Repetitive sequences in plant nuclear DNA: types, distribution, evolution and function.

Repetitive DNA sequences are a major component of eukaryotic genomes and may account for up to 90% of the genome size. They can be divided into minisatellite, microsatellite and satellite sequences. Satellite DNA sequences are considered to be a fast-evolving component of eukaryotic genomes, comprising tandemly-arrayed, highly-repetitive and highly-conserved monomer sequences. The monomer unit of satellite DNA is 150-400 base pairs (bp) in length. Repetitive sequences may be species- or genus-specific, and may be centromeric or subtelomeric in nature. They exhibit cohesive and concerted evolution caused by molecular drive, leading to high sequence homogeneity. Repetitive sequences accumulate variations in sequence and copy number during evolution, hence they are important tools for taxonomic and phylogenetic studies, and are known as "tuning knobs" in the evolution. Therefore, knowledge of repetitive sequences assists our understanding of the organization, evolution and behavior of eukaryotic genomes. Repetitive sequences have cytoplasmic, cellular and developmental effects and play a role in chromosomal recombination. In the post-genomics era, with the introduction of next-generation sequencing technology, it is possible to evaluate complex genomes for analyzing repetitive sequences and deciphering the yet unknown functional potential of repetitive sequences.

Introduction

Genomes of higher eukaryotes contain more DNA than expected when estimates are based on the length and number of coding genes in the genomes. The amount of DNA in the unreplicated genome, or the haploid genome, of a species is known as C-value or Constant-value [1,2]. The lack of correlation between size and complexity of eukaryotic genomes, largely due to the presence of noncoding highly repetitive DNA, is termed as the C-value paradox, which is a common phenomenon observed in higher plants. It is believed that the proportion of protein-coding sequences is generally similar for different plant species, with variation in genome size mainly due to the presence of repetitive DNA [3,4] that has accumulated in the genomes during evolution, since ancestral angiosperms had been indicated to possess small genomes [5]. The term “repetitive sequences” refers to homologous DNA fragments that are present in multiple copies in the genome. Repetitive DNA sequences are present in all higher plants and can account for up to 90% of the genome size in some species. These repetitive DNA sequences are considered to generate major differences between genomes, which may reflect evolutionary distances between species. The repetitive sequences were once thought to be “selfish elements” or “junk DNA” [6,7], since they do not harbor genes.

Knowledge of the distribution, genomic organization, chromosomal location and evolutionary origin of repetitive DNA sequences is necessary for insight into the organization, evolution, behavior and functional potential of repetitive sequences in eukaryotic genomes [8]. In the last few decades, several repetitive sequences have been analyzed (Table 1) to gain more information on primary structure, molecular organization, evolution and function of repetitive sequences, whereas more recently genome-sequencing technology has produced an unprecedented wealth of information about origin, diversity and genomic impact of repetitive sequences [9].

Table 1

Repeat DNA families in various plant species

Species	Repetitive sequence name	Repeat length∗(bp)
Aegilops speltoides	spelt-1	150
Aegilops squarrosa (DD)	pAS1(Afa family)	336–337
	spelt-1	150
Allium cepa	ACSAT 1/ACSAT 2/ACSAT 3	370
	pAc074	314
Allium fistulosum	pAfi100	380
Arabidopsis thaliana	180 bp repeat/HindIII repeat/AtCon/AtCen/pAL1/pAS1/pAtMR/pAtHR/pAa214/AaKB27 family	180
Avena	pAm1	58
Beta vulgaris	pAv34/pAc34/pRp34/pRn34/pRs34	334–362
	pBVI	327–328
	pHC28	149
	pHT30	140–149
	pHT49	162
	ppHC8	162
	pRN1	209–233
	pTS4.1	312
	pTS5	153–160
Brassica	pBcKB4/pBT11/HindIII family/Canrep	175–180
	pBo1.6	203
Brassica campestris	BT4	296
	BT11	175
	CS1	88
	CT10	213
	pBcKB4	360
Brassica napus	Canrep	176
Brassica nigra	pBN4	459
	pBNE8	1732
Brassica oleracea	pBoKB1	360
Camellia sinensis	pMST11	894
Centaurea	HinfI	350
Citrus limon	CL600	600
Crocus vernus	pCvKB4	270
Cucurbita pepo	350 bp satellite	349–352
Cucurbita maxima	170 bp satellite	168–170
Cucumis melo	HindIII repeat	352
Cucumis metuliferus	pMetSat	346
Cucumis sativus	Type I	182
	Type III	177
	Type IV	360
Diplotaxis erucoides	Canrep	175–180
Elaeis guineensis	pEgKB15	355
	pEgKB20	342
Elymus trachycaulus	pEt2	337–339
Glycine max	SB92	92
	STR120	120
Hordeum chilense	pHch1	2.6 kb
	pHch2	2.1 kb
	pHch3	500
	pHch4	2.6 kb
	pHch5	2.0 kb
	pHcKB6	339
Hordeum vulgare	HvRT	118
	pHvMWG2315	331
Leymus racemosus	350 bp family	350
	Lt1	380
	pLrAfa1–6	340
	TaiI	570
Lycopersicon esculentum	GR1	162
	pLEG15	168
Medicago truncatula	MtR1	166
	MtR2/MtR3	183, 166
Musa spp.	Radka1	685
	Radka2	409
	Radka3	808
	Radka4	605
	Radka5	742
	Radka6	193
	Radka7	596
	Radka8	337
	Radka9	334
	Radka10	689
Nicotiana tabacum	HRS60	180
	TAS49	460
Nicotiana rustica	HRS-60 family	180
Oryza australiensis	pOa237	1300
Oryza minuta	pOm1	239
	pOm4	438
	pOmA536	400
	pOmPB10	305
Oryza rufipogon	H2	615
Oryza sativa	C154	352
	C193	353
	CentO-C	126
	CentO/RCS2/TrsD	155
	Os48/TrsA	355
	OsG3	498
	OsG5	756
	TrsC	366
Olea europaea	81 bp family	81
	Oe179	179
	OeTaq80	80
	OLEU	178
	pOS218	218
Pennisetum glaucum	pPgKB19	137
Phaseolus vulgaris	PvMeso 31	3.4 kb
	PvMeso 47	1.7 kb
Pisum	PiSTR-A	211–212
	PiSTR-B	506
Potamogeton pectinatus L.	PpeRsa1	362–364
	PpeRsa2	355–359
Psathyrostachys juncea (NN)	pPjAfa1-3	340
Raphanus sativus	Canrep	175–180
	pRA5/pRB12	177
Rumex	RAE180	180–186
	RAE730 family	727–731
	RAYS1 family	922–932
	RUS1	170
Saccharum officinarum	SCEN family	140
Sorghum bicolor	pSau3A10/pCEN38	137
Secale africanum	pSaD15	887
Secale cereale (Japanese rye)	JNK family	1.2 kb
	pSc34	480
	pSc74	610
	pSc119.1	350
	pSc119.2	350
	pSc200	380
	spelt-1	150
Silene latifolia	15Ssp	159
	SacI	313
	STAR-C	43
	TRAYC	172
	X43.1	335
Sinapis alba	Canrep	175–180
Solanum brevidens	pSB1	322
	pSB7	167
Solanum bulbocastanum	2D8	5.9 kb
Solanum circaefolium	pSCH15	168
Solanum tuberosum	pST3	845
	pST10	121
Trifolium	TrR350	350
Triticum aestivum	dpTa1	340
	TaiI family	570
	WE35	320
Vicia faba	BamH1	250, 1500
	Fok1	59
Vigna unguiculata	pVuKB1	488
Zea mays	Cent4	156
	CentC	156
	H2a/H2b	184–185
	MR68	410
	MR77	1.2 kb
Zingeria biebersteiniana	Zbcen1	755

Note: Plant species are listed alphabetically. ∗ Repeat length is indicated in bp unless otherwise followed with “kb”.

Types and distribution of repetitive DNA sequences

The genomes of most eukaryotes contain a variety of repetitive DNA sequences [10,11]. Repetitive DNA may be dispersed throughout the genome or may be restricted at specific locations in a tandem configuration. According to the length of the repeated unit and array size, tandem repeated DNA sequences can be classified into three groups: (i) microsatellites with 2–5 bp repeats and an array size of the order of 10–100 units, (ii) minisatellites with 6–100 bp (usually around 15 bp) repeats and an array size of 0.5–30 kb and (iii) satellite DNA (satDNA) with a variable AT-rich repeat unit that often forms arrays up to 100 Mb. The monomer length of satDNA sequences ranges from 150–400 bp in majority of plants and animals. satDNA sequences are located at heterochromatic regions, which are found mostly in centromeric and subtelomeric regions in the chromosomes but also at intercalary positions. Figure 1 shows a diagrammatic representation of different types of repetitive sequences on a plant chromosome. Satellite DNAs are portions of the genome that can be separated as satellite peaks from primary DNA peaks [12] using the methods including restriction endonuclease digestion, colony filter hybridization with relic DNA, amplification with specific primers, colony filter hybridization and genomic self-amplification (also known as self-priming) [13].

Figure 1

General distribution of repetitive sequences on plant chromosomes

Distribution of different types of repetitive sequences is represented diagrammatically on a plant chromosome with different colors. Red, centromeric tandem repeats; blue, telomeric repeats; yellow, sub-telomeric tandem repeats; green, intercalary tandem repeats; brown, dispersed repeats; white, genes and low-copy sequences.

Clustered DNA repeats can be detected in centromeric and telomeric heterochromatin, and are transcriptionally inert. Centromeric DNA (such as CENH3) is the most abundant tandem repeats found in both plants and animals [14]. In contrast, subtelomeric repetitive sequence families are usually genus-specific, such as TrsA and Os48 in Oryza species [15,16]; TrsB in Oryza brachyantha [17]; SacI family in Silene latifolia [18]; and pAv34, pAc34, pRp34, pRn34 and pRs34 in Beta species [19], or chromosome (Chr)-specific, such as WE 35 on Chr 5B in Triticum aestivum [20], JNK family on Chr 2R in Secale cereale [21], AfaI family in Poaceae members [22], TRI on Chr Y in Silene latifolia [23] and RUSI on Chr 1 in Rumex species [24].

A particular sequence may be either species-specific or present in many species within a taxonomic family or various families, indicating that some repetitive sequences evolve rapidly, whereas others may be conserved [25]. For instance, the CL600 satellite repeat isolated from Citrus limon was detected in other Rutaceae members (C. aurantium, C. paradisi, Poncirus trifoliata and Fortunella margarita) [26]. Similarly, the PCvKB repetitive DNA isolated from Crocus vernus was present in 16 species of Crocus analyzed and 2 species of Iridaceae, 3 species of Liliaceae and 1 species of Amaryllidaceae [27]. The HindIII repeat of Brassica campestris shows homology not only to that of other Brassica species but also to that of Raphanus sativus, Sinapis alba, Diplotaxis muralis and Erucastrum sp. [28]. Repetitive elements in cauliflower, mustard and radish, all belonging to Brassicaceae family, show 75%–80% homologies [29,30]. Mehrotra et al. [31] recently reported that the pCtKpnI-I satellite repeat, initially isolated from Carthamus tinctorius and other species of Carthamus [31], is present in widely divergent families of angiosperms [32].

Various repetitive sequences from diverse taxa have been integrated into a database for easy accessibility. PlantSat, a database specialized for plant satDNA, has integrated sequence data from several resources such as NCBI and DNA Data Bank of Japan (DDBJ) [33], to provide a list of satDNA sequences for members of many plant families including Poaceae, Brassicaceae, Chenopodiaceae, Cucurbitaceae and Solanaceae [34-36], as well as many other plants. Plant satDNA sequences commonly have monomer unit lengths of 135–195 bp or 315–375 bp, which are consistent with reports that the basic monomer units of plant repetitive DNA sequences correspond to the length of DNA wrapped around a nucleosomal particle [10,33,37].

Different families of repetitive sequences show consistent presence of motifs like AA/TT dinucleotides, pentanucleotide CAAAA, etc. The presence of conserved motifs in unrelated repetitive sequence families suggests their significance for molecular mechanisms underlying the amplification and maintenance of tandem repeats in a genome, and the determination of specific chromatin properties of loci containing the repetitive DNA [33]. The occurrence of short, direct and inverted repeats and short palindromes is a characteristic feature of various plant satellite families. These may act as preferential sites for changes and as potential substrates for homologous recombination allowing rearrangements [38-42]. The repetitive sequences have the nearest-neighbor characteristics with high frequencies of GG, AG and GA nearest neighbors [43], which originate during the repair of heteroduplex intermediates of the exchange [44,45]. Frequent occurrence of GGT and GTT trinucleotides in the monomers of repeat sequences makes the sequence suitable substrate for the de novo telomere synthesis in the repairing process of broken chromosome ends [46].

Another characteristic feature of repetitive sequences is methylation. Methylation of DNA sequences is considered to trigger structural changes in DNA strands [21,22]. Methylation has been observed in satellite repeats following restriction analysis of genomic DNA with MspI and HpaI. Methylation has been reported to occur in a 500 bp satellite repeat family in Arabidopsis [47], in a JNK repeat family in Japanese rye [21], in a pCtKpnI repeat family in Carthamus [31], and many other repeat families [26,48].

Functions of repetitive sequences

Repetitive DNA sequences are present in the heterochromatin region. Heterochromatin has been associated with several functions ranging from regulating gene expression to protecting chromosomal integrity. Heterochromatin can incur different levels of expression of adjacent genes during inheritance, because of either a position effect or juxta-position of heterochromatin and highly active genes, as observed in Drosophila and Saccharomyces [49-52].

Repetitive sequences are implicated in numerous processes such as chromosome movement and pairing, centromeric condensation, chromosome recombination, sister chromatid pairing, chromosome association with the mitotic spindle, chromosome arrangement, interaction of chromatin proteins, histone binding, determination of chromosome structure, karyotypic evolution, regulation of gene expression and genome response to environmental stimuli and physiological changes. These are all considered key components of evolutionary mechanisms and karyotypic differentiation [53,32]. Therefore, repetitive sequences play an important role in the evolution of species [54], and they are all speculated to model the regulatory patterns of genes leading to phenotypic variation [55]. The occurrence of species-specific satDNA enables rapid and reliable identification of species. Grewal and Elgin [56] proposed the transcription of satDNA and its impact on heterochromatin, particularly in terms of the formation and maintenance of heterochromatin structure. Repetitive DNA sequence elements are also involved in cooperative molecular interactions for the formation of nucleoprotein complexes [57]. Repeat sequences may attract some specific nuclear proteins, and the chromatin folding code dictates the DNA–protein interactions, which may underlie the genetic function of the tandem repeats [58]. Tandem repeats are proposed as “tuning knobs” in the evolution due to their ability to adjust the genetic hereditary traits and thereby facilitate the adaptation [59,60]. Melters et al. [14] have suggested that tandem repeats at centromeres may promote the concerted evolution of centromere DNA across chromosomes owing to their high prevalence. Centromere DNA transcripts have been reported to be involved in centromere structure and function [61]. It is speculated that if repetitive DNA is transposable, it may create novel genes [62]. The functions of satellite sequences are still not clear. In the post-genomics era, due to a wealth of resources, it is possible to gain insights into the functional potential of repetitive sequences.

Evolution of repetitive sequences

Repetitive DNA sequences exhibit cohesive and concerted evolution by mechanisms that cause continuous nuclear genome turnover and constitute molecular drive. Such concerted evolution produces high homogeneity in a repetitive DNA family where mutations are diagnostic for species, and are the origin of interspecies genetic divergence [63,64]. satDNA exhibits internal sequence variability depending on the ratio between mutation and homogenization/fixation rates within a species [65]. Levels of sequence identity between the satellite repeats can be attributed by the following factors: the rates and biases of transfer between homologous and nonhomologous chromosomes, the number and distribution of repeats, the physical constraints within the genome, the generation time, the effective population size and various biological and selective constraints [66].

Tandem repetitive sequences are considered to be generated de novo by the combinatorial action of molecular mechanisms such as mutations, unequal crossing over, gene conversion, slippage replication and/or rolling circle replication [44,67], which create and maintain homogeneity of satDNA sequences within species [65,68]. Among them, sequence duplication by aberrant recombination [69,70] or replication slippage [67,71,72] represents the primary event in the formation of a repeat.

Unequal crossing over is assumed to be the primary evolutionary force acting on satellite sequences [44]. Arrays of satDNA are maintained by unequal exchange and intrastrand exchange [73] and unequal crossing over accounts for the alterations in copy numbers of satellite monomers [44]. Individual repetitive units do not evolve independently; instead, the arrays evolve in concert. However, unequal crossing over, by itself, cannot create large tandem arrays of satDNA. Gene amplification and subsequent duplications also play a significant role in satDNA evolution. A few repeats may excise from a tandem array and circularize to provide a template for rolling circle replication [72], and after amplification into a linear array, the repeats may be inserted into a new location in the genome.

Repetitive elements are under different evolutionary constraints as compared to genes, and are considered as fast-evolving components of eukaryotic genomes. The high evolution rates of repetitive sequences can be used to differentiate related species [35,74-78]. Hybrid polyploids are excellent models for studying the evolution of repetitive sequences [37], and variations in their repetitive sequences allow them to be used for taxonomic and phylogenetic studies [79]. Sequence homogeneity and evolution of repetitive sequences are correlated with their copy number. Repetitive sequences with a low copy number are homogeneous and evolve slowly, whereas repetitive sequences with a high copy number are more heterogeneous and evolve quickly [80,81]. Sequence divergence in satDNA proceeds in a gradual manner due to the accumulation of nucleotide substitutions. For instance, sequences were highly conserved in the repetitive satDNA of Palorus ratzeburgii and Palorus subdepressus [82,83]. In addition, high sequence conservation was also observed in human α-satDNA repeat, which is a rare and a highly-conserved repeat in the evolutionarily distant species such as chicken and zebrafish [84]. Similarly, the simple dodeca satDNA repeat is also conserved among evolutionarily distant organisms such as fruitfly, Arabidopsis and human [85]. Evolutionary persistence of large tandem arrays is affected not only by the balance between the rate of amplification and the rate of unequal exchange, but also by a wide range of mechanisms for recombination, replication and gene amplification. However, the amount of bias in these processes acting on satDNA remains unresolved. Nonetheless, natural selection does not have any effect on satellite repeats, because satDNA sequences are not transcribable and are consequently neutral to selection [6,7].

Evolutionary significance of repetitive sequences

Repetitive sequences are speculated to influence cytoplasmic, cellular and developmental processes [86] by increasing genome size and affecting chromosomal recombination. satDNA repeats represent recombination “hotspots” of genome reorganization [87]. The occurrence of satDNA in interstitial and telomeric heterochromatin reduces genetic recombination in adjacent regions [87]. Robertsonian chromosome fusion or fission, the joining of two telo/acrocentric chromosomes at their centromeres to form a metacentric, has been postulated to depend on sequence similarity of regions on the two chromosomes because of large amounts of centromeric, heterochromatic satDNA in eukaryotic genomes [88]. Satellite repeats in eukaryotes are likely involved in sequence-specific interactions and subsequently in epigenetic processes. Nonetheless, repetitive satDNA also has a sequence-independent role in the formation and maintenance of heterochromatin. Transcripts from tandem arrays or satellites are processed by RNA-dependent RNA polymerase (RdRP) and Dicer to produce small interfering RNAs (siRNAs) [89,90]. satDNA-derived siRNAs are probably involved in posttranscriptional gene regulation through the action of the RNA-induced silencing complex (RISC) [91]. Additionally, satDNA-derived siRNAs also have a possible role in gene expression and heterochromatin formation on tandemly-repeated noncoding regions, and in the expression of particular genes with embedded satellite repeats [92].

Recent advances and future perspectives

A remarkable advance in the knowledge of repetitive sequences has occurred in recent years because of the introduction of next-generation sequencing technologies. These technologies can be applied to highly complex populations of repetitive elements in plant genomes, and have been used to characterize genomes and establish phylogenies based on repetitive sequences in Silene latifolia, Helianthus and Orobanche species [93-97]. Various strategies such as single nucleotide polymorphism (SNP) discovery and more sophisticated approaches are being developed to assemble NGS data and to analyze repeats for a better understanding of their contribution to gene function and genome evolution [95]. 454 sequencing and Illumina platforms have been extensively used to comprehensively characterize repetitive DNA in the pea and olive genome as well as the satellite sequences in Silene, including the detection of the most conserved regions, reconstruction of consensus sequences of repeat monomers, identification of major sequence variants and designing of hybridization probes for localization on chromosomes using fluorescence in situ hybridization (FISH) [93,97,98]. 454 pyrosequencing was also used to detect copy-number repeats in the soybean genome [99] and to deduce the repeat composition of genomes for nine species of Orobanchaceae [94]. Recently, Sergeeva et al. [100] presented a detailed account of repetitive sequences of wheat based on 454 sequencing. Whole genome shotgun sequencing has also been employed to identify and analyze the most abundant tandem repeats from diverse animal and plant species [14].

Various web tools have been recently introduced for the analysis of repetitive sequences. A tool called “REViewer” was developed for visualizing and analyzing repetitive elements [101]. A comprehensive toolkit “RepEx” was developed by Gurusaran et al. [102] for extracting various repeats (inverted, everted and mirror) from genomic sequences. Recently, Novák et al. [103] introduced a collection of software tools called “RepeatExplorer” for characterizing repetitive elements and identifying high- and medium-copy repeats in higher plant genomes.

Repetitive sequences are technically challenging to clone and sequence and can be better studied by combining various approaches like mapping and sequence analysis. Repetitive sequences also pose challenges in sequencing and assembling of genomes. Cytogenetics, genomics and bioinformatics tools have allowed the genomes of complex eukaryotes to be investigated. Whole genome resequencing studies, genome-wide analysis, transposon-based sequencing strategies and fine mapping of repetitive sequences can elucidate the structure, evolution, and functional potential of this enigmatic yet indispensable component of the genome. This will in turn assist in sequencing and assembly of complex eukaryotic genomes.

Competing interests

The authors have declared that no competing interests exist.