Rhizoctonia solani AG-1 IA infects both rice and signalgrass in the Colombian Llanos 1

Signalgrass (Urochloa spp.) foliar blight and collar rot diseases are caused by the Basidiomycetes Rhizoctonia solani (sexual stage Thanatephorus cucumeris) fungus, which emerged early in 1990 as an important pathogen for Urochloa pastures in the eastern Colombian Llanos (Ciat 1993, Argel et al. 2005, Alvarez et al. 2013). The emergence of these diseases was first reported in areas previously occupied by rice, a highly susceptible host (Lee & ABSTRACT RESUMO


PALAVRAS-CHAVE
In the south of the United States, Urochloa spp. was reported as host for two distinct AGs of the R. solani species complex: AG-1 IA and AG-1 IB (Black et al. 1996).A recent description from 2013 indicated a prevalence of R. solani AG-1 IA infecting pastures of Urochloa (69 % of total samples) in the warm regions of Colombia: Casane and Córdoba.On the other hand, Rhizoctonia sp.AG-D (Ceratobasidium sp.) was more commonly found in regions with lower temperatures such as Cauca (31 % of the samples) (Alvarez et al. 2013).
It is possible that different pathogens from the Rhizoctonia complex could be associated with foliar blight and death of Urochloa in the Colombian Llanos, especially in areas close to rice fields.
The management of Urochloa foliar blight and collar rot and rice sheath blight complexes relies heavily on varietal resistance (Nunes et al. 2004).Therefore, it is important to understand the relative importance of each Rhizoctonia species/groups, in order to improve disease management.
This study aimed at determining which pathogens from the Rhizoctonia species complex are associated with foliar and sheath blight in Urochloa spp.and rice, in the Colombian Llanos.

MATERIAL AND METHODS
Population samples of infected Urochloa and rice were collected between 2010 and 2011, in the Meta State, eastern Colombian Llanos, in Colombia (Figure 1).Samples were collected in Puerto López, from Urochloa brizantha cv.'Toledo' (BBT1) and hybrid Urochloa cv.'Mulato' (BHM3) with foliar blight symptoms, and in adjacent areas from 'Fedearroz 50' (OS5) rice plants with sheath blight or spot symptoms (Figure 1, Table 1).The transect sampling system, with six lines in each cropping area and seven to eight sampling points per line, was used.Employing the same system, the following hosts with disease symptoms were sampled in the Villavicencio county: U. brizantha cv.'Toledo' (BBT2), hybrid Urochloa cv.'Mulato' (BHM4) and rice cv.'Thailandia' (OS6) (Figure 1, Table 1).Isolations were initially performed by disinfecting leaf fragments in sodium hypochlorite 1 %, followed by transference to PDA media (potato-dextrose-agar, Himédia) with 50 mg mL -1 of chloramphenicol and streptomycin.Plates were kept at 25 ºC in the dark for up to 48 h.Pure cultures were isolated after transference of typical Rhizoctonia-like mycelia to new PDA plates.
For DNA extraction, fungal mycelium was grown in 30 mL of potato dextrose broth (18.5 g L -1 ), for 5 days, on a shaker at 75 rpm, after which the mycelium was collected by filtration and freeze-dried for approximately 48 h.DNA was extracted using a GenElute kit (Sigma-Aldrich), according to the manufacturer's instructions.
The ribosomal DNA ITS-5.8S of 64 sampled fungi was sequenced using the primers ITS4 and ITS5 (Johanson et al. 1998, Matsumoto 2002).PCR product sequence was performed by Macrogen (South Korea).Sequence data from both ends of each amplicon (one forward and one reverse per isolate) were assembled, aligned and concatenated using Chromas (Technelysium, Australia).Phylogenetic analyses were performed through neighbor-joining (NJ) inference and the Kimura model for genetic distance using MEGA v.5.0 (Tamura et al. 2011).The statistical support for each clade in the obtained tree was tested using bootstrap resampling with 1,500 permutations.
For the cross-pathogenicity study, 24 R. solani AG-1 IA isolates were selected, 12 from Urochloa spp.and 12 from rice.Nine R. oryzae-sativae isolates and seven S. hydrophilum isolates from rice were also included, totaling 40 isolates.The rice cultivar 'Fedearroz' and U. brizantha cv.'Toledo' were used as hosts, and uninoculated control plants were included in the assay.Seeds were surface-sterilized with 1 % sodium hypochlorite for 20 min prior to sowing.Four seeds were sown in each pot filled with a soil and sand mix (2:1).After seedling emergence, the pots were thinned to one seedling per pot.Plants were fertilized with 1 g of NPK (15:15:15) granular fertilizer.The pots were watered daily and kept in a greenhouse under natural conditions until inoculation.
Inoculation with R. solani sclerotia, R. oryzaesativae or S. hydrophilum mycelium plug (6 mm) was performed when the plants had four leaves.The inoculum was applied to the base of the last or second-to-last leaf of the main tiller and attached with parafilm.The infected plants were kept in a phytotron in high humidity (95 %), with daytime temperatures maintained between 25 ºC and 27 ºC.Evaluation was performed 6 days after inoculation of rice and 12 days after inoculation of Urochloa by measuring the maximum tiller length and the lesion length on the tiller.A disease index was calculated according to the following equation: 9 x (length of the lesion on the tiller/maximum tiller length) (Jia et al. 2007).The rating of the experimental unit was the combined scores from up to four tillers.To detect evidence of host specialization, analysis of variance followed by a a N = sample size (number of fungi isolates obtained from symptom association in each sampled species); b characterization based upon PCR detection.priori contrast analysis between groups of isolates was performed using SAS 9.1 (SAS System for Windows, SAS Institute, Cary, NC).The statistical analyses were performed independently for each experiment, being completely randomized, with three replications each.

RESULTS AND DISCUSSION
The 234 isolates obtained are described in Table 1.Based on cultural and morphological characteristics, isolates from six plant populations with visual symptoms of foliar blight and death of Urochloa and rice sheath spot and blight were classified in three distinct groups (Table 1).Isolates in the first group showed sasakii-type sclerotia on light brown-colored cultures (Figure 1B), similar to R. solani AG-1 IA (Sneh et al. 1996, Yang et al. 1989).This group contains 161 isolates from Urochloa spp.and 13 from rice plants.
A similar report was made by Alvarez et al. (2013), who found that R. solani AG-1 IA was the predominant pathogen associated with foliar blight in U. brizantha, U. brizantha cv.'Toledo', U. decumbens, U. mutica, Urochloa cv.'Mulato' and 'Mulato II', in Colombia.The second group includes 52 isolates obtained only from rice, with small and irregular-shaped sclerotia and with light brown mycelium (Figure 1C), similar to R. oryzae-sativae (Lanoiselet et al. 2007).The last group consists of eight isolates obtained from rice, showing small round sclerotia with a dark brown to black color (Figure 1D), like S. hydrophilum (Gutiérrez 2007).
The 161 fungi isolates obtained from symptomatic Urochloa app.and 13 isolates obtained from sheath lesions in rice resulted in positive amplification with primers specific to R. solani AG-1 IA.Another 52 isolates from rice plants were positively amplified with primers specific to R. oryzae-sativae, and the remaining eight isolates from the third group did not exhibit amplification for any of the Rhizoctonia specific molecular primers used.
Phylogenetic inferences were based on rDNA ITS sequences of isolates from the three groups (Figure 2).From the first group, 55 isolates (represented by 9 distinct haplotypes in the superior phylogenetic tree) had similar sequences to R. solani AG-1 IA sequences.These isolates are composed of individuals from the populations BBT1 and BBT2 (15 isolates), BHM3 and BHM4 (29 isolates), and OS5 and OS6 (11 isolates).The second biggest group, whose sequences were similar to R. oryzae-sativae, was composed of four isolates from the OS5 and OS6 populations.Finally, a third group of five isolates from the OS5 and OS6 populations showed similar sequences to S. hydrophilum.The phylogenetic inference corroborates the initial identification performed by morphologic characteristics and specific PCR detection.This is the first report of R. oryzae-sativae and S. hydrophilum associated with the sheath blight   1983, Sayler & Yang 2007) and Venezuela (Cedeño et al. 1996).Specifically, R. solani AG-1 IA is associated with sheath blight, R. circinata var.oryzae with sheath spot, R. oryzae-sativae with aggregated sheath spot, S. hydrophilum with sheath spot and S. oryzae with sheath rot (Gutiérrez 2007).The overlapping of similar symptoms reduces assertive diagnosis for these rice sheath diseases (Cedeño et al. 1996, Johanson et al. 1998).As diagnosis is a highly relevant step for establishing proper management strategies, especially with varietal resistance, simple methods are recommended, such as morphological characterization, to identify the associated pathogens.
These findings support the hypothesis that R. solani AG-1 IA isolates infecting Urochloa spp.most likely originated from a population that originally infected rice via host shift (Chavarro Mesa et al. 2015).Similar examples are known to have occurred, such as host jump (from rice to soybean) or host shift (from rice to maize) of R. solani AG-1 IA populations in the southern United States (Assis et al. 2008) andVenezuela (González-Vera et al. 2010).
A recent study indicated that R. solani AG-1 IA isolates from Urochloa (syn.Brachiaria) were also pathogenic to cowpea and soybean (Chavarro Mesa et al. 2015).These results provide evidence that the R. solani AG-1 IA populations adapted to Urochloa spp.most likely are not genetically structured, and therefore keep a wide host range, extending also to the Fabaceae family.3. Isolates of R. solani AG-1 IA from rice or Urochloa are cross pathogenic in both plant species.4. R. oryzae-sativae and S. hydrophilum were detected only in rice, but R. oryzae-sativae could produce very light symptoms in Urochloa.

Figure 1 .
Figure 1.Geographic locations of Urochloa spp.and rice sampled in Colombia, morphological differences among fungi and the disease symptoms of foliar blight and sheath spot or blight.(A) Geographic location of sampled areas close to Puerto López and Villavicencio, in Meta; (B) sasakii-type sclerotia of Rhizoctonia solani AG-1 IA; (C) sclerotia of R. oryzae-sativae; (D) sclerotia of Sclerotium hydrophilum; (E) foliar blight symptoms in Urochloa brizantha cv.'Toledo' related to R. solani AG-1 IA; (F and G) symptoms of rice sheath spot related to R. oryzae-sativae and Sclerotium hydrophilum.

Figure 2 .
Figure 2. Phylogenetic trees, based on neighbor-joining distance, illustrating the relation among ITS-5.8Shaplotypes of Rhizoctonia solani AG-1 IA, R. oryzae-sativae and Sclerotium hydrophilum.The bars indicate genetic distance among clades.Supporting values of bootstrap above 50 % are by each branch.The bold case names are sequences downloaded from NCBI/Genbank, to serve as control, of known species and anastomosis group, to aid classification.

Figure 3 .
Figure 3. Aggressiveness of Rhizoctonia solani AG-1 IA, R. oryzae-sativae and Sclerotium hydrophilum groups of isolates from the Colombian Llanos to Urochloa and rice.Means followed by the same letter are not significantly different by contrast of means (p ≤ 0.01).

Table 1 .
Populations of Urochloa spp.and rice associated with foliar fungi pathogens.