Evolution of sympatric host-specialized lineages of the fungal plant pathogen Zymoseptoria passerinii in natural ecosystems

Introduction

The increasing emergence and severity of infectious fungal diseases threaten food security and natural ecosystems (Fisher et al., 2012; Stukenbrock & Gurr, 2023). Continuous monitoring, prediction modeling of disease spread, and deeper comprehension of fungal pathogens in wild plant hosts have been largely neglected. This is crucial to profile the impact of fungal pathogens on the context of climate change and independent of agricultural environments (Fisher et al., 2012). Current evidence supports that crop wild relatives (CWRs) might serve as reservoirs for domesticated plant pathogens (Monteil et al., 2013, 2016), although still few studies are focused on wild pathogen population processes and dynamics (Rouxel et al., 2013; Penczykowski et al., 2015; Eck et al., 2022; Treindl et al., 2023). CWRs hold higher levels of genetic diversity and have coevolved in sympatry with plant pathogens in natural ecosystems. Moreover, the centers of diversity and domestication of crop plants harbor a wealth of species (Harlan, 1971) that could serve as hosts for plant pathogens (Vavilov, 1992). Despite the latter, natural ecosystems are undervalued economically, which limits funding for studies (Fisher et al., 2012). Furthermore, having access to wild species found in remote locations or immersed in complex geopolitical contexts adds another layer of difficulty, generating a geographical bias toward high-income regions at the expense of exploring the remaining biodiversity (Marks et al., 2023). One way to overcome this is to prioritize neglected areas by collaborating with scientific communities situated in less-represented regions of the globe (Marks et al., 2023), and promoting research on nonmodel species and dynamics in natural ecosystems.

Cumulative evidence supports that ecological divergence of plant pathogens is driven by host specialization. As proposed by Crous & Groenewald (2005) and exemplified by multiple studies (Steenkamp et al., 2002; Choi et al., 2011; Rouxel et al., 2013; Faticov et al., 2022), plant pathogens phylogenies frequently represent multiple closely related sister or cryptic species. In this regard, the Zymoseptoria genus comprises eight ascomycete species, only two of them, Zymoseptoria tritici and Zymoseptoria passerinii (Sacc.) Quaedvlieg & Crous, have been reported to infect domesticated hosts (Quaedvlieg et al., 2011; Stukenbrock et al., 2012b). The origin, population genetics, and plant–pathogen dynamics of the wheat fungal pathogen Z. tritici have been extensively investigated in an agricultural context (Linde et al., 2002; Stukenbrock et al., 2011; McDonald et al., 2022; Orellana-Torrejon et al., 2022; Feurtey et al., 2023). Zymseptoria tritici and Z. passerinii share overlapping geographic ranges and have diversified in sympatry at their center of origin (Rojas-Barrera et al., 2023). However, our understanding of their population genetics and the influence of multiple sympatric hosts on pathogen prevalence in natural ecosystems remain limited (Stukenbrock et al., 2011, 2012a). Population genetic studies on Z. tritici support that the center of diversity of the Zymoseptoria genus is located in the Middle East (Banke et al., 2004), in proximity to the Fertile Crescent recognized as a center of crop domestication (Harlan, 1971), where several CWRs are abundant and naturally distributed (Harlan & Zohary, 1966).

The hemibiotrophic fungus Z. passerinii, which causes the sporadic disease Septoria speckled leaf blotch (SSLB), became significant during its last major outbreak in the late 1990s to early 2000s, in the Upper Midwest of the United States and neighbor provinces in Canada (Toubia-Rahme & Steffenson, 2004). During that period, an extensive study with 309 isolates collected in North Dakota and Western Minnesota in 2003 and 2004 revealed a high genetic diversity, and a shallow population structure for Z. passerinii (Lee & Neate, 2007b), which is supported by an equilibrated frequency of both mating types (Lee & Neate, 2007a), suggesting sexual reproduction. However, the teleomorphic stage of this pathogen has only been reported under experimental conditions and has not been described in the field (Ware et al., 2006). Interestingly, despite the relevance of SSLB in the early 2000s, there are, to our knowledge, no current reports of SSLB outbreaks in North America or elsewhere, including the Middle East.

The disease triangle states that favorable climatic variables are as important as plant characteristics in determining the severity of disease epidemics (Stevens, 1960). Thus, the sporadic nature of SSLB has been attributed to the requirement of > 48 h of continuous moisture for spore germination (Green & Dickson, 1957) and a long period of incubation (16–19 d) (Koble et al., 1959; Cunfer, 2000), suggesting a strong dependence on weather conditions.

Additionally, the development of resistant cultivars with durable resistance traits (Toubia-Rahme & Steffenson, 2004) has been related to the abrupt disappearance of SSLB during recent decades. In contrast to agricultural environments, our sampling of wild grasses during 2018 and 2020 revealed the persistence of SSLB in multiple Hordeum sp. in northwest Iran, which overlaps with the Fertile Crescent region, and the center of origin and diversity for fungal plant pathogens and their wild host species (Harlan & Zohary, 1966; Banke et al., 2004). Although we recovered only a few Z. passerinii isolates from wild hosts, this pathogen has not been reported in North America or Iran since 2004 (Lee & Neate, 2007b). Thus, this dataset provides valuable insights into Z. passerinii populations within a natural ecosystem located at the origin center for both the pathogen and its hosts.

In this work, we aimed to answer the following questions: Have the barley-infecting lineages of Z. passerinii arisen through host tracking, similar to Z. tritici (Stukenbrock et al., 2011) or through host range expansion? If Z. passerinii evolved by host tracking, the pathogen is likely younger than its barley host (Stukenbrock & McDonald, 2008) and may have lost its capacity to infect the wild ancestor of barley (H. spontaneum). By contrast, during host range expansion, the pathogen retains its ability to infect the primary host, and no significant changes in the pathogen's gene pool are expected (Thines, 2019). However, no large-scale study has tested the last condition for host range expansion.

Given our observation of symptoms in what were identified as different Hordeum species, we speculate whether host specificity in sympatric populations has led to population divergence in the center of diversity of the Zymoseptoria genus. Lastly, we ponder whether the persistence of SSLB in natural ecosystems is linked to a shorter disease onset observed in wild-host-infecting lineages compared with the domesticated host. We address these questions using population genomic datasets of host-specialized populations of Z. passerinii.

In addition to the mentioned inquiries, we evaluated the host range of Z. passerinii, establishing a pathosystem in domesticated barley and three of their CWR. An advantageous perspective is that barley is diploid and provides an alternative host model system to study Zymoseptoria-caused diseases in cereals and examining candidate resistance traits in a diploid genome background.