Genetic Diversity, Pathogenicity and Host Resistance of Fusarium Species from Soybean in Canada

1. Introduction

Soybean (Glycine max L.) is one of the most important legume crops worldwide, known as one of the few ‘complete protein’ sources with all nine human-essential amino acids. Cultivation is mainly centered in North and South America, comprising 80% of total world production [1,2,3]. In Canada, soybean production has increased over the past decades, and it has become a crucial crop for human and livestock consumption as well as global exportation. Starting in Ontario and expanding to the Prairie provinces, soybean ranks behind only wheat, canola and barley in acreage [4,5]. In 2022, 1.6 million hectares of soybeans were grown in eastern Canada (mainly in Ontario and Quebec), yielding 5.1 million tonnes, with another 0.5 million hectares (1.4 million tonnes) grown in western Canada (primarily Manitoba and Saskatchewan). The value of the crop in Canada was estimated at more than $33 billion USD [5,6].

In spite of this rapid expansion, several biotic constraints continue to be a major limitation to soybean production. To date, over 300 diseases have been reported on soybean, causing yield losses averaging 11% [7]. A soybean survey conducted in the USA and Ontario found 23 specific diseases from 2015 to 2019, resulting in yield losses of 6-11% [8]. Addressing these biotic constraints is essential for enhancing soybean production, ensuring food security and supporting farmers worldwide.

Fusarium spp. are significant soybean pathogens, causing root rot, wilt, sudden death syndrome (SDS), seed decay and seedling blight [9,10,11]. In some cases, soybean diseases are caused by a single Fusarium species, for instance, Fusarium wilt caused by F. oxysporum Schltdl [12,13,14]. Others involve multiple species, constituting a pathogen ‘complex’. The soybean root rot complex, widespread in North America in recent years, results from infection by multiple Fusarium spp. [9,10,15,16,17]. These species include F. solani, F. oxysporum, F. acuminatum, F. avenaceum, F. cerealis, F. culmorum, F. equiseti, F. graminearum, F. poae, F. proliferatum, F. pseudograminearum, F. redolens, F. sporotrichioides, F. fujikuroi, F. incarnatum-equiseti, F. tricinctum, F. semitectum, F. armeniacum, F. commune and F. verticillioides [9,10,15,16,18,19,20,21,22,23]. While F. solani and F. oxysporum were initially considered as the dominant species in the soybean root rot complex, the composition of Fusarium spp. and their prevalence varied across different geographic regions [9,24,25]. Zhao et al. [19] reported that F. proliferatum is the most virulent species on soybean in Hubei, China. In contrast, in Sichuan province, the most aggressive species were F. oxysporum, F. equiseti and F. graminearum [26]. At the reproductive stage of the crop, F. acuminatum, F. graminearum, and F. solani were more prevalent than other Fusarium spp. in Iowa [10]. Zhang et al. [27] reported eight Fusarium spp. were associated with soybean root rot in Ontario, with the most aggressive species identified as F. graminearum. In western Canada, F. oxysporum and F. acuminatum were the dominant species in surveys of soybean in Manitoba and Alberta, respectively [9,14].

Generally, Fusarium spp. are classified based on various morphological characters, including colony features and pigmentation on different media, as well as the appearance, size and presence of the three spore types: microconidia, macroconidia, and chlamydospores [28]. However, the morphology, ecology, physiology and even genetic traits of Fusarium spp. often exhibit variations across different studies [14,28,29,30]. Currently, the most common and effective approach for identifying Fusarium spp. in soybean relies on DNA Sanger sequencing of PCR amplification products obtained with primers for specific genes or genomic regions. This method offers rapid, precise, sensitive, and convenient results [9,31]. The internal transcribed spacer (ITS) region of nuclear-encoded ribosomal DNA (ITS rDNA) is commonly used to generate primers to identify plant pathogens [32,33]. Both the ITS1/ITS4 and ITS5/ITS4 primer sets are popular for the identification of Fusarium spp. in various crops [14,19,34,35,36]. Additionally, numerous other primers targeting genes within the Fusarium genus have been widely applied in DNA sequence analysis. These alternate targets include the translation elongation factor 1 alpha (TEF1a) genes, mating type locus genes, phosphate permease gene, beta-tubulin gene, as well as the largest and second largest subunit nucleotide sequences of RNA polymerase II, RPB1 and RPB2, respectively [9,37,38]. Furthermore, the most up to date Fusarium database, FUSARIUM-ID v.3.0, primarily identifies Fusarium spp. based on TEF1a sequences, complemented with some other loci, including RPB1, RPB2 and ITS [39]. At present, molecular methods with multiple primers, combined with evaluation of morphological characters, are necessary to differentiate Fusarium spp. and analyze their phylogenetic relationships [19,26,40,41,42].

Fusarium spp. have a wide host range, infecting various crops such as cereals, soybean, other legume crops, canola and corn [43,44,45,46]. Notably, F. graminearum, known for causing severe Fusarium head blight (FHB) in cereal crops, also exhibits high aggressiveness towards soybean [9,47,48]. Cross-pathogenicity among different crops can limit the efficacy of crop rotation in controlling Fusarium diseases [9,44]. As such, seed treatments are widely applied in North America to enhance seedling emergence and provide protection against soilborne pathogens of soybean [49,50,51]. Studies of fungicides and biocontrol agents as soybean seed treatments have also reported potential efficacy against single species of Fusarium under controlled conditions. The effectiveness of these treatments, however, remains limited in field trials conducted in natural environments [25,52,53].

The most promising strategy to manage soybean diseases caused by Fusarium spp. lies in the selection and breeding of resistant cultivars. However, globally available commercial cultivars with complete resistance are yet to be developed [27,54,55]. Incomplete resistance, controlled by polygenetic loci [56,57,58,59], is influenced by the interaction between the cultivar and the environment [60,61,62]. Cultivar screening and detection of resistance to SDS caused by F. virguliforme in soybean has been reported in the United States [58]. Similarly, tolerance or incomplete resistance to virulent isolates of Fusarium also has been identified in Alberta and Ontario [27,52,63]. However, resistance to Fusarium spp. has not been evaluated in Manitoba or other regions of Canada.

To improve management of Fusarium diseases in Canadian soybean cultivation, it is essential to understand the species diversity, distribution, and pathogenicity of Fusarium spp., as well as to assess soybean cultivar resistance in the major production areas. This study aims has three primary objectives: 1) to evaluate the pathogenicity of six common Fusarium spp. on soybean under controlled conditions; 2) to assess resistance/tolerance to these six common Fusarium spp. in a selection of 20 soybean cultivars; and 3) to investigate the genetic diversity and distribution of Fusarium spp. across the major soybean production regions of Canada.

2. Materials and Methods

2.1. Fungal Material and Inoculum Preparation

Twelve isolates representing the six most common Fusarium species (two isolates per species) in Manitoba as reported by Kim et al. [64], consisting of F. oxysporum, F. redolens, F. graminearum, F. solani, F. avenaceum and F. acuminatum, were obtained from the culture collection of the Agriculture and Agri-Food Canada, Brandon Research and Development Centre, Brandon, Manitoba (Table 1). Grain inoculum of each isolate was prepared according to a procedure modified from Chang et al. [15]. Briefly, each of the 12 isolates was cultured on Difco™ potato-dextrose agar (PDA) (Becton, Dickinson and Company, Sparks, MD USA) for one week at 25 °C in the dark. Approximately 10 L of wheat grain was soaked in distilled water overnight, and 600 mL of the soaked grain was transferred to an autoclave bag and autoclaved at 121°C for 90 min. After cooling, the grain was inoculated with 10 pieces of 5-day-old mycelial plugs excised from the PDA colonies of each fungus, and incubated at 25 °C in the dark for 2 weeks. The infested wheat grain was dried and ground into a powder (particle sizes between 0.25 and 1 mm) for use as inoculum.

2.2. Pathogenicity Test

Two soybean cultivars, ‘Akras’ and ‘B150Y1′ were selected to evaluate the pathogenicity of the 12 fungal isolates. Plastic cups (473 mL volume) were filled with soil mix (Sun Gro Horticulture Canada Ltd., Seba Beach, AB), and a layer of grain inoculum powder (10 mL) was applied to each cup and covered with 1 cm of soil mix. Seven seeds of each soybean cultivar were sown into each cup and covered with an additional 1 cm of soil mix. The cups were then placed in a greenhouse maintained at 25 °C with natural light supplemented by artificial lighting (light/dark cycle of 18 h/6 h). Non-inoculated controls of each soybean cultivar received wheat grain powder that had not been inoculated with any of the Fusarium spp. The pathogenicity test was arranged in a randomized complete block design (RCBD) with five replicates and repeated twice.

2.3. Evaluation of Resistance

The most virulent isolate of each Fusarium spp. was selected based on the results of the pathogenicity test and applied to screen a germplasm collection of 20 soybean commercial cultivars, including ‘HS11Ry07′, ‘S15B4′, ‘P15T46R2′, ‘B150Y1′, ‘Williams’, ‘Akras’, ‘TH32004R2Y’, ‘AC Proteus’, ‘AAC Edward’, ‘AAC Springfield’, ‘OAC Prudence’, ‘Misty’, ‘NSC Reston’, ‘OAC Ayton’, ‘Bloomfield’, ‘AC Harmony’, ‘OAC Petrel’, ‘Mandor’, ‘OT15-02′, and ‘NSC Dauphin’. Seeds were provided by the University of Alberta, and Agriculture and Agri-Food Canada (Harrow Research and Development Centre, Morden Research and Development Centre, and Brandon Research and Development Centre). The same seeding and inoculation methods were used as described above for the pathogenicity test. All 20 cultivars were inoculated with each of the six Fusarium spp. Non-inoculated controls, prepared as above, were also included for each cultivar. The experiment was arranged in a RCBD with five replicates and repeated twice.

2.4. Greenhouse Data Collection

Emergence counts were conducted on the 7^th, 14^th and 21^st day after seeding. The plant height from stem to the top leaf was measured on the 14^th day. Root rot severity was evaluated on a 0–4 scale as previously described by Chang et al. [15] and Zhou et al. [14] with minor modifications, where: 0, no symptoms; 1, taproot slightly discolored and fibrous roots healthy, aboveground growth appears normal; 2, taproot mostly discolored and/or fibrous roots with visible lesions, aboveground growth slightly reduced; 3, taproot completely discolored and fibrous roots with obvious lesions, aboveground growth severely reduced; and 4, taproot dark, degraded, few and reddish-brown fibrous roots, plants dead or dying. The 0-4 disease scale was applied to rate symptoms caused by each of the six Fusarium spp. evaluated (Figure 1). The 21-day-old seedlings from each replicate were collected after disease rating; their shoots and roots were separated, dried at 35 °C for 48 h, and weighed.

2.5. Fungal Isolation from Field Samples

Soybean root samples were collected from 106 fields in Manitoba (55 fields), Saskatchewan (18 fields), Ontario (30 fields) and Quebec (3 fields) in the 2022 growing season. Fifteen roots with symptoms of root rot were selected from each field for pathogen isolation. Two symptomatic root pieces (3-5 mm long) were cut from the root tip and crown of each root, surface-sterilized in 1% NaOCl for 60 s, and then rinsed three times with sterilized water. The surface-sterilized root pieces were transferred to PDA containing 0.2 mg/mL streptomycin and incubated for 7 days at 24 °C. The cultures were examined and any colonies of Fusarium were transferred to water agar and incubated for 5 days, at which time a single hyphal tip was cut and transferred to PDA in Petri dishes for purification. The purified cultures were grouped and sub-grouped based on their morphological characteristics, including color, mycelium type, released pigment on both sides of the Petri dishes [28].

2.6. DNA Extraction, PCR Amplification and Sanger Sequencing

The purified cultures were grouped based on their morphology on PDA and one isolate was randomly selected from each group for molecular identification. A total of 336 isolates from 89 fields and six reference isolates of Fusarium spp. were used to extract genomic DNA using a modified CTAB method following O’Donnell et al. [65]. Approximately 100 mg of the mycelium was scraped from the surface of a 10-day-old colony grown on PDA and transferred to a 1.5 ml microcentrifuge tube. The samples were flash-frozen in liquid nitrogen and mechanically homogenized using a TissueLyser II (Qiagen, Hilden, Germany) at 25 Hz with 3 mm beads for 1 min. Seven-hundred µl of CTAB buffer (Teknova, Hollister, CA, United States) was added to each 1.5 mL microcentrifuge tube and incubated at 65°C for 30 min. A 500 µl aliquot of phenol:chloroform:isoamylalcohol (25:24:1) (Invitrogen, Waltham, MA, United States) was added to each extraction tube and centrifuged at 1,107 g (3,000 rpm) for 30 min. The aqueous phase was transferred to a new microcentrifuge tube and mixed with an equal volume of isopropanol, incubated at -20 °C for 1 h and then centrifuged for 30 min at 1,107 g (3,000 rpm). The supernatant was decanted and the DNA pellet was washed twice with 70% ethanol and allowed to dry overnight at room temperature. The DNA pellet was dissolved in 40 µl TE buffer (10 mM Tris-HCl, 1 mM EDTA), the DNA concentration was estimated using a Nanodrop ND2000 (Thermo Fisher Scientific, Wilmington, USA), and diluted to a final concentration of 50 ng DNA/µl in distilled water.

The primer sets ITS4/5 (5′GGAAGTAAAAGTCGTAACAAGG 3′/5′TCCTCCGCTTATTGATATGC 3′; ITS region)[33] and T12 (5′AACATGCGTGAGATTGTAAGT 3′/5′TAGTGACCCTTGGCCCAGTTG 3′; beta-tubulin gene) [35,65] were used for PCR amplification and sequencing of each DNA sample. Amplifications were conducted in a 50 µl reaction volume, which included 1 ng/μl DNA template (5 µl), 10× Go Taq buffer (10 µl), MgCl₂ buffer (5 µl), sterile distilled water (26.75 µl), 2 mM dNTPs (1 µl), 1 μl each of forward and reverse primers (2.5 μM), and 0.25 μl of DNA Taq polymerase (Titanium Taq, Promega, Madison, USA). The PCR conditions were set as follows: initial denaturation at 95°C for 120 s; followed by 40 cycles at 95°C for 45 s, annealing for 45 s, and extension at 72°C for 1 min; and a final extension at 72°C for 10 min followed by cooling at 4 °C until recovery of the samples. The annealing temperatures were 55 °C and 52 °C for the ITS4/5 and T12 beta-tubulin genes, respectively. A 2 µl aliquot from each PCR was examined for presence/absence of amplicons by electrophoresis in 2% agarose. The amplicons in the remaining 48 µl volume were purified using a QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sent for Sanger sequencing at the Molecular Biology Services Unit (MBSU) at the University of Alberta in Edmonton, AB. The sequences obtained were used in BLAST searches of the National Center for Biotechnology Information (NCBI) databases (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The BLAST comparison was conducted in Standard databases (nr etc.) with organism of Nucleotide collection (nr/nt) using highly similar sequences (megablast). The BLAST results were sorted by percent identity of which the first result was used to identify the species of DNA sample.

2.7. Phylogenetic Analysis

All 336 sequences from the isolates obtained from the soybean root samples collected in 2022 and the reference sequences from the six Fusarium spp. were edited using Bio-Edit software v. 7.2.5 [66] with manual adjustment. The six reference sequences that best matched F. oxysporum, F. redolens, F. graminearum, F. solani, F. avenaceum and F. acuminatum were downloaded from GenBank. All of the sequences identified as Fusarium spp. and the six reference sequences were included in a phylogenetic analysis by the maximum likelihood (ML) method using the Jukes Cantor model with default parameters in the CLC main Workbench v.23.0.3 (QIAGEN, Aarhus, Denmark). Bootstrap values (BV) (%) were calculated with 1,000 replicates, and a phylogenetic tree was constructed with BV>70.

2.8. Data Analysis

Analysis of variance (ANOVA) was conducted for all traits analyzed in the pathogenicity test and cultivar resistance evaluation trials in greenhouse using R v. 4.2.0 [67]. The average mean of plants in each cup was calculated for the three seedling germination counts named “count1”, “count2”, “count3”, plant height (cm) named “height”, root rot severity on scale of 0-4 named “DS1”, as well as dry shoot and root weight (g) named “shoot” and “root”, respectively. An additional estimated mean for disease severity that considered the non-germinated seeds as dead plants with a score of 4 was calculated as “DS2”. A least significant difference (LSD) at p< 0.05 was applied to compare the estimated means of all the traits among different treatments using the R package “agricolae”. The reduction percentage was calculated for cluster analysis using R v. 4.2.0 [67]. To demonstrate the level of high tolerance, moderate tolerance, moderate susceptibility and high susceptibility, the cluster group number was set as four. A principal components analysis (PCA) was carried out estimating the reduction percentage of the three germination counts, height, shoot weight and root weight, as well as the increase of two types of disease severity (DS1 and DS2) in R using the package “ggbiplot”. For the Fusarium spp. isolation, the incidence of each species was calculated using the formula: Incidence % = (n/N) ×100, of which n is number of fields where the species was detected in each province and N is the total field number in each province.

3. Results

3.1. Pathogenicity Test

Symptoms caused by the six Fusarium spp. varied, encompassing rotting, girdling and the development of brown sunken lesions (Figure 1). Analysis of variance revealed significant interactions between Fusarium spp. and the soybean cultivars for all the traits, while the repeat effect of two greenhouse experiments was not significant (Table 1). Consequently, data from the two repeated experiments were combined for all traits. Mean and LSD estimates for the six Fusarium spp. were calculated using ‘Akras’ and ‘B150Y1’ for all traits. Fusarium avenaceum isolate 1 exhibited the most adverse effects, resulting in a three-fold reduction in germination counts on both ‘Akras’ and ‘B150Y1′ (Figure 2a).

All Fusarium spp. infections significantly reduced plant height except for F. oxysporum on both cultivars and F. redolens on ‘Akars’ (Figure 2b). For the parameter DS2 (root rot severity), all 12 isolates caused significant severity compared with the non-inoculated control for both ‘Akras’ and ‘B150Y1′, with scores ranging from 1.5 to 3.3 on the 0-4 scale (Figure 2c). The dry shoot weight of infected plants were significantly lower than the non-inoculated controls for both cultivars following inoculation with each of the fungal isolates. A similar trend was observed for dry root weight, although the reductions caused by F. solani isolate 1 on ‘B150Y1,’ F. solani isolate 2 on ‘Akaras,’ F. avenaceum isolate 2 on both cultivars, and both isolates of F. redolens on ‘Akaras’ were not statistically significant relative to the control (Figure 2d).

In the pathogenicity study, the virulence of each isolate of each Fusarium spp. was reflected consistently across all traits. The most virulent isolates of each species were selected based on higher disease severity for the cultivar resistance evaluation trials, including F. oxysporum isolate 2, F. redolens isolate 2, F. solani 1, F. graminearum isolate 2, F. acuminatum isolate 2 and F. avenaceum isolate 1.

3.2. Cultivar Resistance Evaluation

Both greenhouse experiments yielded corresponding results for all variables, with a non-significant repeat effect (p>0.05) (Table 2). Cultivar effect, Fusarium spp. effect, and their interaction were all found to be significant for all of the traits (p<0.05) (Table 2). As such, the reaction of each of the 20 soybean cultivars was evaluated in response to each of the six Fusarium spp. Notably, significant differences among the soybean cultivars were observed for all traits in the non-inoculated control, with ranges of 3.9 - 7.0 for the three germination counts (Count1, Count2, Count3), 8.6 cm - 16.5 cm for plant height, 0.9 g - 1.5 g for dry shoot weight, and 0.2g - 0.4 g for dry root weight. As expected, however, the non-inoculated controls of all 20 cultivars remained completely healthy, with a score of 0 for both DS1 and DS2 (Supplementary Table 1). Consequently, the disease reactions for the six Fusarium spp. were evaluated using percentage reduction for Count1, Count2, Count3, Height, Shoot, and Root, as well as disease severity increase.

Fusarium oxysporum caused the greatest average reduction in germination (80%), while the maximum average height reduction was observed for F. avenaceum (64%). For shoot and root loss, F. oxysporum had the most pronounced effects, with reductions of 81% and 83%, respectively. Regarding disease severity, inoculation with Fusarium oxysporum and F. avenaceum caused greatest severity scores for DS1 (2.8) and DS2 (3.4), respectively. In contrast, F. solani had the lowest impact on all of the traits compared with the other five Fusarium spp.

Principal components analysis with four clusters revealed a positive correlation among variables and cultivar reactions following inoculation with F. graminearum, F. avenaceum, F. acuminatum, F. oxysporum, F. redolens and F. solani (Figure 3). Notably, four cultivars, ‘B150Y1′, ‘P15T46R2′, ‘Misty’ and ‘Mandor’, clustered in the high-tolerance group against F. graminearum. Against F. avenaceum, the most tolerant cultivars were ‘OAC Prudence’, ‘Misty’, ‘NSC Reston’, ‘OAC Ayton’ and ‘AC Harmony’. The cultivars ‘S15B4′, ‘OT15-2′ and ‘Mandor’ displayed the greatest tolerance to F. acuminatum, and ‘P15T46R2′, ‘B150Y1′, ‘Williams’, ‘TH32004R2Y’, ‘AAC Edward’, ‘OAC Prudence’ and ‘OT15-02′ were most tolerant to F. oxysporum. Additionally, the most tolerant cultivars against F. redolens were ‘P15T46R2′, ‘B150Y1′, ‘TH32004R2Y’, ‘AC Proteus’ and ‘OAC Petrel’. The cultivars, ‘HS11Ry07′, ‘S15B4′, ‘P15T46R2′, ‘B150Y1′, ‘Williams’, ‘TH32004R2Y’ and ‘OAC Prudence’ showed the greatest tolerance to F. solani. Overall, the soybean cultivars ‘P15T46R2′ and ‘B150Y1′ displayed suppression against F. graminearum, F. oxysporum, F. redolens and F. solani.

3.3. Fusarium Spp. Identification

A total of 983 purified isolates were obtained from the symptomatic root samples and separated into three morphological groups and nine subgroups (Figure 4). After filtering and grouping based on colony morphology, 336 isolates were selected for molecular identification. The primer sets ITS4/5 (ITS region) and T12 (beta-tubulin gene) produced single bands of ~500 bp [36] and ~580 bp [35], respectively, from the six reference isolates (Figure 5) and 336 isolates collected in this study. Following removal of non-Fusarium species and isolates with poor sequence quality, 221 isolates were confirmed as Fusarium spp., primarily based on the ITS4/5-amplified gene sequences complemented with the T12-amplified sequences (Supplementary Table 2).

The distribution and prevalence of identified Fusarium spp. varied across Canada (Table 3). Based on the ITS sequences, the most prevalent species was F. avenaceum (26%), followed by F. oxysporum (19%), F. acuminatum (18%), F. redolens (9%) and F. equiseti (8%). All of the species identified in this study were grouped into four Fusarium spp. complexes except for the isolates classified as F. solani and F. redolens, which belonged to the F. tricinctum species complex (FTSC), F. oxysporum species complex (FOSC), F. incarnatum-equiseti species complex (FIESC), and F. sambucinum species complex (FSSC). The frequencies of FTSC, FOSC, FIESC and FSSC were 47%, 20%, 12% and 6% of all Fusarium spp. isolates recovered, respectively. Additionally, F. graminearum, and F. solani were exclusively isolated from samples collected in Saskatchewan and Ontario, respectively. Fusarium acuminatum and F. redolens appeared unique to the Manitoba and Saskatchewan. Notably, 91% of FIESC isolates were obtained from western Canada.

Moreover, 36% of the 221 isolates were identified as Fusarium spp. based on the beta-tubulin sequences. Among these, there were 56 isolates of F. acuminatum, four isolates of F. avenaceum, six isolates of F. equiseti, seven isolates of F. flagelliforme, two isolates of F. solani, one isolate each of F. compactum and F. sporotrichioides.

3.4. Phylogenetic Analysis

The ML phylogenetic analyses, based on the ITS4/5-amplified sequences, provided clear separation of the 221 isolates of Fusarium spp. into six distinct groups (Supplementary Figure 1). Group 1 consisted of 118 sequences belonging to FTSC [9,68,69], with 58, 39, 1, 2, 2 and 1 isolates identified as F. avenaceum, F. acuminatum, F. tricinctum, F. reticulatum, F. compactum and F. flocciferum. The remaining 14 sequences were unknown Fusarium spp. and reference sequences from the current study or GeneBank. Group 2 clustered three isolates of F. graminearum, two isolates of F. sporotrichioides, five isolates of F. cerealis and three isolates of F. culmorum into FSSC [9,70]. Group 3 comprised 17 isolates of F. equiseti, eight isolates of F. incarnatum and one of F. longipes, forming FIESC, with a strong bootstrap value (BV) of 91 [9,71]. Group 4 and Group 6 included 42 isolates of F. oxysporum, along with three F. verticillioides isolates and two F. oxysporum reference sequences, all known to be associated with FOSC [9,72]. Group 5 encompassed all three F. solani isolates and F. falciforme from Ontario, along with two reference sequences, with a BV of 100. Group 7 consisted of 19 isolates of F. redolens, one unknown fungal species, one reference isolate and one reference sequence from GeneBank, supported by a BV of 90. However, the grouping separation among FTSC, FSSC and FOSC-2 was not supported strongly (BV<70).

The sequences that were identified as species of FTSC and FIESC by amplification of the beta-tubulin gene were also analyzed for their phylogenetic relationships (Figure 6a and 6b, respectively). Clade 1 included four F. avenaceum isolates, one reference sequence and one reference isolate. Clade 2 comprised 56 isolates of F. acuminatum, one reference sequence and one reference isolate. For FIESC, Clades 3 and 4 consisted of seven isolates of F. flagelliforme and six isolates of F. equiseti, respectively. Additionally, the trees were rooted with one isolate of F. compactum (NSRR22_218) serving as an outgroup.

4. Discussion

Soybean production in western Canada, particularly Manitoba, has been increasing significantly in recent decades [4,5,6]. Root rot of soybean is a global concern involving numerous soilborne pathogens, with Fusarium spp. found to be predominant in disease surveys conducted in eastern [63] and western Canada [9,14,16]. However, studies investigating the pathogenicity of Fusarium spp. in Canada, especially in western Canada, are limited.

Abdelmagid et al. [16] evaluated the pathogenicity of five F. sporotrichioides isolates from Manitoba, which caused up to 70% root rot severity and significant reductions in root and shoot length in soybean. The cross-pathogenicity of five Fusarium spp., including isolates of F. cerealis, F. culmorum, F. graminearum and F. sporotrichioides collected in Manitoba, was also examined on soybean and wheat; the resulting in root rot severity on soybean ranged from 1.89 to 3.33 on a 0-4 scale [9]. In Alberta, F. proliferatum was reported as the most aggressive species on soybean based on greenhouse trials, while other tested Fusarium spp. caused mild to moderate levels of disease [14]. In this study, the pathogenicity of six Fusarium species was compared on soybean, using isolates previously collected from Manitoba. These species included F. oxysporum, F. graminearum, F. solani, F. acuminatum, which are predominant in North America, as well as F. avenaceum, which is also prevalent in Canada, and F. redolens, which has previously reported in the country [9,14]. The severity of root rot caused by these six species varied, with F. oxysporum, F. avenaceum and F. graminearum causing the most severe root rot on the soybean cultivar ‘Akras’. Despite this variability, root rot severity was generally greater to previous studies [14,52,53], suggesting that Fusarium spp. have become more aggressive in recent decades. This trend of increasing virulence should be emphasized to farmers when implementing control measures against Fusarium root rot in soybean.

Several studies have detected horizontal resistance in soybean controlled by polygenetic loci against Fusarium spp. Acharya et al.[57] identified one major and one minor QTL on soybean chromosomes 8 and 6 controlling partial resistance to F. graminearum. Quantitative resistance was also detected against SDS on all 20 soybean chromosomes [56,58]. In this study, 20 commercial soybean cultivars were screened for resistance to virulent isolates of F. oxysporum, F. graminearum, F. solani, F. acuminatum, F. avenaceum and F. redolens selected from the earlier pathogenicity test. The host reactions revealed varying degrees of tolerance. Fusarium oxysporum and F. avenaceum were identified as the most virulent species, while F. solani caused the lowest root rot severity, consistent with the findings of the pathogenicity testing. Complete resistance (root rot severity ≤ 1) was not observed on any of the 20 soybean cultivars evaluated. Moderate tolerance was detected in one, one, five, 15 and nine cultivars tested against F. acuminatum, F. avenaceum, F. graminearum, F. redolens and F. solani, respectively (1 ≤ DS1 ≤2). However, when considering non-germinated seeds as dead plants due to pathogen infection, moderate resistance was only observed against F. redolens and F. solani (1 ≤DS2 ≤2) (Supplementary Table 1). In this context, the level of tolerance was taken into consideration, evaluating all of the traits investigated in this study. The six Fusarium spp. tested also had varying effects on germination counts, plant height, dry shoot weight and dry root weight. In another study, 57 commercial soybean cultivars were evaluated against F. oxysporum, F. graminearum, F. avenaceum, and F. tricinctum, with resistance against all four species identified in the soybean cultivar ‘Maple Amber’, based on root rot severity, rather than emergence, plant height or dry root weight [63]. However, in that study, the root rot severity ranged from 0.5 to 2.7 on a 0-5 scale, and reductions in emergence, plant height and dry root weight were generally < 50%. Nyandoro et al. [52] evaluated the resistance of 12 soybean cultivars against F. avenaceum in greenhouse trials and found high root rot severity among cultivars (ranging from 2.6-3.4) and considerable reductions in emergence (26.7-75.5%). Given the variable performance of soybean cultivars across different trails, PCA was conducted to evaluate tolerance to Fusarium spp., taking into account root rot severity, germination counts, plant height, dry shoot weight and dry root weight. Some tolerant cultivars were identified against each Fusarium spp., with broad-spectrum resistance detected in ‘P15T46R2′ and ‘B150Y1′. The cultivars were tolerant/partially resistant to F. graminearum, F. oxysporum, F. redolens and F. solani.

Numerous studies have highlighted the variability of Fusarium spp. associated with the soybean root rot complex across different environments, locations and years [11,73,74,75]. In the current study, 18 Fusarium spp. were identified from a large collection of symptomatic roots using a combination of morphological and molecular methods (Supplementary Table 2). All the six species included in the pathogenicity test and cultivar evaluation trials were also recovered in the fungal isolation study, with F. avenaceum, F. oxysporum, F. acuminatum and F. redolens among the major groups. Another significant group was F. equiseti, which has been commonly reported in north California and South Dakota in the USA [76,77] as well as in Ontario, Alberta and Manitoba, Canada [9,14,27]. However, species such as F. poae, F. graminearum, F. solani, F. sporotrichioides, F. tricinctum, F. torulosum, F. commune and F. proliferatum , which were reported as predominant in Canada [9,14,27], were only sparsely identified or absent in the current study. Notably, isolates of F. acuminatum, F. solani, and F. redolens displayed location-specific characteristics, consistent with previous reports [9,14].

Colony morphology played a crucial role for the primary grouping and species selection of isolates within the same field, allowing us to narrow down the initial 983 isolates to 336 isolates for molecular identification. The isolates grouped into “White2” were identified as F. redolens except for “NSRR22_186” and “NSRR22_187, which were identified as F. solani. Similarly, all the “White4” isolates and a few “White1” isolates were identified as F. incarnatum-equiseti. Fusarium oxysporum had three morphology types, including “White1”, “Purple1” and “Slimy1”. The similarity in morphology between F. oxysporum and F. equiseti has also been reported in tomato in northeast India [78]. Additionally, three isolates in the “Purple1” subgroup were identified as F. verticillioides. Within the “Red” group, however, the four subgroups (“Red2,” “Red2,” “Red3,” and “Red4”) failed to distinguish whether the isolates belonged to FTSC or FSSC.

Furthermore, sequencing of DNA fragments amplified by PCR based on the ITS region and beta-tubulin gene identified 225 and 79 isolates that belonged to Fusarium spp., respectively. Inconsistencies were frequently observed among species identified within the same Fusarium species complex. For example, while five isolates were classified as F. incarnatum or F. equiseti with the primer set ITS4/5, they were recognized as F. flagelliforme (also a member of FIESC) with the T12 primer set (beta-tubulin gene). Notably, F. flagelliforme was not detected using ITS4/5 in our current study. Chang et al. [26] demonstrated that variable identification of isolates across species complexes or among species within the Fusarium genus was common in studies with multiple primers, including those targeting the beta-tubulin and ITS regions. For instance, ITS sequences have been reported as unsuitable for distinguishing F. equiseti or F. incarnatum [34] . Similarly, the beta-tubulin sequence has been applied to identify species in the FIESC and F. chlamydosporum Species Complexes (FCSC)[79], but failed to distinguish F. armeniacum, F. acuminatum, F. sportrichioides and F. langsethiae [80]. While the TEF1a gene has been widely used to define species and reveal phylogenetic relationships within the genus Fusarium in the past, multiple primers are more frequently employed in recent studies [9,19,41,81,82]. The results of this study underscore the need to use a combination of colony morphology and multiple primers to identify Fusarium spp. accurately.

Phylogenetic analysis successfully grouped most of the Fusarium spp. identified in this study into distinct complexes, such as FTSC, FOSC, FIESC, FSSC, F. redolens and F. solani. However, only FIESC, F. redolens and F. solani had strong bootstrapping support (BV≥70) and displayed location-specific identification. Within groups like FTSC, FOSC-2, and FSSC, the bootstrap values varied, indicating significant genetic variation. For F. oxysporum, the isolates in the morphological subgroups “Slimy1”, “White1” and “Purple1” were not well distinguished by ITS4/5. Meanwhile, isolate overlap was observed in FTSC between F. avenaceum and F. acuminatum, as well as in FIESC between F. equiseti and F. incarnatum. On the other hand, beta-tubulin sequences clearly distinguished F. avenaceum and F. acuminatum in FTSC as well as F. equiseti and F. flagelliforme. O’Donnell et al. [82] concluded that the beta-tubulin gene is not universally informative within Fusarium and is only suitable to distinguish Fusarium spp. forming part of the F. solani and F. incarnatum-equiseti species complexes. This study was the first to use the T12 primer set specific for the beta-tubulin gene to separate Fusarium spp. in FTSC and FIESC. Hafez et al. [9] also generated a phylogenetic tree that clearly distinguished Fusarium spp. of FTSC, FSSC, FOSC, FIESC, F. redolens and F. solani collected from Carman and Melita, Manitoba. In a study investigating Fusarium isolates from central and southern Alberta, overlapping species were found in phylogenetic trees based on both the ITS and TEF1a sequences [14]. Fusarium oxysporum was reported to have extensive genetic variation [10] and morphological variability [83]. A whole genome sequence study demonstrated the large diversity in FOSC in Australia among the identified clades [75]. Overall, phylogenetic analysis based solely on ITS4/5 was insufficient to classify the different Fusarium species complexes or a single species, while beta-tubulin clearly distinguished Fusarium spp. in FTSC and FIESC.

5. Conclusions

Evaluation of the pathogenicity of six Fusarium spp. on soybean indicated that F. avenaceum and F. oxysporum were the most strongly virulent, while F. graminearum, F. acuminatum, and F. redolens also caused significant levels of disease. In contrast, F. solani was weakly virulent, causing mild symptoms of Fusarium root rot on soybean. An assessment of the reaction of a suite of 20 soybean cultivars to inoculation with each of the Fusarium spp., based on root rot severity, emergence, plant height, and dry shoot and root weight, indicated that while no hosts were completely resistant, some cultivars showed partial resistance or tolerance to disease. The soybean cultivars ‘P15T46R2′ and ‘B150Y1′ in particular were consistently tolerant to F. graminearum, F. oxysporum, F. redolens and F. solani, making them promising candidates for farmers to minimize the risk of Fusarium root rot. Furthermore, this study provided valuable insights into the distribution and composition of Fusarium spp. in the major soybean production areas of Canada, updating and complementing existing information on the Fusarium root rot complex in soybean cultivation. These findings may help to guide the development of effective measures to mitigate the risk of Fusarium root rot of soybean.