Sampling and DNA Isolation

The Meloidogyne spp. isolates used in this study were chosen from the living collection maintained at INRA, Sophia Antipolis, France. The geographic origin of the isolates was as follows: M. chitwoodi (Spijkenisse, The Netherlands), M. fallax (Baexem, The Netherlands), M. javanica (Pelotas, Brazil), M. paranaensis (Londrina, Brazil), M. incognita (Antibes, France), M. arenaria (Chappes, France) and M. hapla (La Môle, France). Nematodes were maintained on tomatoes (Lycopersicon esculentum cv. Saint Pierre) grown at 20°C in a greenhouse. They were specifically identified morphologically and according to their isoesterase electrophoretic pattern [20]. Eggs were collected from infested roots, according to the procedure described earlier [21]. Total genomic DNA was purified from 50–100 µl eggs using the DNeasy Tissue Kit (Qiagen) according to the manufacturer’s instructions. Possibility of sample cross-contamination with other nematode DNA was excluded through PCR check of genomic DNA with SCAR (sequence characterized amplified region) primers specific for Meloidogyne chitwoodi and M. fallax species [22].

PCR Analyses, Cloning and Sequencing

The satellite sequences were amplified with specific primers derived from previously published data [18] and from sequences obtained in this work. Primer sequences and their positions on the HOR sequence are indicated in Table S1 and Fig. S1, respectively. The reaction mixture consisted of reaction buffer, 1.5 mM MgCl₂, 0.2 mM dNTPs, 0.5 U GoTaq DNA polymerase (Promega), 0.4 µM of each primer and 20 ng of genomic DNA. The PCR cycling parameters used were as follows: 2 min initial denaturation at 94°C, followed by 30 cycles of: 95°C for 30 sec, 58°C for 30 sec, and 72°C for 1 min. Final extension was at 72°C for 10 min. PCR products were ligated in a pGEM T-Easy vector (Promega) and transformed in Escherichia coli DH5α-competent cells (Invitrogene). Recombinant clones with multimeric arrays of satellite DNA were sequenced by Macrogen (Korea). Monomers and HORs sequences from M. chitwoodi and M. fallax as well as Box 1-containing sequences from M. incognita sequenced genome were deposited in EMBL databank under Accession Numbers: JX186757-JX186849, JX186850-JX186855, JX186856 - JX186877, JX186878 - JX186996, KC968979 - KC969073.

Southern and Dot Blot Analyses

Standard procedures were used for restriction endonuclease digestions, electrophoresis, transfer to nylon membranes [23]. For genomic Southern hybridization analysis, 10 µg of genomic DNAs were partially digested with an appropriate restriction enzyme which cuts once in targeted repetitive unit. Gel electrophoresis was run in a 0.8% agarose, denatured and DNA transferred to Hybond N⁺membrane (Amersham). Hybridizations were performed overnight under high stringency conditions (65°C) in the buffer containing 250 mM Na₂HPO₄ (pH 7.2), 7% SDS, 1 mM EDTA, 0.5% blocking reagent and 50 ng/ml of the probe. Posthybridization washes were done in 20 mM Na₂HPO₄/1 mM EDTA/1% SDS at the temperature 2°C lower than the hybridization temperature. Chemiluminescent detection was carried out using the alkaline phosphatase substrate CDP-Star (Roche Applied Science). Cloned satellite monomers labeled with biotin-dUTP by PCR were used as hybridization probes.

The abundance of satDNA sequences in Meloidogyne species was estimated by quantitative dot blot analysis using a series of genomic DNA dilutions ranging from 50 to 200 ng. Satellite monomers, excised from a plasmid, were dot-blotted in the range between 0.05 and 1 ng, and used as a calibration curve.

Sequence Analyses

DNA sequence data were compared to the GenBank databases by using the BLAST version 2.0 server at the National Center for Biotechnology Information.

The BLAST servers of M. incognita (http://meloidogyne.toulouse.inra.fr/blast/blast.html) and M. hapla (http://www.pngg.org/cbnp/index.php) genome were used to search for Box 1-containing sequences in the sequenced genomes. Initial sequence manipulations were done by using BioEdit v.7.0.5.3 [24]. Multiple alignments and pairwise sequence identity of monomers and HOR sequences were extracted from ClustalW Output (version 1.83) [25]. Lasergene software package v.7.0.0 (DnaStar) was used in further analyses of repetitive sequences including pairwise alignments, dot plot analyses and PCR primer design. Monomers from cloned multimeric arrays were extracted using Key-String Algorithm (KSA) [26]. KSA algorithm is based on the use of a freely chosen short sequence of nucleotides, called the key string, which cuts a given short sequence at each location within multimeric satDNA sequence. Distribution of monomer sequence variability was analyzed by using DnaSP v.4.10.9 [27]. The percent occurrence of the most frequent base at each site was calculated for all monomers repeats; this was plotted with the average percent occurrence and standard deviation (SD). A window length of 15 bp with a step size of 2 was used in the analysis. Due to the large number of monomers, neighbor-joining methods were used to construct phylogenetic tree by PAUP 4.0 (100 bootstrap iterations) [28]. Trees were displayed with MEGA 3.1 [29].

Complex satDNA Arrays in M. chitwoodi and M. fallax

The first goal of this work was to characterize the structure and organization of satDNAs in M. chitwoodi and M. fallax genomes. PCR search for sequences related to M. chitwoodi satDNAs 1a, 1b, 1c, 1d, 2a and 2b revealed, except for 2b, orthologous counterparts in the closely related species M. fallax (Figure 1). However, it was not possible to detect any ortologous satDNA in other analysed Meloidogyne species (M. incognita, M. javanica, M. arenaria, M. javanica, M. paranaensis; data not shown).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Organization of satDNAs in M. chitwoodi (ch) and M. fallax (fa) genome.

(A) Electrophoretic separation of PCR products obtained by amplification of genomic DNAs using primers specific for 1a, 1b, 1c, 2a and 2b satDNAs and U1 sequence are shown on upper panel. (B) Southern hybridizations of genomic DNA partially digested with RE-s and probed with 1a, 1b, 1c, 2a and 2b satDNA monomers and with U1 sequence are shown on lower panel. Approximative contribution of particular sequence in the genome, estimated by dot blot, is shown as a percentage indicated below Southern blots. HORs are indicated with asteriks. M is the DNA ladder marker. ND-not detectable.

https://doi.org/10.1371/journal.pone.0067328.g001

Amplification of M. chitwoodi and M. fallax genomes with primers specific for satDNAs 1a, 2a and 2b produced ladder of bands based on the monomer size while other satDNA amplicons displayed different profiles (Figure 1A). PCR analysis of 1b showed bands of monomeric and dimeric size together with a fragment of about 1.5 kb in length, while amplification with 1c (Figure 1A) and 1d primers revealed complex but similar profiles (shown only for 1c). In order to perform detailed analyses of organizational patterns of these satDNA repeats, first we focused on sequences obtained by amplification of both genomes with 1c satDNA primers. In total, 20 cloned PCR fragments corresponding to multimeric size (i.e. ≥500 bp) were sequenced (Table S2). Two types of satDNA arrays were obtained, distinctive by the composition and organizational complexity of repeat subunits. The first type is characterized by arrays composed of alternating 1c and 1d satDNA monomers which together define the dimeric unit, 338 bp long (169 bp×2), organized in homogenous arrays (8 cloned fragments, M_1cfa_n and M_1cch_n; Table S1). Absence of a 170 bp based ladder in 1c PCR amplification supports this dimeric form as the basic repeating unit of these satDNAs. Multiple sequence alignment of another 12 fragments from M. fallax (H_1cfa_n) and M. chitwoodi (H_1cch_n) (Table S2) revealed complex arrays composed of satDNA monomers 1a, 1b, a new 1b’ variant, 1c, 1d and 2a together with U1, yet uncharacterized sequence segment (Figure 2A and Figure S1). BLAST searches did not indicate any relevant sequence homology of U1 with the studied satDNAs or any other sequence deposited in data bases. In the following experiment, U1 specific PCR primers were constructed in order to extend the segments of complex arrays. In both genomes, obtained PCR products revealed fragments of expected lengths (∼1200 and ∼1400 bp) but also generated a shorter fragment of about 700 bp (Figure 1A). Sequence alignment of fragments obtained with U1 primers (H_uch_n and H_ufa_n) and previously cloned complex arrays (H_1cfa_n and H_1cch_n) is consistent with tandem organization of the HOR unit (Figure S1 and Figure 2A). In contrast to homogeneity of HOR units (84–99% mutual sequence identity) neighboring monomers in HORs show a wide range of relationships: from relatively high sequence identity of 86% between 1b and 1b’ variants to apparently unrelated sequences sharing only 32% identity, such as detected between 2a and 1c monomers (Figure 2A and Table 1). In addition, HOR segments revealed two variants which differ in the presence of 1b-type monomers (Figure 2A and Figure S1). Long HOR variants have two consecutive monomers, 1b and 1b’, that share sequence identity of 86% (Table 1), while short HOR variants lack 1b monomer (Figure 2A). Genomic DNA cut with the REs specific for 1c monomer sequence and probed with the labeled 1c monomer fragment supports the proposed HOR tandem organization (marked with asterisks in Figure 1B). Southern hybridization of genomic DNA with 1c indicates that long HOR variants prevail in M. fallax genome, while short variants seem to be more abundant in M. chitwoodi (Figure 1B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Schematic representation of complex satDNA structure.

(A) The long-L and short-S HOR sequence and (B) complex fragment. The percent identity between monomers is written on arrows above the scheme. Box 1 and Box 2 in junction regions between different monomeres are indicated. 1d* and 1c* represent monomer parts which remain after 2a insertion. The red line below the complex fragment represents the overlapping segment of 1a and 1d monomers. (C) The scheme in the frame represents outcome of the proposed cut-and-paste mechanism of 2a insertion in HOR array.

https://doi.org/10.1371/journal.pone.0067328.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Mean of percent sequence identity between main groups of satDNAs monomers.

https://doi.org/10.1371/journal.pone.0067328.t001

In addition to HORs, the alignment of the 700 bp-long complex fragments amplified with U1 primers revealed one additional homogenous group of sequences common for M. fallax and M. chitwoodi, named h_ufa_n and h_uch_n, respectively (Table S1). These sequences are composed of 1a and 1d complete monomers linked to a novel 170 bp long fragment named U2. The whole composite fragment is flanked by U1 sequences (Figure S2 and Figure 2B). It has to be noted that a 62 bp-long perfectly conserved fragment of U1 is also found as a part of U2 sequence. Additional PCR analyses using U2 specific primers could not prove tandem organization of the 700 bp complex fragment (data not shown). It can be therefore concluded that this fragment probably represents a particular combinatorial form of 1a and 1d satellite repeating units, present in the genome as an interspersed repeat.

Homogenous Monomeric Arrays

PCR with 1a-specific primers produced ladder-like profiles in both genomes, with fragments corresponding to multimers of 170 bp (Figure 1A). Cloning and sequencing (Table S2) revealed homogenous tandem arrays (94% mutual identity) composed of a variant of 1a satDNA sequence, indicated now as 1aM. This variant is different from the HOR variant detected above, which is therefore indicated as 1aH. Average sequence identity between 1aH and 1aM variants is 81% (Table 1). Southern blot hybridization of genomic DNA probed with cloned 1aM-type satDNA repeats confirmed tandem organization of 1aM variants (Figure 1B). In addition, 1aH-specific primers were constructed to check if 1aH builds independent tandem arrays. PCR reaction did not reveal any ladder-like profile (data not shown) indicating that these variants are exclusively present as subunits of HORs.

We also examined if 1b variants could be found in monomeric arrays or are an exclusive component of HOR elements. Southern blot analysis of genomic DNA showed hybridization signals only in bands corresponding to HOR arrays (Figure 1B). PCR with 1b primers revealed fragments whose length corresponds to HOR organization. According to primer position, fragments of monomeric and dimeric forms that appeared in the PCR reaction (Figure 1A) originate from HORs, as they were undetectable in genomic Southern blot. These results emphasize unique organization of 1b monomers exclusively in HORs in both genomes.

It has been published previously that 2a satDNA exists in the M. fallax and M. chitwoodi genome in tandem arrangement, in a high copy number, organized as homogenous monomeric arrays [19]. The results provided in this work show that 2a satellite also exists as the element of HORs in both genomes (Figure 2). No diagnostic sequence differences could be observed with respect to organizational pattern or species of origin (Figure 3 and Figure S3). The only difference is in abundance of 2a satDNA, 3.5% in M. chitwoodi and 20% in M. fallax (Figure 1). Examination of 2b satellite by PCR amplification and Southern blot (Figure 1) confirmed its exclusive presence in the M. chitwoodi genome in the form of high copy homogenous monomeric arrays. The only observed hybridization signal in M. fallax is the faint band (Figure 1B) which could represent a sporadic 2b sequence embedded in a longer DNA segment.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. The phylogenetic tree of 1a, 1b, 1b’ 1c, 1d, 2a and 2b monomers.

Monomers from the HORs (H), dimeric (D) and monomeric arrays (M). Phylogenetic analysis of 212 monomers was performed by neighbor-joining method with bootstrap value of 100. Numbers at nodes indicate bootstrap values (100 replicates; only values greater than 70 % are shown.

https://doi.org/10.1371/journal.pone.0067328.g003

Phylogenetic Analyzes of Monomers

In an effort to assess sequence dynamics of repetitive units in the closely related M. chitwoodi and M. fallax genomes, we examined phylogenetic relationships of all monomers, regardless to their organizational pattern and species origin. A total of 212 monomeric units from M. chitwoodi and M. fallax were included in the multiple sequence alignment (Figure S3). Neighbor-joining phylogenetic analysis showed eight different clusters (1aH, 1aM, 1bH, 1b’H, 1cDH, 1dDH, 2aMH and 2bM; letters H, D, M indicate HOR, dimeric or monomeric organizational form, respectively) distributed in two main branches, satDNAs of group 1 and group 2 (Figure 3). Monomers within clusters could not be distinguished according to the species of origin nor was it possible to differentiate 1c, 1d and 2a monomers according to their array affiliation. In agreement with previous observation, 1a split in 1aM and 1aH according to their organizational origin, while 1b monomers form two distinct groups 1bH and 1b’H related to their position in HORs. It should be noted that 1aH further clusters in two subgroups, based on short and long HOR forms.

Sequence comparisons between monomer groups display three different levels of similarity (Table 1). Similarity is high within 1bH group (86%) and between 1aM and 1aH (81%) monomer variants. Similarities within other satDNAs of group 1 and within satDNAs of group 2 are moderate, ranging from 51 to 66%. Comparison between satDNAs of group 1 and 2 gives negligible similarities, 32–46% (Table 1), and it can be supposed that these two groups might represent sequences of unrelated origin.

Conserved Motifs and Junctions Between Monomers

In contrast to the very low overall sequence similarity between some of the monomer groups (Table 1), pairwise sequence alignment and sliding window analysis of all monomer sequences identified common domains of low variability (Figure 4A, B). The shaded domain in consensus sequences indicates the region of low variability shared among all satDNAs. Part of this region is a conserved 17 bp long segment, named Box 1. It is interesting to note that this sequence segment remains conserved among highly divergent satDNAs. For example, 1c and 2a satDNAs share only 32% identity while in the same time one single change characterizes the Box 1. Comparison of conserved Box 1 sequences (in Figure 4C presented as a reverse complement) with the human CENP-B box shows significant degree of similarity. Six of them have 10–12 out of 17 nucleotides conserved and if bases essential for CENP-B binding in human are considered, 4–5 out of 9 remain conserved. The lowest identity is in exclusively HOR-included elements, 1b’H and 1bH, in which sequences may represent degenerate variants of the motif. This analysis was extended with the search for related motifs in sequenced M. incognita and M. hapla genomes. Preliminary results recovered different repetitive sequences with the Box 1 in unassembled part of M. incognita sequenced genome (Figure S4). However, none of these repeats indicated any sequence similarity with satDNA sequences studied in this work in M. chitwoodi and M. fallax. In addition, detailed analysis of HOR elements in M. chitwoodi and M. fallax revealed that transitions from 1d to 2a monomer and from 2a to 1c are located exactly at the Box 1 (Figure 2A, see Discussion).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Conserved motifs in satDNAs.

(A) Consensus sequences of 1dMH, 1cMH, 1aH, 1bH, 1b’H, 1aM, 2aMH and 2bM satDNAs, determined according to the 50% majority rule. Conserved Box 1 and Box 2 are indicated within the boxed area, and shaded part represents a region of low variability.(B) Identification of low variable domains by sliding window analysis by DnaSP. The average nucleotide variability P is shown by a solid line, and dashed lines represent 2-fold value of standard deviation. (C) Comparison of two variants of Box 1 with the consensus of human CENP-B box. The reverse complementary sequence of Box 1 is presented. Identities between sequences are highlighted in grey, and bases considered essential to bind the CENP-B protein in human [34] are highlighted in red. The number of total conserved bases is reported in brackets. (D) Aligment of Box 2 sequences from HOR related monomers; positions identical to the overall consensus are shown with dots.

https://doi.org/10.1371/journal.pone.0067328.g004

Another common region (Box 2) conserved in HOR-related monomers of group 1 satDNAs (1aH, 1bH, 1b’H, 1cH and 1dH) (Figure 4D and Figure S5). In order to refine alignment of this sequence motif, Box 2 segments were compared with their consensus sequence (Figure 4D). This region is 20 bp-long composed of T, C and A tracts and shows significant degree of mutual sequence identity with only few nucleotide changes (Figure 4D). It must be noted that the Box 2 region is always found in HORs as a transition region between monomers from group 1 (Figure 2A). In addition, detailed analysis of the so-called complex fragment revealed that 1a monomer extends into 1d monomer in the 50 bp long overlapping region shared by both monomers. This whole segment is highly conserved, with only 6 nucleotide substitutions (Figure S2).