Fusarium oxysporum Schltdl. (IBPPM 543) MG593980, Talaromyces sayulitensis Visagie, N. Yilmaz, Seifert and Samson (IBPPM 664) KJ775206, and Azospirillum brasilense SR80 (IBPPM 24) QXHE00000000 were obtained from the IBPPM RAS Collection of Rhizosphere Microorganisms (http://collection.ibppm.ru, accessed on 23.07.2022). Trichoderma harzianum and Trichoderma viride were obtained from the collection of the Severtsov Institute of Ecology and Evolution of the Russian Academy of Sciences.

The fungi were grown on a modified basidiomycetes-rich medium [44] (g/L): NH${}_4$NO${}_3$, 0.724; KH${}_2$PO${}_4$, 1.0; MgSO${}_4$ $\times$ 7H${}_2$O, 1.0; KCl, 0.5; yeast extract, 0.5; FeSO${}_4$ $\times$ 7H${}_2$O, 0.01; ZnSO${}_4$ $\times$ 7H${}_2$O, 0.0028; CaCl${}_2$ $\times$ 2H${}_2$O, 0.033; glucose, 10.0; peptone, 10.0; agar, 15.0; (pH 6.0). The growth of fungal mycelium was controlled by the weight method [45].

Bacteria were grown on R2A medium [46] (g/L): yeast extract, 0.5; peptone, 0.5; casein hydrolysate, 0.5; glucose, 0.5; soluble starch, 0.5; sodium pyruvate, 0.3; K${}_2$HPO${}_4$, 0.3; MgSO${}_4$ $\times$ 7 H${}_2$O, 0.05; agar, 15.0. Growth was controlled by direct plating of multiple dilutions of the culture medium on R2A agar and by counting the CFUs.

The organisms were checked for antagonistic activity by the double-culture method [47].

Phenanthrene (~90%), pyrene (~90%), anthracene ($\ge$95%), fluoranthene ($\ge$97%), fluorene ($\ge$99%), and IAA (Pestanal®, analytical standard) were all purchased from Fluka (Buchs, Switzerland).

For degradation studies, we used a medium previously composed by Tsivileva et al. [48] (g/L): glucose, 4.5; fructose, 4.5; L-asparagine, 1.5; KH${}_2$PO${}_4$, 1.0; K${}_2$HPO${}_4$, 1.0; Na${}_2$MoO${}_4$ $\times$ 2H${}_2$O, 0.002; CaCl${}_2$, 0.02; MgSO${}_4$ $\times$ 7H${}_2$O, 0.2; NaCl, 0.1; NH${}_4$Cl, 1.0. A mixture of PAHs (anthracene, phenanthrene, fluorene, pyrene, and fluoranthene; 1 mg of each) was added to a final PAH concentration of 5 mg per 100 mL. Fungi (20 mg of dry weight or about 10${}^4$ CFU) and A. brasilense (3 $\times$ 10${}^5$ CFU), both singly and in combination, were grown at 24, 30, or 37 °C in 0.3 L Erlenmeyer flasks on an Environmental Shaker–Incubator ES20/60 (BioSan, Latvia) at 130 rpm. At certain time intervals, an aliquot (2 mL) was removed from the flasks and the extracellular enzyme activities and the IAA concentration were determined, as described below.

After 14 days, the contents of the flasks were extracted three times with 5 mL of chloroform, and the extracts were evaporated [43, 45, 49]. The sorption of the PAHs by fungal mycelium did not exceed 5–7%.

High-performance liquid chromatography (HPLC) was performed on an Agilent Technologies 1220 Infinity II LC chromatograph (Agilent Technology, Waldbronn, Germany) equipped with a 254 nm UV detector.

The PAH content was analyzed on a 4.6 $\times$ 150 mm ZORBAX Eclipse PAH 5-Micron column. The solvent system was H${}_2$O:acetonitrile, the linear gradient was 40–100% acetonitrile, and the duration was 17 min. The PAHs and their metabolites were analyzed by comparing the retention times (RT) with those of standard compounds (RT, min): carboxybenzaldehyde (1.938 min), 9,10-phenanthrenequinone (5.145 min), 1-hydroxy-2-naphthoic acid (5.823 min), 9-fluorenone (6.856 min), 9,10-anthraquinone (7.320 min), fluorene (8.844 min), phenanthrene (9.353 min), anthracene (9.888 min), pyrene (10.382 min), and fluoranthene (10.837 min).

IAA content was analyzed on a 4.6 $\times$ 50 mm ZORBAX Eclipse Plus C${}_{18}$ 5-Micron column. The solvent system was H${}_2$O (acidified with H${}_3$PO${}_4$ to pH 2.5): acetonitrile, the linear gradient was 10–90% acetonitrile, and the duration was 15 min.

Enzyme activity was determined by an Evolution 60 (840-208700, Thermo Scientific™, Madison, WI, USA). Laccase activity was measured by the oxidation rate of 2,2${}^{\prime }$-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)diammonium salt (ABTS) at 436 nm ($\epsilon$ = 29,300 M${}^{-1}$cm${}^{-1}$), according to Niku-Paavola et al. [50]. Peroxidase was measured by the oxidation rate of 2,6-dimethoxyphenol (DMP) with H${}_2$O${}_2$ at 468 nm ($\epsilon$ = 14800 M${}^{-1}$cm${}^{-1}$), according to Martínez et al. [51] and by ABTS oxidation, as described above, but in the presence of 0.1 mM H${}_2$O${}_2$. The peroxidase activity was calculated as the difference between the values obtained in the presence and absence of H${}_2$O${}_2$. One unit of enzyme activity was defined as the amount required to catalyze the formation of 1 µmol of product per min and is expressed in relative units (µmol/min/mL of enzyme preparation (U/mL)).

All experimental treatments and all analyses were repeated at least three times. The obtained experimental data were statistically processed, average values were calculated, and the standard deviation (SD) was used for comparisons. Calculations were carried out using Microsoft Excel 2019 (Microsoft, WA, USA).

We first evaluated the mutual influence of the fungi and bacteria by the double-culture method. This method is often used to identify the antagonistic effect of an organism on a particular pathogen. Antagonists grow faster than pathogens and form zones of inhibition, thereby limiting pathogen growth. For example, a similar test was used to examine the ability of a number of Trichoderma species to suppress the growth of Sarocladium oryzae [78]. In our experiments, no antagonistic activity was found in the following pairs: A. brasilense–F. oxysporum, A. brasilense–T. sayulitensis, and A. brasilense–Tr. harzianum. From Fig. 1 we can see the active growth of the mycelium of the studied fungi and A. brasilense. By contrast, Tr. viride suppressed the growth of A. brasilense on agar, as evidenced by the absence of bacterial growth on the agar with active growth of fungal mycelium (Fig. 1). In earlier studies by Kalaiselvi et al. [78], Tr. harizanum was more active than Tr. viride and some other Trichoderma species at inhibiting the growth of Sarocladium oryzae. Apparently, the ability to inhibit the growth of bacteria is determined not only by the Trichoderma species but also by the bacterial species.

Fig. 1.

Growth of bacterial–fungal co-cultures on R2A agar medium.

Anthracene and fluoranthene were the least available of all the fungi (Fig. 2A). This may have been due to the low solubilities and high ionization potentials of these compounds [24, 25, 26, 27, 28]. Among the ascomycetes used, F. oxysporum was the weakest degrader, whereas T. sayulitensis and Tr. harzianum showed the highest degradative activity towards the PAHs. The other Trichoderma member, Tr. viride, intensely metabolized only low-molecular-weight PAHs (fluorene and phenanthrene, 67 $\pm$ 3% and 54 $\pm$ 5%, respectively) (Fig. 2A). Previously, Zafra et al. [35]showed that low-molecular-weight PAHs (naphthalene, phenanthrene, and anthracene) are more accessible for degradation by Trichoderma. These data were confirmed by other authors. For example, Tr. longibrachiatum metabolized between 70 and 90% of the phenanthrene depending on the cultivation conditions [71]. Another study showed that the same fungus metabolized up to 70% of phenanthrene, anthracene, fluoranthene, and pyrene at a starting concentration of 25 mg/L [60]. Tr. harzianum metabolized 27–28% of the initially added 46.8 mg/L of anthracene [79]. Tr. harzianum, Tr. pseudokoningii, and Tr. viride-metabolized pyrene [72]. Tr. reesei cometabolized 54% of the benzo[a]pyrene, at a starting concentration of 20 mg/L [80].

Fig. 2.

Degradation of individual polycyclic aromatic hydrocarbons (PAHs) (A) and their model mixture (B) by the ascomycetes. FLU, fluorene; PHEN, phenanthrene; ANTH, anthracene; FLA, fluoranthene; PYR, pyrene. ◼ F. oxysporum; ◼ T. sayulitensis; ◼ Tr. harzianum; ◼ Tr. viride. The standard deviation (SD) did not exceed 10% of the basic values.

Each of 9-fluorenone, 9,10-phenanthrenquinone, and 9,10-anthraquinone were identified as the main metabolites in the degradation of fluorene, phenanthrene, and anthracene, respectively, in F. oxysporum and T. sayulitensis. This was confirmed by the retention times (RTs) of the identified metabolites against those of the commercial compounds. The formation of quinones as the first metabolites of the PAH degradation by Fusarium is well known. For example, F. solani degrades anthracene and benzo[a]anthracene via the corresponding quinones, which is followed by the formation of phthalic acid [81]. Although Talaromyces members can degrade PAHs, no information is available on the key metabolites [82]. In our study, quinone metabolites of PAH degradation were found in T. sayulitensis for the first time. Moreover, 1-hydroxy-2-naphthoic acid was identified as a metabolite of phenanthrene degradation by Tr. harzianum, whereas no quinone metabolites were found. This is consistent with data from Hadibarata et al. [83], who showed that Trichoderma sp. S019 metabolized up to 72% of the phenanthrene to produce 1-hydroxy-2-naphthoic acid, salicylaldehyde, salicylic acid, and catechol through the pathway described in the bacterium. We did not detect PAH degradation metabolites in Tr. viride, which may have been due to the low degradative activity of this fungus.

All ascomycetes utilized the model mixture of PAHs (equal amounts of fluorene, phenanthrene, anthracene, pyrene, and fluoranthene) by almost 100% (Fig. 2B). A similar effect was described by Cobas et al. [71], who showed that Tr. longibrachiatum immobilized on nylon in a bioreactor utilized a mixture of three PAHs (phenanthrene, benz[a]anthracene, and pyrene) faster than it utilized individual PAHs.

Bacteria in general prefer aerobic PAH degradation to be catalyzed by dioxygenase enzymes. There is no information on the ability of Azospirillum to degrade PAHs in the reviews we described [26, 28]. Previously, we had shown that the PGPR A. brasilense SR-80 can utilize oil as its carbon source [53, 54]. Without denying the presence of a dioxygenase pathway in Azospirillum, we further assumed the presence of an alternative pathway since the formation of PAH quinones at the first stage of degradation testifies in favor of an alternative pathway [43]. Subsequently, we found that this strain could cometabolize only fluorene and anthracene in the presence of a carbon source (glucose, fructose), with the removal of the pollutant identified as 20 $\pm$ 6% and 23 $\pm$ 4%, respectively. During cultivation with a mixture of PAHs, A. brasilense SR-80 could cometabolize all the compounds used. However, the availability and degree of utilization of individual PAHs depended on the cultivation temperature. Utilization was the most intense at 30 °C (Fig. 3A). PAHs are known to be more greatly degraded in a mixture compared to individual PAHs. Thus, the co-metabolism of the PAHs shows synergistic effects on the biodegradation of other PAHs, especially high molecular weight PAHs [29].

Fig. 3.

Degradation of the PAH model mixture (A) and chromatogram of the culture medium after growth of A. brasilense SR-80 (B). FLU, fluorene; PHEN, phenanthrene; ANTH, anthracene; FLA, fluoranthene; PYR, pyrene; FLUQ, 9-fluorenone; ANTHQ, 9,10-anthraquinone.

Fig. 3 shows a yellow-colored extract of the culture liquid (vial) and a chromatogram of metabolites. HPLC allowed the identification of the degradation products of fluorene (9-fluorenone, RT 6.856 min) and anthracene (9,10-anthraquinone RT 7.320 min) (Fig. 3B).

The PAHs are presented in the environment as the mixture; thus, the bioremediation of contaminated sites could be crucially performed by co-metabolism. In our experiments, the PAH mixture was degraded by co-cultures, which included F. oxysporum, T. sayulitensis, Tr. viride, Tr. harzianum, and A. brasilense at temperatures of 24, 30, and 37 °C, which are optimal for the growth of the fungi, co-cultures, and bacterium, respectively. The growth of F. oxysporum and T. sayulitensis was inhibited in the co-cultures, regardless of the cultivation temperature. An increase in the amount of mycelium in the co-cultures was noted only for the Trichoderma fungi (Table 1). Yet, it was impossible to calculate the CFUs after plating the co-cultures on the agar medium owing to the complete overgrowth of agar with mycelium.

Apparently, during submerged cultivation of F. oxysporum and T. sayulitensis, the bacterial and fungal cultures inhibited each other’s growth. It is known that many bacterial–fungal interactions are mediated by the production of various metabolites. For example, fusaric acid produced by F. oxysporum regulates the transcription of a number of genes in Pseudomonas chlororaphis [5]. Competition between F. oxysporum and Arthrobacter oxydans was shown in the presence of PAHs during submerged cultivation [10]. Perhaps similar effects were observed in our experiments. The lack of growth inhibition during solid-state cultivation of the F. oxysporum–A. brasilense and T. sayulitensis–A. brasilense pairs may have resulted from a lack of production of such metabolites. Thus, the absence of antagonistic mutual influence by the bacteria and fungi, found under solid-state cultivation, will not necessarily indicate a similar effect under submerged cultivation.

Studies on the degradation of the PAH mixture showed that all compounds included in the mixture were available for fungal degradation. There were no synergistic effects on PAH degradation in the co-cultures with F. oxysporum (Table 2). A significant decrease in the content of individual PAHs was the result of their fungal degradation. Similar data were obtained by Thion et al. [10], who showed the co-existence of F. oxysporum and Arthrobacter oxydans in sterile sand in the presence of PAHs. However, co-inoculation did not noticeably affect the degradation of the mixture of PAHs (phenanthrene, pyrene, and dibenz[a,h]anthracene) under these conditions [10]. However, synergistic effects of co-cultivation of Fusarium and bacterial strains have also been described. For example, a co-culture of Fusarium sp. and Ochrobactrum anthropi metabolized fluoranthene [84], and a co-culture of F. solani and Rhodococcus erythropolis metabolized benzo[a]pyrene [41] more efficiently than its individual cultures.

T. sayulitensis intensely metabolized the PAHs in the mixture at 30 °C, both alone and in combination with A. brasilense (Table 2). Synergistic effects were found at 24 and 37 °C. At 24 °C, T. sayulitensis did not metabolize the PAHs, despite the vigorous growth of the fungus; in the co-cultures, the degradation increased, compared to A. brasilense alone. At 37 °C, the A. brasilense monoculture did not utilize anthracene, pyrene, or fluoranthene, whereas all PAHs were available in the T. sayulitensis monoculture. In the co-cultures, increased degradation of anthracene, pyrene, and fluoranthene was noted. The synergistic effects we found confirmed the previous data by Shapiro et al. [17], who showed that Talaromyces micromycetes associated with bacteria can grow on kerosene, which suggests a degradative mechanism based on the synergism of bacteria and fungi [17].

Positive effects, resulting in increased PAH degradation, were found when the plant-growth-promoters A. brasilense and Tr. viride or Tr. harzianum were co-cultivated (Table 2). Ascomycetes of the genus Trichoderma are characterized by catabolic degradation of PAHs: Tr. reesei [80], Tr. asperellum [35], Tr. longibrachiatum [60, 71], and Tr. harzianum [72, 79]. The utilization of PAHs (benz[a]anthracene) as the sole carbon source was shown only for T. longibrachiatum [70]. In our experiments, for all ascomycetes, PAH degradation was possible only under the conditions of cometabolism. We identified the degradation products by the co-cultures (fluorene as 9-fluorenone and anthracene as 9,10-anthraquinone), which corresponds to the data presented above.

The initial steps in PAH degradation by many micromycetes are catalyzed by cytochrome P450 monooxygenase and epoxide hydrolase. The reaction products catalyzed by these enzymes can be trans-dihydrodiols, phenols, quinones, and dihydrodiol epoxides, which can further form glucuronide, glucoside, xylose, and sulfate conjugates. However, some species of ascomycete can produce laccases, peroxidases, and dioxygenases that break the aromatic ring [35, 73].

The production of laccases and peroxidases by various Fusarium species is well known [81, 85]. The involvement of these enzymes in the degradation of PAHs by Fusarium has also been discussed [86, 87]. We have previously shown that, unlike other known strains of F. oxysporum, the IBPPM 543 strain of F. oxysporum produces peroxidase (without laccase) during the degradation of pollutants [49]. The activity of this enzyme in the presence of PAHs was induced only at 24 °C, which is consistent with our previous results [49]. The enzyme activity was the highest (2.2 U/mL) during cultivation at 30 °C. However, following an increase in temperature to 37 °C, the extracellular activity of peroxidase decreased, which may be due to both the fungal growth being inhibited at this temperature and a decrease in the enzymatic stability as the temperature increased. In co-cultures, peroxidase activity decreased at 24 and 37 °C and slightly exceeded the control value only at 30 °C (3.3 U/mL).

We found the production of two extracellular enzymes: oxidase (according to the oxidation of ABTS) and peroxidase (according to the oxidation of 2,6-dimethoxyphenol in the presence of H${}_2$O${}_2$) by the fungus T. sayulitensis only at 30 °C (Fig. 4). The activity of both enzymes was low and increased when the PAHs were added to the co-cultures.

Fig. 4.

Production of extracellular enzymes during PAH degradation. ◼ – oxidase; ◼ – peroxidase.

Although the ability of Talaromyces to metabolize lignocellulose and produce hemicellulases is well known [74], there is no information on the production of enzymes that catalyze the degradation of the lignin component. Enzymes have been described in some Talaromyces members that, hypothetically, can be related to the degradation of aromatic compounds and pollutants, such as PAHs. For example, cytochrome C peroxidase, glutathione peroxidase, catalase peroxidase, DyP peroxidase, and a hybrid of ascorbate peroxidase and cytochrome C were found in T. stipitas. Catalase peroxidase was found in T. termophilus [88], while in T. asrtoroseus, which degrades dichlorvos, genes encoding catechol 1,2-dioxygenase were found to be expressed [1].

As mentioned above, Trichoderma members are characterized by the presence of both intra- and extracellular enzymes of the initial attack on the PAH molecules. Trichoderma dioxygenases can catalyze the initial attack on the aromatic ring to form unstable cis-dihydrodiols, which can subsequently be metabolized via the catechol pathway [83]. The presence of PAHs resulted in a significant increase in the activity of catechol 1,2- and 2,3-dioxygenases at the initial stages of degradation [73]. The phenanthrene degradation pathway described by Hadibarata et al. [83], includes the action of catechol-1,2- and 2,3-dioxygenases in the initial stage of oxidation, with 1-hydroxy-2-naphthoic acid, salicylic acid and catechols as the main metabolites. Ultimately, this may explain the ability of some Trichoderma species to utilize PAH as their sole carbon source.

The production of extracellular enzymes, similar to the ligninolytic enzymes in basidiomycetes, has been described in Trichoderma. For example, Tr. tomentosum produced Mn-peroxidase and lignin peroxidase during the degradation of azo dyes [64]. Laccase was found in Tr. harzianum, which degrades bisphenol A and utilizes phenols in Tr. viride [66]. However, in this study, the enzyme was partially purified, and its activity was tested through the oxidation of guaiacol, which is a non-specific substrate of both laccases and tyrosinases. Sowmya et al. [68] described laccase and Mn-peroxidase during the degradation of polyethylene by Tr. harzianum. However, the activity of these enzymes was only checked through the formation of colored zones during the growth of the fungus on an agar medium with guaiacol, which also raises doubts about the nature of the identified enzymes. More correct testing of extracellular laccase (by ABTS) and Mn-peroxidase (by vanillyl acetone) revealed no activities of these enzymes during the degradation of benzo[a]pyrene by Tr. reesei [80]. Additionally, activities of laccase (by ABTS) and peroxidases (by phenol red and veratryl alcohol) during PAH degradation were revealed in Tr. asperellum [74]. The ability of Tr. longibrachiatum to degrade PAHs was accompanied by the production of phenol oxidase, which was identified following the oxidation of gallic acids; this is a more correct name for the extracellular oxidative enzyme in this case [60]. In addition, extracellular laccase production has been described in Tr. atroviride, Tr. harzianum, and Tr. asperellum [74], the oxidation of PAHs by this enzyme may represent an alternative route for the initial attack of the molecules of these compounds.

In our experiments, the activities of extracellular oxidases and peroxidases in Tr. viride or Tr. harzianum under these studied conditions was not found. Apparently, under these conditions, the degradation of PAHs by Tr. viride or Tr. harzianum proceeds along the bacterial pathway described by Hadibarata et al. [83]. The 1-hydroxy-2-naphthoic acid that we identified as the main metabolite of phenanthrene degradation confirms this assumption.

Trichoderma are ubiquitous ascomycetes with a number of growth-promotion properties, including the synthesis of IAA, ACC-deaminase, siderophores, acid phosphatase, and phosphate-dissolving agents under biotic and abiotic stress [60, 89]. For example, IAA production was found in Tr. longibrachiatum [89]. Tr. virens did not produce IAA, yet ACC deaminase was detected [60]. We identified the ability of Tr. viride or Tr. harzianum to produce IAA during PAH degradation, although the IAA concentration was low and did not exceed 0.1 µg/mL. The other two ascomycetes did not produce IAA.