2006 SOUTHEASTERN NATURALIST 5(1):85–92 Population Structure of the Tallapoosa Shiner (Cyprinella gibbsi) and the Tallapoosa Darter (Etheostoma tallapoosae) HEATHER M. CONNELLY1,2, CHRISTOPHER R. TABIT1, AND LEOS G. KRAL1,* Abstract - Cyprinella gibbsi (Tallapoosa shiner) is sympatric with Etheostoma tallapoosae (Tallapoosa darter) and both species are endemic to the Tallapoosa River system of Georgia and Alabama. The darter population has been shown to be divided into genetically divergent populations. In this study, mitochondrial ND4L sequences were analyzed for 10 populations of the Tallapoosa shiner from throughout its distribution. Phylogenetic analysis and analysis of molecular variance show that the shiner population is also divided into genetically divergent populations. These can be designated as management units for future monitoring of the species. The distributions of the genetically divergent populations of the shiner and the darter are similar, and indicate that the two species share a common biogeographic history. Introduction Cyprinella gibbsi (Howell and Williams) (Tallapoosa shiner) is endemic to the Tallapoosa River system in northwestern Georgia and eastern Alabama, and this system is tributary to the Alabama River. This shiner, a midwater dweller, is confined to that part of the Tallapoosa River system above the fall line (Howell and Williams 1971). Etheostoma (Ulocentra) tallapoosae Suttkus and Etnier (Tallapoosa darter), a benthic species, is also endemic to the piedmont portion of the Tallapoosa River system (Suttkus and Etnier 1991). Brogdon et al. (2003) have shown that the Tallapoosa darter population is highly structured genetically, and that genetically divergent populations are confined to specific regions of the Tallapoosa system. This structuring of the darter population is not surprising, since species of the subgenus Ulocentra are mostly allopatrically distributed among and within drainages (Bauer et al. 1995, Porter et al. 2002), presumably due to their limited dispersal abilities. Since the Tallapoosa shiner and its sister species, Cyprinella trichroistia (Jordan and Gilbert) (Jordan and Brayton 1878; tricolor shiner) (Broughton and Gold 2000, Mayden 1989), are allopatric in neighboring tributary systems of the Alabama River, it is likely that the Tallapoosa shiner is subject to the same barriers to dispersal as is the sympatric Tallapoosa darter, and thus both may have a similar genetically divergent population structure. 1Department of Biology, University of West Georgia, Carrollton, GA 30118. 2Current address - School of Genome Science and Technology, University of Tennessee/ Oak Ridge National Laboratory, Oak Ridge, TN 37831. *Corresponding author - lkral@westga.edu. 86 Southeastern Naturalist Vol. 5, No. 1 The purpose of this study is to ascertain the degree of genetic structuring present in populations of the Tallapoosa shiner. Significant structure will require that future monitoring of the species be focused on particular population groups to ensure preservation of maximum genetic heterogeneity. Materials and Methods A total of 98 Tallapoosa shiners was collected by seine and electrofishing from 10 locations throughout the distribution of this species (Table 1, Fig.1). The sample size reflects approximate species abundance at each site. Our goal was to obtain at least 10 specimens from each site, but this was not possible despite an increased collection effort at some sites. Individuals were immediately preserved in 95% ethanol, and voucher specimens deposited in the University of West Georgia collection. Genomic DNA was isolated and quantified as previously described (Brogdon et al. 2003). Single stranded conformation polymorphism (SSCP) Table 1. Site numbers (in reference to Figs. 1 and 3), capture locality, and size of sample of Cyprinella gibbsi sampled in this study. Site Collection locality Sample size 1 Tallapoosa River, Haralson County, GA 11 2 Silas Creek, Cleburne County, AL 12 3 Lockhelooge Creek, Cleburne County, AL 11 4 Buck Creek, Carroll County, GA 9 5 Wedowee Creek, Randolph County, AL 6 6 Cornhouse Creek, Randolph County, AL 10 7 Chikasanoxee Creek, Chambers County, AL 16 8 Jay Bird Creek, Tallapoosa County, AL 10 9 Enitachopco Creek, Clay County, AL 10 10 Oakachoy Creek, Coosa County, AL 3 Figure 1. Cyprinella gibbsi sample sites (circles) in the Tallapoosa River system above the fall line (FL). Numbers outside of circles refer to sample sites and numbers inside circles refer to sample size at each site, both as listed in Table 1. Inset shows the extent of the Tallapoosa River system and the range of the Tallapoosa shiner above the fall line (shaded oval). Other abbreviations: TR = Tallapoosa River, LT = Little Tallapoosa River, HR = Harris Reservoir, and ML = Martin Lake. 2006 H.M. Connelly, C.R. Tabit, and L.G. Kral 87 detection of ND4L haplotypes was performed in 20-μl reactions containing 10 ng of genomic DNA, 10-μl Qiagen HotstarTaq Master Mix, 0.5-μM concentrations of primer tsND4LF, 5' AAA ATT GTG GTT TAA GTC CAC GG 3', and primer tsND4LR, 5' AAG ATT AAG GTT TTG TAA GCG GTC 3' under the following temperature profile: 15-minute hot start at 95 oC followed by 35 cycles of 30 seconds at 94 oC, 1 minute at 55 oC, and 2 minutes at 72 oC. The final cycle was followed by a 10-minute incubation at 72 oC. These primers amplify a 312-bp fragment, and enable the detection of variation within 259 nucleotides of the 297 basepair-long ND4L gene. The amplified products were denatured, fractionated by electrophoresis (Sunnucks et al. 2000) and visualized with SYBRgold staining. Representative samples of SSCP-detected ND4L haplotypes were characterized by DNA sequencing. The entire mitochondrial ND4L gene and a portion of the ND2 gene were PCR-amplified in 50-μl reactions containing 50 ng of genomic DNA, 25-μl Qiagen HotstarTaq Master Mix, 0.5-μM concentrations of primer Arg B-L, 5' CAA GAC CCT TGA TTT CGG CTC A 3' and primer NAP2, 5' TGG AGC TTC TAC GTG RGC TTT 3' (Broughton and Gold 2000) under the same conditions as above. Prior to DNA sequencing, PCR products were gel purified utilizing the Qiagen Qiaquick Gel Extraction Kit. Automated sequencing of the ND4L gene was performed at Davis Sequencing, Davis, CA (www.davissequencing.com), utilizing the Arg B-L amplification primer as the sequencing primer. Sequence alignments were carried out with GeneJockey II DNA analysis software (Biosoft). Only the portions of the DNA sequence data that represented the SSCP discernable portion of the ND4L gene were used to define haplotypes. Haplotype cladogram analysis (Templeton et al. 1992) was conducted using TCS version 1.13 (Clement et al. 2000). Statistical analysis of genetic structuring was conducted by analysis of molecular variance (AMOVA) of SSCP-identified haplotypes utilizing ARLEQUIN version 2.000 (Schneider et al. 2000). Results In all populations of the Tallapoosa shiner sampled, a total of 12 haplotypes of the mitochondrial ND4L gene was identified (Table 2). All but two nucleotide variants are synonymous substitutions that do not alter the amino acid sequence. Position 11 variant (haplotypes C and L) results in a substitution that replaces a valine with an alanine. Position 186 variant (haplotypes H and I) results in a conserved amino acid substitution that replaces a leucine with a phenylalanine. The maximum sequence divergence observed among the haplotypes is 2.02%. This is within the range of intraspecies sequence variation. The ND4L gene sequence divergence between the Tallapoosa shiner and the most closely related species (Broughton and Gold 2000), Cyprinella trichroistia (GenBank accession number: AF111249) is about 4%. Analysis of molecular variance of all populations yielded a FST of 0.54 (P less than 0.001). This indicates a significant degree of genetic structuring of the Tallapoosa shiner populations. The distribution of haplotypes among sample 88 Southeastern Naturalist Vol. 5, No. 1 sites is shown in Table 3. The phylogenetic relationship among the haplotypes is represented as a gene genealogy network in Figure 2. When populations are grouped by haplotype composition and geographic distribution and these various groupings are tested by AMOVA, four groups of populations are identified that maximally partition the molecular variance among the populations (Table 4). Group 1 is comprised of populations located in the upper Tallapoosa River north of Harris Reservoir (sample sites 1, 2, and 3), the Little Tallapoosa River (sample sites 4 and 5), and the Tallapoosa River south of Harris Reservoir (sample sites 6 and 7). Groups 2, 3, and 4 are comprised of individual creek populations near Martin Lake (sample sites 8, 9, and 10, respectively). Those haplotypes found either exclusively or at highest frequency in any one of the four groups are shade-coded in Figure 2, and the geographic distribution of these haplotype clusters is shown in Figure 3. In this grouping of populations, 71.25% of genetic variation is among the four groups, 1.00% of the variation is among populations within groups, and 27.74% of the variation is within the populations (FCT = 0.7125, P < 0.01). While not supported by the overall AMOVA analysis, it is possible that population group 1, as described above, may be comprised of two populations. Haplotypes B and D comprise 19% of all haplotypes observed in Table 2. Mitochondrial ND4L haplotypes for Cyprinella gibbsi from the Tallapoosa River system. Only variable sites are shown at their respective base numbers. GenBank accession numbers: AY869699–AY869710. Haplotype 11 51 55 93 102 111 132 156 186 189 195 246 249 A T A C A G G A G A A T T C B T A C A G A A G A G T T T C C A C A G G A G A A T T C D T A C G G G A G A A T T C E T A C A G A A A A G T T T F T A C A G G A G A A C T C G T A C A A G A G A A T T C H T A T A G G A G T A T T C I T A C A G G A G T A T T C J T A C A G G G G A A T T C K T G C A G A A G A A T T C L C A C A G G A G A A T G C Table 3. Distribution of Cyprinella gibbsi mitochondrial ND4L haplotypes at Tallapoosa River system collection sites as listed in Table 1. Site Haplotype (frequency) 1 A(11) 2 A(11) F(1) G(1) 3 A(11) 4 A(6) B(2) D(1) 5 A(6) 6 A(6) B(1) D(1) F(1) H(1) 7 A(10) B(3) I(1) J(1) K(1) 8 B(7) E(3) 9 C(8) D(1) L(1) 10 D(3) 2006 H.M. Connelly, C.R. Tabit, and L.G. Kral 89 samples from the Tallapoosa River tributaries south of the Harris Reservoir and the Little Tallapoosa River (sample sites 4, 5, 6, and 7: population 1A). These haplotypes are absent in samples from the Tallapoosa River and its tributaries north of the Harris Reservoir (sample sites 1, 2, and 3: population 1B). When these two population groupings alone are compared by AMOVA analysis, weak statistical support is obtained for this subdivision (FCT = 0.085, P = 0.0469 ± 0.0056). Discussion At least four genetically divergent populations of the Tallapoosa shiner have been identified in this study. These populations can be designated as separate Management Units (MUs) by the criteria that these populations differ significantly in mitochondrial allele frequencies, which indicates very low levels of gene flow between populations (Moritz 1994a, b). Table 4. Analysis of molecular variance (AMOVA) among populations of the Tallapoosa shiner. Sum of Variance % of df squares components variation Among groups 3 34.361 0.86136 Va 71.25 Among populations within groups 6 2.772 0.01213 Vb 1.00 Within populations 87 29.176 0.33536 Vc 27.74 Total 96 66.309 1.20885 Fixation indices FSC 0.03490 P = 0.0000 ± 0.0000 FST 0.03490 P = 0.2053 ± 0.0118 FCT 0.03490 P = 0.0088 ± 0.0000 Figure 2. Gene genealogy network of Tallapoosa shiner ND4L haplotypes shown in Table 2. Shading represents geographic prevalence of haplotypes as listed in Table 3. The diameter of the circles is proportional to the relative frequency of each haplotype among all of the haplotypes present in the entire Tallapoosa River system. 90 Southeastern Naturalist Vol. 5, No. 1 The Tallapoosa River system north and south of the Harris Reservoir delineates the geographical boundary of a historically large interbreeding population that can be considered as one MU (but see discussion below). Potential loss of a small number of populations in western Georgia within this MU as a result of habitat degradation (mostly siltation and impoundments; Etnier 1997) due to urbanization is of lesser conservation concern since these populations would not likely be genetically unique. The portion of the Tallapoosa River system in the area of Martin Lake contains at least three genetically divergent populations, each of which can be considered a separate MU. It should be noted that one of these populations (Oakachoy Creek—sample site 10) is defined only by a sample size of 3, and thus the designation of this population as a MU is tentative. Future census data of the Tallapoosa shiner should be interpreted in terms of stability of the MUs identified in this study. Particular attention should be paid to the MUs present in individual streams draining into the Martin Lake portion of the Tallapoosa River. These genetically distinct populations could potentially harbor sequestered components of genetic diversity not present in the larger interbreeding population north of Martin Lake. The genetic population structure of the Tallapoosa shiner is similar to that of the Tallapoosa darter (Brogdon et al. 2003). In particular, in the Martin Lake area, individual creek populations of both the shiner and the darter are nearly or completely homogeneous for individual alleles of the Figure 3. Sample sites (circles) and associated haplotypes of Cyprinella gibbsi in the Tallapoosa River system above the fall line (FL). Numbers refer to sample sites as listed in Table 1. Shades of gray within circles represent frequency of haplotypes as shadecoded in Figure 2. Inset shows the extent of the Tallapoosa River system and the range of the Tallapoosa shiner above the fall line (shaded oval). Other abbreviations: TR = Tallapoosa River, LT = Little Tallapoosa River, HR = Harris Reservoir, and ML = Martin Lake. 2006 H.M. Connelly, C.R. Tabit, and L.G. Kral 91 mitochondrial loci studied. This suggests that the main channel of the southern portion of the Tallapoosa River (historically) and Martin Lake (over the last 76 years) have served as a barrier to migration in both of these species. It is likely, therefore, that other individual creek populations of the shiner within this portion of the Tallapoosa River drainage are also genetically unique, and should be monitored as separate MUs. Further comparison of the genetic structures of the Tallapoosa shiner and the Tallapoosa darter populations indicates an additional similarity. For each species, a historically interbreeding population has been identified by mitochondrial allele distribution to span both the Little Tallapoosa River system north of the Harris Reservoir and the Tallapoosa River system south of the Harris Reservoir. This suggests that the main river channel upstream from Martin Lake in conjunction with the Little Tallapoosa River tributary do not have water flow rates high enough to act as a migration barrier over time for both species. While Harris Reservoir probably is now a migration barrier, its existence for only 20 years would not have affected the current distribution of mitochondrial haplotypes. Different population structures of these two species are observed in the upper Tallapoosa River north of the Harris Reservoir (Brogdon et al. 2003). The darter population in the upper Tallapoosa River is genetically unique compared to the population in the Little Tallapoosa River and the Tallapoosa River south of the Harris Reservoir. This indicates that the confluence of the Upper Tallapoosa River and the Little Tallapoosa River served as a migration barrier for the Tallapoosa darter prior to the construction of Harris Reservoir. In contrast, the shiner populations compared in these same waters are not unique. The A haplotype is the predominant haplotype in all of the Tallapoosa shiner populations in this entire portion of the Tallapoosa River system. However, examination of the distribution of shiner haplotypes reveals that some of the minor haplotypes present in the population range of the Little Tallapoosa River and the Tallapoosa River south of Harris Reservoir do not occur in the populations of the upper Tallapoosa River. While statistical support for genetic differentiation between these populations is not strong, it suggests that the confluence of the Upper Tallapoosa River and the Little Tallapoosa River have served as a barrier to migration for the shiner as well. In conclusion, at least four genetically divergent populations of the Tallapoosa shiner have been identified that can be designated as MUs for future monitoring. From analysis of the distribution of the genetically distinct populations identified in this study, it is likely that populations in other creeks in the Martin Lake area are also genetically distinct. The haplotype distribution of these populations should be determined. The genetic population structures of the Tallapoosa shiner and the Tallapoosa darter suggest that the two species have a common biogeographic history. This is interesting since these species occupy different niches, with the darter being a benthic species and the shiner a mid-water species. Due to their strongly different life histories, it is unexpected that both the shiner and the darter would be impeded by the same barriers to dispersal. 92 Southeastern Naturalist Vol. 5, No. 1 Acknowledgments The authors thank Stephen Brogdon, William Bouthillier, Lahna McGee, Richard Childers, and Heidi Banford for help with stream collections. We thank Heidi Banford, Rudolf Arndt, and four anonymous reviewers for providing valuable comments on the manuscript. This study was supported by Faculty Research Grant funds obtained from the University of West Georgia. Literature Cited Bauer, B.H., D.A. Etnier, and N.M. Burkhead. 1995. Etheostoma (Ulocentra) scotti (Osteichthyes: Percidae), a new darter from the Etowah River system in Georgia. Bulletin of the Alabama Museum of Natural History 17:1–16. Brogdon, S.M., C.R. Tabit, and L.G. Kral. 2003. Population structure of the Tallapoosa darter (Etheostoma tallapoosae). Southeastern Naturalist 2:487–498. Broughton, R.E., and J.R. Gold. 2000. Phylogenetic relationships in the North American cyprinid genus Cyprinella (Actinopterygii: Cyprinidae) based on mitochondrial ND2 and ND4L gene sequences. Copeia 2000:1–10. Clement, M., D. Posada, and K. Crandall. 2000. TCS: A computer program to estimate gene genealogies. Molecular Ecology 9:1657–1660. Etnier, D.A. 1997. Jeopardized southeastern freshwater fishes: A search for causes. Pp. 87–104, In G.W. Benz and D.E. Collins (Eds.). Aquatic Fauna in Peril. Special Publications 1, Southeast Aquatic Research Institute, Lenz Design and Communication, Decatur, GA. 554 pp. Howell, W.M., and J.D. Williams. 1971. Notropis gibbsi, a new cyprinid fish from the Tallapoosa River system in Alabama and Georgia. Copeia 1971:55–64. Jordan, D.S., and A.W. Brayton. 1878. Contributions to North American ichthyology. No. 3. A. On the distribution of the fishes of the allegany region of South Carolina, Georgia, and Tennessee, with descriptions of new or little known species. Buletin of the US National Museum 12:1–95. Mayden, R.L. 1989. Phylogenetic studies of North American minnows, with emphasis on the genus Cyprinella (Teleostei: Cypriniformes). University of Kansas, Museum of Natural History, Miscellaneous Publication 80:1–189. Moritz, C. 1994a. Defining “evolutionarily significant units” for conservation. Trends in Ecology and Evolution 9:373–375. Moritz, C. 1994b. Applications of mitochondrial DNA analysis in conservation: A critical review. Molecular Ecology 3:401–411. Porter, B.A., T.M. Cavender, and P.A. Fuerst. 2002. Molecular phylogeny of the snubnose darters, subgenus Ulocentra (genus Etheostoma, family Percidae). Molecular Phylogenetics and Evolution 22:364–374. Schneider, S., D. Roessli, and L. Excoffier. 2000. Arlequin: A software for population genetics data analysis. Ver 2.000. Genetics and Biometry Lab, Department of Anthropology, University of Geneva, Geneva, Switzerland. Sunnucks, P., A.C.C. Wilson, L.B. Beheregaray, K. Zenger, J. French, and A.C. Taylor. 2000. SSCP is not difficult: The application and utility of single-stranded conformation polymorphism in evolutionary biology and molecular ecology. Molecular Ecology 9:1699–1710. Suttkus, R.D., and D.A. Etnier. 1991. Etheostoma tallapoosae and E. brevirostrum, two new darters, subgenus Ulocentra, from the Alabama river drainage. Tulane Studies in Zoology and Botany 28:1–24. Templeton, A.R., K.A. Crandall, and C.F. Sing. 1992. A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics 132:619–633.