Q

Ka

b

c

a

ARR2A

KRGCQA

1

aao2acoc2

no

nEO

NT

hk

0d

Environmental and Experimental Botany 88 (2013) 100– 106

Contents lists available at SciVerse ScienceDirect

Environmental and Experimental Botany

j o ur nal homep age : www.elsev ier .com/ locate /envexpbot

uantitative trait locus analyses of ozone-induced grain yield reduction in rice

eita Tsukaharaa, Hiroko Sawadab,c, Hideyuki Matsumurab, Yoshihisa Kohnob, Masanori Tamaokia,c,∗

Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, JapanCentral Research Institute of Electric Power Industry, Abiko, Chiba 270-1194, JapanNational Institute of Environmental Studies, Tsukuba, Ibaraki 305-8506, Japan

r t i c l e i n f o

rticle history:eceived 12 September 2011eceived in revised form8 November 2011ccepted 1 December 2011

eywords:

a b s t r a c t

Reduction of grain yield (total seed weight) by ozone in rice (Oryza sativa L.) is believed to be causedby ozone-induced reduction of photosynthetic activity followed by growth inhibition. Here, japonica ricecultivar Sasanishiki and indica rice cultivar Habataki showed different responses to ozone. When exposedto ozone, the leaves of Habataki exhibited no critical damage, whereas those of Sasanishiki developedlesions. Conversely, ozone exposure reduced total seed weight by 19% in Habataki, but not significantlyin Sasanishiki. Chronic ozone exposure also significantly decreased culm length, number of primary

icerain yield (total seed weight)hronic ozone exposureTL analysisPO1

rachis branch, and number of spikelets per panicle in Habataki. QTL analysis in Sasanishiki/Habatakichromosome segment substitution lines identified a single locus associated with the yield loss caused byozone on chromosome 6 of Habataki close to marker RM3430 (107.6 cM). A QTL for reduction of primaryrachis branch number and total spikelet number was found in the same position. These results indicatethat a QTL on chromosome 6 has an important role in ozone-induced yield loss, and is also involved inprimary rachis branch formation and total spikelet number in ozone-exposed rice.

© 2011 Elsevier B.V. All rights reserved.

. Introduction

Ozone is the main photochemical oxidant that causes leaf dam-ge, and exposure to it can decrease the productivity of cropsnd forests. Ashmore et al. (2006) predicted that the troposphericzone concentration will continue to increase in Eastern Asia to020, and that crop yield may decrease by up to 60% in Chinas a result. In Japan, the tropospheric ozone concentration hasontinued to increase, although the concentrations of nitrogenxide and non-methane hydrocarbons, which are the major pre-ursors of ozone, have decreased over the past 20 years (Ohara,007).

Generally, acute exposure to ozone can result in chlorosis andecrosis. Thus, many studies have focused on the mechanismsf ozone-induced leaf injury as a means of understanding the

Abbreviations: QTL, quantitative trait loci; ROS, reactive oxygen species; NA,atural air; LBS, leaf bronzing score; CSSL, chromosome segment substitution line;M, expectation-maximization; LOD, likelihood-of-odds; APO1, ABERRANT PANICLERGANIZATION 1.∗ Corresponding author at: Center for Environmental Biology and Ecosystem,ational Institute of Environmental Studies, Tsukuba, Ibaraki 305-8506, Japan.el.: +81 29 850 2466; fax: +81 29 850 2490.

E-mail addresses: tsukahara@ies.life.tsukuba.ac.jp (K. Tsukahara),i ro s216@yahoo.co.jp (H. Sawada), matz@criepi.denken.or.jp (H. Matsumura),ohno@criepi.denken.or.jp (Y. Kohno), mtamaoki@nies.go.jp (M. Tamaoki).

098-8472/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.envexpbot.2011.12.012

effects of ozone on plants. Consequently, the physiological andmolecular aspects of visible leaf injury by ozone are well under-stood. When ozone enters the plant leaf through the stomata,it is rapidly reacts with components of the cell wall, plasmamembrane and apoplastic fluids (Kangasjärvi et al., 1994). In theapoplast, ozone is decomposed to reactive oxygen species (ROS)such as the hydrogen peroxide (H2O2), superoxide anion (O2

•−),and singlet oxygen (Rao and Davis, 2001; Baier et al., 2005). Fur-thermore, these ROS induce the production of further ROS byplant itself, designated as the oxidative burst (Wohlgemuth et al.,2002). As a consequence, visible ozone damage of leaves can occurdue to (i) direct necrotic tissue damage caused by ROS or (ii)ROS-induced programmed cell death that resemble the hyper-sensitive response observed in pathogen infection (Kangasjärviet al., 2005). Long-term exposure to relatively low concentrationof ozone (chronic ozone exposure) accelerates senescence of plantcell, which leads increase oxidative stress in chloroplasts, anddegradation of ribrose-1,5-bisphosphate carboxylase/oxigenase(Pell et al., 1997). Chronic ozone exposure, followed by distur-bances in sugar metabolism, inhibition of photosynthesis, andimbalances in the redox states results in reduction in plantgrowth and/or crop yield (Schraudner et al., 1997). The ozone

sensitivity, evaluated as visible leaf injury, of many rice culti-vars has been assessed (Sohn and Lee, 1997; Frei et al., 2008),and the mechanism of leaf damage by ozone has been studied(Lin et al., 2001; Frei et al., 2008). The degree of visible leaf

d Experimental Botany 88 (2013) 100– 106 101

iigbor

t2egmotr(iasumuctil2elTdl

2

2

poatoJ2TntPwpageom1ct2uF1eu

K. Tsukahara et al. / Environmental an

njury did not correlate with the grain yield reduction by ozonen 20 rice cultivars (Sawada and Kohno, 2009). This result sug-ests that the leaf injury and grain yield loss may be regulatedy different mechanisms. However, the mechanisms underlyingzone-induced grain yield loss have not been investigated inice.

In recent studies, genes involved in rice yield have been iden-ified by quantitative trait locus (QTL) analysis (Ashikari et al.,005; Fan et al., 2006; Song et al., 2007; Xue et al., 2008). Forxample, the expression level of cytokinin oxidase/dehydrogenaseene (OsCKXs), encoding an enzyme that degrades the phytohor-one cytokinin, in inflorescence meristems regulates the number

f reproductive organs and grain yield (Ashikari et al., 2005). Fur-hermore, at least eleven putative CKX genes have been found inice, but only the OsCKX2 gene has important role for grain yieldAshikari et al., 2005). Despite the identification of several genesnvolved in rice grain yields, no research has clarified the mech-nism of ozone-induced yield loss in rice. The objective of thistudy was to identify the gene loci determining rice yield lossnder high-ozone conditions by using Sasanishiki/Habataki chro-osome segment substitution lines (CSSLs). Toward this end, we

sed 39 chromosome segment substitution lines (CSSLs) from aross between Sasanishiki, as the recurrent parent and Habataki, ashe donor (Ando et al., 2008). Previous study showed that ozone-nduced grain yield loss in Sasanishiki, a parent of CSSLs, wasess than that in an indica variety, Takanari (Sawada and Kohno,009). In addition, Takanari is sibling variety of Habataki (Imbet al., 2004), indicating that ozone responses, including grain yieldoss, are expected to be same between Takanari and Habataki.herefore, we use Sasanishiki/Habataki CSSLs, in this study, toetermine the gene loci that affect ozone-induced grain yield

oss.

. Materials and methods

.1. Plant material and growth conditions

To detect QTLs for ozone resistance, we used a mappingopulation consisting of 39 CSSLs developed from backcrossesf rice (Oryza sativa L.) cultivars Sasanishiki (recurrent parent)nd Habataki (donor parent) (Ando et al., 2008). Seeds of thewo parents and the 39 CSSLs, designated SL401–SL439, werebtained from the Rice Genome Resource Center (RGRC, Tsukuba,apan). Seeds were sown in plastic boxes (80 plants per box;8 cm × 21 cm × 9 cm) filled with seedbed soil on 10 April 2009.he seedlings were first grown for 6 weeks in a glasshouse inatural (ambient) air at the Akagi Testing Center of the Cen-ral Research Institute of the Electric Power Industry (Gunmarefecture, Japan, 36◦28′N, 139◦11′E, 540 m above sea level). Sixeeks after sowing (21 May 2009), the seedlings were trans-lanted into pots (four plants per pot; 0.05 m2 surface areand 0.015 m3 volume) and grown in glasshouses (five pots perlasshouse for each line) under conditions of natural air (NA) orlevated ozone (×2). For the ×2 treatment, artificially generatedzone was added to natural air via a mass flow controller. Theean ozone concentration during the daytime (from 06:00 to

8:00) was 32.0 nl l−1 (ppb) under NA and 76.5 nl l−1 under ×2ondition, with a daily peak of 43.1 nl l−1 and 91.0 nl l−1, respec-ively (Fig. 1). Fertilizer was supplied at N:P:K = 15:15:15 g m−2 at

weeks after transplanting. Plants were grown in a glasshousender natural light until harvest (24 or 25 September 2009).

or total RNA isolation, plants were cut off at the base at0 days before heading, and young panicles (about 5 cm long)nclosed by a leaf sheath were isolated and frozen at −80 ◦C untilse.

Fig. 1. Daily ozone exposure in glasshouse. Values represent mean ozone concen-tration (ppb) each hour of the day averaged from 21 May to 25 September in 2009.NA, natural air condition; ×2, high-ozone condition.

2.2. Measurement of visible leaf injury

At 19 days after transplanting, ozone-induced visible leaf injurywas quantified by assigning a leaf bronzing score (LBS) to the fourthleaf from the base of five plants per treatment. LBS was definedusing the following scale: 0, no visible damage; 1, very few smallchlorotic spots; 2, very few small brown stipples; 3, 10%–20% ofleaf area with chlorotic or brown stipples; 4, 20%–40% of leaf areawith chlorotic or brown stipples; 5, >40% of leaf area with brownlesions; 6, >40% of leaf area with brown lesions and large necroses;7, entire leaf dying (Sawada and Kohno, 2009; Fig. 2a).

2.3. Measurement of gas exchange

On 29 July and 4 August 2009, 3 pots of samples of the twoparental cultivars, Sasanishiki and Habataki, were selected fromeach glasshouse, and gas exchange of two flag leaves per pot wasmeasured with a Li-Cor LI-6400P Portable Photosynthesis Sys-tem with a 3 cm × 2 cm (6 cm2) leaf chamber. Illumination wasprovided by an attached LI-6400-02B LED light (Li-Cor, Inc., Lin-coln, NE, USA); the photosynthetic photon flux density in the leafchamber was 2000 ± 1 �mol m−2 s−1. Temperature, CO2, and watervapor concentration in the leaf chamber were set to 25.2 ± 0.2 ◦C,380 ± 0.5 �mol mol−1, and 22 ± 0.5 �mol mol−1, respectively.

2.4. Measurements of yield and harvest indexes

During cultivation, the day of heading of each plant wasrecorded. Plants were cut off at the base of the culm between 24 and25 September 2009, and the length of the longest leaf blade, culmlength, culm number, and biomass were measured. Panicles werecut off at the neck of the spike, and the panicle length and number ofprimary rachis branch were measured. Then, grains were removedfrom the panicles and sorted into filled and unfilled grains. Thetotal spikelet number, number of spikelets per panicle, number offilled grains, filling rate, total seed weight, and 1000-seed weightwere measured. The total seed weight was calculated from filledgrains only, and was used as the measure of total seed weight inthis report.

2.5. Identification of putative gene loci by QTL analysis

We performed linkage analysis by interval mapping (Lander

and Botstein, 1989) as implemented in the program R/qtl (Bromanet al., 2003), using the expectation-maximization (EM) algo-rithm (Haley and Knott, 1992). The genotype of each CSSLhad been determined previously by using 166 DNA markers

102 K. Tsukahara et al. / Environmental and Experimental Botany 88 (2013) 100– 106

F ys aftb nishikc

da(bta3ct

2

hcUoaUwt(cta3ffiCctw

3

3

w(vdvtci

ig. 2. Leaf lesions in Sasanishiki and Habataki under NA and ×2 conditions 19 daronzing score; LBS). We measured LBS used this standard as index. (b) LBS in Sasaondition; ×2, high-ozone condition.

istributed along the 12 rice chromosomes (Ando et al., 2008),nd the mapping data were obtained from the RGRC web sitehttp://www.rgrc.dna.affrc.go.jp/ineSHCSSL39.html). The recom-ination fractions were converted to centimorgans (cM) by usinghe Haldane mapping function (Haldane, 1919). Putative QTLs werelso detected using R/qtl. A likelihood-of-odds (LOD) threshold of.0 was used for declaring the significance of a putative QTL. Welassified a peak as representing a putative QTL if LOD was >3 underhe NA condition and <3 under the ×2 condition.

.6. Quantitative real-time PCR analysis

Total RNA was isolated from young panicles (5 cm long, beforeeading) of Sasanishiki and Habataki grown under NA and ×2onditions with an RNeasy Plant Mini Kit (Qiagen, Valencia, CA,SA) following the manufacturer’s instructions. A 1 �g aliquotf total RNA was reverse-transcribed using random hexamersnd Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA,SA) in a 20 �l reaction volume, and 2 �l of the reaction mixtureas subsequently used for quantitative real-time PCR. Quantita-

ive real-time PCR was conducted in a LightCycler 480 SystemRoche Applied Science, Mannheim, Germany) and using LightCy-ler 480 SYBR Green I Master reaction mix (Roche) according tohe manufacturer’s specifications. A fragment of APO1 cDNA wasmplified with the PCR primers 5′-CAGGTAAGGGCTCCGTTGGA-′ and 5′-TGCGTAGCATGTTTTGCAGT-3′. As an internal standardor cDNA amounts, a fragment of �-tubulin cDNA was ampli-ed with PCR primers 5′-CATCGACATCAAGTTCGA-3′ and 5′-CGAGTTCGACGATGGTGA-3′. The expression level of APO1 wasalculated by dividing the value of the APO1 signal by that of theubulin signal. Three independent biological samples were usedith the gene-specific primers.

. Results

.1. Ozone-induced leaf lesions occurred only in Sasanishiki

The average LBS in the fourth leaves of cultivar Sasanishikias 1.4 ± 0.55 under the NA condition and three times that level

4.2 ± 0.84) under the ×2 condition (Fig. 2b). On the other hand, noisible injury was detected in leaves of Habataki under either con-ition. All 39 CSSLs exhibited ozone-induced leaf injury, with LBS

alues between 1.6 and 5.6 (data not shown). Under the NA condi-ion, the net photosynthetic rate, transpiration rate, and stomatalonductance were significantly higher in Habataki than in Sasan-shiki (Fig. 3a–c). Although leaf physiological traits showed varietal

er transplanting. (a) Standards for determining the level of visible leaf injury (leafi and Habataki. Values are mean ± SD (n = 5). Error bars indicate SD. NA, natural air

differences, no significant change in these traits was detected underthe ×2 condition (Fig. 3a–c). Taken together, these results indi-cate that indica-type Habataki showed no visible leaf injury andno decrease of photosynthesis when exposed to ×2 ozone, eventhough its stomatal conductance was high under both conditions.

3.2. Ozone exposure reduced total seed weight only in Habataki

To determine traits that potentially affect ozone-induced totalseed weight loss, we investigated changes in vegetative and repro-ductive traits caused by ozone exposure in the parents. Ozonereduced total seed weight by 19% (P = 0.038) in Habataki relativeto NA-grown control plants (Fig. 4b), although no leaf injury wasdetected during the vegetative stage. It reduced total seed weightby 8% (not significant) in Sasanishiki, although it caused leaf injury.Among the vegetative traits, the length of the longest leaf bladewas reduced significantly by 3.2% in Sasanishiki and by 7.4% inHabataki (Fig. 4a). However, culm length was inhibited significantlyonly in Habataki (by 7.5%). Floral traits also showed specific reduc-tion of growth by ozone in Habataki—number of primary rachisbranch (to 83.9% of NA control) and number of spikelets per pani-cle (to 81.6%)—but not in Sasanishiki (Fig. 4b). Ozone significantlydecreased filling rate in both cultivars, but there were no significantdifferences in the other floral traits.

3.3. Putative gene loci associated with yield loss by chronic ozoneexposure

Phenotypic evaluation showed that ozone reduced total seedweight, culm length, number of primary rachis branch, and numberof spikelets per panicle in Habataki. Therefore, we performed QTLanalyses for these traits in the CSSLs. In addition, we also performeda QTL analysis for a reproductive trait, total spikelet number, whichwas more highly reduced by ozone in Habataki (to 81.6% of control;P = 0.058) than in Sasanishiki (to 95.3% of control; P = 0.266). QTLanalysis detected two QTLs for total seed weight under the NA con-dition (Fig. 5a). These QTLs were located on chromosome 6 and 12,but the QTL on chromosome 6 was no longer significant under the×2 condition (Fig. 5b). These results make the QTL on chromosome6 a possible candidate in relation to ozone-induced yield loss. Nosignificant QTL was detected in the analysis of culm length or num-

ber of spikelets per panicle. In the QTL analysis of primary rachisbranch number and total spikelet number, a single QTL peak wasfound on chromosome 6 under the NA condition (Fig. 5c and e),but the peak was no longer significant under ozone (Fig. 5d and f).

K. Tsukahara et al. / Environmental and Experimental Botany 88 (2013) 100– 106 103

Fig. 3. Effects of ozone on physiological traits. Plants were grown under NA and ×2 conditions for about 4 months, and flag leaves were used for measurements of (a) netp are mc

If

3o

sopuatsALsalwalwrfsoS

Fc*

hotosynthetic rate, (b) transpiration rate, and (c) stomatal conductance. Values

ondition.

nterestingly, that QTL was at nearly the same location as the QTLor total seed weight.

.4. Accumulation of APO1 mRNA in Habataki is suppressed byzone

To narrow down the location of the QTL for ozone-induced totaleed weight loss, we compared the LOD scores of the QTL peaksn end of chromosome 6 for total seed weight, the number of therimary rachis branches, and total seed number in plants grownnder the NA condition by following reason. The gene locus thatffects ozone-induced total seed weight loss may found at positionhat has positive effects on harvest indexes under NA condition anduch effects is significantly suppressed in ozone-exposed condition.t the QTL marker for RM3430 (107.6 cM) on chromosome 6, theOD scores ranged narrowly from 3.39 to 3.86. In contrast, the LODcores for number of primary rachis branch at markers RM1370nd RM5436 were below 2 (Fig. 6). These results suggest that geneoci near RM3430 on chromosome 6 have key roles in total seed

eight loss by decreasing the number of primary rachis branchnd total spikelet number under high-ozone conditions. RM3430ies close to the ABERRANT PANICLE ORGANIZATION 1 (APO1) gene,

hich encodes an F-box protein (Ikeda et al., 2007) shown to affectice total seed weight through regulation of primary rachis branch

ormation (Terao et al., 2010). Thus, we analyzed APO1 expres-ion in young panicles. The APO1 transcript level in young paniclesf Habataki grown under the NA condition was 15 times that ofasanishiki (Fig. 7). APO1 expression in ozone-exposed Habataki

ig. 4. Comparison of vegetative and floral traits between Sasanishiki and Habataki. Spidondition compared with plants grown under the NA condition. Each data point represents*P < 0.01 versus NA condition. Asterisks outside the lines represent significant difference

ean ± SD (n = 6). Error bars indicate SD. NA, natural air condition; ×2, high-ozone

decreased by about 50% relative to the NA control; conversely, thatin ozone-exposed Sasanishiki increased to about 6.5 times.

4. Discussion

The high-yielding indica cultivar Habataki yields more than theaverage-yielding japonica cultivar Sasanishiki because it sets moreflowers (spikelets) per panicle and has a higher photosynthetic ratethat might contribute to a higher growth rate (Kobayashi et al.,1990; Asanuma et al., 2008). Here, Habataki showed these samecharacteristics under the NA condition (Figs. 3a and 4b). Habatakialso had no visible leaf injury after growth under the NA condition(Fig. 2b).

Ozone exposure reduced total seed weight by 19% in Habataki,but not significantly in Sasanishiki (Fig. 4b). In contrast, chronicozone exposure caused significant damage in Sasanishiki leaves,but no visible injury in Habataki (Fig. 2b). In addition, ozone expo-sure had no discernible impact on physiological traits, such as netphotosynthetic rate, transpiration rate, and stomatal conductance,in either Sasanishiki or Habataki, despite ozone-independent vari-etal differences in these traits (Fig. 3). Taken together, these resultsindicate that the ozone-induced total seed weight loss was not asso-ciated with the generation of leaf lesions. This confirms previousresults that other rice cultivars (e.g., Kasalath and Takanari) alsoshowed a lack of association between the appearance of leaf injury

and total seed weight loss (Sawada and Kohno, 2009).

In the QTL analysis for total seed weight, we found two peakswith an LOD score of >3 under the NA condition, one close tomarker RM3430 on chromosome 6, and the other on chromosome

er plots of Sasanishiki (solid line) and Habataki (dashed line) grown under the ×2 a percentage of the mean (n = 20). (a) Vegetative traits; (b) floral traits. *P < 0.05 and

s in Sasanishiki; those inside the lines represent significant differences in Habataki.

104 K. Tsukahara et al. / Environmental and Experimental Botany 88 (2013) 100– 106

Fig. 5. Genome scans for quantitative trait loci (QTLs) produced by composite interval mapping. The vertical lines separate the genome into chromosomes 1–12 (labeled atthe bottom) with position in cM progressing left to right along the x-axis. Line graphs of likelihood-of-odds (LOD) scores for (a and b) total seed weight, (c and d) numbero tion arh

1thI6wMggsln(Qlartts

f primary rachis branch, and (e and f) total spikelet number. Graphs for NA condiorizontal rules show LOD = 3.

2. Although both QTLs had significant effects on total seed weight,he specific effect of each was different. The QTL on chromosome 6ad a positive effect on total seed weight under the NA condition.

ndeed, the CSSL that contained a Habataki region of chromosome including RM3430 (line SL421) showed the highest total seedeight among the CSSLs (1.9 times the average yield of the CSSLs).oreover, the QTL on chromosome 6 for total seed weight in plants

rown under the NA condition was no longer significant in plantsrown under the ×2 condition. Thus, the QTL found on chromo-ome 6 appears to be involved in ozone-induced total seed weightoss. In contrast, the QTL found on chromosome 12 may have aegative effect on total seed weight. Among the CSSLs, two linesSL438 and SL439) contained the Habataki allele of the candidateTL on chromosome 12, and their total seed weights were markedly

ower than those of the other lines. Total seed weight in SL438nd SL439 grown under the NA condition was 28.7% and 46.8%,

espectively, of the average total seed weight of the CSSLs, andhis tendency was unchanged in plants grown under the ×2 condi-ion (29.7% and 45.6%, respectively). Such negative effects on totaleed weight might result from the extreme reduction of filling rate

e on the left (a, c and e); those for ×2 condition are on the right (b, d and f). Solid

under the NA condition (31.2% in SL438 and 47.5% in SL439), whichwas unchanged under ozone (data not shown). Thus, the only QTLrelated to total seed weight loss caused by ozone was located onchromosome 6.

In the process of narrowing down the QTL for total seed weightloss caused by ozone, we found that gene loci near RM3430 onchromosome 6 might have key roles via decreases in the number ofprimary rachis branch and total spikelet number under high-ozoneconditions. RM3430 was located close to APO1, which encodes anF-box protein (Ikeda et al., 2007). APO1 was initially identified asa gene that suppresses precocious conversion of rachides branchmeristem to spikelets, ensuring the generation of only a certainnumber of spikelets (Ikeda et al., 2007). A loss-of-function mutantof APO1, apo1, leads to reduction of rice total seed weight througha decrease in number of primary rachis branch (Ikeda-Kawakatsuet al., 2009). Transgenic plants in which APO1 expression was sup-

pressed by the RNAi method also had significantly reduced primaryrachis branch number (76% of control plant) and number of grainsper panicle (82%) (Terao et al., 2010), suggesting that APO1 affectsrice total seed weight through regulation of primary rachis branch

K. Tsukahara et al. / Environmental and Exp

Fig. 6. LOD (likelihood-of-odds) scores for total seed weight (grain yield), number ofprimary rachis branch, and total spikelet number between markers RM6395 (99.2cM) and RM5436 (124.4 cM) on chromosome 6. A schematic of chromosome 6 isshown at the top. The filled box represents the region of chromosome 6 from RM6395to RM5436. The vertical lines in the graph indicate the position of each marker shownat the bottom of the figure.

Fig. 7. Accumulation of APO1 transcript in young panicle. Plants were grown underNA and ×2 conditions, and young panicles (about 5 cm long) were collected for totalRNA isolation. The values obtained for APO1 transcripts were normalized againstti

fshtctttw

K., Uehara, Y., Ishizaka, S., Nakagahra, M., Yamada, T., Koga, Y., 2004. Develop-

he transcription levels of the tubulin gene. Values are mean ± SD (n = 3). Error barsndicate SD. NA, natural air condition; ×2, high-ozone condition.

ormation. Furthermore, Terao et al. (2010) showed that the expres-ion level of APO1 in the inflorescence meristem of Habataki wasigher than that of Sasanishiki. These results indicate that the highotal seed weight of Habataki under the NA condition might beaused by the high level of APO1 expression in inflorescence meris-em, which ozone might reduce. Reduction of APO1 expression in

he inflorescence meristem of ozone-exposed Habataki reducedhe number of primary rachis branch, resulting in lower total seedeight. Indeed, our preliminary gene expression analysis showed

erimental Botany 88 (2013) 100– 106 105

that the accumulation of APO1 transcripts in young panicles ofHabataki grown under the NA condition was remarkably higherthan that of Sasanishiki (Fig. 7), but under the ×2 condition, the levelof APO1 transcript decreased only in Habataki. Further experiments,such as APO1 expression analyses in inflorescence meristem withand without ozone treatment, and phenotypic analyses in near-isogenic lines carrying the Habataki APO1 locus in a Sasanishikibackground, are required to identify whether APO1 is involved inozone-induced total seed weight loss.

5. Conclusions

Two rice cultivars, japonica cv. Sasanishiki and indica cv.Habataki, respond differently to chronic ozone exposure in visibleleaf injury and total seed weight. We found no correlation betweenthe appearance of leaf injury and total seed weight loss, suggestingthat these phenotypes may be regulated by different mechanisms.Our QTL analysis showed that total seed weight loss under high-ozone conditions may be regulated, at least in the cultivars studiedhere, by APO1 expression.

Acknowledgements

The seeds of the 39 Sasanishiki/Habataki CSSL lines(SL401–SL439) were developed by the Rice Genome Projectof the National Institute of Agrobiological Sciences, Japan,and were provided by the Rice Genome Resource Center(http://www.rgrc.dna.affrc.go.jp/). This work was supportedpartly by the Global Environment Research Fund (A-0806) of theMinistry of the Environment, Japan.

References

Ando, T., Yamamoto, T., Shimizu, T., Ma, X.F., Shomura, A., Takeuchi, Y., Lin, S.Y., Yano,M., 2008. Genetic dissection and pyramiding of quantitative traits for paniclearchitecture by using chromosomal segment substitution lines in rice. Theor.Appl. Genet. 116 (6), 881–890.

Asanuma, S., Nito, N., Ookawa, T., Hirasawa, T., 2008. Yield, dry matter productionand ecophysiological characteristics of rice cultivar, Habataki compared with cvSasanishiki. Jpn. J. Crop Sci. 77, 474–480 (in Japanese).

Ashikari, M., Sakakibara, H., Lin, S., Yamamoto, T., Takashi, T., Nishimura, A., Angeles,E.R., Qian, Q., Kitano, H., Matsuoka, M., 2005. Cytokinin oxidase regulates ricegrain production. Science 309, 741–745.

Ashmore, M.R., Toet, S., Emberson, L., 2006. Ozone – a significant threat to futureworld food production? New Phytol. 170 (2), 201–204.

Baier, M., Kandlbinder, A., Golldack, D., Dietz, K.-J., 2005. Oxidative stressand ozone: perception, signalling and response. Plant Cell Environ. 28,1012–1020.

Broman, K.W., Wu, H., Sen, S., Churchill, G.A., 2003. R/qtl: QTL mapping in experi-mental crosses. Bioinformatics 19 (7), 889–890.

Fan, C., Xing, Y., Mao, H., Lu, T., Han, B., Xu, C., Li, X., Zhang, Q., 2006. GS3, a majorQTL for grain length and weight and minor QTL for grain width and thicknessin rice, encodes a putative transmembrane protein. Theor. Appl. Genet. 112,1164–1171.

Frei, M., Tanaka, J.P., Wissuwa, M., 2008. Genotypic variation in tolerance to ele-vated ozone in rice: dissection of distinct genetic factors linked to tolerancemechanisms. J. Exp. Bot. 59, 3741–3752.

Haldane, J.B.S., 1919. The mapping function. J. Genet. 8, 299–309.Haley, C.S., Knott, S.A., 1992. A simple regression method for mapping quan-

titative trait loci in line crosses using flanking markers. Heredity 69,315–324.

Ikeda, K., Ito, M., Nagasawa, N., Kyozuka, J., Nagato, Y., 2007. Rice ABERRANT PANICLEORGANIZATION 1, encoding an F-box protein, regulates meristem fate. Plant J. 51(6), 1030–1040.

Ikeda-Kawakatsu, K., Yasuno, N., Oikawa, T., Iida, S., Nagato, Y., Maekawa, M.,Kyozuka, J., 2009. Expression level of ABERRANT PANICLE ORGANIZATION1 deter-mines rice inflorescence form through control of cell proliferation in themeristem. Plant Physiol. 150, 736–747.

Imbe, T., Akama, Y., Nakane, A., Hata, T., Ise, K., Ando, I., Uchiyamada, H., Nakagawa,N., Furutachi, H., Horisue, N., Noto, M., Fujita, Y., Kimura, K., Mori, K., Takayanagi,

ment of a multipurpose high-yielding rice variety Takanari. Bull. Natl. Inst. CropSci. 5, 35–51 (in Japanese with English Abstract).

Kangasjärvi, J., Talvinen, J., Utriainen, M., Karjalainen, R., 1994. Plant defence systemsinduced by ozone. Plant Cell Environ. 17, 783–794.

1 d Exp

K

K

L

L

O

P

R

S

06 K. Tsukahara et al. / Environmental an

angasjärvi, J., Jaspers, P., Kollist, H., 2005. Signalling and cell death in ozone-exposedplants. Plant Cell Environ. 28, 1021–1036.

obayashi, A., Koga, Y., Uchiyamada, H., Horiuchi, H., Miura, K., Okuno, K.,Fujita, Y., Uehara, Y., Ishizawa, S., Nakagahra, M., Yamada, T., 1990. Breed-ing a new rice variety Habataki. Bull. Hokuriku Agric. Exp. Stat. 32, 65–84(in Japanese).

ander, E.S., Botstein, D., 1989. Mapping Mendelian factors underlying quantitativetraits using RFLP linkage maps. Genetics 121, 185–199.

in, D.-I., Lur, H.-S., Chu, C., 2001. Effects of abscisic acid on ozone tolerance of rice(Oryza sativa L.) seedlings. Plant Growth Regul. 35, 295–300.

hara, T. (Ed.), 2007. Study on Characteristics and Formation Mechanism of Photo-chemical Oxidants Over Japan. Research Report from the National Institute forEnvironmental Studies, Japan, No. 195 (R-195-2007) (in Japanese).

ell, E.J., Schlagnhaufer, C.D., Arteca, R.N., 1997. Ozone-induced oxidative stress:

mechanisms of action and reaction. Physiol. Plant 100, 264–273.

ao, M.V., Davis, K.R., 2001. The physiology of ozone induced cell death. Planta 213,682–690.

awada, H., Kohno, Y., 2009. Differential ozone sensitivity of rice cultivars as indi-cated by visible injury and grain yield. Plant Biol. 11, 70–75.

erimental Botany 88 (2013) 100– 106

Schraudner, M., Langebartels, C., Sandermann, H., 1997. Changes in the biochemicalstatus of plant cells induced by the environmental pollutant ozone. Physiol. Plant100, 274–280.

Sohn, J.K., Lee, S.C., 1997. Varietal difference of resistance to ozone injury in riceplants. Korean J. Crop Sci. 42, 338–343.

Song, X.J., Huang, W., Shi, M., Zhu, M.Z., Lin, H.X.A., 2007. QTL for rice grain widthand weight encodes a previously unknown RING-type E3 ubiquitin ligase. Nat.Genet. 39, 623–630.

Terao, T., Nagara, K., Morino, K., Hirose, T., 2010. A gene controlling the number ofprimary rachis branches also controls the vascular bundle formation and henceis responsible to increase the harvest index and grain yield in rice. Theor. Appl.Genet. 120, 875–893.

Wohlgemuth, H., Mittelstrass, K., Kschieschan, S., Bender, J., Weigel, H.-J., Overmyer,K., Kangasjärvi, J., Sandermann, H., Langebartels, C., 2002. Activation of an oxida-

tive burst is a general feature of sensitive plants exposed to the air pollutantozone. Plant Cell Environ. 25, 717–726.

Xue, W., Xing, Y., Weng, X., Zhao, Y., Tang, W., Wang, L., Zhou, H., Yu, S., Xu, C., Li, X.,Zhang, Q., 2008. Natural variation in Ghd7 is an important regulator of headingdate and yield potential in rice. Nat. Genet. 40, 761–767.