DNA-based Methods for Studying the Diet ofMarine Predators

Diets of large marine predators have been extensively studied to assessinteractions with fisheries, monitor links between diet and reproductive success, andunderstand trophic interactions in marine ecosystems. Since marine species can rarelybe observed foraging directly, most studies rely on the identification of prey remainsin stomach contents or faeces to determine the prey items being consumed. While thisapproach has provided a wealth of information, it has several limitations resultingprimarily from difficulties identifying digested prey and from biased recovery ofremains due to differential digestion. My thesis explores the use of molecular geneticmethods in dietary studies of large marine predators. DNA-based identificationtechniques have been used in several diet studies, but the methods and applicationsare still in the early stages of development. Through a number of studies, Iinvestigated the ability to recover genetic data from various dietary samples using arange of genetic techniques.A) Genetic screening for prey in the gut contents from a giant squid - I assessed theuse of polymerase chain reaction (PCR)-based methods for isolation of prey DNAfrom an Architeuthis gut content sample. A taxonomically informative molecularmarker was selected and a screening method developed using denaturing gradient gelelectrophoresis. The methodology was used to identify prey from otherwiseunidentifiable hard-part remains and the amorphous slurry component of the squid gutsample. The techniques developed here provided a framework for later chapters.B) Analysis of prey DNA in faeces of captive sea lionsPart I: DNA detection, distribution and signal persistence - A feeding trial withcaptive Steller sea lions (Eumetopias jubatus) was carried out to investigate the use ofgenetic faecal analysis as a tool to study diet. I used group-specific PCR detection todetermine: (i) the reliability of prey DNA recovery, (ii) the distribution of prey DNAwithin faeces and (iii) the persistence of the genetic signal after a prey item wasremoved from the diet. The proportions of prey DNA in several samples were alsodetermined using a clone library approach to determine if DNA quantification couldprovide semi-quantitative diet composition data. Results show that the prey DNAcould be reliably detected in sea lion faeces and the genetic signal could persist insamples up to 48 hours after ingestion. Proportions of prey DNA isolated from faeceswere roughly proportional to the mass of the prey items consumed.Part II: DNA quantification - Quantitative real-time PCR was used to furtherinvestigate if quantitative diet composition data could be obtained throughquantification of the DNA present in faeces. I quantified the relative amounts of DNAin three fish species being fed to captive sea lions, then determined the amount ofDNA recovered from these prey items in the sea lions - faeces. The results indicatethat diet composition estimates based on the relative amounts of DNA in faeces canbe biased due to the differential survival of DNA from different fish species; however,these biases may be less than those commonly observed in the conventional analysisof prey hard remains. C) Quantification of damage in DNA recovered from faecal samples - I developed ageneral method to quantify the frequency of DNA damage present in specific generegions. The technique was applied to assess the amount of DNA damage in predatorand prey DNA recovered from sea lion faeces. The estimated frequency of DNAdamage was always higher for the prey DNA than for the predator DNA within afaecal sample. The findings have implications for marker development andcomparison of results obtained in future DNA-based diet studies.D) Studying seabird diet through genetic analysis of faeces - I investigated the diet ofmacaroni penguins (Eudyptes chrysolophus) through conventional analysis ofstomach contents and through the analysis of prey DNA extracted from faeces.Genetic data was obtained from faecal samples using PCR tests to determine thepresence or absence of DNA from potential diet items and also using a clone libraryapproach. Approximately half of the faecal samples tested positive for one or more ofthe prey groups targeted with PCR tests. Euphausiid DNA was most commonlydetected in early stages of chick rearing and DNA from a myctophid fish wasprevalent in faeces collected later; this trend mirrored the data obtained from thestomach contents. Analysis of prey sequences in 'universal'clone libraries revealed ahighly biased recovery of sequences from fish prey; this bias is most likely caused bythe use of degenerate primers with a higher binding affinity for fish DNA templatecompared to DNA from other prey groups. Results obtained from the genetic andtraditional approaches are compared, and potential future applications of the genetictechniques to studying seabird diet are discussed.This series of studies has contributed significantly to our understanding of thestrengths and the limitations of DNA-based diet analysis. The work identifiessituations where genetic methods can be successfully applied to study the diet ofmarine predators and provides guidance for future studies in this emerging field.