Species Diversity and Chemotypes of Fusarium Species Associated With Maize Stalk Rot in Yunnan Province of Southwest China

Maize stalk rot caused by Fusarium species is one of the most important fungal diseases of maize throughout the world. The disease is responsible for considerable yield losses and has also been associated with mycotoxin contamination of the crop. In this study, a survey of maize stalk rot was performed in seven locations of Yunnan Province in China during the cropping season of 2015 and 2016. Based on morphological and molecular characteristics, 204 isolates belonging to 12 Fusarium spp. from symptomatic stalks of maize were identified. Among the isolated strains, 83 were identified as Fusarium meridionale (40.5%), 46 as Fusarium boothii (22.5%), 34 as Fusarium temperatum (16.5%), 12 as Fusarium equiseti (5.9%), 10 as Fusarium asiaticum (4.9%), six as Fusarium proliferatum (3.0%), four as Fusarium verticillioides (2.0%), four as Fusarium incarnatum (2.0%), two as Fusarium avenaceum (1.0%), one as Fusarium cerealis (0.5%), one as Fusarium graminearum (0.5%), and one as Fusarium cortaderiae (0.5%). Fusarium cortaderiae was the first report on the causal agent of maize stalk rot disease in China. These isolates were divided into five chemotypes: nivalenol (NIV), deoxynivalenol (DON), beauvericin (BEA), zearalenone (ZEN), and fumonisin (FUM). Phylogenetic analysis based on partial sequences of the translation elongation factor 1α (TEF1-α) showed a high degree of interspecific polymorphisms among the isolates. Pathogenicity analysis on maize stalks indicated that all the 12 species of Fusarium were able to cause the disease symptoms with different aggressiveness. This study on population, pathogenicity, and toxigenic chemotypes of Fusarium species associated with maize stalk rot in Yunnan Province of southwest China, will help design an effective integrated control strategy for this disease.


INTRODUCTION
In the Yunnan Province of southwest China, maize plays a crucial role in local agricultural production. In this region, the maize crop's yield and quality are particularly affected by stalk rot diseases caused by Fusarium species. Fusarium is an important plant pathogenic fungus with a wide range of hosts, including corn, wheat, rice, and other cereal crops (Boutigny et al., 2011). These pathogens cause ear and stalk rot disease, potentially damaging to crop yield and food safety. Different Fusarium species can produce toxic chemicals known as mycotoxins, which can be an important risk to both animal and human health if accumulated to an unsafe level (Sampietro et al., 2012;Kuhnem et al., 2016).
Fusarium genus has numerous species, which are morphologically indistinguishable, so they are very difficult to identify at the species level (Thomas et al., 2019). Fusarium graminearum species complex (FGSC) has been divided into biogeographically distinct lineages consisting of at least 16 species. Members of the FGSC are also classified into the broader Fusarium sambucinum species complex (FSAMSC; Starkey et al., 2007;O'Donnell et al., 2008O'Donnell et al., , 2013Sarver et al., 2011). Various members of FGSC show different geographic distribution and host preferences (Lee et al., 2015). Among different species in FGSC, F. graminearum is considered as an important pathogen of maize (Moreno-González et al., 2004). Earlier studies have reported that F. graminearum could cause seedling blight and root rot (Du et al., 1997;Munkvold and O'Mara, 2002). However, a previous study reported the presence of Fusarium culmorum, Fusarium solani, Fusarium semitectum, Fusarium verticillioides, and F. graminearum from the lodged maize plants . Another study showed that F. graminearum was the most aggressive strain during pathogenicity tests on maize . In addition, F. graminearum is a dominant pathogen associated with Fusarium head blight (FHB) in North America and Europe (O'Donnell et al., 2004;Starkey et al., 2007), whereas Fusarium asiaticum has been found as a major species in Asia (Qiu et al., 2014). Fusarium graminearum is often found on wheat, but Fusarium boothii and Fusarium meridionale are frequent pathogens of maize, and Fusarium asiaticum is commonly reported from rice (Maier et al., 2006). Besides, F. verticillioides is one of the most common pathogens causing ear and stalk rot in maize. This species is widespread in areas with relatively warm and dry weather (Czembor et al., 2019), including the European and the Kansas state of the United States. In China, many Fusarium species are associated with ear and stalk rot diseases of maize, which resulted in significant yield losses and mycotoxin contamination problems. In China, the notable Fusarium species isolated from maize are F. verticillioides, F. graminearum, F. meridionale, and Fusarium temperatum (Duan et al., 2016). Fusarium temperatum is also an important maize pathogen and described as a new species causing disease in maize crop (Scauflaire et al., 2011). These pathogens can produce different toxigenic chemotypes, demonstrating the tremendous potential of this species for mycotoxin contamination (Duan et al., 2016). Moreover, isolates of F. asiaticum, isolated from head blight infected wheat plants produced  or nivalenol (NIV). It was also showed that the isolates producing different mycotoxins also have differences in growth rate, pathogenicity, conidial length, fecundity, trichothecene accumulation and showed a varying degree of resistance to benzimidazole (Zhang et al., 2012). Although, many Fusarium species have been reported to responsible for maize ear and stalk rot disease in China, no detailed studies have been done in Yunnan Province based on composition, pathogenicity, and toxigenic chemotypes.
As the largest grain crop in Yunnan Province, maize is distributed throughout the province. Yunnan has diverse environmental conditions and topography, where the maize planted in areas have temperature ranging from 9 to 30°C and altitude ranging from 700 to 2,400 m. So, climatic conditions, soil type, water availability, farming system, and planting habits vary significantly throughout the province. Besides, different maize varieties planted in different parts of Yunnan have various growth characteristics, and the yield also varies considerably among the different areas of the province. In this study, diseased stalks of maize were collected from seven locations of Yunnan Province during the cropping season of 2015 and 2016. The study aimed to determine species diversity, pathogenicity, and toxigenic chemotypes of the Fusarium species causing maize stalk rot in Yunnan Province to design an effective integrated control strategy for this disease.

Fungal Isolation, Purification, and Morphological Characterization
Stalks of maize showing typical rot symptoms were collected from seven maize-planting locations in Yunnan Province of China during the cropping season of 2015 and 2016 (Figure 1). The diseased samples were cut into small pieces (approximately 5 mm 2 ) and soaked in 75% ethanol for 2 min. Subsequently, washed three times with sterile water and dried using autoclaved tissue towels. Later, the samples were placed onto potato dextrose agar (PDA) plates, which were supplemented with streptomycin sulfate (150 μg/ml) and kanamycin (150 μg/ml). The PDA plates were incubated at 25°C for 2-3 days in darkness. Fungal colonies showing various morphological features were selected. Fungal isolates were grown on PDA after single spore purification by following the procedure described by Xi et al. (2019). Morphological features of the fungal isolates were observed on PDA and carnation leaf agar (CLA). The appearance of the fungal colonies was recorded after the mycelium fully covered the whole PDA plate. Six Fusarium species including Fusarium avenaceum, Fusarium cerealis, Fusarium equiseti, F. graminearum, F. proliferatum, and F. verticillioides were confirmed by following the details mentioned in The Fusarium Laboratory Manual (Leslie and Summerell, 2006). For the identification of the other six species such as F. meridionale, F. boothii, F. temperatum, F. asiaticum, Fusarium incarnatum, and Fusarium cortaderiae, recently published materials were followed Scauflaire et al., 2011; Frontiers in Microbiology | www.frontiersin.org 3 August 2021 | Volume 12 | Article 652062 Castañares et al., 2016;Walkowiak et al., 2016;Avila et al., 2019). The size of microconidia and macroconidia were taken as average from 50 measurements of each isolate.

Species and Chemotype Determination
About 10 mm mycelial plugs from the colony's edge were inoculated to CM liquid medium and incubated in a shaker without light (175 rpm, 25°C) for 5 days. After incubation, the mycelia were collected by centrifugation (4,000 rpm, 5 min) and stored at −80°C until the subsequent use. Total DNA was extracted using a ZR fungal DNA Kit (ZYMO Research, United States) by following the manufacturer's instruction and stored at −20°C until the subsequent use. Sequences of the translation elongation factor 1α (TEF-1α) from each isolate were amplified using primers EF-1 (5'-ATGGGTAAGGARGACAAGAC-3'), and EF-2 (5'-GGARGTACCAGTSATCATGTT-3'; Geiser et al., 2004). The resulted sequences were compared with the NCBI database 1 1 https://www.ncbi.nlm.nih.gov/ and Fusarium database (FUSARIUM-ID v.1.0 database) 2 for species determination.
To identify each isolate's chemotypes, six specific mycotoxinproducing genes were amplified by PCR using specific primers as previous described (Ward et al., 2002;Jennings et al., 2004;Kulik et al., 2007;Meng et al., 2010;Duan et al., 2016). The sequence of primers used to amplify these genes has been mentioned in Supplementary Table S1. The PCR was done in a 20 μl reaction mixtures including 1 μl of template DNA, 10 μl of 2× DreamTaq PCR Mix (Thermo Fisher Scientific, United States), 7 μl of sterile water, and 1 μl of each primer (10 μM). Amplification reactions were carried out in a C1000 Touch thermal cycler (Applied Biosystems, BIO-RAD, United States).

Phylogenetic Analysis of the TEF-1α Gene Sequence Data
The TEF1-α gene always appeared to have a single copy in Fusarium and showed high levels of sequence polymorphism in closely related species . All of the 2 http://isolate.fusariumdb.org/blast.php sequences (n = 204) were aligned online using the MAFFT alignment program (Katoh and Standley, 2013). Alignments were adjusted manually using Clustal X (Thompson et al., 1994). A phylogenetic tree from multiple alignments of the 204 sequences was constructed using the neighbor-joining method calculated with MEGA X (Sudhir et al., 2018). The Interactive Tree of Life 3 was used to beautify the phylogenetic tree. Clade support was inferred from 1,000 bootstrap replicates.

Pathogenicity Tests on Maize Stalks
B73 maize plants were inoculated at the 10-leaves stage by punching a hole in the stalk at the second or third internode above the soil line using a sterile toothpick. Then 20 μl conidia suspension was injected from representative isolates at a concentration of 10 6 /ml. Mock-inoculated maize stalks were treated with sterilized water. The inoculation site was wrapped using a piece of sterilized gauze to conserve moisture and avoid any contamination. Each representative isolate and control were inoculated on three plants. After 7 days post-inoculation (dpi), the stalks of inoculated plants were split along the longitudinal direction for symptom measurements. The longitudinal brown infected areas were measured as the necrosis area to calculate each identified Fusarium species' virulence using ImageJ software (Zhang et al., 2016).

Analysis of Toxigenic Chemotypes
In FGSC, the Tri genes cluster is responsible for the production of different types of toxins. Three primers based on Tri3, Tri7, and Tri8 intergenic sequences, Tri315F/R, nivPF/R, and MinusTri7F/R, were used to amplify specific 15-AcDON fragments of 864 bp, NIV fragments of 450 bp, and 3-AcDON fragments of 483 bp, respectively. Similarly, the FUM1 gene was used to detect the Fumonisins (FBs) with a fragment of 750 bp. Whereas the esyn1 gene was used to detect the beauvericin (BEA) with a fragment of 600 bp. Also, the PKS4 gene was used to detect the zearalenone (ZEN) with a fragment of 280 bp.
The PCR amplification results showed that all of the 12 Fusarium species can synthesize mycotoxins and the

TEF-1α Sequences
For phylogenetic analysis, a neighbor-joining tree was constructed using the partial TEF-1α gene sequences, including all isolates in this study (Figure 4) . proliferatum, F. temperatum, and F. verticillioides formed an independent branch in the phylogenetic tree owing to these isolates belonging to the FFSC. Similarly, isolates of F. incarnatum and F. equiseti formed an independent branch because of these isolates belonging to the FIESC. Likewise, isolates of F. avenaceum classified into FTSC showed another independent branch in the tree (Figure 4). These results indicated that isolates of Fusarium species showed a high degree of interspecific polymorphisms variation and was unrelated to geographic distribution.

Pathogenicity Tests on Maize Stalks
To test the pathogenicity of the 12 isolated Fusarium species, the stalks of B73 maize plants at the 10-leaf stage were inoculated with each representative fungal species. The symptoms and severity of the disease were recorded at the 7 dpi. The results showed that all of the Fusarium species are pathogenic to maize stalks and showed distinct discoloration of internal stalk tissues around the inoculation site ( Figure 5A). The longitudinal brown infected areas of maize stalks were measured to evaluate the virulence of each identified Fusarium species (Figure 5B). The results indicated that isolates of F. meridionale are the most aggressive among all of the isolates. Pathogenicity of these isolates was confirmed by reisolating the fungus from symptomatic tissues but not from the control plants.

DISCUSSION
Fusarium spp. can cause various diseases at different growth stages of maize, such as root, seedling, stalk, and ear rot, leading to yield losses, and reduction of grain quality (Pfordt et al., 2020). Most of Fusarium spp. can produce different mycotoxins to contaminate small grain crops from pre-harvest to post-harvest stages. The gene of TEF1-α always appeared to have a single copy in Fusarium. It showed high levels of sequence polymorphism in closely related species, and the DNA sequence based on the TEF1-α was often used to identify the putative Fusarium species Berruezo et al., 2018;Fang et al., 2020;Wang et al., 2021). In this study, 12 Fusarium species were isolated and identified from symptomatic maize stalks based on morphological characteristics, phylogenetic analysis (TEF1-α), and Koch's postulates. Among them, F. meridionale (40.5%), F. boothii (22.5%), and F. temperatum (16.5%) were more prevalent. For the consecutive two cropping seasons of 2015 and 2016, we find the F. meridionale, F. boothii, F. asiaticum, F. proliferatum, F. temperatum, F. verticillioides, F. incarnatum, and F. equiseti Munkvold, 2008), through young leaf sheaths, by seed transmission and via wounds caused by hail or insect feeding (Gai et al., 2018). Stalk rot of maize results in defective grain filling, premature senescence, and lodging, which negatively affects production, harvesting, and yield (Quesada-Ocampo et al., 2016). In the present study, the pathogenicity tests showed that all of the Fusarium isolates could cause severe symptoms of maize stalk rot, but the extent of lesion spread was different. Fusarium meridionale was the most aggressive species to infect maize stalks. So, we have reasons to believe that differences in compositions of Fusarium spp. associated with maize stalk rot disease in Yunnan Province were caused by local climatic conditions. To investigate the ability to produce mycotoxins of Fusarium species causing maize stalk rot in Yunnan Province, the toxigenic chemotypes were also evaluated by specific PCR assays. The pathogenicity analysis showed that there was no relationship between the pathogenicity and the type of mycotoxin production. Generally, pathogenicity was not influenced by the type of mycotoxin produced (Adams and Hart, 1989;Goswami and Kistler, 2005). However, the pathogenicity of the F. graminearum to wheat plant has a relationship with the type of the mycotoxin (Shin et al., 2018). Another study on FSAMSC reported that the aggressiveness of the pathogen was related to the type of mycotoxin produced by the pathogen (Laraba et al., 2021). Our results indicated that 45.6% (93/204) strains were NIV producers, whereas 26.0% (53/204) stains were DON producers. So, the contaminations of NIV and DON in maize-related agro-products should be given particular attention in Yunnan Province of China. Besides, the identification of F. cortaderiae, F. cerealis, and F. avenaceum are reported for the first from Yunnan Province, which needs urgent attention to prevent their widespread. These results will provide useful information to design an effective strategy for the control of disease caused by Fusarium species in Yunnan Province of China.

CONCLUSION
In the 2 years of investigation, F. meridionale (40.5%), F. boothii (22.5%), and F. temperatum (16.5%) were the most frequent Fusarium species to cause maize stalk rot disease in Yunnan Province of China. The dominance of the NIV chemotype among isolates needs to pay more attention to food safety and animal health because of the more significant toxic potential of NIV relative to DON. Besides, F. temperatum associated with BEA mycotoxins represented a toxigenic risk for maize production. The current results on species diversity of Fusarium spp. and mycotoxin contaminations associated with maize stalk rot disease will provide valuable information to design effective strategies to control the disease caused by Fusarium spp.

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
WG conceived and designed the experiments. WG and JZ collected the samples in the field. YY and GZ provided substantial assistance to collect the samples in the field. KX and LS performed the experiments. KX and WG wrote and edited the manuscript. All authors contributed to the article and approved the submitted version.