The Fusarium fujikuroi species complex (FFSC) represents one of the most taxonomically diverse groups within the genus Fusarium, comprising over 90 described species, and this complex is phylogenetically structured into five principal clades: the African clades A, B, and C, the American clade, and the Asian clade (Han et al., 2023; Wang et al., 2022; Yilmaz et al., 2021; Zhang et al., 2024, 2025). Members of the FFSC are widely recognized as pathogens of economically important crops and forest species, such as rice, maize, and pine, causing significant agricultural losses globally (Bashyal et al., 2016; Han et al., 2023; Herron et al., 2015; Qiu et al., 2020; Zainudin et al., 2008). In addition to their pathogenicity, several species within the complex, such as F. fujikuroi s. str., F. konzum, F. annulatum (formerly reported as F. proliferatum), F. subglutinans, and F. verticillioides, are known as producers of mycotoxins, including fumonisins, fusaproliferin, moniliformin, and beauvericin, which frequently contaminate cereal grains and other crops, posing serious risks to both animal and human health (Bertuzzi et al., 2020; Choi et al., 2018b; Leslie et al., 2004; Logrieco et al., 1996, 1998; Wu, 2007).
Historically, Fusarium taxonomy has relied on morphological, biological, and phylogenetic species concepts (Leslie et al., 2001). Although morphological characteristics remain important for species identification, they are often insufficient due to high variability and overlap of traits among species (Crous et al., 2021; Geiser et al., 2013; Han et al., 2023; Sempere and Santamarina, 2009). Consequently, multi-locus sequence analysis has become the predominant method for species delimitation within the genus Fusarium (Crous et al., 2021; Geiser et al., 2021; Han et al., 2023). Among the genetic markers employed, the translation elongation factor 1-alpha (tef1), RNA polymerase largest subunit (rpb1), and RNA polymerase second largest subunit (rpb2) are the most informative for phylogenetic resolution (Crous et al., 2021; O’Donnell et al., 2022). Additional markers such as calmodulin (CaM), histone H3 (h3), and beta-tubulin (tub2) have also been used to resolve species boundaries within species complexes (Han et al., 2023; Zhang et al., 2024). A combined multi-locus dataset including tef1, rpb1, rpb2, tub2, and CaM has been applied to clarify generic boundaries and support the current species delimitations within the FFSC (Han et al., 2023; Yilmaz et al., 2021; Zhang et al., 2025).
In Republic of Korea (Korea), several FFSC species such as F. andiyazi, F. circinatum, F. concentricum, F. fujikuroi s. str., F. lactis, F. proliferatum, F. subglutinans, F. thapsinum, F. udum, and F. verticillioides have been reported, primarily based on analyses of tef1 and rpb2 sequences (An et al., 2016; Choi et al., 2019a, 2019b, 2019c, 2021, 2024, 2018b; Ha et al., 2023; Jeong et al., 2013; Kang et al., 2024; Kim et al., 2012; Koo et al., 2023; Lee et al., 2002; Yang et al., 2019). However, many identifications from earlier studies remain uncertain due to limited molecular data and ongoing taxonomic revisions. A notable example is F. proliferatum, which has been widely reported on diverse hosts in Korea. Its taxonomic status was revised in 2021 when F. proliferatum was distinguished from F. annulatum based on morphological and molecular evidence (Yilmaz et al., 2021). Previously, F. proliferatum was synonymized with F. annulatum based on sequence analyses of the large subunit ribosomal, small subunit ribosomal RNA, and tub2 gene regions, along with the ex-type strain of F. annulatum (CBS 258.54), and the representative strains (CBS 217.76, CBS 267.93, CBS 620.80, etc.) of F. proliferatum (O’Donnell et al., 1998a). Despite this, the authors proposed conserving the widely recognized name F. proliferatum over F. annulatum due to the prevalence of isolates identified as F. proliferatum (represented by CBS 217.76) and the fact that only a single isolate (CBS 258.54) of F. annulatum had been identified at that time. Later, Yilmaz et al. (2021) confirmed that F. proliferatum is a distinct species from F. annulatum, based on both morphological characteristics and molecular data derived from its designated epitype (CBS 480.96). Subsequently, numerous isolates initially identified as F. proliferatum were reclassified as F. annulatum through multi-locus sequence analyses (Han et al., 2023; Yilmaz et al., 2021).
A large number of FFSC isolates collected from Korea are preserved in the Korean Agricultural Culture Collection (KACC), including isolates reported in previous studies and others identified only by morphological traits by depositors. Accurate species identification within the FFSC is essential due to its relevance to pathogenicity and mycotoxin production. Therefore, this study aims to: (1) clarify the taxonomic status of FFSC isolates from Korea held in KACC and several recently described species, based on a multi-locus sequence analysis; (2) describe a novel species discovered during this work; and (3) update fungus-host associations within the FFSC in Korea.
Materials and Methods
Fungal isolates
Eighty-one isolates used in this study were obtained from the KACC collection. The cultures were retrieved on synthetic nutrient-poor agar (SNA) (Nirenberg, 1976) from liquid nitrogen preservation. Details of the isolates are listed in Table 1. The holotype (dried culture) of the new species is deposited in the herbarium (KB) of the National Institute of Biological Resources, Korea.
DNA extraction, polymerase chain reaction amplification, and sequencing
Mycelia from 5-day-old colonies grown on potato dextrose agar (PDA; Difco Laboratories, Detroit, MI, USA) were harvested, and genomic DNA was extracted using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol.
Partial sequences of five gene regions, including the tef1, tub2, CaM, rpb1, and rpb2, were amplified and sequenced in both directions using primer pairs listed in Table 2. When amplification of rpb1 in some isolates using published primer pairs (Fa/G2R, Fa/R8, F8/G2R, F7/G2R) (Hofstetter et al., 2007; O’Donnell et al., 2010) failed, a reverse primer, G2Rm (5′-TCATCTGRGTRGCAGGCTCAC-3′), was designed by modifying the G2R primer, and the pair F7/G2Rm was used to amplify and sequence the target fragment.
Polymerase chain reactions (PCR) were conducted in a 25 μL reaction volume containing 12.5 μL of 2× PCR Master Mix (MG Med, Seoul, Korea), 8.5 μL of nuclease-free water, 1 μL of each primer (4.5 pmol), and 2 μL of genomic DNA (100 ng/μL). Amplifications were performed using an AllInOneCycler Thermal Cycler (Bioneer, Daejeon, Korea), following the cycling conditions described in Table 2. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and verified by agarose gel electrophoresis. Purified amplicons were submitted to Macrogen Inc. (Seoul, Korea) for bidirectional Sanger sequencing.
Phylogenetic analysis
Raw sequence data obtained from both forward and reverse primers were trimmed and assembled using MEGA v. 11 (Tamura et al., 2021). All resulting sequences were submitted to the NCBI GenBank database. Sequences from novel species were compared to available sequences from GenBank (https://www.ncbi.nlm.nih.gov/) and FUSARIOID-ID database (https://www.fusarium.org/). Reference sequences representing 101 previously recognized species within the FFSC, along with F. graminearum (CBS 136009) as the outgroup, were retrieved from GenBank and used for phylogenetic analyses (Supplementary Table 1). Each locus was independently aligned using MAFFT v. 7 (Katoh and Standley, 2013) (http://mafft.cbrc.jp/alignment/server/index.html) with the G-INS-i option. The alignments were manually edited where necessary and then concatenated afterwards in MEGA v. 11. Phylogenetic trees based on both individual and concatenated datasets of the five loci were constructed using the maximum likelihood (ML) method in IQ-TREE Web Server (http://iqtree.cibiv.univie.ac.at/) (Trifinopoulos et al., 2016) with 1,000 ultrafast bootstrap replicates. Resulting ML phylogenetic trees were visualized using MEGA v. 11. Bayesian inference was conducted using BEAST2 v. 2.7.5 (Bouckaert et al., 2019). The analysis was run for 10,000,000 Markov Chain Monte Carlo generations, sampling every 1,000 generations. The initial 10% of sampled trees were discarded as burn-in. The final consensus tree was visualized using FigTree v. 1.4.4.
Morphology
Colonies were grown on PDA and oatmeal agar (OA; Difco) for 10 d at 25°C in darkness to assess growth rate and macromorphology; colours were recorded with Rayner (1970) charts. Micromorphology was examined from cultures grown on carnation leaf agar (CLA) (Fisher et al., 1982) incubated at 25°C with 12-h near-UV light and 12-h dark cycles for 7 to 14 days. Slides mounted in lactic acid were observed with a Zeiss Axio Imager A1 microscope equipped with differential interference contrast optics and an AxioCam ICc 3 camera (Carl Zeiss, Göttingen, Germany). Sporodochial structures and chlamydospores were monitored on PDA, OA, CLA, and SNA for up to 30 days. Descriptions follow terminology of Crous et al. (2021). For each structure, at least 30 measurements were recorded. Data are presented as mean ± standard deviation (SD), and summarized in the format (minimum−)mean-SD-mean+SD(−maximum), or alternatively as a range (minimum-maximum).
Rearrangement of species within the FFSC and their hosts in Korea
To investigate the host range and distribution of Fusarium species within the FFSC in Korea, host-fungus associations were compiled from multiple literature and database sources. Information on previously reported Fusarium species and their hosts in Korea was obtained from The List of Plant Diseases in Korea (6.2nd edition, 2024) available at https://genebank.rda.go.kr/english/plntDissInfo.do, and from the National List of Species of Korea (2024) available at https://kbr.go.kr. To assess the global host records of the identified Fusarium species from the present study, additional data were retrieved from the USDA Fungal Databases (https://fungi.ars.usda.gov/) and other publicly available taxonomic and phytopathological databases.
Results
Multi-locus phylogenetic analysis
A concatenated alignment of five loci (CaM, rpb1, rpb2, tef1, and tub2) was constructed, comprising 104 taxa and 5,091 characters including gaps. Of these, 3,342 characters were conserved, 1,696 were variable, and 1,073 were parsimony-informative. The gene boundaries were defined as follows: CaM (1-645), rpb1 (646-2,192), rpb2 (2,193-3,842), tef1 (3,843-4,537), and tub2 (4,538-5,091). ModelFinder, implemented in IQ-TREE, identified TIM2e+I+G4 as the best-fit nucleotide substitution model for the ML analysis under the Bayesian Information Criterion. Phylogenetic analysis based on this concatenated dataset yielded a five-locus ML tree (Fig. 1), and a separate Bayesian analysis (tree not shown) resulted in a congruent topology, with comparable nodal support.
The ML tree revealed that the 81 KACC isolates clustered into nine distinct clades, representing seven known species, a novel species namely F. ipomoeicola, and one potential new species. Among them, 45 isolates clustered with F. fujikuroi with strong support (ML bootstrap support value = 100%/Bayesian posterior probability = 1). One isolate (KACC 410672) was placed in the F. elaeagni clade (100/1). Fifteen isolates grouped within the F. annulatum clade (100/1). One isolate (KACC 49893) formed a distinct clade within the Asian clade, separate from all other species and sister to F. mangiferae; this isolate was considered a candidate for a new species. Eight isolates grouped with F. concentricum (100/1). Two isolates (KACC 47014 and KACC 47015) were assigned to F. thapsinum (100/1). Three isolates (KACC 46426, KACC 47168, and KACC 47318) grouped with F. planum (100/1), while three others (KACC 47732, KACC 47733, and KACC 47735) clustered with F. dendrobii (100/1). Finally, three isolates (KACC 47421, KACC 48680, and KACC 48681) of F. ipomoeicola formed a distinct lineage (100/1) within the American clade. This lineage did not cluster with any known species and formed a sister clade to F. amaranthi, F. pilosicola, and a clade comprising F. circinatum, F. parvisorum, and F. sororula. Phylogenetic analysis also demonstrated that F. hipposidericola, F. jacksoniae, F. xishuangbannaense, and F. oryzigenum are conspecific with F. annulatum, F. babinda, F. hechiense, and F. planum, respectively.
Taxonomy
Fusarium ipomoeicola L. D. Thao, J. W. Yang & S. B. Hong, sp. nov. (Fig. 2).
MycoBank: MB860539.
Etymology: The species epithet is derived from the host plant genus, Ipomoea.
Conidiophores on aerial mycelium, unbranched or irregularly branched, straight or flexuous, bearing terminal single phialides or whorls of 2-3 phialides, or reduced to conidiogenous cells. Aerial conidiogenous cells mono- and polyphialides, subulate to subcylindrical, smooth- and thin-walled, (8−)11.5-16.5(−18) × (2−)2.5-3 μm (av. 14 × 2.6 μm). Microconidia formed on aerial conidiophores, ellipsoidal, oval, reniform, straight or slightly curved, forming small false heads on tips of phialides, 0-1-septate, predominantly 0-septate, hyaline, smooth- and thin-walled, 0-septate conidia: (5−)7-11.5(−14) × (2−)2.5-3(−3.5) μm (av. 9.1 × 2.6 μm); 1-septate conidia: (13−)14-20.5(−24) × (2.5−)3-3.5(−4) μm (av. 17.2 × 3.2 μm). Mesoconidia formed on aerial conidiophores, straight to slightly curved, tapering toward the apical part, 1-3-septate, predominantly 3-septate, hyaline, smooth- and thin-walled, 1-septate conidia: (19−)20.5-25(−26.5) × 3-3.5(−4) μm (av. 22.7 × 3.3 μm); 2-septate conidia: (23.5−)24.5-28.5(−29.5) × 3-3.5(−4) μm (av. 26.5 × 3.4 μm); 3-septate conidia: (24.5−)27.5-36(−41) × 3-3.5(−4) μm (av. 31.8 × 3.4 μm). Sporodochia salmon (41), formed frequently on the surface of carnation leaves, and absent on SNA, PDA, and OA. Sporodochial conidiophores densely aggregated, irregularly branched, bearing terminal solitary monophialides or apical groups of 2-4 monophialides. Sporodochial phialides smooth-walled, subulate to subcylindrical, (11.5−)12.5-16.5(−18) × 2.5-3 (−3.5) μm (av. 14.5 × 2.9 μm). Sporodochial macroconidia slender, moderately curved, tapering towards both ends; apical cell moderately curved, slightly papillate; basal cell papillate non foot-shaped or poorly developed foot-shaped, 3-5-septate, predominantly 3-septate, hyaline, smooth- and thin-walled, 3-septate conidia: (24−)31.5-43(−48.5) × (3−)3.5-4 μm (av. 37.4 × 3.7 μm); 4-septate conidia: (38−)42.5-50(−54) × 3.5-4 μm (av. 46.3 × 3.6 μm); 5-septate conidia: (47.5−)49-57(−61.5) × 3.5-4 μm (av. 53.1 × 3.6 μm). Chlamydospores absent.
Culture characteristics: Colonies on PDA incubated at 25°C in the dark reaching 84-88 mm diameter in 10 days. Aerial mycelia dense, surface rosy vinaceous (58) in the centre, colony margin erose and white. Reverse pale vinaceous (85) in the centre, white at the margin. Colonies on OA incubated at 25°C in the dark reaching 4.5-5 mm diameter in 10 days. Aerial mycelia dense, surface pale vinaceous (85) in the centre, white at the margin. Reverse rosy buff (61).
Typus: Korea, Jeollanam-do, Haenam, from sweet potato (Ipomoea batatas) storage roots, 10 Apr 2015, J. W. Yang (holotype NIBRFG0000521257, ex-holotype culture KACC 48681 = SPL15044).
Additional materials examined: Korea, Jeollabuk-do, Gimje, from sweet potato (Ipomoea batatas) storage roots, 10 Apr 2015 J. W. Yang, culture KACC 48680 = SPL15022; Gyeonggi-do, Yongin, from air, 28 Nov 2011, D. H. Kim, culture KACC 47421.
Notes: Fusarium ipomoeicola was previously misidentified as F. circinatum, the causal agent of sweet potato storage root rot, based on morphological characteristics and a combined internal transcribed spacer (ITS) and tef1 sequence analysis (Yang et al., 2019). The macroconidia described by Yang et al. (2019) on PDA are similar to the mesoconidia observed on CLA in the present study. However, multi-locus phylogenetic analysis based on five loci (CaM, rpb1, rpb2, tef1, and tub2) placed F. ipomoeicola in a distinct and well-supported clade within the FFSC, sister to F. amaranthi, F. pilosicola, and a clade containing F. circinatum, F. parvisorum, and F. sororula. The ex-holotype strain KACC 48681 differs from the ex-type strain of F. amaranthi (NC20865) by 30 nucleotide positions in four loci (1/611 in CaM; 3/876 in rpb1; 25/1,752 in rpb2; and 1/552 in tub2); from F. pilosicola (NRRL 29124, ex-type) by 35 nucleotide positions in four loci (1/719 in CaM; 23/1,701 in rpb2; 8/649 in tef1; and 3/552 in tub2); from F. circinatum (CBS 405.97, ex-type) by 37 nucleotide positions in four loci (3/611 in CaM; 3/876 in rpb1; 23/1,701 in rpb2; and 8/649 in tef1); from F. parvisorum (CMW 25267, ex-type) by 43 nucleotide positions in four loci (4/575 in CaM; 21/1,715 in rpb2; 17/644 in tef1; and 1/539 in tub2); and from F. sororula (CBS 137242, ex-type) by 40 nucleotide positions across five loci (4/612 in CaM; 4/862 in rpb1; 21/1,734 in rpb2; 8/617 in tef1; and 3/539 in tub2).
In addition to molecular distinctions, F. ipomoeicola is also morphologically distinct from its closest relatives. The reverse side of F. amaranthi colonies exhibits a characteristic concentric ring pattern, which was not observed in F. ipomoeicola. Fusarium amaranthi forms sporodochia on PDA and abundantly on SNA, whereas sporodochia in F. ipomoeicola were produced only on CLA and were absent on PDA, SNA, and OA. Chlamydospores were not observed in F. ipomoeicola on any of the tested media (PDA, OA, SNA, CLA), whereas those of F. amaranthi were present on PDA. The sporodochial macroconidia of F. ipomoeicola are also smaller than those of F. amaranthi and F. pilosicola (Hassan et al., 2024; Yilmaz et al., 2021). The formation of coiled sterile hyphae, a typical feature of F. circinatum on SNA and CLA (Nirenberg and O’Donnell, 1998), was not observed in F. ipomoeicola. Additionally, F. ipomoeicola has been identified as a pathogen of sweet potato, a host on which no related species have been reported.
Fusarium annulatum Bugnic., Rev. Gén. Bot. 59: 17 (1952).
Basionym: Fusarium annulatum Bugnic., Rev. Gén. Bot. 59: 17 (1952).
Obligate synonym: Fusarium moniliforme var. annulatum (Bugnic.) F. J. Chen, Variation within Fusarium section Moniliforme (= Liseola) [Ph.D. thesis]: 148 (1991).
New synonym: Fusarium hipposidericola Karun., Tibpromma & X. F. Liu, Mycosphere 14: 554 (2023).
Notes: Fusarium annulatum was originally described by Bugnicourt (1952). It was later considered conspecific with F. proliferatum based on multi-locus sequence analysis (O’Donnell et al., 1998a). However, a study by Yilmaz et al. (2021), using type material of both species, confirmed that F. annulatum and F. proliferatum are distinct taxa. Recently, F. hipposidericola was described from bats (Rhinolophus spp.) in China (Liu et al., 2023). Notably, the rpb2 sequence of F. hipposidericola, including that of the ex-type strain KUMCC 21-0724, does not cluster within the genus Fusarium, and was therefore excluded from our phylogenetic analyses. In contrast, other loci (ITS, CaM, tef1, and tub2) are fully consistent with those of F. annulatum, and the morphological characteristics of F. hipposidericola are indistinguishable from those of F. annulatum. These findings support the synonymization of F. hipposidericola with F. annulatum.
Fusarium babinda Summerell, C. A. Rugg & L. W. Burgess, Mycol. Res. 99: 1345 (1995).
Basionym: Fusarium babinda Summerell, C. A. Rugg & L. W. Burgess, Mycol. Res. 99: 1345 (1995).
New synonym: Fusarium jacksoniae Y. P. Tan & Conroy, Index Austral. Fungi 45: 1 (2024).
Notes: Fusarium babinda was first described from plant debris in soil in Australia (Summerell et al., 1995). More recently, F. jacksoniae was introduced by Tan and Conroy (2024), also from Australia, based solely on the tef1 gene. However, no sequence comparison was made with F. babinda. In our study, the tef1 sequence (PQ393365) of the ex-type strain of F. jacksoniae (BRIP 76473a) was found to be 100% identical to that of the ex-type strain of F. babinda (NRRL 25807), indicating that these names refer to the same taxon.
Fusarium hechiense M. M. Wang & L. Cai, Persoonia 48: 36 (2022).
Basionym: Fusarium hechiense M. M. Wang & L. Cai, Persoonia 48: 36 (2022).
New synonym: Fusarium xishuangbannaense Karun., Tibpromma & X. F. Liu, Mycosphere 14: 561 (2023).
Notes: Fusarium hechiense was described by Wang et al. (2022). Subsequently, F. xishuangbannaense was introduced by Liu et al. (2023), without morphological or molecular comparison to F. hechiense. In our study, multi-locus phylogenetic analysis (Fig. 1) showed that the two species are phylogenetically indistinguishable, forming a strongly supported clade that is sister to F. annulatum. Additionally, the morphological characteristics of F. xishuangbannaense are indistinguishable from those of F. hechiense. Based on these results, we consider F. xishuangbannaense to be a synonym of F. hechiense.
Fusarium planum S. L. Han, M. M. Wang & L. Cai, Stud. Mycol. 104: 124 (2023).
Basionym: Fusarium planum S. L. Han, M. M. Wang & L. Cai, Stud. Mycol. 104: 124 (2023).
New synonym: Fusarium oryzigenum Absalan, Lumyong & K. D. Hyde, J. Fungi 11: 17 (2025).
Notes: Fusarium planum was described in 2023 as the causal agent of maize stalk rot in China (Han et al., 2023). In 2025, F. oryzigenum was described from rice (Oryza sativa) in Thailand, based on a single strain (MFLUCC 24-0637, ex-type) (Absalan et al., 2025). However, no comparison was made with F. planum regarding morphology or sequence data. Our multi-locus phylogenetic analysis (Fig. 1) revealed that both species, including their ex-type strains, cluster within the same well-supported clade, indicating that they represent the same species.
Rearrangement of species within the FFSC and their hosts in Korea
A total of 16 Fusarium species within the FFSC have been identified in Korea, based on findings from this study and previous literature. Among them, four known species namely F. annulatum, F. dendrobii, F. elaeagni, and F. planum are reported for the first time in Korea. These 16 species were associated with 84 plant species, including 22 new plant records in Korea and 15 new plant records worldwide, as documented in this study (Table 3), although their pathogenicity was not confirmed.
New fungus-host associations recorded in Korea: F. annulatum on Allium cepa (onion), Anemarrhena asphodeloides (zhi mu), Cymbidium sp. (boat orchid), Glycine max (soybean), Oryza sativa, Peucedanum japonicum (coastal hog fennel), Sorghum bicolor (sorghum), Zea mays (maize); F. concentricum on Cymbidium sp., Lilium sp. (lily), Momordica charantia (bitter melon), Senna tora (sickle senna); F. dendrobii on Cymbidium sp.; F. elaeagni on Oryza sativa; F. fujikuroi on bamboo (unknown species), Chenopodium quinoa (quinoa), Helianthus annuus (sunflower), Pinus koraiensis (Korean pine); F. ipomoeicola on Ipomoea batatas; F. planum on Glycine max, Momordica charantia; F. thapsinum on Carthamus tinctorius (safflower).
New fungus-host associations recorded worldwide: F. annulatum on Anemarrhena asphodeloides, Cymbidium sp., Peucedanum japonicum; F. concentricum on Cymbidium sp., Momordica charantia, Senna tora; F. dendrobii on Cymbidium sp.; F. fujikuroi on bamboo, Chenopodium quinoa, Helianthus annuus, Pinus koraiensis; F. ipomoeicola on Ipomoea batatas; F. planum on Glycine max, Momordica charantia; F. thapsinum on Carthamus tinctorius.
Discussion
The present study employed multi-locus phylogenetic analyses to clarify the taxonomic identities of 81 Korean Fusarium isolates within the FFSC. Among these isolates, the species names of 35 isolates were revised from those originally assigned by the depositors, highlighting the importance of molecular approaches for accurate species delimitation. In addition, considerable host diversity was observed, revealing previously undocumented associations and underscoring the ecological variability of FFSC members in the region. These findings provide a foundation for discussing species distribution, host relationships, and their taxonomic implications.
In this study, four recently described Fusarium species were identified as synonyms of previously accepted taxa based on molecular data. One important issue is that these newly described species were published without comparison to closely related taxa, including species that had been introduced only shortly before. This lack of comparison led to redundant species names and taxonomic confusion. In addition, the rpb2 sequence from the ex-type strain of F. hipposidericola was found to be inaccurate, raising concerns about the reliability of sequence data used to support new species descriptions. These findings emphasize the importance of validating sequence data and conducting thorough comparisons before proposing new taxa, particularly within species-rich groups such as the FFSC.
Fusarium fujikuroi s. str. is widely recognized as a causative agent of rice bakanae disease and has a broad host range globally (Farr and Rossman, 2025; Han et al., 2023). It is the best-studied FFSC species in Korea, with numerous hosts documented (Table 3). Choi et al. (2018b) confirmed F. fujikuroi s. str. as the dominant species isolated from Korean cereals, particularly rice, and demonstrated its capacity to produce harmful mycotoxins such as fumonisin, fusarins, fusaric acid, apicidin F, and beauvericin (Choi et al., 2018b; Niehaus et al., 2017). Our study reports, for the first time globally, the occurrence of F. fujikuroi s. str. on Chenopodium quinoa (quinoa), bamboo (unknown species), Helianthus annuus (sunflower), and Pinus koraiensis (Korean pine). This finding broadens the known host range of the species and underscores its ecological adaptability to diverse plant taxa, suggesting a potential for cross-host infection and need to monitor its distribution across different cropping systems.
Following the taxonomic revision of F. annulatum and F. proliferatum in 2021, F. annulatum is currently recognized as a multi-host species exhibiting considerable morphological and phylogenetic variability (Han et al., 2023; Yilmaz et al., 2021). Although F. proliferatum has frequently been reported as a pathogen on various host plants in Korea, the previous identifications, including those in recent studies (Ha et al., 2023; Kang et al., 2024; Koo et al., 2023), have now become ambiguous due to taxonomic revisions. In this study, F. annulatum was detected on nine plant species, including major crops such as Oryza sativa, Zea mays, Sorghum bicolor, and Glycine max. Several isolates previously identified as F. proliferatum, for example: KACC 41085 = F6286-2 from Cymbidium ensifolium (spring orchid; formerly C. ginatum), KACC 45822 = V217 from Oryza sativa, KACC 45824 = C45 and KACC 48354 = 15CgGw05-fv3 from Zea mays, and KACC 48817 = S1119 from Sorghum bicolor (Choi et al., 2018b, 2019b; Kim et al., 2012; Lee et al., 2002), were re-identified here as F. annulatum. These findings revealed that F. annulatum has long been prevalent in Korea, whereas the presence of F. proliferatum remains uncertain. This study thus represents the first formal report of F. annulatum in Korea. The fungus has also been reported as a potent mycotoxin producer (Choi et al., 2018b). Hence, greater attention should be given to its monitoring and management.
Fusarium concentricum was first described from Musa sapientum (banana) in Costa Rica, and from Nilaparvata lugens (brown planthopper) in Korea (Nirenberg and O’Donnell, 1998). Subsequently, it has been reported as a highly aggressive pathogen affecting various crops and ornamentals, causing yield losses, particularly in tropical and subtropical regions. (Du et al., 2020; Farr and Rossman, 2025; Wang et al., 2013; Xiao et al., 2019). This species is prevalent in Korea, where it has been found on 14 host plants (Table 3). Given its wide host range and potential economic impact, further studies on the pathogenicity, epidemiology, and mycotoxin production of Korean isolates are warranted to better inform disease management strategies.
Fusarium dendrobii was recently described from Dendrobium sp. (dendrobium orchid) in China (Zhang et al., 2024). This species, causing leaf spot disease on Cymbidium sp. (boat orchid) in Korea, was earlier identified as F. subglutinans based on morphological characteristics and tef1 sequence analysis (Han et al., 2015). In the present study, the F. subglutinans isolates (KACC 47732, KACC 47733, and KACC 47735) studied by Han et al. (2015) were re-identified as F. dendrobii using a combination of tef1, rpb1, rpb2, CaM, and tub2 genes. Therefore, F. subglutinans on Cymbidium sp. has to be changed to F. dendrobii in The List of Plant Diseases in Korea.
Fusarium elaeagni was originally described from Elaeagnus pungens (thorny olive) in China (Wang et al., 2022) and later reported on rice in the same country (Han et al., 2023). This species was also found on rice in Korea in our study. Fusarium elaeagni is indistinguishable from F. fujikuroi s. str. at the tef1 gene sequence, but the two species can be differentiated using the rpb2 gene as indicated in individual phylogenetic trees in this study (not shown). Morphologically, the two species can be readily distinguished. Fusarium elaeagni exhibits greyish-orange sporodochia, whereas those of F. fujikuroi are orange. Macroconidia of F. elaeagni are predominantly 3-4-septate, in contrast to the 3-5-septate macroconidia of F. fujikuroi. Microconidia are ellipsoidal to falcate, occasionally club-shaped in F. elaeagni, while they are ovoid or club-shaped in F. fujikuroi. In addition, aerial phialides are either mono- or polyphialidic in F. elaeagni, but are typically polyphialidic in F. fujikuroi (Wang et al., 2022).
Fusarium planum, recently introduced as a pathogen causing maize stalk rot and found on rice and wheat in China (Han et al., 2023), was recorded in this study on rice straw and two novel hosts in Korea: Glycine max and Momordica charantia (bitter melon). This suggests F. planum may have a broad host range across different plant families. Both F. planum and F. elaeagni are recently introduced species associated with rice and other cereal crops, whose virulence and mycotoxin-producing capabilities remain poorly understood. Therefore, additional studies are needed to better characterize these traits.
Fusarium thapsinum is a well-known species associated with cereal grains worldwide, such as Oryza sativa, Sorghum bicolor, and Zea mays (Farr and Rossman, 2025). Although this fungal species was previously reported on Coix lacryma-jobi (job’s tears), Sorghum bicolor, and Zea mays in Korea, and was identified here on a new host plant, Carthamus tinctorius (safflower), its prevalence and toxigenic potential have been considered limited (An et al., 2016; Choi et al., 2018b, 2019b). The detection of F. thapsinum on Carthamus tinctorius adds to the growing list of non-cereal hosts for this species, highlighting its potential adaptability to a broader range of plants.
The isolate KACC 49893, obtained from Musa × paradisiaca (banana), was identified as a potential new species based on a five-locus analysis and found to be closely related to F. lumajangense and F. mangiferae. Consequently, additional morphological and molecular data from other isolates within this taxon are needed to increase the reliability of this finding and support the formal description of the new species.
The only record of F. circinatum causing disease on sweet potato to date is that reported by Yang et al. (2019). However, the isolates (KACC 48681 = SPL15044 and KACC 48680 = SPL15022) examined in that study as F. circinatum were re-identified in the present work as representing a distinct, newly described species, F. ipomoeicola, based on multi-gene phylogenetic analyses and morphological characteristics. This finding confirms that F. circinatum has, in fact, not been recorded from sweet potato anywhere in the world. Given that sweet potato is one of the most important root crops widely cultivated in Korea, further studies are needed to assess the incidence and distribution of F. ipomoeicola in the country, as well as to elucidate its biological characteristics, including toxin production potential and host range.
Overall, this study provides a comprehensive taxonomic reassessment of Fusarium isolates within the FFSC from Korea, significantly expanding the known species diversity and host range of these agriculturally important fungi. Our findings enhance understanding of FFSC taxonomy and phylogeny both locally and globally and offer valuable insights into the ecological breadth and host associations of these fungi. While laying a strong foundation for future mycological and phytopathological research in Korea, further studies are necessary to elucidate the pathogenicity and mycotoxin profiles of the identified species. Such research will be crucial for developing effective disease management strategies and deepening knowledge of the ecological functions and risks associated with FFSC members in agricultural and natural ecosystems.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Supplement
Print
