Genetic Relationships and Diversity of Weedy Rice (Oryza Sativa L.) and Cultivated Rice Varieties in Okayama Prefecture, Japan

Breeding Science 59: 401–409 (2009)

Genetic relationships and diversity of weedy rice (Oryza sativa L.) and cultivated rice varieties in Okayama Prefecture, Japan

Maiko Akasaka1), Jun Ushiki*1,4), Hiroyoshi Iwata1), Ryuji Ishikawa2) and Toshio Ishii3)

1) National Agricultural Research Center, 3-1-1 Kannondai, Tsukuba, Ibaraki 305-8666, Japan 2) Hirosaki University, 1 Bunkyo, Hirosaki, Aomori 036-8561, Japan 3) Okayama Prefectural General Agricultural Center, 1174-4 Koudaoki, Akaiwa, Okayama 709-0801, Japan 4) Present address: National Agricultural Research Center for Hokkaido Region, 1 Hitsujigaoka, Toyohira, Sapporo, Hokkaido 062- 8555, Japan

In order to clarify the origin of weedy rice (Oryza sativa L.) in Okayama Prefecture, Japan, we used a set of 15 sequence-tagged site primers to analyze polymorphisms among 27 weedy rice accessions and 88 cultivated rice varieties accessions including common rice varieties from Okayama. The band patterns indicated that 13 of the japonica weedy accessions corresponded exactly with one of the japonica cultivated rice varieties. Furthermore, with the exception of a strong shattering habit, these japonica weedy accessions closely resembled each corresponding rice variety in morphophysiological characteristics. Consequently, in view of the close genetic homology and high morphophysiological resemblance between weedy rice and cultivated rice varieties in Okayama, we conclude that weedy rice originated from cultivated rice varieties as an “off- type” caused by genetic mutation accompanied by the development of a strong shattering habit.

Key Words: weedy rice, Oryza sativa, ferality, origin, genotyping, STS markers.

Introduction weedy rice has been noted since the 1940s; however, it has become a more serious problem since the 1980s. Weedy rice Weedy rice is a problematic rice owing to its production of a occurs in dry, direct-seeded rice fields where Akebono, red pericarp and tendency to shatter, traits that often affect Omachi, and “Kibinohana” have been cultivated (Ishii and the quality and yield of cultivated rice (Holm et al. 1997). Akazawa 2003, Ushiki et al. 2005). Recent research has re- Recently, the problem of weedy rice has become more seri- vealed that weedy rice in Okayama consists of two types; ous in the rice-producing areas of Asia, America, and namely, japonica and indica (Ushiki et al. 2005). Japonica Europe where direct seeding of rice is practiced (Delouche weedy rice is widely distributed in Okayama, whereas indica et al. 2007). In Japan, weedy rice has been an occasional weedy rice occurs only within a restricted area. With the ex- problem since the modernization of agriculture during the ception of its shattering habit, japonica weedy rice acces- mid-19th century. Currently, this feral rice is a recognized sions bears a close phenotypic resemblance to specific rice problem in Nagano and Okayama prefectures (Sakai and varieties such as Akebono, Kibinohana, or Omachi (Ushiki Saito 2003, Ishii and Akazawa 2003, Ushiki et al. 2005). et al. 2005). More recently, it was showed that the japonica The weedy rice in different prefectures have different char- weedy rice genetically close to specific rice varieties in acteristics; for example, the pericarp colors of weedy rice Okayama by haplotyping using DNA polymorphisms in in Okayama and Nagano are white and red, respectively. franking regions of two isozyme loci, Acp1 and Sdh1, which However, the accessions of both the prefectures have an ex- are able to differentiate japonica and indica types (Kawasaki tremely strong seed-shattering habit that occurs rarely in rice et al. 2009). However, the haplotype analysis couldn’t be varieties, for example, they drop most of their mature seeds precisely elucidated the genetic relationship between weedy before harvesting naturally in the field (Ushiki et al. 2005). rice and cultivated rice varieties. Rice production in Okayama Prefecture is unique in that During the last decade, DNA markers such as RAPDs dry direct seeding of rice has been practiced since the 1940s (random amplified polymorphic DNAs), AFLPs (amplified and, unlike other prefectures, Okayama continues to use old fragment length polymorphisms), and SSRs (simple se- rice varieties such as “Asahi,” “Akebono,” and “Omachi,” quence repeats), which detect genetic polymorphisms effi- which were bred during the 1920s to 1950s. In Okayama, ciently, were developed to identify rice varieties in Japan (Ashikawa et al. 1999, Akagi et al. 1996, 1997, Olufowote Communicated by T. Imbe et al. 1997, Kubo et al. 2000). Recently, Shinmura et al. Received June 4, 2009. Accepted October 30, 2009. (2005) developed a set of 15 STS (sequence-tagged site) *Corresponding author (e-mail: [email protected]) markers to identify 130 varieties of rice-paddy cultivated in 402 Akasaka, Ushiki, Iwata, Ishikawa and Ishii

Japan covering approximately 99% of the domestic rice varieties, including a few strains of Akebono, Asahi, and planted area. All band pattern data the STS markers for Omachi, were provided by the Okayama Prefectural General 130 domestic rice varieties, including Akebono, Asahi, Agriculture Center, the Nagano Agricultural Experiment Kibinohana, and Omachi, have been described by Shinmura Station, the National Institute of Agrobiological Sciences et al. (2005). Since band pattern similarities are correlated Genebank, and the National Institute of Crop Science in with genetic similarity, these STS markers can be used to Japan (Table 2). evaluate genetic similarity and diversity between weedy rice and cultivated rice varieties. DNA extraction In this study, weedy rice accessions collected from Genomic DNA of the weedy rice accessions and the rice Okayama and rice varieties provided from different sources varieties (a total of 115 accessions/varieties) was extracted in Japan were evaluated by using the STS marker set devel- from individual seedlings using the CTAB method or by us- oped by Shinmura et al. (2005) in order to clarify the origin ing a DNeasy Plant Mini Kit (QIAGEN). Extracted DNA of the weedy rice. The genetic similarity and diversity of the was used as a template for the following analysis after ad- accessions and the varieties were evaluated statistically. On justing the concentration to 100 ng/µL. the basis of the similarity and diversity of weedy rice and cultivated rice varieties, we present a hypothesis for the ori- Amplification of STS markers and detection of band patterns gin of weedy rice and the de-domestication of cultivated rice. Amplification of STS markers was conducted according to the protocol described by Shinmura et al. (2005) with the Materials and Methods following minor modifications. A mixture containing 0.075 µL of Taq polymerase (5 u/µL; TaKaRa), 1 µL of 10× PCR Plant materials buffer [100 mM Tris-HCl (pH 8.3), 500 mM KCl, 15 mM Twenty-seven weedy rice accessions were collected in MgCl2), 1 µL of 2.5 mM dNTP mixture, 0.2 µL of each prim- Okayama from 1994 to 2002 (Table 1). Eighty-eight rice er (10 µM), and 2.5 µL of template DNA (100 ng/µL or

Table 1. Weedy rice accessions collected in Okayama Prefecture Type of japonica/ Morphophysiological Collection Variety cultivated Accessiona ID No.b Collection year indicac typec site name at the site AC6 6 japonica 1 Okimoto Akebono 1998 AC7 7 japonica – Urama Akebono 1994 AC8 8 japonica 1 Shinjyo Akebono 1998 AC9 9 japonica 1 Masatsu Akebono 1996 AC10 10 japonica 1 Koudaoki Akebono 1999 AC11 11 japonica 1 Masatsu Akebono 1996 AC12 12 japonica 1 Nishiichi Akebono 1998 AC13 13 japonica 1 Masatsu Akebono 1996 AC14 14 japonica 3 Ookanda Omachi 1998 AC15 15 japonica 1 Kurata Kibinohana 1998 AC16 16 japonica 1 Kurata Kibinohana 1998 AC17 17 japonica 2 Machikanda Akebono 1999 AC18 18 japonica 8 Koutsu Akebono 1997 AC19 19 indica 9 Koutsu Akebono 1997 AC20 20 japonica – Kurata Kibinohana 1998 AC21 21 japonica 5 Koudaoki Akebono 1999 AC22 22 japonica 1 Koudaoki Akebono 1999 AC23 23 japonica 1 Koudaoki Akebono 1999 AC52 52 japonica – Koudaoki Akebono 1999 AC59 59 indica 7 Masuno Akebono 2000 AC60 60 indica 7 Masuno Akebono 2000 AC61 61 indica 7 Sawada Akebono 2000 AC62 62 japonica 6 Sawada Akebono 2000 AC63 63 japonica 6 Sawada Akebono 2000 AC64 64 japonica 4 Sawada Akebono 2000 AC65 65 indica 7 Sawada Akebono 2000 AC232 232 japonica – Okimoto Akebono 2002 a All accessions were provided by Okayama Prefectural General Agriculture Center. b Serial number of the accessions for identification. c Type of japonica/indica and morphophysiological type referred to Fig. 2 and Ushiki et al. (2005). Genetic relationships of weedy rice and cultivated rice in Japan 403

Table 2. Common rice varieties genotyped by STS markers Variety ID No.a Sourceb Remarksc Variety ID No.a Sourceb Remarksc Aikoku 167 GB 6369 Mihosen 217 GB 7112 Akebono 1 OKA General Mikinishiki 215 GB 9854 Akebono 31 OKA Original Mino 216 GB 10232 Akebono 245 GB 6798 Misuzumochi 162 NAG Original Akihikari 168 GB 6191 Miyamawase 163 NAG Original Asahi 2 OKA General Mochihikari 164 NAG Original Asahi 30 OKA Stock Myotoku 220 GB 7178 Asahi 169 GB 9855 Nakateshinsenbon 198 GB 15118 Asahi 238 GB 6746 Nihonmasari 200 GB 7546 Asahi 239 GB 9328 Nipponbare 27 OKA Original Asahi 241 GB 10736 Okayamamochi 172 GB 6920 Asominori 170 GB 11299 Omachi 3 OKA General Bizennishiki 206 GB 9364 Omachi 25 OKA Original Chubumochi 57 194 GB 9246 Omachi 67 OKA General Fujiminori 211 GB 5565 Omachi 235 GB 4551 Fukuhikari 210 GB 6464 Omachi 173 GB 7529 Hatsunishiki 204 GB 5722 Omachi 236 GB 9872 Hattan 203 GB 5769 Omachi 237 GB 14946 Himenomochi 209 GB 6113 Reimei 226 GB 5993 Hinodesen 207 GB 6873 Saikai 91 187 GB 6859 Hinohikari 29 OKA Original Shinanokogane 150 NAG Original Hinohikari 208 GB 54242 Shinanomochi 3 152 NAG Original Horei 213 GB 9250 Shinonomemochi 188 GB 6892 Hotaka 160 NAG Original Shurei 190 GB 9195 IR36 144 HI — Naganohomare 156 NAG Original Kasalath 80 GB 205306 Norin 17 201 GB 14941 Kibinohana 24 OKA Original Norin 22 202 GB 9839 Kibiyoshi 175 GB 9843 Tosan 38 205 GB 6795 Kinki 25 177 GB 6801 Taishomochi 191 GB 4931 Kinki 33 178 GB 6802 Takanenishiki 153 NAG Original Kimmaze 180 GB 14971 Takanenishiki 192 GB 15043 Kinmonnishiki 147 NAG Original Takedawase 193 GB 14947 Kintokimochi 179 GB 10247 Todorokiwase 283 GB 6490 Koganemasari 184 GB 54235 Towada 197 GB 15123 Koganenami 183 GB 15125 Toyonishiki 195 GB 6112 Kokonoemochi 186 GB 67747 Tsukubanishiki 154 NAG Original Kokuryomiyako 185 GB 7286 Wasewakaba 227 GB 9087 Koshihikari 28 OKA Original Yaegaki 221 GB 9992 Koshihikari 149 NAG Original Yaekogane 165 NAG Original Komyonishiki 182 GB 7418 Yamabiko 224 GB 9255 Kusabue 181 GB 6827 Yamadanishiki 223 GB 6958 Kusanohoshi 229 NICS — Yamafukumochi 225 GB 10206 Manryo 161 NAG Original Yashiromochi 222 GB 9852 Mihonishiki 218 GB 15014 Yutakaho 196 GB 6874 a Serial number of varieties for identification. b HI: Hirosaki University, GB: National Institute of Agrobiological Sciences Genebank, OKA: Okayama Prefectural General Agriculture Center, NAG: Nagano Agricultural Experiment Station, NICS: National Institute of Crop Science. c Numbers indicate registered number (JP No.) in National Institute of Agrobiological Sciences Genebank, “Original” and “General” indicated original seeds stocked and seeds collected from general field in Okayama Prefectural General Agriculture Center. crude) in a volume of 10 µL was prepared. Fifteen sets of annealing at the temperature specific for the primer pair, and STS markers (Shinmura et al. 2005) were used as primers. 2 min extension at 72°C. PCR-amplified products were Each PCR was performed with a single set of STS markers mixed with 1 µL of 6× loading buffer and loaded on a 1.5% using a GeneAmp PCR System 9700 (Applied Biosystems) agarose gel (NipponGene). The gels were electrophoresed at thermal cycler. Multiplex PCR described by Shinmura et al. 100 V for 40 min and stained with ethidium bromide for 30 (2005) was not conducted. The amplification profile was as to 60 min. The DNA bands were visualized under UV light follows: 30–35 cycles of 1 min denaturation at 95°C, 1 min using a Printgraph system (ATTO). 404 Akasaka, Ushiki, Iwata, Ishikawa and Ishii

Cluster analysis not correspond with other common rice varieties examined Cluster analysis was performed on the band pattern data from Okayama. obtained from 245 accessions/varieties consisting of the 27 weedy rice accessions (Table 1) and 88 cultivated rice vari- Cluster 2 (Fig. 1C) eties (Table 2) examined in this study and compared with the One weedy rice accession (AC17) was grouped into 130 rice varieties reported by Shinmura et al. (2005). The Cluster 2. The band patterns of this accession corresponded 130 rice varieties were rice-paddy cultivated varieties in with that of the rice variety “Ginomi.” Japan covering approximately 99% of the domestic rice planted area, and their band patterns were used as references Cluster 3 (Fig. 1D) for comparison. Genetic distance (GD) between a pair of ac- Four weedy rice accessions (AC7, AC14, AC15, and cessions/varieties was based on the presence (1) or absence (0) AC19) were grouped into Cluster 3. The band patterns of of amplification PCR products and was determined by arith- one accession (AC14) corresponded exactly with those of metical complement of simple matching coefficient (Sneath six accessions of Omachi. and Sokal 1973) as defined below: GD = 1 − SC = 1 − (a + d)/ The band patterns of the other three accessions (AC7, (a + b + c + d); SC, similarity coefficient; a, number of matches AC15, and AC19) were different from one strain of Asahi 1,1; b, number of matches 1,0; c, number of matches 0,1 and [Asahi (241)] with respect to two or three markers and did d, number of matches 0,0. Clustering of the accessions/vari- not correspond with other common rice varieties studied eties was made by unweighted pair group method of arith- from Okayama. metic averages (UPGMA) (Sneath and Sokal 1973) using PHYLIP version 3.66 (Felsenstein 2006). The cluster analy- Cluster 4 (Fig. 1E) sis results are illustrated as a dendrogram to compare the Six weedy rice accessions were grouped into Cluster 4 to- genetic similarity and diversity of the accessions/varieties. gether with the indica rice varieties “Kasalath” and “IR36.” Four of these accessions (AC59, AC60, AC61, and AC65) Results were indica-type weedy rice, whereas the other two accessions (AC18 and AC62) were weedy rice with characteris- Cluster analysis of the band patterns for a total of 245 acces- tics intermediate between of japonica and indica (Ushiki et sions/varieties (Fig. 1) resulted in the grouping of most al. 2005). japonica weedy rice accessions into three clusters (Clusters 1, 2, and 3) and most indica weedy rice accessions into one Others (Fig. 1A) cluster (Cluster 4). An overview of the cluster analysis is Two weedy rice accessions having characteristics inter- shown in Fig. 1A, with more detailed views of the analysis mediate between japonica and indica (AC63 and AC64) and shown in Fig. 1B–E. two strains of Omachi [Omachi (235) and Omachi (S)] were distributed in several clusters widely separated from Clus- Cluster 1 (Fig. 1B) ters 1 to 4. Most japonica weedy rice accessions (14 accessions) grouped into Cluster 1. The band patterns of 10 accessions Discussion (AC8, AC9, AC11, AC13, AC16, AC20, AC21, AC22, AC23, and AC52) corresponded exactly with those of A dendrogram generated from the morphophysiological Akebono. There was no difference in band patterns among characteristics of weedy rice accessions and cultivated rice the four Akebono strains examined in this study or among varieties in Okayama that was published in a previous report those reported by Shinmura et al. (2005). (Ushiki et al. 2005) is shown as Fig. 2A. Another dendro- The band patterns of two weedy rice accessions (AC6 and gram generated from the band pattern data of the accessions/ AC232) corresponded exactly with those of Asahi (239), varieties corresponding to those in Fig. 2A is shown for Kimmaze (180) and Nakateshinsenbon (S), parenthetic comparison in Fig. 2B. Type 1 accessions (T1 in Fig. 2) con- numbers and “S” indicate serial numbers of the varieties for sist of 11 accessions that occur in six areas where the rice va- identification and the variety reported by Shinmura et al. rieties Akebono or Kibinohana are cultivated (Table 1). On (2005), respectively. The band patterns of this group dif- the basis of morphophysiological measurements, Type 1 is fered from those of Akebono and the above-listed 10 acces- most typical of the japonica weedy rice types and is similar sions with respect to three markers and differed from those to Akebono or Asahi (Fig. 2A). The results of genetic analy- of 5 other strains of Asahi with respect to one marker. Nota- sis revealed that seven accessions of Type 1 (AC8, AC9, bly, however, one strain of Asahi [Asahi (241)] differed AC11, AC13, AC16, AC22, and AC23) closely resemble from the other strains with respect to either four or five Akebono not only morphophysiologically but also geneti- markers (Fig. 1D). cally. The other two Type 1 accessions (AC10 and AC12) The remaining two weedy rice accessions in this cluster also resemble Akebono, although they have slightly differ- (AC10 and AC12) differed from Akebono and the above- ent band patterns (Figs. 1B and 2B). In contrast, one acces- listed 10 accessions with respect to only one marker and did sion (AC6) collected from the Okimoto area is genetically Genetic relationships of weedy rice and cultivated rice in Japan 405

Fig. 1A. Dendrogram of 245 accessions/varieties consisting of 27 weedy rice accessions and 88 rice varieties studied in this paper, and 130 cultivated rice varieties reported by Shinmura et al. (2005). The figure is derived from UPGMA cluster analysis based on band patterns of 15 STS markers. Dotted frames indicate clusters including weedy rice accessions (Cluster 1, 2, 3 and 4). Arrows indicate position of accessions/varieties as follows; Omachi (S): rice variety “Omachi” reported by Shinmura et al. (2005); AC63, AC64: weedy rice accessions AC63 and AC64 (refer to Table 1); Omachi (235): rice variety “Omachi” strain 235 (refer to Table 2). The bar indicates genetic distance. 406 Akasaka, Ushiki, Iwata, Ishikawa and Ishii

Fig. 1B. Cluster 1 (a part of the dendrogram shown in Fig. 1A). Number in parentheses indicates ID No. of accession/variety (refer to Table 1 and Table 2). “S” in parentheses indicates the data reported by Shinmura et al. (2005). Asterisk indicates weedy rice accession. Underlined variety indicates common rice variety in Okayama Prefecture.

Fig. 1C. Cluster 2 (a part of the dendrogram shown in Fig. 1A). Number in parentheses indicates ID No. of accession/variety (refer to Table 1 and Table 2). “S” in parentheses indicates the data reported by Shinmura et al. (2005). Asterisk indicates weedy rice accession. similar to Asahi (Fig. 1B and Fig. 2). Akebono and Asahi are threshability of Asahi and Akebono was evaluated as “very old varieties that were bred in Mie and Okayama prefectures easy” and “easy,” respectively, by the standard evaluation before the 1950s, respectively. Morphophysiologically, system for rice in Japan (Committee for Advancement of Akebono is relatively similar to Asahi since the former was Rice Yield in Okayama Prefecture 2002). Unlike weedy bred by crossing ‘Norin 12’ and Asahi; however, Akebono rice, these varieties do not shed their grains naturally; how- and Asahi are distinguishable genetically (Fig. 2). The ever, some grains of these varieties are dispersed in the field Genetic relationships of weedy rice and cultivated rice in Japan 407

Fig. 1D. Cluster 3 (a part of the dendrogram shown in Fig. 1A). Number in parentheses indicates ID No. of accession/variety (refer to Table 1 and Table 2). “S” in parentheses indicates the data reported by Shinmura et al. (2005). Asterisk indicates weedy rice accession. Underlined variety indicates common rice variety in Okayama Prefecture. Italic font indicates indica-like weedy rice.

Fig. 1E. Cluster 4 (a part of the dendrogram shown in Fig. 1A). Number in parentheses indicates ID No. of accession/variety (refer to Table 1 and Table 2). “S” in parentheses indicates the data reported by Shinmura et al. (2005). Asterisk indicates weedy rice accession. Italic font indicates indica-like weedy rice. by harvesting operations, and over-wintered seeds can ger- eties was caused by genetic mutation that enhanced the shat- minate under the dry direct seeding system in the following tering ability of the varieties. The weedy-type varieties grew spring (Ishii and Akazawa 2003). in population size and spread throughout a wide area of The foregoing results suggest the following hypothesis: Okayama that had continuously cropped a related variety Asahi or Akebono are assumed to be the origins of most under the dry direct seeding system. japonica weedy rice accessions since, apart from the shatter- The Type 3 accession AC14 (T3 in Fig. 2) is a japonica ing habit, they are closely similar to japonica weedy rice in weedy rice accession that occurred in one area that was cul- both genotype and phenotype. De-domestication of the vari- tivating the variety Omachi (Table 1). Since AC14 closely 408 Akasaka, Ushiki, Iwata, Ishikawa and Ishii

Fig. 2. Comparison of dendrograms based on morphophysiological characteristics (A) and band patterns by STS markers (B). Weedy rice accessions (not underlined) and cultivated rice varieties (underlined) in Fig. 2B are matched to those in Fig. 2A. Italic font indicates indica rice and non-italic font indicates japonica. T1–T9 indicate morphophysiological type, type1–type9, of weedy rice accessions (refer to Table 1). Curly bracket indicates category of same types (A) or same band patterns (B). resembles Omachi in both genotype and phenotype (Fig. 2), Koshihikari (Takeuchi et al. 2006). AC21 is also able to be and since Omachi is an old variety that has been cultivated a genetic variant of Akebono in a mutation of a heading in Okayama since 1922, AC14 probably originated from date gene. Omachi by a de-domestication process similar to that de- The Type 2 accession AC17 (T2 in Fig. 2) is a japonica scribed for Type 1. The band pattern of two strains of Omachi, weedy rice that occurred in a field cultivating Akebono in Omachi (235) and the Omachi (S) in Fig. 1A, differed from 1999 (Table 1). It has been reported that AC17 corresponds the other six strains that were grouped into a distant cluster with the variety Kibinohana morphophysiologically and ge- (Cluster 3 in Fig. 1A and magnified in Fig. 1D). These netically (Ushiki et al. 2005, Kawasaki et al. 2009). How- strains were variable in heading date, culm length, and ever, STS marker band pattern of AC17 was identical to that threshability, although none of the strains had a shattering of Ginomi (Fig. 1C). Kibinohana is a relatively new variety habit similar to weedy rice (Ushiki et al. 2005). These results released in 1989, and its threshability has been evaluated suggest that there are several landraces named Omachi in as medium (Committee for Advancement of Rice Yield in Okayama that differ in genotype and phenotype. Okayama Prefecture 2002). Ginomi was released in Siga Type 5 (T5 in Fig. 2) is a japonica weedy rice accession, Prefecture in 1994, and its threshability has been evaluated AC21, collected from an area cultivating Akebono (Table 1). as hard. There is no official record of Ginomi being cultivat- Although AC21 is closely similar to Akebono in genotype, it ed commercially in Okayama. Consequently, these results clearly differs phenotypically (Fig. 2). For example, AC21 suggest that the origin of AC17 is unclear. was found to be 13 days earlier in heading date and 9 cm We couldn’t also presume the origin of the other weedy shorter in culm length than Akebono (Ushiki et al. 2005). An rice accessions (AC7, AC15, AC18, AC19, AC59, AC60, early heading cultivar, Kanto HD1, which is an isogenic line AC61 AC62, AC63, AC64, AC65) because the band pat- of Koshihikari carrying a Kasalath fragment at one heading terns of the accessions didn’t match with those of rice vari- date gene (Hd1) locus, is shorter in culm length than eties that grown in Okayama (Fig. 1A, Fig. 1D, Fig. 1E and Genetic relationships of weedy rice and cultivated rice in Japan 409

Fig. 2). For this study, we examined only one sample to Cai, H.W. and H. Morishima (2002) QTL clusters reflect character as- determine the genotype of the accession and variety. For sociations in wild and cultivated rice. Theor. Appl. Genet. 104: precise comparison of genotypes between weedy rice and 1217–1228. rice varieties, we have to examine more samples reveal Committee for Advancement of Rice Yield in Okayama Prefecture the extent of the diversity within the accession or variety. (2002) Basic policy for advancement of rice production in Okayama Prefecture, 2002 edn., pp. 42–43. Kawasaki et al. (2009) indicated that a part of weedy rice Delouche, J.C., N. Burgos, D. Gaely, G. Zorrilla and R. Labrada (2007) genetically show a similarity to forage varieties and indica Weedy rices—origin, biology, ecology and control, FAO Plant modern varieties. Genetic analysis of more rice varieties in- Production & Protection paper, Rome, pp. 3–15. cluding forage varieties and indica rice varieties needed to Felsenstein, J. (2006) PHYLIP (Phylogeny Inference Package) Version reveal the origin of these accessions for precise comparison 3.66. Department of Genetics. University of Washington, Seattle, of the genotypes of weedy rice and rice varieties. USA. Distributed by the author. In summary, most japonica weedy rice probably originat- Holm, L., J. Doll, E. Holm, J. Pancho and J. Herberger (1997) THE ed from old japonica rice varieties such as Asahi, Akebono, WILD RICES. In: WORLD WEEDS: Natural Histories and Distri- and Omachi that have been widely grown in Okayama for bution, John Wiley & Sons, Inc., New York, pp. 531–547. more than a half century. Ishii,T. and M.Akazawa (2003) Weedy rice and rice cultivation by dry Acquisition of the shattering habit would be the most im- direct seeding in Okayama Prefecture. In: Proceedings of 18th Symposium of the Weed Science Society of Japan, pp. 7–16. portant factor for the de-domestication of rice varieties. Re- Kawasaki, A., K. Imai, J. Ushiki, T. Ishii and R. Ishikawa (2009) Molec- cently, rice shattering genes, qSH1 (Konishi et al. 2006) and ular constitution of weedy rice (Oryza sativa L.) found in Okayama sh4 (Li et al. 2006), were isolated and their sequences and prefecture. Breed. Sci. 59: 229–236. putative functions were reported. Additionally, some quanti- Konishi, S., T. Izawa, S.Y. Lin, K. Ebana, Y. Fukuta, T. Sasaki and tative trait loci (QTL) for the shattering habit have been de- M.Yano (2006) An SNP caused loss of seed shattering during rice termined (Cai and Morishima 2000, 2002, Xiong et al. domestication. Science 312(5778): 1392–1396. 1999). In our continuing efforts to understand the process of Kubo, T., T. Ogata, A. Yoshimura, Y. Matsue and N. Iwata (2000) DNA rice de-domestication in Japan, we are currently studying fingerprinting of japonica rice varieties using RAPD analysis. Sci. these shattering genes and the related QTLs that might be in- Bull. Fac. Agr. Kyushu Univ. 55: 5–11. volved in the shattering habit of weedy rice. Li,C., A.Zhou and T.Sang (2006) Rice domestication by reducing shattering. Science 311(5769): 1936–1939. Olufowote, J.O., Y. Xu, X. Chen, W.D. Park, H.M. Beachell, R.H. Acknowledgements Dilday, M. Goto and S.R. McCouch (1997) Comparative evaluation of within-cultivar variation of rice (Oryza sativa L.) using

We are grateful to Dr.HiroshiNemoto of the National Institute microsatellite and RFLP markers. Genome 40: 370–378. of Crop Science for providing rice variety “Kusanohoshi.” Sakai, N. and M. Saito (2003) Occurrence of weedy rice and its man- We also thank the Nagano Agricultural Experiment Station agement in Nagano Prefecture. In: Proceedings of 18th Symposium and the National Institute of Agrobiological Sciences Gene- of the Weed Science Society of Japan, pp. 1–6. bank for providing 13 and 62 rice varieties, respectively. Shinmura, K., H. Kanagawa, T. Mikami and T. Fukumori (2005) Devel- opment of multiplex PCR primer sets for the identification of rice Literature Cited varieties. Breed. Res. 7: 87–94. Sneath, D.L. and R.R. Sokal (1973) Numerical Taxonomy, The Princi- Akagi, H., Y. Yokozeki, A. Inagaki and T. Fujiwara (1996) Micro- ples and Practice of Numerical Classification, Freeman & Co., San satellite DNA markers for rice chromosomes. Theor. Appl. Genet. Francisco. 93: 1071–1077. Takeuchi, Y., T. Ebitani, T. Yamamoto, H. Sato, H. Ohta, H. Akagi, H., Y. Yokozeki, A. Inagaki and T. Fujiwara (1997) Highly poly- Hirabayashi, H. Kato, I. Ando, H. Nemoto, T. Imbe and M. Yano morphic microsatellites of rice consist of AT repeats, and a classi- (2006) Development of isogenic lines of rice cultivar Koshihikari fication of closely related cultivars with these microsatellite loci. with early and late heading by marker-assisted selection. Breed. Theor. Appl. Genet. 94: 61–67. Sci. 56: 405–413. Ashikawa, I., Y. Furuta, K. Tamura and T. Yagi (1999) Application of Ushiki, J., T. Ishii and R. Ishikawa (2005) Morpho-physiological char- AFLP technique that uses non-radioactive fluorescent primers to acters and geographical distribution of japonica and indica weedy the detection of genetic diversity in Japanese rice cultivars and rice (Oryza sativa) in Okayama Prefecture, Japan. Breed. Res. 7: cloning of DNA sequences derived from an Indica genome. Breed. 179–187. Sci. 49: 225–231. Xiong, L.Z., K.D. Liu, X.K. Dai, C.G. Xu and Q. Zhang (1999) Identifi- Cai, H.W. and H. Morishima (2000) Genomic regions affecting seed cation of genetic factors controlling domestication-related traits of shattering and seed dormancy in rice. Theor. Appl. Genet. 100: rice using an F2 population of a cross between Oryza sativa and 840–846. O. rufipogon. Theor. Appl. Genet. 98: 243–251.