Genomic and Genetic Control of Phosphate Stress in Legumes

Phosphorus (P), an essential element for growth and development, is taken up by plants as phosphate (Pi), but Pi is unevenly distributed and relatively immobile in soils. As a result, more than 30% of the world's arable land requires the application of P fertilizers for cropping (Vance et al., 2003). Unfortunately, P fertilizers are manufactured from nonrenewable resources that are increasingly becoming more costly and less available. Current estimates indicate that easily mined rock Pi reserves could easily be depleted by 2060 (Steen, 1997; Vance, 2001; Vance et al., 2003). Paradoxically, part of the applied P in intensive cropping systems can enter the waterways through runoff and erosion, contributing to pollution of surrounding lakes and marine environments. Improving P acquisition and use by crops is critical to economical and environmentally friendly crop agriculture. Plants have evolved a variety of adaptive strategies to improve their acquisition, use, and remobilization of P (Vance et al., 2003; Hammond et al., 2004; Lambers et al., 2006). Plant responses to P stress conditions involve changes in root morphology and architecture (Lynch, 1995; Liao et al., 2001; Lynch and Brown, 2001; Yan et al., 2004; Beebe et al., 2006; Hill et al., 2006; Ochoa et al., 2006), as well as changes in shoot and flower development (Bucciarelli et al., 2006). Among the legumes, white lupine (Lupinus albus), common bean (Phaseolus vulgaris), and to a lesser extent barrel medic (Medicago truncatula) and soybean (Glycine max) have been the focus of P stress research. White lupine, a nonmycorrhizal species, is adaptable to scarce P and displays a highly synchronous suite of molecular and biochemical adaptations to P stress by developing proteoid (cluster) roots, increasing organic acid exudation, and enhancing the expression of many genes, such as secreted acid phosphatase (LaSAP1) and Pi transporters (LaPT1; Vance, 2001; Uhde-Stone et al., 2003a, 2003b; Vance et al., 2003, and refs. therein). Common bean is the most important food legume worldwide, and genetic variability for the capacity to produce grain in low soil P conditions has been documented (Broughton et al., 2003; Ochoa et al., 2006). Moreover, several thousand ESTs derived from P-stressed common bean roots have been characterized (Ramírez et al., 2005). Noteworthy in an accompanying article in this legume focus issue, Hernández et al. (2007) have completed a P stress root transcriptome survey in common bean, identifying some 125 genes responsive to P stress. Recent studies of barrel medic, a model legume for plant biology research, showed that P stress delayed: (1) leaf development and leaf expansion along the main and axillary shoots; (2) axillary shoot emergence and elongation, resulting in stunted plants; and (3) timing and frequency of flower emergence (Bucciarelli et al., 2006). P-stressed barrel medic also formed shorter petioles and shorter blade lengths relative to plants in P-sufficient conditions. Whether or not morphological changes seen in P-starved barrel medic plants are attributable to an overall delay in whole plant development or as a P stress response remains to be seen. However, the lack of a standardized approach to describe plant growth and phenotypic responses to P stress (Bucciarelli et al., 2006), together with the plastic nature of plant morphological traits (Beebe et al., 2006; Ochoa et al., 2006), makes result comparisons from different laboratories difficult. Because of the subterranean nature of growth, plant roots have been recalcitrant to phenotypic study. Root adaptations to P limitations include reduced extension of primary roots, highly branched roots with increased lateral roots, and an increased density of root hair formation. Consistent with a general stress response by plants, however, P-stressed plants tend to allocate a greater proportion of biomass to root dry matter compared to P-sufficient plants (López-Bucio et al., 2002, 2003; Hammond et al., 2004; Hill et al., 2006). Most pasture species studied showed reduced total root mass as a response to P stress conditions. On average, a 32% to 86% reduction in root mass was observed in most pasture species with decreasing P concentrations (Hill et al., 2006). Owing to inherent root architecture differences, some pasture species did not respond with root mass reduction to P stress (Hill et al., 2006). Similarly, there were no root architecture differences between barrel medic plants grown under P-sufficient and P-deficient conditions until 28 d after planting, when lateral root length and number of P-limited plants showed a decline (Bucciarelli et al., 2006). By contrast, alfalfa (Medicago sativa) roots show changes in architecture when grown under P stress. Genetic regulation of root architecture changes due to P stress within and among species is not understood and offers a fruitful area of emphasis for future research. Molecular genetic, biochemical, physiological, and morphological responses of plants subjected to P stress have been the subject of many recent reviews (Vance, 2001; López-Bucio et al., 2003; Vance et al., 2003; Franco-Zorrilla et al., 2004; Hammond et al., 2004; Raghothama and Karthikeyan, 2005; Lambers et al., 2006). This Update summarizes the ongoing plant biology research toward the understanding of P stress in legumes such as white lupine, common bean, barrel medic, and soybean. Additional studies using model plant systems such as Arabidopsis (Arabidopsis thaliana) and rice (Oryza sativa) have also provided valuable genomic and genetic evidence in understanding plant responses and adaptations to P limitations (Hammond et al., 2003; Yi et al., 2005; Aung et al., 2006; Bari et al., 2006; Müller et al., 2007). Genetic variability with contrasting degree of root architecture responses to P-limiting conditions has been known for a wide range of plant species (Chevalier et al., 2003; Rubio et al., 2003; Yan et al., 2004). Lynch (1995) has noted a direct correlation between plant productivity and root architecture. As a result, P stress tolerance and adaptation have begun to be analyzed in common bean and Arabidopsis through the identification of quantitative trait loci (QTL) approach (Beebe et al., 2006; Ochoa et al., 2006; Reymond et al., 2006). Ochoa et al. (2006) generated recombinant inbred lines (RIL) from a cross of two common bean accessions with contrasting root architecture traits for adventitious roots. Screening 86 F5:7 RIL under P stress and P-sufficient conditions resulted in the identification of 19 QTLs for adventitious root formation (Ochoa et al., 2006). Because Pi availability is expected to be greater in topsoil compared to subsoil, selection for root trait QTL markers associated with adventitious rooting and topsoil foraging may enhance P acquisition. In a previous study, QTL analysis applied to RIL of a cross of G19833 and DOR 364 common beans showed that root hair formation and root organic acid exudation are important traits for marker-assisted selection and breeding of P stress tolerance and adaptation (Yan et al., 2004). Common bean G19833 is a landrace of the Andean gene pool with superior growth and yield in P stress conditions, and DOR 364 is a Mesoamerican gene pool with low P accumulation efficiency in P-limiting conditions. Recently, a composite interval mapping approach identified 26 more QTLs associated with basal root development and greater P acquisition efficiency in P stress conditions (Beebe et al., 2006). In a recent study involving a RIL population of Arabidopsis, three QTLs involved in root growth response to P stress were identified (Reymond et al., 2006). One of the QTLs, LPR1, explained 52% of the variation associated with primary root length response. QTL analysis of P stress tolerance appears to be a useful approach in determining which root traits are associated with P uptake. With the soon-to-be-completed sequencing of the genomes of Medicago and Lotus accompanied by the strikingly conserved synteny among legume genomes, using positional cloning, it should be possible to identify specific genes that contribute to QTLs affecting adaptation to P stress. The selection and development of P-efficient legume plants using QTL markers would not only be beneficial to low-input agricultural systems but also would enhance environmentally friendly cropping in intensively cultivated systems. Following international collaborations in recent years, more than 25,000 partially sequenced cDNA inserts or ESTs derived from P-starved tissues of four legume species (barrel medic, soybean, common bean, and white lupine) are currently deposited in the public domain (http://compbio.dfci.harvard.edu/tgi/). Microarray and macroarray analysis of P stress in plants showed increased transcript abundance of genes with homology to Pi transporters, organic acid synthesis, purple acid phosphatase, mulitdrug and toxin efflux (MATE), transcription factors, signaling, and defense (Hammond et al., 2003; Uhde-Stone et al., 2003a, 2003b; Wu et al., 2003; Misson et al., 2005; Ramírez et al., 2005; Müller et al., 2007). Gene indices at http://compbio.dfci.harvard.edu/tgi/ derived from EST sequencing efforts of P-stressed tissues have been used as tools for gene discovery, molecular marker generation, and gene transcript pattern analysis. By evaluating available microarray and macroarray data and utilizing bioinformatic analysis of publicly available EST sequencing projects, Graham et al. (2006) identified 52 candidate genes clustered in 22 groups that appear to respond in common to P stress in four legume species and Arabidopsis. This in silico analysis identified P stress-responsive genes that are overrepresented in the gene indices. Transcripts identified annotate to various important functional categories, including MYB and WRKY transcription factors, signal transduction proteins (Ser/Thr kinases, mitogen-activated protein kinases, and calcium-dependent protein kinases), transporters (Pi transporter and ATP-binding cassette transporter family), and purple acid phosphatases. Current research is aimed at: (1) using P stress-induced ESTs in marker-assisted selection for genotypes having improved tolerance to P deficiency; (2) characterizing the functional significance of P stress-induced genes; and (3) identifying candidate genes that may be used to enhance P efficiency. Research in Arabidopsis and rice suggests that an important step in regulation of gene expression during plant stress appears to be the transcriptional activation or repression of genes (Chen et al., 2002; Wu et al., 2003). Transcription factors are key global regulators of gene expression and are known to play critical roles in many biological processes, including the regulation of plant responses to numerous biotic and abiotic stresses (Rubio et al., 2001; Tang et al., 2001; Chen et al., 2002; Singh et al., 2002). In Arabidopsis alone, approximately 6% (about 1,800) of the total number of genes are composed of transcription factors, including about 72 WRKY family of genes, more than 600 zinc finger proteins, and 133 MYB transcription factors (Eulgem et al., 2000; Riechmann et al., 2000; Stracke et al., 2001; Guo et al., 2005). In a microarray analysis, approximately 30% of the 333 transcription factor genes included in the array were up- or down-regulated 2-fold or more during P stress in Arabidopsis (Wu et al., 2003). Misson et al. (2005) and Müller et al. (2007) also reported up to 80 P stress-responsive transcription factor genes in Arabidopsis. P stress-responsive transcription factors belong to several families, including MYB, SCARECROW, APETALA2 domain, homeobox, WRKY, and zinc fingers. A recent bioinformatic analysis of legume gene indices and for genes overrepresented in P-stressed the of several transcription factor genes, including WRKY, MYB, and zinc finger family of genes et al., 2006). and root tissues showed of transcription factor genes (Wu et al., 2003). In of genes that may as and genes during P stress were observed (Hammond et al., 2003; Misson et al., 2005). Recently, a transcription factor involved in Pi stress in rice was and characterized et al., 2005). is in but transcript accumulation was in roots during P of rice plants et al., 2005). P-limiting conditions, of in rice using the resulted in increased P compared to rice et al., 2005). rice also total root length and root resulting in a 30% root and shoot biomass than rice et al., 2005). gene to gene showed in lateral roots, primary root and of rice in P-limited conditions et al., 2005). transcription factors involved in gene expression include the protein in soybean et al., and the MYB transcription factor in Arabidopsis (Rubio et al., the protein to the in acid et al., the Arabidopsis MYB transcription factor with homology to a Pi response gene in was to to an (Rubio et al., transcript abundance of gene was not by Pi (Rubio et al., a recent that a in Arabidopsis et al., 2006). is evidence the regulation of P stress-responsive genes in However, and have observed the within the of many P stress-induced genes, including the white lupine and et al., 2001; et al., 2001; Hammond et al., 2003; Müller et al., 2007). is that of in the common bean, and soybean gene indices. The gene at The for Research ESTs that show homology to zinc finger transcription ESTs were clustered in two and and derived from P stress roots. analysis of ESTs zinc finger transcription factors, observed increased transcript abundance for two of the ESTs in P-starved roots of common bean and Vance, The functional of zinc finger transcription factors remains to be expression of the WRKY family of transcription factor genes in common bean roots. of total was from roots of plants grown under P-sufficient or P-deficient conditions as (Ramírez et al., 2005). and molecular a of that may play important roles in the of many in plants and 2004; and 2004; et al., 2005; Chen et al., 2006). are such about to in length in plants, which as regulators or through to or partially to the et al., 2002; et al., 2002; 2004; and 2004). Most known in plants are to the expression of several of genes, including transcription factors, their in various plant and 2004). Recently, identified in Arabidopsis and rice and was to be by P stress after and of P et al., 2005; et al., 2006). abundance of the of Pi in the et al., and is not at under P-sufficient conditions et al., 2005; Aung et al., 2006; Bari et al., 2006; et al., 2006). On the is to show the gene through et al., 2005; et al., 2006). transcript accumulation of the gene was in P-starved Arabidopsis plants et al., 2005). expression of in Arabidopsis using the resulted in of under P-sufficient conditions and of Pi in Arabidopsis et al., 2005; Bari et al., 2006; et al., 2006). The observed were to by Arabidopsis of the gene from et al., 2005; et al., 2006), as well as to a of a from an pool of Arabidopsis et al., 2006; Bari et al., 2006). a toward the of gene was to to the domain of et al., 2006). As the in was to be a result of a that resulted in an of the were not in of the et al., 2006; Bari et al., 2006). The gene about a is to a including a et al., 2006). Consistent with the of in also for in the et al., 2006). transcript abundance of was not by stress conditions such as or et al., 2005). have that and that under P stress appears to Pi by the expression of in Arabidopsis. A is in common bean P-stressed roots. and have been identified in barrel medic and Lotus and 2004; Bari et al., 2006). A for has been for the transcription factor gene et al., and have noted EST in white lupine P-stressed proteoid root expression and Vance, also have observed expression of a gene gene in P-starved roots of barrel medic and alfalfa and Vance, gene for lateral root development by the expression of two genes, and in Arabidopsis et al., remains to be seen or plant play a in the regulation of genes in The P stress response of plants is to be a and it be to are the of such and have that increased transcript abundance of several genes with P in white lupine and accumulation of white lupine and genes was increased in and in and roots of P-stressed plants, there was no transcript accumulation of genes in P-sufficient plants et al., 2005; Uhde-Stone et al., 2005). The direct of and genes in P has been by studies using lupine and alfalfa plants that the and to a gene et al., 2005; Uhde-Stone et al., 2005). of alfalfa plants grown under P showed and to P-sufficient alfalfa and lupine plants, no was in root of plants grown in P-sufficient abiotic stress such as or of did not produce or show gene in plants and However, the expression of the gene was not to P as increased transcript accumulation of in proteoid roots was during as well as in and stress conditions et al., 2005). Pi stress and expression of genes in proteoid roots of white from proteoid roots of P-deficient white lupine plants at d after emergence under different plants in for plants to for of the shoot of plants to and the in the conditions were et al., 2005). The of Pi and root architecture and root growth have been studied in Arabidopsis. In Arabidopsis, application of Pi and in growth root growth and root architecture et al., 2005; et al., 2006). of Arabidopsis by up to when grown in or conditions et al., 2006). root density increased as as was in P-starved Arabidopsis root density of Arabidopsis was by in in P stress conditions and by in in P-sufficient conditions et al., 2006). have the between and P stress in A signal for Pi in the shoot and the regulation of genes to be at in by the of in A recent microarray analysis by Müller et al. (2007) lines of evidence for a cross between P acquisition and was that more than genes were up- or down-regulated 2-fold or more in Arabidopsis leaf et al., 2007). a number of genes showed expression by Pi and et al., an between Pi and in gene expression during P white lupine genes, including and also showed expression to and genes involved in and et al., 2005). it is that many genes in response to P stress for it remains to be The plant acid has been in the regulation of many of plant growth and root development, including P stress-induced proteoid (cluster) root development et al., et al. (2005) showed that during Pi in the whole primary root and in lateral roots of Arabidopsis. only primary root growth was observed in Arabidopsis et al., 2006). of Pi application of lateral root formation and primary root in Arabidopsis (López-Bucio et al., 2002; et al., 2006). The of lateral root formation was under P-sufficient conditions that did not et al., 2006). application of to P-sufficient white lupine proteoid root formation as seen under P-deficient conditions et al., to P-deficient plants reduced the formation of roots. The data that root development in response to P in white lupine is by In Arabidopsis, a gene is at in root and lateral root and at low in and et al., Arabidopsis plants showed reduced lateral root and the lateral root of could not be observed in roots of plants et al., expression of cDNA in Arabidopsis with increased and gene expression and the of lateral roots et al., that and gene expression could have lateral root in Arabidopsis. applied resulted in more lateral roots formed in plants than remains to be seen the gene has a in root architecture traits associated with P stress adaptations in P stress is known to in many plant species, including bean et al., lupine et al., and Arabidopsis et al., 2003). the of root in bean and Arabidopsis. P application primary and lateral root but not lateral root P can lateral root density to Arabidopsis Root hair formation appears to be in by of in an in root hair density and length et al., 2001; López-Bucio et al., 2003). is that bean, lupine, and barrel medic plants to P stress have increased density and length of root genes are to be involved in root hair development et al., Graham et al. (2006) reported that a key acid in is overrepresented in the ESTs derived from P-stressed roots of lupine, bean, and taken together indicate that plant to a in root adaptation to P The of in root growth and P stress is not are to be regulators of root growth having shoot growth et al., 2003; et al., 2006). of root development and the of increased lateral root Plants that the genes have reduced and show root growth due to more lateral and adventitious root formation et al., 2004). P and result in (López-Bucio et al., accompanied by increased lateral root formation. applied the expression of P stress-induced genes in Arabidopsis roots et al., In P-stressed lupine proteoid roots, gene expression showed a to in expression (Vance et al., 2003). Moreover, application of to P-deficient white lupine proteoid root and is increased in proteoid roots et al., et al., (2006) have a for lateral root in P-sufficient plants that the of and that factors that root and to at lateral root to This would be with the that low P has of primary root growth, or root accompanied by increased lateral root formation. is that changes in root in response to stress. However, the plants P stress signaling, and gene to which evolved with land plants more than are the most important adaptation for plants to scarce P 2004; 2005; et al., 2005). than of plants have A for developing barrel medic, and soybean as model species is due to the that have with and to the from signal between the and plant 2005; et al., 2007). The legume in root that growth, and formation et al., 2005). plant are to formation in root The root growth and from roots the soil the to P and for plant growth, the plant the with The of and between the within root in which The plant and with due to their by the which is an extension of the As by this is an highly between the legume and expression of genes from two different A of is the of this and the is to several recent reviews 2004; 2005; et al., 2005; and et al., et al., 2007). only of genomic studies of be In recent years, several genetic loci in legumes that root and have been characterized through positional loci were as root also 2004; 2005; et al., 2005). Genetic loci identified to and a a and a of loci have been identified in species et al., 2006). Moreover, two proteins with to and and in have been to be for et al., 2005). genetic loci affecting have been identified in various legumes, but et al., 2005). however, that and root some common In there is a of loci that to legume of and in silico of EST a number of plant genes have been identified that respond to and et al., 2002; et al., 2003; et al., 2003; et al., 2004; et al., 2004; et al., 2004; et al., 2005; et al., 2007). gene were identified as highly during and MYB and zinc finger and the functional significance of most of the proteins in legume roots in response to has not been utilizing and functional significance to be the transporter genes from et al., and from barrel medic et al., were to be for Similarly, a calcium-dependent protein that root development in barrel medic was to et al., 2005). functional only the of the As more genes to are the would be to identify genetic strategies that and P acquisition. P is to as the most element for plant With the availability of a wide array of genomic and bioinformatic research P stress research is toward an signal regulation of gene and increased efficiency of cropping that plant identify and in plants that improve P acquisition and P stress adaptations to plants that are at that improve soil P availability to plants contribute to the of economical and environmentally friendly crop agriculture. data from this article can be in the data under and and and for This is a from the Plant Research and the of a or not a or of the by the and not or the of and that also be