Introduction

Some species of Cercospora are etiological agents of leaf spot disease in a wide range of crop plant and wild plant species [1, 2]. Leaf blight and purple seed, caused by the fungal pathogen Cercospora kikuchii (Matsumoto & Tomoyasu) M. W. Gardner, is one of the most important diseases of soybean (Glycine max L. Merr.) worldwide [3]. High yield losses because of leaf blight disease are currently being recorded in Argentina on soybean. Incidence has increased up to 80% in the east-central region of Santa Fe province [4, 5].

C. kikuchii produces the non-host-specific phytotoxin cercosporin, which is very important for the pathogen infecting soybean [6–8]. Cercosporin is a photosensitizer and uses light energy to produce the activated oxygen species superoxide and singlet oxygen. This perylenequinone toxin is photoactivated and lack toxicity in the dark [9–12].

Cercospora leaf spot diseases show no obvious effect on the vigor and/or quality of the host plant at the time of infection. Exposure of plant cells and tissues to cercosporin results in peroxidation of the membrane lipids, leading to membrane breakdown and death of the cells [9]. Membrane damage allows for leakage of nutrients into the leaf intercellular spaces, allowing fungal growth and sporulation [8, 13–15]. Reduction in vigor, quality and yield becomes apparent on the appearance of necrotic lesions commonly found on leaves [16]. The diagnosis of Cercospora infection is possible only from sporulating necrotic lesions or cultures. Traditionally, the identification of Cercospora species has been based mainly on conidial characters and host association [2, 17, 18].

Molecular markers have proven to be powerful tools for the characterization and identification of several plant pathogenic fungi [19]. Several molecular typing methods including ribotyping, randomly amplified polymorphic DNA pattern (RAPD pattern), simple sequence repeats (SSR), inter simple sequence repeats (ISSR), restriction fraction length polymorphism (RFLP), have been successfully used for the identification and epidemiological characterization of different organisms [20–22]. The ribosomal DNA (encoding ribosomal RNA) unit has component sequences that evolve at different rates and that can be used for systematic studies at different taxonomic levels. The internal transcribed spacer (ITS) region within the rDNA units evolves rapidly, but it tends to be uniform in sequence with a particular species, and it differs between species. Therefore, the region separating 18S and 26S rDNA and the 5.8S coding region sequence can be used for a comparison of closely related species and subspecies as well as to characterize across interspecific and intergenic level divergences [23–25]. ISSR consists of the amplification of DNA sequences between SSR by means of anchored or non-anchored SSR homologous primers [26, 27]. ISSR-PCR has also been described as microsatellite-primed PCR in the literature because it specifically amplifies regions of the genome between microsatellites. The method of comparing inter-repeat profiles from the whole genome was first used in the identification of different eukaryotic species because of its accuracy and reproducibility [28].

Preliminary studies with few Argentinian isolates of C. kikuchi demonstrated phenotypic variability [29] and high levels of polymorphisms [30, 31]. While the basis for variability has not been resolved, the utilization of this variability as an exploitable tool has become the major focus of molecular ecology.

The aims of this work were: (a) to confirm and to assess the genetic variability among Cercospora kikuchii isolates collected from different soybean growing areas in Santa Fe province (Argentina) using ISSR markers and sequence information from the internal transcribed spacer (ITS) region of rDNA and (b) to analyze the cercosporin production of the regional C. kikuchi isolates in order to assess whether there was any relationship between the molecular profiles and the toxin production.

Materials and Methods

Fungal Isolates

Samples with characteristic C. kikuchii lesions were collected from soybean growing in different regions of Santa Fe province (Argentina), in years 2005, 2006 and 2007.

Petioles were surface–disinfected by dipping into 3% commercial sodium hypochlorite (55 g Cl l−1) for 3 min and then rinsed twice in fresh sterilized distilled water. Finally petiole pieces (0.5–1 cm) were placed in each wet chamber. Plates were incubated at 26 ± 0.5°C with alternating light–dark cycles of 8 h, for stimulating sporulation [32]. After the third and up to fifth day of incubation, tissues were observed with a stereo zoom microscope (BOECO Germany, BTB 3-C). C. kikuchii conidia from sporulating lesions were picked with the tip of a sterilized dissecting needle, placed onto Potato-Dextrose Agar (PDA) plates and incubated 3–7 days under the same temperature and lightning regimens [29, 30]. In order to obtain monosporic cultures of each C. kikuchii isolate, mycelium from individual germinated conidia was transferred to PDA and cultured fourteen days at 26 ± 0.5°C. Isolates were identified as C. kikuchii by their mycelial aspect and the size and shape of their conidia.

Each isolate was identified with a code, combining letters and numbers. They were maintained on PDA slants and stored at 4°C for daily work.

Strains from the NITE Biological Resource Center (NBRC) Collection (Japan) were used as control.

All studied fungi and their origin are listed in Table 1.

Table 1 Cercospora species isolated from soybean and their origin, genotypes based on internal transcribed spacer regions (ITS) and inter simple sequence repeats (ISSR) analysis and cercosporin production
Full size table

DNA Extraction

DNA extraction was performed as described by Di Conza et al. [33]. In brief, each isolate was subcultured onto a PDA slant, and the slant was incubated at 25°C for 4 days. Conidia suspension in water (105–106 UFC ml−1) was inoculated into 100 ml of Colletotrichum broth [34], which was in turn incubated at 28 ± 0.5°C for 48 h at 180 rpm. Cultures were harvested by filtration, washed with sterile water, blotted dry and finally powdered. The mycelium (0.06 g) was treated with 600 μl lysis buffer [Tris–HCl (pH 7.2) 0.05 M, EDTA 0.05 M, SDS 3%, 2-mercaptoethanol 1%] and 60 μl sarkosyl (10%). The suspension was incubated at 65°C for 1 h and 200 μl of 5 M potassium acetate and 100 μl 4 M sodium chloride were added. After a gentle ten times inversion, the suspension was allowed to rest on ice for 10 min. Following a centrifugation at 15,000 rpm. for 15 min, the upper phase was carefully transferred into a new tube and an equal volume of phenol–chloroform–isoamilic alcohol solution (25:24:1) was added. The suspension was gently mixed by inversion, pelleted by centrifugation for 5 min at 15,000 rpm and the aqueous phase was then carefully removed. Phenol–chloroform–isoamilic alcohol extraction was performed twice. An equal volume of chloroform–isoamilic alcohol solution (24:1, v/v) was added and emulsified by inversion 5 times, followed by centrifugation for 5 min. The aqueous phase was recovered and DNA was precipitated by adding 0.7 volumes of isopropanol. The tube was kept at room temperature for 20–30 min and the DNA was collected by centrifugation. The pellet was washed with 500 μl of 70% ethanol (previously cooled at −20°C) and completely dried at 55°C. Finally, DNA was solubilized in 100 μl of milliQ water, treated with 1 μl RNAsa (Promega, 10 mg ml−1) and stored at −20°C.

DNA concentrations were determined by measuring the A260 with a spectrophotometer and by gel electrophoresis [35].

PCR of Ribosomal DNA. Analysis of the Nucleotide Sequences

Primers designed to amplify the ITS region between 18S and 5.8S and part of the 28S rDNA were used in this case. Each fungal rDNA was amplified by PCR with primers ITS-4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) and ITS-5 (5′-GGA AGT AAA AGT CGT AAC AAG G-3′) [2, 36–39] obtained from FAGOS/Ruralex (Argentina).

The 50 μl PCR reaction mixture contained 5 μl 10× buffer (InbioHighway, Argentina), 5 μl each of 10 mM dNTP (InbioHighway, Argentina), about 2.5 μl each of ITS-4 and ITS-5 (10 μM), 50 ng of genomic DNA, 0.5 μl 5,000 U ml−1 TaqDNA Polymerase (InbioHighway, Argentina) and 4 μl 25 mM MgCl2. The mixture was gently vortexed and centrifuged briefly to collect the sample at the bottom, then overlaid with 15 μl of sterile mineral oil. The amplification reaction was performed in an MJ Research Thermal Cycler. The following cycling parameters were used: initial denaturation at 95°C for 5 min, primer annealing at 55°C for 1 min, enzyme chain extension at 72°C for 1 min, and cycled 34 times with a final extension at 72°C for 10 min.

Control experiments were performed without template DNA.

DNA concentration was determined by horizontal submerged electrophoresis (1× TBE buffer at 100 V) [35] on 0.8% agarose gels by comparison with the 100-bp DNA Ladder (InbioHighway, Argentina) used as molecular weight marker.

For sequencing purposes, amplified PCR products were excised from agarose gels and purified with the Wizard SV Gel & PCR Clean Up System (Promega, Cat.#A9281) and sent to be sequenced at the Biotechnology Institute (Unidad de Genómica) belonging to Instituto Nacional de Tecnología Agropecuaria (INTA), Argentina.

Bioinformatics analyses of sequences with Chromas Lite 2.01 program were carried out. Then, the sequences were aligned using the Vector NTI.9. Align X (http://bioinformatics.unc.edu/software/nti/index.htm) and Blast (http://www.ncbi.nlm.nih.gov/BLAST/) programs and compared with those available in the GenBank data base for Cercospora species (accession number AY633838).

ISSR-PCR DNA Fingerprint Profiling

ISSR PCR-assay proposed by Longato and Bonfante [40] was followed. Briefly, primer (5′-GTG GTG GTG GTG GTG-3′) obtained from FAGOS/Ruralex (Argentina) was used. The reaction mixture (final volume: 50 μl) consisted of approximately 20 ng of DNA and 5 μl adequate buffer (InbioHighway, Argentina) with 1.5 mM MgCl2, 1 mM each of dNTPs (InbioHighway, Argentina), 100 pmol of primer and 1 U of TaqDNA Polymerase (InbioHighway, Argentina) (5 U μl−1). Mineral oil (15 μl) was added to the top of the reaction. Control experiments were performed without template DNA. The amplification reaction was performed in an MJ Research Thermal Cycler programmed as follows: the starting annealing temperature of 70°C, was first decreased to 55°C by 2°C per cycle (repeating this cycle twice), and then 25 extra cycles were run at 55°C. The extension temperature was of 72°C.

The amplification products were analyzed by electrophoresis in 1.3% agarose gel, run in 1× TBE buffer [35] at 100 V and subsequently stained with ethidium bromide. Lambda DNA/EcoRI + Hind III (Promega) and 100-bp DNA Ladder (InbioHighway, Argentina) were used as molecular weight markers.

Penicillium spp. and Cladosporium spp. were included as differentiating genera.

Each assay was performed at least twice.

Gel images were photographed and analyzed with Gel Doc XR system (Bio Rad Life Science Cat.# 170–8170) using the Quantity One Software. Banding pattern similarities were scored by the Dice coefficient (D). The relationships among the different isolates studied were portrayed graphically in the form of dendograms. Bootstrap analysis (1,000 replication) was performed on the resulting tree with WinBoot Program to test the statistical support for each branch. Less than 50% bootstrap values do not represent statistically support and >70% values were considered a strong bootstrap support.

Cercosporin Production Assay

Cercosporin concentration was determined spectrophotometrically according to Jenns et al. [41] modified by González et al. [30]. Briefly, isolates were grown on PDA and incubated at 26 ± 0.5°C for eleven days under 16 h light. Three mycelium plugs (10 mm diameter) were cut from the border of the colonies, transferred to tubes containing 6 ml of 5 N KOH and kept in the dark for 3 h. After centrifugation at 7,500 rpm during 20 min, cercosporin was corroborated by the presence of characteristic peaks at 480, 595 and 640 nm in a spectrophotometer Perkin Elmer Lambda 20 UV/VIS (960 nm min−1 every 1 nm) and analyzed with OriginPro 7 program. A commercial cercosporin toxin (Sigma, lot 35082-49-6) was processed in parallel. Toxin concentration in a spectrophotometer at A480 nm using a molar extinction coefficient of 23,300 [41] was determined. C. kikuchii NBRC 6711 and C. sojina NBRC 6715 were used as control strains. Experiments were repeated twice. Results were reported as nanomoles per agar plug (nmol cyl−1) and standard deviations (SD) were calculated.

Results

ITS Sequence Analysis

The ITS-4 and ITS-5 primers uniformly amplified a fragment of approximately 518 bp from Argentinian isolates and NBRC strains. Amplified rDNA concentrations ranged between 100 and 150 ng μl−1.

Fifteen fungi presented complete sequence identity (100% identical) with C. kikuchii BRCK179 (Gen-Bank, accession number AY633838), C. kikuchii NBRC 6711 and C. sojina NBRC 6715 included. Four isolates were 99% identical to C. kikuchii BRCK179, the difference consisting 1 substitution at position 476 (T instead of C). Two isolates were 99% identical to C. kikuchii BRCK179, the difference consisting of two substitutions at positions 474 and 476 (C and T instead of T and C, respectively). Finally, 16 isolates were 99% identical to C. kikuchii BRCK179 but, in these fungi, C (474 position) substituted T (Fig. 1). The four ITS sequences obtained have been deposited in the GenBank sequence database under accesion numbers HM631725, HM631727, HM631728 and HM631726.

Fig. 1
figure 1

Computer alignment of the internal transcribed spacer region (ITS) of Cercospora kikuchii isolates and NBRC strains. AY633838: Cercospora kikuchii BRCK179 (Gen-Bank, accession number AY633838); CK6711: Cercospora kikuchii NBRC 6711; CS6715: Cercospora sojina NBRC 6715; C0, C2, C3, C4, C5, C6, C7, C8, C20, C23, C26, C32, C41, C14, C25, C39,C40, C35, C38, C9, C15, C16, C17, C18, C19, C21, C22, C24, C27, C28, C29, C30, C31, C36, C37: regional isolates

Full size image

ISSR-PCR DNA Fingerprint Profiling

Clearly detectable amplified ISSR ranged from 159 to 2,450 bp in size. The average number of clear bands generated was 8, with a maximum of 22 and a minimum of two. Figure 2 shows the ISSR patterns of C. kikuchii isolates and NBRC strains.

Fig. 2
figure 2

Sample gel of ISSR (primer GTG5) patterns produced by Cercospora kikuchii isolates and NBRC strains, their respective ITS sequences and cluster analysis dendrogram. C5, C4, C3, C2, C23, C9, C8, C7, C6, C0, C24, C20, C40, C26, C32, C41, C39, C25, C38, C35, C22, C21, C27, C14, C36, C19, C31, C29, C16, C30, C28, C37, C15, C18, C17: regional isolates; CS6715: Cercospora sojina NBRC 6715; CK6711:Cercospora kikuchii NBRC 6711; percentage bootstrap values based on 1,000 replicates are shown on the nodes

Full size image

Cluster analysis generated a dendogram (Fig. 2) with two branches: A and B, with a statistical support between them of 56.9% (not so strong). Branch “A” included four isolates, C2, C3, C4 and C5 (99.7% similarity) from different origin (Table 1) and the same ITS sequence (TAC). Cluster “B” included seven subclusters (B1, B2, B3, B4, B5, B6 and B7). B1 included six isolates, C23, C9, C8, C7, C6 and C0 (100% similarity), all of them with ITS sequence TAC, except C9 (CAC). C6 and C7 were isolated from the Riia Region 5 and A7118 varietal/cultivar, and C8 and C9 were obtained from soybean cultured by particular producers (Table 1). Subcluster B2 included six fungi. C40 presented 61.9% similarity with respect to the other five fungi. Both strains NBRC (C6711 and C6715) presented 88.3% similarity between them and 71.1% with C41. ITS sequence TAC was present in these fungi (B2), except C40 (TAT); C32 and C26, from different origin and cultivar/varietal, presented scarce genetic similarity with respect to C41 and NBRC strains. Subcluster B3 included six isolates, C39, C25, C38, C35, C22, and C21: C39 and C25 with the same pattern bands (100% similarities and ITS sequence TAT), but with less than 50% bootstrap support (45.9% with respect to the other 4 isolates). A 100% similarity between C22 and C21 (ITS sequence CAC) and 99.9% similarity between C38 and C35 (ITS sequence CAT), were detected. Only C21 and C22 were isolated from the same cultivar/varietal and region. B4 included C27 and C14 with the same pattern bands (100% similarity and ITS sequences CAC and TAT, respectively). B5 included C36 and C19 (ITS sequence CAC), both of them from different origin and cultivar/varietal and a low bootstrap value (51.1%). In a similar way, B6 included two isolates: C30 and C28 (ITS sequence CAC) from the same origin and cultivar/varietal (61.7% similarity). Finally, B7 included C18 and C17 (99.5% similarity), both of them from the same origin and cultivar/varietal.

The remaining seven isolates in cluster B: C24, C20, C31, C29, C16, C37 and C15 (different origin), all of them with ITS sequence CAC, except C20 (TAC) (Table 1), could not be assigned to any of these subgroups; further information being needed to solve this situation.

The inclusion of Cladosporium spp. and Penicillium spp. was useful to determine that the technique allowed to differentiate fungal genera.

Cercosporin Production Assay

Thirty three regional isolates and the NBRC strains were cercosporin producers, the greatest concentration being from C. sojina NBRC 6715 and C. kikuchii NBRC 6711. Non-producers were C8 and C20 (all of them with TAC sequence) (Table 1).

Discussion

Soybean (Glycine max (L.) Merr.) is one of the most important crops in Argentina, and it has been characterized by an incredible rate of adoption and growth. In 1970–1971 the soybean production amounted to 59,000 tons, covering a crop area of approximately 38,000 hectares. Nowadays, the crop area has increased to over 16.6 million hectares, making it the world’s third largest producer, the main exporter of soybean oil (30% of the world exports) and the second exporter of soybean flour (27% of the world exports) [42, 43]. Twenty-one percent of the cultivated lands of Argentina are in Santa Fe province whose main crop, soybean, makes it the main national producer. However, soybean crops are affected by several diseases which decrease the total production. The RiiA programme, an interdisciplinary intervention carried out at the Center North of Santa Fe province, was developed by the Facultad de Ciencias Agrarias of Universidad Nacional del Litoral, the Experimental Station INTA Rafaela and some institutions linked to agricultural production. One of its aims is to detect early adverse situations to prevent or mitigate the effects so as to generate information for subsequent campaigns and identify and priorize regional problems. Monitoring, processing and communication of the soybean crop evolution are the goals aimed at by this programme.

This study contributes to the knowledge of some phenotypic and genotypic characteristics of one of the most frequent phytopathogens in order to provide information for the soybean breeding program at RiiA.

Since typing is a necessary first step in knowing pathogens [44], techniques based on DNA polymorphisms are especially valuable to enhance epidemiological studies. PCR of ITS regions and its sequence data have been studied to assess genetic diversity at intraspecific level in different species [2, 25, 45]. As Somai et al. [46] reported, ITS sequence data in this study indicated a closer relationship between C. kikuchii isolates than ISSR data did, even though intraspecific variation was found in the rDNA ITS region of C. kikuchii isolated from geographically widely distributed cultivars of soybean developed in the northern region of Santa Fe province. Therefore, it appears that four genotypes of C. kikuchii may be associated with and isolated from soybean exhibiting symptoms of leaf blight. It has been suggested that the spacer sequences may accumulate mutations in the form of base substitution, duplications, deletions and insertions and chromosomal rearrangements in genomes [47]. In our study, the simplest explanation for this fact is that fungi with ITS sequences CAC, CAT, TAT are genetic subgroups within C. kikuchii that are diverging from C kikuchii BRCK179 sequence (TAC) which is deposited in Gen Bank. ITS sequence analysis results showed high similarity (99–100%) with published sequences of C kikuchii BRCK179 (Gen Bank AY633838). Even in this investigation, CAC sequence was the most frequently detected (43.2%). The number of nucleotide differences of ITS of the rDNA among all sequences ranged from 0 to 2. However, Goodwin et al. [48], in a previous publication, mentioned that the mean number of nucleotide differences between three isolates of C kikuchii ranged from 2 to 7. These differences may be due to different sets of sequences (TREEEBASE) used by those authors for comparing their isolates. Another authors, studying 12 sequences of Brazilian isolates of C. kikuchii, detected that the differences in ITS sequences ranged from 0 to 3 [36].

Data obtained revealed a considerable degree of variation in the Argentinian population of C kikuchii, thus confirming previous results [30, 31]. In agreement with our study, Almeida et al. [36] and Cai [3] determined a high level of variability among isolates of C. kikuchii and these results show no genetic grouping of cercosporin producers or grouping for geographic location. According to Pujol Vieira dos Santos et al. [49] and Stenglein and Ballati [50] there are many factors that could have been affecting polymorphism analysis, e.g. the intraspecific variants of a pathogen, the number of samples selected for analysis, genetic flow between populations, environmental adaptation and selective pressure and migration.

The role of cercosporin in the pathogenicity of C. kikuchii was first demonstrated by Upchurch et al. [8], who considered it crucial for the infection of soybean plants. Ability to produce cercosporin allowed the ancestral Cercospora species to expand its host range. This would explain the occurrence of a large number of closely related species, some with identical ITS sequences, on widely divergent hosts [44, 51].

In this study, isolates that produced high cercosporin concentration belonged to ITS sequence group TAC, as well as non producers did (C8 and C20). It is important to consider that C8 and C20 were isolated from ill soybean. Under our working conditions, C. sojina NBRC 6715 produced a high concentration of cercosporin, unlike Goodwin et al.’s study in which C. sojina isolate was non-producing [48]. It was not clear if the lack of colour in the extract for colorimetric assay observed in some isolates could be due to the sensitivity of the method used since isolates with very low amounts of cercosporin were also able to cause lesions when inoculated on soybean leaves. Upchurch et al. observed that spontaneous and UV-induced mutants which did not produce cercosporin were not able to cause infection when inoculated on soybean leaves [8].

There was no association between the groups of isolates based on ISSR markers and cercosporin content. Isolates that produced high contents of cercosporin were clustered together with those from low cercosporin producers. For example, isolates C14 (101.26 nmol cyl−1) and C25 (114.52 nmol cyl−1) were high producers of cercosporin and were separated in clusters B4 and B3, respectively. Isolates C27 (30.74 nmol cyl−1) and C14 were clustered together but cercosporin content was significantly higher in the latter.

In all analyses it was shown that C. kikuchii isolates from the same geographic region appeared in different groups. Cercosporin production and molecular analyses showed intraspecific variability within C. kikuchii isolates recovered from soybean collected in different RiiA regions from Argentina, so it was difficult to establish a relationship between this variability and the soybean cultivars from which C. kikuchii isolates were obtained. Similar results were previously reported by González et al. and Almeida et al. [30, 36]. According to the obtained results, Argentinian populations of C. kikuchii are phenotypically, genotypically and geographically variable. In agreement with Almeida et al. [36], who consider that this pathogen is easily transmitted by seeds, it is not surprising to find the same haplotypes in different regions. The present study expands on the previous one by using another molecular marker (ISSR) and ITS sequence analysis on a larger collection of isolates. Many groups were present among the 37 studied fungi; however, the groups could not be correlated with cercosporin content or geographic origin of the isolates.

In Argentina, there has been a rapid increase in the soybean producing area since 1970 [42]; therefore, the traffic of seeds from traditional areas to new areas could be responsible for the geographical variability since C. kikuchii is a seed borne pathogen. Unfortunately, an insufficient number of isolates was obtained from each area to permit the evaluation of gene flow among populations more accurately.

For countries with large soybean areas like Argentina, it is very important to know in advance the variability of the pathogen in order to avoid resistant cultivars when sown in different areas. The level of variability among isolates as identified in this work may help to define the breeding method for effective resistance.