Environmental biology of fishes, tập 90, số 3, 2011

Environ Biol Fish (2011) 90:211–222 DOI 10.1007/s10641-010-9733-7 Population genetic structure, diversity and stocking effect of the oriental weatherloach (Misgurnus anguillicaudatus) in an isolated island Yuichi Kano & Katsutoshi Watanabe & Shin Nishida & Ryo Kakioka & Chris Wood & Yukihiro Shimatani & Yôichi Kawaguchi Received: 14 April 2010 / Accepted: 12 October 2010 / Published online: 19 November 2010 # Springer Science+Business Media B.V 2010 Abstract Genetic endemism of island organisms and the threat to such organisms provided by artificially introduced genes are aspects of major interest in evolutionary and conservation studies of fishes In this paper the genetic population structure of the oriental weatherloach, Misgurnus anguillicaudatus, in Sado Island of Japan was elucidated by phylogeographic analysis based on partial mitochondrial control region sequences The specimens were sampled at 62 sites in Sado Island and 14 sites on the mainland close to the island We found various haplotypes of different origins, most of which had already been reported from the mainland and other places of Japan This suggests that the loach has been historically introduced to the island from various regions of Japan Of the 62 sites on the island, cultured/nonnative individuals were confirmed to have been stocked at eight specific sites for feeding of re-introduced Japanese crested ibis (Nipponia nippon) By a Mantel test, geographical and genetic distances were not significantly correlated among the local populations in Sado Island However a significant correlation was found when the eight stocked local populations were excluded from the analysis This implied that the genetic distribution pattern of the loach on the island has been disturbed by the stocking In addition, the nucleotide diversity values of stocked local populations were significantly higher than those of other local populations, also a likely outcome of the stocking In conclusion, the loach on Y Kano (*) : Y Shimatani Department of Urban and Environmental Engineering, Graduate School of Engineering, Kyushu University, Motooka, Nishi-ku, Fukuoka 819-0395, Japan e-mail: kano@species.jp C Wood Department of Biology, Faculty of Sciences, Kyushu University, Higashi-ku, Fukuoka 812-8581, Japan K Watanabe : R Kakioka Department of Zoology, Division of Biological Science, Graduate School of Science, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan S Nishida Department of Biodiversity Sciences, Graduate School of Social and Cultural Studies, Kyushu University, Motooka, Nishi-ku, Fukuoka 819-0395, Japan C Wood College of Life Sciences, Zhejiang University, Zijingang Campus, 338 Yuhangtang Road, Hangzhou 310058 Zhejiang Province, People’s Republic of China Y Kawaguchi Laboratory of Ecosystem Management, Division of Ecosystem Design, Institute of Technology and Science, The University of Tokushima, 2-1 Minami-Josanjima, Tokushima 770-8506, Japan 212 the island likely had their origins in multiple historical introductions and colonizations, where more recent stocking for the ibis has caused further genetic disturbance to their local populations Keywords Genetic disturbance Japanese crested ibis Mantel test Mitochondrial DNA Reintroduction Sado Island Introduction For wild organisms the isolation of species in islands often results in a higher likelihood of endemism where original species or subspecies are sometimes generated (Grant 1998; Avise 2000) This holds true for the freshwater environment where island freshwater endemism is notable in many geographical regions For example the islands of the Greater Antilles (including Cuba, Jamaica, Hispaniola and Puerto Rico) are noted to each have one to several endemic species of freshwater crabs (Cook et al 2008) From New Zealand comes a similar report of an endemic species of freshwater fish Galaxias gollumoides in Stewart Island (McDowall and Chadderton 1999) Possibly most remarkably, a United Nations Environment Programme Report on Madagascar highlighted that, of the island of Madagascar’s 140 known species of freshwater fish, 93 (66% at a species level) were endemic to the island (UNEP 2010) Such taxonomic or genetic endemism is, however, sometimes disturbed by artificially introduced genes Human-mediated genetic disturbance is now a serious issue in wild animal/plant conservation (Hindar et al 1991; Rhymer and Simberloff 1996; Beaumont et al 2001; Randi and Lucchini 2002; Larsen et al 2005) In fish, one fairly extensively studied species that has been investigated for the genetic impact of the introduction of non-indigenous fish-farm bred individuals is the cast of the brown trout Salmo trutta Results were mixed Studies in two countries, Norway (Skaala et al 1996) and Spain (Cagigas et al 1999; Almodovar et al 2001), considered that such introductions had indeed resulted in a significant genetic impact on wild populations However other studies from Norway (Heggenes et al 2002) and Portugal (Santos et al 2006) conversely argued that wild populations of the trout had not been genetically influenced by the artificially introduced individuals Environ Biol Fish (2011) 90:211–222 In a Japanese example the main reason of drastic diminution of the Japanese rosy bitterling Rhodeus ocellatus kurumeus was elucidated to be due to their suffering hybridization with non-indigenous rosy bitterlings Rhodeus ocellatus ocellatus which had been introduced from the Eurasian continent (Kawamura et al 2001) Prevention of releases of domesticated or exogenous individuals, whether intentional or accidental, is now universally recognized as an essential element in nature conservation programs (Allendorf et al 2001) The oriental weatherloach, Misgurnus anguillicaudatus (Cantor 1842), is a one of the most abundant fish species found in traditional paddy fields in Japan and other East Asian regions (Saitoh et al 1988; Tanaka 1999) The loach typically lives in ditches and streams around the paddy fields, migrating into the water-laden paddies to lay its eggs in the summer (Saitoh et al 1988; Tanaka 1999) Previous genetic studies elucidated the phylogeny and phylogeography of the loach in Japan by analysis of mitochondrial DNA (mtDNA), revealing the existence of the two major clades, clades A and B (Morishima et al 2008; Koizumi et al 2009) Clade A is mostly distributed on northern part of Japan and is closely-related to Misgurnus fossilis and Misgurnus mizolepis that live in Europe and China, respectively (Morishima et al 2008; Koizumi et al 2009) Clade B can be further divided into two subclades: subclade B1 is widely distributed all over Japan and seems to be indigenous to Japan Subclade B2 is distributed around Kanto region (middle of Japan) and haplotypes of the same lineage have been found in China (Morishima et al 2008; Koizumi 2009) Clade A is genetically closer to related species such as M fossilis and M mizolepis as apposed to Clade B, and thus the two clades were suggested as different species (Morishima et al 2008; Arias-Rodriguez et al 2009) On the other hand natural triploid and asexually reproducing clonal diploid loaches have been reported at low rates in Japan (0–16%; average: 1.4%) (Zhang and Arai 1999) These were indicated to be due to abnormal meiosis originating from the intercross between different lineages (Morishima et al 2008) With the exceptions of Japan’s four main islands, Sado Island is the second largest island in Japan On Sado Island rice farming has a long history and paddy fields are still spreading throughout the lowlands of the island The oriental weatherloach is thus one of Environ Biol Fish (2011) 90:211–222 the main freshwater fishes on the island (Honma 1961) On the other hand, Sado Island has been famous for being one of the last wild habitats for the Japanese crested ibis (Nipponia nippon Temminck 1835) Hence Sado Island was the site chosen for ibis reintroduction and the species was reintroduced to the island in a state-level restoration program in autumn 2008 and 2009 To aid the bird’s survival in the wild, many restoration initiatives have been implemented on the island In particular, artificial biotopes and organic-farmed paddy fields have been provided in various locations of the island to provide enhanced feeding opportunities for the ibis The preferred food type for the ibis is loach (Chikatsuji 2002; Tei 2007), and therefore cultured individuals were sometimes stocked in the biotopes and paddy fields in Sado Island (Kano et al 2010) This had previously been the case in the Chinese restoration where such stocking was considered essential to aid the recovery of the crested ibis in central China Therefore, for the 20 years since restoration began in China, loach have been purchased every spring and placed into the rice fields to help breeding ibis raise their chicks Even so, in China, the expanding population still faced food shortage problems (Xi et al 1997; Wood et al 2010) Furthermore, there remains a concern for such indiscreet stocking, which had a risk to cause genetic/ecological disturbances in the loach (Koizumi et al 2009) Previously, Koizumi (2009) noted that samples from Sado Island belonged to the subclade B1 However further detailed information, such as sampling places, frequencies and sequences, were not detailed in this study Thus it is necessary to further clarify the detail information regarding the population genetic aspects of the loach in Sado Island As large genetic divergences within freshwater fish species were found in various other species (Avise 2000; Watanabe et al 2006) and such divergences are likely to be accelerated in isolated habitats like islands (Grant 1998), there is a possibility that the loach in the island has some evolutionary/genetic uniqueness, which should therefore be protected as an evolutionary significant unit (Waples 1991) On the other hand, Honma (1961) indicated that most of the freshwater fishes in the island, including the loach, had been artificially introduced from the mainland by humans However Honma’s study failed to provide significant evidence or strong supporting details for this scenario 213 In this paper we determined 838 bp of control region sequences on the mitochondrial DNA (mtDNA) of 295 individuals, which were sampled from throughout Sado Island and from adjacent areas of the mainland These sites included areas where non-native individuals had been known to have been intentionally stocked Based on the results obtained from phylogeographic/population genetic analyses, we discuss the genetic identity of the loach in the island, as well as the effects of the loach stocking associated with ibis reintroduction Materials and methods Sampling and DNA analysis Sado Island (area: 855 km2; coastline: 277 km) is located in the Sea of Japan, 40 km off the coast of the mainland of Niigata Prefecture in Japan (Fig 1) The Island is roughly divided into three regions, namely, Osado, Kuninaka and Kosado (Fig 1) The Osado region is characterized by high mountains and small rural communities interspersed with paddy fields along the coastline The Kuninaka region, by contrast, is flatland with paddy fields extending over almost the entire region In the Kosado region low mountains and shallow sloping valleys are found throughout, with rural communities and local towns scattered in the mountains and along the coastline The specimens were collected at 62 paddy field sites on the island and 14 similar sites in the Niigata region of the mainland (Fig 1), in 2–15 July 2008 and 1–4 June 2009 In each site, to individuals were net captured within a 50 m2, and preserved in absolute ethanol for mtDNA analysis Simultaneously, information of any loach stocking at the sampling sites was collected through local records, interviews with people and government contacts In total, 295 specimens were used for the DNA analysis Using DNA extraction kits (AquaPure Genomic DNA Isolation Kit; BioRad Laboratory, Hercules, CA, USA), total genomic DNA was isolated from a piece of the pectoral fin of the loach Polymerase chain reaction (PCR) amplification was carried out using the primer pair of L15923 (5′TTAAAGCATCGGTCTTGTAA-3′) (Iguchi et al 1997) and 12SAR-H (5′-ATARTRGGGTATC TAATCCYAGTT-3′) (Martin et al 1992) The PCR 214 Environ Biol Fish (2011) 90:211–222 Fig Map of Sado Island and adjacent Niigata region in the mainland of Japan, and haplotype compositions of the loach Misgurnus anguillicaudatus local populations A sixty degree sector of a pie chart indicates one individual (full circle: individuals) with its black, white and grey patterns corresponding to the clades/subclades A, B1 and B2, respec- tively (see also Fig 2) The loach stocking was conducted at the points flagged with the star sign (S1–S8); black, white and grey stars indicate that the stocked individuals were introduced from Sado Island, mainland and unknown places, respectively (see also Fig 3) The grey pattern on the map shows the land use of paddy fields as potential loach habitat condition consisted of 30 cycles of denaturation (72°C, 15 s), annealing (47°C, 15 s), and extension (72°C, 30 s) on a PC808 thermal cycler (ASTEC, Fukuoka, Japan) After purifying the PCR products by treatment with ExoSAP-It (USB Corp., Cleveland, OH, USA), they were sequenced on an automated DNA sequencer (Applied Biosystems 3130xl Genetic Analyzer; Applied Biosystem, Foster City, CA, USA) using amplification primers and the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) were edited with MEGA4 (Tamura et al 2007) and aligned manually (838 bp of the 3′half of the mtDNA control region) The nucleotide sequences were deposited in DDBJ/EMBL/GenBank (Accession numbers: AB543940–AB543971) The information of the local populations was also deposited in GEDIMAP (http:// gedimap.zool.kyoto-u.ac.jp; Watanabe et al 2009) (Population ID: P1087–P1162) Phylogenetic reconstructions were carried out using the following three methods: maximum likelihood (ML) in PhyML ver 3.0 (Guindon and Gascuel 2003), neighbor-joining (NJ) (Saitoh and Nei 1987) in MEGA ver 4.0.2 and the Bayesian method with MrBayes ver 3.1.2 (Ronquist and Huelsenbeck 2003) To identify the best model of evolutionary changes for the ML and Bayesian analyses, the data set was Phylogenetic analysis The phylogenetic analysis was conducted using the overlapped part of our sequences above and those published in Morishima et al (2008) The sequences Environ Biol Fish (2011) 90:211–222 subjected to the Akaike information criterion (AIC; Akaike 1974) and was calculated using MrModeltest ver 2.3 (Nylander 2004); which is a modified version of Modeltest version (Posada and Crandall 1998) The GTR (Yang 1994) + I + Γ was selected The Tamura-Nei (Tamura and Nei 1993) model was applied for NJ method, because the GTR model is not mounted in the MEGA program ML heuristic searches were performed using an SPR (subtree pruning and regrafting) search from 10 random starting trees with four rate categories Bootstrap analyses were performed with 10,000 replicates for NJ and 100 replicates for ML trees MrBayes uses a MCMCMC algorithm that runs four Markov chains simultaneously The Markov chains began from a random tree and ran for 13,000,000 generations, with sampling every 100 generations, to yield a posterior probability distribution The first 10,001 samples were discarded as burnin The appropriateness of burn-in values and the convergence of chains were evaluated using TRACER 1.5.3 (http://tree.bio.ed.ac.uk/software/tracer/) The remaining trees after burn-in were used to generate a 50% majority rule consensus tree The net average distances between clades A and B, clade A and Misgurnus spp (M mizolepis and M fossilis), clade B and M spp., and subclades B1 and B2 are also computed by MEGA Hierarchical genetic structure To know the genetic structure of the loach in Sado Island and Niigata, hierarchical analysis of molecular variance (AMOVA: Excoffier et al 1992) was performed using Arlequin ver 3.1 (Excoffier et al 2005) The variations were partitioned into three levels: among geographical groups (Sado Island and Niigata), among local populations within groups, and within local populations In this analysis, clade A and B were treated separately as different biological units (Morishima et al 2008) The Tamura-Nei model was used as the genetic distance Statistical tests of Φ-statistics were performed at 10 000 permutations Because individuals of clade A were not obtained from Niigata, this analysis was done for the clade B data Genetic isolation by distance A mantel test was conducted for clade B in both Sado Island and Niigata to test the genetic isolation by 215 distance The geographical distance (direct distance) and genetic distance (DA with Tamura-Nei distance) were calculated by ArcGIS 9.3 (ESRI Japan, Tokyo) and MEGA4, respectively, for all possible pairs of the local populations within the groups The Mantel test was implemented using the Arlequin ver 3.1 with 10,000 permutations For Sado Island, the test was further implemented by eliminating the local populations that suffered any known stocking event (S1–S8) in order to isolate such an effect The significance level of P values was compensated by the false discovery rate (FDR) (Benjamini and Hochberg 1995) The test was not conducted for clade A because of the deficiency of data Genetic identity of stocked populations We tried to clarify the genetic identity of the eight local populations in stocked sites by comparing other genetic groups (Sado Island and Niigata) In each clade (A and B), the individuals were pooled together into the Osado, Kuninaka, Kosado or Niigata groups, with the exception of individuals from the stocked sites which were grouped independently Pairwise DA values were calculated to estimate the genetic distance between the groups by MEGA4 According to this genetic distance data, the neighbor-joining trees among the regional groups were constructed using Phylip ver 3.68 (Felsenstein 2010) for respective clades The net average distances between meta-groups Sado (Osado, Kuninaka, Kosado and S7) and Mainland (Niigata, S2, S3, S4 and S8), and between Osado and Kuninaka, Osado and Kosado, and Kuninaka and Kosado were calculated by MEGA in respective clades Nucleotide diversity and stocking effect We examined a hypothesis that the nucleotide diversity (π) would be higher in the places where the stocking had been unambiguously implemented, as in such cases the original and introduced haplotypes would be likely to be mixed together The values of nucleotide diversity of clade B within the local populations were obtained by the MEGA4 with the Tamura-Nei distance method The diversity values were compared between stocked (S1–S8) and other local populations using the Mann–Whitney U-test 216 Table Stocking information of the oriental weatherloach (Misgurnus anguillicaudatus) in Sado Island Environ Biol Fish (2011) 90:211–222 Site Abundance Period Source Note S1 >2,000 2007–2008 Unknown Purchased from a local supermarket S2 >10,000 1970s Mainland Purchased from a fish farm in the mainland S3 2,000–3,000 2006–2007 Mainland Purchased from a fish farm in the mainland S4 ca 5,000 2003–2006 Mainland Purchased from a fish farm in the mainland S5 >5,000 2006–2009 Unknown S6 300–400 2007 Unknown S7 30–50 2007 Sado Island Captured in Kuninaka Region S8 ca 4,000 2004–2007 Mainland Purchased from a fish farm in the mainland The test was not conducted for clade A because of the deficiency of data Results Loach stocking The loach stocking was confirmed to have been conducted on at least sites in Sado Island (Fig 1; S1–S8) Details are shown in Table There were no formal records regarding the stocking and we had no other option other than to depend on local people’s memory The exact abundance, period and source of the stocking therefore, remained rather ambiguous Phylogenetic analysis Analyses based on ML, NJ and Bayesian method yielded very similar topologies (Fig 2), which were approximately consistent to the result of Morishima et al (2008) Four haplotypes of clade A were revealed, all found exclusively on Sado Island None of these haplotypes were identical to any from the Niigata region or to those reported by Morishima et al (2008) However, the four haplotypes were not monophyletic but were scattered in a major subgroup of clade A As for clade B, 25 haplotypes were found in Sado Island, of which 19 haplotypes were identical to those obtained from Niigata region or those in the study of Morishima et al (2008) On the other hand, 15 haplotypes were found from the Niigata region, of which 12 and 10 haplotypes were identical to those obtained from Sado Island and Morishima et al (2008), respectively As in the case of clade A, the haplotypes of clade B on the island were also widely scattered throughout the tree without any apparent cohesiveness (Fig 2) The net average distances between clades A and B, clade A and Misgurnus spp., clade B and M spp., and subclades B1 and B2 are 0.083, 0.051, 0.087, and 0.025, respectively The distance between clades A and B was much higher than that between clade A and M spp Population genetic analyses The AMOVA analyses (Table 2) indicated that the largest genetic variance occurred within local populations (47.9%) and that the lowest value was among the local populations within groups (24.4%) Significant differentiation was found in all levels (P0.05), while there was a significant relationship in the case where the eight stocked local populations were excluded from the analysis (R2 =0.048; P[...]... were filtered out of solution and weighed as a group To obtain an overall description of juvenile tarpon diet, the following values were calculated for each prey taxon: the percent frequency of occurrence of prey taxon i among all stomachs in the sample (%Oi = percent frequency of occurrence), the proportion of the number of prey items present in a prey taxon i to the total number of prey items present... Fluctuation of sea level and paleogeography of Japanese archipelago in the end stage of Quaternary Era Proc Jpn Soc Syst Zool 36:1–3 (In Japanese) Ministry of Environment, Government of Japan (2010) Documents of conference for ibis reintroduction project http://www.env go.jp/nature/toki/torikumi.html Accessed 2 July 2010 Miura I, Sekiya K, Ohtani H, Ogata M, Ichikawa Y (2004) A new type of Rana rugosa... of the Ministry of the Environment, Japan (Subject No F-072, representative: Y Shimatani), Grant-in-Aid for Young Scientists B of the Ministry of Education, Culture, Sports, Science and Technology (Subject No 19710027, representative: Y Kawaguchi), Global COE Programs (Center of excellence for Asian conservation ecology as a basis of human-nature mutualism, representative: T Yahara; and Formation of. .. = percent of total diet by number), and the proportion of the aggregate wet weight of a prey taxon i to the total wet weight of all prey items (%Wi = percent of total diet by weight) Based on these values, an Index of Relative Importance (IRI) was calculated for each prey taxon i, where IRIi ¼ %Oi ð%Ni þ %Wi Þ The IRI is a compound index that incorporates quantity, weight, and frequency of occurrence... estimated by the first reader of the branchiostegal ray was used as the assigned age for all specimens For all other age estimations the index of precision was calculated as equation [1] Index of precision ¼ j½annuli counted on structure À assigned ageŠj assigned age Based on the findings of Den Haas and Mandrak (2004), where the bottom 33% of structures had an index of precision score of 0.29 or higher, we... females of each age Log transformations of the lengths, and regression analysis were then conducted for each size class separately Growth equations were then compared to the growth equations produced by Love (2004) by first extrapolating the raw data from a digital copy of Love’s Fig 3, using the computer software ImageJ (NIH image analysis software, http://rsbweb.nih.gov/ij/) followed by an analysis of. .. determination of sex in Lepisosteidae requires sacrifice of the fish and examination of the internal sex organs (Ferrara and Irwin 2001), we were unable to determine the sex of Environ Biol Fish (2011) 90:235–242 our individuals Love (2004) showed that male and female Spotted Gar had differing length at age and rates of growth, with females growing larger and at a faster rate than males of the same age... the maximum life expectancy of 10 years The high number of individuals in the 5 to 7 year age classes is indicative of strong year classes from the years of 2000 to 2002; however, it is not known at this time 241 what has caused this variation in year classes More research is needed to determine the long-term viability of the Rondeau Bay population of Spotted Gar, the largest of the Canadian populations... body sizes, and a variety of other prey taxa (i.e., not just fishes) were consumed by these larger individuals (Fig 3) The consumption of copepods by relatively large juvenile tarpon was unexpected, and may be unique to the man-made habitats we studied The presence of huge quantities of copepods in the diets of some age-0 tarpon suggests that the fish may employ a combination of suction and ram feeding... than the structurally-simple restoration marsh habitats Additionally, our study sites possessed different levels of connectivity to adjacent mangrove forests and surrounding bodies of water Habitats with higher levels of connectivity often support a greater variety of prey Environ Biol Fish (2011) 90:223–233 species (Poulakis et al 2002; Layman et al 2007; Lewis and Gilmore 2007) Further, physical parameters