Genetic Resources and Crop Evolution

, Volume 60, Issue 1, pp 265–274

Geographical variation of foxtail millet, Setaria italica (L.) P. Beauv. based on rDNA PCR–RFLP

Authors

  • Meiko Eda
    • Faculty of Life and Environmental SciencesPrefectural University of Hiroshima
  • Ayumi Izumitani
    • Faculty of Life and Environmental SciencesPrefectural University of Hiroshima
  • Katsuyuki Ichitani
    • Faculty of AgricultureKagoshima University
  • Makoto Kawase
    • Genetic Resources Center (NIAS Genebank)National Institute of Agrobiological Sciences (NIAS)
    • Faculty of Life and Environmental SciencesPrefectural University of Hiroshima
Research Article

DOI: 10.1007/s10722-012-9832-8

Cite this article as:
Eda, M., Izumitani, A., Ichitani, K. et al. Genet Resour Crop Evol (2013) 60: 265. doi:10.1007/s10722-012-9832-8

Abstract

The rDNA PCR–RFLP of foxtail millet (Setaria italica) germ-plasm collected throughout Eurasia and from a part of Africa was investigated with five restriction enzymes according to our previous study. Foxtail millet germ-plasms were classified by length of the rDNA IGS and RFLP; clear geographical differentiation was observed between East Asia, the Nansei Islands of Japan-Taiwan-the Philippines area, South Asia and Afghanistan-Pakistan. We also found evidence of migration of foxtail millet landraces between the areas. We calculated diversity index (D) for each region and found that center of diversity of this millet is East Asia such as China, Korea and Japan.

Keywords

Foxtail milletGeographical variationPCR–RFLPrDNASetaria italica

Introduction

Foxtail millet, Setaria italica (L.) P. Beauv. ssp. italica is one of the oldest cereals in Eurasia. It has been used in various ways peculiar to different areas of Eurasia (Sakamoto 1987); it is thought to have played an important role in the Old World’s early agriculture. The geographical origin of foxtail millet remains a controversial issue. Cytological and genetic studies have indicated that the wild ancestor of this crop is S. italica ssp. viridis (L.) Thell. (Kihara and Kishimoto 1942; Li et al. 1945; Le Thierry d’Ennequin et al. 2000; Wang et al. 1995). The geographical origin of domesticated ssp. italica cannot be determined by the distribution of ssp. viridis, which is commonly found in different areas in Europe and Asia and colonizing in the New World. Various hypotheses have advanced monophyletic and polyphyletic origins such as a single origin in East Asia including China and Japan (Vavilov 1926) and polyphyletic origins in China and Europe (Harlan 1975). Recently, Li et al. (1995) proposed that foxtail millet was domesticated independently in China and Europe, while landraces in Afghanistan and Lebanon had been domesticated separately in recent times, because they had primitive characteristics such as numerous tillers with small panicles. In contrast to the hypotheses of multiple origins, Sakamoto (1987) suggested that foxtail millet originated somewhere in Central Asia, Afghanistan, Pakistan and northwestern India, because strains with less restricted compatibility (Kawase and Sakamoto 1987) and with primitive morphological traits were found there. This hypothesis was unique in treating China as a secondary center of diversity in foxtail millet and treating Afghanistan and Pakistan as key regions for the origin. Several works were carried out to understand genetic structure of foxtail millet landraces based on isozymes (Kawase and Sakamoto 1984; Jusuf and Pernes 1985), intraspecific hybrid pollen sterility (Kawase and Sakamoto 1987), RFLP (Fukunaga et al. 2002), DNA sequences (Wang et al. 2010), transposon display (Hirano et al. 2011). Most of the works suggested that foxtail millet is clearly differentiated into local landraces groups such as East Asia (China, Korea, Japan), Nansei Islands of Japan-Taiwan-Philippines, India and its surroundings, Afghanistan and Central Asia, and Europe.

Ribosomal DNA (rDNA) is arranged in tandem arrays and contains genes for 25 S, 18 S and 5.8 S rRNAs. The intergenic spacer (IGS) of rDNA is highly variable in length, even within a species or among cultivars, because it contains subrepeats which vary in repeat number (Rogers and Bendich 1987). We also analyzed rDNA RFLP in foxtail millet, cloned a repeat unit (Fukunaga et al. 1997), determined the sequence of the intergenic spacer, and identified the polymorphic region (Fukunaga et al. 2005). In the recent paper, we classified landraces at the IGS sequence level (Fukunaga et al. 2006). We found two major types designated as types I and II. Type I is about 300 bp shorter a ribosomal DNA repeat unit than types II and is distributed broadly in temperate zone while type II is predominant in the subtropical and tropical zones. Type I was further classified into seven subtypes Ia to Ig. Type II was classified further into five subtypes, IIa–IIe (Fukunaga et al. 2006). Among subtypes of type I, subtypes Ia and Ib are distributed broadly from East Asia to Europe but subtype Ic and Id are restrictedly distributed in East Asia and Afhganistan-northwestern Pakistan, respectively (Fukunaga et al. 2006, 2011). We also found types Ie to Ig restricted in the wild ancestor. Several substitutions, a 20-bp deletion and mononucleotide indels are involved in the differences between subtypes. We have investigated distribution of a 20-bp deletion in Pakistani and Afghan landraces and confirmed that many landraces are identical to known subtypes (Fukunaga et al. 2011). Subtype IIa was found in Taiwan and Nansei Islands of Japan, each of subtypes IIb, IIc and IId was found in only one accession, suggesting these three subtypes were rare. Subtype IIe (formly referred as type III) was broadly found in India and sporadically in East Asia (Fukunaga et al. 1997, 2006).

In the present study, we investigated rDNA types and subtypes of 480 accessions from various parts of Eurasia based on rDNA PCR–RFLP based on amplified length polymorphism and RFLP using five restriction enzymes, HpyCH4V, RsaI, AvaII, NdeI and BamHI. We discuss genetic structure of foxtail millet landraces based on the data.

Materials and methods

Plant materials and DNA extraction

As shown in Table 1 and Supplemental Table, a total of 480 foxtail millet accessions were used in this study. All the accessions were maintained at Genetic Resources Center, National Institute of Agrobiological Sciences (NIAS Genebank), Tsukuba, Japan (http://www.gene.affrc.go.jp/databases-plant_search_en.php). These materials cover most of the geographical distribution where foxtail millet had been traditionally cultivated and were also used in a phylogenetic study based on transposon display (TD; Hirano et al. 2011). All of the materials were cultivated in a glasshouse of Faculty of Life and Environmental Sciences, Prefectural University of Hiroshima, Shobara, Hiroshima, Japan and DNA of each accession was extracted from 100 to 200 mg seedling leaves according to Murray and Thompson (1980) with some modifications or using a Qiagen Plant Mini DNeasy kit.
Table 1

Materials used in this study and typing of rDNA PCR–RFLP and diversity index (D)

Country or region

Type I (Hpy−)

Type I (AvaII+)

Type I (AvaII+2)

Type I (AvaII+∆20 bp)

Other type I

Mixture of typeI (Hpy−)+ and other type I

Total type I

Type II (AvaII+ NdeI+)

Type II (AvaII+)

Type II (BamHI+)

Other type II

Total type II

Mixture of types I and IIa

Type V

Total

Db

Japan (w/o Nansei Isls)

29

23

0

0

8

0

60

0

0

40

7

47

5

0

112

0.75

Nansei Islands of Japan

4

0

0

0

3

0

7

5

0

0

1

6

0

0

13

0.70

Taiwan

0

0

0

0

2

0

2

16

0

3

0

19

0

0

21

0.39

Korea

4

17

2

0

8

0

31

0

0

6

3

9

1

0

41

0.75

Primorskaya Prov. (Russia)

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

0.00

Nobosibirsk (Russia)

0

0

0

0

0

0

0

0

1

0

0

1

0

0

1

0.00

Omsk (Russia)

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Saratov (Russia)

1

0

0

0

0

0

1

0

0

0

0

0

0

0

1

0.00

China

8

4

1

0

31

0

44

0

0

30

4

34

4

0

82

0.71

Mongolia

0

0

0

0

0

0

0

0

0

1

0

1

0

0

1

0.00

Philippines

1

0

0

0

0

0

1

6

1

1

0

8

0

0

9

0.52

Indonesia

5

0

0

0

0

0

5

0

2

0

0

2

0

0

7

0.41

Laos

1

0

0

0

0

0

1

0

0

0

0

0

0

0

1

0.00

Thailand

1

0

0

0

0

0

1

0

0

0

0

0

1

0

2

0.50

Myanmar

1

0

0

0

0

0

1

0

1

28

0

29

0

0

30

0.13

Bhutan

0

0

0

0

0

0

0

0

0

1

0

1

0

0

1

0.00

Nepal

7

0

0

1

0

0

8

10

0

6

0

16

1

0

25

0.70

Bangladesh

0

0

0

0

0

0

0

0

0

3

0

3

0

0

3

0.00

India

0

0

0

0

0

0

0

0

0

43

0

43

1

0

44

0.04

Sri Lanka

0

0

0

0

0

0

0

0

0

6

0

6

0

0

6

0.00

Pakistan

11

0

0

8

1

0

20

0

0

1

0

1

0

0

21

0.58

Afghanistan

1

0

0

4

0

0

5

0

0

0

0

0

0

0

5

0.32

Kirghizia

1

0

0

0

0

0

1

0

0

0

0

0

0

0

1

0.00

Kazakhstan

0

0

0

0

4

0

4

0

0

0

0

0

0

0

4

0.00

Tajikistan

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Uzbekistan

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Iran

0

0

1

0

2

0

3

0

0

0

0

0

0

0

3

0.44

Georgia

2

0

0

0

0

0

2

0

0

0

0

0

0

0

2

0.00

Turkey

2

0

0

0

1

1

4

0

0

0

0

0

1

0

5

0.72

Lebanon

3

0

0

0

1

0

4

0

0

1

0

1

1

0

6

0.67

Ukraine

1

0

0

0

4

0

5

0

0

0

0

0

0

0

5

0.32

Poland

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Former Czechoslovakia

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Bulgaria

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Former Yugoslavia

0

0

0

0

1

0

1

0

0

0

0

0

0

0

1

0.00

Hungary

1

0

0

0

4

0

5

0

0

0

0

0

0

0

5

0.32

Germany

1

0

0

0

0

0

1

0

0

0

0

0

0

0

1

0.00

Belgium

1

0

0

0

0

0

1

0

1

0

0

1

1

0

3

0.67

Switzerland

0

0

0

0

0

0

0

0

0

1

0

1

0

0

1

0.00

France

3

0

0

0

0

0

3

0

0

0

0

0

0

0

3

0.00

Spain

1

0

0

0

0

0

1

0

0

0

0

0

0

0

1

0.00

Morocco

0

0

0

0

0

0

0

0

0

0

1

1

0

0

1

0.00

Ethiopia

0

0

0

0

0

0

0

0

0

1

0

1

0

0

1

0.00

Kenya

0

0

0

0

0

0

0

0

0

2

0

2

0

0

2

0.00

South Africa

0

0

0

0

0

0

0

0

0

2

0

2

0

0

2

0.00

Total

90

44

4

13

76

1

228

37

6

176

16

235

16

1

480

0.79

a,bEach of types I and II are also classified into PCR–RFLP type such as type I(Hyp−)/type II(Ava II+). See detail in Supplemental Table. Here we present them as a mixture of types I&II but for calculation of D we treated each combination as a different phenotype

PCR, PCR–RFLP and electrophoresis

PCR was carried out according to Fukunaga et al. (2005) and length polymorphism was detected by electrophoresis on 1.2 % agarose gel. According to Fukunaga et al. (2005, 2006), first we classified them into types I and II. The former study revealed that subtypes Ia, Ib, Ic, Id, IIa and IIe which was renamed from type III were found with high frequencies. Type IIe having additional BamHI site can be distinguished from other subtypes (Fukunaga et al. 1997, 2005, 2006) and subtype Id with 20 bp deletion can be distinguished from other subtypes by 2.0 % agarose gel electrophoresis after RsaI digestion (Fukunaga et al. 2011). We aligned nucleotide sequences of all the subtypes and search for different restriction enzyme sites between subtypes to easily classify subtypes. In addition to BamHI and RsaI, we chose HpyCH4V at position +43 (see Fukunaga et al. 2006), which distinguish subtype Ib from other subtypes of type I, AvaII for subtype Ic, Id from other subtypes of type I at position +33, for subtypes IIa, IIb and IId at positions +33, +267 and +501(the same position of subrepeats 1–3; see Fig. 1 and Fukunaga et al. 2006) from other subtypes having as shown in Fig. 1 and NdeI at position +189 for subtype IIa from other. As there are several other nucleotide substitutions between subtypes, here we designated them as type I(Hpy−), type I(AvaII+), type I(AvaII+∆20 bp) and other type I and type II(AvaII+), type II(BamHI+), typeII(AvaII+NdeI+) and other type II based on Table 2. We also found a few accessions with an additional AvaII site at position +267 to type I (AvaII+) and designated them type I(AvaII+2)(see result). Polymorphism in type I with BamHI or NdeI or polymorphisms in type II with HpyCH4V or RsaI was not detected. Electrophoresis were carried out on 2.5 and 1.2 % agarose gels for PCR products digested with RsaI or HpyCH4V and for those with other restriction enzymes, respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs10722-012-9832-8/MediaObjects/10722_2012_9832_Fig1_HTML.gif
Fig. 1

Structure of the rDNA IGS and differences between types in length and restriction enzyme sites. Black triangles indicate specific restriction sites and a white triangle indicates a restriction site absent to the subtype. A small arrow indicates 20-bp deletion. Other type I lacks AvaII, HpyCH4IV site polymorphism and 20-bp deletion. Other type II lacks AvaII, NdeI and BamHI sites. Boxes indicate subrepeats. Vertical line indicates restriction sites commonly found in all the accessions. H and R stand for HpyCH4IV and RsaI restriction sites, respectively

Table 2

Classification of PCR–RFLP types in this study and their common restriction enzyme recognition sites with known subtypes in the previous study (Fukunaga et al. 2006, 2011)

Type in this study

Ava II(33)

Ava II(267)

HpyCH4IV

20 bp deletion

BamHI

NdeI

Known subtype having common site(s)/deletion

Type I(Hpy−)

Subtype Ib

Type I(Ava II+)

+

+

Subtypes Ic (Ie)

Type I(AvaII+∆20 bp)

+

+

+

Subtype Id

Type I(AvaII+2)

+

+

+

Other type I

+

Other type I(Ia, If, Ig)

Type II(AvaII+)

+

+

+

Subtypes IIb, IId

Type II(AvaII+NdeI)

+

+

+

+

Subtypes IIa

Type II(BamHI+)

+

+

Subtype IIe

Other type II

+

Other type II(IIc)

Gene diversity

Gene diversity (D) in each country or region was calculated by the following equation;
$$ {{D}} = 1 - \sum P_{i}^{2} $$
where Pi is the frequency of ith phenotype (Weir 1990).

Sequencing

Sequence of the unexpected type (type I(AvaII+2)) was determined according to Fukunaga et al. (2006). The sequence was aligned with known subtypes of type I-rDNA by using Clustal W (Thompson et al. 1994; http://clustalw.ddbj.nig.ac.jp) to confirm that there is an additional AvaII recognition site.

Results and discussion

Length polymorphism and geographical distributions of variants

As shown in Fig. 2 and Table 1, length polymorphism of amplified products was observed. Of 480 landraces, 228 (47.5 %) were classified into type I and 235 (49.0 %) into type II. 16 accessions (3.3 %) had both of types I and II and only one accession from Primorskaya Province, ex USSR had type V rDNA (0.2 %; see Fukunaga et al. 1997). As reported in Fukunaga et al. 1997, type I is predominantly observed in Afghanistan, Pakistan, Central Asia and Europe and frequently in East Asia (Japan, Korea, northern China) and type II is predominant in Nansei Islands of Japan, Taiwan, the Philippines, Myanmar, India and Africa. 16 accessions having both of types I and II bands were found in Japan, Korea, China, India, Lebanon and Europe where both of types I and II were distributed. They may have both of two types of rDNA in tandem arrays (Fukunaga et al. 1997). In East Asia such as Japan, Korea and China, both types are found. These results coincided with the previous work (Fukunaga et al. 1997).
https://static-content.springer.com/image/art%3A10.1007%2Fs10722-012-9832-8/MediaObjects/10722_2012_9832_Fig2_HTML.gif
Fig. 2

Restriction patterns of PCR products for subtype classification. a Digestion patterns of type-I PCR product with HpyCH4IV. Blackandwhite triangles indicate type I(Hpy−) and other subtypes of type I. b Digestion patterns of type-I PCR products digested with RsaI. Blackand white triangles indicate type I(AvaII+∆20 bp) and other subtypes of type I, respectively. c Digestion patterns of type-II PCR products digested with NdeI. White and black triangles indicate type II(AvaII+NdeI+) and other subtypes of typeII. d Digestion patterns of PCR products with BamHI. White and black triangles indicate type II (BamHI+) and other subtypes of type II. eLeft PCR products before digestion with restriction enzyme, AvaII. Right PCR products after digestion. A white triangle, a white star, a black star, a black triangle, a gray star indicate other subtype of type I, type I (AvaII+), type I(AvaII+2), other subtype of type II, type II(AvaII+)or type II(AvaII+NdeI+). Type I(AvaII+2) and type II(AvaII+) show the same band patterns after digestion and are distinguishable in length of PCR products (see the restriction map in Fig. 1). M indicates 100 bp ladder size markers. Arrows indicate size

PCR–RFLP typing and geographical distributions of variants

Types I and II were also divided into subtypes based on PCR–RFLP. As several mutations were observed at DNA sequence levels in the previous study (Fukunaga et al. 2006), we picked up a few nucleotide substitutions which are recognized as restriction enzyme sites (Fukunaga et al. 2006) or a 20 bp deletion (Fukunaga et al. 2011) and designated them as type I(Hpy−), type I (AvaII+), type I (AvaII+∆20 bp), and other type I, and type II(AvaII+), typeII(AvaII+NdeI+), type II(BamHI+) and other type II, as shown in Figs. 1 and 2 and Table 2. In addition to these types, we newly found a type I which has an additional AvaII site to type I(AvaII+) and designated it as type I(AvaII+2)(see Figs. 1, 2). We also confirmed that this type has an additional AvaII site at position +267 by sequencing (AB674536). As shown in Table 1, frequencies of type I(Hpy−), type I(AvaII+), typeI(AvaII+2), type I(AvaII+∆20 bp), other type I, typeII(AvaII+NdeI+), type II(AvaII+), type II(BamHI+), and other type II were 18.8, 9.2, 0.8, 3.0, 15.8, 7.7, 1.3, 37 and 3.3 %, respectively. Only one accession from Turkey has mixture of type I(Hpy−) and other typeI. Type II(BamHI+) is the most frequently found, followed by type I(Hpy−), other type I, type I(AvaII+) and type II(AvaII+NdeI+). As shown in the previous studies (Fukunaga et al. 2006, 2011), this study also suggested that type I(Hpy−), having a common site with subtype Ib, is distributed broadly from East Asia to eastern Europe through Central Asia, type I(AvaII+), having a common site with subtype Ic, is found in East Asia such as Japan, Korea and continental China, type I(AvaII+∆20 bp), having a common 20-bp deletion as found in subtype Id (Fukunaga et al. 2011), intensively in northern Pakistan and Afghanistan and rarely in Nepal and in Iran, type II(AvaII+NdeI+), having common sites with subtypes IIa in Nansei Islands, Taiwan, the Philippines and Nepal and type II (BamHI+) having a common site with subtype IIe is intensively in Southeast and South Asia including Myanmar, Bangladesh, India and Sri Lanka and also sporadically in East Asia (Tables 1, 2; Fig. 3). Type I(AvaII+2) were rarely found in Korea, China and Iran. Other type I is found in high frequency in temperate zone such as Japan, Pakistan and western Europe, which probably corresponds with subtype Ia according to Fukunaga et al. (2006).
https://static-content.springer.com/image/art%3A10.1007%2Fs10722-012-9832-8/MediaObjects/10722_2012_9832_Fig3_HTML.gif
Fig. 3

Geographical distributions and frequencies of RFLP types in each country or region. Color for each type/subtype is shown in the box. (Color figure online)

Genetic diversity of foxtail millet accessions

We also calculated diversity index of accessions from each country or region (Table 1). Accessions from Japan to those from Korea showed the highest diversity index 0.75 followed by those from China (0.71), Nepal (0.70) and Nansei Islands of Japan (0.70). Both of accessions from entire Europe (from Bulgaria to Spain in Table 1) and those from Central and West Asia (from Kirghizia to Ukraine in Table 1) show 0.64. Accessions from India showed quite low diversity index (0.04) despite many accessions (n = 44) from various regions from India were used. High values of diversity indices in East Asia support that center of diversity of foxtail millet is East Asia (Vavilov 1926; Fukunaga et al. 2002). Accessions from Nepal which is not considered as the original place of this millet showed high diversity index. It may be because this country is a boarder of East Asia and South Asia and foxtail millet was introduced from both regions.

Genetic structure of foxtail millet and its implication

Geographical variation of foxtail millet landraces were well investigated in morphological characters (Li et al. 1995; Ochiai 1996), phenol color reaction (Kawase and Sakamoto 1982), isozymes (Kawase and Sakamoto 1984; Jusuf and Pernes 1985), intraspecific hybrid pollen sterility (Kawase and Sakamoto 1987), RFLP (Fukunaga et al. 2002), AFLP (Le Thierry d’Ennequin et al. 2000) and transposon display (Hirano et al. 2011). Many works indicated that foxtail millet landraces form geographical groups such East Asia (Japan, Korea, continental China), Nansei Islands of Japan-Taiwan-the Philippines, India and its vicinity (and Africa), Afghanistan–Pakistan-Central Asia and Europe and also suggested that Chinese landraces are highly variable. For examples, East Asian accessions were classified into type A in study of hybrid pollen sterility (HPS) (Kawase and Sakamoto 1987) and have specific isozyme alleles (Kawase and Sakamoto 1984; Jusuf and Pernes 1985), accessions from Nansei Islands of Japan, Taiwan and the Philippines were classified into types B and D and also classified into one cluster by RFLP (Fukunaga et al. 2002) and transposon diplay (TD; Hirano et al. 2011), Indian accessions were classified into type BC in HPS and also classified into one cluster by RFLP and TD, those from Afghanistan and Pakistan were into type AC or C in HPS. The present work coincided with such results; East Asian landraces are characterized by relatively high frequency of type I(AvaII+), landraces from Nansei Islands of Japan-Taiwan-the Philippines by type II(AvaII+NdeI+), South Asian landraces by very high frequency of type II(BamHI+), Afghan and Pakistani landraces by high frequency of type I(AvaII+∆20 bp). In addition to these differentiation patterns, different genetic constitution was also observed between East Europe where other type I is predominant and West Europe where type I (Hpy−) is predominant. The differentiation between eastern Europe and western Europe was also shown in phylogeny based on TD (Hirano et al. 2011).

In the previous studies of RFLPs using limited materials, origin of accessions from Nansei Islands of Japan-Taiwan-the Philippines were not clear (Fukunaga et al. 1997, 2002). Interestingly this study indicates that these landraces have a common genotype of rDNA, type II(AvaII+NdeI+), with some Nepalese landraces. This result coincided with that based on TD markers which are genetic markers from whole genome (Hirano et al. 2011). These results are the evidence of migration of foxtail millet landraces between these two regions.

There are some remote distribution patterns. Accessions from east and south Africa (from Ethiopia, Kenya and South Africa) had type II(BamHI+). This result may reflect relatively recent migration of foxtail millet from India to Africa because some morphological characters of Kenyan landraces are similar to those of Indian landraces (data not shown). In this work, we also found type II(BamHI+) from an accession from Switzerland. As this accession is also classified into the same cluster with Indian landraces (Hirano et al. 2011) and morphological characters (data not shown), it is deduced that this accession was recently brought from India (or Africa).

Although there are some exceptions, clear geographical differentiation patterns imply that this millet has been cultivated in each region in a long history and exchanges of germ-plasms between regions were not so common. As indicated by the previous works, there are large differences between Central-West Asia and South Asia, i.e., distribution pattern of rDNA types such as type I (AvaII+∆20 bp) (probably subtype Id) and type II (BamHI+)(probably subtype IIe) are completely different between these regions. In contrast, intensive distribution of type II(BamHI+) (subtype IIe) in South Asia and its sporadic distribution in East Asia suggest some exchanges or migrations have occurred between these two regions. There have been some migrations of landraces between these two regions in a long history of cultivation of this millet.

Perspectives

In this study, PCR–RFLP of rDNA was applied to a large sample set of world-wide accessions and genetic structure of foxtail millet accessions was clearly shown. This methodology will also be useful to analyze wild ancestors, S. italica subsp. viridis. We have already analyzed Pakistani local populations (Fukunaga et al.2011) of ssp. viridis and revealed that most of them had subtypes Ig, Ie and If. rDNA PCR–RFLP developed in this study can be applied to further study on world-wide populations of the wild ancestor to elucidate geographical origins of foxtail millet.

Acknowledgments

This work was partially supported by NIAS Genebank Project, NIAS, Japan and by a Grant-in-Aid of the Ministry of Education, Culture, Sports, Science and Technology for Young Scientists (B) to KF.

Supplementary material

10722_2012_9832_MOESM1_ESM.xls (66 kb)
Supplementary material 1 (XLS 65 kb)

Copyright information

© Springer Science+Business Media Dordrecht 2012