Leaf Senescence and Nitrogen Loss

The nitrogen content of a leaf reaches the maximum level at around the completion of leaf expansion. When mesophyll cells cease their enlargement, increase of chloroplast number and size also stops (Pyke and Leech, 1987; Baumgartner et al., 1989; Sodmergen et al., 1991). In mature leaves of C3 plants, chloroplast nitrogen accounts for 70-80% of leaf

Plant Cell Death Processes

Copyright 2004, Elsevier, Inc. All rights reserved nitrogen (Morita, 1980; Makino and Osmond, 1991). In the course of leaf senescence, nitrogen (protein) content in the leaf gradually decreases with a concomitant loss of the photosynthetic activity (Wittenbach, 1979; Friedrich and Huffaker, 1980; Peoples et al., 1980; Makino et al, 1983; Crafts-Brandner et al., 1984; Ford and Shibles, 1988). Total apparent loss of chloroplast nitrogen during leaf senescence was estimated to account for 90% of the total nitrogen loss from the leaf in rice (Morita, 1980).

A. Fate of Chloroplasts and their Constituents

Two major hypotheses to explain the decline in chloroplast nitrogen or photosynthetic capability during leaf senescence have been advanced (Huffaker, 1990). It has been proposed that the decline in photosynthesis is mainly attributed to a decrease in the number of chloroplasts in the senescing leaf. Wittenbach et al. (1982), for example, reported a simultaneous decline in photosynthesis, chlorophyll, soluble protein, and total number of chloroplasts in senescing wheat leaves. They also presented ultrastructural evidence indicating that chloro-plasts might be associated with invagination of the vacuole or may move into the vacuole during senescence. The other hypothesis is that the decline is largely attributed to events occurring within chloroplasts that did not change in number until the late stage of leaf senescence. The hypothesis fitted the ultrastructural studies by Thomson and Platt-Aloia (1987) and Inada et al. (1998a,b, 1999), indicating that membrane integrity and subcellular compartmentation of chloroplasts, mitochondria, peroxisomes, and vacuoles in mesophyll cells are maintained until the last stage of senescence. The latter hypothesis might be further supported by observations of non-parallel decline in chloroplast components during senescence. Preferential decline in cytochrome f content (Gepstein, 1988) and preferential retention of LHCII content (Hidema et al., 1991, 1992; Mae et al., 1993) were observed in leaves senesced under relatively low irradiances. The enzymes involved in the photosynthetic carbon reduction cycle did not lose activities or proteins in a synchronized manner during senescence (Makino et al., 1983; Crafts-Brandner et al., 1996, 1998). Stay-green mutants of Lolium temulentum exhibited preferential retention of Chl, P700, and D1 contents until the late stage of senescence, while Rubisco was normally lost during senescence (Thomas et al., 1999). Ford and Shibles (1988) reported that senescence is a two-stage process: first a decline in chloroplast function accompanied by a loss of chloroplast content, followed by a brief terminal phase when whole chloroplasts are lost as well.

III. Protein Metabolism during Leaf Senescence

The amount of a protein in a leaf is a result of its synthesis and degradation during growth. In order to understand protein metabolism in senescing leaves we need to know both synthesis and degradation of the protein throughout the life span of the leaf. Particular attention has been paid to ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), light-harvesting chlorophyll a/b protein complex (LHCII), and D1 protein of photosystem II because of their important roles in photosynthesis and nitrogen economy in plants.

A. Fates of DNA and RNA

In mesophyll cells, there are three sites for protein synthesis: nucleus/cytoplasm, mitochondria, and chloroplasts. Each site has its own DNA and apparatus necessary for RNA

replication and protein synthesis. Chloroplasts contain a large portion of leaf protein, but not all the proteins in chloroplasts are synthesized there. A number of chloroplast proteins are encoded by nuclear DNA, synthesized as precursors in the cytoplasm, imported into the chloroplast, processed, and assembled there. It was shown that the content of nuclear DNA or mitochondrial DNA was unchanged until the last stage of leaf senescence, but the content of chloroplast DNA decreased considerably during senescence in barley (Baumgartner etal., 1989) and rice (Sodmergen etal., 1989, 1991; Inada etal, 1998b, 1999). These data indicate that template availability might be one of the limiting factors for protein synthesis in senescing chloroplasts. Nuclear condensation and DNA fragmentation, which are indices of programmed cell death (Pennell and Lamb, 1997), were observed at the last stage of leaf senescence (Yen and Yang, 1998). Disorganization of the nucleus and degeneration of tono-plast were found at the final stage of senescence followed by complete loss of cytoplasmic components (Thomson and Platt-Aloia, 1987; Inada et al., 1998a,b, 1999).

Cellular RNA is present as ribosomal (rRNA), transfer (tRNA), and messenger RNA (mRNA) components. Ribosomal RNA accounts for 85 to 90% of the total RNA (Brady, 1988). Total RNA or ribosomal RNA content in the leaf gradually decreases after full leaf expansion (Makrides and Goldthwaite, 1981; Brady, 1988; Hensel et al., 1993; Lohman etal., 1994; Crafts-Brandner etal., 1996). For instance, a yellow leaf has about 10-fold less RNA than a green leaf (Bate et al., 1991). Protein synthesis might, therefore, be limited in part by the decreased capacity for protein synthesis in senescing leaves.

B. Synthesis and Degradation of Rubisco

Rubisco is a bifunctional enzyme that catalyzes the first reaction of photosynthetic CO2 fixation and photorespiration. Rubisco is a stromal enzyme of chloroplasts and the most abundant protein in leaves. It accounts for 12-35% of total leaf nitrogen in mature leaves of C3 plants (Evans and Seeman, 1989). Rubisco content in a leaf is a limiting factor for the rate of CO2 assimilation under light-saturated and ambient air conditions (Makino et al., 1985). Therefore, its fate during leaf senescence directly affects the photosynthesis and nitrogen metabolism of the plant.

Rubisco content in a leaf increases during leaf expansion and reaches the maximum amount at around the time of full expansion. Thereafter, it continuously decreases during leaf senescence and reaches a non-measurable level when the leaf is completely senesced (Hall et al, 1978; Wittenbach, 1979; Peoples et al, 1980; Wittenbach et al, 1982; Mae et al., 1983; Ford and Shibles, 1988; Hensel et al, 1993; Crafts-Brandner et al, 1996, 1998). Total apparent loss of Rubisco nitrogen from the leaf blade of rice plants was about 40% of the total nitrogen lost from the leaf during senescence (Makino et al., 1984).

Decline of Rubisco content in senescing leaves can be attributed to the decreased rate of its synthesis, or to the increased rate of its degradation, or to both. In barley (Peterson et al., 1973), Perilla, and Capsicum (Brady, 1988) little or no incorporation of precursors into Rubisco protein could be measured in expanded leaves, so that little or no replacement of the protein occurs. On the other hand, in wheat (Brady, 1988) Rubisco was found to be synthesized in newly expanded leaves, and this was also the case in Zea mays leaves (Simpson etal., 1981). Mae etal. (1983) andMakino etal. (1984) quantitatively examined the synthesis and degradation of Rubisco protein throughout the life span of a rice leaf by using 15N as a tracer (Fig. 10-1). Synthesis of Rubisco peaked during leaf expansion and about 90% of Rubisco synthesized during the life span of the leaf had been formed at about a week after the completion of expansion. After the completion of leaf expansion,

Rice Leaf Senescence

Figure 10-1. Changes in the influx and efflux of nitrogen and changes in the amount of Rubisco synthesized or degraded in the 12th leaf blades on the main stem of rice from leaf emergence through senescence. Plants were grown with the same amount of nitrogen until the emergence of the 12th leaf blades, then on 0.3 mM (15NH4)2SO4 for 3 days and then 0 [-N], 1 mM [C], or 2 mM [S] NH4NO3. The influx and efflux, and the amounts of Rubisco synthesized and degraded were calculated from the changes in 15N content as described by Mae et al. (1983). The solid curve shows changes in the content of the nitrogen or the Rubisco (from Makino et al., 1984).

Figure 10-1. Changes in the influx and efflux of nitrogen and changes in the amount of Rubisco synthesized or degraded in the 12th leaf blades on the main stem of rice from leaf emergence through senescence. Plants were grown with the same amount of nitrogen until the emergence of the 12th leaf blades, then on 0.3 mM (15NH4)2SO4 for 3 days and then 0 [-N], 1 mM [C], or 2 mM [S] NH4NO3. The influx and efflux, and the amounts of Rubisco synthesized and degraded were calculated from the changes in 15N content as described by Mae et al. (1983). The solid curve shows changes in the content of the nitrogen or the Rubisco (from Makino et al., 1984).

Rubisco synthesis sharply decreased. The rate of synthesis was almost proportional to the rate of nitrogen influx into the leaf throughout the leaf's life span, indicating that the availability of the substrate might be involved as one of the factors determining the rate of Rubisco synthesis. The synthesis of Rubisco during senescence was constantly about 3% of the Rubisco amount in the leaf.

Degradation of Rubisco had already started at the time of full leaf expansion and was most active in the early stage of leaf senescence. The degradation rate was much faster than the rate of synthesis throughout the period of senescence.

C. Molecular Basis for Changes in Rubisco Synthesis during Senescence

The holoenzyme of Rubisco in higher plants consists of eight large and eight small subunits. The large subunit (rbcL; Mr ~ 54,000) is coded by chloroplast DNA and translated on chloroplast ribosomes. The small subunit (rbcS; Mr « 13,000) is coded by nuclear DNA, translated on cytoplasmic ribosomes as a precursor molecule, and processed to its mature size within the chloroplasts. The in vivo observed decline in synthesis of Rubisco was shown to correlate with a declining population of mRNAs for each subunit in the leaves of wheat, placing the point of control at either transcription or at mRNA stability (Brady, 1988). In leaves of soybean (Jiang et al., 1993), Arabidopsis (Hensel et al., 1993) and Phaseolus vulgaris (Crafts-Brandner et al., 1996), the decreases in the level of Rubisco protein were correlated with the decreases in the levels of mRNAs for both subunits and in the level of total RNA during senescence. Thus, there is a close correlation between their transcription rate and mRNA abundance, an indication that transcription plays a primary role in determination of mRNA abundance (Rapp etal., 1992). However, analysis of mRNA levels and transcription rates also provided evidence that plastid mRNA stability might be an important factor in determining the level of RNA (Deng and Gruissem, 1987; Mullet and Klein, 1987; Rapp et al, 1992). Kim et al. (1993) analyzed the stability of mRNAs encoded by seven different barley chloroplast genes representing five major chloroplast functions during barley chloroplast development. Their analyses revealed that half-lives of mRNAs in barley chloroplasts range from 3hto over 40 h. The stability of mRNA for rbcL increased 2.5-fold during chloroplast development in the dark, and then decreased 2-fold in chloroplasts of light-grown plants.

Transcriptional activity in plastids varies during chloroplast development. Plastid tran-scriptional activity and DNA copy number increased early in chloroplast development and transcriptional activity per template varied up to 5-fold during barley leaf biogenesis (Baumgartner et al., 1989). Importantly, plastid transcriptional activity per cell and number of plastid DNA copies per cell were highest in the middle part of leaves when they had grown to one-third of full size. Both activity and number in the older top part of the same leaf were lower than those in the middle part. Thereafter, the transcriptional activity sharply decreased during leaf expansion and reached one-tenth of maximum in the leaf at full expansion.

Translational regulation of the genes rbcS and rbcL is also involved as one of the factors determining the levels of Rubisco in leaves. When dark-grown cotyledons of amaranth were transferred into the light, synthesis of the LSU and SSU polypeptides of Rubisco was initiated very rapidly, before any increase in the levels of their corresponding mRNAs (Berry et al., 1988), indicating that the expression of rbcS and rbcL genes could be adjusted at the translational level. To address the question of whether the abundance of the SSU of Rubisco protein influences LSU protein metabolism, Rodermel et al. (1988) generated anti-rbcS transgenic tobacco plants that have reduced amounts of rbcS mRNAs and the SSU proteins. The LSU protein was coordinately reduced in mutant plants, but the rbcL mRNA level was normal. After a short-term pulse, there was less labeled LSU protein in the transgenic plants than in wild-type plants (Rodermel et al., 1996), indicating that the LSU accumulation is controlled in the mutants at the translational and/or posttranslational levels.

D. Light-Harvesting Chlorophyll a/b Binding Protein of PSII

LHCII is the most abundant protein of thylakoid, constitutes approximately 33% of the total mass of membrane protein and binds approximately 50% of the total chlorophyll (Chitniss and Thornber, 1988). In addition to its central role in the light-harvesting process, LHCII is also a key component for regulating and distributing the excitation energy in response to short-term and long-term fluctuation in the light intensity and quality. The apoproteins of LHCII are encoded in the nucleus by a small multigene family. The polypeptides are translated in the cytoplasm on membrane-free polysomes as soluble precursor proteins, which are posttranslationally imported into the chloroplast. Upon entry into the chloroplast, these apoproteins are processed to their mature forms and subsequently inserted and assembled into the thylakoid membranes as pigment-protein complexes.

During greening of higher plants, the abundance of LHCII is primarily a consequence of transcriptional control of CAB genes (Silverthorne and Tobin, 1984). Once the leaf reaches full size, the abundance of mRNAs for LHCII genes as well as those for the small subunits of Rubisco genes sharply falls (Roberts et al., 1987; Hensel et al., 1993; Lohman et al., 1994). The effects of growth irradiance on the levels of LHCII and other photosynthetic proteins in senescing leaves were examined in rice (Hidema et al., 1991, 1992) and Lolium temulentum (Mae et al., 1993). Rubisco protein levels decreased rapidly in both high-light and low-light leaves. The LHCII protein levels in the high-light senescing leaves decreased soon after the decreases in Rubisco level. On the other hand, LHCII levels in the low-light senescing leaves were retained at nearly as high levels as those at the time of full expansion until the late stage of senescence. However, the amounts of LHCII proteins synthesized in those senescing leaves, judged by 15N incorporation, were less than 1% of the existing amount, indicating that there was little synthesis of this protein in senescing leaves. Thus, suppression of its degradation, rather than the acceleration of its synthesis could maintain the high levels of LHCII in the low-light senescing leaves. Those results, together with others by a pulse-labeling technique with isotopes and a Northern-blotting analysis (Roberts et al., 1987; Bate et al., 1991; Humbeck et al., 1996), strongly suggest that the process of protein degradation is responsible for environmentally-induced changes during senescence. In addition, regulation of Rubisco and LHCII degradation in senescing leaves occurs independently at each protein level, rather than random degradation of the whole population. The higher ratio of light-harvesting capacity to Rubisco in the low-light leaves is likely to be an acclimation phenomenon of the senescing leaves similar to that found in expanding leaves.

E. D1 Protein

The D1 protein coded by chloroplast DNA (psbA) is a component of the reaction center of photosystem II (PSII) and known as a rapid turnover protein that is specifically degraded under illumination (Mattoo et al., 1984). Under normal light conditions, degradation of the D1 protein is counteracted by a repair system that includes synthesis of the protein de novo. By contrast, under strong illumination that causes photoinhibition of PSII, the rate of degradation exceeds that of repair so that the amount of D1 protein decreases (Prasil et al., 1992). D1 protein in chloroplasts is far less abundant than Rubisco protein. However, psbA and rbcL mRNAs accumulated similar levels in young leaves, and in mature leaves psbA mRNAwas more abundant than rbcL mRNA (Ellis, 1981; Deng and Gruissem, 1987). The half-life of psbA mRNA was increased more than 2-fold in mature leaves compared with young leaves. The different stability of mRNAs could account for differences in mRNA accumulation during leaf development (Klaff and Gruissem, 1991). In bean leaves total RNA declined 10-fold, but psbA and rbcL mRNAs remained at a constant proportion of total RNA throughout senescence. By contrast, CAB and rbcS mRNAs comprised a progressively decreasing proportion of total RNA as senescence progressed. When 35S-methionine was fed to leaves at late stages of senescence, 65% and 6% of total thylakoid radiolabeled protein, respectively, was incorporated into D1 and LHCII proteins, indicating preferential synthesis of D1 protein in senescing leaves (Bate et al., 1991).

Was this article helpful?

0 0

Post a comment