Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev
Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences Carol A. Stepien *, Amanda E. Haponski Great Lakes Genetics Laboratory, Lake Erie Center and Department of Environmental Sciences, University of Toledo, 6200 Bayshore Road, Toledo, OH 43616, USA
a r t i c l e
i n f o
Article history: Received 17 June 2010 Accepted 23 June 2010 Available online xxxx Keywords: Cytochrome b Etheostoma blennioides Greenside darter Intergrade taxa Percidae Phylogenetic systematics Rooting choice S7 intron Subspecies Taxon inclusion
a b s t r a c t The phylogenetic systematic relationships of the enigmatic greenside darter Etheostoma blennioides complex are analysed using sequences from the mitochondrial (mt) DNA cytochrome b gene and nuclear S7 ribosomal protein intron 1 from putative members of the complex, close relatives, and outgroups (totaling 421 individuals). We compare results from Bayesian and maximum likelihood analysis approaches and a variety of rooting and taxon inclusion scenarios, and include all putative subspecies and intergrade taxa for a new comprehensive analysis. Results reveal that nuclear and mtDNA data congruently, under all scenarios and approaches tested, define a highly-supported restricted greenside darter complex comprising three putative subspecies: E. b. blennioides, E. b. pholidotum, and part of E. b. newmanii (excepting those from the Tennessee/Hiwassee River clade). Within this redefined E. blennioides, only a single putative subspecies – E. b. blennioides – is monophyletic in the mtDNA trees, and none are monophyletic in the nuclear DNA trees. Nuclear and mtDNA results support E. gutselli as a separate species and suggest that the Tennessee/Hiwassee River clade of ‘‘E. b. newmanii” also may constitute a separate species (provisionally ‘‘E. newmanii”), with neither being a part of our redefined E. blennioides complex. The nuclear DNA trees depict the two as highly-supported divergent clades, but the mtDNA results group them together as a single clade, indicating introgression. Future study with greater sample sizes in the southern watersheds, coupling morphological analyses with additional nuclear gene phylogenies, is recommended to further investigate the relationships within the greenside darter complex. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction The greenside darter Etheostoma blennioides Rafinesque 1819 (Teleostei: Percidae) is a historically abundant and diverse fish species that was described by Miller (1968) as containing four subspecies: E. b. blennioides Rafinesque 1819, E. b. newmanii (Agassiz 1854), E. b. gutselli (Hildebrand 1932), and E. b. pholidotum Miller 1968, which variously are distributed in high-gradient streams of the lower Great Lakes and the Potomac and Mississippi River drainage systems (Fig. 1). Based on morphological differences, Etnier and Starnes (2001) elevated the Tuckasegee darter E. gutselli to species-level, which was accepted by the American Fisheries Society’s Committee on the names of fishes (Nelson et al., 2004). Its species-level separation was further supported by our genetic analyses of its DNA sequence divergence (Haponski and Stepien, 2008). In addition to the putative subspecies, Miller (1968) recognized two morphological races of E. b. pholidotum from the Missouri River and Great Lakes, and three races of E. b. newmanii inhabiting the Tennessee, Cumberland, and Arkansas Rivers – * Corresponding author. Fax: +1 419 530 8399. E-mail address:
[email protected] (C.A. Stepien).
along with three possible zones of intergradation in the Gasconade River (E. b. newmanii E. b. pholidotum), Green River (E. b. newmanii E. b. blennioides), and Hiwassee River (E. b. newmanii E. b. gutselli) – whose relationships are examined here. Definition and composition of the greenside darter complex have remained very controversial, despite recent DNA studies. The aim of this investigation is to define E. blennioides and to elucidate its phylogenetic relationships, employing new analyses of mitochondrial (mt) and nuclear DNA sequence data. Our prior study (Haponski and Stepien, 2008) and two others (Piller et al., 2008 and Piller and Bart, 2009) that all were published in this journal (Molecular Phylogenetics and Evolution) left fundamental untested and unresolved questions regarding whether, how, and why resultant mtDNA cytochrome (cyt) b gene phylogenies for the greenside darter complex differed among the three studies. Those studies employed different phylogenetic approaches, outgroup choice, rooting scenarios, and inclusion of taxa, circumventing analyses of consensus. In this new study we uniquely test for congruence among phylogenies that include all proposed component taxa of the greenside darter complex (putative subspecies and intergrades), hypothesized sister groups, and outgroups. We then specifically test whether the mtDNA trees are congruent with
1055-7903/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2010.06.017
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
2
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Fig. 1. Map showing the distribution of taxa examined in this new investigation from the greenside darter Etheostoma blennioides complex sensu Miller (1968) and E. gutselli. Solid line denotes extent of the Great Lakes watershed.
new comprehensive analyses using nuclear DNA sequences, and whether the results differ with genome, phylogenetic method, rooting choice, and inclusive taxa. The goal of our new study thus is to illuminate the systematic status and composition of the greenside darter complex, and address what is needed in future research work. We analyse all known mtDNA cyt b and nuclear S7 intron 1 DNA sequence data to test for congruence and differences among phylogenies derived from: (a) nuclear and mtDNA sequences, (b) Bayesian and maximum likelihood (ML) approaches, (c) various rooting and outgroup scenarios, and (d) inclusion of all putative subspecies and intergrade individuals. We examine these factors in relation to progress in resolving relationships among greenside darter component taxa and sister species, and provide recommendations for future investigation. 2. Materials and methods In this study, as in Haponski and Stepien (2008) and commonly used in phylogenetic systematics, our use of the term ‘‘outgroup” follows Wiley (1981): ‘‘An outgroup is a species or higher monophyletic taxon that is examined in the course of a phylogenetic study to determine which characters may be inferred to be ‘apomorphic’ (i.e., ancestral). One or several outgroups may be examined for each decision. The most critical outgroup comparisons involve the sister group of the taxon studied.” We further adopt the common definition of Wiley (1981) that, ‘‘A sister group is a
species or higher monophyletic taxon that is hypothesized to be the closest genealogical relative of a given taxon exclusive of the ancestral species of both taxa”. Our use of these definitions in Haponski and Stepien (2008) appear to have been misinterpreted by Piller and Bart (2009), leading to possible errors in data analysis and interpretation that are tested here. To define E. blennioides and clarify its intra- and inter-specific relationships, as well as address the allegations made by Piller and Bart (2009) against our previous mtDNA analyses in Haponski and Stepien (2008), we here combine and analyse all available mt cyt b sequence data totaling 401 samples (see Table 1). This new data set comprises the original data used by Haponski and Stepien (2008) (59 haplotypes representing E. blennioides and four for E. gutselli) along with unique haplotypes from Piller et al. (2008) (28 for E. blennioides and two for E. gutselli), including intergrade taxa (see Fig. 1 and Table 1). Moreover, we present new and comprehensive analyses of all nuclear S7 intron DNA sequence data, including 28 E. blennioides and six E. gutselli haplotypes from Haponski and Stepien (2008), and 12 E. blennioides and two E. gutselli haplotypes from Piller et al. (2008). These nuclear sequences encompass the clades determined from the mtDNA data. We then evaluate congruency between results from the nuclear DNA trees and the mtDNA trees. Sequences analysed in this investigation were deposited on the National Institutes of Health (N.I.H.) GenBank () by Haponski and Stepien (2008) as: cyt b GenBank Accession Nos. EF587846-48, EU118843-96, and EU716042-47, and S7 intron 1 Nos.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
3
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Table 1 Samples of greenside darters Etheostoma blennioides sensu Miller (1968) and Etheostoma gutselli, with sampling locations, major drainage systems, latitude (°N), longitude (°W), number of individuals per site, and number of cytochrome b haplotypes per site. Genetic clade
Location
Clade I. E. blennioides A. Ohio/Susquehanna R. = E. b. blennioides
Total Total s. Muddy Fork, IN x. Great Miami R., OH y. Stillwater R., OH cc. Loramie Cr., OH uu. Scioto R., OH zz. Hocking R., OH aaa. Cuyahoga R., OH eee. Allegheny R., NY fff. Monocacy R., MD* hhh. Seeley Cr., NY* iii. Susquehanna R., NY ggg. Ganargua Cr., NY
A. Ohio/Susquehanna R. (42%)* &B. Great lakes/Wa bash R. (58%)* B. Great Lakes/Wabash R. = former E. b. pholidotum
C. Osage R. = former E. b. pholidotum
D. Meramec R. = former E. b. pholidotum E. Cumberland R. = former E. b. newmanii
F. White/Arkansas R. = former E. b. newmanii
G. Ouachita R. = former E. b. newmanii H. Green/Barren R. = former E. b. newmanii E. b. blennioides
I. Gasconade R. = former E. b. newmanii E. b. pholidotum Clade II. E. gutselli
‘‘E. b. newmanii/E. newmanii”
Drainage system
Latitude
Ohio R. Ohio R. Ohio R. Ohio R. Ohio R. Ohio R. Great lakes Ohio R. Potomac R. Susquehanna R. Susquehanna R. Great lakes
38.4383 39.5143 40.1866 40.1942 39.7683 39.4615 41.1446 41.9980 39.4193 42.0145 42.1088 43.0688
Longitude
85.7733 84.7132 84.4732 84.2410 82.9957 82.3129 81.4382 78.2440 77.3903 76.8924 76.2738 77.2981
Total
N individuals
N Haplotypes
370 156 4 11 2 1 18 21 53 21 1 1 23 19
58 22 2 6 1 1 7 6 1 3 1 1 1 2
160
29
1 8 1 11 20 14 32 38 9 24 1 1 4 1 1 1 1 14 8 1 2 5 4
1 4 1 3 10 3 10 2 2 2 1 1 4 1 1 1 1 8 4 1 2 2 4
l. Middle Fork Vermilion R., IL m. Wabash R., IL r. Big Blue R., IN w. Wabash R., IN bb. Auglaize R., OH gg. Ottawa R., OH mm. Blanchard R., OH rr. Portage R., OH ww. Belle R., MI bbb. Grand R., OH ccc. Laurel Cr., Ontario Canada* ddd. Carroll Cr., Ontario Canada* Total b. Little Sac R., MO c. Cole Camp Cr., MO* d. Niangua R., MO i. Big Tavern Cr., MO k. Meramec R., MO Total o. West Fork Stones R., TN v. Obey R., TN dd. Rockcastle R., KY Total
Wabash R. Wabash R. Wabash R. Wabash R. Great lakes Great lakes Great lakes Great lakes Great lakes Great lakes Great lakes Great lakes
40.2333 39.7737 39.3553 40.8396 40.6786 40.7552 41.0362 41.4761 42.9418 41.7349 43.4804 43.5857
87.7667 87.6035 85.9772 85.4380 84.2595 84.0367 83.5767 83.2958 82.8286 81.0474 80.5844 80.2995
Osage R. Osage R. Osage R. Osage R. Meramec R.
37.3036 38.2890 37.7847 38.2777 38.5039
93.3778 93.2260 92.8369 92.2332 90.5909
Cumberland R. Cumberland R. Cumberland R.
35.7225 36.5284 37.3066
86.4453 85.4401 84.1480
a. Kings R., AR f. Middle Fork Little Red R., AR h. Cadron Cr., AR j. South Fork Spring R., AR e. Saline R., AR Total
White R. White R. Arkansas R. White R. Ouachita R.
36.3950 35.8164 35.3800 36.3125 34.5500
93.6353 92.5492 92.2839 91.5278 92.7000
1 1 1 1 1 3
1 1 1 1 1 3
q. Trammel Fork Green R., KY u. East Fork Little Barren R., KY t. South Fork Little Barren R., KY g. Gasconade R., MO
Green R. Barren R. Barren R. Gasconade R.
36.7316 37.0023 37.1003 37.7494
86.2727 85.5078 85.6342 92.3933
1 1 1 1
1 1 1 1
31 15 1 2 1 1 3 2 2 3 7 1 1 1 1 1 1 1
17 6 1 1 1 1 1 1 1 2 7 1 1 1 1 1 1 1
Total Total kk. Cheoah R., NC ll. Little Tennessee R., NC nn. Deep Cr., NC* oo. Cowee Cr., NC* pp. Tuckasegee R., NC qq. Little Tennessee R., NC* tt. Jonathan Cr., NC* vv. Pigeon R., NC Total n. West Fork Sugar Cr., TN p. Duck R., TN ff. Citico Cr., TN ii. Little R., TN ss. Pigeon R., TN xx. French Broad R., NC yy. Nolichucky R., TN
Little Tennessee Little Tennessee Little Tennessee Little Tennessee Little Tennessee Little Tennessee Pigeon R. Pigeon R.
R. R. R. R. R. R.
35.3072 35.3902 35.4516 35.2583 35.4522 35.1422 35.6135 35.5256
83.7944 83.6244 83.4360 83.4126 83.3936 83.3767 83.0397 82.8481
Tennessee R. Duck R. Little Tennessee R. Tennessee R. Pigeon R. Tennessee R. Tennessee R.
35.0978 35.4697 35.5081 35.7700 35.8753 35.1490 36.1173
87.2819 86.3781 84.1047 83.8764 83.1958 82.8000 82.7798
(continued on next page)
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
4
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Table 1 (continued)
*
Genetic clade
Location
Drainage system
Latitude
Longitude
‘‘E. b. newmanii/E. newmanii” E. gutselli
Total z. Big Lost Cr., TN aa. Toccoa R., GA* ee. Coopers Cr., GA* hh. Valley R., NC* jj. Brasstown Cr., GA*
Hiwassee Hiwassee Hiwassee Hiwassee Hiwassee
35.1606 34.9442 34.7437 35.1279 34.9208
84.4686 84.3281 84.1233 83.9844 83.8440
R. R. R. R. R.
N individuals 9 2 2 1 2 2
N Haplotypes 4 2 1 1 1 1
Denote sites sampled by Piller et al. (2008) that lacked reported coordinates, which we approximated from site descriptions.
Table 2 Outgroups used in this study with GenBank accession numbers and publication sources. Taxa
Cytochrome b
Nuclear S7 intron 1
E. E. E. E. E. E. E. E. E. E. E. E. E.
AY374260 (Sloss et al., 2004) AY964698 (Switzer and Wood, unpublished) AY964700 (Switzer and Wood, unpublished) AF045348 (Song et al., 1998) EU296691 (Piller et al., 2008) EU296688 (Piller et al., 2008) AY374273 (Sloss et al., 2004) AF288442 (Porterfield and Page, unpublished) EU296685 (Piller et al., 2008) EU296692 (Piller et al., 2008) EU296694 (Piller et al., 2008) AY374278 (Sloss et al., 2004) AY964706 (Switzer and Wood, unpublished)
AY573274 (Morrison et al., 2006) EU118924 (Haponski and Stepien, 2008) NA AY573273 (Morrison et al., 2006) EU296725 (Piller et al., 2008) NA NA EU118923 (Haponski and Stepien, 2008) NA NA EU296726 (Piller et al., 2008) AY573270 (Morrison et al., 2006) EU118925-26 (Haponski and Stepien, 2008)
bellum blennius blennius b. sequatchiense camurum euzonum lynceum rafinesquei rupestre swannanoa thalassinum tetrazonum variatum zonale
NA denotes that the sequence was not available from GenBank or other sources.
EU118897-922 and EU716054-60, and by Piller et al. (2008) as: cyt b EU296636-684 and EU296695-98, and S7 intron 1 EU296699722). In this new study, we aligned all sequences using BioEdit 7.05 (Hall, 1999) and pruned identical sequences from Piller et al. (2008) so that each haplotype was represented only once. We compare variation in the greenside darter complex with several related taxa and outgroups, including: E. bellum, E. blennius, E. camurum, E. euzonum, E. gutselli, E. lynceum, E. rafinesquei, E. rupestre, E. swannanoa, E. tetrazonum, E. thalassinum, E. variatum, and E. zonale (Table 2). Haponski and Stepien (2008) rooted their trees to the outgroup taxa E. bellum and E. camurum (see Sloss et al., 2004, and Table 2 here), whereas Piller et al. (2008) rooted their trees to E. euzonum, E. tetrazonum, and E. variatum. The reason that we rooted to E. bellum and E. camurum in our earlier study was that data for the mtDNA control region and S7 nuclear intron also were available for those outgroup taxa. In this new study we uniquely explore various rooting scenarios, including the scenario of our work portrayed by Piller and Bart (2009, p. 314) that we rooted to all of the following taxa at once: E. bellum, E. blennius, E. camurum, E. gutselli, E. rafinesquei, E. rupestre, E. zonale, and E. variatum (whereas, in truth, we rooted to only E. bellum and E. camurum). Piller and Bart (2009, p. 314) purported that we were unable to deduce the sister group relationship and violated phylogenetic systematic principles since they claimed (falsely) that we had rooted our trees to a possible sister group (e.g., to E. gutselli). In this new study, we thus explore results of rooting our Bayesian and ML trees based on mtDNA cyt b and nuclear S7 intron sequence data to: (a) the outgroups E. bellum and E. camurum, as originally used by Haponski and Stepien (2008), (b) the scenario erroneously attributed to us by Piller and Bart (2009, p. 314) using E. bellum, E. blennius, E. camurum, E. gutselli, E. rafinesquei, E. rupestre, E. variatum, and E. zonale and, (c) E. euzonum, E. tetrazonum, and E. variatum following Piller et al. (2008). Our new phylogenetic
analyses uniquely compare results from a Bayesian approach implemented in MrBayes v3.1.2 (Ronquist and Huelsenbeck, 2003) and a maximum likelihood (ML) approach in PhyML v3.0 (Guindon and Gascuel, 2003). In Haponski and Stepien (2008), we employed ML and maximum parsimony (PAUP* v4.0 b10; Swofford, 2003) analyses and since Piller and Bart (2009) used solely a Bayesian approach, their trees were not directly comparable to ours. Support for nodes of the ML trees is analysed here using 1000 bootstrap pseudo-replications (Felsenstein, 1985). We employed the Akaike information criteria (AIC) from Modeltest 3.7 (Posada and Crandall, 1998) to determine the most appropriate nucleotide substitution model. The general timereversible model (Lanave et al., 1984), including invariable sites (I) and a gamma distribution (a) was determined to be the most appropriate for the cyt b datasets. For our original cyt b dataset, the model parameters for the invariable sites and gamma distribution followed Haponski and Stepien (2008), with values of I = 0.610 and a = 2.074. Model parameters for our original dataset, rooted to the outgroups E. euzonum, E. tetrazonum, and E. variatum of Piller et al. (2008), are I = 0.609 and a = 1.927. Parameters for the combined Haponski and Stepien (2008) and Piller et al. (2008) datasets, with E. bellum and E. camurum, are I = 0.574 and a = 1.430; values for the combined datasets rooted to the outgroups used by Piller et al. (2008) are I = 0.609 and a = 1.927. For the original S7 data set from Haponski and Stepien (2008), Modeltest selected the Kimura (1980) two-parameter model including a gamma distribution (a = 0.275). For our original dataset rooted to the outgroups E. euzonum, E. tetrazonum, and E. variatum of Piller et al. (2008), Modeltest supported the Hasegawa, Kishino, and Yano (HKY) model (Hasegawa et al., 1985) with a gamma distribution (a = 0.308). The HKY model with a gamma distribution also was selected for the combined Haponski and Stepien (2008) and Piller et al. (2008) datasets, with E. bellum and E. camurum (a = 0.333). For the combined datasets rooted to the
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
outgroups used by Piller et al. (2008), Modeltest chose the HKY model that included a gamma distribution (a = 0.803) and invariable sites (I = 0.446). We compare the ML trees with those obtained using Bayesian analyses with a Metropolis-coupled Markov chain Monte Carlo (MCMCMC) approach in MrBayes. Analyses were initially run for 5 million generations, with sampling every 100 generations. Four separate chains were run simultaneously for each analysis, and
5
two analyses were run at the same time. The burn-in period for the MCMCMC was determined by plotting log likelihood values for each generation to identify when stationarity was reached. However, after plotting the log likelihood values for the 5 million generations, we found that cyt b analyses that included the more distantly related E. bellum and E. camurum had not reached stationarity, whereas those that excluded them had. The cyt b analyses that included E. bellum and E. camurum thus were run
Fig. 2. Bayesian 50% majority rule consensus trees of mitochondrial DNA cytochrome b sequences for the greenside darter complex and relatives. Trees are congruent with those from our maximum likelihood (ML) analyses. Various rooting scenarios are evaluated, including: (a) using the data of Haponski and Stepien (2008) rooted to E. bellum and E. camurum, (b) rooted to E. bellum, E. blennius, E. camurum, E. gutselli, E. rupestre, E. rafinesquei, E. rupestre, E. variatum, and E. zonale as claimed by Piller and Bart (2009), and (c) based on the combined datasets of Haponski and Stepien (2008) and Piller et al. (2008), rooted to E. bellum and E. camurum. Redundant sequences from Piller et al. (2008) are pruned from all analyses. All intergrade taxa from Piller et al. (2008) are included in c. Values at nodes = support from Bayesian posterior probability/1000 bootstrap pseudo-replications in the ML tree; both given as percentages. *100% posterior probability (1.00)/100% bootstrap support. Parentheses = (N haplotypes, N samples). X Type locality of the taxon. Our redefined E. blennioides is labeled ‘‘Clade I”. ‘‘Clade II” denotes E. gutselli and ‘‘E. b. newmanii/E. newmanii” from the Tennessee and Hiwassee Rivers. Clades within E. blennioides are lettered ‘‘A–I”.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
6
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Fig. 2 (continued)
for 10 million generations to ensure that stationarity had been reached. After the MrBayes analyses were completed, plotting the log likelihood values at each generation and determining when the likelihood values had stabilized was used to determine the burn-in period. Twenty-five percent of the generations were discarded as burn-in, and the trees and parameter values sampled prior to the burn-in also were discarded. A 50% majority rule consensus tree was calculated based on the remaining generations and branch support for the trees was calculated via the posterior probability distribution (Holder and Lewis, 2003) in MrBayes. 3. Results 3.1. Phylogenetic relationships and composition of the greenside darter complex based on mtDNA Our phylogenetic trees based on the mtDNA cyt b gene sequence data with both Bayesian and ML analyses, and under all rooting and taxon inclusion scenarios, are 100% congruent in defining an identical set of taxa comprising a restricted greenside darter complex (Figs. 2a–c, Appendix 1a–b). In all trees, this restricted E. blennioides forms a clade (marked as ‘‘Clade I”) with very high support (0.99–1.00 Bayesian posterior probability and 99–100% ML
bootstrap support values) comprising E. b. blennioides, the former E. b. pholidotum, and taxa previously classified as E. b. newmanii sensu Miller (1968) from the Cumberland, White, and Ouachita Rivers, but excluding ‘‘E. b. newmanii” from the subspecies’ type locality of the Tennessee River system and from the Hiwassee River drainage (see further explanation below). These groups also include intergrade taxa sensu Miller (1968) (Fig. 2c, Appendix 1b), and their compositions appear consistent with and without inclusion of the intergrades (Figs. 2a–c, Appendix 1a–b). E. gutselli (labeled as ‘‘Clade II”) is not part of the greenside darter complex in any of our analyses, appearing as a separate taxon supported by high support (0.90–0.97/91–97% in Fig. 2a–b, Appendix 1a), or clustered together with ‘‘E. b. newmanii” from the Tennessee and Hiwassee River systems (Clade II, Fig. 2c, Appendix 1b). Neither E. gutselli nor these ‘‘E. b. newmanii” appear as monophyletic on the trees, according to the cyt b data, likely indicating a more recent evolutionary history shared with each other than with our redefined E. blennioides. On our trees, E. blennius either appears as the apparent sister group to E. gutselli (Fig. 2a–b, Appendix 1a) or as the sister group to the clade containing E. gutselli plus the samples of ‘‘E. b. newmanii” from the Tennessee and Hiwassee River drainages; all of which are outside our redefined greenside darter complex (Fig. 2c, Appendix 1b). In the mtDNA trees, E. rupestre
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
7
Fig. 3. Bayesian 50% majority rule consensus trees of nuclear S7 intron 1 DNA sequences for the greenside darter complex and relatives, rooted to E. bellum and E. camurum, based on: (a) our original dataset from Haponski and Stepien (2008) and (b) the combined Haponski and Stepien (2008) and Piller et al. (2008) datasets. Redundant sequences from Piller et al. (2008) are pruned from all analyses. All intergrade taxa from Piller et al. (2008) are included in b. Values at nodes = support from Bayesian posterior probability/1000 bootstrap pseudo-replications in the ML tree; both given as percentages. *100% posterior probability (1.00)/100% bootstrap support. Parentheses = (N haplotypes, N samples). XType locality of the taxon. Our redefined E. blennioides is labeled ‘‘Clade I”. ‘‘Clade II” includes E. gutselli (labeled as ‘‘IIb”) and our redefined ‘‘E. b. newmanii/E. newmanii” (labeled as ‘‘IIa”). MtDNA clades within E. blennioides (from Fig. 2 and Appendix 1) are lettered ‘‘A–I”, for reference comparisons.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
8
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
appears either as the sister group to E. gutselli + E. blennius (Fig. 2a– b) or as the sister taxon to Clades I (E. blennioides) and II (E. gutselli + ‘‘E. b. newmanii” from the Tennessee and Hiwassee River drainages) + E. blennius (Fig. 2c, Appendix 1a–b). The cyt b trees (Figs. 2a–c, Appendix 1a–b) all demonstrate that according to the mtDNA evidence, E. b. blennioides is the sole monophyletic subspecies of greenside darter supported from Miller’s (1968) subspecies hypothesis, having 1.00/100% support in the Bayesian and ML analyses. A separate grouping within the greenside darter complex contains individuals from the putative E. b. pholidotum mixed with those from the putative E. b. newmanii (from the Cumberland, White, and Ouachita Rivers); neither are monophyletic taxa. Those two subspecies as hypothesized by Miller (1968) thus are unsupported by the mtDNA data. The greenside darter complex includes six well-supported mtDNA population clades, which occur in different watersheds and are lettered A–F (Figs. 2a–c, Appendix 1a–b). In the scenario depicted by Piller and Bart (2009), in which they claimed that Haponski and Stepien (2008) had rooted to all of the following simultaneously: E. gutselli, E. blennius, E. rupestre, E. rafinesquei, E. zonale, E. variatum, E. camurum, and E. bellum (tree depicted in Fig. 2b), the ingroup relationships remain the same as in Fig. 2a. The difference between the topology of the outgroups on the resultant tree (Fig. 2b) versus that obtained with our original rooting scenario to E. camurum and E. bellum (Fig. 2a) should have made it readily apparent that the rooting scenario claimed by Piller and Bart (2009) was not employed by us. Piller and Bart (2009) thus failed to duplicate our analyses. In both of these rooting scenarios, the ingroup relationships remain the same on the trees. Here we show another analysis in which we omit E. camurum and E. bellum since Piller and Bart (2009, p. 341) claimed that Haponski and Stepien (2008) had ‘‘violated the practices of phylogenetic systematics” by using ‘‘too distantly related outgroups”. We thus root the tree to E. euzonum, E. tetrazonum, and E. variatum following Piller et al. (2008), and also include sequence data from other relatives, e.g., E. lynceum, E. swannanoa, and E. thalassinum (shown in Appendix 1a). The resulting tree (Appendix 1a) shows that the redefined greenside darter complex (Clade I) remains the same as that shown in Figs. 2a–c, again with 1.00/100% support. This tree again depicts the sister group of the redefined greenside darter complex (Clade I) as a clade containing E. gutselli (Clade II) and E. blennius, with 1.00/100% support. The nearest taxon to the two clades (the greenside darter complex – Clade I vs. E. gutselli and E. blennius) appears to be E. rupestre, which has 1.00/99% support. Thus, the present results show that the claim of ‘‘too distantly related outgroups” raised by Piller and Bart (2009) towards Haponski and Stepien (2008) is moot since the different rooting scenarios produce analogous results. Trees based on the complete data set of all cyt b haplotypes, including the intergrade taxa (Fig. 2c, Appendix 2b), show 1.00/ 100% support for our redefined greenside darter complex (Clade I) and are congruent between the rooting scenarios of Haponski and Stepien (2008) (Fig. 2c) and Piller et al. (2008) (Appendix 2b). A monophyletic E. b. blennioides (Clade A) again is the sole subspecies supported of Miller’s (1968) putative subspecies. A cluster containing all putative E. b. pholidotum groups with putative E. b. newmanii (from the Cumberland, White/Arkansas, and Ouachita Rivers); neither taxon is monophyletic. All six original clades (lettered A–F) of Haponski and Stepien (2008) remain well-supported. In addition, an intergrade from the Gasconade River (based on a single haplotype and a single individual from Piller et al. (2008) classified as ‘‘E. b. newmanii E. b. pholidotum”) appears related to Clade C (i.e., E. b. pholidotum from the Osage River). The relationship between the redefined greenside darter complex (Clade I) and its apparent sister group of E. gutselli and the restricted ‘‘E. b. new-
manii” (Clade II) + E. blennius remains the same in both rooting scenarios. Thus, Piller and Bart’s (2009) assumptions regarding taxon inclusion and alternative rooting scenarios are unsupported by the analyses. Our trees in the present investigation thus demonstrate that Piller and Bart’s study (2009) was based on invalid assumptions regarding their choices of taxa, outgroups, and rooting; in contrast, our resultant mtDNA trees are robust, stable, and wellsupported. 3.2. Phylogenetic relationships and resolution of the greenside darter complex based on nuclear DNA We here test nuclear DNA S7 intron 1 sequences in new analyses using Bayesian and ML approaches, uniquely including all available data. Piller et al. (2008) did not include any S7 data for the type subspecies, E. b. blennioides, and thus were unable to address the relationships among Miller’s (1968) subspecies. This was corrected in Haponski and Stepien (2008) with our inclusion of substantially more sequences, and is comprehensively analysed here with nuclear DNA sequences that represent all mtDNA clades, additional analysis approaches, and various rooting scenarios. Our most significant finding is that our redefined greenside darter complex resolved from the mtDNA analyses (Clade I) is highly supported by the nuclear data Bayesian analyses with 0.90–0.99 support (Figs. 3a–b, Appendix 2a–b). Within the complex, the subspecies E. b. blennioides is found in two locations on our nuclear DNA trees, and none of Miller’s (1968) putative subspecies are supported as monophyletic. The second most significant conclusion is, in both the nuclear and mtDNA trees, the Tennessee/Hiwassee River restricted ‘‘E. b. newmanii” clade does not belong to the greenside darter complex (Clade I). Individuals of ‘‘E. b. newmanii” from the Tennessee/Hiwassee Rivers are included (introgressed) with E. gutselli in the mtDNA trees (Clade II) and comprise a separate taxon in the nuclear DNA trees (Clade IIb). With the nuclear DNA data, both E. rupestre and E. zonale appear more closely related to the redefined greenside darter complex (0.83 support) than either is to the Tennessee/Hiwassee River ‘‘E. b. newmanii” clade or to E. gutselli. Supporting its species designation, E. gutselli appears as a separate clade with 1.00/96–98% support in the nuclear DNA trees (Clade IIb on Fig. 3a–b, Appendix 2a–b). Unlike the cyt b data, E. gutselli and the restricted ‘‘E. b. newmanii” clade from the Tennessee/Hiwassee Rivers appear as divergent separate taxa in the nuclear DNA trees (as Clades IIb and IIa; Fig. 3b and Appendix 2b). Results remain the same whether the trees are rooted to E. bellum and E. camurum (Figs. 3a–b) or when those outgroups are omitted and E. euzonum, E. tetrazonum, and E. variatum are the designated root (Appendices 2a–b). The Bayesian analysis tree using our original taxa (without the intergrades) is shown in Fig. 3a. In the ML analysis, the tree nests E. rupestre within the greenside darter complex (see Fig. 2c in Haponski and Stepien (2008)). Results from the Bayesian analysis using all available taxa, including the intergrades, are shown here in Fig. 3b. The ML tree of that scenario nests both E. rupestre and E. zonale within the greenside darter complex, which does not occur in the Bayesian trees. Other than that, the relationships supported from the two phylogenetic analysis approaches used here appear congruent. 4. Discussion 4.1. Redefinition of the greenside darter complex Our findings redefine the greenside darter E. blennioides as comprising Miller’s (1968) hypothesized subspecies E. b. blennioides,
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
the former E. b. pholidotum, and some of the former E. b. newmanii (excepting those from the Tennessee and Hiwassee River watersheds), with high support from both the nuclear and mtDNA data sets. E. b. blennioides is the sole subspecies of Miller (1968) that is monophyletic according to the mtDNA data, and is paraphyletic in the nuclear DNA analyses. The putative subspecies E. b. pholidotum and E. b. newmanii sensu Miller (1968) are polyphyletic according to both mtDNA and nuclear DNA data, and thus neither merit taxonomic recognition. Notably, the former E. b. newmanii is divided among two primary clades (labeled ‘‘I” and ‘‘II” on our trees), one of which is interspersed with E. b. pholidotum as part of E. blennioides and the other is not part of E. blennioides. The former ‘‘E. b. newmanii” from the Tennessee and Hiwassee Rivers is markedly divergent from E. blennioides; this result is strongly supported by both mt and nuclear DNA evidence. The Tennessee River is the type locality of ‘‘E. b. newmanii” per Miller (1968). Therefore, according to the International Code of Zoological Nomenclature (Article 61.1; Ride et al., 1999), the name ‘‘newmanii” belongs to this clade that is the designated type locality for the taxon. Its divergence and monophyly in the nuclear DNA data set may warrant its elevation to species-level, as ‘‘E. newmanii”, which we provisionally use here to differentiate it from E. blennioides. MtDNA analyses indicate a history of introgression between E. gutselli and ‘‘E. newmanii” (now restricted to the Tennessee and Hiwassee Rivers), which appear as wellsupported separate taxa using the nuclear data. Interspecies introgression of mtDNA haplotypes is relatively common in Etheostoma (see Bossu and Near, 2009). We therefore recommend an extensive population genetic study of samples from the distribution of this ‘‘E. newmanii” clade to further assess its morphological and genetic characters. Our investigation supports the species-level distinction of E. gutselli following Etnier and Starnes (2001), shown by its significant divergence from our redefined greenside darter complex in both the nuclear and mtDNA data. Results also indicate that E. blennius shares a close evolutionary history with E. gutselli and with ‘‘E. newmanii”, likely reflecting mtDNA introgression and/or incomplete lineage sorting (see Avise, 2004). 4.2. Phylogenetic evidence in relation to the claims of Piller and Bart (2009) Piller and Bart (2009; p. 341) claimed that Haponski and Stepien (2008) ‘‘violated systematic practices” by purportedly rooting to E. gutselli; however, as we have shown here, that scenario was not practiced by us (i.e., we never rooted to E. gutselli). As stated by Piller and Bart (2009), some studies have shown that outgroups used for rooting should be closely related to the ingroup in question and in cases when the outgroups are too distant, spurious topologies and artifacts due to long-branch attractions may result (Felsenstein, 1978; Wheeler, 1990). However, based on the present analyses, trees resulting from the various rooting scenarios yield identical topologies for the greenside darter complex, with similar support values. Thus, taxa used to root the trees in this study, as well as in the previous studies, appear appropriate for the systematic level investigated and Piller and Bart’s claims are unsupported. Piller and Bart (2009; p. 341) additionally claimed that Haponski and Stepien (2008) selectively used only some taxa from the Piller et al. (2008) study that led to ‘‘fallacious evolutionary relationships” of the greenside darter complex. In truth, Haponski and Stepien (2008, p. 71) included all taxa from Piller et al. (2008), pruning the latter’s identical sequences to use each sequence only once. The present study shows results of tests with and without the intergrade taxa. All of our present and past results identify the identical restricted greenside darter clade with high bootstrap and posterior probability support, as evidenced here by
9
mt and nuclear DNA trees. The statement by Piller and Bart (2009) that Haponski and Stepien (2008) derived inappropriate taxonomic conclusions about E. gutselli is unsupported. We, like Piller et al. (2008), recovered E. gutselli clustering with specimens labeled by Piller et al. (2008) as ‘‘E. b. newmanii” and ‘‘E. b. newmanii E. gutselli” in the mtDNA analyses, as also evident in our trees here. In conclusion, results of the present analyses demonstrate that the allegations raised by Piller and Bart (2009) are unsupported and moot. We recommend that future studies thoroughly document and investigate the morphological characters of the greenside darter complex along with their mt and nuclear DNA characters. 4.3. Nuclear DNA phylogeny results and congruence with the mtDNA phylogeny Reanalyses of all of the nuclear S7 intron data support the definitions of our restricted E. blennioides, along with defining E. gutselli, E. blennius, and a restricted ‘‘E. newmanii” (comprising the Tennessee/Hiwassee River clade). Piller et al. (2008) did not include any E. b. blennioides (the type subspecies) in their nuclear S7 DNA dataset. As Piller and Bart (2009) pointed out, it was essential that all ingroup taxa be represented in order to test the taxonomic validity of Miller’s (1968) subspecies. Excluding E. b. blennioides precluded their ability to test their phylogenetic hypothesis of the composition of E. blennioides. In addition, Piller et al. (2008) included only E. blennius, E. euzonum, E. tetrazonum, and E. variatum in their nuclear data analyses, but none of the other taxa that they used in their mtDNA cyt b study. Our present analysis and our original study (Haponski and Stepien, 2008) also include S7 intron sequence data for E. rupestre and E. zonale, which our results show appear more closely related to the ingroup taxa than were their original outgroups (E. euzonum, E. tetrazonum, and E. variatum). In fact, the ML tree of that scenario nests both E. rupestre and E. zonale within the greenside darter complex, which does not occur in the Bayesian trees; this merits further investigation. 4.4. Diversity of the greenside darter complex Miller (1968) recognized four subspecies, five morphological races, and three zones of intergradation comprising the greenside darter complex E. blennioides. Our findings show that the greenside darter complex likely is more diverse, and contains at least six mtDNA clades in our study. The former E. b. pholidotum and the former E. b. newmanii are discerned to be polyphyletic in all analyses, and thus appear invalid as they do not coincide with Miller’s (1968) definitions. Moreover, both of these former groups lacked morphologically diagnostic characters as determined by Haponski and Stepien (2008). If subspecies names are retained in the future, they will need to be restricted to those parts of the component taxa that are monophyletic, with the names matching those designated as the type localities from Miller (1968) per the International Code for Zoological Nomenclature (Ride et al., 1999). We especially recommend further sampling and careful analyses of the greenside darter from the Green-Barren, Gasconade, Cumberland, Ouachita, and White/Arkansas Rivers, where very limited sampling (in most cases only a single individual) was done by Piller et al. (2008). Focused sampling and accurate morphological identification especially are needed in the Tennessee River system, where ‘‘E. newmanii” and E. gutselli show mitochondrial DNA introgression (Fig. 2c, Appendix 1b) and morphological intergradation (Miller, 1968). Additional nuclear genes should be evaluated and additional data sets are needed. Our earlier study, Haponski and Stepien (2008) also assessed variation at the population level
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
10
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
for the northern taxa; this method should be adopted for the southerly populations.
Appendix 1a
In summary, our nuclear and mtDNA results clearly and congruently support a restricted E. blennioides, and the genetic divergence of E. gutselli and ‘‘E. newmanii” as separate from the redefined greenside darter complex. According to these data and our analyses, their evolutionary relationships remain robust regardless of (1) the genome tested, (2) the phylogenetic analysis approach, (3) the inclusion of the intergrade taxa, and (4) the taxa selected to root the trees. Further resolution of relationships within our redefined greenside darter complex and its relatives should incorporate 20–40 individuals or more per watershed, coupled with thorough morphological analyses and additional nuclear genes.
Bayesian 50% majority rule consensus tree for the original greenside darter mtDNA cytochrome b data set from Haponski and Stepien (2008), which here is rooted to E. euzonum, and E. tetrazonum, and E. variatum from Piller et al. (2008). This tree is congruent with the tree from maximum likelihood analysis (ML). Redundant sequences from Piller et al. (2008) are pruned. Values at nodes = support from Bayesian posterior probability (1.00)/ 1000 bootstrap pseudoreplications in the ML tree; both given as percentages. *100% support. XThe type locality. Parentheses = (N haplotypes/N individuals at that branch). Our redefined E. blennioides is labeled ‘‘Clade I”. Clades within E. blennioides are lettered. Etheostoma gutselli is labeled as ‘‘Clade II”, which appears as the sister taxon of E. blennius, reflecting likely mtDNA introgression; these taxa together form the apparent sister group to Clade I (the greenside darter complex).
Acknowledgments
Appendix 1b
This research was funded by Ohio Sea Grant #R/LR-9PD, USEPA #CR-83281401-0, and NSF GK-12 grant DGE-0742395. Collections were aided by M. Bagley, D. Brandon, A. Callahan, J. Callahan, M. Coburn, D. Carlson, T. Crail, J. Faber, B. Fisher, A. Ford, J. Grabarkiewicz, G. Hogue, M. Jedlicka, T. Marth, D. Neely, B. Porter, C. Smith, W. Starnes, R. Strange, S. Tuckerman, and W. Zawiski. Our collections were made under Michigan and Ohio scientific collection permits. Members of the Great Lakes Genetics Laboratory (GLGL), including J. Brown, R. Lohner, D. Murphy, M. Neilson, and O. Sepulveda-Villet aided field and laboratory work, and provided a valuable sounding board for ideas and improvements. This manuscript significantly benefited from critical comments and suggestions by M. Neilson and two anonymous colleagues (who are prominent scientists specializing in darter phylogenetics and biogeography). This is publication #2010-09 from the Lake Erie Research Center.
Bayesian 50% majority rule consensus tree for the combined mtDNA cytochrome b sequence data set from Haponski and Stepien (2008) and Piller et al. (2008), rooted to E. euzonum, E. tetrazonum, and E. variatum following Piller et al. (2008). This tree is congruent with the tree from maximum likelihood analysis (ML). All intergrade taxa are included and redundant sequences from Piller et al. (2008) are pruned. Values at nodes = support from Bayesian posterior probability (1.00)/1000 bootstrap pseudoreplications in the ML tree; both given as percentages. *100% support. XThe type locality. Parentheses = (N haplotypes/N individuals at that branch). Our redefined E. blennioides is labeled ‘‘Clade I”. Clades within E. blennioides are lettered ‘‘A–I”. Etheostoma gutselli and our redefined ‘‘E. b. newmanii/E. newmanii” are labeled ‘‘Clade II”.
4.5. Summary and conclusions
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Appendix 2a Bayesian 50% majority rule consensus tree from the nuclear S7 intron 1 DNA sequence data based on the original dataset from Haponski and Stepien (2008). Tree is rooted to E. euzonum, E. tetrazonum, and E. variatum. Redundant sequences from Piller et al. (2008) are pruned. Values at nodes = support from Bayes-
11
ian posterior probability (1.00)/1000 bootstrap pseudoreplications in the ML tree; both given as percentages. *100% support. XThe type locality. Parentheses = (N haplotypes/N individuals at that branch). Our redefined E. blennioides is labeled ‘‘Clade I”. MtDNA clades within E. blennioides (from Fig. 2 and Appendix 1) are lettered, for reference comparisons. ‘‘Clade II” denotes E. gutselli.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
12
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Appendix 2b Bayesian 50% majority rule consensus tree from the nuclear S7 intron 1 sequence data, based on the combined Haponski and Stepien (2008) and Piller et al. (2008) greenside darter sequence datasets. All intergrade taxa are included and redundant sequences from Piller et al. (2008) are pruned. Tree is rooted to E. euzonum, E. tetrazonum, and E. variatum. Values at nodes = support from
Bayesian posterior probability (1.00)/1000 bootstrap pseudoreplications in the ML tree; both given as percentages. *100% support. X The type locality. Parentheses = (N haplotypes/N individuals at that branch). Our redefined E. blennioides is labeled ‘‘Clade I”. MtDNA clades within E. blennioides (from Fig. 2 and Appendix 1) are lettered for reference comparisons. ‘‘Clade IIb” denotes E. gutselli and ‘‘Clade IIa” denotes our redefined ‘‘E. b. newmanii/E. newmanii”.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
References Agassiz, L., 1854. Notice of a collection of fishes from the southern bend of the Tennessee River (in the state of) Alabama. Am. J. Sci., 2nd. Ser 17, 297–308. Avise, J., 2004. Molecular Markers, Natural History, and Evolution, second ed. Sinauer Associates Inc., Sunderland, NY. Bossu, C.M., Near, T.J., 2009. Gene trees reveal repeated instances of mitochondrial DNA introgression in orangethroat darters (Percidae: Etheostoma). Syst. Biol. 58, 114–129. Etnier, D.A., Starnes, W.C., 2001. The Fishes of Tennessee. University of Tennessee Press, Knoxville, TN. Felsenstein, J., 1978. Cases in which parsimony and compatibility methods will be positively misleading. Syst. Zool. 27, 401–410. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791. Guindon, S., Gascuel, O., 2003. PhyML: a simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704. . Hall, T.A., 1999. BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Sym. Ser. 41, 95. . Haponski, A.E., Stepien, C.A., 2008. Molecular, morphological, and biogeographic resolution of cryptic taxa in the greenside darter Etheostoma blennioides complex. Mol. Phylogenet. Evol. 49, 69–83.
13
Hasegawa, M., Kishino, H., Yano, T., 1985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J. Mol. Evol. 22, 160–174. Hildebrand, S., 1932. On a collection of fishes from the Tuckasegee and upper Catawba River basins, North Carolina, with a description of a new darter. J. Elisha Mitchell Sci. Soc. 48, 50–82. Holder, M., Lewis, P.O., 2003. Phylogeny estimation: traditional and Bayesian approaches. Nat. Rev. Genet. 4, 275–284. Kimura, M., 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16, 111–120. Lanave, C., Preparata, G., Saccone, C., Serio, G., 1984. A new method for calculating evolutionary substitution rates. J. Mol. Evol. 20, 86–93. Miller, R.V., 1968. A systematic study of the Greenside Darter, Etheostoma blennioides Rafinesque (Pisces: Percidae). Copeia 1968, 1–40. Morrison, C.L., Lemarié, D.P., Wood, R.M., King, T.L., 2006. Phylogeographic analyses suggest multiple lineages of Crystallaria asprella (Percidae: Etheostomatinae). Conserv. Genet. 7, 129–147. Nelson, J.S., Crossman, E.J., Espinosa-Pérez, H., Findley, L.T., Gilbert, C.R., Lea, R.N., Williams, J.D., 2004. Common and Scientific Names of Fishes from the United States, Canada, and Mexico. American Fisheries Society, Special Publication 29, Bethesda, MD. Piller, K.R., Bart, H.L., 2009. Incomplete sampling, outgroups, and phylogenetic inaccuracy: a case study of the greenside darter complex (Percidae: Etheostoma blennioides). Mol. Phylogenet. Evol. 53, 340–344.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017
14
C.A. Stepien, A.E. Haponski / Molecular Phylogenetics and Evolution xxx (2010) xxx–xxx
Piller, K.R., Bart, H.L., Hurley, D.L., 2008. Phylogeography of the greenside darter complex, Etheostoma blennioides (Teleostomi: Percidae): a wide-ranging polytypic taxon. Mol. Phylogenet. Evol. 46, 974–985. Posada, D., Crandall, K.A., 1998. Modeltest: testing the model of DNA substitution. Bioinformatics. 14, 817–818. . (v3.7, 2008). Rafinesque, C.S., 1819. Prodrome de 70 nouveau genres d’animaux découverts dans l’interieur des États-Unis d’Amerique durant l’année 1818. J. Phys. Chym., Hist. Nat. 88, 417–429. Ride, W.D.L., Cogger, H.G., Dupuis, C., Kraus, O., Minelli, A., Thompson, F.C., Tubbs, P.K., 1999. International Code of Zoological Nomenclature, fourth ed. International Trust for Zoological Nomenclature, London, UK. . Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574. . (v3.1.2, 2005).
Sloss, B.L., Billington, N., Burr, B.M., 2004. A molecular phylogeny of the Percidae (Teleostei, Perciformes) based on mitochondrial DNA sequence. Mol. Phylogenet. Evol. 32, 545–562. Song, C.B., Near, T.J., Page, L.M., 1998. Phylogenetic relations among percid fishes as inferred from mitochondrial cytochrome b DNA sequence data. Mol. Phylogenet. Evol. 10, 343–353. Swofford, D.L., 2003. PAUP*: Phylogenetic Analysis Using Parsimony (and Other Methods) 4.0. http://paup.csit.fsu.edu/(v4.0b10, 2003). Wheeler, W.C., 1990. Nucleic acid sequence phylogeny and random outgroups. Cladistics 6, 363–367. Wiley, E.O., 1981. Phylogenetics: The Theory and Practice of Phylogenetic Systematics. John Wiley and Sons Inc., New York, NY.
Please cite this article in press as: Stepien, C.A., Haponski, A.E. Systematics of the greenside darter Etheostoma blennioides complex: Consensus from nuclear and mitochondrial DNA sequences. Mol. Phylogenet. Evol. (2010), doi:10.1016/j.ympev.2010.06.017