Portal de Periódicos da CAPES

The nonmycorrhizal root – a strategy for survival in nutrient‐impoverished soils

2005; Wiley; Volume: 165; Issue: 3 Linguagem: Inglês

10.1111/j.1469-8137.2005.01331.x

1469-8137

R. M. Miller,

Plant responses to water stress

One of the more elusive goals in root biology is the integration of structure and function. Despite its importance, this is a goal not easily achieved. Not often does a study elegantly link the natural history of a group of plants with the kinds of root systems the plants produce, and then experimentally demonstrate a mode of action for development and functioning of these root structures. The study of Shane et al. (pp. 887–898 in this issue) has accomplished such a task. Their paper addresses an adaptation of members of the Cyperaceae described as ‘dauciform’ roots, in which a ‘carrot-like’ shape is formed by rootlets densely covered with long root hairs (Lamont, 1974). This research demonstrates that the dauciform root structure is a common characteristic of Cyperaceae in south-western Australia. The research further demonstrates the detrimental effects of elevated phosphorus supply on the development of the dauciform root structure. Until this study, the mechanisms involved in the development and function of this unusual root adaptation had been difficult to explain. ‘Paradoxically, one of the more common root adaptations in nutrient-poor soils is neither mycorrhizal nor nodulated’ Plants growing in nutrient-poor soils have an ability to activate a set of adaptive responses that affect the allocation of internal nutrients and maximize acquisition of external nutrients. The responses include homeostatic adjustments in metabolism and an alteration in root system morphology that accelerates exploitation of limiting nutrients in soil (e.g. Vance et al., 2003; Ticconi & Abel, 2004). A general feature of plants in response to nutrient limitation is increased allocation of photosynthate to root structures and decreased allocation to leaves and stems. Because of the roles played by nitrogen (N) and phosphorus (P) in photosynthetic carbon fixation and cellular processes, plants have evolved various root adaptations for accessing growth-limiting nutrients in soils (Lamont, 1982; Skene, 1998; Vance et al., 2003). In south-western Australia, the soils are highly weathered and quite impoverished in N and P (Lamont, 1982; Pate & Bell, 1999). Many of the plants capable of surviving on these nutrient-poor soils have evolved unusual root adaptations for acquiring from soils nutrients with very low solution concentrations (Lamont, 1982). From an evolutionary perspective, the most common root adaptation for exploiting soils low in P is the mycorrhizal symbiosis (Smith & Read, 1997; Brundrett, 2002), whereas an adaptation associated with N-poor soils is the root nodule symbiosis with N-fixing bacteria (Lamont, 1982; Skene, 1998). These two adaptations are a primary subset of root adaptations that allow plants to survive and speciate in habitats that otherwise would be quite barren (Fitter & Moyersoen, 1996). Paradoxically, one of the more common root adaptations that evolved in nutrient-poor soils is neither mycorrhizal nor nodulated. This adaptation is the production of cluster and dauciform roots, which are structural adaptations in roots that increase nutrient mobilization in nutrient-impoverished soils (Lamont, 1982, 2003; Skene, 1998). On nutrient-impoverished sites in south-western Australia, nonmycorrhizal plants are predominantly members of the Proteaceae and Cyperaceae. The Proteaceae, a group of plants endemic to nutrient-poor soils of the Southern Hemisphere, produce what are referred to as ‘cluster’ or ‘proteoid’ roots. These dense root clusters consist of longitudinal rows of hairy rootlets (Skene, 2001; Lamont, 2003). Of particular interest to the Shane et al. study was determining the prevalence of the ‘dauciform’ root structure in the Cyperaceae and several closely related families that inhabit south-western Australia. Unlike the endemism of the Proteaceae, the Cyperaceae have a worldwide distribution of some 5000 species. Although studies are few, many Cyperaceae species appear to be capable of developing dauciform roots over a wide range of habitats (e.g. Davies et al., 1973; Lamont, 1982; Shane et al.). The short dauciform root system is characterized by laterally swollen, carrot-like structures consisting of rather ephemeral, dense clusters of long root hairs (Davies et al., 1973; Lamont, 1974). Although they are considered functionally analogous with other cluster roots, the research of Shane et al. indicates that dauciform roots are morphologically and physiologically distinct. Our understanding of the development and physiology of dauciform root structures is still quite primitive. However, the research of Shane et al. and earlier studies by Lamont (1974) have shed some light on the development of these unusual root structures found in sedges. These studies demonstrate that dauciform root structures are rather short-lived, lasting no more than a couple of weeks. At extremely low P supply, dauciform structures can compose up to a quarter of a root system's biomass (Shane et al.). The study of Shane et al. demonstrates the importance of habitat in determining their development, with nutrient-impoverished growing conditions being critical to the initiation of dauciform root development. Nutrient supply rates equivalent to those encountered in relatively fertile soils are high enough to suppress dauciform root growth. Both low N and low P supplies favor dauciform root development and growth, whereas increasing N or P supply favors the growth of nondauciform, lateral roots (Lamont, 1974; Shane et al.). These specialized root structures can be suppressed by even the slightest increase in P supply (> 1.0 µm P). Even so, root P concentration appears to remain constant regardless of P supply rate. This particular response suggests that leaf P status – rather than soil P supply – may be the primary control point for development of dauciform root structures (Shane et al.). Shane et al. make a strong argument for the need of similar studies on Cyperaceae species inhabiting other plant communities. The few studies on the distribution of root adaptations of nutrient-impoverished soils of south-western Australia suggest that both plant life form and root adaptation type are habitat-dependent, although root adaptations occur on a smaller spatial scale than does plant life form (Lamont, 1982; Brundrett & Abbott, 1995; Pate & Bell, 1999). Nonmycorrhizal Cyperaceae species were the predominant group of the plant community inhabiting the forest floor at these sites (Lamont, 1982; Pate & Bell, 1999). The existence of spatially segregated root activity within separate soil domains suggests that mycorrhizal type – or lack of mycorrhizas – is most likely dependent on substrate quality (Brundrett & Abbott, 1995). Hence, localization by root type should be based on function and should result in a system in which resources are exploited very efficiently. What is exciting about the research of Shane et al. is the experimental nature of their approach. Rather than basing their study solely on field observations, the researchers screened a wide range of Cyperaceae and allied families for ability to form dauciform root structures under controlled conditions. Field-collected plants were cultivated in aerated nutrient culture at low P supply for up to six months to determine whether dauciform roots would develop. Even though the dauciform root morphology had been reported previously in a few closely related families of the Cyperaceae (e.g. Powell, 1975; Meney et al., 1993), the experimental nature of the survey of Shane et al. demonstrated that this root structure is characteristically formed by Cyperaceae species of the tribes Cariceae and Rhynchosporeae. (Recent classification places most of the genera listed by Shane et al. (table 1) as belonging to the Rhynchosporeae tribe to now belong to the Schoeneae tribe; see Bruhl, 1995.) These observations corroborate and extend the previous surveys on the incidence of dauciform root structures in the Cyperaceae (Davies et al., 1973; Lamont, 1974, 1982). Shane et al. demonstrated that under extremely low P supply dauciform root development was restricted to the local Rhynchosporeae in their study. Furthermore, the genera not producing dauciform roots were confined to the tribe Scirpeae. In part, because of the wide array of plant communities inhabited by the Cariceae members, this tribe may be key to understanding the evolution of the dauciform trait. This rather large tribe of the Cyperaceae is typified by the genus Carex (over 200 species) and has worldwide distribution (Ball, 1990; Reznicek, 1990). What differentiates the Cariceae tribe members surveyed from the Rhynchosporeae tribe members that form dauciform root structures is the temperate and montane distribution of the former (Davies et al., 1973; Ball, 1990). Dauciform structures have been identified in various Carex species growing in vastly different habitats, being found, for example, on roots collected from dunes, alpine pastures, heathland peat and moist grasslands (Davies et al., 1973). The root structures were not as prevalent in more xeric habitats (Davies et al., 1973). Considering the worldwide distribution of the Cariceae, one would expect its members to demonstrate a wide array of root adaptations, whether the species rely on dauciform structures for survival or do not even possess or express the trait. Studies along the lines of Shane et al. should be able to demonstrate whether the expression of dauciform structures is a plesiomorphic condition of the Cyperaceae or is newly derived. The Cyperaceae as a group offer an interesting look into the evolutionary tendency of root systems undergoing the loss of the mycorrhizal association. Arguably, the Cyperaceae are distinctly a nonmycotrophic family within the mycotrophically diverse order of grasses and grass-like plants, the Poales. Phylogenetic studies using rbcL sequence data indicate that the Cyperaceae have close affinity for the Juncaceae (Plunkett et al., 1995), a family also distinctly nonmycotrophic (Powell, 1975). The close phylogenetic affinity between the Juncaceae and Cyperaceae supports a view that their progenitors were already undergoing the loss of mycorrhizas. However, being nonmycotrophic does not necessarily mean that roots are not colonized by mycorrhizal fungi (Tester et al. 1987). A body of literature indicates that many genera of the Cyperaceae can be colonized by arbuscular mycorrhizal fungi (e.g. Lovera & Cuenca, 1996; Miller et al., 1999; Muthukumar, Udaiyan & Shanmughavel, 2004), and several species have been reported to possess ectomycorrhizas, ectomycorrhiza-like structures and dark septate fungi (Haselwandter & Read, 1980; Lipson et al., 1999; Harrington & Mitchell, 2002). What is not well understood is the functional relevancy of the mycorrhizal fungal affinities with Cyperaceae roots. Close examination of the incidence of mycorrhizal colonization in the genus Carex indicates that the major colonization structures are hyphae and vesicles (e.g. Muthukumar & Udaiyan, 2002). In contrast, arbuscule structure, when it occurs, is at very low frequency (Miller et al., 1999; Johnson et al., 2003b), suggesting that the mycorrhizal colonization encountered in Carex roots is likely vestigial. Also, Carex species grown in monoculture are not able to support an active mycorrhizal mycelial network or propagule production (Johnson et al., 2003a; Ruotsalainen & Aikio, 2004). These observations raise the question of whether a functional mycorrhizal symbiosis exists (i.e. whether reciprocal transfer of host carbohydrate for endophyte-supplied phosphate can occur). A more likely scenario is that the colonization experienced by Cyperaceae is a consequence of being part of a community mycelial network (as in Fitter et al., 1998), with the extent of colonization dependent on the activity of the mycotrophic members of the plant community. The existence of hyphal and vesicle colonization within the Cyperaceae plant root system suggests that they are more likely a passive participant. Although generalizing for the vast group of Cyperaceae is perilous, the studies thus far suggest that the habitat soils where the Cyperaceae species are found are just as likely to be N- as P-limiting, and many are organic in nature (e.g. acid and heath peat; Davies et al., 1973). A characteristic of many of the Cyperaceae that inhabit alpine and tundra habitats where soils can be rich in organic matter is the ability to utilize organic sources of N directly (Chapin et al., 1993; Lipson et al., 1999; Raab et al. 1999). In the case of Kobresia, organic N utilization is facilitated by ectomycorrhizas (Lipson et al., 1999). Hence, the Cyperaceae have evolved many different approaches for obtaining sparse nutrients, including adaptations of root structure and the ability to utilize organic sources of N. The array of root adaptations available to the Cyperaceae raises the question of whether the spatially explicit separation of dauciform structures, mycorrhizas and the ability to utilize organic N sources exist. The discussion here raises the following two questions. (1) What are the conditions that have resulted in the selection of the nonmycorrhizal root? (2) More specifically, is the dauciform root an alternative strategy that replaces the mycorrhizal symbiosis, or is it complementary? The study of Shane et al. gives us some clues. The historical and environmental circumstances found in south-western Australia have created an environment in which soils are severely impoverished in P. Paradoxically, nonmycotrophic species dominate many habitats of this region (Lamont, 1982; Pate & Bell, 1999). Plants that can colonize these rather harsh soils instead rely on cluster and dauciform root structures, with fine roots and long, dense root hairs for scavenging the little soil P that is available (Lamont, 1993). The dauciform-root-producing Cyperaceae studied thus far are highly adapted for growth in nutrient-poor soils. The dauciform root structure of the Rhynchosporeae appears to be an adaptation for nutrient exploitation in extremely nutrient-impoverished soils. We can hypothesize from the limited data sets available that Cyperaceae sedges possessing dauciform root structures also have low affinity for mycorrhizas. Sedges that are members of the Rhynchosporeae apparently cannot reduce net P uptake, even under high P supply, suggesting that this tribe evolved as pioneer species in highly disturbed or nutrient-impoverished habitats (Shane et al.). The inability to halt the accumulation of P would most likely limit the dauciform Rhynchosporeae species to soils with very low P content. Because of the diversity of habitats that the Cariceae sedges inhabit, it is unlikely that such a dramatic response would be typical of this tribe, suggesting that the dauciform root system of the Cariceae has different functional constraints. We are still some distance from determining the mechanisms associated with dauciform root function and structure. Research findings thus far describe a root trait that combines changes in developmental pattern with physiology that – unlike mycorrhizas or nodulation – is not initiated by a microbe. Research on the dauciform structure further offers challenging opportunities for gaining mechanistic insights into genetic regulation of morphologic responses to limiting nutrients. The application of molecular approaches offers possibilities for understanding the potential trade-offs, at the gene level, for roots possessing mycorrhizal fungi vs morphological adaptations to limiting nutrients. From an environmental perspective, dauciform roots and their occurrence give us a morphological trait – in addition to mycorrhizas – that has potential to be a transponder for the state of a plant community and, as such, is a sensitive indicator for potential community change. The author's work is supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research, under contract W-31–109-Eng-38. This article has been created by the University of Chicago as operator of Argonne National Laboratory under Contract No. W-31–109-ENG-38 with the U.S. Department of Energy.