DNA Barcoding and the Documentation of Alien Species Establishment on Sub-Antarctic Marion Island

Polar Biol DOI 10.1007/s00300-007-0402-z

ORIGINAL PAPER

DNA barcoding and the documentation of alien species establishment on sub-Antarctic Marion Island

Steven L. Chown · Brent J. Sinclair · Bettine Jansen van Vuuren

Received: 25 October 2007 / Revised: 7 December 2007 / Accepted: 10 December 2007 © Springer-Verlag 2008

Abstract Invasive alien species constitute a substantial Keywords Biological invasion · Establishment · conservation challenge in the terrestrial sub-Antarctic. Lepidoptera · Mitochondrial DNA Management plans, for many of the islands in the region, call for the prevention, early detection, and management of such alien species. However, such management may be Introduction confounded by diYculties of identiWcation of immatures, especially of holometabolous insects. Here we show how a Conservation on sub-Antarctic islands faces several chal- DNA barcoding approach has helped to overcome such a lenges, of which the most signiWcant are the impacts of bio- problem associated with the likely establishment of an alien logical invasions and ongoing climate change. The biotas moth species on Marion Island. The discovery of unidentiW- of several islands in the region have been compromised by able immatures of a noctuid moth species, 5 km from the invasive alien species, and climate change is not only hav- research station, suggested that a new moth species had col- ing direct impacts on the distribution and abundance of spe- onized the island. EVorts to identify the larvae by conven- cies on the islands, but, perhaps more signiWcantly, appears tional means or by rearing to the adult stage failed. to be facilitating the establishment of introduced species However, sequencing of 617 bp of the mitochondrial cyto- and mediating the interaction between indigenous and chrome oxidase subunit I gene, and comparison of the introduced organisms, to the beneWt of the latter (Bergstrom sequence data with sequences on GENBANK and the bar- and Chown 1999; Frenot et al. 2005; Chown et al. 2007). In coding of life database enabled us to identify the species as response to these challenges, management plans have been Agrotis ipsilon (Hufnagel), a species of which adults had developed for the large majority of the sub-Antarctic previously been found regularly at the research station. Dis- islands (de Villiers et al. 2006). covery of immatures of this species, some distance from the All of these plans include an explicit focus on prevention research station, suggests that a population may have estab- of the introduction of alien species, and on the need for the lished. It is recommended that steps to be taken to eradicate early detection of species that may have established despite the species from Marion Island. steps taken to block the known pathways of introduction. However, as has long been recognized, the detection of newly established species at a stage when eradication is still feasible is often problematic owing to small population S. L. Chown (&) · B. J. van Vuuren sizes and the cryptic nature of many species (Wittenberg Department of Botany and Zoology, and Cock 2005). Detection requires a focussed and ongoing Centre for Invasion Biology, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa programme of surveys, especially in areas of high human e-mail: [email protected] activity, and a means to demonstrate that the species in question has established a reproducing population, and is B. J. Sinclair not merely represented by transient adults that may have Department of Biology, The University of Western Ontario, come ashore, but have little likelihood of establishing. London, ON N6A 5B7, Canada Determining whether a reproducing population has a long-term 123 Polar Biol likelihood of invading is more problematic, but a precaution- Materials and methods ary approach would suggest that any established reproducing population should be considered capable of crossing the Background to the problem remaining barriers to invasion. For holometabolous insects, which undergo complete The indigenous Lepidoptera on Marion Island comprises metamorphosis, it may be especially diYcult to establish two (or perhaps three) Xightless species in the families whether a reproducing population exists owing to substan- Yponomeutidae (Embryonopsis halticella) and Tineidae tial diVerences in the morphology of the immature and (Pringleophaga marioni and possibly P. kerguelensis) adult stages, and the focus of many keys and monographs (Chown et al. 2002). However, since detailed entomologi- on the adult stages (e.g. Scholtz and Holm 1985). In conse- cal studies began, reports of winged adult moths, and a sin- quence, when immatures are found during surveys, it may gle species of butterXy, have been made (CraVord et al. be diYcult to determine what species is concerned, even 1986; Hänel et al. 1998). The establishment of one species, though the existence of a reproducing population may be Plutella xylostella (Plutellidae) was veriWed by simulta- considered high and an identiWcation to family or subfamily neous capture of adults and immatures, and by rearing level readily done. Because the likelihood of successful (CraVord and Chown 1987), but for the most part it was invasion in one site is often strongly related to the extent to assumed that adults of the other species (mostly noctuid which a species has successfully invaded elsewhere (e.g. moths and the butterXy Vanessa cardui; CraVord et al. Rejmánek et al. 2005), identiWcation of the animals to spe- 1986; Chown et al. 2002) were transients that arrived either cies level is important. Moreover, it may also assist distin- by natural means or via fresh produce. The latter was deliv- guishing natural immigration from human-assisted ered to the research station on the island in an annual basis processes. (and sometimes more frequently) until the mid-1990s, fol- One way in which the problem with identiWcations of lowing which the Prince Edward Islands Management Plan immatures may be overcome is through DNA barcoding prevented any fresh produce from being brought ashore (Armstrong and Ball 2005). One of the largest hurdles (Prince Edward Islands Management Plan, Department of faced by a DNA-based approach is to standardize methods Environmental AVairs and Tourism, Pretoria, 1996). How- to allow meaningful comparisons between diVerent data ever, in 1996/1997, three larvae of a noctuid moth species sets generated in diVerent laboratories by diVerent research- were discovered at the research station (Hänel et al. 1998). ers. Barcoding protocols provide such a method including They were identiWed provisionally as belonging to an Agro- standardization of the gene fragment targeted. It is based on tis species, but no further identiWcation or concrete veriWca- the notion that most taxonomic groups are characterized by tion of their identity was possible. Because only adult low intraspeciWc variation compared to larger divergences noctuid moths were found in all subsequent regular surveys between species, which allow conWdence in species assign- (see Lee et al. 2007 for detailed descriptions), it had been ments (Moritz and Cicero 2004). The mitochondrial COI assumed that all noctuid moths are transient aliens and that gene was proposed as a region of choice and the usefulness any population at the station had gone extinct (Chown et al. of this region was demonstrated through case studies 2002). The noctuids were considered alien because human (Hebert et al. 2003). Although the approach is controversial activities have artiWcially extended their ranges and (e.g. RubinoV 2006 and references therein), and may fail to increased their abundances (see Pynek et al. 2004 for addi- achieve its aims when evolutionarily complex lineages are tional discussion). involved (Monaghan et al. 2006), it does provide a useful In April 2004, three larvae of an unidentiWed noctuid Wrst step for identifying species in situations where other moth, most resembling an Agrotis species, were collected methods might not be available or where high-diversity from an abandoned Wandering Albatross (Diomedea exu- constrains the eYciency of more traditional techniques. In lans) nest at Archway Bay. The site is some 5 km to the consequence, DNA barcoding has the potential to assist in south–east of the research station, and the immatures were the detection of established alien species, especially where found during sampling for P. marioni caterpillars, which the association of adults and immatures has proven prob- reach high-densities in these nests (Sinclair and Chown lematic, or where the expertise or tools to identify imma- 2006). A signiWcant eVort was made to rear the immatures ture stages or rear them to the adult stage are not available for three reasons. First, they were found some distance or have failed. Here we use an example of the long-stand- downwind (the prevailing winds are north-westerly) from ing detection of adult temperate Lepidoptera on sub-Ant- the research station. Second, their discovery was made at a arctic Marion Island, and the more recent reports of time suggesting, given their stage and assumptions about unidentiWed immatures, as a case study to demonstrate how slow life cycle duration at a relatively low temperature (5– DNA barcoding may serve to assist with the early detection 10°C), that oviposition likely took place when no relief ves- of established populations of alien species. sel was present in the area. Third, they were discovered 123 Polar Biol

8 years after the last larvae had been found. Unfortunately, where the highest similarity was again found to an A. ipsi- these eVorts failed for two of the larvae, one of which died lon sequence, reported from Tanzania (Fig. 1). after pupating in the laboratory after a few months. In con- These results demonstrate unequivocally that the larva sequence, the Wnal larva was immediately killed and pre- collected from the abandoned Wandering Albatross nest served in 99% ethanol for DNA barcoding as an alternative at Archway Bay is that of Agrotis ipsilon (Hufnagel). means to establish its identity. Given the morphological similarity of the other two larvae to the one sequenced, it is assumed that the other noc- Barcoding laboratory protocols tuid moth individuals collected from the nest belong to the same species. Adults of A. ipsilon have been recorded We followed international DNA barcoding protocols as from Marion Island since the early 1980s. Moreover, described by the Consortium for the Barcode of Life given the frequency with which adults have been (CBOL) (http://www.barcoding.si.edu) and targeted 650 bp recorded, Hänel et al. (1998) presumed that the larvae of the COI gene using the primers LCO1490 and HCO2198 captured at the research station were A. ipsilon. Owing to described by Folmer et al. (1994). DNA was extracted the age and preservation method of these individuals, this using a commercial DNA extraction kit (DNeasy Blood & assumption cannot be conWrmed. Tissue Kit, Qiagen). PCR products were gel puriWed with The collection of three immature individuals of A. ipsi- the Wizard SV gel and PCR clean-up system (Promega). lon, 5 km distant from the research station, suggests that a Sequencing reactions were performed using BigDye® small population of this species is established on Marion chemistry (version 3, Applied Biostystems). The puriWed Island. Subsequent searches have failed to reveal additional sequencing product was run on an ABI 3100 automated individuals, but if densities are low, as they tend to be at the sequencer (Applied Biosystems). Electropherograms of the start of a colonization event, then such an absence does not raw data were edited with BioEdit 7.0.5 (Hall 1999). The necessarily indicate that the population has gone extinct. sequence generated in this study was deposited in GenBank Agrotis ipsilon larvae in North America survive periods of (accession number EU 239817). 14 days or more at ¡5°C, and tolerate transient ice forma- tion, while eggs tolerate ¡10°C for periods of days (Beck Sequence analyses

To demonstrate the usefulness of a barcoding approach to Agrotis venerabilis (Canada) address taxonomic uncertainties, the sequence generated Agrotis munda (Australia) from the larval specimen collected on Marion Island was compared with data available through the Barcoding of Life Database (BOLD, http://www.barcodinglife.org). In addi- tion, the sequence obtained from the Marion Island specimen was compared with data available through the Agrotis infusa (Australia) National Center for Biotechnology Information (Genbank, http://www.ncbi.nlm.nih.gov).

Results and discussion

COI data for 18 species of Agrotis are available through BOLD, including A. ipsilon. Data for only two species (A. ipsilon and A. volubilis) are available through Genbank. To Agrotis ipsilon (Canada, USA) identifying the larvae collected from Archway Bay, the 617 bp of the COI gene were compared to the data available through Genbank and BOLD. The Marion Island sequence showed the highest similarity to A. ipsilon 0.5 Agrotis ipsilon (Tanzania, and Marion Island specimen) (AF549736). Indeed, the Marion Island sequence diVered at only two positions from the Genbank reference sequence Fig. 1 The neighbour-joining tree obtained when requesting a “tree- (both were transitional changes). The Marion Island based identiWcation” including the 99 nearest matches in bold (http:// www.barcodinglife.org). The Marion Island specimen is indicated at V A. volubilis sequence di ered from (AF549702) at 24 sites the bottom of the topology. The tree has been edited such that all indi- (»3.8% sequence divergence). This result was conWrmed vidual sequences are not shown as such, but are grouped according to when the Marion Island sequence was submitted to BOLD, species and locality 123 Polar Biol

1988). This species therefore likely has the potential to sur- Beck SD (1988) Cold acclimation of Agrotis ipsilon (Lepidoptera: vive temperature extremes observed at the lower elevations Noctuidae). Ann Entomol Soc Am 81:964–968 Bergstrom D, Chown SL (1999) Life at the front: history, ecology and on Marion Island (see Slabber and Chown 2005). Luck- change on southern ocean islands. Trends Ecol Evol 14:472–477 mann et al. (1976) estimated a minimum developmental Chown SL, McGeoch MA, Marshall DJ (2002) Diversity and conser- threshold of 10.5°C for A. ipsilon, which is lower than the vation of invertebrates on the sub-Antarctic Prince Edward temperatures normally experienced in most habitats on Islands. Afr Entomol 10:67–82 Chown SL, Slabber S, McGeoch MA, Janion C, Leinaas HP (2007) Marion Island. However, these individuals were collected Phenotypic plasticity mediates climate change responses among from an abandoned albatross nest, and work on P. marioni invasive and indigenous arthropods. Proc Roy Soc Lond B has demonstrated that occupied nests have substantially 274:2531–2537 higher temperatures than their surrounding environments, Convey P (2005) Recent lepidopteran records from sub-Antarctic South Georgia. Polar Biol 28:108–110 which certainly promotes the growth and survival of P. CraVord JE, Chown SL (1987) Plutella xylostella L. (Lepidoptera, marioni (Sinclair and Chown 2006), and would likely pro- Plutellidae) on Marion Island. J Entomol Soc S Afr 50:259–260 vide temperatures above the developmental threshold for A. CraVord JE, Scholtz CH, Chown SL (1986) The insects of sub-Antarc- ipsilon. tic Marion and Prince Edward Islands; with a bibliography of entomology of the Kerguelen Biogeographical Province. S Afr J If A. ipsilon does become established more broadly it Antarct Res 16:41–84 may have considerable consequences for ecosystem func- de Villiers MS, Cooper J, Carmichael N, Glass JP, Liddle GM, McIvor tioning, and perhaps also for P. marioni if it prefers the E et al (2006) Conservation management at Southern Ocean Is- nests of albatross species. Generalist herbivorous Lepidop- lands: towards the development of best-practice guidelines. Po- larforschung 75:113–131 tera are not characteristic of the indigenous fauna (Chown de Villiers MS, Cooper J (2008) Conservation and management. In: et al. 2002), and at least on Marion Island, A. ipsilon larvae Chown SL, Froneman PW (eds) The Prince Edward Islands. seem to require a habitat very similar to that preferred by P. Land-sea interactions in a changing ecosystem. Sun, Stellenbosch marioni. Elsewhere, A. ipsilon is known to feed on a wide (in press) Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA prim- variety of plant species, and it is a common pest that can ers for ampliWcation of mitochondrial cytochrome c oxidase sub- reach high-densities and cause substantial damage to crops unit I from diverse metazoan invertebrates. Mol Mar Biol (Annecke and Moran 1982). In consequence, additional Biotechnol 3:294–299 eVorts (especially light-trapping for adults and vegetation Frenot Y, Chown SL, Whinam J, Selkirk PM, Convey P, Skotnicki M, Bergstrom DM (2005) Biological invasions in the Antarctic: surveys for larvae) should be made to determine whether extent, impacts and implications. Biol Rev 80:45–72 the species has established permanently, and if so how Greenslade P, Farrow RA, Smith JMB (1999) Long distance migration widely distributed and abundant it is. Eradication pro- of insects to a subantarctic island. J Biogeogr 26:1161–1167 grammes have been developed for other species at Marion Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Island (the isopod Porcellio scaber and the grasses Agrostis Symp Ser 41:95–98 gigantea and Elymus repens) (de Villiers and Cooper Hänel C, Chown SL, Davies L (1998) Records of alien insect species 2008), and it is reasonable to suggest that such a pro- from sub-Antarctic Marion and South Georgia Islands. Afr Ento- gramme should be investigated for A. ipsilon and imple- mol 6:366–369 Hebert PDN, Cywinska A, Ball SL, de Waard JR (2003) Biological mented if feasible. More generally, the migratory nature of identiWcations through DNA barcodes. Proc Roy Soc Lond B A. ipsilon (Showers et al. 1989), records of this species as a 270:313–321 transient alien at other sub-Antarctic locations (Greenslade Lee JE, Slabber S, Jansen van Vuuren B, van Noort S, Chown SL et al. 1999; Convey 2005), and its ability to survive in the (2007) Colonisation of sub-Antarctic Marion Island by a non- indigenous aphid parasitoid Aphidius matricariae (Hymenoptera, immature stage at Marion Island suggest that it may prove Braconidae). Polar Biol 30:1195–1201 to be one of the most immediate risks in terms of species Luckmann WH, Shaw JT, Sherrod DW, Ruesnick WG (1976) establishment and invasion on the sub-Antarctic islands. Developmental rate of the black cutworm. J Econ Entomol 69:386–388 Acknowledgments We thank Richard Mercer for assistance in the Monaghan MT, Balke M, Pons J, Vogler AP (2006) Beyond barcodes: W Weld, Jennifer E. Lee, Philip Pugh, and two anonymous referees for complex DNA taxonomy of a South Paci c Island radiation. Proc comments on the manuscript, the South African National Antarctic Roy Soc Lond B 273:887–893 Programme for logistic support, and the National Research Foundation Moritz C, Cicero C (2004) DNA barcoding: promise and pitfalls. PLoS for partial funding (SANAP Grant 2069543 to BJvV). Biol 2:1529–1531 Pynek P, Richardson DM, Rejmánek M, Webster GL, Williamson M, Kirschner J (2004) Alien plants in checklists and Xoras: towards better communication between taxonomists and ecologists. Tax- References on 53:131–143 Rejmánek M, Richardson DM, Higgins SI, Pitcairn MJ, Grotkopp E Annecke DP, Moran VC (1982) Insects and mites of cultivated plants (2005) Ecology of invasive plants: state of the art. In: Mooney in South Africa. Butterworths, Durban HA, Mack RA, McNeely JA, Neville LE, Schei PJ, Waage JK Armstrong KF, Ball SL (2005) DNA barcodes for biosecurity: invasive (eds) Invasive alien species. A new synthesis. Island Press, Wash- species identiWcation. Philos Trans R Soc Lond B 360:1813–1823 ington, pp 104–161 123 Polar Biol

RubinoV D (2006) Utility of mitochondrial DNA barcodes in species Slabber S, Chown SL (2005) DiVerential responses of thermal toler- conservation. Conserv Biol 20:1026–1033 ance to acclimation in the sub-Antarctic rove beetle Halmaeusa Scholtz CH, Holm E (1985) Insects of Southern Africa. Butterworths, atriceps. Physiol Entomol 30:195–204 Durban Wittenberg R, Cock MJW (2005) Best practices for the prevention Showers WB, Whitford F, Smelser RB, Keaster AJ, Robinson JF, Lo- and management of invasive alien species. In: Mooney HA, pez JD, Taylor SE (1989) Direct evidence for meteorologically Mack RA, McNeely JA, Neville LE, Schei PJ, Waage JK (eds) driven long-range dispersal of an economically important moth. Invasive alien species. A new synthesis. Island Press, Washington, Ecology 70:987–992 pp 209–232 Sinclair BJ, Chown SL (2006) Catterpillars beneWt from thermal ecosystem engineering by wandering albatrosses on sub-Antarctic Marion Island. Biol Lett 2:51–54

123