Interactions among microbes inhabiting plant tissue can influence plant fitness either positively or negatively. Chestnut blight cankers, caused by the fungal pathogen Cryphonectria parasitica (CP) on American chestnut trees (Castanea dentata), harbor complex communities of mycoviruses and non-pathogenic fungi (Non-CP). Mycovirus infection of CP can cause hypovirulence, which reduces CP growth and reduces the probability that a canker girdles a stem. Girdling cankers kill all tree tissues distal to the canker, maximizing the negative effects of blight epidemics, while non-girdling cankers allow the tree to recover from infection. We investigate whether Non-CP species inhabiting a canker can influence the probability of girdling. Can the composition, distribution, and persistence of Non-CP within a canker act synergistically with mycoviruses to reduce the probability of a canker girdling a stem? Further, can mycoviruses at the edge of cankers slow canker expansion and allow antagonistic Non-CPs to invade the interior of cankers? Chestnut populations in Michigan and Wisconsin are variable in their recovery from blight, with recovery associated with mycovirus presence. However, some cankers girdle trees at recovering populations, and non-girdling cankers occur at populations without mycoviruses. We investigate whether specific Non-CPs are associated with non-girdling cankers in the presence or absence of mycoviruses.
We sampled chestnut blight cankers annually (2012-2016) at six chestnut populations (five in Michigan and one in Wisconsin USA) to track microbial community dynamics within a canker. Overall, non-girdling cankers contained more mycovirus (37.5% of all isolates) than girdling cankers (25.1%). The Non-CP community of non-girdling cankers (37.4%) was composed of Penicillium 4.6%, Trichoderma 2.9%, Pezicula 2.2%, Nectria 1.8%, Valsaceae 1.3%, Umbelopsis 1%, Strasseria 1%, Gnomoniopsis 1%, and dozens of other rare fungal groups. The Non-CP community in non-girdling cankers was more species rich (47 operational taxonomic units [OTUs]) and diverse (Shannon-Weiner Index = 2.18') compared to girdling cankers (39 OTUs; 1.36'). The canker communities at recovering chestnut populations were also more similar to each other than to epidemic populations, based on Sorenson’s quotient of community similarity. We rejected our hypothesis that mycoviruses are distributed along the canker edge and that Non-CPs are found only in the interior of non-girdling cankers. Instead, mycoviruses and Non-CP taxa are distributed as a mosaic in both girdling and non-girdling cankers. Finally, specific Non-CPs were not associated with the presence or absence of mycoviruses in non-girdling cankers, suggesting that Non-CPs may have little influence on stem girdling and recovery from chestnut blight.