Additions to the Inventory of the Genus Alternaria Section Alternaria (Pleosporaceae, Pleosporales) in Italy

The genus Alternaria is comprised of well-known plant pathogens causing various important diseases in plants, as well as being common allergens in animals and humans. Species of Alternaria can be found as saprobes associated with various dead plant materials. This research aims to enhance the taxonomy of saprobic species in the genus Alternaria found on grasses and herbaceous plants from Italy, based on multi-locus phylogenetic analyses of a concatenated ITS, LSU, SSU, tef1-α, rpb2, gapdh and Alt-a1 DNA sequence dataset combined with morphological characteristics. Multi-locus phylogenetic analyses demonstrated six novel species belonging to the genus Alternaria sect. Alternaria as: A. muriformispora sp. nov., A. obpyriconidia sp. nov., A. ovoidea sp. nov., A. pseudoinfectoria sp. nov., A. rostroconidia sp. nov. and A. torilis sp. nov. Detailed morphological descriptions, illustrations and an updated phylogenetic relationship of taxa in the genus Alternaria sect. Alternaria are provided herein.

Alternaria is well-known as dematiaceous hyphomycetes which can be found everywhere. The genus is characterized by mononematous, macro-or micronematous, un-Alternaria species are major plant pathogens that infect a vast array of plant hosts [2,8,10,11,15,30]. Members in Alternaria sect. Alternaria are still confused in their delineation of species which are largely based on morphology and the clarity of their host species. The present study aims to introduce six novel species in Alternaria sect. Alternaria on different specific plant hosts based on a morpho-molecular approach.

Collection, Examination, Isolation, and Conservation
Samples were collected from dead branches, stems, and twigs of several plant hosts in Italy. The samples were dried and preserved in paper bags for further observation and examination under an Olympus SZ61 series stereo microscope. Micro-morphological features were mounted in sterilized distilled water on a clean slide for examination, and captured by a Nikon DS-Ri2 camera under a Nikon ECLIPSE Ni compound microscope. The size of micro-morphological features was measured by using Tarosoft (R) Image FrameWork version 0.9.7. Photographic plates were edited and combined in Adobe Photoshop CS6 software (Adobe Systems Inc., San Jose, CA, USA). The type specimens were deposited at the herbarium of Mae Fah Luang University, Chiang Rai, Thailand (MFLU).
Axenic cultures were obtained from single spore isolation using a spore suspension technique described by Senanayake et al. [31]. Germinated conidia were aseptically cultivated on potato dextrose agar (PDA) or malt extract agar (MEA) media under day/night lighting at room temperature (25-30 • C). The growth of fungal colonies and sporulation in cultures were observed after two weeks and eight weeks of incubation, respectively. The ex-type living cultures were deposited in the Mae Fah Luang University Culture Collection (MFLUCC). The novel species were registered in Index Fungorum (http://www.indexfungorum.org/names/ IndexFungorumRegister.htm, accessed on 15 July 2022).

DNA Extraction, PCR Amplification, and Sequencing
Fungal genomic DNA were extracted from fresh mycelia growing on PDA/MEA for one month using the Biospin Fungus Genomic DNA Extraction Kit (BioFlux ® , Hangzhou, China). The duplicated strain of each species was extracted DNA from fungal fruiting bodies using Forensic DNA Kit (Omega ® , Norcross, GA, USA). DNA fragments were amplified by polymerase chain reaction (PCR) with seven gene loci, including the internal transcribed spacers (ITS: ITS1-5.8S-ITS2) using primers ITS5 and ITS4 [32], the 28S large subunit rDNA (LSU) using primers LR0R and LR5 [33], the 18S small subunit rDNA (SSU) using primers NS1 and NS4 [32], the partial RNA polymerase second largest subunit (rpb2) using primers fRPB2-5F and fRPB2-7cR [34], the translation elongation factor 1-alpha (tef1-α) using primers EF1-728F and EF1-986R [35], Alternaria major allergen (Alt-a1) using primers ALT-F and ALT-R [25] and Glyceraldehyde 3-phosphate Dehydrogenase (gapdh) using primers GDP-1 and GDP-2 [36]. The polymerase chain reaction (PCR) was performed in a Veriti™ 96-Well Fast Thermal Cycler (Applied Biosystem, California, USA) following the protocol described in Li et al. [2]. All PCR products were sent to TsingKe Biological Technology (Beijing) Co., Ltd., China for purification and sequencing. The quality of the sanger DNA sequences and sequence consensus from forward and reward directions was checked and assembled manually in BioEdit v. 7.2.3 [37], and the newly nucleotide sequences were deposited in GenBank (Table 1). Table 1. Taxa used for the phylogenetic analyses in this study and their GenBank accession numbers. The ex-type cultures are indicated with superscript " T " and the newly generated sequences are indicated in bold.

Sequence Alignment and Phylogenetic Analyses
The newly generated ITS, LSU, SSU, tef1-α, rpb2, gapdh and Alt-a1 sequences were subjected to the nucleotide BLAST search engine via the NCBI (https://www.ncbi.nlm. nih.gov/, accessed on 10 April 2022) for checking potential contaminants or erroneous sequences as well as delineating the closely related taxa. All reference sequences were downloaded from GenBank. The multiple sequence matrixes were automatically aligned by MAFFT v. 7.452 (https://mafft.cbrc.jp/alignment/software/, accessed on 20 May 2022) [38]. Manual improvements were made where necessary in BioEdit v. 7.2.3 [37]. Individual gene alignments were separately analyzed by maximum likelihood (ML) in order to check the congruence of tree topology, and, thus, the combined multi-locus phylogenetic trees were inferred based on Bayesian inference (BI) and maximum likelihood (ML) analyses.
Maximum likelihood (ML) analyses were performed by Randomized Axelerated Maximum Likelihood (RAxML) [39,40] implemented in raxmlGUI 1.3 [41] using the default setting, but adjusted with 1000 bootstrap replicates and a GAMMAI model of nucleotide substitution. MrModeltest v. 2.3 [42] was used to determine the best-fit model of nucleotide substitution for each locus and incorporated into the analyses. GTR+I+G was the best-fit model for ITS, LSU and Alt-a1 loci under the Akaike Information Criterion (AIC), while TIM2+I+G was the best-fit model for SSU and rpb2, SYM+I+G was the best-fit model for gapdh and TIM1+I+G was the best-fit model for tef1-α. Bayesian inference (BI) analyses were performed by MrBayes v.3.1.2 [43]. Markov Chain Monte Carlo (MCMC) of six simultaneous Markov chains was run with one million generations to determine posterior probabilities (PP) [44,45], and started from a random tree topology. Trees were frequently sampled at 100th generation and the temperature value of heated chain was set to 0.15. The extra runs were required when the average standard deviation of split frequencies did not lower than 0.01 after one million generation. The first 25% trees represented the burn-in phase of the analyses and were discarded. The remaining trees were used for calculating posterior probabilities (PP) in the majority rule consensus tree. The phylogram were visualized in FigTree v. 1.4.0 [46] and edited in Microsoft Office PowerPoint 2016 (Microsoft Inc., Redmond, WA, USA).

Phylogeny
Six new species collected from dead herbaceous and monocotyledonous plants in Italy were analyzed with other representative Alternaria species in sect.    Alternaria generated by RAxML-based analysis of a combined ITS, LSU, SSU, tef1-α, rpb2, gapdh and Alt-a1 DNA sequence dataset. Bootstrap support values for maximum likelihood (ML, black) equal to or greater than 60% and Bayesian posterior probabilities (PP, red) equal to or greater than 0.95 PP are shown above the nodes. The tree is rooted to Alternaria alternantherae (CBS 124392). Newly species and generated strains are in blue, and the type strains are indicated in bold. Strains obtained from ex-type living culture are indicated by (T) and strains obtained from holotype specimen are indicated by (H).
Culture characteristics: Conidia germinating on PDA within 12 h and germ tubes produced from lateral cells. Colonies cottony, brown to dark brown, reaching 5 cm in 10 days at 25 • C, mycelium superficial, effuse, radially striate, with irregular edge, white to grey hyphae; conidia not sporulated in vitro within 60 days.
The nucleotide pairwise comparison of the ITS showed that Alternaria torilis differs from A. alternata (

Discussion and Conclusions
The aim of the present study was to introduce six novel Alternaria species in sect. Alternaria based on a morpho-molecular approach. These six saprobic species occurred on a variety of host plants in families Apiaceae, Brassicaceae, Chenopodiaceae, Fabaceae, Plantaginaceae, and Poaceae in Italy and could not be ascribed to any known taxa within sect. Alternaria. According to a recent classification provided by Woudenberg et al. [17] and Gannibal [15], we also note the morphological differences among extant species in this section. Hence, six new species: A. muriformispora, A. obpyriconidia, A. ovoidea, A. pseudoinfectoria, A. rostroconidia and A. torilis are introduced, described and illustrated herein.
Multi-locus phylogeny, based on a concatenated ITS, LSU, SSU, tef1-α, rpb2, gapdh and Alt-a1 DNA sequence matrix, revealed that these novel species formed well-resolved subclades within the sect. Alternaria, except for A. obpyriconidia that formed a distinct branch with other closely related species with low support in ML, but well-resolved species in BI analysis (1.00 PP; Figure 1). Based on the phylogenetic analyses and morphological characteristics, coupled with host preferences and nucleotide polymorphisms, A. obpyriconidia is justified as a new species following Jeewon and Hyde [48]. Furthermore, these six new species are distant from A. arborescens species complex (AASC) and A. alternata as well as other species in this section, which provided further evidence to support their phylogenetic affinities within the sect. Alternaria.
In the present analyses, Alternaria doliconidium and A. italica formed subclades, constituted within A. alternata, and that concurred with Li et al. [2]. Even though Woudenberg et al. [17] accepted only 11 phylogenetic species and one species complex in sect. Alternaria, and also treated 35 morphospecies as synonyms of A. alternata, Li et al. [2] re-analyzed the isolates of A. alternata with their new collections and mentioned that A. alternata could be separated to be at least five distinct species. However, more evidence is needed to support this conclusion. Similarly, A. doliconidium and A. italica lack informative cording genes such as Alt-a1, gapdh, rpb2 and tef1-α to justify their heterospecific status, with A. alternata pending further studies.
Woudenberg et al. [17] indicated that Alternaria species, including Alternaria sect. Alternaria, should be delineated by using phylogenomics due to a lack of effective gene sequences; however, the multi-locus phylogenetic analyses could well delineate species in sect. Alternaria (Figure 1) in studies of Wanasinghe et al. [20], Jayawardena et al. [21], Nishikawa and Nakashima [22] and Li et al. [2]. In the present study, phylogenetically analyzed taxa in sect. Alternaria, based on combined the intervening ITS regions, nuclear ribosomal DNA SSU, LSU and protein-coding genes Alt-a1, tef1-α, gapdh and rpb2, demonstrated that the recent taxa in this section formed distinct clades and were well supported in the phylogenetic tree. Nucleotide polymorphic comparisons also show the differences between our new taxa, which support the justifications of the new species described herein. It is interesting to note that in the nucleotide polymorphic comparisons of gene sequences among the species in Alternaria sect. Alternaria, rpb2 contains the most nucleotide differences among the species (up to 3.5%), which implies that this protein-cording gene may be a potentially effective gene region to delineate species in sect. Alternaria.
Nevertheless, species of Alternaria in sect. Alternaria are similar in morphological characteristics, and it is difficult to distinguish these species based solely on morphology. However, the conidial characteristics (e.g., conidial septation and rostrate or non-beak conidia) of our six novel species are significant to distinguish them from other species. Multi-locus phylogenetic analyses also provided further evidence, confirming that these six species are novel. These six species clearly formed a separate branch with significant support values (≥70% ML and 0.95 PP; Figure 1) in the present study, and this concurs with the findings of Li et al. [2]. Jeewon and Hyde [48] suggested that the nucleotide polymorphic comparisons of reliable genes should be more than 1.5% different for justifying the novel species. Even though the ITS, LSU, SSU and tef1-α could not be used to delineate some species in sect. Alternaria, the remaining gene regions (i.e., Alt-a1, gapdh and rpb2) proved sufficient for distinguishing these new species. Therefore, the novel species introduced herein were justified based on the multi-locus phylogeny coupled with morphological characteristics and nucleotide polymorphic comparisons of reliable genes.