Introduction
The genus
Oryza comprises more than 20 species including two cultivated species,
O. sativa L. and
O. glaberrima Steud (
Vaughan, 1994). The number of rice chromosome was 2
n = 24 or 48, and the genomes of species in
Oryza are classified into 10 types: AA, BB, CC, BBCC, CCDD, EE, FF, GG, HHJJ, and HHKK (
Lu, 1999;
Vaughan, 2003). Historically, advances in rice chromosome research are quite limited, as the small chromosome of
Oryza, similar configuration, and unconspicuous centromere make it difficult to identify homologs and distinguish different chromosomes to generate a karyotype by morphological means (
Shishido et al., 2001).
With the first
in situ hybridization on rice chromosomes reported in 1987, using I
125 labeled rRNAs, the genomes
in situ hybridization (GISH) on rice chromosomes and chromosome painting developed very quickly for rice chromosome analysis. Repetitive DNA sequences are either organized in tandem arrays or dispersed throughout the whole genome (
Harrison and Heslop-Harrison, 1995;
Ohmido et al., 2000). The DNA sequences of a repeat and its copy number at each chromosome site can all be evolved rapidly, leading to its specificity to a certain genome species and even a chromosome (
Wang et al., 1995;
Matyásek et al., 1997;
Uozu et al., 1997;
Hall et al., 2004). In this respect, the genome specific repetitive DNA sequence is a proper molecular marker in analyzing the nuclear genome chromosome composition and derivatives (
Cheng et al., 2001a;
Cheng et al., 2002;
Mishima et al., 2002;
Yan et al., 2002;
Jin et al., 2004;
Jin et al., 2005). During the last decade, researches on specific repetitive DNA sequence, such as rDNA, DNA transposon, and centromere DNA showed that specific repetitive DNA site number is one of the most important evolutionary dynamics of genome (
Zwick et al., 2000;
Mishima et al., 2002;
Hall et al., 2004;
Jiang et al., 2004). It has been suggested that one of the main factors that determine the variation in genome size is the quantity of repetitive DNA sequences (
Uozu et al., 1997;
Jiang et al., 2004). Therefore, repetitive sequences are likely to play an important role in genome differentiation during speciation. (
Jiang et al., 1996;
Uozu et al., 1997;
Jiang et al., 2004;
Yan et al., 2005)
C0t-1 DNA contains mainly highly and moderately repetitive DNA sequences, including satellites, microsatellites, and those located in centromeres and telomeres, which determine the chromosome structure and characterization (
Zwick et al., 1997). In our previous study, we used the
C0t-1 DNA of Guang Lu Ai 4 of
O. sativa, as a probe for
in situ hybridization on mitotic metaphase chromosomes of
O. sativa,
O. officinalis, and
O. meyeriana to analyze signals distribution of the repetitive sequences in these three genomes (
Lan et al., 2007). The C genome
C0t-1 DNA was also hybridized to the somatic chromosome preparations of
O. latifolia (CCDD),
O. alta (CCDD), and
O. punctata (BBCC) to analyze the relationships among B, C, and D genomes by distribution of
C0t-1 DNA signals on chromosomes of different genomes (
Lan et al., 2006). However, there are not much reports about the comparative karyotypic analysis of A and C genomes with
C0t-1 DNA signal bands in the genus
Oryza. At present, many RFLP (Restriction Fragment Length Polymorphism) clones have been genetically mapped on rice chromosomes (http://rgp.dna.affrc.go.jp) (
Causse et al., 1994;
Kurata et al., 1994;
Harushima et al., 1998).
The RFLP markers have been used for indicating or localizing genes directly on rice chromosome with fluorescence
in situ hybridization (FISH) (
Song and Gustafson, 1995;
Ohmido et al., 1998;
Cheng et al., 2001b;
Qin et al., 2001). RFLP markers have not been applied extensively in comparative karyotype with FISH method for localizing nucleotide sequences directly to chromosomes or interphase nuclei. The RFLP marker, R2676, from the chromosome 8 of
O. sativa as well as
O.
officinalis through comparative genetic mapping analysis (
Tan et al., 2005). Thus, the same RFLP marker is used as a probe to somatic hybrid on chromosomes for comparative karyotypic analysis between
O. glaberrima and
O. officinalis. This method has been used as a powerful technique to study small plant chromosome comparative karyotype in our laboratory before (
Qin et al., 2001).
In this report, we first demonstrated the comparative karyotypic analysis based on the comparative RFLP map between
O. sativa and
O. officinalis (
Tan et al., 2005). Based on the
C0t-1 DNA banding on their chromosomes, we further analyzed the
O. sativa,
O. glaberrima, and
O. officinalis karyotype. Furthermore, we used total genomic DNA as a probe for hybridization on the chromosomes of their own to contrast with the results of the
C0t-1 DNA distribution on their chromosomes. Then, in light of the
C0t-1 DNA banding on homologous chromosomes in different species, we elucidated what kind of genetic accident occurred during these species genome evolution, and the role of repetitive DNA sequences to the genome size in the genus
Oryza was also discussed.
Materials and methods
Materials and chromosome preparations
The rice variety “Guang Lu Ai 4” (
O. sativa L.) was provided by Professor Zuokui Zeng at the Hubei Academy of Agriculture Sciences, China.
O. glaberrima Steud was provided by academician Yonggen Lu at the Huanan Agricultural University, China. Accession 1589 of
O. officinalis Wall. ex Watt was supplied by the National Wild Rice Garden in Guangdong Province, China, which possesses a number of valuable traits, such as resistance to pests and diseases and high protein content in grains. The RFLP marker R2676 (1.8 kb), hybridized exclusively to the long arm of the chromosome 8 of
O. glaberrima and
O. officinalis, was supplied by the Institute of Genetics, Wuhan University, China. Chromosome preparations were performed following the method by Ren et al. (
1997).
Preparation of C0t-1 DNA
Total genomic DNA of
O. officinalis was extracted by following the CTAB method according to Doyle (
1990). The
C0t-1 DNA preparation was based on the work of Zwick et al. (
1997). Genomic DNA was sterilized under 0.14 Mpa for 3-15 min and broken into the fragments of 800-1500 bp. DNA was reannealed at 65 write units for the required time calculated according to the formula
C0t-1 = mol/L × Ts. Then, S
1 nuclease (2 U/µg DNA, Promega, USA) was added. The tube was placed in a water bath at 37ºC for 1 h. The DNA was extracted twice by equal volumes of phenol–chloroform, precipitated with isopropanol, washed with prechilled 70% ethanol, and dried and resuspended in TE buffer. The
C0t-1 DNA was quantified and stored at -20ºC before it was used.
Probe labeling
The C0t-1 DNA of O. officinalis was labeled with biotin–11–dUTP using a nick translation kit (Roche, Germany) according to the manufacturer’s instructions. The length of the C0t-1 DNA that was used as the probe for FISH was estimated by gel electrophoresis to be 300 bp to 500 bp. The labeled probe was separated from unincorporated nucleotides by passage through Sepharose CL–6B (Sigma, USA) column. Incorporation of biotin–11–dUTP was evaluated by means of dot blots following the alkaline phosphatase conjugate (Roche, Germany) detection.
In situ hybridization and image observation
Fluorescence
in situ hybridization (FISH) was carried out following the method of Li et al. (
2001) with some modifications (
Li et al., 2001). The chromosome preparations were dried at 60ºC for 1 h, pretreated with RNase A (Promega, USA) at 37ºC for 1 h, and followed by washing twice for 5 min in 2 × SSC at 37ºC. Chromosomal DNA was denatured by immersing the slides in 70% deionized formamide (Sigma, USA) in 2 × SSC at 70ºC for 3.5-5.0 min, dehydrated in an ice–cold ethanol series (70%, 95% and 100%), and then air dried.
The hybridization mixture consisted of 50% deionized formamide, 8% (w/v) sodium dextran sulfate (Amresco, USA), 0.5% (w/v) SDS, 0.5 µg of salmon sperm DNA (Sigma, USA) in 2 × SSC, and 80 ng of labeled probes for each slide. After being denatured in boiled water for 10 min, the mixture was immediately chilled in ice for 10 min, 50 µL of the hybridization mixture was applied per slide and covered with a plastic coverslip. The slides were placed in a humidity chamber and denatured at 90ºC for 10 min and then incubated at 37ºC overnight. For detection of the fluorescent signals, the slides were washed first in SSC and then in PBS at room temperature. The detection of the biotin-labeled probe and digoxigenin-labeled probe was achieved with streptavidin-Cy3 (Rockland, USA) and anti-digoxigenin–FITC (Roche), respectively. The preparations were counterstained with 4’, 6-diamidino-2-phenylindole (DAPI) or Propidium iodide (PI). The chromosomes were observed with an Olympus BX61 fluorescence microscope and photographed with Cool–1300QS CCD controlled by Manager Expo 2.1.1 imaging system. Karyotypic analysis was carried out using the chromosome analyzing system FISHView EXPO 2.0 software. The chromosome relative lengths were measured by SPOT advanced software.
Results
Fluorescence in situ hybridization (FISH) with C0t-1 DNA and GISH
O. sativaC0t-1 DNA and their own total genomic DNA were used as probes for
in situ hybridization on mitotic metaphase chromosomes of
O. sativa,
O. glaberrima, and
O. officinalis (Figs. 1A–F) and FISH with its own
C0t-1 DNA to mitotic metaphase chromosomes of
O. officinalis (Figs. 1G and H). The hybridization stringency and the washing stringency were 65%-75% and 80%-90%, respectively (
Li et al., 2001). Under such conditions, the hybridization signals were highly specific. Figures 1A, C, E, F, and H show the red hybridization signals of
C0t-1 DNA probes on chromosomes of
O. sativa,
O. glaberrima, and
O. officinalis, respectively, which indicates that
C0t-1 DNA signals are distributed mainly in areas of centromeres, subcentromeres, and telomeres, with much fewer ones on the middle regions of chromosome arms. There were strong signals on the
O. sativa and
O. glaberrima chromosomes (Figs. 1A and C), but there were rather weak signals on the
O. officinalis chromosomes by using the
O. sativa C0t-1 DNA as probe (Fig. 1E). The
O. sativaC0t-1 DNA signals banding were not distinct on the
O. officinalis chromosomes (Fig. 1E), but the
O. officinalis C0t-1 DNA signals banding were distinct, and the strong hybridization signals were distributed on the
O. officinalis chromosomes (Figs. 1G and H). By contrast, total genomic DNAs were used as probes for GISH on the chromosomes of
O. sativa,
O. glaberrima, and
O. officinalis, respectively, which indicated that strong red signals covered the whole chromosomes (Figs. 1 B, D, and F).
Chromosome karyotype analysis using C0t-1 DNA signals banding
The relative lengths of chromosomes O. sativa, O. officinalis, and O. meyeriana were measured by SPOT advanced software, and the karyotypes were reconstructed using FISHView EXPO 2.0 software (Fig. 1J, K and L). Figures 2A–D are karyotype idiograms of O. sativa, O. glaberrima, and O. officinalis, where the upper part of each is a combined idiogram, and the lower part shows the red hybridization signals of C0t-1 DNA. Karyotypic analysis of O. sativa was performed according to standard karyotype of International Rice Institute, and the conventional method was also adopted in O. glaberrima and O. officinalis. According to FISH images, the signals of C0t-1 DNA have specific band patterns on each chromosome, and the homologous chromosomes show the same banding patterns. Therefore, the ideograms of the species were constructed according to the relative length, arm ratio, and position of C0t-1 DNA signals of the chromosomes (Figs. 2A–D). The relative lengths of chromsomes were measured by SPOT advanced software. These data were summarized in Table 1 and graphically reported in Figs. 2E–H. Chromosomes were conventionally ordered in the idiograms according to Figs. 2A–D. The idiograms show that some chromosomes have the same C0t-1 DNA banding, i.e., chromosomes 1, 4, 5, 6, 8, and 12, of the two cultivated species, namely, O. sativa and O. glaberrima (Figs. 2E and F), whereas there are much more differentiations of distribution of O. sativaC0t-1 DNA on chromosomes between O. officinalis and cultivated rice (Figs. 2E, F and G). Besides, the distribution of O. officinalis C0t-1 DNA and O. sativaC0t-1 DNA on the O. officinalis differs from each other (Figs. 2G and H).
Comparative karyotype analysis by RFLP– FISH
The RFLP marker R2676 (1.8 kb) was found in chromosome 8 of cultivated rice, as well as in chromosomes of
O. officinalis based on the comparative RFLP map (
Tan et al., 2005). Figures 3A, B, E, and F show the results of FISH experiments using R2676 marker. Figures 3A and B show that the R2676 marker signals were detected in metaphase and interphase chromosomes of
O. glaberrima, respectively. The metaphase and prophase chromosomes of
O. officinalis were also probed with R2676 marker, respectively (Figs. 3E and F). Figures 3C and G show enlarged idiograms of the signal-tagged chromosomes. Clear double fluorescent signals were detected at an interstitial region of chromosome and at the end of chromosome of
O. glaberrima and
O. officinalis. It was clear that the two signals were localized at the region on the long arm of chromosome and using five metaphase spreads possessing a similar degree of condensation, and we measured and processed chromosomal lengths through chromosome analyzing system software (Table 2). The data of karyotypic analysis showed that the R2676 marker was located on chromosome 4 of
O. glaberrima and chromosome 7 of
O. officinalis, respectively. The distance of R2676 signals from the centromere of chromosome 4 of
O. glaberrima and chromosome 7 of
O. officinalis were 4.71±1.19 and 3.71±2.51; in addition, the marker was physically mapped at the 86.20% and 91.56% position from the centromere, respectively. The RFLP marker was physically mapped on the chromosome idiograms, as shown in Figs. 3D and H.
Discussion
The repetitive DNA sequences, as an important characteristic of eukaryotic genomes, disperse in the whole genome. They are usually more than 50% in genomes of higher plants (
Flavell et al., 1974;
Bennetzen et al., 2005), such as rice, maize, wheat, rye, and lemon. The content percentages of repetitive sequences are 50%, 78%, 83%, 92%, and 95%, respectively (
Bennetzen et al., 2005).
C0t-1 DNA contains mainly highly and moderately repetitive DNA sequences. In this study, the fluorescence
in situ hybridization was applied to the somatic chromosome preparations of
O. sativa,
O. glaberrima and
O. officinalis with
C0t-1 DNA as a probe. The results showed that the specific signal bands of
C0t-1 DNA developed on the chromosomes of
O. sativa,
O. glaberrima, and
O. officinalis (Figs. 1A, C, E, G, and H).The homologous chromosomes exhibited similar signal bands; therefore, karyotypic analysis was conducted based on
C0t-1 DNA specific band possible in the three species. This karyotypic analysis based on
C0t-1 DNA signal bands differed from the conventional karyotypic analysis. The chromosomes of
Oryza were small and similar in configuration and inconspicuous centromere, so it was difficult to identify homologs and distinguish different chromosomes to generate a karyotype by morphological means. Figures 2D–F and H show the idiograms of chromosomal locations of
C0t-1 DNA in
O. sativa,
O. glaberrima, and
O. officinalis, and the results visually described the location of the repetitive sequences on chromosomes. There were more different distributions of repetitive sequences of each chromosome in a species, even in the two cultivated species of
O. sativa and
O. glaberrima, and there were very few chromosomes with similar
C0t-1 DNA bandings, i.e., chromosomes 1, 4, 5, 6, 8, and 12. However, the similarity and correspondence between
C0t-1 DNA signal patterns and genomic DNA signal patterns indicated that the self-GISH signals actually resulted from the hybridization of genomic repetitive sequences to the chromosomes. On the other hand, nonuniform distribution between
C0t-1 DNA signal on their own chromosome of
O. officinalis and
O. sativa C0t-1 DNA on the chromosomes of
O. officinalis was observed. There were much fewer
O. sativa C0t-1 DNAs on the chromosomes of
O. officinalis (Figs. 1E, G, and H). The genome size of
O. sativa (450 Mb) differed from
O. officinalis (697 Mb) (
Uozu et al., 1997). In the evolutionary process, one of the important reasons for genome aggrandizement was the result of reduplication of highly and moderately repetitive DNA sequences (
Flavell, 1986;
Uozu et al., 1997;
Kumar and Bennetzen, 1999). We speculated that repetitive sequences reduplication of the structure of AA genome (cultivated rice) was distinguished from CC genome (
O. officinalis); in addition, the transposition and recombination of genome were also made differently between two species with the same genome, such as
O. sativa and
O. glaberrima (AA).
Figures 1B, D, and E show the GISH idiograms of
O. sativa,
O. glaberrima, and
O. officinalis hybridized with total genomic DNA probes in their own mitotic metaphase. The hybridization stringency and the washing stringency were 65%-75% and 80%-90%, respectively. The posthybridization stringency is a very important factor to distinguish homology and similarity in the sequences in GISH. Li et al. (
2001) reported that under low stringency (50%-60%), identity could not be distinguished in similarity genomes, and under 78%-86% stringency, the identity could be distinguished mostly (
Li et al., 2001). The AA genome and CC genome existed in highly similar sequences; the results showed that, under the washing stringency at 80%-90%, the hybridization signals of
C0t-1 DNA in target sequences were highly specific. By contrast, the GISH could enhance the reliability of
C0t-1 DNA
in situ hybridization, which indicates that the strong hybridization signals (red) covered the entire length of chromosomes of genome probes (Figs. 1B, D, and F) and also indicates clearly the locations of
C0t-1 DNA on genome. Figures 1A, C, E, G, and H show that variable hybridization signals painted some specific regions of chromosomes.
RFLP marker-based genetic linkage maps have been constructed in numerous plant species. Much less effort has been devoted to physical mapping during the last decade although it can be applied to different species (
Cheng et al., 2001c,
2001d). There are several methods for physical mapping used in plant species, for example, physical maps generated through a BAC contig assembly with large genome. However, it will be technically difficult to correctly achieve specific contigs in polyploid species or extensive sequence duplications. Secondly, the method to generate a physical map is to locate genetically mapped DNA markers to specific chromosomal segments using cytogenetic stocks (
Werner et al., 1992;
Riera-Lizarazu et al., 2000), but it is a daunting task to isolate a large number of cytogenetic stocks. Further cytogenetic stocks cannot be applied in all species. Another method for generation of a physical map is to directly visualize DNA sequences on chromosomes by fluorescence
in situ hybridization (FISH) (
Song and Gustafson, 1995;
Jackson et al., 2000;
Ohmido et al., 2000;
Cheng et al., 2001c,
2001d;
Zhao et al., 2002).
Comparative genetic maps of the cultivated rice (AA) with
O. officinalis (CC) were constructed and used to name the two chromosomes with the strongest colinearity of the two species with the same numbers despite the configuration trait (
Tan et al., 2005). We used the RFLP marker R2676 as the probe to hybridize it on the chromosomes of
O. officinalis and named the chromosome with the detected signal number 8. The configuration trait of the chromosome 8 of
O. officinalis is shown in Table 2. This chromosome 8 of
O. officinalis was exactly the one we named number 7 by using tradition karyotype analysis (Fig. 2J and Fig. 3H). Our results revealed that the relative length of
O. officinalis chromosome 8 does not follow conventional chromosome length in descending order of number. Similarly, chromosome 4 of
O. officinalis named by using molecular marker RG214 was not the one named by using tradition karyotype analysis before (
Qin et al., 2001). Figure 2J shows that
C0t-1 DNA of A genome signals were detected in the end of the short arm of
O. officinalis chromosome 8, indicating a relatively high homology in the repetitive DNA sequences in this region between A genome of cultivated rice and A genome of
O. officinalis. However, the fluorescence intensity on the chromosomes of
C0t-1 DNA of A genome was less than that of its own C genome from
O. officinalis. As
O. officinalis genomes (697 Mb) were much larger than
O. glaberrima genomes (420 Mb) and O. sativa genomes (449 Mb) (
Uozu et al., 1997), which may be the result of reduplication of the highly and moderately repetitive DNA sequences in
O. officinalis or the elimination of those sequences in
O. glaberrima and
O. sativa or species–specific sequences forming in the three species during long period of evolution, we concluded that the highly and moderately repetitive DNA sequences reduplication was one of the important causes due to the genome enlargement in
O. officinalis. No homology signal of
C0t-1 DNA from
O. sativa was detected in the end of the long arm of
O. glaberrima (Fig. 2J).
O. sativa is grown worldwide but mostly in South and South-east Asia. Another cultivated A genome rice
O. glaberrima is grown in western Africa. This suggested that the repetitive DNA sequences of A genome in two cultivated rice were lost for differentiations of growth environment in the evolutional history. Furthermore, our study was based upon comparative genomics and could enhance the efficiency of studying evolution of chromosomes, genomes, and repetitive sequences. It could be beneficial for making comparative analysis of single or repetitive sequences and consequently to accelerate the establishment of the large genetic system for
Oryza plants, even for the genera of grass family.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(305KB)
