Authenticating the use of dried seahorses in the traditional Chinese medicine market in Taiwan using molecular forensics

Original Article

Authenticating the use of dried seahorses in the

traditional Chinese medicine market in Taiwan

using molecular forensics

Chia-Hao Chang

, Nian-Hong Jang-Liaw

, Yeong-Shin Lin

,

Yi-Chiao Fang

, Kwang-Tsao Shao

*

a_{Biodiversity Research Center, Academia Sinica, Taipei, Taiwan, ROC}

b_{Department of Biological Science and Technology, National Chiao Tung University, Hsinchu, Taiwan, ROC} c_{Animal Department, Taipei Zoo, Taipei, Taiwan, ROC}

d_{Institute of Bioinformatics and Systems Biology, National Chiao Tung University, Hsinchu, Taiwan, ROC}

a r t i c l e i n f o

Article history:

Received 1 December 2012 Received in revised form 23 January 2013 Accepted 3 April 2013

Available online 7 August 2013

Keywords: IUCN Red List Molecular forensics Seahorse

a b s t r a c t

Seahorse, which has a unique appearance and exhibits male pregnancy, is a useful component of traditional Chinese medicine (TCM). With the growing demand for TCM, vast amounts of seahorses are harvested from the wild every year and traded internationally. This study investigated 58 dried seahorse samples collected from 23 Chinese herbal medicine stores across Taiwan using molecular forensics. Results showed that eight sea-horse species were present in the Taiwan TCM market. Among them, Knysna seasea-horse (Hippocampus capensis) has an endangered status according to the International Union for Conservation of Nature (IUCN) Red List, while the West African seahorse (Hippocampus algiricus), tiger tail seahorse (Hippocampus comes), thorny seahorse (Hippocampus histrix), great seahorse (Hippocampus kelloggi), yellow seahorse (Hippocampus kuda), hedgehog sea-horse (Hippocampus spinosissimus), and three-spot seasea-horse (Hippocampus trimaculatus) have vulnerable status.

Copyrightª 2013, Food and Drug Administration, Taiwan. Published by Elsevier Taiwan LLC. All rights reserved.

1.

Introduction

Seahorses (Hippocampus spp.), one of the most appealing groups of 300 known species in the family Syngnathidae, are celebrated for their distinctive appearances [1]. They are found in wide longitudinal and latitudinal areas[2] and all exhibit male pregnancy[3]. After receiving eggs from females, male seahorses fertilize eggs in specialized abdominal pouches; the eggs develop into embryos therein, under

suitable conditions[4]. This one-of-a-kind biological feature makes seahorses very appealing to the public, making it even more important to conserve them[5].

Seahorses that feature in traditional medicine, aquarium display, and curiosities are internationally traded; the number of Hippocampus species that are traded in the global market annually has been estimated to be about 20 million[6]. Most of them are sold as dried traditional Chinese medicine (TCM) materials rather than live as aquarium pets [7,8]. A clear * Corresponding author. Fish Ecology & Evolution Laboratory, Biodiversity Research Center, Academia Sinica, 128 Academia Road, Sec. 2, Nankang Taipei 11529 Taiwan, ROC.

E-mail address:zoskt@gate.sinica.edu.tw(K.-T. Shao).

Available online at

www.sciencedirect.com

j o u r n a l h o m e p a g e : w w w . j f d a - o n l i n e . c o m

1021-9498/$ e see front matter Copyright ª 2013, Food and Drug Administration, Taiwan. Published by Elsevier Taiwan LLC. All rights reserved. http://dx.doi.org/10.1016/j.jfda.2013.07.010

picture has emerged about the curative activities of seahorses because researchers have found out that, in addition to their traditional use in treating erectile dysfunction[6,9], seahorses also exhibit antitumor, antiaging, and antifatigue properties and are able to suppress neuroinflammatory responses and collagen release[8,10]. The two-millennium history of TCM guarantees a great demand for seahorses among Chinese people, and dried seahorses are mainly consumed in China, Hong Kong, and Taiwan[7]. The economic development of China, which has increased the demands for TCM, has been catastrophic for wild animals[11]. It is thus predicted that seahorses will face increasingly deleterious conditions in the future.

Even though fishery by-catch, especially from shrimp trawlers, is the principal method of seahorse capture [12], economic enticement has caused an increasing number of fishermen to employ fishing methods that specifically target seahorses in many developing nations, such as Brazil, India, and the Philippines [7]. Fishing has had drastic effects on many aspects of seahorses, such as harming individuals, destroying social structure, reducing reproduction, affecting population structure, and devastating habitats[13]; unfortu-nately, many of their life history traits, such as high mate fi-delity, small brood sizes, lengthy parental care, and low dispersal ability, make them more susceptible to persistent exploitation[14,15].

To allay fishing pressures on seahorses, two measures have been taken by the Conservation on International Trade in Endangered Species (CITES): first, all seahorse species have officially been listed under an Appendix II designation since 2004, in order to prohibit international seahorse trade[16]; and second, a 10-cm minimum height limit (from the top of the head to the tip of the tail) policy for seahorses in trade, except for Hippocampus kelloggi[17], has been enforced to ensure that seahorses have sufficient time to grow up and reproduce before being captured and sold[18]. Some regional organiza-tions or naorganiza-tions, such as the OSPAR Commission, as well as the IUCN Red List of Threatened Species have also cautioned the public against exploiting seahorses by listing several of them as being endangered or vulnerable. If these policies are carried out successfully, harvesting of seahorse resources can be supervised carefully, and utilization of these sensitive creatures will be sustainable.

According to documents obtained from the Bureau of Foreign Trade [19], Taiwan has imported an appreciable amount of desiccated seahorses, in the range of 3181e8797 kg each year, during 2008e2011. These desiccated seahorses have been used as ingredients in TCM and exported mainly from China and Thailand. The CITES trade database also showed that the majority of dried seahorses came from wild populations rather than captive-bred ones. However, the species composition of dried seahorses in the Taiwan market has never been ascertained. The purpose of this study was to investigate the number of seahorse species present in the Taiwan TCM market, and whether Taiwan is taking part in the trade of endangered seahorse species. Because of the difficulty of morphological identification of dried seahorses, especially because they are sometimes sliced longitudinally before being sold, and of the abundant seahorse cytochrome b (cyt b) gene

sequences in GenBank (www.ncbi.nlm.nih.gov), DNA

taxonomy was employed to identify the species of specimens obtained from the Taiwan market.

2.

Methods

2.1. Samples

In total, 58 dried seahorse samples were purchased from 23 Chinese herbal medicine stores in 2012. Because dried sea-horses have exorbitant prices and a detailed morphological examination was not possible on the spot, during collection, dried seashores were classified roughly into different

Table 1 e Sampling locations and results of molecular forensics of 58 dried seahorse (Hippocampus) samples.

Store no. Location Specimen

code

Haplotype code

Molecular identification

1 Taipei City TS01MO-1 TS01MO1 H. trimaculatus 2 Taipei City TS02MO-1 TS02MO1 H. spinosissimus 3 Taipei City TS03MO-1 TS03MO1 NR

4 New Taipei City

TS04MO-1 TS04MO1 H. trimaculatus 5 New Taipei

City

TS05MO-1 TS05MO1 H. kelloggi TS05MO-2 TSH4 H. trimaculatus TS05MO-3 TS05MO3 H. spinosissimus TS05MO-4 TSH1 H. trimaculatus TS05MO-5 TS05MO5 H. trimaculatus 6 Keelung

City

TS14MO-1 TS14MO1 H. spinosissimus TS14MO-2 TSH1 H. trimaculatus 7 Keelung

City

TS15MO-1 TS15MO1 H. trimaculatus TS15MO-2 TS15MO2 H. trimaculatus TS15MO-3 TS15MO3 H. spinosissimus TS15MO-4 TSH5 H. trimaculatus TS15MO-5 TS15MO5 H. trimaculatus 8 Keelung

City

TS19MO-1 TS19MO1 H. kuda TS19MO-2 TS19MO2 H. kuda TS19MO-3 TSH6 H. comes 9 Hsinchu

City

TS11MO-1 TS11MO1 H. trimaculatus TS11MO-2 TSH3 H. trimaculatus TS11MO-3 TS11MO3 H. trimaculatus TS11MO-4 TS11MO4 H. trimaculatus 10 Miaoli County TS06MO-1 TSH2 H. trimaculatus TS06MO-2 TSH3 H. trimaculatus 11 Taichung City

TS12MO-1 TS12MO1 H. kuda TS12MO-2 TS12MO2 H. comes TS12MO-3 TS12MO3 H. trimaculatus TS12MO-4 TSH1 H. trimaculatus 12 Taichung City TS13MO-1 TSH1 H. trimaculatus TS13MO-2 TSH2 H. trimaculatus 13 Taichung City

TS16MO-1 TS16MO1 H. capensis TS16MO-2 TSH6 H. comes TS16MO-3 TS16MO3 H. trimaculatus 14 Changhua

County

TS09MO-1 TS09MO1 H. comes TS09MO-2 TS09MO2 H. histrix TS09MO-3 TS09MO3 H. trimaculatus TS09MO-4 TS09MO4 H. kelloggi TS09MO-5 TS09MO5 H. spinosissimus TS09MO-6 TS09MO6 H. trimaculatus 15 Tainan City TS08MO-1 TS08MO1 H. trimaculatus 16 Kaohsiung

City

TS18MO-1 TS18MO1 H. kelloggi (continued on next page)

morphological forms and one individual of each form was purchased randomly and brought back to the laboratory. De-tails are given inTable 1.

2.2. DNA extraction, polymerase chain reaction, and

sequencing

DNA samples were extracted from fin tissues using a Quick-Gene DNA tissue Kit S (Fujifilm, Tokyo, Japan). Polymerase chain reaction (PCR) amplifications of mitochondrial cyto-chrome b (cyt b) (1140 bp) were performed in a mixture with a final volume of 100 mL, containing 10 ng template DNA; 25mmol of each specific primer; SH-F (50-AAC YAG GAC YAA TGR CTT GA-30) and SH-R (50-GCA SWA GGG AGG RKT TTA AC-30), which were designed on the basis of the mitochondrial genomes of Hippocampus kuda (NC_010272) and Microphis bra-chyurus (NC_010273) and located, respectively, at tRNA-Glu (14311e14330) and tRNA-Thr (15550e15569) according to the sequence (NC_010272); 50 mL of Fast-RunTM Advanced Taq Master Mix (ProTech, Taipei, Taiwan); and distilled water. Thermal cycling began with one cycle at 94C for 4 minutes; followed by 35 cycles of denaturation at 94C for 1 minute, 45e51C for 1 minute, and 72C for 1 minute; and finally, a single extension step at 72C for 10 minutes. PCR products were purified using a PCR DNA Fragments Extraction Kit (Geneaid, Taipei, Taiwan). Approximately 50 ng of the purified PCR products were employed as the template for sequencing, which was performed by following the protocol of the ABI PRISM BigDye Sequencing Kit (PE Applied Biosystems, Foster City, CA, USA) with the SH-F and SH-R primers.

2.3. Data analysis

As suggested by Wilson and Orr[1]and Sanders et al[20], the cyt b DNA sequences of Syngnathus schlegeli (AY786429) and Corythoichthys haematopterus (AY787229) from GenBank served

as outgroups for the phylogenetic analysis. To construct the reference database, data on 30 Hippocampus cyt b haplotypes of 22 Hippocampus species studied by Casey et al[21]and Chang et al[22]were downloaded (Table 2). The entire cyt b gene was 1141 bp; however, the last nucleotide of the incomplete stop codon was deleted from all cyt b gene sequences analyzed in this study so that they had the same length (1140 bp) and could be aligned directly. A maximum-likelihood (ML) tree was constructed utilizing RAxML 7.0.4[23]. In the parameter setting of the software RAxML 7.0.4, data were partitioned by codon position, which was performed under the GTRþ G þ I model. The ML tree was obtained by performing 100 different runs using the default algorithm of the program, and the best ML tree was chosen based on the likelihood scores of subop-timal trees created in each run. Nodal support was certified by a bootstrap analysis with 1000 nonparametric bootstrap iter-ations. The software Mega 4[24]was utilized to calculate the K2P genetic distance among taxa.

3.

Results and discussion

DNA markers, such as 12S rRNA, cytochrome b, and the in-ternal transcribed spacer, are applied widely in taxonomy, and are also used to identify the species compositions of TCMs

Table 1 e (continued )

Store no. Location Specimen

code Haplotype code Molecular identification 17 Kaohsiung City

TS20MO-1 TS20MO1 H. algiricus 18 Kaohsiung

City

TS21MO-1 TS21MO1 H. spinosissimus 19 Kaohsiung

City

TS22MO-1 TSH4 H. trimaculatus 20 Pingtung

County

TS07MO-1 TS07MO1 H. trimaculatus 21 Hualien

County

TS17MO-1 TS17MO1 H. trimaculatus TS17MO-2 TS17MO2 H. comes TS17MO-3 TS17MO3 H. spinosissimus TS17MO-4 TS17MO4 H. trimaculatus 22 Kinmen

County

TS10MO-1 TS10MO1 H. comes TS10MO-2 TS10MO2 H. comes TS10MO-3 TS10MO3 H. trimaculatus TS10MO-4 TS10MO4 H. kuda 23 Penghu

County

TS23MO-1 TS23MO1 H. spinosissimus TS23MO-2 TSH5 H. trimaculatus TS23MO-3 TS23MO3 H. trimaculatus TS23MO-4 TS23MO4 H. trimaculatus NR¼ negative response in the polymerase chain reaction (PCR).

Table 2 e Set of 30 cytochrome b (cyt b) haplotypes from 22 Hippocampus species used as the reference database for the maximum-likelihood phylogenetic analysis.

Scientific name Haplotype

code

Accession no.

Origin

H. abdominalis NZ.98 AF192641 New Zealand AUS.183.1 AF192638 Australia H. algiricus GHA.X1 AF192642 Ghana H. barbouri PH.225 AF192646 Philippines

PH.132 AF192645 Philippines H. breviceps BIR.361 AF192647 Australia H. camelopardalis MOZ.515 AF192648 Mozambique H. capensis K.59 AF192650 South Africa H. comes PB.53 AF192653 Philippines H. coronatus JAP.344.1 AF192658 Japan H. erectus CHS.170.1 AF192661 USA

BRZ.182.4 AF192660 Brazil H. guttulatus EX.221 AF192663 Italy H. hippocampus EX.230 AF192665 England H. histrix JAP.345 AF192667 Japan

VIG.40 AF192668 Vietnam H. ingens PER.184.1 AF192672 Peru H. kelloggi VIE.32 AF192675 Vietnam H. kuda TH.174.9 AF192686 Thailand TA.207 AF192685 Taiwan IND.350.1 AF192680 India H. mohnikei JAP.346.7 AF192688 Japan H. reidi BAR.94 AF192690 Barbados H. spinosissimus PF.152 AF192695 Philippines H. subelongatus AUS.228 AF192697 Australia H. trimaculatus PA.128 AF192701 Philippines

VIC.22 AF192703 Vietnam

TW JX682713 Taiwan

H. whitei AUS.217 AF192704 Australia H. zosterae FK.141 AF192706 USA

and food[25e28]. Dried seahorse samples, although having undergone a manufacturing process and mostly been stored at room temperature, were demonstrated to contain qualified DNA materials that were sufficient for amplification of cyt b gene sequences. In this study, only one (TS03MO-1) of 58 specimens had a negative PCR response, and 49 cyt b haplo-types were sequenced successfully (Table 1). The DNA bar-coding technologies have been applied to prevent the smuggling of endangered species, and many seahorse species

have an endangered or a vulnerable status on the IUCN Red List; consequently, this study suggests that customs au-thorities can utilize a molecular forensic method to verify and oversee imported dried seahorses.

Fig. 1shows the results of the phylogenetic analysis of sea-horse haplotypes from the investigated samples and the Gen-Bank database. The genus Hippocampus is an apparently monophyletic group with high statistical support; however, as noted by Casey et al[21], some seahorse species, such as H. kuda

Fig. 1 e ML tree of Hippocampus inferred from mitochondrial cytochrome b sequences with 1000 bootstrap replicates. Bootstrap values of>70% are indicated. Seven groups were defined. ML [ maximum likelihood.

and Hippocampus capensis, have close genetic relationship owing to recent differentiation (<1 million years ago), which results in an incomplete lineage sorting and blurs resolution of the phylogenetic analysis. The ML tree revealed that haplotypes of investigated samples were distributed in seven groups: Hippo-campus trimaculatus group (K2P genetic distance ranging from 0.1% to 3.8% within group), Hippocampus histrix group (K2P ge-netic distance ranging from 0.6% to 1.1% within group), Hippo-campus comes group (K2P genetic distance ranging from 0.1% to 11.9% within group), H. kelloggi group (K2P genetic distance ranging from 0.2% to 1.3% within group), Hippocampus spino-sissimus group (K2P genetic distance ranging from 0.2% to 2.1% within group), Hippocampus algiricus group (K2P genetic distance being 0.6% within group), and H. kuda/H. capensis group (K2P genetic distance ranging from 0.2% to 3.3% within group). All groups, except for the H. kuda/H. capensis group, represented one Hippocampus species. The close lineage between H. kuda and H. capensis circumscribes genetic authentication. Instead, the morphological character, with or without a coronet, can serve as a good marker to distinguish one from the other[29]. The K2P genetic distances among the 79 Hippocampus haplotypes ranges from 0.1% to 23.2%. The average interspecific K2P genetic dis-tances range from 1.2% (Hippocampus algiricus vs. Hippocampus reidi) to 23.2% (Hippocampus guttulatus vs. Hippocampus coronatus), and the average of intraspecific K2P genetic distances range from 0.3% (H. comes) to 5.9% (H. erectus). In general, the K2P values calculated in this study are tantamount to those of Casey et al[21], as long as there is only 1 bp difference in length be-tween the two studies.

Eight Hippocampus species were found in the Taiwan TCM market: the West African seahorse (H. algiricus) (Fig. 2A), Knysna seahorse (H. capensis) (Fig. 2B), tiger tail seahorse (H. comes) (Fig. 2C), thorny seahorse (H. histrix) (Fig. 2D), great seahorse (H. kelloggi) (Fig. 2E), yellow seahorse (H. kuda) (Fig. 2F), hedgehog seahorse (H. spinosissimus) (Fig. 2G), and three-spot seahorse (H. trimaculatus) (Fig. 2H). Among them, the three-spot seahorse was the most common species in the Taiwan TCM market and was present in 73.9% of investigated stores, in contrast to the rarest species, West African sea-horse, Knysna seasea-horse, and thorny seasea-horse, for each of which only one specimen was found. Compared to the US Californian shops selling seahorses native to Asia, Eastern Africa, American, Oceania, and Europe[20], the Taiwan TCM stores sold seahorses that were mainly from Asia, in addition to the Knysna and West African seahorses.Fig. 2also shows that the 10-cm minimum height limit was violated, and young juveniles were found in the market. After the manufacturing process, samples from a single species may have a diversity of appearances, making morphological identification difficult. The Knysna seahorse has an endangered status in the IUCN Red List, whereas the West African seahorse, tiger tail sea-horse, thorny seasea-horse, great seasea-horse, yellow seasea-horse, hedgehog seahorse, and three-spot seahorse have vulnerable status. All the sampled seahorse species were noted to be under some kind of threat.

The cyt b sequence acted as a good genetic marker to reveal relationships among seahorse species and was also used in

many studies on seahorse population genetics [30e32].

Fig. 2 e Pictures of dried seahorses of different Hippocampus species: (A) H. algiricus (TS20MO-1); (B) H. capensis (TS16MO-1); (C) H. comes (left to right: TS12MO-2, TS10MO-2, and TS09MO-1); (D) H. histrix (TS09MO-2); (E) H. kelloggi (left to right: TS18MO-1, TS05MO-1, and TS09MO-4); (F) H. kuda (left to right: TS19MO-1, TS10MO-4, and TS12MO-1); (G) H. spinosissimus (left to right: TS02MO-1, TS21MO-1, and TS17MO-3); and (H) H. trimaculatus (left to right: TS06MO-2, TS13MO-1, and TS04MO-1).

However, the wide distribution of cyt b haplotypes observed in tiger tail seahorse, three-spot seahorse, and yellow seahorse may hinder it from discriminating genetic structures among different populations[21]. Therefore, it was difficult to deduce the native habitats of these dried seahorses based solely on cyt b sequences.

Distributions of these sampled seahorses are incompatible with official documents, which recorded that, in the most recent 4 years, the imported dried seahorses were mostly from Asia; however, native habitats of both the Knysna seahorse and West African seahorse are in Africa. Except for smuggling, this inconsistency can be explained by the fact that China is a primary consumer and a major exporter of dried seahorses, so it may re-export dried seahorses.

Molecular forensics is a potent tool that can be used to eliminate smuggling of endangered creatures and their prod-ucts, and disclose the real exploitation of certain organisms, which would help enforce the law and promote legislation. A good example of this is a recent DNA barcode study on shark, which demonstrated that up to 22 shark species are consumed in Taiwan, 56% of which have a status ranging from endan-gered to vulnerable (unpublished data). Results of this sea-horse study revealed that seasea-horses face as serious a fishing pressure as do sharks. Threatened seahorse species are traded, and young individuals are also fished. Even though many seahorse species can be raised artificially and aquaculture techniques for rearing them are being developed[33e36], au-thorities should take more measures to supervise the species composition of imported desiccated seahorses and manage the utilization of these organisms until the production from aquaculture can satisfy the demands of the TCM market.

r e f e r e n c e s

[1] Wilson AB, Orr JW. The evolutionary origins of Syngnathidae: pipefishes and seahorses. J Fish Biol 2011;78:1603e23. [2] Lourie SA, Stanley HF, Vincent ACJ, et al. Seahorses: an

identification guide to the world’s species and their conservation. London, UK: Project Seahorse; 1999.

[3] Vincent ACJ. Reproductive ecology of seahorses. Cambridge, UK: University of Cambridge; 1990.

[4] Linton JR, Soloff BL. The physiology of the brood pouch of the male seahorse Hippocampus erectus. Bull Mar Sci Gulf Caribb 1964;14:45e61.

[5] Scales H. Poseidon’s steed: the story of seahorses, from myth to reality. New York, USA: Penguin Group; 2009.

[6] Vincent ACJ. The international trade in seahorses. Cambridge, UK: TRAFFIC International; 1996.

[7] Vincent ACJ, Foster SJ, Koldewey HJ. Conservation and management of seahorses and other Syngnathidae. J Fish Biol 2011;78:1681e724.

[8] Ryu B, Qian ZJ, Kim SK. SHP-1, a novel peptide isolated from seahorse inhibits collagen release through the suppression of collagenases 1 and 3, nitric oxide products regulated by NF-kB/p38 kinase. Peptides 2010;31:79e81.

[9] May B, Tomoda T. Seahorses in the Ben cao gang mu and contemporary Chinese medicine. J Aust Chin Med Educ Res Counc 2002;7:1e12.

[10] Himaya SWA, Ryu B, Qian ZJ, et al. Paeonol from Hippocampus kuda Bleeler suppressed the neuro-inflammatory responses

in vitro via NF-kB and MAPK signaling pathways. Toxicol In Vitro 2012;26:878e87.

[11] Still J. Use of animal products in traditional Chinese medicine: environmental impact and health hazards. Complement Ther Med 2003;11:118e22.

[12] Perry AL, Lunn KE, Vincent ACJ. Fisheries, large-scale trade, and conservation of seahorses in Malaysia and Thailand. Aquatic Conserv Mar Freshw Ecosyst 2010;20:464e75. [13] Baum JK, Meeuwig JJ, Vincent ACJ. Bycatch of lined seahorses

(Hippocampus erectus) in a Gulf of Mexico shrimp trawl fishery. Fish Bull 2003;101:721e31.

[14] Foster SJ, Vincent ACJ. Life history and ecology of seahorses: implications for conservation and management. J Fish Biol 2004;65:1e61.

[15] Curtis JMR, Vincent ACJ. Life history of an unusual marine fish: survival, growth and movement patterns of

Hippocampus guttulatus Cuvier 1829. J Fish Biol 2006;68:707e33.

[16] Conservation on International Trade in Endangered Species (CITES). Amendments to Appendices I and II of the Conservation adopted by the Conference of the Parties at its 12th meeting. Santiago, Chile; 2002.

[17] Foster SJ, Vincent ACJ. Enhancing sustainability of the international trade in seahorses with a single minimum size limit. Conserv Biol 2005;19:1044e50.

[18] Conservation on International Trade in Endangered Species (CITES). Notification to the partiesdtrade in

seahorsesdimplementation of decision 12.54; 2004.http:// www.cites.org/eng/notif/2004/033.pdf.

[19] Bureau of Foreign Trade, Ministry of Economic Affairs, R.O.C. Trade statistics; 2012.http://cus93.trade.gov.tw/FSCI. [20] Sanders JG, Cribbs JE, Fienberg HG, et al. The tip of the tail:

molecular identification of seahorses for sale in apothecary shops and curio stores in California. Conserv Genet 2008;9:65e71.

[21] Casey SP, Hall HJ, Stanley HF, et al. The origin and evolution of seahorses (genus Hippocampus): a phylogenetic study using the cytochrome b gene of mitochondrial DNA. Mol

Phylogenet Evol 2004;30:261e72.

[22] Chang CH, Shao KT, Lin YS, et al. The complete mitochondrial genome of the three-spot seahorse, Hippocampus trimaculatus (Teleostei, Syngnathinae). Mitochondrial DNA 2013 in press.

[23] Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 2006;22:2688e90.

[24] Tamura K, Dudley J, Nei M, et al. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 2007;24:1596e9.

[25] Chan WH, Ling KH, Chiu SW, et al. Molecular analyses of Cordyceps gunnii in China. J Food and Drug Anal

2011;19:18e25.

[26] Huang KM, Liu SM, Huang YW, et al. Identification of species in commercial frozen shrimp meat in Taiwan. J Food Drug Anal 2012;20:839e43.

[27] Lu KT, Lo CF, Lin JH. Identification of testudinis carapax and cervi cornu in Kuei-Lu-Erh-Hsien Chiao by nested PCR and DNA sequencing methods. J Food Drug Anal

2009;17:151e5.

[28] Wu CT, Gupta SK, Wang AZ, et al. Internal transcribed spacer sequence based identification and phylogenic relationship of herba dendrobii. J Food Drug Anal 2012;20:143e51.

[29] Lourie SA, Foster SF, Cooper EWT, et al. A guide to the identification of seahorses. Project Seahorse and TRAFFIC North America. USA: University of British Columbia and World Wildlife Fund; 2004.

[30] Panithanarak T, Karuwancharoen R, Na-Nakorn U, et al. Population genetics of the spotted seahorse (Hippocampus

kuda) in Thai waters: implications for conservation. Zool Stud 2010;49:564e76.

[31] Woodall LC, Koldewey HJ, Shaw PW. Historical and contemporary population genetic connectivity of the European short-snouted seahorse Hippocampus hippocampus and implications for management. J Fish Biol

2011;78:1738e56.

[32] Nickel J, Cursons R. Genetic diversity and population structure of the pot-belly seahorse Hippocampus abdominalis in New Zealand. N Z J Mar Freshwater Res 2012;46:207e18. [33] Koldewey HJ, Martin-Smith KM. A global review of seahorse

aquaculture. Aquaculture 2010;302:131e52.

[34] Palma J, Bureau DP, Andrade JP. Effect of different Artemia enrichments and feeding protocol for rearing juvenile long snout seahorse, Hippocampus guttulatus. Aquaculture 2011;318:439e43.

[35] Martinez-Cardenas L, Purser GJ. Effect of temperature on growth and survival in cultured early juvenile pot-bellied seahorses, Hippocampus abdominalis. J World Aquacult Soc 2011;42:854e62.

[36] Garcia LMB, Hilomen-Garcia GV, Celino FT, et al. Diet composition and feeding periodicity of the seahorse Hippocampus barbouri reared in illuminated sea cages. Aquaculture 2012;358e9:1e5.