Phyllachora species infecting maize and other grass species in the Americas represents a complex of closely related species

Abstract The genus Phyllachora contains numerous obligate fungal parasites that produce raised, melanized structures called stromata on their plant hosts referred to as tar spot. Members of this genus are known to infect many grass species but generally do not cause significant damage or defoliation, with the exception of P. maydis which has emerged as an important pathogen of maize throughout the Americas, but the origin of this pathogen remains unknown. To date, species designations for Phyllachora have been based on host associations and morphology, and most species are assumed to be host specific. We assessed the sequence diversity of 186 single stroma isolates collected from 16 hosts representing 15 countries. Samples included both herbarium and contemporary strains that covered a temporal range from 1905 to 2019. These 186 isolates were grouped into five distinct species with strong bootstrap support. We found three closely related, but genetically distinct groups of Phyllachora are capable of infecting maize in the United States, we refer to these as the P. maydis species complex. Based on herbarium specimens, we hypothesize that these three groups in the P. maydis species complex originated from Central America, Mexico, and the Caribbean. Although two of these groups were only found on maize, the third and largest group contained contemporary strains found on maize and other grass hosts, as well as herbarium specimens from maize and other grasses that include 10 species of Phyllachora. The herbarium specimens were previously identified based on morphology and host association. This work represents the first attempt at molecular characterization of Phyllachora species infecting grass hosts and indicates some Phyllachora species can infect a broad range of host species and there may be significant synonymy in the Phyllachora genus.


| INTRODUC TI ON
Phyllachorales is a monophyletic order of biotrophic fungi comprised of approximately 1,226 recognized species (Maharachchikumbura et al., 2016;Mardones et al., 2017), but global estimates of species within this order approach 160,000 (Cannon, 1997). The Phyllachorales largely contain plant parasitic fungi and are commonly associated with monocotyledonous plants across a range of habitats. These fungi are often referred to as "tar spot" fungi due to the production of stromata on plant hosts that resemble black flecks of tar ( Figure 1) (Mardones et al., 2017). panding each year since the initial report and continuing to have a significant economic impact on maize across many production regions in the United States Valle-Torres et al., 2020). Since first identified in North America in 2015, P. maydis has spread rapidly throughout the United States and Canada , and resulted in yield losses exceeding $US 658 million in 2018 . Although tar spot symptoms caused by members of the genus Phyllachora have been commonly observed on a number of grasses (Figure 1d-f) and shrub species throughout North, Central, and South America, historically the fungus has rarely been known to cause significant plant damage. However, tar spot has been occasionally reported to cause severe damage to maize in Mexico, Central America, and several Caribbean Islands (Valle-Torres et al., 2020).
The origin of P. maydis within the United States is not currently known, although the presence of two distinct epicenters of maize tar spot in the Midwest and Southeast indicates at least two separate emergence events. While tar spot is a new disease on maize in the United States and Canada, it has been present in Mexico, several Caribbean islands including Puerto Rico, Cuba, and the Dominican Republic as well as Central American Countries, such as Guatemala, Honduras, Nicaragua, and Costa Rica for the last century but only caused limited damage. In addition, tar spot signs and symptoms caused by Phyllachora species are common on several native and weedy grass species in North America (Figure 1d-f) (Orton, 1944).
The monographic work by Orton (1944) was completed solely by morphological identification and host affinity. Given our understanding of phenotypic plasticity of many fungi and the ability of biotrophic pathogens to infect multiple hosts (Morris & Moury, 2019), it is possible that cryptic species or species complexes may be present.
Species definitions within the Phyllachorales have historically been based largely on morphological characteristics and assumption of high host specificity due to their presumed biotrophic nature.
However, there are examples in the genus where this assumption of host specificity does not hold true (Cannon, 1991(Cannon, , 1997. Furthermore, species designations based on host specificity are highly dependent on accurate identification of the host species, which may be difficult or impossible in some instances. For example, P. graminis (Pers.) Fuckel is considered a "dustbin" species where many specimens of isolates infecting grasses are deposited with the host not often identified to species (Parbery, 1967). Furthermore, factors such as nutrients available to the fungus, temperature, light quality, light cycles, substrate type, host, and epigenetic factors may also result in alterations in fungal morphology that may result in inaccurate species designations (Francisco et al., 2019;Money, 2013;Slepecky & Starmer, 2009;Stockinger et al., 2009). Thus, our current understanding of the genetic diversity, host range, and species delimitation within the genus Phyllachora is relatively limited and requires reevaluation.
were grouped into five distinct species with strong bootstrap support. We found three closely related, but genetically distinct groups of Phyllachora are capable of infecting maize in the United States, we refer to these as the P. maydis species complex. Based on herbarium specimens, we hypothesize that these three groups in the P. maydis species complex originated from Central America, Mexico, and the Caribbean. Although two of these groups were only found on maize, the third and largest group contained contemporary strains found on maize and other grass hosts, as well as herbarium specimens from maize and other grasses that include 10 species of Phyllachora. The herbarium specimens were previously identified based on morphology and host association. This work represents the first attempt at molecular characterization of Phyllachora species infecting grass hosts and indicates some Phyllachora species can infect a broad range of host species and there may be significant synonymy in the Phyllachora genus.

K E Y W O R D S
biotrophs, pathogen diversity, phyllachorales, phylogeny, sympatric speciation, tar spot

Disease ecology; Microbial ecology
The recent emergence of P. maydis in the United States and Canada may also be associated with the ability of the fungus to better persist and spread than previously thought. Once established, the fungus can survive at least one winter at subzero temperatures on corn residue as ascospores within stromata, which are believed to be the main inoculum source the following season Kleczewski et al., 2019). Under periods of moderate temperatures and wet weather, it is believed that ascospores are dispersed by wind and rain splash where they land on the foliage, stalks, and husks of corn. After spore germination and infection of the host, the fungus remains dormant for at least 2 weeks after which stromata, and associated spermatia and ascospores, are produced. Data from Central America indicated a relatively steep dispersal curve of P. maydis ascospores from a source (Hock et al., 1995). However, the rapid spread of this fungus throughout the Midwest, coupled with observations of "top down" infestations in fields with no history of disease and observations of infestations of isolated plots located 1,200 m from potential inoculum sources, indicate that the pathogen can travel much further across local/regional topographies than estimated previously .
Based on this information, the emergence of P. maydis on corn in the United States and Canada could have been the result of many factors including the introduction of the fungus on infected plant material, natural northern dispersal through wind, establishment in the United States favored by climate change, changes in hybrid genetics, a host jump from a grass species, or a combination of any of these four. This study represents the first attempt to extract and sequence DNA from Phyllachora stroma on fresh and herbariuminfected grass specimens. The goal of this study is to understand the genetic diversity of Phyllachora species causing tar spot disease in contemporary corn production regions in the United States, and compare this to historical specimens of Phyllachora species from herbarium samples of maize and other grasses from Mexico, Central, and South America, the Caribbean, and Europe, as well as contemporary and herbarium species of Phyllachora species associated with grass hosts in the United States. This represents the first attempt at genetic characterization of Phyllachora maydis, Phyllachora graminis, and other grass infecting Phyllachora species. The objectives of this study are to: (1) determine if a single species of Phyllachora is responsible for tar spot disease of maize throughout its range in the Americas over the last century or if distinct genetic groups are responsible for these symptoms; and (2) determine if Phyllachora species infecting native and weedy grasses in close proximity to maize production fields are the same species as those infecting maize.
Understanding the phylogenetic diversity and the potential host and geographic range of Phyllachora species associated with maize and other grasses in the Americas will also help to infer the potential evolutionary origins and speciation patterns in this genus.  Table 1). Field specimens of infested maize and other grasses were collected by numerous individuals from the agricultural community as described in . Samples were pressed, dried at room temperature, and stored at 20°C in manila envelopes until processed. Herbarium F I G U R E 1 Signs and symptoms of Phyllachora spp. on grasses. Phyllachora maydis on maize at severe levels (a); with ascospores being extruded from stroma (b) and showing characteristic tapering ends of mature stromata (c

| DNA extraction, PCR amplification, and sequencing of stroma from leaf tissue
The DNA of individual stroma not surrounded by a necrotic halo were extracted using the X-Tract-N-AMP kit following manufacturer protocols (Sigma). The complete internal transcribed spacer region of ribosomal DNA (ITS1-5.8S-ITS2) with primers ITS1f and ITS4 (White et al., 1990). Stroma without necrotic halos was selected to reduce the potential for contamination by saprophytic fungi that may be present on necrotic tissue within these lesions.
The ITS gene region was amplified from DNA extracted from each stroma using the primer pair ITS1f and ITS4 (Bruns & Gardes, 1993;White et al., 1990) with 35 cycles of the following: 95°C 5 min, 94°C 30s, 52°C 30s, 72°C 1 min, followed by 72°C for 8 min, and a final hold at 4°C in a Thermo Fisher SimpliAmp thermocycler (Thermo Fisher Scientific, Waltham, WA). Individual PCR products from corresponding DNA extractions were loaded into 2% agarose gels and separated via electrophoresis for 40 min at 110V. All gels contained a P. maydis-positive control, a Fusarium graminearum-positive control, and a negative buffer control for quality assurance. Bands on gels were visualized using an Axygen gel imaging station (Axygen, Inc., Union City, CA). Stroma of Phyllachora spp. can be colonized by or associated with several other fungal species (Hock et al., 1992(Hock et al., , 1995McCoy et al., 2019). Consequently, samples returning a single band between 300 and 500 bp were considered free of additional fungal contaminants and used in subsequent analyses.
DNA from samples returning a single ITS band were subject to amplification of the large ribosomal subunit (LSU) region using the primer pair LR0R and LR5 (Dayarathne et al., 2017) using the aforementioned thermocycler conditions. All PCR products were purified using QIAquick PCR kits (Quiagen, Inc., Hilden, Germany), and the ITS and LSU amplicons for all samples were sequenced in the forward and reverse directions at the University of Illinois Core DNA Sequencing Facility (Urbana, Illinois).

| Sequence alignment, phylogenetic analysis, and molecular identification
Sequences generated from this study were combined with sequences obtained from GenBank. Exserohilum turcicum and Cocoicola californica were selected as the outgroups. Sequence data were aligned and concatenated using MAFFT v.7 (www.mafft.cbrc.jp/alignment/server/) using the G-INS-I model and manually inspected. The best = fit partitioning schemes were determined using PartitionFinder (Lanfear et al., 2017) and used to build the phylogenies. Both single gene and concatenated gene sets were analyzed using a maximum likelihood (ML) analysis using RaxML and Bayesian inference with MrBayes. The ML phylogenies were generated by RaxML (Stamatakis, 2014)   This was particularly the case for many of the herbarium samples, some of which were more than 100 years old. The ITS region was the most successfully amplified and sequenced, with 168 sequences generated.

| DNA extraction and PCR amplification from herbarium and contemporary samples
Whereas 91 sequences were generated for the LSU locus (Table 1).

| Phylogenetic diversity of Phyllachora isolates infecting maize and grasses
Based on both ITS + LSU ( Figure 2) and ITS ( Figure S1) phylogenies, we observed five genetically distinct groups that represent individual species of Phyllachora with strong bootstrap and posterior probability support (>70%). The results suggest that tar spot on maize in the United States is caused by three closely related species of Phyllachora ( Figure 2). In all, four species were found on maize but only    The other species recovered from maize was Phyllachora sp. 4. However, samples only included herbarium specimens from Guatemala and Venezuela and did not include any contemporary maize specimens. However, Phyllachora sp. 4 was commonly found among grasses in the United States that are found in proximity to maize production fields in Illinois, South Dakota, and New York ( Table 1). Isolates of Phyllachora sp. 4 were recovered from six grass species in four tribes in the United States representing a broad host range across a breadth of genetically diverse grass species.
Phyllachora sp. 5 was the only species not recovered from maize but was found on many of the same grass species as Phyllachora sp. 4, including rye, triticale, and fall panicum ( Table 1).
While there was limited Phyllachora sequence data in GenBank, we were able to include the ITS sequence of 19 isolates representing six recognized species of Phyllachora to determine any relationship between the isolates used in this study and those submitted previously to GenBank (Figure 3). The genetic cluster was determined as a result of the phylogenetic analysis of the combined DNA sequences from the ITS and LSU regions. These are displayed in Figures 2, 3, and Figure S1. b For contemporary material collected from field samples during this study, specimens of Phyllachora from maize were assumed to be P. maydis and specimens from grass species were treated as unknown Phyllachora sp. For herbarium specimens, we included the species name from the herbarium label.

TA B L E 1 (Continued)
probability (0.99). There was also an isolate of P. graminis from an unknown grass in Canada that grouped together with Phyllachora sp. 5, and the herbarium specimen of P. graminis from Agropyron repens in Germany from this study grouped in Phyllachora sp. 3 (Figure 3). Our results support the findings of previous observations that P. graminis is a poorly defined polyphyletic species that has often been assigned to tar spot symptom on a variety of grass hosts.

| DISCUSS ION
Since P. graminis was described by Persoon in 1785 as Sphaeria graminis and then transferred to the genus Phyllachora by Fuckel (1870), over 300 species have been recorded on graminaceous hosts, and many more on non-grass hosts. However, Parbery (1967) recognized that there are fewer species associated with grasses and established that there were 95 valid graminicolous Phyllachora species world-wide based on morphological characteristics. In the most complete study of Phyllachora species in North America, Orton (1944) identified 45 morphological species from more than 100 host species (Orton, 1944 In this work, we conducted the most comprehensive assessment of Phyllachora maydis reported to date and provided evidence that our understanding of this species and genera is limited and requires sig-

CO N FLI C T O F I NTE R E S T
The authors have declared that no competing interests exist.