• adaptation;
  • domestication;
  • introgression;
  • seed longevity;
  • weedy rice


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • As a weed of rice paddy fields, weedy rice has spread worldwide. In northern China, the expansion of weedy rice has been rapid over the past two decades. Its evolutionary history and adaptive mechanisms are poorly understood.
  • Evolutionary relationships between northern weedy rice and rice cultivars were analyzed using presumed neutral markers sampled across the rice genome. Genes involved in rice domestication were evaluated for their potential roles in weedy rice adaptation. Seed longevity, a critical trait of weedy rice, was examined in an F2 population derived from a cross between weedy rice and a rice cultivar to evaluate weedy rice adaptation and the potential effect of candidate genes.
  • Weedy rice in northern China was not derived directly from closely related wild Oryza species or from the introgression of indica subspecies. Introgression with local cultivars, coupled with selection that maintained weedy identity, shaped the evolution of weedy rice in northern China.
  • Weedy rice is a unique system with which to investigate how weedy plants adapt to an agricultural environment. Our finding that extensive introgression from local cultivars, combined with the continuing ability to maintain weedy genes, is characteristic of weedy rice in northern China provides a clue for the field control of weedy rice.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Weedy rice is distributed worldwide, and is particularly abundant in North America and South and Southeast Asia (Ferrero et al., 1999; Noldin et al., 1999). It has had a detrimental impact on the paddy ecological system and has caused huge yield losses in cultivated rice production. In the USA, yield loss is estimated at $50 million annually (Smith, 1979; Gealy et al., 2002). In China, the area affected by weedy rice has reached 3 333 300 ha and has reduced crop yield by c. 3.4 billion kilograms (Liang & Qiang, 2011). Although long-term intensive rice cultivation in northern China once suppressed weedy rice population expansion, it has become a major problem within the last 20 yr following the application of less intensive management techniques (Xia et al., 2011). In northeast China, 600 000 ha are now infested. The control of weedy rice is difficult because of its close biological similarity to cultivated varieties and its great morphological variability (Ferrero, 2003).

Although many studies have been conducted on weedy rice, its evolutionary origin and adaptive properties have yet to be fully characterized. The genetic background of many US red rice cultivars is closely related to Oryza sativa indica on the basis of simple sequence repeat (SSR) analysis (Vaughan et al., 2001; Londo & Schaal, 2007). Recent studies have confirmed that US weedy rice originated from the Asian O. sativa aus and indica subgroups, which have never been cultivated in the USA (Gross et al., 2010; Reagon et al., 2010; Thurber et al., 2010). Weedy rice in northern China (WRNC) shows extensive morphological diversity. Unlike, in other regions, not all plants possess a red pericarp. WRNC is always associated with white or red awns, seed shattering and the production of many tillers. WRNC sheds mature grains earlier than does cultivated rice. The grains are able to survive through winter and germinate during the cropping season.

Some studies based on morphology and SSR molecular markers have indicated that WRNC is more closely related to japonica subspecies than to either indica cultivars or the wild progenitor O. rufipogon (Cao et al., 2006; Tang et al., 2011). The mechanisms underlying the evolution of WRNC are still unknown. The cloning of rice domestication-related genes, that is, genes targeted by artificial selection to improve traits for cultivation, provides an opportunity to study the adaptation of weedy rice(Konishi et al., 2006; Li et al., 2006; Lin et al., 2007; Sugimoto et al., 2010). In particular, the domestication-related genes Rc, Waxy, Phr1 and Bh4, which are involved in grain quality, may also play a role in weedy rice adaptation (Hirano et al., 1998; Sweeney et al., 2006; Yu et al., 2008; Zhu et al., 2011). Some domestication-related genes, such as GS3 and qSW5, were fixed in certain rice populations during the processes of artificial selection and introgression (Fan et al., 2006; Shomura et al., 2008).

Information on genome-wide variation in Oryza has recently become available, providing a new approach to further elucidate the origins and classification of weedy rice (Garris et al., 2005; Caicedo et al., 2007; Zhao et al., 2010, 2011). Subspecies-specific indel and subspecies-specific intron length polymorphism (SSILP) molecular markers, the latter exploiting indels solely within introns, have been developed (Shen et al., 2004; Wang et al., 2005; Lu et al., 2009; Zhao et al., 2009). Two neutral nuclear p-VATPase, the chloroplast atpB-rbcL intergenic spacer region and the functional SAM (S-adenosyl methionine synthetase) gene have been used successfully to elucidate the domestication history and genetic relationship of cultivated rice (Londo et al., 2006).

In our study, weedy rice was sampled throughout te entire region of rice production in northern China. We assessed the genetic background and evolutionary relationships of weedy rice using genome-wide neutral markers. We investigated domestication-related genes and used a classical genetics approach to examine whether they have played an important role in the adaptation of weedy rice. The results yielded novel insights into the adaptive mechanisms of weedy rice in the paddy fields of northern China, which have valuable implications for the control of weedy rice.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant samples

For analyses involving genome-wide subspecies-specific molecular markers and domestication-related alleles for weedy rice Oryza sativa f. spontanea L. in northern China (WRNC), we collected seeds of 280 accessions directly from natural populations and preserved them at Shenyang Agricultural University, Liaoning Province, China. On the basis of phenotypic variation and cluster analysis of geographic origins, 40 core accessions were selected to cover the range of five provinces and to encompass maximum phenotypic variation and geographic origins. The set of WRNC accessions comprised 19 samples from Liaoning, five from Jilin, three from Ningxia, two from Inner Mongolia and 11 from Heilongjiang. In addition, 51 temperate japonica cultivars formerly cultivated in northern China, 14 indica rice cultivars and one aus cultivar were used as control samples. Plant material information is provided in Supporting Information Table S1.

Haplotype networks for three gene regions (p-VATPase, atpB-rbcL and SAM) provide informative pictures into the genetic relationships and domestication of rice (Londo et al., 2006). We attempted to build a genetic connection between northern weedy rice and other rice populations worldwide. All 40 core accessions of WRNC were selected for sequencing. In addition, we downloaded sequence data for O. sativa representing the major subspecies of rice and to cover the geographic and environmental range of rice cultivation, as well as the wild ancestors O. rufipogon and O. nivara, which span the entire geographic range. The dataset comprised 44 haplotypes for atpB-rbcL, 40 haplotypes for p-VATPase and 67 haplotypes for SAM.

To investigate the relationship between domestication-related genes and seed longevity, an F2 population consisting of 140 individuals was generated by crossing WRNC WR04-6 with the temperate japonica cultivar 02428.

Planting conditions and phenotyping

All sampled rice plants were inbred lines grown at Shenyang Agricultural University, Liaoning Province, China. Seeds were sown on April 15, with seedlings transplanted to their final locations on May 26. Plants were spaced 30.0 cm × 13.4 cm apart, with one seeding per hill, in three replicate plots of 0.9 m × 1.4 m. Harvesting took place on October 10. Fertilizer and water management was carried out according to the normal cultural practices followed in Shenyang.

Panicles were bagged at the heading stage. Shattering was evaluated based on the percentage of shattered seed relative to the total seed weight at maturity (Lin et al., 2007). A rice sample was considered to be ‘nonshattering’ if the percentage was close to zero, similar to that of most temperate japonica cultivars. If the percentage was at or above 90%, as in wild rice, the sample was considered to be ‘high shattering’. The range of percentages comprising ‘low shattering’ rice, which includes most indica cultivars, varied widely depending on the degree to which the panicles were shaken. (Whether or not the panicles were shaken had no effect on ‘nonshattering’ or ‘high shattering’ types.)

After harvest, panicles were stored at 4°C to maintain seed dormancy. Three panicles were incubated on two sheets of moistened filter paper in the dark at 32°C for 1 wk. The emergence of the radicle and/or plumule signified successful germination. The degree of seed dormancy was represented by the germination rates (Sugimoto et al., 2010).

Fifty hulled seeds were randomly selected to determine whether rice hulls darkened after exposure to 1–2% aqueous phenol solution.

Fully filled grains were used for the measurement of grain length, width and weight. Ten grains from each plant were lined up length-wise along a Vernier caliper to measure grain length and then arranged by breadth to measure grain width.

One hundred seeds from each F2 individual were aged at 40°C and 95% relative humidity in a phytotron for 15 d. Germination rates were then measured to assess the degree of aging resistance, an important indicator of seed longevity.

DNA isolation

Fresh leaf tissue was frozen in liquid nitrogen and total genomic DNA was extracted using the Rapid DNA Extraction Kit (Tiangen, China). Chloroplast DNA was extracted using the GMS16011.2.1 Kit (Genmed, Scientifics Inc., USA, MA, USA).

Genome-wide subspecies-specific marker analysis of weedy rice

Two molecular marker systems, indel and SSILP, were developed by comparing the draft genomic sequences of indica cultivar 93-11 and temperate japonica cultivar Nipponbare. The chloroplast DNA markers ORF100 and ORF29 can be employed to identify the subspecies type of the cytoplasm (Nakamura et al., 1998). We accordingly used indel, SSILP, ORF100 and ORF29 markers to scan the genomes of all samples. Information on the primers used is listed in Tables S2 and S3.

PCR electrophoresis bands were scored as AA (Nipponbare type), BB (93-11 type) and AB (heterozygote) for 89 indel and SSILP loci. A neighbor-joining tree was constructed on the basis of Nei's (1972) genetic distances with PowerMarker 3.25 and MEGA 4.0 software (Liu & Muse, 2005; Tamura et al., 2007).

DNA sequence variation in three gene regions

We identified haplotypes for the atpB-rbcL, p-VATPase and SAM genes in WRNC by direct sequencing of PCR products. We integrated these haplotypes with those reported by Londo et al. (2006) to establish genetic connections between WRNC and other rice populations.

We amplified PCR products from the three gene regions for all 40 WRNC accessions. Samples were sequenced with an ABI 3130xl genetic analyzer using the standard BigDye Terminator method. The primers used were published by Londo et al. (2006). Sequence data for O. rufipogon and O. sativa were downloaded from GenBank (accession nos. AM177181AM177311 and AM179944AM179987, respectively). The atpB-rbcL, p-VATPase and SAM sequences for WRNC accessions were aligned with those of O. rufipogon and O. sativa using the MegAlign program in DNASTAR (Burland, 2000).

Identification of domestication-related alleles

The deletion of DNA fragments has given rise to various alleles of domestication-related loci, such as qSW5, Phr1, Bh4 and Rc. We used molecular marker analysis and agarose gel electrophoresis to identify the alleles present for all samples. In addition, the Rc gene region presents subspecific differentiation that can be distinguished using the molecular markers RID19, RID20 and RM625 (Sweeney et al., 2007). Information on the primers used is listed in Table S4.

We used cleaved amplified polymorphic sequence (CAPs) molecular markers to identify alleles at the qSH1, Waxy and GS3 loci for all samples. The PCR products were digested with the matching restriction enzyme. Ancestral-type and derived-type alleles were identified by agarose gel electrophoresis. The primers and restriction enzymes used followed those employed by Konishi et al. (2008) and Yan et al. (2009).

Sequencing of shattering gene SH4 and dormancy gene Sdr4 coding regions was performed with an ABI 3130xl genetic analyzer using the standard BigDye Terminator method. The coding regions of the two genes are 2037 and 1032 bp, respectively (Konishi et al., 2006; Li et al., 2006; Lin et al., 2007; Sugimoto et al., 2010). The primers are listed in Table S4.

After the identification of the alleles of the domestication-related gene loci for all samples, we calculated the frequencies of functional domestication-related alleles in different groups with EXCEL 2007. We performed a correlation analysis between genotype and phenotype of the WRNC accessions to determine whether domestication and domestication-related genes were selected together with phenotypic selection. The correlation analysis was performed with SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA).

Genetic analysis to assess the relationship between domestication-related genes and seed longevity

WRNC has occurred in some abandoned fields for many years (J. Sun, pers. obs.), and is nondormant in northern China. This seed longevity sparked our interest in the mechanism of weedy rice self-production. The dramatic differences between WRNC and its coexisting cultivars involve the domestication-related traits and genes. We designed a classic genetics experiment to explore why high-frequency domestication-related alleles are present in the weedy rice population. A total of 140 F2 individuals were divided into two groups on the basis of the presence of functional or nonfunctional alleles at each of the domestication-related gene loci Rc, Phr1, Bh4 and Waxy. We then compared the germination rates between the two groups after seed aging treatments using a t-test as implemented in SPSS.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Comparative and phylogenetic analysis based on subspecific differentiation loci

The matrix presented in Fig. 1 shows the distribution of subspecies-specific loci on 12 chromosomes for all accessions. We calculated the frequencies of japonica alleles in different groups (Table 1). The average frequency of japonica alleles in WRNC was 0.95, which was almost equal to that of local temperate japonica cultivars. A neighbor-joining tree (Fig. 2) showed that WRNC shares a genetic background with japonica cultivars. All WRNC accessions exhibited japonica-type cytoplasm on the basis of their genotypes at the ORF100 and ORF29 loci.


Figure 1. Genotypic patterns for 89 indica–japonica differentiation loci on 12 chromosomes for weedy rice accessions from five provinces in northern China, japonica rice cultivars from Japan and Heilongjiang, Jilin and Liaoning Provinces, China, and indica rice cultivars. Gray squares, japonica alleles; red squares, indica alleles; yellow squares, heterozygotes.

Download figure to PowerPoint


Figure 2. Unrooted neighbor-joining tree for all sampled accessions constructed from Nei's (1972) genetic distances. Weedy rice accessions sampled from Liaoning, Jilin, Ningxia, Inner Mongolia and Heilongjiang Provinces are indicated in green, orange, gray, pink and black, respectively. Local japonica and indica cultivars are indicated in blue and red, respectively.

Download figure to PowerPoint

Table 1. Frequencies of japonica-type alleles in different groups of rice accessions for loci that differentiate indica and japonica genotypes
GroupLocal originsLocus numberNumber of japonica-type allelesTotal number of allelesFrequency
  1. HLJ, Heilongjiang; JL, Jilin; JP, Japan; LN, Liaoning; IM, Inner Mongolia; NX, Ningxia.

Weedy riceLN89314233820.93
Japonica cultivarJP89140714240.99
Indica cultivar 8935526700.13

The establishment of genetic connection based on the haplotype identification of three genes

The maternally inherited locus, the chloroplast region atpB-rbcL, contains rich genetic variation in cultivated and wild rice. As a result of lost protein-coding ability and under little selective constraint, nuclear p-VATPase presents a high resolution of the genetic relationships among Oryza species. SAM is a functional gene region that has been influenced by the forces of selection.

In this study, we did not detect any novel haplotypes at the atpB-rbcL or SAM loci in WRNC accessions. At the two loci, we detected only one haplotype each, with both haplotypes—atpB-rbcL haplotype-36 and SAM haplotype-52 described in Londo et al. (2006)—having been identified previously in cultivated and wild rice. For the atpB-rbcL gene region, there was a single japonica subspecies cluster, which included a mixture of temperate and tropical japonica varieties with haplotype-36. At the SAM locus, japonica subspecies were split into two clusters; temperate and tropical japonica accessions with haplotype-52 were found in one cluster, whereas temperate japonica samples with haplotype-C1 comprised the vast majority of samples in the other.

The pseudogene p-VATPase showed relatively high informativeness for the taxonomic evaluation of Oryza accessions. Thirty-eight WRNC accessions possessed haplotype-2 and the remaining two accessions possessed haplotype-3, both of which had been identified by Londo et al. (2006). Of the accessions analyzed by Londo et al., haplotype-2 was detected in 96% of the tropical japonica accessions and haplotype-3 was identified in the remaining 4%. Of the temperate japonica cultivars, 82% possessed haplotype-3, with the remaining 18% possessing haplotype-2, haplotype-8, haplotype-5 or haplotype-6. None of the WRNC accessions in this study shared indica (-4, -5, -39 and -40) or aus (-4, -35 and -38) haplotypes. Therefore, we infer that WRNC shared a genetic background with tropical or temperate japonica cultivars, and the similarity between the genetic backgrounds of weedy rice and its coexisting cultivars might be a result of introgression rather than inheritance from a common ancestor.

Characteristics of domestication-related traits and genes

Seed shattering

SH4 and qSH1 are two important genes involved in rice seed shattering. The nonshattering sh4 allele was fixed in all rice cultivars, resulting in a lower frequency of shattering in cultivated rice than in wild rice (Zhang et al., 2009). A single-nucleotide polymorphism (SNP) mutation in the qSH1 gene promoter region caused the loss of shattering in some temperate japonica cultivars.

All 40 WRNC accessions possessed the high seed shattering phenotype. By sequencing and alignment, we found that all samples carried the derived-type allele at the sh4 locus. The identification of the functional nucleotide polymorphism (FNP) at the qSH1 locus using CAPs markers confirmed that all 40 WRNC accessions were of the ancestral type. At the qSH1 locus, the 51 temperate japonica cultivars were of the derived type and the 14 indica and one aus cultivar were of the ancestral type. The derived type is known to exist in only a few temperate japonica populations (Zhang et al., 2009), but almost all cultivars in northern China belong to this category. Not all ancestral-type qSH1 rice plants possessed high seed shattering, such as some indica cultivars; therefore, there are other loci fixed in WRNC populations that contribute to the shattering phenotype. We believe that the high-frequency ancestral-type qSH1 allele in WRNC populations is a necessary rather than sufficient condition for reproduction.

Seed dormancy

After incubation in the dark at 30°C for 1 wk, germination frequency was 68.5 ± 16.4% for temperate japonica cultivars and 74.6 ± 25% for WRNC accessions, both of which were significantly higher (two-way ANOVA; < 0.01) than that of indica cultivars (including aus) (25.5 ± 18.8%). The Sdr4 gene contributes substantially to the differences in seed dormancy between japonica and indica cultivars (Sugimoto et al., 2010), with three identified alleles, namely Sdr4-n (japonica type), Sdr4-k (indica type) and Sdr4-k′ (indica type). By sequencing and alignment, we found that all 40 WRNC accessions carried Sdr4-n rather than the Sdr4-k or Sdr4-k′ alleles. In addition, seven of the 15 indica (including aus) and one of the 51 temperate japonica cultivars possessed an indica allele (Sdr4-k).

Pericarp color

Rc is a domestication-related gene in which a 14-bp deletion causes the rice pericarp color to change from red to white. Most wild rice accessions carried the functional Rc allele with a red pericarp, whereas, in cultivated rice, the nonfunctional allele with a white pericarp was most common. Thirty WRNC accessions possessed a red pericarp and the remaining 10 accessions possessed a white pericarp, which was consistent with the presence of the 14-bp deletion. Regardless of what pericarp color was expressed in WRNC, the Rc region was always the japonica subspecies type, as confirmed by the three molecular markers RID19, RID20 and RM652.

Hull color and phenol reaction

A 22-bp deletion in the Bh4 gene disrupts function and leads to the expression of a straw-white hull in cultivated rice. The selection of bh4 allele was an important event in the evolution of nonshattering grains during rice domestication with the visual phenotype as a selection target (Zhu et al., 2011). A black hull was a common character among WRNC accessions. In 19 WRNC accessions, a black hull was associated with functional Bh4 alleles.

The phenol reaction is a classic diagnostic trait for subspecies differentiation. Indica, aus and wild Oryza species always show a positive phenol reaction, whereas japonica subspecies show a negative reaction caused by an 18- or 29-bp deletion at the Phr1 locus. In the present study, 17 WRNC accessions that showed a positive phenol reaction carried functional alleles at the Phr1 locus. All 15 indica (including aus) cultivars and one temperate japonica cultivar that showed a positive phenol reaction carried functional alleles, and the remaining temperate japonica cultivars carried the 18-bp deletion at the Phr1 locus.

Waxy gene

Because of breeding bias and quality preference, the amylose content in rice grains has experienced different degrees of selection pressure. Derived-type Waxy is widespread in northern temperate japonica cultivars, whereas the ancestral type is dominant in indica and aus cultivars. As revealed by sequencing and CAPs molecular markers, 77.5% of WRNC accessions carried the ancestral-type Waxy allele. Nine of the 14 indica and the one aus sample carried the ancestral-type Waxy allele, and all temperate japonica cultivars were of the derived type. These data show that the ancestral-type Waxy allele is prevalent in the weedy rice population.

Grain size

The grain length : width ratio of WRNC accessions was 2.28 ± 0.31, within the range of the japonica type and significantly smaller (t-test; < 0.01) than that of indica cultivars including aus (3.36 ± 0.67). Selective pressure on grain size was associated with human cultivation during the early history of rice domestication (Purugganan & Fuller, 2009). In a previous study of rice cultivars (Konishi et al., 2008), single mutated FNPs were found for only qSW5 and Rc out of 48 possible FNP combinations of five domestication-related genes. Based on this observation and restriction fragment length polymorphism (RFLP) patterns, it was inferred that either qSW5 or Rc was subjected to artificial selection at the beginning of the japonica rice domestication process (Konishi et al., 2008). In the present study, we detected two deletion genotypes at the qSW5 locus: one had a length of 1212 bp and the other was longer; both were treated as derived-type alleles. We found that 40% of WRNC accessions, 2% of temperate japonica samples and 47% of indica cultivars (including aus) carried an ancestral-type allele at the qSW5 locus. At the GS3 locus, all WRNC accessions carried a derived-type allele. The frequencies of ancestral-type GS3 alleles in temperate japonica and indica cultivars (including aus) were 6% and 20%, respectively.

Classification of domestication-related genes in rice populations

As shown in Table 2, five categorical observations could be made regarding domestication-related genes on the basis of the functional allele frequency in WRNC, temperate japonica, indica and wild rice accessions: (i) weedy rice and cultivated rice populations showed a similar frequency at the Sh4 locus; (ii) frequencies in weedy rice and temperate japonica populations were identical at Sdr4 and GS3 loci; (iii) frequency in the weedy rice population was intermediate between that of indica (including aus) and temperate japonica cultivars at the Phr1 locus; (iv) frequencies at Rc and Bh4 loci in the weedy rice population were more similar to those of wild rice accessions than those of cultivated rice; and (v) frequencies at qSH1, Waxy and qSW5 loci in the weedy rice population were more similar to those of indica cultivars (including aus).

Table 2. Frequencies of functional domestication and domestication-related alleles in different rice populations
PopulationSh4 (i)Sdr4 (ii)GS3 (ii)Phr1 (iii)Rc (iv)Bh4 (iv)qSH1 (v)Waxy (v)qSW5 (v)
  1. Roman numerals in parentheses indicate different categories divided on the basis of the functional allele frequency.

Weedy rice in northern China0000.430.750.4810.780.4
Japonica cultivar00.020.060.0200000.02
Indica 00.470.210.07010.60.47

Ancestral types of Sh4, Sdr4 and GS3 have not been subjected to selection in the WRNC population because the frequencies of related ancestral alleles were zero. In this study, almost all presumably neutral markers in WRNC had frequencies similar to those of local temperate japonica cultivars. Compared with neutral markers sampled genome-wide, however, frequencies of ancestral-type alleles (qSH1, Phr1, Rc, Bh4, qSW5 and Waxy) in WRNC populations were much higher than in their coexisting temperate japonica cultivars (< 0.01; Table 2). In spite of continuous outcrossing with local temperate japonica cultivars, these high-frequency ancestral-type alleles persist in WRNC. The ecotypic divergence and diversification reflected in this frequency deviation from the genomic background must be a result of selection. Correlation analysis of variation between phenotypes and genotypes confirms that domestication-related alleles were selected via phenotypic selection in the weedy rice populations (Table 3). Genotypic and phenotypic data for all samples are presented in Table S1.

Table 3. Evaluation of selection of domestication-related genes and traits in weedy rice using correlation analysis
PhenotypeFunctional alleleFrequencyR Significance levelSelected or not
  1. R indicates the correlation coefficient between phenotype and functional genotype in weedy rice. **, Significant at the 0.01 level. Roman numerals in parentheses indicate different categories divided on the basis of the functional allele frequency. NA, not available.

Pericarp colorRc (iv)0.751**<0.01Selected
Hull colorBh4 (iv)0.481**<0.01Selected
Phenol reactionPhr1 (iii)0.431**<0.01Selected
Grain length : width ratioqSW5 (v)0.40.488**<0.01Selected
Amylose contentWaxy (v)0.78NANANA

Genetics analysis of aging resistance for weedy rice

One hundred and forty F2 individuals generated by crossing WRNC WR04-6 with the temperate japonica cultivar 02428 (which carries a 29-bp deletion at the Phr1 locus) were classified into two groups on the basis of whether they possessed a functional (ancestral-type) or nonfunctional (derived-type) genotype at each of the Rc, Bh4, Phr1 and Waxy loci. A t-test of germination frequency between groups showed that functional and nonfunctional groups were significantly different (P < 1.63 × 10−06) only for the Rc locus (Table 4). This result implies that the contribution of Rc to the grain longevity of weedy rice might be more important than that of the other three genes.

Table 4. Germination rate in groups of rice genotypes derived from 140 F2 individuals classified on the basis of the presence of functional or nonfunctional alleles at four domestication-related gene loci after grain aging treatment
GenotypeGermination rateSignificance level
  1. Differences between means were analyzed using Student's t-test. The sample size is presented in parentheses.Values are ± SD.

Functional Rc0.7 ± 0.27 (98)1.63E-06
Nonfunctional rc0.45 ± 0.27 (42)
Functional Bh40.64 ± 0.28 (106)0.132
Nonfunctional bh40.56 ± 0.32 (34)
Functional Phr10.64 ± 0.28 (106)0.26
Nonfunctional phr10.58 ± 0.33 (34)
Functional Waxy0.64 ± 0.29 (115)0.5
Nonfunctional waxy0.59 ± 0.33 (25)

To clarify further whether Bh4, Phr1 and Waxy genes are associated with aging resistance, we performed t-tests of germination rates in 98 individuals carrying functional Rc alleles (including heterozygotes) and in 42 individuals possessing homozygous nonfunctional rc alleles. As shown in Table 5, in functional Rc individuals, we failed to detect a significant difference in germination rate between functional and nonfunctional groups for Bh4, Phr1 and Waxy. However, for nonfunctional rc individuals, significant differences in germination frequency were observed between functional and nonfunctional groups for Bh4 (P < 0.003) and Phr1 (< 0.039), but not for Waxy. These results reveal that the domestication-related genes Rc, Bh4 and Phr1 strongly influence aging resistance in weedy rice, whereas Waxy has no significant effect on grain longevity.

Table 5. Germination frequency in groups of rice genotypes classified by the presence of functional or nonfunctional alleles at three domestication-related gene loci after grain aging treatment
GenotypeF2 individuals that carried functional Rc allele (= 98)Significance levelF2 individuals that carried nonfunctional Rc allele (= 42)Significance level
  1. Differences between means were analyzed using Student's t-test. The sample size is presented in parentheses.Values are ± SD.

Functional Bh40.70 ± 0.27 (77)0.450.52 ± 0.25 (29)0.003
Nonfunctional bh40.75 ± 0.23 (21)0.27 ± 0.24 (13)
Functional Phr10.70 ± 0.27 (75)0.610.49 ± 0.25 (32)0.039
Nonfunctional phr10.73 ± 0.24 (23)0.29 ± 0.30 (10)
Functional Waxy0.72 ± 0.24 (79)0.30.45 ± 0.28 (36)0.722
Nonfunctional waxy0.65 ± 0.33 (19)0.41 ± 0.27 (6)


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Inferences on the origin of WRNC

There are a number of possible mechanisms for the origin of weedy rice in different regions, including hybridization between cultivated rice and wild rice, reversion from a cultivar to a wild form, incidental selection from wild rice by ongoing ecological adaptation or spread to cultivated fields (De Wet & Harlan, 1975; Harlan, 1992; Bres-Patry et al., 2001; Lu & Snow, 2005) and introduction from exotic germplasm from outside the local region (Londo & Schaal, 2007; Gross et al., 2010; Reagon et al., 2010; Thurber et al., 2010).

Because of geographical isolation—WRNC sampling locations were far distant from the most northern populations of wild rice—there is no opportunity for outcrossing. Combined with our findings, WRNC is not derived directly from closely related wild Oryza species.

Deliberate introduction of germplasm of indica subspecies into temperate japonica cultivars has been largely responsible for the increased rice yield in northern China (Sun et al., 2012). Therefore, it is possible that the donation of genetic material from indica through hybridization may have led to the formation of WRNC. However, our observations regarding genome-wide subspecific differentiation and haplotypes at Rc, atpB-rbcL, SAM and p-VATPase loci do not support the hypothesis that WRNC was derived simply from the introgression of indica.

Landraces of tropical japonica existing before artificial selection, and thus possessing predomestication traits and genes, are called ‘heritage landraces’ (Cheng et al., 2003; Konishi et al., 2008). For these traits and genes, the characteristics of tropical japonica are similar to those of WRNC. In northeast China, rice was once unsuitable for cultivation because of the short frost-free period and the low effective accumulated temperature. To permit rice cultivation in northern China, early Korean immigrants attempted to select cold-tolerant cultivars (Cui, 1872). Selection for cold tolerance has also been carried out in recent breeding programs, including the use of some cold-tolerant tropical japonica cultivars for genetic improvement (Wan, 2009). These cold-tolerant tropical japonica cultivars possessed a long awn, red pericarp and black hull (Liao et al., 1921; Wan, 2009), suggesting that they may have contributed to the genetic makeup of WRNC. We hypothesize that the genomic background of weedy rice was replaced by the local cultivars and, meanwhile, its identity has been retained at loci that made it successful WRNC and elsewhere. Recent genome-wide studies have established the population structure of O. sativa (indica, aus, tropical japonica, temperate japonica and aromatic groups) (Garris et al., 2005; Caicedo et al., 2007; Zhao et al., 2010, 2011). Genome-wide SNP comparative analysis with these genetic groups should further clarify the evolutionary history of WRNC.

A viewpoint on the evolutionary forces of weedy rice populations

Unlike cultivars, wild rice and weedy forms require high levels of grain shattering and dormancy to maintain population reproduction. According to previous studies, a red pericarp, black hull and positive phenol reaction are dormancy-related characteristics (Kuriyama & Kudo, 1967; Sweeney et al., 2006; Yu et al., 2008; Gross et al., 2009). Xia et al. (2011) reported that weedy rice seeds overwinter by means of changes in the critical temperature cues for seed germination. A red pericarp, black hull and positive phenol reaction are associated with seed longevity, which contributes to the long-term maintenance of a considerable number of viable seeds in the soil when conditions are unsuitable for germination.

WRNC has stronger outcrossing habits compared with cultivars, such as higher stigma exsertion rate, stigma vigour and pollen quantity (J. Sun, pers. obs.). Introgression from conspecific crop rice can influence genetic differentiation and possibly the evolution of coexisting weedy rice populations (Xia et al., 2011). After several generations, the WRNC genomic background may be replaced by that of the cultivars. However, the genomic characteristics of weedy rice are shaped in the short term by natural selection, allowing those ancestral-type domestication-related alleles under selection to maintain the properties of weedy rice. For this reason, weedy rice populations worldwide possess certain common phenotypes, such as shattering and a red pericarp. Therefore, introgression and natural selection together shaped the evolution of weedy rice.

In past decades, WRNC was effectively managed as manual transplanting methods were used in rice production. In recent times, weedy rice has spread rapidly with the application of direct seeding and simplified cultivation practices (Yu et al., 2005). The use of less intensive management techniques constantly increases the number of underground seeds, and thus the opportunities for population expansion of WRNC.

Relationship between domestication-related genes and adaptability

The key domestication-related alleles sh4 and prog1 were fixed in cultivated rice by artificial selection. However, Rc, Bh4, Phr1, qSH1 and Waxy are also considered to be domestication-related genes (leading to genetic improvement or diversification) because they do not strongly affect cultivation in terms of the ease of harvest or sowing; they were selected upon after the initial domestication process (Gross & Olsen, 2010). During domestication, the rice genome was altered to maximize reproduction rather than survival, with the decline in population fitness replenished by cultivation management. Once less intensive management techniques are applied, such as direct seeding and simplified cultivation practices, alleles associated with domestication and ecological adaptability will reproduce at high frequency in the weedy rice population. It is therefore easy to understand why domestication-related traits and genes also contribute to the ecological adaptability of rice. Other phenomena are also not difficult to understand, such as the absence of positive selection for derived-type phr1 alleles in indica rice (Yu et al., 2008). Maintenance of seed viability in adverse natural environments may explain why the derived-type phr1 allele was not selected in indica rice, which is cultivated in areas of high temperature and humidity. By contrast, japonica rice cultivation areas are warm and dry, and such environments promote the selection of the derived-type phr1 allele to prevent grain discoloration during storage. The same phenomenon appears with respect to dormancy. In some indica cultivation regions, such as southern China and southeastern Asia, the climate may cause preharvest germination. A certain level of dormancy is desirable and has thus been maintained during cultivation and breeding processes.

Waxy rice grains lose viability more rapidly than nonwaxy rice (Juliano et al., 1990). Waxy gene mutation increases the soluble sugar content in waxy wheat, which leads to a reduction in grain storability. Such a correlation was not observed in our study, whereas the high frequency of the ancestral-type Waxy allele in WRNC was undeniable. This may indicate that other loci mask the function of Waxy or, simply, that the sample size used in the study was too small. Quantitative trait loci mapping and further functional studies should clarify the relationship between domestication-related genes and ecological adaptability.

The best strategy for reducing weed infestation is to prevent the introduction of weedy rice seeds into uncontaminated fields. Once rice fields have been infested, however, agronomic control techniques, such as crop rotation, minimum tillage, hand removal of panicles at the heading stage, seed bed preparation with moldboard plowing and water management, may be helpful (Ferrero, 2003; Delouche et al., 2007). Although agronomic techniques are useful in the control of weedy rice, the potential high costs must be considered. Cultivation of herbicide-resistant genetically modified varieties is effective for the control of weedy rice. However, there is no way to terminate them if the transgenes flow to weedy rice. Compared with transgenic technologies, the utilization of mitigation technologies, which tandemly couple (genetically link) the gene of choice (herbicide resistance) with mitigation genes, are neutral or good for the crop (Gressel & Valverde, 2009). Based on the results of our studies, the evolutionary divergence loci, such as derived-type qSH1 and Rc as mitigation genes and glufosinate, glyphosate and bentazone susceptibilities as mitigating genes, with several seasons of rotation, might be a useful attempt to control, long term, weedy rice in certain areas.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This study was supported by the National Natural Science Foundation of China (NSFC) (31271687), the China Agriculture Research System (CARS-01-13), Open Subject of State Key Laboratory of Rice Biology (090404) and the Natural Science Foundation of Liaoning, China (20102192). The authors thank Dr Yun-Bi Xu of the International Maize and Wheat Improvement Center for a critical review and valuable comments on the manuscript.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Bres-Patry C, Bangratz M, Ghesquiere A. 2001. Genetic diversity and population dynamics of weedy rice in Camargue area [France]. Genetics Selection Evolution 33: 425440.
  • Burland TG. 2000. DNASTAR's Lasergene sequence analysis software. Methods in Molecular Biology 132: 7191.
  • Caicedo AL, Williamson SH, Hernandez RD, Boyko A, Fledel-Alon A, York TL, Polato NR, Olsen KM, Nielsen R, McCouch SR et al. 2007. Genome-wide patterns of nucleotide polymorphism in domesticated rice. PLoS Genetics 3: e163.
  • Cao Q, Lu BR, Xia H, Rong J, Sala F, Spada A, Grassi F. 2006. Genetic diversity and origin of weedy rice (Oryza sativa f. spontanea) populations found in north-eastern China revealed by simple sequence repeat (SSR) markers. Annals of Botany 98: 12411252.
  • Cheng C, Motohashi R, Tsuchimoto S, Fukuta Y, Ohtsubo H, Ohtsubo E. 2003. Polyphyletic origin of cultivated rice: based on the interspersion pattern of SINEs. Molecular Biology and Evolution 20: 6775.
  • Cui Z. 1872. Jiangbei Diary (in Chinese).
  • De Wet JMJ, Harlan JR. 1975. Weeds and domesticates: evolution in the man-made habitat. Economic Botany 29: 99108.
  • Delouche JC, Labrada R, Rosell C. 2007. Weedy rices: origin, biology, ecology and control. Rome, Italy: FAO.
  • Fan C, Xing Y, Mao H, Lu T, Han B, Xu C, Li X, Zhang Q. 2006. GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theoretical and Applied Genetics 112: 11641171.
  • Ferrero A, Vidotto F, Balsari P, Airoldi G. 1999. Mechanical and chemical control of red rice (Oryza sativa L. var. sylvatica) in rice (Oryza sativa L.) pre-planting. Crop Protection 18: 251.
  • Garris AJ, Tai TH, Coburn J, Kresovich S, McCouch S. 2005. Genetic structure and diversity in Oryza sativa L. Genetics 169: 16311638.
  • Gealy DR, Tai TH, Sneller CH. 2002. Identification of red rice, rice, and hybrid populations using microsatellite markers. Weed Science 50: 333339.
  • Gressel J, Valverde BE. 2009. A strategy to provide long-term control of weedy rice while mitigating herbicide resistance transgene flow, and its potential use for other crops with related weeds. Pest Management Science 65: 723731.
  • Gross BL, Olsen KM. 2010. Genetic perspectives on crop domestication. Trends in Plant Science 15: 529537.
  • Gross BL, Reagon M, Hsu SC, Caicedo AL, Jia Y, Olsen KM. 2010. Seeing red: the origin of grain pigmentation in US weedy rice. Molecular Ecology 19: 33803393.
  • Gross BL, Skare KJ, Olsen KM. 2009. Novel Phr1 mutations and the evolution of phenol reaction variation in US weedy rice (Oryza sativa). New Phytologist 184: 842850.
  • Harlan JR. 1992. Crops and man. Crop Science Society of America. Madison, WI, USA: American Society of Agronomy.
  • Hirano HY, Eiguchi M, Sano Y. 1998. A single base change altered the regulation of the Waxy gene at the posttranscriptional level during the domestication of rice. Molecular Biology and Evolution 15: 978987.
  • Juliano BO, Perez C, Chang T. 1990. Varietal differences in longevity of tropical rough rice stored under ambient conditions. Seed Science and Technology 18: 361369.
  • Konishi S, Ebana K, Izawa T. 2008. Inference of the japonica rice domestication process from the distribution of six functional nucleotide polymorphisms of domestication-related genes in various landraces and modern cultivars. Plant and Cell Physiology 49: 12831293.
  • Konishi S, Izawa T, Lin SY, Ebana K, Fukuta Y, Sasaki T, Yano M. 2006. An SNP caused loss of seed shattering during rice domestication. Science 312: 13921396.
  • Kuriyama H, Kudo M. 1967. Complementary genes Ph and Bh controlling ripening black coloration of rice hulls and their geographical distribution. Japanese Journal of Breeding 17: 1319.
  • Labrada R. 2003. Weed Management for Developing Countries (Addendum 1). Rome, Italy: FAO.
  • Li C, Zhou A, Sang T. 2006. Rice domestication by reducing shattering. Science 311: 19361939.
  • Liang DY, Qiang S. 2011. Current situation and control strategy of weedy rice in China. China Plant Protection 31: 2124 (in Chinese).
  • Liao P, Li S, Song L. 1921. The log of Zhuanghe County (in Chinese).
  • Lin Z, Griffith ME, Li X, Zhu Z, Tan L, Fu Y, Zhang W, Wang X, Xie D, Sun C. 2007. Origin of seed shattering in rice (Oryza sativa L.). Planta 226: 1120.
  • Liu K, Muse SV. 2005. PowerMarker: an integrated analysis environment for genetic marker analysis. Bioinformatics 21: 2128.
  • Londo JP, Chiang YC, Hung KH, Chiang TY, Schaal BA. 2006. Phylogeography of Asian wild rice, Oryza rufipogon, reveals multiple independent domestications of cultivated rice, Oryza sativa. Proceedings of the National Academy of Sciences, USA 103: 9578.
  • Londo JP, Schaal BA. 2007. Origins and population genetics of weedy red rice in the USA. Molecular Ecology 16: 45234535.
  • Lu BR, Cai X, Jin X. 2009. Efficient indica and japonica rice identification based on the InDel molecular method: its implication in rice breeding and evolutionary research. Progress in Natural Science 19: 12411252.
  • Lu BR, Snow AA. 2005. Gene flow from genetically modified rice and its environmental consequences. BioScience 55: 669678.
  • Nakamura I, Urairong H, Kameya N, Fukuta Y, Chitrakon S, Sato Y. 1998. Six different plastid subtypes were found in O. sativaO. rufipogon complex. Rice Genetics Newsletter 15: 8081.
  • Nei M. 1972. Genetic distance between populations. American Naturalist 106: 283292.
  • Noldin JÉA, Chandler JM, McCauley GN. 1999. Red rice (Oryza sativa) biology. I. Characterization of red rice ecotypes. Weed Technology 13: 1218.
  • Purugganan MD, Fuller DQ. 2009. The nature of selection during plant domestication. Nature 457: 843848.
  • Reagon M, Thurber CS, Gross BL, Olsen KM, Jia Y, Caicedo AL. 2010. Genomic patterns of nucleotide diversity in divergent populations of U.S. weedy rice. BMC Evolutionary Biology 10: 180.
  • Shen YJ, Jiang H, Jin JP, Zhang ZB, Xi B, He YY, Wang G, Wang C, Qian L, Li X et al. 2004. Development of genome-wide DNA polymorphism database for map-based cloning of rice genes. Plant Physiology 135: 11981205.
  • Shomura A, Izawa T, Ebana K, Ebitani T, Kanegae H, Konishi S, Yano M. 2008. Deletion in a gene associated with grain size increased yields during rice domestication. Nature Genetics 40: 10231028.
  • Smith R Jr. 1979. How to control the hard-to-kill weeds in rice. Weeds Today 10: 1214.
  • Sugimoto K, Takeuchi Y, Ebana K, Miyao A, Hirochika H, Hara N, Ishiyama K, Kobayashi M, Ban Y, Hattori T et al. 2010. Molecular cloning of Sdr4, a regulator involved in seed dormancy and domestication of rice. Proceedings of the National Academy of Sciences, USA 107: 57925797.
  • Sun J, Liu D, Wang JY, Ma DR, Tang L, Gao H, Xu ZJ, Chen WF. 2012. The contribution of intersubspecific hybridization to the breeding of super-high-yielding japonica rice in northeast China. Theoretical and Applied Genetics 125: 11491157.
  • Sweeney MT, Thomson MJ, Cho YG, Park YJ, Williamson SH, Bustamante CD, McCouch SR. 2007. Global dissemination of a single mutation conferring white pericarp in rice. PLoS Genetics 3: e133.
  • Sweeney MT, Thomson MJ, Pfeil BE, McCouch S. 2006. Caught red-handed: Rc encodes a basic helix–loop–helix protein conditioning red pericarp in rice. Plant Cell 18: 283294.
  • Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 15961599.
  • Tang L, Ma DR, Xu ZJ, Deng HF, Chen WF, Yuan LP. 2011. Utilization of weedy rice for development of japonica hybrid rice (Oryza sativa L.). Plant Science 180: 733740.
  • Thurber CS, Reagon M, Gross BL, Olsen KM, Jia Y, Caicedo AL. 2010. Molecular evolution of shattering loci in U.S. weedy rice. Molecular Ecology 19: 32713284.
  • Vaughan LK, Ottis BV, Prazak-Havey AM, Bormans CA, Sneller C, Chandler JM, Park WD. 2001. Is all red rice found in commercial rice really Oryza sativa? Weed Science 49: 468476.
  • Wan JM, 2009. Breeding and pedigree of Chinese rice. Beijing, China: China Agriculture Press, 10: 312314 (in Chinese).
  • Wang X, Zhao X, Zhu J, Wu W. 2005. Genome-wide investigation of intron length polymorphisms and their potential as molecular markers in rice (Oryza sativa L.). DNA Research 12: 417427.
  • Xia HB, Xia H, Ellstrand NC, Yang C, Lu BR. 2011. Rapid evolutionary divergence and ecotypic diversification of germination behavior in weedy rice populations. New Phytologist 191: 11191127.
  • Yan CJ, Yan S, Yang YC, Zeng XH, Fang YW, Zeng SY, Tian CY, Sun YW, Tang SZ, Gu MH. 2009. Development of gene-tagged markers for quantitative trait loci underlying rice yield components. Euphytica 169: 215226.
  • Yu G, Bao Y, Shi C, Dong C, Ge S. 2005. Genetic diversity and population differentiation of Liaoning weedy rice detected by RAPD and SSR markers. Biochemical Genetics 43: 261270.
  • Yu Y, Tang T, Qian Q, Wang Y, Yan M, Zeng D, Han B, Wu CI, Shi S, Li J. 2008. Independent losses of function in a polyphenol oxidase in rice: differentiation in grain discoloration between subspecies and the role of positive selection under domestication. Plant Cell 20: 29462959.
  • Zhang LB, Zhu Q, Wu ZQ, Ross-Ibarra J, Gaut BS, Ge S, Sang T. 2009. Selection on grain shattering genes and rates of rice domestication. New Phytologist 184: 708720.
  • Zhao K, Tung CW, Eizenga GC, Wright MH, Ali ML, Price AH, Norton GJ, Islam MR, Reynolds A, Mezey J. 2011. Genome-wide association mapping reveals a rich genetic architecture of complex traits in Oryza sativa. Nature Communications 2: 467.
  • Zhao K, Wright M, Kimball J, Eizenga G, McClung A, Kovach M, Tyagi W, Ali ML, Tung CW, Reynolds A. 2010. Genomic diversity and introgression in O. sativa reveal the impact of domestication and breeding on the rice genome. PLoS ONE 5: e10780.
  • Zhao X, Yang L, Zheng Y, Xu Z, Wu W. 2009. Subspecies-specific intron length polymorphism markers reveal clear genetic differentiation in common wild rice (Oryza rufipogon L.) in relation to the domestication of cultivated rice (O. sativa L.). Journal of Genetics and Genomics 36: 435442.
  • Zhu BF, Si L, Wang Z, Zhou Y, Zhu J, Shangguan Y, Lu D, Fan D, Li C, Lin H et al. 2011. Genetic control of a transition from black to straw-white seed hull in rice domestication. Plant Physiology 155: 13011311.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

nph12012-sup-0001-TableS1.xlsapplication/msexcel56KTable S1 Properties of weedy rice and Oryza accessions used in this study (genotypic and phenotypic characteristics of rice accessions are also listed)
nph12012-sup-0002-TableS2.xlsapplication/msexcel27KTable S2 Information on subspecies-specific indel markers used for genome-wide scanning, including chromosomal location, primer sequences and PCR product size
nph12012-sup-0003-TableS3.xlsapplication/msexcel33KTable S3 Information on subspecies-specific intron length polymorphism (ILP) markers used for genome-wide scanning, including chromosomal location, primer sequences and PCR product size
nph12012-sup-0004-TableS4.xlsapplication/msexcel27KTable S4 Molecular marker information used for the identification of domestication-related and taxonomic alleles, including marker types, primer sequence and restriction enzyme