Is plant genetic control of ectomycorrhizal fungal communities an untapped source of stable soil carbon in managed forests?
- First Online:
- Cite this article as:
- Hoeksema, J.D. & Classen, A.T. Plant Soil (2012) 359: 197. doi:10.1007/s11104-012-1201-z
Ectomycorrhizal (ECM) fungi provide one of the main pathways for carbon (C) to move from trees into soils, where these fungi make significant contributions to microbial biomass and soil respiration.
ECM fungal species vary significantly in traits that likely influence C sequestration, such that forest C sequestration potential may be driven in part by the existing community composition of ECM fungi. Moreover, accumulating experimental data show that tree genotypes differ in their compatibility with particular ECM fungal species, i.e. mycorrhizal traits of forest trees are heritable. Those traits are genetically correlated with other traits for which tree breeders commonly select, suggesting that selection for traits of interest, such as disease resistance or growth rate, could lead to indirect selection for or against particular mycorrhizal traits of trees in forest plantations.
Altogether, these observations suggest that selection of particular tree genotypes could alter the community composition of symbiotic ECM fungi in managed forests, with cascading effects on soil functioning and soil C sequestration.
KeywordsCarbon sequestrationEctomycorrhizal fungiPinusExtracellular enzymes
Understanding how above- and below-ground processes shape ecosystem processes such as carbon (C) sequestration has been a major research area in forest ecology and ecosystem management over the last 10 years, yet little is still known about how these processes might interact with climatic change to shape managed ecosystems and their ability to sequester C in the future (e.g., Bardgett and Wardle 2010; Orwin et al. 2011; Schroter et al. 2004; van der Putten et al. 2009; Wardle et al. 2004). Selecting tree species and genotypes for plantation forestry based purely on their above-ground characteristics, such as wood density or growth rate, may not facilitate alternative management goals such as increased soil C sequestration. In particular, belowground characteristics of trees, such as their compatibility with particular taxa of mycorrhizal fungi, could substantially accelerate or decelerate the soil C cycle (Chapela et al. 2001; Orwin et al. 2011), but these relationships need much more exploration. While managing plant genotypes and their fungal symbionts is clearly an emerging area of research (e.g., Singh et al. 2011), here we develop the hypothesis that because mycorrhizal fungal taxa vary in characteristics that may influence soil C cycles, and because tree genotypes often vary in their compatibility with particular mycorrhizal fungal taxa, such ‘mycorrhizal traits’ of trees should be considered when making selections of tree genotypes for plantation forestry.
Increased carbon (C) sequestration in plantation forests holds significant potential to reduce net CO2 emissions (Goodale et al. 2002; Jackson and Schlesinger 2004). In the southeastern U.S., managed pine ecosystems occupy nearly 20 % of land area (Joyce et al. 2001), a proportion that is predicted to continue to grow (Prestemon and Abt 2002), and these ecosystems have been proposed as major contributors to C sequestration in the U.S. (Johnsen et al. 2001). Although significant C is stored in the woody stems of forest trees, the mineral soils of managed forests may hold the greatest potential for below-ground C storage, as the global average soil C pool is 3.3 times the size of the atmospheric pool and 4.5 times the size of the biotic pool (Lal 2004). Despite this potential, processes controlling soil C pools and fluxes in tree plantations are much more poorly understood than those controlling aboveground C dynamics (Goodale et al. 2002; Johnsen et al. 2001), leaving substantial uncertainty about the mechanisms of C sequestration in these systems (Leake et al. 2004), especially under climatic change (Drake et al. 2011; Parrent et al. 2006; Schafer et al. 2003).
The degree to which managed forests contribute to long-term C sequestration in terrestrial ecosystems will be influenced by management decisions (Galik and Jackson 2009), including site selection and fertilization and the selection of tree species and genotypes with favorable traits for C sequestration (Garten et al. 2011; Karlinski et al. 2010). Numerous tree traits under genetic control may have direct and indirect effects on forest C sequestration, ranging from leaf and root chemistry, to rooting or wood properties that confer resistance to wind damage in hurricanes, to compatibility with different species of root symbiotic ectomycorrhizal (ECM) fungi (e.g., Lukac et al. 2003). Understanding how these traits are correlated among tree genotypes may allow optimal selection of genotypes that favor increased C sequestration while still providing commercial sylvicultural benefits.
Ectomycorrhizal fungi potentially facilitate long-term C storage in pine forests
At a local scale, native populations of trees such as pines typically host approximately dozens to hundreds of different species of ectomycorrhizal (ECM) fungi as root symbionts (Buee et al. 2009; Horton and Bruns 2001; Tedersoo et al. 2010). In these symbioses, the host plant provides the vast majority of C for fungal growth, sending 10–20 % of fixed C to the fungal symbionts (Allen 1991; Hobbie 2006), although some ECM fungi may also garner significant C from non-host sources in the soil (Smith and Read 2008). The fungi use this C to build a hyphal network in the soil, acting as extended host root systems, exploring small pore spaces in soils and typically enhancing plant nutrient uptake (Smith and Read 2008). As a consequence, ECM fungi provide one of the main pathways for C to move from ECM trees into soils (Godbold et al. 2006), where these fungi have been estimated to harbor approximately 30 % of microbial biomass (Hogberg and Hogberg 2002) and to contribute up to 35 % of total soil respiration (Heinemeyer et al. 2007). Moreover, fungal hyphae decompose more slowly than plant roots and ECM colonization can slow root decomposition by 65 %, likely due to greater recalcitrance of fungal cell wall components (Langley et al. 2006). ECM fungi can also competitively suppress saprobic fungi in soils, further slowing decomposition processes in litter and soil (Gadgil and Gadgil 1971; Leake et al. 2002; Lindahl et al. 2010). In addition, extracellular polysaccharides and hyphae from arbuscular mycorrhizal (AM) fungi are known to contribute significantly to formation of soil macro-aggregates which protect plant derived soil organic matter (SOM), and hydrophobin proteins in ECM fungi are hypothesized to play a similar role (Rillig and Mummey 2006). ECM fungi have also been indicated as facilitating the response at Free Air Carbon Enrichment (FACE) sites where soil C sequestration has increased under simulated elevated atmospheric CO2 concentrations (Lukac et al. 2003; Pritchard et al. 2008; Schafer et al. 2003). On the other hand, it is also possible that ECM fungi could act as conduits for returning surplus plant C to the atmosphere via decomposition processes, thus limiting the potential for soils to store increased C under elevated atmospheric CO2 (Heinemeyer et al. 2007). Altogether, these factors suggest that ECM fungi have the potential to significantly alter soil C sequestration in forests (Rooney et al. 2009; Treseder and Allen 2000).
Candidate characteristics of a soil C storage-promoting ECM fungal species
Hypothesized C sequestration trait
Candidate ECM fungal taxa
High-biomass exploration strategy
Low respiration rate and/or low N concentration
Low decomposition rate and low N concentration
Cenococcum geophilum, Lactarius chrysorheus (Koide and Malcolm 2009)
Low fine root turnover rate
Variation inconsistent among taxa (Treseder et al. 2004)
High tissue chitin concentration
Russula sp. (Wallander et al. 1997)
High chitinase enzyme activity, potentially allowing suppression of saprobic fungi in woody debris
Lactarius quietus and Tomentella spp. (Buee et al. 2007)
In addition to influencing inputs, a few studies indicate that ECM fungal species vary in traits that influence the rate at which C leaves the ECM fungal biomass pool and enters the SOM pool (Leake et al. 2004). These traits include respiration rate (Bidartondo et al. 2001; Trocha et al. 2010), fine root turnover rate (Treseder et al. 2004), tissue chitin content (Wallander et al. 1997), and decomposition rate (Koide and Malcolm 2009). While we clearly need more studies to determine whether these traits vary in consistent ways among ECM fungal taxa, or whether such variation is more influenced by experimental context (e.g., lab vs. field, field fertility, genetic variation among individuals, plasticity of individual genotypes, or variation in traits among different tissues) or seasonal variation, we suggest these traits are good candidates to target. Decomposition rates and respiration rates, for example, can vary significantly among mycorrhizal species and are positively correlated with tissue N concentrations (Koide and Malcolm 2009; Trocha et al. 2010). Such data indicate that fostering specific mycorrhizal taxa in plantations may speed or slow the rate of C decomposition and turnover in soils.
ECM fungal species may function in ways that actively shape C dynamics in soils around them, especially through production of extracellular enzymes that may degrade existing soil carbon. There is some indication that ECM fungal species vary in their ability to use extracellular enzymes to break down recalcitrant C-containing compounds in the soil, and in their direct or indirect suppression of saprobic fungi that might be competing for these resources (Buee et al. 2007; Courty et al. 2010; Jones et al. 2010; Taylor et al. 2004). Orwin et al. (2011) used a modeling approach to show that removal of organic nutrients by mycorrhizal fungi from moderately recalcitrant (‘slow’) pools of soil organic matter can lead to increased C storage because the remaining organic matter quality is reduced and subsequent decomposition rates by saprobic microbes are slowed. Buee et al. (2007) found approximately 10-fold variation among ECM fungal species in potential activities of extracellular enzymes involved in accessing organic nutrients in slow organic matter pools. They also found that several ECM fungal taxa exhibited substantial chitinase activity when growing in dead woody debris that was also colonized by saprobic white-rot fungi, suggesting an ability of these ECM fungi to break down dead or living tissue of saprobic fungi, potentially slowing the decomposition of woody debris by saprobic fungi.
Although differential production of some classes of extracellular enzymes by ECM fungi may increase C storage, other classes of enzymes may contribute to C loss from soils by speeding decomposition of highly recalcitrant components of litter or soil organic matter such as lignin, and ECM fungal species seem to vary in their production of such enzymes. For example, Buee et al. (2007) found that two Russula species and one Tomentella species exhibited 4- to 7-fold higher than average (compared to other taxa observed in the community) activity of laccase, an enzyme involved in degradation of lignin and other complex polyphenolic molecules. However, it is important to note that studies are also revealing that specific ECM fungal taxa vary in potential activity of particular enzymes over time and when colonizing different substrates, highlighting the strong context-dependency in production of these extracellular enzymes (Buee et al. 2007; Courty et al. 2010). For example, Lactarius quietus exhibited high activity of cellobiohydrolase enzymes when colonizing a soil horizon rich in leaf litter, but exhibited high chitinase activity when colonizing dead woody debris (Buee et al. 2007).
Genotypic variation in trees for ECM symbiosis traits
Accumulating studies suggest that populations of forest tree species have significant genetic variation in their compatibility with particular ECM fungal species, suggesting that tree genotype selection could influence ECM fungal community composition. For example, Dixon et al. (1987) found that colonization by the ECM fungus Pisolithus tinctorius varied among genotypes of loblolly pine. Three Pinus species on the West Coast of North America (P. contorta, P. muricata, and P. radiata) harbor significant genetic variation between and within populations for compatibility with the ECM fungal species Rhizopogon occidentalis (Hoeksema et al. 2009; Hoeksema and Thompson 2007; Piculell et al. 2008). Korkama et al. (2006) showed that Norway spruce (Picea abies) genotypes vary in community composition of ECM fungi. Similarly, Leski et al. (2010) found that Scots pine (P. sylvestris) genotypes varied significantly in ECM fungal community composition, especially with respect to relative abundances of a low-biomass Wilcoxina species versus high-biomass Suillus species, which exhibit orders of magnitude greater standing biomass in soils compared to Wilcoxina species. The latter two studies provide examples of what has been called a “community phenotype” whereby the genotype of one species determines the composition of other species with which it interacts (Whitham et al. 2006), and such results suggest that tree populations could respond to natural or artificial selection by shifting the communities of ECM fungi with which they interact.
Although the evidence is still somewhat sparse, several kinds of studies across a diversity of plant taxa suggest that mycorrhizal traits of plants, e.g., plant compatibility with particular mycorrhizal fungal species or overall responsiveness of plants to mycorrhizal fungi, are genetically correlated with other plant traits of concern to managers such as growth rates and herbivore or disease resistance (e.g., Piculell et al. 2008; Sthultz et al. 2009; Tagu et al. 2005; Toth et al. 1990; Zhu et al. 2001), perhaps due to the influence of shared pleiotropic loci in the plant genome. For example, evidence from genomic studies of poplars (Populus spp.) suggests that loci contributing to quantitative variation in colonization by the ECM fungus Laccaria bicolor are loci that have previously been found to be involved in pathogen resistance (Labbe et al. 2011). The implication of such observations is that genetic selection for conventional traits of interest such as disease resistance or growth rate could lead to indirect selection for or against particular mycorrhizal traits of trees, including ECM fungal community phenotypes. Studies by Piculell et al. (2008) on Pinus muricata, Leski et al. (2010) on Pinus sylvestris, and Hoeksema and Thompson (2007) on three pine species found that pine seedling genotypes differed in symbiotic compatibility with particular ECM fungal species, as well as in several seedling morphological traits such as relative growth rate. Sthultz et al. (2009) found that different genotypes of Pinus edulis varied in both their ECM fungal communities and in their resistance to aboveground scale insect herbivores. Putative quantitative trait loci (QTLs) controlling colonization intensity by the ECM fungus Laccaria bicolor mapped onto the Populus trichocarpa genome very near a previously discovered QTL that contributes to mediation of poplar resistance to fungal rust pathogens (Tagu et al. 2005). Such findings raise the possibility that one or more closely linked loci may have simultaneous (pleiotropic) effects on both mycorrhizal colonization and other key tree traits such as pest resistance, and that selection for one of those traits could result in indirect selection on the other trait.
If tree breeders are concerned with soil C dynamics, they need to be cognizant of how selection on conventional traits may indirectly affect mycorrhizal traits of trees. The general idea that artificial selection on conventional plant traits may lead to correlated selection on traits governing plant interactions with root symbiotic microorganisms is supported by studies of arbuscular mycorrhizal (AM) and rhizobial symbioses. For example, at least two studies (Toth et al. 1990; Zhu et al. 2001) demonstrated reduced AM fungal colonization intensity in highly selected modern cultivars of crop plants compared to older cultivars, showing that artificial selection for crop traits (such as disease resistance) has resulted in correlated selection against mycorrhizal colonization intensity. These results parallel a recent study showing how artificial selection has reduced the ability of modern cultivars of soya beans to benefit from root symbiosis with rhizobia bacteria (Kiers et al. 2007).
For achieving maximum C sequestration in managed forests, it is essential to understand how tree species and genotype (i.e., seed source, family, or clone) selection may affect soil C dynamics (Garten et al. 2011). At the local scale, soil C accrual is a function of C inputs from plants and the decomposition of these products by the soil microbial community. Tree species and genotype selection may directly or indirectly alter the amount and quality of C entering the soil in several significant ways that are currently poorly understood. Tree species and genotypes may differ in community composition of ECM fungi, and species of these fungi vary in amounts of C garnered from host plants, in standing biomass, in the recalcitrance of fungal tissues, in baseline respiration rates, root exudate production, and in their effects on carbon cycling through enzymatic activities or suppression of saprobic fungal species. Thus, we hypothesize that selection of plant species and genotypes will influence soil C sequestration through alterations of community composition of ECM fungal symbionts. Such management of indigenous mycorrhizal fungal communities may prove more effective than nursery inoculation programs, since failures of the latter have often been attributed to displacement of inoculated fungi by indigenous fungi after outplanting. For example, Malajczuk (1987) found that when Eucalyptus seedlings were inoculated with ECM fungi, inoculated fungi largely failed after outplanting and success of outplanted seedlings was more correlated with colonization by indigenous ECM fungi.
Testing this hypothesis will require field studies to assess (1) whether variation among tree genotypes in ECM fungal community composition is consistent, (2) whether whole suites of ECM fungal species associated with particular tree genotypes have traits that are more favorable than average for C sequestration (as summarized in Table 1), (3) whether those tree genotypes and associated suites of ECM fungi are strongly correlated with measurable increases in long lived C pools, and (4) which characteristics of ECM fungal species contribute to these effects. The likelihood of such an outcome will depend on two assumptions, both of which still need to be validated. First, variation in traits linked to C sequestration among ECM fungal taxa would ideally be greater than variation within taxa, e.g. among different genetic strains of fungal taxa, among tissue types (sporocarps, rhizomorphs, and mycorrhizal root tips), or among environmental contexts. If within-species variation in such traits overwhelms between-species variation, then managing ECM fungal communities based on species-level traits would not be advisable; rather, managing ECM fungal genetic composition should be the goal. Alternatively, consistent ECM fungal species-level traits influencing C sequestration may not be necessary, if plant genotypes can control variation in expression of those traits as suggested by the findings of Courty et al. (2011), who showed that potential activity levels of extracellular enzymes excreted by ECM fungi, such as cellobiohydrolase and beta-glucosidase, vary substantially among poplar genotypes. Second, multiple traits that enhance C sequestration must be correlated across tree genotypes and ECM fungal taxa. For example, if ECM fungal taxa with high biomass exploration strategies (potentially favorable for C sequestration) always have high respiration rates (unfavorable for C sequestration) and vice versa, then suites of ECM fungal taxa favorable for C sequestration may not be possible. As more studies accumulate assessing C sequestration traits of ECM fungal taxa, will any taxa arise that consistently have multiple favorable C sequestration traits? Answering such questions, especially in a management framework that pairs specific tree genotypes with distinct mycorrhizal taxa or genotypes, would reveal the potential for managing mycorrhizal traits of trees to influence soil C dynamics. We are clearly not yet close to being able to manage soil C dynamics in forests through manipulation of ECM fungal communities via tree genotype selection, and other tools may offer more immediate short-term solutions; however, we suggest that interest in sequestering C in forest soils will only increase in the future, and that research efforts should be devoted to understanding how mycorrhizal management may contribute to long-term strategies for C sequestration.