Colonization History, Host Distribution, Anthropogenic Influence and Landscape Features Shape Populations of White Pine Blister Rust, an Invasive Alien Tree Pathogen

Abstract

White pine blister rust is caused by the fungal pathogen Cronartium ribicola J.C. Fisch (Basidiomycota, Pucciniales). This invasive alien pathogen was introduced into North America at the beginning of the 20th century on pine seedlings imported from Europe and has caused serious economic and ecological impacts. In this study, we applied a population and landscape genetics approach to understand the patterns of introduction and colonization as well as population structure and migration of C. ribicola. We characterized 1,292 samples of C. ribicola from 66 geographic locations in North America using single nucleotide polymorphisms (SNPs) and evaluated the effect of landscape features, host distribution, and colonization history on the structure of these pathogen populations. We identified eastern and western genetic populations in North America that are strongly differentiated. Genetic diversity is two to five times higher in eastern populations than in western ones, which can be explained by the repeated accidental introductions of the pathogen into northeastern North America compared with a single documented introduction into western North America. These distinct genetic populations are maintained by a barrier to gene flow that corresponds to a region where host connectivity is interrupted. Furthermore, additional cryptic spatial differentiation was identified in western populations. This differentiation corresponds to landscape features, such as mountain ranges, and also to host connectivity. We also detected genetic differentiation between the pathogen populations in natural stands and plantations, an indication that anthropogenic movement of this pathogen still takes place. These results highlight the importance of monitoring this invasive alien tree pathogen to prevent admixture of eastern and western populations where different pathogen races occur.

Effects of Host Variability on the Spread of Invasive Forest Diseases

“…Similarly, a fragmented distribution of some white pines (e.g., Pinus strobiformis Engelm.) in western North America could only retard, but not prevent their infection by the white pine blister rust fungus C. ribicola [75]. When hosts are scattered, landscape connectivity (i.e., how the landscape structure facilitates or impedes the disease spread among patches [76]) plays an important role for disease spread.…”

Pathogens in space: Advancing understanding of pathogen dynamics and disease ecology through landscape genetics

“…Landscape genetic clustering and assignment methods have largely built upon classical methods from population genetics (e.g., principal components analysis, STRUCTURE, Pritchard, Stephens, & Donnelly, ) by incorporating spatial information (e.g., GENELAND, Guillot, Mortier, & Estoup, ; sPCA, Jombart, Devillard, Dufour, & Pontier, ) and environmental heterogeneity (e.g., constrained ordination, Anderson & Willis, ; POPS, Jay, ) into estimates of population structure and providing quantitative estimates of ancestry for each individual (François & Waits, ). Clustering methods have been relatively popular in studying pathogens and implemented for the inference of landscape barriers affecting both host (Addis, Lowe, Hossack, & Allendorf, ; Cote, Garant, Robert, Mainguy, & Pelletier, ; Cullingham, Kyle, Pond, Rees, & White, ; Frantz, Cellina, Krier, Schley, & Burke, ) and microparasite (Brar et al., ; Rieux et al., ) spatial genetic variation. Edge detection methods, such as Monmonier's maximum difference algorithm, (Monmonier, ) have also been used to detect landscape barriers to transmission in pathogen studies (Carrel et al., ; Joannon et al., ).…”

“…However, it must be noted that many of the genetic clustering and assignment methods presented here rely on classical population genetics models that generally do not apply to microorganisms. Therefore, the use of such methods in the study of pathogens is often limited to the inference of pathogen movement using host gene flow, with the exception of some fungal pathogens (Brar et al., ; Rieux et al., ). Some recent methods for identifying spatial population structure are free of classical population genetic assumptions, such as LOCALDIFF (Duforet‐Frebourg & Blum, ), EEMS (Petkova, Novembre, & Stephens, ) and MAPI (Piry et al., ), and can be applied to pathogens directly without these potential constraints.…”

“…Landscape effects on viral transmission are typically studied using phylogenetic approaches (Fountain‐Jones, Craft et al., ; Joannon, Lavigne, Lecoq, & Desbiez, ; Streicker et al., ; Young et al., ). To date, pathogens with larger and more slowly mutating genomes, such as protozoans (Carrel et al., ; Lo et al., ) and fungi (Brar et al., ; Rieux, De Bellaire, Zapater, Ravigne, & Carlier, ), have been studied using population genetics‐based methods with highly variable microsatellite and SNP loci. New methods based on next‐generation sequencing technologies such as targeted enrichment techniques (Lee et al., ) are helping to address challenges with sequencing viruses and other pathogens, facilitating greater use of pathogen genetic data in future landscape genetic studies.…”

Direct and Indirect Interactions between Landscape Structure and Invasive or Overabundant Species

“…In particular, the spatial distribution of host species may define the landscape structure for pathogens and pests at large spatial scales and, consequently, strongly influence movements across regions [30,31]. For example, landscape connectivity, partly due to how landscape structure affects host distribution, can affect the spread of pathogens [32] and pests, including the maritime pine bast scale (Matuscoccus feytaudi) [33] and white pine blister rust (Cronartium ribicola) [34].…”