Introduction

The polycyclic aromatic hydrocarbons (PAHs) are molecules with two or more fused benzene rings. The common sources of PAHs in the environment include natural as well as anthropogenic. Natural sources are forest and rangeland fires, oil seeps, volcanic eruptions, and exudates from trees, while anthropogenic sources of PAHs include burning of fossil fuel, coal tar, wood, garbage, and used lubricating oil, among others. PAHs are considered priority pollutants due to their carcinogenic and mutagenic effects (Cerniglia and Sutherland 2010) and thus represent a public health risk. High molecular weight PAHs (HMW-PAHs), containing six or more fused aromatic rings, are recalcitrant to microbial attack (Das and Chandran 2011) limiting their biodegradation in different ecosystems. Recently, the term “recalcitrant” has been revisited and suggested that solubility is more important than molecular complexity in determining compound persistence in nature (Marín-Spiotta et al. 2013). This has been proved for the enzymatic transformation of asphaltenes in which the transformation was carried out in a one-phase ternary medium constituted of different proportions of toluene, isopropanol, and water. According to the reaction media used, it seems that a main limiting factor for the asphaltene transformation is its high hydrophobicity, and thus the biotransformation may occur only when the asphaltenes and the enzyme are in the same phase, reducing the mass transfer limitations (Fedorak et al. 1993). Some authors consider HMW-PAHs compounds with more than four aromatic rings; however, several reports are available in literature describing the microbial transformation of compounds with four and five aromatic rings. Here, we consider as HMW-PAHs those compounds containing six or more aromatic rings. The recalcitrance of PAHs to microbial degradation increases directly with their molecular weight and their hydrophobicity (octanol-water partition). Microbial metabolism of low molecular weight PAHs containing up to three rings (naphthalene, phenanthrene, anthracene and fluorene) has been extensively studied. However, information about the microbial degradation of PAHs with more than five benzene rings is scarce (Juhasz and Naidu 2000). In spite of abundant information in the literature about the biodegradation of PAHs containing less than five aromatic rings, to our knowledge, there are very few reports with HMW-PAHs of six or more aromatic rings. Recently, the conferred capacity to metabolize a six-ring PAH, benzo(g,h,i)perylene, in a yeast expressing the cytochrome P450 (CYP)63A2 gene from Phanerochaete chrysosporium has been reported (Syed et al. 2013).

Asphaltenes are present in most petroleum materials and in all heavy oils and bitumens from oil sands. This material is defined by solubility as the components that dissolve in toluene and precipitate from crude oil in n-alkane solvents (either n-pentane or n-heptane). The asphaltene concentration in a light crude oil may be less than 1 %, while the concentration in heavy bitumen can be more than 20 % of the total weight. The low solubility of the asphaltene fraction in the crude oil poses challenges during the production and processing of petroleum (Scotti and Montanari 1998). The asphaltene composition includes aromatic molecules of very high molecular weight (Mullins 2011), and their complex molecular structure and high hydrophobicity makes them resistant to biodegradation. Very few microorganisms able to degrade asphaltenes have been reported (Hernández-López et al. 2015a). Some old reports on oil biodegradation claimed the degradation of the asphaltene fraction by mixed bacteria (Bertrand et al. 1983; Rontani et al. 1985; Lacotte et al. 1996). Nevertheless, none of these reports described the analytical results of extractable materials recovered from appropriate sterile controls. The complex nature of asphaltenes possesses challenges to the biodegradation studies. Unfortunately, in the majority of studies, the extent of asphaltene degradation has been determined using gravimetric measurements, without considering the effect of the presence of oxidized molecules or the production of biosurfactants in the precipitation behavior of the asphaltene fraction.

Recently, Neosartorya fischeri was the first microorganism that evidenced the mineralization of petroleum asphaltenes with rigorous substrate preparation and controls (Uribe-Alvarez et al. 2011). After 11 weeks of growth in a medium containing asphaltenes as a sole source of carbon and energy, the fungus was able to metabolize 15.5 % of the asphaltenic carbon, including 13.2 % mineralization to CO2. Pestalotiopsis sp., a fungus from another genus, was also reported able to degrade asphaltenes as a sole carbon source. This fungus, which is halotolerant, was able to degrade 21.4 % of asphaltenes after 15-day incubation (Yanto and Tachibana 2013, 2014). The results obtained in these works clearly demonstrated that there are microorganisms capable to metabolize and mineralize asphaltenes, which is considered the most recalcitrant petroleum fraction.

Beside its capacity to metabolize asphaltenes, here, we report the capacity of N. fischeri to metabolize HMW-PAHs. Additionally, we present experimental evidence that strongly suggest cytochrome P450 system involved in the oxidation of these recalcitrant compounds. The implications of these findings are discussed.

Materials and methods

Chemicals

Coronene (99 %), benzo(g,h,i)perylene (98 %) and indeno(1,2,3-cd)pyrene (98.8 %), TRI reagent® RNA isolation reagent, ethanol, and trifluoroacetic acid were purchased from Sigma-Aldrich (Saint Louis, MO). NaNO3, K2HPO4, MgSO4•7H2O, KCl, FeSO4•7H2O, and dextrose anhydrous powder were obtained from J.T. Baker (Center Valley, PA). Bacto peptone, bacto potato dextrose broth (PDB), and potato dextrose agar (PDA) were obtained from DIFCO (Sparks, MD). Dichloromethane, methanol, hexane, hydrochloric acid, and KOH were supplied by Fermont (Monterrey, Mexico).

Petroporphyrin-free asphaltenes were purified from Mexican Maya crude oil as previously described (Uribe-Alvarez et al. 2011). First, asphaltenes were obtained by precipitation with n-pentane. Crude oil (100 g) was suspended in 5 l of n-pentane and stirred for 1 h. The mixture was left in repose overnight at 20 °C in a stopped flask, then the suspension was centrifuged at 10,000g and the pellet was redissolved in 100 ml of dichloromethane. The asphaltene solution was reprecipitated with 5 l of n-pentane, maintained overnight in repose, and centrifuged. This procedure was repeated one more time and the precipitate was washed with n-pentane obtaining the asphaltenes fraction. Next, the petroporphyrin-free asphaltenes were prepared from 6 g of asphaltenes diluted in 300 ml of toluene and stirred for 2 h at 20 °C. Then, 3 l of acetone was added and the mixture stirred for 3 h at 20 °C and maintained in repose for 48 h. Afterwards, the mixture was centrifuged at 12,000g for 30 min and the pellet was washed three times with acetone. The red-brownish supernatant and subsequent washes contained the petroporphyrins, whereas the pellet contained the petroporphyrin-free asphaltenes.

Organism and culture media

Neosartorya fischeri strain used in this study was isolated and characterized as previously reported (Uribe-Alvarez et al. 2011). This strain is available in the fungal collection of the Instituto de Ecologia (WDCM782) as N. fischeri IE 978. The strain was maintained on PDA or Saboraud medium. In order to evaluate the hydrocarbons (HMW-PAHs or asphaltenes) as sole carbon source, a modified Czapek minimal medium without original carbon source was used (3.0 g l−1 NaNO3; 1.0 g l−1 K2HPO4; 0.5 g l−1 MgSO4•7H2O; 0.5 g l−1 KCl; and 0.01 g l−1 FeSO4•7H2O) supplemented with either asphaltenes, coronene, benzo(g,h,i)perylene, or indeno(1,2,3-cd)pyrene. The insoluble substrates were ground in fine particles before adding to medium. In order to avoid fungal growth on the residual carbon source from the original rich cultures, PAH and asphaltenes cultures were inoculated with an inoculum from 20-day cultures with PAH or asphaltenes as sole carbon source. All cultures were inoculated with 5 % inoculum volume. The cultures were incubated at 37 °C and shaken at 100 rpm.

Determination of oxygen consumption

The substrate fungal metabolization was monitored by measuring the oxygen consumption with an OxiTop® measuring system equipped with a KOH trap (CO2 trap). This method is based on the monitoring of pressure changes with a piezoresistive electronic pressure sensor. N. fischeri was inoculated in 40 ml of Czapek medium with 10 mg of coronene, indeno(1,2,3-cd)pyrene, benzo(g,h,i)perylene, or asphaltene. Control cultures were carried out without carbon source. The culture bottles were incubated at 37 °C and 100 rpm for 17–30 days. All the cultures and controls were performed in triplicates.

Ergosterol production

The ergosterol content in fungal biomass was measured using a methodology previously reported (Barajas-Aceves et al. 2002). N. fischeri was inoculated in 20 ml of Czapek medium containing 5 mg of coronene, indeno(1,2,3-cd)pyrene, benzo(g,h,i)perylene, or asphaltene as a sole source of carbon. Control cultures without carbon source were also analyzed. The cultures were incubated at 37 °C and shaken at 100 rpm for 14 days; then, they were centrifuged into 50-ml Teflon centrifuge bottles. Methanol (10 ml) was added to the biomass pellet, and the suspension sample was sonicated for 3 min (1 min × 3) in an ice-water bath and then centrifuged at 9000 rpm for 15 min at 4 °C. The biomass pellet was washed three times with 10 mL of methanol. Supernatants were transferred to 50-ml Teflon tubes, and 5 ml of absolute ethanol and 2 g of KOH were added. This mixture was saponified in a water bath at 80 °C for 1 h. The tubes were then removed from the water and allowed to cool and 10 ml of hexane were added. The tubes were closed and shaken vigorously. Two phases are formed and the upper hexane layer was removed and transferred into a round-bottom flasks. The alcoholic phase was washed twice with 10 ml of hexane. The combined hexane phases were then evaporated to dryness on a rotatory evaporator at 40 °C under vacuum. The extracted ergosterol was redissolved in 1 ml HPLC-grade methanol and transferred to an amber vial and stored at −20 °C until HPLC analysis.

The extracts were analyzed on an HPLC (Agilent 1100 series) equipped with a reverse-phase column Phenomenex Kinetex 5 μ C18 100 Å, 150 × 4.60 mm. The samples were eluted with 97:3 methanol/water with 0.1 % trifluoroacetic acid with a flow rate of 0.5 ml min−1, and monitored at 282 nm for 20 min. The ergosterol content was estimated against a standard curve obtained from ergosterol solutions of 0 to 1000 μg ml−1.

Metabolite identification

The metabolites from HMW-PAHs transformation were analyzed by HPLC and mass spectrometry. N. fischeri was inoculated in 20 ml of Czapek with 5 mg of coronene, indeno(1,2,3-cd)pyrene, or benzo(g,h,i)perylene and incubated for 17 days at 37 °C and 100 rpm. Then, the culture medium was acidified with a few drops of HCl and extracted three times with 10 ml dichloromethane. The mixture was shaken and centrifuged, and the organic phase was collected and concentrated under vacuum and analyzed by high-performance liquid chromatography (HPLC). The chromatograph (Agilent 1100 series) was equipped with UV detector and a reverse-phase C18 column 5 μ, 100 Å (150 × 4.6 mm) supplied by Phenomenex (Los Angeles, CA). The elution was performed with a gradient from methanol to 40 % dichloromethane in methanol over a period of 15 min, at a flow rate of 0.4 mL min−1. The substrate and metabolites were monitored at 280 nm.

Products of transformation were also analyzed by nLC-MS at the Queen’s Mass Spectrometry and Proteomics Unit (MSPU) at the Queen’s University in Kingston, Ontario. A nano LC high-resolution MS (Orbitrap Velos Pro) hybrid mass spectrophotometer and Agilent capillary HPLC 1100 system with a mass resolution of 100,000 Da and a mass accuracy of 1 ppm RMS for organic compounds, were used. The samples were eluted with a gradient from methanol to dichloromethane in 25 min at a rate of 300 nL min−1.

Fourier transform infrared spectroscopy with attenuated total reflection (FTIR-ATR) spectra were recorded by a PerkinElmer Spectrum 100 (PerkinElmer Cetus Instruments, Norwalk CT) from 650 to 4000 cm−1, using a universal ATR sampling accessory and a diamond tip. The Spectrum software supplied with the equipment was used for baseline subtraction.

Microsomal transformation of HMW-PAHs

Microsomes were obtained according to Jauregui et al. (2003). Fungal cells (10 mg dry weight) were grounded in a mortar in liquid nitrogen with protease inhibitor and suspended in 25 mL of 20 mM Tris–HCl buffer pH 7.0 containing 1 mM EDTA, 0.5 mM DTT, and 10 % glycerol. The mixture was homogenized in a blender (Dremel Moto-tool) at 28,000 rpm. The microsomal fraction was isolated by differential centrifugation. The homogenized biomass was first centrifuged at 600g for 5 min and then the supernatant was centrifuged at 12,000g for 10 min. Then, 8 mM CaCl2 was added to the supernatant to allow complete sedimentation of microsomes at 27,000g for 15 min. The pellet was washed two times with 20 mM Tris–HCl buffer containing 8 mM CaCl2.

The microsomes were assayed for the NADH oxidation in the presence of each polycyclic aromatic hydrocarbons; coronene, benzo(g,h,i)perylene, or indeno(1,2,3-cd)pyrene. Reactions were carried out in 1 ml reaction mixtures containing 10 μM of PAH, 0.4 mM NADH, 100 mM phosphate buffer, pH 5.0, containing 10 % acetonitrile, and 20 ng of microsomal protein. The PAH transformation was monitored spectrophotometrically for NADH oxidation at 340 nm. No NADH oxidation was detected in controls without microsomes or PAHs.

Microscopy studies

Atomic force microscopy

After 5-day growth in either Czapek medium with asphaltenes or glucose-peptone medium, one drop of the culture was placed on the slide and dehydrated at room temperature. The fungal cells were examined directly for atomic force microscopy (AFM) in a scanning probe microscope (SPM) equipped with an AFM XE-70 Park system. The method used was non-contact with a cantilever NC HR non-contact of Park system.

Laser scanning confocal microscopy

Fluorescence microscopy of the strain under both culture conditions (asphaltenes and glucose-peptone) was performed on an inverted laser scanning microscope FV1000 FluoView™ (Olympus, Japan) equipped with a Multi-line Ar laser for excitation at 405 for asphaltenes and PAHs and 488 nm for FM4-64 and Sybrgreen (SYBRG) and 543 nm for BODIPY. An UPlanFLN 60X, 1.42 NA oil immersion objective was used. Images were captured simultaneously by fluorescence and differential interference contrast (DIC) channels with FluoView™ software (Olympus, Japan). Final figures were created using Adobe Photoshop CS5 Extended (Adobe Systems Inc., San Jose, CA).

Cell staining methods

For different cell compartment staining, mycelia cultured with asphaltenes or glucose-peptone was incubated in separate assays with 2.5 μM FM4-64 (Molecular Probes, Eugene, OR), to label the plasma membrane and organelle membranes, 1:10,000 SYBR® Green I (Molecular Probes, Eugene, OR) to label nuclei and mitochondria and 2.5 μg ml−1 BODIPY (Sigma-Aldrich, Saint Louis, MO) to label lipids.

Results

Fungal growth on HMW-PAHs

In order to determine the ability of N. fischeri to metabolize HMW-PAHs, the mycelium was inoculated in Czapek media containing coronene, indeno(1,2,3-cd)pyrene, benzo(g,h,i)perylene, or petroleum asphaltenes as sole carbon source. Kinetics of oxygen consumption was monitored during the fungal growth (Table 1). N. fischeri was able to grow in a medium containing HMW-PAHs as a sole carbon source. Asphaltene showed the highest oxygen consumption (138 mg l−1 day−1), while coronene and indeno(1,2,3-cd)pyrene also showed an average oxygen consumption with values of 82 and 81 mg l−1 day−1, respectively, and without statistic (p < 0.05) differences between them. Finally, benzo(g,h,i,)perylene showed the lowest oxygen consumption with 62 mg l−1 day−1. Inoculated controls without carbon source showed no oxygen consumption (Table 1).

Table 1 Maximal oxygen consumption rate and ergosterol production of N. fischeri cultures with HMW-PAHs and asphaltenes as sole carbon source
Full size table

The fungal growth was estimated by measuring the ergosterol production (Table 1). Asphaltenes and coronene showed the highest average erosterol production during the first 14 days of culture with 34 and 31 μg ml−1 day−1, respectively. While cultures with indeno(1,2,3-cd)pyrene and benzo(g,h,i,)perylene showed lower ergosterol production of 22 and 27 μg ml−1 day−1, respectively. Controls without carbon source showed, as expected, no growth with a slight decrease in the ergosterol content due to endogenic metabolism (Table 1). The oxygen consumption and the ergosterol production showed a linear correlation coefficient of 0.81.

Fungal transformation of HMW-PAHs

Metabolic products were extracted from 17-day-old cultures of N. fischeri with different HMW-PAHs as sole source of carbon and energy. The extracts were analyzed by HPLC, infrared spectroscopy, and mass spectrometry to determine their chemical nature. Polar compounds were detected by HPLC chromatography as shown for coronene (Fig. S1). No products were detected in non-inoculated control cultures.

The production of oxidized products was confirmed by the appearance of new bands of O–H and C–O groups in the FTIR-ATR spectra (Fig. S2). Pure coronene (Fig. S2a), benzo(g,h,i)perylene (Fig. S2c), and indeno(1,2,3,cd)pyrene (Fig. S2e) show characteristic bands at 3044–3024 (C–H stretching), 1610–1598 (C=C stretching), 1340–1318 (C–C stretching), and 840–675 cm−1 (out-of-plane C–H vibrations). C–C stretch vibrations produce generally intense peaks at 1569 and 1309 cm−1, while the peaks close to 1200 cm−1 are mainly contributed by C–H in-plane vibrations. These signals are associated to the presence of aromatic hydrocarbon. After biotransformation of coronene (Fig. S2b) and benzo(g,h,i)perylene (Fig. S2d), the appearance of weak bands in the range of 3730–3700 cm−1 and between 1400 and 1100 cm−1, which can be assigned at O–H stretching and in-plane O–H vibration, respectively. On the other hand, the biotransformed extract from indeno(1,2,3c,d)pyrene (Fig. S2f) showed new strong bands localized at 3350–3200 (O–H stretching), 1460–1360 (in-plane bend O–H vibrations) and 1050–950 cm−1 (C–O stretching), indicating that oxidation reaction occurs at higher extent on this HMW-PAH compound.

The reaction products were also analyzed by nano-LC high-resolution MS hybrid mass spectrometry. Hydroxycoronene (C24H12O, [M+H]+ m/z 317.09652 Da), dihydroxycoronene (C24H12O2, [M+H]+ m/z 333.09721 Da), and coronenodione (C24H10O2, [M+H]+ m/z 331.07612 Da) were detected from coronene metabolization (Fig. 1). In fungal cultures with benzo(g,h,i)perylene, the monohydroxy (C22H12O, [M+H]+ m/z 293.09609 Da), dihydroxy (C22H12O2, [M+H]+ m/z 309.09028 Da), monoketone (C22H11O, [M+H]+ m/z 292.08752 Da), and dione (C22H10O2, [M+H]+ m/z 307.07536 Da) derivatives were detected (Fig. 2). Finally, N. fischeri metabolization of indeno(1,2,3-cd)pyrene showed also hydroxyindeno(1,2,3-cd)pyrene (C22H12O, [M+H]+ m/z 293.09567 Da), dihydroxyindeno(1,2,3-cd)pyrene (C22H12O2, [M+H]+ m/z 309.09101 Da), indeno(1,2,3-cd)pyrenone C22H11O, [M+H]+ m/z 292.08758 Da), and indeno(1,2,3-cd)pyrenodione (C22H10O2, [M+H]+ m/z 307.07470 Da) (Fig. 3). All these metabolic products strongly suggest the participation of a cytochrome P450 monooxygenase. It is important to point out that the exact position of the hydroxyl and ketone substituents on the PAH structure has not been determined. Thus, Figs. 1, 2, and 3 are schematic representations of a possible isomeric structure.

Fig. 1
figure 1

Coronene metabolites identified by nano-LC-MS. The exact position of the hydroxyl and ketone substituents on the PAH structure has not been determined. Thus, the figure is a schematic representation of a possible isomeric structure

Full size image
Fig. 2
figure 2

Benzo(g,h,i,)perylene metabolites identified by nano-LC-MS. The exact position of the hydroxyl and ketone substituents on the PAH structure has not been determined. Thus, the figure is a schematic representation of a possible isomeric structure

Full size image
Fig. 3
figure 3

Indeno(1,2,3-cd)pyrene metabolites identified by nano-LC-MS. The exact position of the hydroxyl and ketone substituents on the PAH structure has not been determined. Thus, the figure is a schematic representation of a possible isomeric structure

Full size image

In vitro transformation (microsomal fraction)

Concentrated extracellular medium (100 times) by ultrafiltration (cutoff 10,000 Da) was unable to transform any of the HMW-PAHs tested. This extracellular extract showed extremely low laccase activity and no peroxidase activity (Uribe-Alvarez et al. 2011). Then, microsomal fractions were obtained from N. fischeri grown in two different media; potato dextrose broth (PDB) and Czapek, both supplemented with phenanthrene. The microsomes were then assayed for the NADH oxidation in the presence of HMW-PAHs. The microsomal fraction obtained from Czapek was able to oxidize NADH with rates of 25 nmol μg prot−1 min−1 in indeno(1,2,3-cd)pyrene, 20 nmol μg prot−1 min−1 in coronene, and 6 nmol μg prot−1 min−1 in benzo(g,h,i)perylene. No NADH oxidation could be detected with microsomes from PDB medium and in control experiments without the addition of PAH substrate. These results clearly indicate that in the presence of PAH, as sole carbon source, specific CYPs are induced and thus microsomes became able to transform HMW-PAHs. The CYP participation in the xenobiotics oxidation, including PAHs, has been well documented and the increased expression of CYP has been demonstrated in microsomal fractions from Aspergillus fumigatus cultures in the presence of xenobiotics (Baillie et al. 1996). Syed et al. (2010) performed a genome-wide CYP microarray screen of the white-rot fungus Ph. chrysosporium, and six PAH-inducible CYP genes were identified. All six CYP were cloned and expressed in Pichia pastoris in conjunction with a homologous P450 oxidoreductase. Each of the six recombinant CYP monooxygenases showed PAH-oxidizing activity albeit with varying substrate specificity toward PAHs from three to five rings. The same group demonstrated that the cloned CYP63A2 conferred the unique catalytic capability to oxidize the six-ring PAH benzo(g,h,i)perylene (Syed et al. 2013).

Microscopy studies

The participation of CYP in the HMW-PAHs and asphaltene metabolism should involve substrate cell internalization. Confocal microscopy studies were conducted to compare observations of N. fischeri grown in asphaltenes and glucose-peptone, respectively. Laser scanning confocal microscopy without any dye addition was used, taking advantage of the aromatic fluorescence property of asphaltenes. Under these conditions, confocal images showed an intense fluorescence in fungal hyphae grown with asphaltenes different to those grown in glucose-peptone condition that did not have any fluorescence (Fig. 4). This could be attributed to the accumulation of aromatic compounds inside the cell. Fungal hyphae were also stained with FM4-64 for membranous organelles, with BODIPY for lipid bodies, and with SYBR green for nuclei and mitochondria. As demonstrated by the confocal images in Fig. 4, the asphaltene fluorescence does not match with membranous organelles or with lipid bodies. The HMW-PAH internalization has been also confirmed in fungal cultures grown with coronene (Fig. 5).

Fig. 4
figure 4

Confocal images of the intracellular localization of asphaltenes. Membranous organelles are stained with FM4-64, lipid bodies were stained with BODIPY, and nuclei and mitochondria were stained with SYBR green (SYBRG). Scale bar = 10 μm

Full size image
Fig. 5
figure 5

Confocal images of the intracellular localization of coronene. a Fluorescence at 405 nm. b Merge of fluorescence and differential interference contrast (DIC). Scale bar = 10 μm

Full size image

The presence of PAH inside cells was also detected in Fusarium solani (Rafin el al. 2000; Verdin et al. 2005). BaP accumulation in lipid bodies in F. solani was later confirmed and a PAHs passive uptake mechanism was suggested (Verdin et al. 2005; Wu et al. 2009; Thion et al. 2012). These authors observed the same blue fluorescence that observed in Fig. 4. Fayeulle et al. (2014) also found BaP fluorescence inside the hyphae and claimed that internal lipid bodies constitute the intracellular storage sites for PAHs, but contrary to previous studies, they demonstrated an energy-dependent uptake process for PAHs.

Fungal hyphae from asphaltene and glucose cultures were observed by atomic force microscopy (Fig. 6). Morphological differences on the surface of N. fischeri were observed. The hypha of cells from glucose-peptone media showed to be smoother than hyphae grown in asphaltenes as carbon source. Also, larger protuberances on the surface of fungal mycelia from asphaltene cultures were detected. The roughness of each hyphae was also determined. The hyphae grown in asphaltenes showed a roughness average of 22.932 versus 0.112 nm observed in hyphae from glucose-peptone cultures. Differences in the hyphae size were also found. Hyphae from glucose-peptone were 10-μm width, while hyphae from asphaltenes cultures were only 5-μm width. Wu et al. (2009), recently found morphological differences on the surface of strain Aspergillus sp. BAP14 between the glucose and benzo(a)pyrene (BaP) treatment. Consistently, by AFM they observed that the hyphae grown in glucose were smoother than those from BaP treatment. In addition, they detected many larger protuberances in cells grown in BaP, which might refer to the lipid vesicles as observed by fluorescence microscope after Sudan III staining. This phenomenon had been also reported by Verdin et al. (2005) for F. solani cultures with benzo(a)perylene.

Fig. 6
figure 6

Atomic force microscopic (AFM) analysis showing morphological differences of Neosartorya fischeri hyphae grown in a asphaltenes and b glucose-peptone: 1 Hyphae, 2 Hyphae surface section

Full size image

Discussion

Neosartorya fischeri has been previously reported as able to grow with asphaltenes as carbon source (Uribe-Alvarez et al. 2011) and it has been estimated that at least 15.5 % of asphaltene were mineralized to CO2 in fungal cultures. Asphaltene biodegradation has been recently reviewed (Hernández-López et al. 2015a) and from literature, it is known that there are few microorganisms, both bacteria and fungi, able to metabolize this recalcitrant petroleum fraction. Asphaltene composition includes aromatics with very high molecular weight, which are considered recalcitrant to microbial attack. Recalcitrance of PAHs seems to increase directly with the molecular weight (Juhasz and Naidu 2000). Microbial metabolism of PAHs containing until three rings (i.e., naphthalene, phenanthrene, anthracene and fluorene) has been extensively studied; however, information about microbial degradation of those PAH’s containing more than six benzene rings is scarce (Juhasz and Naidu 2000). Benzo(a)pyrene biodegradation, containing five benzene rings, has been extensively studied (Cerniglia and Yang 1984; Verdin et al. 2005; Ullrich and Hofrichter 2007; Fayeulle et al. 2014). In our work, three HMW-PAH, two with six benzene rings and one with seven, have been tested for N. fischeri metabolization. The non-ligninolytic fungi N. fischeri is able to metabolize benzo(g,h,i)perylene, indeno(1,2,3-cd)pyrene, and coronene. Other non-ligninolytic fungi such as Cunninghamella elegans have been reported as able to metabolize PAHs that range in size from naphthalene to benzo(a)pyrene (two to five aromatic rings) (Sutherland 1992; Pothuluri et al. 1993; Cerniglia et al. 1994); but before this work, scarce information is available on the fungal transformation of HMW-PAHs with more than five aromatic rings. Transformation of HMW-PAHs of various ring sizes, including 6 rings, benzo(g,h,i)perylene was obtained using recombinantly expressed CYP63A2 (Syed et al. 2013). No activity toward the three-ring PAHs was found, suggested that this fungal CYP has a preference for HMW-PAHs. This selectivity could be related to the size of active-site cavity as suggested from modeling and docking estimations (Syed et al. 2013).

FTIR and nLC-MS analyses (Figs. 2, 3, and 4 and S2) clearly showed the production of hydroxylated derivatives from the N. fischeri metabolism on HMW-PAHs. Two mechanisms have been proposed for the incorporation of oxygen before aromatic ring cleavage; one involving dioxygenases, mainly in bacterial degradation, and monooxygenases for fungal PAH degradation. Several enzymatic systems could be involved in the fungal metabolism of PAHs, including intracellular cytochrome P450 and extracellular lignin peroxidase, manganese peroxidase, and laccase. The initial transformation of PAHs by non-basidiomycetes fungi results in the production of dihydrodiols, phenols, quinones, and dihydrodiol epoxides (Cerniglia and Yang 1984; Cerniglia 1997; Vazquez-Duhalt 1998; Haritash and Kaushik 2009).

N. fischeri, as a non-ligninolytic fungus, does not produce extracellular peroxidases and only a very low laccase-like activity has been detected (Uribe-Alvarez et al. 2011). Thus, it seems that a CYP enzyme is involved in the first oxidation transformation of HMW-PAHs. In addition to the chemical nature of products, this suggestion is supported by the HMW-PAH transformation by microsomal preparations from cultures in the presence of PAHs, while no activity was found with microsomes from cultures without PAH, suggesting the induction of specific CYPs by the presence of PAH.

Moreover, an analysis of microarray data from comparative expression performed of fungal cells grown either in asphaltenes or in glucose as a sole carbon source, which is available in the Gene Expression Omnibus (GEO) with the number GSE68146 (Hernández-López et al. b), shows three upregulated CYP monooxygenases genes and one flavin monooxygenase gene (Table 2). This microarray data showed that 287 genes were upregulated when fungi were growth in asphaltenes; meanwhile, 118 were downregulated. A number of 38 of the 287 upregulated genes encoded enzymes, from which 14 were oxidoreductases and 3 of them were identified as cytochrome P450 monooxygenases (CYP) and one as flavin binding monooxygenase that could be involved in the asphaltenes oxidation in N. fischeri (Table 2). The Blast2GO tool was used to assign function to each of the genes (Götz et al. 2008). The upregulated genes showed an induction between 22 and 27 times, and the same criterion was used for downregulated genes (>4 times). This was determined by calculating the logarithm base two of the ratio of expression between the two conditions. The conventional practice of setting arbitrary cutoff point is the changing twofold change in gene expression (Doddapaneni and Yadav 2005); however, in this study, we considered differential expression greater than four times.

Table 2 Upregulated monooxygenase genes identified from the analysis of differential microarray data of N. fischeri grown either in asphaltenes (Asph) or in glucose (Glu) as a sole carbon source
Full size table

CYP monooxygenases have been reported to be involved in the fungal biotransformation of dioxins, endocrine disruptors, pesticides, and PAHs (Masaphy et al. 1996; Jauregui et al. 2003; Subramanian and Yadav 2009; Kasai et al. 2010; Cerniglia and Sutherland 2010). Despite several decades of research demonstrating involvement of CYP monooxygenation reactions in fungal metabolism of PAHs, it is until recently that a specific CYP responsible for oxidation of these compounds has been described (Syed et al. 2013). From genomic P450 microarray screening, six PAH-responsive CYP genes inducible by PAHs were identified (Syed et al. 2010). Specifically, the role of CYP63A2 from Ph. chrysosporium on the transformation the six-ring PAH benzo(g,h,i)perylene was clearly demonstrated showing specificity for HMW-PAH (Syed et al. 2013).

Substrate-inducible CYP-mediated reactions have been reported in filamentous fungi during detoxification of environmental low and medium molecular weight PAHs (da Silva et al. 2004). In addition, the CYP monooxygenase participation in the transformation of PAHs has been described in fungi lacking extracellular oxidoreductases (Cerniglia and Sutherland 2010). The role of CYP has been also investigated in vivo and in vitro using CYP inhibition during phenanthrene metabolism in Ph. chrysosporium and Pestalotiopsis sp. Piperonyl butoxide, a well-known CYP inhibitor, can significantly inhibit the phenanthrene transformation to trans-9,10-dihydrodiol and asphaltene degradation indicating the important role of CYP catalyzed oxidation in the initial metabolism of PAH (Yanto and Tachibana 2014; Ning et al. 2010).

On the other hand, the flavin monooxygenases (FMOs) are a family of NADPH- and oxygen-dependent enzymes that readily N- and S-oxygenate a diverse group of xenobiotics and certain endogenous amines. Each isoform demonstrates species- and tissue-specific expression patterns as well as distinguishable but overlapping substrate specificities with other FMOs and cytochrome P450s (Furnes and Schlenk 2004).

The white-rot basidiomycete Ph. chrysosporium was investigated in its cellular responses against vanillin. Based upon a proteomic survey, it was demonstrated that two flavin-containing monooxygenases (PcFMO1 and PcFMO2) are translationally upregulated in response to exogenous addition of vanillin. PcFMO1 was cloned and heterologously expressed in Escherichia coli in order to elucidate their catalytic functions. Cloned bacteria showed activity against monocyclic phenols such as phenol, hydroquinone, and 4-chlorophenol. Interestingly, the product from hydroquinone was identified as 1,2,4-trihydroxybenzene, an important intermediate in a metabolic pathway of aromatic compounds in which the aromatic ring can be further cleaved by fungal dioxygenases for mineralization. The authors suggest that the ortho cleavage pathway of phenolic compounds would presumably be associated with PcFMO1 (Nakamura et al. 2012).

Our results strongly suggest that a CYP is involved in the metabolic oxidation of HMW-PAHs and asphaltenes. We have identified three CYP monooxygenases candidates that could be involved in the asphaltene oxidation by N. fischeri (Table 2). These monooxygenase genes showed a significant induction (between 8- and 16-fold) when the fungus was cultured in the presence of asphaltenes. The induction of CYPs encoding genes in response to the presence of different PAHs, and the role of CYP from Ph. chrysosporium on the PAH biotransformation, including six-ring PAH has been reported (Syed et al. 2010, 2011, 2013).

Finally, the CYP transformation of HMW-PAHs implicates an intracellular metabolism. Thus, cell uptake and transport mechanisms should be involved. Taking advantage of the intrinsic aromatic fluorescence of asphaltenes and HMW-PAHs, we performed laser scanning confocal microscopy, showing an intense fluorescence in fungal hyphae when grown in asphaltenes (Fig. 5) and in coronene (Fig. 6). This observation strongly suggests a fungal asphaltenes uptake. Verdin et al. (2005) observed the same blue fluorescence in F. solani grown with benzo(a)pyrene, and proposed a passive mechanism of uptake. On the other hand, Fayeulle et al. (2014) found benzo(a)pyrene fluorescence inside the hyphae, and suggested that the lipid bodies constitute the intracellular storage sites of PAHs with an energy-dependent uptake.

Thus, we can conclude so far that N. fischeri is able to growth in a medium containing HMW-PAHs (six and seven rings) as sole source of carbon and energy. The metabolites produced during the biodegradation include hydroxyl and ketone derivatives. In addition to the chemical nature of metabolic products, the involvement of CYP enzymatic system is supported by (i) the fact that the highly hydrophobic substrates are internalized in the fungal hyphae cells, (ii) at least three CYP genes are upregulated when the fungus is growing with aromatic hydrocarbons as sole source of carbon, and (iii) the microsomal preparations from cultures in the presence of PAH are able to oxidize HMW-PAHs.