Plant Soil (2010) 326:3–20 DOI 10.1007/s11104-009-9981-5
REGULAR ARTICLE
Plant performance in stressful environments: interpreting new and established knowledge of the roles of arbuscular mycorrhizas Sally E. Smith & Evelina Facelli & Suzanne Pope & F. Andrew Smith
Received: 19 January 2009 / Accepted: 24 March 2009 / Published online: 19 May 2009 # Springer Science + Business Media B.V. 2009
Abstract Arbuscular mycorrhizal (AM) symbioses are formed by approximately 80% of vascular plant species in all major terrestrial biomes. In consequence an understanding of their functions is critical in any study of sustainable agricultural or natural ecosystems. Here we discuss the implications of recent results and ideas on AM symbioses that are likely to be of particular significance for plants dealing with abiotic stresses such as nutrient deficiency and especially water stress. In order to ensure balanced coverage, we also include brief consideration of the ways in which AM fungi may influence soil structure, carbon deposition in soil and interactions with the soil microbial and animal populations, as well as plantplant competition. These interlinked outcomes of AM symbioses go well beyond effects in increasing nutrient uptake that are commonly discussed and all require to be taken into consideration in future work designed to understand the complex and multifaceted
Responsible Editor: Peter Christie. S. E. Smith (*) : E. Facelli : S. Pope : F. Andrew Smith School of Earth and Environmental Sciences, University of Adelaide, Adelaide, South Australia 5005, Australia e-mail:
[email protected] S. Pope School of Agriculture, Food and Wine, University of Adelaide, Adelaide, South Australia 5005, Australia
responses of plants to abiotic and biotic stresses in agricultural and natural environments. Keywords Arbuscular mycorrhizas . Nutrient uptake . Water relations . Soil structure . Plant competition . Carbon deposition in soil . Soil microorganisms
Introduction Terrestrial plants have evolved a range of strategies which allow them to access nutrients effectively. These strategies include development of extensive root systems, with fine roots and long root hairs, and highly specialised cluster roots, as well as symbiotic interactions (Lambers et al. 2008a). Although here we focus on arbuscular mycorrhizal (AM) symbioses as by far the most prevalent, it is useful to briefly consider them in the context of the whole range of root and symbiotic strategies that are important in soils containing different types of organic and inorganic nutrients (Table 1). The occurrence, evolution and ecological significance of the different strategies will of course differ in different plants and different soils (Lambers et al. 2008b). Long, finely branched root systems and roots with long root hairs are important in increasing effectiveness in acquisition of relatively immobile inorganic nutrients such as P and Zn (Gahoonia and Nielsen 2004a, b). Where soils are severely impoverished in labile inorganic P, production of exudates containing
4
Plant Soil (2010) 326:3–20
Table 1 Diagrammatic representation of plant and symbiotic mycorrhizal strategies involved in acquisition of soil nutrient resources (particularly N, P and Zn) of different availabilities
Strategy
Plant strategies Roots and root
Resource
hairs
Soluble
all nutrients
Symbiotic strategies
Exudates/clusters AMa
ECMb and ERMc
P
P, Zn (N)
P&N
P
(P)
P&N
(P)
(P)
P&N
inorganic Insoluble inorganic Labile/soluble all nutrients organic Recalcitrant
P&N
organic Increased intensity of shading indicates increased importance of the adaptation. Modified from Smith and Read (2008) a
Arbuscular mycorrhiza
b
Ectomycorrhiza
c
Ericoid mycorrhiza
tricarboxylic acid anions, either by simple roots or specialised cluster roots, can enhance the availability of insoluble inorganic P (Lambers et al. 2008a; Lambers and Poot 2003). Such plant adaptations are particularly prevalent in the Proteaceae, often found on ancient and highly weathered soils where P and N are very scarce, but are also produced by some species in other families (Lambers and Poot 2003). Similar low-nutrient soils support plants that live symbiotically with organisms that enhance acquisition of nutrients. The least common are plants that form N2-fixing symbioses with bacteria and cyanobacteria (N acquired from the atmosphere), and by far the most common and widespread are plants forming mycorrhizal symbioses. These are of four main types, which involve approximately 90% of terrestrial vascular plants and specialised soil-borne fungi (Table 1). The overall significance of different types of mycorrhiza in nutrition of plants is well established (Smith and Read 2008). Arbuscular mycorrhizal (AM) symbioses are the most frequent
symbiotic plant adaptation for plant growth in low-P soils; they occur in nearly all terrestrial plant ecosystems worldwide and involve probably as many as 80% of vascular plant species, together with a conservative estimate of 150 fungal species (Fitter 2005). The main physiological basis for this symbiosis is usually considered to be bidirectional transfer of nutrients: uptake of inorganic soil P (and also Zn and N) by AM fungi via their extensive external hyphae, and transfer to the plant, in exchange for organic carbon (C). The fungi are obligate symbionts and cannot survive without this supply. The end result in experiments using low-P soil is often (but certainly not always) a large increase in plant growth, compared with nonmycorrhizal (NM) control plants, because the activities of the fungi in absorbing and transferring soil-derived nutrients result in relief of nutrient deficiencies. However, as will be discussed later, the AM pathway for P uptake has also been shown to operate significantly in plants which do not show an increase in total
Plant Soil (2010) 326:3–20
P uptake and/or growth. This ‘hidden’ P uptake (which may also occur in AM-responsive plants) has important implications for interpretation of outcomes of AM symbiosis at a range of scales from cellular to ecological (Smith et al. 2009a). In soils with a large accumulation of organic matter the ectomycorrhizal (ECM) and ericoid mycorrhizal (ERM) symbioses come into play, because in these the fungal partners have a strong capacity to mobilise both P and N from both available and recalcitrant organic forms. Thus ECM trees are often important components of both boreal and tropical forests, and ERM shrubs are important in heathlands. Both groups of symbiont also transfer inorganic P and N to plants, and consequently also occur and play roles in mineral soils, that are low in organic C, P and N. However, it has to be remembered that many types of trees form AM symbioses, while some form both AM and ECM symbioses. Further, most plants with N2-fixing symbioses are also AM, though there are exceptions, e.g. lupins, which form cluster-roots (Lambers et al. 2008a; Lambers and Poot 2003; Smith and Read 2008). A diversity of outcomes of the symbioses between different plant and AM fungal species or isolates is increasingly recognised, and this will impact on ecological interactions in plant and fungal communities, as well as among the broader soil microbial community. This review will highlight activities of arbuscular mycorrhizas, which are important for both natural and agro-ecosystems, including most major crops. In particular, we will provide an update on new and less widely appreciated aspects of the symbioses. Recent research has revealed subtleties in the way AM fungi influence plant nutrition, and has implications for the diversity of plant responses ranging from highly positive to negative, in terms of P uptake and growth. We will also focus on aspects of the symbioses that are only indirectly related to plant nutrition, including roles of the external mycelium in stabilising soil structure and delivering organic C to soil, effects on water relations of plants and interactions with some other soil organisms. Low soil water content is strongly linked to low nutrient availability and to poor soil structure, so that mycorrhiza-development may play significant roles in alleviation of both water and nutrient deficiency. These features of arbuscular mycorrhizas may have special significance in arid environments or in fragile or compacted soils, which were a particular focus of the International Workshop
5
in Shihezi, China, and provided the original rationale for writing this paper. Our intention is to extend discussion of the diversity of outcomes of symbioses involving different plant-fungus combinations (Newsham et al. 1995b) and to highlight aspects which are likely to have significant ecological implications in a range of different environments.
Arbuscular mycorrhizas and plant nutrient uptake: new research shows that the two uptake pathways do not act additively It is well established that AM plants have two pathways by which nutrients (particularly P) can be absorbed: the direct uptake pathway through epidermis and root hairs, and an AM pathway, in which P absorbed by external fungal mycelium is translocated to structures inside the root and thence across the symbiotic interface to plant cortical cells (Fig. 1). Recent work, both physiological and molecular, has provided new insights into the integration of these two uptake pathways and how they influence plant nutrient acquisition (Bucher 2006; Javot et al. 2007; Smith and Read 2008). As indicated above, a large number of field and glasshouse investigations have shown that the outcome of establishment of AM symbioses in low-P soil is a marked increase in plant growth and P uptake, compared with NM control plants of the same species, which of course do not usually exist in nature. The traditional explanation of these effects is that direct uptake of P (and probably other nutrients such as Zn and N) is supplemented by uptake via the AM pathway. Relief of P stress in the plant is considered to be the basis for increased growth (i.e. the two pathways act additively). This simple view is now being questioned (Smith et al. 2009a, b). In an increasingly well recognised number of cases, AM colonisation does not result in any increases in growth or in total plant P, and sometimes the AM plants are smaller than the NM controls (Johnson et al. 1997; Smith et al. 2009a). There is therefore a continuum of responses from strongly positive to negative, indicating considerable ‘functional diversity’ in AM symbioses (Jakobsen et al. 2002). The AM-responsiveness in terms of plant growth is determined by properties of 1) the plant genome: e.g. development of extensive root systems and long root-hairs that enhance P uptake by the plant when it is NM; 2) the AM fungal genome:
6
Plant Soil (2010) 326:3–20 P depletion zone
AM pathway
Arb
CC P plant high-affinity P transporters (AM-inducible)
Direct pathway
plant high-affinity P transporters (P-responsive)
fungal high-affinity P transporters
P
root growth
Fig. 1 Diagrammatic representation of the integration of the direct and AM pathways of uptake of P as orthophosphate in a root growing through soil. The direct pathway involves expression of plant orthophosphate transporters in root hairs and epidermis, resulting in progressive depletion of P close to the root surface. The AM pathway involves the expression of fungal high affinity transporters in external mycelium and plant
transporters in the interface between fungus and plant. In this diagram the intracellular fungus is depicted as an arbuscule (Arb) inside a cortical cell (CC), but other structures such as intracellular coils may also be important. Diagram drawn by Emily Grace and reproduced from Smith et al. (2009b), with permission of the ATSE Crawford Fund, Australia
e.g. inherent extensiveness of external hyphae and other features, as yet not clearly identified; and 3) plant-fungus genomic interactions: e.g. speed and extent of plant colonisation and upregulation of orthophosphate transporter genes in colonised root cortical cells (see Jakobsen et al. 2002; Smith et al. 2009a; Smith and Read 2008, and references therein). It is crucial to recognise that for a single plant species responsiveness varies considerably with the identity of the fungal symbiont. This has important consequences for discussions of whether AM fungi which do not promote a positive growth response can be viewed as parasitic, ‘cheating’ their plant partners by receiving C, but delivering little or no P (Johnson et al. 1997; Jones and Smith 2004; Smith et al. 2009a). It is now clear that the AM pathway plays an important role in P uptake by plants that are colonised by AM fungi, but do not respond in terms of increased P content. Experiments with compartmented pots, in which radioactive P (32P or 33P) is available only to the external mycelium, have repeatedly shown delivery of P via the AM pathway to plants, not only in positively responsive plant-fungus combinations, but also when the plants show no benefit in terms of increased growth or P uptake (e.g. Hetrick et al. 1996; Ravnskov and Jakobsen 1995; Smith et al. 2003; 2004; Zhu et al. 2003). The operation of the AM P uptake
pathway is accompanied by expression of genes for individual transporters for orthophosphate (Pi) in root cortical cells that are strongly upregulated in AM plants, regardless of the plant responsiveness (Bucher 2006; Poulsen et al. 2005). Where growth responses are positive it is still possible to suggest that the direct and AM uptake pathways act additively, but when responses are neutral or negative this cannot be true. Measurement of the contributions of the two pathways has now shown unequivocally that in many cases of non-responsive symbioses (as well as responsive ones) most of the plant P enters via the AM pathway. The corollary must be that the direct uptake pathway makes a reduced or negligible contribution (Li et al. 2006; Poulsen et al. 2005; Smith et al. 2003, 2004). There are three possible explanations (not mutually exclusive) for the loss of activity of the direct pathway. The first is that the concentration of P in the soil solution adjacent to the root becomes depleted to the point that no further uptake of P can occur, regardless of the activity of the direct P transport system. Depletion at the epidermal uptake surface (i.e. in the rhizosphere) is particularly important for immobile nutrients such as P (Silberbush and Barber 1983; Tinker and Nye 2000), and is worse in dry soil (see below). The second explanation is that the orthophosphate transporters in the direct pathway have reduced expression (measured
Plant Soil (2010) 326:3–20
as RNA accumulation) and hence activity; this has been observed in some, but by no means all, experiments (Burleigh et al. 2002; Glassop et al. 2005; Liu et al. 1998; Paszkowski et al. 2002). The third explanation is that there is downstream regulation of transporter synthesis and activity. The last two explanations assume that transporter activity is important in controlling the rate of Pi uptake, although there are some doubts about this (Rae et al. 2004). Elucidation of mechanisms underlying reduced contributions of the direct uptake pathway will be a worthwhile area for future research. Clearly plants need to maintain the option of utilising the direct pathway, because very young seedlings and root apices may not be mycorrhizal, due to delays in colonisation. Some soils (e.g. following disturbance or burning) have low mycorrhizal infectivity, so that plants may not become colonised at all. Furthermore, only one investigation of the relative contributions of direct and AM uptake has, as far as we are aware, involved a wild plant species (Hetrick et al. 1994) and again extension of our knowledge will be important in understanding roles of AM symbioses in plant interactions and competition.
Mycorrhiza-induced growth depressions The obligately symbiotic AM fungi are of course completely dependent on plants for organic C. Explanations of the outcomes of symbiosis for the plant are often given simply in terms of the balance between net costs (C loss to the fungus) and net benefits (additional P supply via the fungus). Where net costs exceed net benefits, and plant growth depressions follow, it is then conventionally assumed that the fungus is a parasite that ‘cheats’ or ‘exploits’ its host by obtaining C but providing little or no soil P. This conventional explanation is tenable when the fungi colonise the roots extensively. However, recent data and review of previous results show that large growth depressions are not necessarily associated with high AM fungal colonisation, but also occur when there is very low internal root colonisation, and in some cases also low external mycelium in soil (Grace et al. 2009a, b; Li et al. 2008a; Smith et al. 2009a, b). Furthermore, the explanation is very plantoriented, and ignores questions that are important in contexts of fungus-plant physiological and molecular interactions, ecology and evolution. Importantly, our
7
increased understanding of the pathways of P uptake in AM plants are beginning to show that simple explanations for growth depressions based on excessive C drain are almost certainly incomplete (Grace et al. 2009b; Li et al. 2008a; Smith et al. 2009a). It is possible that growth depressions in the absence of high fungal biomass are the result of P deficiency, induced by reduced activity of the direct P uptake pathway and inadequate contribution of the AM pathway because of the low root colonisation or hyphal development in soil (Li et al. 2008a; Smith et al. 2009a). The regulatory interplay between AM P uptake and direct uptake through epidermis and root hairs has yet to be fully revealed. Understanding the controlling factors has the potential to inform plant breeding strategies to eliminate growth depressions where this might prove advantageous to crop production (Grace et al. 2009b; Li et al. 2008a; Smith et al. 2009b). The existence of growth depressions has long perplexed those interested in the evolution of AM symbioses, because it has been assumed that larger size is a selective advantage. Hence, it was hard to comprehend the evolutionary persistence of a symbiosis that did not confer a positive growth benefit. A number of explanations have been brought forward which may partially explain the conundrum. Some researchers have drawn on the (undoubted) artificiality of pot experiments to suggest that growth depressions are artefacts, most probably caused by low light and hence low photosynthesis and limiting C supply (Koide and Schreiner 1992). However, this explanation is unlikely to be correct where depressions are associated with low colonisation and hence minimal C drain. In any event, it is hard to show that depressions do not occur in nature, because of the extreme difficulty of producing (again artifactual) NM controls under field situations (Plenchette et al. 1983a, b). Other explanations for the evolutionary persistence of AM symbioses in non-responsive plants suggest benefits of AM symbioses that are not directly related to plant growth, such as tolerance to pathogens or drought and effects on soil structure. These are certainly important in some circumstances (Miller and Jastrow 2002; Newsham et al. 1995a, b; Tisdall and Oades 1982). However, now that it is known that the AM P uptake pathway operates in virtually all AM plants, whether responsive or not in terms of P uptake and growth, additional explanations based on AM P uptake become feasible, including effects on plant competi-
8
Plant Soil (2010) 326:3–20
tion. Furthermore, the effects that the external AM fungal mycelium has on soil structure cannot be ignored, given that the latter plays a critical role in plant water relations and maintenance of soil fertility.
Exploration of soil by external mycelium and interactions with soil structure In all types of mycorrhiza the fungi produce large amounts of mycelium in soil. Hyphal length density associated with an AM root can vary considerably, from as low as 300 µma
2–20 µm
Length density in soil a
Fig. 2 Hyphae of AM fungi have very much smaller diameters than many soil pores. Scanning electron micrograph of a hypha of the AM fungus Glomus intraradices (arrowhead) growing through a sand matrix with most common pore diameter of 38 µm. Bar=20 µM. Photo courtesy Elizabeth Drew (Drew 2002)
