Laserfiche WebLink
<br /> <br />1328 <br /> <br />dams and sires. These parameters have not been <br />assessed to date and are essential in order to determine <br />the potential for a Ryman-Laikre effect on the wild red <br />drum population in Texas waters. <br />In this study, we used (1) parentage data based on <br />progeny produced from 13 separate spawning events <br />occurring over a 2-week period in a TPWD hatchery <br />during the spring of 2002 and (2) hatchery spawning <br />and release records over the 2003 spawning season to <br />estimate the average effective size of a single spawn <br />and of a release of hatchery-reared fingerlings into the <br />wild. The average reduction in effective size per spawn <br />was estimated by identifying genetically the number of <br />dams and sires and the number of offspring generated <br />by each dam X sire combination contributing to a <br />spawn. We then used simulation to estimate the <br />average effective size of a released population <br />comprised of mixes of progeny from different spawns. <br /> <br /> <br />Methods <br /> <br />A total of 45 adult red drum broodfish (27 dams and <br />18 sires) maintained in nine 13-m3 spawning tanks at <br />the TPWD Marine Development Center (MDC) in <br />Flour Bluff were used in the study. Broodfish (dams <br />and sires) were obtained by TPWD personnel from the <br />wild red drum population offshore of the south Texas <br />coast. Each spawning tank putatively contained three <br />dams and two sires. Temperature and photoperiod were <br />manipulated following a 150-d maturation cycle <br />(McCarty 1987) in order to achieve spontaneous <br />spawning. Approximately 15,000-30,000 offspring <br />were sampled randomly from each of 13 separate <br />spawning events occurring at night over a 2-week <br />period (11-24 April) in the spring of 2002. Fertilized <br />(buoyant) eggs were collected at the effluent of each <br />spawning tank and incubated separately for about n h <br />under conditions described in Henderson-Arzapalo <br />(1987). Newly hatched larvae from each spawn were <br />transferred to separate 1- or 2-acre, prefertilized ponds <br />(Colura 1987). Harvest of ponds was conducted 45-60 <br />d postfertilization, when fingerlings had reached an <br />average length of approximately 30mm. A random <br />sample of 125 fingerlings from each of the 13 <br />spawning events were placed individually into labeled <br />cryopreservation (Nunc) tubes and frozen in liquid <br />nitrogen for subsequent genetic analysis. <br />Genomic DNA was extracted from caudal fin tissue <br />of all broodfish via standard phenol-chloroform <br />procedures, as described in Gold and Richardson <br />(1991). Genomic DNA from all experimental progeny <br />also was extracted from caudal fin tissue but using an <br />alkaline lysis method (Saillant et al. 2002). A small <br />piece (-2 mm3) of fin from each individual was <br />digested for 2 h at 650C in 50 ~L of a sodium <br /> <br />GOLD ET AL. <br /> <br />hydroxide solution (200 mM). The fin~] pH of the <br />solution was adjusted to 8 by adding 50 ~L tris-HCI <br />(200 mM, pH 8.0). <br />Genotypes at four microsatellite loci (Socl9, Soc85, <br />Soc402, and So(428) were obtained via polymerase <br />chain reaction (PCR) amplification of genomic DNA <br />from all broodfish and all experimental offspring. <br />Details, including PCR primer sequences, annealing <br />temperatures, and observed or expected heterozygosity <br />for these four microsatellites may be found in SailIant <br />et al. (2004). One of each pair of PCR primers was <br />fluorescently labeled with one fluorescent dye of set D <br />(Applied Biosystems, Foster City, California): 6-Fam <br />for SocJ9 and Soc85 and Hex for Soc402 and Soc428. <br />Polymerase chain reactions were performed in I O-~L <br />volumes containing I ilL (25 ng) of genomic DNA, I <br />~L of lOX reaction buffer (500 mM KCI, 100 mM tris- <br />HCI [pH 9.0], 1% TritonX-lOo), 1.5 mM MgCI2, 2.0 <br />mM of each deoxynucleotide triphosphate, 5 pM of <br />each primer, 0.5 units of Taq DNA polymerase (Gibco <br />BRL). Thermal cycling was carried out as follows: <br />initial denaturation at 940C for 3 min followed by 35 <br />cycles consisting of 30-s denaturation at 940C, 45-s <br />annealing at the optimized annealing temperature, 1- <br />min extension at noc, with a final extension of 10 min <br />at 720C. Polymerase chain reaction products were <br />loaded on 5% Long Ranger (Cambrex) single-pack <br />gels and electrophoresed on an ABI PRISM 377 DNA <br />automatic sequencer (Applied Biosystems). All gels <br />were analyzed using Genescan Analysis 3.1.2 (Applied <br />Biosystems); allele-calling was performed with Geno- <br />typer software, version 2.5 (Applied Biosystems). <br />Genotypes at each microsatellite for each individual <br />(broodfish and offspring) were scored and entered into <br />a database. Assignment of offspring to an individual <br />dam and sire based on the genotypes at the four <br />micro satellites was implemented via the program <br />Probmax version 1.2 (Danzmann 1997; available at <br />http://www.uoguelph.ca/-rdanzman/software/ <br />PROBMAX/). Parentage assignment was unequivocal <br />in all cases. <br />The genetic effective population size (N) of a <br />released population is influenced in part by (1) the <br />number of dams and sires in a spawning tank <br />contributing to a spawn and (2) the variation in family <br />size among mating (dam X sire) combinations. <br />Following Crow and Kimura (1973), the value of Ne <br />for the former was estimated from the equation <br /> <br />4NdNs <br />Ne = (Nd +N.,)' <br /> <br />(I) <br /> <br />.If .. <br /> <br />respectively, ContI <br />based on the nu <br />produced from e, <br />accounts for une <br />offspring. This va <br />equation <br /> <br />where Ned and Nes <br />and sires, respectivl <br />for Ned and Nes " <br />(Lacy 1989) <br /> <br />where nf and nm aJ <br />respectively, that <br />represents the prop <br />each dam or sire to <br />The genetic effec <br />also is influenced <br />progeny from differ, <br />spawns contribute to <br />this, we used record <br />2003 to generate : <br />mixtures of spawns <br />mixed and the num1 <br />that were transferred <br />prerelease grow out. <br />spawn during the nig <br />400,000 progeny (r <br />spawning tank are cc <br />to prefertilized grow- <br />about 150,000 per re <br />from individual gro\\ <br />populations are del <br />night's spawn. Duri: <br />from individual spav <br />transferred to indivi( <br />from one to seven; t] <br />varied from about 20 <br />a bootstrap resampi <br />estimate the distributi <br />size of 10,000 rele <br />empirical distributiOl <br />number of progeny p <br /> <br /> <br />where N is the genetic effective population size and Nd <br />and N e represent the number of dams and sires, <br />s <br />