Biological Control 76 (2014) 87–94
Contents lists available at ScienceDirect
Biological Control journal homepage: www.elsevier.com/locate/ybcon
Mycoparasitism of Phakopsora pachyrhizi, the soybean rust pathogen, by Simplicillium lanosoniveum Nicole Ward Gauthier a,⇑, Karunakaran Maruthachalam b, Krishna V. Subbarao b, Matthew Brown c, Ying Xiao c, Clark L. Robertson a, Raymond W. Schneider a a b c
Louisiana State University Agricultural Center, Department of Plant Pathology and Crop Physiology, Baton Rouge, LA, United States University of California at Davis, Salinas, CA, United States Louisiana State University, Department of Biological Sciences, Baton Rouge, LA, United States
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Simplicillium lanosoniveum wrapped
around urediniospores before colonizing urediniospores. S. lanosoniveum penetrated through germ pores within 24 h after inoculation. Hyphae of S. lanosoniveum erupted from colonized urediniospores 7 days after inoculation. Over 90% of urediniospores were colonized within 5 days after inoculation with S. lanosoniveum.
a r t i c l e
i n f o
Article history: Received 12 July 2013 Accepted 27 May 2014 Available online 4 June 2014 Keywords: Cordycipitaceae Pucciniales Uredinales Glycine max
a b s t r a c t Rust of soybean, caused by Phakopsora pachyrhizi, was first reported in the southeastern US in 2004 where it quickly became established. Yield losses ranged from 35% to more than 80%. The mycophilic fungus Simplicillium lanosoniveum was previously shown to colonize rust pustules on soybean leaves and prevent germination of urediniospores. The number of pustules on soybean leaves also was significantly reduced when S. lanosoniveum colonized leaves. This study examined the antagonistic interactions between P. pachyrhizi and S. lanosoniveum using confocal, transmission (TEM) and scanning electron microscopy (SEM) in order to determine if S. lanosoniveum was mycoparasitic. Co-inoculated detached soybean leaves were examined using TEM and SEM to examine changes in urediniospores colonized by S. lanosoniveum. An isolate of S. lanosoniveum, previously shown to be an antagonist of P. pachyrhizi, was transformed with the green fluorescent protein (GFP) gene and used to document the infection process using confocal microscopy. S. lanosoniveum colonized pustules and coiled around urediniospores before infection. The mycoparasite penetrated urediniospores through germ pores within 24 h after inoculation, at which time organelles showed signs of degradation. By 2 days after inoculation, there was extensive colonization of urediniospores by hyphae of S. lanosoniveum. Hyphae of the mycoparasite erupted from the colonized urediniospores at 7 days after inoculation, and there was extensive sporulation on the surface of urediniospores. Over 90% of urediniospores were colonized within 5 days after inoculation with the mycoparasite. We showed in previous studies that S. lanosoniveum is an antagonist of P. pachyrhizi as well as an effective biocontrol agent, but we were unable to document its mode of action. We now present microscopic evidence that S. lanosoniveum is a mycoparasite. Published by Elsevier Inc.
⇑ Corresponding author. Present address: 201 Plant Science, Lexingon, KY 40546, United States. E-mail address:
[email protected] (N.W. Gauthier). http://dx.doi.org/10.1016/j.biocontrol.2014.05.008 1049-9644/Published by Elsevier Inc.
88
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
1. Introduction Soybean (Glycine max L.) rust, caused by Phakopsora pachyrhizi Syd. & P. Syd., was first reported in Japan in 1902 (Bromfield, 1984; Pivonia and Yang, 2004). Since then, it has spread throughout the soybean-growing regions of Asia, Africa, and the Americas. The disease was reported in the continental United States in 2004 (Schneider et al., 2005). Yield losses in the southeastern United States, where the disease has become endemic, ranged between 35% and >80% in soybeans not protected with fungicides (Dorrance et al., 2008; Rupe and Sconyers, 2008; Walker et al., 2011). P. pachyrhizi overwinters on kudzu ([Pueraria lobata (Willd.) Ohwi]) in the lower Gulf South, where the invasive vine serves as a green bridge during winter months. Thus, protective fungicides are critical for effective disease management as overwintering inoculum is readily available during the entire soybean cropping season. Simplicillium lanosoniveum (J.F.H. Beyma) Zare & Gams was documented as an inhabitant of sori of soybean rust in 2007 (Ward et al., 2011). Under laboratory conditions, diseased soybean leaves inoculated with S. lanosoniveum produced fewer sori, and urediniospores from these sori failed to germinate (Ward and Schneider, 2012). These effects were accompanied by significant reductions in disease severity, and there was an increase in the proportion of brown urediniospores (in contrast to the noninfected hyaline urediniospores) that failed to germinate (Ward and Schneider, 2012). Additionally, field studies demonstrated that S. lanosoniveum colonized soybean leaves infected by the pathogen and reduced overall disease incidence and severity (Ward and Schneider, 2012). Even during periods of latent infection, less DNA of P. pachyrhizi was detected in soybean leaves when S. lanosoniveum was introduced (Ward and Schneider, 2012). S. lanosoniveum belongs to the family Cordycipitaceae, which includes mycoparasitic and entomopathogenic fungi (Humber, 2007; Sung et al., 2007; Zare and Gams, 2001). This fungus is discussed in taxonomical references and collections but has never been examined as a mycoparasite (Bischoff and White, 2004; Centralbureau voor Schimmelcultures; Zare and Gams, 2001). Our previous work, using scanning electron microscopy (SEM), indicated that S. lanosoniveum colonized soybean rust sori, and hyphae ramified throughout sori and coiled around individual urediniospores (Ward and Schneider, 2012). These studies also confirmed that the fungus failed to establish on healthy leaf surfaces, which suggested a high degree of specificity towards P. pachyrhizi and potentially to other rust pathogens. Hyphae traversed leaf surfaces until reaching sori, which were then rapidly colonized (Ward and Schneider, 2012; Ward et al., 2011). Examination of these hyphae of S. lanosoniveum stained with Calcofluor white and viewed with a fluorescence microscope revealed that the antagonist apparently penetrated and coiled inside urediniospores after 3–5 days (Ward, 2009). This preliminary study indicated that S. lanosoniveum may be a mycoparasite. Whether the observed hyphae were from S. lanosoniveum or another colonist from fieldcollected urediniospores could not be ascertained with this method, and confirmation of this possible mycoparasitic interaction remained elusive. In the work presented here, scanning electron microscopy (SEM), transmission electron microscopy (TEM), and confocal laser-scanning (CLSM) microscopy were employed to examine interactions between S. lanosoniveum and P. pachyrhizi in order to determine if the association was mycoparasitic or if S. lanosoniveum was an opportunist that colonized urediniospores after they had been killed. Observations using Calcofluor white, a fluorescent brightener, were inconclusive in a previous study because we could not perform optical sectioning, which would be required to show that hyphae were coiled within urediniospores and that they
originated from the introduced inoculum rather than from a contaminant associated with urediniospores (Ward et al., 2011). Also, it was difficult to discern the extent of internal colonization (infection), which would be required as evidence for mycoparasitism, as opposed to an external association that may or may not have killed urediniospores with an excreted toxin or cell wall degrading enzymes. The objectives of this study were to determine: (1) the time course of infection of urediniospores by S. lanosoniveum, and (2) whether S. lanosoniveum directly penetrated urediniospores of P. pachyrhizi. Determining whether or not the interaction is mycoparasitic has implications for further work regarding mechanisms of infection, host:mycoparasite recognition, and host specificity. In order to address these issues, it was necessary to visually document (1) the initial colonization phase using SEM, (2) the time course of infection in which the introduced putative mycoparasite could be identified with certainty following optical sectioning using CLSM, and (3) ultrastructural effects of the putative mycoparasite on urediniospores using TEM. Taken together, these microscopic observations would provide clear evidence for mycoparasitism. 2. Materials and methods 2.1. Urediniospore collection Soybean plants were grown at the Louisiana State University Agricultural Center’s Ben Hur Research Farm near Baton Rouge, LA, in 2009 as described previously (Ward and Schneider, 2012; Ward et al., 2011). Plants were infected with naturally occurring inoculum of P. pachirhizi. Urediniospores were collected with a spore collector (G&R Manufacturing, Manhattan, KS) retrofitted with a hand-held battery powered vacuum (Craftsman, Sears, Hoffman Estates, IL) in October and November 2009. Urediniospores were collected into 20 ml scintillation vials and stored at 80 °C until use. Upon removal from storage, approximately 5 mg urediniospores were rehydrated by placing them in an uncovered petri dish lid in a 15 cm glass desiccator with 2 cm distilled water in the bottom (Bonde et al., 2006; Bromfield, 1964). Urediniospores were rehydrated for 24 h, and germination was assessed by placing them on 1.5% water agar for 4 h in the dark at 25 °C (Twizeyimana and Hartman, 2010). Urediniospores were considered germinated when germ tubes were at least as long as the length of the urediniospore. Urediniospore collections with germination rates above 80% were used in all experiments. 2.2. Collection of isolates of S. lanosonivem Isolates of S. lanosonivem were collected in 2007 from soybean plants near Baton Rouge, Louisiana and Quincy, Florida that were infected with P. pachyrhizi. The colonist was isolated from detached, diseased leaves as described previously, and cultures were maintained on cornmeal agar and stored in the dark at 4 °C until use (Ward et al., 2011). In previous experiments, isolate D082307-2A was determined to be antagonistic under laboratory and field conditions; therefore, it was used in all experiments described here (30). 2.3. Scanning electron microscopy SEM was used to examine the topography of colonized sori, to determine whether urediniospores remained intact, and to confirm that hyphae were confined to sori. Suspensions of urediniospore of P. pachyrhizi were prepared by suspending rehydrated spores in 0.01% Tween 20 and adjusting to 106 spores ml 1 with the aid of a hemacytometer. Inoculum of
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
S. lanosoniveum was prepared from 14-day-old cultures on potato dextrose agar (PDA). Dishes were flooded with sterile phosphate buffer (0.5 mM, pH 7.1) and rubbed with a glass rod to dislodge conidia. Conidial suspensions were diluted in 0.01% Tween 20 and adjusted to 106 spores ml 1 using a hemacytometer (Ward et al., 2011). Soybean cultivar Asgrow 6202 (Monsanto; St. Louis, MO) was grown outdoors in 22 cm diameter nursery containers in sterile potting mix (Miracle Gro; Scotts Co., Marysville, OH). At least 50 leaves were harvested after flowering and placed in 10 23 25 cm clear polystyrene boxes lined with moist paper towels. Suspensions of urediniospores were spread on the abaxial surface of leaflets at 1 ml per leaf with a glass rod. Leaves were allowed to dry (approximately one hour), and then moist chambers were closed and incubated at 25 °C under a 12-h photoperiod (fluorescent lights, 850–1000 lux) as described previously (Ward et al., 2011). Sori developed within 7–10 days. On day 14, conidial suspensions of S. lanosoniveum were inoculated onto leaf surfaces at 1 ml per leaf and spread with a glass rod. On days 5, 7, 10, and 14 after inoculation with S. lanosoniveum, colonized sori were cut from leaves and fixed in formalin-acetic acid-alcohol (FAA) overnight (Ruzin, 1999; Ward et al., 2011). After dehydration in an ethanol series culminating in 100% EtOH, samples were critical point dried and affixed to stubs (Sass, 1958; Ward et al., 2011). Stubs were sputter-coated with 60:40 gold:palladium. Samples were viewed and photographed with an SEM (JEOL JSM-6610; JEOL; Tokyo, Japan). Experiments were repeated twice. 2.4. Confocal microscopy Because internal structures of colonized urediniospores could not be visualized using SEM, optical sectioning with a confocal microscope was used to view both inner and outer surfaces of urediniospores colonized by a green fluorescent protein (GFP)transformed isolate of S. lanosoniveum. This microscope, along with the transformed isolate, allowed us to determine the proportion of urediniospores that became colonized during co-cultivation and to document the time course of colonization events. For visualization with the confocal microscope, S. lanosoniveum isolate D082307-2A was transformed with GFP as follows. The fungus was grown on PDA at 25 °C. It was initially tested for sensitivity to the antibiotic geneticin (Sigma, St. Louis, MO) at different concentrations (up to 15,000 lg/ml). Growth was completely inhibited at 15,000 lg/ml, and this concentration was used to select transformants. Transformants were generated using Agrobacterium tumefaciens strain EHA105 harboring the pSK2265 TDNA binary vector (Maruthachalam and Subbarao, unpublished). Conidia of the individual transformants were harvested, resuspended in sterile distilled water, and plated on water agar to obtain monoconidial cultures. A single conidium from each of the transformants was picked and transferred to a petri dish containing PDA. The fluorescence of ZsGreen from all the transformants was confirmed with a compound microscope (Olympus BX60) equipped with a GFP filter (450–490 nm excitation, 500 nm emission). Conidia of transformants were stored in glycerol at 80 °C and regrown on PDA for laboratory experiments. About 50 transformants were tested for their ability to colonize sori, and one isolate with high GFP signal, D082307-2A-GFP15, was selected for further studies. S. lanosoniveum isolate D082307-2A-GFP15 was grown on 1% cornmeal agar topped with sterile Whatman #10 filter paper by spreading 1 ml of a 106 conidial suspension over the filter paper as described below. Urediniospores of P. pachyrhizi were removed from storage and rehydrated as described above, and urediniospores with germination rates of at least 80% were used in these
89
studies. Approximately one mg of dry urediniospores was spread across a 5 cm2 area on 14-day-old filter paper cultures of S. lanosoniveum D082307-2A-GFP15. About 50 randomly selected urediniospores were picked from the filter paper-covered dish with a needle, placed on microscope slides and examined immediately. Samples were viewed with the confocal microscope every 12 h for 5 days after incubation as described below. This experiment was repeated twice. Confocal images were generated using a Leica TCS SP2 (Leica Microsystems; Wetzlar, Germany) confocal laser scanning microscope with a 63 1.4 NA Apo lens. A 488 nm krypton/argon laser was used to excite both the GFP in the hyphae of S. lanosoniveum and the autofluorescent compounds in the urediniospores of P. pachyrhizi. GFP images were obtained at the 510–535 nm emission spectra, and images of the red autofluorescence from the urediniospores were made using the 550–650 nm spectra. Images were processed using the Leica TCS software. 2.5. Transmission electron microscopy TEM was used to examine whether the fungus directly penetrated urediniospore walls, entered through germ pores, or degraded outer membranes of urediniospores before colonization and infection. Furthermore, subcellular membranes were examined after colonization of urediniospores by S. lanosoniveum. Filter paper cultures were prepared as described above. After 1–5 days, filter paper sections with urediniospores were cut into 1 mm squares and prepared as described below. Controls consisted of urediniospores spread onto cornmeal agar topped with filter paper as described above except that the cultures did not include S. lanosoniveum. Samples were fixed, embedded, and stained as described below. Fixation was in formalin-acetic acid-alcohol (FAA) for 4 h, followed by four washings in cacodylate buffer, and rinsed overnight. Fixation in OsO4 followed for 2 h, and samples were washed twice each in cacodylate buffer and distilled water. Samples were stained with uranyl acetate blocking stain for 1 h, and then washed twice with distilled water. Samples were then dehydrated in an EtOH series ending in 100% EtOH. Infiltration with LR White resin culminated with 100% resin, and then samples were embedded in 100% LR White resin and polymerized overnight at 70 °C. Embedded samples were sectioned to obtain 1.5-lm-thick sections using an ultramicrotome equipped with a diamond knife. Ultra-thin sections were viewed with a TEM (JEOL 100-CX TEM, Peabody, MA) at an acceleration voltage of 80 kV. This experiment was repeated twice. 3. Results 3.1. Scanning electron microscopy Hyphae of S. lanosoniveum traversed leaf surfaces and colonized rust sori, but they failed to colonize leaf surfaces that were free of sori. Within 3 days, hyphae ramified throughout sori and coiled tightly around individual urediniospores (Fig. 1A). Most urediniospores were collapsed within 5 days. Possible penetration points were composed of numerous hyphae growing together in a radial pattern, and these structures resembled appressoria (Fig. 1B). The putative mycoparasite produced an amorphous material between hyphae and urediniospores that may have facilitated their attachment to urediniospores. Five days after inoculation, sunken germ pores were observed, and after 14 days, there were numerous holes in apparently degraded urediniospores (not shown). S. lanosoniveum produced conidia from hyphae as early as 3 days after inoculation, and after 7–10 days, conidia were produced on long single phialides emerging from collapsed urediniospores (Fig. 2). Urediniospores not
90
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
Fig. 3. Confocal microscope image of a urediniospore of Phakopsora pachyrhizi colonized by an isolate of Simplicillium lanosoniveum that was transformed with the green fluorescent protein gene. The antagonist ramified throughout the urediniospore within 3 days after co-culture. Hypha (H) of S. lanosoniveum that penetrated the spore wall through a germ pore of a urediniospore (U) of P. pachyrhizi.
Fig. 1. Scanning electron microscope view of (A) hyphae of Simplicillium lanosoniveum coiled around urediniospores of Phakopsora pachyrhizi after 3 days and (B) putative appresorium-like penetration sites produced by Simplicillium lanosoniveum after 7 days. Conidia (C) and hyphae (H) of S. lanosoniveum, urediniospore (U) of P. pachyrhizi.
of 514 nm, and this facilitated the observation that urediniospore walls had thinned over germ pores by 3 days after inoculation (not shown). Hyphae of S. lanosoniveum frequently entered or emerged through germ pores (Fig. 4). Hyphae branched extensively upon entering urediniospores where they formed tight coils. Three-day-old co-inoculations revealed heavy infection with over 60% of urediniospores colonized by the antagonist. By day five, over 90% of urediniospores were colonized with GFP-transformed S. lanosoniveum. Urediniospores not exposed to S. lanosoniveum did not have breaks in spore coats, and no green fluorescence was observed (not shown). 3.3. Transmission electron microscopy
Fig. 2. Scanning electron microscope view of conidia of Simplicillium lanosoniveum produced on long phialides extending from collapsed urediniospores of Phakopsora pachyrhizi. Conidia (C) of S. lanosoniveum, (U) urediniospore of P. pachyrhizi.
exposed to S. lanosoniveum appeared turgid, intact, and there was no hyphal colonization of sori (not shown). 3.2. Confocal microscopy S. lanosoniveum isolate D082307-2A-GFP15 colonized urediniospores 3 days after co-culture with P. pachyrhizi, and hyphae subsequently ramified throughout uredinia (Fig. 3). While hyphae coiled around urediniospores, they did not appear to constrict urediniospores. Urediniospores autofluoresced red at a wavelength
Ultra-thin sections of infected urediniospores revealed sites of penetration and degradation within urediniospores. One day after co-inoculation, urediniospore organelles were not recognizable, and cytoplasm was restricted to undefined masses of granulated material (Fig. 5). Germ pores were often visible as a thinning spore wall in as little as 24 h after co-inoculation with S. lanosoniveum. Degradation of germ pores was observed as hyphae of S. lanosoniveum entered urediniospores through germ pores, and some urediniospores contained multiple bulging germ pores (Fig. 6A). In these urediniospores, cytoplasm was granulated and cellular contents were disrupted. This condition was in contrast to germinating urediniospores in the absence of S. lanosoniveum in which organelles aggregated proximal to germ tubes (Fig. 6B). As time elapsed, S. lanosoniveum continued to colonize urediniospores of P. pachyrhizi. On day two, urediniospores appeared to be evacuolated. After 3 days, co-inoculated urediniospores were heavily colonized by S. lanosoniveum. Ultra-thin sections revealed that hyphae of S. lanosoniveum entered and exited germ pores, and organelles were completely absent (Fig. 7). On day 10, all urediniospores were emptied of their contents, intact organelles were not apparent, and there was a near absence of hyphae of S. lanosoniveum. Urediniospores were contorted and collapsed (Fig. 8). Throughout the infection and colonization process, the integrity of the urediniospore wall was preserved until spores were completely emptied of their contents (5–10 days). Urediniospores not
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
91
Fig. 4. Corresponding confocal microscope (B & D) and brightfield images (A & C) of urediniospores of Phakopsora pachyrhizi colonized by Simplicillium lanosoniveum 3 days after co-culture. Confocal images consist of 50–60 focal planes through a section approximately 20 lm thick. After 3 days, hyphae (H) of S. lanosoniveum entered and exited germ pores (G) of P. pachyrhizi. Putative penetration site (P), urediniospores (U) of P. pachyrhizi.
exposed to S. lanosoniveum were normally shaped and were surrounded by a thick spore wall (Fig. 9). All organelles appeared intact, and germ pores were not sunken and their walls were relatively thick. There were no bulging germ pores and no indication of altered turgor pressure. Additionally, urediniospores that were not exposed to S. lanosoniveum germinated within 4–6 h, while inoculated urediniospores failed to germinate.
4. Discussion In previous studies, we showed that S. lanosoniveum aggressively colonized soybean pustules following directed hyphal growth across leaf surfaces toward pustules (Ward et al., 2011). Hyphae of S. lanosoniveum then intertwined among urediniospores resulting in substantially reduced germination rates. Furthermore, we showed that S. lanosoniveum effectively suppressed development of soybean rust under field conditions following foliar applications of conidial suspensions (Ward and Schneider, 2012). However, because of the limitations imposed by light microscopy, we did not determine if S. lanosoniveum was mycopar-
asitic, or if it inhibited urediniospore germination by other means. Documenting a mycoparasitic mode of action has implications for further research such as host:mycoparasite recognition and host specificity. Mycoparasitism, defined by Alexopoulos et al. (1) as parasitism of a fungus by another fungus, was clearly documented in this study. S. lanosoniveum infected urediniospores of P. pachyrhizi within 24 h after co-inoculation, and parasitized urediniospores were not viable. To our knowledge, this is the first report of mycoparasitism of any rust fungus by S. lanosoniveum. Previous reports documented rust pathogens parasitized by other fungi. Examples include Cladosporium uredinicola on Puccinia spp., Eudarluca caricis on species of Puccinia and Melampsora and Verticillium psalliotae on P. pachyrhizi (Nischwitz et al., 2005; Saksirirat and Hoppe, 1990; Yuan et al., 1999). The appresorium-like structures observed in this study were similar to putative penetration sites formed when Cladosporium parasitized the bean rust pathogen (Uromyces appendiculatus) and the western gall rust pathogen (Endocronartium harknessii), as observed in previous studies (Assante et al., 2004; Tsuneda and Hiratsuka, 1979). However, no indentations or impressions were
92
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
Fig. 5. Transmission electron microscope views of urediniospores of Phakopsora pachyrhizi 1 day after co-culturing with Simplicillium lanosoniveum. Degradation of germ pores is indicated by white arrows. Cytoplasm often appeared granulated and hyphae of S. lanosoniveum were often present. Hyphae (H) of S. lanosoniveum entering germ pores (G) of urediniospores (U) of P. pachyrhizi.
Fig. 7. Transmission electron microscope view of a urediniospore (U) of Phakopsora pachyrhizi 3-days after co-culturing with Simplicillium lanosoniveum. Urediniospores were colonized with hyphae of S. lanosoniveum, while organelles were degraded and cytoplasm was aggregated or often nonexistent. Possible adhesive matrix (A) and hyphae (H) of S. lanosoniveum.
Fig. 8. Transmission electron microscope view of a urediniospore (U) of Phakopsora pachyrhizi 10-days after co-culturing with Simplicillium lanosoniveum. Urediniospores were misshapen, and less hyphae of the antagonist was observed as compared to observations earlier in the infection process. Possible adhesive matrix (A) and hyphae (H) of S. lanosoniveum.
Fig. 6. Transmission electron micrographs of (A) urediniospores of Phakopsora pachyrhizi with apparently altered turgor pressure as evidenced by the bulging germ pore (G) after 24 h exposure to Simplicillium lanosoniveum. (B) Germinating urediniospore in the absence of S. lanosoniveum with a germ tube (T).
Fig. 9. Urediniospore of Phakopsora pachyrhizi not exposed to Simplicillium. lanosoniveum. (N) Nucleus, (V) Vacuole.
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
observed in this study, which suggests that mechanical pressure was not applied by S. lanosoniveum, as was observed in previous studies (2, 23). The confocal microscopic images clearly showed that S. lanosoniveum quickly penetrated and colonized more than 60% of urediniospores within 3 days. This confirms our previous work with field-grown soybeans in which we monitored the infection process using quantitative real time PCR (27). This rapid infection process clearly has the potential to reduce secondary inoculum availability and to greatly suppress the rate of epidemic development. While germ pores of urediniospores of P. pachyrhizi were degraded by S. lanosoniveum, organelles within urediniospores also disintegrated. Cytoplasm of urediniospores of P. pachyrhizi aggregated, and cellular contents were unidentifiable 6 h after co-inoculation before significant amounts of hyphae of S. lanosoniveum were observed within urediniospores. This phenomenon is similar to mycoparasitism of Uromyces dianthi by Verticillium lecanii (currently classified as Lecanicillium lecanii) (Spencer, 1980; Spencer and Atkey, 1981). Lecanicillium lecanii is a member of the Cordyciptaceae family, which also includes Simplicillium, as well as Beauveria, Cordyceps, and Isaria species (Sung et al., 2007; Zare and Gams, 2001). Verticillium lecanii either directly penetrated germ pores of urediniospores of U. dianthi or penetrated spore walls; these penetration events were preceded by enzymatic degradation (22, 23). Other studies reported that V. lecanii was an effective mycoparasite against Puccinia horiana on greenhouse chrysanthemums, and that it was hyperparasitic on Sphaerotheca fuliginea, the causal agent of powdery mildew on greenhousegrown cucumber (Askary et al., 1997). Another member of the Cordycipitaceae family, Lecanicillium psalliotae (Treschew) Zare & W. Gams) (formerly Verticillium psalliotae), was reported to degrade urediniospores of P. pachyrhizi by means of b-glucanase, chitinase, and protease (19, 20). In most interactions in the present study, there were indications that cell-wall degrading enzymes may have aided the pathogen in entering germ pores of urediniospores of P. pachyrhizi. For example, degradation of germ pore material adjacent to hyphae of S. lanosoniveum was often observed. Fibrous material between conidia of S. lanosoniveum and urediniospores of P. pachyrhizi, which resembled the adhesive material reported by others, was often observed (2, 5, 14). Studies by others suggested that adhesive material may have allowed close contact of the antagonist to urediniospores and provided the physical means by which lytic enzymes or toxic metabolites attacked the host at the entry sites (Askary et al., 1997; Benhamou et al., 1999; Moricca et al., 2001). Assante et al. (2004) determined that this enzymatic degradation of cell wall polysaccharides and breakage of their secondary linkages with polymeric chitin microfibrils must occur in order to supply Cladosporium tenuissimum with nutrients during the first steps of the interaction. Future work should examine this material to determine whether it is the same as those studied by others. Additionally, this may be an indication of whether conidia of S. lanosoniveum adhere to urediniospores when the latter disperse. This would be an effective mechanism by which the mycoparasite could spread long distances while still maintaining a nutrient source. Outer surfaces of urediniospores were examined to assess the integrity of spore walls following colonization by S. lanosoniveum. Spore coats remained rigid during colonization, and indentations or collapsed areas of urediniospores of P. pachyrhizi may have resulted from penetration pegs during infection. However, urediniospores did not collapse until contents were completely consumed. Occasionally, however, bulging germ pores were observed. We speculate that this may have been caused by disruption in osmoregulatory processes leading to increases in turgor pressure. In summary, these results clearly demonstrated for the first time that S. lanosoniveum is a mycoparasite of P. pachyrhizi. These
93
findings support our contention that S. lanosoniveum may be an effective biological control agent in that it quickly colonized pustules and parasitized a high percentage of urediniospores. Resulting parasitized urediniospores were not viable and did not germinate. These discoveries could be exploited in both conventional and organic soybean production systems to control soybean rust. Aside from direct application of S. lanosoniveum as a biological control agent, we need to determine the roles that this organism and other naturally occurring mycoparasites have on disease development on a local and regional basis. For example, it is possible that fungicides applied to manage other foliar diseases of soybean, such as those caused by Cercospora spp., may inadvertently reduce populations of mycoparasites such that incidence and severity of soybean rust may be enhanced. Another intriguing question concerns the source of inoculum of S. lanosoniveum. Is it disseminated with urediniospores of the pathogen, or do rust pathogens of other host plant species serve as reservoirs? Acknowledgments The authors would like to thank Periasamy Chitrampalam of the University of California – Davis, Salinas, CA for his assistance and insight into the transformation of S. lanosoniveum with GFP. References Askary, H., Benhamou, N., Brodeur, J., 1997. Ultrastructural and cytochemical investigations of the antagonistic effect of Verticillium lecanii on cucumber powdery mildew. Phytopathology 87, 359–368. Assante, G., MaffiI, D., Saracchi, M., Farina, G., Moricca, S., Ragazzi, A., 2004. Histological studies on the mycoparasitism of Cladosporium tenuissimum on urediniospores of Uromyces appendiculatus. Mycol. Res. 108, 170–182. Benhamou, N., Rey, P., Picard, K., Tirilly, Y., 1999. Ultrastructural and cytochemical aspects of the interaction between the mycoparasite Pythium oligandrum and soilborne plant pathogens. Phytopathology 89, 506–517. Bischoff, J.F., White, J.F., 2004. Torrubiella piperis sp nov (Clavicipitaceae, Hypocreales), a new teleomorph of the Lecanicillium complex. Stud. Mycol., 89–94. Bonde, M.R., Nester, S.E., Austin, C.N., Stone, C.L., Frederick, R.D., Hartman, G.L., Miles, M.R., 2006. Evaluation of virulence of Phakopsora pachyrhizi and P. meibomiae isolates. Plant Dis. 90, 708–716. Bromfield, K.R., 1964. Cold-induced dormancy and its reversal of uredospores of Puccinia graminis var. tritici. Phytopathology 54, 68–74. Bromfield, K.R., 1984. Soybean Rust, Monograph No. 11. American Phytopathological Society, St. Paul, MN. Centralbureau voor Schimmelcultures, C., Fungal Biodiversity Centre. Dorrance, A.E., Hershman, D.E., Draper, M.A., 2008. Economic Importance of Soybean Rust. In: Dorrance, A.E., Draper, M.A., Hershman, D.E. (Eds.), Using Foliar Fungicides to Manage Soybean Rust. The Ohio State University. Humber, R.A., 2007. Recent Phylogenetically Based Reclassifications of Fungal Pathogens of Invertebrates. Society for Invertebrate Pathology. Moricca, S., Ragazzi, A., Mitchelson, K.R., Assante, G., 2001. Antagonism of the twoneedle pine stem rust fungi Cronartium flaccidum and Peridermium pini by Cladosporium tenuissimum in vitro and in planta. Phytopathology 91, 457–468. Nischwitz, C., Newcombe, G., Anderson, C.L., 2005. Host specialization of the mycoparasite Eudarluca caricis and its evolutionary relationship to Ampelomyces. Mycol. Res. 109, 421–428. Pivonia, S., Yang, X.B., 2004. Assessment of the potential year-round establishment of soybean rust throughout the world. Plant Dis. 88, 523–529. Rupe, J., Sconyers, L., 2008. Soybean Rust. The Plant Health Instructor. Ruzin, S.E., 1999. Plant Microtechnique and Microscopy. Osford University Press Inc, New York. Saksirirat, W., Hoppe, H.H., 1990. Light- and scanning electron microscopic studies on the development of the mycoparasite Verticillium psalliotae Treschow on uredospores of the soybean rust (Phakopsora pachyrhizi Syd.). J. Phytopathol. 128, 340–344. Sass, J.E., 1958. Botanical Microtechnique. The Iowa State University Press, Ames, IA. Schneider, R.W., Hollier, C.A., Whitam, H.K., Palm, M.E., McKemy, J.M., Hernández, J.R., Levy, L., DeVries-Paterson, R., 2005. First report of soybean rust caused by Phakopsora pachyrhizi in the continental United States. Plant Dis. 89, 774. Spencer, D.M., 1980. Parasitism of carnation rust (Uromyces dianthi) by Verticillium lecanii. Trans. Br. Mycol. Soc. 74, 191–194. Spencer, D.M., Atkey, P.T., 1981. Parasitic effects of Verticillium lecanii on two rust fungi. Trans. Br. Mycol. Soc. 77, 535–542. Sung, G., Hywel-Jones, N., Sung, J., Luangsa-ard, J.J., Shrestha, B., Spatafora, J.W., 2007. Phylogenetic classification of Cordyceps and the clavicipitaceous fungi. Stud. Mycol. 57, 5–59.
94
N.W. Gauthier et al. / Biological Control 76 (2014) 87–94
Tsuneda, A., Hiratsuka, Y., 1979. Mode of parasitism of a mycoparasite, Cladosporium gallicola, on western gall rust, Ednocronartium harknessii. Can. J. Plant Pathol. 1, 31–36. Twizeyimana, M., Hartman, G.L., 2010. Culturing Phakopsora pachyrhizi on detached Leaves and urediniospore survival at different temperatures and relative humidities. Plant Dis. 94, 1453–1460. Walker, D.R., Boerma, H.R., Phillips, D.V., Schneider, R.W., Buckley, J.B., Shipe, E.R., Mueller, J.D., Weaver, D.B., Sikora, E.J., Moore, S.H., Hartman, G.L., Miles, M.R., Harris, D.K., Wright, D.L., Marois, J.J., Nelson, R.L., 2011. Evaluation of USDA soybean germplasm accessions for resistance to soybean rust in the southern United States. Crop Sci. 51, 678–693. Ward, N.A., 2009. Mycoparasitism of Phakopsora pachyrhizi by Simplicillium lanosoniveum. 2009 National Soybean Rust Symposium. Plant Management Network International.
Ward, N.A., Schneider, R.W., 2012. Effects of Simplicillium lanosoniveum on Phakopsora pachyrhizi, the soybean rust pathogen, and its use as a biological control agent. Phytopathology 102, 749–760. Ward, N.A., Schneider, R.W., Aime, M.C., 2011. Colonization of soybean rust sori by Simplicillium lanosoniveum. Fungal Ecol. 4, 303–308. Yuan, Z.W., Pei, M.H., Hunter, T., Ruiz, C., Royle, D.J., 1999. Pathogenicity to willow rust, Melampsora epitea, of the mycoparasite Sphaerellopsis filum from different sources. Mycol. Res. 103, 509–512. Zare, R., Gams, W., 2001. A revision of Verticillium section Prostrata. IV. The genera Lecanicillium and Simplicillium gen. nov. Nova Hedwigia 73, 1–50.
