Selective screening: isolation of fungal strains from contaminated soils in Austria

Unsieved soil of each sample (approximately 0.15 g) was dissolved in 1 mL of 0.9% sterile NaCl solution and vortexed to gain a homogenized liquid inoculant. 20 μl of each inoculant were used for the microtiter plate screening method using a 96 wells plate. Three different contaminants were used to identify new fungal species being able to grow in the presence of PCB 126 (10 ng μl^-1 PCP 126 in isooctane, Dr. Ehrenstorfer GmbH), hexadecane (99% analytical grade, Alfa Aesar) or toluene (Merck KGaA, Darmstadt, DE). Per contaminant 150 μl of media, 50 μl of contaminant solution and 20 μl of the inoculum inoculant were used for hexadecane and PCB 126. Plates were kept at room temperature on a shaker. Toluene (5%) was diluted with dibutyl phthalate in a beaker in a glass vacuum desiccator and plates were stored next to it to expose them to the volatile phase. Toluene plates were filled with 200 μl of medium plus 20 μl of inoculant. All tests were performed in triplicates, a negative control (medium without inoculant) was included on each plate. In addition, a positive control was carried out for each soil: 200 μl of a glucose medium and 20 μl of inoculant to investigate the number of fungal colonies deriving from the contaminated soil samples with glucose as carbon source. Growth of fungal species was measured through changes in the optical density (OD) at 700 nm with a microtiter plate reader Infinite M 1000 (Tecan, CH, set at 24°C, 70.8 rpm, wavelength: 700 nm, bandwidth: 5 nm) and OD values were corrected with values of the negative control. After 10 days, fungal growth was observed in all wells, the screening method was stopped and the liquid of the microtiter plate wells was plated on 2% malt extract agar (MEA: 2% malt extract, 2% D-glucose, 0.1% bacto peptone, 2% agar) and Rose-Bengal Chloramphenicol agar (Merck) from the wells with the contaminant but also from the positive control to isolate fungal strains growing on glucose and PAHs. Fungal strains were isolated and pure cultures were cultivated.

In order to test the ability of each isolated strain to live in the presence of the contaminants and to identify the best growing strains, the microtiter plate screening was repeated using pure cultures as inoculant rather than PAH contaminated soil. For this experiment, 1 cm² of fungal biomass was diluted in 1 mL of 0.9% of sterile NaCl containing glass beads (Carl Roth, DE) and was homogenized on a Ribolyzer (FastPrep-24 Instrument, MP Biomedicals, CA) for 5 sec at 4 m s^-1. The screening protocol for PAH contaminated soils was also used to grow strains in triplicates at room temperature in the presence of hexadecane, PCB 126 or toluene. Additionally, positive and negative controls, as previously described, were set up and the plates were incubated for 27 days.

As described above, soil samples varied in PAH content (Figure 3 and Table 2). This had a direct effect on the number of colonies isolated from the soil (Figure 1a, b). In the first plate screening, mixed cultures were obtained and resulted in 93 pure cultures. The highest amount of PAH was measured in S6. Nevertheless we could isolate two fungal colonies, one of them resulting in a successful candidate of the screening. In 6 out of 12 samples, a link between the proportion of HMW to total PAH concentration and the total number of isolated fungal communities was indicated by a strong linear regression (Figure 1a). The proportion of HMW / total PAH can be linked to the age of the contamination (Lladó et al., 2013). Accordingly, microorganisms in soil samples with lower proportion of HMW PAHs tend to be less adapted to contaminant degradation compared to microorganisms that could adapt to the contamination over a longer period of time in soil samples with a higher proportion of HMW PAHs. The 6 samples excluded from the regression analysis (S 7-12) were samples of very high HMW proportion, indicating high age. In these samples the microbial composition may have further changed because most degradable contaminants were already degraded. No correlation between individual PAHs and the ability to survive one of the three target contaminants could be found.

Table 2

GC Results: Measured US EPA PAH concentrations and standard deviations of the 12 soil samples (S1-12). Values are presented as mg/kg dry weight and standard deviations were calculated (mg/kg dryweight±standard deviation)

Tabelle 2. GC-Ergebnisse: Gemessene US EPA PAK-Konzentrationen und Standardabweichungen der 12 Bodenproben (S1-12) sind in der Tabelle dargestellt. Die Werte sind also mg/kg Trockengeweicht dargestellt und die Standardabweichung wurde berechnet (mg/kg Trockengewicht±Standardabbweichung)

Figure 1

After the cultivation of pure cultures, the microtiter plate screening was repeated. Results are represented in Table 3. All values were corrected with the values of the negative controls.

Table 3

Results of the second microtiter plate screening: Number of colonies of having the same growth pattern of all 93 fungal isolates on toluene, hexadecane and PCB 126. + = good growth, ~ = very slow growth and - = no growth. Hex = hexadecane, Tol = toluene, PCB = PCB 126

Tabelle 3. Ergebnisse des zweiten Mikrotiterplatten-Screenings: Anzahl der Kolonien mit demselben Wachstumsmuster innerhalb der 93 isolierten Kolonien in Gegenwart von Toluol, Hexadekan und PCB 126. + = gutes Wachstum, ~ = sehr langsames Wachstum und - = kein Wachstum. Hex = Hexadekan Tol = Toluol, PCB = PCB 126

For the 93 isolated fungal colonies, survival was highest in the presence of hexadecane (76%) and toluene (56%), likely due to a lower toxicity compared to the chlorinated PCB (LD50 Rat oral: Toluene 2.6 to 7.5 g/kg, PCB 4250 mg/kg, Hexadecane not lethal; cf., U.S. National Library of Medicine). 20% of the colonies showed slow or no growth in the pure strain screening, which can be explained by mutualism or by an additional carbon source such as dissolved organic carbon from the soil which was missing in the second screening using pure cultures.

Interactions and mutualism of different strains in the microbiota might play an important role for degradation as a study by Boonchan et al. (2000) showed.

After eliminating similar strains based on their macroscopic and microscopic appearance, we chose 11 fungal strains (Table 4) that showed a high increase in biomass (Absorbance ≥ 0.5 for hexadecane or toluene, increasing absorbance with PCB) and more than one microtiter plate setup (Figure 2a, b, c). These strains were chosen for identification through Sanger sequencing of the ITS or 18S rDNA. We used different primer pairs to get the optimal coverage and identity using NCBI BLAST (Table 1). For most of the strains, growth in the presence of PCB 126 was poorly visible. Only strains BL3 and U, identified as Ochroconis longiforum and Pyrenochaeta inflorescentiae, showed clear increase of OD values over 27 days. In the presence of hexadecane, all the strains but alpha 14 and BL4 showed high increase of OD values.

Table 4

Growth pattern of the chosen fungal isolates for sequencing: + = good growth, ~ = very slow growth and - = no growth. Hex = hexadecane, Tol = toluene, PCB = PCB 126

Tabelle 4. Wachstumsmuster der zum Sequenzieren ausgewählten Pilzisolate: + = gutes Wachstum, ~ = sehr langsames Wachstum und - = kein Wachstum. Hex = Hexadekan Tol = Toluol, PCB = PCB 126

Figure 2

Figure 3

The two fungal strains only increased their biomass at the end of the test, and therefore, might need longer growth time. For 9 strains, an increase of OD in the presence of toluene was visible.

Nine out of 11 strains (as given in Table 1) have previously been reported as isolated from the contaminated sites. Especially good results in the microtiter plate screening showed Purpureocillium (Paecilomyces) lilacinum. The strain showed the greatest increase of biomass in the presence of hexadecane and toluene. It was reported to be isolated from various contaminated sites: plastics contaminated soil (Pradeep et al., 2013), from the acidic river Rio Tinto (Oggerin et al., 2013) and — classified as Penicillium lilacinum — from mine drainage containing high amounts of cadmium (Tatsuyama et al., 1975). The strain is very stress resistant, survives extreme conditions concerning pH values and heavy metal concentration but is also medically relevant as opportunistic pathogen (Luangsa-Ard et al., 2011).

Ochroconis longiphorum (CBS 435.75) was isolated from the Canadian soil underneath the asphalt of a car park (Samerpitak et al., 2014), but also from the soil contaminated by a leaking diesel tank (Andreolli et al., 2016). Ochroconis longiphorum was formerly also described as Scolecobasidium longiphorum. In the genus Ochroconis, opportunistic infections of cold blooded vertebrates are reported but also for some thermophilic species warm-blood vertebrates might be a susceptible host (Samerpitak et al., 2014).

Pyrenochaeta inflorescentiae showed good growth in the presence of hexadecane and toluene. The reference listed in the Blast result (GenBank: EU552153.1) is deposited at the strain collection of the CBS-KNAW fungal biodiversity Center (CBS 119222), originating from Cape Town, from a senescent flowerhead. No degradation of contaminants of P. inflorescentiae was reported yet, but Pyrenochaeta sp. DS3sAY3a, was isolated from passive coal mine drainage treatment systems in central Pennsylvania, where toxic metals are removed from the drainage water (Santelli et al., 2010). Zeiner et al. (2016) concluded that Pyrenochaeta sp. could contribute to lignocellulose degradation by woodrot Basidiomycete fungi using cellulases and hemicellulases together with redox-active accessory enzymes in their cellulose-degrading capacity. This degrading capacity may be based on the direct enzymatic carbohydrate breakdown as well as indirect carbon oxidation via Fenton-based hydroxyl radical formation (Zeiner et al., 2016). Results from our study show that P. inflorescentiae can live in the presence of contaminants and might even be able to degrade them not only through co-metabolism with wood-rot fungi.

The isolation of Roussoella intermedia in this study pointed out as there is no report on the degradation of contaminants, but still the strain grew in the presence of all three contaminants. Roussoella intermedia was isolated from bamboo tree trunks in Papua New Guinea (CBS 170.96) and Hong Kong (ZHOU et al.) and it was isolated from north temperate lakes (Fallah-Moghaddam, 1999). Ligninolytic fungi as Roussoella are mostly wood degraders and they can be enriched through lignin-rich additions such as tree branches to soil (Bamforth and Singleton, 2005). Lignocellulosic material, which was colonized by the fungal strain originally, may have been added to the soil and the ligninolytic enzymes for wood degradation may have helped to degrade PAHs and the other contaminants tested.

Penicillium janthinellum shows significant growth in the presence of hexadecane, toluene and PCB 126, although the latter at a lower rate (Figure 2c). Penicillium janthinellum, renamed Penicillium simplicissimum was isolated from S 1 with the highest concentration of Pyrene, Chrysene and Benzo[a]pyrene. This is in line with literature where degradation of Pyrene, Chrysene, Benzo[a]pyrene (Cerniglia, 1997) by this strain is reported. Penicillium janthinellum has the ability to immobilize aluminum (Leitão, 2009) and was isolated from extremely acidic soil (Hujslová et al., 2017). Boonchan et al. (2000) suggest that P. janthinellum needs a bacterial co-culture to mineralize benzo[a]pyrene as a sole carbon and energy source. It is reported to degrade the aliphatic and aromatic fractions of crude oil (Oudot et al., 1987), oxidize chrysene (Kiehlmann et al., 1996) and April et al. (1999) report that 31% of the aliphatic fraction of Norman Well crude oil was degraded. Compared to other isolated fungal strains, it is not reported to be pathogenic for human and therefore, is a good candidate for bioremediation purposes.

Another good degrader for hexadecane might be Penicillium canescens. Penicillium canescens showed increased OD values in the presence of hexadecane and slightly with PCB. It was described by Say et al. (2003) as biosorbent for heavy metals from aqueous solutions and was therefore suggested as a strain for the cleanup of the environment, especially the aqueous systems. Further, it was shown to degrade dibenzofuran (Hammer et al., 2001). Garon et al. (2004) reported the isolation of the strain from a heavily PAH contaminated soil. Interestingly, it was isolated from S6, the sample with the highest amount of PAHs.

Doratomyces purpureofuscus was described by Bello Mouhamadou et al. (2013) as being isolated from the PCB contaminated soil and was shown to be able to degrade a mixture of 7 PCBs. It shows very slow growth with PCB 126 in this screening but increased growth in the presence of toluene. This non-pathogenic strain is successfully growing also in presence of hexadecane. Interestingly, 7 out of the 11 strains are opportunistic pathogens or have a close pathogenic strain within the genus. In the genus Roussoella, there is one reported opportunistic pathogen, Roussoella percutanea (Ahmed et al., 2014). In the genus Pyrenochaeta, the strain Pyrenochaeta romeroi (Yadav et al., 2015; Kulkarni et al., 2016) is reported to cause subcutaneous phaeohyphomycosis in transplant or diabetes patients (Yadav et al., 2015; Kulkarni et al., 2016). Purpureocillium (Paecilomyces) lilacinum is also reported as an opportunistic pathogen (Luangsa-Ard et al., 2011). This is in line with the hypothesis of Prenafeta et al. (2006), which suggests a linkage of microbial degradation potential and pathogenicity that might be due to the similarity of chemical structures of several brain components resembling alkylbenzene and lignin biodegradation intermediates.