Forest tree breeding in face of climate change

Yousry A. El-Kassaby

Department of Forest Sciences and Conservation, Faculty of Forestry

The University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4

Email: y.el-kassaby@ubc.ca

Traditional tree breeding programs are long-term and resource-dependent.  They often follow the classical recurrent selection scheme with its repeated cycles of breeding, testing, and selection (Allard, 1999).  Each breeding cycle requires the intensive management of multiple populations (breeding, testing, and production), so gain, diversity, and the build-up of co-ancestry are effectively managed (Namkoong et al., 1988).  These programs are dedicated to meeting the planting demands of specific ecological targets (breeding zones), thus progeny testing and seed deployment are often restricted within these targets (Ying and Yanchuk, 2006).  Substantial and unrestricted genetic movement is exercised within the confines of these specific breeding zones as offspring (new recombinations) are planted throughout.  Notwithstanding the man-made unrestricted genetic movement within breeding zones, these breeding programs are in essence, spatially static and might be slow in dealing with the increased mobility required to cope with rapid climate change.  Forest tree populations may not be able to track shifting favourable environmental conditions mediated by climate change considering the documented misalignment between rates of gene flow (via seed and pollen) and environmental change (bioclimatic envelopes) (Kremer et al., 2012).  Even under the best scenarios where gene flow is not constrained by fragmentation or restricted by physical barriers, human intervention is warranted considering the anticipated intensity and directionality of climate change (Vitt et al., 2010).

The extent of genetic diversity in species’ peripheral populations is often considered to be lower than that present in their central counterparts; however, there is increased evidence supporting the role of gene flow as a replenishing factor leading to increased genetic diversity as compared to “standing” diversity (Yeaman and Jarvis 2006) with some suggestions that gene flow will introduce better adapted genes as compared to local ones, specifically under changing climate scenarios (Alleaume-Benharira et al., 2006).  Additionally, it is expected that the warming trend will increase growth and fecundity, improve survival, and promote germination and recruitment (Reich and Oleksn 2008; Andalo et al. 2005; Savolainen et al. 2011; Kellomaki et al., 1997), thus chances for increased productivity is anticipated in the new favourable environmental conditions.

Here, I propose an in situ breeding approach that is particularly suitable to meeting the increased pace of genetic material movement dictated by changes in response to climate.  In this scheme, testing and selection are concurrently and naturally conducted at the species’ front edge, thus the selected material would be most adapted to the anticipated future range expansion with minimal photoperiodic transfer.  In this scheme, existing natural populations and/or plantations at the species’ latitudinal front will play a dual role in providing already tested genotypes (testing) and selection will be exclusively based on adaptive attributes related to climate such as cold and drought tolerance and timing of growth and reproduction (selection).

Genetic evaluation of individuals will be based on using large batteries of anonymous DNA markers preferably distributed throughout the genome obtained through Next Generation Sequencing (NGS).  The proposed method capitalizes on the availability of: 1) affordable NGS platforms for providing large numbers of randomly distributed genomic markers for any species without the need for their reference genomes and 2) innovative quantitative genetics analytical approaches that are anchored around utilizing genomic markers.  The simultaneous retrospective testing and selection among cohorts in natural unstructured settings faces two major challenges caused by variation in age and spatial arrangement among the evaluated individuals; however, these could be overcome by using age as covariate and the utilization of spatial statistics to mitigate spatially-caused differences.  It is noteworthy to mention that most adaptive attributes are minimally affected by age and/or special arrangements.

DNA fingerprinting will be exclusively based on using one of the NGS methods such as genotyping-by-sequencing (Chen et al., 2013) or whole-exome targeted sequencing (Neves et al., 2013) as they are effective in the identification of variation at the single nucleotide level (SNP: single nucleotide polymorphisms) needed for the subsequent genetic evaluation among naturally tested unstructured cohorts.  Genetic evaluation will be based on either: 1) the genomic best linear unbiased prediction (GBLUP) method that utilizes the “realized kinship” matrix produced from the DNA fingerprinting data which does not require prior genealogical information about the individuals under evaluation (vanRaden, 2008) and/or 2) genome-based prediction methods (genomic selection) which utilize information across the entire genome simultaneously to explain the observed phenotypic variability of complex polygenic traits (Meuwissen, 2001).  The proposed genetic analyses are perfectly suited to unstructured populations where prior knowledge on genealogy is often lacking.  The selected individuals, in turn, would form the production populations for the production of adapted stock for planting at the new favourable environmental conditions often located at the species’ latitudinal front edge.

In this presentation, I will briefly review the classical recurrent selection scheme and provide examples of the genomic-based methods using data from structured and unstructured forest tree populations.

 

Allard (1999) Principles of plant breeding, Wiley. NY;

Alleaume-Benharira et al. (2006) J Evoly Biol 19:203-215;

Andalo et al. (2005) For Ecol Manage 205:169-182;

Chen et al. (2013) Tree Genet genomes (in press);

Kellomäki et al. (1997) Climatic Change 37:683-708;

Kremer et al. (2012) Ecol Letters 15:378-392;

Meuwissen et al. (2001) Genetics 157:1819-1829;

Namkoong et al.(1988) Theor Appl Genet Monog 11;

Neves et al. (2013) Plant J 75:146-156;

Reich and Oleksyn (2008) Ecol Letters 11:588-597;

Savolainen et al. (2011) J Hered 102:526-536;

vanRaden (2008) J Dairy Sci 91:4414-4423;

Vitt et al. (2010) Biol Conserv 143:18-27;

Yeaman and Jarvis (2006) Proc Royal Soc B-Biol Sci 273:1587-1593;

Ying and Yanchuk(2006) For Ecol Manage 227:1-13