Revolution in food web analysis and trophic ecology: diet analysis by DNA and stable isotope analysis



D. D. Heath, Fax: (519) 971 3616;


Characterization of energy flow in ecosystems is one of the primary goals of ecology, and the analysis of trophic interactions and food web dynamics is key to quantifying energy flow. Predator-prey interactions define the majority of trophic interactions and food web dynamics, and visual analysis of stomach, gut or fecal content composition is the technique traditionally used to quantify predator-prey interactions. Unfortunately such techniques may be biased and inaccurate due to variation in digestion rates (Sheppard & Hardwood 2005); however, those limitations can be largely overcome with new technology. In the last 20 years, the use of molecular genetic techniques in ecology has exploded (King et al. 2008). The growing availability of molecular genetic methods and data has fostered the use of PCR-based techniques to accurately distinguish and identify prey items in stomach, gut and fecal samples. In this month’s issue of Molecular Ecology Resources, Corse et al. (2010) describe and apply a new approach to quantifying predator-prey relationships using an ecosystem-level genetic characterization of available and consumed prey in European freshwater habitats (Fig. 1a). In this issue of Molecular Ecology, Hardy et al. (2010) marry the molecular genetic analysis of prey with a stable isotope (SI) analysis of trophic interactions in an Australian reservoir community (Fig. 1b). Both papers demonstrate novel and innovative approaches to an old problem – how do we effectively explore food webs and energy movement in ecosystems?

Figure 1.

 The aquatic habitats used for two studies of diet and trophic interactions that employed molecular genetic and stable isotope analyses. Panel a: Example of Rhone basin habitat (France) where fish diet was determined using PCR to classify prey to a series of ecological clades (photo by Emmanuel Corse). Panel b: A weir pool on the lower Murray River (Australia) where food web and prey use was evaluated using a combination of advanced molecular genetic and stable isotope analyses (photo credit: CSIRO).

Alternative and imaginative methods for diet analysis have been used since 1946, starting with immunological techniques and continuing with sophisticated DNA-based methods, paralleling technological advances in molecular ecology (Fig. 2). Immunological techniques used to identify prey are diverse and include antigen-antibody interactions in solution (e.g. agglutination, precipitation reactions, immuno- electrophoresis) as well as solid-phase techniques (e.g. ELISA, radio-immune assays; Boreham & Ohiagu 1978). Indeed, immunological techniques are still used and are extremely helpful (Fig. 2) even though the technique is labor intensive and costly. Stable isotope (SI) analysis (primarily carbon (reported as ∂13C) and nitrogen (∂15N)) is another, more widely applied technique that has been used since the 1970s to characterize food webs (Deniro & Epstein 1978, 1981). Isotope fractionation events in living organisms most often result in the enrichment of the heavier isotope of nitrogen (increase in ∂15N) relative to food items, providing a relative estimate of trophic position. Carbon SI values show less enrichment between diet and organism but often vary between photosynthetic sources (e.g. aquatic phytoplankton vs. terrestrial plants) and habitats (i.e. marine vs. freshwater) and thus can be used to characterize carbon sources or organisms and food webs (see Peterson & Fry 1987; Hardy et al. 2010). DNA-PCR based techniques have been utilized for the identification of prey items from stomach, gut or fecal contents using DNA hybridization (Rosel & Kocher 2002), cloning and sequencing (Deagle et al. 2005) and presence/absence of diagnostic PCR products on agarose gels (Gorokhova 2006), among others (see Teletchea 2009). Molecular genetic techniques used for diet analyses range from straightforward to more complex approaches that utilize cutting edge molecular genetic technology, such as DNA microarrays (Hardy et al. 2010) and high-throughput parallel sequencing (Pegard et al. 2009). However, technically simple and widely accessible approaches such as that developed by Corse et al. (2010) provides a powerful tool for the characterization of complex food webs that can potentially be used by ecologists not familiar with advanced molecular techniques. Furthermore, the approach developed by Corse et al. (2010) is faster and less expensive than more sophisticated molecular genetic techniques, and thus could be used for applied management or conservation purposes. Technological advances are ongoing and new methods may emerge that we have not yet considered; moreover, by combining existing technologies we can overcome limitations inherent in some techniques and gain new and clearer insights into ecological processes.

Figure 2.

 The number of published studies (per 5- or 10-year interval) using indirect methods for diet determination from gut or fecal content analysis. The solid line shows the number of studies using molecular genetic techniques, while the dashed line shows the studies using immunological techniques. Reference list is available as online Supplementary Material.

In the January issue of Molecular Ecology Resources Corse et al. (2010) describe an innovative technique to analyze trophic interactions in complex food webs via diet analysis. Corse et al. (2010) characterized prey communities by grouping them into ecological clades, where the clades were defined by molecular genetic and habitat similarities. Hence their goal was to characterize diet, and thus trophic interactions, by functional and genetically similar prey groups. They designed 34 sets of 18S ribosomal RNA gene (rDNA) PCR primers that identify all ecological clades in seven different microhabitats in a European river ecosystem. By having a complete database of prey types available in each habitat, they minimized the possibility of underestimating (or missing altogether) diet components not amplified by existing or novel ‘universal’ primers. Furthermore, their work allows analysis of which prey is preferred, that is, predator selectivity. Using this approach Corse et al. (2010) successfully show subtly different feeding habits in three closely related species of fish in the same ecosystem.

In a related study in this issue of Molecular Ecology, Hardy et al. (2010) combined SI and molecular genetic diet analysis to achieve greater resolution in their food web analysis of a freshwater pool community on the lower Murray River in southern Australia. Their goals were to determine the source of organic matter entering their study ecosystem and whether there was significant seasonal variation in the source of the organic matter. For these ambitious goals, they used two conceptually different approaches; SI analysis and DNA-PCR based diet analysis. SI analysis can detect trophic interactions that are not expected, but often fails to provide specific trophic interactions because isotopic values in potential prey can overlap. Hardy et al. (2010) also used PCR to amplify rDNA subunit regions and the subsequent sub-cloning and sequencing allowed the authors to identify diet to available metazoan, fungal, protozoan and plant taxa. The authors also develop a DNA microarray printed with synthetic rDNA oligonucleotides which is then hybridized with gut content PCR products to identify prey species. The advantage of a microarray approach is that it is cost effective and large numbers of environmental or gut samples can be screened quickly. Although the technological aspects of Hardy et al.’s (2010) food web analyses are impressive, of more conceptual interest are their comparisons of the results from the two approaches. The SI analysis allowed them to identify food web anomalies that were not evident based on the DNA-PCR approach (i.e. terrestrial carbon input and seasonal changes driven by the action of methanogens). However, their DNA-PCR approach provided more specific trophic (predator-prey) interaction information than would have been possible from SI analysis alone. One of the most important aspects of this study is the potential for using such a sensitive approach for early detection of anthropological and natural environmental changes in the ecosystem. Although Hardy et al. (2010) were able to get a more detailed picture of food web dynamics (transfer of carbon and energy through the food web) than in Corse et al.’s (2010) study, Hardy et al. (2010) technologically advanced approach is perhaps better suited to experimental applications to model systems, since it requires substantially greater technical infrastructure and expertise.

Ongoing change in the environment is inevitable, especially in the face of new environmental challenges such as climate change or invasive species. It is thus critically important to have tools to effectively quantify early responses in the community. Characterizing predator-prey interactions is a very important component of ecosystem-level studies, particularly because some species will modify their diet in response to environmental change or perturbation. Corse et al. (2010) and Hardy et al. (2010) provide novel and innovative approaches to indirect diet analyses, and both studies highlight the potential for such trophic analyses to detect environmental changes due to anthropological effects. Indirect diet analyses are becoming increasingly common in ecological research (Fig. 2) and this reflects the critical need for such information in ecology and conservation. A quantitative understanding of predator-prey dynamics and potential food sources will not only better define trophic interactions and food web structure, it will help us better understand community ecology at a fundamental level.

L.C.-M. is a doctoral student studying the ecological factors driving fisheries recruitment success in the Great Lakes. Her main interests are related to community ecology and resource management. D.D.H. is a professor whose main research interests fall within evolutionary and ecological genetics. Both authors are very involved in applying and developing diverse molecular genetic techniques to aid in the understanding of biotic and abiotic interactions in the environment.