Skip to main content

Elucidating the relationship between heterozygosity and fitness in a natural marine mammal population

Final Report Summary - FURSEALFITNESS (Elucidating the relationship between heterozygosity and fitness in a natural marine mammal population)

Many important fitness traits including parasite resistance, survivorship and reproductive success often correlate with heterozygosity in natural populations. Where documented, such heterozygosity-fitness correlations (HFCs) have the clear potential to influence interactions between pathogens and their hosts and the evolution of mate choice. However, because most studies use only around ten microsatellite markers, the proximate mechanisms underlying HFCs remain obscure. My four-year CIG proposal aimed to deploy 768 Single Nucleotide Polymorphisms (SNPs) to elucidate the primary mechanism underlying HFCs for a variety of fitness traits in a model pinniped system, the Antarctic fur seal.

Research objectives
To uncover the basis of HFCs in wild animal populations requires a model system in which a range of HFCs have already been documented. An ideal opportunity is presented by an exceptionally detailed long-term study of a breeding colony of Antarctic fur seals (Arctocephalus gazella) at Bird Island, South Georgia. Here, my colleagues at the British Antarctic Survey have constructed an aerial walkway that provides access to all of the animals that come ashore, thereby allowing the collection of detailed individual-based observational and genetic data. By genotyping 4900 samples collected over 17 years at 9 microsatellites, I have uncovered HFCs for virtually every fitness trait so far examined, although admittedly most of these have been measured in males.
The main aim of this proposal was to determine whether HFCs in this species are due to inbreeding depression or whether they could involve a small number of genes of large effect size (so called 'local effects'). The following objectives were established: (Objective 1) Extend the current microsatellite dataset to cover 21 consecutive breeding seasons and test whether HFCs are also present in females; (Objective 2) Generate a panel of 768 SNPs including 100–200 targeted within candidate genes, and genotype these in 1440 fur seal individuals; (Objective 3) Analyse the resulting SNP data for associations with fitness. Test the hypothesis that inbreeding is the primary mechanism accounting for HFCs; (Objective 4) If inbreeding is found not to play a major role, test the hypothesis that genes involved in immunity and growth are disproportionately important; (Objective 5) Construct a linkage map to identify any genomic regions that may be disproportionately important to fitness.

Description of work performed
The period of the CIG has been highly productive, resulting in no fewer than 41 publications in peer reviewed international journals, 13 of which were directly supported by this grant. These include three publications as either first and senior author in Nature and Proceedings of the National Academy of Sciences of the United States of America as well as numerous papers in leading international field journals such as BMC Genomics, Molecular Ecology and Evolution. Almost all of the objectives set out in the original proposal have been met and the remaining ones should be complete within the next twelve months. Much additional work has also been undertaken that feeds into and compliments the proposed research.
To briefly summarise, genetic sampling and microsatellite genotyping have both progressed according to schedule. Analysis of the enlarged microsatellite dataset together with three decades of detailed individual-based data from the study colony is complete and shows that heterozygosity is strongly associated with multiple fitness traits in females, ranging from early survival to breeding success. In order to develop SNPs that can be assayed in the study population, I used next-generation sequencing to generate a vastly improved transcriptome. By including tissues collected from the spleens of animals that died of natural causes, I was able to increase the representation of immune genes five-fold relative to the original transcriptome based on skin, and thereby enrich the SNP set for key candidate genes including the Major Histocompatibility Complex (MHC). I also identified and validated additional SNPs in the fur seal by cross-amplifying SNPs from a canine SNP chip.
Initially I planned to use Illumina's BeadXpress system for the SNP genotyping. However, validation rates of custom SNPs, defined as the proportion of putative SNPs that are verified to be polymorphic in a population, are often very low. I therefore undertook extensive exploration of the SNPs residing within the fur seal transcriptome and developed a protocol for maximising the probability that they would successfully validate. This involved assembling a draft genome of the Antarctic fur seal (assembly size: 2.41Gb; scaffold/contig N50: 3.1Mb/27.5kb) and using this to map the probe sequences of 144 putative SNPs previously genotyped in 480 individuals. By doing so, I was able both to demonstrate that sequence uniqueness and proximity to intron-exon boundaries play an important role and to generate a predictive model allowing the selection of transcriptomic SNPs with a higher than average probability (95%) of successfully validating.
Unfortunately, the BeadXpress was recently discontinued so I had to switch genotyping technologies. I therefore developed a method based on Restriction Site Associated DNA (RAD) sequencing (that is capable of accurately estimating inbreeding in any organism. A pilot study that I carried out in a related pinniped species, the harbour seal, was recently published in the high-ranking international journal, Proceedings of the National Academy of Sciences of the United States of America. My research group also recently developed an R package for analysing RAD and other types of high density SNP data for inbreeding. My PhD student, recruited during the course of the CIG thanks to a successful grant application to the German Science Foundation, has now generated a large RAD dataset. This is currently being analysed for inbreeding and a linkage map will shortly be constructed.

Main results achieved so far
Among the many achievements of this project, two main research highlights emerge. First, my colleague Jaume Forcada and I analysed three decades of individual-based life history and genetic data from the fur seal study population, which has been declining. We have found that climate change has reduced prey abundance causing a significant reduction in birth weight over this period. However, the mean age and size of adult females recruiting into the breeding population are both increasing over time. Such females are significantly more heterozygous than both their non-recruiting siblings and their own mothers, and have driven a long-term increase in breeding female heterozygosity of 8.5% per generation. Thus, climate change has led to increasing selection against lower quality, homozygous females. This work resulted in a senior author publication in Nature. Second, I developed a novel approach for estimating genome-wide heterozygosity using high-throughput sequencing. Applied to an HFC for parasite load in harbour seals, genome-wide heterozygosity explained five times more fitness variation than the heterozygosity of a small panel of microsatellites. This demonstrates unequivocally that the HFC is due to inbreeding depression, and also suggests that heterozygosity may explain far more variation in fitness than previously envisaged. This work was published as a first author paper in Proceedings of the National Academy of Sciences of the United States of America.

Expected final results, impact and use
The harbour seal results strongly suggest that RAD sequencing will be very powerful for detecting inbreeding depression if this is indeed the mechanism underlying HFCs in the Antarctic fur seal. Furthermore, if inbreeding depression accounts for the reduced fitness of homozygous individuals, the proportion of trait variance explained by heterozygosity should increase in proportion to the number of additional markers deployed. If so, this would make an important contribution to the classic ‘nature versus nurture’ debate, strengthening greatly the role of nature.
In conclusion, the Antarctic fur seal study offers an unprecedented opportunity to understand the relationship between genetic diversity and fitness in free-ranging, natural population of an important marine mammal. This CIG proposal has made significant process towards elucidating how key fitness traits, including female survivorship under increasing climate stress, correlate with an individual’s genotype. Not only will this work contribute towards understanding and managing a key player in the fragile Antarctic ecosystem, but it also has the potential to make a major impact on our understanding of the link between genotype and fitness and how this feeds into general principles that underpin the evolution of breeding behaviour.

Dr Joe Hoffman

University of Bielefeld
Department of Animal Behaviour
PO Box 10 01 31
33501 Bielefeld

Phone: +49 521 106-2711