Physiological analysis of leaf senescence of two rice cultivars with different yield potential

The objective of this work was to evaluate the physiological changes that occur in different leaves during the early and late grain-fi lling stages of two rice genotypes (Oryza sativa subsp. indica, BRS Pelota cultivar, and O. sativa subsp. japonica, BRS Firmeza cultivar), which present differences in grain yield potential. The plants were cultivated in greenhouse. Pigment content, chlorophyll fl uorescence, electron transport and oxygen evolution rate were determined in the grain-fi lling stage, from the fi rst to the forth leaf (top to bottom). Pigment content, photochemical effi ciency of photosystem II and electron transport decreased signifi cantly according to the position of leaves in 'BRS Pelota'. The BRS Firmeza cultivar shows higher pigment content and higher activity of the photosynthetic apparatus in comparison to 'BRS Pelota' during the grain-fi lling stage.


Introduction
Senescence is a normal event in the life cycle of plants (Borrás et al., 2003).It is a sequence of complex degenerative processes that are initiated at full maturity and may ultimately lead to leaf death (Camp et al., 1982;Thomas et al., 2003;Tang et al., 2005).The most remarkable events in leaf senescence are the loss of chlorophyll and the disassembly of the photosynthetic apparatus, which result in decreases in the photosynthetic energy conversion capacity and effi ciency.Also, there is a concurrent decline in chain electron transport for those components remaining in the leaf (Adams et al., 1990;Weng et al., 2005;Zhang et al., 2006).The decrease in electron transport along photosystem II may be due to an inactivation of the oxygen evolution system or of the photosystem II reaction center complex, as well as to the inhibition of energy transfer from carotenoids to chlorophyll (Lu et al., 2002).In addition, senescence also affects the composition of the antenna system of photosynthesis.According to Kura-Hotta et al. (1987), chlorophyll a disappears more Pesq. agropec. bras., Brasília, v.44, n.7, p.695-700, jul. 2009 rapidly than chlorophyll b during senescence, resulting in a decrease in the chlorophyll a/b ratio.
Many studies related to the senescence of rice leaves have been reported (Biswas & Choudruri, 1980;Kura-Hotta et al., 1987;Hidema et al., 1991;Yamazaki et al., 1999;Tang et al., 2005;Weng et al., 2005).Historically, researches on biochemical changes that occur during leaf senescence have focused on loss of photosynthetic pigments, degradation of protein, and reabsorption of mineral nutrients (Zhang et al., 2006).Reductions in photosystem II (PSII) activity, expressed through the analyses of chlorophyll fl uorescence and electron transport chain, have recently been examined during the senescence of leaves in different cultivars of super-high-yielding hybrid rice (Jiao et al., 2003;Weng et al., 2005).However, most of these investigations have focused on senescence processes at the single-leaf level.
Chlorophyll a fl uorescence has been shown to be a noninvasive and reliable method to assess the changes in PSII function under different environmental conditions (Krause & Weis, 1991).The use of this method to assess PSII photochemistry, during leaf senescence in cultivars with differences in grain yield potential, is important because it gives new insights about the fundamental processes of energy absorption and excess excitation energy use and dissipation by PSII in cereal crop plants during senescence.Moreover, leaf senescence pattern is commonly variable, and differences in canopy depth and leaf senescence should exist among cultivars, for they might contribute to the differences observed in grain yield.
Tropical Oryza sativa subsp.japonica rice cultivar BRS Firmeza presents stay-green phenotype and has shown later senescence than O. sativa subsp.indica rice cv.BRS Pelota.The stay-green trait may result from a delay in the onset of leaf senescence, from a reduced rate of senescence, or from the inhibition of one of the partial processes involved in chlorophyll breakdown (Thomas & Howarth, 2000;Luquez & Guiamét, 2001).On the theoretical basis, the stay-green trait may cause, different impacts on grain yield as a result of a delay in the onset of leaf senescence or of a lower photosynthetic decline rate, which extends the assimilatory capacity of the canopy, contributing to higher grain formation (Luquez & Guiamét, 2001;Rampino et al., 2006).
The objective of this work was to evaluate the physiological changes that occur in different leaves during the early and late grain-fi lling stages of two rice genotypes (O.sativa subsp.indica BRS Pelota cultivar -high grain production; and O. sativa subsp.japonica BRS Firmeza cultivar -low grain production).

Materials and Methods
Plants of rice (O.sativa L.) cultivars BRS Firmeza (O.sativa subsp.japonica) and BRS Pelota (O.sativa subsp.indica) were grown to the ripening stage (about 4-5 months) in a greenhouse under natural light conditions.The experiment was carried out in summer, from november 2007 to february 2009 at Pelotas, RS, Brazil.These cultivars were chosen because of their differences in grain-yield potential (7.5 and 10 Mg ha -1 for 'BRS Firmeza' and 'BRS Pelota', respectively).'BRS Firmeza' presents stay-green phenotype, characterized by an extended green leaf color period, which is longer than the one for BRS Pelota cultivar.
The experimental design was completely randomized with two rice cultivars and two grain-fi lling stages, with ten replicates for each rice cultivar.Each replicate was composed of one plastic pot (12 L) fi lled with soil fertilized according to Santos et al. (2006).Twenty-fi ve seeds were sown in each pot.In order to obtain uniform plants, the seedlings of each pot were thinned from 25 to 5 per pot after the appearance of the second leaf.Thereafter, a water depth of 3 to 5 cm was maintained until maturity.Leaf samples were collected from the plants during the early (100 days) and late (115 days) grain-fi lling stages in both cultivars.The changes in photosynthetic pigment content, chlorophyll fl uorescence parameters and oxygen evolution rate in the fl ag leaf (further referred to as 1 st leaf) and the second, third and fourth leaves were considered as a senescence index.
Chlorophyll and carotenoids were extracted with 80% acetone (Arnon, 1949) and determined using a spectrophotometer (Ultrapec 2100 pro, Biochrom Ltd., England) according to Lichtenthaler (1987).Chlorophyll a fl uorescence was measured using a fl uorescence monitoring system (FMS2, Hansatech, England) after the leaves were dark-adapted for 20 min.The minimal fl uorescence level (F 0 ) with all PSII reaction centers opened was determined by measuring modulated light, which was enough to avoid any signifi cant variation in fl uorescence.The maximal Pesq. agropec. bras., Brasília, v.44, n.7, p.695-700, jul. 2009 fl uorescence level (F M ) with all PSII reaction centers closed was determined by applying a saturating pulse on dark-adapted leaves.Then, the leaf was continuously illuminated with a white actinic light.The steady-state value of fl uorescence (F s ) was reached and a second saturating pulse was applied to determine the maximal fl uorescence level in the light-adapted state (F M ').The actinic light was then removed and the minimal fl uorescence level in the light-adapted state (F 0 ') was determined by illuminating the leaf with far-red light.
With the measurement of fl uorescence, the following parameters were calculated: maximal effi ciency of PSII photochemistry in dark-adapted state, F V /F M ; photochemical quenching coeffi cient, qP = (F M ' -F s )/(F M ' -F 0 '); effective effi ciency of PSII photochemistry, F V ' -F M '; actual quantum yield of PSII electron transport in light-adapted state, φ PSII = (F M ' -F s )/F M ' (Genty et al., 1989).The electron transfer rate through photosystem II (ETR) was calculated from the fl uorescence data using the formula from Krall & Edwards (1992): ETR = Φ PSII × 0.85 × 0.5 × PPFD, where 0.85 represents the estimated proportion of incident photons absorbed by the leaf (usually 80%), 0.5 indicates an estimated value for the distribution proportion of energy in PSII (usually 50% in C 3 plants), and PPFD indicates the photosynthetic photon fl ux density.
The photosynthetic rate (maximum oxygen evolution rate) was determined using a gas-phase oxygen electrode (Leaf Lab, Hansatech, England).A piece of about 3 cm 2 was cut from the center of the leaf blade.The determinations of oxygen evolution rate were conducted under the following conditions: air temperature of 30 o C and 2,000 μmol m -2 s -1 irradiation.The CO 2 was supplied as a 1 mol L -1 solution of sodium bicarbonate.The photosynthetic rate, expressed in μmol (O 2 ) m -2 s -1 , was calculated when the oxygen evolution curve became stable.The effect of different parameters was analyzed by ANOVA using a SPSS 11.5 statistical package (SPSS, USA).

Results and Discussion
Total chlorophyll content, chlorophyll a/b ratio and total carotenoids in leaves decreased according to the position of leaves in 'BRS Pelota' in both stages (Table 1).In 'BRS Firmeza', however, these values did not decrease markedly.These results indicate that the senescence of rice leaves may not be a simple function of leaf age, as indicated by Biswas et al. (1980), but may primarily be under genetic control (Okada, 1998;Yamazaki et al., 1999).'BRS Firmeza' shows stay-green phenotype (Magalhães Júnior et al., 2003), characterized by an extended leaf green color period, which can arise from delays in the beginning of senescence and in senescence progress, as previously indicated by Rampino et al. (2006).
The loss of chlorophyll and the concurrent yellowing of the leaves are convenient and distinctive indicators of leaf senescence.That characteristic phenotype can be uncoupled from physiological leaf senescence.The well-known phenomenon of leaf senescence is the loss of chlorophyll content.According to Wiedemuth et al. (2005), chlorophyll content is a widely used parameter for measuring the degradation of the photosynthetic apparatus in leaves during senescence.The changes in chlorophyll a/b ratio indicate that there are differential changes in the photosynthetic pigment stoichiometry during late senescence between O. sativa subsp.indica BRS Pelota cultivar and O. sativa subsp.japonica BRS Firmeza cultivar.It is assumed that the conversion of chlorophyll b to a represents a critical factor for changes in chlorophyll a/b ratio (Zhang et al., 2006).Gossauer & Engel (1996) stated that the conversion of chlorophyll b to a should precede chlorophyll degradation in higher plants.Thus, the differences in chlorophyll a/b ratio observed between 'BRS Pelota' and 'BRS Firmeza' should refl ect genetic differences Table 1.Chlorophyll content (µmol g -1 of fresh matter), chlorophyll a/b ratio and total carotenoids (µmol g -1 of fresh matter) from the fi rst to the forth leaf of BRS Pelota and BRS Firmeza rice cultivars at early (GF1) and late (GF2) grain-fi lling stages (1) .
The decline of photosynthetic pigment content was consistent with a loss of maximal photochemical efficiency of PSII (F V /F M ) in O. sativa subsp.indica BRS Pelota cultivar, in both stages (Table 2).The F V /F M values were similar between the 1 st (flag leaf) and 2 nd leaves during the early and late grain-filling stages, but they decreased in the 3 rd and 4 th leaves.These results were also verifi ed in the F V /F 0 ratio, which can be used as a parameter indicator of maximum effi ciency of the PSII photochemical process as well as of the activity of photosynthetic potential (Krause & Weis, 1991).In 'BRS Firmeza', no difference was observed in these parameters.This result refl ects the presence of stay-green gene(s) in 'BRS Firmeza', which causes the persistence of leaf green color, retarding the decrease of photosynthetic light reactions and extending the assimilatory capacity of the canopy, and this fact could indicate that the PSII apparatus remains functional for longer periods in senescent leaves of the rice cultivar characterized by a stay-green phenotype, i.e. 'BRS Firmeza' is able to maintain a continued effi cient energy transfer and capture in the remaining PSII apparatus.
The F V '/F M ', which expresses the effective effi ciency of PSII photochemistry measured in light-adapted samples (Roháĉek, 2002), did not differ in O. sativa subsp.indica BRS Pelota cultivar according to leaf position and grain-fi lling stage (Table 2).In contrast, F V '/F M ' was higher at the early grain-fi lling stage in 'BRS Firmeza' in comparison to the late stage.No differences were observed between leaves at the early grain-fi lling stage, but at the late stage higher values were observed in the fl ag and 3 rd leaves.
The earlier loss of effective photochemical effi ciency should be a good indicator of senescence in Pelota' plants, showing that the rice cultivar with higher grain-yield capacity senesces before the one with lower grain yield.Concerning the physiological signifi cance of the senescence and its relationship with assimilate partitioning in plants, these results suggest that the higher productivity of BRS Pelota rice cultivar results from its higher assimilate mobilization ability, in spite of its lower light absorption capacity.These results are supported by the lowest chlorophyll content and by the F V /F M values obtained in 'BRS Pelota' in both developmental stages, which showed reduction according to the leaf position (Tables 1 and 2).
In tropical O. sativa subsp.japonica BRS Pelota cultivar, the effective photochemical effi ciency of PSII (F V '/F M ') and the actual quantum yield of PSII electron transport in the light-adapted state (φ PSII ) were not different between either grain-fi lling stage or leaf position (Table 2).However, in 'BRS Firmeza', F V '/F M ' decreased with the leaf position in the late grain-fi lling stage.No variation was observed in φ PSII and F V '/F M ' at the early grain-fi lling stage.The variation dynamics of photochemical quenching (qP) values was similar to that observed Table 2. Maximal photochemical effi ciency of PSII (F V /F M ), maximum quantum yield ratio of photochemical and concurrent non-photochemical processes in PSII (F V /F 0 ), effective photochemical effi ciency of PSII (F V '/F M '), photochemical extinction coeffi cient (qP), actual photochemical effi ciency of PSII (φ PSII ), electron transport through PSII (ETR, µmol electrons m -2 s -1 ) and oxygen evolution rate (OER, µmol O 2 m -2 s -1 ) from the fi rst to the forth leaf of BRS Pelota and BRS Firmeza rice cultivars at early (GF 1 ) and late (GF 2 ) grain-fi lling stages (1) .
(1) Means (n = 5) followed by equal letters in the parameters do not differ by Tukey's test, at 5% probability. (2)Completely senescent leaves.Pesq. agropec. bras., Brasília, v.44, n.7, p.695-700, jul. 2009 in φ PSII .The photochemical quenching coeffi cient indicates the photon fraction absorbed by the antenna pigment used by PSII or the photochemical electron transport energy (Jiao et al., 2003).These results could indicate that BRS Firmeza rice cultivar would use the energy absorbed by PSII more effi ciently and convert this energy into a photochemical effect at the late grain-development stage.Furthermore, the electron transport rate through PSII decreased from the upper to the lower leaves in both rice cultivars during both developmental stages (Table 2).The gradually decreasing changes in ETR were consistent with the changes in F V /F M and the chlorophyll levels during plant senescence.
There were no significant differences regarding leaf position in the rates of photosynthetic oxygen evolution (OER) in both grain-filling stages of both rice cultivars (Table 2).These results were consistent with those obtained by Kura-Hotta et al. (1987), which showed that OER remained fairly constant among leaves at different positions.Also, no differences were observed in OER between the 1 st and 2 nd leaves in both IR58025a and IR58025B normal and fertility rice lines during the flowering and late grain-filling stages (Bacarin et al., 2008).However, opposite results had been observed previously in rice plants.Yamazaki et al. (1999) reported that the OER decreased with the depth of the canopy, except for the 2 nd leaves, which showed an OER similar to that of the 1 st leaves (flag leaves).That result was consistent with the increase in chlorophyll levels, which were higher in the 5 th leaves and decreased from the top to the bottom leaves.The oxygen evolution rate appears to be closely related to chlorophyll content, at least during the first week of plant development.
Although conventional plant breeding continues to produce new varieties with increased yield, the magnitude of these increases is falling, indicating that a plateau is being reached with the major yield-limiting factor being grain number (Rampino et al., 2006).The exploitation of stay-green phenotypes of the type described in the present work should potentially increase yields above this plateau through a decrease in the photosynthetic decline rate and an extension of the assimilatory capacity of the plant canopy.

Conclusion
The BRS Firmeza cultivar shows higher pigment content and higher activity of the photosynthetic apparatus in comparison to 'BRS Pelota' during the grain-fi lling stage.