Dieback of Euonymus alatus (Celastraceae) Caused by Cytospora haidianensis sp. nov. in China

: Euonymus alatus (Celastraceae) is widely cultivated in China for its economic value and landscape beneﬁts. Euonymus alatus dieback occurs due to members of Cytospora and has become one of the most severe diseases affecting its cultivation in China. In this study, we examined the causal agent of bough dieback on campuses of University Road, Beijing, China. Among the strains, three were morphologically consistent with Cytospora , showing hyaline and allantoid conidia. Based on phylogenetic analyses of the concatenated actin (ACT), internal transcribed spacer (ITS), RNA polymerase II second largest subunit (RPB2), translation elongation factor 1-alpha (TEF1- α ) and beta-tubulin (TUB2) gene sequences, along with morphological and physiological features, we propose C. haidianensis as a novel species. It was conﬁrmed as a causal agent of dieback of E. alatus by pathogenicity tests. Mycelial growth of Cytospora haidianensis occurred at pH values ranging from 3.0 to 11.0, with optimum growth at 8.3, and at temperatures from 5 to 35 ◦ C, with optimum growth at 19.8 ◦ C. We also tested the growth of C. haidianensis in the presence of six carbon sources. Sucrose, maltose and glucose were highly e ﬃ cient and xylose was the least. The ability of C. haidianensis to grow at 19.8 ◦ C may help to explain its occurrence causing dieback of E. alatus in Beijing during the autumn season. Three replicates conducted for PDA plugs


Introduction
Euonymus alatus (Celastraceae) has been widely cultivated for ornamental landscaping in China because of its tolerance to many environmental conditions [1]. At present, the related research on the fungal diseases of Euonymus is mainly on anthracnose caused by Colletotrichum gloeosporioides, powdery mildew by Oidium euonymi-japonici and dieback by Cytospora euonymicola and C. euonymina [2,3].
The genus Cytospora has wide distribution and has often been regarded as comprising phytopathogens, endophytes or saprobes occurring on a broad range of hosts [3,4]. Several species have been reported as pathogens causing severe branch or trunk dieback disease on monocotyledonous, dicotyledonous and gymnosperm hosts (e.g., Anacardiaceae, Elaeagnaceae, Fabaceae, Juglandaceae, Myrtaceae, Rosaceae, Salicaceae and Ulmaceae) [5,6]. The symptoms of Cytospora canker are elongate, slightly sunken and discoloured areas in the bark at first, then the forming of several prominent black fruit bodies [5]. Conidia emerge from the fructifications in the form of yellow to orange or red gelatinous tendrils under moist conditions [3]. Cytospora species have single or multiple locules (and/or diaporthalean-like perithecia), filamentous conidiophores (and/or clavate to elongate obovoid asci) and allantoid hyaline conidia (and/or ascospores) [5]. As plant pathogens, Cytospora species have also been reported to be associated with other diseases, such as root rot of Chinese jujube and collar rot of pomegranate [7,8].

Phylogenetic Analysis
The current isolates were initially identified as Cytospora sp. based on morphological observations and BLAST results. To clarify their further phylogenetic position, an analysis based on the 5 combined genes (ACT, ITS, RPB2, TEF1-α and TUB2) was constructed to compare Cytospora species from the current study with other strains in the GenBank database. Diaporthe vaccinii CBS 160.32 was selected as the outgroup in all analyses. Subsequent alignments for each gene were generated using MAFFT v.7 [16] and manually adjusted using MEGA v.6 [17]. Ambiguously aligned sequences were excluded from the analysis. Reference sequences were selected based on ex-type or ex-epitype sequences available from recently published literature [5,7,[18][19][20][21][22][23][24] (Table 1).
Phylogenetic analyses were formed by PAUP v.4.0b10 for the maximum parsimony (MP) method [25], MrBayes v.3.1.2 for the Bayesian inference (BI) method [26] and RAxML v.7.2.8 for the maximum likelihood (ML) method [27]. Tree length (TL), consistency index (CI), retention index (RI) and rescaled consistency (RC) were calculated [25]. ML analysis was generated using a GTR+G+I model of site substitution following recent study [4], including estimation of gamma distributed rate heterogeneity and proportion of invariant sites [27]. Branch support was evaluated with a bootstrapping method of 1000 replicates [28]. BI analysis was performed using a Markov chain Monte Carlo (MCMC) algorithm with Bayesian posterior probabilities [29]. A nucleotide substitution model was estimated by MrModeltest v.2.3 [30] and a weighted Bayesian analysis was considered. Two MCMC chains were run from random trees for 10,000,000 generations and trees were sampled each 100th generation. The first 40% of trees were discarded as the burn-in phase of each analysis and the Bayesian posterior probability (BPP) was calculated to assess the remaining trees [29]. The branch support from MP and ML analysis was evaluated with a bootstrapping (BS) method of 1000 replicates [28]. Phylograms were constructed using Figtree v.1.3.1 [31]. Sequence data were deposited in GenBank. The aligned matrices used for phylogenetic analysis were submitted through TreeBASE (www.treebase.org; study ID S26000).

Pathogenicity Test
Three Cytospora strains (CFCC 54184, CFCC 54056 and CFCC 54057) obtained in this study were used to conduct the pathogenicity test. The pathogenicity test was performed on 1-year-old E. alatus plants obtained from seeds kept in a greenhouse at constant 28 • C and 99% relative humidity. On healthy plants, twigs to be used for inoculation were surface disinfected with 75% ethanol for 1 min. The bark surface of each disinfected twig was scalded with a sterilized inoculating loop within a region 5 mm in length to a depth of 2 mm. For mycelial inoculation, a 5 mm diameter PDA plug with mycelium was taken from a 3-day-old colony and inoculated onto the wounded twigs. Three replicates were conducted for each isolate. Non-colonized PDA plugs and sterile water were used as negative controls. Pathogenicity was determined by the length of the necrotic lesion caused by the tested isolates, which was measured 3 weeks after inoculation. Fungal isolates were re-isolated from the infected tissue, and morphological characterization and DNA sequence comparisons were conducted to follow Koch's postulates.

Temperature and pH Tests
The 3 Cytospora isolates showed similar growth characteristics, so we used the type strain of the new species (CFCC 54057) to evaluate the effects of temperature and pH on colony growth using PDA plates. Tested temperatures ranged from 0 to 40 • C at intervals of 5 • C (i.e., 0, 5,10,15,20,25,30,35 and 40 • C). In order to clarify the effect of pH on radial mycelial growth, PDA medium was adjusted with 0.1 M NaOH and 0.1 M HCl to obtain pH values from 2.0 to 12.0 at intervals of 1.0 (i.e., 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0 and 12.0). A 5 mm diameter mycelial plug was placed in the centre of a 90 mm petri dish with PDA medium and incubated at 28 • C in the dark, with 3 replicates for each treatment. The effects of pH and temperature on mycelial growth were determined by measuring the colony diameter after 24, 48, 72 and 96 h of incubation and the data were converted to radial growth in millimetres [32]. Data were analysed in IBM SPSS Statistics v.22.0 (IBM Inc., Armonk, NY, USA) to select the model that best fit the individual data points, and SPSS was used to confirm the selected model. The optimal temperature and pH value of the regression curves were calculated based on the regression equations generated by IBM SPSS Statistics, and output figures with Origin v.8.0.

Carbon Colony Growth Test
To investigate the utilization of carbon sources, the type strain of the new species (CFCC 54057) was incubated in the dark at 28 • C on PDA medium for 4 days. PDA medium was used as the base medium (potato 20 g, sucrose 20 g, agar 17 g, distilled water to complete 1000 mL). The 20 g of sucrose was replaced by 20 g of fructose, galactose, glucose, maltose, sucrose or xylose to test these compounds as carbon sources. A 5 mm diameter PDA plug of mycelium was transferred to the centre of each sole carbon source medium. Colony growth was determined by measuring the colony diameters after incubation for 24, 48, 72 and 96 h at 28 • C in the dark, and the results were subsequently converted to radial growth [32]. Mean comparisons were conducted using Tukey's honestly significant difference (HSD) test (α = 0.05) in SigmaPlot v.14.0.

Phylogenetic Analyses
A combined matrix of five gene sequences of Cytospora species was constructed. The combined alignment matrices (ACT, ITS, RPB2, TEF1-α and TUB2) included 192 accessions (3 from this study and 189 retrieved from GenBank) and counted 3056 characters including gaps (350 characters for ACT, 631 for ITS, 726 for RPB2, 725 for TEF1-α and 624 for TUB2), of whih 1594 characters were constant, 130 variable characters were parsimony-uninformative and 1349 (44.14%) characters were variable and parsimony-informative. The MP analysis generated 200 parsimonious trees, the first of which is presented in Figure 1 (TL = 8,573, CI = 0.312, RI = 0.788, RC = 0.246). The tree topologies of ML and BI analyses were similar to that of the MP tree.
Based on the initial analysis, a second, more inclusive combined matrix was constructed using 27 accessions from the first dataset. The second combined alignment matrix counted 2531 characters including gaps (274 characters for ACT, 529 for ITS, 726 for RPB2, 553 for TEF1-α and 449 for TUB2). In total, 1,819 characters were constant, 182 variable characters were parsimony-uninformative and 547 (21.61%) characters were variable and parsimony-informative. The MP analysis generated one parsimonious tree and the best tree (TL = 1,225, CI = 0.768, RI = 0.853, RC = 0.656) is presented in Figure 2. The tree topologies of ML and BI analyses were similar to that of the MP tree.
Based on the multilocus phylogeny and morphology, all three strains were assigned to one new species, named Cytospora haidianensis, representing a monophyletic clade with high support value (MP/ML/BI = 100/100/1).

Pathogenicity Test
The three Cytospora haidianensis strains (CFCC 54184, CFCC 54056 and CFCC 54057) tested in this study were pathogenic on the Euonymus alatus twigs. No symptoms were observed in the non-inoculated controls. Brown lesions appeared at the inoculated points after 7 days of inoculation. The diseased spots turned brown and lesion areas were up to 16 mm long at 14 days after inoculation. By the third week after inoculation, the length of the brown necrotic lesions ranged from 36 to 45 mm (Figure 4). Koch's postulates

Pathogenicity Test
The three Cytospora haidianensis strains (CFCC 54184, CFCC 54056 and CFCC 54057) tested in this study were pathogenic on the Euonymus alatus twigs. No symptoms were observed in the non-inoculated controls. Brown lesions appeared at the inoculated points after 7 days of inoculation. The diseased spots turned brown and lesion areas were up to 16 mm long at 14 days after inoculation. By the third week after inoculation, the length of the brown necrotic lesions ranged from 36 to 45 mm ( Figure 4). Koch's postulates were performed by successful re-isolation of fungal strains from all necrotic twigs inoculated with Cytospora haidianensis. The morphology and DNA sequences of the isolates retrieved from the inoculated twigs were consistent with those of the strains used for inoculation.

Effects of Temperature and pH on Mycelial Growth
Colonies of C. haidianensis grew on PDA in the temperature range from 5 to 35
Forests 2020, 3, x www.mdpi.com/journal/forests regression analysis, the optimal growth for C. haidianensis after incubation was estimated to occur at 19.8 °C ( Figure 5).   Figure 6). Based on these regression equations, the optimal growth of C. haidianensis after 24 and 48 h incubation was estimated to be at pH 8.3.

Effects of Carbon Sources on Mycelial Growth
Cytospora haidianensis was able to grow using all six carbon sources tested. After 24 h, the utilization of sucrose was significantly greater than galactose, while there was no difference among fructose, glucose, xylose and maltose, which were slightly less well utilized than the other three carbon sources. The utilization of galactose was significantly lower than that of all other carbon sources tested. However, after 96 h, sucrose utilization was significantly higher than galactose and xylose, while there was no difference between fructose and glucose. Galactose had the lowest level of carbon utilization (Figure 7).

Effects of Carbon Sources on Mycelial Growth
Cytospora haidianensis was able to grow using all six carbon sources tested. After 24 h, the utilization of sucrose was significantly greater than galactose, while there was no difference among fructose, glucose, xylose and maltose, which were slightly less well utilized than the other three carbon sources. The utilization of galactose was significantly lower than that of all other carbon sources tested. However, after 96 h, sucrose utilization was significantly higher than galactose and xylose, while there was no difference between fructose and glucose. Galactose had the lowest level of carbon utilization (Figure 7).

Effects of Carbon Sources on Mycelial Growth
Cytospora haidianensis was able to grow using all six carbon sources tested. After 24 h, the utilization of sucrose was significantly greater than galactose, while there was no difference among fructose, glucose, xylose and maltose, which were slightly less well utilized than the other three carbon sources. The utilization of galactose was significantly lower than that of all other carbon sources tested. However, after 96 h, sucrose utilization was significantly higher than galactose and xylose, while there was no difference between fructose and glucose. Galactose had the lowest level of carbon utilization (Figure 7).

Discussion
In the present study, three specimens were collected from symptomatic branches and twigs associated with dieback disease of Euonymus alatus in Beijing, China. A novel fungal species, C. haidianensis, was introduced based on molecular, morphological and physiological data, and confirmed as the causal agent after pathogenic tests.
Pathogenicity tests were conducted on 1-year potted E. alatus plants in a greenhouse. The results indicated that C. haidianensis was pathogenic on E. alatus twigs. According to Pan et al. [7], Cytospora species invade the xylem and cause mortality of the whole branch, similar to the results obtained in this study within three weeks, showing the typical stem blight that occurred in the sampled place ( Figure 4). The growth temperature for phytopathogenic fungi is generally from 10 to 35 • C, optimally from 20 to 30 • C [33]. For instance, the optimal growth temperature of Penicillium cellarum causing rot in stored sugar beet roots was reported as 22 • C [34]; for Diaporthe neotheicola and D. ambigua causing dieback blueberry in Chile, it was 25 • C; for Diaporthe sp., it was 22 • C [35]; and for Phoma sorghina, which was found to cause twisted leaf disease in sugarcane in China, it was 20-25 • C [36]. The mycelia of C. haidianensis grew from 5 to 35 • C, with an optimal temperature of 19.8 • C ( Figure 5).
Most phytopathogenic fungi grow optimally in a pH range between 5 and 6.5 [37]. For Lasiodiplodia vaccinii, the range was 5.0 to 7.0, though it could still grow slowly at pH of 4.0 or 10.0 [33]. Similar results have been reported for L. theobromae, which could grow on media with a pH range from 4.0 to 10.0, with the optimal pH in the range of 5.0 to 7.0 [36]. The optimal pH value for C. haidianensis was from 8.0 to 10.0, though it could still grow slowly at pH of 4.0 or 11.0 ( Figure 6). All six carbon sources tested in this study contributed to the growth of C. haidianensis, with less utilization of xylose than all the other carbon sources used (Figure 7).
The dieback in Euonymus alatus caused by C. haidianensis damages the plants. Cytospora haidianensis blights many branches and leaves discolouration, causing gradual death of a large number of E. alatus (Figure 4). This phenomenon is not confined to Beijing; Cytospora euonymicola was also reported as a pathogenic fungus from Euonymus in Shaanxi Province, and Cytospora euonymina was also found in Euonymus in Shanxi Province [3]. A similar phenomenon also happens in other countries; Cytospora euonymi was also associated with the blight of Euonymus twigs in the USA and Europe. Other genera such as Cercospora, Colletotrichum, Coniothyrium and Fusarium were also reported to be pathogenic fungi in Euonymus [38].
To date, C. haidianensis has been found only from Euonymus alatus in Beijing. Management practices, including better ventilation and lighting, might help to alleviate the damage resulting from stem dieback caused by C. haidianensis. The distribution and host spectrum of C. haidianensis need further study.

Conclusions
A novel fungal species, Cytospora haidianensis, is an emerging pathogen on Euonymus alatus dieback disease in Beijing, China. The new species is the causal agent for E. alatus by Koch's postulates that grows best at 19.8 • C, pH 8.3. All the six carbon sources tested support the growth of C. haidianensis with the sucrose utilization is significantly higher than others.
Author Contributions: Experiments: X.Z., M.P. and H.L.; writing-original draft preparation: X.Z.; writing-review and editing: X.F. and C.T. All authors have read and agreed to the published version of the manuscript.
Funding: This study was financed by the Fundamental Research Funds for the Central Universities (2019ZY23) and College Student Research and Career-creation Program of Beijing (X201910022006).