Interplay of Selective Pressure and Stochastic Events Directs Evolution of the
MEL172 Satellite DNA Library in Root-Knot Nematodes
Nevenka Meštrović,* Philippe Castagnone-Sereno, and Miroslav Plohl*
*Department of Molecular Biology, Ru;er Bošković Institute, Zagreb, Croatia; and Interactions Plantes-Microorganismes et Santé
Végétale, INRA/UNSA/CNRS, UMR1064, Sophia Antipolis, France
According to the library model, related species can have in common satellite DNA (satDNA) families amplified in differing
abundances, but reasons for persistence of particular sequences in the library during long periods of time are poorly understood. In this paper, we characterize 3 related satDNAs coexisting in the form of a library in mitotic parthenogenetic
root-knot nematodes of the genus Meloidogyne. Due to sequence similarity and conserved monomer length of 172 bp, this
group of satDNAs is named MEL172. Analysis of sequence variability patterns among monomers of the 3 MEL172 satellites revealed 2 low-variable (LV) domains highly reluctant to sequence changes, 2 moderately variable (MV) domains
characterized by limited number of mutations, and 1 highly variable (HV) domain. The latter domain is prone to rapid
spread and homogenization of changes. Comparison of the 3 MEL172 consensus sequences shows that the LV domains
have 6% changed nucleotide positions, the MV domains have 48%, whereas 78% divergence is concentrated in the HV
domain. Conserved distribution of intersatellite variability might indicate a complex pattern of interactions in heterochromatin, which limits the range and phasing of allowed changes, implying a possible selection imposed on monomer sequences. The lack of fixed species-diagnostic mutations in each of the examined MEL172 satellites suggests that they
existed in unaltered form in a common ancestor of extant species. Consequently, the evolution of these satellites seems
to be driven by interplay of selective constraints and stochastic events. We propose that new satellites were derived from an
ancestral progenitor sequence by nonrandom accumulation of mutations due to selective pressure on particular sequence
segments. In the library of particular taxa, established satellites might be subject to differential amplification at chance due
to stochastic mechanisms of concerted evolution.
Introduction
Satellite DNA (satDNA) sequences are highly reiterated
genomic components located in heterochromatic compartments including functional centromeres. They often differ
in copy number, nucleotide sequence, and composition, making entire profile of satDNA sequences species specific.
However, specificity of satellite profiles can only be the result
of rapid fluctuations in copy number within a set of satDNAs,
without changes in nucleotide sequences (reviewed in
Ugarković and Plohl 2002). According to the library model,
an ancestor of closely related species contains a set of
satDNAs within its genome, and as new species emerge,
one or few satDNAs become predominant whereas other repeats remain present as low–copy number sequences (Fry
and Salser 1977; Meštrović et al. 1998). Differential amplification/deletion of (peri)centromeric satellites from the library can be linked with speciation processes (Meštrović
et al. 1998). However, the library model does not predict
if the choice of diverse satellite families able to enter and
persist in a library is driven solely by random processes or
the selection of satellite families is constrained.
SatDNA sequences evolve according to principles of
concerted evolution, in which mutations are homogenized
throughout a repetitive family and fixed within a group of
reproductively linked organisms in a stochastic process of
molecular drive (Dover 1986). However, study of sequence
variability in Arabidopsis centromeric satellite and in human a-satellite (Hall et al. 2003) showed nonrandom patterns of evolution within their monomer sequences. This
study highlighted the hypothesis that the number and distribution of changes in satDNA monomers are the result of
Key words: satellite DNA library, sequence variability, concerted
evolution, selective pressure, parthenogenesis, root-knot nematode.
E-mail: [email protected].
Mol. Biol. Evol. 23(12):2316–2325. 2006
doi:10.1093/molbev/msl119
Advance Access publication September 18, 2006
Ó The Author 2006. Published by Oxford University Press on behalf of
the Society for Molecular Biology and Evolution. All rights reserved.
For permissions, please e-mail: [email protected]
a balance between forces that generate and spread sequence
changes and the selective pressure(s) acting on satellite
sequences. The hypothesis is supported by recent observations that satDNAs may be involved in various functional
interactions in heterochromatin. For example, satDNAs located in a region of functional centromere are proposed
to interact specifically with centromere-specific, histonelike variants and drive their evolution (Talbert et al.
2004; Henikoff and Dalal 2005, and references therein).
Recent analyses showed that transcripts of satDNAs can
be functional in the form of small interfering RNAs, which
act as signals in heterochromatin formation and maintenance (Martienssen 2003).
Root-knot nematodes (RKNs) of the genus Meloidogyne are agriculturally important, widely distributed plantparasitic nematodes, with species that mostly reproduce
by parthenogenesis (mitotic or meiotic), whereas amphimixis is encountered less frequently (Castagnone-Sereno
2006). In our previous study, we characterized 2 high–copy
number satDNAs in the mitotic nematodes Meloidogyne
javanica and Meloidogyne paranaensis (Meštrović et al.
2005). The satellite repeats from M. javanica could not be
resolved from those described earlier in Meloidogyne arenaria (Castagnone-Sereno et al. 2000) and were therefore indicated as a single satDNA family named MARJA. In addition,
this satDNA showed 65% similarity in comparison with the
MPA1 satDNA characterized in M. paranaensis (Meštrović
et al. 2005). Intra- and intervariability analysis of MARJA
and MPA1 satDNAs disclosed one domain of high and
one of low variability in segments of monomer sequences,
equilocated within and between the 2 satellite families.
In order to study sequence dynamics of satDNAs in
congeneric Meloidogyne species, in this paper, we characterize additional MARJA and MPA1 monomers from Meloidogyne incognita and Meloidogyne hapla B and a new
related class of satDNA, AJL, identified in M. arenaria,
M. javanica, and M. incognita. New monomer sequences
Evolution of the MEL172 Satellite DNA Library 2317
were analyzed together with those obtained previously,
MARJA form M. arenaria and M. javanica and MPA1
from M. paranaensis (Castagnone-Sereno et al. 2000;
Meštrović et al. 2005) by phylogenetic approach, by analysis of distribution of polymorphic nucleotide positions and
by the study of transition stages in the process of sequence
homogenization. To our knowledge, for the first time, analysis of spread and homogenization of mutated nucleotides
along the monomer sequences of MEL172 satDNAs disclosed domains with 3 different levels of variability conserved across the set of related satDNAs.
Materials and Methods
Sampling and DNA Isolation
Five parthenogenetic Meloidogyne species were used
in this study: M. javanica (Brazil), M. paranaensis (Brazil),
M. incognita (France), M. arenaria (France), and M. hapla
strains A and B (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 (Carneiro et al. 2000). Eggs were collected
from infested roots, according to the procedure described
earlier (Castagnone-Sereno et al. 2000). For each nematode
isolate, total genomic DNA was purified from 100 to 200 ll
eggs using the DNeasy Tissue Kit (Qiagen, Courtaboeuf,
France) according to the manufacturer’s instructions.
Detection and Cloning of Satellite Sequences
Satellite sequences were amplified in Meloidogyne
species with specific primers designed on consensus sequences derived from MARJA and MPA1 satellites analyzed previously (Meštrović et al. 2005) and from the
AJL satDNA characterized in this study. The following
primer pairs were used to amplify sequences homologous
to MARJA (aj 5#-ACGAT AAACTTTGTGGGGTAA-3#
and aj1 5#-GGGTACAATTCGTCTTGGTAA-3#), MPA1
(p1 5#-TGATTTTCTGGATGGTGATACG-3# and p2
5#-GATTGGGTACAGGTCGTGGT-3#), and AJL (ajL1
5#-AATGACGTTTAAATTGTTTTAATATCG-3# and ajL2
5#-TTTATGACTGGCACTATTCAAAG-3#) satDNAs.
Positions of primers are indicated in figure 5. Amplifications
were performed with HotMaster Taq DNA polymerase
(Eppendorf, Hamburg, Germany) using annealing temperature of 55 °C and 1-min extension at 65 °C for 35 cycles.
Southern hybridization analysis of polymerase chain reaction (PCR) products was performed with cloned satellite
monomers (MARJA, MPA1, and AJL) as probes. Hybridization was done using the DIG DNA Labeling and Detection kit
(Roche Biochemical, Mannheim, Germany) according to the
manufacturer’s instructions under high stringency conditions (hybridization temperature: 68 °C; washing: 0.1 3
SSC, 1% sodium dodecyl sulfate (SDS) at 66 °C). Under relaxed conditions, PCR amplification (annealing at 52 °C) and
Southern hybridization (60 °C; washing: 0.1 3 SSC, 1% SDS
at 58 °C) revealed the same pattern (data not shown). The
control of PCR reaction was done with internal transcribed
spacer (ITS) primers (Zijlstra 1997). The possibility of contamination with nonspecific DNA was excluded through re-
actions with sequence-characterizedamplifiedregion(SCAR)
primers specific for M. javanica, M. arenaria, and M. paranaensis. SCAR primers produce randomly amplified polymorphic DNA fragments that are specific for 1 species only
(Randig et al. 2002). The amount of satDNAs in Meloidogyne
species was estimated by quantitative dot-blot analysis, as described previously (Meštrović et al. 2005).
The selected products amplified by PCR with primer pair
specific for MPA1, MARJA, and AJL satellites were ligated
into a pGEM T-vector System I (Promega, Madison, WI) and
transformed in Escherichia coli DH5a-competent cells. Recombinant clones with dimers and trimers of satDNA were
sequenced on both strands by Genome Express (France). The
sequences were trimmed to the monomer length avoiding
primer-binding sites to prevent any effect on the analysis
of sequence variability. Monomer sequences obtained in this
work were deposited in European Molecular Biology Laboratory databank under accession numbers: DQ435688–
DQ435766 and DQ834644–DQ834669.
Sequence Analysis
Multiple sequence alignments were done using ClustalX. Phylogenetic analyses were performed using the parsimony criterion in PAUP*, version 4b10 (Swofford 1998).
All nucleotide positions were considered equivalent and
weighted equally, using gaps as missing data. A parsimony
tree was obtained after 500 random searches with Tree Bisection-Reconnection branch swapping and the Multrees option. Bootstrap values were calculated on 500 replicates.
The nucleotide diversity was estimated using DnaSP program v.3.99 (Rozas J and Rozas R 1999). The nucleotide diversity Pi is defined as the average number of nucleotide
differences per site between any 2 DNA sequences chosen
randomly from the sample population. The conserved and
variable regions in satDNAs were defined by sliding window
analysis using a window length of 15 bp and a step size of 2 bp.
Windows exhibiting variability Pi (average 1 2 standard
deviation) or Pi (average 2 standard deviation) were considered significant.
The variation pattern at each nucleotide position of
monomer sequences shared between 2 groups of sequences
was analyzed by the method of Strachan et al. (1985), recently reintroduced in analysis of tandemly repeated
sequences (Pons et al. 2002). This method enables comparisons between 2 sets of sequences. Each nucleotide position is classified into 1 of the 6 homogenization classes
(Strachan et al. 1985), which we further grouped into 3 transition stages: homogenized (class 1), intermediate (classes
2–4), and advanced (classes 5 and 6). The homogenized
stage represents complete homogeneity of a nucleotide position across all monomers from 2 satDNAs. The intermediate stage includes gradual spreading of a new mutation in
1 satDNA family, whereas the other family remains homogeneous for a progenitor base. The advanced stage includes
positions where the 2 satDNAs are completely homogeneous for different nucleotides (class 5) and positions in
which homogenized mutation in one set of sequences is
subsequently replaced with a new one (class 6).
Detection of internal repeats was performed on the
consensus sequence through the online program OligoRep
2318 Meštrović et al.
FIG. 1.—Electrophoretic separation of PCR products obtained by amplification of genomic DNAs (20 ng) from Meloidogyne species using primers
specific for (a) MARJA, (c) MPA1, and (e) AJL satDNAs, and corresponding Southern blots after hybridization with cloned (b) MARJA, (d) MPA1, and
(f )AJLsatDNAmonomers.ThelanesaremarkedaccordingtoisolatesofgenomicDNAs:A1andA2,Meloidogynearenaria;J1andJ2,Meloidogynejavanica;I1
and I2, Meloidogyne incognita; P1 and P2, Meloidogyne paranaensis; HA, Meloidogyne hapla A; and HB, M. hapla B. Amplified products from PCR
reactions indicated with asterisk were cloned and sequenced. Line N represents a control of PCR reaction (primer pair without DNA), and M is the size marker.
(g) Control hybridization of cloned repeats of MARJA (lane A), MPA1 (lane B), and AJL (lane C) with corresponding probes. PCR reactions on genomic DNAs
with SCAR primers specific for (h) M. javanica, (i) M. arenaria, and (j) M. paranaensis. (k) Control PCR reaction with primers specific for ITS.
(http://www.mgs.bionet.nsc.ru/mgs/programs/oligorep/).
Tertiary structure was studied by constructing 3-dimensional
models of DNA helix axis by the online program model.it
(http://www.icgeb.trieste.it/dna/).
Results
Distribution of MARJA and MPA1 SatDNAs in
Meloidogyne spp.
PCR amplification with primers specific for MARJA
(Meštrović et al. 2005) and the corresponding Southern hybridization analysis revealed a well-defined ladder of bands
with expected fragment length in M. incognita, M. paranaensis, and M. hapla B but not in meiotic M. hapla A
(fig. 1a and b). Genomic DNAs from M. arenaria and
M. javanica were not included in this analysis because they
harbor already characterized high-copy MARJA repeats
(Castagnone-Sereno et al. 2000; Meštrović et al. 2005). Results of PCR amplification of M. arenaria, M. javanica, M.
incognita, and M. hapla A and B genomic DNAs with primers specific for MPA1 satDNA and subsequent Southern
hybridization analysis of amplified fragments are presented
in figure 1c and d. Again, the source of previously determined high-copy MPA1 repeats, M. paranaensis, was
not included in the PCR screen (Meštrović et al. 2005).
Bands of expected size were obtained in the amplification
of M. incognita and M. hapla A and B genomic DNAs, although in the latter signals were of very low intensity. After
amplification of M. arenaria and M. javanica genomic
DNAs, bands of expected size appeared masked within intensive smear of fragments of various size. Because primer
construction was limited by monomer length, high A 1 T
content (.70%), and sequence similarity, the smeared signal might be a consequence of nonspecific annealing of p1
to MARJA satellite (see fig. 5), which is present in high–
copy number in these species.
Characterization of Satellite Repeats
PCR products obtained after amplification of genomic
DNA from M. incognita and M. hapla B with primers specific for MARJA (fig. 1a) and for MPA1 (fig. 1c) were
cloned and sequenced. MARJAi and MARJAhB monomers
were obtained after amplification of M. incognita and M.
hapla B genomic DNA with MARJA primers, respectively.
Nucleotide diversity within MARJAi (Pi 5 0.050) and
MARJAhB (Pi 5 0.056) monomers is slightly higher
with respect to the diversity of original MARJA monomers
(Pi 5 0.040) from M. arenaria and M. javanica determined
previously (Meštrović et al. 2005), whereas the overall nucleotide diversity is 0.048 (table 1). Alignment of all MARJA
monomers did not reveal any species-diagnostic mutations,
and the whole set is therefore indicated as MARJAall
(Supplement 1, Supplementary Materials online).
MPA1 homologs from M. incognita and M. hapla B
were cloned and sequenced. Monomers from M. hapla B
(MPA1hB), with internal variability of 0.045, did not show
any specific features compared with the original MPA1
satDNA determined previously (Meštrović et al. 2005).
On the contrary, sequence analysis of monomers obtained
after amplification of M. incognita genomic DNA with primers specific for MPA1 disclosed 2 groups of repeating
units. One group of monomers (MPA1i) falls into the range
of internal variability (Pi 5 0.050) of original MPA1 monomer variants. All MPA1 variants determined from examined taxa are indicated as MPA1all (see Supplement 1,
Supplementary Materials online). Interestingly, the second
group of highly homogenous set of monomers (AJL1i)
shows lower divergence when compared with MARJAall
than with MPA1all repeats, despite the fact that AJL1i
monomers were obtained after amplification with primers
specific for MPA1 (fig. 5). To reduce a potential bias
induced by nonspecific annealing of MPA1 primers, new
Evolution of the MEL172 Satellite DNA Library 2319
Table 1
Summary of Nucleotide Differences between MEL172
Monomers
MARJAall
AJLall
MPA1all
MARJAall
AJLall
MPA1all
0.048 6 0.004
0.130 6 0.002
0.203 6 0.007
0.018 6 0.002
0.166 6 0.007
0.052 6 0.004
NOTE.—Intersatellite variability is indicated in bold.
primers specific for AJL1i monomers (fig. 5) were designed
with changed 3# end with respect to the MPA1 and MARJA
satDNA sequences. To check for the presence of AJL1i
monomers in Meloidogyne species, we performed PCR
with specific primers and subsequent hybridization analysis. The results revealed a regular ladder-like pattern in M.
arenaria, M. javanica, and M. incognita, but it was not possible to detect any band in M. paranaensis and in both
strains of M. hapla (fig. 1e and f). Phylogenetic analysis
(fig. 2) of all AJL monomers from M. incognita (AJL1i
and AJL2i), M. arenaria (AJL2a), and M. javanica (AJL2j)
does not show any subclustering. This new homogeneous
group of repeats is indicated as AJLall.
Sequence alignment of all monomer variants analyzed
above with the original MARJA and MPA1 monomers described previously (Meštrović et al. 2005) showed that the
divergence is mainly due to nucleotide substitutions, with
120 variable sites out of 172 nt positions (see Supplement 1,
Supplementary Materials online). The phylogenetic analysis (fig. 2) showed clustering of monomers into 3 groups
(MARJAall, MPA1all, and AJLall) in the maximum parsimony tree, as described above. The 3 groups of sequences
are resolved with high confidence, showing clearly that
they belong to the 3 homogeneous satDNAs. Within each
satDNA, monomer sequences could not be grouped according to the species of origin. Due to similarities in nucleotide
sequence and monomer length, examined satellites are indicated as MEL172 group. Sequence analysis in order to
detect substructure of satDNA monomers based on direct
or inverted repeats as well as analysis of tertiary structure
induced by bent DNA helix axis did not reveal any prominent feature (not shown).
The relative contribution of MARJA, MPA1, and AJL,
assessed from dot-blot experiments, show 2 distinctive levels of abundance in congeneric Meloidogyne species (table
2): the high–copy number (.10,000 copies per genome) and
the low–copy number (between 100 and 1,000 copies per
genome). Under conditions used in performed experiments,
it was not possible to detect particular satDNAs in some species, either in dot-blot or in PCR analyses. However, level of
sensitivity could not definitively exclude presence of residual copies of examined satDNAs in those species. It can be
concluded that all 3 related satDNAs described above are
differentially amplified in the set of congeneric Meloidogyne
species, making a library of conserved satellite sequences.
Distribution of Variability among MEL172 SatDNAs
The study of the distribution of nucleotide diversity
among MEL172 satDNAs was performed by sliding window analysis. Sets of satDNA monomers were already de-
fined in the phylogenetic analysis as MARJAall, AJLall,
and MPA1all satDNAs (fig. 2). First, we performed pairwise analysis of examined satDNAs (fig. 3a–c) and then,
using the same criteria, analysis on all 3 sequence sets together (fig. 3d). The analysis revealed 2 low-variable (LV)
(LV1 and LV2), 2 moderately variable (MV) (MV1 and
MV2), and 1 highly variable (HV) domain. It is difficult
to define precisely the range of particular domains in monomer sequences, and approximate limits of each domain were
determined as interception of the curve in figure 3d and the
corresponding Pi 6 2 standard deviation value. Both LV domains remained unaffected irrespectively of the nucleotide
divergence between compared satDNAs taken in the variability analysis. Between them, the most conserved sequence
segment is the 19 bp long LV2 domain, with only 1 homogenized mutation (see fig. 5). MV domains showed a similar
level of variability irrespectively of the sequence set taken in
the analysis (fig. 3). In the HV domain, the variability increases in parallel with overall nucleotide divergence between satDNAs included in the analysis (fig. 3). However,
overall intrasatellite variability is low in all examined satDNAs (table 1), and mutations are in general uniformly distributed along the monomer (not shown). The only exceptions
are 2 segments in MARJA and MPA1 satDNAs in which intrasatellite variability roughly corresponds to intersatellite
distribution of mutations (Meštrović et al. 2005).
In order to study dynamics of nucleotide changes in
the process of sequence homogenization in domains defined above, we analyzed transition stages by the method
proposed by Strachan et al. (1985), modified as described
in Materials and Methods. Each position in the pairwise
analysis of above-defined sets of satDNA monomers was
classified into 1 of the 3 transitional stages: homogenized,
intermediate, and advanced stage. The results of the analysis are presented in figure 4a for LV domains LV1 and
LV2 together, in figure 4b for both MV domains MV1
and MV2, and in figure 4c for the HV domain. Limits of
domains were as in sliding window analysis (fig. 3) and
are also shown in figure 5. Most positions (.70%) in
LV domains in all comparisons (fig. 4a, columns I–III; divergence Pi over the entire monomer length is indicated
at the top) belong to the homogenized stage, and the rest
of positions can be mostly classified in the intermediate
stage (18–22% of mutated positions). In MV domains
(fig. 4b, columns I–III), the number of sites in homogenized
stage decreases, whereas the number of sites in intermediate
stage increases in parallel with divergence between sets of
satDNAs. Interestingly, the portion of mutations in advanced stage (from 21% to 25%) remains similar regardless
of the divergence between sets of satDNAs, indicating a
limited homogenization of new mutations in the MV domain. On the contrary, the number of sites falling in the
advanced stage increases dramatically in the HV domain
(from 20% to 60%; fig. 4c, columns I–III) in parallel with
increased sequence divergence.
The comparison among the consensus sequences of
each MEL172 satDNA (fig. 5) shows that sequence divergence is concentrated in the HV domain, in which 29 nt
positions out of 37 are changed (78%). In MV domains,
42 positions out of 88 (48%) are changed. However, only
2320 Meštrović et al.
FIG. 2.—Maximum parsimony tree of the 175 MEL172 monomers: group MARJAall: MARJA monomers from Meloidogyne arenaria
(MAR-n; Castagnone-Sereno et al. 2000), Meloidogyne javanica (MJA-n; Meštrović et al. 2005), Meloidogyne incognita (MARJAi-n), and Meloidogyne
hapla B (MARJAhB-n); group AJLall: monomers from M. incognita (AJL1i-n and AJL2i-n obtained with 2 different primer pairs), M. arenaria (AJL2an), and M. javanica (AJL2j-n); group MPA1all: monomers from M. paranaensis (MPA1-n; Meštrović et al. 2005), M. incognita (MPA1i-n), and M. hapla
B (MPA1hB-n). Low-copy satDNAs, sequenced in this work, are indicated in italics.
Evolution of the MEL172 Satellite DNA Library 2321
Table 2
The Copy Number of MARJA, MPA1, and AJL SatDNAs
in Meloidogyne Species Estimated by Dot-Blot and PCR
Analyses
Meloidogyne arenaria
Meloidogyne javanica
Meloidogyne incognita
M. paranaensis
Meloidogyne hapla B
M. hapla A
MARJA
MPA1
AJL
15,700a
12,750b
100–1,000
100–1,000
100–1,000
ND
100–1,000
100–1,000
100–1,000
19,800b
100–1,000
,100
100–1,000
100–1,000
100–1,000
ND
ND
ND
NOTE.—ND, not detected.
a
Castagnone-Sereno et al. (2000).
b
Meštrović et al. (2005).
3 positions are changed in 48 nt long segments of LV
domains (6%).
Discussion
The satDNAs analyzed here in RKNs of the genus Meloidogyne have 2 important features concerning evolution
of satDNA sequences. First, variability analysis revealed
nonrandom accumulation of mutations along monomer sequences of the 3 related MEL172 satDNAs, showing domains with different variability levels preserved across
the set of satDNAs. Second, related satDNAs analyzed
in this work coexist in the majority of examined parthenogenetic taxa in differing abundances, thus forming a library
of conserved repeats.
Intersatellite comparisons show that the entire monomer sequence of MEL172 satDNAs can be subdivided into 2 LV domains, 2 MV domains, and 1 HV domain. The
LW domains are highly homogeneous between satDNAs.
Whereas the MV domains are characterized by the limited
number of mutations homogenized within each satDNA
family, the number of homogenized changes in the HV domain increases in parallel with sequence divergence between MEL172 satDNAs. This observation supports the
idea about extensive homogenization processes in the fixation of changes on ‘‘allowed’’ positions in limited sequence segments of MEL172 satDNAs. Domains with
different variability levels alternating in the monomer sequence of MEL172 satDNAs suggest a complex pattern
of interactions that might be involved, for example, in nucleosome phasing and heterochromatin formation. In that
case, LV domains might represent the yet uncharacterized
DNA sequence motifs involved in DNA–protein interactions. Unfortunately, the small size of nematode chromosomes (,0.5 lm in metaphase) and their holocentric
nature (Triantaphyllou 1985) impose serious limitations
on the precise localization of characterized satDNAs, and
in addition, nothing is known about the heterochromatin
structure and histone variants in holocentric RKN chromosomes (reviewed in Castagnone-Sereno 2006). Polymorphism observed recently in Arabidopsis thaliana
centromeric satDNA and in human a-satellite was interpreted as an indication of constraints imposed on particular
segments of monomer sequence (Hall et al. 2003). It was
suggested that the effect may be caused by functional inter-
actions between the satDNA and various protein components in heterochromatin and/or in the functional
centromere, but direct experimental evidence for this hypothesis is still lacking. Alternatively, the lack of conservation of LV domains in satDNAs from different Arabidopsis
species led to an assumption that regions with reduced polymorphism may serve as sites that promote recombination
within repeats (Hall et al. 2005). A similar alternative function of LV domains might be important in the spread of
MEL172 satDNAs along holocentric chromosomes. Structural characteristics of satDNAs (e.g., dyad structures and
intrinsically bent helix axis) might represent additional features involved in tight packing of DNA in heterochromatin
(Ugarković and Plohl 2002). The fact that no prominent secondary or tertiary structure was detected in any of MEL172
satDNAs reinforces the hypothetical significance of nucleotide sequence variability pattern. The repeat length could
also be an important determinant required for uniform nucleosome phasing and heterochromatin propagation (Henikoff
et al. 2001). The repeat length of MEL172 satellite monomers is well conserved and equals that of human a-satellite
(Durfy and Willard 1990) and A. thaliana satellite monomers
(Hall et al. 2003).
Based on presented results, we propose 2 phases in the
evolution of MEL172 satDNAs, formation of satDNAs,
and persistence of satDNAs in the library. Constitution
of the library must be a process that occurred in ancestral
organism(s), before the onset of extant species. Molecular
studies based on mitochondrial DNA markers have shown
that mitotic parthenogenetic RKN species share a common
lineage and that they diverged from amphimictic/meiotic
species (Hugall et al. 1997). A set of divergent satDNAs
could be formed from an ancestral repetitive sequence
by amplification of mutated monomer variants nascent in
regions of restricted sequence homogenization. In accord
with Dover’s (1986) hypothesis, it has been suggested
that limited sequence homogenization between distantly
located arrays can explain origin of divergence among subsets of Arabidopsis centromeric satellites (Hall et al. 2005).
Similarly, the source of novel (sub)families can be in aberrant satDNA repeats accumulated at array ends due to low
homogenization efficiency in bordering regions (Schueler
et al. 2005). In our case, we propose that mutated monomers
of the ancestral MEL172 sequence can be amplified and
constitute the library of satDNAs only if they have sequence characteristics necessary for putative interactions
in heterochromatin. Therefore, the hypothesis could be
raised that monomers of MEL172 satDNAs were initially
shaped and spread under selective pressure due to functional interactions. Once established, these satDNAs might
be subject to differential amplification in particular taxa
at chance, due to dynamics of recombinational mechanisms in the process of concerted evolution. Consistent
with the library hypothesis, the whole set can be in distant
species replaced with satDNA families of unrelated origin, what might be followed by adaptive coevolution of
DNA-binding components. Results indicating adaptive coevolution of centromere-specific, histone-like variants and
satDNAs in several animal and plant experimental systems
(Henikoff and Dalal 2005) add to the reliability of the proposed scenario.
2322 Meštrović et al.
FIG. 3.—Identification of LV (LV1 and LV2), MV (MV1 and MV2), and HV domains along monomers of MEL172 satDNAs by sliding window
analysis. The joint variability analysis of monomers; (a) MARJAall and AJLall, (b) AJLall and MPA1all, (c) MARJAall and MPA1all, and (d) of all 3
sets of monomers together; MARJAall, MPA1all, and AJLall. The average nucleotide variability Pi calculated by DnaSP is shown by a solid line, and
dashed lines represent 2-fold value of standard deviation.
Based on the hypothesis of ancient origin of M. hapla,
dated about 43 MYA (Hugall et al. 1997), we propose
a long-term conservation of MEL172 satDNAs. Paradoxically, MEL172 satDNAs in the library are interspecifically
conserved in a way that their constitutive monomers remain
unaltered in the entire length, even in domains characterized
as HV. This would indicate that homogenization mechanisms act on the satellite monomer as a whole, favoring
Evolution of the MEL172 Satellite DNA Library 2323
FIG. 4.—Graphic representations of the distribution of 3 transition
stages (modified from Strachan et al. 1985) during spread of new mutations
in pairwise comparisons of MEL172 satDNAs. Comparisons are presented
in column I (MARJAall vs. AJLall), column II (MPA1all vs. AJLall), and
column III (MARJAall vs. MPA1all) for (a) LV, (b) MV, and (c) HV domains of satellite monomers. Nucleotide divergence Pi between monomers
of compared satDNAs is indicated at the top. Percentage of unclassifiable
positions (UP) according to Strachan et al. (1985) is indicated for each transition stage.
persistence of the ancestral monomer sequence. The theory
of concerted evolution predicts that a small bias in favor of
some set of monomer variants can result in extreme sequence conservation, preventing the fixation of new se-
quence variants at the species level (Ohta and Dover
1984; Strachan et al. 1985). It is possible that this effect
of concerted evolution can be enhanced by functional constraints imposed on the sequence. Alternatively, conservation of satellite sequences in sturgeons for more than 100
Myr is explained as an effect of slow rate of both sequence
changes and concerted evolution (Robles et al. 2004).
Long-term conservation of satDNAs among RKN species
presented in this paper and in recent analysis (Meštrović
et al. 2006) is comparable to that found in highly conserved
satDNAs in sexual species (Mravinac et al. 2002, 2005). In
accord with this observation, satellite monomer variants in
parthenogenetic Bacillus taxa constitute an ancestral library, conserved during significant evolutionary periods
(Cesari et al. 2003). Evolution of MEL172 satDNAs does
not show any specificities that could be linked directly with
the parthenogenetic mode of reproduction, similar to that observed in comparative analysis of Bag320 in bisexual and
unisexual Bacillus taxa (Luchetti et al. 2003). These results
indicate that turnover mechanisms in mitotic parthenogens
might be sufficient to preserve some ‘‘optimal’’ set of monomer variants for long periods of time.
Phylogenetic studies suggested that the ancestor of
meiotic/mitotic M. hapla separated first from the other examined species and represents the most divergent species in
the group (Hugall et al. 1997). M. incognita, M. javanica,
and M. arenaria form a sister species cluster, whereas M.
paranaensis is more distant (De Ley et al. 2002; Tenente
et al. 2004). An additional hypothesis based on analysis of
ITS fragments indicates hybridization and subsequent polyploidization during evolution of mitotic lineages (Hugall
et al. 1999). Nevertheless, besides deep ancestry of M. hapla, exact dating and sequence of events in Meloidogyne is
only partially understood (reviewed in Castagnone-Sereno
2006). It has been observed that gradual accumulation of
divergences among satDNA sequences of related species
or populations can generate useful molecular marker on different taxonomic levels, depending on the rate of accumulation of nucleotide substitutions (reviewed in Ugarković
and Plohl 2002). In contrast to that, sequence comparison
of related satDNAs in the MEL172 library has no relevance
to the phylogeny of Meloidogyne species. In that context,
presence/absence of satDNAs in particular Meloidogyne
species can be more informative because taxonomic relationships might be mirrored by the time of satDNA variant
birth or elimination in course of genomic turnover processes. For example, the AJL satDNA is distributed within
the sister cluster M. arenaria, M. javanica, and M.
incognita, whereas it could not be detected in more distant
M. paranaensis and in both races of M. hapla. Inability to
detect most, if not all, MEL172 satDNAs in M. hapla A
may be related to a higher level of sequence dynamics
FIG. 5.—Consensus sequences for each satDNA; MARJAall, MPA1all, and AJLall as determined according to the 50% majority rule. LV (LV1 and
LV2), MV (MV1 and MV2), and HV domains are indicated by boxes. Arrows indicate positions of primer pairs used to amplify satDNA repeats.
2324 Meštrović et al.
due to meiotic reproductive lifestyle and/or specific species/
taxon traits (Luchetti et al. 2003).
In conclusion, we suggest that interplay of stochastic
events and selective pressure can explain constitution and
persistence of related satDNAs in the library. The sequence
features might represent a limiting factor in the formation
and survival of new satDNA families in the library, although
their ultimate sequence dynamics will depend on stochastic
turnover processes in a particular species. In addition to the
examined MEL172 satellites, presence of other satellites of
this group could not be excluded in Meloidogyne species.
In the future, the characterization of nucleosomal phasing
and DNA–protein interactions in heterochromatin compartments of RKN species could represent a convenient experimental system to test constraints that limit changes in satellite
monomers and possible role of these satDNAs in centromeric
function of holocentric chromosomes.
Supplementary Material
Supplement 1 is available at Molecular Biology and
Evolution online (http://www.mbe.oxfordjournals.org/).
Acknowledgments
This work has been supported by the Research Fund of
Republic of Croatia and French Institut National de la
Recherche Agronomique. Short exchange visits were funded
through the Croatian–French bilateral project ‘‘COGITO’’
#06410ZL. We would like to thank Daphne Preuss and Joan
Pons for critical reading and useful suggestions during preparation of the manuscript, as well as to 2 anonymous referees,
whose comments helped to significantly improve the manuscript.
Literature Cited
Carneiro RMDG, Almeida RMA, Qhénéhervé P. 2000. Enzyme
phenotype of Meloidogyne spp. populations. Nematology.
2:645–654.
Castagnone-Sereno P. 2006. Genetic variability and adaptive evolution in parthenogenetic root-knot nematodes. Heredity.
96:282–289.
Castagnone-Sereno P, Leroy F, Abad P. 2000. Cloning and characterization of an extremely conserved satellite DNA family
from the root-knot nematode Meloidogyne arenaria. Genome.
43:346–353.
Cesari M, Luchetti A, Passamonti M, Scali V, Mantovani B. 2003.
Polymerase chain reaction amplification of the bag320 satellite
family reveals the ancestral library and past gene conversion
events in Bacillus rossius (insecta phasmatodea). Gene.
312:289–295.
De Ley IT, De Ley P, Vierstraete A, Kassen G, Moens M,
Vanfleteren J. 2002. Phylogenetic analyses of Meloidogyne
small subunit rDNA. J Nematol. 34:319–327.
Dover GA. 1986. Molecular drive in multigene families: how biological novelties arise, spread and are assimilated. Trends
Genet. 2:159–165.
Durfy SJ, Willard HF. 1990. Concerted evolution of primate alpha
satellite DNA: evidence for an ancestral sequence shared by
gorilla and human X chromosome alpha satellite. J Mol Biol.
216:555–566.
Fry K, Salser W. 1977. Nucleotide sequences of HS-a satellite
DNA from kangaroo rat Dipodomys ordii and characterization
of similar sequences in other rodents. Cell. 12:1069–1084.
Hall SE, Kettler G, Preuss D. 2003. Centromere satellites from
Arabidopsis populations: maintenance of conserved and variable domains. Genome Res. 13:195–205.
Hall SE, Luo S, Hall AE, Preuss D. 2005. Differential rates of local
and global homogenization in centromere satellites from
Arabidopsis relatives. Genetics. 170:1913–1927.
Henikoff S, Ahmad K, Malik HS. 2001. The centromere paradox:
stable inheritance with rapidly evolving DNA. Science. 293:
1098–1102.
Henikoff S, Dalal Y. 2005. Centromeric chromatin: what makes it
unique? Curr Opin Genet Dev. 15:177–184.
Hugall A, Stanton J, Moritz C. 1997. Evolution of the AT-rich
mitochondrial DNA of the root knot nematode, Meloidogyne
hapla. Mol Biol Evol. 14:40–48.
Hugall A, Stanton J, Moritz C. 1999. Reticulate evolution and the
origins of ribosomal internal transcribed spacer diversity in
apomictic Meloidogyne. Mol Biol Evol. 16:157–164.
Luchetti A, Cesari M, Carrara G, Cavicchi S, Passamonti M, Scali
V, Mantovani B. 2003. Unisexuality and molecular drive:
bag320 sequence diversity in Bacillus taxa (insecta; phasmatodea). J Mol Evol. 56:587–596.
Martienssen RA. 2003. Maintenance of heterochromatin by RNA
interference of tandem repeats. Nat Genet. 35:213–214.
Meštrović N, Castagnone-Sereno P, Plohl M. 2006. High conservation of the differentially amplified MPA2 satellite DNA
family in parthenogenetic root-knot nematodes. Gene. 376:
260–267.
Meštrović N, Plohl M, Mravinac B, Ugarković Ð. 1998. Evolution
of satellite DNAs from the genus Palorus—experimental
evidence for the ‘‘library’’ hypothesis. Mol Biol Evol. 15:
1062–1068.
Meštrović N, Randig O, Abad A, Plohl M, Castagnone-Sereno P.
2005. Conserved and variable domains in satellite DNAs of
mitotic parthenogenetic root-knot nematode species. Gene.
362:44–50.
Mravinac B, Plohl M, Meštrović N, Ugarković Ð. 2002. Sequence
of PRAT satellite DNA frozen in some coleopteran species.
J Mol Evol. 54:774–783.
Mravinac B, Plohl M, Ugarković Ð. 2005. Preservation and high
sequence conservation of satellite DNAs suggest functional
constraints. J Mol Evol. 61:542–550.
Ohta T, Dover GA. 1984. The cohesive population genetics of
molecular drive. Genetics. 108:501–521.
Pons J, Petitpierre E, Juan C. 2002. Evolutionary dynamics of
satellite DNA family PIM357 in species of the genus Pimelia
(Tenebrionidae, Coleoptera). Mol Biol Evol. 19:1329–1340.
Randig O, Bongiovanni M, Carneiro RMDG, Castagnone-Sereno
P. 2002. Genetic diversity of root-knot nematodes from Brazil
and development of SCAR markers specific for the coffeedamaging species. Genome. 45:862–870.
Robles F, de la Herran R, Ludwig A, Ruiz Rejon C, Ruiz Rejon M,
Garrido-Ramos MA. 2004. Evolution of ancient satellite
DNAs in sturgeon genomes. Gene. 338:133–142.
Rozas J, Rozas R. 1999. DnaSP version 3: an integrated program
for molecular population genetics and molecular evolution
analysis. Bioinformatics. 15:174–175.
Schueler MG, Dunn JM, Bird CP, Ross MT, Viggiano L, NISC
Comparative Sequencing Program, Rochi M, Willard
HF, Green ED. 2005. Progressive proximal expansion of
the primate X chromosome centromere. Proc Natl Acad Sci
USA. 102:10563–10568.
Strachan T, Webb D, Dover GA. 1985. Transition stages of molecular drive in multiple-copy DNA families in Drosophila.
EMBO J. 4:1701–1708.
Swofford DL. 1998. PAUP*: phylogenetic analysis using parsimony (* and other methods). Version 4. Sunderland (MA):
Sinauer Associates.
Evolution of the MEL172 Satellite DNA Library 2325
Talbert PB, Bryson TD, Henikoff S. 2004. Adaptive evolution of
centromere proteins in plants and animals. J Biol. 3:18.
Tenente GCMV, De Ley P, De Ley IT, Karssen G, Vanfleteren JR.
2004. Sequence analysis of the D2/D3 region of the large subunit rDNA from different Meloidogyne isolates. Nematropica.
34:1–12.
Triantaphyllou AC. 1985. Cytogenetics, cytotaxonomy and
phylogeny of root-knot nematodes. In: Sasser JN, Carter
CC, editors. An advanced treatise on Meloidogyne. Vol. 1.
Raleigh (NC): North Carolina State University Graphics.
p. 113–126.
Ugarković Ð, Plohl M. 2002. Variation in satellite DNA
profiles—causes and effects. EMBO J. 21:5955–5959.
Zijlstra C. 1997. A fast PCR assay to identify Meloidogyne hapla,
M. chitwoodi, and M. fallax, and to sensitively differentiate
them from each other and from M. incognita in mixtures. Fundam Appl Nematol. 20:505–511.
Aoife McLysaght, Associate Editor
Accepted August 23, 2006