Chloroplast ultrastructure regeneration with protection of photosystem II is responsible for the functional ‘stay-green’ trait in wheat

Authors

  • P. G. LUO,

    Corresponding author
    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    2. College of Agriculture and Biotechnology, China Agricultural University, Beijing 100193, China
      P. G. Luo. Fax: +86 28 86290978; e-mail: lpglab@sicau.edu.cn. Z. L. Ren. Fax: +86 28 86290873; e-mail: zhlrenlab@yahoo.com.cn
    Search for more papers by this author
  • K. J. DENG,

    1. School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu 610054, China
    Search for more papers by this author
  • X. Y. HU,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • L. Q. LI,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • X. LI,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • J. B. CHEN,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • H. Y. ZHANG,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • Z. X. TANG,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • Y. ZHANG,

    1. School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu 610054, China
    Search for more papers by this author
  • Q. X. SUN,

    1. College of Agriculture and Biotechnology, China Agricultural University, Beijing 100193, China
    Search for more papers by this author
  • F. Q. TAN,

    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    Search for more papers by this author
  • Z. L. REN

    Corresponding author
    1. State Key Laboratory of Plant Breeding and Genetics, Sichuan Agricultural University, Ya'an, Sichuan 625014, China
    2. School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu 610054, China
      P. G. Luo. Fax: +86 28 86290978; e-mail: lpglab@sicau.edu.cn. Z. L. Ren. Fax: +86 28 86290873; e-mail: zhlrenlab@yahoo.com.cn
    Search for more papers by this author

P. G. Luo. Fax: +86 28 86290978; e-mail: lpglab@sicau.edu.cn. Z. L. Ren. Fax: +86 28 86290873; e-mail: zhlrenlab@yahoo.com.cn

ABSTRACT

CN17 is a functional stay-green wheat variety that exhibits delayed leaf senescence and enhanced photosynthetic competence. To better understand these valuable traits, levels of chlorophyll a and b, soluble proteins, unsaturated fatty acids, and other components of CN17 were assayed. In addition, chloroplast ultrastructure, chloroplast number, and differences in gene expression between CN17 and a control variety, MY11, were examined. By 21 d post-anthesis (DPA), CN17 leaves exhibited a significantly higher maximal photochemical efficiency for photosystem II (PSII) (Fv/Fm) and a significantly higher efficiency of excitation capture by open PSII reaction centres (Fv′/Fm). In addition, chlorophyll degradation in CN17 was delayed by approximately 14 d, and was not blocked as observed in cosmetic stay-green phenotypes. The soluble protein content (Ps) of CN17 was higher than MY11 at all timepoints assayed, and the ratio of unsaturated to saturated fatty acids was significantly higher. CN17 also exhibited isolated granal lamellae associated with vesicles and diminished peroxidation, and between 35 and 42 DPA, a sharp decrease in chloroplast number was detected. Taken together, these results strongly support the hypothesis that chloroplast ultrastructure regeneration is responsible for the functional stay-green trait of CN17, and gene expression data provide insight into the mechanistic details.

INTRODUCTION

Leaf senescence involves an extensive amount of cell death (Gan & Amasino 1995) in a plant, and potentially reduces photosynthesis (Gut et al. 1987). In general, a reduction in photosynthesis is attributed to changes in chloroplast ultrastructure (Grover & Mohanty 1992) which can lead to a decline in the photochemical activities of photosystems and the disassembly of the photosynthetic apparatus (Thomas & Stoddart 1980; Grover & Mohanty 1992). Overall, there have been significant genetic variations in the rate and degree of leaf senescence observed during important growth stages (e.g. the grain-filling period) both between species and within species (Thomas & Howarth 2000). In addition, delayed leaf senescence resulting from a stay-green genotype has been reported in many plants, including several crop species (Hörtensteiner 2009).

Currently, there are five types of stay-green plant varieties that have been identified: type A – which exhibits a loss of pigment and function at a normal rate with delayed initiation of senescence; type B – which exhibits normal initiation yet a slower rate of senescence; type C – which undergoes functional senescence at a normal rate yet involves defective chlorophyll degradation; type D – which undergoes rapid tissue death; and type E – which has a normal epigenetics pattern to support the photosynthetic capacity needed for an intensely green genotype, yet contains stay-green pigment (Thomas & Howarth 2000). Furthermore, these five varieties can be assigned to two principal categories: functional and cosmetic (Hörtensteiner 2009). In functional stay-green varieties, photosynthetic activity is retained during leaf senescence, while cosmetic stay-green varieties maintain their green colour due to a decrease in chlorophyll degradation in the absence of photosynthetic activity. Functional stay-green genotypes are potentially valuable in agronomy as a delay in the initiation or progression of senescence can promote the transfer of photosynthetic products from energy sources (e.g. photosynthetic tissues such as leaves) to energy sinks (e.g. developmental organs such as seeds) (Borras, Slafer & Otegui 2004). As a result of this transfer, crop yields are increased.

A common feature of stay-green genotypes is their capacity to retain chlorophyll in the photosynthetic apparatus during senescence. In general, chlorophyll is maintained in protein complexes to avoid the photoxidative damage that free chlorophyll can cause in cells (Kusaba et al. 2007). These protein complexes include photosystem I (PSI), photosystem II (PSII) and the cytochrome b6f complex, which are all found in thylakoid membranes. In higher plants, two types of chlorophyll are present: chlorophyll a (Chl a) and chlorophyll b (Chl b). Chl a is a component of all chlorophyll–protein complexes, while Chl b is localized primarily in the light-harvesting Chl a/b–protein complex (LHCP). In addition, Chl b is hypothesized to be important for the stability of the LHCP (Bellemare, Bartlett & Chua 1982). Similarly, PSII-associated light-harvesting proteins (LHCII) are localized primarily in the grana (the appressed regions of the thylakoid membrane) and are thought to play an important role in grana formation (Allen & Forsberg 2001).

Previous studies of several cosmetic stay-green mutants (e.g. nyc1, nil, cyc3, sgr and pph-1) have demonstrated that their mutated genes are only involved in the chlorophyll degradation pathway, which is not coupled to effective photosynthesis (Armstead et al. 2007; Jiang et al. 2007; Kusaba et al. 2007; Park et al. 2007; Sato et al. 2007, 2009; Morita et al. 2009; Schelbert et al. 2009). Therefore, the functional stay-green trait is of particular interest to crop breeding researchers as it has the potential to increase grain yields of cereal crops. However, although functional stay-green varieties have been characterized physiologically and genetically in several plant species (Thomas & Howarth 2000; Spano et al. 2003; Yoo et al. 2007), the mechanism by which leaf greenness is coupled to effective photosynthetic competence remains unclear.

The wheat cultivar, CN17, was derived from a wheat-rye wide cross (Tang et al. 2008) and is a functional stay-green variety that retains high photosynthetic activity during the grain-filling stage (Luo et al. 2006). Many regions in southwest China grow CN17 due to its high yield (Luo et al. 2009) and its resistance to diseases such as stripe rust (Luo et al. 2008). CN17 plants also exhibit the type B stay-green phenotype and retain higher photosynthetic activity during senescence. The latter is determined by measuring the net photosynthetic rate during senescence. These traits are known to be controlled by a gene, or a group of genes, that are located primarily on the wheat-rye 1BL/1RS translocated chromosome (Luo et al. 2009). However, it remains unknown how, and by which pathway(s), these gene(s) regulate the stay-green traits of CN17. Therefore, the objective of the present study was to assay chlorophylls a and b, soluble proteins, unsaturated fatty acids, and other components of CN17, and also to compare the chloroplast ultrastructure, chloroplast number and gene expression of CN17 with a control variety.

MATERIALS AND METHODS

Plant genotypes and growth conditions

All of the experiments were conducted in the fields of the Yaan Agricultural Research Station of Sichuan Agricultural University in southwest China (27°17′ N, 120°16′ E) during the 2008–2009 wheat-growing season (total rainfall approximately 430 mm). The wheat cultivars used in the study included the functional stay-green cultivar, CN17 (Luo et al. 2009), and a control cultivar, MY11. The latter is an agronomic parent of the CN17 pedigree (Luo et al. 2008).

Cultivars of CN17 and MY11 were sown in clay soil on 2 November 2008. Each experiment was conducted as a randomized complete block design with three replicates. Each plot was 3.0 m long and 12 rows wide with a row spacing of 0.25 m. The rainfall received was sufficient for the entire growth stage, and no symptoms of water deficit were observed during the study. The average air temperatures from sowing to anthesis, and from anthesis to grain maturity, were 9.5 and 21 °C, respectively. Anthesis for CN17 and MY11 were recorded on 23 and 24 March 2009, respectively. Plants received 4, 4, 5 and 8 g N m−2 ammonium nitrate at the one-node, meiosis, heading and anthesis stages, respectively. Following the initial planting, the fungicide, JS399-19 (chemical name 2-cyano-3-amino-3-phenylancryic acetate; obtained from the Jiangsu Branch of National Pesticide Research & Development South Center of China, Jiangsu, China), and imidacloprid 10% wettable powder [chemical name 1-(6-chloro-3-pyridylmethyl)-N-nitromidazolidin -2-ylideneamine; obtained from the Yangnong Chemical Group Company, Jiangsu, China], were applied to control disease and pest stresses, respectively. Thirty plants of each cultivar that exhibited identical growth and developmental conditions for ear emergence were marked for all subsequent measurements and observations.

Measurement of chlorophyll fluorescence and quantum yields in PSII

Chlorophyll fluorescence of marked plants was measured using a modulated fluorescence system (Li-6400-40 LCF, Li-Cor, Lincoln, NE, USA) as described previously (Chen et al. 2010). Six measurements were obtained for flag leaves during the flowering stage, and the results represent the average value of three measurements recorded for 10 CN17 and 10 MY11 plants on six sampling dates. Maximal photochemical efficiency of PSII in dark-adapted leaves, Fv/Fm = (Fm − Fo)/Fm; the efficiency of excitation capture by open PSII reaction centres, Fv/Fm = (Fm − F0/Fm); the photochemical quenching coefficient, qP = (Fm − Fs / Fm − F0); the quantum yield of photochemical energy conversion in PSII (ΦPSII); the quantum yield of regulated non-photochemical energy loss in PSII (ΦNPQ); and the quantum yield of non-regulated, non-photochemical energy loss in PSII (ΦNO) were calculated as previously described (Kramer et al. 2004).

Measurement of chlorophyll and soluble protein concentrations in leaf tissues

Chlorophyll was extracted from leaf tissues homogenized in liquid nitrogen and then extracted with 80% (v/v) acetone containing 1 µm KOH three times. After centrifugation (2 min; 16 000 g), the combined supernatants were diluted 1:10 with acetone, and the absorbance of the mixture was measured by spectrophotometry at 645 and 663 nm (Strain, Cope & Svec 1971). Chl a and Chl b contents were calculated as previously described (Hill et al. 1985). In addition, soluble protein concentrations of flag leaf samples were determined using the Bradford (1976) method.

Analysis of lipids and fatty acids

Thylakoid membranes were isolated as previously described (Sprague & Staehelin 1987), and levels of lipids and fatty acids were analysed using the method of Chen & Anderson (1992) with minor modifications. Aliquots (400 µL) of the extracted lipids were dried under nitrogen and resuspended in 100 µL of 2-bromoacetophenone (10 mg mL−1 in acetone). After vortexing, 100 µL triethylamine (10 mg mL−1 in acetone) was added to the reaction mixture. The mixture was then sealed immediately in a screw-capped glass tube and heated for 5 min in a boiling water bath. After cooling, 160 µL acetic acid (2 mg mL−1 in acetone) was added to the mixture, and the tube was heated for 5 min. The resulting fatty acid phenacyl esters were filtered through a 0.22 µm microfilter membrane and redissolved in 500 µL methanol for analysis by high-performance liquid chromatography HPLC (Shimadzu 20 AT, Kyoto, Japan). Samples were separated on a Kromasil C18 column (Akzo Nobel, Hörneborgsvägen, Sweden) (5 µm, 4.6 mm × 250 mm). The mobile phase consisted of acetonitrile and 5% cetyltrimethylammonium bromide in water (95:5, v/v), and the mobile phase flow rate was monitored at 1.5 mL min−1. The detector was set at 242 nm and the column temperature was 18 °C.

Transmission electron microscopy of chloroplasts

Transmission electron microscopy (TEM) of wheat flag leaf chloroplasts of CN17 and MY11 plants was performed as described previously (Jung et al. 2003) using a 75 kV PEI-Tecnai 12 transmission electron microscope (PEI, Eindhoven, the Netherlands). Leaves were sampled weekly from anthesis to maturation.

Determination of chloroplast numbers

The number of chloroplasts present in each genotype was determined as described previously (Pyke & Leech 1987) using a Nikon Optiphot microscope (Tokyo, Japan).

Gene expression analysis

Total RNA was extracted from CN17 and MY11 plants using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. The Dynabeads oligo dT25 system (Dynal A.S., Oslo, Norway) was then used to purify the mRNA and to construct cDNA libraries which were prepared using RT-PCR. A suppression subtractive hybridization (SSH) library was constructed according to a protocol described previously (Diatchenko et al. 1996) that used a PCR-Select cDNA Subtraction Kit (Clontech, Palo Alto, CA, USA). Subtracted cDNA was then ligated into the pGEM-T vector (Promega, Madison, WI, USA) and transformed into Escherichia coli strain, DH5α. A total of 49 positive clones were sequenced using an ABI Prism 3100 automated sequencer (Perkin Elmer ABD, Santa Clara, CA, USA), and the sequences obtained were analysed against the GenBank database using BLASTX and BLASTN programs. Putative physiological and functional categories of the identified genes were assigned based on GoFigure Gene Ontology annotations available at http://www.geneontology.org.

Statistical analysis

Correlation coefficients between the values of the variable were determined using bivariate analyses performed by Sigmaplot 2001 software (SPSS Inc., Chicago, IL, USA). Changes in the mean of each index during the process of leaf senescence were used as variables for both CN17 and MY11.

RESULTS

PSII activity of CN17 is protected during leaf senescence

Using a modulated fluorescence system, chlorophyll fluorescence was measured for CN17 and MY11 plants. In these assays, CN17 exhibited a significantly higher maximal photochemical efficiency for PSII (Fv/Fm) and a significantly higher efficiency of excitation capture by open PSII reaction centres (Fv′/Fm) at 21 d post-anthesis (DPA). This timepoint represents the normal onset of leaf senescence in wheat. In contrast, MY11 exhibited a noticeable decrease in both Fv/Fm and Fv′/Fm values at 21 DPA (Fig. 1a,b). Moreover, values for Fv/Fm and Fv′/Fm for CN17 at 42 DPA were more than 80% as high as at 21 DPA, while the Fv/Fm ratio at 42 DPA for MY11 was approximately 40% of the 21 DPA value, and the Fv′/Fm ratio was less than 10% of the 21 DPA value. Furthermore, between 14 and 21 DPA, the Fv′/Fm ratio for CN17 was significantly reduced (P < 0.001) concomitant with a significant increase in the Fv/Fm ratio (P = 0.04). In contrast, only a slight decrease in the Fv′/Fm ratio was observed for MY11 during this interval (P = 0.054), accompanied by a slight increase in the Fv/Fm ratio (P = 0.82). During the same interval, CN17 exhibited a higher photochemical quenching coefficient (qP), a higher quantum efficiency of photosystem electron transport (ΦPSII) value, a lower quantum yield of regulated non-photochemical energy loss in PSII (ΦNPQ), and a lower quantum yield of unregulated, non-photochemical energy loss in PSII (ΦNO), particularly at the end of leaf senescence compared with MY11 (Fig. 1c–f). Taken together, these findings suggest that CN17 is better able to protect itself compared with MY11, especially after 21 DPA.

Figure 1.

Differences in photosystem II (PSII) activity between the stay-green variety (CN17) and the control variety (MY11) after flowering. Values for maximal photochemical efficiency of PSII (Fv/Fm) (a), efficiency of excitation capture by open PSII reaction centres (Fv′/Fm) (b), photochemical quenching coefficient (qP) (c), quantum yield of photochemical energy conversion in PSII (ΦPSII) (d), quantum yield of regulated non-photochemical energy loss in PSII (ΦNPQ) (e), and quantum yield of non-regulated, non-photochemical energy loss in PSII (ΦNO) (f) in wheat flag leaves during late leaf senescence are provided. For all values, n = 10 and the data represent the mean ± SEM. ΦPSII, ΦNPQ, and ΦNO were derived from the average of related parameters. DPA, days post-anthesis; flowering was designated as 0 DPA.

CN17 has a strong ROS degradation system that maintains the stability of PSII

In CN17, Chl a levels decreased somewhat between 14 and 21 DPA, whereas Chl b levels increased slightly, resulting in a smaller Chl a/b ratio compared with control MY11 (Fig. 2a,c,d). As Chl b is mainly localized to PSII, whereas Chl a is distributed between both PSII and PSI (Schelbert et al. 2009), this finding indicates that PSII is protected to a greater extent than PSI in CN17. Moreover, chlorophyll levels, particularly those of Chl b, were clearly reduced between 35 and 42 DPA in CN17. This observation suggests that chlorophyll degradation in CN17 was delayed by approximately 14 d, and was not blocked as observed in cosmetic stay-green phenotypes. In our previous study, CN17 also exhibited a greater antioxidative capacity than MY11 (Luo et al. 2006). The soluble protein content (Ps) of CN17, which consists mainly of ribulose 1,5-bisphosphate carboxylase/oxygenase (RuBisCo), was higher than MY11 at all timepoints assayed following anthesis (Fig. 2d). It is possible that this soluble protein content positively regulates photosynthesis and reduces the accumulation of electrons to minimize the formation of reactive oxygen species (ROS).

Figure 2.

Changes in Chl a (a) and Chl b content (b), as well as the Chl a/Chl b ratio (c) and soluble protein (Ps) content (d), after flowering in wheat flag leaves of CN17 and MY11. For all values, n = 10 and the data represent the mean ± SEM. DPA, days post-anthesis; flowering was designated as 0 DPA.

Protection of the grana in CN17 is required for reconstruction of the photosynthetic apparatus during delayed leaf senescence

Using TEM analysis, changes in chloroplast ultrastructure were found to be similar in CN17 and MY11 up to 21 DPA (Fig. 3a,b). However, at 21 DPA (the onset of normal leaf senescence), CN17 exhibited isolated granal lamellae that were associated with vesicles and diminished peroxidation (Fig. 3c,d), as indicated by densely stained osmiophilic plastoglobuli (Op) and granal oligomers present (Fig. 3c). These changes are indicative of mechanisms that protect the ultrastructure of chloroplasts (Lee et al. 2009). Furthermore, the long chain of grana connected by stromal thylakoids in the ultrastructure of the chloroplasts were observed to reappear by 28 DPA in CN17. Specifically, connections between granal oligomers were detected (Fig. 3f), changes in vesicle shape were observed that represented a restoration of thylakoids to a well-organized arrangement (Fig. 3g) and densely stained Op were detected (Fig. 3h,i). In contrast, large starch grains (S) and chloroplasts separated from the cell membrane were detected in MY11 around 28 DPA (Fig. 3j). In Fig. 3e, dismantled chloroplasts in MY11 at 21 DPA are shown, compared with CN17 at 35 DPA (Fig. 3l). These observations suggest that grana in CN17 were protected to a greater extent than the stromal thylakoids when leaf senescence was initiated at 21 DPA, and that the retention of stacked granal thylakoid may be crucial for reconstruction of the photosynthetic apparatus. Moreover, this reconstruction process appears to delay leaf senescence by approximately 14 d.

Figure 3.

Changes in chloroplast ultrastructure before and after initiation of leaf senescence in CN17 and MY11. Representative electron micrographs of flag leaves collected from CN17 (the black arrow indicates the osmiophilic plastoglobuli) (a) and MY11 (the black arrow indicates the grana thylakoid) (b) at approximately 14 DPA (the white arrow indicates the cell wall). (c) and (d): a CN17 flag leaf cells from CN17 at 21 DPA. (e): A MY11 flag leaf cell at approximately 21 DPA. For CN17 flag leaf cells at 28 DPA, connections between oligomers of the photosynthetic apparatus are shown (the black arrow indicates the reforming stoma thylakoid) (f), thylakoids regain a well-organized arrangement and are separated by vesicles (g), and chloroplast ultrastructure is reformed and includes production of Mw (h and i, with the white arrows in panel I indicating the dissolution of Ops previously associated with dense staining). (j) The appearance of large starch grains (s) and separation of a chloroplast from the cell membrane at approximately 28 DPA in a MY11 flag leaf cell (the white arrow indicates a dissolving, separated chloroplast). (k) Production of membrane whorls (Mw) during the dismantling of chloroplast ultrastructure in MY11 plants at approximately 28 DPA. (l): Dismantling of chloroplast ultrastructure after a period of regeneration in a CN17 plant at approximately 35 DPA. St, stroma; Cw, cell wall; N, nucleus; Cp, chloroplast; Gt, grana thylakoid; St, stoma thylakoid; Op, osmiophilic plastoglobuli; v, vesicles. Scale bars = 1 µm.

The high polyunsaturated fatty acid content of CN17 is important for protection of the photosynthetic apparatus

An analysis of the fatty acid composition of thylakoid membranes in CN17 and MY11 (Fig. 4a–f) found that linolenic acid, linoleic acid, palmitic acid and stearic acid were the main lipid components in both genotypes. However, the ratio of unsaturated to saturated fatty acids was found to be significantly higher in CN17 than in MY11 at 21 DPA. Furthermore, upon completion of leaf senescence in CN17, this ratio was markedly higher than the initial value. A parallel increase in the content of unsaturated fatty acids in CN17 was also detected between 21 and 28 DPA. These findings are consistent with the hypothesis that the chloroplast ultrastructure is regenerated during this stage in CN17, and that the increased ratio of unsaturated to saturated fatty acids enhances the stability of the thylakoid membrane by the end of leaf senescence (McConn & Browse 1998).

Figure 4.

Primary fatty acid levels detected in thylakoid membranes of NC17 and MY11 after flowering. The amounts of linolenic acid (a), linoleic acid (b), oleic acid (c), palmitoleic acid (d), stearic acid (e) and palmitic acid (f) were assayed. For all the values, n = 3, and the data represent the mean ± SEM. DPA, days post-anthesis; flowering was designated as 0 DPA; mg per g refers to fresh weight.

Synchronous degradation of chloroplasts is delayed in CN17

When the numbers of chloroplasts present were compared for CN17 and MY11, little difference was observed before 28 DPA. However, between 35 and 42 DPA, a sharp decrease in chloroplast number was detected in CN17, while a similar decrease was observed between 28 and 35 DPA for MY11 (Fig. 5). These findings indicate that CN17 is able to block, or delay, the degradation of its constituent internal chloroplasts when leaf senescence begins. These findings also support the hypothesis that the stay-green phenotype of CN17 does not result from a blockage of the chlorophyll degradation pathway as it does in cosmetic stay-green varieties. Rather, the functional stay-green phenotype of CN17 is due to prolonged maintenance of a stable chloroplast ultrastructure, thereby potentially delaying the dissociation of chlorophyll from light-harvesting complexes.

Figure 5.

Changes in the numbers of chloroplasts per cell detected for CN17 and MY11. The mean ± SEM is provided for 30 cells for each timepoint and cell line. Nc, number of chloroplasts.

BLAST analysis of EST sequences against leaf senescence-induced cDNA libraries

Using the BLAST program, a total of 49 EST sequences were aligned against three NCBI databases. As a result, 32/49 sequences showed homology with known or annotated genes present in cDNA libraries derived from leaf senescence-related processes in wheat, rice, barley, pea and Arabidopsis (Table 1). Of the genes up-regulated in CN17, 31.3% belonged to photosynthesis and leaf senescence pathways, 12.5% belonged to the hydrolase pathway, 9.4% belonged to the oxidoreduction pathway, 9.4% belonged to cellular processes pathways, and 3.1% belonged to the photorespiration, lipid transport and protein biosynthesis pathways. The remaining 28.1% of the up-regulated genes detected had unspecified or unknown functions. In addition, 2/32 sequences matched putative senescence-associated genes. These included JK738986 and JK739001 (Table 1) which were up-regulated in CN17 and were found to share high homology with the cDNA of ssa-13.

Table 1. BLAST search performed for forward subtraction clones present in the functional stay-green genotype, CN17
PathwayExpressed sequence TAG accession no.GenBank accession no.Results of BlastxGene nameCopy of ESTsLength (bp)Max identity (%)
Photosynthesis (31.3%)JK738973CAA44027.1Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit rbcL 2296100
JK738985AAY54130.112596
JK738975BAA35176.1Ribulose-1,5-bisphosphate carboxylase/oxygenase small subunit rbcS 17492
JK738982ABI96906.1Ribulose bisphosphate carboxylase/oxygenase activase RCA 281100
JK738989ABG22615.115696
JK738994ABY52939.1PSII oxygen-evolving complex 23 kDa protein, chloroplast psbP 128290
JK738978CAP72041.1PSII protein D1 psbA 126699
JK738993ACS68717.1PSI reaction centre subunit XI psaL 14192
JK738996P20143.1PSI reaction centre subunit VI psaH 166100
JK738981ADB97492.1NADPH-protochlorophyllide oxidoreductase B porB 1297100
Hydrolase (12.5%)JK738987XP_003604156.1Cell wall-associated hydrolase 485100
JK738988XP_003604156.185100
JK738992XP_003637074.1292100
JK739000XP_003637074.1282100
Lipid transport (3.1%)JK738984AAB32998.1Lipid transfer protein homologltp9.413792
Oxidoreduction (9.4%)JK738999ADK94876.1Putative cytosolic ascorbate peroxidase 1105100
JK738990ACZ73612.1Cinnamate 4-hydroxylase 113477
JK739002ABB48040.1POT family proteinLOC_Os10g42870118385
Protein biosynthesis (3.1%)JK738976CAO02550.1Putative ribosomal protein S3rrn23114498
Photorespiration (3.1%)JK738983ACB69759.1Hydroxypyruvate reductaseHPR110294
Cellular processes (9.4%)JK738980BAA10929.1Cytochrome P450-like TBP proteinCYP P45038370
JK738997XP_003614400.113580
JK738998ABO20848.1154100
Unknown pathway (28.1%)JK738986ABO20848.1Putative senescence-associated proteinssa-132292100
JK739001BAB33421.125893
JK738974NP_001054429.1Unknown protein 157100
JK738991BAK03023.1Predicted protein 113490
JK739003XP_002489002.1Hypothetical protein SORBIDRAFT_0531s002010 118383
JK738977 No significant similarity found 1  
JK738972 No significant similarity found 1  
JK738979 No significant similarity found 1  
JK738995 No significant similarity found 1  

DISCUSSION

Photochemical efficiency and pigment components in CN17

Previous studies have shown that CN17 plants display greater photosynthetic competence during the later stages of leaf senescence compared with control MY11 plants (Luo et al. 2006). However, it was unclear whether this increase in photosynthetic competence was due to an increase in the photochemical efficiency of PSII, or an increase in the efficiency of excitation capture by open PSII reaction centres. In the present study, CN17 exhibited significantly higher values for maximal photochemical efficiency of PSII (e.g. Fv/Fm), and efficiency of excitation capture by open PSII reaction centres (e.g. Fv′/Fm), particularly in the later stages of leaf senescence (Fig. 1a,b). CN17 also displayed a higher qP, a higher proportion of absorbed energy was used in photochemistry (ΦPSII), a lower quantum yield of regulated non-photochemical energy (ΦNPQ) loss in PSII was detected, often considered a marker of photoprotection (Hendrickson, Furbank & Chow 2004; Klughammer & Schreiber 2008), and less non-regulated, non-photochemical energy was lost in PSII (ΦNO), especially during the later stages of leaf senescence (Fig. 1c–f). Based on these results, we hypothesize that protection of PSII, as well as effective energy utilization by CN17, contribute to the functional stay-green phenotype of CN17.

During leaf senescence, CN17 was also found to retain higher levels of Chl a and Chl b than MY11 (Fig. 2a–c). Moreover, the ratio of Chl a to Chl b was markedly reduced between 14 and 21 DPA in CN17, yet significantly increased during the same interval in MY11 (Fig. 2c). As Chl a and Chl b are components of LHCII, these observations suggest that LHCII is retained in CN17 at 21 DPA, possibly due to state transitions in the membrane architecture (Chuartzman et al. 2008). Consistent with this hypothesis, Chl b has been shown to be important for maintaining LHCII stability (Bellemare et al. 1982; Sato et al. 2009). However, during the later stages of leaf senescence in CN17, Chl b and Chl a levels were observed to decrease (Fig. 2a,b). Based on these results, it appears that the stay-green phenotype of CN17 correlates with a retention of chlorophyll, and therefore, the chlorophyll degradation pathway was not completely inhibited.

Granal stability and PSII activity

Ultrastructural analyses performed for CN17 and MY11 identified several notable differences in chloroplast degeneration. Firstly, the thylakoid system in CN17 at 21 DPA was observed to include a disorganized set of granal lamellae, with interior granal lamellae being isolated within several pronounced lamellar regions by vesicles. These observations are consistent with the effects of an osmotic regulation mechanism whereby granal lamellae have been shown to be protected from oxidative degradation by being separated (Robinson 1985). Secondly, a higher number of oligomeric granal lamellae was present in CN17 than in MY11. Thirdly, while the number of Op was found to be similar for the two genotypes, these structures were more strongly stained with osmium tetroxide in CN17 cells at 21 DPA. Osmium tetroxide reacts primarily with the double bonds of unsaturated fatty acid residues in lipids (Ongun, Thomson & Mudd 1968), particularly glycolipids and phospholipids which are important components of chloroplast membranes (McConn & Browse 1998). Therefore, the difference in osmium tetroxide staining between CN17 and MY11 may represent a reduction in the number of unsaturated lipids that are undergoing oxidative degradation in CN17 due to an enhanced antioxidant defence system (Luo et al. 2006).

Previous reports have suggested that the restoration of chloroplast ultrastructure in senescent leaves is associated with changes in the distribution of thylakoid membranes and granal stacking (Greening, Butterfield & Harris 1982). Accordingly, the results of the present study, and those of a previous study (Kusaba et al. 2007), support the hypothesis that the high connectivity between thylakoid membranes in the grana physically contributes to the retention of stacked thylakoid membranes. Moreover, the significant decrease in the Fv′/Fm ratio (P < 0.001) and the significant increase in the Fv/Fm ratio (P = 0.04) from 14 to 21 DPA in CN17, while changes in both Fv′/Fm and Fv/Fm ratios were not significant (P = 0.05) (Fig. 1a,b), support the hypothesis that this retention delays degradation of the reaction centre in PSII by sequestering photoinactivated PSII in appressed granal domains (Chow et al. 2005). This retention may also play a key role in the regeneration of the photosynthetic apparatus as less energy is needed to form unstacked thylakoid membranes versus grana (Kim et al. 2005). Furthermore, protection of PSII and the grana may provide an opportunity for complete reconstruction of the photosynthetic apparatus to occur following leaf senescence, and this could be facilitated by the delay associated with leaf senescence that potentially inhibits a further increase in photosystem disorder. It is also well known that oxidative stress caused by ROS can suppress the repair of photo damage to PSII by suppressing the de novo synthesis of most proteins present in thylakoid membranes (Nishiyama et al. 2001, 2004). Correspondingly, CN17 was previously found to have an increased antioxidative capacity (Luo et al. 2006). Therefore, in combination, these results indicate that maintenance of an active PSII in CN17 involves both sustained repair of PSII and protection of granal structures.

Thylakoid membrane regeneration and unsaturated fatty acid composition

Findings in Arabidopsis mutants indicate that plants require high levels of polyunsaturated lipids to maintain their photosynthetic machinery (McConn & Browse 1998). A fatty acid analysis of thylakoid membranes in CN17 revealed that a high ratio of unsaturated to saturated fatty acids was present during leaf senescence. In addition, a transition between the appearance of darkly stained Op and thylakoids was also observed to be reversible, and to be accompanied by developmental changes such as a reduction in photoassimilates (Luo et al. 2009) in flag leaf cells. This observation suggests that the function of Op depends on the degree of oxidation of the fatty acid residue carbon double bonds of lipids in thylakoid membranes. This hypothesis is also consistent with a previous study in which a reduction in polyunsaturated membrane lipids was found to significantly decrease the number of stacked and non-stacked thylakoids present (Hugly et al. 1989). Overall, these observations suggest that the unsaturated fatty acid content of thylakoids has two effects: it helps alleviate damage to PSII, and it enhances repair of PSII following damage. Similar observations were made during salt-induced leaf senescence in Synechococcus (Allakhverdiev et al. 2001). Moreover, in rice leaf, unsaturated fatty acids have been hypothesized to protect chloroplast structures (Zhang et al. 2010). Thus, the thylakoid membranes in CN17 may have been regenerated between 21 and 28 DPA through the activity of Op which depends on the presence of unsaturated fatty acids.

Reconstruction of the photosynthetic apparatus and the functional stay-green trait

To protect itself against damage induced by an excess of absorbed light energy, a plant will reorganize the structure of its photosynthetic membrane. This process has been shown to involve the dissociation of PSII and the aggregation of stacked thylakoid membranes (Johnson et al. 2011). In the present study, the disorganization and isolation of lamellar stacking observed in CN17 were restored to well-defined and regularly distributed granal stacks connected by parallel stromal thylakoids through the osmotic regulation of vesicles. Furthermore, the high levels of unsaturated fatty acids present promoted the ability of the lipid matrix to mediate the assembly of chloroplast membrane components (Tsvetkova, Brain & Quinn 1994). This facilitated the reforming and restacking of thylakoid lamellae as indicated by the dark staining obtained with Op. Therefore, chloroplast ultrastructure in CN17 appears to be progressively regenerated when low levels of peroxidation of polyunsaturated membrane lipids are maintained, thereby permitting the subsequent reassembly of the functional oligomeric components of the photosynthetic apparatus. The highly connected morphology of this apparatus is manifested by the enhanced photosynthetic competence of CN17 exhibited following leaf senescence (Fig. 1).

Proposed mechanism of delayed leaf senescence in CN17

Recent reports indicate that the overexpression of the C-repeat/dehydration-responsive element binding factor 2 (CBF2) gene in Arabidopsis delays the onset of leaf senescence and extends the life span of plants by approximately 2 weeks (Sharabi-Schwager et al. 2010). These observations imply that the regeneration of the photosynthetic apparatus can be regulated by stress-responsive genes. In Fig. 6, we summarize the relationship between genes previously shown to be involved in leaf senescence in CN17. Moreover, in the present study, two up-regulated sequences (JK738986 and JK739001; Table 1) were found to exhibit high homology with the cDNA of ssa-13, a gene induction factor of leaf senescence. Further studies will be needed to characterize these newly discovered genes associated with leaf senescence, and to identify the indices that may be affected by their expression. Indices relevant to leaf senescence are provided in Table 2, which summarize the differences observed between CN17 and MY11.

Figure 6.

A diagram representing the relationships between genes found to be involved in leaf senescence and the resulting changes in chloroplast structure and function in CN17. Thick and thin arrows represent primary versus secondary causes and effects, respectively, based on the gene expression analysis performed.

Table 2. A comparison of the physiological and biochemical indices of CN17 and MY11 associated with leaf senescence
 Chl a/bChl aChl bNcPs F v/Fm F v′/FmqP ΦPSIIC18:3C18:2C16:1C16:0C18:0
  1. Note: For each item, the top line represents MY11 (control) data and the bottom line represents CN17 data.

  2. *, different at P = 0.05. **, different at P = 0.01.

  3. – indicates a negative relationship.

  4. Chl a/b, ratio of Chl a to Chl b; Nc, number of chloroplasts per cell; Ps, soluble protein; Fv/Fm, maximal photochemical efficiency of photosystem II (PSII); Fv′/Fm′, efficiency of excitation capture by open PSII reaction centres; qP, photochemical quenching coefficient; ΦPSII, quantum efficiency of photosystem electron transport; C18:2, linoleic acid; C16:1, palmitoleic acid; C16:0, palmitic acid; C18:0, stearic acid.

Chl a/b **      * *  
  * **     * 
Chl a   ** ** ** ** **   ** * *   
        *       
Chl b     *     * ** *    
              
Nc     **   **   ** ** **    
     *          
Ps      ** **   ** ** *    
       *       *  
F v/ F m        **   * *     
       *      * *  
F v′/Fm         ** ** *    
         *    * * *
qP              
         *      
ΦPSII          ** ** * * *
              
C18:3            *   *
              
C18:2            *   **
              
C16:1             * *
              
C16:0              **
              **

CONCLUSIONS

Based on the results of the present study, we propose a model for leaf senescence in CN17 whereby protection of PSII and granal stability, as well as an increase in levels of unsaturated fatty acids, provide an opportunity to regenerate chloroplast ultrastructure. Furthermore, the stronger antioxidative capacity of CN17 sustains the effective repair of PSII by maintaining the normal synthesis of proteins de novo, thereby ensuring that reconstruction can be accomplished. Our results suggest that reconstruction of the chloroplast ultrastructure delays leaf senescence, and therefore, is responsible for enhanced photosynthetic competence. It is also hypothesized that in functional stay-green varieties, the delay in leaf senescence is regulated by the ROS degradation system, protein synthesis and fatty acid metabolism. This is in contrast with cosmetic stay-green varieties of wheat where only the chlorophyll degradation pathway has been shown to mediate leaf senescence.

ACKNOWLEDGMENTS

We express our gratitude for the financial support of the National Natural Science Foundation of China (No. 30971787 and 30730065) and the Provincial Science and Technology Foundation for Young Scientists of Sichuan, China (2010JQ0042).

Ancillary