Introduction

Chickpea (Cicer arietinum L.) is an important pulse crop cultivated worldwide. Wilt caused by Fusarium oxysporum f. sp. ciceris (Padwick) Matuo and K. Sato is one of the major factors responsible for low productivity of the crop. The disease is distributed in all the chickpea-growing areas of the world including India. Eight races of F. oxysporum f. sp. ciceris were identified by reactions on a set of differential chickpea cultivars. Of these, 4 races, namely 1A, 2, 3, and 4, were reported from India (Haware and Nene 1982), while races 0, 1B/C, 5, and 6 were reported from Mediterranean region and the USA (Phillips 1988). In addition to differences in races, the Indian populations of the pathogen produce typical vascular wilt symptoms in chickpea. Honnareddy and Dubey (2006) reported the occurrence of more than one race from each state of India along with a few new races. Further, it was suggested that some of the new chickpea cultivars showing clear-cut differential reaction for Indian population of the pathogen possibly due to the development of new variants in a field should be incorporated into earlier reported set of differentials (Dubey and Singh 2008; Dubey et al. 2010).

Different molecular markers were used for the characterization of genetic diversity of the pathogen and the evolutionary origin of a particular forma specialis (Jimenez-Gasco et al. 2002; Dubey and Singh 2008). The phenotypic and genetic characterization of an area-specific pathogenic population of F. oxysporum f. sp. ciceris is essential for efficient disease management. Different genetic markers such as ITS or intergenic spacer regions of ribosomal DNA (rDNA) and the TEF-1α gene have been used to resolve evolutionary relationships in Fusarium species and for developing PCR (polymerase chain reaction)-RFLP assays for the differentiation of Fusarium at the species level (Gurjar et al. 2009). The TEF-1α gene (Stengleina et al. 2010) and β-tubulin gene sequences have been used for phylogenetic analysis in fungi (Thon and Royse 1999). Jimenez-Gasco et al. (2002) analyzed 7 F. oxysporum f. sp. ciceris isolates representing four Indian races for TEF-1α and β-tubulin gene sequences. Gurjar et al. (2009) used ITS-RFLP, ISSR, and AFLP for distinguishing Indian races of the pathogen. In both the studies, less number of isolates were used and the virulence analysis of these isolates has not been done for race determination as now it has been proved that each state of India has more than one race (Dubey and Singh 2008) and new distribution pattern of the race in India has been determined (Dubey et al. 2012). The isolates included in Jimenez-Gasco et al. (2002) study may be representative of the same race, because earlier the race distribution was area/state specific, but presently, it has been observed that each state/area has more than one race (Dubey et al. 2012). The three genes TEF-1α, β-tubulin, and ITS sequences of Indian populations of F. oxysporum f. sp. ciceris representing 8 races have not been analyzed earlier.

The aims of the present study were to analyze the genetic diversity of new Indian populations of F. oxysporum f. sp. ciceris originating from different chickpea-growing areas on the basis of TEF-1α, β-tubulin, and ITS gene sequences and to examine whether the variation in the sequences of these genes could be correlated with their geographic origin (state) and/or races of the pathogen.

Materials and methods

Fungal cultures and DNA extraction

Seventy isolates of F. oxysporum f. sp. ciceris collected from 13 states of India were included in the present study (Table 1). These isolates are representative populations of 640 isolates of the pathogen, which have already been characterized for their morphological feature as well as virulence (Dubey and Singh 2008; Dubey et al. 2010). The isolates were maintained at 4 °C on potato dextrose agar. The single spore pure cultures of F. oxysporum f. sp. ciceris isolates were multiplied in potato dextrose broth (20 g l−1; Hi-media) at 25 ± 1 °C on shaking incubator for 120 rpm for 7 days. Harvested mycelial mats were used for DNA extraction following modified CTAB method (Murray and Thompson 1980; Dubey and Singh 2008) and stored at −20 °C in small aliquots.

Table 1 The details of the isolates of Fusarium oxysporum f. sp. ciceris and their GenBank accession numbers for various genes used in the present study
Full size table

PCR amplification and sequencing

The genomic DNA of 70 isolates was analyzed with gene-specific primers, namely TEF-1α, β-tubulin, and ITS. PCR mixture (25 μL) contained 50 ng template DNA, 1.0 U Taq DNA polymerase, 1.5 mmol/L of MgCl2, 0.6 mmol/L of each dNTPs (Bangalore Genei, India), 10 pmol each of primers for TEF-1α and β-tubulin, and 5 pmol each for ITS in 10× reaction buffer. The PCR was performed using Eppendorf ep™ gradient thermal cycler with initial denaturation at 94 °C for 5 min followed by 35 cycles of denaturation at 94 °C for 1 min, annealing at 60 °C for TEF-1α, 62 °C for β-tubulin and 55 °C for ITS for 1 min, and extension at 72 °C for 1 min with final extension at 72 °C for 7 min followed by cooling at 4 °C. Amplification products were electrophoresed on 1.25 % agarose gel in 1x TAE buffer subsequently eluted, purified using Genei Pure™ gel extraction and purification kits (Bangalore Genei, India), and sequenced (Xcelris Labs Limited, Ahmedabad, India). The nucleotide sequences were subjected to BLAST analysis (http://www.ncbi.nih.gov/index.html) and submitted to GenBank at NCBI and accession numbers (JN231134 to JN231186 for TEF-1α gene, JN231187 to JN231236 for β-tubulin gene, and JN400675 to JN400720 for the ITS region) obtained. The multiple sequence alignment and pairwise alignment were made using BioEdit version 7.0.5 (Hall 1999). The phylogenetic tree was constructed by CLUSTALW 1.8 sequence alignment and bootstrap neighbor joining by MEGA 4.1 program with 1,000 bootstrap replicates.

Results

TEF-1α analysis

The genomic DNA of 70 isolates of F. oxysporum f. sp. ciceris was successfully amplified with a set of TEF-1α gene-specific primers EF1 (5′-ATGGGTAAGGA(A/G)GACAAGAC-3′) and EF2 (5′-GGA(G/A)GTACCAGT(G/C)ATCATGTT-3′) (O’Donnell et al. 1998) and obtained ~720-bp amplicons. The nucleotide sequences of 53 representative isolates of F. oxysporum f. sp. ciceris varied from 630 to 737 bp and shared more than 90 % nucleotide sequence similarity with the TEF-1α sequences of F. oxysporum available in GenBank. The phylogenetic tree constructed using bootstrap neighbor-joining analysis of 53 isolates of the present study along with 11 isolates of F. oxysporum from other countries clearly grouped the isolates into two major clusters (Fig. 1). Thirty-nine isolates of the present study representing eight races were grouped into first major cluster along with 11 isolates from other countries. Second major cluster had 14 isolates from the present study. Out of 12 isolates of race 1, 8 were placed into this cluster along with two isolates of race 6 and one isolate each from race 2, 3, 5, and 7. The both major clusters were further subdivided into two subclusters.

Fig. 1
figure 1

Neighbor-joining tree showing the phylogenetic relationships among the isolates of F. oxysporum f. sp. ciceris (Foc) based on their TEF-1α sequences. The sequences generated during this study were labeled (labeled diamond) Fo- F. oxysporum, Fb- F. oxysporum f. sp. cubense, F- Fusarium, Fop- F. oxysporum f. sp. phaseol, and R-race number

Full size image

β-Tubulin analysis

β-Tubulin gene-specific primers (Jimenez-Gasco et al. 2002) successfully amplified the genomic DNA of 70 isolates and produced ~500-bp amplicons. β-Tubulin nucleotide sequences of 50 area/state representative isolates of the present study ranged from 479 to 517 bp. The β-tubulin gene sequences of the isolates used in the present study shared more than 90 % nucleotide sequence similarity with β-tubulin sequences of F. oxysporum available in GenBank. The phylogenetic tree constructed using bootstrap neighbor-joining analysis of 50 isolates of the present study representing eight races along with eight foreign isolates grouped them into two major clusters. The first major cluster had all 50 isolates of the present study along with one isolate from Spain. The remaining seven foreign isolates were grouped into the second major cluster (Fig. 2).

Fig. 2
figure 2

Neighbor-joining tree showing the phylogenetic relationships among isolates of F. oxysporum f. sp. ciceris (Foc) based on their β-tubulin sequences. The sequences generated during this study were labeled (labeled diamond) Fol- F. oxysporum f. sp. lycopersici, Fo- F. oxysporum, Foe- F. oxysporum f. sp. erythroxyli, Focb- F. oxysporum f. sp. cubense ,and R-race number

Full size image

ITS analysis

PCR amplification with specific primers ITS1 and ITS4 given by White et al. (1990) yielded a single DNA fragment ~550 bp in size present in all the isolates. Nucleotide sequences of ITS region of 46 area representative isolates ranged from 528 to 564 bp. The phylogenetic tree of 46 F. oxysporum f. sp. ciceris isolates of the present study and 10 isolates of F. oxysporum originated from other parts of the world grouped the isolates into two major clusters. All Indian isolates representing 8 races of the present study were placed in the first major cluster along with 8 isolates of other countries. The second major cluster had only two isolates one each from Japan and Greece (Fig. 3).

Fig. 3
figure 3

Neighbor-joining tree showing the phylogenetic relationships among isolates of F. oxysporum f. sp. ciceris (Foc) based on their ITS sequences. The sequences generated during this study were labeled (labeled diamond) Fo- F. oxysporum, Focb- F. oxysporum f. sp. cubense, Folu- F. oxysporum f. sp. luffae, Foly- F. oxysporum f. sp. lycopersici, Fov- F. oxysporum f. sp. vasinfectum, and R-race number

Full size image

Discussion

The phylogenetic tree analysis of TEF-1α, β-tubulin, and ITS gene sequences of F. oxysporum f. sp. ciceris showed that the each clade had the isolates from different states of India representing different races of the pathogen, indicating considerable variability in some of the isolates belonging to the same areas. The sequence analysis of TEF-1α gene clearly indicated that more than 50 % Indian populations representing 8 races of the pathogen clustered along with 11 foreign isolates of Fusarium. The remaining Indian populations (14 isolates) representing 6 races of the pathogen are considered distinct from others in respect of TEF-1α gene sequences and predominated by the population of race 1, which was prevalent mainly in southern part of India. Earlier to this, TEF-1α gene region was used for variability and identification of various Fusarium species (Yli-Mattila et al. 2011). The sequence analysis of β-tubulin gene clearly showed high level of similarity, and all the isolates were clustered with an isolate of Spain in a single group. The isolates belonging to other forma specialis were clustered separately; therefore, in respect of β-tubulin gene, forma specialis-specific grouping could be made. The present findings are in accordance with the earlier observation that the β-tubulin genes are potentially useful in phylogeographic investigations and characterization at species level (O’Donnell and Cigelnik 1997; Chung et al. 2008). The β-tubulin gene sequences did not correspond to the place of origin and race of the isolates. Similarly, Hill et al. (2010) while working on F. oxysporum from sugar beet concluded that the phylogenetic group did not correlate with pathogenicity and/or geographic origin. The phylogenetic tree from ITS sequences showed a high level of similarity and grouped into a single cluster in spite of the differences in the places of origin of the isolates and the races of the pathogen. Similarly, Hill et al. (2010) while working on F. oxysporum concluded that the phylogenetic group did not correlate with pathogenicity and/or geographic origin.

The high level of sequence similarity (90 %) in respect of these three gene regions is attributed to the origin of F. oxysporum f. sp. ciceris from a small founder population that was pathogenic to Cicer spp. (Jimenez-Gasco et al. 2002). The present findings are in accordance with the observations of Jimenez-Gasco et al. (2002) that the isolates representing the two pathotypes and the geographic range of the pathogen showed identical sequences for TEF-1α, β-tubulin, histone 3, actin, and calmodulin. Their observations were based on seven typical wilt producing F. oxysporum f. sp. ciceris isolates representing 4 Indian races, but presently, the racial scenario of the pathogen has been entirely changed, and a new set of differential cultivars were identified for race profiling of the pathogen (Dubey et al. 2012). The isolates included in Jimenez-Gasco et al. (2002) study were not tested for their virulence to determine races, and however, earlier the race distribution was area/state specific, and now in each state/area, more than one race was present (Dubey et al. 2012). The phylogenetic analysis of β-tubulin sequences only supported the theory of monophyletic origin of the pathogen given by Jimenez-Gasco et al. (2002). The present finding is differ with the results of Jimenez-Gasco et al. (2002) in respect of phylogenetic analysis of the sequences of TEF-1α and ITS, which grouped the other F. oxysporum isolates separately. This difference might be due to less number of isolates included in the earlier studies. Earlier to this, Gurjar et al. (2009) identified the race 3 of F. oxysporum f. sp. ciceris as F. proliferatum based on phylogenetic analysis with EF-1α sequences without determining the races of the isolates used in the study. The EF-1α sequences generated from the original race 3 representative isolates deposited in ITCC, New Delhi, India, by Haware and Nene (1982) showed sequence similarity to F. oxysporum f. sp. ciceris isolates of EF-1α. Therefore, the identity of the culture represented race 3 of F. oxysporum f. sp ciceris used by Gurjar et al. (2009) might be relooked. However, recently, the Indian populations have been characterized into eight races based on a new set of differential cultivars of chickpea (Dubey et al. 2012).

The critical analysis of the sequences of TEF-1α, β-tubulin, and ITS genes revealed that most of the isolates shared more than 90 % similarity among themselves for each gene sequence. The isolates showed less variability in respect of gene-specific regions, indicating the development of variants due to introduction of large numbers of area-specific resistant varieties of chickpea in the recent past. The phylogenetic groups generated in the present study did not correspond to the origin of the isolates and races of F. oxysporum f. sp. ciceris in respect of β-tubulin and ITS sequences, whereas it was partially corresponded in case of TEF-1α sequence. This is might be due to introduction of various race representative populations of the pathogen through infected seeds and cultivation of chickpea cultivars having different genetic background in an area. The markers/genes to determine virulence factors for different races of the pathogen have to be worked out. The low level of variability observed among the populations in respect of these three gene sequences could be used for designing F. oxysporum f. sp. ciceris-specific markers. The β-tubulin gene showed advantage over other two genes studied due to high level of similarity and could be more suitable for F. oxysporum f. sp. ciceris-specific studies. The information generated in the present study on F. oxysporum f. sp. ciceris is new because it is based on Indian populations representing 8 new races of the pathogen recently characterized using new set of chickpea differentials (Dubey et al. 2012). The information based on TEF-1α, β-tubulin, and ITS genes generated in the present study is useful for understanding of evolutionary relationship of wilt producing populations of the pathogen and development of specific diagnostics.