Acccepted:April 17, 2017
Corresponding author: Ryo Ishikawa ([email protected])
Investigation of genetic loci controlling non-shattering behaviour in an Indica rice cultivar ‘IR36’
Yuki Tsujimura, Chizuru Inoue, Than Myint Htun, Yumi Oka, Takashige Ishii, Ryo Ishikawa
Graduate School of Agricultural Science, Kobe University
（1-1 Rokkodai, Nada-ku, Kobe 657-8501, Japan）
Summary: Asian cultivated rice Oryza sativa L. has been known to be domesticated from wild rice, O.
rufi pogon. During domestication, agronomically suitable traits were selected, including a loss of seed shattering.
In previous studies, three quantitative trait loci (QTLs), sh4, qSH1, and qSH3, were found to be involved in the non-shattering behaviour of O. sativa Japonica Nipponbare . In contrast, O. sativa Indica IR36 has cultivated alleles at sh4 and qSH3, but a partial abscission layer formation has been observed. Differences in the seed-shattering degree between Japonica and Indica rice are mainly accounted for by the genotypes at qSH1. However, the allele effects at the loci controlling the non-shattering behaviour of Indica rice are not known clearly. We previously produced an introgression line (IL) carrying O. sativa Japonica Nipponbare cultivated alleles at sh4 and qSH3 in the genetic background of wild rice O. rufi pogon W630. Although this introgression line (hereafter IL(3+4)) has the same genotypes at sh4 and qSH3 as O. sativa Indica IR36 , it has a well-formed abscission layer compared to that of IR36, resulting in very easy-shattering like wild rice. To understand the independent effects of sh4 and qSH3 on non-shattering behaviour of IR36 , we evaluated the F1 and F2 plants between IR36 and IL(3+4). In F1 plants, the formation of a clear abscission layer was observed in a manner similar to that of IL(3+4), suggesting that IR36 may harbour recessive allele(s) at unidentifi ed seed-shattering loci. We also observed the segregation of seed-shattering behaviour in the F2 population, confi rming that other unknown mutation(s) may underlie the non-shattering behaviour of IR36 .
Key words: Rice (Oryza sativa L.), Oryza rufi pogon, seed shattering, sh4, qSH3
Introduction
During crop domestication, ancient people targeted the selection of several agronomically useful traits of wild plants for their needs (Harris 1989). These traits were related to seed size, plant shape, a loss of seed shattering, and seed dormancy (Allaby 2010). Of these, a loss of seed shattering is regarded as one of the most important domestication traits (Fuller 2007, Harlan 1975). Wild plants have strong seed-shattering behaviour caused by an abscission layer formed between the grain and rachis. As seed shattering is a major cause of yield loss, hunter-gatherers and early farmers may have selected non-shattering plants for more effi cient harvests (Dong and Wang 2015).
Asian cultivated rice Oryza sativa L. was domesticated from its wild ancestor, O. rufi pogon (Fuller 2007, Oka 1998). Through domestication, rice plants weakened or lost seed-shattering ability due to an inhibition of abscission layer formation. Previous studies have identified three quantitative trait loci (QTLs), sh4, qSH1 and qSH3 involved in non-shattering behaviour of cultivated rice (Htun et al. 2014, Konishi et al. 2006, Li et al. 2006). A major QTL, sh4, was identifi ed from a
cross between O. nivara and O. sativa Indica. The cultivated allele of sh4 is widely conserved in all cultivars tested and inhibits the formation of the abscission layer, resulting in a reduction of seed shattering (Li et al. 2006). A second QTL,
qSH1, was identified from a cross between O. sativa Indica
Kasalath and O. sativa Japonica Nipponbare . An SNP was shown to reduce the expression of a downstream gene, resulting in the absence of abscission layer formation (Konishi et al. 2006). A third QTL, qSH3, was detected from a cross between O.
rufi pogon and O. sativa Japonica (Onishi et al. 2007), although
the effect of qSH3 on seed shattering was relatively smaller than those of sh4 and qSH1. In our recent study, qSH3 was also identifi ed as an important locus for seed shattering using an F2 population between O. sativa Japonica Nipponbare and an introgression line (IL) carrying the Nipponbare alleles at qSH1 and sh4 in the genetic background of O. rufi pogon W630 (Htun
et al. 2014).
In general, Japonica rice has no abscission layer formation, whereas Indica rice has partial abscission layer formation. O.
sativa Japonica Nipponbare has cultivated alleles at the three
loci, resulting in the complete inhibition of abscission layer formation (Htun et al. 2014). In contrast, O. sativa Indica IR36 has cultivated alleles at sh4 and qSH3, and partial abscission
Research Article
layer formation is observed. The difference in the degree of seed shattering between Japonica and Indica rice is mainly accounted for by the genotypes at qSH1 (Konishi et al. 2006). To better understand the variation of non-shattering behaviour of rice cultivars, we previously produced the IL carrying Nipponbare -cultivated alleles at sh4 and qSH3 in the genetic background of O.
rufipogon W630 (Inoue et al. 2015). The IL showed a slight
inhibition of abscission layer formation around the vascular bundle with very easy seed-shattering behaviour similar to wild rice. Despite having cultivated alleles at sh4 and qSH3 in both IR36 and the IL, a clear difference in the seed-shattering degree was observed, implying that other loci independently of sh4 and
qSH3 may still be involved in the non-shattering behaviour of O. sativa Indica IR36 . In the present study, we conducted a genetic
analysis of the seed-shattering behaviour in an F2 population between O. sativa Indica IR36 and an IL having cultivated alleles at sh4 and qSH3 in the genetic background of wild rice.
Materials and Methods
Plant materials
An Indica rice cultivar, O. sativa Indica IR36 and wild rice O.
rufipogon W630 originating from Myanmar were used in this
study. An IL carrying O. sativa Japonica Nipponbare alleles at
sh4 and qSH3 in the genetic background of O. rufi pogon W630
was previously developed and named as IL(3+4) in this study. IR36 was crossed with the IL(3+4) and the resulting F1 plants were self-pollinated to obtain an F2 population. These F2 plants were grown in the paddy fi eld at Kobe University in 2015. Eight
IR36 and four F1 plants were grown in the same condition.
Genotyping of IL(3+4) at the three known seed-shattering loci
Genotypes at the three seed-shattering loci, qSH1, sh4, and
qSH3, were surveyed for Nipponbare , O. rufipogon W630,
IR36 , IL(3+4), as well as F1 plants between IR36 and IL(3+4). We employed dCAPS markers for qSH1 and sh4 (Htun et al. 2011) and two flanking SSR markers, RM15539 (F: 5 - T C T T G T T G G G C A A G T AT G T C C - 3 a n d R : 5 -TCATGTGTTGCTTTCCGTTC-3 ) and RM3601 (F: 5 - A C C G G C A C G A G A C A G A T T A G - 3 a n d R : 5 -GGGAAGGGATTGTTTCTTCA-3'), for the qSH3 candidate region (Inoue et al. 2015). IL(3+4) carries the functional and the non-functional alleles at the causal mutations of qSH1 and sh4, respectively (Fig. 1). At qSH3 locus, IL(3+4) had a cultivated chromosomal segment as shown by two SSR marker genotypes. The F1 plant between IR36 and IL(3+4) also showed the three seed-shattering loci are fi xed with the same alleles at the causal SNPs of sh4 and qSH1 as well as with the chromosomal segments of cultivated rice covering qSH3 (Fig. 1).
T6+ VK
T6+ 50 50Fig. 1. Genotypes of the plant materials at sh4 , and qSH3 used in this study.
dCAPS markers for causative mutations at qSH1 and sh4 and two SSR markers flanking the qSH3 candidate region were used. Npb: Oryza sativa Japonica Nipponbare , W630: O. rufi pogon W630, IR36: O. sativa Indica IR36 , IL: IL(3+4), F1: F1 plant between IR36 and IL(3+4). (-) and (+) indicate without and with digestion by restriction enzymes (Ava III for qSH1 and Taq I for sh4 ), respectively.
Histological analysis
Histological analysis was conducted to investigate the abscission layer formation of parental lines and their F1 plants. Samples were collected from the pedicel tissue of grains just before the heading date. Collected tissues were fi xed in an FAA solution (formaldehyde:acetic acid:70% ethanol = 1:1:18 in a volume ratio) with vacuum infiltration and preserved at 4°C. They were dehydrated in an ethanol series of 70, 80, and 90% for 2 days each and then embedded in Technovit 7100 resin (Heraeus Kulzer, Germany) according to the manufacturer s instructions. Samples were cut into 3-μm sections with a rotary microtome, RM1215RT (Leica Biosystems, Germany) and stained with toluidine blue O solution. Finally, they were photographed with a microscope digital camera using imaging software, ToupView (Amscope, USA).
Evaluation of seed-shattering behaviour
Matured rice plants are preferable for the evaluation of seed shattering. We recorded the heading date of each F2 individual weekly, from 10 August to 15 September 2015. Most F2 plants flowered in this period but 21 plants did not flower. After 30 days of the last record of heading date (i.e., 15 October), the seed-shattering behaviour of F2 individuals was evaluated. Three panicles were observed for the presence or absence of seeds left in the panicle. Among the mature plants, those with no seeds
(i.e., those whose seeds had already dropped from the panicle) were recorded as shattering . Those with no seeds by shaking panicles were also recorded as shattering . The others with seeds left on panicles were recorded as non-shattering .
Results and Discussion
Chromosomal constitution of an introgression line, IL(3+4)
In our previous study, we produced ILs carrying Nipponbare chromosomal segments covering qSH3 and sh4 in the genetic background of wild rice, O. rufi pogon W630 (Inoue et al. 2015). We focused on these lines, because they show seed-shattering behaviour, but carry the same genotypes at qSH3 and sh4 as IR36 . In this study, we used IL2, one of the two ILs(3+4), from Inoue et al. (2015). We surveyed other chromosomal regions of IL(3+4) using 78 microsatellite markers covering whole rice chromosomes. As a result, eight SSR marker loci were found to be Nipponbare alleles, showing fi ve Nipponbare chromosomal segments introgressed in the genetic background of wild rice (Fig. 2). No seed-shattering loci have been found to locate on the introgressed chromosomal segments except for sh4 and qSH3 loci (Ishikawa et al. 2017), implying that the difference in seed-shattering degree between IR36 and IL(3+4) is likely to be regulated by novel loci independently of the causal mutations at
qSH1, sh4, and qSH3. Therefore, IL(3+4) may be good material
to detect novel loci involved in non-shattering behaviour of Indica cultivated rice.
50 50 50 50 50 50 50 50 50 50 50% 50$ 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 1LSSRQEDUHFKURPRVRPDOVHJPHQW :FKURPRVRPDOVHJPHQW VK T6+ T6+
Fig. 2. Graphical genotype of IL(3+4).
Abscission layer formation of the parental lines and their F1 plant
In our previous study, IL(3+4) was found to have partial inhibition of the abscission layer formation around vascular bundles resulting in a slight reduction in the degree of seed shattering (Inoue et al. 2015). Although seeds are connected to the pedicels, they are easily shattered when grown in paddy fields; however, IR36 does not shed seeds at maturation in paddy fields. To clearly understand the difference in
seed-shattering behaviour between W630, IR36 , and IL(3+4), we produced vertical section of the abscission layer of these plants. In wild rice, the complete abscission layer was observed and was formed from the outside to vascular bundles (Fig. 3A). IR36 showed a partial abscission layer formation with approximately 70% of it being inhibited (Fig. 3B, Ishikawa et al. 2017). Although IR36 had a partial abscission layer formation, it held seeds at maturation. As observed in our previous study, IL(3+4) showed the slight inhibition of the abscission layer around vascular bundles. A few cells were not stained with toluidine blue O, showing that these cells contributed to connecting seeds on the pedicel (Inoue et al. 2015). The abscission layer formation of IL (3+4) was longer than that of IR36 ; and IL(3+4) had fewer cells unstained with toluidine blue O around vascular bundles than IR36 (Fig. 3B and 3C). These observations suggest that the degree of the abscission layer formation is one of the factors involved in the difference in seed-shattering degree between the two parents. We then investigated abscission layer formation of the F1 plants between IR36 and IL(3+4). All F1 plants showed clear abscission layer formation with strong seed-shattering behaviour as observed for IL(3+4) (Fig. 3D). These results indicate that IR36 may still carry some novel allele(s) at unidentifi ed seed-shattering loci independently of qSH3 and sh4.
' $ % &
9% 9% 9% 9%
Fig. 3. Abscission layer formation of F1 plant between IR36 and IL(3+4).
(A) O. rufi pogon W630, (B) O. sativa Indica IR36 , (C) IL(3+4), (D) F1 plant between IR36 and IL(3+4). A pair of triangles indicates the region where the abscission layer exists. VB: vascular bundle. Bars = 50 μm.
Segregation of a seed-shattering behaviour observed in an F2 population
To evaluate seed-shattering behaviour of F2 individual plants between IR36 and IL(3+4), we grew 288 plants together with the parents and their F1 plants in the paddy fi eld. A total of 21 plants did not fl ower until 15 September 2015 (Table 1); these plants were therefore not evaluated for seed-shattering behaviour owing to the presence of immature seeds. At the evaluation of the seed-shattering behaviour, 40 F2 plants were missing or could not be scored owing to the growth defects. Some of them accounted for hybrid weakness or vigour between the two parental lines. Therefore, the seed-shattering behaviour of 227 plants was scored. Of these, 168 F2 plants were shattering and 59 were non-shattering (Table 1). The ratio of shattering to non-shattering was approximately 3:1 (P > 0.05), suggesting
that the difference in seeds-shattering behaviour between the two parental lines might be regulated by a single gene. However, as the phenotypic evaluation of seed-shattering behaviour is based on the observation of presence or absence of seeds left in the panicle, and the difference in fl owering time in an F2 population may affect seed-shattering degree, we do not know how many loci are involved in the control of the seed-shattering degree between IR36 and IL(3+4). To evaluate the seed-shattering degree more precisely in the future, breaking tensile strength value should be measured to clarify the phenotypic differences.
Table 1. Segregation of seed-shattering behaviour in the F2 population
3KHQRW\SH 1RRISODQWV 6KDWWHULQJ 1RQVKDWWHULQJ /DWHIORZHULQJ *URZWKGHIHFWV
In summary, the present results indicated that IR36 might still harbour novel mutation(s) at unidentified seed-shattering loci that contribute to non-shattering behaviour. A generation of a backcrossed population would be ideal to reduce the effects of other traits on seed shattering and enable seed-shattering behaviour to be evaluated more precisely. Identification of the novel seed-shattering loci would be useful for understanding the process of rice domestication. Furthermore, the determination of novel loci will be helpful to control seed shattering to maximise yield in the future.
Acknowledgements
We thank Mr. A. Nishimura, Kobe University, for his support of marker genotyping. The wild rice accession, O. rufipogon W630, was provided by the National Institute of Genetics supported by the National Bioresource Project, MEXT, Japan. This study was supported in part by a Grant-in-Aid from JSPS (No. 26450003 to R.I.).
References
Allaby, R. (2010) Integrating the processes in the evolutionary system of domestication. J. Exp. Bot. 61: 935-944.
Dong, Y. and Y. Wang (2015) Seed shattering: from models to crops. Front. Plant. Sci. 6: 476.
Fuller, D. Q. (2007) Contrasting patterns in crop domestication and domestication rates: recent archaeobotanical insights from the Old World. Ann. Bot. 100: 903-924.
Harlan, J. R. (1975) Crops & Man. American Society of Agronomy. Madison, Wisconsin.
Harris, D. R. (1989) An evolutionary continuum of people-plant interaction. In Foraging and farming Harris, D. R. & G. C. Hillman (eds.), Routledge, London and New York. 11-26. Htun, T. M., T. Ishii and R. Ishikawa (2011) Temporal changes
of seed shattering degree of substitution lines having non-shattering alleles from cultivated rice (Oryza sativa) in the genetic background of wild rice (O. rufi pogon). J. Crop Res. 56: 39-44.
Htun, T. M., C. Inoue, O. Chourn, T. Ishii and R. Ishikawa (2014) Effect of quantitative trait loci for seed shattering on abscission layer formation in Asian wild rice Oryza rufi pogon. Breed. Sci. 64: 199-205.
Inoue, C., T. M. Htun, K. Inoue, K. Ikeda, T. Ishii and R. Ishikawa (2015) Inhibition of abscission layer formation by an interaction of seed-shattering loci, sh4 and qSH3, in rice. Genes Genet. Syst. 90: 1-9.
Ishikawa, R., A. Nishimura, T. M. Htun, R. Nishioka, Y. Oka, Y. Tsujimura, C. Inoue and T. Ishii (2017) Estimation of loci i n v o l v e d i n n o n - s h a t t e r i n g o f s e e d s i n e a r l y r i c e domestication. Genetica 145: 201-207.
Konishi, S., T. Izawa, S. Y. Lin, K. Ebana, Y. Fukuta, T. Sasaki and M. Yano (2006) An SNP caused loss of seed shattering during rice domestication. Science 312: 1392-1396.
Li, C., A. Zhou and T. Sang (2006) Rice domestication by reducing shattering. Science 311: 1936-1939.
Oka, H. (1988) The ancestors of cultivate rice. In Origin of cultivated rice , Japan Scientific Societies Press, Tokyo/ Elsevier, Amsterdam. 15-24.
Onishi, K., K. Takagi, M. Kontani, T. Tanaka and Y. Sano (2007) Different patterns of genealogical relationships found in the two major QTLs causing reduction of seed shattering during rice domestication. Genome 50: 757-766.
インディカ型栽培イネ IR36 の非脱粒性を支配する遺伝子座の解析
村雄紀・井上千鶴・Than Myint Htun・岡 佑美・石井尊生・石川 亮
神戸大学大学院農学研究科（〒 657-8501 神戸市灘区六甲台町 1-1）
要旨：栽培イネ（Oryza sativa L.）はアジアの野生イネ O. rufi pogon から栽培化されたことが知られている．栽培化では種子脱粒 性の喪失など農業上好ましい形質が選抜された．これまでの研究から，ジャポニカ型栽培イネ日本晴の非脱粒性には 3 つの遺伝 子座（qSH1，sh4，qSH3）が報告されている．一方，部分的な離層形成を示すインディカ型栽培イネの IR36 の非脱粒性には sh4 と qSH3 座が関与していると考えられる．ジャポニカ型栽培イネとインディカ型栽培イネの脱粒程度の違いを説明する主要因は qSH1 座の遺伝子型である．しかしながら，インディカ型栽培イネの非脱粒性を支配する遺伝子座については sh4 と qSH3 座以外 に詳しいことは分かっていない．我々はこれまでの研究において，野生イネ O. rufi pogon W630 の遺伝的背景において，qSH3 と sh4 座において日本晴の対立遺伝子を持つイントログレッション系統 IL（3+4）を作出している．この IL（3+4）は qSH3 と sh4 座 においてインディカ型栽培イネ IR36 と同様に栽培イネの対立遺伝子を持っているが，野生イネに似た強い脱粒性を持つ．本研 究では，これらの脱粒性の違いに着目し両者の F1ならびに F2植物の脱粒性の評価を行った．F1植物の離層形成について樹脂切 片を用いて観察したところ，IL（3+4）と同様の離層形成が見られたため，IR36 には未知の遺伝子座における劣性変異が存在する 可能性が示唆された．続いて，F2植物を評価したところ，種子脱粒性に明瞭な分離が観察された．以上の結果から，インディカ 型栽培イネ IR36 の非脱粒性には，未知の遺伝子座における変異が存在する可能性が確認された．今後は戻し交雑系統などを作 出し，これらの新規遺伝子座の同定を進める必要がある．
キーワード：イネ Oryza sativa L.，Oryza rufi pogon，種子脱粒性，sh4，qSH3
作物研究 62 号（2017） 連絡責任者：石川 亮（[email protected]）
