INTRODUCTION
Due to globalization and climate change an increasing number of fungal species are colonizing new areas worldwide. Numerous plant pathogenic fungi have benefited from modern agriculture practices and trade, and are recognized as growing threats to food security, conservation of biodiversity and global economy (Rosenblum et al. 2010; Fisheret al. 2012; Islam et al. 2016). These fungi often show complex population genetic structure due to multiple dispersal and back-dispersal events, varying demographic histories and hybridization events (Brasier & Kirk, 2010; Gladieux et al . 2018, 2014; Stukenbrock et al . 2007). Modern agricultural practices involve large-scale monocultures where well-adapted fungal pathogens can colonize and spread rapidly (Stukenbrock & McDonald 2008; Croll & McDonald 2017).
It is less known to what degree fungi adapt to other dimensions of human-made habitats. A few fungi have colonized buildings where the environment is dry. Such species may cause a reduction in indoor air quality or the decomposition of wooden structures with substantial economic losses (Schmidt 2007). In this study, we focus on the dry rot fungus Serpula lacrymans (Serpulaceae, Boletales, Agaricomycetes, Basidiomycota). Serpula lacrymans is known to be a primary decomposer of large substrates both in nature and in buildings (Harmsen 1960; Kauserud et al. 2012). It has a natural distribution in conifer woodlands in high altitude and/or latitude in Asia, from which it has invaded human-made constructions in temperate regions all over the world (Kauserud et al . 2007).
Previous population genetic studies based on microsatellites showed that the genetic diversity of the European population is very low (Kauserudet al. 2007; Engh et al. 2010b; Maurice et al.2014), suggesting that a few individuals established through a founder event. Further evidence for a narrow population bottleneck in Europe stems from the observations of a limited number of mating type (MAT) and self-recognition vegetative incompatibility (vic ) alleles present in European isolates (Kauserud 2004; Kauserud et al. 2006; Enghet al. 2010b; Skrede et al. 2013; Maurice et al.2014). Observation of natural Agaricomycete populations indicates high numbers of MAT and vic alleles are expected due to frequency-dependent selection acting on both loci (Raper 1996; Enghet al. 2010b; Coelho et al. 2017).
The European population of S. lacrymans has most likely dispersed via human vectors to North and South America, Australia and New Zealand (Kauserud et al. 2007). Population genetic analyses indicate that the Japanese dry rot population represents a separate transition from the natural habitat in mainland Asia to the built environment. In addition, the Japanese population possesses significantly higher genetic diversity compared to the European population (Kauserud et al.2007), also manifested in the presence of a higher number of MAT andvic alleles (Engh et al. 2010b). Evidence of admixture between Asian and European populations was identified in isolates from New Zealand (Kauserud et al . 2007), suggesting multiple introduction events from different source populations. However, which populations and genetic material have contributed to the variation in the New Zealand population is unknown.
The indoor habitat exploited by S. lacrymans resembles its natural habitat in terms of relatively dry conditions during most of the year and scattered presence of large resource units that it colonizes. Nevertheless, the transition to the indoor environment probably imposed a series of novel selective regimes. Based on analyses of allelic variation in a few neutral microsatellite loci, European and Japanese indoor populations appear to be genetically highly differentiated (Kauserud et al . 2007). Recent gene expression analyses suggested a wider niche related to competitive ability and substrate breath for a Japanese strain compared to a European strain (Hess et al . 2021). However, it is not known whether the two indoor populations possess different adaptations or physiological characteristics as a consequence of their independent evolutionary histories. Adaptive similarities between the two independent founder populations (Europe versus Japan) could indicate which characteristics are crucial for long-term survival in the built environment. Isolates of S. lacrymans possess significantly more efficient wood decay on spruce compared to the sister species S . himantioides (Balasundaram et al . 2018; Hess et al . 2021). Further, genomic analyses of these three isolates suggested that the extremely effective brown rot decay mechanism of S. lacrymans was due to increased capacity of a chelator mediated Fenton reaction (CMF). The evolution of specifically efficient CMF may be linked to the reduced enzymatic machinery compared to other brown rot fungi (Eastwood et al. 2011; Presley & Schilling 2017; Balasundaram et al. 2018), including its sister species S. himantioides (Balasundaram et al. 2018). These genomic analyses suggested that the selection of genes related to intracellular transport and growth were important for the colonization of large substrates.
In this study, we confirmed the presence of two divergent populations ofS. lacrymans in Europe and Japan. We aimed to estimate the time since establishment in the built environment and the number of haplotypes founding each of these two populations using demographic inference. Previous analyses indidcated that isolates from New Zealand possessed admixed genotypes between European and Japanese isolates. We aimed to investigate whether this is due to a recent admixture event. Finally, we aimed to test whether the European and Japanese populations are locally adapted to the built environment, and whether growth and wood decay are important during local adaptation. Thus, we searched for genomic islands of differentiation and selective sweeps as signals of selection. We also investigated the wood decay ability of some isolates as a measure of fitness differences across the populations.