- © 2018 American Society of Plant Biologists. All rights reserved.
Abstract
The nitrogen (N)-rich ureides allantoin and allantoate, which are products of purine catabolism, play a role in N delivery in Leguminosae. Here, we examined their role as an N source in nonlegume plants using Arabidopsis (Arabidopsis thaliana) plants mutated in XANTHINE DEHYDROGENASE1 (AtXDH1), a catalytic bottleneck in purine catabolism. Older leaves of the Atxdh1 mutant exhibited early senescence, lower soluble protein, and lower organic N levels as compared with wild-type older leaves when grown with 1 mm nitrate but were comparable to the wild type under 5 mm nitrate. Similar nitrate-dependent senescence phenotypes were evident in the older leaves of allantoinase (Ataln) and allantoate amidohydrolase (Ataah) mutants, which also are impaired in purine catabolism. Under low-nitrate conditions, xanthine accumulated in older leaves of Atxdh1, whereas allantoin accumulated in both older and younger leaves of Ataln but not in wild-type leaves, indicating the remobilization of xanthine-degraded products from older to younger leaves. Supporting this notion, ureide transporter expression was enhanced in older leaves of the wild type in low-nitrate as compared with high-nitrate conditions. Elevated transcripts and proteins of AtXDH and AtAAH were detected in low-nitrate-grown wild-type plants, indicating regulation at protein and transcript levels. The higher nitrate reductase activity in Atxdh1 leaves compared with wild-type leaves indicated a need for nitrate assimilation products. Together, these results indicate that the absence of remobilized purine-degraded N from older leaves of Atxdh1 caused senescence symptoms, a result of higher chloroplastic protein degradation in older leaves of low-nitrate-grown plants.
In plants, the degradation of purine compounds starts with the conversion of AMP to inosine monophosphate by AMP deaminase (AMPD; EC 3.5.4.6), which leads, by multiple pathways, to the production of oxypurines such as xanthine and hypoxanthine (Yoshino et al., 1979; Xu et al., 2005; Zrenner et al., 2006; Sabina et al., 2007). In the degradation pathway, xanthine is initially oxidized by xanthine dehydrogenase (XDH; EC 1.1.1.204) to urate, which is further converted by urate oxidase (UOX; EC 1.7.3.3) and a transthyretin-like protein to allantoin, the main end product in most mammals (Zrenner et al., 2006; Reumann et al., 2007; Werner and Witte, 2011; Hauck et al., 2014). Conversely, plants possess a set of enzymes that further break down allantoin to allantoate to ureidoglycolate, catalyzed by allantoinase (ALN; EC 3.5.2.5), allantoate amidihydrolase (AAH, EC 3.5.3.9.), and ureidoglycine aminohydrolase (EC 3.5.3-.), respectively (Werner et al., 2010, 2013). The ureidoglycolate amidohydrolase (EC 3.5.1.116.) converts ureidoglycolate to the basic metabolic building blocks, glyoxylate and ammonium (Werner et al., 2013). The release of four ammonium molecules, which essentially should be reassimilated, parallels the sequential hydrolysis of purines (Werner et al., 2010).
Environmental stimuli can induce premature leaf senescence (Miller et al., 1999; Munné-Bosch and Alegre, 2004; Pageau et al., 2006; Lim et al., 2007). Among these stimuli is nitrogen (N) deficiency, which leads to accelerated yellowing and senescence of old leaves (Smart, 1994; Thomas and DeVilliers, 1996; Pourtau et al., 2004; Kato et al., 2005; Diaz et al., 2006; Peng et al., 2007), whereas younger leaves remain green, as a result of nutrient mobilization from the older senescing leaves (Smart, 1994)
N remobilization from older leaves is mainly in the form of amino acids originated from degraded proteins (Hirel et al., 2001; Hörtensteiner and Feller, 2002; Mickelson et al., 2003; Jukanti and Fischer, 2008), and nutrient recycling in older leaves is considered the common destiny of two processes: plant senescence and proteolysis (Diaz-Mendoza et al., 2016). Proteolysis of plastidial proteins in older leaves provides the main source of N for remobilization, and most studies have been focused on chloroplast rather than on the other cellular compartments. The proteolysis and breakdown of Rubisco during senescence and its significance in N translocation have been shown (Kato et al., 2005; Diaz-Mendoza et al., 2016). In terms of plastidial protein breakdown and the resulting amino acid degradation and remobilization to young leaves, a role for the N-rich ureides allantoin and allantoate as a potential N internal source was rarely suggested and never demonstrated before. If ureides are employed as an N source by nonlegume plants, then impairments that prevent ureide generation or its degradation may result in premature senescence as a result of the degradation of plastidial proteins, such as Rubisco in the older leaves, to substitute the missing ureides and to remobilize the resulting N metabolites to the younger growing leaves.
Leaf senescence occurs in a coordinated manner (Lim et al., 2003); it starts from the inhibition of leaf expansion (Diaz et al., 2005), followed by the induction of metabolic changes that result in nutrient degradation and remobilization (Pate, 1980; Simpson et al., 1983). This strategy of recycling endogenous nutrients from the senescing leaves is used by plants to maintain the growth of younger leaves and reproductive organs under nutrient-limiting stress (Aerts, 1990; Buchanan-Wollaston and Ainsworth, 1997; Hörtensteiner and Feller, 2002; Eckhardt et al., 2004). Proteins are the major source for N in the senescing leaves, contributing more than 50% of leaf N remobilization (Masclaux-Daubresse et al., 2010), yet nucleobases such as purines are rich in N (Thomas et al., 1980; Atkins et al., 1982; Schubert, 1986; Brychkova et al., 2015) and, thus, are a potential source for N recycling. In legumes, the N fixed in the form of ureides was indeed shown to be translocated from the nodules, where the ureides are synthesized de novo, to the aerial plant tissues, where they are degraded and used as an N source (Stebbins and Polacco, 1995; Smith and Atkins, 2002; Todd et al., 2006). Importantly, a role for nucleic acid/purine degradation products in plant N metabolism also was presented in non-N-fixing legumes, and the recycling of nucleic acids in tissues undergoing stress-induced premature senescence was considered a possible source for increased ureides in nonnodulated common bean (Phaseolus vulgaris) plants (Alamillo et al., 2010). The high levels of ureides evident in shoot and leaves of nonnodulated common bean plants fertilized with nitrate was suggested to be the result of remobilized N from senescent leaves to be employed for new growing tissue (Díaz-Leal et al., 2012).
The results with nonnodulated legume plants may suggest a role for ureides as an endogenous N source in nonlegume plants such as Arabidopsis (Arabidopsis thaliana). Indeed, when supplemented externally to the growth medium, uric acid, allantoin, and allantoate, the products of purine degradation, serve as sole N sources during the growth of Arabidopsis plants (Desimone et al., 2002; Todd and Polacco, 2004; Nakagawa et al., 2007). However, Arabidopsis plants grown with allantoin as a sole N source (Werner et al., 2008) exhibited reduced growth (Desimone et al., 2002).
The role of senescence-induced endogenous purine degradation products (ureides) has not been fully examined in plants, and no functional analysis employing relevant mutants has yet proved that purine degradation products have a role in N metabolism (Havé et al., 2017; Oszvald et al., 2018). The facts that leaf senescence is paralleled by a decrease in RNA (Crafts-Brandner et al., 1996, 1998) and an increase in the level of uriedes, whereas enzymes/transcripts of genes involved in the purine degradation pathway are up-regulated during leaf senescence (Brychkova et al., 2008), suggest the involvement of degraded products of purine pools, such as RNA, in N metabolism (Werner and Witte, 2011; Havé et al., 2017).
To examine the role of purine-degraded metabolites as an important N source in Arabidopsis plant development, the knockout mutants xanthine dehydrogenase1 (Atxdh1), allantoinase (Ataln), and allantoate amidohydrolase (Ataah) were studied under sufficient and limited N conditions. Under N-deficient conditions, the purine degradation pathway was activated at transcript and protein levels to provide an additional source of N from older senescent leaves to young leaves. Growth of the purine catabolism mutants under nitrate limitation resulted in premature senescence symptoms in old leaves of mutants but not in those of wild-type plants. In contrast, sufficient supply of nitrate resulted in the disappearance of the premature senescence symptoms, with parallel enhancement of organic N and soluble protein content in mutant older leaves. This was achieved by higher nitrate reductase activity in Atxdh1 than in wild-type leaves. Furthermore, it was demonstrated that the absence of remobilized purine-degraded N from older leaves was the cause for senescence symptoms in the nitrate-starved Atxdh1, a result of the higher degradation of chloroplastic proteins, such as Rubisco large subunit, and D1, the component of the reaction center of PSII. Importantly, this was followed by increased remobilization from the older leaves of low-nitrate-grown Atxdh1 plants, indicated by the enhancement of autophagy-related protein5 (ATG5) and ATG8A, which are essential for N and carbon remobilization (Thompson et al., 2005; Phillips et al., 2008; Honig et al., 2012).
RESULTS
High-Nitrate Supplementation Prevents Senescence Symptoms in Atxdh1 Older Leaves
A mutation in AtXDH1, a key enzyme in the purine catabolism process, confers early leaf senescence (Brychkova et al., 2008). To examine the role of purine-degraded metabolites as an important endogenous N source, we studied the effect of sufficient and limited N supplementation to the knockout mutants Atxdh1, Ataln, and Ataah with 25-d-old plants. Total chlorophyll levels in Atxdh1, Ataln, and Ataah old leaves supplemented with low nitrate (1 mm) were lower than in wild-type old leaves, whereas there was no difference in the young leaves. Importantly, increasing nitrate levels in the growth medium (5 mm) enhanced the total chlorophyll level in mutant old leaves (Fig. 1, A and B; Supplemental Fig. S1). Additionally, expression levels of the senescence marker Cys protease SENESCENCE-ASSOCIATED GENE12 (SAG12; Gepstein et al., 2003), the chlorophyll-degradation gene ACCELERATED CELL DEATH2 (ACD2; Tanaka et al., 2003), and STAY-GREEN PROTEIN1 (SGN1; Park et al., 2007) were up-regulated significantly in the old leaves of the mutants supplemented with the lowest nitrate level as compared with the wild type but not in mutant old leaves in plants grown with high nitrate (Fig. 1C; Supplemental Fig. S1, C and D). Similar senescence symptoms in the older leaves of low-nitrate-supplied mutants impaired in XDH1, ALN, or AAH genes were observed, while the wild type was not affected (Fig. 1; Supplemental Fig. S1), indicating a shortage of an endogenous N source related to the purin catabolism pathway in the three mutants.
Effects of different nitrate levels on senescence symptoms in wild-type (Columbia [Col]) and Atxdh1 leaves. A, Leaf appearance from left to right is old to young, where the first and last four leaves are designated young and old, respectively. B, Total chlorophyll content. FW, Fresh weight. C, Relative expression of senescence-related markers in old leaves. SAG12, ACD2, and SGN1 are At5G45890, At4G37000, and At4G22920, respectively. The chlorophyll content data represent means obtained from a representative experiment from six independent biological replications. Values denoted by different letters are significantly different (Tukey-Kramer honestly significant difference [HSD] test, P < 0.05). The expression of each transcript was compared with that of young leaves of the wild type in 5 mm nitrate treatment after normalization to ELONGATION FACTOR1-α (EF-1α; At5g60390). Values marked with asterisks denote significant differences (Student’s t test, n = 3, P < 0.05) between treatment and genotypes for each transcript, and the data represent means obtained from three independent experiments. Error bars are defined as se.
The Atxdh1 Mutation Confers Lower Organic N and Soluble Protein Levels But Higher RNA Than in Wild-Type Old Leaves Grown under N Deficiency
The levels of total N in its various forms are key physiological indicators of plant health. Examination of N levels in young and old leaves grown under low (1 mm) and high (5 mm) nitrate revealed a total N decrease in wild-type and Atxdh1 leaves with decreasing nitrate application. However, total N was significantly lower in old leaves of Atxdh1 as compared with the wild type under low N (Fig. 2A). Interestingly, under low-N supplementation, organic N was considerably lower in old and young leaves of the Atxdh1 mutant as compared with the wild type (decrease of 1.03 and 0.83 mmol N g−1 dry weight, respectively), whereas no significant difference was noticed between the leaves of these two genotypes when fed with high nitrate (Fig. 2B). This indicates that, under low-nitrate conditions, the mutation in XDH1 stimulates the degradation of organic N and protein in the older leaves (Fig. 2) to supply N essential for the growth of the younger leaves. This is consistent with the early-senescence phenotype in mutant older leaves and its absence in the younger leaves, albeit the organic N in the latter was lower than in the wild type but still significantly higher as compared with the old leaves in Atxdh1 plant grown under low nitrate supply (Figs. 1 and 2).
Effects of different nitrate levels on protein and RNA levels in old and young leaves of the wild type (Col) and Atxdh1. Total N (A), total organic N (B), nitrate (C), ammonium (D), soluble protein content (E), and total RNA content (F) were determined in 25-d-old plants grown on N-deficient soil supplemented with 1 or 5 mm NaNO3 as the only N source. The data represent means obtained from six independent experiments. Values denoted with different letters are significantly different according to the Tukey-Kramer HSD test (P < 0.05). Different uppercase letters in A to D indicate differences between treatments. Different uppercase letters in E and F indicate differences between treated young leaves. Different lowercase letters in E and F indicate differences between treated old leaves. Error bars are defined as se. DW, Dry weight; FW, fresh weight.
The high-nitrate application increased nitrate content in the young and old wild-type and Atxdh1 leaves, being lower in young leaves of the Atxdh1 mutant as compared with wild-type plants (Fig. 2C). The significantly lower nitrate accumulation in Atxdh1 younger leaves indicated a higher rate of nitrate assimilation in the mutant younger leaves to overcome the absence of ammonium originated from unimpaired purine catabolism and employed for organic N biosynthesis in the wild type. Indeed, the ammonium content in the Atxdh1 mutant was similar to that in the wild type grown under high-nitrate conditions (Fig. 2D), indicating that nitrate was assimilated to ammonium for incorporation into organic molecules (Somerville and Ogren, 1980; Joy, 1988; Stitt, 1999; Coruzzi, 2003; Wang et al., 2003). Interestingly, total soluble protein content in young leaves was similar in wild-type and mutant plants and was significantly higher than that of the old leaves of both genotypes. However, soluble proteins were significantly lower in old leaves of 1 mm nitrate-fed mutant plants as compared with the wild type (decreased by 33% [by 0.13 mg soluble protein g−1 fresh weight]), the latter having similar soluble protein levels as the 5 mm-supplied wild-type and mutant plants (Fig. 2E). The lowest soluble protein and organic N in old leaves of the mutant fed with low nitrate (Fig. 2, B and E) indicated the remobilization of degraded protein products from these leaves.
The estimation of total RNA level does not represent the whole pool of purine-degraded compounds, since there are additional purine pools in plants, such as nucleosides and bases (e.g. AMP, ADP, ATP, GMP, and GDP), purine alkaloids (e.g. 3-methylxanthine, 7-methylxanthosine, and theobromine), coenzymes (e.g. NAD, NADP, FAD, and CoA), and adenylosuccinate, as well as isopentenyl AMP, S-adenosyl-l-Met, S-adenosyl-l-homo-Cys, and more (Meyer and Wagner, 1986; Ranocha et al., 2001; Smith and Atkins, 2002; Koyama et al., 2003; Sabina et al., 2007; Ashihara et al., 2008; Lange et al., 2008; Agrimi et al., 2012). Yet, RNA estimation can act as an indicator. With this caveat in mind, significantly lower RNA levels in wild-type old leaves fed with low nitrate as compared with the RNA levels in the wild type supplemented with high nitrate or mutant leaves fed with low or high nitrate indicated the employment of a purine degradation product for N remobilization from older leaves to younger leaves, where no differences in RNA levels were evident within nitrate treatments or between genotypes (Fig. 2F). Considering the significant level of xanthine normally accumulated in Atxdh1 leaves (Brychkova et al., 2008; Ma et al., 2016), a significant decrease in RNA level would be expected in Atxdh1 old leaves if a similar purine degradation, as in low-nitrate-grown wild-type old leaves, had occurred. The lack of a significant decrease in low-nitrate-grown mutant old leaves likely indicates a certain level of feedback-inhibited purine degradation as a result of accumulated xanthine and/or possible preference for non-RNA purines at the initial stage of purine catabolism in the mutant old leaves. Overall, the significant rate of decrease in total N, organic N, and soluble proteins and the higher RNA in the old mutant leaves as compared with wild-type old leaves in low-nitrate-fed plants suggest negligible N remobilization from purines and significant N degradation and protein remobilization from old leaves in the mutant.
Low N Supplementation Confers Enhancement of Ureide Transporter Transcripts in Old Leaves of the Wild Type
Ureides, which are purine degradation products, are transported from the nodules of legume roots via the xylem (Collier and Tegeder, 2012) to the shoot (Schubert, 1981). Ortholog ureide permease (AtUPS) gene expression in yeast (Saccharomyces cerevisiae) and/or Xenopus laevis oocytes shows that AtUPS1 acts as xanthine and allantoin permeases, whereas AtUPS2 is a uracil transporter and AtUPS5 is more likely a xanthine and allantoin permease (Desimone et al., 2002; Schmidt et al., 2004, 2006). Recently, AtUPS5 was suggested to act as a key component in allantoin transport to the shoots (Schmidt et al., 2006; Lescano et al., 2016), whereas AtUPS1 is significantly up-regulated in adult Arabidopsis shoots in response to sudden total N starvation (Krapp et al., 2011). The lower total RNA level in old leaves of low-nitrate-supplied wild-type plants as compared with Atxdh1 (Fig. 2F) led us to explore the transcript expression of the UPS transporters related to the transport of allantoin. Overall, older leaves had higher levels of all UPS transporters compared with younger leaves. The expression of AtUPS1, AtUPS2, and AtUPS5 was significantly higher in old wild-type leaves supplemented with 1 mm nitrate compared with plants supplemented with 5 mm nitrate or with Atxdh1 plants supplemented with 1 or 5 mm nitrate (Fig. 3). Yet, the expression of AtUPS1, AtUPS2, and AtUPS5 in the old leaves of low-nitrate-fed Ataln plants was similar to the expression in wild-type old leaves (Fig. 3).
Effects of different nitrate levels on ureide permease expression in old and young leaves of the wild type (Col), Atxdh1, and Ataln. Quantitative analysis of AtUPS1 (A), AtUPS2 (B), and AtUPS5 (C) transcript levels by reverse transcription quantitative PCR (RT-qPCR) was performed using 25-d-old plants grown on N-deficient soil supplemented with 1 or 5 mm NaNO3 as the only N source. The expression of each treated line was compared with that of young leaves of Col in 5 mm nitrate treatment after normalization to EF-1α (At5g60390). The data represent means obtained from a representative experiment from two independent experiments (Tukey-Kramer HSD test, P < 0.05). Error bars are defined as se.
These results support the notion that UPS transporters mediate allantoin transport (Desimone et al., 2002; Pélissier et al., 2004; Schmidt et al., 2004, 2006; Lescano et al., 2016) and suggest a role for the UPS transporters in allantoin transport from old to young leaves. They also suggest that the induction of the UPS transcripts is sensitive to low nitrate supply and to the presence or flux of allantoin (Desimone et al., 2002); either the allantoin presence or its flux is absent in the Atxdh1 mutant (Fig. 4).
Effects of different nitrate levels on ureide content in old and young leaves of the wild type (Col), Atxdh1, and Ataln. The levels of xanthine (A) and allantoin (B) were quantified from plants grown on N-deficient soil supplemented with 1 or 5 mm NaNO3. FW, Fresh weight; ND, not detected. The data represent means obtained from one of five independent experiments with similar results. Values denoted with different letters are significantly different according to the Tukey-Kramer HSD test (P < 0.05). Different uppercase letters indicate differences between mutants and wild-type plants. Error bars are defined as se.
Comparing the Catabolism of Xanthine and Allantoin in the Wild Type and Atxdh1, Ataln, and Ataah Mutants
During catabolic activity, substrates are expected to accumulate within the Atxdh1, Ataln, and Ataah mutant lines. Therefore, it is of interest to examine the possibility of differential accumulation in wild-type and mutant young and old leaves. Importantly, xanthine accumulated chiefly in the old leaves of Atxdh1, was several fold higher than in the young leaves, and was significantly higher in old leaves in plants supplemented with low nitrate compared with old leaves of Atxdh1 fed with high nitrate. The xanthine level in wild-type old leaves was more than 13-fold lower than in the mutant (Fig. 4A). The possible stress effect of xanthine toxicity was examined in leaf discs sampled from sixth to tenth rosette leaves (counted from the bottom and being without senescence symptoms) exposed to water (mock) and 1 mm xanthine or allantoin for 48 h. Anthocyanin accumulation was used as a sign of stress (Chalker-Scott, 1999; Gould et al., 2002; Schüssler et al., 2008). Higher anthocyanin levels were evident in the presence of xanthine as compared with allantoin, especially in Atxdh1 when compared with wild-type leaf discs (Supplemental Fig. S2). This indicates that xanthine accumulation could contribute to stress and hastening of the senescence phenotype in Atxdh1 old leaves supplemented with low N (Fig. 1). However, this is likely not the case here (Fig. 1; Supplemental Fig. S1), since no senescence symptoms were evident in the xanthine-treated leaf discs (Supplemental Fig. S2). Significantly, lower xanthine levels were evident in the old leaves of 18- and 25-d-old Ataln and Ataah mutants and wild-type plants as compared with Atxdh1 (Fig. 4A; Supplemental Fig. S3). Yet, Ataln and Ataah also displayed enhanced senescence symptoms relative to the wild type under low-nitrate conditions (Fig. 1; Supplemental Fig. S1). The results (Fig. 4A; Supplemental Fig. S2) suggest that the senescence symptoms in the mutants are not the result of toxic xanthine accumulation.
Xanthine accumulation in the Atxdh1 mutant indicates a higher purine catabolic activity rate (Ma et al., 2016), and the higher xanthine accumulation in Atxdh1 old leaves, being the highest in the low-nitrate-fed plants, indicates high demand for purine degradation products blocked from being further catabolized and remobilized for young leaf growth (Fig. 4A). The low xanthine level in wild-type old leaves is a result of xanthine degradation and the further consumption of the degraded ureide products, as indicated by the low level of allantoin in the wild type and its enhancement in the old leaves of the low-nitrate-fed Ataln plants (Fig. 4B). Notably, the negligible allantoin levels evident in Atxdh1 leaves, which were much lower than in wild-type leaves (Fig. 4B), were shown by others (Brychkova et al., 2008; Watanabe et al., 2014) and can be explained by a slight undetectable XDH activity resulting from AtXDH2 or another yet unknown source. Importantly, the levels of allantoin accumulated in both young and old Ataln leaves were much higher than the level of xanthine accumulated in Atxdh1 leaves, indicating that the accumulated xanthine may further feedback inhibit purine degradation in Atxdh1 leaves. These results suggest that the range between the accumulated xanthine and allantoin, in Atxdh1 and Ataln, respectively, may be used as an indicator for the rate of purine catabolism in wild-type plants.
Interestingly, while a large difference between older and younger leaves is evident for xanthine accumulation, ureide accumulation was much less different (Fig. 4; Supplemental Fig. S3). This indicates that significant purine breakdown takes place in the older leaves and may be why allantoin, rather than xanthine, is more readily exported from the older leaves. These results further indicate that subsequent catabolic steps may take place in all leaves and are consistent with the observed elevated UPS transcript levels in wild-type older leaves (Fig. 3).
Expression of the Purine Degradation Network Is Up-Regulated by Low-Nitrate and Down-Regulated by High-Nitrate Application
During natural senescence and dark-induced senescence, an orchestrated regulation of transcripts related to purine catabolism was observed that includes the up-regulation of upstream purine catabolism transcripts, such as AtXDH1 and AtUOX, with parallel down-regulation of the downstream transcripts AtALN and AtAAH (Brychkova et al., 2008). We wished to elucidate how the purine catabolism gene network is orchestrated either by N deficiency-induced senescence and/or by sufficient N that prevents senescence in purine mutant leaves. Therefore, the transcripts of the purine-degrading enzymes were analyzed in plants supplemented with low and high nitrate as the only N source. The low-nitrate treatment resulted in an increase of purine catabolism gene transcripts in wild-type plants as compared with the high-N treatment. Whereas AtAMPD, AtAAH, and AtALN expression was increased in old and young leaves, the expression of AtXDH1 and AtUOX increased only in old leaves (Fig. 5). Interestingly, except for the AtXDH1 transcript, the Atxdh1 mutant showed a similar expression pattern of the purine degradation gene network to the wild type (Fig. 5). Notably, Ataln and Ataah mutants exhibited a similar expression pattern to the wild type and Atxdh1 (Fig. 5; Supplemental Fig. S4). Thus, we can conclude that the purine degradation transcript network is generally up-regulated by N deficiency and down-regulated by high-nitrate application.
Transcript expression of the purine catabolism genes in young and old leaves of wild-type and Atxdh1 plants supplemented with 1 or 5 mm nitrate as the only N source. Quantitative analysis of AtAMPD (A), AtXDH (B), AtUOX (C), AtALN (D), and AtAAH (E) transcripts by RT-qPCR was performed using wild-type (Col) and Atxdh1 25-d-old plants grown on N-deficient soil. The expression of each treated line was compared with that of young leaves of the wild type in 5 mm nitrate treatment after normalization to EF-1α (At5g60390). The data represent means obtained from three independent experiments. Values denoted by different letters are significantly different (Tukey-Kramer HSD test, P < 0.05). Error bars are defined as se.
Low-Nitrate Supplementation Confers Enhanced XDH Activity and AAH Protein in Leaves
To examine the levels of protein expression, we chose XDH1 and AAH as representative proteins of the upstream and downstream components of purine catabolism, to be estimated by activity gels and immunodetection, respectively (Fig. 6). Two XDH activity bands appeared in the wild type, whereas no activity bands were noticed in proteins extracted from Atxdh1 leaves. These results indicated elevated levels of XDH activity in low nitrate in both old and young leaves (Fig. 6A).
Expression analyses of XDH and AAH proteins in young and old leaves of wild-type (Col) and Atxdh1 plants supplemented with 1 or 5 mm nitrate as the only N source. Activity of XDH (A) and immunoblot analysis of AAH (B) are shown. Protein was extracted from old and young leaves of the wild type (Col) and Atxdh1 grown on 1 or 5 mm NaNO3 as the only N source. The general activity of XDH on a Native-SDS-PAGE gel was detected by using phenazine methosulfate (PMS) as the electron carrier intermediate and 3(4,5-dimethylthiazolyl-2)2,5-diphenyltetrazolium bromide as the electron acceptor. AAH protein level was analyzed by immunoblotting employing a specific antiserum against AAH. Protein extracted from Ataah mutant leaves was used as the negative control. The data represent one of three independent experiments with similar results.
AAH protein expression was evaluated by immunoblotting after Native-PAGE, employing highly specific antisera (a gift from Claus-Peter Witte). The AAH protein was detected as one band in the wild type and was verified as absent in Ataah (Fig. 6B). Interestingly, analyses of AAH levels in the wild type and in Atxdh1 revealed that the AAH level increased in old and young leaves with decreasing nitrate supplementation (Fig. 6B). These results indicate that XDH1 and AAH protein and activity generally are in agreement with the results of the transcript expression, demonstrating the enhanced generation of ureides and their degradation for further deployment of N during deficiency (Figs. 5 and 6; Supplemental Fig. S4).
Nitrate Reductase Activity in 18- and 25-d-Old Atxdh1 and Wild-Type Plants Fed with High and Low Nitrate Levels
Nitrate reductase (NR) catalyzes the first step of nitrate assimilation toward the biosynthesis of ammonia by generating the intermediate nitrite (Campbell, 1988; Kaiser and Huber, 1994; Sivasankar and Oaks, 1996). To examine the influence of purine catabolism potential, NR activity was estimated in young and old leaves of 18-d-old wild-type and Atxdh1 mutant plants supplemented with high and low nitrate as the only N source (Fig. 7A). In general, NR activity was significantly higher in younger leaves than in older leaves. These results are consistent with the finding that primary N-assimilating enzymes decrease with aging (Masclaux et al., 2000). In addition, the level of NR activity was always higher in the mutant line. Lower nitrate levels were detected in the young as compared with the old leaves of 25- and 18-d-old plants (Figs. 2C and Fig. 7B, respectively) and in the young leaves of Atxdh1 compared with the wild type in the 18-d-old plants (Fig. 7B), indicating higher nitrate assimilation by NR (Fig. 7A). The higher nitrate content in old Atxdh1 as compared with low-nitrate-grown wild-type old leaves (Fig. 7B) is likely the result of the higher uptake of nitrate by the mutant plant in the absence of allantoin in the young growing leaves, whereas the NR activity rate in the old mutant leaves is lower than in the young mutant leaves (Fig. 7A). To this end, the higher NR activity in the mutant younger leaves, followed by lower nitrate, indicates a compensation activity for the lack of a remobilized purine-dependent N source transported from the older to the younger leaves in the low-nitrate-grown Atxdh1 plants.
Effects of different nitrate levels on NR activity and nitrate content in old and young leaves of the wild type (Col) and Atxdh1. NR activity (A) and nitrate content (B) were analyzed in 18-d-old plants grown on N-deficient soil supplemented with 1 or 5 mm NaNO3 as the only N source. The data represent means obtained from three independent experiments. Values denoted by different letters are significantly different (Tukey-Kramer HSD test, P < 0.05). Error bars are defined as se. DW, Dry weight.
The Destiny of Allantoin Exogenously Applied to Old Leaves of the Wild Type and Atxdh1
Atxdh1 is unable to generate ureides as a result of a mutation in the purine catabolism pathway, which is upstream of ureide biosynthesis, and thus is an excellent tool with which to evaluate the flux and destiny of exogenously applied allantoin. We infiltrated an equal volume of 5 mm allantoin into the four oldest leaves of 18-d-old nitrate-starved Atxdh1 mutants and wild-type plants and detected allantoin and its degraded products allantoate, glyoxylate, and ammonium in the oldest (leaves 1–4 from the bottom), middle (leaves 5–8), and youngest (leaves 9–12) leaves during the 3 h after application. The net contribution of the infiltrated allantoin was evaluated by deducting the level of the metabolites detected in the various leaves of plants similarly infiltrated with solution without allantoin from the level detected when infiltration was performed with allantoin.
Of the applied allantoin (1.53 and 1.55 µmol g−1 fresh weight), approximately 22% and 20% was degraded to allantoate (0.34 and 0.32 µmol g−1 fresh weight) in wild-type and mutant leaves, respectively. Additional degradation of 3.5% (0.053 µmol g−1 fresh weight) to glyoxylate occurred in mutant leaves 30 min after the application. The distribution of these metabolites did not change much in the mutant after 3 h, whereas an additional amount of endogenously generated allantoin, allantoate, and glyoxylate (1.47, 0.53, and 0.03 µmol g−1 fresh weight, respectively) was noticed in the wild type (Fig. 8A).
Remobilization of infiltrated allantoin and its degradation to downstream metabolites in leaves of wild-type (Col) and Atxdh1 plants. A, Distribution of allantoin, allantoate, and glyoxylate in wild-type and Atxdh1 plants 30 min and 3 h after allantoin infiltration. The infiltrated amount of allantoin is presented. B to E, Allantoin (B), allantoate (C), glyoxylate (D), and ammonium (E) levels in old (leaves 1–4 from the bottom), middle (leaves 5–8), and young (leaves 9–12) leaves. The allantoin (5 mm) solution or water was infiltrated by injection into the old leaves (the first four leaves from the bottom) of 18-d-old 1 mm nitrate-treated plants (see sulfite infiltration by injection [Brychkova et al., 2012]). The data are presented as the net level of each metabolite, calculated by subtracting the level detected in the water-infiltrated plant from the detected metabolite. Values denoted with different letters are significantly different according to the Tukey-Kramer HSD test (P < 0.05). Error bars are defined as se. FW, Fresh weight.
The net distributed allantoin and its degradation products allantoate, glyoxylate, and ammonium in the various leaves showed that, 30 min after allantoin infiltration, approximately 30% of the allantoin was transported from the older leaves and allocated mainly (67% of the transport) to the younger leaves and the rest to the middle leaves in both genotypes. Importantly, 3 h after the application, the level of allantoin decreased in the older leaves and, as a result, was enhanced in the younger leaves but not in the middle leaves of Atxdh1. In contrast, the level of allantoin was enhanced in wild-type older leaves by internal purine-degraded sources, yet the increase in the younger leaves was at the same level as in the mutant (Fig. 8B). These results indicated a preference of allantoin transport from older to younger leaves. Notably, the allantoate that originated in Atxdh1 leaves only as a result of the infiltrated allantoin degradation generally followed the distribution of allantoin (Fig. 8, B and C). These results indicate that allantoate generated in the older leaves either was transported or the transported allantoin was degraded in the young leaves.
The detection of one molecule of glyoxylate, resulting from the net contribution of infiltrated allantoin, implies the release of four molecules of N in the form of ammonia, a result of the degradation of one allantoin molecule (Werner and Witte, 2011). Importantly, net glyoxylate enhancement was evident in the middle and younger leaves of the mutant but not in the wild type 30 min after the application. Yet, a significant amount of glyoxylate was detected in both wild-type and mutant younger leaves as well as in mutant older leaves after 3 h (Fig. 8D). The allantoin likely originated from the older leaves, whereas the steps of allantoate degradation to glyoxylate occurred in all leaves. This was further supported by the enhanced net ammonium level, the result of the infiltrated allantoin-degraded products in all leaves, especially in the older and younger wild-type and mutant leaves 30 min after the application. Three hours after the application, the net ammonium was more evident in wild-type old and young leaves than in mutant leaves (Fig. 8E), likely as a result of the higher ureide level (endogenous plus infiltrated) in wild-type leaves and/or the higher need for an N source by the nitrate-starved mutant.
DISCUSSION
In tropical legumes such as soybean (Glycine max), cowpea (Vigna unguiculata), and common bean, symbiotic N fixation in plant nodules leads to the incorporation of the fixed N into purine nucleotides to be converted into ureides, which are used as major nodule (root)-to-shoot N transport compounds (Schubert, 1986; do Amarante et al., 2006). In other plants, leaf senescence is accompanied by a decrease in nucleic acid content (Masclaux et al., 2000). Hence, it is reasonable to consider that relatively N-enriched ureides (1:1 carbon:N ratio) could serve as possible endogenous conduits for remobilized N during plant growth and development in nonlegumes (Havé et al., 2017). In support of this notion is the observation that ureides accumulate in several nonlegume shrubs and trees, likely for storage and translocation of N (Schmidt and Stewart, 1998).
The presence of all components necessary for purine catabolism and its recycling and remobilization are well established in nonlegume plants (Todd and Polacco, 2006; Todd et al., 2006; Zrenner et al., 2006; Werner and Witte, 2011). However, the physiological role of purine degradation, how it contributes to N remobilization, and overall N nutrition and the conditions under which it becomes a critical limiting step in organ survival are not known (Havé et al., 2017; Oszvald et al., 2018).
By analyzing the central mutations of the purine degradation pathway under moderate N-limiting conditions that did not affect biomass accumulation as compared with the wild type grown with sufficient N (Supplemental Fig. S5), we show the critical role of the purine-degraded product in providing N to the young leaves.
The Integration of N Starvation and Impairment in Purine Catabolism Results in Early Senescence Symptoms in Old Leaves of Arabidopsis Plants
Leaf senescence hallmarks, such as decreases in chlorophyll level, soluble protein content, and organic N content and enhancement of senescence molecular markers, were evident in the Arabidopsis mutant plants Atxdh1, Ataln, and Ataah supplemented with 1 mm nitrate as the only N source (Figs. 1 and 2; Supplemental Fig. S1). Leaf senescence has an important role in N metabolite management to remobilize important degraded N-containing components for the reassimilation of N resulting from chloroplast degradation, hydrolysis of stromal proteins, and other degraded organelles and cell components (Masclaux et al., 2000; Hörtensteiner and Feller, 2002; Eckhardt et al., 2004; Fischer, 2007; Liu et al., 2008). A similar metabolic strategy was suggested to take place in senescence resulting from N limitation (Aerts, 1990). Significantly, old leaves of wild-type plants that underwent nitrate starvation did not show senescence symptoms, whereas the various independent mutants (Atxdh1, Ataln, and Ataah) impaired in the purine catabolism pathway did (Figs. 1 and 2; Supplemental Fig. S1). Senescence symptoms and elevated expression of the senescence marker genes SAG12, SGN1, and ACD2 were noted in old leaves of the purine pathway mutants grown with low nitrate (Fig. 1; Supplemental Fig. S1). These leaves also exhibited significantly higher expression of GLN1.1, GDH1, and GDH2 transcripts as compared with the wild type (Supplemental Fig. S6). Enhanced expression of the latter transcripts was shown previously to be associated with the protein degradation necessary to remobilize N from senescent leaves in tobacco (Nicotiana tabacum; Masclaux et al., 2000; Pageau et al., 2006). Considering the absence of all described senescence symptoms in old leaves of nitrate-starved wild-type plants, one must conclude that the impairment in purine catabolism forced the mutant plants to compensate for N shortage by activating chloroplast protein degradation in the old leaves (Matile et al., 1996; Suzuki and Shioi, 1999; Pruzinská et al., 2005). Hence, an active purine catabolic pathway is necessary to prevent premature senescence.
Allantoin, a Degraded N-Enriched Purine Metabolite, Is Remobilized from Old to Young Growing Leaves
A significantly lower nucleic acid level, expressed as total RNA, was detected in old leaves of the nitrate-starved wild type, whereas no senescence symptoms were noticed (Fig. 2). These results support remobilization of the degraded nucleic base purine metabolites, such as ureides, from old to young leaves. This notion was additionally supported by allantoin infiltration into the old leaves of the wild type and the Atxdh1 mutant, where a significant enhancement of allantoin levels was evident not only in the treated leaves but in the younger leaves of Atxdh1, as well as the wild type, as compared with control plants treated with solution without allantoin (Fig. 8). These results indicate that allantoin likely originated from the older leaves, yet the steps of allantoate degradation to glyoxylate and ammonia occurred in all leaves (Fig. 8), playing a role as an N source for the younger leaves. This scenario is further supported by the significantly enhanced expression of AtUPS1, AtUPS2, and AtUPS5 transporters in old wild-type leaves of nitrate-starved plants compared with plants supplemented with high-nitrate or Atxdh1 plants (Fig. 3). The up-regulation of the transporters indicated their participation in the remobilization of organic N forms from the senescing tissues (Kojima et al., 2007) and may indicate that ureides from old leaves supply young leaves in nitrate-starved plants. Furthermore, the enhanced transcript and protein levels of the upstream and downstream purine catabolism genes were detected preferentially in old leaves of nitrate-starved plants (Figs. 5 and 6; Supplemental Fig. S4). Such an orchestration of purine catabolism should normally result in the full degradation of purine metabolites, and indeed, xanthine and allantoin did not accumulate in wild-type leaves (Fig. 4), whereas they accumulated in their related mutants. Higher accumulation of xanthine was evident in the old as compared with the young leaves of Atxdh1, and significantly higher allantoin was noticed in the young leaves of Ataln. This indicates that xanthine is mostly degraded in the old leaves, and the majority of the generated allantoin is remobilized to the young leaves (Fig. 4). These results support ureide remobilization from old leaves to young growing leaves.
Premature Senescence in the Older Leaves of Low-N Atxdh1 Plants
In maize (Zea mays), N remobilization begins earlier and with an increased rate when plants are grown with low as compared with high N supply (Ta and Weiland, 1992; Uhart and Andrade, 1995). Emerging and growing organs, such as young leaves, are a potential sink to trigger N remobilization from older plant parts, which includes N remobilization from leaf to leaf during the vegetative phase (Wendler et al., 1995; Masclaux-Daubresse et al., 2008). Proteins in the mature leaves are potential N storage to be degraded and remobilized to the young growing leaves (Hensel et al., 1993; Masclaux et al., 2000; Hörtensteiner and Feller, 2002; Fischer, 2007; Liu et al., 2008). Among these are chloroplastic proteins, such as Rubisco, which represents 50% of the total proteins in mature leaves of C3 plants (Staswick, 1997). Indeed, the chloroplast-located Rubisco large subunit and D1 protein, a component of the reaction center of PSII (Keren et al., 1997), were decreased in old leaves of the N-limited Atxdh1 mutant, whereas the level of the autophagy-related proteins ATG5 and ATG8A, which are essential for N and carbon remobilization (Thompson et al., 2005; Phillips et al., 2008; Honig et al., 2012), were increased (Supplemental Fig. S7). These data indicate an enhanced remobilization of degraded protein and are consistent with the observed 33% or more decrease in the soluble protein in old mutant leaves compared with old leaves of nitrate-starved wild-type plants (Fig. 2E).
Importantly, no difference in chlorophyll content was evident, and no yellowing of leaves was observed in the old leaves of the low-nitrate-grown Atxdh1 plants at 18 d (Supplemental Fig. S8, A and B). Furthermore, the mutation in AtXDH1, or the level of the supplemented nitrate, had no effect on the level of the soluble proteins in the old leaves at this growth stage (Supplemental Fig. S8C). Taking into consideration that the level of the soluble proteins decreased in the older leaves and increased in the younger leaves of the 25-d-old plants as compared with the 18-d-old plants (compare Fig. 2E with Supplemental Fig. S8C), our results indicate a higher soluble protein degradation and remobilization rate from the older to the younger leaves of the nitrate-starved Atxdh1 as compared with wild-type plants. The results further indicate that the higher degradation rate of 0.13 mg protein g−1 fresh weight in Atxdh1 as compared with the wild type (Fig. 2E; Supplemental Table S1) among the proteins Rubisco and D1 (Supplemental Fig. S7, A and B) is the cause of the senescence symptoms in the older leaves of the nitrate-starved Atxdh1. In contrast, the N supplied via the degradation of ureides in the nitrate-starved wild-type old leaves prevented similar levels of protein degradation and the senescence symptoms.
The estimation of total RNA level does not represent the whole pool of purine-degraded compounds. Thus, to evaluate whether the N remobilized from the degraded ureides in nitrate-starved wild-type old leaves is sufficient to prevent the degradation of 0.13 mg protein g−1 fresh weight, the level of the accumulated xanthine and allantoin in the old leaves of the nitrate-starved Atxdh1 and Ataln plants was employed as an estimate for purine catabolism rate. Considering that the N-to-protein conversion factor ranged in plant leaves from 6.25 to 4.43 (Yeoh and Wee, 1994), the decrease of 0.13 mg protein g−1 fresh weight in mutant old leaves indicated a higher remobilization of 20.8 to 29.3 µg of N from the low-nitrate-grown mutant old leaves originating from degraded soluble protein (Fig. 2E). Essentially, the 0.37 or 0.78 µmol g−1 fresh weight xanthine or allantoin, accumulated in the old leaves of the nitrate-starved Atxdh1 and Ataln mutants, respectively (Fig. 4), contains 20.7 or 43.6 µg N g−1 fresh weight (see calculation in Supplemental Table S1).
In old leaves, the estimated N level from purine catabolism (20.7 or 43.7 μg N g−1 fresh weight) is similar to or higher than the N generated by protein degradation in low-nitrate-grown Atxdh1 and was similar to the wild-type level (20.8–29.3 μg N g−1 fresh weight). These results indicate that purine catabolism could prevent protein degradation at a level similar to that in the Atxdh1 mutant and that this is the reason for the absence of senescence symptoms in the low-nitrate-grown wild-type old leaves.
The Level of the Applied Nitrate Negatively Regulates Purine Degradation in Arabidopsis Leaves
The application of nitrate was shown to down-regulate AtALN and AtAAH expression in two N-starved Arabidopsis plants (Werner et al., 2008) and inhibit nodule formation as well as the fixation of atmospheric N2 in legume plants (Murray et al., 2017, and refs. therein), indicating that nitrate supply negatively affects purine metabolite use in legumes and nonlegume plants. This was clearly demonstrated by the lower accumulated xanthine and allantoin in leaves of 5 mm nitrate-supplemented Atxdh1 and Ataln than in the 1 mm nitrate-fed mutants (Fig. 4). Nitrate supplementation negatively regulated purine catabolism at both the transcript (Figs. 3, 5, and 7) and protein (Fig. 6) expression levels. However, another level of posttranslational modification was shown before in ryegrass (Lolium multiflorum). In that case, the lower XDH activity as well as the lower allantoin and allantoate levels in leaves of nitrate-supplied annual ryegrass as compared with ammonium-supplied plants (Sagi et al., 1998) were attributed to the preferred allocation of the molybdenum cofactor (Moco). The catalytic center of NR and XDH1, plant molybdoenzymes, requires Moco. The preferential allocation of Moco to NR supports nitrate assimilation in the presence of high nitrate over AtXDH1 activity (Sagi et al., 1997, 1998; Sagi and Lips, 1998). While not examined here directly, the enhanced AtXDH1 activity and decreased NR activity in old leaves and vice versa in young leaves, as well as the enhanced AtXDH1 activity and decreased NR activity in nitrate-starved plants and vice versa in leaves of high-nitrate-supplied plants, supported this notion (Figs. 6 and 7).
High-nitrate application almost fully abrogated the senescence symptoms evident in the old leaves of low-nitrate-supplied mutants by enhancing the organic N level and soluble protein content in these leaves (Figs. 1 and 2; Supplemental Fig. S1). This was attributed to NR activity that was significantly higher in mutant than in wild-type plants in leaves of 18-d-old plants (Fig. 7) before the appearance of senescence symptoms (Supplemental Fig. S8). The lower nitrate (Figs. 2 and 7B) and higher NR activities in Atxdh1 relative to the wild type when both types of plants were supplemented with high nitrate (Fig. 7) suggest that the shortage of purine N in the Atxdh1 mutant is compensated by higher nitrate assimilation under these conditions.
The Role of Ureides as N Storage Sources Rather Than as Stress Protectants
The ureides allantoin and allantoate accumulate in wild-type leaves exposed to periods of extended dark, protecting the leaves from the dark-induced oxidative stress by scavenging reactive oxygen species. The exposure of Atxdh1, lacking allantoin and allantoate, to the same stress conditions results in severe senescence symptoms in the leaves (Brychkova et al., 2008). The protective role of allantoin was further demonstrated recently using Ataln mutants that accumulate high levels of allantoin and exhibit higher tolerance to drought and osmotic stress when compared with wild-type and Atxdh1 plants (Watanabe et al., 2014). Under both stresses, the extended-dark-grown (Brychkova et al., 2008) and low-nitrate-grown Atxdh1 exhibited senescence symptoms, whereas the wild type did not.
In wild-type plants exposed to extended dark periods, allantoin and allantoate levels are increased through the down-regulation of AtALN and AtAAH (Brychkova et al., 2008), whereas under the low-nitrate treatment described here, the accumulation of allantoin in wild-type old leaves was absent (Fig. 4) due to enhanced purine catabolism, demonstrated at both protein and transcript levels (Figs. 5 and 6; Supplemental Fig. S4). The lack of allantoin accumulation during low-nitrate supply suggests either a lower level of stress compared with dark exposure or the lack of a role as a stress protectant.
Significantly, the three independent mutants impaired in XDH1, ALN, and AAH gene transcripts exhibited similar senescence symptoms in their older leaves in response to low nitrate supply, whereas the supply of sufficient nitrate abrogated their senescence symptoms (Fig. 1; Supplemental Fig. S1). Yet, among the three mutants, Atxdh1 did not accumulate the antioxidants allantoin and allantoate (Brychkova et al., 2008), whereas Ataln accumulated allantoin and Ataah accumulated both allantoin and allantoate (Fig. 4; Watanabe et al., 2014). Under low nitrate, the mutants accumulating the antioxidants allantoate and/or allantoin (Ataah and Ataln) shared identical phenotypes (senescence in the older leaves), with the low-nitrate-grown Atxdh1 mutant lacking the ureides allantoin and allantoate. Together, these results for the mutants indicate that any positive effect is not from ureide accumulation acting as a stress protectant but due to its function as an N storage form.
CONCLUSION
The following data support a role for ureides as an endogenous N source in Arabidopsis plants grown under limited nitrate supply. (1) The three independent mutant plants impaired in the purine catabolism pathway all showed premature senescence symptoms in the old leaves when grown under insufficient nitrate supply, whereas the wild type did not show signs of senescence. Supplementation of high nitrate prevented the N-dependent senescence symptoms in the three mutants, indicating the essentiality of normal purine catabolism for N metabolism (Fig. 1; Supplemental Fig. S1). (2) The significant decrease in organic N and soluble proteins in the old Atxdh1 leaves (see the scheme in Fig. 9A, bottom left) as compared with wild-type old leaves (Fig. 9A, bottom right) in the low-nitrate-fed plants suggests significant N degradation and remobilization from mutant old leaf proteins (Fig. 2). (3) The soluble protein level decrease in the older leaves and increase in the younger leaves of the 25-d-old plants, as compared with the 18-d-old plants (compare Fig. 2E and Supplemental Fig. S8C), indicate a higher soluble protein degradation rate and remobilization from the older to the younger leaves in the nitrate-starved Atxdh1 than in wild-type plants (see schemes in Fig. 9A). The rate of purine catabolism and remobilization of the resulting N from older to younger growing wild-type leaves (summarized in Supplemental Table S1) shows that the absence of the purine-degraded N remobilized from the older leaves is the cause of the senescence symptoms, a result of a higher soluble protein degradation rate in older leaves of nitrate-starved 25-d-old Atxdh1 plants (Fig. 9A, left). (4) The higher degradation rate of the chloroplastic proteins (see scheme in Fig. 9A, bottom left), such as Rubisco large subunit and D1, and the enhanced remobilization of their degradation product, indicated by the increased ATG8A and ATG5, cause premature senescence symptoms in the older leaves of nitrate-starved Atxdh1 plants (Supplemental Fig. S7). (5) The lower nitrate (Figs. 2 and 7B) and higher NR activity rates in Atxdh1 compared with wild-type leaves under high nitrate (Fig. 7) suggest a possible shortage of ureide-originated N to be remobilized from older to younger Atxdh1 growing leaves under high nitrate availability. This is compensated by the higher nitrate assimilation rate (see schemes in Fig. 9A and B, left), resulting in enhanced organic N and soluble protein that delays the senescence symptoms (Figs. 1 and 2; Supplemental Fig. S1). (6) The higher purine catabolism rate evident in nitrate-starved wild-type old leaves as compared with high-nitrate-fed wild-type plants (compare Fig. 9, A and B, right) indicates the higher purine degradation and remobilization activity rates in the nitrate-starved wild-type old leaves and higher levels of allantoin transport from the old to younger leaves. These are indicated by the following features: (i) significantly higher up-regulation of purine degradation transcripts, functioning upstream and downstream of ureide generation; (ii) enhanced AtXDH1 and AtAAH expression, as representative of upstream and downstream proteins, respectively (Figs. 5 and 6); (iii) significantly up-regulated transcript levels of the ureide transporters UPS1, UPS2, and UPS5 (Fig. 3); (iv) higher xanthine accumulation rate in nitrate-starved Atxdh1 old leaves compared with high-nitrate-fed old leaves (Fig. 9, A and B, bottom left) and the accumulation of higher allantoin levels in nitrate-starved Ataln young and old leaves (Fig. 4); and (v) significant transport of allantoin infiltrated to the old leaves of nitrate-starved wild-type and Atxdh1 plants to the younger leaves, as well as its degradation to allantoate and glyoxylate, resulting in the release of four NH3 molecules for each molecule of allantoin (Fig. 8), thus supporting a role for ureides as an N source in Arabidopsis plants grown under limited nitrate.
Scheme of N mobilization in Atxdh1 and wild-type (WT) plants in low- and high-N conditions. The destiny of nitrate as well as the purine catabolism and protein degradation products in low-nitrate-fed (A) and high-nitrate-fed (B) wild-type and mutant plants is illustrated. The mutation in Atxdh1 blocks purine catabolism, resulting in the accumulation of xanthine in the oldest leaves (bottom row), being lower in the oldest leaves of the high-nitrate-fed than the low-nitrate-fed Atxdh1 plants. The low nitrate level in the Atxdh1 mutant and the lack of the purine catabolism-degraded N source for the growth of the young leaves (top row) results in the degradation of chloroplastic protein to compensate for the small N pool. At low supplemented nitrate in the wild type, purine-degraded enzymes and UPS transporters are major players in N remobilization to sink tissues. NR is less active, but most importantly, a high chloroplastic protein degradation rate is not activated; thus, no senescence symptoms are evident. The Atxdh1 mutant supplemented with high nitrate exhibits a high nitrate assimilation rate by enhanced NR activity (up-pointing black arrowheads), sufficient for the N pool in the oldest and youngest leaves, and the chloroplastic protein degradation in the oldest leaves is prevented. Under high-nitrate application in wild-type plants, nitrate assimilated by NR provides the major N source, down-regulating purine catabolism (down-pointing black arrowheads of purine catabolic enzymes as well as UPSs), and protein degradation (thin arrows) and the N source from the N pool are remobilized to the youngest leaves’ N pool.
MATERIALS AND METHODS
Plant Material Growth Conditions
Arabidopsis (Arabidopsis thaliana) wild-type and mutant plants used in this study were of the Columbia-0 background. The following homozygous T-DNA-inserted mutants were employed: Atxdh1 (GABI_049D04, SALK_148366; accession no. At4g34890) described previously by us (Yesbergenova et al., 2005; Brychkova et al., 2008); Ataln (SALK_ 013427, SALK_146783; accession no. At4g04955) as shown before (Todd and Polacco, 2006; Watanabe et al., 2014); and Ataah (SALK_112631; accession no. At4g20070) as described previously (Todd and Polacco, 2006).
Seeds were surface sterilized in 80 (v/v) alcohol for 2 min, washed three times in sterile water, and sown on one-half-strength Murashige and Skoog agar plates (Murashige and Skoog, 1962). The plates were placed at 4°C for 3 d to synchronize germination and then were transferred to a controlled growth room at 22°C, 14/10-h light/dark photoperiod, and light intensity of 150 µE m−2 s −1. Six-day-old seedlings were transferred each to a 0.128-L pot containing a 1:1 mixture of perlite and nutrient-free soil (Sun Gro Horticulture Canada). Plants were irrigated twice per week with 0.5× Hoagland solution (Hoagland and Arnon, 1950) modified to contain 1 or 5 mm NaNO3 as the only N source, where the sodium level was balanced to contain 5 mm sodium in all the treatments by the supplementation of NaCl. Salinization was avoided by irrigation performed to leach out 50% of the irrigated nutrient solution. The leaves of plants at 18 or 25 d after germination (the latter just before bolting) were harvested, snap frozen in liquid N, and stored at −80°C for further use. The first four rosette leaves from the bottom were designated as old leaves, and the uppermost four leaves from the top were designated as young leaves.
Determination of Chlorophyll and Anthocyanin
For chlorophyll determination, four leaf discs were sampled from old and young rosette leaves of wild-type, Atxdh1, Ataln, and Ataah plants grown under the low- and high-nitrate conditions. The leaf discs (7 mm diameter) were immersed in 90% ethanol and incubated at 4°C for 2 d in the dark. Absorbance of the extracted chlorophyll was measured at 652 nm, and total chlorophyll was estimated (Ritchie, 2006). To assess the responses to external xanthine and allantoin, 7-mm leaf discs were sampled from rosette leaves of wild-type and Atxdh1 mutant plants and put in petri dishes, on filter paper soaked with a solution containing water (mock) and 1 mm xanthine or allantoin, for 2 d in permanent light. Thereafter, the discs were washed and the total anthocyanin was measured as described by Laby et al. (2000). In addition, the green area of the leaf disc was estimated by employing Digimizer 3.2.1.0 (http://www.digimizer.com), presented as the ratio of the green part to the total area of the leaf disc, as the chlorophyll damage indicator.
Metabolite Analysis
Samples (100 mg) were grounded in 25 mm K2PO4:KH2PO4 buffer (1:4, w/v), pH 7.5, using a chilled mortar and pestle (Brychkova et al., 2008; Lescano et al., 2016). The resulting homogenates were transferred to 1.5-mL microcentrifuge tubes, centrifuged at 15,000g for 20 min at 4°C, and the supernatant was used for analyses. Quantification of the ureides allantoin and allantoate was performed using the differential conversion of ureide compounds to glyoxylate and colorimetric detection of ureides and glyoxylate described by Vogels and Van der Drift (1970) and employed by others (Todd et al., 2006; Brychkova et al., 2008; Werner et al., 2008, 2013; Watanabe et al., 2014; Lescano et al., 2016; Takagi et al., 2016).
To follow the destiny of allantoin and its transport in plants, allantoin solution (5 mm) or water (control) was infiltrated, employing a needleless syringe (see sulfite infiltration by injection [Brychkova et al., 2012]) into the old leaves (the first four leaves from the bottom) of 18-d-old 1 mm nitrate-treated Atxdh1 mutants and wild-type plants. Then, after 30 min and 3 h of allantoin application, the allantoin and the degraded products allantoate, glyoxylate, and ammonium were measured in the oldest (leaves 1–4 from the bottom), middle (leaves 5–8), and youngest (leaves 9–12) leaves. The data are presented as net metabolite levels, which are calculated by subtracting the internal levels of the respective metabolites measured from the water-infiltrated plant.
Xanthine was detected using the xanthine oxidase assay as described previously (Brychkova et al., 2008). Ammonium was detected by the Nessler method (Molins-Legua et al., 2006). Nitrate content was analyzed according to Cataldo et al. (1975). Total N in the dried tissues was measured by an elemental analyzer (Thermo Scientific FLASH 2000 CHNS/O Analyzers).
Protein Extraction and Fractionation
Total protein from Arabidopsis rosette leaves was extracted as described before (Sagi et al., 1998; Kurmanbayeva et al., 2017). Concentrations of total soluble protein in the resulting supernatant were determined according to Bradford (1976). Native-SDS-PAGE was carried out as described previously (Sagi and Fluhr, 2001; Srivastava et al., 2017). Samples containing the extracted proteins were incubated on ice for 30 min in sample buffer containing 47 mm Tris-HCl (pH 7.5), 2% (w/v) SDS, 7.5% (v/v) glycerol, 40 mm DTT as the thiol-reducing agent, and 0.002% (w/v) Bromophenol Blue. The incubated samples were centrifuged at 15,000g for 3 min before loading the supernatant and resolved subsequently on a 7.5% (w/v) SDS-polyacrylamide separating gel and 4% (w/v) stacking gels. Native-SDS-PAGE was carried out using 1.5-mm-thick slabs loaded with 50 µg of old leaf or 100 µg of young leaf proteins unless mentioned otherwise.
XDH In-Gel Activity and Nitrate Reductase Kinetic Activity
Regeneration of the active proteins after denaturing PAGE was carried out by removal of the SDS by shaking the gel for 1 h in 10 mm Tris-HCl buffer (pH 7.8) solution (65 mL of buffer per mL of gel) containing 2 mm EDTA and 1% (w/v) Triton X-100 (Sagi and Fluhr, 2001; Srivastava et al., 2017). Following the regeneration process, the gels were assayed for normal in-gel XDH activities using 0.1 mm PMS, 1 mm 3(4,5-dimethylthiazolyl-2)2,5-diphenyltetrazolium bromide, and the addition of 0.5 mm xanthine mixed with 1 mm hypoxanthine in 0.1 mm Tris-HCl buffer (pH 8.5) at 25°C under dark conditions. To detect the superoxide generation activity of XDH, PMS was omitted and the mix of xanthine with hypoxanthine as a specific substrate was employed (Yesbergenova et al., 2005). The quantity of the resulting formazan was directly proportional to enzyme activity during a given incubation time in the presence of excess substrate and tetrazolium salt (Rothe, 1974; Srivastava et al., 2017).
For NR activity, the samples were extracted in a buffer containing 3 mm EDTA, 3.6 mm DTT, 0.25 m Tris-HCl (pH 8.48), 3 mg l-Cys, 3 mm NaMoO4, and protease inhibitors, including aprotenin (10 μg mL−1) and pepstatin (10 μg mL−1), and the activity was detected as described previously (Sagi et al., 1997).
Immunoblotting
Protein crude extract samples (20–50 μg) extracted as described by Sagi et al. (1998) were subjected to Native-PAGE to detect AAH and SDS-PAGE for the other proteins. The fractionated proteins were transferred onto polyvinylidene difluoride membranes (Immun-Blot membranes; Bio-Rad). The membrane was probed first with the following primary antibodies: anti-AAH (a gift from Claus-Peter Witte) at a 1:500 dilution ratio, antibody specific to D1 protein (a component of the reaction center of PSII; Agrisera) at a 1:10,000 ratio, specific antibodies to ATG8A (Abcam) at a 1:1,000 dilution ratio and ATG5 (Agrisera) at a 1:1,000 dilution ratio, and an antibody recognizing large subunits of Rubisco (a gift from Michal Shapira) at a dilution ratio of 1:3,000. Thereafter, the proteins underwent binding with secondary antibodies diluted 5,000-fold in phosphate-buffered saline (anti-rabbit IgG; Sigma-Aldrich). Protein bands were visualized by staining with Clarity Western ECL Substrate (Bio-Rad) and quantified by ImageLab (version 5.2; Bio-Rad).
RT-qPCR
Total RNA was extracted from plants using the Aurum Total RNA Kit according to the manufacturer’s instructions (Bio-Rad). First-strand cDNA was synthesized in a 10-µL volume containing 350 ng of plant total RNA that was reverse transcribed employing an iScript cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions. The generated cDNA was diluted 10 times, and quantitative analysis of transcripts was performed employing the set of primers presented in Supplemental Table S2 designed to overlap exon junctions. Gene expression was normalized to POLYUBIQUITIN10 (At4g05320) and EF-1α (At5g60390) as housekeeping genes, employed as described previously (Brychkova et al., 2008).
Statistical Analysis
All results are presented as means and se. The data for total N, total organic N, nitrate, and ammonium represent means obtained from at least three independent biological experiments. Metabolite, protein content, and transcript measurements represent means obtained through at least three independent biological experiments. Each treatment was evaluated using ANOVA (JMP 8.0 software). Comparisons among three or more groups were made using one-way ANOVA with Tukey’s multiple comparison tests.
Accession Numbers
Sequence data for this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: At4g34890 (XDH1), At4g04955 (ALN), and At4g20070 (AAH).
Supplemental Data
The following supplemental materials are available.
Supplemental Figure S1. Effects of nitrate level supplemented to the growth medium of wild-type and mutant plants impaired in the purine catabolic pathway.
Supplemental Figure S2. Effects of exogenous application of xanthine and allantoin on leaf disc appearance and anthocyanin content.
Supplemental Figure S3. Xanthine content in old and young leaves of 18-d-old Atxdh1, Ataln, and Ataah mutant and wild-type plants grown in N-deficient soil supplemented with 1 mm NaNO3.
Supplemental Figure S4. Expression levels of purine catabolism transcripts in young and old leaves of wild-type and purine catabolism-impaired plants supplemented with 1 or 5 mm nitrate as the only N source.
Supplemental Figure S5. Total leaf fresh weight in 25-d-old wild-type plants grown in N-deficient soil supplemented with 1 or 5 mm NaNO3.
Supplemental Figure S6. Relative expression of N assimilation senescence-related transcripts in old and young leaves of wild-type and Atxdh1 plants.
Supplemental Figure S7. Immunoblot analysis of Rubisco large subunits, a component of the PSII reaction center, and autophagy proteins ATG8A and ATG5.
Supplemental Figure S8. Leaf appearance, total chlorophyll content, and soluble protein content of the old and young leaves of 18-d-old wild-type and Atxdh1 mutant plants.
Supplemental Table S1. Calculation of N content in xanthine and allantoin accumulated in the old leaves of 1 mm nitrate-grown Atxdh1 and Ataln mutants, respectively, as the estimate of the endogenous N source used by the wild type to prevent senescence symptoms
Supplemental Table S2. Gene-specific primer sequences used for expression analyses.
Footnotes
↵1 This research was supported by the Israel Center of Research Excellence (ICORE) ‘Plant Adaptation’ (ISF Grant no. 757/12). Dr. Sudhakar Srivastava is the recipient of a postdoctoral fellowship from the Jacob Blaustein Center for Scientific Cooperation. M.S. and R.F. thankfully acknowledge a grant from the Israel Science Foundation (Grant no. 417/03) in partial coverage of the costs.
↵[OPEN] Articles can be viewed without a subscription.
- Received June 29, 2018.
- Accepted August 25, 2018.
- Published September 6, 2018.
REFERENCES
