MARINE ECOLOGY PROGRESS SERIES Vol. 291: 103–113, 2005 Published April 28 Mar Ecol Prog Ser Denaturing gradient gel electrophoresis (DGGE) as a tool for identification of marine nematodes A. A. Cook1, 2, 6, P. Bhadury3, N. J. Debenham1, B. H. M. Meldal1, 4, M. L. Blaxter5, G. R. Smerdon3, M. C. Austen3, P. J. D. Lambshead1, A. D. Rogers2,* 1Nematode Research Group, Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD, UK 2Natural Environment Research Council, British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK 3Plymouth Marine Laboratory, Prospect Place, West Hoe, Plymouth PL1 3DH, UK 4School of Ocean and Earth Science, Southampton Oceanography Centre, University of Southampton, Waterfront Campus, European Way, Southampton SO14 3ZH, UK 5Institute of Cell, Animal and Population Biology, Ashworth Laboratories, University of Edinburgh, King’s Buildings, West Mains Road, Edinburgh EH9 3JT, UK 6Present address: The International Seabed Authority, 14–20 Port Royal Street, Kingston, Jamaica ABSTRACT: Many phyla of marine invertebrates are difficult to identify using conventional morpho- logical taxonomy. Larvae of a wider set of phyla are also difficult to identify as a result of conserva- tion of morphology between species or because morphological characters are destroyed during sam- pling and preservation. DNA sequence analysis has the potential for identification of marine organisms to the species level. However, sequence analysis of specimens is time-consuming and impractical when species diversity is very high and densities of individuals huge, as they are in many marine habitats. The effectiveness of the 18S rRNA gene sequences for identification of one species- rich marine group, the Nematoda, is analysed. Following identification of variable regions of the 18S rRNA gene, primers were designed to amplify a small segment of sequences suitable for denaturing gradient gel electrophoresis (DGGE). The effectiveness of DGGE for identifying individual species is 2 analysed. DGGE analysis of natural communities of nematodes detected less than ⁄3 of the species present. This fraction of the community probably represents the abundant species in the original samples. It is concluded that DGGE is not a useful tool for analysis of species richness in marine com- munities as it fails to detect rare species of which there are usually many in the marine benthic envi- ronment. However, DGGE may be a useful method for detecting changes in communities that influ- ence the abundance of the most common taxa. KEY WORDS: Marine nematodes · 18S ribosomal DNA · Denaturing gradient gel electrophoresis Resale or republication not permitted without written consent of the publisher INTRODUCTION represent some of the most species-rich communities of metazoans on the planet and include, in order of The taxonomic crisis is well recognised throughout numerical importance, nematodes, crustaceans, mol- the scientific community (e.g. May 1997, Hebert et al. luscs and polychaetes (e.g. Grassle & Maciolek 1992). 2003a). It arises from a falling number of taxonomists The fact that these groups are studied by relatively few and a narrow focus of existing expertise that neglects taxonomists is a significant limitation in marine ecol- many highly diverse groups of organisms (May 1997, ogy and the assessment of the impacts on biodiversity Tautz et al. 2003). This is particularly the case for of global climate change and human activities. organisms that live in many marine habitats, especially As stated by Tautz et al. (2003), ‘Insights into the in the sediments of the seabed. These habitats may stability or change of animal and plant guilds require *Corresponding author. Email: [email protected] © Inter-Research 2005 · www.int-res.com 104 Mar Ecol Prog Ser 291: 103–113, 2005 species identification on a broad scale.’ Such broad- these are thought to exhibit concerted evolution there scale species identification for any sample of marine are many cases where intragenomic variation has been sediment requires a huge effort on the part of several detected, especially in the internal transcribed spacer taxonomic experts to sort and identify individual spec- regions (e.g. ITS 1 and ITS 2; Harris & Crandall 2000, imens. The time, effort and costs of doing this for rou- Chu et al. 2001). tine biomonitoring associated with specific impact The second major step in using DNA sequences for assessments, or longer-term environmental studies, species identification is to accelerate the screening are multiplied by the number of samples. However, the process so that it truly becomes a more rapid and cost- problems associated with such studies cannot merely effective method than manual identification of speci- be reduced to practical considerations of scale. Mor- mens. Without such approaches, DNA barcodes will phological identification of marine species, belonging only be an adjunct to traditional methods for species to the most speciose groups, is problematic through identification from environmental samples. Previously, subjectivity of opinion of individual taxonomists, dif- screening methods have been developed for small ferences in the skills of taxonomists (compare special- numbers of species and vary from the use of species- ist museum taxonomists with non-specialist parataxon- specific PCRs to PCR product-mobility assays using omists) and problems of nomenclature and the electrophoresis (for review on identification of marine existence of cryptic species. Synonyms, for example, larvae see Rogers 2001; for terrestrial nematodes, are thought to over-inflate estimates of species diver- Foucher & Wilson 2002, Foucher et al. 2004). sity by 20% (May 1997) and sibling species have been Free-living marine nematodes typify many of the commonly found across numerous marine taxa in all taxonomic problems associated with poorly studied habitats (Knowlton 1993). Such problems become invertebrate groups. Their identification to species particularly intractable when samples from different level can be difficult, time-consuming and expensive. geographic localities are compared, especially on Specimens have to be mounted on slides and examined regional, oceanic or global scales. individually using high-power interference microscopy The adoption of DNA-based technologies offers one (Warwick et al. 1998). Moreover, only a small fraction route to increase the speed and cost-effectiveness of of species have been or will be described (Lambshead species identification for ecological surveys or biomon- 1993, Hugot et al. 2001). Therefore, only a few special- itoring. This approach is based on the use of DNA ists with extensive taxonomic knowledge can work sequences as ‘barcodes’ for species (Floyd et al. 2002, with the group. In addition, identification of juvenile Hebert et al. 2003a). The first step in this process is the meiofaunal taxa is often only possible by inference identification of regions of the genome that offer suffi- based on the presence of adults in the same samples cient variation to resolve closely related species as well (Litvaitis et al. 1994). The morphological characters as higher taxa (e.g. Tautz et al. 2003). Such approaches used for classification may also be unreliable, and require considerable effort in pilot studies that ‘cali- known species, especially those used for biomonitor- brate’ sequence information by comparing the genetic ing, may be complexes of sibling species with different distances expected between known species and higher functional responses (Warwick & Robinson 2000). taxa. The efficacy of mitochondrial genes as molecular Free-living nematodes are an important component barcodes for species identification has been explored of the marine ecosystem and are a potentially ideal recently (Hebert et al. 2003a,b). However, mitochondr- taxocene for biomonitoring and testing ecological ial DNA can be subject to a number of problems. These theory. This is because they are a ubiquitous, abun- include mitochondrial transfer between species, trans- dant and hyper-diverse group of organisms in marine, posing of mitochondrial genes to the nucleus (Tautz et terrestrial and freshwater environments. For ecological al. 2003), slow rates of mitochondrial evolution leading studies, work at the species level provides a greater to low divergences in sequences between species (e.g. resolution than that at higher taxonomic levels (Cook Cnidaria; France & Hoover 2002, Hebert et al. 2003b) 2001). Whilst there have been many investigations into and differing arrangements of the mitochondrial the use of functional grouping as a means of classifica- genome in different taxa that prevent ‘universal’ tion (Wieser 1952, Romeyn & Bouwman 1983, Jensen primers from amplifying target regions (e.g. nema- 1987, Thistle et al. 1995, Moens & Vincx 1997), recent todes; Okimoto et al. 1991, 1992, Blouin 1998, Keddie work has suggested that, until a better understanding et al. 1998, Lavrov & Brown 2001). Nuclear genes may of species biology is reached, this approach is of lim- provide an alternative or companion to mitochondrial ited use (Cook 2001). Moreover, the functional biology barcode sequences (e.g. Floyd et al. 2002). However, of marine nematode species in the deep sea is often the most commonly used barcode regions of nuclear unknown. DNA—e.g. small subunit (ssu) and large subunit (lsu) The aim of this study was to assess sequence varia- rDNA—belong to multigene families, and although tion in the ribosomal ssu for its power to resolve