Large-scale biodiversity inventory projects are becoming increasingly popular (see Janzen et al. 2009, deWaard et al. 2009, Basset et al. 2012, Tänzler et al. 2012, Hausmann et al. 2013, Telfer et al. 2015, Aagaard et al. 2016, Geiger et al. 2016, Miller et al. 2016, Wilson et al. 2016), particularly after the method of DNA barcoding has been established as a fast and efficient species discovery and identification tool (see Hebert et al. 2003, Packer et al. 2009, Ratnasingham and Hebert 2013).

The present paper introduces the ongoing large-scale biodiversity discovery project IndoBioSys (Indonesian Biodiversity Discovery and Information System). The project is focusing on establishing a vertebrate and invertebrate diversity discovery pipeline and a biodiversity information system in Indonesia. Samples are processed through an integrated sorting pipeline that has been set up and optimized at the Zoologische Staatssammlung in Munich, Germany, and tested at the Museum Zoologicum Bogoriense in Cibinong, Indonesia. For DNA barcoding, samples were submitted to the Canadian Centre for DNA Barcoding in Guelph, Canada.

A targeted, biodiversity knowledge-based screening approach for the identification of novel active biological compounds is a central aspect of the project. For this purpose, the initiative aims at providing a comprehensive biodiversity inventory, including descriptions of new species with sequence data that are publicly available in the Barcode of Life Data Systems (BOLD) and also integrated in the Indonesian Biodiversity Information System (IBIS), aimed at providing access to existing and new information on Indonesia’s biological resources. The present paper focuses on the insect order Lepidoptera that is among the largest and economically most important groups of insects.

Although general collection protocols compiled for molecular studies are available (e.g. http://lepbarcoding.org/protocols.php, Hajibabaei et al. 2005, Ivanova and Kuzmina 2013), they are rather standard and focused mostly on the pipelines for the DNA sequencing data analyses using already available material (i.e. pinned or papered museum specimens and field samples). Considering that DNA sequencing techniques are advanced and well-established but strongly depend on the quality of specimens, our collection protocol helps overcoming challenges in obtaining high-quality samples suitable for both morphological and DNA analyses and presents a workflow that secures availability of tissues, abstracts and data for future studies.

The highly diverse biota of Indonesia comprise both Oriental and Australian elements, with a high proportion of endemic species (see Holloway 1985, Holloway 1994, Holloway 1996, Holloway 1997, Schmidt 2005, Schmidt 2013, Schmidt 2015). The IndoBioSys study area, where extensive sampling has been carried out, is located in the Mount Halimun-Salak National Park, a conservation area in the Indonesian province of West Java (Fig. 1). Covering 113,357 hectares, the National Park harbours the largest sub-mountain forest in West Java (Fig. 2). The vegetation comprises primary and secondary forest and cultivated areas, with primary forest covering almost 70% of the area (Hartono et al. 2007).

Figure 1.  

Map of Western Java showing the Halimun-Salak national Park (hatched). The study area is located in the Western Javan montane rain forest ecozone (dark green). Red dot in inset map shows the location of the study area in the Sundaland region. Map created with SimpleMappr (http://www.simplemappr.net).

Figure 2.  

Halimun National Park (Indonesia, West Java), one of the collecting sites.

Moths of Indonesia are poorly studied (see Schmidt 2015) and no checklists of moths occuring in Indonesia have been published. Some ecological studies have been conducted on Indonesian Lepidoptera, including several recent surveys of Macrolepidoptera in secondary forests, national parks, protected forests and mountane populated areas of Maluku Islands, West Papua, Central Kalimantan, and Java. As a result, lists of Macrolepidoptera have been compiled, comprising 160 species from 14 families collected on the island of Ternate, 178 species from 19 families collected at the base of the Foja Mountain Nature Reserve (Sutrisno 2012), up to 214 species from 21 families on different plots of Gunung Patuha protected forest (Sutrisno 2009), up to 278 species from 19 families on different plots of Central Kalimantan (Sutrisno 2005), up to 297 species from 19 families in the Nusa Barong Nature Reserve (Sutrisno 2007), and up to 846 species on different plots of Halimun-Salak National Park (Sutrisno 2008). No definite conclusions have been drawn about the total number of species in certain localities as further studies were needed (see Sutrisno 2005, Sutrisno 2007, Sutrisno 2012). Many undescribed species were expected to occur in the study areas (see Sutrisno 2005) but no new species were formally described, and the identification of specimens was mainly based on external morphological characters and study of the types was beyond the frame of these primarily environmental studies. It is highly probable that sibling species may have been overlooked (Schmidt, unpubl. data). For comparison, biodiversity studies conducted on Borneo revealed more than 1,000 species of Geometridae in this area (Holloway 1994, Holloway 1996, Holloway 1997). Considering the country size and diversity of habitats, Indonesia is expected to be one of the main biodiversity hotspots, and further biodiversity assessment studies based on DNA barcoding would rapidly increase the knowlege of the largely unknown Indonesian moth fauna.

Sampling of Lepidoptera has been conducted at three plots in the study area at different elevations during one dry and one wet season. Material has been collected during the day using a sweep net and at night using standard light sources having a strong emission in the ultraviolet range of the spectrum. At each plot, two light traps at a distance of about 30-40 meters from each other were operated simultaneously: (1) a UV light trap with two light sources, including black light lamps and fluorescent light tubes (8 W each) running from rechargeable 12 V batteries, (2) a mercury vapour light trap (125 W) running from a generator. The light sources were placed in front of a white sheet and protected from the rain by an umbrella. Active sampling (no killing traps) was preferred. Moths were one of the major target groups for the project. Collection- and general preparation methods, as well as digital imaging of specimens, have been described in numerous studies (e.g. Common and Waterhouse 1972, Klots 1973, Common 1990, Landry and Landry 1994, Prendini et al. 2002, Häuser et al. 2005, Paulson 2005, Gibb and Oseto 2006, Krogmann and Holstein 2010, Infusino et al. 2017) and in various contributions on the web (e.g. Wheeler et al. 2001, Warren 2015). Obtained specimens were treated according to the barcoding protocol developed in the Biodiversity Institute of Ontario (Guelph, Canada) (Ivanova et al. 2006, Wilson 2012) and are included in BOLD (Ratnasingham and Hebert 2007).

Collecting equipment

Equipment for preservation

Equipment for DNA barcoding

Fieldwork. Specimen sampling

Post-fieldwork

Pre-lab preparation protocol

Post-lab activity. Storing of the vouchers

Frequently asked questions

The workflow. The collection protocol presented in this article has been successfully employed for field- and pre-lab activities that were part of the IndoBioSys project in the years 2015 and 2016. The workflow is presented in Fig. 3. During the first stage of the survey of the Indonesia’s Lepidoptera diversity we focused on a few target groups, including the Geometridae. More than thirty 96-well lysis plates that are routinely used for DNA barcoding by the Canadian Centre for DNA Barcoding (CCDB) were processed using a high-throughput protocol, and several plates were additionally processed at the Zoologische Staatssammlung (ZSM, Munich).

Figure 3.  

The workflow from collecting to storage of specimens of Lepidoptera in our IndoBioSys project.

Success rates. About 98% of the sequenced specimens of the Geometridae processed using the present collecting and preparation protocol yielded sequences with more than 500 base pairs, meeting the length requirement for DNA barcode status (see Ratnasingham and Hebert 2007). When other protocols were implemented (e.g. for Malaise trap samples or samples stored without using sufficient amounts of silica gel) less than 88% of the specimens yielded sequences with more than 500 base pairs. Performing a statistical analysis to compare different protocols is beyond the scope of the present study.

Treatment of specimen with ethyl acetate. Sequencing for inventory projects like IndoBioSys requires killing and preserving the specimens in a DNA-friendly way. The Lepidoptera were killed with ethyl acetate vapours and in most cases removed from the killing jars after about 20-30 minutes (but not longer than 40 minutes) to make sure that the DNA is not damaged. Our results confirm the findings by Willows-Munro and Schoeman (2014) suggesting that ethyl acetate can be successfully used to collect specimens for DNA analysis. However, the impact of long-term preservation of Lepidoptera in ethyl acetate has not been studied.

Remarks to a recently publised alternative preserving and vouchering method. In a recent publication by Cho et al. (2016) a procedure for preserving and storing Lepidoptera tissues has been presented. This procedure aims at creating accessible and easily visualized “wing vouchers” of individual Lepidoptera specimens while preserving the remainder of the insect in a cryogenic freezer for molecular research, with the wings preserved in protective plastic holders so that both dorsal and ventral patterns and colours can be easily viewed. However, this method involves removing a pair of wings (in some cases all wings) from a specimen. It is not just the “aesthetically pleasing display” that matters. Scissors are used to cut the wings at the base, whereby the wing base gets damaged which hampers the study of the morphological characters (e.g. venation in both sexes and androconial scales in males). Besides, this procedure is obviously not recommended for treatment of type specimens, considering the amount of undescribed rare taxa processed while conducting research related to a large-scale biodiversity discovery projects in a diverse and poorly studied region. Another drawback of the “wing vouchering” approach is its inapplicability to small moths (Cho et al. 2016). Our proposed method does not suffer from these drawbacks because we remove legs from the freshly collected specimens and preserve them for molecular analysis. An important part of the procedure is that the tissues get dry as fast as possible, either in a well closing box with silica gel beads or in a drying oven. Our method is applicable to all Lepidoptera.

Additionally, reliable storage methods were described and discussed by Knölke et al. (2005) and Lopez-Vaamonde et al. (2012).

Towards future outcomes. The ultimate objective of our study is to present a methodological pipeline assisting in successful sampling, preparation, preservation, morphological and molecular analyses and secure storage of high-quality material for a biodiversity assessment which combines the expertise gained through the DNA barcoding and the taxonomist’s knowledge.

The protocol has been prepared as part of the IndoBioSys project funded by the German Federal Ministry of Education and Research (BMBF) and supported by the MfN and ZSM. The Indonesian government also partly contributed to the financial support from the DIPA PROJECT of the Research Center for Biologi 2016-2017. Thanks to the head of Halimun-Salak national park and the Ministry of Environment and Forestry for giving access permits to the national park. The project was carried out by the German side (Museum für Naturkunde, Leibniz-Institut für Evolutions- und Biodiversitätsforschung, Berlin and Zoologische Staatssammlung, Munich) in cooperation with the Indonesian participants (Museum Zoologicum Bogoriense, LIPI, Bogor). Many thanks to Christoph Häuser, Thomas von Rintelen, Kristina von Rintelen, Wolfram Mey and Johannes Frisch (Museum für Naturkunde – Leibniz-Institut für Evolutions- und Biodiversitätsforschung, Berlin), Rosichon Ubaidillah, Darmawan, Pramesa, Sarino, Sih Kahono and Giyanto (Museum Zoologicum Bogoriense, LIPI, Bogor), Hasmiandy Hamid (Andalas University, Padang), Andrew Polaszek (Natural History Museum, London), Andreas Floren (University of Würzburg), Michael Balke and Ditta Amran Balke (ZSM, Munich) for planning the logistics and/or taking part in the field trips. Special thanks to Gabriela Silva, John Wilson, Pavel Stoev and Yasen Mutafchiev (BDJ) for positive and helpful comments on the manuscript. Sincere thanks to the director of the Biodiversity Institute of Ontario, Paul Hebert, and the barcoding team (Guelph, Canada) for DNA sequencing and providing the bioinformatic tools for data-analysis.