Quantitative Trait Loci, Epigenetics, Sugars, and MicroRNAs: Quaternaries in Phosphate Acquisition and Use

Phosphorus (P) is a critical element for plant growth and is frequently the limiting nutrient in many soils. Continued production and application of P fertilizer relies on a nonrenewable resource that will peak in about 2050. This will result in significantly increased cost, particularly for developing countries. Significant research efforts in the genomics of P stress have shown that many suites of genes regulated in a coordinated fashion are involved in plant acclimation to P deficiency. These genomic studies, in conjunction with traditional plant breeding, have shown that P-acclimation traits are controlled by multiple genes, most probably in quantitative trait loci (QTLs). Future development of near isogenic lines (NILs) and recombinant inbred lines (RILs) coupled to next-generation sequencing will facilitate the cloning of genes in QTLs regulating P-deficiency acclimation. Defining the role of epigenetic regulation of gene expression in adaptation to abiotic stress will provide new targets for improving plant adaptation to P starvation. Cross talk between sugars, microRNAs (miRNAs), and P-starvation-induced gene expression may be significant to understanding the fundamental underpinning of plant adaptation to nutrient stresses. Plants with highly efficient P acquisition and use could reduce the need for P fertilizer in the developed world, thereby ameliorating the overuse of P while concurrently enhancing yield in the developing world, where P is frequently unavailable.

IMPORTANCE OF P

P is one of 17 essential elements (nutrients) required for plant growth (Tiessen, 2008; Cordell et al., 2009). The P concentration in plants ranges from 0.05% to 0.50% dry weight. The concentration gradient from the soil solution P to plant cells exceeds 2,000-fold, with an average free P of 1 to 5 μm in the soil solution (Bieleski, 1973; Schachtman et al., 1998). Although bound P is quite abundant in many soils, it is largely unavailable for uptake, and P is frequently the most limiting element for plant growth and development (Tiessen, 2008; Cordell et al., 2009). The acid-weathered soils of the tropics and subtropics are particularly prone to P deficiency.

Mined rock phosphate, a nonrenewable resource, is the primary source of P fertilizer (Steen, 1998; Tiessen, 2008; Cordell et al., 2009). Easily mined, high-quality rock phosphate sources are projected to be depleted within 30 to 50 years. In addition, the world’s major reserves of rock phosphate are located in Morocco and China. Uncertain political issues could limit access to the world’s P resources. Coalescence of these factors as well as production of food for energy has seen P-fertilizer costs increase 6- to 9-fold in the past few years. A potential phosphate crisis looms for agriculture in the 21st century (Abelson, 1999; Tiessen, 2008).

Application of P fertilizer, however, is problematic for both the intensive and extensive agriculture of the developed and developing worlds, respectively. Under adequate P fertilization, only 20% or less of that applied is removed in the first year’s growth. This results in P loading of prime agricultural land and increased P runoff (Kirkby and Johnston, 2008; White and Hammond, 2008). An even greater concern is the lack of available P fertilizers for extensive agriculture in the tropics and subtropics, where the majority of Earth’s people live. Lack of infrastructure, money, and transportation make P fertilization unattainable for these areas. It is imperative that within the next 40 years plant biologists understand the genetic basis for acclimation to P deficiency and through traditional selection and/or biotechnology develop germplasm sources with improved P-use efficiency.

Plant responses to P-stress conditions involve changes in both shoot and root development (Lynch, 1995; Lynch and Brown, 2001; López-Bucio et al., 2003; Bucciarelli et al., 2006; Lambers et al., 2006; Hernández et al., 2007). P-deficient plants display (1) delayed leaf development and reduced photosynthetic capacity, (2) reduced axillary shoot emergence and elongation (stunted plants), (3) impaired flower development, (4) increased anthocyanin accumulation, (5) increased root-shoot ratio, (6) altered root architecture, and (7) increased exudation from roots of organic acids, phenolics, protons, and enzymes.

Because roots are the primary site for acquiring P, they have become a rich topic for developmental and genetic studies. Soil P limitation is a primary effector of root architecture (Dinkelaker et al., 1995; Williamson et al., 2001; López-Bucio et al., 2003; Lambers et al., 2006) and is known to impact all aspects of root growth and development. Phenotypic and genotypic adaptations to P deficiency involve changes in root architecture that facilitate the acquisition of P from the topsoil. Adaptations that enhance acquisition of P from topsoil are important because of the relative immobility of P in soil, with the highest concentrations usually found in the topsoil. Lynch and Brown (2001) refer to P-deficiency-induced modifications of root architecture as adaptations for topsoil foraging. Root characteristics associated with improved topsoil foraging for scarce P are a more shallow horizontal basal root growth angle, increased adventitious root formation, enhanced lateral root proliferation, and increased root hair density and length. The phenotypic complexity of plant acclimation to P deficiency reflects the polygenic nature of these processes.

In recent years, significant inroads into understanding biochemical and molecular events involved in plant acclimation to P stress have been made through comparative microarray and macroarray studies of P-deficient plants as compared with P-sufficient plants. In addition, definition of single-gene mutants in Arabidopsis (Arabidopsis thaliana) that show impaired P signaling has demonstrated the role of transcription factors involved in P-induced gene expression. The importance of miR399 in P signal transduction has also been a fruitful avenue of research. In coming years, complementary research approaches that utilize next-generation sequencing coupled to analysis of well-defined plant germplasm containing QTLs for enhanced P-stress tolerance will lead to the identity of genome regulatory elements. Moreover, plants exposed to biotic stress, such as diseases, undergo epigenetic changes leading to resistance responses, but the importance of epigenetic mechanisms in adaptation to abiotic stress is an underdeveloped discipline. Lastly, while sugars can act as signal molecules in several developmental responses and have been implicated in P-stress responses, it is unclear how sugars moderate the transcriptional regulation of gene expression. This contribution will address the implications of next-generation sequencing, QTLs, epigenetics, and sugar-miRNA cross talk in plant acclimation to P deficiency.

QTLS ASSOCIATED WITH P-STRESS TOLERANCE AND ACCLIMATION

Genetic variability for the complex root and shoot responses to P-limiting conditions has been demonstrated for a wide range of species. Variation for complex phenotypic traits are frequently controlled by many genetic loci, QTLs, scattered throughout the genome (Price, 2006). QTLs for traits related to P-deficiency tolerance have been found in rice (Oryza sativa; Wissuwa et al., 2002), wheat (Triticum aestivum; Su et al., 2009), common bean (Phaseolus vulgaris; Beebe et al., 2006; Ochoa et al., 2006), Arabidopsis (Reymond et al., 2006), soybean (Glycine max; Li et al., 2005; Zhang et al., 2009), barley (Hordeum vulgare; Gahoonia and Nielsen, 2004), and maize (Zea mays; Zhu et al., 2005; Chen et al., 2009; Hochholdinger and Tuberosa, 2009).

Utilizing a mapping population derived from a cross of the intolerant ‘Nipponbare’ cultivar with the tolerant landrace ‘Kasalath,’ Wissuwa et al. (2002) identified a QTL for tolerance to low P in rice designated Phosphorus uptake1 (Pup1). Introgression of Pup1 into NILs allowed fine mapping of the Pup1 locus to the long arm of chromosome 12 (15.31–15.47 Mb) on the basis of the Nipponbare reference genome (Heuer et al., 2009). Next-generation sequencing was used to characterize the genomic (278 kb) introgression region. Although the gene(s) regulating the Pup1 phenotype has yet to be identified, it is worthwhile to note that the Pup1 region is found in 50% of the rice accessions adapted to stress-prone environments (Chin et al., 2010).

In common bean, RILs were developed for shallow-rooted and deep-rooted phenotypes (Rubio et al., 2003). Under field conditions, when available P was concentrated in the topsoil layer, the shallow-rooted RILs were more productive and had a competitive advantage over deep-rooted RILs. Further analysis of the RILs by Liao et al. (2004) showed 16 QTLs controlling the root traits. Adventitious root formation in 84 RILs grown under limiting P conditions was shown to be important for P acquisition in common bean (Ochoa et al., 2006). The QTLs for root traits related to low P tolerance were mainly located on linkage groups B2 and B9. One QTL on linkage group B9 accounted for 61% of the total phenotypic variation. Beebe et al. (2006) evaluated 71 RILs grown under either low P or high P and identified 26 QTLs affecting P accumulation and root architecture. Enhanced P uptake was associated with basal root development. Development of NILs of common bean having enhanced P tolerance and shallow basal roots will provide the germplasm resources for further fine mapping of important traits. The common bean genome is currently being sequenced. Having a reference genome and NIL genetic resources for common bean coupled with next-generation high-throughput RNA sequencing (RNA-seq) will facilitate fine mapping and candidate gene(s) identification regulating P tolerance and root traits in common bean.

Tuberosa and Salvi (2007) developed a library of maize introgressed lines from B73 (recurrent parent) crossed with Gaspe Flint (donor parent) to identify major QTLs for root architecture and growth that map to maize bin 1.06. Utilizing other maize RILs grown under low and high P, Chen et al. (2008) identified a QTL at bin 1.06 for P efficiency and topsoil root dry weight. They found several QTLs for interactions (epistasis) located near bin 1.06. The bin 1.06 region on maize chromosome 1 has been reported to control a QTL for root architecture in five maize genetic backgrounds (Hochholdinger and Tuberosa, 2009). The QTL at bin 1.06 has also been associated with nitrogen-use efficiency. This region of the maize genome is currently under evaluation for candidate genes.

Building upon a number of approaches, a gene has been identified in Arabidopsis that is involved in a QTL controlling root growth response (Reymond et al., 2006; Svistoonoff et al., 2007). As noted earlier, when grown on low P, Arabidopsis primary root growth is inhibited while lateral root growth is stimulated. Reymond et al. (2006) generated NILs by segregating an F6 RIL mapped for root growth response to low P. Fine mapping of the low-phosphate root (LPR1) QTL trait located it to a 2.5-Mb region at the top of chromosome 1. Further fine mapping by Svistoonoff et al. (2007) refined the LPR1 trait to a 36-kb region. They then mutagenized the LPR1 line and screened for progeny with long roots on low-P medium. Moreover, Svistoonoff et al. (2007) also developed T-DNA insertion mutants for LPR1. Utilizing fine mapping of the locus and mutagenesis, they defined the LPR1 QTL locus as encoding a multicopper oxidase enzyme. Transcripts for LPR1 were most abundant in the root meristem and root cap. Ticconi et al. (2009) have recently proposed that LPR1 interacts with phosphate deficiency 2 (PDR2) to adjust root meristem activity.The authors suggested that the root cap played a critical role in local P sensing.

As evidenced by the cloning of LPR1 in Arabidopsis, complementary approaches will lead to cloning of genes controlling QTLs involved in tolerance to P deficiency. Progress on cloning these QTLs will be dependent upon several factors, including (1) the development of RILs segregating for the desired trait, followed by the development of NILs having the introgressed trait; (2) fine mapping of the trait utilizing molecular markers, derived preferably from single nucleotide polymorphisms generated by RNA-seq comparisons of the NILs; (3) mutations in or near the locus; and (4) a reference genome. In the near future, low-cost next-generation sequencing will rapidly advance cloning of QTLs for not only P deficiency tolerance but also other traits.

ACCLIMATION TO PHOSPHATE STRESS INVOLVES EPIGENETIC CHANGES?

Acclimation and resistance to abiotic and biotic stresses involve significant biochemical and developmental plasticity. While much of this plasticity is the direct result of either increased or decreased transcription of several suites of genes, some may be derived from epigenetic modifications that alter gene expression (Lukens and Zhan, 2007; Zhang, 2008; Chinnusamy and Zhu, 2009). In recent years, epigenetic changes have been noted as salient features in adaptation to abiotic and biotic stresses. Changes in DNA methylation are a hallmark of the epigenetic regulation of gene expression (Boyko and Kovalchuk, 2008; Zhang, 2008). In plants, DNA hypermethylation is generally linked with repressive chromatin in gene promoters and repression of gene expression, while hypomethylation leads to enhanced transcription. Enhanced expression of a glycerophodiesterase (GPXPD) gene in tobacco (Nicotiana tabacum) in response to aluminum, salt, and cold stresses has been associated with demethylation in the coding region of the GPXPD gene (Choi and Sano, 2007). These authors, however, did not find demethylation in the promoter region of the gene. Many genes showing enhanced transcription in response to aluminum stress are also induced during P stress. As with aluminum stress, GPXPDs are known to be highly expressed in response to P starvation and return to basal levels as P stress is relieved (Misson et al., 2005; Morcuende et al., 2007). The methylation status of GPXPDs has not been evaluated during P stress, but it as well as many other genes that respond quickly to P status may be under a similar form of epigenetic regulation. With the soon to be available single-molecule real-time DNA sequencing capabilities, laboratories will be able to do direct sequencing for methylated DNA rather than bisulfite sequencing (Flusberg et al., 2010). This will allow for immediate and direct evaluation of the epigenetic status of plants grown under any stress condition.

Smith et al. (2010) have implicated the actin-related protein 6 (APR6) as an epigenetic modulator of some P-starvation response (PSR) genes. APR6 is a key component of the SWR1 complex involved in chromatin remodeling and is required for histone H2A.Z incorporation into chromatin (Jarillo et al., 2009). The physiological and molecular phenotypes of apr6 mutants were noticeably similar to those displayed by P-starved Arabidopsis plants (Smith et al., 2010). Loss of function of APR6 resulted in a dramatic decrease in H2A.Z abundance at several PSR gene sites accompanied by an increase in gene transcription. Chromatin remodeling is an integral mechanism in regulating yeast structural phosphate regulon (PHO) gene expression (Barbaric et al., 2007; Wippo et al., 2009). In addition, chromatin remodeling has also been implicated as a component of adaptation to pathogen stress in Arabidopsis (March-Diaz et al., 2008).

Another epigenetic mechanism involved in adaptation to stress appears to involve posttranslational modifications to the N-terminal region of nucleosome core complex histones through acetylation, phosphorylation, ubiquitination, and sumolaytion (Boyko and Kovalchuk, 2008; Chinnusamy and Zhu, 2009). The WD-40 protein gene HOS15 of Arabidopsis has been shown to be important in histone deacetylation and is crucial for the repression of genes associated with plant acclimation and tolerance to cold stress (Zhu et al., 2008). HOS15 mutants accumulate higher amounts of stress-related transcripts and are hypersensitive to cold temperatures. Phosphorylation of histone H3,S10 and acetylation of histone H4 is correlated with increased abundance of salt tolerance transcripts in tobacco and Arabidopsis (Sokol et al., 2007). The limited information available on the epigenetic regulation of plant response to P deficiency makes this topic rich for exploration.

SIGNALING OF PHOSPHATE STRESS: SUGARS AND MIRNAS CROSS TALK?

The plethora of biochemical and developmental adaptations displayed in plants subjected to P deficiency result from both local and systemic signaling, which activates the coordinated expression of a medley of genes (Franco-Zorrilla et al., 2004; Müller et al., 2007; Tesfaye et al., 2007; Hammond and White, 2008). Suc derived from photosynthate and miRNAs have been implicated as critical molecules signaling P status of the plant. Chiou and Bush (1998) showed that Suc could act as a signal molecule in assimilate partitioning. A growing body of evidence now supports Suc derived from photosynthate as part of the systemic signaling leading to P-deficiency-induced increase in lateral root formation and increased root hair density (Hermans et al., 2006; Jain et al., 2007; Karthikeyan et al., 2007; Zhou et al., 2008). Moreover, Suc has been shown to be required for enhanced expression of P-starvation-induced genes. To test the role of photosynthate and phloem Suc on P-stress transcript induction, shoots of white lupin (Lupinus albus) plants were either darkened or stems were girdled to block phloem transport, and the starvation-enhanced expression of genes in roots was evaluated (Liu et al., 2005; Tesfaye et al., 2007). Both treatments reduced gene expression in P-stressed roots to nondetectable levels within a few hours. Returning darkened plants to light restored P-starvation-induced gene expression in roots. In P-stressed Arabidopsis roots, P-starvation-induced genes showed further enhanced expression when supplemented with 3% Suc (Franco-Zorrilla et al., 2005; Karthikeyan et al., 2007). Müller et al. (2007) evaluated the interaction between P and Suc in Arabidopsis leaves. Using a 2-fold cutoff, they found that 187 transcripts responded to P starvation while 644 responded to Suc. They identified 149 transcripts that were regulated by the interaction between P starvation and Suc availability. One group of 47 genes having increased expression in response to P deficiency was further enhanced by Suc. Many of the transcripts in this group encode proteins involved in P remobilization and carbohydrate metabolism. Although Suc appears to be important in signaling P status and the full expression of P-starvation genes, the mechanism remains elusive. The sucrose nonfermenting1 kinase:calcineurin B-like protein kinase (SNF1:CIPK) pathway has been implicated as the transduction system for sugar signaling (Hummel et al., 2009; Rosa et al., 2009). Whether the SNF1:CIPK pathway regulates sugar signaling during P starvation deserves attention.

Computational and molecular cloning approaches revealed a group of endogenous noncoding small RNAs that may play important roles in the control of many developmental processes in plants and animals (Bartel, 2004; Jones-Rhodes and Bartel, 2004; Xie et al., 2005). miRNAs are noncoding small RNAs, about 20 to 24 nucleotides in length in plants, that function as posttranscriptional negative regulators or repressors through base pairing to complementary or partially complementary sequences in target mRNAs, leading to cleavage of that RNA. Most known miRNAs in plants are predicted to target the expression of several classes of genes, including transcription factors, indicating their importance in regulating various plant developmental aspects (Bartel and Bartel, 2004). Recently, miR399, first identified in Arabidopsis and rice (Sunkar and Zhu, 2004), was shown to be induced by P stress after 24 and 48 h of P starvation (Fujii et al., 2005; Aung et al., 2006; Bari et al., 2006; Chiou et al., 2006). Transcript abundance of miR399 declines rapidly following the addition of P in the medium (Bari et al., 2006) and is not detected at all under P-sufficient conditions. Chiou et al. (2006) found that while miR399 was highly up-regulated during P stress, transcripts for a ubiquitin-conjugating E2 enzyme (UBC24) were reduced by 5-fold. By comparison, plants grown under P sufficiency had significantly reduced miR399 but UBC24 was abundant. Computational analysis of the 5′ upstream region of UBC24 showed several target-binding sites for miR399. Overexpression of miR399 suppressed the accumulation of UBC24 transcripts and resulted in enhanced accumulation of P in the shoot. Moreover, P remobilization was impaired in plants overexpressing miR399. Enhanced accumulation of P in shoots and impaired P remobilization in plants overexpressing miR399 phenocopied the Arabidopsis pho2 mutant (Dong et al., 1998).

In concurrent studies, Bari et al. (2006) using map-based cloning identified the impaired gene in the pho2 mutant as the ubiquitin-conjugating enzyme UBC24. They also found that UBC24 had five complementary miR399-binding sites in the 5′ upstream region. Bari et al. (2006) extended the understanding of the PHO2-miR399 interaction by showing that plants with the phr1-MYB protein mutation failed to accumulate miR399 transcripts. Their observations showed that PHR1-MYB was required for miR399 expression. Thus, PHO2, miR399, and PHR1 define a critical phosphate signaling pathway (Doerner, 2008). The MYB transcription factor PHR1 regulates miR399 expression, which in turn regulates the UBC24 E2 ligase PHO2. PHO2 then regulates a subset of P-starvation genes. Studies have now clearly shown that miR399 can move in the phloem sap and serve as a long-distance signal for phosphate homeostasis (Lin et al., 2008; Pant et al., 2008). A novel genetic concept superimposed upon miR399 and P signaling has been demonstrated by Franco-Zorrilla et al. (2007). They found that along with transcriptional control, miR399 activity is also regulated by “target mimicry.” The non-protein-coding gene Induced by Phosphate Starvation (IPS1) contains a motif with sequence complementarity to miR399. IPS1 is induced along with miR399 during P stress. However, rather than being cleaved by miR399, IPS1 transcripts can bind to and sequester miR399, thereby acting to attenuate the inhibitory activity of miR399. Target mimicry by IPS1 and other non-protein-coding genes may be a mechanism to coregulate numerous miRNAs.

A recently proposed unique aspect of P signaling and miRNA involves potential cross talk between photosynthate (Suc) availability and miRNA expression during P deficiency (Liu et al., 2010). The authors found that expression of miR399 in either shoots or roots required photosynthetic carbon assimilation. When P-sufficient bean plants were subjected to P starvation, miR399 was strongly induced in roots and shoots within 24 h. Surprisingly, miR399 was not expressed in roots of dark-treated and stem-girdled P-starved plants. Moreover, expression of miR399 transcript expression was blocked in dark-treated leaves of P-deficient plants. Whether light and sugars modulate the expression of miR399 and other miRNAs known to be involved in abiotic and biotic stress needs to be addressed.

CONCLUSION

P is required for plant growth and development, but its availability is frequently limiting. Plants have evolved numerous adaptive mechanisms for acclimation to P deficiency. These mechanisms involve the activation of metabolic, molecular, developmental, and regulatory processes that modify root architecture to increase soil volume exploration and recycling of internal P. Modification of root architecture is frequently accompanied by increased exudation of organic acids, protons, and enzymes to increase P availability. Recent advances in genomics and genetics suggest that plant acclimation to P deficiency involves cross talk between sugars and gene expression, including expression of miR399. The development of well-defined RILs and NILs having P tolerance coupled to next-generation sequencing will lead to the identification of genes regulating adaptation to P stress. Next-generation sequencing will also be critical to defining whether epigenetic changes are involved in P-stress responses. Further understanding of the biochemical and genetic regulation of these quaternaries in plant acclimation to P stress will pave the way to developing crop plants with enhanced P acquisition and use.