Using SINEs to Probe Ancient Explosive Speciation: ‘‘Hidden’’ Radiation
of African Cichlids?
Yohey Terai,*1 Kazuhiko Takahashi,*1 Mutsumi Nishida, Tetsu Sato,à and Norihiro Okada*
*Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, Midori-ku, Yokohama, Japan;
Division of Molecular Marine Biology, Ocean Research Institute, University of Tokyo, Nakano, Tokyo, Japan; and
àConservation Division, WWF Japan, Minato-ku, Tokyo, Japan
Cichlid fishes of the east African Great Lakes represent a paradigm of adaptive radiation. We conducted a phylogenetic
analysis of cichlids including pan-African and west African species by using insertion patterns of short interspersed
elements (SINEs) at orthologous loci. The monophyly of the east African cichlids was consistently supported by seven
independent insertions of SINE sequences that are uniquely shared by these species. In addition, data from four other loci
indicated that the genera Tilapia (pan-African) and Steatocranus (west African) are the closest relatives to east African
cichlids. However, relationships among Tilapia, Steatocranus, and the east African clade were ambiguous because of
incongruencies among topologies suggested by insertion patterns of SINEs at six other loci. One plausible explanation
for this phenomenon is incomplete lineage sorting of alleles containing or missing a SINE insertion at these loci during
ancestral speciation. Such incomplete sorting may have taken place earlier than 14 MYA, followed by random and
stochastic fixation of the alleles in subsequent lineages. These observations prompted us to consider the possibility that
cichlid speciation occurred at an accelerated rate during this period when the African Great Lakes did not exist. The SINE
method could be useful for detecting ancient exclusive speciation events that tend to remain hidden during conventional
sequence analyses because of accumulated point mutations.
Introduction
Cichlid flocks of the east African Great Lakes, which
consist of Lakes Victoria, Malawi, and Tanganyika, have
attracted the interest of evolutionary biologists for more
than a century. These species exhibit extraordinary levels
of diversity and high species endemicity to each lake as the
result of independent explosive adaptive radiation (Fryer
and Iles 1972; Greenwood 1984, 1991; Coulter 1991).
During the last decade, many molecular phylogenetic
studies were conducted to elucidate the phylogenetic
relationships among cichlids (for review see Meyer 1993
and Nishida 1997). These studies suggested that cichlids in
Lakes Victoria and Malawi are closely related, and that
species in both lakes are related to only a portion of the
lineages found in Lake Tanganyika, the oldest Great Lake,
estimated at 9–12 Myr (Cohen, Soreghan, and Scholz
1993). Until recently, little attention was paid to riverine
cichlid species that exhibit pan-African and west African
distribution. However, these riverine species constitute an
important element of any comprehensive phylogenetic reconstruction of African cichlids; knowledge of their origins is indispensable for the elucidation of founder species
in east African lakes and hence for inference of the genetic
basis for the adaptive radiation that occurred in those
lakes. Recent studies (Sültmann et al. 1995; Zardoya et al.
1996; Streelman and Karl 1997; Mayer, Tichy, and Klein
1998; Streelman et al. 1998) have focused on pan-African
and west African species through the use of nuclear
markers. Such markers evolve more slowly than mitochondrial DNA, which had been employed extensively in
previous phylogenetic analyses of east African cichlids.
Recent molecular studies designed to gain inferences from
1
These two authors contributed equally to this work.
Key words: exclusive speciation, incomplete lineage sorting,
African Great Lakes, cichlid, retroposon, SINE, AFC family.
E-mail: [email protected].
Mol. Biol. Evol. 20(6):924–930. 2003
DOI: 10.1093/molbev/msg104
Ó 2003 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038
924
the large framework of cichlid phylogeny again included
a combination of analyses of mitochondrial and nuclear
genes and morphological characters (Farias, Orti, and
Meyer 2000), as well as a reassessment of the mitochondrial cytochrome b gene (Farias et al. 2001). With respect
to achieving an accurate African cichlid phylogeny, however, these studies remain preliminary, either because
taxon sampling of west African cichlids was insufficient
and/or because of poor bootstrap supports for a significant
fraction of their relationships.
Another topic of interest is whether the evolution of
pan- and west African cichlids was accompanied by incomplete lineage sorting and/or interspecific hybridization,
which appears to have occurred multiple times during the
evolution of cichlids in the African Great Lakes (Moran and
Kornfield 1993, 1995; Parker and Kornfield 1997; Nagl et al.
1998; Van Oppen et al. 2000; Takahashi et al. 2001a,
2001b). If this was indeed the case for the pan- and west
African cichlids, then caution must be exercised when
attempting to elucidate their phylogeny, given that the
genealogies of individual loci may not coincide with their
phylogeny (Nei 1987; Pamilo and Nei 1988; Takahata 1989;
Avise 2000).
Recently, a series of successful phylogenetic analyses
was conducted for African Great Lakes cichlids by investigating insertions of the AFC (African cichlid) family
(Takahashi et al. 1998, 2001a, 2001b) of short interspersed
elements (SINEs) into orthologous loci in the nuclear genome (for reviews of the SINE method see Okada 1991;
Cook and Tristem 1997; Miyamoto 1999; Shedlock, Milinkovitch, and Okada 2000; Shedlock and Okada 2000). Short
interspersed elements multiply via retroposition, and their
choice of insertion sites is nearly random (Weiner,
Deininger, and Efstratiadis 1986). In the absence of any
known specific excision mechanism, SINEs remain at the
integration locus indefinitely. These characteristics of SINEs
make them advantageous for elucidating phylogenetic
relationships; the sharing of a SINE sequence at a specific
site can be regarded essentially as a synapomorphy. Short
Using SINEs to Probe Ancient Explosive Speciation 925
Table 1
Fish Species Analyzed by PCR in this Study
Species (tribe)
Haplochromis nyerereia
Labidochromis caeruleusa
Bathybates/Hemibatesb (Bathybatini)
Boulengerochromis microlepisa
(Tilapiini)
Ctenochromis horei (Haplochromini)
Cyprichromis leptosomaa
(Cyprichromini)
Lepidiolamprologus elongatus
(Lamprologini)
Tanganicodus irsacaea (Eretmodini)
Trematocara unimaculatum
(Trematocarini)
Anomalochromis thomasia
Chromidotilapia finleyia
Hemichromis lifalilia
Nanochromis parilusa
Pelvicachromis taeniatusa
Sarotherodon melanotheron
Steatocranus casuariusa
Teleogramma brichardia
Tilapia buttikoferia
Oreochromis niloticus
Oreochromis tanganicae
Tilapia rendallia
Tylochromis polylepisa
Geographic Distribution
(sampling locality)
Lake
Lake
Lake
Lake
Victoria
Malawi
Tanganyika
Tanganyika
Lake Tanganyika
Lake Tanganyika
Lake Tanganyika
Lake Tanganyika
Lake Tanganyika
West Africa
West Africa
West Africa
West Africa
West Africa
West Africa
(Cote d’Ivoire)
West Africa
West Africa
West Africa
Pan-Africa
(Lake Tanaganyika)
Pan-Africa
(Lake Tanaganyikac)
Pan-Africa
West Africa
(Lake Tanganyikac)
NOTE.—The nomenclature of each species follows that of Poll (1986).
a
Specimen purchased from commercial sources in Japan. All the other
specimens were collected in the field.
b
One or two species from Bathybates ferox, Bathybates fasciatus, and Hemibates stenosoma was/were analyzed for each locus as representatives of the tribe
Bathybatini.
c
Although Oreochromis tanganicae and Tylochromis polylepis are unique
to Lake Tanganyika, other closely related species of the same genera show panand west African distribution, respectively.
interspersed elements have also been useful for detection of
both recent (Hamada et al. 1998; Takahashi et al. 2001a)
and ancient (Takahashi et al. 2001b) rapid speciation
accompanied by incomplete lineage sorting of alleles. In
this article we report use of the SINE method to elucidate the
phylogenetic relationships among the ancient cichlid
lineages in Africa, as well as testing for the existence of
incomplete lineage sorting among these lineages to gain
insights into their mode of speciation.
Materials and Methods
Analyses were conducted on 13 species belonging to
11 different genera of pan- and west African cichlids, 7
species belonging to different tribes in Lake Tanganyika,
and 2 representative species from Lakes Malawi and
Victoria (table 1). Tylochromis polylepis, endemic to Lake
Tanganyika, was categorized with pan-African cichlids
because this genus exhibits pan-African distribution.
Eighteen loci at which AFC SINEs (Takahashi et al.
1998; Terai, Takahashi, and Okada 1998) had been inserted were isolated randomly from genomic libraries of
Ophthalmotilapia ventralis (loci 223, 234, 241, 255, and
260), Tropheus moorii (loci 304 and 316), Haplotaxodon
microlepis (loci 450 and 454), Tanganicodus irsacae (loci
504 and 509), Tilapia rendalli (loci 904 and 906), Dimidiochromis compressiceps (loci 1223, 1240, 1280, and
1544), and Haplochromis machadoi (locus 1708). Primers
to sequences flanking SINEs that were inserted into the
above individual loci (table 2) were used in polymerase
chain reactions (PCRs) to amplify these loci from the
genomes of species listed in table 1. The PCR products
were sequenced when confirmation of results was necessary
(accession numbers AB100503-AB100592). Construction
of the phylogenetic tree and detection of possible incomplete lineage sorting was conducted by hand on the
basis of the presence or absence of a SINE sequence at
the individual loci listed above (see Results and Discussion).
All phylogenetic and molecular biology experiments were
performed using standard techniques as described previously (Takahashi et al. 1998).
Results and Discussion
During the analysis of locus 504, long PCR products
were observed among all the endemic species of Lake
Tanganyika examined, suggesting the existence of a SINE
sequence at this locus (fig. 1A, panel a). Similar results
were observed for the representative species from Lakes
Malawi and Victoria (table 3). For all other species, either
pan- or west African, only short PCR products were
observed for locus 504, suggesting the absence of a SINE
insertion. The PCR results were confirmed by Southern
hybridization using two different types of probes. The first
hybridization was performed using a probe specific for
sequences of the AFC SINE, and each long PCR product
yielded a positive signal (fig. 1A, panel b). In contrast, the
shorter PCR products yielded no signal and therefore were
considered not to contain a sequence homologous to AFC
SINE. The second hybridization was conducted using
a probe specific for the flanking sequence of the locus 504
SINE, and all short and long PCR products yielded
signals. We confirmed that all PCR products were
amplified specifically from locus 504 (fig. 1A, panel c).
Although a PCR product of intermediate length was
observed for Boulengerochromis microlepis (panel a), the
Southern blots confirmed that this locus indeed contained
a SINE and that its shorter length was due to the existence
of a small deletion near that locus (data not shown).
Similarly, locus 241 also showed that species endemic to
the east African lakes share a unique AFC SINE sequence
(table 3). The above results indicate that the SINE sequences had undergone independent insertion at these two
loci in a common ancestor of east African cichlids, and
hence suggest their monophyly (hereafter, this clade is
referred to as ‘‘the east African clade’’). The east African
clade consists of Trematocara unimaculatum (Trematocarini tribe), Boulengerochromis microlepis (Tilapiini
tribe), and Bathybates/Hemibates (Bathybatini tribe),
as well as the MVhL lineage, which by an earlier SINE
analysis (Takahashi et al. 2001b) is proposed to consist of
all the other cichlid tribes in Lake Tanganyika and the
926 Terai et al.
Table 2
PCR Primers
Locus
59-Flanking
223
234
241
255
260
304
316
450
454
504
509
904
59-TCTCAGCACAGAAAGTTTACACA-39
59-CTTTGTGATGATGTGAAACTGT-39
59-CAGATGGCTAGTCAGCACTG-39
59-CAGATAGTTGTCAGAAGGTT-39
59-GAGGTACTCTCACTTGGATAGGTT-39
59-TGGAGAAGCCCCATGAATCC-39
59-CTCCACATTTCCCAGGAATC-39
59-AAACAAGACACTGAGACAGCAA-39
59-TGCATGTATCCTGAGGGTAAGA-39
59-CTTAAACGACACCAGAACTGAT-39
59-TATACTGCCACTGTTCAATGGA-39
59-TGTGCATTGTCAACATCCAT-39
59-CCATCAGCAAAGTTGCAGAGT-39
59-TGTTCATAAATAACTGAGCATGTCTTC-39
906
1223
1240
1280
1544
1708
59-CCAGAAACGGGAGGGTCACAT-39
59-CTGTATCCTGACAATGCCTCTCTA-39
59-CCGTTACAGTGGCATGTCAAG-39
59-CACTGAATGGTGAATTTGAATGAGTA-39
59-GCAGTCACACTCAATCAGCCTCTTA-39
59-TCAAATGGATTCCCAAGTCAGTC-39
39-Flanking
59-AGAGGAGGATGCTCGAACTT-39
59-CACATATTATCCAGGACTTTAAG-39
59-GGCCACATGATCAGATATCCTT-39
59-TTTCTACACACTCTCAGTCTGAGG-39
59-GAGGGAAAGAGGAATAAGTTGGA-39
59-GGAGTTACAGTCATTTAGAGAATACTAG-39
59-AACCAGACCAGCAGTGACATAAG-39
59-GCTGTGAGGCTCAACCATTA-39
59-CATTCTGCATCTCTAGTCAAGTCT-39
59-CATTTTTGACACCTAAATGTACAAC-39
59-CATCACTGCTACTGAAACTG-39
59-CCACTTTGCCTTAAGCTTCT-39
59-CTCTCACTTGCTGCCCAGAT-39
59-ACTGTGTAGACTGGGTAAGCAA-39
59-TCGCTGCTGTACTGTTCACTT-39
59-CTACACTCAGCTTCTCACCCTC-39
59-TAGTCCAACATCCAACTTGCAG-39
59-GATTAATGAGGGAAGTCTATGTCTGG-39
59-GGCAGCACAGCTTAGGTTTACA-39
59-TGGTCGAAGTCTAATCGTTTGGTG-39
59-AGGGATACCTGGATGTAAAAGCA-39
NOTE.—Where two primers are indicated for one or both of the 59-/39-flanking regions, PCR was conducted using different combinations of the primers depending on
the species.
endemic cichlids in Lakes Victoria and Malawi. The results for loci 255, 450, 509, 1240, and 1708 were also
consistent with monophyly among the above cichlids,
although in several species these loci failed to yield a PCR
product (table 3, blanks). In all cases these ‘‘failed’’ species
were pan- or west African and appear to be old lineages
based on the monophyly of the east African clade. Thus,
the PCR primers may have failed to anneal to the flanking region of the isolated loci as a result of accumulated
mutations.
A different pattern of SINE insertion was observed at
loci 234 and 454 (fig. 1B and table 3), in which PCR
products containing SINE sequences were observed
among species of the east African clade as well as the
pan/west African species Tilapia rendalli, Tilapia buttikoferi, and Steatocranus casuarius. The remaining pan/
west African species showed no AFC SINE insertions.
These results suggest that the two species of Tilapia and S.
casuarius are the closest relatives to the east African clade.
The SINE insertion patterns obtained from loci 904 and
1280 were consistent with the above relationship, implying
that the two species of Tilapia and S. casuarius are closer
to the east African clade than Oreochromis niloticus,
Sarotherodon melanotheron, and Tylochromis polylepis.
The data for locus 906 also suggested that T. polylepis and
Chromidotilapia finleyi may be more distantly related to
the east African clade than O. niloticus and S. melanotheron. However, this relationship needs to be confirmed by
characterizing positive insertions of other markers at additional loci.
On the basis of the above results, a phylogenetic tree
for the cichlids was constructed (fig. 2). Monophyly of the
east African clade was suggested by earlier studies using
nuclear DNA and/or mitochondrial sequences. However,
these studies involved only a limited number of genera
from pan- and west Africa (3–6 genera; Sültmann et al.
1995; Zardoya et al. 1996; Streelman and Karl 1997;
Mayer, Tichy, and Klein 1998; Streelman et al. 1998;
Farias, Orti, and Meyer 2000; Farias et al. 2001). Our data
from 11 pan- and west African genera for two loci (241
and 504) as well as from 3 to 10 genera for the other five
loci (255, 450, 509, 1240, and 1708; table 3) provide the
most comprehensive support for monophyly of the east
African clade. The sister group relationship between
Tilapia/Steatocranus and the east African clade is
consistent with the phylogeny proposed by a sequence
analysis of the noncoding nuclear locus DXTU1, in which
Oreochromis, Tylochromis, Pelvicachromis, and Hemichromis were included as well (bootstrap value of 75%;
Mayer, Tichy, and Klein 1998). That study also suggested
(bootstrap value of 66%) that the clade consisting of 7
species of Oreochromis is the sister group to the clade
corresponding to our east African þ Tilapia=Steatocranus
clade. The results obtained for locus 906 are also consistent with this hypothesis (table 3). Our results support the
hypothesis that the ancestral stock, which may formerly
have populated west Africa, gave rise to lineages that were
the forebears of Anomalochromis, Chromidotilapia, Hemichromis, Nanochromis, Oreochromis, Pelvicachromis,
Sarotherodon, Teleogramma, and Tylochromis. The divergence of Tilapia, Steatocranus, and the ancestor of the
east African clade then followed this event (see Mayer,
Tichy, and Klein 1998, Discussion). Clarification of the
detailed relationships among the west African lineages will
require further characterization of other loci into which the
AFC SINE has inserted.
Although most of the polytomies in our phylogenetic
tree were due to the lack of phylogenetically informative
loci, those observed among Tilapia, Steatocranus, and the
east African clade were due to incongruent patterns of
SINE insertion at six loci (223, 260, 304, 316, 1223, and
1544; table 3). At locus 260, an AFC sequence was
identified only in T. rendalli and the east African clade
(fig. 3, panel a). At locus 316, an AFC SINE was found
Using SINEs to Probe Ancient Explosive Speciation 927
FIG. 1.—Detection of orthologous loci in cichlid species using PCR.
The three panels (a–c) in A and B show an agarose gel (top) and two
autoradiograms. Panel a, electrophoretic profile of PCR products. Panel b,
Southern hybridization of AFC SINE sequences within the PCR products
shown in panel a. Panel c, rehybridization of the blot in panel b for
detection of a locus-specific sequence flanking the site at which the SINE
unit was inserted. Closed and open arrowheads indicate expected
mobilities of amplified fragments containing or lacking a SINE insertion,
respectively.
exclusively in the above cichlid lineages, as wells as in
T. buttikoferi (fig. 3, panel b), and a similar result was observed for locus 223 (table 3). The most straightforward
interpretation of these results is that T. rendalli represents
the closest relative to the east African clade, with
T. buttikoferi as the next closest relative. However, other data
in our study contradict this interpretation. T. buttikoferi
and S. casuarius shared the locus 1544 SINE insertion
with the east African clade, but no such insertion was
found in T. rendalli (fig. 3, panel c). Locus 1223 showed
a similar pattern of AFC SINE insertion. Here, the simplest
interpretation is that T. rendalli is not the closest relative to
the east African clade. Data for locus 304 further support
the premise that S. casuarius, and not T. rendalli, is the
closest relative of the east African clade. Taken together,
the SINE insertion data for the above six loci suggest conflicting interpretations of which lineage(s) is closest to
the east African clade. Importantly, however, none of the
insertion patterns were in conflict with the relationships
suggested by the other 11 loci in our study, namely, the
monophyly of the east African clade and its close relationship with Steatocranus and the two species of Tilapia.
Genealogies of multiple loci that show mutual
discordance can generally be explained by either interspecific hybridization or incomplete lineage sorting. In
particular, lineage sorting has occurred or is ongoing in
cichlids of the African Great Lakes. Ongoing incomplete
lineage sorting has been reported for fish in Lake Victoria,
based on nuclear sequence analysis (Nagl et al. 1998); for
rock-dwelling cichlids (Mbuna) in Lake Malawi, based on
mitochondrial haplotypes (Moran and Kornfield 1993,
1995; Parker and Kornfield 1997) and a microsatellite
locus (Van Oppen et al. 2000); and for non-Mbuna
cichlids in Lake Malawi, based on analysis of AFC SINE
insertions (Takahashi et al. 2001a). All of these reports of
incomplete lineage sorting were based on observations of
trans-species polymorphisms of multiple alleles or haplotypes. By investigating discordant patterns of AFC SINE
insertion among 14 loci for the Tanganyikan cichlids,
Takahashi et al. (2001b) suggested that there may have
been ‘‘ancient’’ incomplete lineage sorting ;5–10 MYA,
when the major lineages of this lake diverged, and they
proposed that alleles that had been polymorphic during
that period have since become fixed stochastically in each
lineage (see fig. 4 in Takahashi et al. [2001b] for a scheme
showing an example of putative incomplete lineage sorting
of SINE insertion). With respect to the present work, the
divergence of the ancestors of Tilapia, Steatocranus, and
the east African clade may have occurred sufficiently long
ago so as to allow such alleles to become fixed, given that
their speciation events must be older than the age of the
east African clade itself, which includes the major lineages
of the Tanganyikan cichlids analyzed in this study (fig. 2).
Thus, the present incongruencies among the genealogies of
the loci we analyzed are most likely due to ‘‘ancient’’
incomplete lineage sorting, which is more like the phenomenon proposed for cichlids in Lake Tanganyika than the
situation observed for these fish in Lakes Malawi and
Victoria.
Under what circumstances could the speciation of the
ancestral lineages of Tilapia, Steatocranus, and the east
African clade have accompanied incomplete lineage
sorting? Lineage sorting tends to be incomplete when
successive speciation events occur rapidly (Nei 1987;
Pamilo and Nei 1988; Takahata 1989). All previous examples of incomplete lineage sorting in cichlids (Moran
and Kornfield 1993, 1995; Parker and Kornfield 1997;
Nagl et al. 1998; Van Oppen et al. 2000; Takahashi et al.
928 Terai et al.
Table 3
Presence/Absence of an AFC SINE at Each Locus
Suggested Clades
East African þ
Steatocranus þ
Tilapia
East African
Distribution
Species
Lake Malawi
Labidochromis caeruleus
Lake Victoria
Haplochromis nyererei
Lake Tanganyika Boulengerochromis
microlepis
Ctenochromis horei
Cyprichromis leptosoma
Hemibates/Bathybates
Lepidiolamprologus
elongatus
Tanganicodus irsacae
Trematocara
unimaculatum
Pan- and
west Africa
Tilapia rendalli
Tilapia buttikoferi
Steatocranus casuarius
Oreochromis niloticus
Oreochromis tanganicae
Sarotherodon melanotheron
Anomalochromis thomasi
Chromidotilapia finleyi
Hemichromis lifalili
Nanochromis parilus
Pelvicachromis taeniatus
Teleogramma brichardi
Tylochromis polylepis
‘‘Incongruent’’
241 255 450 504 509 1240 1708 234 454 904 1280 906 223 260 304 316 1223 1544
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
þ
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
þ
þ
þ
–
þ
þ
þ
þ
–
þ
þ
–
–
–
–
–
–
–
–
–
–
–
þ
–
–
–
þ
þ
–
–
–
–
–
–
–
–
–
–
–
–
þ
þ
–
–
þ
þ
–
–
–
–
–
–
–
–
–
–
–
–
þ
þ
–
–
–
–
þ
þ
þ
–
–
–
–
–
–
–
–
–
–
þ
þ
þ
–
–
þ
þ
þ
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
þ
–
–
–
–
–
–
–
–
–
–
þ
NOTE.—The presence or absence of an AFC SINE is indicated by (þ) or (–), respectively. Blanks indicate that no PCR product was observed.
2001a, 2001b) occurred in the African Great Lakes which
are known to exhibit explosive rates of speciation. The
speciation events that are the focus of the present study,
however, preceded the divergence of the oldest lineages in
Lake Tanganyika, which corresponds to the basal node of
the east African clade (fig. 2). Thus, it is possible that rapid
speciation occurred in riverine ecosystems before Lake
Tanganyika was formed. An alternate explanation is that
speciation may have taken place in a putative ancient lake
that existed prior to the formation of Lake Tanganyika.
Phylogenetic analysis of central and east African riverine
and lacustrine cichlids using mitochondrial DNA sequences
(Salzburger et al. 2002) suggests that some lineages that
diverged during the primary lacustrine radiation of Tanganyikan cichlids may have been capable of colonizing
surrounding rivers secondarily. If a similar process was involved in the more ancient lacustrine radiation hypothesized in the present study, then the ancestors of Tilapia,
Steatocranus, and the east African clade might have been
the lineages that fled the ancient lake for the riverine
habitat prior to the lake’s disappearance. Inclusion of more
species into future analyses would be helpful to gain
greater insight into how extensively the assumed incomplete lineage sorting or interspecific hybridization event
occurred, and thus to test the above hypotheses. Especially,
analyses of various lineages of tilapiines may be
important, either for this purpose or for reconstruction
of a more detailed phylogeny, given that they may
consist of diverse lineages (Nagl et al. 2001).
FIG. 2.—A phylogenetic tree based on data from the present study.
Numbers in boxes at the nodes and internodes indicate the loci to which
SINE sequences were assumed to have retroposed during the indicated
period. The PCR products were successfully amplified for all the
examined species in table 3 from the loci indicated by boldface numbers,
whereas other loci failed to yield a product from one or more of these
species. The MVhL lineage, whose monophyly was suggested by an
earlier SINE study (Takahashi et al. 2001b), includes cichlids endemic
to Lakes Victoria and Malawi, as well as those in Lake Tanganyika, with
the exclusion of lineages such as Trematocara (Trematocarini tribe),
Boulengerochromis (Tilapiini tribe), Bathybates (Bathybatini tribe), and
Hemibates (Bathybatini tribe). The gray circle at the node indicates the
period when retention of ancestral polymorphisms (presence or absence
of a SINE) was assumed to have occurred at loci 223, 260, 304, 316,
1223, and 1544.
Using SINEs to Probe Ancient Explosive Speciation 929
from multiple loci. Our data suggest that SINEs could be
useful as probes for the analysis of explosive speciation,
including events that were ‘‘ancient.’’
Acknowledgments
The authors thank the Lake Tanganyika Research
Group of Kyoto University for the collection and identification of cichlid specimens. This work was supported
by a grant-in-aid (to N.O.) for Specially Promoted Research and for Overseas Scientific Surveys from the
Ministry of Education, Culture, Sports, Science and Technology of Japan.
Literature Cited
FIG. 3.—Incongruent patterns of SINE insertion observed among
Steatocranus, two species of Tilapia, and the east African clade. The east
African clade is represented by two cichlid species from Lake
Tanganyika. Panels a–c show electrophoretic profiles of PCR products
amplified from loci 260, 316, and 1544, respectively. Closed and open
arrowheads indicate expected mobilities of amplified fragments containing or lacking a SINE insertion, respectively.
Our work demonstrates that the analysis of SINE
insertions is capable of detecting possible ancient incomplete lineage sorting (or interspecific hybridization)
that took place even before the divergence of the major
cichlid lineages of Lake Tanganyika, during an event that
occurred ;14 MYA, as estimated by an earlier allozyme
analysis (Nishida 1997). Previous analyses of nuclear and/
or mitochondrial markers (Sültmann et al. 1995; Zardoya
et al. 1996; Streelman and Karl 1997; Mayer, Tichy, and
Klein 1998; Streelman et al. 1998; Farias, Orti, and Meyer
2000, Farias et al. 2001) did not recognize this phenomenon. Given that ancient incomplete lineage sorting
(or interspecific hybridization) can be detected only by
analyzing genealogic discordance among multiple loci, the
failure of several of these studies to recognize the phenomenon can be explained by the fact that only a single locus
(or two, in the case of Farias et al. [2000]) was analyzed.
Even if multiple loci are analyzed, conventional methods
of sequence analysis tend to produce relatively large
statistical errors when the subjects are ancient lineages that
diverged over a relatively short period (e.g., 30% of the
bootstrap support for the sister group relationship between
Steatocranus and Tilapia; see fig. 2 in Mayer, Tichy, and
Klein 1998). Thus, it may often be difficult to discern such
statistical errors from bona fide incomplete lineage sorting
in analyses of discordance among topologies obtained
Avise, J. C. 2000. Phylogeography: the history and formation of
species. Harvard University Press, Cambridge.
Cohen, A. S., M. G. Soreghan, and C. A. Scholz. 1993. Estimating the age of formation of lakes: an example from Lake
Tanganyika, east African rift system. Geology 21:511–514.
Cook, J. M., and M. Tristem. 1997. ‘SINEs of the times’—
transposable elements as clade markers for their hosts. Trends
Ecol. Evol. 12:295–297.
Coulter, G. W. 1991. The benthic fish community. Pp. 151–199
in G. W. Coulter, ed. Lake Tanganyika and its life. Oxford
University Press, New York.
Farias, I. P., G. Orti, and A. Meyer. 2000. Total evidence:
molecules, morphology, and the phylogenetics of cichlid
fishes. J. Exp. Zool. 288:76–92.
Farias, I. P., G. Orti, I. Sampaio, H. Schneider, and A. Meyer.
2001. The cytochrome b gene as a phylogenetic marker: the
limits of resolution for analyzing relationships among cichlid
fishes. J. Mol. Evol. 53:89–103.
Fryer, G., and T. D. Iles. 1972. The cichlid fishes of the Great
Lakes of Africa: their biology and evolution. Oliver & Boyd,
Edinburgh.
Greenwood, P. H. 1984. African cichlids and evolutionary
theories. Pp. 141–154 in A. A. Echelle and I. Kornfield, eds.
Evolution of fish species flocks. University of Maine at Orono
Press, Orono.
Greenwood, P. H. 1991. Speciation. Pp. 86–102 in M. H. A.
Keenleyside, ed. Cichlid fishes: behaviour, ecology and
evolution. Chapman & Hall, London.
Hamada, M., N. Takasaki, J. D. Reist, A. L. DeCicco, A. Goto, and
N. Okada. 1998. Detection of the ongoing sorting of ancestrally
polymorphic SINEs toward fixation or loss in populations of
two species of char during speciation. Genetics 150:301–311.
Mayer, W., H. Tichy, and J. Klein. 1998. Phylogeny of African
cichlid fishes as revealed by molecular markers. Heredity
80:702–714.
Meyer, A. 1993. Phylogenetic relationships and evolutionary
processes in east African cichlid fishes. Trends Ecol. Evol. 8:
279–284.
Miyamoto, M. M. 1999. Perfect SINEs of evolutionary history?
Curr. Biol. 9:R816–R819.
Moran, P., and I. Kornfield. 1993. Retention of an ancestral
polymorphism in the Mbuna species flock (Teleostei:
Cichlidae) of Lake Malawi. Mol. Biol. Evol. 10:1015–1029.
Moran, P., and I. Kornfield. 1995. Were population bottlenecks
associated with the radiation of the Mbuna species flock
(Teleostei: Cichlidae) of Lake Malawi? Mol. Biol. Evol. 12:
1085–1093.
Nagl, S., H. Tichy, W. E. Mayer, N. Takahata, and J. Klein.
1998. Persistence of neutral polymorphisms in Lake Victoria
cichlid fish. Proc. Natl. Acad. Sci. USA 95:14238–14243.
930 Terai et al.
Nagl, S., H. Tichy, W. E. Mayer, I. E. Samonte, B. J. McAndrew,
and J. Klein. 2001. Classification and phylogenetic relationships of African Tilapiine fishes inferred from mitochondrial DNA sequences. Mol. Phylogenet. Evol. 20:361–374.
Nei, M. 1987. Molecular evolutionary genetics. Columbia
University Press, New York.
Nishida, M. 1997. Phylogenetic relationships and evolution of
Tanganyikan cichlids: a molecular perspective. Pp. 1–24
in H. Kawanabe, M. Hori, and M. Nagoshi, eds. Fish
communities in Lake Tanganyika. Kyoto University Press,
Kyoto, Japan.
Okada, N. 1991. SINEs: short interspersed repeated elements of
the eukaryotic genome. Trends Ecol. Evol. 6:358–361.
Pamilo, P., and M. Nei. 1988. Relationships between gene trees
and species trees. Mol. Biol. Evol. 5:568–583.
Parker, A., and I. Kornfield. 1997. Evolution of the mitochondrial
DNA control region in the Mbuna (Cichlidae) species flock of
Lake Malawi, east Africa. J. Mol. Evol. 45:70–83.
Poll, M. 1986. Classification des Cichlidae du lac Tanganyika:
tribus, genres et espèces. Mém. Acad. Roy. Belg. Cl. Sci.
45:5–163.
Salzburger, W., A. Meyer, S. Baric, E. Verheyen, and C.
Sturmbauer. 2002. Phylogeny of the Lake Tanganyika cichlid
species flock and its relationship to the central and east African
haplochromine cichlid fish faunas. Syst. Biol. 51:113–135.
Shedlock, A. M., M. C. Milinkovitch, and N. Okada. 2000. SINE
evolution, missing data, and the origin of whales. Syst. Biol.
49:808–817.
Shedlock, A. M., and N. Okada. 2000. SINE insertions: powerful
tools for molecular systematics. BioEssays 22:148–160.
Streelman, J. T., and S. A. Karl. 1997. Reconstructing labroid
evolution with single-copy nuclear DNA. Proc. R. Soc. Lond.
Ser. B 264:1011–1020.
Streelman, J. T., R. Zardoya, A. Meyer, and S. A. Karl. 1998.
Multilocus phylogeny of cichlid fishes (Pisces: Perciformes):
evolutionary comparison of microsatellite and single-copy
nuclear loci. Mol. Biol. Evol. 15:798–808.
Sültmann, H., W. E. Mayer, F. Figueroa, H. Tichy, and J. Klein.
1995. Phylogenetic analysis of cichlid fishes using nuclear
DNA markers. Mol. Biol. Evol. 12:1033–1047.
Takahashi, K., Y. Terai, M. Nishida, and N. Okada. 1998.
A novel family of short interspersed repetitive elements
(SINEs) from cichlids: the patterns of insertion of SINEs at
orthologous loci support the proposed monophyly of four
major groups of cichlid fishes in Lake Tanganyika. Mol. Biol.
Evol. 15:391–407.
Takahashi, K., M. Nishida, M. Yuma, and N. Okada. 2001a.
Retroposition of the AFC family of SINEs (short interspersed
repetitive elements) before and during the adaptive radiation
of cichlid fishes in Lake Malawi and related inferences about
phylogeny. J. Mol. Evol. 53:496–507.
Takahashi, K., Y. Terai, M. Nishida, and N. Okada. 2001b.
Phylogenetic relationships and ancient incomplete lineage
sorting among cichlid fishes in Lake Tanganyika as revealed
by analysis of the insertion of retroposons. Mol. Biol. Evol.
18:2057–2066.
Takahata, N. 1989. Gene genealogy in three related populations:
consistency probability between gene and population trees.
Genetics 122:957–966.
Terai, Y., K. Takahashi, and N. Okada. 1998. SINE cousins: The
39-end tails of the two oldest and distantly related families of
SINEs are descended from the 39 ends of LINEs with the same
genealogical origin. Mol. Biol. Evol. 15:1460–1471.
Van Oppen, M. J. H., C. Rico, G. F. Turner, and G. M. Hewitt.
2000. Extensive homoplasy, nonstepwise mutations, and
shared ancestral polymorphism at a complex microsatellite locus in Lake Malawi cichlids. Mol. Biol. Evol.
17:489–498.
Weiner, A. M., P. L. Deininger, and A. Efstratiadis. 1986.
Nonviral retroposons: genes, pseudogenes, and transposable
elements generated by the reverse flow of genetic information.
Annu. Rev. Biochem. 55:631–661.
Zardoya, R., D. M. Vollmer, C. Craddock, J. T. Streelman,
S. Karl, and A. Meyer. 1996. Evolutionary conservation
of microsatellite flanking regions and their use in resolving
the phylogeny of cichlid fishes (Pisces: Perciformes). Proc.
R. Soc. Lond. Ser. B 263:1589–1598.
Pierre Capy, Associate Editor
Accepted February 9, 2003