• The mechanism behind the 13C enrichment of fungi relative to plant materials is unclear and constrains the use of stable isotopes in studies of the carbon cycle in soils.
• Here, we examined whether isotopic fractionation during respiration contributes to this pattern by comparing δ13C signatures of respired CO2, sporocarps and their associated plant materials, from 16 species of ectomycorrhizal or saprotrophic fungi collected in a Norway spruce forest.
• The isotopic composition of respired CO2 and sporocarps was positively correlated. The differences in δ13C between CO2 and sporocarps were generally small, < ±1‰ in nine out of 16 species, and the average shift for all investigated species was 0.04‰. However, when fungal groups were analysed separately, three out of six species of ectomycorrhizal basidiomycetes respired 13C-enriched CO2 (up to 1.6‰), whereas three out of five species of polypores respired 13C-depleted CO2 (up to 1.7‰; P < 0.05). The CO2 and sporocarps were always 13C-enriched compared with wood, litter or roots.
• Loss of 13C-depleted CO2 may have enriched some species in 13C. However, that the CO2 was consistently 13C-enriched compared with plant materials implies that other processes must be found to explain the consistent 13C-enrichment of fungal biomass compared with plant materials.
It has been suggested that the 13C enrichment of ectomycorrhizal fungi relative to the host leaf is the result of isotopic fractionation during synthesis of compounds in host plants and a subsequent transfer of 13C-enriched carbon to the fungal symbiont (Hobbie & Colpaert, 2004). Similarly, in wood saprotrophs, the 13C enrichment has been proposed to derive partly from a preferential use of 13C-enriched wood carbohydrates (Hobbie, 2005). Wood carbohydrates have c. 1–2‰ higher δ13C values compared with bulk wood in C3 plants (Benner et al, 1987; Gleixner et al., 1993). However, other internal metabolic processes must also contribute to the 13C enrichment since wood saprotrophs have been reported to have c. 2‰ higher δ13C values than wood cellulose (Gleixner et al., 1993). Recently, fungal mycelia grown in pure culture showed a dynamic 13C enrichment of fungal tissue compared with substrate that was dependent on growth stage, substrate and species (Henn et al., 2002; Kohzu et al., 2005).
The rule of isotopic mass balance dictates that any 13C enrichment of fungal biomass relative to its presumed substrate must be balanced by either a loss of relatively 13C-depleted carbon or an uptake of relatively 13C-enriched carbon from an unknown C source. In the first case, 13C-depleted carbon could potentially be lost in respiration as CO2 or by exudation, for example, as organic acids. This loss must be preceded by an isotopic fractionation at metabolic branch points (Gleixner et al., 1993; Schmidt & Gleixner, 1998), producing 13C-depleted compounds used in respiration or exudation, and 13C-enriched structural compounds, such as chitin. Alternatively, the loss of 13C-depleted CO2 could be the result of isotopic fractionation during decarboxylation reactions. In the second case, relatively 13C-enriched organic carbon could potentially be supplied by the plant or be preferentially taken up from dead organic matter. In ectomycorrhizal fungi, C delivered from the host plant is the main C source. Alternatively, it has been hypothesized that anaplerotic fixation of 13C-enriched CO2 increases the δ13C of soil microbes by 1–1.5‰, assuming 5% of microbial cellular C is derived from 13C-enriched soil CO2 (Ehleringer et al., 2000).
At present there are only a few reports in which the δ13C values from fungal respiration have been compared with the associated biomass, but these studies typically consider only one or two saprotrophic species, and there are, to our knowledge, no reports from ectomycorrhizal fungi. In an early study, the CO2 respired by the soil-dwelling saprotrophic filamentous fungus Fusarium roseum, grown in liquid culture, was 1–2‰ higher in δ13C than the substrates glucose and palmitic acid (Jacobson et al., 1970). Likewise, the CO2 respired by the yeast fungus (Candidus lipolytica) was enriched by 2‰ relative to the biomass and the substrate glucose (Zyakun, 1996), while the CO2 respired by the wood saprotroph Cryptoporus volvatus and the litter saprotroph Marasmius androsaceus was depleted by 0–1‰ compared with the biomass and the substrate, 10% malt extract and 90% C4 sucrose (Henn & Chapela, 2000). By contrast, the wood saprotroph Trametes versicolor grown on wood produced CO2 that was c. 3‰ depleted compared with the fungal biomass, resembling the values of the wood it was growing on (Kohzu et al., 1999).
In this study, our aim was to test whether 13C discrimination during fungal respiration contributes to the 13C enrichment of the fungal biomass relative to its substrate. We compared the δ13C of CO2 respired by fungal sporocarps with the δ13C and δ15N of sporocarps and their associated plant materials collected in a mature northern spruce forest. In total, 16 species of both ectomycorrhizal and saprotrophic fungi were investigated.
Material and Methods
Fungal sporocarps were sampled from a 70-yr-old, 22-m-tall, Norway spruce (Picea abies (L.) Karst.) forest stand, established on formerly cultivated land close to Brevens bruk, southern Sweden (59°00′N, 15°35′E, 125 m above sea level). The soil consists of a thin litter layer on top of a mineral soil, which is a washed till dominated by sand (90%), and is an entisol (psamment) according to FAO nomenclature. The mean annual temperature for the site is 5.8°C and the annual precipitation is 550 mm (for further details of the site, see Ekblad et al., 2005; Boström et al., 2007). Sporocarps were also sampled in an adjacent Norway spruce stand (70-yr-old) that was similar except for the fact that the forest ground was partly covered by mosses (Bryophyta) and a sparse field layer of dwarf shrubs (Vaccinium myrtillus L. and V. vitis-idaea L.).
Sporocarps were sampled twice in July and twice in December 2003, and six times in September 2004 (Table 1). Sporocarps were sampled from the soil surface, wood and litter. Below-ground sporocarps of Elaphomyces muricatus were sampled by raking the upper 10 cm of soil. The distance between each sample was generally > 10 m, to minimize the risk of sampling the same individual mycelium, but occasionally, several sporocarps of the same species were sampled from a single log (Table 1). Only apparently healthy sporocarps were sampled. Generally a single sporocarp was sampled from each spot and regarded as a sample, but to get enough material for analysis, 50–200 sporocarps of Micromphale perforans, three sporocarps of Lycoperdon perlatum, and two–five sporocarps of E. muricatus were lumped together to represent one sample.
Table 1. Stable isotope ratios (mean ± SE) of ectomycorrhizal and saprotrophic sporocarps and respired CO2
Date of sampling
Root, wood, litter
δ13C of CO2 respired by sporocarp is significantly different from δ13C of sporocarp material (95% CI).
Values for fine roots sampled 2003 and 2004 are from A. Ekblad (unpublished; see the ‘Materials and Methods’ section for details).
Wood and litter material were sampled adjacent to the saprotrophic sporocarps, newly fallen leaf litter from the ground, and discs of wood from logs or tree stumps. Substratum was not sampled for L. perlatum. Living fine roots (< 2 mm) were sampled with a soil corer (inner diameter, 50 mm) down to 10 cm depth on 17 occasions from May to September in 2003 and on four occasions from May to September in 2004 (A. Ekblad, unpublished). Most of the root tips were ectomycorrhizal as judged from ocular estimates and chitin analyses. Mycelia were harvested in September 2004 from hyphal ingrowth nylon mesh bags (Wallander et al., 2001) that were inserted at 5 and 10 cm soil depth in September 2003 (data from Boström et al., 2007).
The CO2 respired by a fungal sporocarp was generally collected in the forest c. 15 min after excision of the sporocarp. But sporocarps of E. muricatus and wood saprotrophic polypores sampled in 2003 were brought to the laboratory (20°C) for collection of CO2, within 5 h after excision. Generally, whole sporocarps were analysed for isotopic composition of solid matter and CO2, but large sporocarps had to be cut into smaller pieces to fit into the respiration chamber. Pieces were cut from the top to the base of fleshy sporocarps, or from the present year's pore layer in wood saprotrophic polypores that were several years old. To sample the respired CO2, sporocarp(s) or a piece of sporocarp was enclosed in a 1 l opaque PVC jar, and gas samples were taken at intervals during 15–30 min by pumping the air in the jar into 12 ml gas vials prefilled with CO2-free air, via a sampling loop as described by Boström et al. (2007). To test that the δ13C signature of the CO2 produced did not change during the storage of sporocarps, CO2 was repeatedly collected from a sample of L. perlatum (three sporocarps) and a sample of Fomes fomentarius over 3 h. To clarify if the isotopic signature was different between pore layers from different years, the respired δ13C signature was determined from three successive pore layers in a sporocarp of Fomitopsis pinicola. The sporocarps were placed in liquid nitrogen directly after collection of CO2. Both sporocarps and substrate were stored in a freezer (–20°C) until freeze-dried and ground to fine powder in a ball mill for isotopic analysis.
Taxonomic identification and categorization of fungi by functional groups was mainly conducted after Hallingbäck & Aronsson (1998), and authority of species is given in Table 1. All collected fungal species except the ascomycete E. muricatus belong to the basidiomycetes. Of the 16 fungal species examined, seven were ectomycorrhizal fungi, two were litter saprotrophs, and seven were wood saprotrophs, of which two were brown-rot fungi and three were white-rot fungi (Table 1). Sporocarps of ectomycorrhizal E. muricatus and wood saprotrophic Hypholoma capnoides were categorized into three different growth stages: (i) small undeveloped sporocarps with no visible spores; (ii) more developed sporocarps containing spores; and (iii) fully developed sporocarps with spores in abundance. Altogether, 89 fungal samples were taken for determination of δ13C of respired CO2, and δ13C and δ15N of solid sporocarps.
The C and nitrogen concentrations and isotopic composition (δ13C or δ15N = ((Rsample – Rstandard)/Rstandard) × 1000 (‰), and R is the molar ratio 13C/12C or 15N/15N) of the fine ground material was determined as described by Boström et al. (2007). The standard deviation of 10 replicated samples was ≤ 0.09‰ for δ13C and ≤ 0.3‰ for the δ15N.
The CO2 concentrations and 13C abundance of the CO2 in the gas samples was determined within less than 2 d as described by Ekblad et al. (2005). The standard deviation of 10 replicated samples was ≤ 0.08‰ for δ13C and ≤ 2% for the CO2 concentration. The rate of evolution and the isotopic composition of the CO2 from the fungi were calculated as previously described (Ekblad & Högberg, 2000).
In the text, the mean ± 95% confidence intervals of each species are based on samples taken on a single day for most species, but samples were collected on two or three occasions for some species. Mean and range values for functional groups (mycorrhizal, wood saprotrophs) are based on mean values for species. Linear regression analyses were carried out in SigmaPlot 8.0 (Systat Software GmBH, Erkrath, Germany).
δ13C and δ15N of sporocarps
When plotting the δ13C and δ15N values of sporocarps against each other, the ectomycorrhizal fungi and the saprotrophs fell into clearly distinct groups, but separation of litter and wood saprotrophs was not possible (Fig. 1).
The mean δ13C value of the ectomycorrhizal sporocarps was –26.0 ± 0.7‰, (range –27.1 to –25.0‰), with the lowest value in Gomphidius glutinosus and the highest value in Hygrophorus piceae (Table 1). By contrast, the saprotrophs showed a higher mean δ13C value of –23.6 ± 0.8‰ (range –25.1 to –22.4‰), with the lowest value in Trametes hirsuta and the highest value in H. capnoides and F. fomentarius (Table 1). No difference in δ13C value was found between white- and brown-rot wood saprotrophs (Table 1).
The δ13C values of saprotrophic sporocarps were positively correlated (R2 = 0.91, P < 0.001, n = 8 spp.) to the δ13C values of their substrate, wood or litter (Fig. 2). Saprotrophic sporocarps were on average 3.0 ± 0.3‰ enriched in 13C relative to wood or litter, ranging from 2.6‰ in the Pholiota alnicola to 3.7‰ in the H. capnoides. In ectomycorrhizal fungi the sporocarps were 13C-enriched by between 1.2‰ and 3.4‰ relative to the fine roots (Table 1). The smallest isotopic shift was found in E. muricatus and the largest shift in H. piceae.
The mean δ15N value of the ectomycorrhizal sporocarps was 6.3 ± 1.8‰ (range 3.5–8.6‰), with the lowest value in H. piceae and the highest value in G. glutinosus (Table 1). By contrast, the saprotrophs showed a lower mean δ15N value of 0.8 ± 0.7‰ (range –2.6–3.8‰), with the lowest value in M. perforans and the highest value in P. alnicola (Table 1). That the δ15N in a single species can vary a lot was found in M. perforans, showing δ15N values from –4.5‰ to 1.0‰ (Table 1). The wood saprotrophic sporocarp was, on average, 2.3 ± 1.9‰ enriched in δ15N relative to wood, whereas the litter saprotroph M. perforans showed a similar value to the litter (data not shown).
δ13C of respired CO2
When plotting the δ13C of CO2 respired by sporocarps against the δ13C of sporocarps dry matter, the ectomycorrhizal fungi and the wood saprotrophs fell into clearly distinct groups, but the two litter saprotrophs fell within the other two groups (Fig. 3). The mean δ13C value of the CO2 respired by ectomycorrhizal fungi was –25.3 ± 1.4‰ (range –27.7 to –23.6‰), with the lowest value in E. muricatus and the highest value in H. piceae (Table 1). The saprotrophs showed a mean δ13C value of –24.0 ± 0.9‰ (range –26.4 to –22.3‰), with the lowest value in T. hirsuta and the highest value in H. capnoides (Table 1).
The respired δ13C was positively correlated to the δ13C values of sporocarps, both when ectomycorrhizal fungi (y = 1.81x + 21.84, R2 = 0.88, P = 0.002, n = 7) and wood saprotrophs (y = 1.03x + 0.06, R2 = 0.61, P = 0.04, n = 7) were treated separately, as well as when all species were taken together (y = 0.71x – 7.19, R2 = 0.57, P < 0.001, n = 16, Fig. 3). The differences in δ13C between the CO2 and the sporocarps were generally small, < ±1‰ in nine out of 16 species, and the average isotopic shift (0.04 ± 0.56‰) for all species taken together was not significantly different from zero. However, when analysing the fungal groups separately, a slightly different pattern emerged. In ectomycorrhizal basidiomycetes, the CO2 was 13C-enriched compared with the sporocarps by up to 1.6‰ in three (Cantharellus cibarius, Paxillus involutus and H. piceae) out of six species (P < 0.05; Table 1). By contrast, the CO2 was 13C-depleted by up to 1.7‰ in three (F. fomentarius, Ganoderma applanatum and T. hirsuta) out of five species of wood saprotrophic polypores. The CO2 was slightly 13C-depleted (0.7‰) in the ectomycorrhizal E. muricatus, which was exceptional in several ways: it was the only ascomycete, the only hypogeous fungus and the only ectomycorrhizal fungus that was sampled in December. The polypores and E. muricatus (long-lived sporocarps) were separated from all other fungal species (short-lived fleshy sporocarps), when plotting the δ13C differences between the respired CO2 and the sporocarps against the respiration rate (Fig. 4).
Despite the fact that the CO2 produced by polypores was 13C-depleted compared with the sporocarps, the CO2 was 13C-enriched relative to plant materials in all fungal species. Thus, in saprotrophs, the CO2 was 13C-enriched relative to wood or litter by between 1.1 and 3.8‰, with the lowest value in G. applanatum and the highest value in H. capnoides (Table 1; Fig. 2). In ectomycorrhizal fungi, the CO2 was 13C-enriched by between 0.4 and 4.8‰ relative to the fine roots. The smallest isotopic shift was found in E. muricatus and the largest shift in H. piceae (Table 1).
Effects of sporocarp growth stage, age and storage on isotopic compositions
Sporocarp growth stage did not affect the δ13C or the δ15N of the sporocarps, or the δ13C of the respired CO2 of the two species studied (Fig. 5); nor did the age of the pore layer of F. pinicola affect the δ13C values of the sporocarps or their respired CO2. The δ13C values for sporocarp materials of three successive pore layers were –22.3‰ for the current year, –22.9‰ for 1 yr old, and –23.0‰ for 2 yr old. The values for the CO2 were –22.2 ± 0.9‰ for the present year, –21.6 ± 0.9‰ for 1 yr old, and –22.0 ± 2.1‰ for 2 yr old.
Storage of sporocarps did not affect the δ13C of the respired CO2, as found from 3 h of repeated CO2 sampling from F. fomentarius (range –24.1 to –23.2‰) and L. perlatum (range –24.5 to –23.1‰; Fig. 6).
The ectomycorrhizal sporocarp respiration rate ranged from 17 to 222 µmol C g−1 h−1, with the lowest value in E. muricatus at 20°C and the highest value in H. piceae at 14°C. A smaller range in respiration rate was found in saprotrophic fungi, showing values from 34 to 139 µmol C g−1 DW h−1, with the lowest value in G. applanatum and F. pinicola, both sampled at 20°C, and the highest value in H. capnoides, sampled at 11–13°C (Fig. 4). The respiration rate was positively correlated to the water concentration in sporocarps of both ectomycorrhizal fungi (R2 = 0.84, P = 0.004, n = 7 spp.) and saprotrophs (R2 = 0.72, P = 0.008, n = 8 spp.) (data not shown). The sporocarp respiration rate did not correlate with the size of the sporocarps. However, in a sporocarp of the polypore F. pinicola several years old, the respiration rate decreased with increasing age of three successive pore layers (present year, 25.2 ± 1.4 µmol C g−1 DW h−1; 1 yr old, 21.0 ± 3.6 µmol C g−1 DW h−1; and 2 yr old. 13.4 ± 1.4 µmol C g−1 DW h−1).
The differences in δ13C values between the CO2 and the sporocarps were not consistent; the CO2 respired by some saprotrophic species was 13C-depleted compared with the sporocarps, while that of some ectomycorrhizal species was 13C-enriched (Table 1). The range and inconsistency of the δ13C data in the present study compare with the few published reports known to us (Zyakun, 1996; Kohzu et al., 1999; Henn & Chapela, 2000). Thus, respiration of 13C-depleted CO2 may have enriched some fungal species in 13C, but because the CO2 was consistently 13C-enriched compared with the wood, litter or roots, other mechanisms have to be found to explain the consistent 13C enrichment of fungal sporocarps relative to the plant materials.
Preferential use of carbohydrates could be one such mechanism. It has been suggested that wood saprotrophic fungi primarily incorporate cellulose derived break down products, regardless of lignin-degrading capability (Hobbie et al. 2001). The similar δ13C values of sporocarps from white-rot and brown-rot fungi (Table 1) are in agreement with this hypothesis. A preference for cellulose, which show around 4‰ higher δ13C values than lignin and c. 2‰ higher δ13C values than bulk wood in C3-plants (Benner et al. 1987; Gleixner et al., 1993), may explain some or all of the observed 2–3‰13C enrichment of the CO2 and the sporocarps relative to wood in the investigated wood saprotrophs (Fig. 2), but since we have not measured isotopic composition of cellulose we cannot say how much this contributes. Except for preferential use of carbohydrates, other mechanisms, such as anaplerotic fixation of 13C-enriched CO2 and exudation of 13C-depleted carbon, may also play a role to a variable extent.
In ectomycorrhizal fungi, similar mechanisms as those in the saprotrophs may explain the 13C enrichment of the sporocarps and the respired CO2 relative to plant materials (Table 1). Thus, it has been suggested that the carbohydrates supplied by the host are 13C-enriched compared with plant biomass (Hobbie & Colpaert, 2004), although exudation of 13C-depleted carbon and anaplerotic fixation of 13C-enriched CO2 could potentially contribute in this case as well. The size of the isotopic shift between ectomycorrhizal sporocarps and roots found in the present study is comparable to the 1.2–2.9‰ shift previously observed between ectomycorrhizal sporocarps and plant leaves (Högberg et al., 1999).
We do not know why the polypores and E. muricatus were separated from the other fungal species, when the sizes of the δ13C differences between the CO2 and the sporocarps were plotted against the respiration rate (Fig. 4), but one explanation could be that the different fungi respired metabolites that had variable δ13C values as the result of isotopic effects at metabolic branch points. Alternatively, the δ13C of the exogenous C source changed over time such that the C used during sporocarp formation was different in δ13C compared with the C used for respiration when the sporocarps were sampled. The photosynthates as well as root respiration are known to vary in isotopic composition from day to day, depending on weather and environmental conditions (Brugnoli & Farquhar, 2000; Ekblad & Högberg, 2001; Ekblad et al., 2005; Brandes et al., 2006). Hence, if the respired CO2 and the fungal tissue originate from photosynthates fixed at different times with varying isotopic composition, it seems inevitable that the CO2 and the fungal tissue sometimes differ in isotopic composition. It is tempting to speculate that in more long-lived mycorrhizal sporocarps, such as E. muricatus, the relative contribution of stored C to sporocarp respiration could potentially vary over the growing season, which may be one of several reasons why E. muricatus respired 13C-depleted CO2 relative to the sporocarp. The time of the year of sampling is probably less important in wood saprotrophs growing on a stable reliable C source – a log. In support of this view, sporocarps of F. fomentarius sampled in September and December showed a similar isotopic shift between CO2 and sporocarps (Table 1). This stability over time was also illustrated in the δ13C values of CO2 and solid matter from three successive pore layers of F. pinicola.
Stable isotope studies of fungi commonly involve the sampling of sporocarps, because of the impracticality of sampling and taxonomic identification of mycelia. An important question is whether isotopic patterns in sporocarps are comparable to their respective mycelia. In the present study, the δ13C and δ15N ranges of ectomycorrhizal sporocarps compare with the values of mycelia sampled in 2004 from the same forest (Boström et al., 2007; Fig. 1). This is also in agreement with Wallander et al. (2001), reporting similar δ13C values in ectomycorrhizal sporocarps and mycelia. It is unknown whether the δ13C of CO2 respired by sporocarps and mycelia are comparable.
To our knowledge, this is the most complete report so far comparing the δ13C values of fungal respired CO2 and biomass, and the first comparison of values from CO2 and biomass of ectomycorrhizal fungi. Respiration of CO2 that is 13C-depleted compared with the sporocarps may have caused isotopic enrichment in some species. However, the contrasting isotopic shifts between the respired CO2 and the sporocarps of different fungal groups, and the fact that the CO2 respired by the sporocarps was consistently 13C-enriched compared with wood, litter or roots, suggest that respiration of 13C-depleted CO2 cannot entirely explain why the fungi are consistently 13C-enriched compared with plant materials. We hypothesize that this 13C enrichment of sporocarps is mainly caused by preferential uptake of carbohydrates that are 13C-enriched compared with plant materials, but respiration or exudation of 13C-depleted carbon or uptake of 13C-enriched carbon through anaplerotic CO2 fixation may also contribute to a variable extent. Future studies, in which isotope compositions of fungal metabolites are measured, may help us to elucidate whether variation in the fungal metabolism causes the small but variable isotopic shift between fungal tissue and respired CO2.
This study was financially supported by the Swedish Research Council and the Swedish Research Council for the Environment, Agricultural Sciences and Spatial Planning. We are grateful to Andy Taylor and Ingrid Högström for help with identifying fungal sporocarps, and to Gray Gatehouse for correcting the language.