77 Science Diliman (July-December 2020) 32:2, 77-96 Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) from Mindoro and Leyte Islands, Philippines Raffy Jay C. Fornillos* Gerard Clinton L. Que Rogel Victor D. Mendoza Ian Kendrich C. Fontanilla DNA Barcoding Laboratory, Institute of Biology College of Science University of the Philippines Diliman Perry S. Ong† Biodiversity Research Laboratory, Institute of Biology College of Science University of the Philippines Diliman ABSTRACT The Chinese pond mussel Sinanodonta woodiana (Lea, 1834) is a large freshwater bivalve species of the family Unionidae and a known invasive alien species. Proper verification of its identity as well as its source population is crucial for the control of its spread. However, its high plastic shell morphology that resembles other non-invasive species of unionids can be an obstacle. The distribution and ecological impact of this invasive unionid is not fully understood and should be further investigated to prevent further spread in the Philippines. In this study, we used the cytochrome oxidase I (cox1) gene to verify the identity of putative S. woodiana samples collected from Bato Creek in Oriental Mindoro and Lake Danao in Leyte, Philippines and elucidate their source populations. Eighteen cytochrome oxidase subunit 1 (cox1) barcodes were generated from samples collected from Lake Danao, Leyte (n=13) and Bato Creek, Oriental Mindoro (n=5). These barcodes were subjected to Basic Local Alignment Search Tool (BLAST) analysis, which showed that the cox1 sequences from the Philippine samples matched with those of S. woodiana (>94%) found in GenBank. The sequences were then aligned with cox1 sequences of S. woodiana and other unionid representatives from GenBank. * Corresponding Author Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 78 Phylogenetic and haplotype network analyses also showed three haplotypes (Hap 1, 2, and 4) of S. woodiana samples from Lake Danao and Bato Creek. Hap 1 and 2 are distinct haplotypes observed in Lake Danao samples while Hap 4 is shared between Lake Danao and Bato Creek samples and have clustered with conspecific specimens from Malaysia and Indonesia, suggesting their potential Island Southeast Asian origin. Keywords: Unionidae, DNA barcoding, Invasive Alien Species, Sinanodonta woodiana INTRODUCTION The study of invasive alien species (IAS) cannot be overemphasized, and the negative impact of these organisms cannot be underestimated as they often affect the agricultural sector and cause significant public health problems (Andersen et al. 2004). IAS have been introduced either accidentally or purposely, and many are now uncontrolled in newly colonized ecosystems where they thrive and continually cause significant economic and ecological damage (Anderson 1993). A common example is the introduction of the golden apple snail Pomacea canaliculata in the Philippines, which was originally imported as an alternative protein source for farmers but are now considered an agricultural pest due to their uncontrolled proliferation in agricultural lands (Anderson 1993; Halwart 1994). Another freshwater mollusk, the invasive unionid Chinese pond mussel Sinanodonta woodiana (Lea, 1834), is now considered a major problem in many countries in Europe, North America and other parts of Southeast Asia (Kraszewski and Zdanowski 2007; Popa and Murariu 2009; Lajtner and Crnčan 2011; Colomba et al. 2013; Kamburska et al. 2013; Soroka et al. 2014). This bivalve species, which closely resembles its non-invasive relatives in the genus Anodonta, negatively affects local anodontine populations including A. anatina and other unionids in Europe (Guarneri et al. 2014). S. woodiana’s invasive capacity is primarily attributed to its high filtering capacity, and the cross- resistance induction of its parasitic glochidium larva to its fish host (Donrovich et al. 2017; Douda and Čadková 2017). These characteristics show that these invasive mussels can become “masters of invasion” due to their high tolerance of various environmental factors such as changes in temperature, moisture, humidity, and utilization of hosts, thus making them able to survive and reproduce in many types of habitats (Mwatawala et al. 2006; Davidson et al. 2011). R. C. Fornillos et al. 79 Sinanodonta woodiana is a large freshwater dioecious bivalve species of the family Unionidae. This family is a large group of freshwater mussels with several member species endemic in freshwaters of North America, Europe and East Asia. S. woodiana’s native distribution include the freshwater basins of Amur River, Hanka Lake, China, Hong Kong, Taiwan, Cambodia, Thailand, and Japan but were also reported in other non-endemic territories in Europe where they are now widespread and have established stable populations (Kraszewski and Zdanowski 2007; Popa and Murariu 2009; Latjner and Crnčan 2011; Colomba et al. 2013; Kamburska et al. 2013; Soroka et al. 2014). The invasion of S. woodiana in European freshwaters was due to the introduction of fish stocks infested with the bivalve’s parasitic larval glochidium which primarily attaches to the gills of its fish host (Kraszewski 2006; Kraszewski and Zdanowski 2007; Guarneri et al. 2014). S. woodiana produces large biomass both in natural and colonized areas, and has the ability to tolerate a wide range of physical and chemical factors which contribute to its capacity to invade new aquatic habitats. This bivalve is also a broad host generalist, often able to develop in both co-invasive and native fish hosts (Douda et al. 2012). Moreover, its invasiveness is also associated with its ability to grow fast and produce high numbers of offspring, and to its long life span ranging from 10-15 years (Dudgeon and Morton 1983; Afanasjev et al. 2001; Kraszewski and Zdanowski 2007; Guarneri et al. 2014). Aside from food and fish host competition, S. woodiana can quickly dominate both natural and pre-colonized habitat by establishing a strong benthic-pelagic coupling that may result in changes in the biocenoses of organisms affecting the physical characteristics of freshwater systems (Kraszewski and Zdanowsk 2001; Kraszewski and Zdanowski 2007; Guarneri et al. 2014). Proper identification and detection of S. woodiana have been a challenge since its shell morphology resembles that of the native species of anodontines such as Anodonta anatina (Guarneri et al. 2014). In fact, the taxonomy of S. woodiana was originally assigned to the genus Anodonta but was later grouped to the more appropriate genus Sinanodonta to standardize differences in nomenclature and classification especially for inland water mollusks where anodontines and S. woodiana coexist (Kraszewski and Zdanowski 2007; Guarneri et al. 2014). Morphology-based analysis and molecular tools for species identification have been used for S. woodiana in the past. Linear and geometric morphometrics were used for S. woodiana samples from Europe and other parts of Asia such as the Philippines (Demayo et al. 2012; Guarneri et al. 2014). Results showed that S. woodiana’s phenotype is highly plastic, and the variations in shell dimensions measured were due to the effects of various environmental factors. In the study Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 80 of Guarneri et al. (2014), populations of S. woodiana in two of Italy’s largest lakes, Po and Maggiore, indicated that the species exhibits variation even between populations and may potentially lead to species misidentification. Similar results were reported on geometric morphometrics of S. woodiana specimens collected from two separate freshwater bodies (Lake Lanao and Lawis Stream) in Mindanao, Philippines. Variations observed in its shell were attributed to allometry within populations and varying environmental factors were hypothesized to play a major role. Molecular analysis for species identification was also carried out in European S. woodiana samples using the general barcode gene cytochrome c oxidase subunit 1 (cox1) (Guarneri et al. 2014), confirming 99.84-100% similarity of the samples compared using S. woodiana accessions in the Barcode of Life Database (BOLD). A more comprehensive analysis using various S. woodiana cox1 gene accessions in GenBank (https://www.ncbi.nlm.nih.gov/) was generated using Bayesian Inference, which indicated two major lineages detected, namely temperate and tropical, in which samples from the Philippines clustered together with other S. woodiana from Indonesia and Malaysia to form the tropical lineage (Vikhrev et al. 2017). The status of S. woodiana distribution in the Philippines is limited and is only known in few locations based on previous reports such as in Mindanao (Demayo et al. 2012). Another introduced unionid species with shell form and shape similarity with S. woodiana is Cristaria plicata, which is also present in the country but is used for pearl farming (Guerrero et al. 2002). The possibility of introducing the wrong unionid species for pearl farming is highly likely due to the striking resemblance and plastic shell morphology of the two species. Just like S. woodiana, C. plicata is considered an invasive exotic species and may pose occupational hazard such as injuring workers due to its sharp shell, but is also a source of food and income to people who culture it (Cagauan et al. 2007). C. plicata was detected in Taal Lake (Mutia et al. 2017), Lake Oro in Agusan del Sur (Sularte and Jumawan, 2016), and has been reportedly cultured in Nueva Ecija and Laguna for pearl farming (Guerrero 2002). In this study, we conducted molecular identification through DNA barcoding by using the cox1 gene from putative S. woodiana samples collected from Bato Creek and Lake Danao in the absence of a useful morphological key to confirm species identity. These aforementioned sites have no prior reports on the occurrence of S. woodiana. This study also utilized neighbor-joining and maximum likelihood (ML) algorithms and median-joining haplotype network analysis to further reveal the relationships of the two subpopulations in the Philippines with other specimens of S. woodiana from other countries accessed from GenBank. R. C. Fornillos et al. 81 MATERIALS AND METHODS Sample Collection S. woodiana samples were bought from local vendors along Lake Danao in the municipality of Ormoc, Leyte and handpicked from the waters along Bato Creek in the municipality of Victoria, Oriental Mindoro (Figure 1). The difference in sampling methodology arose due to a serendipitous discovery of S. woodiana in Oriental Mindoro during fieldwork for another unrelated study. Samples were stored in re-sealable plastic bags, placed in a styrofoam chest filled with ice cubes and then transported to the DNA Barcoding Laboratory at the Institute of Biology of the University of the Philippines Diliman in Quezon City. Photographs of the samples were likewise taken. Morphometrics of the Bato Creek, Oriental Mindoro specimens were also taken (Table 1). Figure 1. Location of the (A) study sites in Bato Creek, Mindoro and Lake Danao, Leyte. Sinanodonta woodiana sample collected from (B) Bato Creek and (C) Lake Danao. Map adapted from ArcGIS base map downloaded from http://www.geoportal.gov.ph/viewer. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 82 Table 1. Measurements of the dorsal shell length of 18 S. woodiana specimens presented in this study as well as their corresponding GenBank Accession Numbers. Specimen GenBank Accession Number Dorsal Length (cm) mindorobv1 MN322559 7.51 mindorobv2 MN322558 4.20 mindorobv3 MN322555 4.33 mindorobv4 MN322557 6.28 mindorobv5 MN322554 4.90 UN1 KX424967 11.5 UN2 KX424968 10.3 UN4 KX424969 11.5 UN5 KX424970 10.0 UN6 KX424976 11.0 UN8 KX424977 11.0 UN10 KX424978 10.0 UN12 KX424979 12.0 SW3 KX424971 * SW4A KX424972 * SW5B KX424973 * SW7A KX424974 * SW8B KX424975 * *No Data DNA Extraction and Polymerase Chain Reaction Foot muscle tissue from 13 individuals of S. woodiana samples from Lake Danao and 5 individuals from Bato Creek were used. DNA extraction was performed using a commercial DNA extraction kit (Purelink® Genomic DNA Extraction Kit, Invitrogen Life Technologies or PureLink™ Genomic DNA Mini Kit, ThermoFisher Scientific) following the manufacturer’s protocol with a final elution of 150 microliters (uL). For DNA quantitation using Nanodrop 2000c (Thermo Scientific), 1 uL was used for each sample. Two primer sets were used for cox1 amplification, namely HCO-LCO (HCO: 5’ TAAACTTCAGGGTGACCAAAAAATCA -3’, LCO: 5’- GGTCAACAAATCATAAAGATATTGG -3’) (Folmer et al. 1994) and StyHCO- StyLCOii (StyHCO: 5’- GAATTAAAAATATATACTTCTGGGTG -3’, StyLCOii: 5’- ACGAATCATAAGGATATTGGTAC -3’) (Fontanilla et al. 2017). Either primer pair, which targets the same region of cox1, was used for amplification. R. C. Fornillos et al. 83 For Lake Danao samples, polymerase chain reaction (PCR) was done by preparing a master mixture with the following components: 5 uL 10X PCR Buffer (-MgCl 2 ), 0.5 uL 0.1875 mM dNTP, 1.25 uL for each primer (StyHCO, StyLCOii, respectively), 5 uL Q buffer, 1 uL 25 mM MgCl 2 , 0.125 uL Taq Polymerase (5U/uL), 8.875 uL nuclease-free dH 2 O, and 2 uL DNA for a total of 25 uL per sample using a PCR condition described by Ma et al. (2012) and run in a 96-well thermocycler (MultiGene Optimax). For the Bato Creek samples, the PCR master mix consisted of 5 uL 10X MyTaq™ PCR Buffer (with 5 mM dNTPs and 15 mM MgCl 2 ), 1 uL of forward and reverse primers (LCO, HCO, respectively), 1 uL of 50 mM MgCl 2 , 0.125 uL of MyTaq™ DNA Polymerase (5U/uL) (Bioline, United Kingdom), 14.875 uL nuclease dH 2 O, and 2 uL of DNA template for a total of 25 uL. PCR conditions consisted of an initial denaturation step at 95°C for 5 minutes, followed by 36 cycles of denaturation at 92°C for 30 seconds, annealing at 51°C for 30 seconds, extension at 68°C for 2 minutes, and a final extension step at 68°C for 5 minutes run in a 96-well thermocycler (SimpliAmp, Applied Biosystems). The difference in reagents is not expected to affect results since specificity for the target area of cox1 is dependent on the primers. Agarose Gel Electrophoresis, Purification and Sequencing Each PCR product was loaded in a 1% agarose gel stained with 1% ethidium bromide (EtBr), submerged in 0.5X TBE (1 L 5x TBE = 54 g Tris, 27.5 g Boric Acid, 20 mL of 0.5 M EDTA pH 8.0) and exposed to 100 volts of electricity for 30 minutes to separate amplicons (~650-700 bp) using a horizontal gel electrophoresis apparatus (Gel XL ultra v-2 Labnet). The gel was visualized in an ultraviolet transilluminator. Visible bands were cut carefully using sterile scalpel blades and were stored in 2 uL microcentrifuge tubes until purification. The purification process was done using a commercial gel extraction kit (QIAquick® Gel Extraction Kit, QIAGEN and Thermo Scientific™ GeneJET Gel Extraction Kit) following the manufacturer’s protocol. A final elution volume of 50 uL was prepared per sample. Samples were sent to 1st Base Malaysia and Macrogen South Korea together with the primers used for single pass sequencing for both forward and reverse sequences. DNA Sequences In silico Analyses Forward and reverse sequences were assembled using Staden Package version 4.0 (Staden et al. 2000), and each consensus sequence was subjected to nucleotide BLAST® (Altschul and Koonin 1998) for determining related sequences. One hundred ninety cox1 sequences of unionids were downloaded from GenBank (http://blast. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 84 ncbi.nlm.nih.gov/) and included in the analyses. S. woodiana cox1 sequences and other unionid cox1 sequences were downloaded and aligned in BioEdit Sequence Alignment Editor 7.0.9.0 (Hall 1999) using the feature ClustalW (Gibson et al. 1996). Uniform gaps were deleted and the shortest sequence was used as a reference for cutting the edges of the alignment. The sequence alignment file was converted to nexus (.nex) format using DAMBE v. 6.4.81 (Xia 2013, 2017), a format readable by PAUP* version 4.01b10 (Swofford 2002) for tree construction and genetic distance calculation. The substitution model was determined using jModelTest v.2.1.10 (Darriba et al. 2012), and the model with the best log-likelihood score was chosen using the Akaike Information Criterion (AIC) (Akaike 1973, 1974; Hurvich and Tsai 1993). Neighbor-joining (Saitou and Nei 1987) and maximum likelihood (ML) (Felsenstein 1981) tree construction methods were applied. Bootstrap replication was executed to determine branch support and reliability (bootstrap nreps=1000) (Felsenstein 1985). The generated ML tree was visualized and rooted using Tree Explorer (Tamura 1999), with the bootstrap supports for both tree construction methods incorporated. The tree was rooted on two species under Margaritiferidae, a sister family of Unionidae (Combosch et al. 2017), Margaritifera margaritifera (JN243891, DQ060171) and Margaritifera auricularia (KC703969, JX046574). Unique haplotypes were then subjected to median-joining haplotype network analysis using NETWORK v5.0.1.1 (Bandelt et al. 1999) to determine the possible origin of Philippine S. woodiana. Haplotype network analysis aimed to determine the relationship between observed haplotypes in a dataset depending on a number of single nucleotide polymorphisms (SNPs) observed in the cox1 gene. RESULTS A total of 18 cox1 sequences obtained from S. woodiana individuals collected from Bato Creek, Mindoro and Lake Danao, Leyte (Figure 1, Table 1) were generated in this study with 21 identified haplotypes from the generated cox 1 alignment (Figure 3, Table 2). BLAST results (Table 3) showed that the cox1 sequences matched those of S. woodiana found in the GenBank (>94%). ML tree based on TRN+I model of DNA substitution as determined by jModelTest showed the distinct clustering of the Philippine specimens together with other S. woodiana specimens from Malaysia and Indonesia with 91% ML bootstraps (Figure 2). R. C. Fornillos et al. 85 Figure 2. Maximum likelihood tree of S. woodiana geographical isolates using the mitochondrial cytochrome oxidase subunit 1 gene (cox1) tested in PAUP*. Bootstrap replication was done to test branch reliability (bootstrap nreps = 1000) to nodes with Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 86 bootstrap support < 50 were only shown. The tree was rooted on four cox1 sequences of two species under the Margaritiferidae family [Margaritifera margaritifera (GenBank Accession No. JN243891, DQ060171)] [Margaritifera auricularia (GenBank Accession No. KC703969, JX046574)]. ML tree on the right is an inset that shows the branches and bootstrap supports of the Tropical Invasive Lineage. Figure 3. Median-joining network of S. woodiana haplotypes of the cytochrome c oxidase subunit I gene (cox1) from GenBank and sequences generated in this study. A total of 21 haplotypes were observed from the generated 520 bases long cox1 DNA alignment. Red dots are putative haplotypes. Sequences from Lake Danao split into three haplotypes (Hap 1, Hap 2, Hap 4), while those from Bato Creek, Mindoro Oriental grouped with one of the Lake Danao haplotypes and other S. woodiana samples from Malaysia and Indonesia (Hap 4). Table 2. Composition of the 21 distinct haplotypes used in the median-joining haplotype network analysis. Hap 1, 2, and 4 are haplotypes associated with S. woodiana samples collected from Bato Creek and Lake Danao in the islands of Mindoro and Leyte, respectively. Samples with missing locations have no sampling details indicated in GenBank. Haplotype No. GenBank Accession Number of Sequences Species ID Location 1 KX424968 Sinanodonta woodiana Lake Danao, Leyte, Philippines 2 KX424979, KX424978, KX424973, KX424972, KX424971 Sinanodonta woodiana Lake Danao, Leyte, Philippines 3 KX051325, KX051324, KX051316 Sinanodonta woodiana Malaysia R. C. Fornillos et al. 87 4 KX424977, KX424976, KX424975, KX424974, KX424970, KX424969, KX424967 Sinanodonta woodiana Lake Danao, Leyte, Philippines MN322554 MN322559, MN322558, MN322555, MN322557, Sinanodonta woodiana Bato Creek, Mindoro, Philippines KX051326, KX051320, KX051318, KX051321 Sinanodonta woodiana Malaysia KU891642, KU891641 Sinanodonta woodiana Indonesia 5 KX051323, KX051319 KX051317 Sinanodonta woodiana Malaysia 6 KX051315 Sinanodonta woodiana Malaysia 7 KX051328 Sinanodonta woodiana Malaysia 8 KJ434487 Sinanodonta woodiana China 9 KY561633 Anemina sp. Russia 10 GQ451870 Anodonta archaeformis South Korea 11 GQ451869 Anodonta archaeformis South Korea 12 AB055627 Sinanodonta woodiana Japan 13 KY561635 Sinanodonta sp. Vietnam 14 KX822668 Sinanodonta woodiana Vietnam 15 KM272949 Sinanodonta woodiana China 16 KJ434486 Sinanodonta woodiana China 17 GQ451868 Sinanodonta woodiana South Korea 18 GQ451867 Sinanodonta woodiana South Korea 19 KU853269, KU853268 KU853267, KU853266 Sinanodonta woodiana Russia 20 KJ434489, KJ434488, KJ434490 Sinanodonta woodiana China 21 KJ125079 Sinanodonta woodiana Ukraine JQ253894 Anodonta woodiana Ukraine KJ125078 Sinanodonta woodiana Poland EF440349, AF468683 Anodonta woodiana Poland KF731775, KF731776 Anodonta woodiana Italy KJ434483, KJ434482, KJ434485, KJ434484 Sinanodonta woodiana JQ253893, HQ283345, HQ283344 Anodonta woodiana Table 2. Composition of the 21 distinct haplotypes used in the median-joining haplotype network analysis. Hap 1, 2, and 4 are haplotypes associated with S. woodiana samples collected from Bato Creek and Lake Danao in the islands of Mindoro and Leyte, respectively. Samples with missing locations have no sampling details indicated in GenBank. (Cont'n.) Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 88 Table 3. BLASTn results of 18 S. woodiana sequences generated from this study. Accession Number BLASTn Result Query Cover Percent Identity KX424977 KX424976 KX424975 KX424970 KX424974 KX424969 KX424967 MN322559 MN322558 MN322555 MN322557 MN322554 MH319868 Sinanodonta woodiana UGSB 19578 isolate 24480 cytochrome c oxidase subunit I (COI) 100% 100% KX424968 MH319868 Sinanodonta woodiana UGSB 19578 isolate 24480 cytochrome c oxidase subunit I (COI) 99% 100% KX424972 KX424971 KX424979 KX424978 KX424973 MG742232 Sinanodonta woodiana cytochrome oxidase subunit I (COI) gene, partial cds; mitochondrial 100% 100% A total of three haplotypes were observed in all S. woodiana cox1 sequences generated from this study, two were from Lake Danao (Hap 1 and Hap 2) and one was shared between Lake Danao and Bato Creek (Hap 4). These unique cox1 sequences were used for haplotype network analysis (Figure 3). Results showed that the Lake Danao and Bato Creek samples of S. woodiana probably came from a population of S. woodiana from Malaysia introduced to the lake, though the mode of introduction is uncertain and still a subject for investigation. Among all haplotypes, S. woodiana from Malaysia and Indonesia are closest to the Lake Danao and Bato Creek specimens based on the level of support for branch reliability on the node where the Philippine specimens diverged from the Chinese S. woodiana and on the minimal SNPs observed among Philippine, Chinese, and other Island Southeast Asian (Malaysia and Indonesia) specimens as compared to European specimens (Figure 2). R. C. Fornillos et al. 89 DISCUSSION The use of molecular data is crucial for species delineation if traditional taxonomy, which is based on morphological traits, is insufficient. In this study, the presence of the S. woodiana is confirmed using DNA barcoding using cox1 (Hebert et al. 2003). Phylogenetic and median-joining haplotype network analyses are useful tools for visualizing the relationships of samples with different geographical distributions. The ML tree of cox1 of unionid taxa showed that the Lake Danao and Bato Creek specimens are conspecific with S. woodiana. Median-joining haplotype network analysis visualizes the possible origin of the Philippine S. woodiana populations using all distinct haplotypes of S. woodiana and creates a network based on mutation events that occurred among haplotypes. The Philippine specimens probably originated from Malaysia. The low haplotype diversity observed from the Lake Danao and Bato Creek samples might be due to founder effect; the same observation has been reported by Soroka et al. (2014) on their analysis of S. woodiana samples from Hungary and Poland. DNA barcoding for species identification is proven useful for S. woodiana as the high plasticity of the morphology of S. woodiana hinders accurate identification of the species (Kraszewski 2006; Guarneri et al. 2014). In addition, other molecular markers and even the use of microsatellites have been suggested, which may provide useful information on its source and path of invasion and better resolution on the phylogenetic relationship of S. woodiana to other unionids (Bogan and Roe 2008; Popa et al. 2011). The potential host fish species of S. woodiana within Lake Danao and Bato Creek system is also a promising area of research. Its possible interaction with local fish stocks such as Oreochromis niloticus (Nile tilapia), Chanos chanos (milkfish), as well as its known host Cyprinus carpio may be of particular interest due to the mussel being a generalist towards its host (Douda et al. 2012). Identification of the host fish species responsible for its invasion success by bearing its glochidium larva could reveal how the species was introduced in the Philippines. Such was the case for Europe where the bivalve utilized Aristichthus nobilis (bighead carp), Ctenopharyngodon idella (grass carp) and Hypophthalamichthys molitrix (silver carp) for its dispersal and propagation (Colomba et al. 2013). The introduction of many species for aquaculture could be the most likely route of introduction of the mussel in the Philippines. Therefore, a survey of freshwater habitats, particularly those utilized for aquaculture using introduced species, is necessary to assess the distribution and spread of this invasive species. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 90 Moreover, further surveys on other endemic unionids with detailed descriptions on morphology, molecular identity, and life history is encouraged so that accurate comparative assessments could be made such as on how invasive exotic unionids like S. woodiana impact their distribution and dispersal. This is the case of Rectidens sumatrensis, a native unionid in Borneo, for which S. woodiana is a major competitor and a significant threat that has dominated freshwater habitats in the island and is likely associated with intentional introductions made for food source and ornamental purposes (Zieritz et al. 2018). CONCLUSIONS Invasive alien species are an emerging threat to global biodiversity. Due to the increase in mobility and access to agricultural goods and aquaculture products, opportunistic transport of organisms highly associated with these commodities may likewise occur. In this study, S. woodiana was detected from two sites in the Philippines. Phylogenetic and haplotype network analyses revealed that the Philippine populations are most closely related to other Island Southeast Asian haplotypes, such as those from Indonesia and Malaysia, and may even suggest their likely origin from these areas. Introduced fish species bearing the glochidium larva may be the most likely route of introduction in the Philippines. The mussel’s resemblance to other native unionid species could also facilitate its further spread. Other potential freshwater areas in the Philippines can be surveyed for the presence of S. woodiana to determine its distribution and potential effect to local unionid species. Moreover, S. woodiana’s resemblance to its relatives may facilitate its further spread as it could be mistaken for another unionid, such as the bivalve Cristaria plicata, which also belongs to family Unionidae and resembles S. woodiana morphologically. Like S. woodiana, it is native to East Asia and has been introduced in the Philippines. Recommendations Increasing sample size and adding more locations can be performed to provide more comprehensive data on the distribution of S. woodiana in the Philippines. Population genetic analysis can also shed light on the potential origin and route of spread in the Philippines. Detection and survey of the presence of the glochidium larva in fish hosts could aid in mitigating the spread of the IAS in the Philippines through its host. Areas with reported C. plicata should be prioritized to confirm species identity through barcoding and to survey other native unionid species inhabiting the sites. R. C. Fornillos et al. 91 Additionally, further investigations should be made on how S. woodiana was dispersed in these non-native territories such as in Lake Danao and Bato Creek. Around Lake Danao, for example, S. woodiana is being sold and consumed as a major shellfish commodity. Short informal interviews conducted by the authors with the vendors reported that S. woodiana was purposely introduced in the lake primarily as a source of food and income for the locals, though these reports should be carefully interpreted and further confirmed by implementing more scientific and systematic methods of extracting information from the locals such as interviews using questionnaires, focus group discussions, and key informant interviews. This can be explored in the future and information from these surveys will supplement biological data in tracing the origin and path of invasion of S. woodiana. Furthermore, examining the distribution network of both S. woodiana and its potential host fish species in the locality will provide significant information on the path and spread of invasion most especially if there are restocking practices being made using S. woodiana or of fish hosts infected with the mussel’s glochidium to other freshwater bodies, thus contributing to the further spread. In the study, S. woodiana samples collected from Lake Danao were bought from vendors in the same clump, making it difficult to trace the specific spot in the lake where these mussels were collected. A geographic information systems approach will help us understand its dispersion by mapping sites with stable S. woodiana populations in freshwater habitats and in sites where they are recently reported. In this manner, authorities will be able to manage S. woodiana’s invasion, protecting more freshwater habitats from colonization. Acknowledgements This study was funded by the Emerging Interdisciplinary Research (EIDR) Program of the Office of the Vice President for Academic Affairs of the University of the Philippines provided to PSO and the Department of Science and Technology- Accelerated Science and Technology Human Resources Development Program (DOST-ASTHRDP) provided to RJCF. Moreover, the authors would like to express their gratitude to all research assistants and principal investigators of DNA Barcoding Laboratory, Molecular Population Genetics Laboratory, and Biodiversity Research Laboratory of the Institute of Biology, University of the Philippines Diliman for their support in the undertaking of this study. Consent for Publication All authors gave their consent for the publication of this work. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 92 Financial Statement Disclosure This study was funded by the Emerging Interdisciplinary Research (EIDR) Program of the Office of the Vice Chancellor for Academic Affairs of the University of the Philippines and the Department of Science and Technology-Accelerated Science and Technology Human Resources Development Program (DOST-ASTHRDP). Conflict of Interest The authors declare that they have no competing interests. REFERENCES Afanasjev SA, Zdanowski B, Kraszewski A. 2001. Growth and population structure of the mussel Anodonta woodiana (Lea, 1834)(Bivalvia, Unionidae) in the heated Konin lakes system. Fish Aquat Lif. 9(1):123-131. Akaike H. 1973. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, editors. 2nd International Symposium on Information Theory; Tsahkadsor, Armenia. Budapest: Akademia Kiado. p. 267-281. Akaike H. 1974. A new look at the statistical model identification. IEEE T Automat Contr. 19(6):716-723. Altschul SF, Koonin EV. 1998. Iterated profile searches with PSI-BLAST--a tool for discovery in protein databases. Trends Biochem Sci. 23(11):444-7. Anderson B. 1993. The Philippine snail disaster. Ecologist. 23(2):70. Andersen MC, Adams H, Hope B, Powell, M. 2004. Risk assessment for invasive species. Risk Analysis. 24(4):787-793. Bandelt HJ, Forster P, Röhl A. 1999. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 16:37-48. Bogan AE, Roe KJ. 2008. Freshwater bivalve (Unioniformes) diversity, systematics, and evolution: status and future directions. J N Am Benthol Soc. 27(2):349-369. Cagauan AG. 2007. Exotic aquatic species introduction in the Philippines for aquaculture – a threat to biodiversity or a boon to the economy? J Environ Sci Manag. 10(1):48-62. Colomba MS, Liberto F, Reitano A, Grasso R, Di Franco D, Sparacio I. 2013. On the presence of Dreissena polymorpha (Pallas, 1771) and Sinanodonta woodiana woodiana (Lea 1834) in Sicily (Bivalvia). Biodivers J. 4(4):571-580. R. C. Fornillos et al. 93 Combosch DJ, Collins TM, Glover EA, Graf DL, Harper EM, Healy JM, Kawauchi GY, Lemer S, McIntyre E, Strong EE, Taylor JD. 2017. A family-level tree of life for bivalves based on a Sanger-sequencing approach. Mol Phylogenet Evol. 107:191-208. Darriba D, Taboada, G, Doallo R, Posada D. 2012. JModelTest 2: More models, new heuristics and parallel computing. Nat Methods. 9(8):772-772. Davidson AM, Jennions M, Nicotra AB. 2011. Do invasive species show higher phenotypic plasticity than native species and, if so, is it adaptive? A meta-analysis. Ecol Lett. 14(4):419- 431. Demayo C, Cabacaba KM, Torres MA. 2012. Shell shapes of the Chinese pond mussel Sinanodonta woodiana (Lea 1834) from Lawis Stream in Iligan City and Lake Lanao in Mindanao, Philippines. Adv Environ Biol. 6(4):1468-1473. Douda K, Vrtílek M, Slavík O, Reichard M. 2012. The role of host specificity in explaining the invasion success of the freshwater mussel Anodonta woodiana in Europe. Biol Invasions. 14(1):127-137. Douda K, Čadková Z. 2017. Water clearance efficiency indicates potential filter-feeding interactions between invasive Sinanodonta woodiana and native freshwater mussels. Biol Invasions. 20(5):1093-1098. Donrovich SW, Douda K, Plechingerová V, Rylková K, Horký P, Slavík O, Liu HZ, Reichard M, Lopes-Lima M, Sousa R. 2017. Invasive Chinese pond mussel Sinanodonta woodiana threatens native mussel reproduction by inducing cross-resistance of host fish. Aquat Conserv. 27(6):1325-1333. Dudgeon D, Morton B. (1983). The population dynamics and sexual strategy of Anodonta woodiana (Bivalvia: Unionacea) in Plover Cove Reservoir, Hong Kong. J Zool. 201(2):161-183. Felsenstein J. 1981. Evolutionary trees from DNA sequences: a maximum likelihood approach. J Mol Evol. 17(6):368-376. Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evol. 39(4):783-791. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol. 3(5):294-299. Fontanilla IKC, Naggs F, Wade CM. 2017. Molecular phylogeny of the Achatinoidea (Mollusca: Gastropoda). Mol Phyloget Evol. 114:382 385. Gibson T, Higgins D, Thompson J. 1996. ClustalW. EMBL, Heidelberg, Germany and EMBL/EBI, Hinxton, UK. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 94 Guarneri I, Popa OP, Gola L, Kamburska L, Kauceri R, Lopes-Lima M, Popa LO, Riccardi N. 2014. A morphometric and genetic comparison of Sinanodonta woodiana (Lea 1834) populations: does shape really matter? Aquat Invasions. 9(2):183-194. Guerrero III RD. 2002. Freshwater mollusk farming in the Philippines: potential and prospects. International Congress on Medical and Applied Malacology, Los Baños; 2002 Oct 21-24; Laguna, Philippines. Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series. 41:95-98. Halwart, M. 1994. The golden apple snail Pomacea canaliculata in Asian rice farming systems: present impact and future threat. Int J Pest Manage. 40(2):199-206. Hebert PDN, Cywinska A, Ball SL, deWaard JR. 2003. Biological identifications through DNA barcodes. Proc Royal Soc B. 270(1512):313-321.  Hurvich C, Tsai C. 1993. A corrected akaike information criterion for vector autoregressive model selection. J Time Ser Anal. 14(3):271-279. Kamburska L, Lauceri R, Riccardi N. 2013. Establishment of a new alien species in Lake Maggiore (Northern Italy): Anodonta (Sinanodonta) woodiana (Lea 1834) (Bivalvia: Uniondiae). Aquat Invasions. 8(1):111-116. Kraszewski A, Zdanowski B. 2001. The distribution and abundance of the Chinese pond mussel Anodonta woodiana (Lea 1834) in the heated Konin Lakes. Arch Polish Fish. 9(2):253-265. Kraszewski A. 2006. Morphological variation in the Chinese clam Sinanodonta woodiana (Lea 1834) in the heterogenous conditions of the Konin Heated Lake System in Central Poland. Folia Malacol. 14(1):11-23. Kraszewski A, Zdanowski B. 2007. Sinanodonta woodiana (Lea 1834) (Mollusca) - a new mussel species in Poland: occurrence and habitat preferences in a heated lake system. Pol J Ecol. 55(2):337-356. Lajtner J, Crnčan P. 2011. Distribution of the invasive bivalve Sinanodonta woodiana (Lea 1834) in Croatia. Aquat Invasions. 6(1):119-124. Ma H, Ma C, Ma L, 2012. Molecular identification of genus Scylla (Decapoda: Portunidae) based on DNA barcoding and polymerase chain reaction. Biochem Syst Ecol. 41:41-47. Mutia MTM, Muyot MC, Torres Jr. F, Faminialagao CM. 2018. Status of Taal Lake fishery resources with emphasis on the endemic freshwater sardine, Sardinella tawilis (Herre, 1927). Phillip J Fish. 25(1):128-135. R. C. Fornillos et al. 95 Mwatawala MW, De Meyer M, Makundi RH, Maerere AP. 2006. Seasonality and host utilization of the invasive fruit fly, Bactrocera invadens (Dipt., Tephritidae) in central Tanzania. J Appl Entomol. 130(9-10):530-537. Popa OP, Murariu D. 2009. Freshwater bivalve mollusks invasive in Romania. Biological invasions: towards a synthesis. NeoBiota. 8:123-133. Popa OP, Popa LO, Krapal AM, Murariu D, Iorgu EI, Costache M. 2011. Sinanodonta woodiana (Mollusca: Bivalvia: Unionidae): isolation and characterization of the first microsatellite markers. Int J Mol Sci. 12:5255-5260. Saitou N, Nei M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol. 4(4):406-425. Soroka M, Urbańska M, Andrzejewski W. 2014. Chinese pond mussel Sinanodonta woodiana (Lea 1834) (Bivalvia): origin of the Polish population and GenBank data. J Limnol. 73(3):454- 458. Staden R, Beal KF, Bonfield JK. 2000. The Staden package, 1998. Methods Mol Biol 132:115- 130. Sularte RP, Jumawan JC. 2016. Freshwater mollusks assemblage (Mollusca: Gastropoda) with notes on invasive species and its environmental parameters in Lake Oro, Esperanza, Mindanao, Philippines. Asian J Conserv Biol. 5(1):25-30. Swofford DL. 2002. PAUP* version 4.0. Phylogenetic analysis using parsimony (and other methods). Tamura K. 1999. Tree Explorer, version 2.12. Tokyo, Japan: Department of Biology, Tokyo Metropolitan University. http://evol.gen.biol.metro-u.ac.jp/TE/TE_man.html Vikhrev IV, Konopleva ES, Gofarov MY, Kondakov AV, Chapurina YE, Bolotov IN. 2017. A tropical biodiversity hotspot under the new threat: discovery and DNA barcoding of the invasive Chinese pond mussel Sinanodonta woodiana in Myanmar. Trop Conserv Sci. 10:1-11. https:// doi.org/10.1177/1940082917738151. Xia X. 2013. DAMBE5: a comprehensive software package for data analysis in molecular biology and evolution. Mol Biol Evol. [cited 2019 July 11]; 30(7):1720-1728. http://dx.doi. org/10.1093/molbev/mst064. Xia X. 2017. DAMBE6: new tools for microbial genomics, phylogenetics, and molecular evolution. J Hered. [cited 2019 July 11]; 108(4):431-437. http://dx.doi.org/10.1093/jhered/ esx033. Zieritz A, Bogan AE, Rahim KAA, Sousa R, Jainih L, Harun S, Abd Razak NF, Gallardo B, McGowan S, Hassan R, Lopes-Lima, M. 2018. Changes and drivers of freshwater mussel diversity and distribution in northern Borneo. Bio Conserv. 219:126-137. Molecular Identification of the Chinese Pond Mussel Sinanodonta woodiana (Lea, 1834) 96 ______ Raffy Jay C. Fornillos is an instructor at the Institute of Biology, UP Diliman. He obtained his MS Biology at the same institution. He specializes in Molecular Parasitology. Rogel Victor D. Mendoza is currently a research associate under the DNA Barcoding Laboratory at the Institute of Biology, UP Diliman. He obtained his MS in Biology from the Institute of Biology, UP Diliman with his thesis focusing on the population and molecular genetics of bat species. Gerard Clinton L. Que is a member of the Molecular Population Genetics Laboratory of the Institute of Biology, UP Diliman. He did his thesis work for his Bachelor of Science and Master of Science in Biology at the same institute under his adviser, Dr. Ian Kendrich C. Fontanilla. His research interest is in Molecular Phylogenetics and Molecular Ecology. Perry S. Ong† was a Professor and Head of the Biodiversity Research Laboratory at the Institute of Biology, UP Diliman. He obtained his Ph.D. at the Monash University Department of Ecology and Wildlife Biology. His expertise was in Philippine Wildlife Biology. He passed away on 2 March 2019. Ian Kendrich C. Fontanilla is a Professor and Head of the DNA Barcoding Laboratory at the Institute of Biology, UP Diliman. He received his Ph.D. in Genetics from the University of Nottingham, United Kingdom. He specializes in Molecular Genetics, Phylogenetics, and Malacology.