Polyphyletic Origin of Cultivated Rice Based on the Interspersion Pattern Os SINEs

7/23/2019 Polyphyletic Origin of Cultivated Rice Based on the Interspersion Pattern Os SINEs

http://slidepdf.com/reader/full/polyphyletic-origin-of-cultivated-rice-based-on-the-interspersion-pattern-os 1/9

Polyphyletic Origin of Cultivated Rice: Based on the Interspersion

Pattern of SINEs

Chaoyang Cheng,* Reiko Motohashi,*1 Suguru Tsuchimoto,* Yoshimichi Fukuta, Hisako Ohtsubo,* and Eiichi Ohtsubo*

*Institute of Molecular and Cellular Biosciences, the University of Tokyo, Bunkyo-ku, Tokyo, Japan; andHokuriku National Agricultural Experimental Station, Joetsu, Japan

The wild rice species Oryza rufipogon with wide intraspecific variation is thought to be the progenitor of the cultivatedrice species Oryza sativa with two ecotypes, japonica and indica. To determine the origin of cultivated rice, subfamilymembers of the rice retroposon p-SINE1, which show insertion polymorphism in the O. sativa – O. rufipogon population,were identified and used to ‘‘bar code’’ each of 101 cultivated and wild rice strains based on the presence or absence of the p-SINE1 members at the respective loci. A phylogenetic tree constructed based on the bar codes given to the ricestrains showed that O. sativa strains were classified into two groups corresponding to japonica and indica, whereasO. rufipogon strains were in four groups, in which annual O. rufipogon strains formed a single group, differing from theperennial O. rufipogon strains of the other three groups. Japonica strains were closely related to the O. rufipogonperennial strains of one group, and the indica strains were closely related to the O. rufipogon annual strains, indicatingthat O. sativa has been derived polyphyletically from O. rufipogon. The subfamily members of p-SINE1 constitutea powerful tool for studying the classification and relationship of rice strains, even when one has limited knowledge of morphology, taxonomy, physiology, and biochemistry of rice strains.

Introduction

Oryza sativa is the species of rice cultivatedworldwide. It has five wild relatives which belong to theOryza genus with the AA genome. Of these, Oryzarufipogon is the species closest to O. sativa and is gen-erally thought to be its progenitor (Oka 1988; Morishima,Sano, and Oka 1992; Khush 1997). O. sativa and O.rufipogon, however, show high intraspecific variation (seeMorishima, Sano, and Oka 1992). Morphologically, O.sativa strains are classified into two ecotypes, indica and japonica (Kato, Kosaka, and Hara 1928). A third ecotype, javanica, has also been reported for certain strains (Matsuo1952; Morinaga 1954), but these are thought to be tropicalcomponents of a single japonica group (Oka 1958). Eventhe ecotypes indica and japonica show overlapping

morphological variations, and there is no single criterionby which to distinguish them with certainty (Morishima,Sano, and Oka 1992). O. rufipogon strains are classified astwo ecotypes, perennial and annual, which differ markedlyin life-history traits and habitat preference (Oka andMorishima 1967; Oka 1988; Morishima, Sano, and Oka 1984, 1992). An intermediate type has been noted for some O. rufipogon strains (Morishima, Oka, and Chang1961; Sano, Morishima, and Oka 1980). The intraspecificvariation of O. sativa has been investigated on a large scaleby restriction fragment length polymorphism (RFLP)analysis (Wang and Tanksley 1989) and by isozymeanalysis (Glaszmann 1987), whereas few extensivemolecular investigations have been conducted on variation

in O. rufipogon.The origin of O. sativa has been questioned.

Perennial type O. rufipogon is considered to be a possible

progenitor, because the high genetic variability in theperennial population would provide higher evolutionarypotential than that in the annual population, and becausesome wild-cultivated hybrid plants have been shown toform a bridge that connects the perennial type of O. rufipogon with O. sativa in various traits (Oka 1964,1974). Annual type O. rufipogon is also considered to bea possible progenitor of O. sativa (Chang 1976), becauseannual strains have high seed productivity and certainother characters similar to O. sativa. Some O. rufipogonstrains have been shown to be capable of evolving intoboth the indica and japonica types of O. sativa (Oka 1974;Oka and Morishima 1982), which supports the idea that the two ecotype strains have been derived monophyleti-

cally by domestication. Biochemical and molecular studies, however, show that the indica and japonica strainsare closer in some characters or loci to different O.rufipogon strains than they are to each other (Second 1982;Morishima 1986; Dally and Second 1990; Wang, Second,and Tanksley 1992; Mochizuki et al. 1993; Hirano et al.1994), which may support the idea that the two ecotypestrains originated diphyletically. Note, however, that thenumber of genetic markers or O. sativa and O. rufipogonstrains used in these studies are not sufficient to discuss theorigin of the two ecotype strains.

Short interspersed elements (SINEs) are retroele-ments found in eukaryotic genomes (see Maraia 1995).SINEs range from 70 to 500 bp, and each has an internalpromoter for RNA polymerase III (pol III). Many SINEsappear to be related to tRNAs, whereas a few, such as theprimate Alu and rodent B1 family elements, are related to7SL RNA. SINEs have been used as markers for phylogenetic studies owing to their character: Theprobability of independent retroposition at the samechromosome site is virtually nil, and a SINE insertion isirreversible (for a review, see Shedlock and Okada 2000).The insertion polymorphism of SINEs has been used toinfer the phylogenetic relationship among species of primates (Bailey and Shen 1993, 1997; Hamdi et al.

1Present address: R. Motohashi, RIKEN Tsukuba Institute, Tsukuba,Ibaraki, 305-0074, Japan.

Key words: Oryza sativa, indica and japonica rice, Oryza rufipogon,SINE, insertion polymorphism, polyphyletic origin.

E-mail: [email protected].

Mol. Biol. Evol. 20(1):67–75. 2003DOI: 10.1093/molbev/msg004

2003 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038

67



1999), other animals (Murata et al. 1993; Shimamura et al.1997; Nikaido, Rooney, and Okada 1999), and plants(Mochizuki et al. 1993; Tatout et al. 1999). AlthoughSINEs are useful for phylogenetic studies, most publishedstudies have been limited to an analysis of interspeciesrelationships. This mainly reflects the fact that most of theidentified SINEs are fixed at particular loci in thepopulation of a given species. Some Alu members,however, show insertion polymorphism in humans (Batzer et al. 1994; Roy et al. 1999), and several members of thesalmon SmaI family show intraspecific polymorphism inchum salmon (Takasaki et al. 1997). These polymorphicmembers must have retroposed recently in the evolution-ary time scale, and therefore they are not fixed in thehuman and salmon populations.

Here, we report the identification of members of a subfamily of p-SINE1, the first plant SINE identified inO. sativa and its related rice species (Umeda, Ohtsubo, andOhtsubo 1991; Mochizuki et al. 1992, 1993; Motohashi et al. 1997). We show that these p-SINE1 members havebeen recently amplified in the O. sativa and O. rufipogonpopulations. A phylogenetic tree of the various O. sativaand O. rufipogon strains was constructed from ‘‘bar codes’’based on the presence or absence of p-SINE1 members at particular loci in each strain. In this tree, the ecotypes of O. sativa or O. rufipogon strains are clearly distinguished.The O. sativa japonica strains and some perennialO. rufipogon strains appear to constitute one group. Most O. sativa indica strains and the annual O. rufipogon strainsare separated but appear to constitute another group. Thisindicates that the japonica and indica strains are descendedfrom different ancestors. The phylogenetic tree alsosuggests that temperate japonica strains originated froma common source and that a few indica strains originatedby recent domestication in the annual O. rufipogonpopulation. We also show that two types of chloroplast DNA are distributed in the rice strains tested and discussthe origin of the maternal source of O. sativa.

Materials and MethodsRice Strains

The rice strains used are listed below. O. sativa cv.Nipponbare, IR36, and C5924 have been describedpreviously (Motohashi et al. 1997). Some of the wild ricestrains were obtained from Dr. N. Kurata, the GeneticStock Research Center of the National Genetic Institute(Japan). Total genomic DNAs were isolated from thesestrains, as described previously (Ohtsubo, Umeda, and

Ohtsubo 1991). DNA samples of O. sativa cv. Zheda 22,Zheda 8044, and Minghui 86 were supplied by Dr. Q.Xue, Zhejiang University (China). DNA samples of theother rice strains were obtained from the HokurikuAgricultural Experiment Station (Japan).

Polymerase Chain Reaction

The polymerase chain reaction (PCR) analysis wasdone with Ex Taq DNA polymerase (Takara), as describedelsewhere (Motohashi et al. 1997). The presence or ab-sence of each p-SINE1 member was determined by identi-

fying one unique PCR fragment with or without a p-SINE1member after electrophoresis in the agarose gel. When thefragments differed in size or when two or more bands werepresent, the presence or absence of p-SINE1 in thefragments was confirmed by Southern hybridization or by direct sequencing of the PCR products (Mochizuki et al.1993; Motohashi et al. 1997). Primer sequences will beprovided upon request.

Adaptor-ligation based–PCR (ADL-PCR) (Spertini,Beliveau, and Bellemare 1999) was performed with ExTaq DNA polymerase (Takara) as follows: The total DNAof O. sativa cv. IR36 or C5924 was digested with EcoRI, Hind III, BamHI, XbaI, NheI or SpeI (New EnglandBiolabs), none of which cut the p-SINE1 sequence. T4DNA ligase (New England Biolabs) was used to ligate thedigested DNA with the oligonucleotide adaptor. PCR first was performed with a ligated sample as the template andwith primers that hybridize to the adaptor and the p-SINE1sequence to obtain fragments with the proximal portion of p-SINE1 and its flanking sequence. PCR then wasaccomplished with primers that hybridize to the adaptor and a different portion of the p-SINE1 sequence (see fig. 1for primers used). Fragments that included the entire p-SINE1 sequence were obtained by an ADL-PCR withprimers that hybridize to the flanking sequence of eachidentified member and the adaptor. Two pairs of primers[P1 and P2; P1(RA) and P2(RA) (see fig. 1)] were used toisolate general p-SINE1 members and RA subfamilymembers, respectively.

Inverse PCR (IPCR) was performed as describedelsewhere (Tenzen et al. 1994) with the total DNA from O.sativa cv. C5924 as the template and with primers that hybridize to the additional sequence of each of the three p-SINE1 transcripts that were identified as cDNA by 59

RACE-PCR (unpublished results).

Cloning and DNA Sequencing

In this study, PCR products were cloned into thepGEM-T Easy Vector plasmid (Promega), according to thesupplier ’s instructions. DNA sequencing was done witha BigDye Terminator kit (PE Biosystem) and an ABIPRISM DNA sequencing system. Nucleotide sequencedata reported are available in the DDBJ/EMBL/GenBankdatabases under the accession numbers AB086069– AB086091.

Computer Analysis

Primary nucleotide sequences were analyzed with theGENETYX-Mac 10.1 system program. Nucleotide se-quence searches of the DDBJ/Genbank/EMBL databaseswere done with the program BLAST (Altschul et al. 1990)and a Smith-Waterman search (Smith and Waterman1981). Multiple sequences were aligned by use of theprogram CLUSTAL W (version 1.7).

To construct the phylogenetic tree for the various ricestrains, the presence or absence of the p-SINE1 members at particular loci in the strains was organized into a data matrix, such that the presence of a p-SINE1 member at a given locus was coded 1, and its absence at the same

68 Cheng et al.



locus was coded 0. Some strains generated two PCR-amplified fragments with or without a p-SINE1 member,indicative of both the presence and the absence of themember. Such a case was coded 1, the presence state. The

Neighbor-Joining method was used with the computer program PAUP* 4.0b8 (Swofford 1998).

Results and DiscussionIdentification of p-SINE1 Members Recently Amplifiedin O. sativa and O. rufipogon

Forty-seven members of p-SINE1, present at different loci in the genome of O. sativa cv. Nipponbare (japonica),IR36, or C5924 (indica), were collected. Of these, 34 hadbeen identified previously (Motohashi et al. 1997) and 13were newly found by ADL-PCR or IPCR (see Materials

and Methods) or by a computer-based homology search of the databases. The presence or absence of these membersat particular loci in strains of six rice species with the AAgenome, O. sativa, O. rufipogon, O. glumaepatula, O.

meridionalis, O. longistaminata, and O. barthii, wasdetermined by PCR analysis with a pair of primers that hybridize to the flanking regions of each p-SINE1 member.Seventeen p-SINE1 members were found to be absent at corresponding loci in one or more of the rice strainsexamined (table 1), suggesting that they are polymorphic p-SINE1 members which may have been amplified duringthe divergence of the rice species.

Of the polymorphic members, five (r2, r30, r210,r215, and r216) were absent in rice species other than O.sativa and O. rufipogon and showed insertion poly-morphism in O. sativa and O. rufipogon (table 1). This

FIG. 1.—Nucleotide sequences of the p-SINE1 members, most showing insertion polymorphism among the rice species. A, Seventeen p-SINE1members described in the first section of Results (see table 1). A nonpolymorphic p-SINE1 member, r1, is shown for comparison. A consensus sequence(CON) of p-SINE1 (Motohashi et al. 1997) is shown at the top. Sequences corresponding to the A-box and the B-box of the pol III promoter are shown

in boldface letters. Note that the p-SINE1 members include six RA subfamily members with three diagnostic mutations, as indicated by the squarebrackets. B, All RA-subfamily p-SINE1 members, except for r64 and r65. p-SINE1 members with an 3 are those present in all the rice strains examined;those with a circle are specifically present in O. sativa, O. rufipogon, or both. The other members showing insertion polymorphism are marked with anasterisk. In each p-SINE1 sequence, nucleotides identical to those in the consensus sequence are shown by dashes; deleted nucleotides are indicated byslashes. RA-subfamily diagnostic mutations are boxed. RA-subfamily members may be further divided into four groups, as indicated by the squarebrackets. Group-specific mutations are boxed. Member r63 exceptionally has a solo long terminal repeat of a retrotransposon whose target site sequenceis boxed. Positions of two pairs of primers [P1 and P2; P1(RA) and P2(RA)] used to isolate the p-SINE1 members by ADL-PCR are shown at thebottom by horizontal arrows. Note that the 39 ends of the primers P1(RA) and P2(RA) match the mutated nucleotides in the p-SINE1 sequence of theRA subfamily.

Polyphyletic Origin of Cultivated Rice 69



means that these p-SINE1 members were amplifiedrecently during speciation of O. sativa and O. rufipogonand/or after the divergence of the two species. Acomparison of nucleotide sequences of the above memberswith a p-SINE1 consensus sequence derived previously(Motohashi et al. 1997) showed that these members sharethree common mutations; two in the A-box and B-boxsequences in the pol III promoter and one in the distal endregion (fig. 1 A). None of the other polymorphic members,except for r34, nor any of the nonpolymorphic membershad three mutations (fig. 1 A). Member r34 showedinsertion polymorphism among the O. sativa and

O. rufipogon strains, but it was also present in the strainsof several species other than O. sativa and O. rufipogon,unlike the above five members (table 1). The insertionpolymorphism of r34 may represent genetic introgressionduring speciation. Alternatively, it may reflect differentialsorting of an ancestral polymorphism, as previouslyproposed (Wu 1991; Hamada et al. 1998). We haverecently identified other polymorphic members like r34and analyzed their distribution among many wild ricestrains of all the species with the AA genome, and wediscussed two possibilities based on the results obtained(Cheng et al. 2002).

Table 1The Presence or Absence of p-SINE1 Members in Oryza Strains with the AA Genome

pSINE1 (Os-1 Os-2 Os-3) (Or-1 Or-2) Og Om Ol Ob Reference

3 r1 1 1 1 1 1 1 1 1 1 1

* r6 1 1 1 1 1 1 1 1 2 2* r25 6 6 6 6 6 2 6 1 1 3* r29 1 1 1 1 1 1 1 2 1 3

* r32 1 1 1 1 1 1 2 1 1 3* r46 1 1 1 1 1 1 2 1 1 3* r68 1 1 1 1 1 1 1 2 1

* r69 2 1 2 2 1 1 1 1 1

* r70 1 1 1 1 1 1 2 2 1

* r72 1 1 1 1 1 1 2 2 1

* r50 1 1 1 1 1 1 / 2 1

* r509 1 1 1 1 1 1 2 1 1

8 r2 1 1 1 1 2 2 2 2 2 18 r30 1 1 1 2 1 2 2 2 2 38 r210 2 2 1 2 2 2 / 2 2

8 r215 2 2 1 2 2 2 2 2 2

8 r216 2 2 6 2 2 2 2 2 2

* r34 1 1 2 1 2 1 2 2 2 3

8 r51 1 1 2 2 1 2 / / 2

8 r52 1 1 1 1 1 / / / 2

8 r53 2 1 2 1 1 / / / 28 r54 2 1 1 2 2 2 2 2 2

8 r55 2 1 1 2 1 2 2 2 2

8 r56 2 1 1 2 2 2 / / 2

8 r57 2 1 2 1 2 2 2 2 2

8 r58 2 1 1 2 2 2 2 2 2

8 r59 2 1 2 / 2 / 2 2 2

8 r60 2 1 1 2 2 2 2 / 2

8 r61 2 1 1 1 2 2 2 2 2

8 r62 1 1 2 2 1 2 2 2 2

* r63 1 1 1 1 1 1 2 2 1

3 r64 1 1 1 1 1 1 1 / 1

3 r65 1 1 1 1 1 1 1 / 1

8 r66 1 1 / 1 2 2 / / 2

8 r501 1 1 1 2 2 2 2 2 2

8 r502 1 1 1 1 1 2 2 2 2

8 r503 1 1 1 2 1 2 2 2 2

8 r504 1 2 / 1 2 2 / 2 28 r505 1 2 2 2 2 2 2 / 2

8 r506 1 2 2 2 1 2 2 / 2

8 r507 1 2 2 2 2 / 2 / /

NOTE.— p-SINE1 members identified in the O. sativa genome are shown in the first column. A p-SINE1 member, r1, is shown separately from the others, because

it is the representative one of nonpolymorphic members. Seventeen polymorphic p-SINE1 members described in the first section of Results are also shown separately

from the other members. Members with an 3 are those present in all the rice strains examined. Those with a circle are specifically present in O. sativa, O. rufipogon,

or both. The other members showing insertion polymorphism are marked with an asterisk. The presence ( 1) or absence (2) of p-SINE1 members at particular loci was

determined by analysis of PCR-amplified fragments (see Materials and Methods). A 6 sign indicates a strain that generates fragments both with and without p-SINE1.

A slash shows that no PCR fragments were amplified by PCR. The rice strains used were: Os-1, Os-2, and Os-3, O. sativa cv. Nipponbare, IR36, and C5924; Or-1

and Or-2, O. rufipogon W593 and W1954; Og, O. glumaepatula W1192; Om, O. meridionalis W1625; Ol, O. longistaminata W1451; and Ob, O. barthii W1581.

References (1, Umeda, Ohtsubo, and Ohtsubo 1991; 2, Mochizuki et al. 1992; 3, Motohashi et al. 1997) describing nine p-SINE1 members are given in the last col-

umn. Another nine members (r501–r509) were found in the databases under the accession numbers, AQ363431, AQ861438, AQ865056, AC026815, AC077693,

AC079022, AP003539, AC069145, and AC069300. The remaining members were identified in this study by ADL-PCR or IPCR.

70 Cheng et al.



Note that three members (r210, r215, and r216) wereisolated by IPCR based on the sequences of transcripts insuspension-cultured cells of O. sativa cv. C5924 (see Mate rials and Meth ods). This indicates that somemembers with the common mutations are transcribed andsupports the speculation that the six p-SINE1 membershave recently been amplified.

Recently Amplified p-SINE1 Members Forming a Sub-family

To ascertain whether members other than the six withthe three diagnostic mutations are present in O. sativa, weused ADL-PCR to isolate p-SINE1 members withmutations in the pol III promoter and obtained 16 members(r51–r66; table 1). Except for two members, r64 and r65,these members carried the same three mutations as those inthe six p-SINE1 members identified above (fig. 1 B),suggesting that these and the six members form a sub-family. Two members (r64 and r65) did not have the thirddiagnostic mutation in the distal end region of p-SINE1(fig. 1 B), suggesting that they are not subfamily members.A computer-aided homology search of the databasesfurther identified seven members (named r501–r507) that have the three diagnostic mutations (table 1 and fig. 1 B).Alignment of all members with three diagnostic mutationsshowed that they are highly homologous but divided intoseveral groups, the members of which carry an additionaldiagnostic mutation(s) (fig. 1 B).

To determine whether the newly identified23 membershave been recently amplified or not, their presence or absence at their respective loci in the rice strains wasinvestigated. Twenty members were absent from the four rice species other than O. sativa and O. rufipogon andshowed insertion polymorphism in O. sativa and O.rufipogon (table 1). One member, r63, which had a retro-transposon insertion within the p-SINE1 sequence, however,was present not onlyin O. sativa and O. rufipogon but also intwo of the other four species (table 1), as was one (r34) of thesix members identified earlier. Two members, r64 and r65,which were considered not to be subfamily membersbecause of lack of a third diagnostic mutation in the distalend region of p-SINE1, were present in all the rice species(table 1), as expected. These findings indicate that p-SINE1members that carry all three diagnostic mutations have beenamplified in recent evolutionary time. We therefore namedthem RA (recently amplified) subfamily members.

As stated above, RA subfamily members have twomutations in the A-box and B-box sequences of theinternal pol III promoter. These members have another mutation in the distal end region, which is thought to beimportant for priming reverse transcription and integrationinto the target site. All mutations that occurred in p-SINE1may have had a positive influence on the transcription andretroposition of p-SINE1, leading to the beneficialamplification of RA subfamily members in the O. sativaand O. rufipogon populations.

Phylogenetic Analysis of O. sativa Strains Based on p-SINE1 Insertion Polymorphism

We collected 101 strains of O. sativa and O. rufipogon.O. sativa strains have been classified morphologically into

FIG. 2.—The rice strains used. Ecotypes of the O. sativa strains, japonica (JA) and indica (IN), are shown. Ecotypes of the O. rufipogon

strains, annual (A), perennial (P), and intermediate (I) (Morishima, Oka,and Chang 1961; Morishima, Kurata, and Sano, personal communication)are also shown. O. rufipogon strains, whose ecotypes are reported asperennial based on one classifier or annual based on another, are indicatedby (P, A). Twenty-four columns indicate the loci where p-SINE1members are present. Each rice strain shows a characteristic pattern of thepresence (1) or absence (2) of p-SINE1 members. The presence of p-SINE1 members identified originally in Nipponbare, IR36, or C5924 isshown by a boxed 1. The plus-or-minus signs indicate that PCR-amplified fragments both with and without a p-SINE1 member weregenerated. A slash indicates that no PCR fragments were amplified.Relationships among the rice strains are shown at left by the cladogrambased on the results in figure 3. The four groups (I–IV) defined in figure 3are shown.




indica and japonica (fig. 2). Except for seven strains, theO. rufipogon strains have been classified into two ecotypes,annual and perennial (fig. 2). Of the exceptional strains, fivehave been classified as an intermediate type and two as theannual type based on one classifier or as the perennial typebased on another (fig. 2).

The presence or absence of the 22 RA polymorphicmembers, one non-RA polymorphic member (r69), andone non-RA, nonpolymorphic member (r1) of p-SINE1 at particular loci in each rice strain was investigated by PCR.Note here that of the p-SINE1 members examined, eight are those identified in one japonica strain, O. sativa cv.Nipponbare, whereas 16 are those identified in two indica strains, O. sativa cv. IR36 and C5924 (see fig. 2, in whichthe presence of p-SINE1 members identified originally ineach of the three rice strains is shown by a boxed 1). Asexpected, the p-SINE1 member r1 was present in all ricestrains examined, but the other p-SINE1 members showedinsertion polymorphism (fig. 2). None of the polymorphic p-SINE1 members, however, exclusively distinguished the

strains of one ecotype from those of another (fig. 2), aswould be expected from the gene flow that may haveoccurred spontaneously among the strains of closelyrelated species O. sativa and O. rufipogon (see Morishima et al. 1992). To show the relationships among the ricestrains, each was bar-coded based on the presence or absence of p-SINE1 members at the respective loci, anda phylogenetic tree of the rice strains was constructed bythe Neighbor-Joining method based on the bar codes givento all the strains (fig. 3). In the phylogenetic tree, the ricestrains are divided into four major groups, I–IV. O. sativastrains are clearly separated into two groups, I and II.Group I comprises 32 japonica strains and one indica strain, whereas group II comprises 32 indica and 4 japonica strains (fig. 3). This means that the two ecotypestrains of O. sativa can be distinguished almost exclusivelyby the presence or absence of p-SINE1 members and that the present method provides a means of studying therelationships of these rice strains. The exceptional oneindica strain in group I and four japonica strains in group II

FIG. 3.—A phylogenetic tree showing the relationships among the O. sativa and O. rufipogon strains. The four groups (I–IV) identified are shownby four deep internal branches indicated by thick lines. The large star indicates a hypothetical ancestor, whose position is shown by the dashed arrow.All the rice strains are listed in figure 2. The japonica strains of O. sativa are indicated by letters in blue, and the indica strains are in pink. Tropical

japonica strains are underscored. The annual, perennial and intermediate strains of O. rufipogon are respectively shown by letters in green, red, andorange. The O. rufipogon strains, whose ecotypes are reported as annual based on one classifier or perennial based on another, are shown by blackletters. The wild rice strains of the four species other than O. rufipogon are circled. Rice strains with an asterisk are those carrying the chloroplast DNAwith a deletion. The scale bar equals a distance of 0.1.

72 Cheng et al.



may be the result of a mishap in handling the rice strains or of wrong classification, because indica and japonica showoverlapping morphological variations.

The japonica strains in group I include those from thetropical area of South Asia and those from the temperatearea of East Asia (fig. 2). The tropical and temperate japonica strains form a branch of the tree on which thetemperate strains are clustered together (fig. 3). Thissuggests that the temperate japonica strains originatedfrom a common source. Note that eight japonica strains,including Nipponbare and J67, cannot be distinguishedfrom one another (fig. 3). This suggests that these japonica strains are derived from a narrow source.

Phylogenetic Analysis of O. rufipogon Strains

O. rufipogon strains are divided among all four groups on the tree (fig. 3). The five O. rufipogonintermediate-type strains examined are in group II or IV,whereas the two O. rufipogon strains, whose ecotypeidentification is inconsistent among different classifiers,also are in groups II and IV (fig. 3). This is significant

because O. rufipogon shows remarkably high intraspecificvariation (Morishima, Sano, and Oka 1992). Interestingly,all the annual O. rufipogon strains, except for one strain(W2004 in group IV), are present in group II, whereas theperennial strains are in each of the other three groups, I, III,and IV (fig. 3). This finding supports the speculation that the annual strains are derived from the primitive perennialstrain (Morishima, Sano, and Oka 1992). The annual O.rufipogon strains have been referred to as an independent species, Oryza nivara (Sharma and Shastry 1965), whoseclassification is still accepted by some rice geneticists.Because annual type strains appear to form a distinct groupin the O. rufipogon population (fig. 3), differentiation intothe annual type should be considered an intraspecific

variation, as reported previously (Oka 1988).The insertion of SINE is thought to be irreversible.

Therefore, it is reasonable that the ancestral state of thespecies with the AA genome would have had no insertionof any p-SINE1 member. The perennial strains of group IVhave only a few p-SINE1 members at particular loci (fig.2), which suggests that these strains may represent theprogenitor of the other perennial strains as well as of theannual strains. In the phylogenetic tree, the hypotheticalancestor with no p-SINE1 members at the respective locihas been placed with the strains of group IV (fig. 3).

Polyphyletic Origin of O. sativa

The phylogenetic tree shows the japonica strains of O. sativa and six perennial strains of O. rufipogon are ingroup I (fig. 3), indicating that the japonica and perennialstrains of this group originated from a common source,probably O. rufipogon of the perennial type. Note that fiveof the six perennial O. rufipogon strains in this group arefrom China (fig. 2).

The indica strains of O. sativa and annual strains of O. rufipogon are both in group II (fig. 3). Most indica strains are clearly distinct from the annual strains,indicating that both types of strains are originated fromanother common source. When we consider the fact that

the O. sativa cultivars are essentially perennial plants(Morishima, Sano, and Oka 1992), the source is most likely O. rufipogon of the perennial type. However, a fewindica strains, including those from Assam (India) andNepal, are clustered with the annual strains (fig. 3). Thissuggests that these indica strains may have originated inthe annual O. rufipogon population by recent domestica-tion. These findings show that the O. sativa strainsoriginated polyphyletically.

Relationship Between O. rufipogon and Other Wild RiceSpecies with the AA Genome

The presence or absence of all the p-SINE1 membersat particular loci in the four strains belonging to the other f ou r w il d r ic e spe ci es w it h t he A A g en om e (O.glumaepatula, O. meridionalis, O. longistaminata, andO. barthii ) was also investigated by PCR. These strainscarry only some of the p-SINE1 members at thecorresponding loci (fig. 2), and are clustered together withthe perennial strains of O. rufipogon of group IV and withthe hypothetical ancestor (fig. 3).

The four strains in the tree appear to be closelyrelated (see fig. 3). This is due to the use of RA-subfamily p-SINE1 members, which are specifically present in strainsof the O. rufipogon – O. sativa population, but not in thoseof the other species. Therefore, to analyze the relationshipsamong the strains of the other species in more detail, p-SINE1 members showing insertion polymorphism in oneor more of the other species have to be used. In fact, thenon-RA polymorphic p-SINE1 members identified in thisstudy (see table 1) proved to be useful for the phylogeneticstudy of all the rice species with the AA genome, whichshowed that the O. sativa and O. rufipogon strains appear as one of several major branches of the phylogenetic tree(Cheng et al. 2002).

Analysis of the Maternal Source of the Rice StrainsInvestigated

Previous RFLP analysis indicated that there areoverlaps in variation in the chloroplast DNA of O. sativaand O. rufipogon (Murata et al. 1993). It is reported that indica strains carry chloroplasts whose genome has a 69-bp deletion at the ORF100 locus, whereas japonica strainsdo not (Kanno et al. 1993; Kaneda et al. 1996). To clarifythe variation in chloroplast DNA in the rice strains used inthis study, the presence or absence of a deletion inchloroplast DNA was investigated by PCR with a pair of primers hybridizing to the sequences flanking ORF100. O.

rufipogon strains of each of the four groups havechloroplast DNA with or without the deletion (see fig. 3for strains [asterisk] with deleted chloroplast DNA). Allfour strains of the other wild rice species with the AAgenome had chloroplast DNA without deletion (fig. 3).This suggests that chloroplast DNA deletion occurred after O. rufipogon diverged from its ancestral rice species.

As for O. sativa, 63% of the indica strains hadchloroplast DNA with deletion, but the rest had thechloroplast DNA without deletion (fig. 3), suggesting that the indica strains have a polyphyletic maternal source. Allthe japonica strains, however, had chloroplast DNA




without deletion (fig. 3), suggesting that japonica strainshave a monophyletic maternal source. This finding isinteresting but inconsistent, considering that some japonica strains have been derived by hybridization withindica strains in the course of variety improvement. Thecombination of cytoplasm carrying the deleted chloroplast DNA with the japonica-type nucleus may be restrictedduring natural selection by a mechanism as yet unknown.Moreover, nuclear substitution in japonica strains byindica strains may be frequent during variety improve-ment, whereas nuclear substitution in indica strains by japonica strains is rare. This rare form of nuclea r substitution would lead to the production of japonica strains carrying chroloplast DNA without deletion. Thistheory may be supported by distorted segregation inindica–japonica hybrids, in which the segregation ratiosusually deviate from the mendelian ratio owing to anincrease in the number of genes derived from the indica parent (Oka 1988).

Acknowledgments

We thank Drs. N. Kurata and Q. Xue for providingthe rice strains. We are grateful to Drs. H. Morishima andY. Sano for the information on the ecotypes of the O.rufipogon strains and for their critical reading of themanuscript. This work was supported by grants from theMinistry of Education, Culture, Sports, Science andTechnology of Japan and from the Ministry of Agriculture,Forestry and Fisheries of Japan.

Literature Cited

Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J.Lipman. 1990. Basic local alignment search tool. J. Mol. Biol.215:403–410.

Bailey, A. D., and C. K. J. Shen. 1993. Sequential insertion of Alu family repeats into specific genomic sites of higher primates. Proc. Natl. Acad. Sci. USA 90:7205–7209.

———. 1997. Molecu lar origi n of the mosai c seque ncearrangements of higher primate a-globin duplication units.Proc. Natl. Acad. Sci. USA 94:5177–5182.

Batzer, M. A., M. Stoneking, M. Alegria-Hartman et al.(11 co-authors). 1994. African origin of human-specificpolymorphic Alu insertions. Proc. Natl. Acad. Sci. USA 91:12288–12292.

Chang, T.-T. 1976. The origin, evolution, cultivation, dissem-inationand diversification of Asian and African rice. Euphy-tica 25:425–441.

Cheng, C., S. Tsuchimoto, H. Ohtsubo, and E. Ohtsubo. 2002.Evolutionary relationships among rice species with AA

genome based on SINE insertion analysis. Genes Genet. Syst.77:323–334.Dally, A. M., and G. Second. 1990. Chloroplast DNA diversity in

wild and cultivated species of rice (Genus Oryza, sectionOryza): cladistic-mutation and genetic-distance analysis.Theor. Appl. Genet. 80:209–222.

Glaszmann, J. C. 1987. Isozymes and classification of Asian ricevarieties. Theor. Appl. Genet. 74:21–30.

Hamada, M., N. Takasaki, J. D. Reist, A. L. Decicco, A. Goto,and N. Okada. 1998. Detection of the ongoing sorting of ancestrally polymorphic SINEs toward fixation or loss inpopulations of two species of charr during speciation.Genetics 150:301–311.

Hamdi, H., H. Nishio, R. Zielinski, and A. Dugaiczyk. 1999.Origin and phylogenetic distribution of Alu DNA repeats:irreversible events in the evolution of primates. J. Mol. Biol.289:861–871.

Hirano, H.-Y., K. Mochizuki, M. Umeda, H. Ohtsubo, E.Ohtsubo, and Y. Sano. 1994. Retrotransposition of a plant SINE into the wx locus during evolution of rice. J. Mol. Evol.38:132–137.

Kaneda, C., M. Umikawa, M. R. Singh, C. Nakamura, and N.Mori. 1996. Genetic diversity and subspecies differentiation inlocal rice cultivars from Manipur state of India. Breed. Sci.46:159–166.

Kanno, A., N. Watanabe, I. Nakamura, and A. Hirai. 1993.Variations in chloroplast DNA from rice (Oryza sativa):differences between deletions mediated by short direct-repeat sequences within a single species. Theor. Appl. Genet. 86:579–584.

Kato, S., H. Kosaka, and S. Hara. 1928. On the affinity of ricevarieties as shown by fertility of hybrid plants (in Japanese).Bull. Sci. Fac. Agric. Kyushu Univ. 3:132–147.

Khush, G. S. 1997. Origin, dispersal, cultivation and variation of rice. Plant Mol. Biol. 35:25–34.

M arai a, R. J. 1995. T he i mpac t of short i nt ersperse delements (SINEs) on the host genome. Springer-Verlag,

New York.Matsuo, T. 1952. Genealogical studies on cultivated rice (in

Japanese). Bull. Natl. Inst. Agric. Sci. Jpn. 3:1–111.Mochizuki, K., H. Ohtsubo, H. Hirano, Y. Sano, and E. Ohtsubo.

1993. Classification and relationships of rice strains with AAgenome by identification of transposable elements at nine loci.Jpn. J. Genet. 68:205–217.

Mochizuki, K., M. Umeda, H. Ohtsubo, and E. Ohtsubo. 1992.Characterization of a plant SINE, p-SINE1, in rice genomes.Jpn. J. Genet. 67:155–166.

Morinaga, T. 1954. Classification of rice varieties on the basis of affinity. Pp. 1–14 in Reports of the 5th Meeting of theInternational Rice Commission’s Working Party on RiceBreeding. Ministry of Agriculture and Forestry, Tokyo.

Morishima, H. 1986. Wild progenitors of cultivated rice and their

population dynamics. Pp. 3–14 in Rice genetics. InternationalRice Research Institute, Manila.

Morishima, H., H. I. Oka, and W.-T. Chang. 1961. Directions of differentiation in populations of wild rice, Oryza perennis andO. sativa F. spontanea. Evolution 15:326–339.

Morishima, H., Y. Sano, and H. I. Oka. 1984. Differentiation of perennial and annual types due to habitat conditions in thewild rice Oryza perennis. Plant Syst. Evol. 144:119–135.

———. 1992. Evolutionary studies in cultivated rice and its wildrelatives. Oxford Surveys Evol. Biol. 8:135–184.

Motohashi, R., K. Mochizuki, H. Ohtsubo, and E. Ohtsubo.1997. Structures and distribution of p-SINE1 members in ricegenomes. Theor. Appl. Genet. 95:359–368.

Murata, S., N. Takasaki, M. Saitoh, and N. Okada. 1993.Determination of the phylogenetic relationships among pacific

salmonids by using short interspersed elements (SINEs) astemporal landmarks of evolution. Proc. Natl. Acad. Sci. USA90:6995–6999.

Nikaido, M., A. P. Rooney, and N. Okada. 1999. Phylogeneticrelationships among cetartiodactyls based on insertions of short and long interspersed elements: hippopotamuses are theclosest extant relatives of whales. Proc. Natl. Acad. Sci. USA96:10261–10266.

Ohtsubo, H., M. Umeda, and E. Ohtsubo. 1991. Organization of DNA sequences highly repeated in tandem in rice genomes.Jpn. J. Genet. 66:241–254.

Oka, H. I. 1958. Intervarietal variation and classification of cultivated rice. Indian J. Genet. Plant Breed. 18 :79–89.

74 Cheng et al.

Polyphyletic Origin of Cultivated Rice Based on the Interspersion Pattern Os SINEs

Documents

Transcript of Polyphyletic Origin of Cultivated Rice Based on the Interspersion Pattern Os SINEs