Genomics of the expanding pine pathogen Lecanosticta acicola reveals patterns of ongoing genetic admixture

Abstract

Lecanosticta acicola is the causal agent for brown spot needle blight that affects pine trees across the northern hemisphere. Based on marker genes and microsatellite data, two distinct lineages have been identified that were introduced into Europe on two separate occasions. Despite their overall distinct geographic distribution, they have been found to coexist in regions of northern Spain and France. Here, we present the first genome-wide study of Lecanosticta acicola, including assembly of the reference genome and a population genomics analysis of 70 natural isolates from northern Spain. We show that most of the isolates belong to the southern lineage but show signs of introgression with northern lineage isolates, indicating mating between the two lineages. We also identify phenotypic differences between the two lineages based on the activity profiles of 20 enzymes, with introgressed strains being more phenotypically similar to members of the southern lineage. In conclusion, we show undergoing genetic admixture between the two main lineages of L. acicola in a region of recent expansion.

Importance: Lecanosticta acicola is a fungal pathogen causing severe defoliation, growth reduction, and even death in more than 70 conifer species. Despite the increasing incidence of this species, little is known about its population dynamics. Two divergent lineages have been described that have now been found together in regions of France and Spain, but it is unknown how these mixed populations evolve. Here we present the first reference genome for this important plant pathogenic fungi and use it to study the population genomics of 70 isolates from an affected forest in the north of Spain. We find signs of introgression between the two main lineages, indicating that active mating is occurring in this region which could propitiate the appearance of novel traits in this species. We also study the phenotypic differences across this population based on enzymatic activities on 20 compounds.

Keywords: Lecanosticta acicola; admixture; comparative genomics; needle blight; plant pathogen; population genomics.

Fig 1

(A)- Phylogenetic tree representing the evolutionary relationships among the species included in the phylome. The tree is based on a maximum likelihood analysis of a concatenated alignment of 1,092 genes (see Methods), trimmed to delete columns with more than 20% gaps. The phylogenetic tree was reconstructed using IQTREE using the LG + F + R8 model according to the BIC criterion. Rapid bootstrap was calculated and nodes with less than 100% support are indicated. Species indicated in bold are pathogens of pine trees. (B)- Representation of the gene cluster found shared between L. acicola and D. septosporum. (C)- Gene tree representing the evolution of ToxD as obtained from the phylome reconstruction. Bootstrap values below 90 are marked on the tree.

Fig 2

Phylogenetic tree based on the concatenation of marker genes. Strains sharing identical marker sequences were collapsed (indicated by triangles). In bold are strains sequenced in this study. On the right of the tree marker sequences are represented: each column represents a marker gene. Different marker haplotypes of the NL or SL are shown in shades of blue and red, respectively. Different shades of blue or red indicate strains that share the same pattern of SNPs for a given marker gene. Invariable markers are depicted in gray.

Fig 3

(A)- Frequency graph depicting the distribution of SNPs/Kb for each strain for windows of 5 Kb of the contigs longer than 100,000 base pairs. Green is the distribution belonging to the strain with an overall SNPs/Kb of 5. Blue is the distribution of the two NL strains. Red is an example of a SL strain (B)- On top is an unrooted tree showing the evolution of the 70 strains used in this study. Below it is a close-up image from the part of the phylogenetic tree holding strains belonging to the SL. Red are the strains considered purely from the SL. Green are the strains detected as admixed by Structure. Black are the strains that would have been classified as being from the NL based on marker genes. (C) Structure graph showing the two populations in blue and red. The green square indicates admixed isolates.

Fig 4

Phylogenetic network of sequenced L. acicola isolates. Isolates colored in blue represent those belonging to the NL, those in red belong to the SL, and those in green contain introgression from the NL to the SL based on Structure.

Fig 5

Heatmap showing genomic windows assigned to the SL (red), NL (blue), undetermined (gray), or unmapped (black) on a similarity threshold of 2 SNPs/Kb to the reference. Columns indicate 5 Kb windows and are clustered together according to the pattern of ancestry observed in the different isolates. Rows indicate each of the isolates sequenced in this study.

Fig 6

Geographic distribution of sampling sites for some of the strains used in this study. Blue, red, and green rectangles mark isolation sites for NL, SL, or introgressed (according to Structure) strains, respectively. The background area covered in green tones corresponds to the area of distribution of Pinus radiata arboretum in the Basque Country.

Fig 7

PCA based on enzymatic activity of 20 enzymes. Markers are based on the lineage: circles for the SL, squares for the NL, and crosses for the strains that are introgressed according to Structure. Colors relate to the host pine species.