The fungal pathogen Colletotrichum acutatum SM017 was identified (Yu and Yun, 2007) and isolated from field-grown hot pepper in Asan. C. acutatum was cultured on potato dextrose agar (PDA, Difco, Grand Island, NY) (24 g potato dextrose, 15 g agar/l) for 10 days at 25°C under fluorescent lamps with a 12 h light/dark period. Conidia of C. acutatum were collected from cultures using 2 ml of sterile water. Colonies were then streaked on plates using a brush, and the mixtures of mycelia and conidia were filtered through four layers of cheesecloth. The concentration of the conidial suspension was adjusted to 1×10⁵ conidia/ml using a light microscope (Eclipse 80i, Nikon, Tokyo, Japan) with a hemacytometer.

In total, 813 myxobacterial extracts were obtained from Myxobank (http://www.myxobank.or.kr/), which had been previously (extracted in each myxobacterial) isolated from South Korean soils between 2002 and 2008. Among these isolates, 455 consisted of 318 cellulolytic and 137 bacteriolytic myxobacteria that had been previously studied (Table 2; Kim and Yun, 2011b). Five hundred thirty-six isolates, or 65.9% of the tested extracts, were cellulolytic and identified as S. cellulosum. The remaining isolates were bacteriolytic myxobacteria and identified as Corallococcus spp., Myxococcus spp., Stigmatella spp., and Citrobacter sp., which comprised 19.4%, 12.9%, 0.4%, and 0.1%, respectively, of the total isolates tested. S. cellulosum extracts were provided from Myxobank and prepared with ST21CXP agar medium (Hyun et al., 2009). Crystals were obtained from 50 ml of the CSG/1 liquid medium and dissolved in dimethyl sulfoxide (DMSO) to make 1 ml of myxobacterial extract. After selecting four isolates of cellulolytic myxobacteria (KYC 3512, 3584, 3279, and 3262), cell-cultured filter paper of the four isolates was cultured on ST21P agar medium and transferred to cell-cultured filter paper to CSG/1 liquid medium by shaking at 32°C to obtain liquid culture filtrates (Hyun et al., 2009).

A dual culture assay for antagonism was conducted on PDA against C. acutatum with antagonistic myxobacterial extracts for 10 days. Mycelial agar plugs (5 mm in diameter) of the pathogen were placed in the middle of the PDA medium and 12 μl of myxobacterial extract was placed on the mycelia for 7 to 10 days at 25°C. Mycelial growth was then measured. Each replication consisted of three plates. The clear zone around conidia C. acutatum was examined to assess blocking of germination by myxobacterial extracts. To test for germination, a 250 μl conidial suspension adjusted to 1×10⁵ conidia/ml was smeared on PDA plates, and an 8-mm paper disk with 12 μl myxobacterial extract was placed in the middle of the culture and incubated in darkness for 7 days at 25°C. The control was 12 μl DMSO on an 8-mm paper disk for each fungal culture. Diameter of the clear zone (no conidial germination) was measured. Among the tested myxobacterial extracts, 19 extracts with the greatest antifungal activity were selected based on least mycelial growth and largest clear zone (Table 1). To compare antagonistic activity against hot pepper anthracnose in the bioassays, dithianon (75% active ingredient, wettable powder) was applied as a positive control.

A bioassay for antifungal activity of the selected 19 cellulolytic myxobacteria was conducted on hot pepper fruits. The surface of treated fruits was sterilized with 1% sodium hypochlorite for 2 min and washed twice with sterilized water. Prepared fruits were treated 5-μl drops of extract of the 19 cellulolytic myxobacteria with 50 times dilution after puncturing green pepper with a needle. Fruits were then inoculated with 3 μl of conidial suspensions of anthracnose, and their symptoms were observed at 6 and 7 days after infection. The incidences of anthracnose and control values for the myxobacterium and the fungicide treatments were determined as follows:

Anthracnose incidence (%) = No. of diseased fruits/Total no. of treated fruits × 100;

Control value (%) = (Incidence on untreated plot - Incidence on treated plot)/(Incidence on untreated plot) × 100.

Following screening, four cellulolytic isolates were selected, KYC 3512, 3584, 3279, 3262, and liquid filtrates were prepared to use in the next experiment.

A bioassay for antifungal activity of myxobacteria KYC 3512, 3584, 3279, and 3262 on hot pepper anthracnose was conducted on detached fruits in test tubes and pots in a greenhouse. The prepared culture filtrates of the four cellulolytic myxobacteria cultured in GSG/1 liquid medium were sprayed on detached green pepper fruit following surface sterilization with 70% ethanol. After treated fruits were air-dried for 12 h, the conidial suspension was sprayed on the fruit with an airbrush until runoff. Each of the inoculated fruits was suspended inside a test tube with 4 cm diameter by the fruit stalk tied to a cross-shaped metal brace in the test tube opening. To maintain humid conditions inside the test tube, a piece of paper towel was soaked in sterile distilled water and placed at the bottom of the test tubes. After 7 to 10 days of inoculation, symptoms on the detached fruits were measured. To compare the selected myxobacterial extracts, dithianon was applied as a positive control. The control value was calculated in each treatment as described above.

A bioassay of the four selected myxobacteria in pots was conducted in a greenhouse located in Asan (36° 47′ 57.45 N, 127° 04′ 05.57 E) in July 2011. KYC 3262 was selected and tested in the greenhouse in July 2012. Seedlings of pepper cultivar (cv.) ‘Supermannita’ were grown until developing 10 or more green fruits. Fertilizer and pesticides for aphid control were applied as needed. In 2011, culture filtrates of KYC 3512, 3584, 3279, and 3262 were sprayed on the fruits until the filtrate ran off. Treated plants were dried for 1 day, and then 10⁶ conidia/ml of suspension was sprayed onto treated fruits, and fruits were then placed in a 100% relative humidity (RH) incubator (Vision Korea, Suwon, Republic of Korea) for 24 h at 25°C. Treated plants were grown in the greenhouse at 50% to 55% RH and 25°C to 30°C, and disease progress was observed for 14 to 21 days, with watering at 3-day intervals. As a positive control, 1 mg/ml dithianon was sprayed on plants, with three replications. Each treatment had more than 30 fruits and 4 to 5 pots. In 2012, the selected KYC 3262 was treated 1 day before and after inoculation to compare preventive and curative effects with those of the commercial fungicides. To determine preventive effects, the culture filtrate of KYC 3262 and dithianon were applied to the pots 1 day before the pathogen inoculation. KYC 3262 filtrate and tebuconazole were applied 1 day after the pathogen inoculation for the curative test. Three experiments were repeated in the greenhouse in 2012 using 10 or more plants for each treatment.

Bioassay for KYC 3262 in the field in Asan where was 100 m apart from the greenhouse was performed in 2011, 2012, and 2013. The three treatments consisted of an untreated control, a fungicide spray, and the myxobacteria KYC 3262 culture filtrate. Cultivar ‘Supermannita’ was chosen and transplanted on 10 May 2011, 12 May 2012, and 11 May 2013. Fungicides were sprayed alternatively at 0.5 mg/ml tebuconazole, 1 mg/ml dithianon, 0.5 mg/ml azoxystrobin, or 1 mg/ml chlorothalonil. The spray schedule was generally at 7 to 10 day intervals (Table 3). Insecticide was applied when needed, but was applied less than three times in each growing season. Each plot consisted of at least 90 plants, and 20 plants were selected early in the season for measurement of disease and yield. Disease incidence was the percentage of diseased fruits, and yield was the total fresh weight of red peppers of the selected 20 plants per plot. Anthracnose incidence was measured four times in 2011, two times in 2012, and seven times in 2013. Harvests were conducted three times each year. Weeds were controlled as needed using chemical and cultural methods.

The myxobacterial extracts that completely inhibited mycelial growth and produced a clear zone by blocking the germination of C. acutatum were considered to have antifungal activity (Fig. 1). Of 536 cellulolytic extracts, 45 extracts blocked both C. acutatum germination and inhibited mycelial growth. Of 277 bacteriolytic extracts, only 11 exhibited antifungal activity. The cellulolytic myxobacteria had much more antifungal activity than the bacteriolytic myxobacteria (Fig. 1). Obviously, more cellulolytic extracts displayed antifungal activity when compared with bacteriolytic extracts.

The effect of the antagonists on mycelial growth could be divided into three levels: complete, delayed, and non-inhibition (Table 2). Of 536 cellulolytic myxobacterial extracts, 41.6% were non-inhibited, 40.3% were delayed, and only 18.1% completely inhibited the mycelial growth of C. acutatum. Among 277 bacteriolytic myxobacterial extracts, 82.7% did not inhibit the mycelial growth of C. acutatum, 8.3% completely inhibited mycelial growth, and 9% delayed growth. Blocking spore germination, as shown by a clear zone, was divided into four levels according to size (Table 2). Of 536 cellulolytic myxobacterial extracts, 82.8% of the cellulolytic myxobacterial extracts against C. acutatum had had no clear zone (Table 2). Only 1.1% had clear zones of 20 to 30 mm against C. acutatum, which was the most inhibited fungus. Among the tested 277 extracts of bacteriolytic myxobacteria, one extract blocked conidial germination with clear zones of 20 mm or more (Table 2).

After 5 days of inoculation with anthracnose conidial spraying, disease incidences on detached hot pepper fruits were examined. The incidence of anthracnose in the non-treated control was 85% after wound inoculation. Multiple comparisons among the tested 19 isolates were conducted on the final day of examination. The four lowest disease incidences among the filtrate treatments were with KYC 3279, 3584, 3512, and 3262, which had control values of 33%, 38%, 71%, and 87%, respectively, at 5 post-inoculation (Table 1). These cellulolytic isolates were selected to be used in the next experiment by preparing liquid culture filtrates.

The culture filtrates of KYC 3279, 3584, 3512, and 3262 were sprayed on detached green pepper fruits. Control values of KYC 3262, 3279, 3584, and 3512 measured on the first day were 60%, 53.3%, 53.3%, and 31.6%, respectively (Fig. 2A). Control values of the myxobacteria were slightly lower than the dithianon control value of 61.6%. On the second measured day, control values for KYC 3262, 3279, 3584, and 3512 were 57.7%, 43.6%, 40.4%, and 25.9%, respectively, and that of the positive control dithianon was 43.6% (Fig. 2A).

Hot pepper plants with 10 or more green fruits were used to compare the four selected cellulolytic culture filtrates in the greenhouse at Asan in 2010. The control values with KYC 3262, 3279, 3584, and 3512 on the first measured day were 78.8%, 45.2%, 60.9%, and 45.7%, respectively. The control value for the commercial fungicide dithianon was 86.8% (Fig. 2B). Control values on the second measured day for KYC 3262, 3279, 3584, and 3512 were 59.2%, 45.5% 36.7%, and 31.2%, respectively, while that for dithianon was 55.1%; KYC 3262 had slightly higher control values than the commercial fungicide (Fig. 2B).

To evaluate the preventive and curative effects of the selected KYC 3262, a greenhouse experiment was conducted in 2012. Disease incidence on the final measured day in the non-treated control was 46.6%. The incidence with dithianon (a protective fungicide) was 2.0% after being treated 24 h before the pathogen inoculation, and the incidence with tebuconazole (a curative fungicide) was 7.3% after being treated 24 h after pathogen inoculation. The incidence rates with KYC 3262 at 24 h before and after inoculation were 12.3% and 16.3%, respectively. Control values of KYC 3262 24 h before and after inoculation were 74.6% and 66.0%, respectively, while the respective control values with dithianon and tebuconazole were 92.3% and 82.6% (Fig. 3).

KYC 3262 filtrate applications to control anthracnose in hot pepper were evaluated in the field from 2011 to 2013. Final disease incidence was 93% for unsprayed control plants, 64% for KYC 3262 sprayed plants, and 75% for fungicide sprayed plants in 2011 (Fig. 4A). The final disease incidence was 32% for unsprayed control plants, 3% for KYC 3262 sprayed plants, and 1% for fungicide sprayed plants in 2012 (Fig. 4B). In 2013, the final incidence was 27% for unsprayed control plants, 5% for KYC 3262 sprayed plants, and 2.5% for the fungicide sprayed plants (Fig. 4C). Control values for KYC 3262 were 31%, 89% and 82%, and those for the fungicide were 19%, 97%, and 91% in 2011, 2012, and 2013, respectively (Fig. 4D). Obviously, the 93% anthracnose incidence in 2011 in the unsprayed treatment was the highest disease incidence among the 3 years. The final yield of red pepper fruits from 20 plants at harvest were recorded three times in 2011 and 2013 (Fig. 5). The total yield of the 20 plants measured was 0.1 kg in the unsprayed control, 2.66 kg in the KYC 3262 treatment, and 2.0 kg in the fungicide treatment in 2009. In 2013, the total yield of the unsprayed control was 12.5 kg in the unsprayed control treatment, 18.6 kg in the KYC 3262 treatment, and 20.2 kg in the fungicide treatment.