Qilin Zhu，Zexi Cai，Qilin Tang，Weiwei Jin*

aNational Maize Improvement Center of China，Beijing Key Laboratory of Crop Genetic Improvement，Coordinated Research Center for Crop Biology，China Agricultural University，Beijing 100193，China

bMaize Research Institute，Sichuan Agricultural University，Wenjiang 611130，China

?

Repetitive sequence analysis and karyotyping reveal different genome evolution and speciation of diploid and tetraploid Tripsacum dactyloides

Qilin Zhua，1，Zexi Caia，1，Qilin Tangb，Weiwei Jina，*

aNational Maize Improvement Center of China，Beijing Key Laboratory of Crop Genetic Improvement，Coordinated Research Center for Crop Biology，China Agricultural University，Beijing 100193，China

bMaize Research Institute，Sichuan Agricultural University，Wenjiang 611130，China

A R T I C L E I N F O

Article history:

in revised form 7 April 2016

Accepted 6 June 2016

Available online 16 June 2016

Tripsacum

Low-coverage sequencing

Repetitive sequence

Karyotype

Phylogenetic analysis

A B S T R A C T

In the subtribe Maydeae，Tripsacum and Zea are closely related genera.Tripsacum is a horticultural crop widely used as pasture forage.Previous studies suggested that Tripsacum might play an important role in maize origin and evolution.However，our understanding of the genomics and the evolution of Tripsacum remains limited.In this study，two diploids，T.dactyloides var.meridionale（2n=36，MR）and T.dactyloides（2n=36，DD），and one tetraploid，T.dactyloides（2n=72，DL）were sequenced by low-coverage genome sequencing followed by graph-based cluster analysis.The results showed that 63.23%，59.20%，and 61.57%of the respective genome of MR，DD，and DL were repetitive DNA sequence.The proportions of different repetitive sequences varied greatly among the three species.Fluorescence in situ hybridization（FISH）analysis of mitotic metaphase chromosomes with satellite repeats as the probes showed that the FISH signal patterns of DL were more similar to that of DD than to that of MR.Comparative analysis of the repeats also showed that DL shared more common repeat families with DD than with MR.Phylogenetic analysis of internal transcribed spacer region sequences further supported the evolutionary relationship among the three species.Repetitive sequences comparison showed that Tripsacum shared more repeat families with Zea than with Coix and Sorghum.Our study sheds new light on the genomics of Tripsacum and differential speciation in the Poaceae family.

?2016 Crop Science Society of China and Institute of Crop Science，CAAS.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

The genus Tripsacum belongs to Maydeae of the Poaceae family［1］.The basic number of chromosomes in Tripsacum is n=18［2］.However，the nine species in genus Tripsacum show different ploidy:diploid（2n=2×=36）and tetraploid（2n= 4×=72）［3］.Tripsacum dactyloides，also commonly known as Eastern gamagrass，is a horticultural and forage crop that originated in temperate regions in America［4］.Tripsacum has been proposed as one of the progenitors of maize（Zea mays L.）［5］，or as having played a pivotal role in maize（Zea mays L.）evolution［6-9］.As a high-protein forage crop with good resistance to insects and drought［10，11］，Tripsacum has been proposed to have great potential in agriculture and plant breeding［12］.Previous studies of Tripsacum focused mainly on using Tripsacum to improve maize and on karyotype comparison［13，14］.However，genomic information for Tripsacum is extremely sparse.

In the genomes of most eukaryotes，repetitive DNA sequences are abundant［15］.Because repetitive sequences do not harbor genes，they were once considered useless［16，17］.However，more recent studies have shown that the accumulation of repetitive sequences played an important role in fueling genome expansion and variation［18，19］. Eukaryotic genomes contain various repetitive DNA sequences，suchasinterspersedrepeatedsequencesand tandemly repeated sequences［20，21］.Interspersed repeated sequencesareusuallydispersed throughoutagenome，whereas tandemly repeated sequences are arranged in a tandem configuration and always occupy a small proportion of a genome［22］.Satellite DNA，which is one type of tandemly repeated DNA sequence，is always located in subtelomeric and pericentromeric regions and sometimes appears at a functional centromere［23］.Satellite DNA evolves rapidly in eukaryotic genomes and thus shows varied sequences，copy numbers，and chromosome locations in different species［24］. It is thus useful for karyotyping and comparative genome analysis［25，26］.

Genome-wide characterization of repetitive sequences can be accomplished only with large volumes of sequencing data，which have been acquired in only a few model species by classical sequencing methods［27］.Bioinformatic scientists formerly faced a large challenge in identifying repetitive sequences in short sequencing reads of non-reference species because of the difficulty in assembling repetitive regions，particularly tandemly repeated regions［28］.Fortunately，the graph-based cluster algorithm［29］can be used to characterize repetitive sequences from low-coverage sequencing data efficiently.

In this study，approximately 7GB，10GB，and 8GB genomic sequences of T.dactyloides var.meridionale（2n=36，MR），T.dactyloides（2n=36，DD），and T.dactyloides（2n=72，DL）were sequenced.We used bioinformatic and cytological analyses to investigate the genome structure and the composition of repetitive sequences in the three Tripsacum species. Our comparative analysis of repetitive sequences sheds new light on the genomics of MR，DD and DL and the phylogenetic relationships between the three Tripsacum species and their relatives in Maydeae of the Poaceae family.

2.Materials and methods

2.1.Plant materials

T.dactyloides（2n=36，DD），T.dactyloides var.meridionale（2n=36，MR），and T.dactyloides（2n=72，DL）were used in this study.T.dactyloides（2n=36，DD）and T.dactyloides var. meridionale（2n=36，MR）are from Germplasm Resources Information Network.T.dactyloides（2n=72，DL）were obtained from Professor Qilin Tang，Maize Research Institute，Sichuan Agricultural University.Plants grown from seeds were cultivated in a greenhouse at China Agricultural University.

2.2.Genome size estimation

The genome sizes of T.dactyloides（2n=36，DD），T.dactyloides var.meridionale（2n=36，MR），andT.dactyloides（2n=72，DL）were estimated by flow cytometry［30］.Fresh leaves of Tripsacum and Zea mays B73 were harvested and cut up in lysis buffer（0.74% Na2EDTA，0.18%Tris，1.1%β-mercaptoethanol，5.8%KCl，0.01% spermine and 0.1%Triton X-100）for 10 min.The mixture was filtered through nylon mesh（30 μm）and centrifuged for 10 min and the supernatant was discarded.The PI solution（50 μg mL-1）wasaddedtothepellettostainDNA.TheG1nuclei fluorescence intensity was measured with a flow cytometer（FACSCalibur，BD，USA）with 10，000 particles per run.

2.3.DNA extraction，genome sequencing，and internal transcribed spacer（ITS）sequence amplification

DNA was extracted using the standard cetyltrimethyl ammonium bromide（CTAB）extraction protocol［31］.The DNA was sequenced using a HiSeq 2000 platform （BerryGenomics）. 100 bp paired-end reads were obtained.ITS sequences were amplified with the following primers，Pl:5′-TCGTAACAAGG TTTCCGTAGG-3′and P2:5′-TCCTCCGCTTATTGATATGC-3′［32］.Approximately 600 bp fragments were sequenced.The phylogenetic tree was constructed with MEGA6.0.

2.4.Data analysis

After removing the low-quality reads，repetitive sequence assembly was conducted by graph-based clustering following Novak et al.［29］.RepeatMasker［33］was used to identify repeat type，and Reversed Position Specific-Blast（RPS-Blast）was used to detect conserved protein domains［34］.Tandem Repeats Finder was used to identify satellite monomer within contig sequences［35］.The number of reads of each cluster represented the frequency of the cluster in the genome.

To confirm the distribution of repeats in different species，the reads were labeled with species names and a combined datasetwasconstructedforthegraph-basedclustering analysis.To analyze the repetitive sequences among different subtribes，another combined dataset was built.Paired-end data of Zea mays B73，Sorghum bicolor BTX623，and Z.luxurians were obtained from DNAnexus（http://sra.dnanexus.com/）. These two combined datasets were analyzed according to the methods described above.

Using the Z.mays B73 as the reference，the data of the species of Tripsacum were masked with RepeatMasker［33］.

2.5.Probe labeling and fluorescence in situ hybridization（FISH）

Root tips of T.dactyloides（2n=36），T.dactyloides var.meridionale（2n=36），and T.dactyloides（2n=72）were harvested and treated with N2O at 2 atm for 2 h，then kept in fixing solution（ethanol:glacial acetic acid=3:1）for at least 4 h.Root tips were digested in cellulose and pectinase for 2 h at 37°C，squashed on glass slides in the fixing solution with finepointed forceps，and dried over an alcohol flame.

Table 1-Genome size estimation of T.dactyloides（2n=36），T.dactyloides var.meridionale（2n=36），and T.dactyloides（2n=72）.

Satellite repeats were amplified by PCR using genomic DNA of the three species.Cloning of satellite repeats was performed using primers designed according to the extracted repeatclusters.Satelliteprobeswerelabeledwithdigoxigenin-11-dUTP or biotin-16-dUTP（Roche Diagnostics）.Probe labeling was performed by standard nick translation.

FISH was performed according to the published protocol［36］.Slides and probes were denatured in solutions（FAD and SSC）for 2 and 10 min，respectively，at 85°C.Hybridizations were performed at 37°C overnight.Signals were detected with anti-biotin-FITC and anti-digoxigenin-rhodamine.Slides were counterstainedwithDAPI（4′-6′-diamidino-2-phenylindole，Vector Laboratories）.An epifluorescence microscope（Olympus BX61，Japan）equipped with a CCD camera（Qimaging）was used to capture images.FISH signals were measured by the Image-Pro plus 6.5 software.The images were processed using Adobe Photoshop CS4 software（Adobe Systems San Jose，CA）.

3.Results

3.1.Different species of Tripsacum differed in genome size

We used flow cytometry［30］to determine the genome size of MR，DD，andDL.BecauseTripsacumandmaizearecloselyrelated［3］，we used the maize genome（approximately 2300 Mb［37］）as the reference.As shown in Table 1，the estimated genome sizes of MR，DD，and DL were 3103，3814，and 6892 Mb，respectively. The genome size of DL was slightly smaller than twice that of DD and slightly greater than twice that of MR.

3.2.Repetitive sequences were abundant in the Tripsacum genomes

To analyze the genome structures of diploid and tetraploid Tripsacum，we used Hiseq2000 to perform low-coverage sequencing on MR，DD，and DL.We collected 7457.58 Mb， 10，216.58 Mb，and 8485.93 Mb of sequencing data for MR，DD，and DL，respectively，with the respective coverages of 1.20×，1.34×，and 0.62×（Table 2）.

Table 2-Sequencing features of T.dactyloides（2n=36），T.dactyloides var.meridionale，and T.dactyloides（2n=72）.

Using cluster-based repeat identification and classification［38］，we found that repetitive sequences accounted for 63.23% and 59.20%of the genome sequence of MR and DD，respectively.The proportion of repetitive sequences in the genome sequence of tetraploid DL is 61.57%.Similar to maize［37］，Tripsacum contained several repetitive sequence families. Longterminalrepeat（LTR）retrotransposons，long/short interspersed element（LINE/SINE）retrotransposons，En-Spm transposons，and satellites were widespread in the genomes of the three species（Table 3）.

Table 3-Repeat element composition of T.dactyloides（2n=36），T.dactyloides var.meridionale，T.dactyloides（2n=72），and Zea mays.

As displayed in Table 3，retrotransposons occupied 46.53%，42.26%，and 51.86%of the genomes of MR，DD，and DL，respectively.As in maize，most of them were Ty1/Copia and Ty3/Gypsy families.Ty1/Copia accounted for 17.27%，21.34%，and 19.06%in MR，DD，and DL，respectively.The contents of Ty3/Gypsy in MR，DD，and DL were 29.20%，21.11%，and 32.60%，respectively.The En-Spm were the most predominant type of transposons and occupied 1.75%，1.77%，and 1.67%of the MR，DD，and DL genomes，respectively.Satellite DNA accounted for more than 10%of the genome in the diploid species（14.66%in MR and 13.99%in DD）but only 7.19%in the tetraploid DL.Although the genome size varied among the three Tripsacum species，the genome composition was similar.

Table 4-The proportion and monomer lengths of satellite repeats in T.dactyloides（2n=36），T.dactyloides var. meridionale，and T.dactyloides（2n=72）.

3.3.Satelliterepeatidentificationandkaryotypingof T.dactyloides Var.meridionale and T.dactyloides

In addition to 5S rDNA and 45S rDNA，three tandem repeats SatS1，SatS2，and CentT were identified by bioinformatic analysis.As shown in Table 4，the most abundant type of tandem repeats in the three species was SatS1，which is a homolog of maize the 180-bp Knob sequence.SatS1 accounted for 10.70%，12.60%，and 5.64%of the genomes of MR，DD，and DL，respectively.SatS2 is a homolog of the other maize Knob sequence TR1，and CentT is a homolog of the maize centromere satellite CentC.Both SatS2 and CentT occupied less than 2%of the genomes.The proportions of the common satellite repeats were different in the three genomes，indicating that the satellites evolve rapidly in Tripsacum.

Fig.1-The karyotype for T.dactyloides var.meridionale（MR）mitotic metaphase chromosomes.A.The first FISH with CentC（green），microsatellite TAG repeat［39］（red），and 45S rDNA（yellow）as the probes.B.The second FISH with SatS1（green），5S rDNA（red），and SatS2（yellow）as the probes.C.The 36 chromosomes of MR were cut out from the figures（A，B）and displayed. D.Ideogram of the karyotype of MR.Ideogram of the distribution pattern of microsatellite TAG repeat（purple），SatS1（green），SatS2（orange），5S rDNA（yellow），CentT（blue），45S rDNA（red）and on the 18 pairs of chromosomes.Bars，10 μm.

To further characterize the distribution of these satellites，we performed karyotype analysis of diploid MR and DD. As shown in Figs.1 and 2，the 45S rDNA in MR and DD showed similar distribution patterns and were located on the short-arm terminus of chromosome 18.CentT signals were detected on the centromeres of each chromosome in both MR and DD.However，the signal intensity of CentT varied dramatically among different MR chromosomes（Fig.1）.CentT signal intensity was stronger on nine pairs of chromosomes than on the other nine pairs.In contrast，no difference in CentT signal intensity was detected in DD（Fig.2）.In addition，SatS1 signals were located on one or both termini of each chromosome in both MR and DD.In most chromosomes，SatS1 signals appeared only on the long-arm terminus. However，five pairs of chromosomes in MR and seven in DDshowed SatS1 signals at both termini.Notably，the signal patterns of 5S rDNA and SatS2 were considerably different in MR and DD.In MR，only one pair of 5S rDNA signals appeared on the chromosomes（Figs.1，3-A），whereas in DD，there were two pairs of 5S rDNA signals in subcentromeric regions（Figs.2，3-B）.SatS2 signals also presented different distribution patterns in MR and DD.Five pairs of SatS2 signals were detected near the SatS1 signals on chromosome pairs 12，14，15，16，and 18 in MR（Figs.1，3-D）.However，only three pairs of TR-1 signals were detected on chromosome pairs 7，17，and 18 in DD（Figs.2，3-E）.

Fig.2-The karyotype for T.dactyloides（2n=36，DD）mitotic metaphase chromosomes.A.The first FISH with SatS2（red），45S rDNA（yellow），and 5S rDNA（green）as the probes.B.The second FISH with SatS1（green），5S rDNA（green），CentT（red），and SatS2（yellow）as the probes.C.The 36 chromosomes of DD were cut out from the figures（A，B）and displayed.D.Ideogram of the karyotype of DD.Ideogram of the distribution pattern of SatT1（green），SatS2（orange），CentT（blue），5S rDNA（yellow），and 45S rDNA（red）on the 18 pairs of chromosomes.Bars，10 μm.

We examined the distribution patterns of 5S rDNA and SatS2 in the tetraploid DL.Four pairs of 5S rDNA signals were examined on the chromosomes in DL，and the signal intensity of one pair was clearly weaker than that of the other three pairs（Fig.3-C）.Six pairs of SatS2 signals were located on chromosome termini in DL（Fig.3-F）.The number of signal pairs of 5S rDNA and SatS2 in DL was twice that in DD.Based on the distribution patterns of the satellites，we proposed that DL was phylogenetically closer to DD than to MR.

3.4.ITS phylogenetic tree showed that DD and DL are related

ITS sequences are phylogenetically useful tools for the taxonomic analysis of species［40］.In our study，approximately 600 bp of ITS regions（ITS1，5.8S rDNA，and ITS2）in DD，MR，DL，and Z.mays were amplified and used for phylogenetic analysis.

As shown in Table 5，the sequence lengths（591 bp to 595 bp）of the ITS regions were similar among the fourspecies.ITS 1 varied from 207 bp to 211 bp.ITS 2 varied from 220 bp to 221 bp.The mean guanine+cytosine（G+C）content was higher in ITS2 than in ITS1.The length and（G+C）% of ITS1 and ITS2 were identical in DD and DL but were greater than those in MR.Alignment of the sequences of ITS1，5.8S rDNA，and ITS2 of B73，MR，DD，and DL showed that the length and structure of 5.8S rDNA were conserved among MR，DD，and DL，and four nucleotides in the 5.8S rDNA distinguished Tripsacum from Z.mays B73（Table 5）.However，the variations in ITS1 and ITS2 regions were detected in Zea and Tripsacum. More than 40 and 20 variation sites in ITS1 and ITS2，respectively，were detected in Zea and Tripsacum.Compared with DL，MR showed 2，0，and 5 variation sites in ITS1，5.8S，and ITS2，respectively，whereas DD showed no variation in these regions（Table 6）.We constructed a phylogenetic tree to determine their relationship using whole ITS sequences（Fig.4）.The results showed that DL is closer to DD than to MR.

Fig.3-FISH with repetitive sequence 5S rDNA and SatS2 as the probes in T.dactyloides（2n=36，DD），T.dactyloides var. meridionale（MR），and T.dactyloides（2n=72，DL），and the three species showed different distribution patterns.A.The signals of 5S rDNA in MR，red.B.The signals of 5S rDNA in DD，red.C.The signals of 5S rDNA in DL，red.D.The signals of SatS2 in MR，red. E.The signals of SatS2 in DD，red.F.The signals of SatS2 in DL，green.More abundant signals（arrowhead），less abundant signals（arrow）.Bars，10 μm.

Table 5-Lengths（bp）and G+C content of ITSI，5.8S rDNA，and ITS2 of Zea and Tripsacum.

Table 6-Variation sites in ITS1，5.8S，and ITS2 regions of the ITS sequences among Zea mays and Tripsacum.

3.5.Comparative analysis of Tripsacum，Coix，Zea，and Sorghum based on the repetitive sequences

We first performed a comparative analysis of the repetitive sequences of the three species of Tripsacum，DD，MR，and DL. We built a dataset by combining the labeled reads with sample names and then performed graph-based clustering. As shown in Figs.5，175 common clusters were detected in the three species，occupying more than 56%of their genomes.DD and MR shared 16 common clusters accounting for 0.67%and 0.76%of the DD and MR genomes，respectively.DL shared 20 common clusters with DD accounting for 0.34%and 1.65%of the DD and DL genomes，respectively.In contrast，MR and DL shared only 4 common clusters accounting for only 0.05%of the MR and 0.07%of the DL genome.In addition，15 clusters occupying 0.36%of the DL genome were unique to DL and not found in DD or MR.Thus，more repeat families were common to DL and DD than to DL and MR.

WealsocomparedtherepetitivesequencesofZea，Tripsacum，Coix，and Sorghum.We downloaded the pair-ends sequence data of Z.mays B73 and Sorghum Tx378 from DNAnexus and combined them with our sequencing data.As shown in Fig.6，there were only 18 common clusters of repeat families in all the nine species，accounting for 9.40%of the Z.mays B73 genome，8.16%of the Z.mays ssp.parviglumis genome，6.75%of T.dactyloides genome，7.38%of the Coixlacryma-jobi genome，2.78%of theCoixaquatica genome，1.43% of the Sorghum bicolor genome，and 1.88%of the Sorghum propinquum genome.We found that Zea and Tripsacum had 116 clusters in common occupying almost 50%of their genomes. However，only 47 common clusters were found in Tripsacum and Coix and only 30 common clusters in Tripsacum and Sorghum（Fig.6）.Thus，Tripsacum shared the most common repeat families with Zea，and shared more repeat families with Coix than with Sorghum.

Fig.4-ITS phylogenetic tree based on the neighbor-joining method.DD，MR，and DL represent T.dactyloides（2n=36），T.dactyloides var.meridionale，and T.dactyloides（2n=72），respectively.

4.Discussion

4.1.Repetitive sequences varied among Tripsacum species

Repetitive DNA sequences may have important implications in genome differences and may occupied up to 90%of the genome in some species［41］.The type and copy number of repetitive DNA sequences vary greatly in different genomes. In general，larger genomes contain higher proportions of repetitive sequences.In Arabidopsis and rice，with respective genome sizes of 120 and 466 Mb［42］，repetitive sequences occupy only approximately 25%and 40%，respectively，of the genomes of these species［43］.In larger genomes，repetitive sequences occupy approximately 62%of the Sorghum genome（740 Mb）［44］and 85%of the Zea genome（2300 Mb）［37］.In our study，the genome sizes of diploid Tripsacum MR and DD were approximately 3103 b and 3814 Mb，respectively，larger than that of maize.However，the proportions of repetitive sequences in DD and MR genomes were only 59.20%and 63.23%，respectively，substantially less than that of maize（85%）.The proportion of repetitive sequences in Tripsacum may have been underestimated in our study，owing to the limitations of low-coverage sequencing and our analysis methods.

OurstudyshowedthatthethreeTripsacumspecies contained different amounts of repetitive sequences.Notably，the genome size of tetraploid Tripsacum DL was slightly smaller than twice that of diploid DD，while slightly larger than twice that of diploid MR.However，the proportion of repetitive sequences was slightly higher in DL（61.57%）than in DD（59.20%）but slightly lower than in MR（63.23%）.Moreover，the three Tripsacum species contained different repeat families. For example，DL had 15 unique clusters of repetitive sequences accounting for 0.36%of the DL genome but absent from the two diploid species.These differences in the repetitive sequences of the three Tripsacum species may have resulted from expansion and rearrangement of repetitive sequences after genome divergence and original tetraploidization［45］.The repetitive elementsinthesedifferentspeciesexpanded tovaryingextents and at varying speeds［46］.

Fig.5-Comparative analysis of the repetitive sequences in T.dactyloides（2n=36，DD），T.dactyloides var.meridionale（MR），and T.dactyloides（2n=72，DL）.Distribution of repeat families in the three species of Tripsacum and numbers of common clusters. DD，MR，and DL represent T.dactyloides（2n=36），T.dactyloides var.meridionale，and T.dactyloides（2n=72），respectively.

4.2.The phylogenetic relationship of Tripsacum and closely related species based on comparative analysis of repetitive sequences

Satellites evolve rapidly among different，even related，species.Karyotype analysis using satellites as probes can identify chromosome homology in polyploid species by characterizing chromosomal structures［47］.In our study，by use of satellite repeat probes，we found that the number of probe signals on DL chromosomes was twice that on DD chromosomes and that the distribution patterns of the signals were similar in the two species.Thus，we speculated that DL might have a closerevolutionary relationship with DD than with MR and that tetraploid DL evolved from diploid DD.Moreover，the comparativeanalysisoftherepetitivesequencesshowedthatDLshared more repeat families with DD than with MR，suggesting that DL is more closely related to DD than to MR.The results of ITS sequence phylogenetic analysis also support our speculation.

In the Poaceae family，Zea，Tripsacum，Coix，and Sorghum are close relatives，although they originated in different regions［2］.Previous studies have shown that the divergence time of the ancestors of maize and sorghum was approximately 11.9 million years ago［48］.The evolutionary divergence of Tripsacum andZea occurred approximately 4.5-4.8 million years ago［49］.In this study，we compared all repeat families of the species from Tripsacum and its closely related genera.Similar with previous studies，our results supported the proposition that Tripsacum and Zea were most closely related and that the evolutionary relationship between Tripsacum and Sorghum was relatively distant.Zea is more closely related to Tripsacum than to Coix.

Fig.6-Comparative analysis of the repetitive sequences in Tripsacum and closely related species.Distribution of repeat families and corresponding numbers of common clusters.

Acknowledgments

This research was supported by the National Natural Science Foundation of China（Nos.31471499，91535206）.The authors thank Dr.Gui Su for valuable comments in the preparation of this manuscript.

R E F E R E N C E S

［1］E.A.Kellogg，L.Watson，Phylogenetic studies of a large data set:I.Bambusoideae，Andropogodeae and Pooideae（Gramineae），Bot.Rev.59（1993）273-343.

［2］D.Koo，J.Jiang，Extraordinary tertiary constrictions of Tripsacum dactyloides chromosomes:implications for karyotype evolution of polyploids driven by segmental chromosome losses，Genetics 179（2008）1119-1123.

［3］R.V.Tantravahi，Cytology and Crossability Relationships of Tripsacum，Bussey Institute，Harvard University，Cambridge，MA，1968.

［4］B.Kindiger，C.L.Dewald，The reproductive versatility of eastern gamagrass，Crop Sci.37（1997）1351-1360.

［5］M.Eubanks，A cross between two maize relatives:Tripsacum dactyloides and Zea diploperennis（Poaceae），Econ.Bot.49（1995）172-182.

［6］P.C.Mangelsdorph，R.G.Reeves，The Origin of Indian Corn and its Relatives，Texas Agricultural Experiment Station Texas，Bulletin，5741939.

［7］G.W.Beadle，Teosinte and the origin of maize，J.Hered.30（1939）245-247.

［8］W.C.Galinat，Theoriginofmaizeasshownbykeymorphological traits of its ancestor，teosinte，Maydica 28（1983）121-138.

［9］J.Doebley，Molecular evidence and the evolution of maize，Econ.Bot.44（1990）6-27.

［10］L.L.Jackson，C.L.Dewald，C.C.Bohlen，A macromutation in Tripsacum dactyloides（Poaceae）:consequences for seed size，germination，and seedling establishment，Am.J.Bot.79（1992）1031-1038.

［11］L.L.Jackson，C.L.Dewald，Predictive evolutionary consequences of greater reproductive effort in Tripsacum dactyloides，a perennial grass，Ecology 75（1994）627-641.

［12］C.Spillane，C.Baroux，J.M.Escobar-Restrepo，D.R.Page，S. Laoueille，U.Grossniklaus，Transposons and tandem repeats are not involved in the control of genomic imprinting at the MEDEA locus in Arabidopsis，Cold Spring Harb.Symp.Quant. Biol.69（2004）465-476.

［13］A.R.Orr，R.Kaparthi，C.L.Dewald，M.D.Sundberg，Analysis of inflorescence organogenesis in eastern gamagrass，Tripsacum dactyloides（Poaceae）:the wild type and the gynomonoecious GSF1 mutant，Am.J.Bot.88（2001）363-381.

［14］J.C.Lamb，J.A.Birchler，Retroelement genome painting: cytological visualization of retroelement expansions in the genera Zea and Tripsacum，Genetics 173（2006）1007-1021.

［15］Z.X.Cai，H.J.Liu，Q.Y.He，M.W.Pu，J.Chen，J.S.Lai，X.X.Li，W.W.Jin，Differential genome evolution and speciation of Coix lacryma-jobi L.and Coix aquatica Roxb.Hybrid Guangxi revealed by repetitive sequence analysis and fine karyotyping，BMC Genomics 15（2014）1025.

［16］W.F.Doolittle，C.Sapienza，Selfish genes，the phenotype paradigm and genome evolution，Nature 284（1980）601-603.

［17］L.E.Orgel，F(xiàn).H.Crick，Selfish DNA:the ultimate parasite，Nature 284（1980）604-607.

［18］M.A.Garrido-Ramos，I.H.R.De，R.M.Ruiz，R.C.Ruiz，Subtelomeric satellite DNA family isolated from the genome of the dioecious plant Silene latifolia，Genome 42（1999）442-446.

［19］D.Dechyeva，T.Schmidt，Molecular organization of the terminal repetitive DNA in Beta species，Chromosom.Res.14（2006）881-897.

［20］T.Schmidt，J.S.Heslop-Harrison，Genomes，genes and junk: the large-scale organization of plant genomes，Trends Plant Sci.3（1998）195-199.

［21］J.M.Jiang，J.A.Birchler，W.A.Parrott，R.K.Dawe，A molecular view of plant centromeres，Trends Plant Sci.8（2003）570-575.

［22］R.B.Flavell，Repetitive DNA and chromosome evolution in plants，Philos.Trans.R.Soc.Lond.B 312（1986）227-242.

［23］J.Macas，D.Pozárková，A.Navrátilová，M.Nouzová，P. Neumann，Two new families of tandem repeats isolated from genus Vicia using genomic self-priming PCR，Mol.Gen.Genet. 263（2000）741-751.

［24］S.Mehrotra，V.Goyal，Repetitive sequences in plant nuclear DNA:types，distribution，evolution and function，Genomics Proteomics Bioinformatics 12（2014）164-171.

［25］T.Cermak，Z.Kubat，R.Hobza，A.Koblizkova，A.Widmer，J. Macas，B.Vyskot，E.Kejnovsky，Survey of repetitive sequences in Silene latifolia with respect to their distribution on sex chromosomes，Chromosom.Res.16（2008）961-976.

［26］R.Navajas-Pérez，T.Schwarzacher，M.Ruiz Rejón，M.A. Garrido-Ramos，Characterization of RUSI，a telomere-associated satellite DNA，in the genus Rumex（Polygonaceae），Cytogenet.Genome Res.124（2009）81-89.

［27］J.Jurka，V.V.Kapitonov，O.Kohany，M.V.Jurka，Repetitive sequences in complex genomes:structure and evolution，Annu.Rev.Genomics Hum.Genet.8（2007）241-259.

［28］T.J.Treangen，S.L.Salzberg，Repetitive DNA and next-generation sequencing:computational challenges and solutions，Nat.Rev.Genet.13（2012）36-46.

［29］P.Novák，P.Neumann，J.Macas，Graph-based clustering and characterization of repetitive sequences in next-generation sequencing data，BMC Bioinformatics 11（2010）378.

［30］J.Dolezel，J.Greilhuber，J.Suda，Estimation of nuclear DNA content in plants using flow cytometry，Nat.Protoc.2（2007）2233-2244.

［31］J.J.Doyle，A rapid DNA isolation procedure for small quantities of fresh leaf tissue，Phytochem.Bull.19（1987）11-15.

［33］A.F.Smit，R.Hubley，P.Green，RepeatMasker open-3.0，http://www.repeatmasker.org 1996-2010.

［34］S.F.Altschul，T.L.Madden，A.A.Sch?ffer，J.Zhang，Z.Zhang，W.Miller，D.J.Lipman，Gapped BLAST and PSI-BLAST:a new generation of protein database search programs，Nucleic Acids Res.25（1997）3389-3402.

［35］G.Benson，Tandem repeats finder:a program to analyze DNA sequences，Nucleic Acids Res.27（1999）573-580.

［36］C.Liu，J.Liu，H.Li，Z.Zhang，Y.Han，S.Huang，W.Jin，Karyotyping in melon（Cucumis melo L.）by cross-species fosmid fluorescence in situ hybridization，Cytogenet. Genome Res.129（2010）241-249.

［37］P.S.Schnable，D.Ware，R.S.Fulton，J.C.Stein，F(xiàn).S.Wei，S. Pasternak，C.Z.Liang，J.W.Zhang，L.Fulton，T.A.Graves，P. Minx，A.D.Reily，L.Courtney，S.S.Kruchowski，C.Tomlinson，C.Strong，K.Delehaunty，C.Fronick，B.Courtney，S.M.Rock，E. Belter，F(xiàn).Y.Du，K.Kim，R.M.Abbott，M.Cotton，A.Levy，P. Marchetto，K.Ochoa，S.M.Jackson，B.Gillam，W.Z.Chen，L. Yan，J.Higginbotham，M.Cardenas，J.Waligorski，E. Applebaum，L.Phelps，J.Falcone，K.Kanchi，T.Thane，A. Scimone，N.Thane，J.Henke，T.Wang，J.Ruppert，N.Shah，K. Rotter，J.Hodges，E.Ingenthron，M.Cordes，S.Kohlberg，J. Sgro，B.Delgado，K.Mead，A.Chinwalla，S.Leonard，K.Crouse，K.Collura，D.Kudrna，J.Currie，R.F.He，A.Angelova，S. Rajasekar，T.Mueller，R.Lomeli，G.Scara，A.Ko，K.Delaney，M.Wissotski，G.Lopez，D.Campos，M.Braidotti，E.Ashley，W. Golser，H.Kim，S.Lee，J.K.Lin，Z.Dujmic，W.Kim，J.Talag，A. Zuccolo，C.Z.Fan，A.Sebastian，M.Kramer，L.Spiegel，L. Nascimento，T.Zutavern，B.Miller，C.Ambroise，S.Muller，W. Spooner，A.Narechania，L.Y.Ren，S.Wei，S.Kumari，B.Faga，M.J.Levy，L.McMahan，P.Van Buren，M.W.Vaughn，K.Ying，C.T.Yeh，S.J.Emrich，9，Y.Jia，A.Kalyanaraman，A.P.Hsia，W.B.Barbazuk，R.S.Baucom，T.P.Brutnell，N.C.Carpita，C. Chaparro，J.M.Chia，J.M.Deragon，J.C.Estill，Y.Fu，J.A. Jeddeloh，Y.J.Han，H.Lee，P.H.Li，D.R.Lisch，S.Z.Liu，Z.J.Liu，D.H.Nagel，M.C.McCann，P.SanMiguel，A.M.Myers，D. Nettleton，J.Nguyen，B.W.Penning，L.Ponnala，K.L. Schneider，D.C.Schwartz，A.Sharma，C.Soderlund，N.M. Springer，Q.Sun，H.Wang，M.Waterman，R.Westerman，T.K. Wolfgruber，L.X.Yang，Y.Yu，L.F.Zhang，S.G.Zhou，Q.H.Zhu，J.L.Bennetzen，R.K.Dawe，J.M.Jiang，N.Jiang，G.G.Presting，S.R.Wessler，S.Aluru，R.A.Martienssen，S.W.Clifton，W.R. McCombie，R.A.Wing，R.K.Wilson，The B73 maize genome: complexity，diversity，and dynamics，Science 326（2009）1112-1115.

［38］B.De Felice，R.R.Wilson，L.Ciarmiello，C.Conicella，A novel repetitive DNA sequence in lemon（Citrus limon（L.）Burm.）and related species，J.Appl.Genet.35（2004）315-320.

［39］A.Kato，J.C.Lamb，J.A.Birchler，Chromosome painting using repetitive DNA sequences as probes for somatic chromosome identification in maize，Proc.Natl.Acad.Sci.U.S.A.37（2004）13554-13559.

［40］L.A.Alice，C.S.Campbell，Phylogeny of Rubus（Rosaceae）based on nuclear ribosomal DNA internal transcribed spacer region sequences，Am.J.Bot.86（1999）81-97.

［41］J.Macas，T.Meszaros，M.Nouzova，PlantSat:a specialized database for plant satellite repeats，Bioinformatics 18（2002）28-35.

［42］K.Arumuganathan，E.D.Earle，Nuclear DNA content of some important plant species，Plant Mol.Biol.Rep.9（1991）208-218.

［43］S.Ouyang，C.R.Buell，The TIGR plant repeat databases: a collective resource for the identification of repetitive sequences in plants，Nucleic Acids Res.32（2004）360-363.

［44］A.H.Paterson，J.E.Bowers，R.Bruggmann，I.Dubchak，J. Grimwood，H.Gundlach，G.Haberer，U.Hellsten，T.Mitros，A. Poliakov，J.Schmutz，M.Spannagl，H.B.Tang，X.Y.Wang，T. Wicker，A.K.Bharti，J.Chapman，F(xiàn).A.Feltus，U.Gowik，I.V. Grigoriev，E.Lyons，C.A.Maher，M.Martis，A.Narechania，R.P. Otillar，B.W.Penning，A.A.Salamov，Y.Wang，L.F.Zhang，N.C. Carpita，M.Freeling，A.R.Gingle1，C.T.Hash，B.Keller，P.Klein，S.Kresovich，M.C.McCann，R.Ming，D.G.Peterson，Mehboob-ur-Rahman，D.Ware，P.Westhoff，K.F.X.Mayer，J. Messing，D.S.Rokhsar，The Sorghum bicolor genome and the diversification of grasses，Nature 457（2009）551-556.

［45］I.J.Leitch，M.D.Bennett，Genome downsizing in polyploid plants，Biol.J.Linn.Soc.82（2004）651-663.

［46］M.G.Kidwell，Transposable elements and the evolution of genome size in eukaryotes，Genetica 115（2002）49-63.

［47］Z.Y.Xiong，J.C.Pires，Karyotype and identification of all homoeologous chromosomes of allopolyploid Brassica napus and its diploid progenitors，Genetics 187（2011）37-49.

［48］Z.Swigonová，J.S.Lai，J.X.Ma，W.Ramakrishna，V.Llaca，J.L. Bennetzen，J.Messing，Close split of sorghum and maize genome progenitors，Genome Res.14（2004）1916-1923.

［49］H.Hilton，B.S.Gaut，Speciation and domestication in maize and its wild relatives:evidence from the globulin-1 gene，Genetics 150（1998）863-872.

29 March 2016

*Corresponding author.

E-mail address:weiweijin@cau.edu.cn（W.Jin）.

Peer review under responsibility of Crop Science Society of China and Institute of Crop Science，CAAS.1These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.cj.2016.04.003

2214-5141/?2016 Crop Science Society of China and Institute of Crop Science，CAAS.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.