selected to be used in this study because it is structurally closely related to the authorized MCPA-thioethyl ester (Lewis et al. 2016) and was commercially available in the pure form (contrary to MCPA-thioethyl). The ester forms of MCPA are generally chemically stable and more lipophilic than the MCPA salts (Lewis et al. 2016), but they can undergo hydrolysis to the acid ionizable form MCPA when present in environments at pH 7 or higher (van Ravenzwaay et al. 2004). In soil pore water and natural waters with pH values higher than the pKa value of MCPA (~3.73; Lewis et al. 2016), this molecule can undergo dissociation into the more water-soluble ionic form. These features make MCPA herbicides generally mobile in the environment with ability to reach aquatic ecosystems (Lewis et al. 2016) where they can potentially exert adverse effects in non-target plants (Peixoto et al. 2009), yeast and frogs (Papaefthimiou et al. 2004). Finally, the fungicide PYR is one of the most frequently used fungicides in vineyards, fruits, vegetables and ornamentals mostly against the grey mold Botrytis cinerea and other fungi (Lewis et al. 2016). The MoA of PYR in target phytopathogenic fungi has been mainly attributed to inhibition of methionine biosynthesis and secretion of hydrolases involved in the infection process but the exact primary mechanism is still not completely clear (Fritz et al. 2003; Milling and Richardson 1995). Moderate mobility in soil as well as high potential for surface runoff on the steep slopes of grapevine plantations together with high stability in water make PYR a relevant water and sediment contaminant (Seeland et al. 2012; Verdisson et al. 2001). Concerns have been raised on PYR potential toxicity over non-target organisms, particularly when mixed with other currently used pesticides (Coleman et al. 2012; Seeland et al. 2012; Verdisson et al. 2001).

As far as we are aware, there is still limited knowledge over MoA and (eco)toxicological side-effects of these moderately lipophilic compounds (Table S1) in non-target microbial or higher eukaryotes from ecosystems that may be exposed to high inadvertent pesticide pollution. The present study intends to make a contribution for better understanding these issues by using a transcriptomic approach with the microbial eukaryotic model Saccharomyces cerevisiae. Changes in gene expression reflect rapid and sensitive responses of organisms to environmental insults. Toxicogenomics using DNA microarray platforms for ecologically relevant eukaryotes has been an auspicious tool in predictive (eco)toxicology, disclosing molecular biomarkers of toxicity and xenobiotic MoA in biological systems (reviewed in Steinberg et al. 2008). A number of gene expression profiling studies have used the yeast S. cerevisiae, revealing toxicity mechanisms of bioavailable xenobiotics, drugs and industrial effluents (reviewed in Braconi et al. 2016; Santos and Sá-Correia 2015; Yasokawa and Iwahashi 2010) and disclosing molecular indicators of xenobiotic toxicity to be used in environmental biomonitoring (Gil et al. 2015; Kim et al. 2006). Despite the existence of drawbacks associated with yeast unicellularity and cross-species extrapolation, S. cerevisiae is considered a valuable test organism in environmental toxicology because it is easy to cultivate, has a fully annotated genome and shares a significant degree of metabolic and regulatory conservation with more complex eukaryotes (Braconi et al. 2016; Cascorbi et al. 1993; Santos and Sá-Correia 2015). Linking gene expression profiles elicited by toxicants with toxicity mechanisms in the yeast may help predicting possible MoA associated with toxicological outcomes of exposure and orient more complex and expensive ecotoxicology studies in environmentally relevant eukaryotes (Braconi et al. 2016; Santos and Sá-Correia 2015; Teixeira et al. 2007). As such, this simple model is considered to contribute to reduce the use of animal models in (eco)toxicology (Santos and Sá-Correia 2015, Yasokawa and Iwahashi 2010), which is mandatory at regulatory level (Edwards et al. 2016).

Therefore, the present study aimed at comparing transcriptomic responses to deleterious effects of sub-lethal concentrations of the six pesticide active substances in yeast cells, using Affymetrix Yeast 2.0 Genome arrays. To ensure comparable pesticide-triggered transcriptional responses and correlating with yeast outcomes at higher level of biological organization (growth inhibition), the genome-wide transcriptional profiles were obtained in a standardized yeast population upon 2 h of exposure to concentrations of each pesticide provoking equivalent toxic effects, namely pesticide concentrations inhibiting yeast growth by 20% relative to the growth of the untreated control cells (hereafter designated as the 20%-effective concentration IC20). These short-term exposure conditions associated with an impairment of low magnitude in yeast physiology were chosen in order to unveil potential early-alarm indicators of pesticide aggression before signs of eventual deterioration of cellular function (Steinberg et al. 2008).

Materials and methods

Pesticides, yeast strain and culture conditions

The six pesticide active substances under study (ALA, CAB, DIU, MCPA-ME, PYR and S-MET) were purchased from Sigma-Aldrich (PestanalR product line). The limit of solubility in water for these compounds is as follows (Lewis et al. 2016): 0.89 mM for ALA, 1.46 mM for CAB, 0.15 mM for DIU, 0.61 mM for PYR and 1.69 mM for S-MET, at 20–25 °C. The limit of solubility in water for MCPA-ME was not available in the literature, and therefore its value was assumed to be similar to the value for the MCPA-thioethyl ester (9.3 μM; Lewis et al. 2016). Stock solutions of each pesticide were prepared in the carrier solvent dimethylsulfoxide (DMSO) and added to the aqueous culture and exposure media to obtain the desired concentrations of compound supplementation (Gil et al. 2011, 2014). The final concentration of DMSO in all media including the controls untreated with pesticide (hereafter designated as CT02) was kept equal to 0.2% v/v, a DMSO concentration with no detectable effect on yeast growth or gene expression (Gil et al. 2011).

The S. cerevisiae parental strain BY4741 (Mata, his3∆1, leu2∆0, met15∆0, ura3∆0) (Euroscarf collection) was used. Unless otherwise indicated, yeast cells were batch-cultured with orbital agitation (250 rpm) at 30 °C in liquid MMB basal medium (pH 6.5) that contained, per liter: 1.7 g of YNB without amino acids or NH4+ (Difco), 20 g of glucose, 2.65 g of (NH4)2SO4 (Merck), 20 mg of histidine, 20 mg of methionine, 60 mg of leucine, 20 mg of uracil (all from Sigma) (Gil et al. 2011, 2014). To get preliminary data on the IC20 of each pesticide, examination of pesticide effects in yeast growth was carried out using standardized exponential yeast cells grown in MMB liquid medium (OD640nm = 0.25 ± 0.05) supplemented with adequate volumes of pesticide stock solutions in DMSO (final pesticide concentrations up to 1.1, 1.6, 0.4, 4.7, 0.7 and 1.5 mM of ALA, CAB, DIU, MCPA-ME, PYR and S-MET, respectively, and final DMSO concentration equal to 0.2% v/v). The yeast specific growth rate was estimated from the exponential phase of the growth curves (culture OD640nm versus time) obtained for each combination of pesticide and concentration tested, and the percentage of inhibition of yeast growth was estimated as the ratio between the growth rate in the presence of the pesticide concentration and in its absence (CT02 control); further details are available in Gil et al. (2011).

Pesticide exposure assays and RNA preparation

To obtain comparable transcriptional profiles representing the yeast responses to the six pesticides, exposure assays were performed with a standardized yeast cell population incubated in the presence of the toxic equivalent IC20 of each pesticide. Therefore, to obtain samples for transcriptomic analysis, exponential yeast cells in MMB liquid medium (OD640nm = 0.25 ± 0.05) were supplemented with 0.10, 0.34, 0.54, 0.72, 0.74 or 0.78 mM of PYR, DIU, MCPA-ME, CAB, ALA or S-MET, respectively or without any pesticide (CT02 cells). On the other hand, to obtain samples for Real-Time quantitative reverse transcription real-time polymerase chain reaction (qRT-PCR) analysis, the standardized cells in MMB medium were supplemented with two (CAB and DIU) or three (the other four pesticides) pesticide concentrations exerting increasing percentages of growth-inhibition, namely ~10%, 20% (IC20) or 50%, and without any pesticide (CT02 cells) (Gil et al. 2011, 2014). Those inhibitory concentrations were as follows (respectively): 0.53, 0.78 and 1.06 mM for S-MET; 0.41, 0.74 and 0.93 mM for ALA (Gil et al. 2011); 0.19, 0.54 and 2.4 mM for MCPA-ME; 0.05, 0.10 and 0.55 mM for PYR (Gil et al. 2014). For DIU and CAB, solely ~10% and 20%-inhibitory concentrations were used (0.034 and 0.34 mM for DIU; 0.36 and 0.72 mM for CAB, respectively) because, in both cases, concentrations high enough to cause 50% growth-inhibition could not be achieved in the aqueous growth media (not shown). In both the above described microarray and Real-Time qRT-PCR exposure assays, independent biological triplicates for each combination of pesticide and exposure condition and for the control CT02 cells were prepared. The cell suspensions supplemented with the pesticides were incubated at 30 °C and 250 rpm (orbital agitation) during a 2-h exposure period. The cells were then harvested by centrifugation (12,857g, 5 min, 4 °C), immediately frozen in liquid nitrogen, and stored at −80 °C. Total RNA isolation was performed using the hot-phenol method (Kohrer and Domdey 1991). Determination of RNA concentration and purity by spectrophotometry and confirmation of RNA integrity in a Bioanalyzer were performed as described elsewhere (Gil et al. 2011, 2014).

DNA microarray processing and data analysis

Microarray processing was performed at the Gene Expression Unit of Instituto Gulbenkian de Ciência (Oeiras, Portugal). For each biological triplicate at each exposure condition, 100 ng of total RNA was processed for use on Affymetrix (Santa Clara, CA, USA) GeneChip Yeast Genome 2.0 Arrays, according to the manufacturer’s GeneChip3’ IVT Express kit user manual. RNA processing, array hybridization, washes and double-stain protocols (FS450_0003; GeneChip HWS kit, Affymetrix) performed on an Affymetrix GeneChip Fluidics Station 450 are described elsewhere (Gil et al. 2011, 2014). Arrays were scanned on a GeneChip scanner (3000 7 G, Affymetrix) and the 36 scanned arrays (18 from the six pesticide exposure samples and 18 from the CT02 samples) were analyzed for quality control, Absent/Present calls, normalization and gene expression calculations (DNA-Chip Analyzer dChip 2010, Wong Lab, Harvard). Genes were considered as differentially expressed if the 90% lower confidence bound of the fold change (lbFC) between the pesticide-exposure experiments and the baseline (CT02 samples) was above 1.2, with a median False Discovery Rate (FDR) < 5%. An exception was the herbicide MCPA-ME, for which the genes were considered differentially expressed if the lbFC between the pesticide-exposure experiments and the baseline was above 1.3, to obtain an FDR < 10%. Microarray data are deposited in the Gene Expression Omnibus (GEO) repository under accession number GSE53125.

Annotations for the 5716 S. cerevisiae transcripts on the arrays were obtained from Affymetrix database (; November 2009). Gene functional description is also available in the SGD - Saccharomyces Genome Database (Cherry et al. 2012) and the MIPS Functional Catalogue Database (FunCatDB) (Ruepp et al. 2004). To visualize the gene expression profiles for each pesticide in a dendrogram where they are organized in clusters according to similarities, hierarchical clustering analysis was performed based on Partek® Genomics Suite software version 6.3 (Partek Inc. St. Louis, MO, USA, 2008) using as input the datasets displaying gene expression values for the significantly differentially expressed genes in response to each pesticide. A comparative spread-sheet analysis of the six datasets of differentially-expressed genes relative to the six different pesticide exposure contexts was also carried out using the software VENNTURE (version; available for download online at according to author instructions (Martin et al. 2012). VENNTURE offers an interface for Venn diagram generation that enables the analysis of up to six datasets of any length allowing the identification of genes in each intersection (representing pesticide-shared responsive genes) or in disjoint groups (representing pesticide-unique responsive genes) (Martin et al. 2012). Gene enrichment within functional clusters of the datasets of up- and downregulated genes in response to each pesticide was performed using the hierarchically structured Functional Distribution of Gene Lists tool available online at the FunCatDB resource (; Ruepp et al. 2004) and the hypergeometric-distribution-based FunSpec bioinformatics tool with a Bonferroni correction for multiple testing (available online at; Robinson et al. 2002), according to authors’ instructions. The functional categories found to be statistically overrepresented (p value < 0.01) in both analyses were considered in the present study, unless otherwise indicated.

Real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR)

To confirm modifications in transcript levels in RNA samples upon yeast cells exposure to the IC20 and the ~10 and 50% growth-inhibitory concentrations of each pesticide (indicated above), ten genes were chosen (Table S2) from functional categories statistically overrepresented within the pesticide datasets with a focus on transcripts that could reflect relevant mechanistic information over yeast response to pesticide toxic effects. Processing of total RNA samples and qRT-PCR assays (based on the two-step SYBR Green I dye assay) were performed as described elsewhere (Gil et al. 2011, 2014). Briefly, total RNA samples were treated with Ambion TURBO DNA-free kit to remove genomic DNA contamination, synthesis of complementary DNA was performed with the TaqMan Reverse Transcription Reagents kit and subsequent PCR step used the Power SYBRRGreen PCR Master Mix in a 7500 RT-PCR system (Applied Biosystems), according to Applied Biosystems Gene Expression Analysis Technical Manual. Specific primers were designed using Primer Express Software (Applied Biosystems) and are available in Supplementary Table S2. For each test gene and pesticide-exposure condition, transcript fold change (FC) values were calculated by the ∆∆CT method (according to the Applied Biosystems instructions). ACT1 transcript level was used as endogenous control. Normalized FC values are relative to the CT02 cells (used as calibrator). The FC values of a given gene for the different pesticide exposure concentrations are thus relative to the control CT02 cells (set as 1). Data reported are mean±SD values of at least three FC determinations for each independent biological triplicate of total RNA samples from each exposure condition. Statistical analysis to identify differences in FC values between the yeast cells exposed to different growth inhibitory effects of each pesticide (i.e., ~10, 20 and 50%), and between each one of these and the CT02 cells, was based on one-way ANOVA followed by the post-hoc Tukey test for unequal N, using Statistica 7.0 software (StaSoft, Inc., Tulsa, US). Differences were considered statistically significant for p values < 0.05.


Outline of the transcriptomic responses to pesticide stress in yeast cells

We tested the six pesticides at their equitoxic IC20 towards a standardized S. cerevisiae cell suspension (measured mean ± SD percentage of inhibition of yeast specific growth rate equal to 19.7 ± 2.6%) and obtained the respective transcriptional profiles. The complete lists of pesticide-responsive genes together with their respective FC values and description of protein function are provided as Supplementary Materials for CAB (Table S3), S-MET (Table S4), ALA (Table S5), DIU (Table S6), MCPA-ME (Table S7) and PYR (Table S8). The total numbers of genes with transcript levels significantly increased or decreased in the yeast cells upon exposure to the pesticide IC20 are outlined in Fig. 1a. Apparently, a larger transcriptomic response in terms of total numbers of differentially expressed genes was observed upon yeast exposure to the insecticide CAB (1063 total genes, corresponding to around 20% of all transcripts encoded in the yeast genome), when compared with the responses to the four herbicides and the fungicide (in average, ~8%) (Fig. 1a). For all the pesticides, approximately 80% of the differentially expressed genes in the datasets have known or inferred biological functions, the rest corresponding to uncharacterized open reading frames (Table S3 to S8).
Fig. 1

Overview of the effect of 2 h of exposure to equivalent toxic concentrations (20%-effective towards growth rate IC20) of the pesticides carbofuran (CAB), pyrimethanil (PYR), alachlor (ALA), diuron (DIU), S-metolachlor (S-MET) or methyl(4-chloro-2-methylphenoxy)acetate (MCPA-ME) in the transcriptome of Saccharomyces cerevisiae BY4741 cells. a Proportional diagrams illustrating the total numbers of genes with expression increased or decreased in yeast cells in response to pesticide-exposure compared to untreated control cells. b Dendrogram illustrating the comparison of the profiles of the differentially expressed genes between the six pesticides, obtained with hierarchical clustering analysis. Each square represents a single transcriptional profile, meaning that squares with the same pattern represent independent biological triplicates for each pesticide. The relative distance between clusters provides a measure of similarity among the represented gene expression profiles

The incorporation of all data relative to the differentially expressed genes (Table S3 to S8) in a hierarchical clustering analysis enabled visual interpretation of comparisons between the transcriptional profiles associated to the yeast responses to the six pesticides, which is presented in Fig. 1b. This analysis showed the yeast transcriptomic responses to be clearly distinguishable from each other, with the insecticide CAB having the most distinct profile (Fig. 1b). The fungicide PYR formed a separate cluster together with the four herbicides even though it was in a clearly separate branch (Fig. 1b). Regarding the four herbicides (ALA, S-MET, DIU, and MCPA-ME), they clustered close to each other but in distinct branches; it seems interesting that despite ALA and S-MET belonging to the same chemical family of chloroacetanilide herbicides (Table S1), their branches were clearly separated within the herbicide cluster (Fig. 1b).

Functional clustering and characterization of pesticide-responses

To disclose mechanistic aspects underlying yeast responses to the pesticides at the conditions under study, functional enrichment analyses of the datasets of differentially expressed genes was carried out and detailed analyses are available in Supplementary Tables S9, S10 and S11. Tables 1 and 2 summarize and compare the main overrepresented functional categories (p value < 0.01) found enriched in the datasets of repressed or induced genes, respectively. Overall, these analyses emphasized considerable differences in the yeast transcriptional response to the insecticide, the fungicide and the four herbicides (Tables 1 and 2), consistent with the hierarchical clustering results (Fig. 1b). Some major aspects of the comparative analyses relevant to provide biological profiles of the yeast response to each pesticide and possibly meaningful in the scope of (eco)toxicology are highlighted below and further interpreted in the discussion.
Table 1

Comparison of main functional categories enriched within the datasets of genes with expression significantly decreased in Saccharomyces cerevisiae BY4741 cells upon 2 h of exposure to the sub-lethal equitoxic 20%-effective concentrations of the pesticides carbofuran (CAB), pyrimethanil (PYR) (Gil et al. 2014), alachlor (ALA) (Gil et al. 2011), S-metolachlor (S-MET), methyl(4-chloro-2-methylphenoxy)acetate (MCPA-ME) or diuron (DIU)

Functional category











Amino acid metabolism

degradation of arginine



Phosphate metabolism



C-Compound and carbohydrate metabolism



C-2 compound and organic acid metabolism



Regulation of C-compound and carbohydrate metabolism



rRNA synthesis


tRNA synthesis


General transcription activities


Transcription termination


Transcriptional control



rRNA processing


mRNA processing (splicing, 5′, 3′-end proc)


Protein synthesis

Ribosome biogenesis


Translation initiation


Biogenesis of cellular components

Cell wall



Cytoskeleton//structural proteins


Transport, transport routes


Sugar transport



Amine/polyamine transport



RNA transport


Vitamin/cofactor transport


Drug/toxin transport



Heavy metal ion transport



Vesicular cellular import/endocytosis




Cell rescue, defence

Stress response



oxidative stress response



osmotic and salt stress response




Cell growth / morphogenesis


Budding, cell polarity, filament formation



Pheromone response, mating-type determination


Protein binding


RNA binding


Only classes found statistically overrepresented in the datasets compared with the frequency in the whole yeast genome (p value < 0.01) are represented

Table 2

Comparison of main functional categories enriched within the datasets of genes with expression significantly increased in Saccharomyces cerevisiae BY4741 cells upon 2 h of exposure to the sub-lethal equitoxic 20%-effective concentrations of the pesticides carbofuran (CAB), pyrimethanil (PYR) (Gil et al. 2014), alachlor (ALA) (Gil et al. 2011), S-metolachlor (S-MET), methyl(4-chloro-2-methylphenoxy)acetate (MCPA-ME) or diuron (DIU)

Functional category











Amino acid metabolism


Biosynthesis of glutamate


Biosynthesis of arginine



Metabolism of urea (urea cycle)



Metabolism of methionine



Biosynthesis of homocystheine


Metabolism of cysteine



Metabolism of tryptophan


Biosynthesis of tryptophan



Nitrogen, sulfur and selenium metabolism


Catabolism of nitrogenous compounds



Sulfur metabolism/sulfate assimilation



C-compound and carbohydrate metabolism



Lipid, fatty acid and isoprenoid metabolism



ergosterol biosynthesis



Metabolism of vitamins, cofactors, and prosthetic groups







metabolism of nonprotein amino acids




Glycolysis and gluconeogenesis



Pentose-phosphate pathway


Tricarboxylic acid pathway






Transport routes

Siderophore-iron transport/metal homeostasis


Anion transport



phosphate transport


C-compound and carbohydrate transport



C4-dicarboxylate transport


Amino acid/amino acid derivatives transport


Amine/polyamine transport


Lipid/fatty acid transport



Alantoin and allantoate transport


Drug/toxin transport



Cell rescue, defence

Oxidative stress response



Resistance proteins/chemical agente resistance




Homeostasis of cations / metal ions


Homeostasis of anions (e.g., phosphate)


Only classes found statistically overrepresented in the datasets compared with the frequency in the whole yeast genome (p value < 0.01) are represented.

Regarding the datasets of repressed genes, functional enrichment indicated the yeast response to CAB to be clearly distinct from the responses to the other five pesticides under study, given that it was the only one comprising remarkable numbers of overrepresented categories related with transcription, protein synthesis, RNA transport, cytoskeleton/structural proteins, cell growth/morphogenesis, budding, cell polarity and pheromone/mating-type (Table 1). Concretely, some of the CAB-repressed genes assigned to the transcription and protein synthesis functional categories included (Table S3, S10 and S11): (i) genes involved in transcription initiation at RNA polymerase protein promoters (CCL1) and encoding subunits of RNA polymerase (RPA34, RPB5, RPB10); (ii) the gene REF2 encoding a RNA-binding protein involved in the cleavage step of mRNA 3’-end formation prior to polyadenylation; (iii) genes involved in mRNA splicing (PBP4, MFM1, DCP1); (iv) genes encoding subunits of transcription factors involved in translation initiation (TIF35, TIF11, GDC2); and (v) genes involved in synthesis and processing of rRNA (RRP7) and ribosome biogenesis (RLP24, REX4). In addition, diverse genes found down-regulated by CAB were within overrepresented categories related with cell cycle process (SWI5 and ACE2 encoding transcription factors required for the expression of proteins related to cell cycle), cell growth and morphogenesis (FAR1, HOF1, RHO5), budding/cell polarity (DIG1, PHD1, FAR1, STE12, TEC1), pheromone/mating-type determination (AGA2, FUS1, FUS3, KAR4, STE2, STE12, STE4, STE6) and cytoskeleton/structural proteins (END3, HSP42) (Table S3 and S10). In contrast, a comparatively smaller number of PYR-repressed genes were related with translation initiation and cell growth/morphogenesis (Table S10 and S11). Notably, the PYR-repressed genes were mostly assigned to the enriched categories amino acid metabolism (degradation of arginine), phosphate metabolism or C-compound/carbohydrate metabolism (Table 1 and S10). These results substantiate distinguishable PYR- and CAB- transcriptional responses. In addition, clear differences between the major CAB- and PYR-enriched functional categories and those in the datasets of herbicide-responsive repressed genes seem also relevant (Table 1). However, characterization and comparisons among the four herbicide responses are not so clear due to the comparatively smaller numbers of enriched functional categories that could be retrieved from the respective datasets of repressed genes (Table 1, S10 and S11). In addition, within these datasets, there was no clear pattern of overrepresented functional categories shared by the four herbicides or all six pesticides (Table 1).

With respect to the datasets of induced genes, in general the major functional categories significantly enriched upon pesticide exposure were metabolism, energy, cellular transport/transported substrates and cell rescue/response to stress (Table 2). Yet, a remarkable number of differences in the responses elicited by CAB or PYR in the yeast were evident. Indeed, up-regulated transcriptional response to PYR seems mainly related with amino acid metabolism (mostly focused on biosynthesis of arginine and metabolism of sulfur amino acids), sulfate assimilation and secondary (non-protein amino acids) metabolism (Table 2, S9 and S11, and Gil et al. 2014). On the contrary, CAB-induced genes were mostly assigned to the functional categories energy metabolism, catabolism of nitrogenous compounds, and lipid, fatty acid and isoprenoid metabolism (Table 2, S9 and S11). In particular, remarkable numbers of genes encoding enzymes of the ergosterol biosynthesis pathway (ERG2, ERG3, ERG4, ERG5, ERG7, ERG9, ERG24, ERG25, ERG26, ERG27) and implicated in regulation of sterol biosynthesis (ERG28, HES1/OSH5) were significantly induced in yeast upon CAB exposure (Table 2 and S9). Functional categories related with homeostasis of metal ions, homeostasis of cations and homeostasis of anions were also overrepresented amongst genes induced in the CAB-stressed cells (Table 2, S9 and S11). For example, genes encoding proteins involved in the transport of ionic/ionizable solutes like phosphate (PHO89), sulfite (SSU1), ammonium (MEP1 and MEP2), or in the transport of urea/polyamines (DUR3) and siderophore-iron (SIT1 and FRE1), were significantly induced by CAB (Table S3 and S9). Comparatively, oxidative stress response and drug/toxin transport were functional categories significantly enriched within the PYR-dataset but not within the CAB’s dataset (Table 2, S9 and S11, and Gil et al. 2014). It is worth noting that, in general, many of the functional categories enriched in the CAB- and the PYR-datasets of induced genes were not found in the herbicide ones (Table 2). There is also no clear pattern of overrepresented functional categories in common to the four herbicide-responses (Table 2). Notably, the up-regulated transcriptional responses to ALA and S-MET, two structurally related herbicides (Table S1), comprised a number of interesting differences with respect to gene functional enrichment in the respective datasets as follows (Table S4, S5, S9 and S11, and Gil et al. 2011): (i) ergosterol biosynthesis in the case of S-MET (ERG3, ERG4, ERG5, ERG25, ERG26, ERG27, ERG28 and HES1/OSH5) but not in ALA’s; (ii) metabolism of phytosphingosine, a precursor of phytoceramide, and sphingolipids (RSB1, SUR, YPC1, LAC1) in ALA-response but not in S-MET’s; (iii) catabolism of allantoin (complete set of genes DAL1, DAL2, DAL3, DAL7, DUR1,2) in the case of S-MET, whereas most of these genes not significantly responsive to ALA; and (iv) biosynthesis of homocysteine, sulfur metabolism/sulfate assimilation and metabolism of vitamins/cofactors for ALA but not for S-MET (Table 2).

A few common features were however found shared by some of the pesticides (Tables 1 and 2). For example, transcriptional profiles of induced genes enriched in functional categories related with lipid/fatty acid metabolism and lipid/fatty acid transport were identified in CAB, S-MET-, ALA-, MCPA-ME- and DIU-stressed cells but not in PYR´s (Table 2). For example, the responses to stress exerted by the insecticide CAB and the herbicides S-MET- and MCPA-ME showed the induction of genes within the enriched category ergosterol biosynthesis in common (Table 2, S9 and S11). On the other hand, with respect to the major functional category nitrogen, sulfur and selenium metabolism, the pesticides ALA, S-MET, DIU and CAB increased the expression of genes involved in the catabolism of nitrogenous compounds, while only ALA and PYR induced genes comprising sulfur amino acid metabolism/sulfate assimilation (Table 2 and Gil et al. 2011, 2014). Transcriptional responses to three herbicides (ALA, S-MET and DIU) and the insecticide were enriched in C-compound and carbohydrate metabolism categories related with energy biosynthesis, which included the sub-categories tricarboxylic acid pathway (CAB, ALA and DIU), fermentation (CAB and ALA), glycolysis (ALA and DIU) and the pentose-phosphate pathway (DIU) (Table 2, Table S9 and S11). In addition, all pesticide-responses shared the overrepresented functional category detoxification and almost all (except CAB) had in common the overrepresented category resistance proteins/chemical agent resistance and drug/toxin transport (Table 2, Table S9 and S11). For example, among the pesticide-shared responsive genes were PDR5, PDR16, YOR1, SNQ2, PDR15 (all up-regulated) and PDR12 (down-regulated), which encode putative drug-efflux transporters of the ATP-binding-cassette-superfamily; also, induction of gene PDR3, a major transcriptional regulator of those genes, was shared (Table S9 and S11). These types of transcriptional modifications are related with pleiotropic drug resistance (PDR) in yeast cells (Kolaczowska and Goffeau 1999). In addition, all pesticide transcriptional responses elicited induction of genes involved in antioxidant response (Table S3 to S8), even though the functional categories oxidative stress response and peroxidase reaction were statistically significantly overrepresented only in ALA, S-MET, DIU and PYR responses (Table 2, S9 and S11); examples of these genes are HSP12 (encode a plasma membrane 12 kDa heat shock protein induced in response to heat-shock, oxidative stress and other stress conditions), GRE2 (encode a NADPH-dependent methylglyoxal reductase with a role in general stress response), SNQ2 (encode a putative PDR exporter also known to determine resistance to singlet oxygen species), GPX2 (encode a thioredoxin-dependent phospholipid hydroperoxide peroxidase) and GRX1/GRX2 (encode GSH-dependent disulfide oxidoreductases responsive to reactive oxygen species and depletion of the reduced glutathione GSH) (Table S3 to S8).

Pesticide specific responses

We further examined whether subsets of pesticide-unique differentially expressed genes could be of biological relevance to assist in the characterization of response mechanisms specific for each pesticide in the yeast. A simultaneous Venn comparison of the six complete datasets of pesticide-repressed or pesticide-induced genes was performed with VENNTURE software (Martin et al. 2012) (Fig. 2). The pesticide-responsive genes in each of the 63 pesticide-unique and pesticide-shared (all possible combinations) intersection subsets (Fig. 2) are included in Supplementary Tables S12 and S13 (induced and repressed genes, respectively). Of the total numbers of pesticide-responsive up-regulated genes (Fig. 1a), about 77, 47, 44, 34, 17 and 12% were apparently unique for CAB, PYR, S-MET, DIU, MCPA-ME and ALA responses, respectively (Fig. 2, Table S12). Consistently, similar percentages (87, 57, 50, 36, 18 and 14%, respectively) were found in the datasets of down-regulated genes (Fig. 2, Table S13). It became thus evident that variable numbers of induced or repressed genes were apparently specific of each pesticide response, with the highest numbers achieved in the CAB-stressed cells (>75%) and relatively lower values for the other five cases in the order PYR > S-MET > DIU > MCPA-ME, ALA (Fig. 2, Table S12 and S13).
Fig. 2

Venn diagrams illustrating the distribution and overlap of the six datasets of genes with expression significantly repressed (middle diagram) or induced (lower diagram) in yeast cells exposed for 2 h to the equitoxic 20%-effective concentration (IC20) of each pesticide. The upper diagram indicates the correspondence between the 63 pesticide-shared (57) or pesticide-unique (6) Venn intersections and the pesticide dataset(s) represented in each one (each letter corresponds to one pesticide as follows: A alachlor, C carbofuran, D diuron, M MCPA—methyl ester, P pyrimethanil, S S-metolachlor)

Regarding the CAB-response, the functional categories found significantly overrepresented within the CAB-unique subsets of differentially expresses genes (Table S12 and S13) were similar to the ones in the CAB-complete datasets (Tables 1, 2 and S11), with the exception of biosynthesis of homocysteine and catabolism of nitrogenous compounds (Table 2 and S11) which were not enriched in the former (Table S14). On what concerns the PYR-response, there was in general consistency in the functional categorization between the complete datasets (Tables 1 and 2) and the PYR-unique subsets (Table S14). On the contrary, in general, the smaller herbicide-unique subsets comprised only very few overrepresented functional categories (Table S14) and therefore further discussion (below) of the biological relevance of the yeast response to each herbicide toxicity is mostly based on the functional characterization of the complete microarray datasets (Table S9, S10 and S11). Interestingly, only few pesticide-responsive genes were common to all the pesticides or shared by the four herbicides (Fig. 2, Table S12 and S13).

Confirmation of gene expression changes

For yeast cells challenged with the IC20 of each pesticide, modifications in transcription of selected genes obtained by using Real-Time qRT-PCR in samples from independent exposure experiments confirmed the microarray data (Fig. 3). Notably, for almost all pesticide/gene combinations, the transcript levels varied consistently with the concentration-dependent effects in yeast growth (Fig. 3). These yeast genes were from overrepresented functional categories (Table S9 and S10) related with cell rescue and defense (YGP1), oxidative/general stress response (GRE2), sulfur amino acid metabolism/biosynthesis (MET28 and STR3), arginine metabolism/biosynthesis (ARG5,6), lipid/fatty acid metabolism and transport (ICT1, CRC1 and RSB1), carbohydrate transmembrane transport (HXT7) and iron transmembrane transport (FTR1) (detailed functional descriptions in Table S3 to S8). Some interesting particular features of their transcriptional profiles can be highlighted as follows (Fig. 3): (i) transcription of the gene ARG5,6 significantly increased only by PYR; (ii) genes ICT1, YGP1, MET28 and GRE2 induced, and genes HXT7 and FTR1 repressed, for all pesticides except CAB; (iii) RSB1 transcription increased by the four herbicides and the fungicide, but reduced by CAB (Fig. 3).
Fig. 3

Comparison of the effect of each pesticide (abbreviations as in Fig. 1) on the transcript levels obtained from Real-Time qRT-PCR assays for the indicated genes in Saccharomyces cerevisiae BY4741 cells. Cells were exposed during 2 h to concentrations of each pesticide causing increasing percentages of growth inhibition, namely ~10% (), 20%, IC20 (■) or 50% (□) relative to the untreated control cells. All fold change values were normalized based on transcript levels for the endogenous control gene ACT1 and are relative to the ratio of gene messenger RNA/ACT1 messenger RNA in the untreated control cells (set as 1). Error bars represent ± standard deviation. *, # or ⊗ indicate the mean is significantly different from the control cells or from the ~10% or 20%-inhibited cells, respectively. Numbers between brackets above each IC20’s column correspond to the fold change value for each gene in the respective microarray dataset and is indicated for comparison purposes; abs means absent or not significantly modified in the microarray dataset; nd means not determined by qRT-PCR


Prediction of mechanisms of response to pesticide toxicity

Overall results obtained from the comparative transcriptomic analyses upon short-term exposure to the six active substances tested at equitoxic sub-lethal levels highlighted the potential of gene expression profiling to distinguish the responses to toxicological effects of the pesticides of different chemical families in yeast cells. Yet some considerations related with the compounds’ lipophilic nature are noteworthy in the (eco)toxicological context. Of the six compounds under study, five have similar moderate log P values (between 2.68 and 3.09 at pH 7; Table S1) and CAB shows a slightly lower value of 1.8 (Table S1). Based on these values, biomembranes can be anticipated to be important cellular targets for all the substances at the herein used exposure conditions (pH 6.5). This means the pesticide molecules may accumulate in the lipid bilayers perturbing membrane function as selective permeability barrier and interfering with lipid bilayers and transmembrane enzymes, and/or may passively diffuse through the plasma membrane and reach target sites in different locations inside the cell (Cascorbi et al. 1993). This would dictate their similar tendencies to bioconcentrate and provoke cytotoxicity and adverse outcomes not only in yeast cells but also other biological systems (Cascorbi et al. 1993; Junghans et al. 2003; Papaefthimiou et al. 2004). Therefore, it would be expected to find a considerable number of gene expression changes related with remodeling of membrane lipid composition and limitation of membrane damage shared by all the six pesticides (Cascorbi et al. 1993; Viegas et al. 2005), which however was not the case. For instance, overrepresented functional categories like lipid/fatty acid metabolism and/or ergosterol metabolism and/or lipid/fatty acid transport were not identified in the cells challenged with PYR (log P = 2.84; Table S1). In addition, despite the similar lipophilic nature of the six molecules, the comparative transcriptomic analyses performed emphasized distinguishable yeast responses to each pesticide. Taken together, these observations thus suggest no clear correlation between transcriptomic responses to pesticide toxicity in yeast cells and pesticide lipophilicities, at least for the six active substances and environmental conditions herein studied. Therefore, the different pesticide molecules may presumably transverse the plasma membrane and cell wall at similar rates but reaching different molecular targets and evoking different response mechanisms in the yeast cells, even though ultimately eliciting a similar pesticide adverse outcome at phenotypic level (i.e., 20% growth inhibition).

That almost all pesticide IC20 exposures elicited induction of genes related with detoxification (all), resistance proteins/chemical agent resistance/toxin transport (except CAB), lipid/fatty acid/ergosterol membrane composition (except PYR) or oxidative stress response (except CAB and MCPA-ME) could be compatible with the notion of initiation of the so called environmental stress response (ESR) in the pesticide-stressed cells. The ESR is known to offer transient cellular protection and remodeling during early times regardless of the type of chemical stress involved (Gasch and Werner-Washburne 2002). However, this aspect remains unclear because other characteristic features of this general stress response, such as repression of genes involved in essential processes like nucleic acid biosynthesis, protein synthesis, ribosome biogenesis or cytoskeletal functions (Gasch and Werner-Washburne 2002) were not shared by all pesticides being mostly significantly triggered by CAB only. In addition, in general, expression of the genes MSN2 and/or MSN4 encoding two relevant ESR transcription factors (Gasch and Werner-Washburne 2002) was not pesticide-responsive for any of the active substances tested.

In the present work, patterns of biological responses arising from functional characterization of the transcriptional profiles in yeast cells might assist prediction of novel pesticide-specific mechanistic information. Even though the analysis is restricted to the transcriptional response and further information on post-transcription and translation is necessary to gain a more complete picture, possible relevance of these functional predictions in terms of novel clues on pesticide MoA is discussed below for each pesticide.

The carbamate insecticide carbofuran (CAB)

In the present work, yeast cells exposure to CAB produced, at the IC20, the most extreme modifications in gene expression, regarding both total numbers of differentially expressed genes (with ~75% being CAB-unique) and types of enriched functional categories. The primary MoA of CAB as a carbamates insecticide rely in the inhibition of the enzyme acetylcholinesterase (Bocquené et al. 1995; Hernández-Moreno et al. 2011) and as far as it is known the yeast does not possess a neurological system. Therefore, these results may reflect CAB potential to impact yeast and other eukaryotic cells in diverse ways which may account for additional cytotoxicity and side-effects in both target and non-target organisms. This aspect is of ecotoxicological relevance and, therefore, biological interpretation of some of these transcriptional alterations in yeast cells is further discussed below.

The CAB’s transcriptional profile in yeast was found clearly distinct from the transcriptional responses to the fungicide and the four herbicides in that it was the only one involving repression of a remarkably high number of genes related with transcription, protein synthesis, cell growth/morphogenesis, budding, cell polarity and pheromone/mating-type. These results suggest CAB may be able to cause a generalized drop in transcription and protein synthesis in the stressed yeast cells. Declines in these types of biological processes have been reported as part of the general ESR, which could contribute for energy savings under stress conditions while yeast cells adapt to new growth conditions (Gasch and Werner-Washburne 2002). Yet, in the present work, this type of explanation seems unlikely because all the six pesticides exerted similar effects towards yeast growth but only the CAB-repressed transcriptional profile was enriched in transcription and protein synthesis functional categories. It can be hypothesized that impacts on those pathways may constitute secondary effects of CAB action in eukaryotic cells and possibly contributing, at least partially, to observed reductions in, for instance, cell density and physiology in freshwater flagellate microalga (Azizullah et al. 2011), detoxification enzyme activity in fish (Hernández-Moreno et al. 2011) or hematological and biochemical parameters in fish (Harabawy and Ibrahim 2014). Given the existence of numerous reports of CAB adverse outcomes in diverse terrestrial and aquatic microbial (Azizullah et al. 2011; Mansano et al. 2016; Megharaj et al. 1993) and higher eukaryotes (Bocquené et al. 1995, Chelinho et al. 2012; Harabawy and Ibrahim 2014; Hernández-Moreno et al. 2011; Kaur et al. 2012; Saxena et al. 2014), this hypothesis merits further investigation.

In the present work, other apparently relevant transcriptional responses to CAB in yeast cells can be highlighted. First, the repression of genes related with cytoskeleton/structural proteins suggest that CAB may impact cytoskeleton related processes at least in yeast cells. Actin cytoskeleton plays critical roles in processes such as cytokinesis, cell polarity, cell morphogenesis and endocytosis, in both fungal and animal cells (Pruyne et al. 2004). Second, the concomitant induction of a remarkable number of genes from the ergosterol biosynthesis pathway, suggest CAB may impact membrane fluidity and permeability in yeast cells through ergosterol biosynthesis. It seems relevant that some of these yeast genes (e.g., ERG28 and HES1) have homologs implicated in the regulation of sterol biosynthesis in mammalian cells and that sterol biosynthesis share conserved steps in fungi, plants and mammals (Gachotte et al. 2001, Jiang et al. 1994). Interestingly, the freshwater flagellate Euglena gracilis was reported to become more compact and with rounder shapes upon CAB exposure and this was attributed to CAB interference with non-identified plasma membrane characteristics (Azizullah et al. 2011; Megharaj et al. 1993). Additionally, a possible toxicity mechanism of CAB towards soil earthworm species was proposed to involve its interference with protein/lipid molecules of earthworm body wall (Saxena et al. 2014). Third, the transcriptomic data also suggest CAB may disturb ionic homeostasis in yeast cells, as proposed regarding inhibition of E. gracilis mobility by CAB (Azizullah et al. 2011). On the other hand, in the present work, the datasets of CAB-responsive genes did not appear to be significantly enriched in categories related with oxidative stress. However, neurobehavioral alterations in rats exposed to CAB were reported before as being possibly due to non-cholinergic mechanisms associated with impairment of mitochondrial respiratory chain functions leading to oxidative stress (Kamboj et al. 2008). Other authors suggested toxicity of CAB in rats to possibly occur through the induction of oxidative stress by changing the normal activity of antioxidant enzymes and/or through the depletion of antioxidants in the cell (Kaur et al. 2012). Given the biological importance of the above referred gene expression modifications in terms of potential adverse outcomes of CAB exposure their meaningfulness in more complex organisms deserves further in depth exploration.

The chloroacetanilide herbicides alachlor (ALA) and S -metolachlor ( S -MET)

In the present study, the hierarchical clustering analysis and the comparative categorization of the microarray data highlighted a number of differences between the ALA and S-MET transcriptional responses in yeast cells, which may be relevant in the ecotoxicological context and are further discussed below. Mechanistic predictions from the transcriptomic response to the IC20 of ALA in yeast cells were analyzed in detail in previous reports (Gil et al. 2011, 2017). Briefly, the short-term toxic response elicited by ALA action in yeast cells appeared to be related with: (i) turnover of ceramide and metabolism/transport of sphingolipids, which we proposed to be presumably associated with possible inhibition of very-long-chain fatty acid enlongation in yeast cells (Gil et al. 2011), similarly to the MoA proposed for chloroacetanilides in algae and plants (Junghans et al. 2003; Schmalfuβ et al. 2000); (ii) sulfate assimilation and metabolism/biosynthesis of sulfur amino acids, in particular of methionine and S-Adenosylmethionine formation (Gil et al. 2011), which can have a link with oxidative stress and appears to be associated with sulfur metabolism disorders in diverse organisms (Cavallaro et al. 2010); (iii) iron homeostasis, which is highly conserved from fungi to animals and critical for their wellbeing (Gil et al. 2017, Kaplan and Kaplan 2009); and (iv) detoxification (e.g., oxidative stress, transmembrane toxin efflux) (Gil et al. 2011). Notably, functional categories referred in (i) to (iii) were not found significantly enriched in the S-MET datasets.

In particular, both ALA and S-MET evoked responses associated with lipid metabolism and transport but apparently involving different pathways, namely biosynthesis of phytosphingosine (precursor of phytoceramide) and sphingolipids by ALA but not by S-MET, and biosynthesis of ergosterol in S-MET-response but not in ALA’s (Gil et al. 2011 and the present work). Ergosterol is a cholesterol-like lipid that is within the major fungal membrane lipids, also including sphingolipids and glycerophospholipids (van der Rest et al. 1995). These types of lipids are vital for maintenance of appropriate membrane fluidity and functioning, from yeast to animal cells (Gachotte et al. 2001). In the yeast there are evidences of compensatory cross-talk between ergosterol and sphingolipid metabolism involving the (very)-long-chain fatty acid elongase system (Eisenkolb et al. 2002; Swain et al. 2002). Whether the latter may be affected by both herbicides in yeast cells, as it is proposed for plants and microalgae (Junghans et al. 2003; Schmalfuβ et al. 2000), is not clear, but it could lead to imbalances in lipid composition thus affecting membranes fluidity and permeability. Based on these transcriptional clues, we speculate that the ALA- and the S-MET-stressed cells could respond with different strategies to modulate sphingolipid and sterol levels, respectively, in order to achieve proper membrane fluidity and function under chemical stress.

Other distinct pattern in the transcriptomic response to the two herbicides was the induction by S-MET of the complete set of genes involved in the catabolism of allantoin, whereas most of them were not found in the dataset of ALA-responsive genes. Allantoin is a normal cellular constituent that can function as a reserve source of intracellular nitrogen (Yoo et al. 1985). These results may suggest ALA and S-MET stressed cells to somehow sense ammonium and/or amino acid limitation differently.

Overall, the generally distinct transcriptomic responses elicited by ALA and S-MET in the yeast model suggest different features regarding their molecular targets in biological systems. Specific structural differences of chloroacetanilide molecules were proposed as main cause for their different reactivity towards enzyme active sites (Eckermann et al. 2003) and impact in different biological systems (Genter et al. 2009; Kale et al. 2008; Saha et al. 2012), even though the actual mechanisms remain unclear. We hypothesize the different chemical structure and size of the alkoxyalkyl groups attached to the N-atom in the two herbicide molecules, i.e., the methoxymethyl in ALA and the bulkier methoxymethylethyl in S-MET (Table S1), as possibly imparting different levels of interaction/binding with molecular target sites in the cell. These differences appear to be meaningful with respect to adverse outcomes in non-target eukaryotic cells. ALA and MET (racemic mixture) have been reported to differ significantly as substrates for biotransformation enzymes in rat and human liver microsomes (Coleman et al. 2000; Kale et al. 2008). A study of specific interactions of chloroacetanilide herbicides with human ABC transmembrane transporters (MDR1, MRP1, MRP2 and BCRP) indicated both ALA and MET, among other chloroacetanilides, to be able to modulate intestinal xenobiotic absorption through inhibition of MDR1 but their affinity for MDR1 and the range of concentrations that stimulated its ATPase activity were slightly different between each other (Oosterhuis et al. 2008). At higher levels of biological organization, ALA and MET also diverged with respect to cytotoxicity towards either rat or human hepatocytes (Kale et al. 2008) as well as to site-specificity (and potency) concerning tumor induction in rats, namely liver for MET or nasal, thyroid and gastric for ALA (Genter et al. 2009). Importantly, these observations are consistent with the idea that the two herbicides must be divided into separate risk groups (Coleman et al. 2000; Kale et al. 2008).

The phenylurea herbicide diuron (DIU)

DIU is known to block electron transfer at the level of photosystem II in plants and photosynthetic microorganisms (Giacomazzi and Cochet 2004), but its MoA in non-photosynthetic organisms is not fully understood. In the present work, the functional category metabolism of tryptophan was overrepresented only within the DIU dataset. It involved the induction of ARO9 (first step of tryptophan catabolism) and of the genes encoding enzymes (Bna1p to Bna7p) that in S. cerevisiae comprise the de novo pathway for biosynthesis of nicotinic acid mononucleotide from the amino acid tryptophan (Lin and Guarente 2003). The nicotinic acid mononucleotide is a precursor of the cofactor nicotinamide adenine dinucleotide (NAD) that is essential to keep balanced intracellular redox state and is known to participate in many essential biological processes such as transcription, energy metabolism and DNA repair in biological systems (Lin and Guarente 2003). In both microbial and higher eukaryotes, synthesis of NAD is generally carried out through the de novo pathway (synthesis from tryptophan) or by recycling degraded NAD products (the salvage pathway) (Lin and Guarente 2003). We thus speculate these results may suggest existence of increased NAD requirements in the DIU stressed yeast cells, possibly due to a need to maintain NAD in sufficient amount to function as metabolic regulator of the NAD:NADH redox ratio in response to oxidative stress and/or energy metabolism alterations. Interestingly, among the DIU-induced genes found in overrepresented functional categories, some were associated with oxidative stress detoxification and others with energy metabolism. As far as we are aware, the prediction that DIU may interfere with tryptophan metabolism and with de novo biosynthesis of NAD in yeast cells is new and deserves further investigation. In this respect it is noteworthy that NAD levels and NAD:NADH ratio imbalances may have adverse outcomes related with not only yeast life-span, but also longevity and several age-associated diseases in higher eukaryotes (Lin and Guarente 2003).

Consistently, induction of genes related with protection of cells against oxidative and nitrosative stress and/or depletion of endogenous antioxidants like GSH (Herrero et al. 2008) were within the transcriptional response evoked by DIU. Even though DIU shared these types of gene expression modifications with ALA, S-MET and PYR (Gil et al. 2011, 2014, 2017 and present work) and the presence of reactive oxygen species was not determined experimentally in the DIU-stressed cells, we suggest a putative antioxidant response might be attributable to side-effects of DIU possibly associated with interference with electron transfer in oxidative phosphorylation process and/or imbalance of NAD:NADH redox ratio (discussed above). A number of reports showed evidences of DIU-induced in vitro cytogenetic effects in chinese hamster cell-lines (Federico et al. 2011) and dose-dependent (125–2500 ppm) systemic and organ-specific toxicity (e.g., urinary bladder, spleen, liver) in male Wistar rats (Domingues et al. 2011; Ihlaseh et al. 2011). In particular, increased levels of reactive oxygen species were suggested to contribute to bladder cell necrosis and urothelial hyperplasia (Ihlaseh et al. 2011).

The chlorophenoxyherbicide MCPA-methyl ester (MCPA-ME)

Like other MCPA ester forms, MCPA-ME can undergo hydrolysis to the ionizable acid MCPA form in environments at neutral or higher pH (van Ravenzwaay et al. 2004). Toxicokinetic and biotransformation studies of diverse MCPA derivatives (acid, salt or ester forms) in rats showed them all to be rapidly converted and excreted as acid MCPA, regardless of the chemical form the animals were exposed to (van Ravenzwaay et al. 2004). The acid MCPA have potential for exerting toxicological effects in non-target plants (Peixoto et al. 2009), yeast and frogs (Papaefthimiou et al. 2004).

In the present work, exposure to MCPA-ME elicited, at the IC20, the least intense and functionally least diverse transcriptional modifications in yeast cells. Despite showing a transcriptional profile clearly distinct from the other five equitoxic exposure conditions, indicated by hierarchical clustering analysis, very few functional categories were however found significantly enriched amongst the datasets of genes which were repressed or induced by MCPA-ME; even the few ones found enriched comprised low numbers of differentially expressed genes. This fact hampered ability to retrieve biological relevance from the MCPA-ME-responsive transcriptome.

Functional categories related with yeast pleiotropic drug resistance (PDR) phenomena (Kolaczowska and Goffeau 1999) were however significantly populated with MCPA-ME responsive genes. This is consistent with the previously reported yeast transcriptional response to the structurally similar herbicide 2,4-D (Teixeira et al. 2007). Whether one or more of those PDR transmembrane efflux pumps have a role in protecting yeast cells from MCPA-ME stress remains to be experimentally demonstrated. It is possible that the lipophilic MCPA-ME (logP ~2.7;Table S1) may rapidly permeate the cell plasma membrane by passive diffusion, followed by its intracellular hydrolysis to the acid form MCPA by non-specific esterases, as proposed to occur in rat cells (van Ravenzwaay et al. 2004). Then, in the almost neutral cytosol, the weak acid MCPA dissociates leading to intracellular acidification and accumulation of the respective anion which cannot passively cross the plasma membrane and can cause adverse effects in enzymes and cell physiology (Teixeira et al. 2007; Peixoto et al. 2009). Consistently, yeast cells exposed to moderately toxic levels of acid MCPA (0.14 mM) were reported before to have their intracellular pH (pHi) dropped from ~6.7 to 5.6 in 5 h (Cabral et al. 2004). In these MCPA-stressed cells, plasma membrane H+-ATPase activity was stimulated and cellular volume decreased, presumably as a way to restore pHi homeostasis during recovery from MCPA stress (Cabral et al. 2004). We suggest that yeast cells adaptation under MCPA-ME stress could also involve pumping out the herbicide molecule and/or the acid and anionic forms resulting from its hydrolysis and dissociation in the neutral cytosol (van Ravenzwaay et al. 2004) presumably through transmembrane PDR efflux pumps, as observed in response to diverse xenobiotics (Kolaczowska and Goffeau 1999; Teixeira et al. 2007). It should be recognized that, in the present work, this type of functional response was elicited by not only MCPA-ME but also the other three herbicides and the fungicide. This stress response does not generally occur without expenditure of metabolic energy and redox balance and can thus contribute to slight-moderate adverse outcomes at physiological level depending on the severity of stress conditions (Gasch and Werner-Washburne 2002; Teixeira et al. 2007).

The anilinopyrimidine fungicide pyrimethanil (PYR)

The comparative transcriptomic analyses performed in the present work emphasized the significant up-regulation of genes involved in sulfate assimilation and metabolism/synthesis of sulfur amino acid metabolism (methionine, homocysteine, cysteine, S-adenosylmethionine) by PYR (Gil et al. 2014 and present work). Despite this observation seeming consistent with the MoA related with inhibition of methionine biosynthesis proposed for PYR in the target-fungi B. cinerea (Fritz et al. 2003), we should be aware of the fact that this type of transcriptional response was shared with the non-related herbicide ALA.

Remarkably, the transcriptional response evoked by this fungicide showed a unique feature in that it was the only active substance herein tested encompassing the enriched functional category metabolism/biosynthesis of L-arginine. Indeed, it included the notable coordinated induction of the whole set of ARG genes encoding enzymes of the arginine biosynthesis pathway and the repression of the two genes involved in arginine degradation; mechanistic predictions from transcriptomic response to the IC20 of PYR in yeast cells were analyzed in detail in Gil et al. (2014). In particular, the predicted involvement of arginine metabolism as a specific target of PYR toxicity may reflect a novel aspect of the PYR MoA as fungicide (Gil et al. 2014) and deserves further studies in target and non-target eukaryotes. Arginine and metabolites (e.g., citruline and ornithine) have important biological roles in cells, and imbalances in arginine availability have been considered to influence development and health in mammals (Wu et al. 2009) and fish (Capiotti et al. 2013).

In conclusion, with the comparative genomic approach performed, we came across a number of predictions of biological pathways and mechanisms that may contribute to improve understanding of responses to pesticide toxicity and revealing potential new aspects of the MoA of each pesticide in the yeast model. It should however be recognized that this knowledge is not necessarily enough in its own right to devise adverse outcomes at higher levels of biological organization relevant to ecological risk assessment (Edwards et al. 2016), given the existence of differences in xenobiotic susceptibility across species. In addition, there are differences between the pesticide concentrations herein used and those possibly present in the environment when pesticides are applied in line with good agricultural practices and regulatory frameworks. Nevertheless, we suggest the molecular hints disclosed in the present work in yeast cells to contribute to predict possible secondary MoA and potential toxicological outcomes of pesticide exposure meaningful for other eukaryotes from ecosystems, especially in situations of inadvertent high environmental contamination. They might contribute to guide future research in the (eco)toxicological field when planning toxicological studies in more complex and experimentally less accessible eukaryotes, which may surely be needed to unravel uncertainties, fill gaps and refine MoA (Edwards et al. 2016). It seems relevant in this context that proteins encoded by a number of these pesticide-responsive genes share significant sequence and functional conservation with proteins of diverse fungal and higher eukaryotes (Gil et al. 2011, 2014, 2017). The obtained data also have potential in disclosing molecular indicators of toxicity that may be useful in development of yeast-based bioassays for biomonitoring. Recently, a yeast-based gene expression assay built on measurements of FC in transcription of the PYR-responsive genes ARG3 and ARG5,6 was found suitable for the preliminary screening of toxicity of eluates from PYR-sprayed soils (Gil et al. 2015).



We acknowledge financial support by grant UID/BIO/04565/2013 from Portuguese FCT (Fundação para a Ciência e Tecnologia) and grant LISBOA-01-0145-FEDER-007317 from POR Lisboa 2020 (Programa Operacional Regional de Lisboa 2020) awarded to iBB-Institute for Bioengineering and Biosciences, and by funds from FEDER (European Funds for Economic and Regional Development), COMPETE (Competitiveness Factors Operational Programme) and Portuguese National Funds through the FCT funded research project PTDC/AMB/64230/2006 (ended in 2011) and fellowship SFRH/BD/60933/2009 (for F.N.G., ended in 2014).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Supplementary material

10646_2018_1929_MOESM1_ESM.pdf (2 mb)
Supplementary Material
10646_2018_1929_MOESM2_ESM.xlsx (72 kb)
Supplementary Material


  1. Abrantes N, Pereira R, Gonçalves F (2010) Occurrence of pesticides in water, sediments, and fish tissues in a lake surrounded by agricultural lands: concerning risks to humans and ecological receptors. Water Air Soil Pollut 212:77–88CrossRefGoogle Scholar
  2. Azizullah A, Richter P, Häder DP (2011) Comparative toxicity of the pesticides carbofuran and malathion to the freshwater flagellate Euglena gracilis. Ecotoxicology 20:1442–1454CrossRefGoogle Scholar
  3. Bocquené G, Bellanger C, Cadiou Y, Galgani F (1995) Joint action of combinations of pollutants on the acetylcholinesterase activity of several marine species. Ecotoxicology 4:266–279CrossRefGoogle Scholar
  4. Braconi D, Bernardini G, Santucci A (2016) Saccharomyces cerevisiae as a model in ecotoxicological studies: a post-genomics perspective. J Proteom 137:19–34CrossRefGoogle Scholar
  5. Cabral MG, Sá-Correia I, Viegas CA (2004) Adaptative responses in yeast to the herbicide 2-methyl-4-chlorophenoxy acetic acid at the level of intracellular pH homeostasis. J Appl Microbiol 96:603–612CrossRefGoogle Scholar
  6. Capiotti KM, Fazenda L, Nazario LR, Menezes FP, Kist LW, Bogo MR, Bonan CD (2013) Arginine exposure alters ectonucleotidase activities and morphology of zebrafish larvae (Danio rerio), Int J Devl Neurosci 31:75–81CrossRefGoogle Scholar
  7. Cascorbi I, Bittrich H, Ricklinkat J, Voss W, Seyfarth A, Forêt M (1993) Effects of a heterogenous set of xenobiotics on growth and plasma membranes of mammalian and fungal cell cultures. Ecotoxicol Environ Saf 26:113–126CrossRefGoogle Scholar
  8. Cavallaro RA, Fuso A, Nicolia V, Scarpa S (2010) S-adenosylmethionine prevents oxidative stress and modulates glutathione metabolism in TgCRND8 mice fed a B-vitamin deficient diet. J Alzheimer Dis 20:997–1002CrossRefGoogle Scholar
  9. Chelinho S, Lopes I, Natal-da-Luz T, Domene X, Nunes MET, Espíndola ELG, Ribeiro R, Sousa JP (2012) Integrated ecological risk assessment of pesticides in tropical ecosystems: a case study with carbofuran in Brazil. Environ Toxicol Chem 31:437–445CrossRefGoogle Scholar
  10. Chirnside AEM, Ritter WF, Radosevich M (2007) Isolation of a selected microbial consortium from a pesticide mix-load site soil capable of degrading the herbicides atrazine and alachlor. Soil Biol Biochem 39:3056–3065CrossRefGoogle Scholar
  11. Cherry JM, Hong EL, Amundsen C, Balakrishnan R, Binkley G, Chan ET, Christie KR, Costanzo MC, Dwight SS, Engel SR, Fisk DG, Hirschman JE, Hitz BC, Karra K, Krieger CJ, Miyasato SR, Nash RS, Park J, Skrzypek MS, Simison M, Weng S, Wong ED (2012) Saccharomyces Genome Database: the genomics resource of budding yeast. Nucleic Acids Res 40:D700–D705. Database issueCrossRefGoogle Scholar
  12. Coleman MD, Neil JDO, Woehrling EK, Bate O, Ndunge A, Hill EJ, Menach A, Reiss CJ (2012) A preliminary investigation into the impact of a pesticide combination on human neuronal and glial cell lines in vitro. PLoS ONE 7:e42768CrossRefGoogle Scholar
  13. Coleman S, Linderman R, Hodgson E, Rose RL (2000) Comparative metabolism of chloroacetamide herbicides and selected metabolites in human and rat liver microsomes. Environ Health Perspect 108:1151–1157Google Scholar
  14. DeLorenzo ME, Danese LE, Baird TD (2013) Influence of increasing temperature and salinity on herbicide hoxicity in estuarine phytoplankton. Environ Toxicol 28:359–371CrossRefGoogle Scholar
  15. Domingues A, Barbisan LF, Martins PR, Spinardi-Barbisan AL (2011) Diuron exposure induces systemic and organ-specific toxicity following acute and sub-chronic exposure in male Wistar rats. Environ Toxicol Pharmacol 31:387–396CrossRefGoogle Scholar
  16. EC—European Community (2013) Directive 2013/39/EU of the European Parliament and of the Council of 12August2013 amending directives 2000/60/EC and 2008/105/EC as regards priority substances in the field of water policy. Official Journal of the EU, L226 of 24.08.2013. Accessed 13 July 2017
  17. EC—European Community (2009) Regulation No 1107/2009 of the European Parliament and of the Council of 21 October 2009 concerning the placing of plant protection products on the market and repealing Council Directives 79/117/EEC and 91/414/EEC. Official Journal of the EU, L309 of 24.11.2009. Accessed 13 July 2017
  18. Eckermann C, Matthes B, Nimtz M, Reiser V, Lederer B, Böger P, Schröder J (2003) Covalent binding of chloroacetamide herbicides to the active site cysteine of plant type III polyketide synthases. Phytochemistry 64:1045–1054CrossRefGoogle Scholar
  19. Eisenkolb M, Zenzmaier C, Leitner E, Schneiter R (2002) A specific structural requirement for ergosterol in long-chain fatty acid synthesis mutants important for maintaining raft domains in yeast. Mol Biol Cell 13:4414–4428CrossRefGoogle Scholar
  20. Edwards SW, Tan YM, Villeneuve DL, Meek ME, McQueen CA (2016) Adverse outcome pathways—organizing toxicological information to improve decision making. J Pharmacol Exp Ther 356:170–181CrossRefGoogle Scholar
  21. Federico C, Motta S, Palmieri C, Pappalardo M, Librando V, Saccone S (2011) Phenylurea herbicides induce cytogenetic effects in Chinese hamster cell lines. Mutat Res 721:89–94CrossRefGoogle Scholar
  22. Fritz R, Lanen C, Chapeland-Leclerc F, Leroux P (2003) Effect of the anilinopyrimidine fungicide pyrimethanil to the cystathionine β-lyase of Botrytis cinerea. Pest Biochem Physiol 77:54–65CrossRefGoogle Scholar
  23. Fulton MH, Key PB (2001) Acetylcholinesterase inhibition in estuarine fish and invertebrates as an indicator of organophosphorus insecticide exposure and effects. Environ Toxicol Chem 20:37–45CrossRefGoogle Scholar
  24. Gachotte D, Eckstein J, Barbuch R, Hughes T, Roberts C, Bard M (2001) A novel gene conserved from yeast to humans is involved in sterol biosynthesis. J Lipid Res 42:150–154Google Scholar
  25. Gasch AP, Werner-Washburne M (2002) The genomics of yeast responses to environmental stress and starvation. Funct Integr Genom 2:181–192CrossRefGoogle Scholar
  26. Genter MB, Warner BM, Medvedovic M, Sartor MA (2009) Comparison of rat olfactory mucosal responses to carcinogenic and non-carcinogenic chloracetanilides. Food Chem Toxicol 47:1051–1057CrossRefGoogle Scholar
  27. Geret F, Burgeot T, Haure J, Gagnaire B, Renault T, Communal PY, Samain JF (2011) Effects of low-dose exposure to pesticide mixture on physiological responses of the Pacific oyster (Crassostrea gigas) Environ Toxicol 28:689–699CrossRefGoogle Scholar
  28. Giacomazzi S, Cochet N (2004) Environmental impact of diuron transformation: a review. Chemosphere 56:1021–1032CrossRefGoogle Scholar
  29. Gil FN, Becker JD, Viegas CA (2014) Potential mechanisms underlying response to effects of the fungicide pyrimethanil from gene expression profiling in Saccharomyces cerevisiae. J Agric Food Chem 62:5237–5247CrossRefGoogle Scholar
  30. Gil FN, Belli G, Viegas CA (2017) The Saccharomyces cerevisiae response to stress caused by the herbicidal active substance alachlor requires the iron regulon transcription factor Aft1p. Environ Microbiol 19:485–499CrossRefGoogle Scholar
  31. Gil FN, Gonçalves AC, Jacinto MJ, Becker JD, Viegas CA (2011) Transcriptional profiling in Saccharomycescerevisiae relevant for predicting alachlor mechanisms of toxicity. Environ Toxicol Chem 30:2506–2518CrossRefGoogle Scholar
  32. Gil FN, Moreira-Santos M, Chelinho S, Pereira C, Feliciano JR, Leitão JH, Sousa JP, Ribeiro R, Viegas CA (2015) Suitability of a Saccharomyces cerevisiae-based assay to assess the toxicity of pyrimethanil sprayed soils via surface runoff: comparison with standard aquatic and soil toxicity assays. Sci Total Environ 505:161–171CrossRefGoogle Scholar
  33. Harabawy ASA, Ibrahim ATA (2014) Sublethal toxicity of carbofuran pesticide on the African catfish Clarias gariepinus (Burchell, 1822): hematological, biochemical and cytogenetic response. Ecotoxicol Environ Saf 103:61–67CrossRefGoogle Scholar
  34. Herrero E, Ros J, Bellí G, Cabiscol E (2008) Redox control and oxidative stress in yeast cells. Biochim Biophys Acta 1780:1217–1235CrossRefGoogle Scholar
  35. Hernández-Moreno D, Pérez-López M, Soler F, Gravato C, Guilhermino L (2011) Effects of carbofuran on the sea bass (Dicentrarchus labrax L.): study of biomarkers and behaviour alterations. Ecotoxicol Environ Saf 74:1905–1912CrossRefGoogle Scholar
  36. Ihlaseh SM, Bailey KA, Hester SD, Jones C, Ren H, Cardoso APF, MLCS Oliveira, Wolf DC, de Camargo JLV (2011) Transcriptional profile of diuron-induced toxicity on the urinary bladder of male Wistar rats to inform mode of action. Toxicol Sci 122:330–338CrossRefGoogle Scholar
  37. Jiang B, Brown JL, Sheraton J, Fortin N, Bussey H (1994) A new family of yeast genes implicated in ergosterol synthesis is related to the human oxysterol binding protein. Yeast 10(3):341–353CrossRefGoogle Scholar
  38. Junghans M, Backhaus T, Faust M, Scholze M, Grimme LH (2003) Predictability of combined effects of eight chloroacetanilide herbicides on algal reproduction. Pest Manag Sci 59:1101–1110CrossRefGoogle Scholar
  39. Kale VM, Miranda SR, Wilbanks MS, Meyer SA (2008) Comparative cytotoxicity of alachlor, acetochlor, and metolachlor herbicides in isolated rat and cryopreserved human hepatocytes. J Biochem Mol Toxicol 22:41–50CrossRefGoogle Scholar
  40. Kamboj SS, Kumar V, Kamboj A, Sandhir R (2008) Mitochondrial oxidative stress and dysfunction in rat brain induced by carbofuran exposure. Cell Mol Neurobiol 28:961–969CrossRefGoogle Scholar
  41. Kaplan CD, Kaplan J (2009) Iron acquisition and transcriptional regulation. Chem Rev 10:4536–4552CrossRefGoogle Scholar
  42. Kaur B, Khera A, Sandhir R (2012) Attenuation of cellular antioxidant defense mechanisms in kidney of rats intoxicated with carbofuran. J Biochem Mol Toxicol 26:393–398CrossRefGoogle Scholar
  43. Kim HJ, Rakwal R, Shibato J, Iwahashi H, Choi JS, Kim DH (2006) Effect of textile wastewaters on Saccharomycescerevisiae using DNA microarray as a tool for genome-wide transcriptomics analysis. Water Res 40:1773–1782CrossRefGoogle Scholar
  44. Kohrer K, Domdey H (1991) Preparation of high molecular weight RNA. Methods Enzymol 194:398–405CrossRefGoogle Scholar
  45. Kolaczowska A, Goffeau A (1999) Regulation of pleiotropic drug resistance in yeast. Drug Resist Updat 2:403–414CrossRefGoogle Scholar
  46. Lewis KA, Tzilivakis J, Warner D, Green A (2016) An international database for pesticide risk assessments and management. Hum Ecol Risk Assess 22:1050–1064CrossRefGoogle Scholar
  47. Lin SJ, Guarente L (2003) Nicotinamide adenine dinucleotide, a metabolic regulator of transcription, longevity and disease. Curr Opin Cell Biol 15:241–246CrossRefGoogle Scholar
  48. Liu H, Ye W, Zhan X, Liu W (2006) A comparative study of rac- and S-metolachlor toxicity to Daphnia magna. Ecotoxicol Environ Saf 63:451–455CrossRefGoogle Scholar
  49. Mansano AS, Moreira RA, Pierozzi M, Oliveira TMA, Vieira EM, Rocha O, Regali-Seleghim MH (2016) Effects of diuron and carbofuran pesticides in their pure and commercial forms on Paramecium caudatum: the use of protozoan in ecotoxicology. Environ Pollut 213:160–172CrossRefGoogle Scholar
  50. Martin B, Chadwick W, Yi T, Park S-S, Lu D, Ni B, Gadkaree S, Farhang K, Becker KG, Maudsley S (2012) VENNTURE—A novel Venn diagram investigation tool for multiple pharmacological dataset analysis. PLoS ONE 7:e36911CrossRefGoogle Scholar
  51. Megharaj M, Pearson HW, Venkateswarlu K (1993) Toxicity of carbofuran to soil isolates of Chlorella vulgaris, Nostoc linckia and N. muscorum. Appl Microbiol Biotechnol 39:644–648CrossRefGoogle Scholar
  52. Milling RJ, Richardson CJ (1995) Mode of action of the anilino-pyrimidine fungicide pyrimethanil. 2. Effects on enzyme secretion in Botrytis cinerea. Pest Sci 45:43–48CrossRefGoogle Scholar
  53. Oosterhuis B, Vukman K, Vági E, Glavinas H, Jablonkai I, Krajcsi P (2008) Specific interactions of chloroacetanilide herbicides with human ABC transporter proteins. Toxicology 248:45–51CrossRefGoogle Scholar
  54. Papaefthimiou C, Cabral MDG, Mixailidou C, Viegas CA, Sá-Correia I, Theophilidis G (2004) Comparison of two screening bioassays, based on the frog sciatic nerve and yeast cells, for the assessment of herbicide toxicity. Environ Toxicol Chem 23:1211–1218CrossRefGoogle Scholar
  55. Peixoto FP, Lopes ML, Madeira VMC, Vicente JAF (2009) Toxicity of MCPA on non-green potato tuber calli. Acta Physiol Plant 31:103–109CrossRefGoogle Scholar
  56. Pruyne D, Legesse-Miller A, Gao L, Dong Y, Bretscher A (2004) Mechanisms of polarized growth and organelle segregation in yeast. Annu Rev Cell Dev Biol 20:559–591CrossRefGoogle Scholar
  57. Robinson MD, Grigull J, Mohammad N, Hughes TR (2002) FunSpec: a web-based cluster interpreter for yeast. BMC Bioinforma 3:35–39CrossRefGoogle Scholar
  58. Ruepp A, Zollner A, Maier D, Albermann K, Hani J, Mokrejs M, Tetko I, Guldener U, Mannhaupt G, Munsterkotter M, Mewes HW (2004) The FunCat, a functional annotation scheme for systematic classification of proteins from whole genomes. Nucleic Acids Res 32:5539–5545CrossRefGoogle Scholar
  59. Saha S, Dutta D, Karmakar R, Ray DP (2012) Structure-toxicity relationship of chloroacetanilide herbicides: relative impact on soil microorganisms. Environ Toxicol Pharmacol 34:307–314CrossRefGoogle Scholar
  60. Santos SC, Sá-correia I (2015) Yeast toxicogenomics: lessons from a eukaryotic cell model and cell factory. Curr Opin Biotechnol 33:183–191CrossRefGoogle Scholar
  61. Saxena PN, Gupta SK, Murthy RC (2014) Comparative toxicity of carbaryl, carbofuran, cypermethrin and fenvalerate in Metaphire posthuma and Eisenia fetida—a possible mechanism. Ecotoxicol Environ Saf 100:218–225CrossRefGoogle Scholar
  62. Schmalfuβ J, Matthes B, Knuth K, Böger P (2000) Inhibition of Acyl-CoA elongation by chloroacetamide herbicides in microsomes from leek seedlings. Pest Biochem Physiol 67:25–35CrossRefGoogle Scholar
  63. Seeland A, Oehlmann J, Müller R (2012) Aquatic ecotoxicity of the fungicide pyrimethanil: effect profile under optimal and thermal stress conditions. Environ Pollut 168:161–169CrossRefGoogle Scholar
  64. Sousa JCG, Ribeiro AR, Barbosa MO, Pereira MFR, Silva AMT (2018) A review on environmental monitoring of water organic pollutants identified by EU guidelines. J Hazard Mat 344:146–162CrossRefGoogle Scholar
  65. Swain E, Baudry K, Stukey J, McDonough V, Germann M, Nickels JT (2002) Sterol-dependent regulation of sphingolipid metabolism in Saccharomyces cerevisiae. J Biol Chem 277:26177–26184CrossRefGoogle Scholar
  66. Steinberg CEW, Stürzenbaum SR, Menzel R (2008) Genes and environment—striking the fine balance between sophisticated biomonitoring and true functional environmental genomics. Sci Total Environ 400:142–161CrossRefGoogle Scholar
  67. Tadonléké RD, Leberre B, Perreau F, Humbert JF (2009) Responses of lake bacterioplankton activities and composition to the herbicide diuron. Aquat Toxicol 94:103–113CrossRefGoogle Scholar
  68. Teixeira MC, Duque P, Sá-Correia I (2007) Environmental genomics: mechanistic insights into toxicity of and resistance to the herbicide 2,4-D. Trends Biotechnol 25:363–370CrossRefGoogle Scholar
  69. USEPA—United States Environmental Protection Agency (2009) National primary drinking water regulations. Accessed 10 July 2017
  70. van der Rest ME, Kamminga AH, Nakano A, Anraku Y, Poolman B, Konings WN (1995) The plasma membrane of Saccharomyces cerevisiae: structure, function and biogenesis. Microbiol Rev 59:304–322Google Scholar
  71. van Ravenzwaay B, Pigott G, Leibold E (2004) Absorption, distribution, metabolism and excretion of 4-chloro-2-methylphenoxyacetic acid (MCPA) in rats. Food Chem Toxicol 42:115–125CrossRefGoogle Scholar
  72. Vega FA, Covelo EF, Andrade ML (2007) Accidental organochlorine pesticide contamination of soil in Porrino, Spain. J Environ Qual 36:272–279CrossRefGoogle Scholar
  73. Verdisson S, Couderchet M, Vernet G (2001) Effects of procymidone, fludioxonil and pyrimethanil on two non-target aquatic plants. Chemosphere 44:467–474CrossRefGoogle Scholar
  74. Viegas CA, Cabral MG, Teixeira MC, Neumann G, Heipieper HJ, Sá-Correia I (2005) Yeast adaptation to 2,4-dichlorophenoxyacetic acid involves increased membrane fatty acid saturation degree and decreased OLE1 transcription. Biochem Biophys Res Commun 330:271–278CrossRefGoogle Scholar
  75. Wu G, Bazer FW, Davis TA, Kim SW, Li P, Rhoads JM, Satterfield MC, Smith SB, Spencer TE, YinY (2009) Arginine metabolism and nutrition in growth, health and disease. Amino Acids 37:153–168CrossRefGoogle Scholar
  76. Yasokawa D, Iwahashi H (2010) Toxicogenomics using yeast DNA microarrays. J Biosci Bioeng 110:511–522CrossRefGoogle Scholar
  77. Yoo HS, Genbauffe FS, Cooper TG (1985) Identification of the ureidoglycolate hydrolase gene in the DAL gene cluster of Saccharomyces cerevisiae. Mol Cell Biol 5:2279–2288CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.iBB-Institute for Bioengineering and Biosciences, Instituto Superior Técnico (IST)Universidade de Lisboa (UL), Av Rovisco PaisLisboaPortugal
  2. 2.Instituto Gulbenkian de Ciência, Rua da Quinta Grande N°6OeirasPortugal
  3. 3.Department of Bioengineering, IST, UL, Av Rovisco PaisLisboaPortugal

Personalised recommendations