Mutagenic effects of a commercial glyphosate-based herbicide formulation on the soil filamentous fungus Aspergillus nidulans depending on the mode of exposure.

Glyphosate-based herbicides (GBH) are the most used pesticides worldwide. This widespread dissemination raises the question of non-target effects on a wide range of organisms, including soil micro-organisms. Despite a large body of scientific studies reporting the harmful effects of GBHs, the health and environmental safety of glyphosate and its commercial formulations remains controversial. In particular, contradictory results have been obtained on the possible genotoxicity of these herbicides depending on the organisms or biological systems tested, the modes and durations of exposure and the sensitivity of the detection technique used. We previously showed that the well-characterized soil filamentous fungus Aspergillus nidulans was highly affected by a commercial GBH formulation containing 450 g/L of glyphosate (R450), even when used at doses far below the agricultural application rate. In the present study, we analysed the possible mutagenicity of R450 in A. nidulans by screening for specific mutants after different modes of exposure to the herbicide. R450 was found to exert a mutagenic effect only after repeated exposure during growth on agar-medium, and depending on the metabolic status of the tested strain. The nature of some mutants and their ability to tolerate the herbicide better than did the wild-type strain suggested that their emergence may reflect an adaptive response of the fungus to offset the herbicide effects. The use of a non-selective molecular approach, the quantitative random amplified polymorphic DNA (RAPD-qPCR), showed that R450 could also exert a mutagenic effect after a one-shot overnight exposure during growth in liquid culture. However, this effect was subtle and no longer detectable when the fungus had previously been repeatedly exposed to the herbicide on a solid medium. This indicated an elevation of the sensitivity threshold of A. nidulans to the R450 mutagenicity, and thus confirmed the adaptive capacity of the fungus to the herbicide.

Transcriptome profiling of the fungus Aspergillus nidulans exposed to a commercial glyphosate-based herbicide under conditions of apparent herbicide tolerance.

Glyphosate-based herbicides, such as Roundup®, are the most widely used non-selective, broad-spectrum herbicides. The release of these compounds in large amounts into the environment is susceptible to affect soil quality and health, especially because of the non-target effects on a large range of organisms including soil microorganisms. The soil filamentous fungus Aspergillus nidulans, a well-characterized experimental model organism that can be used as a bio-indicator for agricultural soil health, has been previously shown to be highly affected by Roundup GT Plus (R450: 450 g/L of glyphosate) at concentrations far below recommended agricultural application rate, including at a dose that does not cause any macroscopic effect. In this study, we determined alterations in the transcriptome of A. nidulans when exposed to R450 at a dose corresponding to the no-observed-adverse-effect level (NOAEL) for macroscopic parameters. A total of 1816 distinct genes had their expression altered. The most affected biological functions were protein synthesis, amino acids and secondary metabolisms, stress response, as well as detoxification pathways through cytochromes P450, glutathione-S-transferases, and ABC transporters. These results partly explain the molecular mechanisms underlying alterations in growth parameters detected at higher concentrations for this ascomycete fungus. In conclusion, our results highlight molecular disturbances in a soil fungus under conditions of apparent tolerance to the herbicide, and thus confirm the need to question the principle of "substantial equivalence" when applied to plants made tolerant to herbicides.

Proteomic analysis of the soil filamentous fungus Aspergillus nidulans exposed to a Roundup formulation at a dose causing no macroscopic effect: a functional study.

Roundup® is a glyphosate-based herbicide (GBH) used worldwide both in agriculture and private gardens. Thus, it constitutes a substantial source of environmental contaminations, especially for water and soil, and may impact a number of non-target organisms essential for ecosystem balance. The soil filamentous fungus Aspergillus nidulans has been shown to be highly affected by a commercial formulation of Roundup® (R450), containing 450 g/L of glyphosate (GLY), at doses far below recommended agricultural application rate. In the present study, we used two-dimensional gel electrophoresis combined to mass spectrometry to analyze proteomic pattern changes in A. nidulans exposed to R450 at a dose corresponding to the no-observed-adverse-effect level (NOAEL) for macroscopic parameters (31.5 mg/L GLY among adjuvants). Comparative analysis revealed a total of 82 differentially expressed proteins between control and R450-treated samples, and 85% of them (70) were unambiguously identified. Their molecular functions were mainly assigned to cell detoxification and stress response (16%), protein synthesis (14%), amino acid metabolism (13%), glycolysis/gluconeogenesis/glycerol metabolism/pentose phosphate pathway (13%) and Krebs TCA cycle/acetyl-CoA synthesis/ATP metabolism (10%). These results bring new insights into the understanding of the toxicity induced by higher doses of this herbicide in the soil model organism A. nidulans. To our knowledge, this study represents the first evidence of protein expression modulation and, thus, possible metabolic disturbance, in response to an herbicide treatment at a dose that does not cause any visible effect. These data are likely to challenge the concept of "substantial equivalence" when applied to herbicide-tolerant plants.

Multiple effects of a commercial Roundup® formulation on the soil filamentous fungus Aspergillus nidulans at low doses: evidence of an unexpected impact on energetic metabolism.

Soil microorganisms are highly exposed to glyphosate-based herbicides (GBH), especially to Roundup® which is widely used worldwide. However, studies on the effects of GBH formulations on specific non-rhizosphere soil microbial species are scarce. We evaluated the toxicity of a commercial formulation of Roundup® (R450), containing 450 g/L of glyphosate (GLY), on the soil filamentous fungus Aspergillus nidulans, an experimental model microorganism. The median lethal dose (LD50) on solid media was between 90 and 112 mg/L GLY (among adjuvants, which are also included in the Roundup® formulation), which corresponds to a dilution percentage about 100 times lower than that used in agriculture. The LOAEL and NOAEL (lowest- and no-observed-adverse-effect levels) associated to morphology and growth were 33.75 and 31.5 mg/L GLY among adjuvants, respectively. The formulation R450 proved to be much more active than technical GLY. At the LD50 and lower concentrations, R450 impaired growth, cellular polarity, endocytosis, and mitochondria (average number, total volume and metabolism). In contrast with the depletion of mitochondrial activities reported in animal studies, R450 caused a stimulation of mitochondrial enzyme activities, thus revealing a different mode of action of Roundup® on energetic metabolism. These mitochondrial disruptions were also evident at a low dose corresponding to the NOAEL for macroscopic parameters, indicating that these mitochondrial biomarkers are more sensitive than those for growth and morphological ones. Altogether, our data indicate that GBH toxic effects on soil filamentous fungi, and thus potential impairment of soil ecosystems, may occur at doses far below recommended agricultural application rate.

Development and validation of a custom microarray for global transcriptome profiling of the fungus Aspergillus nidulans.

Transcriptome profiling is a powerful tool for identifying gene networks from whole genome expression analysis in many living species. Here is described the first extensively characterized platform using Agilent microarray technology for transcriptome analysis in the filamentous fungus Aspergillus (Emericella) nidulans. We developed and validated a reliable gene expression microarray in 8 × 15 K format, with predictive and experimental data establishing its specificity and sensitivity. Either one or two 60-mer oligonucleotide probes were selected for each of 10,550 nuclear as well as 20 mitochondrial coding sequences. More than 99 % of probes were predicted to hybridize with 100 % identity to their aimed specific A. nidulans target only. Probe sensitivity was supported by a highly narrow distribution of melting temperatures together with thermodynamic features, which strongly favored probe-target perfect match hybridization, in comparison with predicted secondary structures. Array quality was evaluated through transcriptome comparison of two A. nidulans strains, differing by the presence or not of Escherichia coli LacZ transgene. High signal-to-noise ratios were measured, and signal reproducibility was established at intra-probe and inter-probe levels. Reproducibility of microarray performances was assessed by high correlation between two-color dye signals and between technical replicates. Results were confirmed by RT-qPCR analysis on five genes. Though it covers 100 % of the A. nidulans targeted coding sequences, this low density array allows limited experimental costs and simplified data analysis process, making it suitable for studying gene expression in this model organism through large numbers of experimental conditions, in basic, biomedical or industrial microbiology research fields.

Purine utilization proteins in the Eurotiales: cellular compartmentalization, phylogenetic conservation and divergence.

The purine utilization pathway has been thoroughly characterized in Aspergillus nidulans. We establish here the subcellular distribution of seven key intracellular enzymes, xanthine dehydrogenase (HxA), urate oxidase (UaZ), 5-hydroxy-isourate hydrolase (UaX), 2-oxo-4-hydroxy-4-carboxy ureido imidazoline decarboxylase (UaW), allantoinase (AlX), allantoicase (AaX), ureidoglycolate lyase (UglA), and the fungal-specific α-ketoglutarate Fe(II)-dependent dioxygenase (XanA). HxA, AlX, AaX, UaW and XanA are cytosolic, while UaZ, UaX and UglA are peroxisomal. Peroxisomal localization was confirmed by using appropriate pex mutants. The pathway is largely, but not completely conserved in the Eurotiomycetes, noticeably in some species AaX is substituted by an alternative enzyme of probable bacterial origin. UaZ and the urate-xanthine UapA and UapC transporters, are also localized in specific cells of the conidiophore. We show that metabolic accumulation of uric acid occurring in uaZ null mutations is associated with an increased frequency of appearance of morphologically distinct colony sectors, diminished conidiospore production, UV resistance and an altered response to oxidation stress, which may provide a rationale for the conidiophore-specific localization. The pathway-specific transcription factor UaY is localized in both the cytoplasm and nuclei under non-inducing conditions, but it rapidly accumulates exclusively to the nuclei upon induction by uric acid.

The Aspergillus nidulans acuL gene encodes a mitochondrial carrier required for the utilization of carbon sources that are metabolized via the TCA cycle.

In Aspergillus nidulans, the utilization of acetate as sole carbon source requires several genes (acu). Most of them are also required for the utilization of fatty acids. This is the case for acuD and acuE, which encode the two glyoxylate cycle-specific enzymes, isocitrate lyase and malate synthase, respectively, but also for acuL that we have identified as AN7287, and characterized in this study. Deletion of acuL resulted in the same phenotype as the original acuL217 mutant. acuL encodes a 322-amino acid protein which displays all the structural features of a mitochondrial membrane carrier, and shares 60% identity with the Saccharomyces cerevisiae succinate/fumarate mitochondrial antiporter Sfc1p (also named Acr1p). Consistently, the AcuL protein was shown to localize in mitochondria, and partial cross-complementation was observed between the S. cerevisiae and A. nidulans homologues. Extensive phenotypic characterization suggested that the acuL gene is involved in the utilization of carbon sources that are catabolized via the TCA cycle, and therefore require gluconeogenesis. In addition, acuL proves to be co-regulated with acuD and acuE. Overall, our data suggest that AcuL could link the glyoxylate cycle to gluconeogenesis by exchanging cytoplasmic succinate for mitochondrial fumarate.

Mutations in the basic loop of the Zn binuclear cluster of the UaY transcriptional activator suppress mutations in the dimerisation domain.

UaY is the specific ZnII(2)Cys(6) transcriptional activator of the purine utilisation pathway in Aspergillus nidulans. Previous work has determined the consensus binding sequence by EMSA and foot-printing. We determine here that it binds as a dimer to its cognate CGG-N(6)-CCG sites. We identify the uaY109 mutation, which has been shown to affect differentially the expression of a number of UaY-regulated genes, as a F112I substitution in the DNA-binding motif dimerisation domain. We isolated back mutants, revertants carrying different residues at the same position (I112N and I112M) and intragenic suppressors mapping in the first loop of the Zn cluster (N75T and N75K). We have analysed the original mutant and its revertants by growth tests and by their effects on the mRNA steady states of five UaY-regulated genes. We have determined the effect of the different mutations on UaY dimerisation, on the apparent Kdiss of the UaY DNA-binding domain to appropriate DNA sequences and on the methylation interference pattern. We have attempted to rationalise these phenotypes by modelling the UaY DNA binding domain on the structure of the highly similar Ppr1p. However, modelling of the wild-type and mutant proteins provides only a partial explanation for the observed phenotypes. This suggests that the mutated residues may have other roles besides the obvious ones inferred from their position in the sequence and by the similarity of UaY and Ppr1p.

Completing the purine utilisation pathway of Aspergillus nidulans.

We have previously identified by classical genetics and biochemistry, all the genes of Aspergillus nidulans predicted to be involved in purine utilisation, together with cognate regulatory genes and one gene encoding a novel xanthine hydroxylation activity. In this article we complete the description of the purine utilisation pathway with the identification of the two genes (uaX and uaW) encoding the enzymes catalysing the conversion of the product of urate oxidation by urate oxidase, 5-hydroxyisourate, to optically active allantoin. The identification of these additional genes confirms the complete absence of clustering of the genes involved in purine utilisation in A. nidulans.

Gene silencing of transgenes inserted in the Aspergillus nidulans alcM and/or alcS loci.

While carrying out a systematic disruption of the genes of unknown function in the alc gene cluster from the filamentous fungus Aspergillus nidulans, we observed a strong diminution of the transcription of markers inserted in the alcS gene. This was found to be the case for the two markers tested, nadA (from A. nidulans) and pyrG (from A. fumigatus) involved in purine utilization and uracil/uridine biosynthetic pathway, respectively. The same phenomenon was also observed with insertion of the nadA gene in the alcM locus, another gene of the alc cluster. In the case of nadA, the level of expression was directly correlated to the ability of the corresponding strains to grow on adenine as a sole nitrogen source. The insertion of the pyrG marker into alcS complemented perfectly vegetative growth, but did not allow a proper sexual cycle. This suggests that the lowered pyrG expression is not sufficient to provide the intracellular concentration of pyrimidines required for the sexual cycle. Thus, due caution must be exercised when disrupting genes with pyrG, one of the most commonly employed markers, especially if the gene to be disrupted is involved or suspected to be involved in the sexual cycle.

Phenotypes of mutations in the 5'-UTR of a limiting transcription factor in Aspergillus nidulans can be accounted for by translational inhibition and leaky scanning.

The uaY gene encodes the transcriptional activator of purine catabolism genes in Aspergillus nidulans. uaY12 results in strongly defective growth on purines as nitrogen sources and in strongly diminished transcription of UaY-regulated genes. This mutation introduces an ATG codon 64 bp upstream of the uaY ATG, generating a 68-codon open reading frame (uORFA), overlapping with the uaY ORF. uaY12 revertants fall into three categories: i. The majority eliminate the aberrant ATG. The growth and transcriptional phenotypes of these revertants are identical to those of the wild type. i. Two revertants create a stop codon in frame with the uaY12 aberrant ATG, shortening the length of the uORFA, thus uORFA no longer overlaps the uaY ORF. The latter are partial suppressors of the uaY12 mutation, while chain termination suppressors, in turn, suppress this novel phenotype. iii. Two partial suppressors are unlinked to uaY. These two mutations result in a pleiotropic phenotype usually associated with ribosomal proteins. We hypothesize that uORFA strongly diminishes translation of the uaY ORF and that revertants negate this effect by a number of different mechanisms. The first-AUG rule and the phenomena of translational inhibition and leaky scanning provide a coherent explanation of the results presented in this article.

The nadA gene of Aspergillus nidulans, encoding adenine deaminase, is subject to a unique regulatory pattern.

The adenine deaminase of A. nidulans, encoded by nadA, can be considered both as a catabolic and a purine salvage enzyme. We show that its transcriptional regulation reflects this double metabolic role. As all other genes involved in purine utilisation it is induced by uric acid, and this induction is mediated by the UaY transcription factor. However, it is also independently and synergistically induced by adenosine by a UaY-independent mechanism. At variance with all other enzymes of purine catabolism it is not repressed but induced by ammonium. This is at least partly due to the ammonium responsive GATA factor, AreA, acting in the nadA promoter as a competitor rather than in synergy with UaY. The adB gene, encoding adenylo-succinate synthetase, which can be considered both a biosynthetic and a salvage pathway enzyme, shares with nadA both ammonium and adenosine induction.

Organization of the biosynthetic gene cluster for the macrolide antibiotic spiramycin in Streptomyces ambofaciens.

Spiramycin, a 16-membered macrolide antibiotic used in human medicine, is produced by Streptomyces ambofaciens; it comprises a polyketide lactone, platenolide, to which three deoxyhexose sugars are attached. In order to characterize the gene cluster governing the biosynthesis of spiramycin, several overlapping cosmids were isolated from an S. ambofaciens gene library, by hybridization with various probes (spiramycin resistance or biosynthetic genes, tylosin biosynthetic genes), and the sequences of their inserts were determined. Sequence analysis showed that the spiramycin biosynthetic gene cluster spanned a region of over 85 kb of contiguous DNA. In addition to the five previously described genes that encode the type I polyketide synthase involved in platenolide biosynthesis, 45 other genes have been identified. It was possible to propose a function for most of the inferred proteins in spiramycin biosynthesis, in its regulation, in resistance to the produced antibiotic or in the provision of extender units for the polyketide synthase. Two of these genes, predicted to be involved in deoxysugar biosynthesis, were inactivated by gene replacement, and the resulting mutants were unable to produce spiramycin, thus confirming their involvement in spiramycin biosynthesis. This work reveals the main features of spiramycin biosynthesis and constitutes a first step towards a detailed molecular analysis of the production of this medically important antibiotic.

Sub-families of alpha/beta barrel enzymes: a new adenine deaminase family.

No gene coding for an adenine deaminase has been described in eukaryotes. However, physiological and genetical evidence indicates that adenine deaminases are present in the ascomycetes. We have cloned and characterised the genes coding for the adenine deaminases of Aspergillus nidulans, Saccharomyces cerevisiae and Schizosaccharomyces pombe. The A.nidulans gene was expressed in Escherichia coli and the purified enzyme shows adenine but not adenosine deaminase activity. The open reading frames coded by the three genes are very similar and obviously related to the bacterial and eukaryotic adenosine deaminases rather than to the bacterial adenine deaminases. The latter are related to allantoinases, ureases and dihydroorotases. The fungal adenine deaminases and the homologous adenosine deaminases differ in a number of residues, some of these being clearly involved in substrate specificity. Other prokaryotic enzymes in the database, while clearly related to the above, do not fit into either sub-class, and may even have a different specificity. These results imply that adenine deaminases have appeared twice in the course of evolution, from different ancestral enzymes constructed both around the alpha/beta barrel scaffold.