Multigene Phylogeny and Morphology Reveal Three Novel Species and a Novel Record of Agaricus From Northern Thailand

Agaricus is a saprophytic mushroom genus widely distributed throughout the world. In this study, a survey of the Agaricus species carried out around Chiang Mai University in northern Thailand from 2018 to 2019 yielded 12 collections. Morphological characteristics and phylogenic analyses based on the internal transcribed spacers (ITS) and a fragment of the large subunit (LSU) of the nuclear ribosomal DNA (rDNA), and a fragment of the translation elongation factor 1-alpha (tef1) genes were investigated. The results revealed that these collections belong to six species including Agaricus erectosquamosus, Agaricus pallidobrunneus, Agaricus subrufescens, and three new species. Agaricus thailandensis sp. nov. was found to belong to Agaricus sect. Minores, which is placed in Agaricus subg. Minores. Aagricus pseudoerectosquamosus sp. nov. was placed in Agaricus sect. Brunneopicti within Agaricus subg. Pseudochitonia. Furthermore, Agaricus lannaensis remains an incertae sedis in Agaricus subg. Pseudochitonia. Additionally, this study was proposed that A. pallidobrunneus was discovered in Thailand for the first time. Full descriptions, color photographs, illustrations, and phylogenetic trees are provided.

There are approximately 6,000 records of Agaricus in the Species Fungorum (accessed in April 2021). However, these records have been found to include synonyms, some misidentifications, and a number of species that have not yet been well-documented. To date, a total of over 500 species of Agaricus are currently recognized. These include many new species from America, Asia, Australia, and Europe (Wang et al., 2015;Kerrigan, 2016;He et al., 2017;Callac and Chen, 2018) and few new species from Africa (Hama et al., 2010;Zhao et al., 2012b;Ling et al., 2021) and Oceania (Geml et al., 2007;Lebel and Syme, 2012;Lebel, 2013). Among the approximately 200 new species described since 2000, more than half have been reported to be from Asia, and a quarter have been reported in America (particularly North America) (Zhao et al., 2011Thongklang et al., 2014a;Liu et al., 2015;Karunarathna et al., 2016;Kerrigan, 2016;Chen et al., 2017;Callac and Chen, 2018;Hyde et al., 2018). Morphological characteristics, odor, and the chemical reactions of Schäffer's reagent and KOH have mainly been used in the traditional identification of the Agaricus species (Parra, 2008;Karunarathna et al., 2014;Chen et al., 2015Chen et al., , 2017Zhou et al., 2016). However, high variations of phenotypes, varying environmental factors, differing geographic conditions, and the separate developmental stages of basidiomata may influence the morphological identification process. This could make it difficult to distinguish this particular species from other closely related Agaricus species (Heinemann, 1978;Kerrigan, 1986;Singer, 1986;Callac et al., 1998a,b;Challen et al., 2003;Parra, 2008). Therefore, the application of molecular tools that are based on DNA analyses has proven to be essential in identifying the Agaricus species (Challen et al., 2003;Kerrigan et al., 2005Kerrigan et al., , 2008Zhao et al., 2011;Chen et al., 2015;Thongklang et al., 2016). Zhao et al. (2016) classified Agaricus into five subgenera and 20 sections based on morphological characteristics, multigene molecular phylogeny (ITS, LSU, and tef 1), and the divergence time. The revised classification of Agaricus by Chen et al. (2017), andParra et al. (2018)  Over the last decade, the study of Agaricus has expanded rapidly, especially in tropical regions. Notably, Thailand is proving to be a hot spot for the discovery of novel species. This is evidenced by the discovery of many new species of macrofungi, 45 of which are new Agaricus species that have been discovered since 2011 (Zhao et al., 2011(Zhao et al., , 2012aKarunarathna et al., 2014;Thongklang et al., 2014a,b;Ariyawansa et al., 2015;Chen et al., 2015Chen et al., , 2017Liu et al., 2015;Li et al., 2016;Hyde et al., 2017Hyde et al., , 2018He et al., 2018a). This study outlines how we found twelve Agaricus specimens during the course of our investigation of macrofungi in northern Thailand. Among these, we describe three new species and one new record. This investigation introduces the taxa based on studies of morphology and multigene analyses of combined ITS, LSU, and tef 1 sequences. Additionally, this study included a mini-review of Agaricus species that are found in Thailand.

Sample Collection
Agaricus were surveyed at Chiang Mai University, Chiang Mai Province, Thailand during the rainy seasons of the years 2018 and 2019. Photographs were immediately taken in the field. Basidiomata were collected and wrapped in aluminum foil and kept in plastic boxes while being transferred to the laboratory within 24 h of collection. Specimens were dried in a hot air oven at 45 • C until they were completely dried. They were then kept in a plastic zip-locked bag and deposited in the Biology Department's Herbarium (CMUB) along with the Herbarium of Sustainable Development of Biological Resources (SDBR-CMU), Faculty of Science, Chiang Mai University, Thailand. Facesoffungi and MycoBank numbers have also been provided (Robert et al., 2013;Jayasiri et al., 2015).

Morphological Observation
Macroscopic descriptions were made based on fresh specimens. Color, name, and codes were given according to the methods employed by Kornerup and Wanscher (1978). Chemical reactions were determined following the methods described by Chen et al. (2015) and He et al. (2017) including Melzer's reagent, 10% potassium hydroxide (KOH) in water, and Schäffer's reaction. Microscopic characteristics, including basidiospores, basidia, cystidia, and pileipellis, were observed from dried specimens that had been rehydrated in 95% ethanol followed by distilled water, 5% aqueous KOH, or Melzer's reagent. A minimum of 50 basidiospores, 20 basidia, and cystidia were measured using a compound light microscope (Olympus CX31, Japan). Basidiospores are presented in the following format: (a)b-c-d(e), for which "c" represents the average, "b" and "d" represent the average + and -standard deviation (SD), respectively, and "a" and "e" represent the minimum and maximum values, respectively. For spore statistics, Q represents the ratio of length divided by the width of the basidiospore, and Q m represented the average Q of all specimens ± standard deviation.

DNA Extraction, PCR Amplification, and Sequencing
DNA was extracted from the fresh tissue of each specimen using the FAVOGEN Genomic DNA Extraction Mini Kit (Taiwan) by following the manufacturer's instructions. The ITS region of the rDNA was amplified using the ITS4/ITS5 primers (White et al., 1990) by polymerase chain reaction (PCR). The LSU of the rDNA gene was amplified with LR5/LROR primers (Vilgalys and Hester, 1990) and the tef 1 gene was amplified with primers EF1-983F/EF1-1567R (Rehner and Buckley, 2005). The PCR programs of ITS, LSU, and tef 1 genes were established by following the methods employed by He et al. (2017). PCR products were checked by electrophoresis on 1% agarose gels stained with ethidium bromide and observed under UV light. PCR products were purified using NucleoSpin Gel and a PCR Cleanup Kit (Macherey-Nagel, Germany). PCR products were then sent to a commercial sequencing provider (1 ST BASE Company, Kembangan, Malaysia). The obtained sequences were ultimately subjected to BLASTn search in GenBank. 1

Sequence Alignment and Phylogenetic Analyses
Newly generated sequences were assembled using the Sequencher program. Details of the sequences used for phylogenetic analysis obtained from this study and previous other studies are provided in Supplementary Table 1. Two sequence datasets were prepared for phylogenetic analyses. The first dataset was comprised of sequences of two subgenera, namely Agaricus subg. Flavoagaricus and Agaricus subg. Minores. The second dataset contains sequences of six sections in Agaricus subg. Pseudochitonia namely, Agaricus sect. Bohusia, Agaricus sect. Brunneopicti, Agaricus sect. Floculenti, Agaricus sect. Nigrobrunnescentes, Agaricus sect. Rubricosi, and Agaricus sect. Sanguinolenti. The datasets were then aligned using MAFFT version 7 (Katoh and Standley, 2013). The first and second aligned datasets were deposited in TreeBASE under the numbers 27426 and 28087, respectively. Maximum Likelihood (ML) phylogenetic tree inference was performed for each dataset using RAxML-HPC2 version 8.2.10 (Stamatakis, 2006) on the CIPRES web portal (Miller et al., 2009). The phylogenetic tree was inferred from a four-partitions (ITS, LSU, tef 1 exons, and tef 1 introns) combined dataset using the GTRCAT model with 25 categories. A. campestris LAPAG370 was used as an outgroup for both datasets. Statistical support of the clades was obtained with 1,000 rapid bootstrap replicates. For Bayesian Inference (BI), the best-fit model of substitution amongst those implementable in MrBayes was estimated separately for each region using jModelTest 2 (Darriba et al., 2012) on the CIPRES portal based on the Bayesian Information Criterion (BIC). The selected models that are similar in both data sets, were HKY + I + G for ITS, SYM + G for tef 1 exons, K80 + G for introns 1 http://blast.ncbi.nlm.nih.gov of tef 1. While for LSU, were K80 + I + G in the first dataset, and K80 + I in the second dataset. Partitioned BI was performed with MrBayes 3.2.6 software for Windows (Ronquist et al., 2012). Two runs of five chains were conducted for eleven million (first dataset) and three hundred thousand generations (second dataset), which were sampled every 200 generations. At the end of the runs, the average deviations of split frequencies were 0.007058 (first dataset) and 0.009527 (second dataset). The potential scale reduction factor values of all parameters were close to 1. The burn-in phase (25%) was estimated by checking the stationarity in the plot generated by the sump command.

Taxonomic Description of New Species
Spores print dark brown (8F5) and the odor is phenol-like. Macrochemical reactions; KOH reaction yellow and Schäffer's reactions negative.
Ecology and distribution: Fruiting solitary or gregarious on sandy loam soil during the rainy season (mid-May to October). Known only from Thailand.
Note: Agaricus thailandensis was placed within Agaricus sect. Minores of Agaricus subg. Minores based on the morphological and molecular data. The grayish magenta to dark ruby fibrils on the pileus surface and the reaction of KOH Schäffer's reactions of A. thailandensis are morphologically similar to those of A. purpureofibrillosus (Linda J. Chen, R.L. Zhao & K.D. Hyde) in Agaricus sect. Minores of Agaricus subg. Minores (Fr.) . In contrast, A. purpureofibrillosus has shorter basidiospores (4.5-5.3 × 2.7-3 µm) than A. thailandensis and their clear separation was supported by the phylogeny . Phylogenetically, A. thailandensis formed a sister clade to the unnamed species Agaricus sp. voucher CA935 collected from Thailand, and they formed a sister clade to A. flammicolor and A. badioniveus (Figure 1). Both species have been collected from China and Thailand . Morphologically, A. flammicolor differed from A. thailandensis by the presence of orange bright fibrils on the pileus, smaller basidiospores (4.4-6.2 × 2.5-3.2 µm) and narrower basidia (12-16 × 5-6 µm) . Agaricus badioniveus is different from A. thailandensis by the yellowish-brown fibrils on the pileus and the mostly smaller basidiospores (5-6.2 × 3.1-3.8 µm) .

DISCUSSION
Agaricus is widely distributed in both temperate and tropical areas throughout the world (Kerrigan et al., 2005, Kerrigan, 2016Chen et al., 2015;Zhao et al., 2016;He et al., 2017He et al., , 2018aCallac and Chen, 2018;Ling et al., 2021). Morphological characteristics have been traditionally used in the identification of specimens of the Agaricus species. However, identification can be difficult as some species have similar features. Thus, identification can be limited by the morphological characteristics as well as the different environmental conditions that affect those morphological characteristics (Heinemann, 1978;Kerrigan, 1986;Singer, 1986;Callac et al., 1998a,b;Parra, 2008). Over the last two decades, molecular phylogeny has been an essential tool in the identification of Agaricus (Challen et al., 2003;Kerrigan et al., 2005Kerrigan et al., , 2008Zhao et al., 2011;Parra, 2013;Chen et al., 2015). The current classification of the genus Agaricus consists of six subgenera and twenty-four sections based on the combined data of morphological characteristics, the multigene phylogenetic analysis, and an estimation of divergence times Chen et al., 2017;He et al., 2018a;Parra et al., 2018).
In Thailand, the fourteen Agaricus species that were recorded by Thai taxonomists were identified only by their morphological characteristics. Of these, most have previously been found in temperate areas. However, there has been a lack of available molecular data on this species (Chandrasrikul et al., 2011) and there may have been incidences of mis-classification because tropical microflora are poorly understood. Thus, we are not sure these fourteen species have been accurately identified, expect A. subrufescens (Wisitrassameewong et al., 2012). Agaricus species richness is high in Thailand but many species have not yet been described based on a single collection and some of them have not been confirmed by phylogenetic data. Zhao et al. (2011) classified Agaricus collected from temperate and tropical regions, while ten Agaricus species collected from Thailand have been described based on their morphological and molecular characteristics.
In this study, six species of Agaricus including three new species (A. lannaensis, A. pseudoerectosquamosus, and A. thailandensis), one new record (A. pallidobrunneus), and two previously known species (A. erectosquamosus and A. subrufescens) collected from Chiang Mai Province, Thailand were identified based on their morphological characteristics and a multigene phylogenetic analysis. Agaricus lannaensis and A. pseudoerectosquamosus belong to Agaricus subg. Pseudochitonia in an incertae sedis clade and Agaricus sect. Brunneopicti, respectively. Agaricus thailandensis has been placed in Agaricus sect. Minores of Agaricus subg. Minores. Based on the phylogenetic analyses, A. lannaensis (including Agaricus sp. CA820) formed a sister clade to unnamed species Agaricus voucher LD2012162 (Figure 2); however, they cannot be assigned to Agaricus sect. Flocculenti according to the classification system of Agaricus . Agaricus sp. CA820 collected from Thailand should be recognized in A. lannaensis, however, its morphological characteristics should be further confirmed. Agaricus pseudoerectosquamosus and A. thailandensis are closely related to the unnamed species Agaricus sp. voucher NTT117 and CA935, respectively which have been collected in Thailand (Figures 1, 2). However, there is a lack of available information on the morphological characteristics of these unnamed species. Thus, their species definition will be required in future studies. Notably, A. subrufescens and A. erectosquamosus have been previously reported to be from Thailand by Wisitrassameewong et al. (2012) and Zhao et al. (2016), respectively. Agaricus pallidobrunneus has been reported from China , however, it has now been found for the first time in Thailand. Thus, to our knowledge, the Agaricus species recorded in Thailand has been raised to 62 species, 13 sections in five subgenera by the morphological and molecular evidence. Nevertheless, 13 Agaricus species listed by Chandrasrikul et al. (2011) require further confirmation by molecular data.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

AUTHOR CONTRIBUTIONS
CJ, JK, and NS: conceptualization and resources. CJ, JK, SV, and NS: methodology, formal analysis, and writing-review and editing. SV and CJ: software. NS and SL: validation. CJ, JK, and SV: investigation, data curation, and writing-original draft. NS and SL: supervision. All authors read, revised, and approved the final manuscript.

FUNDING
This work was supported by Chiang Mai University, Thailand.

ACKNOWLEDGMENTS
We acknowledged the Department of Biology and Research Center of Microbial Diversity and Sustainable Utilization, Faculty of Science, Chiang Mai University for providing the laboratories and molecular work. We are grateful to Russell Kirk Hollis for assisting with English language editing of this manuscript.