Bioremediation of Dichlorodiphenyltrichloroethane (DDT)-Contaminated Agricultural Soils: Potential of Two Autochthonous Saprotrophic Fungal Strains

The spread and environmental accumulation of DDT over the years represent not only a threat to human health and ecological security but also a major challenge because of the complex chemical processes and technologies required for remediation. Saprotrophic fungi, isolated from contaminated sites, hold promise for their bioremediation potential toward toxic organic compounds, since they might provide an environment-friendly solution to contamination. Once we verified the high tolerance of autochthonous fungal strains to high concentrations of DDT, we showed how fungi from different phyla demonstrate a high metabolic versatility in the presence of DDT. The isolates showed the singular ability to keep their functionality, despite the DDT-induced production of reactive oxygen species.

ABSTRACT

DDT (dichlorodiphenyltrichloroethane) was used worldwide as an organochlorine insecticide to control agricultural pests and vectors of several insect-borne human diseases. It was banned in most industrialized countries; however, due to its persistence in the environment, DDT residues remain in environmental compartments, becoming long-term sources of exposure. To identify and select fungal species suitable for bioremediation of DDT-contaminated sites, soil samples were collected from DDT-contaminated agricultural soils in Poland, and 38 fungal taxa among 18 genera were isolated. Two of them, Trichoderma hamatum FBL 587 and Rhizopus arrhizus FBL 578, were tested for tolerance in the presence of 1-mg liter⁻¹ DDT concentration by using two indices based on fungal growth rate and biomass production (the tolerance indices Rt:Rc and TI), showing a clear tolerance to DDT. The two selected strains were studied to evaluate catabolic versatility on 95 carbon sources with or without DDT by using the Phenotype MicroArray system and to investigate the induced oxidative stress responses. The two strains were able to use most of the substrates provided, resulting in both high metabolic versatility and ecological functionality in the use of carbon sources, despite the presence of DDT. The activation of specific metabolic responses with species-dependent antioxidant enzymes to cope with the induced chemical stress has been hypothesized, since the presence of DDT promoted a higher formation of reactive oxygen species in fungal cells than the controls. The tested fungi represent attractive potential candidates for bioremediation of DDT-contaminated soil and are worthy of further investigations.

IMPORTANCE The spread and environmental accumulation of DDT over the years represent not only a threat to human health and ecological security but also a major challenge because of the complex chemical processes and technologies required for remediation. Saprotrophic fungi, isolated from contaminated sites, hold promise for their bioremediation potential toward toxic organic compounds, since they might provide an environment-friendly solution to contamination. Once we verified the high tolerance of autochthonous fungal strains to high concentrations of DDT, we showed how fungi from different phyla demonstrate a high metabolic versatility in the presence of DDT. The isolates showed the singular ability to keep their functionality, despite the DDT-induced production of reactive oxygen species.

INTRODUCTION

The use of dichlorodiphenyltrichloroethane (DDT) was banned in most industrialized countries in the early 1970s due to toxicological concerns (1–3); DDT had been used in agriculture and in public health as an insecticide for several decades. It is still used in some countries to control malaria vectors (2, 4). However, notwithstanding the general prohibition in use, its residues are still consistently found in soil, food, human, and animal samples (5–9). DDT is classified as a persistent organic pollutant (POP) because of its high stability in the environment, which makes it a lasting source of ecosystem contamination (10–12). To this specificity must be added that DDT is easily transportable over long distances by air and ocean currents, and it is biomagnifiable through the food chain (13–16). Crops, in fact, can accumulate compounds belonging to the group of chlorinated pesticides (17), which pose a high risk to human health since DDT and other POPs have proven to be cancerogenic, mutagenic, and endocrine disruptors (2, 18). Over the past decade, research on DDT degradation pathways in soils made significant progress, showing that the formation of further toxic metabolites during the aging and transformation of the same compound also poses environmental challenges and makes the remediation of DDT contaminated soils a difficult matter (19). Remediation approaches based on the addition of chelators, low-molecular-mass organic acids, cosolvent washing, and the use of sodium and seaweeds to promote and accelerate DDT degradation have been tested experimentally (20). However, the number and variety of toxic molecules that can populate a long-DDT-contaminated soil require a multifunctional strategy that can both buffer the adverse effects of DDT natural breakdown and decrease the overall concentration of the contaminant.

Restoration of polluted soils may provide an added value in maintaining ecoservices given by natural resources and, consequently, supporting human well-being (21, 22). Soil contamination leads to decreased biodiversity and activity of soil biota (23, 24), suppressing microorganisms and producing alteration of ecosystem functionality.

The use of microorganisms to biotransform toxic compounds, such as DDT, has the advantage of being environment-friendly and cost-effective compared to chemicophysical approaches, which require large investments and generate wastes during the process that are difficult to manage (25, 26). Even though the protracted use of DDT has caused over time a strong contamination of agricultural soils (27), the level of contamination in these soils is far lower than that of industrially polluted sites (28), which makes it economically feasible to apply only “nature-based solutions” that are potentially able to restore agro-ecosystem services.

Given that in terrestrial habitats the presence of any organic compound, even pollutants, represents sources of carbon and energy for specialized species of saprotrophs, the maintenance of a high fungal diversity can ensure the recycling of organic contaminants. Fungi, in fact, respond to environmental changes, evolving resistance to chemical stresses or developing metabolic adaptations to new conditions that confer a competitive advantage to some species over others (29). Such species acquire an important functional role after pollution stress and so should be considered for their potential in soil remediation.

Fungi are nature’s unique recyclers, and their ability to convert toxic organic compounds to harmless products has already been well documented (12, 29–31). Therefore, the use of fungal taxa for bioremediation purposes is a promising alternative to conventional approaches, particularly in applying indigenous fungi isolated from chemical-contaminated soils (32, 33).

The results of a survey carried out on 53 agricultural soils in Poland showed that more than 80% of the samples contained DDT or its isomers and metabolites in trace amounts (28). In this context, the isolation from some DDT-contaminated soils and biological characterization of fungal assemblages were carried out in order to select indigenous species potentially applicable for bioremediation. Among the isolated strains which were determined to be the most tolerant to high DDT concentration according to two tolerance indices (Rt:Rc and TI), two strains were selected on the basis of metabolic features for further investigations. We investigated the oxidative stress responses of the selected strains, considering the changes in reactive oxygen species (ROS) production and activities of the different active oxygen-scavenging enzymes in the presence of DDT. Finally, their phenotypic profiles and metabolic responses under this abiotic stress condition were investigated by using the Phenotype MicroArray (PM) system.

RESULTS

DDT content in soil.

The soil samples used for the isolation and selection of the fungal species contained on average 57 and 43 μg kg⁻¹ DDT in all, respectively, for the Myśliwska site and both Rabata sites. The sample from Myśliwka contained three isomers (p,p′-DDE, 24.7 μg kg⁻¹; p,p′-DDD, 8.3 μg kg⁻¹; and p,p′-DDT, 20.3 μg kg⁻¹), while only two isomers were detected in the soil samples from Rabata (p,p′-DDE, 29.2 μg kg⁻¹; and p,p′-DDD, 9.5 μg kg⁻¹).

Identification of fungal isolates.

The identification of the two fungal strains could be confirmed on the basis of internal transcribed spacer 1 (ITS1), the 5.8S gene, and the ITS2 regions of the ribosomal DNA (rDNA) (34), in conjunction with the gene encoding translation elongation factor 1α (tef1) (35). Moreover, tef1 intron 4, in combination with intron 5 regions, was successful in the molecular identification of the Trichoderma strain (36).

The pairwise comparison of fungal ITS and tef1 sequences with those available in the public online databases confirmed the identity at the species level of Trichoderma FBL 587 and Rhizopus FBL 578 as Trichoderma hamatum (Bonord.) Bainier (accession no. MK890773 and MK895140) and Rhizopus arrhizus A. Fisch. (accession no. MK890772 and MK895139), respectively. The best hits according to the alignments obtained using the BLAST search program (37) with the NCBI (38), UNITE (39), and TrichOKEY (40) databases are reported in Table 1.

TABLE 1.

Molecular identification of the two fungal strains

Isolate	Locus	Accession no.	Sequence length (bp)	Highest similarity (%)a match to:
				NCBI database	UNITE ITS database	TrichOKEY/Trichoblast database
Rhizopus, isolate 1	ITS1, 5.8S, ITS2 rDNA	MK890773	608	R. oryzae, 99.67% (MH865587.1)	R. arrhizus, 99% (MH715977.1, SH1145919.08FU)
Rhizopus, isolate 2	tef1	MK895139	645	R. oryzae 98.91% (KX196176.1)
Trichoderma, isolate 1	ITS1, 5.8S, ITS2 rDNA	MK890772	583	T. hamatum, 100% (KP009347.1)	T. hamatum, 100% (KP009347, SH1568608.08FU)	T. hamatum, 5 genus-specific hallmarks
Trichoderma, isolate 2	tef1, 4th intron, 5th intron, 6th exon	MK895140	1,239	T. hamatum, 98.31% (KJ665513.1)		T. hamatum, 4 of 12 known Tef anchors

^aBest hits were determined according to the alignments obtained using the BLAST search program (37) with the NCBI (38), UNITE (39), and TrichOKEY (40) databases. Where applicable, percent similarity values are given and accession numbers are indicated in parentheses.

Evaluation of fungal growth rate and tolerance.

Exposure to DDT did not significantly reduce the diametric growth of the two selected species.

Both tolerance indices used to evaluate the capacity/adaptation of the strains to grow in the presence of DDT showed a high range of values (Table 2). A value equal to 100% for the tolerance index Rt:Rc was observed in both fungal species and for the tolerance index (TI), a value of >100% occurred for both species (Table 2).

TABLE 2.

Evaluation of fungal tolerance (Rt:Rc and TI) to DDT for the two selected taxa^a

Tested taxon/Fungal Biodiversity Laboratory (FBL) accession no.b	Sampling site	Rt:Rc (%)	TI (%)	Mean pH ± SE		ΔpH
				Control	Treatment
Rhizopus arrhizus A. Fisch. FBL 578	Rabata 2	100	103	2.40 ± 0.1	2.28 ± 0.0	0.12
Trichoderma hamatum (Bonord.) Bainier FBL 587	Mysliwska	100	108	3.98 ± 0.3	3.87 ± 0.4	0.11

^aRt:Rc is defined as the ratio of the colony extension rates in the presence (Rt) or absence (Rc) of DDT. TI is defined as the ratio of the dry weight of the fungal biomass in the presence (Rt) or absence (Rc) of DDT. Surface pH values of the medium and the difference in comparison to cultures not amended with DDT after growth of the fungal species for 7 days at 25°C are also shown (n = 3).

^bR. arrhizus belongs to the phylum Mucoromycota; T. hamatum belongs to the phylum Ascomycota.

R. arrhizus showed a tendency to induce alkalinization of the medium after fungal growth, whereas T. hamatum tended to reduce the medium pH, although the ΔpH values were not statistically significant (Table 2).

DDT effect on fungal metabolic profile.

The two strains (T. hamatum and R. arrhizus) analyzed to evaluate the effect of DDT on fungal metabolic behavior in the presence of different carbon sources showed very diverse responses. Differences between the fungal species in the total number of substrates used and in the maximum absorbance value were observed (Fig. 1). The total number of substrates that R. arrhizus was able to metabolize was lower than those used by T. hamatum (41 and 67 substrates, respectively), underlying a narrower carbon source utilization spectrum for R. arrhizus. On the other hand, the maximum optical density (OD) for the substrates metabolized by R. arrhizus was generally greater than 1.5, underlying a higher metabolism (fungal growth and respiration) than that for T. hamatum. T. hamatum was able to effectively use 32 of 95 carbon substrates (34%), including glucose, mannose, xylose, trehalose, cellobiose, and xylitol, when grown without the addition of DDT. Poor growth was observed for 35 substrates. T. hamatum was unable to utilize 29 substrates (30%), including d-glucuronic acid, d-lactic acid methyl ester and l-alaninamide (Fig. 1). R. arrhizus was able to growth well on 22 substrates (23%), including N-acetyl-d-glucosamine, l-arabinose, d-arabitol, arbutin, d-cellobiose, d-fructose, and d-galactose. R. arrhizus poorly metabolized 19 substrates, such as d-mannitol, d-mannose, d-raffinose, d-ribose, stachyose, and γ-amino-n-butyric acid. It was unable to utilize 57 substrates (60%), e.g., m-erythritol and l-fucose (Fig. 1).

FIG 1.

Visualization of PM curves for OD₇₅₀ data using the function level plot in opm package for the comparison of the phenotypes of both tested species. Each curve is displayed as a thin horizontal line in which the curve height, as measured in color value units, is represented by the color intensity of red, yellow, and blue (red parts indicate higher curves). The x axes correspond to the measurement time in hours.

Seventeen substrates (18%) were used commonly by both strains, although R. arrhizus demonstrated in all of them a higher growth rate (Fig. 1). Among them, l-amino-n-butyric acid-arabinose, d-cellobiose, d-fructose, d-galactose, gentiobiose, and glycerol can be mentioned. On the other hand, neither fungus was able to metabolize 28 substrates (30%), such as l-lactic acid, d-malic acid, succinamic acid, l-phenylalanine, ethanolamine, putrescine, and uridine. Furthermore, some species-specific catabolic abilities were observed. For instance, T. hamatum was able to use l-fucose, α-d-lactose, d-melezitose, d-melibiose, and turanose, which were not used by R. arrhizus, whereas the latter strain metabolized l-proline, which was not used by T. hamatum (Fig. 1).

Discriminant analysis showed that the centroids of data at 144 h of incubation (Fig. 2) were significantly separated in both T. hamatum and R. arrhizus (although for R. arrhizus the incubation time with the best separation between control and treatment was 216 h of incubation), thus highlighting the overall differences between treated and untreated inocula (Fig. 2). This specific time point was considered for subsequent investigations on metabolic ratio and substrate categories.

FIG 2.

Discriminant analysis of substrates used by T. hamatum and R. arrhizus for all specific time intervals of incubation. The plots show the centroids as synthetic values for all of the observations at each time point (288 observations per centroid). Each DA was run using the first three components of the relevant PCA. The cumulative explained variance for the sum of the three components was 78.11 for R. arrhizus and 80.31 for T. hamatum.

Both T. hamatum and R. arrhizus after 144 h of incubation used polyols (category 7), as well as oligosaccharides (category 9) and glucosides (category 10) (Fig. 3). Hexoses (category 3) were better used by T. hamatum than by R. arrhizus, which in contrast metabolized pentoses more efficiently (category 4) (Fig. 3). Figure 3 shows also that the OD₄₉₀ to OD₇₅₀ coordinates of R. arrhizus for eight categories of substrates (3, 4, 6, 7, 8, 9, 10, and 12) were higher than those of the remaining categories, which clustered close to the origin of the axes in the plot. In T. hamatum, instead, nearly all the categories of substrates were to some extent used, since a significant respiration activity for all of them was observed, even if no corresponding biomass development was recorded. The gradient of the regression line, representing the general metabolic quotients, was higher in R. arrhizus than in T. hamatum (Fig. 3).

FIG 3.

Scatter plot of OD₇₅₀ and OD₄₉₀ values grouped into 16 substrate categories with the relevant regression lines, calculated for both T. hamatum and R. arrhizus.

The kinetic parameters describing the respiration and growth curves were statistically different in R. arrhizus grown on +DDT substrates compared to the control (−DDT) substrates, with the exception of the parameter λ (describing the lag phase) (P < 0.05) (Table 3). On the contrary, no difference between the average values of λ, μ (the slope), A (the maximum optical density reached), and AUC (the area under the curve) between control and treatment (+DDT) was evidenced for T. hamatum (Table 3).

TABLE 3.

Average values of curve kinetic parameters in T. hamatum and R. arrhizus inocula, with or without the addition of DDT^a

Inoculum or P value	Respiration				Biomass
	μ	λ	A	AUC	μ	λ	A	AUC
R. arrhizus
R. arrhizus control	0.014 A	3.32 A	0.99 A	174.96 A	0.01 A	77.36 A	0.62 A	98.22 A
R. arrhizus + DDT	0.012 B	−227.21 A	0.85 B	151.04 B	0.01 B	91.14 A	0.48 B	74.82 B
P value	0.04	0.48	0.01	0.03	0.01	0.90	0.00	0.01
T. hamatum
T. hamatum control	0.01 A	−25.84 A	0.90 A	160.12 A	0.01 A	36.43 A	0.45 A	68.28 A
T. hamatum + DDT	0.01 A	8.17 A	0.92 A	164.25 A	0.01 A	66.06 A	0.46 A	71.04 A
P value	0.36	0.39	0.68	0.59	0.40	0.44	0.60	0.57

^aThe descriptive curve parameters used to discriminate between the metabolisms of R. arrhizus and T. hamatum were lambda (λ), which represents the lag phase; mu (μ), which is the slope; A (maximum), representing the OD value at the end of incubation when the organism reached a plateau in growth and substrate use; and the area under the curve (AUC). Differences were evaluated for significance with post hoc Student-Newman-Keuls test. The values are means from n = 384 curves per inoculum (4 replicas, 96 substrates each). Average values of curve parameters for control and test inocula with the same letter (A or B) are not significantly different (P < 0.05).

However, it is worth noting that differences in the use of some substrates were observed for T. hamatum in the presence or absence of DDT after 144 h of incubation (Table 4). Conversely, no differences in single carbon use were significant for R. arrhizus (data not shown).

TABLE 4.

Average optical densities for biomass and respiration of T. hamatum grown with or without DDT^a

Substrate	Respiration			Biomass
	Avg OD		P	Avg OD		P
	Control	+DDT		Control	+DDT
A3 N-acetyl-d-galactosamine	0.56 B	0.60 A	0.03	0.07 A	0.08 A	0.29
A11 arbutin	1.70 B	1.77 A	0.03	0.84 A	0.86 A	0.27
B1 α-cyclodextrin	0.32 A	0.35 A	0.07	0.30 B	0.32 A	0.04
B4 i-erythritol	1.70 A	1.49 B	0.04	0.95 A	0.72 B	0.04
C1 glucose-1-phosphate	0.85 B	0.88 A	0.05	0.19 B	0.24 A	0.00
H6 l-serine	1.05 A	1.07 A	0.67	0.35 B	0.37 A	0.04
H9 putrescine	0.79 A	0.81 A	0.64	0.07 B	0.11 A	0.01

^aOnly the substrates with significant different values (post hoc Student-Newman-Keuls test) for either respiration or biomass or both are listed. Average values for control and +DDT with the same letter are not significantly different (P < 0.05). OD, optical density. n = 4 per inoculum.

Oxidative stress analysis.

The addition of DDT to the growth medium promoted, as expected, a significant increase in ROS production by the two tested strains: about 34% more in T. hamatum and 52% more in R. arrhizus than for the −DDT control (Table 5).

TABLE 5.

ROS production by the two tested strains (R. arrhizus and T. hamatum) grown with or without DDT in the substrate^a

Strain	Mean ROS production ± SE		Rate of increase (%)
	Control	Treatment (+DDT)
R. arrhizus	1,269.67 ± 22.10	1,917.67 ± 151.46*	34
T. hamatum	1,036.67 ± 15.06	2,181.33 ± 60.2*	52

^aROS production was assessed as the DCF fluorescence. Asterisks denote significant differences between treatment and control values (n = 3; *, P < 0.05).

The activities of the four ROS-scavenging enzymes tested were not all increased by the oxidative stress, with superoxide dismutase (SOD) and catalase (CAT) showing a strain-dependent behavior (Fig. 4). The SOD activity increased in T. hamatum, but not in R. arrhizus, compared to the control. On the other hand, the CAT activity increased only in R. arrhizus in comparison to the control. The glutathione S-transferase (GST) and peroxidase (POX) activities increased compared to the control in both species (Fig. 4).

FIG 4.

Antioxidant enzyme activities in R. arrhizus and T. hamatum grown in control (−DDT) and +DDT medium. Means ± standard errors are shown (n = 3). Asterisks denote significant differences between treatment and control samples (*, P < 0.05).

DISCUSSION

Soil fungal species richness.

Saprotrophic soil fungi from DDT-contaminated soils were isolated in order to select two potential candidates for bioremediation purposes, relying on the assumption that indigenous fungal strains naturally selected in polluted soils are often more able to survive and metabolize DDT than organisms introduced from elsewhere (32, 33, 41–43). The two investigated strains belong to the phyla Ascomycota and Mucoromycota, respectively. Most pollutant degraders belong to the phylum Ascomycota, which is reported as one of the most common groups in contaminated soils, being able to colonize most ecological niches (42, 44, 45). On the other hand, species belonging to the phylum Mucoromycota, being pioneer saprotrophic fungi characterized by rapid growth and short explorative phase, commonly exhibited high tolerance to environmental stresses (30, 46). The soil microbial activities affect the environmental fate of pesticides, since one of the routes for residue degradation is the metabolization of the active substance by microorganisms (47). Therefore, microbial functions such as enzymatic activity, possessed by fungal species isolated from historically contaminated sites, would restore soil health and fertility (48). However, while the fungal community is reduced by the presence of contaminants, the fungal species adapted to their presence should be more capable of removing them (30). As a possible consequence, in soils with relatively low levels of pollution, particular fungal taxa can become the dominant populations due to their adaptation in using pollutants as the C source (49). Furthermore, in recent studies the ability of some species to biodegrade pollutants has been linked to mycotoxin synthesis (50); in this way, it is important to choose species known as not producing mycotoxins to exploit the bioremediation potential and avoid the risk of mycotoxin contamination.

Evaluation of fungal growth rate and tolerance.

The acquired tolerance to DDT of the two fungal species T. hamatum and R. arrhizus was confirmed by the tolerance tests performed: the analyzed strains grew without significant inhibition in terms of growth rate and biomass production, resulting in high tolerance index values, thus highlighting a strong tolerance to the presence of the pesticide (Table 2).

DDT effect on fungal metabolic profile.

By the PM analysis, T. hamatum was able to metabolize and effectively grow on about 32 carbon sources, many of them carbohydrates such as glucose, fructose, mannose, galactose, xylose, trehalose, and cellobiose (Fig. 1). These substrates have been previously reported as being the most frequently used by Trichoderma species (51). Domsch et al. (52) reported different C sources for T. hamatum, which include starch, d-fructose, d-galactose, d-xylose, d-cellobiose, d-trehalose, glycerol, m-erythritol, d-mannitol and dextrin. All these compounds were confirmed by the phenotypic profile obtained from OD data of the investigated T. hamatum. Domsch et al. (52) also reported the degradation of diesel oil components for T. hamatum. However, the phenotypic profile of the investigated T. hamatum showed that it was able to grow efficiently using substrates that are generally not used by this genus, such as lactic acid and ethanolamine (51). The wide spectrum of carbon source utilization by Trichoderma species, such as T. harzianum, indicates a capacity to adapt to a wide range of habitats (51, 53–55). Nevertheless, a species-specific utilization of some substrates has been observed for some Trichoderma species using the same PM method. For example, T. harzianum readily assimilated N-acetyl-β-d-mannosamine and l-phenylalanine; T. asperellum grew on d-melezitose, raffinose, maltitol, dextrine, melibiose, and d-glucose; T. viridescens metabolized N-acetyl-l-glutamic acid; and Hypocrea jecorina and T. koningii grew on gentiobiose and salicine (51, 53–55). In this research, T. hamatum was able to grow on substrates not generally used by T. harzianum but used by T. asperellum, Hypocrea jecorina, and T. koningii, such as melezitose, gentiobiose, and salicine. On the other hand, T. hamatum could not grow on substrates that have been reported as specific for T. harzianum, such as N-acetylmannosamine. T. hamatum was able to use sucrose, d-melibiose, d-melezitose, d-mannitol, and d-raffinose, which have been reported as the best preferred substrates for T. asperellum, underlining a possible genetic and functional closeness (51). In fact, T. hamatum is included in the same section (Trichoderma) as the cosmopolitan species T. atroviride and T. asperellum and possesses a versatile phenotype, as do other Trichoderma species, such as T. parareesei and T. asperellum (51, 56). Such traits may be related to taxonomical distance between species belonging to the same genus, as observed by Kordowska-Wiater et al. (57), resulting in significant differences in their substrate utilization profiles. Previous studies have also revealed an overlap in substrate utilization by different Trichoderma species, which resulted in grouping of phylogenetically unrelated species. Even though this feature is limiting to the PM system power for taxonomic identification (58), the study of fungal metabolic profiles can provide ecological insights and be a valuable integrative tool for characterization of individual strains, species, and ecological groups (51, 58, 59).

R. arrhizus was able to grow on several substrates, underlying also for this species a broad ecological niche width. Phenotypic data on the genus Rhizopus using the PM system are scant, since only with Rhizopus oryzae and Rhizopus microsporus strains PM has been previously used to support protoplasm fusion for enhancing fumaric acid production (data available only for hybrids) (57).

The growth rate of R. arrhizus was higher than that of T. hamatum as pointed out by the higher maximum OD at 750 nm (OD₇₅₀), despite the fact that it was able to use a lower number of substrates (Fig. 1). This may be related to specialization and the efficient use of substrates, which may result in a more efficient use of the resource exhibiting a habitat specialization with a potential competitive advantage for the species. Among the substrates inducing strong growth are some amino acids (l-proline and l-ornithine) and several carbohydrates (monosaccharides such as d-xylose and l-arabinose; disaccharides such as d-trehalose, sucrose, β-gentiobiose, and cellobiose; trisaccharides such as maltotriose; and sugar alcohols such as xylitol and arabitol). Interestingly, most of these sugars are derived from glucose (sucrose, cellobiose, maltotriose, sorbitol, d-trehalose, β-gentiobiose, β-methyl-d-glucosamide, d-salicin, and N-acetyl-d-glucosamide), though glucose itself was not among the substrates inducing strong growth. These evidences are in line with those reported for the metabolism of R. oryzae (currently R. arrhizus). In fact, R. oryzae can use starch and cellulose, and d-fructose, d-glucose, d-galactose, and d-mannose are typical components of mycelium (52). The organophosphorus insecticide dyfonate, the phenylamide herbicides, and the fungicide carboxin can be metabolized by R. oryzae (52). With a particularly versatile enzyme machinery, R. oryzae is generally considered an effective saprotrophic fungus with specific metabolic abilities to use components of plant and fungal cells, such as starch, hexose sugars, and chitin, and an astonishing importance for potential industrial applications (60–62).

The analysis of the OD curves showed that 17 substrates were used by both species, which may indicate an overlap in niches of the decomposition process. Conversely, both fungi showed species-specific substrate metabolism: l-proline was only used by R. arrhizus, whereas l-fucose, α-d-lactose, d-melezitose, d-melibiose, and turanose were only used by T. hamatum. Proline metabolism in yeasts, plants, and bacteria is related to stress responses (63), and its presence in R. arrhizus may suggest an adaptation to habitats with extreme environmental conditions. In fact, R. oryzae is a cosmopolitan saprotrophic fungus that can grow in a wide variety of habitats, including alkaline soils, salt marshes, sewage sludge, and a uranium mine, in a wide range of temperatures (15 to 45°C) and pHs (6.3 to 7.2 or higher) (52). The metabolism of l-fucose, α-d-lactose, d-melezitose, d-melibiose, and turanose by T. hamatum points to a capacity of Trichoderma species to occupy different ecological niches, not only as decomposers but also as mycoparasites, plant symbionts, and endophytes, as previously reported by several studies (58, 59, 64). For example, melezitose, which can come from different sources, e.g., the sap of conifers and the honeydew produced by aphids, can be metabolized by wilt fungi and entomopathogenic fungi and hydrolyzed by some yeasts (65, 66).

The differences in carbon sources used by each fungus were also evident in the presence of DDT on the substrate, as emerged from the discriminant analysis (DA) (Fig. 2). Interestingly, the two fungi used the carbon sources with different dynamics and reacted to the presence of DDT at different time intervals of incubation (Fig. 2).

The points of scatter plot for the 16 substrate categories at the selected incubation time (144 h), identified as coordinates of the OD₄₉₀ and OD₇₅₀ values (Fig. 3), are an expression of the efficiency of the fungus in the use of carbon sources, as well as its general fitness (67). The metabolic quotient (i.e., the respiration/biomass ratio) confirmed the evidences of higher OD maxima for the singly used substrates, as observed for R. arrhizus in comparison to T. hamatum. The differences in the metabolism of substrate categories are related to the differences observed in the curves of the single substrates for R. arrhizus and T. hamatum (Fig. 1 and 3). Both species used important substrate categories, such as hexoses (e.g., glucose, fructose, galactose, or mannose) and oligosaccharides (e.g., maltose or sucrose), which definitely are the main substrate groups used by fungi (68). However, a general higher ratio may also indicate that a relatively small fungal biomass is developed without consuming much substrate. Conversely, low OD values at 750 nm (biomass) that yield very high OD values at 490 nm (respiration rates) indicate a stressing metabolic situation (67). Color formation in the water well was observed in all the replicated inoculums of T. hamatum, while with R. arrhizus the water well always remained colorless. Water was not used as a blank but as a variable based on the recommendation of Vaas et al. (69). The fungal behavior in the water well (category 1, in Fig. 3) which contains the redox dye but no carbon sources can be very species specific and depends on the germination peculiarities of the inoculated fungal spores. Conidial germination in most filamentous fungi requires the presence of low-molecular-mass nutrients (70), but in others the presence of water alone is sufficient to induce conidial respiration and isotropic growth. During this swelling process due to water uptake, some metabolic activities, such as respiration and DNA and RNA synthesis with breakdown of stored lipids and sugars, can occur (71).

Consistent differences in the behavior of the two fungi, both tolerant to DDT, were found based on comparison between the overall metabolic activities in the presence of DDT in the in vitro inocula. The kinetic parameters obtained from all the curves of respiration and growth in the presence or absence of DDT showed differences in R. arrhizus, whereas they were similar for the majority of substrates in T. hamatum (Table 3). In fact, in the presence of DDT the values of curve parameters of R. arrhizus were lower than those of the control, indicating a possible stressful condition due to the xenobiotic presence. Despite the high tolerance indices from growth data, these significant differences take account of the average values of curve parameters of PM substrate and may represent a wider picture of the phenotypic response of R. arrhizus to DDT.

For many substrates the differences between control and treatment, although statistically significant, remain very light. The strong tolerance of these strains to the xenobiotic compound may suggest physiological responses to keep internal homeostasis conditions close to those in control. Moreover, the positive effect of DDT on growth with the other substrates can be related to their role in fungal metabolism and functions. In the case of T. hamatum, the seven substrates inducing a different metabolism in the presence of DDT deserve some additional comments. With respect to the control, DDT induced a higher biomass production for six substrates (α-cyclodextrin, glucose 1-phosphate, l-serine, N-acetyl-d-galactosamine, arbutin, and putrescine) and reduced it on i-erythritol. The lower average OD value of T. hamatum grown on i-erythritol plus DDT highlights a quite strong stress from DDT on the fungus metabolism. Erythritol, an osmolyte in yeasts, belongs to the pentose phosphate pathways and is produced by dephosphorylation and reduction of d-erythrose-4-phosphate catalyzed by erythrose reductase through a NADPH-dependent and reversible reaction (72). This enzyme, found in T. reesei, might be present in other Trichoderma species, which were observed to use this substrate (51, 72). It is thus possible to hypothesize that DDT may have a negative impact on the erythritol catabolism in T. hamatum, reducing its metabolic performance. Cyclodextrins can be metabolized by Trichoderma species, which prefer the α type, the one present in the PM system, even though different Trichoderma species showed considerable deviations in their capacity to decompose cyclodextrins (73). Glucose 1-phosphate is the activated form of glucose, which is among the most-used substrates by T. hamatum. Serine is the key amino acid in the active site of serine proteases, enzymes catalyzing the hydrolysis of peptide bonds (74). These enzymes are important in the proteolysis of fungal cell wall, which is a mechanism for mycoparasitism, a feature of several Trichoderma species exploited in the biocontrol of soilborne plant-pathogenic fungi (64). Possibly also involved in mechanisms of plant protection, putrescine, a polyamine, is known to support the growth of fungi, and changes in its metabolism precede a wide variety of morphogenetic events (75). N-Acetyl-d-galactosamine is a component of the fungal wall and, along with galactopyranose and galactosamine residues, constitutes the GAG heteropolysaccharide with an important role in conidial germination (76). Its significant use under test conditions by fungus may be a response to stressful conditions due to DDT presence; this may be the case for the phenolic glycoside arbutin. In fact, an increased catabolism of arbutin may be related to an elevated carbohydrate uptake and metabolism, as has been observed in the white rot fungus Cerrena unicolor (77).

Oxidative stress analysis.

The DDT supplemented to the growth medium induced an increase in ROS production in both tested strains (Table 5), which was paralleled by an increased activity of some scavenging enzymes (Fig. 4). Fungi, like other organisms, adapt to an increase in ROS production by upregulating the activity of their antioxidative machinery (78, 79). Although it is known that filamentous fungi have supplementary mechanisms to handle ROS, and studies have examined some fungal genera, such as Aspergillus, Penicillium, Paecilomyces, and others (80–83), there are still limited data available on such mechanisms due to the difficulty in generating defective mutants (84). There is, however, a widely held view that ROS in fungi serve as sources of oxidative stress, defense purposes, and signaling to induce cell differentiation (85, 86).

The protection of mycelia from oxidative stress by the increased activity of the tested scavenging enzymes could be functional to provide detoxification of ROS under DDT exposure. Interestingly, the two analyzed species behaved differently with regard to the expression of SOD and CAT activities, while showing a similar increasing pattern for POX and GST (Fig. 4). This enzymatic activity trend may be due to the different types of produced and accumulated ROS that are dealt with by these enzymes (87). However, a possible explanation of the different patterns of enzyme activity in the two species could be also linked to the functions of ROS as signal transduction molecules, the degradation of lignocellulose, and the Fenton-type mechanisms, enzymatic processes linked to POP degradation (86, 88). However, data are scant about the specific effect of DDT on oxidative stress in fungi.

In T. hamatum, DDT enhanced the SOD activities with significant changes compared to the control but not did not alter the CAT activities. The antioxidant enzyme SOD plays a crucial role in its ability to dismutate superoxide anion to H₂O₂ and O₂. Hydrogen peroxide is also degraded by other enzymes such as CAT (84). Other reports proposed that SOD activity, although is much more prompted by superoxide, could be increased by H₂O₂ (80, 89). However, an increase in superoxide does not necessarily result in the induction of catalase. In fact, H₂O₂ is the spontaneous and enzymatic dismutation product of O₂⁻; in addition, the presence of H₂O₂ along with O₂⁻ may result in the generation of toxic hydroxyl radicals via Fenton chemistry (90). In a recent study, a species of the genus Trichoderma was used as a model to investigate the influence of oxidative stress induced by an organophosphate pesticide on the mitochondrial respiratory chain (91). It was found that, despite the increased lipid peroxidation induced by the bromopropylate exposure, the analyzed fungal strain adapted to the growth conditions.

By comparing the results of treatment to the control, the divergent response in CAT and SOD activity in R. arrhizus was seen to be consistent with those concerning the effect on these enzymes by different pesticides in other fungal species (82). In Russo et al. (92), ROS production and antioxidant enzyme activities were investigated for two fungal species in the presence of HCH isomers. These researchers observed an increase in ROS production and CAT and GST activities in both of the tested fungi: in Penicillium simplicissimum the SOD activity significantly increased under HCH stress conditions compared to the control, whereas in T. harzianum the SOD activity did not show the same trend (92). As reported by Kreiner et al. (82), exposure to menadione (MD) in Aspergillus niger also first resulted in an increase in CAT activity, whereas the SOD activity did not increase. This result was explained by suggesting that the SOD expression at the start of MD stress was sufficiently high to dismutate the produced superoxide radicals, resulting in an immediate increase in CAT activity (82).

On the other hand, both GST and POX showed significantly increased activity, suggesting a possible common role to cope with the oxidative stress. Glutathione transferases are a family of multifunctional enzymes widely known to have important functions in preventing oxidative stresses and maintaining the redox balance in fungi, although the structural, functional, and evolutionary relationships between them are still unclear (93). POX also belongs to a large family of enzymes that play a role in various biological processes, including the transformation of a variety of xenobiotics according to a free radical mechanism, thereby fostering biodegradation processes (94). Studies on adaptive responses to oxidative stress in A. niger have shown that the decline in the activity of antioxidant enzymes corresponds to a shift from oxidative defense back to fungal growth (89). In this sense, the decrease in enzymatic activities represents a fungal strategy in order to prevent energy waste (89). Considering our results regarding the high tolerance values and the different metabolic profiles of the tested fungal species, it is possible to hypothesize that they are able to provide different mechanisms for adaptation in the presence of DDT.

Conclusions.

Fungal species living in contaminated soils often acquire physiological and morphological adaptation strategies to overcome the different forms of stress posed by xenobiotics. Our research highlighted the wide tolerance of a quite diverse autochthonous soil fungal community, including several species with different ecological features, to prolonged DDT contamination. Several of these species merit further investigation for their potential in bioremediation. The two fungal species selected as models of tolerance in this study (T. hamatum and R. arrhizus) have shown high metabolic adaptability and the maintenance of a wide trophic niche, even in the presence of elevated doses of a toxic compound. The observed changes in the fungal redox equilibrium provoked by DDT could be attributed to processes of biodegradation or the stress induced by DDT.

In conclusion, it is possible that the two tested strains, T. hamatum and R. arrhizus, are able to provide different mechanisms for adaptation in the presence of DDT and thus can be considered for potential applications in bioremediation or rhizobioremediation.

MATERIALS AND METHODS

Soil sampling methodology and analytical determination of DDT content.

Contaminated soils with DDT or its metabolites were collected from two fields worked according to organic farming methods (named Myśliwska [GPS 51°56′10.9″N, 20°10′48.8″E] and Rabata [GPS 51°56′36.8″N, 20°10′20.2″E]) located in the vicinity of Skierniewice, Voivodship Lodzki, Poland. The fields were known to contain DDT as a result of a previous survey conducted at several locations in Poland (28), identified as Skierniewice 1 and Skierniewice 3 in that reference. Both fields were characterized by a sandy loamy soil with a content of organic matter of about 3%.

In spring 2016, from a bare soil, one sample was collected from Myśliwska, and two were gathered from Rabata for the work reported by the present paper, due to the size of the fields: about 0.5 ha for the former and 1.5 ha for the latter location. Soil samples were collected according to the official methodology adopted in Poland for sampling of materials for testing for residues of plant protection products (95). The method foresees a typical soil sampling methodology, gathering randomly about 20 subsamples of about 200 g each, which are mixed to form the laboratory sample of about 1 kg. The samples were sent immediately to the laboratory for isolation purposes under refrigerated conditions.

Residue determinations of DDT and its isomers and metabolites (p,p′-DDT, o,p′-DDT, p,p′-DDD, o,p′-DDD, o,p′-DDE, and p,p′-DDE) in soil were performed using a gas chromatograph (Agilent Technologies, 6890N), equipped with a Zebron ZB-MultiResidue-1 chromatographic column, and a mass detector (5975B Inert XL MSD). Extraction of the compounds was carried out according to the QuEChERS method (EN 15662:2008). In brief, after sample comminution and homogenization in the presence of dry ice, 10-g portions of soil aliquots were taken for extraction. After the addition of an extraction solution (10 ml of water, 10 ml of acetonitrile, 4 g of magnesium sulfate, 1 g of sodium chloride, 1 g of trisodium citrate dihydrate, and 0.5 g of disodium hydrogen citrate sesquihydrate), the sample was intensively shaken and centrifuged for phase separation. An aliquot of the organic phase was cleaned up by dispersive solid-phase extraction with 25 mg of the sorbent PSA (Supelco; catalog no. 52738-U) and 150 mg of magnesium sulfate for removal of residual water. Triphenyl phosphate solution was used as an internal standard. All data were determined using certified analytical standards, taking into account specific matrix effects, and corrected by an internal standard. Data were adjusted to soil dry mass (soil drying was obtained by heating for 24 h at 80°C).

Isolation of potential DDT-degrading fungal strains.

The isolation of fungal strains was performed using a dilution plate method (soil/water ratio of 1:1,000) as described by Persiani et al. (96). From each suspension, 0.5-ml portions were plated in five petri dishes (0.1 ml/dish) (96). The culture medium used was soil extract agar prepared using soil from the sampling area (97). After the incubation period (7 days at 25°C) on soil extract agar, the isolates were transferred as pure cultures to be identified by conventional taxonomic keys based on macro- and microscopic characteristics (data not shown) (52, 58, 98). They were stored at 4°C as pure cultures on malt extract agar (MEA) in the collection of the Fungal Biodiversity Laboratory (FBL; Sapienza, University of Rome, Italy).

Screening for evaluation of fungal tolerance to DDT.

Tolerance screenings to DDT were carried out in petri plates containing solid culture medium (MEA; malt extract, 20 g liter⁻¹; peptone, 1 g liter⁻¹; glucose, 20 g liter⁻¹; agar, 20 g liter⁻¹; distilled water). A solution of DDT isomers (4 mg of o,p′-DDT and 16 mg of p,p′-DDT) in acetone was prepared and supplemented into the medium at a final concentration of 1 mg liter⁻¹. The concentration of acetone in the final test medium was below the threshold of 0.1 ml liter⁻¹, as suggested according to the Organization for Economic Cooperation and Development (99).

The DDT concentration was set in order to expose fungal species to strong stress conditions. Assays without the addition of pesticide were used as a control. Prior to inoculation, 84-mm-diameter sterile cellophane membranes were placed in sterile conditions on the surface of the agar in each petri dish in order to separate the fungi from the medium, allow the passage of nutrients and metabolites between the medium and the colony, and facilitate the recovery of the mycelium (100). After 7 days of growth, mycelium inoculation was carried out using stock cultures of each fungal strain with a 5-mm-diameter cork borer. After incubation for 7 days at 25°C in the dark, the fungal colonies were removed from the medium, and mycelia were oven dried at 100°C until reaching a constant weight for at least 2 days. All the assays were carried out in triplicate.

Fungal tolerance to DDTs was evaluated by using two indices: (i) the tolerance index (Rt:Rc), defined as ratio of the colony extension rates in the presence (Rt) or absence (Rc) of DDTs, and (ii) the tolerance index (TI), based on the dry weights (DW) of fungal biomass (101, 102) as follows: TI (%) = (DW of treated mycelium/DW of control mycelium) × 100.

Furthermore, the surface pH of the culture medium was measured across the diameter of the petri dish using a conical tip FC 202D pH electrode (Hanna Instruments, Woonsocket, RI) and a portable pH meter (HI 99161; Hanna Instruments). The pH measurements were used to show the pH profile for treatments; the average surface pH values were used to calculate the differences (ΔpH) between control and test samples.

Focus on the fungal strains of interest.

Among the most tolerant strains tested (data not shown), two fungal strains (Trichoderma hamatum [Bonord.] Bainier FBL 587 and Rhizopus arrhizus A. Fisch. FBL 578) were identified by both conventional taxonomic keys based on macro- and microscopic characteristics and molecular analyses. Their phenotype profiles and oxidative stress responses in the presence of DDT were also characterized. Moreover, the two strains were also selected on the basis of growth and metabolic parameters such as sporulation capacity, diametric growth, biomass production, and absence of mycotoxin production according to data reported in the literature (52).

Genetic analyses.

Fungal DNA was extracted from fresh mycelium by using a DNeasy PowerSoil kit. The purity and quantity of DNA were checked by using agarose gel electrophoresis and a NanoDrop 8000 spectrophotometer (Thermo Scientific, Waltham, MA). The DNA concentration was adjusted to 10 ng μl⁻¹.

All PCRs were performed in a Veriti 96-well thermal cycler (Applied Biosystems) using GoTaq DNA polymerase (Promega). The reactions were carried out in a 25-μl volume containing 1× reaction buffer, 1.5 mM MgCl₂, 0.2 mM concentrations of each deoxynucleoside triphosphate, a 1.0 μM concentration of upstream primer, a 1.0 μM concentration of downstream primer, 1.25 U of GoTaq DNA polymerase, and 1 μg of bovine serum albumin, which is <0.3 μg 25 μl⁻¹ of the DNA template.

The internal transcribed spacer (ITS) region of the rRNA (∼600 bp) was amplified using the primer pair ITS1 and ITS4 (34); in the case of Rhizopus, the translation elongation factor 1α gene (tef1; ∼600 bp) was amplified using the primer pair EF1-1018F and EF1-1620R (35), while tef1 from Trichoderma (∼1,300 bp) was amplified using the primer pair EF1-728F and TEF1LLErev (36), so that the fragment includes the fourth and fifth introns and a significant portion of the last large exon.

Thermocycling programs.

(i) ITS. The thermocycling program for ITS was as follows: 5 min of denaturation at 94°C, followed by 30 cycles of 1 min of denaturation at 94°C, 1 min of annealing at 60°C, and a 1-min extension at 72°C. Ten minutes at 72°C was used as the final extension step.

(ii) tef1. The thermocycling program for tef1 (touchdown cycle) was as follows: 2 min of denaturation at 94°C and one amplification cycle with annealing at 66°C, which was then incrementally reduced by 1°C per cycle over the next nine cycles. An additional 36 amplification cycles were then performed, each consisting of 30 s of denaturation at 94°C, a 30-s annealing step at 56°C, and a 1-min extension at 72°C, concluding with a 10-min incubation at 72°C as a final extension step, followed by a quick cooling to room temperature.

PCR products were checked by agarose gel electrophoresis, purified with a QIAquick PCR purification kit (Qiagen), and sent out for Sanger sequencing (NHM, Molecular Laboratory, London, UK).

The forward and reverse electropherograms obtained for each fungal isolate were verified visually and aligned using CLUSTALW (v2.0) to obtain consensus sequences that were then compared using the BLAST search program (37) with the NCBI database (38). The ITS sequences were also compared using the UNITE database (39, 103) and the TrichOKEY v1.0 barcode sequence identification program for Trichoderma species with the web interface (www.isth.info) (40). The Trichoderma tef1 sequence was used in TrichoBLAST v1.0, and the tef1 fourth, fifth, and sixth exons were searched with BLAST within the Multilocus Database of Phylogenetic Markers using this component of the TrichOKEY database (http://isth.info/).

Analysis of fungal metabolic profile in the presence of DDT.

The phenotypes of the T. hamatum and R. arrhizus strains and the effects of DDT on their carbon metabolism were determined using the PM system (104). The inoculation was carried out in FF MicroPlates (Biolog, Inc., Hayward, CA) with suspensions of fungal conidia (for T. hamatum) or sporangiospores (Rhizopus arrhizus) according to a previously described method (104, 105). In triplicate experiments, inocula of the two fungal strains amended with 1 mg liter⁻¹ DDT in acetone were compared to controls, carried out on the same medium but without the addition of DDT.

The fungal species tested were initially grown on 2% MEA plates in the dark at 25°C for 7 days to obtain fresh conidia or spores for inoculum preparation. The operating conditions were as reported by Ceci et al. (100).

Optical measures were recorded in order to evaluate the overall differences in the carbon metabolism of the two fungal species and to assess the mechanisms of cometabolism or inhibition by studying the utilization of different substrates in the absence or presence of the DDT. The OD₄₉₀ and OD₇₅₀ of Biolog wells were measured to assess the respiratory activity and fungal biomass of mycelial growth, respectively, using a microplate reader (Molecular Devices Vmax) (100, 102, 105). Fungal growth (biomass) was measured as the turbidity increase in the wells (OD₇₅₀). The fungal respiratory activity was measured by the intensity of the purple color (OD₄₉₀) resulting from the reduction of the tetrazolium redox dye (p-iodonitrotetrazolium), present in the wells of the Biolog FF plates, through the action of fungal succinate dehydrogenase as a proxy for respiratory activity. Immediately after inoculation, both ODs were measured to zero the spectrophotometer, specifically for each Biolog plate. The plates were then read at various time intervals (0, 24, 48, 72, 96, 144, 168, 192, 216, and 240 h) (100, 102, 105).

Oxidative stress analysis.

The species R. arrhizus and T. hamatum were monitored with respect to the changes in the ROS production and activities of the following active oxygen-scavenging enzymes in the presence of DDT: SOD, CAT, GST, and POX. The selected strains were grown for 7 days on MEA at 25°C. A sporangiospore/conidium suspension at a concentration of 2 × 10⁸ spores ml⁻¹ was prepared by using a Thoma counting chamber. Spores/conidia were removed from culture dishes with the aid of a sterile loop by adding sterile distilled water amended with Tween 20. The suspension was filtered through some sterile gauze several times to remove residues of hyphae or conidiophores, and 5 ml (containing 2 × 10⁸ spores/conidia ml⁻¹) of the final spore/conidium suspension was inoculated in 75 ml Czapek-Dox medium (sucrose, 30 g liter⁻¹; NaNO₃, 3 g liter⁻¹; KH₂PO₄, 1 g liter⁻¹; MgSO₄⋅7H₂O, 0.5 g liter⁻¹; KCl, 0.5 g liter⁻¹; FeSO₄⋅H₂O, 0.01 g liter⁻¹; distilled water, 1 liter), adjusted to pH 5.6 with 1 M HCl, before autoclaving. The mycelium was harvested after 24 h by centrifugation and resuspended in fresh Czapek-Dox medium with or without the addition of the oxidative stress agent (DDT isomeric mixture at a final concentration of 1 mg liter⁻¹). According to the methods of Angelova et al. (80), the mycelium was harvested by filtration, washed in sterile-distilled H₂O and then in cold 50 mM potassium buffer (pH 7.8), and resuspended in the same buffer. The cell suspension was disrupted using a Tissue Lyser LT (Qiagen) and, after centrifugation (12,000 × g for 20 min at 4°C), the supernatant was collected to determine ROS production and enzymatic activities.

ROS production was determined using the oxidative stress cell permeant 2′,7′-dichlorodihydrofluorescein diacetate dye (H₂DCFDA; Sigma-Aldrich), which is oxidized to a fluorescencing dye, 2,7-dichlorofluorescein (DCF), with excitation/emission wavelengths of 350 nm/600 nm (106, 107).

CAT, SOD, GST, and POX activities were determined using commercial assay kits (Sigma-Aldrich) according to the manufacturer’s protocols. CAT activity was determined by measuring the decrease of H₂O₂ at 240 nm (108) with a spectrophotometer (Hach Lange DR5000) at different times (30, 60, 90, and 120 s). One unit of CAT was defined as the amount of enzyme that caused a reduction in absorbance at 240 nm of 0.01 per min. SOD, GST, and POX activities were measured by a microplate reader (BioTek Synergy H4 Hybrid). SOD was expressed by determining the absorbance at 440 nm, and one unit of enzyme activity is defined as the amount of enzyme required for 50% inhibition of nitrotetrazolium blue reduction. GST activity was calculated by measuring the variations in absorbance recorded at 340 nm corresponding to the conjugation of 1-chloro-2,4-dinitrobenzene/ethacrynic acid by GST. POX activity was measured by recording the absorbance at 570 nm, and one unit of POX is defined as the amount of enzyme that reduces 1.0 mmol of H₂O₂ per min.

Protein content was measured using the protocol reported by Bradford (109), with bovine serum albumin as the standard, to normalize the results of enzymatic activities.

All tests were performed in triplicate.

Data treatment.

Fungal growth (diametric growth and dry weight) and oxidative stress data (ROS production and antioxidant enzyme activity) were tested for statistical significance using one-way analysis of variance (ANOVA), followed by a Newman-Keuls test at P < 0.05, using XLSTAT software (Addinsoft 2007-Pro, v2018.1.49386; Addinsoft, Paris, France). Tolerance indices and oxidative stress values are presented as means ± standard errors of the mean.

The opm package (69, 110), v1.3.72 for R software (v3.4.3), was used to compare the carbon source utilizations of the two fungal strains, and the results are presented with level plots; the same package was used to evaluate the effects of DDT on fungal metabolic profile. The data set constituted by all OD values, comprising three replicates × 96 substrates × two treatments (control and DDT in acetone) × two metabolic parameters (respiration and mycelial growth), gave rise to 1,152 individual PM curves. The descriptive curve parameters used to discriminate between the metabolism of R. arrhizus and T. hamatum were lambda (λ), which represents the lag phase; mu (μ), which is the slope; A (maximum), representing the OD value reached at the end of incubation when the organism reached a plateau in growth and substrate use; and the area under the curve (AUC) (69, 110).

The absorbance readings obtained at 750 and 490 nm for each fungus throughout the whole incubation period with or without DDT were first analyzed with an exploratory principal component analysis (PCA; not shown; XLSTAT, Addinsoft, v 2018). The first three components resulting from the PCA were then used to run a discriminant analysis (DA). This procedure was aimed at choosing the incubation time with the highest variance, providing data useful for further comparisons.

ANOVA was carried out on (i) single time points in the data set, where it was recorded for both the fungi, for the maximum variance between inocula (144 h of incubation), and (ii) the aggregate mean of AUC estimates obtained for each well (69, 110) as synthetic values of fungal use of the carbon sources. Furthermore, the 95 substrates were divided into 16 categories according to chemical or metabolic affinities (water, heptoses, hexoses, pentoses, sugar acids, hexosamines, polyols, polysaccharides, oligosaccharides, glucosides, peptides, l-amino acids, biogenic and heterocyclic amines, tricarboxylic acid [TCA] cycle intermediates, aliphatic organic acids, and others) as described by Pinzari et al. (111). The average absorbance for all wells within each category was calculated. The correlation between values of substrate use (respiration, OD₄₉₀) and growth pattern (biomass/turbidity production, OD₇₅₀) for each group of compounds was calculated with data obtained after 144 h of incubation. The resulting scatter plotting with regression analysis was used to compare the metabolic phenotypes of the two fungi for the treatment versus the control conditions. Moreover, the ratios between the values of substrate use (respiration, OD₄₉₀) and growth pattern (biomass/turbidity production, OD₇₅₀) indicated the specific fungal respiration rate and were used to show the metabolic efforts of each organism to use certain substrates (111).

Chemicals and analytical reagents.

Two DDT isomers (o,p′-DDT and p,p′-DDT) were added to the media for the different biological tests: o,p′-DDT (97.0% purity; Ehrenstorfer GmbH, Wesel, Germany) and p,p′-DDT (99.1% purity; Sigma-Aldrich, Seelze, Germany). Fungal culture media were amended with a final concentration of 1 mg liter⁻¹ DDT solution (o,p′-DDT/p,p′-DDT = 1:4) prepared in pure acetone.

Data availability.

The data sets generated and analyzed during the current study are available from the corresponding author on reasonable request. Sequence data that support the findings of this study have been deposited in GenBank with the accession numbers MK890773, MK895140, MK890772, and MK895139.