|Home | About | Journals | Submit | Contact Us | Français|
Industrially produced carbon-based nanomaterials (CNM), including fullerenes and nanotubes, will be introduced into the environment in increasing amounts in the next decades. One likely environmental chemical transformation of C60 is oxidation to C60 fullerol through both abiotic- and biotic-mediated means. Unfortunately, knowledge of the environmental fate of oxidized CNM is lacking. This study used bulk and compound specific 13C stable isotope ratio mass spectrometry techniques and spectroradiometry analysis to examine the ability of two white rot basidiomycete fungi (Phlebia tremellosa and Trametes versicolor) to metabolize and degrade an oxygenated CNM, C60 fullerol. After 32 weeks of decay, both fungi were able to bleach and oxidize fullerol to CO2. Additionally, the fungi incorporated minor amounts of the fullerol carbon into lipid biomass. These findings are significant in that they represent the first report of direct biodegradation and utilization of any fullerene derivative and provide valuable information about the possible environmental fates of other CNM.
It is anticipated that industrially-produced carbon-based nanomaterials (CNM), including fullerene (C60) and its functionalized derivatives, will become widely distributed in the environment in this next century as the number of commercial applications of these products is expanding (1,2). Toxicological research on C60 and functionalized C60 has demonstrated a wide range of possible impacts to microbial and mammalian physiologies. There are a limited set of studies assessing the toxicity of polyhydroxylated fullerene, or C60 fullerols (3–5). For example, zebrafish embryos exposed to an aqueous solution (50 mg/L fullerols) exhibited no toxic effects (6); however, human dermal fibroblasts and liver carcinoma cells exhibited increased toxic effects as the hydroxyl group density on the fullerol decreased (4). Fullerols have also been shown to induce membrane damage in mammalian liver tissue through the production of reactive oxygen species under UV light (3).
The physicochemical properties of fullerenes and their derivatives is one of the main determinants of their potential environmental fate. For example, the ability of C60 to cluster into nanoparticles (nC60) in water greatly enhances their aqueous concentration above dispersed molecular C60 (6). Additionally, aqueous nC60 can be oxidized through exposure to ozone to chemical forms that include carbonyl, vinyl ether, and hydroxyl groups (7). The conversion of C60 to oxygenated forms such as fullerols results in a fundamental change to its potential environmental fate as fullerols are highly water soluble (8–10) and form reactive oxygen species upon exposure to light (11). The potential for enzymatic biodegradation of CNM to produce oxygenated forms has recently been demonstrated in experiments using horseradish peroxidase and nanotubes (12). Additionally, there is potential for the direct release of oxygenated CNM, given their commercial application in areas such as such as drug delivery (13–15), an industry that is anticipated to expand. A schematic of the potential means of converting fullerenes to fullerols in the environment and likely environmental fates is shown in Figure 1. Because of their high water solubility and highly reactive surface chemistry, it is likely that the mean residence time of oxygenated CNM would be dramatically lower than that of underivatized analogs in soils, sediments, or the aquatic systems.
It is important to note that C60 fullerol contains numerous isomers and that the type of C-O linkages can be in a variety of functional groups including sp3-C-OH, sp2-C-OH, hemiketals and carbonyls, influencing potential toxicity and environmental reactivity. Supporting Information Table S1 contains a review of reported FTIR analysis of commercially available and synthesized fullerols highlighting this point.
Given the great diversity of microbes capable of decomposing highly condensed aromatic structures (16–18), the biodegradation of oxygenated CNM is likely to be one of the more important mechanisms for environmental transformation. Within soil and sediment systems the basidiomycete fungi are likely to be important degraders of these compounds as they are capable of degrading a wide variety of aromatic species, such as polycyclic aromatic hydrocarbons (e.g. 19), lignin (20, 21), coal (22, 23), and biochar (24). Some white rot basidiomycete fungi, such as Trametes versicolor, can simultaneously degrade lignin and other aromatic compounds along with polysaccharides while others, such as Phlebia tremellosa, selectively decay lignin (20). White rot decay of such aromatic substances is facilitated by a variety of extracellular nonspecific enzymes, such as manganese peroxidase, lignin peroxidase, and laccase (25, 26) which not only oxidize but also promote polymerization. The latter possibility is an important point when considering the potential of chemical cross-linking of fullerols to soil and sediment organic matter (Figure 1). By their enzymatic action white rot basidiomycete fungi also bleach dark-colored aromatic compounds as has been illustrated in paper pulp processing (27, 28). It has also been noted that the presence of lignin and lignin-like compounds induces the production of peroxidases and speeds up the bleaching and delignification process (29).
In this study, we examine the ability of two species of white rot basidiomycete fungi to degrade and metabolize synthetic C60 fullerols (C60(OH)19–27), either directly or as a result of cometabolic decay, in the presence of growth media and wood wafers, in order to further elucidate the fate of these important nanomaterials in natural environments.
The fungi Trametes versicolor (strain MAD 697-R) and Phlebia tremellosa (strain PRL 2845), maintained in the Forest Pathology and Wood Microbiology culture collection at the University of Minnesota, St. Paul, Minnesota, were chosen to test the ability of white rot basidiomycetes to degrade and metabolize fullerols. For each inoculation, approximately 15 mL of autoclaved 2% malt extract/agar media was added to a 50 mL clear glass jar (Scientific Specialties, Inc, Lodi, California) with hard plastic screw cap closures and a Teflon coated rubber septum. The influence of an autoclaved birch wood wafer was tested in half of the experiments, as wood can act as a trigger for the production of a wide array of ligninase enzymes as compared to growth in only culture media (29, 30). Wood (~350 mg dry weight) and fullerol (~35 mg at 22.32 atom % 13C, synthesized at TDA Research as described in Supporting Information) were added on top of a glass fiber filter disk (GF/F) which rested on top of the congealed media. The C60OxHy, determined to contain 19–27 oxygen (C60(OH)19–27) atoms by solid state 13CNMR as shown in the Supporting Information Table S2, was placed either on the birch wafer (media + wood + fullerol experiments or MWF) or on top of the glass fiber filters (media + fullerol experiments or MF). The resultant percentage of fullerol C to total C in each jar was about 2.4% for MWF experiments and about 5.7% for the MF experiments. Fungal plugs were then added to all jars except controls such that they touched the fullerol and/or wood wafers as well as media. Experiments without fullerol and just media (M) and media with wood (MW) we also conducted. Three replicates of each experiment type (M, MW, MF, and MWF) for both fungal species (T. versicolor and P. tremellosa), as well as three replicates of MW and MWF for no fungal controls, were inoculated. One of the T. versicolor MF replicates, however, failed to grow and was not considered in the final analysis. The fungi were allowed to colonize the fullerol for 32 weeks. Experiments were kept in the dark to limit light exposure and reactor lids were kept closed and opened only in a clean air bench to provide aeration every three weeks and allow inspection to ensure no visible bacterial colonies were present. Oxygen concentrations were not monitored but opening the reactors at this interval appeared to be sufficient to maintain growth rates, as determined from previous experiments, and also minimized possible bacterial contamination. At 16 weeks the headspace was sampled for CO2 and a portion of the fungal hyphae were harvested for compound-specific stable carbon isotope analysis of fungal lipids (see below).
The dark colored C60(OH)19–27 is very hygroscopic and immediately upon its addition to the inoculation jars it began to dissolve. Within two weeks of inoculation, fullerol appeared to be uniformly distributed throughout the growth media in each experiment.
Spectral reflectance measurements (Analytical Spectral Device [ASD] Fieldspec 3 spectroradiometer) of inoculation media contents were taken after 32 weeks in order to assess the potential chemical alteration of the fullerols (details concerning spectral analysis found in Supporting Information). Chemical change was manifested primarily in the 350 – 900nm range. Two significant features were analyzed in that wavelength range in each sample (see Figure 2): 1) an absorbance minimum feature with differing wavelength values and depth for each sample, and 2) an inflection point leading out of the absorbance feature back up to highly reflective values for higher wavelengths. Continuum removal data manipulation (described in Supporting Information) was performed on the raw spectral data, which allowed these inflection points to be analyzed as maxima on the absorbance spectra (31).
Hyphae were sampled from the jars at 16 weeks and base extracted to remove and concentrate fatty acid lipids according to procedures modified from Wakeham and Pease (32). For gas chromatographic analysis fatty acids were converted to trimethylsiloxyl derivatives and their structure analyzed using an HP 5890 gas chromatograph, containing a 5% phenyl polymethylsiloxane capillary column (30m, 0.25 mm i.d. HP-5) interfaced with an HP 5971 quadropole mass spectrometer. The GC oven was programmed from 40°C to 260°C at 7°C per minute and maintained for 6 minutes. Commercially available standards were used for reference. Both fungal species contain abundant 9,12-octadecadienoic acid, C18:2 (33), and this was chosen as a marker to assess 13C uptake into fungal biomass using analysis with compound specific gas chromatography – combustion – isotope ratio mass spectrometry (GC-C-IRMS).
The fate of fullerol carbon was tracked using stable carbon isotope ratio monitoring of headspace CO2, residual media and wood after 32 weeks decay, and the isolated fatty acid C18:2 from the hyphae at the Purdue Stable Isotope (PSI) Facility. The fate of the 13C-fullerol is discussed in terms of the % of C in the CO2 and C18:2 derived from the 13C label in the MF and MWF experiments (equation 2-shown for C18:2 MF-M comparison). Additionally, the percent of initial 13C-fullerol C (IFC) remaining in the media or wood of each experiment after 32 weeks of decay was calculated by dividing the mass of fullerol C remaining in the media after 32 weeks by the mass of the initial fullerol C added (equation 3-shown for media). Details on the derivation of these equations can be found in Supporting Information.
Analysis of 13C content of individual fatty acids was performed by GC-C-IRMS (34) on an Agilent 6890 GC interfaced to a PDZ-Europa 20/20 (SerCon Ltd., Crewe, United Kingdom) IRMS via a microcombustion furnace. The GC operating conditions were the same as described for structural analysis. Co-injection of eicosane as well as CO2 pulses of known isotopic composition during the GC run were used to determine isotope drift and 13C composition. The 13C content of CO2 produced by the fungi was measured at the end of 16 weeks at the time of sampling the hyphae. Headspace gases were removed in 500 µL aliquots from the inoculation jars with a gastight syringe and injected into helium flushed 12 mL exetainer vials. The stable carbon isotope composition was determined using a SerCon (Crewe, United Kingdom) Cryoprep TG2 interfaced to a PDZ Europa 20/20 IRMS. Vials of CO2 of known isotope composition were used as calibrants. Comparison of the 13C content of CO2 allowed for calculation of the fraction of CO2 carbon that came from fullerol C as shown in equation 2. Wood and media from the jars were harvested after 32 weeks decay, dried at 60°C and ground for isotope analysis using a SerCon EA-CN 1 elemental analyzer interfaced to a PDZ-Europa 20/20 IRMS.
Differences in the extent of fullerol oxidation and incorporation of fullerol carbon into lipids were evaluated using two-tailed Student’s t-tests.
The two fungi used in the experiment have the capability to produce oxidative enzymes such as lignin peroxidase, manganese peroxidase, and laccase (26, 35, 36), which are known to oxidize aromatic plant-derived substances such as lignin (20, 25) as well as polycyclic aromatic substances (19, 37, 38). We anticipated that similar activity would be expressed in these experiments, and that fullerols would be oxidized and hydroxylated, causing a bleaching of the dark brown fullerol compound (39). By 32 weeks this bleaching was apparent both visually (see Supporting Information Figure S1) and by the total reflectance measurements (see Figure 3). In fact, even within three weeks of starting the fungal inoculations, it was noted that in the inoculated MWF experiments the media, darkened by fullerol, was lighter in color than the fullerol-containing media without fungi. Figure 3a–b illustrates the ability of these fungi to bleach the fullerol in media both in the presence of wood (3a) and without wood (3b). For P. tremellosa MW and T. versicolor M experiments (i.e. no fullerol added) samples, absorbance regions were shifted below 350 nm due to bleaching of the nutrients in the media, which is below the range of the spectroradiometer. These points are therefore not plotted in Figure 3.
In all experiments, both fungal species were able to bleach the fullerol, indicated by the shift to lower wavelength values for both the maximum and absorbance of the media after 32 weeks decay. Whereas the presence of the wood wafer was expected to increase the capacity of both fungi to bleach fullerol, only P. tremellosa exhibited enhanced bleaching in the presence of wood. This may be a function of the fungi responding differently to the nutrient levels provided by the growth media. By extrapolation, fullerols in the environment might be expected to exhibit greater alteration by some basidiomycete fungi when vascular plant carbon, most likely derived from lignin residues, is present. There are reported observations of enhanced oxidative decay in the presence of lignin (29), as well as polycyclic aromatic hydrocarbons and chlorinated phenols by white rot fungi (40, 41). During bleaching and delignification of wood pulps, a one electron oxidation of aromatics within the lignin is the first cause decolorization (29, 42), and it is likely a similar phenomenon is occurring with the fullerol in the present experiments, with initial attack first on the hydroxyl functionality. In liquid culture agitated systems, the onset of lignin bleaching is reported to occur in one to two weeks (36), and solid-state cultures take longer for delignification to occur (29). A similar time frame for bleaching in our solid media samples was observed—after two weeks of fungal growth, the media had started to lighten.
Although the chemical structure of the altered fullerol cannot be directly assessed by spectroradiometry, the spectral shift seen in these experiments must be due to at least one of two phenomena: 1) removal of the sp2-carbon on the fullerol by oxidation and cage rupture, resulting in lower molecular weight metabolites, or 2) increased hydroxylation of the sp2-carbon, which has been shown to lighten the color of aqueous fullerol solutions (39). If the cage structure of the fullerol is oxidized into smaller fragments there exists the possibility that the fungi, or possibly other microbes in the natural environment, could utilize the fullerol carbon for biomass production or energy. The 13C content of the CO2 produced and the fungal lipids was used to test these possibilities (below).
Metabolized fullerol in this system has three possible fates: uptake into fungal biomass, oxidation to CO2, or build-up of structurally- and chemically-altered products. Figure 4a shows the percentage of the C in the target lipid (C18:2) derived from the 13C fullerol. The data demonstrate these fungi are incorporating the fullerol carbon into fungal biomass, although only to a small degree. The fungus P. tremellosa incorporated a relatively greater percentage of 13C fullerol carbon into C18:2 than T. versicolor for both the MF (p=0.018) and MWF (p=0.002) experiments. For P. tremellosa, the C18:2 from the MWF are comprised of less 13C-fullerol C (0.078 ± 0.090%) than the MF experiments (0.14 ± 0.09%) (p=0.022) while T. versicolor exhibited no difference between experiments with and without wood (~0.04%, p=0.987). These differences may be a combination of different enzyme activities and the fact that the MWF experiments had more non-fullerol carbon available to the fungi. Considering this data and the relative loading of fullerol C in the decay experiments (ranging from 2.4 to 5.7% of total C) we can estimate that the fungi have a 20 fold (for the P. tremellosa MF experiment) to 142 fold (for the T. versicolor MWF experiment) preference for uptake of media or wood carbon over fullerol carbon. For the fullerol C to be taken up into this lipid the fullerol cage structure must be first broken down into acetate and reassembled to palmitic acid (C16:0), which is the starting material for longer chain fatty acids (43). It is also possible that other anabolic products, e.g. sugars, proteins, etc, were formed in part from the fullerol C, although they were not analyzed. The minor capability of the fungi to convert the fullerol carbon to biomass in the presence of media is consistent with other findings on the activity of white rot fungi on lignin (44) and coal (22, 24, 45). Therefore, the bleaching of the fullerols (Figure 2 and Figure S1) observed in this study must be, in part, a function of rupture of the cage and not simply increased hydroxylation (39). To the best of our knowledge, this is the first evidence for the biodegradability of these structures as well as the first evidence for their utilization in anabolic processes. This has important implications for the production of highly water soluble molecular fragments which may have an enhanced capability for further microbial consumption (46).
Sixteen weeks after the fungal inoculation began it was evident that some of the cage structure was being oxidized to CO2 by the 13C content of the CO2 formed by the fungi (Figure 4b), which is consistent with the observed uptake of fullerol carbon into the fungal lipids. The calculated percentages of fullerol 13C contributing to the CO2 range from 1.35 ± 0.01% (T. versicolor) to 2.86 ± 0.78% (P. tremellosa) for MF experiments and 0.924 ± 0.516% (T. versicolor) to 1.36 ± 0.73% (P. tremellosa) for MWF experiments. P. tremellosa appears to be more aggressive towards oxidation of the fullerol cage carbon to CO2 than T. versicolor (p=0.081) for MWF experiments but not for the MF experiments. The greater capability of P. tremellosa to decompose as well as bleach the fullerol structure is consistent with other findings on the aggressive activity of this particular fungus on natural wood lignin (20) and also on synthetic lignin compounds (47). In contrast to the findings for the large discrimination against fullerol C uptake into lipids, the fungi show a lower preference for the media C over that of fullerol C when it comes to oxidation to CO2. Specifically, P. tremellosa in the media experiment was only 2.0 times as likely to oxidize media C as it was fullerol C to CO2 while T. versicolor, in its media fullerol decay experiment, exhibited the maximum discrimination against fullerol of 4.2 times. Carbon dioxide production from mineralization of substrate material by white rot fungi has also been observed during degradation of polycyclic aromatic hydrocarbons such as phenanthrene and pyrene (48), pentachlorophenol (49), and also of synthetic lignins (47).
The primary fate of the fullerol carbon after 32 weeks decay, even though the spectral observations indicate substantial chemical bleaching, was as altered metabolites remaining in the media. Figure 5 shows distribution of the fullerol C remaining in these measured forms in the reaction jars at the end of the experiment. Among the replicates there was a wide range in the percentage of fullerol C remaining in the media but, in general, the two fungi showed similar responses among experiments with and without wood. Specifically, T. versicolor ranged from 61.1% to 96.1% initial fullerol C remaining dissolved in the media for the MF experiment and 54.4% to 74.1% for the MWF experiment while P. tremellosa ranged from 67.1% to 71.8% for MF and 63.5% to 78.7% for MWF. The remainder of the 13C label, not accounted for in Figure 5, must be associated with the CO2 loss and fungal hyphae produced, but unfortunately, neither was quantified for mass.
In general, the presence of wood in the experiments did not exhibit an influence on the amount of fullerol carbon remaining in the media as metabolites, although this was anticipated based upon known activity of fungi (29). The amount of the original 13C-fullerol C remaining on the wood residue was always a minor component, <2.2% for all samples which was a function of both low wood recoveries because of enhanced decay, and dissolution of fullerols through the wood to the media. The no fungus fullerol control experiments indicate that although 100% of the fullerol was added to the top of the wood only 2.1% remained in the wood structure by 32 weeks. The remainder of the fullerol carbon accumulated in the media once it had diffused through the system.
This study provides the first evidence of fullerol biodegradation and utilization. Both species of white rot fungi under these conditions could bleach fullerol as well as oxidize a small portion of it to CO2, and incorporate it into fungal biomass. For P. tremellosa, the presence of wood appeared to enhance its ability to bleach the fullerol but not to decompose it to CO2 nor incorporate it into biomass. The presence of wood for T. versicolor appeared to have little influence on decay dynamics. We can anticipate a range of activities among different species in nature and under different nutrient conditions but it is evident that if fullerols are produced in or released to the environment they have a great likelihood of being chemically altered by white rot fungi. Certainly further study is needed to determine the chemical structures of the bleached fullerol products as these will have a range of structural-dependent environmental fates. Additionally, fullerol degradation studies with natural soil communities and under nutrient limiting conditions would help determine the likely environmental fate of such compounds and inform future regulatory efforts.
Within the supporting information are included the methodological details of the fullerol synthesis used in this study and its analysis by solid state 13C-NMR, spectroradiometry, and stable isotope measurements. We also include a table summarizing published FTIR spectral data on fullerols as well as table comparing the 13C-NMR of the fullerol synthesized for this study and a commercially available fullerol. Additionally, a color figure illustrating the visual and spectroscopic evidence of fungal bleaching of fullerols from the inoculation experiments is provided. This information is available free of charge via the Internet at http://pubs.acs.org.
The authors would like to thank Benjamin Held and Joel Jurgens at the University of Minnesota for their help with the fungal laboratory experiments, David Gamblin and Sergey Oleynik at Purdue University for help on stable isotope measurements and lipids analysis, and Julie Petersen at TDA research for assistance with the fullerol synthesis. The authors also acknowledge support from the National Science Foundation (NSF) under Award EEC-0404006, the United States Environmental Protection Agency (EPA) under STAR Grant, Award RD-83172001-0, and the National Institutes of Health (NIH) under Grant R01EB000703.