Citrus trees (Clementine mandarin variety: Gungcheon) were grafted to a trifoliate orange tree and transplant to pots (ø = 30 cm) filled with commercial soils (Number-One, Hongseong, Korea) and 10% perlite (Parat, Sam Son, Seoul, Korea). These citrus plants were placed in an incubation room maintained at 25°C with 90% humidity and 14 h of photosynthesis period. Each leaf of sprout at approximately 10 days after budding was cut with a sterilized razor blade and used for test.

Diaporthe citri causing melanose in citrus plants (Clementine mandarin variety: Gungcheon) was isolated in the laboratory as described previously (Ko et al., 2012). For sporulation, D. citri was grown on a potato dextrose agar (PDA; Becton, Dickinson and Company, Claix, France) medium under aerobic condition for 14 days according to published method (Ko et al., 2012). For inoculums, 10 ml of distilled water was added onto PDA formed with pycnidia of fungus. Conidia were harvested using a loop and filtered with a double folded mira-cloth. The concentration of inoculum was adjusted to 1 × 10⁵ conidia/ml using a hemocytometer (Hausser Scientific Inc., Horsham, PA, USA).

Bio-sulfur, consist of Thiobacillus sp. and element sulfur, was taken from Ecobio Holding Co. Ltd. (Incheon, Korea). To test the antifungal effect, bio-sulfur was added to PDA or potato dextrose broth (PDB; Becton, Dickinson and Company, Claix, France) to the concentration at 1,000 ppm. PDA was used to test antifungal effect of bio-sulfur. An agar block containing mycelium of D. citri was inoculated onto the center of PDA medium followed by incubation at 25°C for 7 days. Diameters of mycelia were measured with a ruler for untreated groups and groups treated with bio-sulfur or with Dithianon. For liquid medium, three mycelium blocks of D. citri were dropped into PDB followed by incubation at 25°C with shaking at 110 rpm in a shaking incubator (HB-201SL, Hanbaek Scientific Co., Bucheon, Korea) for 7 days. Fresh weight of mycelia was measured with an electronic balance. Experiments were separately replicated three times and every experiment contained three media both PDA and PDB.

Leaves of citrus sprouts of 10 days old were inoculated with D. citri. These leaves were cut and laid in a Petri-dish (ø = 90 mm). The end of petiole was wrapped with cotton wool soaked with sterilized water. These citrus leaves were sprayed with sterilized water, bio-sulfur solution at 1,000 ppm, or Dithianon at 0.75 g/l. After drying for 3 h, the inoculum added with 0.01% Tween 20 (Yakuri Pure Chemicals Co., Kyoto, Japan) was dropped onto leaves 4 points every 10 μl each followed by incubation in an incubator (DA MIL-2500, Dong-A, Siheung, Korea) at 25°C for 7 days. Diameters of infected sites showing symptoms of melanose were measured with a ruler. Experiments were separately replicated three times and every treatment contained three leaves.

To determine the mechanism involved in disease suppression of bio-sulfur on citrus leaves, both treated with chemicals and untreated leaves were observed with a fluorescent microscope (BX60, Olympus, Tokyo, Japan) at 1, 3, and 5 days after inoculation. Ten microliters of inoculum suspension adjusted to 1 × 10⁵ conidia/ml was dropped on citrus leaves. Inoculated sites of samples were cut with a razor blade in size of 5 × 5 mm² and fixed with 2% glutaraldehyde in phosphate buffer (pH 7.2) at 4°C for 2 h. Fixed samples were washed with phosphate buffer three times (10 min each) and dyed with 2% diethanol (UVtex-2B, Fungiqual A, Muellheim, Germany) for 40 m at room temperature. These samples were washed phosphate buffer three times again (10 min each) and mounted on glass slides with 70% glycerin (OCI Company Ltd., Seoul, Korea). Infected sites were observed with a fluorescent filter set (exciter filter, BP 400-440; interference beam splitter, FT 460; barrier filter, LP 470). The number of germinated conidia and hyphal length were determined for all treatment groups. Experiments were separately replicated three times using three samples for each treatment.

Surfaces of inoculated citrus leaves both treated with chemicals and untreated were observed with a scanning electron microscope to see fine structures of D. citri. Inoculated parts of leaves were cut to sizes of 5 × 5 mm² and fixed with a 2% glutaraldehyde in cacodylic buffer (pH 7.2) at 4°C for 2 h. Samples were washed with cacodylic buffer three times (10 min each). Post fixation was carried out with 1% osmium tetroxide in cacodylic buffer at 4°C for 2 h. These fixed samples were washed with distilled water two times for 10 min each. Dehydration was performed with 30%, 50%, 70%, 90%, and 100% ethanol for 30 min each followed by treatment with 100% twice for 30 min each. Samples were dried with a critical point dryer (CPD 030, BAL-Tec, Los Angeles, CA, USA) and coated with platinum by sputter coater-platinum (Q150R Plus - Rotary Pumped Coater, Quorum Technologies Ltd., Sussex, UK) at 20 mA for 90 s. Coated samples were observed with a field emission scanning electron microscope (FE-SEM Mira3, Tescan Ltd., Brno, Czech Republic).

Data of diameters of mycelia on the media for determining the antifungal effect of bio-sulfur, sizes of inoculated citrus leaves showing symptoms, the number of germinated conidia, and hyphal lengths on surfaces of citrus leaves were analyzed with Duncan’s multiple range test using SAS program version 9.0 (SAS Institute Inc., Cary, NC, USA). Statistical significance was considered at P < 0.05.

At 7 days after inoculation, mycelial growth on PDA added with bio-sulfur was inhibited compared to that on untreated PDA (Fig. 1A and B). On PDA containing a commercial fungicide Dithianon, mycelia were rarely detected (Fig. 1C). Diameters of mycelia on PDA treated with bio-sulfur were decreased by 24% compared to those on untreated PDA, indicating that bio-sulfur had antifungal activity against D. citri. Dithianon also showed significant suppression effect on mycelia (Fig. 1D). The antifungal effect of bio-sulfur was apparent in PDB in which the contact extent of bio-sulfur was increased. At 7 days after inoculation, mycelial ball was smaller in the group treated with bio-sulfur than that of untreated group (Fig. 2A and B). Similarly, mycelial ball in the group treated with Dithianon was much smaller than that of the untreated group (Fig. 2C). Fresh weight of mycelia grown in bio-sulfur treated liquid media was also decreased to 65% compared to that of mycelia in the untreated group (Fig. 2D).

It has been reported that sulfur compound has direct antifungal effect on an artificial medium. Kim et al. (2018) have shown that paper disk containing loess-sulfur on PDA medium has antifungal activities against Colletotrichum gloeosporioides known to cause anthracnose in jujube. Also, inorganic sulfur compound on PDA media added with an inorganic sulfur has a distinctive growth inhibition effect on Cylindrocarpon destructans and Fusarium solani known to cause root rot in ginseng plant (Lee et al., 2014b). Furthermore, mycelium growth of Fusarium oxysporum causing vascular wilts in various plants is decreased on PDA medium added with sulfur nanoparticles (Choudhury et al., 2011). In the present study, both diameter of D. citri on bio-sulfur added PDA medium and weight of mycelia in PDB medium containing bio-sulfur were decreased clearly compared to those of untreated group (Figs. 1 and 2). In order to evaluate the antifungal activity of bio-sulfur comparing to the commercial fungicide Dithianon in the farm, concentration of the fungicide applying in the field was applied in this study. These results revealed a direct antifungal effect of bio-sulfur against D. citri on artificial media.

Disease severity on bio-sulfur pretreated leaves was highly inhibited compared to that of untreated one (Fig. 3A and B), indicating that bio-sulfur had disease suppression capacity. However, the protection rate by bio-sulfur was not as high as Dithianon. After pretreatment with Dithianon, disease symptoms were hardly observed (Fig. 3C). Average sizes of leaves showing disease symptoms in both treatment groups (bio-sulfur or Dithianon) were significantly reduced compared to those of untreated group (Fig. 3D).

It has been known that sulfur contains substance that could suppress disease severity of some crop plants. It has been reported that disease severity of powdery mildew is decreased by about 90% in tomato plants treated with loess-sulfur mixture (Shim et al., 2014). Ginseng anthracnose caused by Colletotrichum gloeosporioides is also suppressed by treatment with lime sulfur diluted 400 times (Lim et al., 2015). Sulfur containing salts can also inhibit postharvest diseases such as carrot cavity spot caused by Pythium sulcatum and potato dry rot caused by Fusarium sambucinum (Kolaei et al., 2012). Likewise, disease severity caused by D. citri on citrus leaves was decreased by pretreatment with bio-sulfur in the present study (Fig. 3), suggesting that bio-sulfur could be a candidate substance as a new fungicide.

To compare the infection behavior of D. citri, citrus leaves pretreated with bio-sulfur or Dithianon were observed with a fluorescent microscope. Most conidia were germinated at 1 day after inoculation. They grew fast to form mycelium at 5 days after inoculation on untreated citrus leaves (Fig. 4A, D, and G). However, germination rate of conidia on bio-sulfur pretreated leaves were apparently decreased (Fig. 5A). In addition, most of hyphal growths were limited compared to untreated group (Figs. 4B, E, H, and 5B). These results indicate that conidial germination and hyphal growth could be inhibited by bio-sulfur. Most conidia could not germinate on leaves pretreated with Dithianon until 3 days after inoculation (Figs. 4C, F, and 5A). Lengths of hyphae in the group treated with Dithianon were smaller than those in the group treated with bio-sulfur (Fig. 5B), explaining why Dithianon had stronger inhibition effect on melanose disease severity (Fig. 3).

Inhibition of sulfur on spore germination has been reported in many plants. For example, conidial germination rate of Stemphylium botryosum causing garlic leaf blight is decreased more than 90% on garlic leaves treated with sulfur compound (Ryu et al., 2015). Less conidial germination rate of Erysiphe cichoracearum causing powdery mildew on okra treated with sulfur fungicides such as nanosulphur, Canadian nanosulphur, Merck Sulphur, and sulphur 80 WP has also been reported (Gogoi et al., 2013). Conidial germination rate of Botrytis cinerea is also inhibited 99% by buffered grape juice containing sulfur dioxide at 2.8 μg/ml (Smilanick et al., 1990).

It is well-known that inhibition of hyphal growth is a major cause of disease suppression on many crop plants. Mycelium growth of gray mold caused by B. cinerea on grapefruits is inhibited by treatment with sulfur dioxide fumigation, resulting in decreased severity of gray mold (Gabler et al., 2010). Also, inhibition of mycelium growth by sulfur can control disease caused by Acremonium acutatum or Trichothecium roseum on grapes (Oh et al., 2014). In case of peach fruit fumigated with sodium hydrosulfide, mycelium development of Monilinia fructicola causing brown rot is reduced, resulting in less disease incidence (Wu et al., 2018). These observations imply that low germination rate of conidia and inhibition of hyphal growth by bio-sulfur might be major factors decreasing melanose severity on citrus leaves.

Citrus leaves were observed with a scanning electron microscope to reveal the ultra-structure of D. citri. At 1 and 3 days after inoculation, morphologies of most conidia and hyphae were intact on untreated citrus leaves (Fig. 6A and D). At 5 days after inoculation, a lot of hyphae were entangled. Tangles of hyphae showed a smooth surface on untreated citrus leaves (data not shown). However, on citrus leaves pretreated with bio-sulfur, most conidia were not germinated. Some germinated conidia were shrunk at 1 day after inoculation (Fig. 6B). At 3 days after inoculation, some conidia were flat and tidy compared to those of untreated one. Also, some hyphae were morphologically deformed and branched (Fig. 6E). Two days later, morphologically changed conidia and branched hyphae were frequently observed (data not shown), indicating that conidia and hyphae of D. citri were directly affected by bio-sulfur. More than twofold numbers was counted in the hyphal branch on the observed scanning electron microscopy images pretreated with bio-sulfur compared with those of untreated one (data not shown). On Dithianon pretreated leaves, conidia were rarely found at all time of observations (Fig. 6C). Most surfaces of conidia were puckered and reduced in size at 3 days after inoculation (Fig. 6F). These phases were also shown similarly on citrus leaves at 5 days after inoculation, showing shrunk and roughed conidia (data not shown).

Some reports have revealed that fungal structures are modified by treatment with sulfur compounds. For examples, hyphae of Penicillium italicum on mandarin fruit fumigated with hydrogen sulfide are modified, leading to decrease of citrus blue mold (Fu et al., 2014). Also, modification of conidia and hyphae of Aspergillus niger by spray with sulfur might lead to inhibition of powdery mildew on mango (Reuveni et al., 2018). Morphological changes or forming branch of filamentary fungus indicate harmful environment such as exposure to chemicals or fungicide. Therefore, bio-sulfur may play a role as a fungicide to control melanose.

In conclusion, bio-sulfur treatment can significantly decrease conidia germination and hyphal growth of D. citri. When fungus had contact with bio-sulfur, modification of conidia and hyphae was detected morphologically. These results suggest that bio-sulfur can be used to control melanose in eco-friendly citrus farms where the use of chemicals is limited. However, in order to use bio-sulfur in the farm practically it should be prior to test either its toxicity in agricultural environment or its influence by chemical residue in human health.