Evaluating the genetic and epigenetic contributions to porcine maternal infanticide and their potential use to identify at risk animals

Lead Research Organisation: University of Cambridge
Department Name: Pathology


Maternal infanticide behaviour towards the newborn has been observed in several mammalian species including mouse, rat, pig and human. There is considerable evidence that this complex abnormal maternal behaviour has an important inherited component but is also influenced by interacting environmental factors.

In the pig breeding industry, piglet loss as a result of maternal infanticide has both an economic and welfare impact. From the size of the breeding herds in the UK and Europe (400,000 and 15,000,000 sows respectively - DEFRA estimates in 2007) it is conservatively estimated that 20,000-30,000 piglets are at risk in the UK and 750,000-1,125,000 in Europe at each parity. It would, therefore, be important to use any understanding of the causes of porcine maternal infanticide to reduce the welfare and economic impact in the pig breeding industry.

We have been studying porcine maternal infanticide in order to identify regions of the pig genetic material that influence maternal behaviour. These genetic studies have uncovered a number of chromosome segments that carry genes and DNA sequences that may be important in determining this infanticide behaviour. The activity of genes or DNA sequences that regulate genes can be altered either by direct changes in the sequence of DNA (mutation) or by a modification of the DNA that does not alter the sequence but does change its activity; this is known as an epigenetic modification and can be both transmissible and established by interacting environmental factors.

In this project we aim to investigate the candidate genes and sequences identified by our genetic studies for evidence of DNA sequences change (mutation) and DNA modification (epigenetic change) to provide direct evidence linking these genes and sequences to the maternal infanticide behaviour. This will be done by comparing the DNA in these candidate regions between affected and unaffected sows. The variations in DNA sequence or DNA modification observed between the two groups of animals may provide the basis for developing predictive tests that could be useful in identifying sows predisposed to maternal infanticide behaviour. We will evaluate this possibility by testing significant differences between affected and unaffected animals on a large collection of animals where we have a well documented record of their maternal behaviour at birth and in the immediate postnatal period. In addition to potentially allowing the design of predictive tests, an improved understanding of the genes, pathways and underlying biology of maternal infanticide in pigs offers routes to the development of possible treatment strategies.

Technical Summary

Maternal infanticide (piglet savaging) is a significant welfare and economic issue for the pig breeding industry. Our Genome-wide case/control association studies (cohort of 1200 sows covering 4 pig breeds analysed on the Illumina Single Nucleotide Polymorphism (SNP) array) have revealed several loci that make a genetic contribution to savaging behaviour. Previous heritability estimates show greater daughter-dam than daughter-sire transmission of the phenotype indicating a parent-of-origin effect and involvement of epigenetic mechanisms.

This project seeks to use:

1. Sequence capture and deep sequencing of the most significantly associated genomic intervals (collectively 3.9 Mb of DNA) to provide evidence of DNA sequence variation (mutation) of candidate genes and regulatory sequences in savaging sows.
2. Methylation studies (using sequencing of bisulphite converted DNA) of promoter regions of significantly associated genes and regulatory sequences to assess the involvement of epigenetic modification in the savaging phenotype.
3. Extended genotyping of savaging pedigrees to exploit long range phasing to assess partitioning of alleles of paternal or maternal origin between affected and unaffected animals. This will indicate genomic intervals where there is a parent-of-origin effect that may coincide with genetic intervals that have already been identified or define new contributing loci.

These studies will provide information of sequence variation and epigenetic modification that could be used to design predictive tests to identify sows predisposed to savaging. We will evaluate the usefulness of savaging-associated haplotypes and epigenetic change in a wider cohort of animals across 4 pig breeds used in our original genome-wide association studies. The data will also indicate what genes and genetic pathways are involved in the biology of maternal infanticide and, therefore, potentially suggest strategies for developing therapeutic intervention.

Planned Impact

The outputs from this research proposal can potentially impact on and benefit a wide spectrum of users: other research workers in the field; industrial consumers of research findings; clinical services and healthcare systems; government and agencies concerned with the development and implementation of policy in a number of areas; the general public as a result of the research output driven actions of these various categories of user.
This work will be of interest to other research workers interested in behavioural genetics and the role of the interaction of genes with environment in shaping behavioural outcomes. There are several facets that the research community working in this area may wish to develop. First, they will wish to establish a better understanding of the genes, mechanisms and biology underpinning behavioural responses. This opens up possibilities of delineating bio-markers that may be useful diagnostic tools to identify individuals and organisms predisposed to certain types of behaviour and routes to the design of new therapeutic drugs to modify behaviour. Second, they will wish to gain a better understanding of the involvement of epigenetic modification of the genome in determining disease outcomes. This will reveal what genes and genetic pathways are important in this respect and begin to throw light on what environmental cues are important and at what time in life these act to determine disease phenotype in the immediate and future trajectory of an organism or individual.
There are a number of ways in which the industrial sector can use and benefit from this research and the stimulus it may have on other researchers in the field. First, the pig breeding industry will benefit from any development of bio -markers that will aid breeding management and avoid the welfare costs and economic losses incurred through savaging sows. Reductions in piglet mortality, associated with improved maternal care, represent a rare scenario within the agriculture sector where benefits can simultaneously be achieved in production efficiency (improving profitability and reducing environmental impact) and animal welfare standards. Improvements in sow maternal behaviour have the additional advantage of being applicable across different system types and changes in these in the future. Second, the pharmaceutical industry: the provision of drug targets (whether they be specific genes or pathways critical to a particular behavioural response) offers the opportunity of the development of new drugs that could be used in the treatment of disease and the management of animal breeding. These are both clear economic contributions to the country. Finally, the public will benefit from the above through better management of livestock in the pig and potentially other animal husbandry industries.
In the longer term, results may benefit the public through social and public health policy aimed at improved management of mental health issues if the same genes are found to be involved in human neuropsychiatric disorders.
The research outcomes also have relevance to broader government and its agencies developing food production policy. Proactive policy can be developed designed to avoid economic loss and welfare costs.


10 25 50
publication icon
Quilter CR (2012) An association and haplotype analysis of porcine maternal infanticide: a model for human puerperal psychosis? in American journal of medical genetics. Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics

Description This study has used a three different genome wide mapping studies to identify regions of the porcine genome associated with maternal infanticide. These studies have revealed several intervals on different pig chromosomes that:
1. Show genetic association and, therefore, may harbour genes that are involved in maternal aggressive behaviour.
2. Show a parent of origin effect indicating preferential transmission from one or other parent. This is consistent with the involvement of epigenetic mechanisms influencing maternal behaviour and with previous observation that the maternal aggressive trait shows a high dam/daughter heritability.

To investigate the molecular basis of maternal infanticidal behaviour, the project has undertaken a systematic sequence and methylation analysis of the chromosome intervals defined by the genetic analysis described above. The purpose of these molecular studies is to identify (a) any sequence variation in genes or functional non-coding RNAs and (b) methylation (epigenetic) patterns that are associated with aberrant maternal behaviour.

We have completed gathering of all the sequence and methylation data and are currently undertaking detailed bioinformatics analysis of this large body of data.
Exploitation Route On completion of analysis of the sequence and methylation data, the outputs may reveal sequence variants and methylation patterns that identify key causative genes. These markers have the potential to form the basis of a predictive test for the pig breeding industry and to inform studies on puerperal psychosis in humans.
Sectors Agriculture, Food and Drink,Healthcare