Rhizoctonia species are basidiomycete fungi associated with important agricultural and horticultural crops grown all over the world. These fungi are able to colonize and/or infect seeds, roots, leaves, stems and fruits of many crops. Rhizoctonia spp. comprise an important multinucleate species complex represented by R. solani and several others which are either mononucleates or binucleates. Up to now, there are 14 anastomosis groups (AG) of R. solani described: AG-1 to AG-13 and AG-BI (Carling 2000). Binucleate Rhizoctonia species are grouped either in seven AGs (CAG1 to CAG7) in the USA (Burpee et al. 1980) or 19 AGs (AGA to AGS) around the world (Ogoshi 1987).

While most of the R. solani AGs are recognized as important plant pathogens, the binucleate Rhizoctonia species are mostly considered as mycorrhizal or having a role as biocontrol agents (Burpee and Goulty 1984; Cardoso and Echandi 1987a, b; Herr 1988; Escanbde and Echandi 1991; Harris and Adkins 1999). Thus, only a few of the known binucleate Rhizoctonia spp. AG are pathogenic. Non-pathogenic binucleate Rhizoctonia spp. are commonly found in soil organic matter or in plant debris. In a few Brazilian fields sampled, they were mostly associated with beans, peanuts (Ceresini and Souza 1997) and soybeans (Fenille et al. 2002) primarily infected by R. solani AG 4 HG-I or AG 2.2 IIIB, causing damping-off and root rot symptoms. Most of these binucleate Rhizoctonia spp. were non-pathogenic and were considered to have potential for biocontrol (Harris 2000). Pathogenic binucleate Rhizoctonia spp. isolates belonging to AG E have been observed associated with soybean in the USA (Ploetz et al. 1985) and Indonesia (Naito et al. 1993). A few other binucleate Rhizoctonia spp. isolates were pathogenic to potato roots (Carling and Leiner 1990) and yacon (Fenille et al. 2005). Distinct AGs of R. solani have already been identified causing disease on different vegetable crops in Brazil (Bolkan and Ribeiro 1985; Kuramae et al. 2003). However, there are no reports on the occurrence of pathogenic binucleate Rhizoctonia causing disease on vegetable crops in Brazil. Particularly on kale, a Rhizoctonia-like disease has been observed causing outbreaks in São Paulo State vegetable-growing areas. Hypotcotyl and root rot symptoms have been observed on kale.

Considering the local importance of this Rhizoctonia-like disease for kale production, the objective of this study was to characterize the Rhizoctonia spp. isolates infecting kale by determining their nuclear condition, hyphal anastomosis grouping, virulence and genetic identity with other Rhizoctonia spp. based on sequencing analysis of the ITS-5.8S rDNA region. Our hypothesis was that a binucleate Rhizoctonia spp. is the primary agent of a hypotcotyl and root rot disease detected on kale.

The isolates were collected from kale (Brassica oleracea var. acephala) cultivated in Lençois Paulista, SP, Brazil, with hypocotyl and root rot symptoms. Small sections (0.5–0.7 cm long) were cut from the edge of an advancing lesion on each diseased root sample. These samples were surface-sterilised by immersion in 70% ethanol for 30 s, 2% sodium hypochlorite for 30 s, washed in sterile distilled water (SDW) and placed directly on Ko and Hora medium (1971). The nuclear condition and hyphal anastomosis grouping of the isolates were determined using the procedure of Ceresini et al. (1996). Hyphal cells were examined microscopically at 200 and 400× and nuclei were counted in 20 cells per isolate. Kale binucleate Rhizoctonia spp. isolates were paired against the AG testers (AG A, AG Ba, AG Bb, AG Bo, AG C, AG D, AG F, AG G, AG O, AG P and AG Q) available in our laboratory. Hyphal anastomosis was observed at 400× magnification using light microscopy after staining the vegetative cells with a 0.03% safranin-O aqueous solution and a 3% KOH aqueous solution. All isolates of Rhizoctonia spp. obtained were binucleate and did not show anastomosis reactions with any of the binucleate Rhizoctonia spp. testers used. However, in the absence of a few AG testers of binucleate Rhizoctonia spp. being not available in our laboratory (CAG 1, CAG 3, CAG 4, CAG 5, CAG 6, CAG 7, AG L, AG T and AG U), a phylogenetic approach was used for characterizing the unknown kale isolates. We based this approach on the comparative analysis of sequences of the ITS-5.8S rDNA region from all the AGs of binucleate Rhizoctonia spp. as well as the Rhizoctonia solani available and the unknown kale isolates. The sequences of the unavailable AGs were obtained from the GenBank. This approach provided us with some information about the placement of the unknown kale isolates among the AGs described for binucleate Rhizoctonia spp. worldwide. The use of a molecular tool is not currently accepted as a replacement criterion for the hyphal anastomosis characterization of AGs of Rhizoctonia spp. However, in certain instances the sequencing of the ITS-5.8S rDNA region has been used to separate subgroups of R. solani when anastomosis itself cannot separate them (Kuninaga et al. 1997).

The pathogenicity of the isolates was tested under greenhouse conditions at 25 ± 2°C using the substrate and inoculum preparation recommended by Fenille and Souza (1999). A completely randomized block design with five replications was used. After transferring the inoculum to the substrate, thirty seeds of kale cv. Portuguesa were sown per pot. Disease severity was assessed, 20 days after inoculation, using the scale described by Chung et al. (1988) and the fungus was re-isolated. All isolates caused hypocotyl and root rot symptoms on kale (Fig. 1) with different symptom severities. CO4 and CO2 isolates were the most virulent isolates (P = 0.05) (Table 1).

Fig. 1
figure 1

(A) is the control. (B) Pathogenicity of isolate CO4 of binucleate Rhizoctonia spp. to kale after 20 days of inoculation

Full size image
Table 1 Virulence of binucleate Rhizoctonia spp. on kale 20 days after inoculation
Full size table

To characterize the genetic identity of Rhizoctonia species infecting kale with other binucleate Rhizoctonia spp., four kale isolates (CO1, CO2, CO3, CO4) were compared with 12 binucleate Rhizoctonia spp. testers available in our laboratory (AG A, AG Ba, AG Bb, AG Bo, AG C, AG D, AG F, AG G, AG O, AG P and AG Q), and 39 testers (13 binucleate Rhizoctonia spp. and 25 R. solani) publicly available sequences. Total genomic DNA was extracted from the isolates using the methodology of Kuramae-Izioka (1997). The ITS4/ITS5 primer set (White et al. 1990) was used for PCR amplification of nuclear ITS1 and ITS2 regions and the 5.8S rRNA gene.

Each PCR product was purified using MicroSpin S-400 HR columns (Amersham Pharmacia), according to the instructions of the manufacturer, and sequenced using 10 ng of PCR product and 1 μM each of the ITS2, ITS3, ITS4 or ITS5 primers following the protocol supplied with the Amersham Premix Terminator (Amersham Pharmacia). Sequencing was performed using a PE Applied Biosystems Model 377 DNA Sequencer. The four sequenced fragments generated from each isolate were assembled using Phred/Phrap (Ewing et al. 1998) and Consed (Gordon et al. 1998). All consensus bases were of high quality with a Phred value >20. The consensus sequence for each isolate was trimmed in order to have the ITS1-5.8S-ITS2 sequence analyzed. First, the sequences were aligned using the computer software package CLUSTAL X (Thompson et al. 1997). The alignment parameters were gap opening = 10, gap extention = 0.20, delay divergent sequences = 30%, DNA transition weight = 0.50. Second, poorly aligned positions and divergent regions in the alignment were removed by using Gblocks 0.91b (Castresana 2000). The threshold parameters used were: minimum number of sequences for a conserved position = 50% of the number of sequences + 1, minimum number of sequences for a flank position = 85% of the number of sequences, maximum number of contiguous non-conserved positions = 8, minimum length of a block = 10. Thirty, distances (% divergence) between all pairs of sequences from the multiple alignment were calculated. The trees showing the phylogenetic relatedness between isolates and AG testers (Rhizoctonia binucleate and R. solani) were constructed using three different phylogenetic methods: maximum likelihood (PHYML) (Guindon and Gascuel 2003), maximum parsimony (MP), and neighbour-joining (NJ). For PHYML, the nucleotide model was HKY85 (Hasegawa et al. 1985), the transition/transversion ratio was 4, and the Gamma distribution parameter was 1.0. MP and NJ analyses were done using PROTPARS (heuristic search with characters equally weighted) and PROTDIST (Kimura formula) from Phylip (Felsenstein 1996), respectively. Non-parametric bootstrap support for MP and NJ was calculated from 100 resampling rounds.

All phylogenetic methods applied, viz. PHYML, MP, and NJ resulted in trees supporting the same phylogenetic position of the four kale isolates (Fig. 2). The kale isolates showed genetic identities ranging from 99.3 to 99.8% (Table 2) between them, even though they were collected from plants in the same field. The kale isolates were phylogenetically closely related to CAG 7 (AF354084) (Fig. 2), with identities of 98.5 and 98.7% (Table 2). The number of nucleotide differences among kale isolates and CAG 7 was higher at the ITS1 region (10 nucleotides) than at the ITS2 region (5 nucleotides). It is suggested that the binucleate Rhizoctonia spp. causing hypocotyl and root rot on kale Brazil comprises a new AG not yet described.

Fig. 2
figure 2

Phylogenetic trees constructed by Maximum Likelihood (I), Maximum Parsimony (II), and Neighbour-Joining (III) methods illustrating the phylogenetic relationship of four isolates from kale (CO1, CO2, CO3, CO4) and 50 testers [26 binucleate Rhizoctonia spp. (blue colour), and 24 R. solani (red colour)]. The sequences of AF354086 (CAG1), AF354080 (CAG3), AF354081 (CAG4), AF354082 (CAG5), AF354083 (CAG6), AF354084 (CAG7), AF354093 (AGL), AF354094 (AGO), AB196663 (AGT) and AB196664 (AGU) are from GenBank. The numbers at branches indicate % of 100 bootstrap replications. Only branches with >75% are shown. The ITS1-5.8S-ITS2 sequences of Rhizoctonia binucleate spp. of kale were deposited at GenBank: CO1 (DQ279021), CO2 (DQ301760), CO3 (DQ279022) and CO4 (DQ301761)

Full size image
Table 2 Similarity (%) between Rhizoctonia testers (Rhizoctonia binucleate species and R. solani) and isolates of kale (CO1, CO2, CO3, CO4)
Full size table