Traditionally, the soil nitrogen cycle in forests has been thought to be governed by mineralization processes driven by bacteria. However, increasing evidence indicates that macrofungi, particularly those involved in mycorrhizal symbioses, play important roles in N and C cycling in temperate and boreal forests, and that the N cycle in these systems is more complex than previously thought (Lindahl et al., 2002; Read & Pérez-Moreno, 2003). For instance, it is now known that many plants can access organic forms of N, usually via their mycorrhizal symbionts (Näsholm et al., 1998, 2000; Aerts & Chapin, 2000). Ectomycorrhizal (EcM) fungi have been shown to utilize N directly from organic materials such as proteins (Abuzinadah & Read, 1986); leaf litter (Bending & Read, 1995; Pérez-Moreno & Read, 2000); pollen (Pérez-Moreno & Read, 2001a); collembolans (Klironomos & Hart, 2001); and nematodes (Pérez-Moreno & Read, 2001b), and to transfer acquired N to their plant partners. This allows ‘short-circuiting’ of the traditional N cycle, as the mineralization pathway need not be involved, and provides a means for close coupling of the N and C cycles.
Ectomycorrhizal fungi are taxonomically diverse, comprising at least 5000–6000 species. Most EcM trees are able to associate with many different mycobionts, and most EcM fungi are able to associate with many different phytobionts (Molina et al., 1992). This leads to a great taxonomic diversity of EcM associations. However, the factors responsible for this diversity are little understood and it is uncertain whether important links exist between taxonomic and functional diversity. Often it has been assumed that the apparently low degree of host-specificity in EcM associations indicates a high degree of functional redundancy among the mycobionts. At the same time, it has been proposed that the taxonomic diversity of EcM fungi reflects great diversity in function (Cairney, 1999). That some, but not all, EcM fungi can obtain N directly from soil organic matter suggests the latter view. However, ‘establishing biodiversity–function relationships remains one of the most intractable challenges in ecological research’ (Leake, 2001), and many more data are needed before the importance of such relationships can be assessed fully.
Unfortunately, soil organisms such as fungi and the processes they mediate are difficult to study. Direct field observation entails disturbance and provides little control over the myriad of variables in the system. Laboratory experiments allow more control and clearer observation but, because of the necessary simplification, their relevance to nature often is not clear (Read, 2002). Thus, elucidating the functional roles of fungi will require a variety of creative approaches. One promising methodology is natural abundance stable isotope ratios. Most biologically important elements occur as two or more stable isotopes, with one being far more abundant than the other(s). Fractionation of the isotopes by biological and physical processes leads to concentration differences in substances of biological interest, and these differences can provide insights into fluxes among organisms; between organisms and their abiotic environment; and among compartments of the abiotic environment. An important advantage in using natural abundance stable isotope ratios for ecosystem studies is their ability to present a time-integrated picture of functional processes which often are difficult to measure directly (Robinson, 2001).
Over the past 10 yr, numerous surveys of N and C stable isotope ratios in sporocarps of macrofungi have been reported (Gebauer & Dietrich, 1993; Handley et al., 1996; Lilleskov et al., 1997, 2002; Taylor et al., 1997; Gebauer & Taylor, 1999; Hobbie et al., 1999a, 2001; Högberg et al., 1999b; Kohzu et al., 1999; Chapela et al., 2001; Henn & Chapela, 2001; Trudell et al., 2001, 2003; Griffith et al., 2002; Horwath et al., 2002; Taylor et al., 2003), most carried out in north temperate or boreal, often conifer-dominated, forests. These have suggested several patterns: (1) δ15N in EcM fungi is comparable to that in mineral soil and greater than that in plant foliage and saprotrophic fungi; (2) δ15N and δ13C in both EcM and saprotrophic fungi are greater than those in their bulk substrates; (3) δ15N and δ13C in soils increase with depth; (4) δ13C in EcM fungi is less than that in saprotrophic fungi, but greater than that in plant foliage. However, confidence in the generality of these patterns, and our ability to draw inferences about the sources and magnitude of variation in stable isotope ratios, are constrained because, with the exception of Taylor et al. (2003), each of these studies has at least one of the following limitations: (1) many sporocarp samples are not identified beyond family or genus; (2) sample sizes are small (one or two observations per species, with each observation representing a single sporocarp); (3) few or no analyses of saprotrophic fungi and/or associated ecosystem pools such as plants and soil are included; and (4) data from different forest types are pooled.
The recent study by Taylor et al. (2003) represents a major advance, by including a large number of analyses with all fungi identified to species, replicate analyses of most species, and inclusion of many saprotrophic fungi plus leaves from associated phytobionts. Three important findings from that study are: (1) isotope signatures of fungus sporocarps varied by family, genus and species; (2) isotope data were most informative at the species level; and (3) species composition of the EcM fungi was important for determining their aggregate isotope signatures. Earlier, Lilleskov et al. (2002) provided evidence that isotope signatures can reflect ecophysiological function – specifically, that EcM fungi that can utilize organic N exhibited higher δ15N than did species restricted to mineral N sources. Hobbie and colleagues (Hobbie et al., 1999a, 1999b, 2000, 2001; Hobbie & Colpaert, 2003) have presented a series of increasingly refined models relating δ15N in fungus sporocarps and plant foliage to soil N status and ecosystem processes. For instance, based on theoretical grounds, field measurements and a laboratory study, they propose that EcM fungus–plant differences in δ15N in part reflect the proportion of mycobiont N transferred to phytobiont(s) and that this, in turn, is influenced by allocation of C from the phytobiont(s) to the mycobiont. Other investigations have focused on the mechanisms that underlie the isotope effects (Högberg et al., 1999a; Kohzu et al., 2000; Emmerton et al., 2001; Henn & Chapela, 2004). Taken together, the results of these studies allow us to begin to understand the processes that produce the different isotope signatures of EcM and saprotrophic fungi, and their associated soil and plant pools. However, there remains a need for more extensive data on sporocarps and other pools from a greater variety of habitats to test the generality of the existing observations and conclusions.
To further assess the usefulness of stable isotopes for understanding elemental cycling and fungal ecology in forests, we have been studying the N and C stable isotope patterns in macrofungi, plants and soils in two old-growth conifer forests from climatically different areas of the Olympic Peninsula, western Washington, USA. Here we: (1) present the largest set of N and C stable isotope data on macrofungi and associated ecosystem pools reported to date; (2) provide additional support for the principal conclusions of Taylor et al. (2003) and extend their observations to additional taxa in a different forest setting; (3) relate the observed stable isotope patterns to differences in the species composition of the macrofungi and discuss possible ecophysiological bases for the relationship; and (4) compare our data with the model of Hobbie and Colpaert (2003) to assess its usefulness for interpreting field isotope measurements.