post-anthesis development of inferior and superior spikelets in rice in relation to abscisic acid...

RESEARCH PAPER

Post-anthesis development of inferior andsuperior spikelets in rice in relation toabscisic acid and ethylene

Jianchang Yang1, Jianhua Zhang2,*, Zhiqing Wang1, Kai Liu1 and Peng Wang1

1 Key Laboratory of Crop Genetics and Physiology of Jiangsu Province, Yangzhou University, Yangzhou,Jiangsu, China2 Department of Biology, Hong Kong Baptist University, Hong Kong, China

Received 4 May 2005; Accepted 18 October 2005

Abstract

The purpose of this studywas to test the hypothesis that

the interaction between abscisic acid (ABA) and ethy-

lene may be involved in mediating the post-anthesis

development of spikelets in rice (Oryza sativa L.). Two

rice genotypes were field-grown, and the changes of

ABA, ethylene, and 1-aminocylopropane-1-carboxylic

acid (ACC) levels in spikelets during grain filling and

their relationships with endosperm-division and grain-

filling rates were investigated. The results showed that

earlier-flowering superior spikelets exerted dominance

over later-flowering inferior spikelets in endosperm

cell-division and grain-filling rates. The two genotypes

behaved the same. Later-flowering spikelets had higher

levels of ethylene and ACC than earlier-flowering

spikelets. The ethylene evolution rate was significantly

and negatively correlated with the cell division and grain

filling rates. By contrast to ethylene, later-flowering

spikelets contained a lower ABA content/concentration

and showed a low content ratio of ABA to ACC than

earlier-flowering ones. The cell-division and grain-filling

rates were significantly and positively correlated with

both ABA contents and the ratio of ABA to ACC.

Application of cobalt ion (inhibitor of ethylene synthe-

sis) or ABA at an early grain-filling stage significantly

increased endosperm cell division rate and cell number,

grain-filling rate, and grain weight of inferior spikelets.

Application of ethephon (an ethylene-releasing agent)

or fluridone (an inhibitor of carotenoid synthesis) had

the opposite effect. The results suggest that antagonis-

tic interactions between ABA and ethylene mediate

endosperm cell-division and grain-filling in rice. A

higher ratio of ABA to ethylene in rice spikelets is

required to maintain a faster grain-filling rate.

Key words: Abscisic acid, 1-aminocylopropane-1-carboxylic

acid, endosperm cell, ethylene, grain filling, inferior spikelets,

rice, superior spikelets.

Introduction

A rice (Oryza sativa L.) panicle is composed of a largenumber of spikelets, and each spikelet is considered anindividual unit in the complex inflorescence (Mohapatraand Sahu, 1991; Cao et al., 1992). During ontogeny, thedevelopment of primary branches on the panicle axis,opposite to their initiation, is in basipetal succession fromthe top to base, and spikelet development on a primary orsecondary branch is acropetal, with the exception of the oneat the tip of the branch which flowers first (Xu and Vergara,1986). Lack of synchronization in spikelet developmentcauses variations in kernel quality and weight (Mohapatraand Sahu, 1991; Yang et al., 2000). In general, earlier-flowering superior spikelets (grains), usually located onapical primary branches, fill fast and produce larger andheavier grains. While later-flowering inferior spikelets(grains), usually located on proximal secondary branches,are either sterile or fill slowly and poorly to produce grainsunsuitable for human consumption (Mohapatra et al., 1993;Naik and Mohapatra, 1999). The slow grain-filling rate andlow grain weight of inferior spikelets have often beenattributed to a limitation in carbohydrate supply (Sikder and

* To whom correspondence should be addressed. E-mail: [email protected]: ABA, abscisic acid; ACC, 1-aminocylopropane-1-carboxylic acid; DPA, days post-anthesis; IAA, indole-3-acetic acid.

Journal of Experimental Botany, Vol. 57, No. 1, pp. 149–160, 2006

doi:10.1093/jxb/erj018 Advance Access publication 5 December, 2005

ª The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.For Permissions, please e-mail: [email protected]

at Monash U

niversity on Decem

ber 4, 2014http://jxb.oxfordjournals.org/

Dow

nloaded from

http://jxb.oxfordjournals.org/

Gupta, 1976; Wang, 1981; Murty and Murty, 1982; Zhuet al., 1988). More recent work has shown, however, thatthere is no clear causative relationship between assimilateconcentration and spikelet development in rice (Mohapatraand Sahu, 1991; Mohapatra et al., 1993, 2000). So far, theintrinsic factors responsible for variations in grain fillingbetween the superior and inferior spikelets remain elusive.

It has been proposed that spikelet development may bemediated through endogenous hormones (Naik andMohapatra, 1999; Yang et al., 2000, 2001), and a low ratiobetween promotive and inhibitory hormones in inferiorspikelets may lead to their poor development (Naik andMohapatra, 1999). Among phytohormones, both ethyleneand abscisic acid (ABA) are generally regarded as in-hibitory growth regulators (Walton, 1980; Trewavas andJones, 1991; Chen and Lur, 1996; Mohapatra et al., 2000).A high ethylene evolution rate has frequently been relatedto abortion in maize (Zea mays) (Chen and Lur, 1996) andreduction in grain weight in wheat (Triticum aestivum) (Xuet al., 1995; Beltrano et al., 1999). Application of ethyleneinhibitors improves drymatter partitioning and developmentof later-flowering spikelets on rice panicles (Mohapatraet al., 2000). However, the cause-and-effect relationshipbetween ethylene emission from rice spikelets and thedevelopment of the spikelets has not been investigated.

There is a report that the anthers located on the distalfloret positions of a wheat spikelet contain higher ABAlevels than those on the proximal positions, and that thehigh ABA concentration is linked to reduced grain set (Leeet al., 1988). Under water stress, the reduction in grain setand kernel growth in wheat (Morgan, 1980; Saini andAspinall, 1982; Ahmadi and Baker, 1999) and a decreasedrate of endosperm cell division rate in maize (Myers et al.,1990; Ober et al., 1991) have been observed to beassociated with elevated levels of ABA. There are manyobservations, however, that ABA can promote dry matteraccumulation in the sink organ and its level is correlatedwith the growth rate of fruits or seeds (Eeuwens andSchwabe, 1975; Browning, 1980; Beruter, 1983; Schussleret al., 1984, 1991; Wang et al., 1987; Ross and MacWha,1990; Kato et al., 1993; Wang et al., 1998; Yang et al.,2001).

The endosperm of rice represents >90% of the finalweight of a kernel (Murata and Matsushima, 1975; Caoet al., 1992). The process of grain filling is actually theincrease in both cell number and cell filling in theendosperm. There are generally positive relationshipsbetween endosperm cell division rate, endosperm cellnumber, and grain weight (Cao et al., 1992; Yang et al.,2002a). It is highly possible that both ABA and ethylenemay exert their effects on cell division at the endospermdevelopment stage, but information linking ABA andethylene actions to cell division is lacking in the literature.

The purpose of this study was to test the hypothesis thatABA and ethylene and their interactions may be involved

in mediating the post-anthesis development of inferior andsuperior spikelets on a rice panicle. The changing patternsof ABA and ethylene concentrations in rice grains duringgrain filling and their relationships with endosperm-divisionand grain-filling rates were investigated. The effects ofchemical regulators on the levels of ABA and ethylene inthe grains were also studied to verify the roles of the twohormones.

Materials and methods

Plant materials

The study was conducted at a farm belonging to YangzhouUniversity, Jiangsu Province, China (328 309 N, 1198 259 E) duringthe rice growing season (May to October) of 2002, and repeated in2003. Two rice genotypes, YD-4 (Yangdao 4, an indica inbredcultivar) and LY-9 (Pei-ai 64S/Yangdao 7, an indica/indica F1hybrid), were grown in the paddy field. Seedlings were raised inthe field and were sown on 10 May and transplanted on 11 June ata hill spacing of 0.230.2 m with two seedlings per hill. The plotdimension was 638 m. Each of the genotypes had four plots asrepetitions in a complete randomized block design. The soil in thefield was sandy loam (Typic fluvaquents, Entisols; US classifica-tion) that contained organic matter at 2.45% and available N-P-K at108, 34.2, and 66.9 mg kg�1, respectively. N (60 kg ha�1 as urea),P (30 kg ha�1 as single superphosphate), and K (40 kg ha�1 as KCl)were applied and incorporated before transplanting. N as urea wasalso applied at mid-tillering (40 kg ha�1) and at panicle initiation(50 kg ha�1). Both genotypes (50% of plants) headed on 21–23August, and were harvested on 20 October. Except for drainage at theend of tillering (11–15 July), the water level in the field was kept at1–2 cm during the whole growth period. The temperatures, averagedper 10 d from anthesis (21–23 August) to harvest, were 27.4, 26.5,25.3, 24.1, 23.5, and 21.8 8C.

Sampling

Eight hundred panicles that headed on the same day were chosen andtagged for each plot, and for 50 of them the flowering date and theposition of each spikelet on the panicle were recorded. The durationof anthesis from the first spikelet to the last on a panicle was 7 d forboth genotypes. The spikelets that flowered on the same day weresampled as the same group and, accordingly, seven groups ofspikelets on a panicle were identified. The spikelets that floweredon the first day on a panicle were considered as the first group (D1),and those that flowered on the second day as the second group (D2),and so on. Each group of spikelets, sampled at 2-d intervals fromanthesis to 30 d post-anthesis (DPA) and at 6-d intervals from 30DPA to maturity, consisted of 280–300 spikelets (grains) at eachsampling time. Two-thirds of sampled grains were used for ABA,ethylene, and 1-aminocyclopropane-1-carboxylic acid (ACC) assays.Eighty to ninety sampled grains were used for measurements of graindry weight and soluble carbohydrate. These were dried at 70 8C toconstant weight, dehulled, and weighed. Soluble carbohydrate in thegrains was determined as described by Yoshida et al. (1976). Ten totwelve grains, with a small hole cut on the edge of the hull, were fixedin Carnoy’s solution (absolute ethanol:glacial acetic acid:chloroform;9:3:1; v/v/v) for 48 h, and then kept in 70% (v/v) ethanol pendingexamination of endosperm cell number.

Nuclear/cell counting

The method for isolation and counting of endosperm cells wasmodified from Singh and Jenner (1982). Briefly, fixed grains were

150 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


dehulled and transferred into 50% (v/v) and 25% (v/v) ethanol,respectively, and finally into distilled water for 5–7 h prior todissection of the endosperm. The endosperm was isolated undera dissecting microscope and dyed with a Delafied’s haematoxylinsolution for 24–30 h, washed several times with distilled water, andthen hydrolysed in 0.1% (w/v) cellulase (No. c-2415; SigmaChemical Co., St Louis, MO, USA) solution (pH 5.0) at 40 8C for4–6 h and oscillated. The isolated endosperm cells were diluted to2–10 ml according to the development stage of the endosperm,from which 8–10 subsamples (20 ll for each subsample) weretaken to a counting chamber (1 cm2 area). Using a light micro-scope, the endosperm cell number of 10 fields of view for eachcounting chamber was noted. Within 2–4 DPA, the number ofnuclei was counted as the endosperm cell number. The total cellnumber per endosperm was calculated according to Liang et al.(2001). Eight grains (endosperms) were examined for each geno-type at each measurement.The division processes of endosperm cells, as well as the processes

of grain filling, were fitted by Richards’ growth equation (Richards,1959) as described by Zhu et al. (1988):

MðWÞ= A

ð1 +Be�ktÞ1N

ð1Þ

Endosperm cell division rate or grain-filling rate (R) was calculatedas the derivative of equation (1)

R=AkBe

�kt

Nð1 +Be�ktÞðN + 1Þ

N

ð2Þ

where M is the cell number and W is the grain weight, A is themaximum cell number/grain weight, t is the time after anthesis (d),and B, k, and N are coefficients determined by regression. The activecell-division/grain-filling period was defined as that when M or Wwas from 5% (t1) to 95% (t2) of A. The average cell-division/grain-filling rate during this period was therefore calculated from t1 to t2.

ABA extraction, purification, and quantification

The methods for extraction and purification of abscisic acid [(+)-ABA] were modified from those described by Bollmark et al. (1988)and He (1993). Samples of 20–30 grains were ground in a mortar (at0 8C) in 10 ml 80% (v/v) methanol extraction medium containing1 mM butylated hydroxytoluence as an antioxidant. The extract wasincubated at 4 8C for 4 h and centrifuged at 4800 g for 15 min at thesame temperature. The supernatants were passed through ChromosepC18 columns (C18 Sep-Park Cartridge, Waters Corp, Millford, MA,USA), pre-washed with 10 ml 100% and 5 ml 80% methanol,respectively. The hormone fractions were dried under N2, anddissolved in 2 ml phosphate-buffered saline, containing 0.1% (v/v)Tween 20 and 0.1% (w/v) gelatin (pH 7.5), for analysis by anenzyme-linked immunosorbent assay (ELISA).The mouse monoclonal antigen and antibody against ABA, and Ig

G-HRP (immunoglobulin G–horseradish peroxidase) used in ELISAwere produced at the Phytohormones Research Institute, ChinaAgricultural University, China (He, 1993). The methods for quanti-fication of ABA by ELISA and recovery test were as describedpreviously (Yang et al., 2001, 2002b). The recovery percentage ofABA in grains was 82.565.6. The specificity of the monoclonalantibody and other possible non-specific immunoreactive interfer-ence had been checked previously and proved reliable (Wu et al.,1988; Yang et al., 2002b).

Ethylene and ACC analysis

Ethylene evolved from grains was determined according to Beltranoet al. (1994) with modifications. Briefly, sampled grains were placed

between two sheets of moist paper for 1 h at 27 8C in darkness toallow wound ethylene to subside. Each sample contained 60–80grains. Grains were then transferred into 15 ml glass vials containingmoist filter paper and immediately sealed with airtight subasealstoppers, and incubated in the dark for 24 h at 27 8C. A 1 ml gassample was withdrawn through the subaseal with a gas-tight syringe,and ethylene was assayed using a gas chromatograph (HP5890 SeriesII; Hewlett Packard Com, Palo Alto, CA, USA) equipped witha Porapak Q column (0.33200 cm, 50–80 mesh) and flame ionizationdetector. Temperatures for the injection port, column, and detectorwere kept constant at 140, 100, and 200 8C, respectively. Nitrogenwas used as the carrier at a flow rate of 30 ml min�1, and hydrogenand air were used for flame ionization detection at the rate of 30ml min�1 and 300 ml min�1, respectively. The rate of ethylene evolu-tion was expressed as a function of per grain.ACC in the grains was determined according to Chen and Lur

(1996). Ethylene evolved from ACC was assayed by using gaschromatography as described above. The transformation rates asa percentage from ACC to ethylene were 8664.2, 9265.4, and7865.1, respectively, at early, mid- and late grain-filling stages.

Chemical applications

Both the genotypes were used for chemical application. Plants weregrown in eight cement tanks in open-field conditions. Each tank (0.3cm high, 1.6 m wide, 8.8 m long) was filled with sandy loam soil withthe same nutrient content as the field experiment. Thirty-day-oldseedlings raised in the field were transplanted on 12 June into thetanks at a hill spacing of 0.1530.20 m with one seedling per hill.N (6 g m�2 as urea), P (3 g m�2 as single superphosphate), and K(3 g m�2 as KCl) were applied and incorporated before transplanting.N as urea was also applied at mid-tillering (3 g m�2) and atpanicle initiation (3 g m�2). The water level in the tank was kept at1–2 cm during the whole growth period.Synthetic (6)-ABA, ethephon (an ethylene-releasing agent), co-

baltous nitrate [Co(NO3)2, which inhibits ethylene synthesis byinhibiting ACC oxidase] (all from Sigma, St Louis, MO, USA) andfluridone (Fluka, Riedel-de Haen, Germany), an inhibitor of carot-enoid biosynthesis, which may indirectly also reduce ABA synthesis,were applied to the plants. Preparation of the chemical solutions wasas described elsewhere (Ober and Sharp, 1994; Chen and Lur, 1996).Starting at initiation of heading (panicles begin to appear out thesheath of the flag leaf), 20310�6 M (6)-ABA, 20310�6 M fluridone,50310�3 M ethephon, or 5310�5 M Co(NO3)2 was applied topanicles by using a writing brush which had been dipped in thesolutions. The chemicals were applied daily for 8 d at the rate of2–5 ml per panicle at each application. All the solutions containedethanol and Tween 20 at final concentrations of 0.1% (v/v) and0.01 (v/v), respectively. The same volume of deionized water con-taining the same concentrations of ethanol and Tween 20 wasapplied to the control plants. Each chemical treatment was on an areaof 2.4 m2 with four replications.For all the chemical treatments, spikelets on a panicle were divided

into two groups, i.e. the superior and the inferior. Superior spikeletswere those that flowered on the first two days and inferior ones werethose that flowered on the last two days on a panicle. Levels of ABAand ethylene in the grains were determined 9 and 16 DPA,respectively. Endosperm cell number was measured at 2-d intervalsfrom anthesis to 30 DPA, and grain weight at 6-d intervals fromanthesis to maturity. Measurement methods were the same as de-scribed above. Twenty plants (170–176 panicles) for each treatmentwere harvested at maturity to determine final grain weight.

Statistical analysis

The results were analysed for variance using SAS/STAT statisticalanalysis package (version 6.12, SAS Institute, Cary, NC, USA). Data

ABA and ethylene in rice spikelets 151

at Monash U

niversity on Decem


Dow

nloaded from


from each sampling date were analysed separately. Means were testedby least significant difference at the P 60.05 level (LSD0.05). Linearregression was used to evaluate the relationships between concen-trations of ABA and ethylene in the grains and the endosperm cell-division rate or grain-filling rate. As the data from both years werevery similar they were averaged.

Results

Cell division and grain filling

Figure 1 illustrates the division progression and divisionrate of endosperm cells for the spikelets that flowered ondifferent dates. The earlier the spikelets flowered, the earlierthe maximum cell division rate was reached, the shorter theactive cell division period, and the greater the cell divisionrate and final cell number. There were no significantdifferences in the cell numbers and cell division periodsbetween the spikelets that flowered on the first day (D1)and those on the second day (D2).The two genotypes be-haved the same (Fig. 1).

Very similar to the changing pattern of endosperm cellnumber, grain dry weight increased much faster for theearlier-flowering superior spikelets than for later-floweringinferior ones (Fig. 2). The earlier the spikelets flowered, thegreater the grain-filling rate and the heavier the grain. Asshown in Figs 1 and 2, the grain weight was closely relatedto the cell number, i.e. the more endosperm cells inspikelets, the higher the grain weight, suggesting thatendosperm cell number plays a dominant role in thedetermination of grain weight in rice.

By contrast to the cell-division rate and grain-filling rate,later-flowering spikelets contained a higher soluble carbo-hydrate concentration than earlier-flowering ones at the earlygrain-filling stage (Fig. 3a, b), indicating that assimilatesupply is not a factor limiting endosperm cell division andgrain filling in inferior spikelets. The changes in solublecarbohydrate were closely associated with those of sucrosein the spikelets (Fig. 3c, d), and they were significantlycorrelated (r=0.98**, P=0.01), suggesting that the highconcentration of soluble carbohydrate was mainly attributedto a high concentration of sucrose in the inferior spikelets.

Levels of ABA, ethylene, and ACC in spikelets

ABA levels were much higher, while ethylene and ACClevels were much lower, in superior spikelets than ininferior spikelets (expressed either as per unit weight oras per grain during the grain-filling period; Table 1).Correlation of hormone content (per grain) and hormoneconcentration (per unit weight) during grain filling wassignificant with r=0.91**, 0.93**, and 0.96** (n=30,P=0.01), respectively, for ABA, ethylene, and ACC. Asendosperm cell-division and grain-filling rates were ex-pressed on the basis of per grain, data presented as hormonecontent would be more comparable in the analysis of therelationship between hormones and grain development.

At the initial stage of grain filling, the ABA content ofspikelets was low, but it increased with the grain-fillingprocess, declining after reaching a maximum (Fig. 3a, b).The concentrations and the peak values of ABA varied with

Fig. 1. Cell numbers (a, c) and cell division rates (b, d) in the endosperm of rice spikelets. The indica cultivar YD-4 (a, b) and indica hybrid LY-9 (c, d)were field-grown. D1, D2, D3, D4, D5, D6, and D7 are the spikelets that flowered on the first, second, third, fourth, fifth, sixth, and seventh days,respectively, on a panicle. The division rate of endosperm cells was calculated according to Richards’ (1959) equation. Vertical bars in (a) and (c)represent 6standard error of the mean (n=8) where these exceed the size of the symbol.

152 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


the flowering dates of spikelets. The earlier the spikeletsflowered, the earlier they reached a maximum and thegreater the peak value of ABA.

In sharp contrast to ABA, ethylene evolution fromspikelets was very high at the early grain-filling stage, butrapidly decreased with the grain-filling processes (Fig. 4c,d). The later the spikelets flowered, the greater the ethylene

evolution rate at early and mid-grain-filling stages. A verysimilar pattern was observed for ACC content in thespikelets (Fig. 4e, f). There was a significant correlationbetween ACC content and the ethylene evolution rate(r=0.99**, P=0.001), suggesting that the high ethyl-ene evolution rate is attributed to the high ACC contentin inferior spikelets.

Fig. 3. Concentrations of soluble carbohydrate (a, b) and sucrose (c, d) in rice spikelets. The indica cultivar YD-4 (a, c) and indica hybrid LY-9 (b, d)were field-grown. D1, D2, D3, D4, D5, D6, and D7 are the spikelets that flowered on the first, second, third, fourth, fifth, sixth, and seventh days,respectively, on a panicle. Vertical bars represent 6standard error of the mean (n=4) where these exceed the size of the symbol.

Fig. 2. Grain weight (a, c) and grain-filling rates (b, d) of rice. The indica cultivar YD-4 (a, b) and indica hybrid LY-9 (c, d) were field-grown. D1, D2,D3, D4, D5, D6, and D7 are the spikelets that flowered on the first, second, third, fourth, fifth, sixth, and seventh days, respectively, on a panicle.The grain-filling rate was calculated according to Richards’ (1959) equation. Vertical bars in (a) and (c) represent 6standard error of the mean (n=4)where these exceed the size of the symbol.


at Monash U

niversity on Decem


Dow

nloaded from


As shown in Fig. 5, earlier-flowering spikelets alsoshowed a higher ratio of ABA to ACC than later-floweringspikelets. The earlier the spikelets flowered, the greater theratio (Fig. 5). The peak values of both ABA and the ratio ofABA to ACC were associated with the maximum cell-division rates. Regression analysis demonstrated that therewas a significant and positive correlation between themaximum rate of endosperm cell division and the average

ABA content and the ratio of ABA to ACC (r=0.89** andr=0.94**, respectively, P=0.01), whereas the correlationbetween the maximum rate of endosperm cell division,during the active cell division period or the phase of linearendosperm cell increase, and the ethylene evolution rate wassignificant and negative (r=–0.95**, P=0.01) (Fig. 6a–c).

During active grain filling or linear growth period,changes in ABA content and the ratio of ABA to ACC in

Table 1. Levels of ABA, ethylene, and ACC expressed on the basis of dry weight and per grain in the superior and inferior spikeletsduring grain filling of rice

Superior spikelets are those that flowered on the first two days and inferior ones are those that flowered on the last two days on a panicle. Each valuerepresents the mean of the 15 measurements taken at 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, and 30 DPA6standard deviation of four replicationsfor each measurement.

Cultivar Groups ofspikelets

ABA Ethylene ACC

(pmol grain�1) (pmol g�1 DW) (pmol grain�1 h�1) (pmol g�1 DW h�1) (pmol grain�1) (pmol g�1 DW)

YD-4 Superior 10.660.57 735656.8 0.1260.01 8.5360.76 4.6460.32 343625.8Inferior 5.1460.42 365627.3 0.5160.03 60.264.15 20.861.14 24356137

LY-9 Superior 7.6860.51 521641.6 0.1460.01 8.3260.64 5.7460.43 431632.9Inferior 3.9760.28 269618.5 0.5360.04 76.166.18 21.661.35 30976242

Fig. 4. Changes in levels of ABA (a, b), ethylene (c, d), ACC (e, f) in rice spikelets. The indica cultivar YD-4 (a, c, e) and indica hybrid LY-9 (b, d, f)were field-grown. D1, D2, D3, D4, D5, D6, and D7 are the spikelets that flowered on the first, second, third, fourth, fifth, sixth, and seventh days,respectively, on a panicle. Vertical bars represent 6standard error of the mean (n=4) where these exceed the size of the symbol.

154 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


grains (Figs 4a, b, 5a, b) paralleled the grain-filling rate(Fig. 2a, b). Figure 7 shows that there was a significantand positive correlation between ABA content and ABA toACC ratios, whereas the correlation between ethyleneevolution rates and grain-filling rates was significant andinverse (r=0.94**, r=0.82**, and r=�0.55**, respectively,P=0.01) (Fig. 7a–c).

Effects of chemical applications

Application of Co(NO3)2, an inhibitor of ethylene synthe-sis, significantly reduced the ethylene evolution rate andconcentration of soluble sugars in inferior spikelets, andincreased the cell-division rate, maximum cell number,grain-filling rate and grain weight (Tables 2, 3). Applicationof ethephon, an ethylene-releasing substance, exhibited theopposite effects. Application of fluridone, an indirectinhibitor of ABA synthesis, reduced ABA content butincreased the ethylene evolution rate of both superior andinferior spikelets (Table 2). It significantly reduced the cell-division rate, maximum cell number, grain-filling rate, andgrain weight of both types of spikelets (Table 3). By

contrast to fluridone, application of ABA significantlyincreased ABA and reduced ethylene levels in inferiorgrains, and the cell division rate, maximum cell number,grain filling rate, and grain weight of these grains increasedsignificantly. All the chemicals affected the superiorspikelets less than the inferior ones (Tables 2, 3).

Discussion

It is generally presumed that constraints in assimilateavailability for inferior spikelets results in their poor grainfilling (Sikder and Gupta, 1976; Wang, 1981; Murty andMurty, 1982; Zhu et al., 1988; Cao et al., 1992). The

Fig. 5. Changes in the ratio of ABA to ACC in rice spikelets. Data arefrom Fig. 4a, b, e, f. The indica cultivar YD-4 (a) and indica hybrid LY-9(b) were field-grown. D1, D2, D3, D4, D5, D6, and D7 are the spikeletsthat flowered on the first, second, third, fourth, fifth, sixth, and seventhdays, respectively, on a panicle. Vertical bars represent 6standard errorof the mean (n=4) where these exceed the size of the symbol.

Fig. 6. Relationships of the mean division rates of endosperm cells withthe mean levels of ABA (a) and ethylene (b) and the ratio of ABA to ACC(c) in rice spikelets during the active cell division period. Data are fromFigs 1c, d, 4a–d, and 5a, b. The indica cultivar YD-4 (closed circles) andindica hybrid LY-9 (open circles) were field-grown. Correlationcoefficients (r) are calculated and asterisks (**) represent statisticalsignificance at P=0.01 (n=14).


at Monash U

niversity on Decem


Dow

nloaded from


present results showed, however, that later-flowering in-ferior spikelets contained much higher levels of solublesugars than earlier-flowering superior ones at the earlygrain-filling stage (Fig. 3a, b). The results demonstrate thatcarbohydrates are not the limiting factor, at least at theinitial grain-filling stage, to the development of inferiorspikelets.

The phenomenon that earlier-flowering superior spike-lets exert dominance over later-flowering inferior spikeletsis often explained as apical dominance or primigenicdominance (Bangerth, 1989). In the hypothesis of primi-genic dominance, it is suggested that indole-3-acetic acid(IAA) export of the earlier developed sink inhibits the IAA

export of later developed sinks, and this depressed IAAexport of the subordinated fruit/sink acts as the signal thatleads to inhibited development (Bangerth, 1989). Earlierwork (Yang et al., 2000) has shown, however, that IAAconcentration in later-flowering inferior grains is notnecessarily lower than that in the earlier-flowering superiorgrains in rice, and the grain-filling rate is not significantlycorrelated with the IAA concentration. Duan et al. (1999)reported a similar observation. The data on the effect ofexogenous IAA on the grain-filling rate are rathercontroversial (Labrana et al., 1991; Patel and Mohapatra,1992; Zhou et al., 1999; Wang et al., 2001; Bhatia andSingh, 2002).The present data, for the first time, haveshown a close association between changes of ethyleneproduction and ABA content in grains and the rates ofendosperm cell division and grain filling (Figs 1, 2, 4).Later-flowering inferior spikelets had a much greaterethylene evolution rate and higher ACC content thanearlier-flowering superior spikelets (Fig. 4c–f). There wassignificant and negative correlation between ethyleneevolution rate and the cell-division and grain-filling rates(Figs 6b, 7b). Application of the inhibitor of ethylenesynthesis (cobalt ion) increased, and while applying ethylene-releasing agent (ethephon) reduced, the cell division-rate, grain-filling rate, and grain weight (Tables 2, 3). Thedata suggest that ethylene plays a role in inhibitingendosperm cell division and grain filling, and high ethyleneproduction in inferior spikelets contributes to their poordevelopment.

Little is known how ethylene affects cell division andgrain growth. It is suggested that ethylene may enhance thebreakdown of cytokinins (Bollmark and Eliasson, 1990),which play an important role in maintaining cell division inthe endosperm (Morris et al., 1993; Yang et al., 2002a).Naik and Mohapatra (2000) have reported that ethyleneinhibitors applied to rice panicles at booting stage couldsignificantly enhance the activities of sucrose synthase,which is generally regarded as a biochemical marker of sinkstrength (Wang et al., 1993), in the grains and promotegrain filling. Their work may suggest that ethylene limitssink strength by inhibiting the key enzymes involved.

It has been proposed that ethylene produced from thedominant basal spikelets of maize at the time of pollinationinhibits growth of the poorly developed kernels at the distalend of the inflorescence (Reed and Singletary, 1989). It wasobserved that ethylene and ACC levels in later-floweringinferior spikelets were much higher than those in earlier-flowering superior spikelets at the initial stage of grainfilling (Fig. 4c–f). Such a result implies that high ethyleneand ACC levels in inferior spikelets might be due tosynthesis per se. One explanation is that the upper leavesof rice produce a large amount of ethylene during theperiod of grain filling (Debata and Murty, 1983; Khanand Choudhury, 1992), and ethylene may exert more ofan inhibitory action on the inferior spikelets than on the

Fig. 7. Relationships of grain-filling rates with the levels of ABA (a) andethylene (b) and the ratio of ABA to ACC (c) in rice spikelets during theactive grain-filling period. Data are from Figs 2c, d, 4a–d, and 5a, b. Theindica cultivar YD-4 (closed circles) and indica hybrid LY-9 (opencircles) were field-grown. Correlation coefficients (r) are calculated andasterisks (**) represent statistical significance at P=0.01 (n=152).

156 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


superior ones, because the former are usually located onthe lower part of a panicle and remain confined to the enclos-ure formed by the flag leaf for a longer time than thelatter (Mohapatra et al., 2000). However, the causal roleof leaf-released ethylene in inhibiting inferior spikelets isdoubtful because, in wheat, leaves also evolve a largeamount ethylene during the grain filling period (Labranaet al., 1991; Beltrano et al., 1994), but proximal spikeletsusually produce heavier grains than the apical ones (Penget al., 1992; Jiang et al., 2003).

It was observed that the higher levels of ethylene andACC were closely associated with the lower ABA con-tent in inferior grains (Fig. 4a, b). The change in ABA

concentration in the grains followed a similar pattern to thegrain filling rate (Figs 2c, d, 4a, b). The cell-division rateand grain-filling rate were significantly and positively cor-related with the contents of ABA (Figs 6a, 7a). Applicationof ABA to panicles significantly increased the rates of celldivision and grain filling and grain weight of inferiorspikelets, whereas application of fluridone, an indirectinhibitor of ABA synthesis, had the opposite effect (Tables2, 3). From the above therefore it is speculated that,contrary to ethylene, ABA plays a role in enhancing grainfilling. It is possible that insufficient ABA to suppressethylene emission leads, at least partly, to the poor grainfilling of inferior spikelets.

Table 3. Effects of applied ethephon, cobatous nitrate [Co(NO3)2], ABA, and fluridone on soluble sugars (mg g�1DW) in grains,endosperm cell division rate (3103 cells per endosperm d�1), maximum endosperm cell number (3103 cells per endosperm), grain-filling rate (mg grain�1 d�1), and grain weight (mg grain�1) of rice

The plants were grown in cement tanks that were filled with soil. The panicles were applied with 50310�3 M ethephon, 5310�5 M Co(NO3)2, 20310�6

M ABA, or 20310�6 M fluridone daily for 8 d starting at the initiation of heading with four replications for each chemical treatment. Soluble sugars inthe grain were measured at 9 and 16 DPA. Endosperm cell-division rate and grain-filling rate were calculated according to Richards’ (1959) equation.Grain weight was determined at maturity. The data presented are the average between the two genotypes because they behaved the same. Letters after thevalues indicate statistical significance at the P=0.05 level within the same column and within the same type of spikelet.

Spikelet type and treatment Soluble sugars Cell-division rate Maximum cell number Grain-filling rate Grain weight

9 DPA 16 DPA

Superior spikeletsControl 18.4 b 8.52 bc 25.1 a 278 a 1.11 a 27.1 a50 3 10�3 M ethephon 18.6 b 9.21 b 24.5 a 272 a 1.05 a 26.6 a5 3 10�5 M Co(NO3)2 17.9 b 8.02 c 25.3 a 281 a 1.12 a 27.5 a20 3 10�6 M ABA 18.1 b 8.14 c 24.7 a 276 a 1.10 a 26.9 a20 3 10�6 M fluridone 32.4 a 14.8 a 19.6 b 224 b 0.88 b 21.4 bInferior spikeletsControl 208 b 112 c 13.2 b 190 c 0.59 b 18.8 c50 3 10�3 M ethephon 234 a 151 b 11.8 c 163 d 0.46 c 14.9 d5 3 10�5 M Co(NO3)2 176 d 79.6 e 14.9 a 225 a 0.64 a 21.5 a20 3 10�6 M ABA 189 c 91.5 d 14.7 a 210 b 0.63 a 20.0 b20 3 10�6 M fluridone 241 a 178 a 9.89 d 138 e 0.41 d 12.7 e

Table 2. Effects of applied ethephon, cobatous nitrate [Co(NO3)2], ABA, and fluridone on levels of ABA and ethylene in the grainsof rice

The plants were grown in cement tanks that were filled with soil. The panicles were applied with 50310�3 M ethephon, 5310�5 M Co(NO3)2, 20310�6

M ABA, or 20310�6 M fluridone daily for 8 d starting at the initiation of heading with four replications for each chemical treatment. Levels of ABA andethylene in grains were determined at 9 and 16 DPA, respectively. Control plants were supplied with deionized water. The data presented are the averagebetween the two genotypes because they behaved the same. Letters after the values indicate statistical significance at the P=0.05 level within the samecolumn and within the same type of spikelet.

Spikelet type and treatment 9 DPA 16 DPA

ABA (pmol grain�1) Ethylene (pmol grain�1 h�1) ABA (pmol grain�1) Ethylene (pmol grain�1 h�1)

Superior spikeletsControl 15.7 b 0.17 b 12.8 ab 0.09 b50 3 10�3 M ethephon 15.9 b 0.23 a 12.3 b 0.09 b5 3 10�5 M Co(NO3)2 15.2 b 0.11 c 12.7 ab 0.08 b20 3 10�6 M ABA 17.9 a 0.16 b 13.2 a 0.08 b20 3 10�6 M fluridone 10.8 c 0.24 a 8.71 c 0.13 aInferior spikeletsControl 1.42 b 0.67 c 6.93 b 0.45 b50 3 10�3 M ethephon 1.46 b 0.82 b 6.89 b 0.65 a5 3 10�5 M Co(NO3)2 1.39 b 0.41 e 6.41 b 0.31 d20 3 10�6 M ABA 6.78 a 0.54 d 10.5 a 0.38 c20 3 10�6 M fluridone 0.41 c 0.92 a 3.87 c 0.64 a


at Monash U

niversity on Decem


Dow

nloaded from


The mechanism by which ABA facilitates endospermcell division and grain filling is not understood. It has beenobserved that ABA regulates epidermal cell type-specificgene expression in the meristematic zone of Arabidopsis(van Hengel et al., 2004). Studies of several speciesindicate that an important role of endogenous ABA is tolimit ethylene production, and thereby functions to main-tain plant growth (Sharp et al., 2000; LeNoble et al., 2004).It is also proposed that ABA plays an important role inrelation to sugar-signalling pathways and enhances theability of plant tissues to respond to subsequent sugarsignals (Rook et al., 2001). There are many reports thatABA enhances the movement of photosynthetic assimilatestowards developing seeds (Dewdney and McWha, 1979;Ackerson, 1985; Brenner and Cheikh, 1995; Yang et al.,2004a). Earlier work (Yang et al., 2004b) has shown thatthe activities of key enzymes involved in sucrose-to-starchconversion, such as sucrose synthase and starch synthase,are enhanced in wheat grains by a mild water stressimposed during grain filling, and are highly correlatedwith elevated ABA levels there. Furthermore, the activitiesof the enzymes and remobilization of assimilates from thestem to the grains are significantly increased when ABAwas applied to plants at an early grain-filling stage. Theseresults suggest, by contrast to ethylene, that ABA maypromote grain filling through regulating sink strength.

The present results showed that the changes in the ratioof ABA to ACC in grains were also associated with theendosperm cell-division rate and the grain-filling rate (Figs1b, d, 2b, d, 5a, b). The rates of cell division and grainfilling not only correlated with the levels of ABA andethylene, but they also correlated with the ratio of ABA toACC (Figs 6c, 7c). Such results imply that antagonisticinteractions between ABA and ethylene may mediate celldivision and grain filling in rice.

In conclusion, earlier-flowering superior spikelets ex-hibit dominance over later-flowering inferior ones. Carbo-hydrates may not be the major limiting factor to thedevelopment of inferior spikelets. High ethylene productionand low ABA content in inferior spikelets result in slowendosperm cell division and poor grain filling, which leadto low grain weight. On the other hand, a high ABAconcentration in superior spikelets contributes to a fastgrain-filling rate. The rates of cell division and grain fillingare not only correlated with the levels of ABA and ethylene,but are also correlated with the ratio of ABA to ACC ingrains. Antagonistic interactions between ABA and ethyl-ene may be involved in mediating cell division and grainfilling. A higher ratio of ABA and ethylene in rice spikeletswould be required to enhance the grain-filling rate. Furtherinvestigation is needed to establish a possible cause-and-effect relationship between ABA/ethylene and grain de-velopment in rice using mutants or transgenic plants withan attenuated capacity to respond to, or synthesize, thehormones.

Acknowledgements

We are grateful for grants from the National Natural ScienceFoundation of China (Project No. 30270778, 30370828,BK2003041), the Research Grant Council of Hong Kong (RGC2149/04M, 2165/05M), and the Area of Excellence for Plant andFungal Biotechnology in the Chinese University of Hong Kong.

References

Ackerson RC. 1985. Invertase activity and abscisic acid in relation tocarbohydrate status in developing soybean reproductive structures.Crop Science 25, 615–618.

Ahmadi A, Baker DA. 1999. Effects of abscisic acid (ABA) ongrain filling processes in wheat. Plant Growth Regulation 28,187–197.

Bangerth F. 1989. Dominance among fruits/sinks and the search fora correlative signal. Physiologia Plantarum 76, 608–614.

Beltrano J, Carbone A,Montaldi ER, Guiamet JJ. 1994. Ethyleneas promoter of wheat grain maturation and ear senescence. PlantGrowth Regulation 15, 107–112.

Beltrano J, Ronco MG, Montaldi ER. 1999. Drought stresssyndrome in wheat is provoked ethylene evolution imbalanceand reversed by rewatering, aminoethoxyvinylglycine, or sodiumbenzoate. Journal of Plant Growth Regulation 18, 59–64.

Beruter J. 1983. Effect of abscisic acid on sorbitol uptake in growingapple fruits. Journal of Experimental Botany 34, 737–743.

Bhatia S, Singh R. 2002. Phytohormone-mediated transformation ofsugars to starch in relation to the activities of amylases, sucrose-metabolizing enzymes in sorghum grain. Plant Growth Regulation36, 97–104.

Bollmark M, Eliasson L. 1990. Ethylene accelerates the breakdownof cytokinin and thereby stimulates rooting in Norway sprucehypocotyl cuttings. Physiologia Plantarum 80, 534–540.

Bollmark M, Kubat B, Eliasson L. 1988. Variations in endogenouscytokinin content during adventitious root formation in peacuttings. Journal of Plant Physiology 132, 262–265.

Brenner ML, Cheikh N. 1995. The role of hormones in photosyn-thate partitioning and seed filling. In: Davies PJ, ed. Planthormones, physiology, biochemistry and molecular biology.Dordrecht: Kluwer Academic Publishers, 649–670.

Browning G. 1980. Endogenous cis, trans-abscisic acid and peaseed development: evidence for a role in seed growth fromchanges induced by temperature. Journal of Experimental Botany31, 185–197.

Cao X, Zhu Q, Yang J. 1992. Classification of source-sink types inrice varieties with corresponding cultivated ways. In: Xiong Z,Min S, eds. Prospects of rice farming for 2000. Hangzhou:Zhejiang Science & Technology Press, 361–372.

Chen CY, Lur HS. 1996. Ethylene may be involved in abortion ofthe maize caryopsis. Physiologia Plantarum 98, 245–252.

Debata A, Murty KS. 1983. Endogenous ethylene content in riceleaves during senescence. Indian Journal of Plant Physiology 26,425–427.

Dewdney SJ, McWha JA. 1979. Abscisic acid and the movementof photosynthetic assimilates towards developing wheat (Triticumaestivum L.) grains. Zeitschrift fur Pflanzenphysiologie 92,186–193.

Duan J, Tian C, Liang C, Huang Y, Liu H. 1999. Dynamic changesof endogenous plant hormones in rice grains in different parts ofpanicle at grain filling stage. Acta Botanica Sinica 41, 75–79.

Eeuwens CJ, Schwabe WW. 1975. Seed and pod wall developmentin Pisum sativum L. in relation to exacted and applied hormones.Journal of Experimental Botany 26, 1–14.

158 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


He Z. 1993. Method for an indirect enzyme-linked immunosorbentasssay. In: He ZP, ed. Guidance on experimental chemicalcontrol in crop plants. Beijing: Beijing Agricultural UniversityPublishers, 60–68.

Jiang D, Cao W, Dai T, Jing Q. 2003. Activities of key enzymes forstarch synthesis in relation to growth of superior and inferior grainson winter wheat (Triticum aestivum L.) spike. Plant GrowthRegulation 41, 247–257.

Kato T, Sakurai N, Kuraishi S. 1993. The changes of endogenousabscisic acid in developing grains of two rice cultivars withdifferent grain size. Japanese Journal of Crop Science 62,456–461.

Khan RI, Choudhury MA. 1992. Role of endogenous hormones inthe regulation of whole plant senescence in rice. Indian Journal ofExperimental Biology 30, 131–134.

Labrana X, Vendrell M, Araus JL. 1991. Ethylene production inwheat flag leaves and ears during grain filling. Plant Physiologyand Biochemistry 29, 349–354.

Lee BT, Martin P, Bangerth F. 1988. Phytohormone levels in theflorets of a single wheat spikelet during pre-anthesis developmentand relationships to grain set. Journal of Experimental Botany 39,927–933.

LeNoble ME, Spollen WG, Sharp RE. 2004. Maintenance of shootgrowth by endogenous ABA: genetic assessment of the involve-ment of ethylene suppression. Journal of Experimental Botany55, 237–245.

Liang J, Zhang J, Cao X. 2001. Grain sink strength may be relatedto the poor grain filling of indica-japonica rice (Oryza sativa)hybrids. Physiologia Plantarum 112, 470–477.

Mohapatra PK, Naik PK, Patel R. 2000. Ethylene inhibitorsimprove dry matter partitioning and development of late floweringspikelets on rice panicles. Australian Journal of Plant Physiology27, 311–323.

Mohapatra PK, Patel R, Sahu SK. 1993. Time of flowering affectsgrain quality and spikelet partitioning within the rice panicle.Australian Journal of Plant Physiology 20, 231–242.

Mohapatra PK, Sahu SK. 1991. Heterogeneity of primary branchdevelopment and spikelet survival in rice in relation to assimi-lates of primary branches. Journal of Experimental Botany 42,871–879.

Morgan JM. 1980. Possible role of abscisic acid in reducing seed setin water-stressed wheat plants. Nature 285, 655–657.

Morris RD, Blevins DG, Dietrich JT, et al. 1993. Cytokinins inplant pathogenic bacteria and developing cereal grains. AustralianJournal of Plant Physiology 20, 621–637.

Murata Y, Matsushima S. 1975. Rice. In: Evans LT, ed. Cropphysiology. Cambridge: Cambridge University Press, 75–99.

Murty PSS, Murty KS. 1982. Spikelet sterility in relation tonitrogen and carbohydrate contents in rice. Indian Journal ofPlant Physiology 25, 40–48.

Myers PN, Setter TL, Madson JT, Thompson JF. 1990. Abscisicacid inhibition of endosperm cell division in cultured maizekernels. Plant Physiology 94, 1330–1336.

Naik PK, Mohapatra PK. 1999. Ethylene inhibitors promote malegametophyte survival in rice. Plant Growth Regulation 28, 29–39.

Naik PK, Mohapatra PK. 2000. Ethylene inhibitors enhancedsucrose synthase activity and promoted grain filling of basal ricekernels. Australian Journal of Plant Physiology 27, 997–1008.

Ober ES, Setter TL, Madison JT, Thompson JF, Shapiro PS.1991. Influence of water deficit on maize endosperm development:enzyme activities and RNA transcripts of starch and zein synthesis,abscisic acid, and cell division. Plant Physiology 97, 154–164.

Ober ES, Sharp RE. 1994. Proline accumulation in maize (Zeamays L.) primary roots at low water potentials. I. Requirement forincreased levels of abscisic acid. Plant Physiology 105, 981–987.

Patel R, Mohapatra PK. 1992. Regulation of spikelet developmentin rice by hormones. Journal of Experimental Botany 43, 257–262.

Peng Y, GuoW, Yan L. 1992. Source–sink relationship in wheat andits regulation. In: Peng Y, ed. Wheat production and physiology.Nanjing: Dongnan University Publishers, 20–62.

Reed AJ, Singletary GW. 1989. Roles of carbohydrate supply andphytohormones in maize kernel abortion. Plant Physiology 91,986–992.

Richards FJ. 1959. A flexible growth function for empirical use.Journal of Experimental Botany 10, 290–300.

Rook F, Corke F, Card R, Munz G, Smith C, Bevan MW. 2001.Impaired sucrose-induction mutants reveal the modulation ofsugar-induced starch biosynthetic gene expression by abscisicacid signaling. The Plant Journal 26, 421–433.

Ross GS, McWha JA. 1990. The distribution of abscisic acid inPisum sativum plants during seed development. Journal of PlantPhysiology 136, 137–142.

Saini HS, Aspinall DA. 1982. Sterility in wheat (Triticum aestivumL.) induced by water deficit or high temperature: possiblemediation by abscisic acid. Australian Journal of Plant Physiology9, 529–537.

Schussler JR, Brenner ML, Brun WA. 1984. Abscisic acid andits relationship to seed filling in soybeans. Plant Physiology 76,301–306.

Schussler JR, Brenner ML, Brun WA. 1991. Relationship ofendogenous abscisic acid to sucrose level and seed growth rate ofsoybeans. Plant Physiology 96, 1308–1313.

Sharp RE, LeNoble ME, Else MA, Thorne ET, Gherardi F. 2000.Endogenous ABA maintains shoot growth in tomato independ-ently of effects on plant water balance: evidence for an interactionwith ethylene. Journal of Experimental Botany 51, 1575–1584.

Sikder HP, Gupta DKD. 1976. Physiology of grain in rice. IndianAgriculture 20, 133–141.

Singh BM, Jenner CF. 1982. A modified method for the de-termination of cell number in wheat endosperm. Plant ScienceLetters 26, 273–278.

Trewavas AJ, Jones HG. 1991. An assessment of the role of ABA inplant development. In: Davies WJ, Jones HG, eds. Abscisic acid:physiology and biochemistry. Oxford: Bios Scientific Publishers,169–188.

van Hengel AJ, Barber C, Roberts K. 2004. The expressionpatterns of arabinogalactan-protein AtAGP30 and GLABRA2reveal a role for abscisic acid in the early stages of root epidermalpatterning. The Plant Journal 39, 70–83.

Walton DC. 1980. Biochemistry and physiology of abscisic acid.Annual Review of Plant Physiology 31, 453–489.

Wang F, Sanz A, Brenner ML, Smith A. 1993. Sucrose synthase,starch accumulation, and tomato fruit sink strength. Plant Phys-iology 101, 321–327.

Wang TL, Cook SK, Francis RJ, Ambrose MJ, Hedley CL. 1987.An analysis of seed development in Pisum sativum. VI. Abscisicacid accumulation. Journal of Experimental Botany 38,1921–1932.

Wang Y. 1981. Effectiveness of supplied nitrogen at the primordialpanicle stage on rice characteristics and yields. International RiceResearch News Letter 6, 23–24.

Wang Z, Cao W, Dai T, Zhu Q. 2001. Effects of exogenoushormones on floret development and grain set in wheat. PlantGrowth Regulation 35, 225–231.

Wang Z, Yang J, Zhu Q, Zhang Z, Lang Y, Wang X. 1998.Reasons for poor grain filling in intersubspecific hybrid rice. ActaAgronomica Sinica 24, 782–787.

Wu S, Chen W, Zhou X. 1988. Enzyme linked immunosorbentassay for endogenous plant hormones. Plant Physiology Com-munication 5, 53–57.


at Monash U

niversity on Decem


Dow

nloaded from


Xu X, Vergara BS. 1986. Morphological changes in rice panicledevelopment: a review of literature. International Rice ResearchInstitute Research Paper Series 117, 1–16.

Xu ZZ, Yu ZW, Qi XH, Yu SL. 1995. Effect of soil drought onethylene evolution, polyamine accumulation and cell membrane inflag leaf of winter wheat. Acta Physiologia Sinica 21, 295–301.

Yang J, Peng S, Visperas RM, Sanico AL, Zhu Q, Gu S. 2000.Grain filling pattern and cytokinin content in the grains and roots ofrice plants. Plant Growth Regulation 30, 261–270.

Yang J, Zhang J, Huang Z, Wang Z, Zhu Q, Liu L. 2002a.Correlation of cytokinin levels in the endosperm and roots withcell number division activity during endosperm development inrice. Annals of Botany 90, 369–377.

Yang J, Zhang J, Huang Z, Wang Z, Zhu Q, Liu L. 2002b.Abscisic acid and cytokinins in the root exudates and leaves andtheir relationship to senescence and remobilization of carbonreserves in rice subjected to water stress during grain filling. Planta215, 645–652.

Yang J, Zhang J, Wang Z, Xu G, Zhu Q. 2004a. Activities ofkey enzymes in sucrose-to-starch conversion in wheat grains sub-

jected to water deficit during grain filling. Plant Physiology 135,1621–1629.

Yang J, Zhang J, Wang Z, Zhu Q, Liu L. 2004b. Activities offructan and sucrose-metabolizing enzymes in wheat stems sub-jected to water stress during grain filling. Planta 220, 331–343.

Yang J, Zhang J, Wang Z, Zhu Q, Wang W. 2001. Hormonalchanges in the grains of rice subjected to water stress during grainfilling. Plant Physiology 127, 315–323.

Yoshida S, Forno D, Cock J, Gomez K. 1976. Determination ofsugar and starch in plant tissue. In: Yoshida S, ed. Laboratorymanual for physiological studies of rice. Los Banos, ThePhilippines: International Rice Research Institute, 46–49.

Zhou XM, MacKenzie AF, Madramootoo CA, Smith DL. 1999.Effects of stem-injected plant growth regulators, with or withoutsucrose, on grain production, biomass and photosynthetic activityof field-grown corn plants. Journal of Agronomy and Crop Science183, 103–110.

Zhu Q, Cao X, Luo Y. 1988. Growth analysis in the process of grainfilling in rice [in Chinese with English abstract]. Acta AgronomicaSinica 14, 182–192.

160 Yang et al.

at Monash U

niversity on Decem


Dow

nloaded from


post-anthesis development of inferior and superior spikelets in rice in relation to abscisic acid...

Documents