Monophyletic origin of multiple clonal lineages in an asexual fish

Transcription

Molecular Ecology (2010)
doi: 10.1111/j.1365-294X.2010.04869.x
Monophyletic origin of multiple clonal lineages in an
asexual fish (Poecilia formosa)
M A T T H I A S S T Ö C K , * 1 K A T H R I N P . L A M P E R T , †1 D I R K M Ö L L E R , ‡ I N G O S C H L U P P ‡§ and
MANFRED SCHARTL–
*Department of Ecology and Evolution, University of Lausanne, Biophore, CH-1015 Lausanne, Switzerland, †Evolutionary
Ecology and Biodiversity of Animals, University of Bochum, Universitaetsstr. 150, Bld. ND05 Room 785, D-44780 Bochum,
Germany, ‡Department of Zoology, University of Hamburg, Martin-Luther-King Pl. 3, D-20146 Hamburg, Germany,
§Department of Zoology, University of Oklahoma, 730 Van Vleet Oval, Norman, OK 73019, USA, –Theodor-Boveri-Institut,
Physiologische Chemie I, University of Würzburg, Biozentrum, Am Hubland, D-97074 Würzburg, Germany
Abstract
Despite the advantage of avoiding the costs of sexual reproduction, asexual vertebrates
are very rare and often considered evolutionarily disadvantaged when compared to
sexual species. Asexual species, however, may have advantages when colonizing (new)
habitats or competing with sexual counterparts. They are also evolutionary older than
expected, leaving the question whether asexual vertebrates are not only rare because of
their ‘inferior’ mode of reproduction but also because of other reasons. A paradigmatic
model system is the unisexual Amazon molly, Poecilia formosa, that arose by
hybridization of the Atlantic molly, Poecilia mexicana, as the maternal ancestor, and
the sailfin molly, Poecilia latipinna, as the paternal ancestor. Our extensive crossing
experiments failed to resynthesize asexually reproducing (gynogenetic) hybrids confirming results of previous studies. However, by producing diploid eggs, female F1hybrids showed apparent preadaptation to gynogenesis. In a range-wide analysis of
mitochondrial sequences, we examined the origin of P. formosa. Our analyses point to
very few or even a single origin(s) of its lineage, which is estimated to be approximately
120 000 years old. A monophyletic origin was supported from nuclear microsatellite data.
Furthermore, a considerable degree of genetic variation, apparent by high levels of clonal
microsatellite diversity, was found. Our molecular phylogenetic evidence and the failure
to resynthesize the gynogenetic P. formosa together with the old age of the species
indicate that some unisexual vertebrates might be rare not because they suffer the longterm consequences of clonal reproduction but because they are only very rarely formed as
a result of complex genetic preconditions necessary to produce viable and fertile clonal
genomes and phenotypes (‘rare formation hypothesis’).
Keywords: clonal reproduction, genotypic variability, hybrid origin, microsatellites, mtDNA,
teleosts, unisexual
Received 30 November 2009; revision received 25 May 2010; accepted 11 June 2010
Introduction
The existence of asexual species is challenging to the
view that sexual reproduction is the most evolutionary
advantageous mode for organisms to propagate, to
Correspondence: Matthias Stöck, Fax: +41 21 692 4265;
E-mail: Matthias.stoeck@unil.ch
1
Equal contribution.
! 2010 Blackwell Publishing Ltd
persist as a species and even to diversify into distinct
species. The short-term advantages of asexual reproduction, most notably the shorter population doubling times
(‘twofold cost of males’; Maynard Smith 1978), are generally thought to be outweighed by a long-term fitness
decrease because of the absence of genetic segregation
and recombination. Thus, nonrecombining (‘asexual’)
lineages should have low genotypic variability,
2 M . S T Ö C K E T A L .
accumulate deleterious mutations and are therefore considered to be short-lived in evolutionary terms, also
explaining their scarcity (Muller 1932; Bell 1982; Otto &
Lenormand 2002). However, asexual organisms may
have advantages when colonizing new habitats or competing with sexual counterparts once they arise (Avise
2008) and therefore would be expected to evolve more
frequently than they do, suggesting that additional factors are relevant for explaining their scarcity. In vertebrates, which are supposed to be ancestrally sexually
reproducing, the rarity of asexuals may instead be
because of constraints on the reversion to asexuality
(only 0.1% of all vertebrate species, Avise 2008).
So-called asexual fishes and amphibians display
reproductive modes that deviate from true parthenogenesis (Avise 2008) and thus avoid, at least to some
degree, the disadvantages from not having meiotic
recombination by incorporating ‘fresh’ genetic material
either every generation (hybridogenesis) or in unpredictable intervals through paternal leakage (‘leaky
gynogenesis’; Avise 2008). In hybridogenesis, the complete paternal genome is eliminated. Fertilization with a
new, recombined genome replaces the eliminated one
and restores the hybrid state. After the elimination of
one complete chromosome set, some triploid lineages of
fishes and amphibians produce clonal diploid eggs,
while others even enter a normal meiosis and produce
recombined gametes (‘meiotic hybridogenesis’; Alves
et al. 2001; for a recent review: Lamatsch & Stöck 2009).
In ‘kleptogenesis’, females acquire full or partial genomes by a not fully understood mechanism from their
mates. Female kleptogens thus secure highly adapted
genomes and purge genomes with deleterious alleles
(Bogart et al. 2007). Taking into account their ecological
success, the question arises why only approx. 50 of the
more than 31 000 described species of teleost fishes and
6500 amphibians are unisexual (Vrijenhoek et al. 1989).
The clonal fish Poecilia formosa (Amazon molly), the
first unisexual vertebrate to be recognized (Hubbs &
Hubbs 1932), reproduces by gynogenesis. Here, unreduced, ameiotic eggs are triggered for parthenogenetic
development by sperm of males from closely related
species (Hubbs & Hubbs 1932; Monaco et al. 1984;
Dawley 1989; Vrijenhoek 1994). Typically, the sperm
DNA is degraded, and the offspring are clones of their
mother. Sometimes, however, the exclusion of the paternal chromosomes fails, and genomic fragments (microchromosomes) or the full sperm genome are included
permanently into the clonal germ line (Balsano et al.
1972; Rasch & Balsano 1974; Turner 1982; Schartl et al.
1995a; Lamatsch et al. 2004; Nanda et al. 2007). Thus,
‘fresh’ subgenomic amounts of DNA or even a whole
chromosome set can be incorporated into the ameiotic
lineage, giving rise to new clones.
Like all other unisexual vertebrates, P. formosa arose
by hybridization. In general, hybridization combines
two different genotypes, which may be advantageous
because it introduces genetic variation and functional
novelty, but may also cause problems, especially during meiosis, because the pairing of divergent homologues might be difficult to accomplish. Even though
gynogens reproduce via ameiotic eggs, the combination of independently evolved genomes forcing genes
to function outside their normal genetic background
(Dobzhansky-Muller effect) might still be problematic
for the organism (Orr & Turelli 2001). In the case of
P. formosa, the hybrid origin has been recognized very
early based on morphology (Hubbs & Hubbs 1932).
Molecular analyses confirmed the hybrid state of the
species and found that the maternal ancestor is the
Atlantic molly, Poecilia mexicana, and the paternal
ancestor the sailfin molly, Poecilia latipinna (Avise et al.
1991; Schartl et al. 1995b). Previous estimates put
the initial hybridization event at approximately
280 000 years ago (Lampert & Schartl 2008). Estimating three generations per year (Hubbs & Hubbs 1932),
this would mean that Amazon mollies have persisted
as a clonal species for approximately 840 000 generations.
So far, the question whether one, a few or multiple
hybridization events led to the formation of the Amazon molly could not be answered satisfactorily. Both
parental species are sympatric in the Atlantic coastal
drainages of northern Mexico (Darnell & Abramoff
1968; Schlupp et al. 2002). Consequently, there has
been and still might be opportunity for de novo
hybridizations. Earlier genetic analyses indicated the
existence of multiple clones in P. formosa that might
have been the result of several independent
hybridization events (Turner et al. 1983; Lampert et al.
2005; Schories et al. 2007). On the other hand, previous attempts to resynthesize P. formosa in the laboratory have been unsuccessful (Turner et al. 1980;
Turner 1982), indicating complications for hybridization.
In this study, we addressed the question if P. formosa
is of mono- or polyphyletic origin to better understand
how clonal vertebrate species evolve from their sexual
ancestors and to test the hypotheses that evolutionary
constraints may limit the formation of unisexuals (Vrijenhoek et al. 1989). We found support for a monophyletic origin of P. formosa, a rather high level of clonal
diversity, and an old age of the species. Our data indicate that unisexual vertebrates might be rare not
because they suffer the long-term consequences of clonal reproduction but that they are only very rarely
formed because of complex genetic preconditions necessary to produce viable and fertile clonal genomes.
SINGLE ORIGIN OF MULTIPLE POECILIA CLONES 3
Material and methods
Experimental animals from field collections and
laboratory crosses
In an attempt to resynthesize P. formosa in the laboratory, we performed interspecific crosses (19 crosses of
P. mexicana females with P. latipinna males and three
reciprocal crosses) employing three different populations of each parental species. P. mexicana were from
Rio Verde (San Louis Potosi; allopatric), Laguna
Champaxan (loc. 6, sympatric), Rio Oxolotan (Tabasco,
allopatric), and P. latipinna from Laguna Champaxan
(loc. 6, sympatric), Tampico (loc. 7, allopatric) and Florida (loc. 36, allopatric). A total of 41 females were
mated either individually or in groups of 2–3 to single
males. Populations were chosen to represent the current
co-occurrence of P. mexicana and P. latipinna as well as
geographically close and widely separated genotypes
(for information on species’ ranges see: Darnell &
Abramoff 1968; Schlupp et al. 2002). F1 offspring were
raised, and F1 females were mated to Black Mollies
(ornamental breed of Poecilia cf. sphenops), which are
homozygous at two dominant black pigmentation pattern loci [(Schröder 1964) and own unpublished data],
to test for clonal reproduction. Absence of the dominant
pigmentation phenotype (a clear indicator of paternal
contribution to the offspring’s genotype) was used to
check for gynogenetically produced offspring (Schartl
et al. 1995a; Lampert et al. 2007). Ploidy levels were
determined using flow cytometry (Lamatsch et al.,
2000). Phylogenetic and population genetic analyses
were performed by analysing individuals from 25 populations all over the natural range of all species (Fig. 1,
Table S1, Supporting Information). A total of 143 field
collected individuals of the species P. latipinna, P. mexicana mexicana, P. mexicana limantouri and P. formosa
were analysed for mitochondrial haplotype, and 108 for
nuclear microsatellites.
MtDNA amplification
Genomic DNA was extracted with standard methods
from dorsal fin clips or pooled organs. Mitochondrial
DNA (mtDNA) was amplified in two fragments. For
the mtDNA control region (D-loop), we used primers
L-Pro (5¢-AAC CTC CAC CCC TAA CTC CCA AAG)
or alternatively a partly overlapping version L-Pro2 (5¢GAT TCT AAC CTC CAC CCC TAACTC CCA-3¢) combined with H-Phe1 (5¢-GGT ACA ATT GAT AGT AAA
GTCAGG ACC A-3¢) in a PCR with 95 "C 5 min, 38·
(95 "C 30 s, 65 "C 30 s, 72 "C 1.5 min), 72 "C 5 min. For
a fragment of cytochrome b, we used L15513 (5¢-CTR
GGA GAC CCN GAA AAC TT-3¢) with H16498
(5¢-CCT GAA GTA GGA ACC AGA TG-3¢) using the
same PCR protocol, but with an annealing temperature
of 53 "C. PCR products were sequenced in both directions, edited using SEQUENCHER version 4.7 and aligned
with SE-AL version 2.0a11 (http://tree.bio.ed.ac.uk/soft
ware/seal/). All sequences were submitted to GenBank,
accession numbers HM567171–HM567311. For most
analyses, a minisatellite ATTTATA (mitochondrial Dloop) that was present 1–2 times in P. mexicana and 2–3
times in P. formosa, was replaced by G, GG or GGG,
assuming that the entire repeat appears as insertion or
deletion, and therefore its mutational behaviour is like
that of a single base pair.
Phylogenetic analyses of mtDNA data
Phylogenetic trees were calculated using four different
methods. With PAUP* (Swofford 2001), maximum parsimony (MP) and neighbour joining (NJ) methods were
applied. The best fitting model of sequence evolution
was selected using MRMODELTEST (Nylander 2004). We
used the program MRBAYES (MB; version 3.0b4; Huelsenbeck & Ronquist 2001), running four chains for 5 or
10 million generations, with tree sampling every 1000
generations. The ‘burnin’-value was selected by visualizing the log likelihoods associated with the posterior
distribution of trees in the program TRACER (version 1.4;
Rambaud & Drummond, available from: http://beast.
bio.ed.ac.uk/). All trees generated before the log likelihood curve flattened out were discarded.
Maximum likelihood (ML) phylogenies of mitochondrial sequence alignments were generated using PHYML
version 2.4.5 (Guindon & Gascuel 2003) using the HKY
(mtDNA) model. In each case, we choose a BioNJ as a
starting tree, and the options to optimize the topology,
branch length and rate parameters. All other parameters
were used as in the default of the program (http://
atgc.lirmm.fr/phyml/ for details). We generated bootstrap values based on 1000 resampled data sets.
We applied two different methods to calculate parsimony networks. We used TCS version 1.21 (Clement
et al. 2000) to connect mtDNA haplotypes. In this
approach, we used the gaps as ‘5th base’ with a 95%
connection limit (Fig. 1c). In addition, we used the program SPLITSTREE version 4.10 (Hudson & Bryant 2006) to
apply parsimony splits (Bandelt & Dress 1992).
Demographic analyses of mtDNA data and divergence
time estimates
We applied the program FLUCTUATE (Kuhner et al. 1998),
a maximum likelihood estimator of the parameters
Theta (h) and g (h = 2Nel, where l is the DNA substitution rate per site per generation and Ne is the female
4 M . S T Ö C K E T A L .
(a)
(b)
0
(b)
P.formosa
250
500
1000 km
(c)
P.mexicana
0
75
150
200
300 km
Fig. 1 Sampling localities (for details see Appendix S1 and Table S1, Supporting Information) in Mexico and the United States and
mitochondrial relationships in the Poecilia formosa breeding complex. (a) Overview of the sampling localities with framed region as
zoomed in b. (b) Sampling locations in Eastern Mexico with four major river systems. (c) Parsimony-based haplotype network of the
mitochondrial sequence data from the control region of 85 P. formosa and 45 P. mexicana as obtained with program TCS; numbers
represent localities shown in a and b; circle size in c depends on individual numbers with the respective haplotype; dots represent
hypothetical haplotypes that were not empirically sampled. The shortest connection between networks of P. formosa and P. mexicana
consists of a haplotype of P. mexicana from the Laguna Champaxan (loc. 6) to the most widespread mtDNA haplotype of P. formosa
as found at localities 4, 8, 9, 13–16, 18, 21 and 24. Three more remote connections (mexicana at loc. 10 with formosa at loc. 12, 22–24;
mexicana at loc. 6 and 15 with formosa at loc. 4, 8, 9, 13–16, 18, 21, 24) exist but each mexicana haplotype is two (or more) mutational
steps apart from the next formosa haplotype, and all in-between haplotypes have never been found empirically. Note that localities
with closely related P. formosa are situated in several different major river drainages, while those of P. mexicana are strongly geographically structured: P. mexicana from Rio Panuco and its tributaries from the Sierra Madre belong to one phylogenetic group (see
also separate subclades in Figs S1 and S2), while the other P. mexicana group is restricted to the Rio Soto la Marina. Widespread
P. formosa lineages suggest that they, in contrast to P. mexicana, were able to overcome zoogeographical borders. e.g. by ‘river mouth
hopping’.
effective population size and g the exponential growth
parameter in units of l)1). Repeated analyses to ensure
stability of estimates were run. Growth was inferred
using logarithmic likelihood ratio tests with one degree
of freedom (Huelsenbeck & Rannala 1997). Divergence
times of the P. formosa mitochondrial lineage were estimated using a Bayesian-coalescence approach (Drummond et al. 2006; Drummond & Rambaut 2007) as
implemented in BEAST version 1.4.8. In analysis of
884 bp from the control region, we used a matrix from
85 P. formosa and 45 P. mexicana individuals. We constrained P. formosa to be monophyletic and assumed a
strict molecular clock, with substitution rates from the
range for Actinopterygii (by exclusion of Acipenseriformes) reported by Burridge et al. (Burridge et al.
2008a,b) for the control region of various teleost fishes.
We ran two independent analyses for 10 · 106 generations and checked for convergence and stationarity of
the different analyses in Tracer 1.4 and combined the
results in the BEAST module LOGCOMBINER version 1.4.8
(after removing the first 106 generations from each analysis as ‘burnin’).
Microsatellite amplification
A total of 40 P. formosa, 39 P. mexicana and 28 P. latipinna from fifteen different locations [Tamasopo and
Cascadas de Tamasopo (loc. 3), Ojo Frio (loc. 5), Laguna
Chamapaxan (loc. 6), Mante (loc. 10), Rio Guayalejo
(loc. 14), Rio Corona (loc. 15), Rio Purificacion (loc. 17),
Rio Oxolotan (loc. 46), Texas (loc. 25–27), Florida (loc.
26, 34–40), Louisiana (loc. 28–31, Mississippi (loc. 32),
Georgia (loc. 41–43), South Carolina (loc. 44) and North
Carolina (loc. 45)] were analysed. We used ten microsatellite loci (Sat1, KonD15, PR39, mCA16, mCA20,
mATG31, mATG38, mATG44, mATG61, mATG78)
which had been shown to be variable in P. formosa
(Lampert et al. 2005, 2006; Schories et al. 2007). PCRs
were performed in a total volume of 10 lL, containing
10 mM Tris–HCl (pH 8.85), 50 mM KCl, 0.1% Triton X100, 1.5 mM MgCl2, 0.2 mM of each dNTP, 10 pmol of
each primer and 0.05 U Taq polymerase. PCR conditions were as follows: 5 min of denaturing at 94 "C, 30
cycles of (94 "C denaturing, 55 "C annealing [all primers
except Sat1: 58 "C and KonD15: 52 "C) and 72 "C for
extension (each step for 30 s)], followed by a final
extension of 5 min at 72 "C. PCR products were analysed on a Licor 4300 DNA Analyzer (Licor Biosciences,
NE, USA).
Microsatellite analyses
Fragment size analysis was carried out with the SAGA2
software (Licor Biosciences). Only individuals that were
successfully scored at a minimum of 8 of 10 microsatellite loci were used for statistical comparisons. Allelic
richness, calibrated for a single individual to compensate for varying sample sizes in different field sites, was
calculated using the program FSTAT (Goudet 2001).
Genotypic diversity was calculated using the PDC (Proportion of distinguishable clones) measurement suggested by Menken et al. (1995). This is a relative
measurement of clonal diversity and compensates for
different sample sizes in different populations. For phylogenetic analysis, Cavalli-Sforza and Edwards chord
genetic distance between individual genotypes (as recommended by Takezaki & Nei 1996) was calculated
using the program MSA version 4.05 including 1000
bootstrap replicates (Dieringer & Schlötterer 2003). The
resulting genetic distances were further analysed using
the PHYLIP version 3.6 program package (Felsenstein
2004): program NEIGHBOR to generate 1000 neighbour
joining trees, program CONSENSUS to calculate the majority rule consensus tree from the 1000 neighbour joining
trees. The programs FIGTREE (version 13.1 Rambaut, A.,
http://tree.bio.ed.ac.uk/software/figtree/) and TREEGRAPH (Stöver & Müller, 2010) were used for the graphic display of the tree.
Results
Interspecific hybrids obtained in the laboratory
In an attempt to resynthesize P. formosa, fish from the
parental species of various geographical origin were
mated. This resulted in 636 F1 offspring, of which only
17% were males. Fecundity was apparently not considerably different from intraspecific matings (7.95 vs. 9.35
offspring ⁄ female). The 206 female F1 were used for testcrosses with Black molly males. A total number of 3118
offspring were obtained from these crosses. All fish
showed a black spotting phenotype. The inheritance of
the dominant paternal pigmentation genes indicated
that none of the offspring was produced asexually.
Remarkably, 66.1% of the progeny were triploids. It has
been reported earlier (Lampert et al. 2007) that in such
crosses, triploid offspring are produced from the fertilization of diploid eggs, which are generated by automixis. All tested triploids (n = 43; 2 males, 41 females)
were infertile.
Genotypic diversity in wild populations
We found that microsatellite loci display astonishingly
high allelic and genotypic richness in all populations
investigated (Table 1). Although most microsatellite
alleles in P. formosa represent a subset of their homologues in P. mexicana and P. latipinna, we also found
6 M . S T Ö C K E T A L .
high genotypic diversity in P. formosa populations (proportion of distinguishable clones: 0.77–0.95; Table 2).
Phylogenetic analyses
Phylogenetic analyses of 60 mtDNAs comprising all
four species ⁄ subspecies confirmed the close affinity of
mtDNA between P. mexicana and P. formosa (Avise
et al. 1991) and excluded P. latipinna as maternal ancestors (data not shown). Additional phylogenetic analyses
(NJ, MP, ML, MB; for the latter two analyses: see
Figs S1 and S2, Supporting Information) were conducted in P. mexicana (n = 45) and P. formosa (n = 85)
from representative populations. These analyses
resulted in large polytomic clades containing both
P. formosa and a single subclade of P. mexicana haplotypes. While the resolution of these dichotomic phylogenetic methods was insufficient to tell a multiple
origin scenario apart from a rare-origin setting, it never
contradicted a rare or even a single-origin scenario. To
better resolve this phylogeny, we analysed the mtDNA
sequences in parsimony networks that take multiple
connections between individual haplotypes into account
(SPLITSTREE, data not shown; TCS: Fig. 1c). Among the
empirically found mtDNA haplotypes, the most parsimonious connection between sequences of the bisexual
Table 1 Mean allelic richness for all microsatellite loci in different species and localities. Numbers in parenthesis refer to
collection sites (see Table S1, Supporting Information)
G (14)
P (17)
LC (6)
Cascadas
Rio Corona
Florida (25)
Poecilia
formosa
Poecilia
mexicana
Poecilia
latipinna
1.6780
1.6246
1.6447
1.2334
1.3013
1.5166
1.2178
1.3000
1.6734
1.5001
Table 2 Results of the Menken analysis of proportion distinguishable clones (PDC) for all populations and sampling localities with a sample size larger than five individuals
LCPf LCPm PPf
Number of individuals 20
Number of genotypes 19
PDC
0.95
13
13
1
All All
PPm Pm Pf
15
19
11
19
0.73 1
29
29
1
40
35
0.875
Localities: LC, Laguna Champaxan (6); P, Rio Purificacion (17);
G, Rio Guayalejo (14); species: Pf, Poecilia formosa; Pm, Poecilia
mexicana.
ancestor P. mexicana and hybrid all-female P. formosa
consists in a single mutational step representing the
transition from a haplotype of P. mexicana from Laguna
Champaxan (Fig. 1b: loc. 6) to the geographically most
widespread mtDNA haplotype of P. formosa found at
ten different localities throughout its range. There are
three additional but more remote connections (Fig. 1c)
between the networks of P. formosa and P. mexicana.
However, each of those mexicana haplotypes is two or
more mutational steps apart from the next formosa haplotype. In addition, all in-between haplotypes were only
theoretically generated by the analyses but have never
been found empirically. In addition, these ‘theoretical’
mexicana haplotypes would lead to formosa haplotypes—via several steps—that are themselves more parsimoniously explained by single-step mutations from
empirically found formosa haplotypes.
Thus, a probable (and empirically best documented)
derivation of the P. formosa lineage could be a singlestep mutation that led from a haplotype at Laguna
Champaxan to the most widespread haplotype of
P. formosa. Coalescence theory considers the most frequent haplotype as ancestral (Watterson & Guess 1977;
Clement et al. 2000). This and the fact that the progenitor haplotype of P. mexicana is also found here suggest
that P. formosa arose in this region.
Nuclear microsatellite data support the mitochondrial
data. A phylogenetic tree built on individual genetic
distances (Fig. 2) shows a highly supported (bootstrap
87%) monophyletic origin of all P. formosa genotypes,
clearly rejecting multiple recent or ongoing hybridizations to generate new clonal P. formosa lineages.
Although their hybrid nature might be contributing, the
monophyly of the P. formosa clade stems also from
intensive allele sharing within P. formosa (Table S2,
Supporting Information). Shared alleles could be traced
back to the parental species but were not necessarily
common in sexual populations (Fig. S3, Supporting
Information), again supporting a few rather than multiple origins, as the latter would tend to generate similar
allele frequencies in parental and hybrid lineages. In
addition, P. formosa from different sites show lower
genetic distances within and between populations than
P. mexicana (Table 3).
Population dynamics and species’ age
Despite intense sampling, our analyses did not disclose
a single shared mtDNA haplotype between P. mexicana
and P. formosa. This fact by itself again supports rare or
single ancient, as opposed to multiple recent or ongoing, hybrid origin(s). Using mtDNA sequences, the program FLUCTUATE disclosed a population increase for
both species. For Theta (2Nel) estimated to have the
same order of magnitude in both species (similar: 0.0138
for asexual P. formosa, 0.0149 for sexual P. mexicana),
the significant growth value is an order of magnitude
larger in the asexual (g = 3273) than in the sexual species (g = 554). When testing the same assumed growth
value (g = 1100) in both species (Table 4), we found a
significant difference from the ‘no-growth assumption’
in P. formosa but not in P. mexicana, confirming that
growth in P. formosa is greater than in P. mexicana. The
higher growth for clonal P. formosa (Table 4) suggests
population expansion, consistent with its derivation
from few ancestors and colonization of a large range by
this lineage. Age estimates of the P. formosa haplotype
group, using BEAST and rates for calibrated mitochondrial molecular clocks from various actinopterygian
fishes (Burridge et al. 2008a,b), ranged from 65 ka (highest rate: 0.0385 mutations ⁄ site ⁄ My, from the fast evolving East African cichlid) to 580 ka (lowest rate: 0.00425)
with an average of 118 ka for the mean D-loop rate of
0.021. Other estimates from mitochondrial or nuclear
DNA dated the origin of P. formosa at approximately
280 ka ago (Schartl et al. 1995b; Lampert & Schartl
2008).
Discussion
Our failure to resynthesize P. formosa by crossing its
parental species, despite the large amount of crosses
and individuals involved, fits several earlier unsuccessful attempts to resynthesize P. formosa (Turner et al.
1980; Turner 1982). However, all except two of our F1hybrid females produced diploid eggs (Lampert et al.
2007), plus additional data from this study, which
would be a prerequisite for the evolution of gynogenesis.
High clonal diversity as found by the microsatellite
markers in this study could indeed be taken as indication of numerous and potentially recent hybrid origins,
however, allele sharing between clonal lineages was still
very high (Table S2, Supporting Information) and phylogenetic analyses found P. formosa to be a monophyletic group. In some other taxa, it is well known that
multiple hybridization events led to multiple clones
(Avise 2008; Lamatsch & Stöck 2009). The existence of
multiple nuclear P. formosa clones, detected by microsatellite and multilocus fingerprint analyses, is in accordance with simulations (Balloux et al. 2003) that
suggest a higher allelic diversity but lower genotypic
diversity in clonally reproducing compared to sexual
populations. Both effects result from the lack of recombination in asexuals. The high number of genotypes
detected in gynogenetic P. formosa, however, is still
puzzling but was also found in earlier multilocus fingerprint analyses and possibly results from a high
mutation rate (Turner et al. 1990). A relatively high
genetic diversity was also confirmed in other population genetic studies of P. formosa including neutral microsatellite and multilocus DNA fingerprint data
(Lampert et al. 2005) but also functional MHC (major
histocompatibility complex) genes (Schaschl et al. 2008;
Lampert et al. 2009). Clonal diversity in P. formosa,
detected by DNA fingerprinting, was explained as a
result of mutations and subsequent selection and is documented by hypervariable loci (Schartl et al., 1991; see
also: Turner et al. 1990). Obviously, P. formosa does not
fulfil the negative prediction of low genetic diversity in
clonal species leading to the conclusion that its fitness
might also not be reduced compared to its sexual ancestors. Considering the relatively old age of the species
estimated from other markers, mutations seem a likely
source of genotypic variability, especially at loci with
generally high mutation rates such as microsatellites
and MHC. This notion is supported by other studies
(e.g. Angers & Schlosser 2007), which found that in the
phylogenetically young Phoxinus eos-neogaeus complex
genotypic diversity was relatively low. Earlier studies
using biochemical markers also support a single origin
of P. formosa. Most of the allozyme electromorphs
within P. formosa have also been found in P. latipinna
and P. mexicana (Turner 1982). The maximum genetic
distance within P. formosa was much smaller than the
average within gonochoristic teleosts and led to the
conclusion that P. formosa ‘are ultimately descended
from hybridization events involving the same or very
closely related individuals’ (Turner & Steeves 1989;
Turner et al. 1990; Avise et al. 1991).
A quite ancient, monophyletic origin together with
the inability to resynthesize P. formosa support that unisexual vertebrates are rare not necessarily because they
suffer serious disadvantages, but because the conditions, under which they can arise, are rare. The switch
from sexual meiotic reproduction to ameiotic unisexuality is apparently extremely difficult to achieve.
Although hybrids between P. formosa’s ancestors show
a prerequisite for unisexual reproduction, namely the
production of diploid eggs, they suffer from very low
reproductive rates as the diploid eggs that are fertilized
during mating result in mostly sterile triploid offspring
(Lampert et al. 2007). In some other ‘unisexual’ species,
more than one origin has been inferred (Avise 2008;
Lamatsch & Stöck 2009). However, several attempts to
resynthesize hybridogenetic Poeciliopsis fish in the laboratory have, despite occasional success, exhibited ‘considerable variance of one parental genome (monacha) in
the production of viable, fertile and developmentally
stable interspecific hybrids’ (Wetherington et al. 1987).
Clearly, successful hybridization of the known parental
species is rare and exceptional even under the enforced
8 M . S T Ö C K E T A L .
Table 3 Genetic distance (Cavalli-Sforza and Edwards chord distance) within (diagonal) and between populations. On average, Poecilia formosa from different field sites are more closely related to each other (mean ± SD: 0.504 ± 0.074, min: 0.13, max: 0.64) than Poecilia mexicana from different field sites (mean ± SD: 0.797 ± 0.067, min: 0.45, max: 0.87)
PfG
PfG
Mean ± SD
Min–max
PmG
Mean ± SD
Min–max
PfP
Mean ± SD
Min–max
PmP
Mean ± SD
Min–max
PfLC
Mean ± SD
Min–max
PmLC
Mean ± SD
Min–max
PmG
PfP
PmP
PfLC
PmLC
0.393 ± 0.090
0.125–0.472
0.806 ± 0.017
0.763–0.818
0.624 ± 0.266
0.317–0.791
0.431 ± 0.095
0.127–0.595
0.709 ± 0.063
0.642–0.818
0.310 ± 0.244
0.000–0.809
0.781 ± 0.056
0.641–0.867
0.796 ± 0.034
0.664–0.854
0.857 ± 0.016
0.806–0.874
0.371 ± 0.223
0.049–0.874
0.516 ± 0.085
0.255–0.641
0.783 ± 0.031
0.694–0.833
0.520 ± 0.048
0.371–0.641
0.823 ± 0.057
0.625–0.900
0.425 ± 0.182
0.064–0.841
0.750 ± 0.076
0.591–0.871
0.690 ± 0.117
0.454–0.854
0.724 ± 0.097
0.476–0.871
0.813 ± 0.045
0.633–0.874
0.793 ± 0.053
0.531–0.900
0.615 ± 0.219
0.061–0.900
LC, Laguna Champaxan (6); P, Rio Purificacion (17); G, Rio Guayalejo (14); Pf, Poecilia formosa; Pm, Poecilia mexicana.
Table 4 Estimates of historical demographic parameters of the mitochondrial lineages of Poecilia formosa and its maternal ancestor
Poecilia mexicana from the program FLUCTUATE
Mitochondrial
control region
clade ⁄ taxon
N
Theta =
2Nel
P. mexicana
P. formosa
45
87
0.0149
0.0138
g
Ln
(likelihood)
for Lmax
Ln
(likelihood)
for zero
growth
554.70
3273.63
0.0677
1.2876
)4.0377
)21.6622
2 (Lmax)
Lg = 0)
No growth
can be
rejected
Ln
(likelihood)
for growth
g = 1100
2 (Lmax)
Lg = 1100)
Growth
value
significantly
different
8.21
45.89
Yes
Yes
)0.3057
)4.9919
0.3734
6.2719
Yes
*Theta = 2Nel, g, exponential growth parameter; L, maximum likelihood.
crossing conditions. Similarly, in the hybridogenetic
Rana esculenta frog complex, R. ridibunda genomes varied geographically in inducing clonal gametogenesis in
interspecies hybrids (Hotz et al. 1985). Hybrid minnows
of the Phoxinus eos-neogaeus complex were found to
be not monophyletic, suggesting multiple independent
origins of lineages (Angers & Schlosser 2007). As they
are of relatively recent origin (<50 000 years ago,
although not the result of current hybridization events),
this may represent a different situation than in the
rather old lineages of P. formosa.
It is reasonable to assume that beyond the production
of unreduced eggs more genomic changes have to occur
to bring about the full process of gynogenesis (Vrijenhoek 1994; Schlupp 2005). If we consider that several of
such changes need to be present in a single hybrid animal to become the ancestor of an asexual lineage, the
probability of such an event to occur becomes extremely
small. The relatively old age of P. formosa and its obvious success, evident from the enormous historic population growth and its present range, may be partly
derived from the peculiar mode of gynogenetic repro-
Fig. 2 Neighbour joining tree of individual genetic distances based on 10 nuclear microsatellite genotypes in Poecilia formosa and its
parental species. Consensus tree from 1000 replicates. bootstrap support is given for branches if larger than 50%; branches with a
bootstrap support of less than 50% were collapsed. P. formosa (Pform)—green, Poecilia mexicana (Pmex)—blue, Poecilia latipinna
(Plat)—black.
10 M . S T Ö C K E T A L .
duction that allows for introgression of fresh genetic
material (paternal leakage) from sexually reproducing
mating partners (Schartl et al. 1995a; Lampert & Schartl
2008). The high number of different alleles in a nonrecombining (‘frozen’) hybrid genome, the large effective
population size and adaptive mutations are other mechanisms that may be responsible for the obvious escape
of P. formosa from a long-term fitness decline.
For the formation of asexuals through hybridization
of gonochoristic lineages, the ‘balance hypothesis’
(Wetherington et al. 1987; Moritz et al. 1989) proposes
that the genetic divergence between parental genomes
has to be large enough to affect meiosis in hybrids to
produce a sufficient proportion of gametes without
ploidy reduction, but not too big to avoid a significant
hybrid viability or fertility decrease. However, not only
a certain phylogenetic distance between hybridizing
species is required to affect meiosis, which by itself
greatly reduces the chance of appropriate species to
occur sympatrically (as most speciation events occur in
allopatry). Beyond this precondition, our data suggest
that solely the combination of very specific genotypes
might lead to the successful formation of an asexual
lineage.
Our results of a most likely monophyletic origin of
P. formosa, the rather high level of clonal diversity and
the relatively old age of the species indicate that unisexual vertebrates might be rare not only because they suffer the long-term consequences of clonal reproduction
but because they might be also very rarely formed—because of complex genetic preconditions necessary to
produce viable and fertile clonal genomes. Our ‘rare
formation hypothesis’, as exemplified by gynogenetic
P. formosa, provides an explanation for the rarity of
asexual species, which—once arisen—can be ecologically very successful and persistent. This idea is in line
with earlier assumptions by Vrijenhoek (1989) who proposed that the formation of unisexual lineages faces
severe genetic, developmental and ecological constraints.
Acknowledgements
We thank Thomas Broquet, Jérôme Goudet, Nicolas Perrin and
Nicolas Salamin (Lausanne), Michael Hickerson (New York),
Florian Leese (Bochum) as well as a two anonymous reviewers
and Robert J. Vrijenhoek (Monterey) for discussion and suggestions on the manuscript; Craig Moritz (Berkeley) for support;
Jakob Parzefall, Maximilian Schartl, Dunja K. Lamatsch, Kay
Körner, Marion Döbler, Martin Plath, Katja Heubel, Ute Hornung, Alexander Froschauer, Rüdiger Riesch and Jan Schlupp
for help in the field; Zoé Dumas, Monika Niklaus-Ruiz and
Karim Ghali for technical help. We are grateful to the Mexican
government and several US authorities for permission to
collect fishes. This work was supported by grants from the
Deutsche Forschungsgemeinschaft (SFB 567) to M.Sch., SCHL
344 ⁄ 5-6 to I.S., and by a research fellowship (Sto 493 ⁄ 1-2) from
the Deutsche Forschungsgemeinschaft (DFG) to M.St.
References
Alves MJ, Coelho MM, Collares-Pereira MJ (2001) Evolution in
action through hybridisation and polyploidy in an Iberian
freshwater fish: a genetic review. Genetica, 111, 375–385.
Angers B, Schlosser IJ (2007) The origin of Phoxinus
eos-neogaeus unisexual hybrids. Molecular Ecology, 16, 4562–
4571.
Avise J (2008) Clonality. Oxford University Press, Oxford.
Avise J, Trexler J, Travis J, Nelson W (1991) Poecilia mexicana is
the recent female parent of the unisexual fish P. formosa.
Evolution, 46, 1530–1533.
Balloux F, Lehmann L, de Meeûs T (2003) The population
genetics of clonal and partially clonal diploids. Genetics, 164,
1635–1644.
Balsano J, Darnell R, Abramoff P (1972) Electrophoretic
evidence of triploidy associated with populations of the
gynogenetic teleost Poecilia formosa. Copeia, 2, 292–297.
Bandelt H, Dress A (1992) A canonical decomposition theory
for metrics on a finite set. Advances in Mathematics, 92, 47–105.
Bell G (1982) The Masterpiece of Nature: The Evolution and
Genetics of Sexuality. University of California Press, Berkeley.
Bogart J, Bi K, Fu J, Noble D, Niedzwiecki J (2007) Unisexual
salamanders (genus Ambystoma) present a new reproductive
mode for eukaryotes. Genome, 50, 119–136.
Burridge C, Craw D, Fletcher D, Waters M (2008a) Geological
dates and molecular rates: Fish DNA sheds light on time
dependency. Molecular Biology and Evolution, 25, 624–633.
Burridge C, Craw D, Jack D, King T, Waters J (2008b) Does
fish ecology predict dispersal across a river drainage divide?
Evolution, 62, 1484–1499.
Clement M, Posada D, Cradell K (2000) TCS: a computer
program to estimate gene genealogies. Molecular Ecology, 9,
1657–1659.
Darnell RM, Abramoff P (1968) Distribution of the gynogenetic
fish, Poecilia formosa, with remarks on the evolution of the
species. Copeia, 2, 354–361.
Dawley R (1989) An introduction to unisexual vertebrates. In:
Evolution and Ecology of Unisexual Vertebrates (eds Dawley R,
Bogart J), pp. 1–18. New York State Museum, Albany, NY.
Dieringer D, Schlötterer C (2003) Microsatellite analyser (MSA):
a platform independent analysis tool for large microsatellite
data sets. Molecular Ecology Notes, 3, 167–169.
Drummond A, Rambaut A (2007) BEAST: Bayesian evolutionary
analysis by sampling trees. BMC Evolutionary Biology, 7, 214.
Drummond A, Ho S, Phillips M, Rambaut A (2006) Relaxed
phylogenetics and dating with confidence. PLOS Biology, 4,
699–710.
Felsenstein J (2004) PHYLIP (Phylogeny Inference Package) Version
3.6. Distributed by the author, Seattle.
Goudet J (2001) Fstat, a program to estimate and test gene
diversities and fixation indices (Version 2.9.3). Available at:
http://www2.unil.ch/popgen/softwares/fstat.htm
Guindon S, Gascuel O (2003) A simple, fast, and accurate
algorithm to estimate large phylogenies by maximum
likelihood. Systematic Biology, 52, 696–704.
S I N G L E O R I G I N O F M U L T I P L E P O E C I L I A C L O N E S 11
Hotz H, Mancino G, Bucciinnocenti S et al. (1985) Rana ridibunda
varies geographically in inducing clonal gametogenesis in
interspecies hybrids. Journal of Experimental Zoology, 236, 199–
210.
Hubbs C, Hubbs L (1932) Apparent parthenogenesis in nature,
in a form of fish of hybrid origin. Science, 76, 628–630.
Hudson D, Bryant D (2006) Application of phylogenetic
networks in evolutionary studies. Molecular Phylogenetics and
Evolution, 23, 254–267.
Huelsenbeck J, Rannala B (1997) Phylogenetic methods come of
age, testing hypotheses in an evolutionary context. Science,
276, 227–232.
Huelsenbeck J, Ronquist F (2001) MrBayes, Bayesian inference
of phylogenetic trees. Bioinformatics, 17, 754–755.
Kuhner M, Yamato J, Felsenstein J (1998) Maximum likelihood
estimation of population growth rates based on the coalescent.
Genetics, 149, 429–434.
Lamatsch D, Stöck M (2009) Sperm-dependent parthenogenesis
and hybridogenesis in teleost fishes. In: Lost Sex—The
Evolutionary Biology of Parthenogenesis (eds Schoen I, Martens
K, van Dijk P), pp. 399–432. Springer, Heidelberg, Berlin.
Lamatsch DK, Steinlein C, Schmid M, Schartl M (2000)
Noninvasive determination of genome size and ploidy
level in fishes by flow cytometry: detection of triploid
Poecilia formosa. Cytometry, 39, 91–95.
Lamatsch D, Nanda I, Schlupp I et al. (2004) Distribution and
stability of supernumerary microchromosomes on natural
populations of the Amazon molly, Poecilia formosa. Cytogenetic and Genome Research, 106, 189–194.
Lampert K, Schartl M (2008) The origin and evolution of a
unisexual hybrid: Poecilia formosa. Philosophical Transactions of
the Royal Society B, 363, 2901–2909.
Lampert K, Lamatsch D, Epplen J, Schartl M (2005) Evidence
for a monophyletic origin of triploid clones of the Amazon
molly, Poecilia formosa. Evolution, 59, 881–889.
Lampert K, Lamatsch D, Schories S et al. (2006) Microsatellites
for the gynogenetic Amazon molly, Poecilia formosa: useful
tools for detection of mutation rate, ploidy determination
and overall genetic diversity. Journal of Genetics, 1, 67–71.
Lampert K, Lamatsch D, Fischer P et al. (2007) Automictic
reproduction in interspecific hybrids of poeciliid fish. Current
Biology, 17, 1948–1953.
Lampert K, Fischer P, Schartl M (2009) Major histocompatibility
complex variability in the clonal Amazon molly, Poecilia
formosa: is copy number less important than genotype?
Molecular Ecology, 18, 1124–1136.
Maynard Smith J (1978) The Evolution of Sex. Cambridge
University Press, Cambridge.
Menken S, Smit E, Den Nijs H (1995) Genetical population
structure in plants: gene flow between diploid sexual and
triploid asexual dandelions (Taraxacum section Ruderalia).
Evolution, 49, 1108–1118.
Monaco P, Rasch E, Balsano J (1984) Apomictic reproduction in
the Amazon molly, Poecilia formosa, and its triploid hybrids.
In: Evolutionary Genetics of Fishes (ed. Turner B), pp. 311–318.
Plenum Press, New York.
Moritz C, Brown W, Densmore L et al. (1989) Genetic diversity
and the dynamics of hybrid parthenogenesis in Cnemidophorus (Teeidae) and Heteronotia (Gekkonidae). In:
Evolution and Ecology of Unisexual Vertebrates (eds Dawley R,
Bogart J), pp. 87–112. New York State Museum, Albany.
Muller H (1932) Some genetic aspects of sex. American
Naturalist, 66, 118–138.
Nanda I, Schlupp I, Lamatsch D et al. (2007) Stable inheritance
of host species-derived microchromosomes in the gynogenetic
fish, Poecilia formosa. Genetics, 177, 917–926.
Nylander JAA (2004) MrModeltest v2. Program distributed by
the author, Uppsala.
Orr H, Turelli M (2001) The evolution of postzygotic isolation:
accumulating Dobzhansky-Muller incompatibilities. Evolution,
55, 1085–1094.
Otto S, Lenormand T (2002) Resolving the paradox of sex and
recombination. Nature Review Genetics, 3, 252–261.
Rasch E, Balsano J (1974) Biochemical and cytogenetic studies
of Poecilia from eastern Mexico. II. Frequency, perpetuation
and probable origin of triploid genomes in females
associated with Poecilia formosa. Revista de Biologı̀a Tropicale,
21, 351–381.
Schartl M, Schlupp I, Schartl A, Meyer MM, Nanda I, Schmid
M, Epplen JT, Parzefall J (1991) On the stability of
dispensable constituents of the eukaryotic genome: Stability
of coding sequences versus truly hypervariable sequences in
a clonal vertebrate, the amazon molly, Poecilia formosa.
Proceedings of the National Academy of Sciences of the United
States of America, 88, 8759–8763.
Schartl M, Nanda I, Schlupp I, Wilde B, Epplen JT, Schmid M,
Parzefall J (1995a) Incorporation of subgenomic amounts of
DNA as compensation for mutational load in a gynogenetic
fish. Nature, 373, 68–71.
Schartl M, Wilde B, Schlupp I, Parzefall J (1995b) Evolutionary
origin of a parthenoform, the Amazon molly Poecilia formosa,
on the basis of a molecular genealogy. Evolution, 49, 827–
835.
Schaschl H, Tobler M, Plath M, Penn DJ, Schlupp I (2008)
Polymorphic MHC loci in an asexual fish, the Amazon molly
(Poecilia formosa; Poeciliidae). Molecular Ecology, 17, 5220–
5230.
Schlupp I (2005) The evolutionary ecology of gynogenesis.
Annual Review of Ecology, Evolution, and Systematics, 36, 399–
417.
Schlupp I, Parzefall J, Schartl M (2002) Biogeography of the
Amazon molly, Poecilia formosa. Journal of Biogeography, 29,
1–6.
Schories S, Lampert KP, Lamatsch DK, Garcia de Leon FJ,
Schartl M (2007) Analysis of a possible independent origin of
triploid P. formosa outside of the Rio Purificacion river
system. Frontiers in Zoology, 4, 13.
Schröder J (1964) Genetische Untersuchungen an domestizierten
Stämmen der Gattung Mollienesia (Poeciliidae). Zoologische
Beiträge, 10, 369–463.
Stöver BC, Müller KF (2010) TreeGraph 2: Combining and
visualizing evidence from different phylogenetic analyses.
BMC Bioinformatics, 11, 7.
Swofford, DL (2003) PAUP*. Phylogenetic Analysis Using
Parsimony (*and other Methods). Version 4. Sinauer Associates,
Sunderland, Massachusetts.
Takezaki N, Nei M (1996) Genetic distances and reconstruction
of phylogenetic trees from microsatellite DNA. Genetics, 144,
389–399.
Turner BJ (1982) The evolutionary genetics of a unisexual fish,
Poecilia formosa. In: Mechanisms of Speciation (ed. Barigozzi C),
pp. 265–305. Liss, New York.
12 M . S T Ö C K E T A L .
Turner B, Steeves HR III (1989) Induction of spermatogenesis
in an all-female fish species by treatment with an
exogeneous androgen. In: Evolution and Ecology of Unisexual
Vertebrates (eds Dawley R, Bogart J), pp. 113–122. New York
State Museum, Albany.
Turner BJ, Brett BLH, Rasch EM, Balsano JS (1980)
Evolutionary genetics of a gynogenetic fish, Poecilia formosa,
the Amazon Molly. Evolution, 34, 246–258.
Turner BJ, Balsano JS, Monaco PJ, Rasch EM (1983) Clonal
diversity and evolutionary dynamics in a diploid-triploid
breeding complex of unisexual fishes (Poecilia). Evolution, 37,
798–809.
Turner BJ, Elder JF, Laughlin TF, Davis WP (1990) Geneticvariation in clonal vertebrates detected by simple-sequence
DNA fingerprinting. Proceedings of the National Academy of
Sciences of the United States of America, 87, 5653–5657.
Vrijenhoek R (1994) Unisexual fish: model systems for
studying ecology and evolution. Annual Review of Ecology and
Systematics, 25, 71–96.
Vrijenhoek RC (1989) Genetic and ecological constraints in the
origins and establishment of unisexual vertebrates. In:
Evolution and Ecology of Unisexual Vertebrates (Dowley RM,
Bogart JB.s eds), pp. 24—31.New York State Museum,
Albany, New York.
Vrijenhoek R, Dawley R, Cole C, Bogart J (1989) A list of the
known unisexual vertebrates. In: Evolution and Ecology of
Unisexual Vertebrates (eds Dawley R, Bogart J), pp. 19–23.
New York State Museum, Albany, NY.
Watterson G, Guess H (1977) Is the most frequent allele the
oldest? Theoretical Population Biology, 11, 141–160.
Wetherington J, Kotora K, Vrijenhoek R (1987) A test of the
spontaneous heterosis hypothesis for unisexual vertebrates.
Evolution, 41, 721–723.
the molecular processes in organismic development and their
malfunction in cancerogenesis. Ingo Schlupp is an Associate
Professor at the University of Oklahoma. He is interested in
organisms living in extreme environments, and the evolution
and maintenance of recombination. His former graduate student Dirk Möller, PhD, is now a businessman. Matthias Stöck
investigates vertebrate speciation, hybridization, polyploidization, phylogeography and the evolution of sex-linked markers,
especially in amphibians and fishes.
Supporting information
Additional supporting information may be found in the online
version of this article.
Fig. S1 Unrooted Bayesian tree obtained from sequence data of
the mitochondrial D-loop of 130 individuals of P. formosa and
its maternal ancestor P. mexicana, using the program MrBayes,
10 Mio generations, GTR+I model.
Fig. S2 Unrooted Maximum Likelihood tree obtained from
sequence data of the mitochondrial D-loop of 130 individuals
of P. formosa and its maternal ancestor P. mexicana, using the
program PhyML, GTR model, bootstrap values for 1000 resampled datasets are given for major branches.
Fig. S3 Depicted are the most common alleles for each locus
that Poecilia formosa had in common with Poecilia mexicana.
Table S1 Locality information, IDs as in Fig. 1 and Figs S1
and S2, Supporting Information
Table S2 Multilocus microsatellite genotypes for all individuals
analysed for the phylogenetic tree (Fig. 2)
Appendix S1 Sample information.
This work contributes to an integrative research program on
inter-specific interactions between organisms. Kathrin P. Lampert uses molecular and field techniques to investigate the evolution and maintenance of sexual reproduction in vertebrates
and invertebrates, Manfred Schartl’s main research areas are
Please note: Wiley-Blackwell are not responsible for the content
or functionality of any supporting information supplied by the
authors. Any queries (other than missing material) should be
directed to the corresponding author for the article.
1601mex_III_12_Tamasopo (3)
M_VIII_Moe444_Huchihuayan (2)
M_VIII_Moe492_Huchihuayan (2)
Poe91mex_236CascadasTamasopo (3)
0.5937
M_VII_Moe443_Axtlapexco (1)
1543mex_III_2_Mante (10)
Poe272bmex_233RioGuyalejoEZapata (14)
Poe273mex_233RioGuyalejoEZapata (14)
M_IV_Moe48_SofTampico (4)
M_VI_Moe122_Altamira (6)
M_V_Moe57_SofTampico (4)
M_III_Moe108_Altamira (6)
M_II_Moe1_Altamira (6)
1764form_WBrownsvilleTexas (19)
Poe62bform_233RioGuyalejoEZapata (14)
F_IX_Moe6_Gonzales (11)
F_I_Moe119_Floodway (23)
F_I_Moe192_NofMante (9)
F_I_Moe296_SofLinares (18)
F_I_Moe355_NuevoPadilla (16)
F_I_Moe362_RioCorona (15)
F_I_Moe49_SofTampico (4)
F_I_Moe557_RioBarbarena (8)
F_I_Moe97_E48 (20)
F_VII_Moe327_RioCorona (15)
F_XII_Moe116_Floodway (23)
F_XII_Moe160_Floodway(23)
F_XII_Moe445_SanFernando (12)
F_XII_Moe76_BayView (21)
F_XII_Moe87_Olmito (22)
F_XII__Moe115_Floodway (23)
F_XIV_Moe358_CuidadMante (13)
F_XI_Moe11_Altamira (6)
F_X_Moe184_Barretal (17)
F_X_Moe222_Olimito (22)
F_X_Moe81_E48 (20)
1546form_III_2_Mante (10)
F_XIII_Moe293_SofLinares (18)
Poe15form_232_RioPurificacion (17)
Poe32mex_OjoFrio (5)
Poe271form_233RioGuyalejoEZapata (14)
F_VIII_Moe10_loc.I (22)
F_VI_Moe53_SofTampico (4)
1765form_SanMarcosTexas (24)
F_III_Moe64_Ditch_I (22)
F_II_Moe113_Floodway (23)
F_II_Moe143_Ditch_I (22)
F_II_Moe145_SanMarcos (24)
F_II_Moe84_Olmito (22)
F_IV_Moe245_SanMarcos (24)
F_IV_Moe426_Olmito (22)
F_V_Moe146_SanMarcos(24)
F_V_Moe92_E48 (20)
formosa
0.4212
mexicana
1661mex_III_8_WRioCorona (15)
M_IX_Moe322_Barretal (17)
M_XI_Moe324_Barretal (17)
mexicana
0.5937
1644mex_III_10_RioPurificacion (16)
M_XII_Moe320_Barretal (17)
M_X_Moe229_Barretal (17)
Poe10mex_232_RioPurificacion (16)
0.03
M_XI_Moe3242_Barretal (17)
1644mex1_III_10_RioPurificacion (16)
850
M_XII_Moe3_Moe320_Barretal (17)
1660mex_III_8_WRioCorona (15) Poe11mex_232_RioPurificacion (17)
703
1662mex1_III_8_WRioCorona (15)
Poe32mex_Ojo frio (5)
512
608
1543mex_III_2_Mante (10)
851
Poe272bmex_233RioGuy._EZapata (14)
Poe92mex_236Casc.Tamasopo (3) Poe273mex_233RioGuyal_EZapata (14)
Poe91mex_236Casc.Tamasopo (3)
1601mex_III_12_Tamasopo (3)
M_VIII_Moe444_III_12_Tamasopo (3)
M_VIII_Moe492_III_12_T amasopo (3)
978
661
680
644
M_V_Moe57_SofTampico (4)
M_VI_Moe122_Altamira (6)
F_VIII__Moe193_loc.I (22)
616
mexicana
974
F_XII_Moe445_SanFernando (12)
F_XII_Moe76_BayView (21)
F_XII_Moe87_Olmito (22)
862
1765form_SanMarcosTexas (24)
F_IV_Moe426_Olmito (22)
F_V_Moe146_SanMarcos (24)
F_V_Moe92_E48 (20)
831
F_X_Moe81_E48 (20)_
F_XI_Moe15_232_RioPurificacion (17)
F_I_Moe557_RioBarbarena (8)
1764form1_III_2_Mante (10)
F_I_Moe355_NuevoPadilla (16)
F_I_Moe49_SofTampico (4)
F_IX_Moe6_Gonzales (11)
F_I_Moe119_Floodway (23)
F_I_Moe296_SofLinares (18)
F_XIV_Moe358_CuidadMante (13)
F_I_Moe97_E48 (20)
F_VII_Moe327_RioCorona (15)
F_X_Moe222_Olmito (22)
493
formosa
0.001
mexicana
629
mATG31
mATG38
mATG44
mATG78
mATG61
mCA16
mCA20
KonD15
Sat1
PR39
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
P. formosa
P. mexicana
APPENDIX 1 - SAMPLE INFORMATION
Sample ID
Poe1
Poe2
Poe3
Poe4
Poe5
Poe6
Poe7
Poe8
Poe9
Poe10
Poe11
Poe12
Poe13
Poe14
Poe15
Poe16
Poe17
Poe18
Poe19
Poe20
Poe21
Poe22
Poe23
Poe24
Poe25
Poe26
Poe27
Poe28
Poe29
Poe30
Poe31
Poe32
Poe62b
Poe63b
Poe64b
Poe65b
Poe66b
Poe91
Poe92
Poe93
Poe94
Poe271b
Poe272b
Poe273b
1541
1543
1546
1598
1601
1644
1660
1661
1662
1757
1764
1765
1793
1794
1808
1809
1816
2710
2768
2769
2771
LC7
LC13
LC22
LC24
LC28
LC31
LC35
Moe11
Moe12
Moe13
Moe140
Moe14
Moe15
Moe4
Moe108
Moe71
Moe1
Moe20
Moe122
Moe443
Moe572
Moe78
Moe184
Moe324
Moe229
Moe319
Moe322
Moe323
Moe76
Moe358
Moe64
Moe68
Moe143
Moe97
Moe92
Moe81
Moe113
Moe117
Moe391
Moe119
Moe116
Moe160
Moe6
Moe444
Moe492
Moe10
Moe193
phenotypic sex
m
m
m
m
m
m
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
juv
f
f
f
f
f
juv
juv
juv
juv
f
f
m
f
m
f
m
juv
f
m
m
m
m
f
f
m
f
m
m
f
f
m
m
m
m
m
m
w
w
m
juv
f
f
f
f
f
f
f
f
m
m
m
juv
f
f
m
m
m
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
Species
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
Gambusia spec.
mexicana
formosa
formosa
formosa
formosa
formosa
mexicana
mexicana
mexicana
mexicana
formosa (10 fin rays!)
mexicana (8 fin rays)
mexicana
formosa
mexicana
formosa
latipinna
mexicana
mexicana
mexicana
mexicana
mexicana
latipinna
formosa
formosa
Black Molly
Black Molly
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
formosa
formosa
formosa
formosa
formosa
formosa
formosa
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
formosa
formosa
mexicana
mexicana
mexicana
mexicana
mexicana
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
mexicana
mexicana
formosa
formosa
Locality ID
232 Rio Purificacion/Barretal
235 Ojo frio
233 Rio Guayalejo Emilio Zapata
236 Cascadas Tamasopo
III/2 Mante
III/2 Mante
III/2 Mante
Texas, Olmito
III/12 Tamasopo
III/10 Rio Purification
III/8 WF Rio Corona
III/8 WF Rio Corona
III/8 WF Rio Corona
IV/5 New Altamira
Brownsville WF Texas
Texas, San Marcos WF
from breeder
from breeder
II/7 Florida
II/7 Florida
IV/2 New Altamira
Florida
Texas, San Marcos
IX/24 near Tampico
V/4 Rio Oxolotan
IV/5, Altamira, Laguna Champayan
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Altamira
Axtlapexco
Axtlapexco
Barretal
Barretal
Barretal
Barretal
Barretal
Barretal
Barretal
BayView
CiudadMante
Ditch
Ditch
Ditch
E48
E48
E48
Floodway
Floodway
Floodway
Floodway
Floodway
Floodway
Gonzales
Huchihuayan
Huchihuayan
loc.I
loc.I
lat
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
22.1905
23.2771
23.2771
23.2771
23.2771
23.2771
21.9402
21.9402
21.9402
21.9402
23.2771
23.2771
23.2771
22.8123
22.8123
22.8123
25.9865
21.9402
24.0432
23.9506
23.9506
23.9506
22.3915
25.8997
29.8580
25.1256
25.1256
22.3915
22.3915
29.8580
22.4242
22.1905
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
21.0002
21.0002
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
24.0785
25.9500
22.8664
25.9865
25.9865
25.9865
25.9417
25.9417
25.9417
26.1200
26.1200
26.1200
26.1200
26.1200
26.1200
22.8167
21.4801
21.4801
25.9865
25.9865
long
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.3221
-98.9386
-98.9386
-98.9386
-98.9386
-98.9386
-99.3951
-99.3951
-99.3951
-99.3951
-98.9386
-98.9386
-98.9386
-99.0125
-99.0125
-99.0125
-97.5313
-99.3951
-98.9044
-99.1194
-99.1194
-99.1194
-97.9311
-97.4795
-97.8642
-80.4072
-80.4072
-97.9311
-97.9311
-97.8642
-97.8866
-99.3221
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-98.3294
-98.3294
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-99.1235
-97.3500
-99.0258
-97.5313
-97.5313
-97.5313
-97.4310
-97.4310
-97.4310
-97.9612
-97.9612
-97.9612
-97.9612
-97.9612
-97.9612
-98.4300
-98.9668
-98.9668
-97.5313
-97.5313
Moe8
Moe73
Moe192
Moe70
Moe355
Moe222
Moe84
Moe88
Moe90
Moe426
Moe87
Moe557
Moe362
Moe363
Moe327
Moe445
Moe145
Moe147
Moe148
Moe149
Moe154
Moe204
Moe208
Moe246
Moe346
Moe347
Moe245
Moe342
Moe146
Moe296
Moe293
Moe300
Moe523
Moe49
Moe53
Moe55
Moe56
Moe48
Moe52
Moe57
LC1
LC10
LC12
LC14
LC15
LC17
LC18
LC19
LC2
LC21
LC23
LC25
LC26
LC27
LC29
LC32
LC6
LC8
LC9
LC91
LC11
LC30
LC13
LC24
LC28
LC31
LCI
LCII
LCIII
LCIV
LCV
LCVI
IS08009
IS08012
IS08016
IS08018
IS08019
IS08024
IS08027
IS08036
IS08038
IS08042
IS08043
IS08046
IS08048
IS08050
IS08052
IS08054
IS08055
IS08056
IS08060
IS08062
f
m
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
m
juv
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
m
f
f
m
f
m
m
m
f
f
m
m
m
formosa
mexicana
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
mexicana
mexicana
mexicana
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
formosa
latipinna
latipinna
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
mexicana
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
latipinna
loc.I
Altamira
NofMante
NofMante
NuevoPadilla
Olmito
Olmito
Olmito
Olmito
Olmito
Olmito
RioBarbarena
Rio Corona
Rio Corona
Rio Corona
SanFernando
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SanMarcos
SofLinares
SofLinares
SofLinares
SofLinares
SofTampico
SofTampico
SofTampico
SofTampico
SofTampico
SofTampico
SofTampico
IV/5, Altamira, Laguna Champaxan
Angleton,Texas
Cameron,Louisiana
Hwy 182,Louisiana
Houma,Louisiana
DesAlemands,Louisiana
Hwy 613,Mississippi
Fort Walton Beach,Florida
Otter Creek,Florida
Weeki Wachee,Florida
Kissimmee,Florida
Satellite Beach,Florida
Lake Eustins,Florida
Bulow Creek,Florida
St. Mary's,Georgia
Crescent,Georgia
Bennetts Point,Georgia
James Island,South Carolina
Wilmington,North Carolina
Cavasso Creek,Texas
Lincoln Park,Texas
25.9865
22.3915
22.7670
22.7670
24.0408
25.9865
25.9865
25.9865
25.9865
25.9865
25.9865
22.5670
23.9506
23.9506
23.9506
22.8330
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
29.8580
24.6500
24.6500
24.6500
24.6500
22.1830
22.1830
22.1830
22.1830
22.1830
22.1830
22.1830
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
22.3915
29.1772
29.7605
29.5791
29.6200
29.8156
30.5120
30.4802
29.3195
28.4988
28.1458
28.1724
28.8508
29.4073
30.7535
31.5293
32.5929
32.7341
34.2690
28.2173
25.8996
-97.5313
-97.9311
-99.0000
-99.0000
-98.9024
-97.5313
-97.5313
-97.5313
-97.5313
-97.5313
-97.5313
-97.9000
-99.1194
-99.1194
-99.1194
-98.1670
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-97.8642
-99.3330
-99.3330
-99.3330
-99.3330
-97.8670
-97.8670
-97.8670
-97.8670
-97.8670
-97.8670
-97.8670
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-97.9311
-95.4425
-93.8503
-90.0800
-90.6663
-90.4629
-88.5512
-86.5853
-82.7830
-82.6478
-81.4661
-80.5963
-81.2404
-81.1220
-81.5846
-81.4403
-80.4631
-79.9959
-77.7948
-96.9878
-97.4794
Table S1 Locality information, IDs as in Figs. 1, S1 and S2
ID lat
long
1
21.000
-98.329
2
21.480
-98.967
3
21.940
-99.395
3a
N.A.
N.A.
4
22.183
-97.867
5
22.191
-99.322
6
22.392
-97.931
7
22.424
-97.887
8
22.567
-97.900
9
22.767
-99.000
10
22.812
-99.012
11
22.817
-98.430
12
22.833
-98.167
13
22.866
-99.026
14
23.277
-98.939
15
23.951
-99.119
16
24.043
-98.904
17
24.079
-99.124
18
24.650
-99.333
19
25.900
-97.480
20
25.942
-97.431
21
25.950
-97.350
22
25.987
-97.531
23
26.120
-97.961
24
29.858
-97.864
25
25.900
-97.479
26
28.217
-96.988
27
29.177
-95.443
28
29.761
-93.850
29
29.620
-90.666
30
29.816
-90.463
31
29.579
-90.080
32
30.512
-88.551
33
30.480
-86.585
34
29.320
-82.783
35
28.499
-82.648
36
25.126
-80.407
37
28.146
-81.466
38
28.172
-80.596
39
28.851
-81.240
40
29.407
-81.122
41
30.754
-81.585
42
31.529
-81.440
43
32.593
-80.463
44
32.734
-79.996
45
34.269
-77.795
46
17.438
-92.772
Locality
Axtlapexco
Huchihuayan
236 Cascadas Tamasopo & III/12 Tamasopo
Rio Verde
S of Tampico
235 Ojo frio
Altamira & IV/5 New Altamira, Laguna Champaxan
IX/24 near Tampico
Rio Barbarena
N of Mante
III/2 Mante
Gonzales
SanFernando
CiudadMante
III/8 WF Rio Corona
III/10 Rio Purification & Nuevo Padilla
SofLinares
Brownsville WF Texas
E48
BayView
Texas, Olmito & loc. 1 & Ditch
Floodway
Texas, San Marcos
Lincoln Park, Texas
Cavasso Creek, Texas
Angleton, Texas
Cameron, Louisiana
Houma, Louisiana
DesAlemands, Louisiana
Hwy 182, Louisiana
Hwy 613, Mississippi
Fort Walton Beach, Florida
Otter Creek, Florida
Weeki Wachee, Florida
II/7 Florida
Kissimmee, Florida
Satellite Beach, Florida
Lake Eustins, Florida
Bulow Creek, Florida
St. Mary's, Georgia
Crescent, Georgia
Bennetts Point, Georgia
James Island, South Carolina
Wilmington, North Carolina
Rio Oxolotan
Table S3: Genetic distance within and between populations. P. formosa are more closely
related (mean +/- stdev: 0.504 +/- 0.074, min: 0.13, max: 0.64) than P. mexicana from
different field sites (mean +/- stdev: 0.797 +/- 0.067, min: 0.45, max: 0.87)
PfG
PmG
PfP
PmP
PfLC
PfG
mean+/-stdev
min - max
0.393 +/- 0.090
0.125 – 0.472
PmG
mean+/-stdev
min - max
0.806 +/- 0.017
0.763 - 0.818
0.624 +/- 0.266
0.317 - 0.791
PfP
mean+/-stdev
min - max
0.431 +/-0.095
0.127 - 0.595
0.709 +/- 0.063
0.642 - 0.818
0.310 +/- 0.244
0.000 - 0.809
PmP
mean+/-stdev
min - max
0.781 +/- 0.056
0.641 - 0.867
0.796 +/- 0.034
0.664 - 0.854
0.857 +/- 0.016
0.806 - 0.874
0.371 +/- 0.223
0.049 - 0.874
PfLC
mean+/-stdev
min - max
0.516 +/- 0.085
0.255 - 0.641
0.783 +/- 0.031
0.694 - 0.833
0.520 +/- 0.048
0.371 - 0.641
0.823 +/- 0.057
0.625 - 0.900
0.425 +/- 0.182
0.064 - 0.841
PmLC
mean+/-stdev
min - max
0.750 +/- 0.076
0.591 - 0.871
0.690 +/- 0.117
0.454 - 0.854
0.724 +/- 0.097
0.476- 0.871
0.813 +/- 0.045
0.633 - 0.874
0.793 +/- 0.053
0.531 - 0.900
PmLC
0.615 +/- 0.219
0.061 - 0.900

Monophyletic origin of multiple clonal lineages in an asexual fish

Transcription

Similar documents

Molecular phylogeny of the softshell turtle genus

LSN : Seed : Favela chic: Artists give slums a brighter

1410 Rio BRazos DRive | Waterstone

File - Investor

Genetic evidence for wild-living Aspideretes nigricans and a

Laguna de Bay Region/Philippines Laguna de Bay Region

Sample file

General Meeting Friday, 27 April Board 2012

Content AgendaEvent: 5th Viennese Neighbour`s Day

Pharr District Regional Characteristics and Inventory