Application of Genomics to Understand Forest-Pest Interactions

Catherine I. Cullingham, Janice E.K. Cooke, David W. Coltman Department of Biological Sciences, University of Alberta, Edmonton

Pine forests in western North America have been devastated by an unprecedented outbreak of mountain pine beetle-fungal complex. Primarily, lodgepole pine forests in Canada have been affected resulting in economic loss and ecosystem damage. More concerning is the range expansion of the beetle/fungi both north and east resulting in exposure to a new host species, jack pine (Cullingham et al. 2011), a sister species to lodgepole pine. This naïve host is distributed throughout the Boreal forest across Canada and into the eastern United States putting an entire ecosystem at risk.

Are there genetic differences within and between lodgepole and jack pine?

Do these genetic differences affect mountain pine beetle colonizing or reproductive success (i.e. spread potential)?

How do environmental factors interact with the host-insect relationship?

Questions

Mountain pine beetle attack data in British Columbia and Alberta in three time periods: 1959-1998, 1999-2005 and 2006-2009 to demonstrate the magnitude of the recent outbreak. Attack data is overlaid pine distribution and density.

Introduction

Results

USE GENOMICS TO IDENTIFY AND UNDERSTAND HOST-GENETIC RISK FACTORS

Wound Drought

Genes of potential importance are expressed

in response to the change in environment

Extract RNA which represents the expressed

portion of the genome (transcriptome)

Needle

Stem

Root

RNA used as a source for high-throughput

sequencing

Methods Lodgepole Jack pine

1 597 295 fragments 1 558 772 fragments

Combined

Assembly (MIRA, Newbler)

SNP detection/selection Minimum coverage: 15X

MAF: 10% SNP > 100bp

Flanking region

1536 SNPs

399 457 362

Many thanks to all of the technical and intellectual support from Sophie Dang, Matt Bryman, Cory Davis, Stephanie Boychuk, Joël Fillon, Dominik Royko, Jasmine Janes, Patrick James, Jill Hamilton, Adriana Arango, Josh Miller, René Malenfant, William Peachman, and Lucky Chauhan.

Acknowledgements

References Balding (2003) Theor. Popul. Biol. 63: 221-230. Beaumont & Balding (2004) Mol. Ecol. 13: 969-980. Cullingham et al. (2011) Mol. Ecol. 20, 2157-2171. Foll & Gaggiotti (2008) Genetics, 180, 977-993. Gompert & Buerkle (2009) Mol. Ecol. 18, 1207-1224.

Implications

Characterize SNPs in lodgepole, jack pine and their hybrids across naïve and non-naïve stands

OUTLIERS

Identify loci that deviate from neutral expectations within and between species, “outlier analysis” (Beaumont & Balding 2004)

Spatially informed o MatSAM (Joost et al. 2008)

Spatially uninformed o BayeSCAN (Foll & Gaggiotti

2008) o BayesFST (Balding 2003) o Introgress (Gompert &

Buerkle 2009)

Across data-sets, 136 SNP loci showed patterns of non-neutral inheritance. The majority of which were found when comparing lodgepole and jack pine (“All”). These include transcription factors, protein degradation, growth regulators, synthesis, energy production and water transport.

All

Jack

8 130

2

0 0 0

Lodge

4

Interpolated surfaces of three SNP loci that that encode for proteins of potential interest that displayed patterns of non-neutral inheritance, and an association with the environment.

This research addresses knowledge gaps that exist given the range-expansion of the beetle to novel environments. From here we will assess the function of these candidate loci to identify whether they influence spread potential. We can then use our interpolated surfaces as inputs in spread-risk and economic models to mitigate the impacts of this outbreak, and assess potential future outbreaks.

Population Genomics

Physiology & Function

Ecology

MPB vector Fungal Pathogen

Pine host

Spread-risk models

Millions of DNA fragments were generated and processed using a bioinformatic work-flow