Etiology of Cyclocarya paliurus Anthracnose in Jiangsu Province, China

Cyclocarya paliurus is an extremely valuable and multifunctional tree species whose leaves have traditionally been used in used in medicine or as a medicinal tea in China. In recent years, anthracnose has been frequently observed on young leaves of C. paliurus in several nurseries located in Jiangsu Province, resulting in great yield and quality losses. To date, no information is available about the prevalence of C. paliurus anthracnose in China. The main purpose of the present study was to characterize the etiology of C. paliurus anthracnose. Phylogenetic analysis of the eight-loci concatenated dataset revealed that all 44 single-spore Colletotrichum isolates belonged to three species in the Colletotrichum gloeosporioides species complex, namely, Colletotrichum aenigma, Colletotrichum fructicola, and C. gloeosporioides sensu stricto. Phenotypic features, including the colony appearance and the morphology of conidia, appressoria, and ascospores, were consistent with the phylogenetic grouping. Virulence tests validated that the three Colletotrichum species could cause typical symptoms of anthracnose on C. paliurus leaves, similar to those observed in the field. The optimum mycelial growth temperature ranged from 25 to 30°C for all representative isolates, while C. gloeosporioides s. s. isolates exhibited greater tolerance to high temperature (40°C). Fungicide sensitivity assays indicated that all three Colletotrichum species were sensitive to tetramycin, which may be a potential alternative for the management of C. paliurus anthracnose. To our knowledge, this study provides the first report of C. aenigma, C. fructicola, and C. gloeosporioides s. s. causing C. paliurus anthracnose in China as well as in the world.


INTRODUCTION
Cultivated for fine timber and as a medicinal plant, Cyclocarya paliurus is the sole extant species in the genus Cyclocarya and is native to China, naturally distributed in mountainous regions in the middle and lower reaches of the Yangtze River (Fang et al., 2011;Deng et al., 2015;Xie et al., 2015;Zheng et al., 2020). In Chinese folklore, C. paliurus is commonly called the "sweet tea tree, " and its leaves have traditionally been used as drug formulations for the treatment of obesity or diabetes mellitus (Fang et al., 2011;Cao et al., 2017;Xie et al., 2018). In recent years, increasing attention has been paid to C. paliurus because phytochemical studies have demonstrated that the extracts of its leaves possess a wide range of biological activities beneficial to human beings, such as antihypertensive (Xie et al., 2006), hypoglycemic (Wang et al., 2013), anti-HIV-1 , antioxidant (Xie et al., 2010;Wang et al., 2013;Liu et al., 2018a,b), antitumor (Liu et al., 2018b), and anticancer (Xie et al., 2013) activities. Current focal studies of C. paliurus have concentrated on producing or identifying the bioactive components in its leaves. Unfortunately, to date, no information about C. paliurus anthracnose is available, and this disease could become a limiting factor affecting the C. paliurus tea industry.
The Coelomycetous genus Colletotrichum Corda includes plant pathogens responsible for anthracnose diseases with a global distribution Wikee et al., 2011;Cannon et al., 2012;Dean et al., 2012;He et al., 2019). From the perspective of economic and scientific importance, Colletotrichum was denoted the eighth most significant fungal phytopathogen group worldwide (Dean et al., 2012), attacking over 3200 dicot and monocot plant species (Manire et al., 2002;O'Connell et al., 2012). The morphological taxonomy of Colletotrichum species has historically been arduous owing to overlapping characteristics, and the morphology sometimes varies with environmental factors (e.g., temperature, illumination, etc.) in culture (Freeman et al., 1998;Cannon et al., 2012;Damm et al., 2019). Molecular tools have been widely applied to effectively identify and define fungi at the species level. In recent years, the majority of studies regarding anthracnose were conducted principally via morphology and multigene phylogeny based on modern taxonomic concepts, which provides a more precise and robust solution Liu et al., 2014Liu et al., , 2015De Silva et al., 2017a;Diao et al., 2017;Guarnaccia et al., 2017;Fu et al., 2019).
Colletotrichum spp. infections initially occur via the attachment of spores to the host plant surface, followed by spore germination and the formation of an appressorium, which penetrates the plant cuticle. This process suggests that appressoria and spores play a critical role in the infection cycle and that certain highly inhibitory substances against spore germination and appressorium production would be potential alternatives to control anthracnose (De Silva et al., 2017b;Gao et al., 2020;Konsue et al., 2020). Currently, chemical pesticides are identified as the principal agents used for anthracnose management (Bi et al., 2011). However, excessive use of such chemicals has also brought a series of challenges over time Hu et al., 2015;Duan et al., 2018), including pathogen resistance and residual toxicity that affects human health and the environment (Kim et al., 2015;Alijani et al., 2019). The selection of environmentally safe, high-efficacy and relatively new fungicides is therefore imperative.
The application of antibiotic fungicides derived from metabolites of beneficial microbes to control phytopathogens has recently attracted increased attention since these compounds have been found to be environmentally friendly and may help to overcome pesticide resistance due to their low toxicity to non-target organisms and structural versatility (Moreira and May De Mio, 2015;Simionato et al., 2017;Han et al., 2020), offering a safe and effective way to circumvent the drawbacks of chemically synthesized pesticides and decreasing the environmental risks associated with their contamination (Ma et al., 2018a;Zhu et al., 2018).
Tetramycin, the fermentation metabolite of Streptomyces ahygroscopicus, exhibits excellent inhibitory activity against many plant pathogens, including Botrytis cinera, Passalora fulva, Phytophthora capsici, and Pyricularia oryzae Ren et al., 2014;Song et al., 2016;Ma et al., 2018a), which has been registered to manage rice and fruit crop diseases in China . On the other hand, a previous study reported that tetramycin has the potential to elicit disease resistance by activating plant defensive enzymes, including polyphenol oxidase (PPO), peroxidase (POD), and phenylalanine ammonia lyase (PAL) . Owing to its environmental friendliness and high efficiency, tetramycin has become the preferred fungicide in recent years (Song et al., 2016;Ma et al., 2018a,b).
Kasugamycin, the fermentation product of Streptomyces kasugaensis, is a member of the aminoglycoside antibiotic (Uppala and Zhou, 2018). It was originally developed as a biofungicide for the management of rice blast caused by P. oryzae. Kasugamycin inhibits protein biosynthesis, with both fungicidal and bactericidal activities (McGhee and Sundin, 2011). Due to it is high efficiency and friendliness to environment, the use of Kasugamycin in United States has been approved by EPA for controlling diseases of several pome fruits in the past decade 1 .
In 2018, during an investigation of C. paliurus, serious anthracnose symptoms ( Figure 1A) were observed in several nurseries located in the scientific research base of Nanjing Forestry University in Baima town (Baima), Nanjing. Over a half of the leaves were infected in Baima based on our observation. This anthracnose has been considered an emerging disease, but it is becoming endemic; nevertheless, the etiology, epidemiology, and management of this disease are uncertain. Hence, the objectives of the present study were to (1) accurately identify the Colletotrichum spp. causing C. paliurus anthracnose in Jiangsu Province, China, combining morphological and biological characteristics with molecular phylogenetic analyses; (2) examine the virulence of these fungi on C. paliurus leaves in vitro; and (3) characterize and compare the inhibitory effects of biofungicides against different Colletotrichum spp. in vitro.

Field Survey and Sampling
A field survey of C. paliurus anthracnose was carried out in Nanjing (five nurseries), Changzhou (four nurseries), and Yancheng (four nurseries) in September and October 2018 during the late growing season. Disease incidence was calculated as the percentage of trees displaying anthracnose symptoms out of the total number of evaluated trees (Bautista-Cruz et al., 2019). Three leaves exhibiting typical symptoms of anthracnose were randomly sampled per plant, and at least 10 symptomatic plants were sampled per nursery. All samples were then packaged in self-sealing bags and transported in an ice chest to the laboratory and then stored at 5 • C prior to isolation.

Colletotrichum Isolation
To isolate the fungus, small sections (4-by-4 mm pieces) were removed from the margin of leaf lesions, surface disinfected in 1% (vol/vol) NaClO 3 for 45 s and 75% ethanol for 30 s, rinsed in sterile distilled water three times, and air-dried on sterilized paper. The sections were then cultured onto 2% potato dextrose agar (PDA) (five sections per plate) amended with 100 µg/mL ampicillin to inhibit bacterial growth and incubated at 25 • C in the dark. The emerging edges of the fungal mycelium were observed daily and transferred aseptically onto new PDA plates. Colonies similar in morphology to Colletotrichum spp. were purified using the monosporic isolation procedure described by Cai et al. (2009), and single-spore cultures were preserved in PDA slant test tubes at 4 • C for follow-up studies. All isolates used in this study were deposit in State Key Laboratory of Forest Protection in Nanjing Forestry University.

Molecular Identification and Phylogenetic Analysis
For further characterization of the Colletotrichum spp., total genomic DNA (gDNA) of all single-spore isolates was extracted following the CTAB method described by Than et al. (2008). The concentrations of gDNA extracts were adjusted to 100 ng/µL with autoclaved double distilled water (ddH 2 O) using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Madison, WI, United States) and stored at −20 • C before use. Polymerase chain reaction (PCR) amplification was performed for the following loci: the ITS region, calmodulin (CAL), β-tubulin (TUB), actin (ACT), chitin synthase 1 (CHS-1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), glutamine synthetase (GS), and Apn2-Mat1-2 intergenic spacer (ApMat) genes. PCR amplifications were conducted in a 25 µL volume, mixed with 8.5 µL of ddH 2 O, 1 µL of each primer (10 µM), 2 µL of template DNA, and 12.5 µL of 2 × PCR Taq Master Mix (Applied Biological Materials Inc., Canada), using an Eppendorf Nexus Thermal Cycler (Germany). A negative control was added in all amplifications, where an equal volume of ddH 2 O replaced the template DNA. The primers and PCR settings for each locus are shown in Table 1. Amplification products were purified and sequenced by Jie Li Biotech Company (Shanghai, China). Forward and reverse DNA sequences were assembled and manually edited where necessary using Bioedit software (version 7.0.5 2 ), and the consensus sequences were deposited in GenBank ( Table 2). Reference sequences from extype or other authoritative specimens of Colletotrichum spp. were retrieved from GenBank and aligned with sequences generated herein for constructing phylogenetic trees, with C. boninense (MAFF 305972) used as an outgroup (Tables 2, 3).
The phylogenetic analysis for each individual locus and the concatenated matrix were inferred under the Bayesian inference (BI) and maximum-likelihood (ML) criteria in MrBayes 3.2.6 (Ronquist et al., 2012) and MEGA 7 (Kumar et al., 2016), respectively. For BI analysis, the best nucleotide substitution model of each locus was ascertained by MrModeltest 2.3 according to AICc, with K2 + I identified for CHS, TN93 identified for GADPH, K2 + G identified for ACT and ApMat, GTR + G identified for CAL and GS, and TN93 + G identified for ITS and TUB. Four Markov chains were run for 30 million generations simultaneously, with trees sampled every 1000 generations. The first 25% of trees were discarded as the burn-in phase of the analyses, while the remaining trees were used for calculating posterior probabilities (PPs) in the majority rule consensus tree. ML analysis was performed based on the GTR + G + I model, and clade support was determined by 1000 bootstrap replicates, with gaps treated as missing data.

Morphological and Biological Characterization
Fourteen representative isolates were selected for further studies according to BI/ML phylogenetic analysis ( Mycelial blocks (2 mm in side length) aseptically taken from actively growing cultures were transferred to new PDA plates and incubated at 25 • C in darkness. Colony characteristics, including conidiomata or ascomata production, were determined up to 30 days post-inoculation (dpi). Conidia, appressoria, ascospores, and asci for microscopy were obtained and examined according to the procedure described by Weir et al. (2012). At least 30 measurements per structure were recorded at ×100 magnification using a ZEISS Axio Imager A2m microscope (Carl Zeiss, Göttingen, Germany) equipped with differential interference contrast (DIC) optics. To observe fungal structures developed on infected tissue, leaves showing typical symptoms of anthracnose were collected and prepared using the method of Huang et al. (2018), with photomicrographs taken by a Regulus 8100 field emission scanning electron microscope (FE-SEM, Japan).
To determine the optimal temperature for colony growth, mycelial blocks (2 mm in side length) of 14 representative isolates were cultured as described above and incubated at temperatures of 5-40 • C with 5 • C intervals. The colony diameter was measured at two perpendicular angles, and the average was taken at 4 dpi. Five replicates per isolate were examined at all eight temperatures, and the experiment was conducted twice. Differences in the morphological and biological characteristics of the isolates were determined by one-way analysis of variance (ANOVA) using IBM SPSS Statistics 24.0 software (SPSS, Inc., Chicago, IL, United States).

Virulence Tests of Colletotrichum Isolates
Virulence tests were conducted with reference to previous reports with minor modifications Chen Y. et al., 2017;Xue et al., 2019). Fourteen representative Colletotrichum isolates were selected and cultured on PDA and used for virulence tests on detached C. paliurus leaves under controlled conditions ( Table 2). Conidial suspensions of each isolate were prepared as previously described and adjusted to two concentrations of 1 × 10 6 and 1 × 10 8 conidia/mL with ddH 2 O.
Asymptomatic C. paliurus leaves were surface disinfected and air-dried as mentioned above, and then one piercing wound was made on the right side of each leaf using a sterile needle (insect pin, 0.71 mm in diameter), or the leaves were left unwounded. Wound inoculation was performed by placing an 8 µL conidial suspension (1 × 10 6 conidia/mL) or mycelial blocks (5 mm in length) from margins of actively growing colonies onto each stab wound. Non-wound inoculation was conducted by placing an 8 µL spore suspension (1 × 10 8 conidia/mL) or mycelial blocks onto the mid-right region of the leaves without pin pricking. Leaves inoculated with ddH 2 O or non-colonized PDA blocks were treated as negative controls. The experiment was conducted in triplicate for each treatment and control, involving five leaves per replicate. All treatments and controls were placed into transparent containers (334 × 215 × 87 mm) lined with moist sterile filter paper and sealed by plastic wrap to maintain a high relative humidity and then incubated at 25 • C under a 12 h photoperiod in a growth chamber. The whole experiment was carried out twice. LG2 LG4  Citrus sinensis b Sequences in italics were generated in this study. ITS, internal transcribed spacers 1 and 2 together with 5.8S nrDNA; GAPDH, partial glyceraldehyde-3-phosphate dehydrogenase gene; CAL, partial calmodulin gene; ACT, partial actin gene; CHS-1, partial chitin synthase 1 gene; TUB2, partial beta-tubulin gene. c Isolates used for morphological and biological analysis, virulence tests, and biofungicide sensitivity assays. Disease incidence was determined at 10 dpi, while the incubated leaves were monitored for the onset of anthracnose lesions for up to 20 dpi. Virulence was determined by measuring the diameter of the necrotic lesions in two perpendicular directions at 7 and 10 dpi for the wounded and non-wounded leaves, respectively. Differences in the virulence of the isolates were determined by ANOVA, and mean values were compared by Tukey's test (P < 0.05) using SPSS as previously described. Each Colletotrichum isolate involved in the virulence test was reisolated from the inoculated leaves, and their identity was confirmed by morphological and molecular approaches as previously described to fulfill Koch's postulates.

Effects on Mycelial Growth
Phenazine-1-carboxylic acid [1% active ingredient (a.i.); Shanghai Non-gle Biological Products Co., Ltd., Shanghai, China], tetramycin (0.3% a.i.; Liaoning Wkioc Bioengineering Co., Ltd., Liaoning, China), and kasugamycin (4% a.i.; Shaanxi Microbe Biotechnology Co., Ltd., Shaanxi, China) were used. Fourteen representative isolates were selected based on the above studies. The fungicide sensitivity of each isolate was tested on complete medium (CM) plates (Yeast extract 10 g/L, Casamino-acid 5 g/L, Agar 15 g/L, 1% sterile glucose after autoclaving) amended with fungicides. Mycelial blocks (2 mm in side length) aseptically taken from actively growing cultures were placed onto CM with or without (control) fungicide amendments. The final concentrations of each a.i. in the amended media were 0.1, 0.25, 0.5, 1, 2.5, and 5 µg/mL for tetramycin and 1, 2.5, 5, 10, 25, and 50 µg/mL for PCA and kasugamycin. Each treatment was tested in triplicate, and the entire experiment was repeated twice. The mean colony diameter was measured at 4 dpi, and the formula for percent inhibition was [(radial growth of the control -radial growth at fungicide concentration)/radial growth of the control] × 100%. Half of the maximal effective concentration (EC 50 ) was estimated by regression to the log 10 probability conversion of the percentage of inhibition of the fungicide concentrations.

Effects on Spore Germination
Tetramycin was selected to test its ability to inhibit conidia germination. Spore suspensions and fungicide solutions were mixed with sterilized water to 10 mL volume. The final fungicide concentrations were 0.005, 0.01, 0.05, 0.1, 0.5, and 1 µg/mL, while spore suspension was adjusted to 1 × 10 5 spores/mL for each treatment. A 20 µL droplet of each suspension was placed on a hydrophobic cover slip and incubated at 25 • C for 18-20 h in a humidity chamber according to Fang et al. (2018). Each treatment was conducted in triplicate, and the entire experiment was repeated twice. Conidia were then observed at ×100 magnification using a ZEISS microscope and scored as germinated if the length of the germ tube was longer than half of the conidial length. The conidial germination inhibition rate was calculated as previously described (Munir et al., 2016).

Field Symptoms and Colletotrichum Isolates
In May 2018, typical symptoms of anthracnose were first observed on newly emerged leaves of C. paliurus in a commercial nursery in Baima (Figure 1A), and the infection quickly spread to all C. paliurus nurseries within the growing season, with the infection rate reaching 64% (150 trees were investigated). Similar symptoms were observed in plant bases at Changzhou and Yancheng, with infection rates over 35 and 45% (100 trees were investigated), respectively. The initial symptoms appeared in the form of subcircular or irregular pale-brown spots scattered on the leaves ( Figure 1C). Gradually, the lesions enlarged and coalesced to form large necrotic areas, which turned off-white surrounded by a dark-brown border as symptoms progressed ( Figure 1B). The dead tissue withered, resulting in premature defoliation of the plant in severe cases ( Figure 1A). Under high-moisture conditions, a number of acervuli were formed in concentric rings and oozed gelatinous orange spore masses ( Figure 1D). Photomicrographs further corroborated the presence of conidiophores and conidia on the surfaces of leaf lesions under optical or SEM microscopy (Figures 1E,F). A total of 44 monosporic Colletotrichum isolates were recovered from symptomatic tissues and used for further molecular identification ( Table 1). The shapes and sizes of conidia of these cultures were basically concordant with the sporulation on the lesions (Figures 2C, 3C, 4C). The general morphological characteristics of all isolates resembled those of Colletotrichum species.

Molecular Identification and Phylogenetic Analysis
In the present study, the ITS, CAL, ACT, GPDH, TUB, CHS-1, GS, and ApMat region/genes of all 44 monosporic isolates were successfully amplified and sequenced ( Table 1). Sequences generated herein along with reference sequences from ex-type or other authoritative specimens were concatenated for phylogeny construction, composing a dataset of 3192 characters, with 1828 constant characters, 574 parsimony-uninformative characters, and 790 parsimony-informative characters.
The topological structure of the phylogenetic trees constructed using BI and ML criteria was basically consistent, demonstrating that the evolutionary relationships of the experimental strains were statistically supported. A consensus tree with clade support from bootstrap proportions (BPs) and PP values was generated (Figure 5). The phylogenetic tree revealed that all 44 Colletotrichum isolates belonged to three well-separated clades and nested within the C. gloeosporioides species complex.

Morphological and Biological Analyses
Fourteen representative isolates clustered in three clades in the ML/BI phylogenetic analysis, including three of C. aenigma, seven of C. fructicola, and four of C. gloeosporioides s. s., were selected for further studies ( Table 1).
All 14 representative isolates tested exhibited a similar growth pattern on PDA at the different treatment temperatures. No mycelial growth of any tested isolates was observed in vitro at 5 • C. The optimum mycelial growth temperature of the three Colletotrichum species was 25-30 • C, but the high temperature tolerance of the three species was different. Isolates of C. aenigma and C. fructicola were more sensitive to high temperature and grew very slowly (or could not grow) at 40 • C, with mean growth rates lower than those of C. gloeosporioides s. s. isolates.

Virulence Tests of Colletotrichum Isolates
All 14 selected isolates were pathogenic on leaves of C. paliurus and reproduced typical symptoms of anthracnose. Seven days after wounded or non-wounded inoculation, distinct brown or off-white necrotic lesions with dark-brown boundaries developed (Figure 6), while no symptoms developed on the corresponding mock controls.
The severity of disease caused by these isolates showed significant differences ( Table 4). Isolates of C. gloeosporioides s. s. generally showed strong virulence, with mean lesion diameters ranging from 17.88 to 23.16 and 17.52 to 22.11 mm with wounded and non-wounded inoculation using mycelial plugs as inocula, respectively. There was no significant difference in virulence among C. fructicola isolates in C. paliurus leaves, with mean lesion diameters ranging from 16.65 to 20.52 and 17.41 to 21.09 mm with wounded and non-wounded inoculation using mycelial plugs as inocula, respectively. C. aenigma isolates showed much weaker virulence, with mean lesion diameters ranging from 12.38 to 14.89 and 11.78 to 14.12 mm with wounded and non-wounded inoculation using mycelial plugs as inocula, respectively. The lesions produced by mycelial inoculation were generally larger than those produced by spore suspension inoculation among the three Colletotrichum species (Table 4). Colletotrichum gloeosporioides s. s. isolate BM6 and C. fructicola isolate BM5 produced reproductive structures of the fungus on the necrotic lesions (Figure 6). C. fructicola isolate YH6 produced lesions with a wheel-shaped pattern on C. paliurus leaves (Figure 6). The Colletotrichum species were reisolated from all inoculated symptomatic leaves and were found to be morphologically and molecularly identical to the original isolates using the aforementioned methods, thus fulfilling Koch's postulates.

Sensitivity of Colletotrichum Isolates to Biofungicides
Fourteen representative isolates evaluated showed similar biological responses to all tested biofungicides. Kasugamycin at 50 mg/mL showed no suppressive activity against the mycelial growth of the three Colletotrichum spp. on CM medium (EC 50 > 100 µg/mL). PCA showed moderate inhibition of the  mycelial growth of the three Colletotrichum spp., with isolates of C. fructicola exhibiting more sensitivity to this biofungicide. The EC 50 of tetramycin against the mycelial growth of all representative isolates was lower than that of any of the other biofungicides, including the low EC 50 of tetramycin against spore germination, indicating that tetramycin was the most effective biofungicide against the three Colletotrichum spp. used in this study.

DISCUSSION
In recent years, the cultivation of C. paliurus has undergone a major expansion to meet the increasing demand for young leaves of this species for medical use or C. paliurus tea production in China, which may have caused the high incidence of foliar diseases in these newly established plantations. Therefore, it is of great importance to diagnose and control these fungal diseases of C. paliurus. Unfortunately, little information was available about these diseases, i.e., C. paliurus anthracnose. Hitherto, this study is first comprehensive analysis demonstrating the etiology of C. paliurus anthracnose in China, providing valuable information about the phenotypic and molecular characteristics, virulence, and fungicide sensitivity of the causal agents associated with this disease. Moreover, this study provides the first report of C. aenigma, C. fructicola, and C. gloeosporioides s. s. causing C. paliurus anthracnose in China as well as in the world.
Colletotrichum gloeosporioides species complex is regarded as the most challenging taxa within the Colletotrichum genus (Silva et al., 2012). Although polyphasic method is recommended for characterizing Colletotrichum species, there is still lack of consensus among taxonomists on the selection of markers for phylogenetic studies (Cao et al., 2020;Vieira et al., 2020). Recent studies revealed that concatenated GS and ApMat alignment can achieve a satisfactory Colletotrichum species identification Sharma et al., 2017). Conservative region/genes (ITS, GAPDH, CAL, CHS-1, ACT, and TUB) have been previously accepted for delimiting species in this species complex . Therefore, in the present study, eight loci (ITS, GAPDH, CAL, CHS-1, ACT, TUB, including GS and ApMat) were selected in phylogenetic analysis for Colletotrichum isolates classification. Based on BI/ML multilocus concatenated phylogenetic analyses, including sequences from 28 authentic specimens in the C. gloeosporioides species complex, the 44 isolates were categorized into three well-separated clades: six isolates clustered in the C. aenigma clade (14%), 28 isolates clustered in the C. fructicola clade (64%), and 10 isolates clustered in the C. gloeosporioides s. s. clade (22%). With respect to phenotypic characterization based on colony morphology, characteristics of conidia, appressoria, ascospores, and asci were entirely in line with the results of the molecular data.
Colletotrichum fructicola was first described by Prihastuti et al. (2009), causing coffee berry disease in Thailand. The species is geographically diverse and threatens a wide range of hosts, which has been reported on Fragaria × ananassa  , and Mangifera indica (China) (Mo et al., 2018). In the current study, C. fructicola was the most predominant species and exhibited strong pathogenicity (Table 4), which seems to be the most economically harmful species of C. paliurus anthracnose in Jiangsu Province, China. Colletotrichum gloeosporioides s. s., a genetically and biologically diverse species, previously reported to infect fruits in tropical area (Sangeetha and Rawal, 2008;Udayanga et al., 2013), which is probably related to its ability to tolerate high temperatures (Table 5). However, this species was recently reported increasingly prevalent in the temperate region, such as Hebei, Shandong, and Shanxi Provinces of China (Jayawardena et al., 2016;Wang et al., 2020). Results in the present study demonstrated that C. gloeosporioides showed the strongest pathogenicity to C. paliurus ( Table 4). The prevalence and ecological adaptation zone of C. gloeosporioides on C. paliurus in China would be further studied.
Interestingly, multiple Colletotrichum species were isolated and identified from the same leaf and even within the same lesion of single C. paliurus trees. As reported in previous studies, several Colletotrichum species can cause anthracnose on the same host (Munir et al., 2016;Chen Y. et al., 2017;De Silva et al., 2017a;Diao et al., 2017;Guarnaccia et al., 2017;Fu et al., 2019;Xue et al., 2019). It is reasonable to believe that C. paliurus anthracnose may be a complex disease. With more samples collected, it is possible that even more Colletotrichum species, or even novel species, will be characterized as responsible for this disease. Consequently, future attention should be given to probe Colletotrichum species collected from C. paliurus anthracnose in different geographical areas with different latitudes or elevations in China.
Temperature is an indispensable factor that affects epidemics of anthracnose or other plant diseases (Dubrulle et al., 2020). High temperatures and their frequency may be the factors leading to the delay or non-occurrence of plant diseases (Han et al., 2016;Xue et al., 2019). In the present study, no significant differences occurred in the optimum and minimum mycelial growth temperatures of C. aenigma and C. fructicola, while C. gloeosporioides s. s. isolates exhibited more tolerance to high temperature, which was in concordance with previous study results (Han et al., 2016). These data may provide useful information for C. paliurus anthracnose control strategies: fungicide applications should be timed before the optimum growth temperature is reached. Once infection occurs, the suppression of spore germination and mycelial growth within the plant tissue plays a crucial role in anthracnose management. In the present study, tetramycin showed excellent inhibitory effect on the mycelial growth and spore germination of the three Colletotrichum species ( Table 6). The satisfactory inhibitory activity against different life stages of Colletotrichum species indicates that tetramycin may be a potential alternative for the management of C. paliurus anthracnose. In previous studies, the excellent curative and protective activity of tetramycin has been widely reported in Phytophthora blight, rice blast, tomato leaf mold, Corynespora leaf spot, and cucumber gray mold Miao et al., 2015;Song et al., 2016;Ma et al., 2018a,b), demonstrating that tetramycin would be helpful to prevent the occurrence and spread of plant diseases throughout the field. Accordingly, protective and curative activity of tetramycin on C. paliurus anthracnose in the field trials would be further studied before it is put into use.

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
X-RZ was responsible for the entire process of experimentation and writing the manuscript. M-JZ helped perform the experiment and analyze the results. X-LS and S-ZF provided experimental materials. F-MC supervised the work. All authors contributed to manuscript revision and read and approved the submitted version.

FUNDING
This study was financially supported by the Postgraduate Research and Practice Innovation Program of Jiangsu Province (KYCX20_0875).