Fungal strains were isolated from freshwater environments by collecting samples from soil sediments, water, and plant litter in various regions. Details of the strains used in this study are provided in Table 1. The freshwater samples were immediately filtered onsite using a hand pump to pass 50 mL of the sample through an MCE membrane filter (HAWP04700; MF-Millipore, Merck Millipore, Burlington, MA, USA). The filters were then placed onto water agar (WA; 20 g/L agar, 1 L distilled water, 100 ppm streptomycin) plates and incubated at 15°C for 2 days. The collected freshwater sediments and plant materials were washed twice with sterile distilled water (SDW) and placed in SDW at 20°C for 1 week. After incubation, 100 μL of the cultured water was spread onto WA and incubated at 15°C for 2 days. The collected soil samples were serially diluted to 10⁻², 10⁻³, and 10⁻⁴ using SDW, and 200 μL of each dilution was spread onto potato dextrose agar (PDA; Difco, Franklin Lakes, NJ, USA) plates and incubated at 15°C for 2 days. All the fungi cultured from the isolates were subjected to single-spore isolation. The pure isolates were inoculated onto PDA and incubated at 20°C. The isolated fungi were suspended in 15% glycerol and stored at −80°C.

To identify the isolated strains, fungal mycelia cultured on PDA were harvested and placed in tubes containing glass beads for homogenization. Fungal genomic DNA was isolated using the NucleoSpin Plant II DNA extraction Kit (Macherey-Nagel, Düren, Germany) and MagListo 5M Plant Genomic DNA Extraction Kit (Bioneer, Daejeon, Korea). For molecular identification of the fungi, PCR amplification of the internal transcribed spacer (ITS) rDNA region was performed using primers ITS5 (5′-GGAAGTAAAAGTCGTAACAAGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) (White et al., 1990). The translational elongation factor 1-alpha (TEF1α) region was amplified using primers EF1 (5′-ATGGGTAAGGARGACAAGAC-3′) and EF2 (5′-GGARGTACCAGTSATCAT-3′) (O’Donnell et al., 1998). Amplicons were sequenced by Macrogen (Seoul, Korea) using the same primers. MEGA-X software (Biodesign Institute, Tempe, AZ, USA) was used for sequence alignment and phylogenetic tree construction.

The isolates were cultured on PDA medium and incubated at 25°C for 5 days. The morphological characteristics of the fungal mycelia were examined based on the spore characteristics described by Jaklitsch (2009). For microscopic observation, a drop of distilled water was placed on a glass slide, and a small section of mycelia was cut. The specimen was then covered with a coverslip and observed under a microscope (Goh et al., 2018). Conidiophores and conidia were examined using an Eclipse Ni microscope (H550S, Nikon, Tokyo, Japan) equipped with a Ds-Ri2 digital camera (Nikon). Conidia dimensions are presented as the mean values of a minimum of 50 conidia, with extreme values in parentheses.

Pathogenic fungal cultures of Fusarium oxysporum (Fo), F. solani (Fs), and Pc were obtained from the Korean Agricultural Culture Collection (KACC), National Agrobiodiversity Center (NAAC), Rural Development Administration (RDA), Korea. Additionally, Ca, Cg, Csi, and Cfr were sourced from the Plant Clinical Pathology Laboratory at Andong National University.

A screening experiment was conducted to evaluate the antifungal activity of freshwater fungal isolates against phytopathogenic fungi, using a dual culture plate assay as per the procedure described by Hernandez Castillo et al. (2011). Freshwater fungal isolates were cultured on PDA medium and incubated at 25°C for 7 days. Pathogenic fungi, Pc and Fs, were also cultured on PDA medium under identical conditions. Using a 5-mm cork borer, mycelial plugs were collected from 7-day-old cultures of both the freshwater isolates and the pathogens. These plugs were placed 4 cm apart on opposite sides of a 6-cm-diameter Petri dish. The control group comprised pathogenic fungi on PDA plates paired with agar plugs without fungi. The plates were incubated at 25°C for 10 days. Following incubation, the degree of growth inhibition between the mycelial plugs and the pathogen was measured and compared with that of the untreated control. The inhibition rate was calculated using the method provided by Rahman et al. (2009), following formula:

Siderophore production and phosphate-solubilizing ability, which have positive effects on plant growth, were evaluated. The ability of the fungal strains to produce siderophores was assessed using the Chrome Azurol Sulfate (CAS) assay developed by Schwyn and Neilands (1987). To prepare CAS agar, 30.24 g/L PIPES and 1.5% agar were dissolved in 900 mL double-distilled water (ddH₂O). The pH was adjusted to 6.5 using 10 N NaOH, and the solution was sterilized by autoclaving at 121°C for 15 min. After sterilization, the CAS agar was supplemented with a sterile blue dye solution (CAS, 2 Mm; FeCl₃·6H₂O, 1 mM; HDTMA [CTAB], 5 mM). The prepared CAS was poured into six-well plates. After solidification, the agar was cut in half and placed in contact with the cut surface of a 7-day-old PDA fungal culture using the method described by Osman et al. (2019). Siderophore production was indicated by the degree of discoloration of the CAS agar in contact with the fungal mycelium. The discoloration rate was calculated using the following formula:

Furthermore, phosphate-solubilizing ability was assessed using Pikovskaya’s (PVK) agar plate method described by Zhang et al. (2018). Briefly, 31.3 g/L PVK agar and 0.025 g/L bromophenol blue were dissolved in ddH₂O and sterilized by autoclaving at 121°C for 15 min. After placing the fungal plug in the center of the plate, it was incubated at 25°C for 7 days. The solubilization index was calculated using the method described by Doilom et al. (2020), according to the following formula:

The ability of fungi to produce various enzymes, including amylase, chitinase, and glycosidase, was assessed in vitro. The freshwater fungi used in the experiment were cultured on PDA medium at 25°C for 7 days, after which 5-mm plugs were cut from the cultures using a cork borer. The ability of fungi to produce various enzymes, including amylase, chitinase, and glycosidase, was assessed in vitro. The freshwater fungi used in the experiment were cultured on PDA medium at 25°C for 7 days, after which 5-mm plugs were excised from the cultures using a cork borer. Amylase and chitinase activities were determined following the method described by Mun et al. (2023). For the amylase activity assay, the medium was prepared by dissolving 0.1% AZCL-amylose, 0.5% xanthan gum, and 2% agar in ddH₂O. The chitinase activity assay medium was prepared by adding 4.5 g of colloidal chitin, 0.3 g of MgSO ·7H₂O, 3 g of (NH₄)₂SO₄, 1 g of KH₂PO₄, 0.15 g of bromocresol purple, 0.2 mL of Tween 80, and 15 g of agar to 1 L of ddH₂O. The glycosidase activity assay medium was prepared by dissolving Esculin Iron Agar (16.5 g/L) and agar (10 g/L) in ddH₂O, following the method of Mathlouthi et al. (2002). All media were sterilized by autoclaving at 121°C for 15 min. A fungal plug was placed at the center of each medium, and the plates were incubated at 25°C. After 3 days, enzyme activity was assessed. Amylase and chitinase activities were evaluated by observing the presence of clear zones around the fungal plug. Glycosidase activity was measured based on the length of the darkened zone, signifying hydrolysis by the fungi. The enzyme activity index was determined following the method outlined by Azuddin et al. (2024), using the following formula:

Trichoderma strains with strong antifungal activity were selected for further experiments. The antifungal efficacy of these isolates were evaluated against six plant pathogens: apple bitter rot (Cg, Csi, and Cfr), red pepper anthracnose (Ca), red pepper Phytophthora blight (Pc), and red pepper Fusarium wilt (Fo). Dual-culture assays were performed as described by Hernandez Castillo et al. (2011). The pathogens were dual-cultured with the Trichoderma isolates on PDA medium and incubated at 25°C for 10 days. The antifungal activity of the Trichoderma isolates was determined by measuring the inhibition of pathogen mycelial growth. The inhibition rate was calculated using the method provided by Rahman et al. (2009), following formula:

An experiment was conducted to determine whether Trichoderma spp. isolated from freshwater can produce VOCs that inhibit the growth of plant pathogenic fungi. This experiment was based on the methodology described by Kim et al. (2023b) using a sandwich plate assay. Trichoderma and the pathogenic fungi were inoculated at the center of separate PDA plates and incubated at 25°C for 10 days. After incubation, 5 mm fungal plugs were collected from the cultures using a cork borer. The plugs of the pathogenic fungi and Trichoderma isolates were inoculated in the center of 6 cm PDA plates, which were then assembled in a sandwich format. In the control group, PDA plates containing pathogenic fungi were overlapped without Trichoderma. All plates were incubated at 25°C for 4 days, and the inhibition of mycelial growth of the pathogenic fungi was observed. The inhibition rate was calculated using the method provided by Rahman et al. (2009), following formula:

To evaluate the plant growth-promoting potential of Trichoderma spp., red pepper seeds (cv. Millennium) were sown in plant trays (570 × 390 × 90 mm) filled with commercial garden soil (Shinsung Minera Ltd., Goesan, Korea). After 4 weeks, the germinated seedlings were transplanted into pots (12 × 10 cm). Trichoderma cultured on PDA medium for 10 days were cut into 5 mm plugs, and three plugs were inoculated into potato dextrose broth liquid medium. After 7 days, 25 mL of Trichoderma suspension (10⁵ spores/mL) was applied to the soil in the pots. The Trichoderma suspension was drenched three times at 14-day intervals. The plants were grown in a greenhouse maintained at 25 ± 5°C. Shoot length, stem diameter, and number of flowers were measured and recorded 45 days after inoculation.

We evaluated the preventive effects of Trichoderma suspensions against Fusarium wilt and Phytophthora blight on red pepper seedlings. Fo was cultured on PDA medium at 25°C for 10 days, and the pathogen concentration was adjusted to 1 × 10⁵ spores/mL using SDW. Pc was cultured on V8 medium (80 mL V8 juice, 10 g agar, 1 L ddH₂O, pH adjusted to 6.5 with 10 N NaOH) at 25°C for 10 days. To induce zoospore formation, SDW was added to the V8 plates, followed by a cold shock (10°C for 24 h) and exposure to broad-spectrum light at room temperature for 24 h. Red pepper seedlings were transplanted to plastic pots (12 × 10 cm) containing commercial garden soil. After 1 week, 25 mL of Trichoderma suspension from seven different strains was applied three times at 7-day intervals. Two days after the third application, 25 mL of Fo suspension was applied to the soil to cause Fusarium wilt. To induce Phytophthora blight, Pc was inoculated according to the method of Bartual et al. (1993), in which a plug (Ø 6 mm) of mycelium was placed on a wounded stem. The control group consisted of seedlings treated with SDW only. Pots were maintained in a greenhouse at 25°C. Disease severity caused by both pathogens was assessed 14 days after inoculation. Disease severity (%) was calculated using the method described by Ahmed et al. (2022), and the control value (%) was calculated using the method described by Heo et al. (2024), according to the following formula:

Disease severity (%) was calculated based on the disease index scale (from 0 to 4), where 0 = no visible disease symptoms; 1 = slightly wilted leaves with brownish lesions beginning to appear on the stems; 2 = 30-50% of plants diseased; 3 = 51-70% of plants diseased; 4 = 71-100% of plants diseased. The following equations were used:

An experiment was conducted to determine whether Trichoderma suspension could control anthracnose in red pepper fruits. This experiment was performed based on the method described by Kim and Kim (2020). Red peppers of similar size were surface sterilized by immersion in 1% NaOCl for 3 min and 70% ethanol for 2 min. Subsequently, the red peppers were rinsed twice with SDW and air-dried. Three wounds were made on the surface of each red pepper using a sterile needle. A 10 μL suspension of Trichoderma (diluted 100-fold and 500-fold with SDW) was applied to the wounded sites. After drying, a 10 μL suspension of Ca (10⁵ conidia/mL) was applied to the same sites. The red peppers were then placed in a plastic tray lined with a wet towel at the bottom and incubated at 25°C. Seven days after inoculation, the disease index was measured to calculate the disease severity and control values. The disease index rated on a scale from 0 to 4, where 0 = no symptoms; 1 = 1-25% with disease lesions ≥ 2 mm; 2 = 26-50% with disease lesions ≥ 4 mm; 3 = 51-75% with disease lesions < 8 mm; and 4 = 76-100% with sunken lesions and spore production. Disease severity (%) was calculated using the method described by Ahmed et al. (2022), and the control value (%) was calculated using the method described by Heo et al. (2024), according to the following formula:

Data were analyzed using analysis of variance (ANOVA) with R software (RStudio “Prairie Trillium” Release [2022-05-20] for Windows). An independence test was performed using Pearson’s chi-squared test (Agresti, 2007). Significant differences between treatment means were determined using the least significant difference method, with statistical significance set at P < 0.05. All experiments were performed at least twice and the data were analyzed separately for each experiment. The results from representative experiments are provided.

The phylogenetic relationships of Trichoderma spp. were determined using sequences from two molecular markers: ITS and TEF1α. Seven Korean isolates from freshwater environments (Table 1) and previously published authentic isolates were included in the analysis. Phylogenetic trees based on ITS and TEF1α clearly resolved the seven isolates, all belonging to Longibrachiatum Clade, and were identified as T. longibrachiatum and T. capillare. Strains FBCC-F1514, FBCC-F1537, FBCC-F1645, FBCC-F1688, and FBCC-F1835 formed a well-supported clade (100% bootstrap value) with the type isolate T. longibrachiatum ATCC18648. Similarly, strains FBCC-F1634 and FBCC-F1547 clustered with the type isolate T. capillare C.P.K.2883, which was supported by 100% bootstrap values (Fig. 1).

The morphological characteristics of T. longibrachiatum and T. capillare were examined by culturing isolates on PDA and observing colony morphology and conidial structures over time (Fig. 2).

T. longibrachiatum, isolates FBCC-F1514, FBCC-F1688, FBCC-F1645, and FBCC-F1835 produced white aerial mycelia that were densest at the center and became sparser toward the periphery after 5 days of incubation. The colony reverse was also white in these isolates. FBCC-F1537 exhibited white aerial mycelia with a yellow reverse. After 7 days (Fig. 2A-E), the aerial mycelia and reverse of FBCC-F1835 had turned green. Microscopic observation revealed lageniform phialides, mostly straight, nearly cylindrical or distinctly swollen below the middle (Fig. 2H and I). The conidia were green, slightly rough-walled, subglobose to broadly ellipsoidal, and measured 3.0-4.5 × 2.2-3.2 μm (length × width) (Fig. 2J). The morphology observed in this study matches that of T. longibrachiatum as described in earlier reports (Kim et al., 2023b; Park et al., 2005).

T. capillare, isolate FBCC-F1634 displayed white to green aerial mycelia that were dense at the center and became lighter towards the margin after 5 days of growth, with the reverse side showing a yellow coloration. Isolate FBCC-F1547 also showed white to green aerial mycelia, although growth was relatively shallow and the reverse was similarly yellow. After 7 days (Fig. 2F and G), FBCC-F1634 and FBCC-F1547 showed a distinct green coloration of the aerial mycelia. Phialides were lageniform, mostly straight, nearly cylindrical, or swollen below the middle (Fig. 2K and L). The conidia were green, slightly roughened, subglobose to broadly ellipsoidal, and measured 2.3-4.0 × 2.0-3.0 μm (length × width) (Fig. 2M). The observed morphological traits correspond well with the descriptions of T. capillare provided by Goh et al. (2018).

An experiment was conducted to evaluate the antifungal activity of 44 Trichoderma strains against pathogenic fungi Pc and Fs. The results showed that all 44 strains exhibited antifungal activity against these pathogens, with inhibition rates ranging from 40-100%. Among these, the seven selected Trichoderma strains demonstrated high antifungal activity, with FBCC-F1835 exhibiting a particularly strong antagonistic ability, achieving 100% inhibition of both pathogens (Table 2).

The ability of 44 Trichoderma strains isolated from freshwater to produce siderophores and solubilize phosphate was evaluated using agar plate assays. Among them, 30 strains showed siderophore production, whereas all strains exhibited phosphate solubilization activity. Seven selected Trichoderma strains demonstrated high activity, with siderophore production > 80% and phosphate solubilization > 60% (Table 2). These findings suggest that the isolated Trichoderma strains not compete with pathogens for Fe³⁺ by producing siderophores but also possess the ability to solubilize insoluble phosphate, thereby enhancing nutrient availability.

A total of 44 Trichoderma strains were evaluated for chitinase, glycosidase, and amylase activities using enzyme activity plate assays. The results showed that most strains exhibited glycosidase and amylase activities, whereas chitinase activity was minimal. Among the selected strains, six displayed glycosidase activity and four exhibited amylase activity (Table 2). These findings suggest that glycosidase activity may contribute to the activation of plant defense mechanisms under stressful conditions.

Seven strains were selected through screening assays (Table 2). Dual-culture assays were conducted to test the antimicrobial activities of the selected strains against various pathogenic fungi. All strains showed > 80% antifungal activity against Colletotrichum spp. Notably, strain FBCC-F1835 exhibited high antifungal activity, with 94.17%, 87.50%, and 94.17% inhibition of Csi, Cfr, and Cg, respectively. For Ca, strain FBCC-F1537 showed the highest control rate (89.17%). For Fo, the strains demonstrated > 70% antifungal activity, ranging from 72.50-93.33%. Additionally, all seven strains exhibited 100% inhibition of mycelial growth of Pc (Fig. 3). An overlapping plate assay was conducted to determine whether the VOCs produced by the selected Trichoderma strains inhibited pathogenic fungi. The results showed that VOCs from Trichoderma strains inhibited mycelial growth in all pathogens, except Csi. Strain FBCC-F1514 exhibited superior inhibitory effects against five pathogenic fungi: Ca, Csi, Cfr, Cg, and Fo. Strain FBCC-F1645 showed 100% inhibition of Pc (Fig. 4).

The growth-promoting effects of the Trichoderma isolates were evaluated on red pepper seedlings under greenhouse conditions over 45 days (Fig. 5). All treatments demonstrated an increase in shoot growth. The control group exhibited an average shoot length of 448.33 mm, whereas the FBCC-F1688 treatment resulted in the highest shoot length, measuring 616.67 mm. Similarly, stem diameter increased in all treatments compared to that in the control, with FBCC-F1514 yielding the thickest stems. The control group did not produce any flowers, whereas all treatments, except FBCC-F1645, induced flowering. The FBCC-F1634 treatment produced the highest number of flowers with an average of seven (Table 3). These results suggest that the tested Trichoderma strains have possess substantial plant growth-promoting potential, making them promising candidates for use as BCAs with added agronomic benefits.

The biocontrol efficacy of Trichoderma strains against Fusarium wilt in red pepper was evaluated 14 days after inoculation with a Fo suspension under greenhouse conditions. All seven Trichoderma strains exhibited control rates > 40%. Among these, FBCC-F1547 demonstrated the highest efficacy with a complete disease control rate of 100%. The remaining strains, excluding FBCC-F1634, showed control rates > 70%, with the lowest efficacy of 43.75% (Fig. 6). These findings underscore the potential of FBCC-F1547 as a highly effective biocontrol agent for the management of Fusarium wilt in red pepper.

Further evaluation of biological control efficacy was conducted by applying Pc mycelial plugs to wounds on pepper plants inoculated three times with a Trichoderma suspension. All seven Trichoderma strains tested demonstrated disease control rates > 50%, with FBCC-F1547 and FBCC-F1688 showing the highest efficacy of 75%. FBCC-F1514 exhibited the lowest efficacy of 50%, with wilting and defoliation consistently observed in the leaves. Cross-sectional stem analysis revealed that the control group was completely desiccated, whereas the treated groups displayed only minor browning of the stem tips (Fig. 7). These results suggest that the tested strains have the potential to act as BCAs to directly prevent Fusarium wilt and Phytophthora blight in plants.

The efficacy of Trichoderma suspensions in controlling anthracnose in red peppers was evaluated using 100-fold and 500-fold dilutions. After 7 days of incubation in a plant growth chamber, both dilutions effectively suppressed anthracnose. The 500-fold dilution achieved preventative effects ranging from 40-86.67%, whereas the 100-fold dilution showed higher efficacy, with preventive effects ranging from 66.67-86.68%. Notably, the 500-fold suspensions of FBCC-F1514 and FBCC-F1645 exhibited greater control efficacy than their respective 100-fold suspensions. All treatments, except for FBCC-F1688 (100-fold and 500-fold) and FBCC-F1835 (500-fold), exhibited anthracnose control values > 70%. Moreover, FBCC-F1634 demonstrated consistent efficacy across both dilutions (Fig. 8). These findings highlight the potential of the seven Trichoderma isolates as effective BCAs against red pepper anthracnose.