Prathap V, Suresh KUMAR, Nand Lal MEENA,2, Chirag MAHESHWARI, Monika DALAL,Aruna TYAGI

(1Division of Biochemistry, Indian Council of Agricultural Research-Indian Agricultural Research Institute, New Delhi 110012,India; 2Division of Germplasm Evaluation, National Bureau of Plant Genetic Resources, New Delhi 110012, India; 3National Institute for Plant Biotechnology, New Delhi 110012, India)

Abstract: Phosphorus (P) deficiency limits the growth, development, and productivity of rice. To better understand the underlying mechanisms in P-deficiency tolerance and the role of Pup1 QTL in enhancing P use efficiency (PUE) for the development of P-efficient rice cultivars, a pair of contrasting rice genotypes(Pusa-44 and NIL-23) was applied to investigate the morpho-physio-biochemical and proteomic variation under P-starvation stress. The rice genotypes were grown hydroponically in a PusaRich medium with adequate P (16 mg/kg, +P) or without P (0 mg/kg, -P) for 30 d. P-starvation manifested a significant reductions in root and shoot biomass, shoot length, leaf area, total chlorophyll, and P, nitrogen and starch contents, as well as protein kinase activity. The stress increased root-to-shoot biomass ratio, root length,sucrose content, and acid phosphatase activity, particularly in the P-tolerant genotype (NIL-23).Comparative proteome analysis revealed several P metabolism-associated proteins (including OsCDPKs,OsMAPKs, OsCPKs, OsLecRK2, and OsSAPks) to be expressed in the shoot of NIL-23, indicating that multiple protein kinases were involved in P-starvation/deficiency tolerance. Moreover, the up-regulated expression of OsrbcL, OsABCG32, OsSUS5, OsPolI-like B, and ClpC2 proteins in the shoot, and OsACA9,OsACA8, OsSPS2F, OsPP2C15, and OsBiP3 in the root of NIL-23, indicated their role in P-starvation stress control through the Pup1 QTL. Thus, our findings indicated that -P stress-responsive proteins, in conjunction with morpho-physio-biochemical modulations, improved PUE and made NIL-23 a P-deficiency tolerant genotype due to the introgression of the Pup1 QTL in the Pusa-44 background.

Key words: phosphorus deficiency; Pup1 QTL; protein kinase; acid phosphatase; proteome analysis; rice

Phosphorus (P) is the second most important macronutrient for plant growth and nutrition, following nitrogen (N), which exists in the soil in more than 170 inorganic and mineral forms (Holford, 1997). In a cell,P is one of the indispensable structural components for several important biological macromolecules, including nucleic acids, proteins, membrane-lipids, sugar-phosphates,adenylates, and it is necessary for regulatory and energy transfer processes (Plaxton and Lambers, 2015;Kumar et al, 2021a). Therefore, any deficiency in P availability leads to severe consequences on crop productivity, particularly in acidic soils (Chiou and Lin,2011). Though a sufficient amount of P is present in soils, it’s existence in a variety of insoluble inorganic form (oxides and hydroxides of aluminum and iron)and organic form (phytate, lecithin and glycerophosphate)renders it unavailable to plants (Li et al, 2008;Hinsinger et al, 2011). A considerably high proportion of soil P in the soil (50%-80%) exists in inaccessible organic forms. As plants absorb P in inorganic phosphate forms such as H2PO4-and HPO42-(Wang et al, 2010),soils, containing higher amounts of P in organic and inorganic forms, may suffer from P deficiency (Hammond et al, 2004; Li et al, 2008).

According to an estimate, about 30% of arable land globally is deficient in P availability to varying extents(Muralidharudu et al, 2011). To meet the P requirements of crop plants, a massive amount of P fertilizers is profligately applied in agricultural fields each year aiming at increased crop production. Due to the increasing demand/consumption of P fertilizers, global demand for P fertilizers reached approximately 49 million tons in 2022 as per the report of FAO, United Nations, in 2022. India consumes about 19% of the global P fertilizers, with a majority (90%) being imported (Sharma and Thaker, 2011). Unfortunately,such indiscriminate use of P fertilizers is neither favorable for sustainable agriculture nor environmentally sustainable,as non-renewable rock phosphate reserves are depleting rapidly (Hallama et al, 2019). Moreover, due to the low P acquisition efficiency (PAE) and P use efficiency(PUE) of crop plants, only 30% of applied phosphatic fertilizers are utilized by plants, while the remaining 70% participates in eutrophication of water bodies, soil hardening, and environmental pollution (López-Arredondo et al, 2014).

Rice (Oryza sativaL.) is one of the staple food crops for the global population, accounting for 80% of the food requirements for more than half of the world’s population. Generally, rice exhibits low PUE (about 25%) and displays typical P-deficiency symptoms such as stunted growth, reduced tillering, narrow and short dark green leaves, fewer panicles and grains, and thin stems (Vinod and Heuer, 2012; Das et al, 2017). Plants have evolved various adaptive strategies to acquire P from the soil, particularly under P-deficiency conditions,including anatomical, morpho-physio-biochemical,and genetic adaptations (Vance et al, 2003; Hammond et al, 2004). Some plants can convert unavailable forms of P into available forms, enhancing PUE (Rengel and Marschner, 2005). P deficiency-induced secretion of nucleases, organic acids, protons, and/or acid phosphatase(APase) assists P solubilization, mineralization, desorption,or and enhances PAE (Plaxton and Tran, 2011;Scheible and Rojas-Triana, 2015). Furthermore, plants can modulate their root system architecture (RSA),including root-to-shoot biomass ratio (RSR), lateral root formation, and root hair development to augment PAE by scavenging P from decaying biomass in the soil (Péret et al, 2011; Kumar et al, 2021a). Other P mining strategies deployed by plants include symbiotic associations with mycorrhiza and up-regulating the expression of high-affinity P transporters (Rausch and Bucher, 2002; Harrison et al, 2010; Kumar et al, 2022a).Moreover, under P-starvation/deficiency, internal PUE is modulated by replacing plasma membrane phospholipids with amphipathic sulfolipids and galactolipids (Nussaume et al, 2011; Plaxton and Lambers, 2015). Furthermore,P starvation triggers the remobilization of P from cellular reserves, older/dead cells to younger cells/tissues through scavenging P from biomolecules like DNA and RNA (Siebers et al, 2015; Dissanayaka et al, 2018).

Plant responses to P starvation stress involve an elaborate and complex network of molecular signaling transduction pathways. The plasma membrane-localized P transporter 1 (PHT1) has been identified as one of the major transporters involved in P uptake and redistribution.Some P starvation-induced (PSI) PHT1 members are transcriptionally regulated by the P starvationresponsive PHR-SPX module (Gu et al, 2016; Prathap et al, 2022). PHRs are MYB family transcription factors that directly bind to the PHR1 binding sequence(P1BS) in thecis-elements of the promoter of PSI genes, up-regulating their expression (Rubio et al, 2001;Zhou et al, 2008; Kumar et al, 2021b). A class of negative regulators known as SPX modulates the activities of PHRs post-translationally. SPXs physically interact with PHRs to prevent them from binding to P1BS(Zhong et al, 2018). P-starvation responses in plants also involve interactions with other metabolic processes such as carbon (C) and N. Recent studies have shown that nitrate is required to activate PSI signaling (Maeda et al, 2018; Hu et al, 2019; Hu and Chu, 2020).OsSPX4 (a repressor protein) regulates N and P starvation signal transduction by controlling OsNLP3 and OsPHR2, respectively, which are the central regulators of these signaling pathways (Hu et al, 2019).

Accumulation of sugar (sucrose and starch) to encourage carbon allocation and root growth has been reported in plants in response to P-starvation stress(Hermans et al, 2006). Sucrose, one of the systemic signaling molecules, has been reported to get transported from the shoot to the root and activate PSI responses(White and Hammond, 2008; Liu and Vance, 2010).Furthermore, the exogenous application of sucrose was reported to promote PSI gene expression under both P-deficient and P-sufficient conditions (Karthikeyan et al,2007). Up-regulated expression of the phloem-specific sucrose transporter AtSUS2 is linked to increased sucrose transport and accumulation, which activates PSI genes, resulting in P acquisition (Lei et al, 2011).The sucrose effect is most likely mediated through PHR1/PHL (Bustos et al, 2010). However, it remains unknown whether P-starvation causes an increase in sucrose uploading in the root and what the mechanism is behind sucrose-influenced phosphate starvation response 1 (PHR1)-mediated regulation of PSI genes.

Over the last two decades, considerable breeding efforts have been made to identify potential QTLs for low P tolerance toward improving the productivity of modern crop varieties in P-deficient soil (Wissuwa and Ae, 2001; Chin et al, 2010; Gamuyao et al, 2012).Several QTLs associated with agronomic traits like P-uptake, P-translocation and PUE that contribute to higher yield under P-deficient conditions in rice have been identified (Ni et al, 1998; Wissuwa et al, 1998,2002; Li et al, 2008; Shimizu et al, 2008). P-uptake 1(Pup1) QTL for P-deficiency tolerance is about a 50-kb DNA fragment on the long arm of chromosome 12 identified initially in the P-efficientausrice variety Kasalath (a traditionalindicalandrace) (Wissuwa et al,1998; Gamuyao et al, 2012). Though other QTLs associated with P-uptake/PUE have been identified in recent years,Pup1is the only QTL that has been extensively studied for P-deficiency tolerance in rice.ThePup1QTL has been reported to contribute up to 54.5% of the phenotypic variation in P-deficiency tolerance in rice (Ni et al, 1998). The candidate gene responsible for P-deficiency tolerance isP-starvation tolerance 1(PSTOL1), which codes for a protein kinase and acts as an early root growth enhancer,allowing the plant to absorb more P from the soil(Wissuwa et al, 2002; Gamuyao et al, 2012). However,the underlying regulatory mechanism for P-starvation tolerance remains unknown. Therefore, gaining a better understanding of plant responses to P-starvation stress at morphological, physiological, biochemical,and molecular levels would assist in developing P-starvation/deficiency-tolerant rice varieties.

A promising strategy to decipher the mechanisms involved in P-starvation/deficiency tolerance is to select a pair of contrasting rice genotypes for PAE/PUE grown under the most contrasting (P-starved and P-sufficient) conditions (Kumar et al, 2021a). Since the soil contains either high, medium, or low P, and rice is a water-loving plant, hydroponic culture is a better option to grow the plants under varying nutrient regimes for experimental purposes wherein no/exact supply of P can be easily managed or controlled(Kumar et al, 2022a). Developing a rice variety with enhancing PAE and PUE might assist in optimizing the cost to benefit ratio, particularly on part of the application of phosphatic fertilizers by minimizing the expensive and excessive use of P fertilizers. Excessive use of P fertilizers results in the eutrophication of water bodies as well as the depletion of non-renewable (rock phosphate) natural resources (Kumar et al, 2022b).Until recently, most studies on plant responses to P-starvation/deficiency stress have been conducted at the early (seedling) stage of plant growth, and only a few studies have been performed at the advanced(vegetative) stage of growth (Kumar et al, 2021a,2022a). Since P is required for plant growth and development at every developmental stage (in fact, an increasing amount of P is required with the advancing stage of growth and development), we focused on investigating and understanding the morpho-physio-biochemical mechanisms adopted by rice at the vegetative stage to cope with P-starvation stress.

RESULTS

Morpho-physiological responses to P starvation stress

Substantial variations in morpho-physiological parameters were observed in the rice genotypes (Pusa-44 and NIL-23) without P (0 mg/kg, -P) for 30 d. Some typical-P stress-responsive characteristics, such as decreased shoot and root biomass, increased RSR, longer roots,shorter-erect leaves, and coloration of leaves due to accumulations of anthocyanin, were observed in the plants grown under -P stress conditions (Fig. 1). The stress had significant effects on shoot morphology with a significantly reduced plant height, smaller and erect leaves, and considerably less biomass (Fig. 1-A).Similarly, significant variations in root morphology were also observed (Fig. 1-B and -C). However, the reduction in growth and development was more prominent in Pusa-44 (P-sensitive genotype) than in NIL-23 (P-tolerant genotype).

P-starvation stress alters root system architecture

P-starvation stress significantly affected the RSA of plants by modulating root length, total root projected area, root surface area, the number of root tips, the number of root forks, etc. (Fig. 2). After 30 d of -P treatment, some RSA parameters such as total root length (61.2%) (Table S1), total root projected area(60.0%), total root surface area (60.2%), the number of root tips (69.1%), and the number of root forks (70.9%)were significantly higher in NIL-23 (Fig. 2). These results indicated that the introgression of thePup1QTL makes NIL-23 more responsive to -P stress.Significantly higher total root surface area, and the numbers of root tips and forks (Fig. 2-B to -D)indicated the effect ofPup1QTL on RSA even under normal (+P, 16 mg/kg) conditions.

Fig. 1. Shoot and root morphologies of rice genotypes grown hydroponically under P-starvation stress for 30 d.

P-starvation stress affects leaf area and chlorophyll content

Leaf is a key determinant of photosynthate accumulation which determines the growth and development of the plant. A considerable reduction in leaf area was observed in contrasting rice genotypes under -P stress.The reduction in leaf area was up to 75.4% in Pusa-44,while it was comparatively lower (72.8%) in NIL-23(Table 1). Moreover, a comparison of the total chlorophyll content in the leaves of the contrasting rice genotypes indicated a considerable (61.3%) reduction in chlorophyll content in Pusa-44 compared with that(50.0%) in NIL-23 under -P stress (Table 1). Thus, the effect of -P stress was more pronounced in Pusa-44.These indicated that NIL-23 can better manage photosynthesis under -P stress compared with Pusa-44.

P-starvation stress affects RSR

A comparative analysis of biomass production by the contrasting rice genotypes under -P and +P treatment revealed that -P stress had a more prominent effect on shoot biomass production, particularly in Pusa-44(69.5% reduction) compared with NIL-23 (65.6%)(Table 1). Although an overall reduction in root biomass was recorded in the contrasting rice genotypes,the reduction was significantly greater in Pusa-44(61.7%) compared with that in NIL-23 (30.9%). In contrast, RSR increased under -P stress, with a higher ratio in the case of NIL-23 (0.80) compared with that in Pusa-44 (0.45) under -P stress. Despite a decrease in shoot length and biomass under -P stress, an increase in total root length was observed, which is one of the characteristic features of plant adaptive responses.

P-starvation modulates accumulation and mobilization of carbohydrates

For a more comprehensive understanding of the allocation and redistribution of photosynthate/carbohydrate between the shoot and root, starch and sucrose contents in these tissues were analyzed. Starch content in the shoot of Pusa-44 was reduced by 50.0%, while a 40.8%reduction was observed in NIL-23 under -P stress. A similar trend was also observed for starch content in the root, with a 47.7% reduction in Pusa-44 and a 41.2%reduction in NIL-23 under -P stress (Fig. 3-A). The impact of -P stress was greater on the reduction of starch content in the shoot and root of Pusa-44 compared with NIL-23. Moreover, the comparative analysis showed that -P stress significantly increased sucrose content in the shoot and root of the contrasting rice genotypes. Sucrose content increased by 26.1% in the shoot and 53.0% in the root of Pusa-44, while it was 70.0% in the shoot and 20.3% in the root of NIL-23, respectively, under -P stress. Compared with Pusa-44, the increase in sucrose content was 43.9%higher in the shoot, whereas it was 32.7% lower in the root of NIL-23 (Fig. 3-B).

Fig. 2. Modulation in root system architecture of rice genotypes grown hydroponically under P-starvation stress for 30 d.

Table 1. Morpho-physiological variations in rice genotypes [Pusa-44 (P-sensitive genotype) and NIL-23 (P-tolerant genotype)] grown hydroponically under P-starvation stress for 30 d.

P-starvation affects P remobilization in plant

Although the P content in the shoot, as well as in the root, was significantly reduced under -P stress, the P content was comparatively higher in NIL-23 than in Pusa-44 (Fig. 3-C). Interestingly, the P content in the shoot as well as in root of NIL-23 was significantly higher than that of Pusa-44 under -P stress. Our findings indicated that NIL-23 more efficiently mobilizes P from the root to the shoot under -P stress.

P-starvation stress affects N uptake by plants

Under -P stress, the total N content in the root as well as in the shoot of NIL-23 was recorded to be higher,with a more significant increase in the root (Fig. 3-D).Under the unstressed conditions (16 mg/kg P), no difference in N content between the two rice cultivars was observed in root as well as in shoot tissues. Under-P stress, NIL-23 could maintain optimum N content in the root but it was significantly less in Pusa-44. Thus,-P stress showed a considerable decrease (21.16 %) in N content in the shoot of Pusa-44, while it was only 14.8% in the shoot of NIL-23 (Fig. 3-D).

P-starvation stress-induced variations in APase activity

The -P stress was observed to significantly increase APase activity in the root and in the shoot of the contrasting rice genotypes. The -P stress significantly increased APase activity in both the shoots and roots of the contrasting rice genotypes. However, the increase in APase activity was considerably higher (2-fold) in the roots compared with the shoots. Moreover, the increase was significantly higher in NIL-23 compared with Pusa-44 (Fig. 4-A). No significant difference in APase activity was observed between shoot and root as well as between the rice genotypes under normal (+P,16 mg/kg) condition.

P-starvation stress-induced variations in protein kinase activity

A significant decrease in total protein kinase activity was observed in the shoots of Pusa-44, while only a minor variation in activity was observed in NIL-23,under -P stress. However, in the roots of the contrasting rice genotypes, a significant reduction in total protein kinase activity was recorded under -P stress (Fig. 4-B).The reduction was 42.8% in the roots of Pusa-44, it was only 27.3% in the roots of NIL-23. Thus, a significantly higher reduction in protein kinase activity, which affects stress signals, was recorded in the shoots as well as the roots of the stress-sensitive genotype (Pusa-44)compared with the stress-tolerant genotype (NIL-23).

Fig. 3. Effects of P-starvation stress on biochemical parameters in rice genotypes grown hydroponically under P-starvation stress.

Fig. 4. Effects of P-starvation stress on acid phosphatase (A) and total protein kinase activities (B) in rice genotypes grown hydroponically under P-starvation stress for 30 d.

Correlation analysis of physio-biochemical parameters

The degree of linear relationships among various traits based on their correlation coefficients showed significant associations between many physiological and biochemical characteristics, as revealed by Pearson correlation analysis (Fig. 5). Except for the correlation between shoot APase activity and total protein kinase activity, all shoot and root traits were significantly correlated. Root length, APase activity, and sucrose content showed a significant negative correlation with the N, P, and starch content in both the shoot and root,leaf area, shoot length, chlorophyll content, and shoot and root biomass. It was also observed that RSR had no significant correlation with root length, sucrose content,and APase activity in both the shoot and root.

Stress-induced differential expression of P metabolism-associated proteins

Fig. 5. Pearson correlation (PC) analysis depicting correlation coefficient and an association between physiological and biochemical characteristics.

In the shoot of NIL-23, 634 proteins were expressed under -P stress, while it was 1 293 proteins in Pusa-44.Thus, a considerably smaller (50%) number of proteins were expressed in the shoot of NIL-23 under -P stress.However, in the root of NIL-23, more (559) proteins were observed to be expressed under -P stress compared with only 358 proteins expressed in the root of Pusa-44. This indicated a significantly more (＞ 56%)number of proteins were expressed in the root of NIL-23, which might be involved in P uptake, while approximately 51% fewer proteins expressed in the shoot of NIL-23 under -P stress, which might be responsible for managing the stressed conditions (Fig.6-A).

In the root of Pusa-44, 27 proteins were down-regulated(with 39 proteins up-regulated) compared with roughly 2-fold more (53) proteins down-regulated (along with 40 proteins up-regulated) in the root of NIL-23. Moreover,in the shoot of Pusa-44, only 204 proteins were upregulated (with 477 proteins down-regulated) compared with about 2.7-fold more (555) proteins up-regulated(along with only 12 proteins down-regulated) in the root of NIL-23. Thus, our findings indicated that -P stress caused a greater number of proteins to be up-regulated in the shoot but a greater number of proteins to be down-regulated in the root of NIL-23(compared with Pusa-44) because of the regulatory effects of the introgression ofPup1QTL (Fig. 6-B).

Fig. 6. Comparative analysis of differentially expressed proteins (DEPs) in shoot and root of in rice genotypes [ Pusa-44 (P-sensitive genotype)and NIL-23 (P-tolerant genotype)] under a most contrasting condition [16 mg/kg (+P) and 0 mg/kg (-P)].

Furthermore, a four-way analysis of the up- and down-regulated proteins in the shoot of the rice genotypes indicated that 64.1% (499) proteins were exclusively up-regulated in the shoot of NIL-23 compared with 22.3% (174) proteins exclusively up-regulated in the shoot of Pusa-44. However, the exclusively up-regulated proteins in the root of NIL-23 were different (with no common proteins) from those in the root of Pusa-44 (Fig. 6-C). The exclusively down-regulated proteins in the shoot of NIL-23 were very fewer (9) compared with 459 down-regulated proteins in the shoot of Pusa-44. However, the exclusively down-regulated proteins in the root of NIL-23 were more in number (53) compared with those (40) in the root of Pusa-44 (Fig. 6-D).

Gene Ontology (GO) analysis of P-starvation stress-induced proteins

GO analysis of the differently expressed proteins(DEPs) in the shoots of the contrasting rice genotypes indicated that proteins related to the P metabolic process were significantly affected by P-starvation stress (Fig. S1, Tables S1 and S2). The DEPs associated with the P metabolic process induced by P-starvation stress are presented in Table S3. Out of the 45 DEPs related to the P metabolic process, most(43) of them were highly up-regulated (Log2Fold Change ＞ 2), while only 2 were observed to be downregulated. In contrast, 5 out of the 15 DEPs associated with the P metabolic process were significantly up-regulated, whereas 10 of them were significantly down-regulated. Thus, our findings revealed that the up-regulated expression of proteins played an important role in the stress tolerance of NIL-23 mediated by the introgression of thePup1QTL,corroborating earlier reports (Kumar et al, 2021a,2022a). The proteins associated with the P metabolic process were observed in greater numbers and had a higher up-regulated expression in the shoots of NIL-23 compared with those in the shoots of Pusa-44 when grown under P-starvation stress.

The highly up-regulated proteins related to the P metabolic process in the shoots of NIL-23 include pyruvate phosphate dikinase 1, serine/threonine-protein kinase Nek2, mitogen-activated protein kinase 10,cyclin-dependent kinase C-1, and calcium-dependent protein kinase 21. In contrast, ethylene receptor and a putative protein phosphatase 2C were the downregulated proteins in the NIL-23 shoot. On the contrary, a different set of five P metabolism-related proteins, namely serine/threonine-protein kinase 6,probable protein phosphatase 2C, serine/threonine-protein kinase Nek6, receptor kinase-like protein Xa21, and brassinosteroid leucine-rich repeat receptor kinase 1,were up-regulated (Log2Fold Change ＞ 2) in the shoots of Pusa-44. However, no such GO term or DEPs associated with the P metabolism under the biological process category was observed in the roots of the contrasting rice genotypes.

Stress-induced top-five highly up-regulated proteins in tolerant genotype

A large number of proteins (555, accounting for 97.8% of the proteins) were significantly up-regulated in the shoots of NIL-23 in response to -P stress,whereas only 204 (30%) of the proteins were up-regulated in the roots of Pusa-44. Among these, the Rubisco large subunit, ABC transporter G family member 32, sucrose synthase 5, DNA polymerase I B,and chaperone protein ClpC2 were the highly up-regulated proteins in the shoots of NIL-23 (Table S4). Similarly, in the roots of NIL-23, about 43% (40)of the proteins were significantly up-regulated under -P stress. Among these, Ca2+-ATPase isoform 9, OsACA8,sucrose phosphate synthase 2F, protein phosphatase 2C,and OsBiP3 were the highly up-regulated proteins.

Stress-induced highly down-regulated proteins in tolerant genotype

Under -P stress, 53 (57% of the DEPs) were significantly down-regulated in the roots of NIL-23,while 27 (41% of the DEPs) were down-regulated in the roots of Pusa-44. Among the down-regulated proteins in the shoots of NIL-23, some important players were the ethylene receptor 3, serine/threonine protein phosphatase 2A, ADP-glucose pyrophosphorylase,and putative protein phosphatase 2C (Table S5). The heat shock 70 kDa protein and calpain-type cysteine protease ADL1 were among the highly down-regulated proteins in the roots of NIL-23 (Table S5).

DISCUSSION

Plants adopt different morphological, physiological,and molecular strategies to cope with low P stress.These adaptive variations assist plants in improving both PAE and PUE (Yang et al, 2019; Ding et al,2021). P availability affects leaf morphology (size and angle), tiller number, and yield (Mghase et al, 2011;Ruan et al, 2018; Kumar et al, 2021a). P deficiency even at early growth stage significantly affects the tiller ability and thus the yield of the plant (Ruan et al, 2018;Kumar et al, 2022a). In addition, several physiological and biochemical processes (accumulation of purple pigment in leaves, decreased chlorophyll content,APase activity, phytohormone signaling, photosynthesis,cell-wall structure, etc.) in plants are affected due to P-starvation/deficiency stress (Yuan and Liu, 2008;Mghase et al, 2011; Ruan et al, 2018; Kumar et al, 2021a,2022a). Recently, Irfan et al (2020) characterized several rice cultivars for their responses to -P stress and reported a significant reduction in plant height and biomass production. Our findings corroborate the findings of the above-mentioned researchers.

Comparative analysis indicated that NIL-23 had longer roots and better root architecture than Pusa-44 for efficient P acquisition under P-starvation stress.The morpho-physiological properties of RSA greatly influence nutrient and water absorption from the soil(Hodge et al, 2009). The P utilization rate of rice has been reported to be higher (13.1%) than that of maize(11%) and wheat (10.7%) (Ma et al, 2003; Shabnam and Iqbal, 2016). Alterations in RSA (formation of shallow roots, increased lateral roots, root-hair density)have been well-documented to play crucial roles in modulating PAE (Lynch, 2011; Kafle et al, 2019; Liu et al, 2021; Li et al, 2022). InArabidopsis, P-starvation stress promotes the formation of highly branched roots at the expense of primary roots (Péret et al, 2011). Such adaptive changes in RSA and increased RSR improve PAE by exploring more soils to scavenge P from decomposing organic matters in the top soil layers (Ma et al, 2003; Péret et al, 2011; Aziz et al, 2014; Kumar et al, 2021a). In rice and maize, P-starvation stress was reported to significantly increase the development of lateral roots, root hairs, and the formation of root clusters as well as the modulation of root angle (Jin et al,2012; Lambers et al, 2011). ThePup1QTL harbors thePSTOLgene that acts as an early root growth enhancer in rice (Wissuwa et al, 1998; Gamuyao et al, 2012);thus, our findings confirmed the role of QTL in improving PAE through modulation in RSA traits on its introgression in the Pusa-44 genetic background.

Our results indicated that NIL-23 can better manage photosynthesis under -P stress compared with Pusa-44.P-starvation stress in a plant is known to impair chlorophyll biosynthesis (Zhang et al, 2014; Secco et al,2015; Yang et al, 2020). Kumar et al (2021a) reported the down-regulated expression of several genes associated with the photosynthetic process in rice under P-starvation stress. Thus, our findings are in agreement with previous reports (Pieters et al, 2001;Xu et al, 2007; Xing and Wu, 2014; Kumar et al, 2021a).Different rice cultivars exhibit varying responses in terms of shoot biomass under P-starvation/deficiency stress (Irfan et al, 2020). The reduction in shoot and root biomass production in response to P-starvation stress has been reported in different plants likeArabidopsis, rice, wheat, and chickpea (Narang and Altmann, 2001; Alloush, 2003; Kumar et al, 2021a;Wang et al, 2022). RSR is considered to be an excellent indicator of the partitioning of photosynthates between the shoot and root of the plants in response to -P stress.Increased root growth and length in response to -P stress have been reported in several plants like maize,rice, sorghum, onion, and tomato (F?hse et al, 1991; Sun et al, 2016; Zhang et al, 2019; Kumar et al, 2021a).Thus, our findings corroborate those of the earlier reports.

Studies suggested that -P stress impacts starch and sucrose metabolism (Rao et al, 1990; Nanamori et al,2004). An increase in the triose-phosphate to phosphate ratio under low P concentration stimulates the key enzyme of the starch biosynthesis pathway, ADP-glucose phosphorylase, which increases starch synthesis in the chloroplast (Stitt and Quick, 1989; Nielsen et al, 1998).However, the effect of -P stress on starch and sucrose metabolism varies between crop species. We observed a decrease in starch content in the shoots as well as the roots of the contrasting rice genotypes (Fig. 3-A), but a significant increase in sucrose content in shoots and roots of the rice genotypes under -P stress (Fig. 3-B).The increased sucrose content was more in NIL-23 compared with Pusa-44. Our findings corroborate those of Li et al (2006), who reported that -P stress increase sucrose content while decreasing starch content in the shoot and the root of rice. This can be explained by the fact that -P stress does not always restrict the export of triose-phosphate from the chloroplasts to the cytosols(Nanamori et al, 2004), and it stimulates some of the enzymes involved in sucrose synthesis (Rao et al,1990). Hermans et al (2006) reported that plants modulate sucrose and starch to encourage carbon allocation in the root, in response to P-starvation stress.Sucrose acts as a systemic signaling molecule that activates the PSI responses in the root to transport it from the shoot to the root (White and Hammond, 2008;Liu et al, 2010). Up-regulation of the AtSUS2 was associated with increased sucrose transport and accumulation-mediated activation of PSI genes (Lei et al, 2011). The sucrose effect was proposed to be mediated by PHR1/PHL (Bustos et al, 2010).

P-starvation/deficiency stress causes efficient mobilization of P from the root to the shoot, remobilization of P from older to younger tissues/organs as well as redistribution of cellular P within the cell by scavenging P from various biomolecules (DNA and RNAs) (Siebers et al, 2015; Dissanayaka et al, 2018).Our findings, showing higher P content in the shoot compared with that in the root in the genotypes,particularly in NIL-23, under +P conditions, corroborate the reports of Yugandhar et al (2017). This indicated better PAE and PUE of NIL-23 as reported earlier by Kumar et al (2021a). For optimal growth and development, including initiation, expansion and elongation of leaves under -P stress, mobilization of P from older/dying organs to younger/growing tissues is crucial for the survival of plants (Assuero et al, 2004;Kavanová et al, 2006). Moreover, a P stress-tolerant rice cultivar (Akamai) was reported to have lower P content in the roots under P-starvation stress compared with a P stress-sensitive cultivar (Koshihikari)(Dissanayaka et al, 2017, 2018), which indicates that P is more efficiently/quickly mobilized from the root to the shoot. Our findings were in agreement with those of Wissuwa and Ae (2001) and Pariasca-Tanaka et al(2009), who reported higher P content in the Nipponbare near-isogenic lines harboring thePup1QTL.

During the vegetative stage of growth, plants require a higher amount of N and P for various metabolic processes, including nucleic acid, amino acid, and protein biosynthesis as well as energy metabolism(?gren et al, 2012). Therefore, plant growth and survival under P-starvation stress depend on its capability for C-N-P homeostasis. Though cross-talk between P and C (Yuan and Liu, 2008), as well as N and C(Müller et al, 2007), for signal transduction pathways is well-understood, the cross-talk between N and P metabolism, particularly under -P stress, remains elusive, and requires further investigations. Studies indicated that the interaction between N and P signaling is mediated by nitrogen limitation adaptation(NLA) and PHO2, which regulate phosphate transporter activity, resulting in the N-dependent accumulation of P in the shoot (Peng et al, 2007; Lin et al, 2013). PHR1 has been reported to be a central regulator of nitrate-inducible GARP-type translational repressor 1 (NIGT1) (Kiba et al, 2018). N-P interaction was reported to modulate RSA through a regulatory component of N and P signaling mediated by cytokinin(Cerutti and Delatorre, 2013). Several reports suggested that the availability of N modulates P-starvation responses (Kant et al, 2011; Liang et al, 2015; Medici et al, 2019). Under P-starvation stress, N availability activates P acquisition. The N-P interactions involve three major signaling factors, including SPXs, PHRs,and PHO2 (Kumar et al, 2021b; Prathap et al, 2022). In rice, a nitrate sensor (nitrate transporter NRT1.1B) was reported to interact with SPX4 (a phosphate-signaling repressor) (Hu et al, 2019). PHR is positively regulated by N availability at both the transcriptional and post-transcriptional levels (Sun et al, 2018; Varala et al,2018).

Under P-starvation/deficiency stress, P is liberated from a wide range of P-rich organic molecules, including nucleic acids, P-monoesters, and P-anhydrides (Plaxton and Tran, 2011; Liang et al, 2014; Stigter and Plaxton,2015). The secretion of APase, ribonucleases, and organic acids assists in the mobilization of fixed P in the rhizosphere, thereby, improving the availability of P for the plant (Zhang et al, 2014; Wang et al, 2018).Moreover, intracellular purple APs play a key role in recycling P (Bozzo and Plaxton, 2008; Tran et al, 2010).Among the 26 PAPs identified in rice, 10 are significantly increased under P-starvation stress (Zhang et al, 2011).OsPAP10a is a key APase isoform associated with roots that is stimulated by P stress and assists in scavenging organic P forms in the rhizosphere (Tian et al,2012). Thus, our findings corroborate earlier results of Morcuende et al (2007) and Calderon-Vazquez et al(2008). The increased protein kinase activity in response to -P stress has been reported earlier in several crops likeArabidopsis, rice, wheat, and tomato (Sano and Youssefian, 1994; Wang et al, 2002; Scheible et al,2004). Our findings corroborated those of Deng et al(2022), who reported an increase in the number of expressed proteins in the root of wild rice (Oryza rufipogonGriff.) seedlings under -P stress. Such a differential protein profile induced by -P stress has been reported in several crop plants like rice, maize,soybean, andBrassica napus(Yao et al, 2011; Zhang et al, 2014; Tantray et al, 2020; Cheng et al, 2021).Thus, our findings indicated that -P stress caused a greater number of proteins to be up-regulated in the shoots but a greater number of proteins to be down-regulated in the roots of NIL-23 (compared with that in Pusa-44) because of the regulatory effects of the introgression of thePup1QTL. A considerable variation in the number of differentially expressed genes (Kumar 2021), as well as the proteins in NIL-23(compared with those in Pusa-44) under -P stress,confirms the role of thePup1QTL in P-starvation stress tolerance, which corroborates the findings reported earlier in rice (Tantray et al, 2020).

Several kinases of the serine/threonine-protein kinase, mitogen-activated protein kinase, calciumdependent protein kinase, histidine kinase, putative CBL-interacting protein kinase, and cyclin-dependent kinase families were highly up-regulated in the shoot of NIL-23. Protein kinases have been reported to be extremely important in signal transduction and cellular processes, including cell growth, differentiation,metabolism, and transcription (Cai et al, 2013). The kinases catalyze the phosphorylation reaction in which a P from ATP is transferred to a hydroxyl groupcontaining amino acid (serine, threonine, and tyrosine)in the substrate. However, P-starvation stress causes considerable decreases in ATP content in plant cells(Raghothama, 1999), and hence only a few protein kinases respond to the -P stress. The up-regulated expression of several protein kinases in wheat, tomato,andArabidopsishas been reported under P-starvation stress (Sano and Youssefian, 1994; Wang et al, 2002;Scheible et al, 2004). In response to P-starvation stress,a stress-activated protein kinase 3 (a member of the SNF-1-related protein kinase-2, SnRK2) was reported to be dephosphorylated (Chen et al, 2015). Recently,the role of receptor-like cytoplasmic kinases in P homeostasis was reported (Zhang et al, 2018). These findings suggest that OsCDPKs, OsMAPKs, OsCPKs,OsLecRK2, and OsSAPks play important roles in -P stress management in the shoot of NIL-23 through physiological, biochemical, and molecular adaptive responses mediated by thePup1QTL. In response to P deficiency stress, theMPK2Btranscript is up-regulated in wheat shoots and roots (Bai et al, 2022).Overexpression ofTaMPK2Benhances tolerance to low P stress in tobacco by promoting P uptake, better RSA and cellular reactive oxygen species homeostasis.Phosphate transporters (PT3/PT4), PIN9, and antioxidants (MnSOD1/POD1;7) are transcriptionally regulated and underpin TaMPK2B, which favorably regulate RSA, P uptake, formation, and ROS scavenging under P deficiency (Bai et al, 2022).

It has been previously reported that P-starvation stress suppresses ATP synthesis as well as Rubisco activation and regeneration in the chloroplast (Fredeen et al, 1990; Wissuwa et al, 2005). P-starvation stress also affects carbon flux via the Calvin cycle due to the down-regulated expression of Rubisco large and small subunits appearing as its reduced activity (Lauer et al,1989; Jacob and Lawlor, 1993). In contrast to the down-regulation of OsrbcL observed in the shoot of Pusa-44, the Rubisco large subunit was considerably up-regulated in NIL-23, indicating that the stress-tolerant genotype could maintain a higher photosynthetic rate even under -P stress. The highly up-regulated expression of ABC transporter G family member 32 (OsABCG32,LOC_Os01g24010) in the shoot of NIL-23 under -P stress is in agreement with the findings of Nguyen et al(2018), who reported OsABCG family transporters to be regulated by abiotic stresses like drought, salt, heat,etc. The OsABCg32 is one of the members of ATP-binding cassette transporter proteins, which are engaged in the transport of a wide range of compounds across the membranes utilizing ATP. Similarly, the up-regulated expression of sucrose synthase 5 (OsSUS5)in the shoot of NIL-23 under -P stress corroborates the earlier findings of Hermans et al (2006), who reported-P stress-induced up-regulation of sucrose biosynthetic enzymes increase translocation of sugars from the shoot to the root, helpful in promoting root growth for improved uptake of P from the soil. This was supported by our finding of increased sucrose content in the shoot of NIL-23 under -P stress.

Interestingly, a mitochondrial DNA polymerase I B(OsPolI-like B, LOC_Os08g07850) (involved in DNA replication in mitochondria) and casein lytic proteinase/heat shock protein 100 (Clp/Hsp100) (a class of chaperone proteins involved in the ATP-dependent remodel/disassembly of protein complexes) were highly up-regulated under -P stress in the shoot of NIL-23.Such a chaperone has been reported to play essential roles in chloroplast development and heat stress tolerance inArabidopsis(Lee et al, 2006). This suggests that up-regulation of ClpC2 (LOC_Os12g12850) chaperon is involved in the proper development of chloroplast as well as in an abundance of OsrbcL in the shoot of NIL-23, thereby improving carbon fixation under -P stress.

Ca2+-ATPases (OsACA8 and OsACA9), highly up-regulated and involved in ABA-mediated P stress signaling, which we observed in root of NIL-23, have also been reported to function as transporters in restoring/maintaining ion homeostasis by pumping Ca2+out of the cytosol (Cerana et al, 2006). ACA8 and ACA9 were also reported to be up-regulated and involved in stress-induced abscisic acid (ABA) stimulus as well as stress signaling inArabidopsis(Cerana et al, 2006).The up-regulation of sucrose phosphate synthase(OsSPS2F), an enzyme involved in sucrose biosynthesis,regulates the availability of sucrose for plant growth and development observed in the root of NIL-23,corroborating the reports of Lutfiyya et al (2007). The up-regulated expression of protein phosphatase 2C(OsPP2C15) in the root of NIL-23 indicated that ABA-signaling is involved in the -P stress tolerance.Several protein kinases have been reported to be involved in P homeostasis (Fragoso et al, 2009; Lei et al, 2014; Chen et al, 2015; Zhang et al, 2016),whereas no protein phosphatase has been reported to be involved in P acquisition and homeostasis in plants(Yang et al, 2020). OsPP2C is a family of serine/threonine phosphatases, requiring Mg2+or Mn2+for activity, and they are involved in the ABA-mediated stress signaling pathway, which negatively modulates protein kinases (Xue et al, 2008; Shobbar and Bennett al,2012). OsPP95, a PP2C protein phosphatase, interacts with and dephosphorylates OsPT8 at Ser-517. OsPP95 overexpression in rice was reported to decrease OsPT8 phosphorylation and increase OsPT2/OsPT8 trafficking from the endoplasmic reticulum to the plasma membrane,resulting in P deposition (Yang et al, 2020).

The ethylene receptor 3, observed to be downregulated in the shoot of NIL-23, is a two-component regulator, acting as a negative modulator of the ethylene signaling pathway (Yau et al, 2004). Under -P stress,ethylene has been reported to limit primary root growth while favoring lateral root growth (He et al, 1992; Ma et al, 2003). Thus, the down-regulated expression of ETR3 in the root improves -P stress tolerance in NIL-23. We observed down-regulated expression of AGPL1 and reduced starch content in the shoot of NIL-23, correlate with the ADP-glucose pyrophosphorylase enzyme activity in starch metabolism, which is helpful in P-starvation signaling and P homeostasis (Qi et al,2020). Rice genome encodes for two small subunits(OsAGPS1 and OsAGPS2) and four large subunits(OsAGPL1-OsAGPL4) of AGPase (Lee et al, 2007).OsAGPL1 plays an important role in starch synthesis in culm, developing embryo as well as in starch accumulation in the endosperm (Cook et al, 2012).Mutation inAGPL1orAGPS1was reported to down-regulate the expression ofOsSPX2, but not the expression ofOsSPX1orOsIPS1. Mutants foragpl1andagps1decreases starch but increases sucrose content with suppressesSPX2, which suppressed PHR-mediated activation of phosphate transporters (PHT1s) and P uptake (Meng et al, 2020). The down-regulated expression of OsPP2Awe observed in the shoot of NIL-23 is in agreement with earlier report, which is involved in dephosphorylation of SIT1, inhibiting its activity, and positively affects abiotic stress tolerance(Zhao et al, 2019).

The need of the day is to improve the PUE of crop plants to minimize the cost of cultivation, decrease excessive application of phosphatic fertilizers in crops,reduce environmental pollution and save natural (rock phosphate) resources. In view of the known impacts ofPup1QTL on PUE of rice, we aimed at deciphering the molecular basis ofPup1-mediated PUE in rice. Therefore,in the present study, we focused on understanding the physiological, biochemical and proteomic parameters involved in P-starvation tolerance. Our findings indicated that the introgression ofPup1QTL in Pusa-44 genetic background (NIL-23) exhibits P-deficiency/starvation tolerance in terms of better shoot and root growth, RSR, increased APase activity compared with that of the parental genotype (Pusa-44). Moreover,alteration in starch and sucrose content in the shoot and root of NIL-23 favors physio-biochemical parameters for better PUE. Proteome profiling indicated that introgression ofPup1QTL improves PUE through modulation in the expression of P metabolism-related proteins, particularly in the shoot. The protein kinases,such as OsCDPKs, OsMAPKs, OsCPKs, OsLecRK2,and OsSAPks, were the major players in the signaling cascade. However, functional validation of the candidate proteins will be necessary before they can be utilized for improving the PUE of crop plants and developing P-efficient rice cultivars suitable for P-deficient soils.

METHODS

Rice materials and growth conditions

A pair of contrasting rice genotypes [Pusa-44 (P-deficiency sensitive)and a near-isogenic line (NIL)-23 (P-deficiency tolerant,harboringPup1QTL)] were utilized in the present study.NIL-23 is one of the BC3F6lines generated by introgression ofPup1QTL originally from Kasalath into the Pusa-44 genetic background [through initial backcrossing of IR64 as a recurrent parent (IR64-Pup1-F)]. Mature seeds of the contrasting rice genotype (NIL-23 and Pusa-44) were disinfected with 70%ethanol, followed by 0.1% HgCl2. After soaking the seeds overnight in sterile water, they germinated in the dark at 37 °C for 3 d. The germinated seeds were placed on moist germination paper sheets with the coleoptile facing upward to facilitate the vertical growth of the shoot and downward growth of roots(Kumar et al, 2021a). Fifteen-day-old seedlings were transferred to a hydroponic culture in a 20 L plastic container filled with Pusa Rice Hydroponics (PusaRicH) medium with either P (16 mg/kg, +P, control) or no P (0 mg/kg, -P, treatment) and grown until the plants became 45-day-old (Kumar et al, 2021a) in a greenhouse at the National Phytotron Facility, Indian Agricultural Research Institute, New Delhi, under natural light at 30 °C during the day, 22 °C at night, and 85% relative humidity. Morphological data were collected on 45-day-old plants in three replications. Shoot and root tissues were collected in three replications from the plants to analyze the effects of P-starvation stress on physiological and biochemical parameters,as well as to record the variations in the proteome profile. The pH of the nutrient (PusaRicH) medium was maintained at 5.2 using KOH/HCl (1.0 mol/L) and the medium was changed every 3 d (Sharma et al, 2018).

Measurement of plant height, biomass, and RSR

The height of the 45-day-old rice plants of the contrasting rice genotypes grown under +P or -P treatment in hydroponic culture was measured using a measuring scale in centimeters (cm). The freshly collected shoots and roots from the plants of contrasting rice genotypes were used to measure the fresh weight of the tissues, while dry biomass was recorded in milligrams (mg) after drying the tissues in a hot air oven at 65 °C for 4 d until a constant weight was achieved. The RSR was calculated using the formula: RSR = (Root dry biomass / Shoot dry biomass).

Analysis of RSA

To analyze RSA of the contrasting rice genotypes grown hydroponically under +P and -P conditions, a La2400 scanner(WinRHIZO Pro; Regent Instruments Inc, Montreal, Canada)was used to estimate various morphological features of RSA in triplicate. The root system of the contrasting rice genotypes grown under the two different conditions was separated from the shoots using a scalpel blade. The individual root sample was thoroughly washed with water and spread on the acrylic root scanning tray (40 cm × 30 cm) containing 0.5 cm deep water using the dissecting brush and forceps. The root system was scanned in greyscale at 600 dpi and the images were processed and analyzed using WinRHIZO Pro software v2009(Regent Instruments Inc, Montreal, Canada) for various root morphological traits [total root length (cm), total root projected area (cm2), total root surface area (cm2), the number of root tips,and the number of forks] on per plant basis.

Measurement of leaf area and chlorophyll content

The area of individual leaves for the contrasting rice genotypes was measured in three replications using a portable Leaf Area Meter (LiCOR 3100, Lincoln, Nebraska, USA) for the contrasting rice genotypes. The leaves were detached and passed through the leaf area meter to determine leaf area per plant. Total chlorophyll content in the leaves was estimated by using the dimethyl sulfoxide (DMSO) method as described previously(Kaur et al, 2023). Briefly, 0.5 g leaf tissues were cut into small(5 mm) pieces and incubated in 2 mL DMSO at 60 °C for 20 min. After repeated extraction with DMSO, the final volume of the chlorophyll extract was made up to 10 mL with DMSO.The absorbance of the extract was measured at 645 and 663 nm using a UV-Visible Spectrophotometer (EvolutionTM220,Thermo ScientificTM, NY, USA), and total chlorophyll content was calculated using the formula: Total chlorophyll content(mg/g) = (20.2 ×A645) + (8.02 ×A663).

Estimation of starch content

Starch content in shoot and root tissues was determined using the Anthrone method according to Clegg et al (1956). Starch was extracted from 100 mg of freshly collected tissues using hot 70% ethanol. After centrifugation at 10 000 ×gat 4 °C for 10 min, the supernatant was discarded and the pellet was washed with 70% hot ethanol. To the washed pellet, 5 mL water and 6.5 mL of 52% perchloric acid were added and mixed for 5 min followed by centrifuging at 10 000 rpm at 4 °C for 10 min. This step was repeated three times with 5 mL of fresh perchloric acid, the supernatants were pooled together,and the final volume was made up to 100 mL with double distilled water (DDW). Then, 0.1 mL of the aliquot and 5 mL of Anthrone reagent (0.2%) were mixed and heated in a boiling water bath for 10 min. Finally, it was cooled rapidly and the intensity of color was measured at 620 nm using a UV-Visible Spectrophotometer (EvolutionTM220, Thermo ScientificTM, NY,USA). The starch content was calculated using the standard curve prepared with glucose.

Estimation of sucrose content

Sucrose content in leaf and root tissues was determined using the modified Anthrone method (Finley and DA, 1973). The tissue samples (1 g) were powdered using liquid nitrogen and homogenized with 50 mL of 80% ethanol. The homogenate was incubated in a boiling water bath for 15 min and then cooled immediately. The volume was made up to 100 mL with 80% ethanol, 1 mL of aliquot was added with 9 mL of Fehling solution, and the mixture was placed in a boiling water bath for 15 min, followed by cooling down to room temperature (RT).Finally, 10 mL of Anthrone reagent (0.2%) was added to 1 mL of the aliquot and incubated for 30 min at 40 °C. The absorbance of the resultant was measured at 610 nm against a reagent blank using the UV-Visible Spectrophotometer(EvolutionTM220, Thermo ScientificTM, NY, USA). The sucrose content in tissue samples was calculated using a standard curve prepared with sucrose.

Analysis of N and protein contents

N and protein contents in shoot and root tissue samples were determined using the Kjeldahl method. To estimate N content,100 mg of dried tissue was digested with concentrated H2SO4(25 mL) in the presence of sodium thiosulphate as described earlier (Kumar et al, 2022a). The N and protein contents (%)were calculated using the formula:N(%) = (1.4 ×NV) /W,whereNrepresents normality of acid,Vrepresents volume of acid used to complete normalization of NH3, andWrepresents weight of the sample. For protein content, protein content (%) =N(%) × 6.25.

Analysis of P content

To determine the total P content in shoot and root tissues, a vanadate-molybdate method was used (Hanson, 1950). The oven-dried tissue samples (1 g) were digested with the double acid mixture containing 3.5 mL of HClO4and 10 mL of HNO3at 200 °C for 90 min. The digest was filtered through Whatman filter paper No. 1, and the volume was made up to 100 mL with DDW. In a 50 mL volumetric flask containing 10 mL of vanadate-molybdate solution, 25 mL of the extract/digest was added and the volume was made up to 50 mL with DDW. The mixture was vortexed, incubated at RT for 10 min for color development, and absorbance was recorded at 420 nm using an atomic absorption spectroscopy. The total P content in the tissue sample was calculated from the standard curve prepared using 1-5 mg P and expressed as mg/g of dry weight.

Estimation of APase activity

APase activities in shoot and root tissues were estimated using the method described by Johnson et al (1973). Fresh tissues(100 mg) were homogenized in 5 mL of ice-chilled citrate buffer (0.1 mol/L, pH 5.2). The homogenate was centrifuged at 8 000 ×gat 4 °C for 15 min and the supernatant was used as the enzyme extract for APase activity assay. A reaction mixture(3 mL), consisting of 0.1 mL enzyme extract, 0.4 mL ice-cold citrate buffer (0.1 mol/L, pH 5.2), and 0.5 mLp-nitrophenol (10 mmol/L p-NP, pH 5.2), was incubated at RT for 10 min for the enzymatic reaction to complete. The reaction was terminated by adding 2 mL of Na2CO3(0.2 mol/L) and absorbance was measured at 405 nm. Enzyme activity was calculated by plotting OD values against a standard curve for p-NP (10-100 mmol/L) and expressed as mmol/L of p-NP released per min per mg of protein.

Estimation of protein kinase activity

Total protein kinase activity in shoot and root tissues was determined using the ADP-GloTM Kinase Assay kit V6930(Promega, USA). Fresh tissues were powdered using liquid nitrogen and homogenized in 0.5 mL of freshly prepared 1×kinase reaction buffer followed by centrifugation at 13 000 ×gat 4 °C for 15 min. The supernatant was used as an enzyme extract for kinase activity assay. The assay was performed in a 96-well sterile microplate using 25 μL of enzyme extract. The plate was incubated at RT for the kinase reaction and the reaction was terminated by adding 25 μL of ADP-Glo reagent.The reaction mixture was incubated at RT for 40 min, and 50 μL of kinase detection buffer was added. After adding luciferase and luciferin, the reaction mixture was incubated at RT for 60 min and luminescence was measured using a luminometer (GloMax?Discover System, Promega, USA).Protein content in the tissue samples was determined by the Bradford method (Bradford, 1976). Protein kinase activity was calculated using the standard curve prepared using different concentrations of ATP-ADP ratio as described by the manufacturer. Total protein kinase activity was expressed in RLU (μmol/L of ADP/min/mg).

Protein extraction, liquid chromatography tandem mass spectrometry, and data analysis

For proteome analysis, proteins were extracted from shoot and root tissues (1 g) collected from 45-day-old plants of contrasting rice genotypes grown hydroponically under -P and +P conditions using the trichloroacetic acid (TCA)/acetone precipitation method.The sample tissues (1 g) were homogenized using liquid nitrogen in 5 mL sucrose buffer (30% of sucrose, 2% of SDS, 0.1 mol/L of Tris-HCl, pH 8) containing 5% of β-mercaptoethanol, 1 mmol/L of PMSF, and phenyl isothiocyanate (C7H5NS). After vertexing and mixing with a tissue homogenizer, the mixture was placed on a platform rocker for 1 h. The contents were centrifuged at 10 000 ×gat 4 °C for 15 min and an equal volume of Tris-buffered phenol was added followed by incubation on the platform rocker at RT for 30 min. The content was centrifuged again at 10 000 ×gat 4 °C for 30 min.The lower phenol layer was collected and added with 6 volumes of 100 mmol/L ammonium acetate/methanol solution,incubated overnight at -20 °C. Subsequently, the content was centrifuged at 10 000 ×gat 4 °C for 15 min, and the pellet was washed with pre-chilled acetone, air-dried and dissolved in 50 mmol/L ammonium bicarbonate solution. Then, TCA was added to the concentration of 10% and kept on ice for 10 min.The content was centrifuged at 13 000 ×gat 4 °C for 10 min and the whitish fluffy pellet was washed 3-4 times with acetone. Finally, the pellet was air-dried and dissolved in 100 μL Tris-buffer. One-dimension (1D) electrophoresis was performed using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The protein bands were visualized by using a silver-staining kit (Sigma, St. Louis, USA),and trypsin digestion was performed on a 100 mg protein sample diluted with 50 mmol/L NH4HCO3. After pretreatment of the proteins with 100 mmol/L DTT at 95 °C for 1 h and 250 mmol/L iodoacetamide at RT in the dark for 45 min,in-solution protein digestion (1:100, enzyme : protein) was performed at 37 °C overnight. The digested protein sample was vacuum dried and dissolved in 50 μL of 0.1% formic acid.

Liquid chromatograph-tandem mass spectrometry (LC-MS/MS)analysis was performed by the Ultra-Performance Liquid Chromatography (Nano ACQUITY, Waters, UK). For peptide separation, a 10-μL sample was injected into the BEH C18 column (75 μm × 150 cm × 1.7 μm). For gradient elution,buffers A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) were used, and the separated peptides were directed to SYNAPT?G2 Q-TOF (Waters, MA, USA) equipped with Nano-Electrospray Ionization detector. MS/MS spectra thus generated were searched against the UniProtKB proteome database (https://www.uniprot.org/) using MassLynx 4.1 WATERS ver3.0.2 software [Protein Lynx Global Server(PLGS) Waters, MA, USA]. For quantitative analysis, an label free quantitation algorithm was used with peptide tolerance set to 50 mg/kg on the complete scan and the fragment tolerance set to 100 mg/kg (Tsai et al, 2016). The minimum number of fragment matches for peptides and proteins was set to 2 and 5,respectively. The maximum missed cleavages allowed for data search was set to 1. The fixed changes were identified as carbamidomethylation of Cys and oxidation of Met. The proteins with a PLGS score ＞ 50 and adjustedPvalue ≤ 0.05 were considered as DEPs between the -P and +P samples.

Statistical analysis

Statistical analysis of the physiological and biochemical data was performed using R studio 4.2.1. (https://www.rstudio.com/).The acquired data were subjected to a two-factor (Genotype ×Treatment) analysis of variance (ANOVA) in a general linear model using the package. The mean differences were evaluated using the Tukey’s HSD test with the help of library Agricola.Descriptive statistics of the traits were graphically represented using a bar plot representing the standard deviation (± SD) for the mean of three replications (n= 3).

ACKNOWLEDGMENTS

The study was funded by the financial support received from the Centre of Advanced Agricultural Science and Technology-National Agricultural Higher Education Project jointly funded by the World Bank and ICAR (Grant No. 8776-IN-P151072).Prathap V acknowledges the Senior Research Fellowship from Council for Scientific and Industrial Research. Suresh KUMAR acknowledges financial support from the Indian Council of Agricultural Research [Grant No. 18(3)/2018-O&P], New Delhi, for research on P deficiency stress in rice.

SUPPLEMENTAL DATA

The following materials are available in the online version of this article at http://www.sciencedirect.com/journal/rice-science;http://www.ricescience.org.

Fig. S1. Gene Ontology (GO) functional enrichment analysis of differentially expressed proteins (DEPs) belonging to biological process under P-starvation stress in shoot of contrasting rice genotypes.

Table S1. Gene Ontology (GO) analysis of differentially expressed proteins in shoot of Pusa-44 under phosphorus starvation.

Table S2. Gene Ontology (GO) analysis of differentially expressed proteins in shoot of NIL-23 under phosphorus starvation.

Table S3. Differentially expressed proteins associated with phosphorus metabolic process in shoot of contrasting rice genotypes under phosphorus-starvation stress.

Table S4. Some of top five highly up-regulated proteins in shoot and root of NIL-23 in response to phosphorus-starvation stress.

Table S5. Highly down-regulated proteins in shoot and root of

NIL-23 in response to phosphorus-starvation stress.