Morphological and molecular identification of two new Alternaria species (Ascomycota, Pleosporaceae) in section Radicina from China

Abstract The fungal genus Alternaria was distributed widely and found in different habitats such as plant or indoor environment. During an investigation into this genus in China, two new Alternaria species, Alternaria vulgarae and A. divaricatae were respectively isolated from diseased leaves of Foeniculum vulgare and Saposhnikovia divaricata, which both belonged to Umbelliferae. Phylogenetically, they were determined as new species belonging in the section Radicina of Alternaria based on the combined four gene fragments of ITS, TEF1, GAPDH and RPB2. Morphologically, the two species were illustrated and compared with other relevant Alternaria species in section Radicina.


Introduction
Alternaria Nees (1816) was typified by Alternaria tenuis (the synonym of A. alternata), a species with muriform and catenulate conidia. Since then, hundreds of new species were proposed in the genus. Meanwhile, because of unstable taxonomic standards (morphological characteristics, host and growing environment, etc.), the controversies about species boundary started and never stopped (Elliott 1917;Fries 1832;Neergaard 1945;Joly 1964;Wiltshire 1933;Simmons 1967Simmons , 1992. In 1992, Simmons introduced reasonable standards to get unified taxonomic concepts on Alternaria species based on colony and conidial morphology. At the same time, the concept of speciesgroup was introduced, the small-spored, catenulate taxa of Alternaria were divided into six morphological groups by Simmons and Roberts (1993). More recently, around 300 Alternaria morphospecies have been accepted based on the shape, size, septation of conidia, as well as sporulation patterns. Small-spored Alternaria species were also redefined and divided into 10 subsections characterized by short (>50(-60) μm) or medium (50-100(-105) μm) conidia produced in various patterns of branched and unbranched chains or solitary (Simmons 2007). However, the identification remained challenging due to the impact of environmental conditions and other unknown factors.
On the other hand, multigene phylogenetic analyses have provided strong support for the re-definition of the Alternaria genus. Many sequences of gene regions such as the internal transcribed spacer region of rDNA (ITS), large subunit ribosomal DNA (LSU), mitochondrial small subunit (mtSSU), Alternaria major allergen (ALT), glyceraldehydes-3-phosphate dehydrogenase (GAPDH), translation elongation factor 1-alpha (TEF1), RNA polymerase second largest subunit (RPB2), and ATPase etc. were applied to delimit the genus (Pryor and Gilbertson 2000;Hong et al. 2005;Lawrence et al. 2012Lawrence et al. , 2013Woudenberg et al. 2013Woudenberg et al. , 2014Poursafar et al. 2018). In recent studies, both morphological and molecular analyses were used for the delimitation of the genus Alternaria, which has been divided into 28 sections and eight monotypic lineages (Woudenberg et al. 2013;Lawrence et al. 2016;Ghafri et al. 2019;Marin-Felix et al. 2019). The number of Alternaria species has been continuously growing after re-descriptions and new discovery (Deng et al. 2018;Ahmadpour 2019: Liu et al. 2019Tao et al. 2019;Bessadat et al. 2020;He et al. 2020). Coincidentally, several phylogenetic lineages have strongly supported morphology-based sections but others not (Simmons 2007;Woudenberg et al. 2015).
During the investigation into Alternaria species in China, two new taxa were isolated from umbelliferous plants, Foeniculum vulgare and Saposhnikovia divaricata. The study was designed to determine them based on a polyphasic approach including morphology and phylogenetic analyses.

Isolation and morphological studies
Leaves of Foeniculum vulgare and Saposhnikovia divaricata with necrotic spots were respectively collected from Wenjiang district (Chengdu, Sichuan in June, 2015) and Badong county (Yichang, Hubei in July, 2016) in China. For fungal isolation, the samples were stored in sterile plastic bags and transported to the laboratory. The tissues were cut into small segments and placed on moist filter papers within Petri dishes then incubated at 25 °C to stimulate sporulation. After 24 h, the samples were examined under a stereomicroscope. Alternaria-like spores were picked up and inoculated to potato dextrose agar (PDA: Difco, Montreal, Canada) using sterilized glass needles. All isolated pure cultures were inoculated to test-tube slants and stored at 4 °C. Dried cultures from the single spore and ex-type strains were deposited in the Fungi Herbarium of Yangtze University (YZU), Jingzhou, Hubei, China.
To determine colonial characteristics (size, color and texture of colony), the strains were cultured on PDA at 25 °C for 7 days in darkness. To analyze the morphological features of conidia (conidial size, shape, sporulation, etc.), fresh mycelia were transferred on potato carrot agar (PCA) and V8 juice agar (V8A) then incubated at 22 °C under an 8 hour photoperiod for 7 days (Simmons 2007). Conidia were mounted into a lactophenol picric acid solution and digital images were captured under a Nikon ECLIPSE Ni-U microscope system (Nikon, Japan). Conidia (n = 50) were randomly selected for determining the morphology and sporulation patterns were also photographed at the same time.

Phylogenetic analyses
Preliminary BLAST searches in GenBank with ITS and TEF1 sequences of the present isolates indicated that they had a close phylogenetic relationship with species from section Radicina of Alternaria. Subsequently, sequence data of 19 Alternaria species and Stemphylium herbarum CBS 191.86 (outgroup) were retrieved from National Center for Biotechnology Table 1). The gene sequences were concatenated and edited manually according to ITS+TEF1+GAPDH+RPB2 for YZU 161234 and YZU 161235 and ITS+TEF1+RPB2 for YZU 151055 and YZU 151059 with equal weight in MEGA v.7.0.26 (Kumar et al. 2016). Maximum parsimony (MP) analysis was performed in PAUP 4.0 (Swofford 2002) using the heuristic search option of 1000 random-addition sequences and tree bisection and reconnection (TBR) as the branch-swapping algorithm. Gaps were treated as missing data. The bootstrap values (BS) with 1000 replicates were performed to determine branch support. Parsimony scores of tree length (TL), consistency index (CI), retention index (RI) and rescaled consistency (RC) were calculated for each generated tree. The Bayesian inference (BI) analysis was performed with a Markov Chain Monte Carlo (MCMC) algorithm with Bayesian posterior probabilities in MrBayes v. 3.2.1 (Ronquist et al. 2012). The bestfit evolutionary models (GTR+I+G) were determined in MrModel-test v. 2.3 (Posada and Crandall 1998) using the Akaike Information Criterion (AIC). Two independent analyses with four chains each were run for 10,000,000 generations. Trees were sampled every 100 th generation. The run was stopped until the standard deviation of split frequencies reaches < 0.01 and the initial 25 % of the trees were discarded as the burn-in phase of each analysis. Maximum likelihood (ML) analysis was performed using RAxML v.7.2.8 (Stamatakis 2006), implementing GTRCAT model and executing 1000 rapid ML bootstrap replications. Branch support equal to or above 0.70/70%/70% for PP (posterior probability of BI analysis) and BS (bootstrap for ML and MP analyses) values were shown at the nodes in the phylogram.
Additional isolate examined. China, Sichuan Province, Chengdu City, Wenjiang District, Herb Garden of Chengdu University of Traditional Chinese Medicine, from leaf spot of Saposhnikovia divaricata. 17 June, 2015, L He, living culture YZU 151059.
Notes. Phylogenetically, Alternaria divaricatae forms a distinct clade in section Radicina, which appears to be sister to a clade including A. petroselini, A. selini and A. vulgarae (Fig. 1). Morphologically, A. divaricatae was different from A. petroselini, A. selini and A. vulgarae by producing smaller conidia (Table 2) and special sporulation from apex or geniculate loci of lateral or apical of aerial hyphae. Moreover, A. chlamydosporifera, A. glehniae and A. smyrnii grouped together and clustered as a sister clade with A. divaricatae, A. petroselini, A. selini and A. vulgarae (Fig. 1). Obviously, the conidia of A. divaricatae was smaller than A. smyrnii (Table 2) and A. divaricatae could be also easily differentiated from A. chlamydosporifera by the lack of chlamydospores in culture (Marin-Felix et al. 2019). Meanwhile, A. glehniae was distinguished from A. divaricatae by its single conidium on apex of conidiophore (there was no geniculate sporulation loci) and production of secondary conidium (Tao et al. 2019). In addition, A. radicina and A. carotiincultae were distinguished from present species by distant phylogenetic relationship in section Radicina.
Additional isolate examined. China, Hubei Province, Yichang city, Badong county on infected leaves of Foeniculum vulgare. 19 July, 2016, L He, living culture YZU 161235.  (2007)  Notes. Phylogenetic analysis based on combining four gene fragments indicated that Alternaria vulgarae fell in an individual branch in section Radicina of Alternaria and displayed a close relationship with A. petroselini and A. selini with high supported values (Fig. 1). Morphologically, A. vulgarae could be easily distinguished from A. petroselini and A. selini by their sporulation and length of conidiophores. Conidia of A. petroselini were solitary or cluster a small clump with 2-4 spores near the tips or lateral of conidiophores. Occasionally, the secondary conidium could be observed. Meanwhile, the single conidium or conidial chains (1-3) of A. selini grew from numerous lateral conidiophores, which produced from wire-like hyphae (Simmons 2007). Differently, conidia of A. vulgarae were erected from apex of conidiophores or terminal hyphae. There were no small conidial clumps and secondary conidium formed (Fig. 3B, C, D, E). Moreover, the conidiophores of A. vulgarae (12-80 × 4-6 μm) was longer than A. petroselini (30-60 × 5-6.5 μm) and shorter than A. selini (200-400 × 4-6 μm) (Simmons 2007). Besides, A. vulgarae differed from A. petroselini in conidial shape. Conidium populations of A. petroselini were dominated by shot-ovoid to subsphaerical spores though, the shapes of A. vulgarae were mainly ovoid, ellipsoid or long-ellipsoid (Simmons 2007).

Discussion
Morphologically, Alternaria radicina species-group was one of the 10 subsections (A-1) and comprised 8 species described by Simmons (2007): A. atrocariis, A. carotiincultae, A. japonica, A. petroselini, A. radicina, A. selini, A. smyrnii and A. soliaridae. With the development of molecular studies, the species-group was re-defined and the section Radicina was introduced and perfected (Pryor and Gilbertson 2000;Lawrence et. al 2013;Woudenberg et al. 2013). Uniformly, species in this section had some similar morphological characters, such as conidiophores, sporulation, conidial shape and etc. The phylogenetic analysis showed that only five species were clustered in section Radicina. Except for A. atrocariis, which had no published sequence data, the two other species were shown to belong to other sections: A. japonica felled in the section Japonicae and A. soliaridae formed a separate monophyletic lineage (Woudenberg et al. 2013). Recently, two more species A. chlamydosporifera and A. glehniae were reported in the section Radicina (Marin-Felix et al. 2019;Tao et al. 2019).
In the current study, two new Alternaria species belonged to the section Radicina based on morphological and phylogenetic analysis. Alternaria divaricatae was identified as a novel species based on unique morphological and well-supported phylogenetic analysis ( Fig. 2 and Table 2). Phylogenetically, A. vulgarae clustered with A. petroselini and A. selini. Although its phylogenetic position was not well-supported, A. vulgarae can be distinguished from these two species in section Radicina by morphological characteristics (Table 2). Except the length of conidiophores, A. vulgarae was characterized by its sporulation. Meanwhile, A. vulgarae won't form secondary conidium (Table 2). These characters were important standards to identify Alternaria species (Simmons 2007). And, according to Jeewon and Hyde (2016), a fungal species can be defined based on the distinctive morphological characters even though the phylogenies were not well-supported, because the phylogeny cannot really reflect all morphologies.

Acknowledgments
The financial support of the work was given by the National Natural Science Foundation of China (31400014) and Guangxi Key Laboratory of Biology for Crop Disease and Insect Pests (2019-KF-01).