De novo genome assembly of Geosmithia morbida, the causal agent of thousand cankers disease

Introduction

Studying molecular evolution of any phenotype is now made possible by the analysis of large amounts of sequence data generated by next-generation sequencing platforms. This is particularly beneficial for the study of emerging fungal pathogens, which are progressively recognized as a threat to global biodiversity and food security. Furthermore, in many cases their expansion is a result of anthropogenic activities and an increase in trade of fungal-infected goods (Fisher et al., 2012). Fungal pathogens are capable of evolving rapidly in order to overcome host resistance, fungicides, and to adapt to new hosts and environments. Whole genome sequence data are useful in identifying the mechanisms of adaptive evolution within fungi (Stukenbrock et al., 2011; Gardiner et al., 2012; Condon et al., 2013). For instance, Stukenbrock et al. (2011) investigated the patterns of evolution in fungal pathogens during the process of domestication in wheat using all aligned genes within the genomes of wheat pathogens. They found that Zymoseptoria tritici, a domesticated wheat pathogen (formerly known as Mycosphaerella graminicola), underwent adaptive evolution at a higher rate than its wild relatives, Z. pseudotritici and Z. ardabiliae (Stukenbrock et al., 2012). The study also revealed that many of the pathogen’s 802 secreted proteins were under positive selection. A study by Gardiner et al. (2012), identified genes encoding aminotransferases, hydrolases, and kinases that were shared between Fusarium pseudograminearum and other cereal pathogens. Using phylogenomic analyses, the researchers demonstrated that these genes had bacterial origins. These studies highlight the various evolutionary means that fungal species employ in order to adapt to specific hosts, as well as the importance of genomics and bioinformatics in elucidating evolutionary mechanisms within the fungal kingdom.

Many tree fungal pathogens associate with bark beetles in the family Scolytinae (Six & Wingfield, 2011). With climate change, beetles and their fungal symbionts can invade new territory and become major invasive forest pests on a global scale (Kurz et al., 2008; Sambaraju et al., 2012). A well-known example of an invasive pest is the mountain pine beetle and its symbiont, Grosmannia clavigera that has affected approximately 3.4 million of acres of lodgepole, ponderosa, and five-needle pine trees in Colorado alone since the outbreak began in 1996 (Massoumi Alamouti et al., 2014; Colorado State Forest Service, 2015). Another beetle pest in the western U.S., Pityophthorus juglandis (walnut twig beetle), associates with several fungal species, including the emergent fungal pathogen Geosmithia morbida (Tisserat et al., 2009; Kolarik et al., 2011).

Reports of tree mortality triggered by G. morbida infections first surfaced in 2009 (Kolarik et al., 2011), while the fungus was described as a new species in 2011 (Tisserat et al., 2009). This fungus is vectored into the host via P. juglandis and is the causal agent of thousand cankers disease (TCD) in Julgans nigra (eastern black walnut) (Zerillo et al., 2014). This walnut species is valued for its wood, which is used for furniture, cabinetry, and veneer. Although J. nigra trees are planted throughout western U.S. as a decorative species, they are indigenous to eastern North America where the walnut industry is worth hundreds of millions of dollars (Rugman-Jones et al., 2015; Zerillo et al., 2014). In addition to being a major threat to the eastern populations of J. nigra, TCD is of great concern because certain western walnut species including J. regia (the Persian walnut), J. californica, and J. hindsii are also susceptible to the fungus according to greenhouse inoculation studies (Utley et al., 2013).

The etiology of TCD is complex because it is a consequence of a fungal-beetle symbiosis. The walnut twig beetle, which is only known to attack members of genera Juglans and Pterocarya, is the most common vector of G. morbida (Kolarik et al., 2011). Nevertheless, other beetles are able to disperse the fungus from infested trees (Kolařík, Kostovčík & Pažoutová, 2007; Kolařík, & Jankowiak, 2013). As vast numbers of beetles concentrate in the bark of infested trees, fungal cankers form and coalesce around beetle galleries and entrance holes. As the infection progresses, the phloem and cambium discolor and the leaves wilt and yellow. These symptoms are followed by branch dieback and eventual tree death, which can occur within three years of the initial infection (Kolarik et al., 2011). Currently, 15 states in the U.S. have reported one or more incidences of TCD, reflecting the expansion of WTB’s geographic range from its presumed native range in a few southwestern states (Rugman-Jones et al., 2015). Additionally, TCD has also been found in Europe where walnut species are planted for timber (Montecchio & Faccoli, 2014).

To date, G. morbida is one of only two known pathogens within the genus Geosmithia, which consists of mostly saprotrophic beetle-associated species (the other pathogen is G. pallida) (Lynch et al., 2014). The ecological complexity this vector-host-pathogen system exhibits makes it an intriguing lens for studying the evolution of pathogenicity. A well-assembled reference genome will enable us to identify genes unique to G. morbida that may be utilized to develop sequence-based tools for detecting and monitoring epidemics of TCD and for exploring the genomic features of Geosmithia species, which may help explain the evolution of pathogenicity. Here, we present a de novo genome assembly of Geosmithia morbida. The objectives of this study are to: 1) assemble the first, high-quality draft genome of this pathogen; 2) annotate the genome to better understand the genomic composition of Geosmithia species; and 3) briefly compare the genome of G. morbida to two other fungal pathogens for which genomic data are available: Fusarium solani, a root pathogen that infects soybean, and Grosmannia calvigera, a pathogenic ascomycete that associates with the mountain pine beetle and kills lodgepole pines in North America.

Methods

Results and Discussion

Conclusion

This work introduces the first genome assembly and analysis of Geosmithia morbida, a fungal pathogen of the black walnut tree that is vectored into the host via the walnut twig beetle. The de novo assembly is composed of 73 scaffolds totaling in 26.5 Mbp. There are 6,273 predicted proteins, and 4.41% of these are unknown. In comparison, 68.27% of F. solani and 26.70% of G. clavigera predicted proteins are unknown. We assessed the quality of our genome assembly and the predicted protein set using BUSCO, and found that 98 and 95% of the single copy orthologs specific to the fungal lineage were present in both, respectively. These data are indicative of our assembly’s high quality and completeness. Our BLASTp search against the PHI-base database revealed that G. morbida possesses 1,974 genes that are homologous to proteins involved in pathogenicity. Furthermore, G. morbida shares several domains with known effector proteins that are key for fungal pathogens during the infection process.

Geosmithia morbida is one of only two known fungal pathogens within the Geosmithia genus (Lynch et al., 2014). The genome assembly introduced in this study can be leveraged to explore the molecular mechanisms behind pathogenesis within this genus. The putative list of pathogenicity genes provided in this study can be used for future comparative genomic analyses, knock-out, and inoculation experiments. Moreover, genes unique to G. morbida may be utilized to develop DNA sequence-based tools for detecting and monitoring ongoing and future TCD epidemics.

Supplemental Information