INTRODUCTION

Cumin (Cuminum cyminum L.) locally known as Zeera in Hindi, belongs to order umbellales and family Umbelliferae and believed to have originated from Egypt (Edison et al., 1991). It is grown extensively in South-Eastern Europe and North Africa bordering the Mediterranean Sea and in India and China (Chattopadhyay and Maiti, 1990). In India, cumin is one of the most important seed spice crop grown in the states of Rajasthan, Gujarat, Uttar Pradesh and Tamil Nadu. The area under cumin cultivation in India is about 430840 ha with annual production of 227829 tones. Rajasthan stands first in acreage (202980 ha) and production (120981 tones). Major cumin producing districts in Rajasthan are Barmer, Jalor, Nagaur and Jodhpur. During 2004-05, 13750 metric tones of cumin seed were exported earning a foreign exchange of Rs. 10190.00 lakhs. Cumin seeds preferred for their typical pleasant aroma due to an aromatic alcohol (animol) and spicy taste. It is largely used in mixed spices and curry powders.

The average yield of this crop is however low due to lack of superior varieties, scientific crop production technology and vulnerability to diseases like Wilt, blight and Powdery mildew incited by Fusarium oxysporum f. sp. cumini, Alternaria burnsii and Erysiphe polygoni, respectively. In these diseases wilt is most common, results in yield losses up to 35% in cumin in some districts of Rajasthan (Vyas and Mathur, 2002). Many other plants are also adversely affected by these pathogens (Gupta et al., 2010; Srivastava et al., 2011; Kelaniyangoda et al., 2011).

Certain difficulties have been encountered for the management of this disease in plant because of scanty availability of known and recommended fungicides in the market. In addition, these chemicals are costly and not fully efficacious in some cases. In fact, during last two decades, problems of resistance against bactericides and fungicides (Silva and Singh, 1974; Nene and Thapliyal, 2000) have increased as well as some new important pathogenic strain have been identified and documented (Champawat and Pathak, 1991; Chakraborty, 2005; Mamatha and Rai, 2004). Management of wilt by resistance breeding is an economically viable and ecologically desirable strategy. Many scientists have also tried to manage it by use of biocontrol agents (Reid et al., 2002; Mujeebur et al., 2004; Jegathambigai et al., 2009). Knowledge of diversity and relatedness among the pathogen populations is, however, a prerequisite for exploitation of resistance breeding. F. oxysporum isolates from different agroclimatic areas exhibit considerable diversity (Singh and Kumar, 2011) with respect to cultural characteristics (Elwakil and Ghoneem, 1999; Haglund, 1984) also seen most often in areas of fields that are low and compacted, especially during periods of high temperatures (Hanson and Jacobsen, 2009).

There are a number of techniques used to observe genetic variation within fungal pathogens. One of the techniques is VCG which is based on the ability of the mycelium to anastomose to form heterokaryon to determine genetic relatedness (Mohammadi and Mofrad, 2009). Now a day’s attention has been focused on studying polymorphism at the DNA level for genetic characterization because different molecular markers, apart from elucidation of genetic variability, can also be used to study evolution and monitoring movement/shift of pathogen population over time and space Previously study on genetic relationship and phylogeny among Fusarium species have been conducted at DNA levels by many scientists (Bogale et al., 2006, 2007; Suga et al., 2000; Benthy et al., 1998; Attitalla et al., 2011; Hill et al., 2011).

The aim of this study was to better understand phylogenetic relationships among F. oxysporum isolates collected from different regions of Rajasthan and determine if this diversity correlates with pathogenicity or geographic origin.

MATERIALS AND METHODS

Collection and identification: Collection of fungal isolates from cumin plants exhibiting symptoms of F. oxysporum were collected from different cumin cultivated areas of Rajasthan. Tap root and side root samples were trimmed, washed in running tap water to remove soil, blotted dry and cut into 10 mm segments. Root segments were sterilized with 0.5% NaCl for 1 min Each longitudinal section was transferred to Petri dishes containing potato dextrose agar (PDA; Hi media). Plates were incubated on the laboratory bench for 5 days at 25±2°C. Fungi were transferred to freshly prepared PDA to eliminate contamination. Single spore isolates were either prepared from cultures of Fusarium species (Nelson et al., 1983; Leslie et al., 2006) or isolates were subcultured onto 2% (w/v) water agar (Hi media) for hyphal tip transfer (Nelson et al., 1983) to obtain pure cultures. Single spore isolation made from each colony. Isolates were identified morphologically to species based on characteristics of macroconidia, phialides, microconidia, chlamydospores and colony growth traits (Leslie et al., 2006).

Pathogenicity screening: Ten seeds of cumin were planted in each of 304 cm diameter plastic pots containing a soil mixture. After seed germination, a plug of F. oxysporum culture was put on the base of a seedling. Negative controls were treated same way but lacked fungi. The pots were placed in a greenhouse at 25±2°C. Symptoms were evaluated after 3-6 days of incubation (Kavak and Boydak, 2006; Keinath, 1994; Valizadeh et al., 2007).

Vegetative compatibility grouping (VCG): Nitrate non-utilizing (nit) mutants were generated and the phenotypes were characterized as nit1, nit3 or nitM as described by Correll et al. (1987). Pairings of all possible combinations or complementation of the nit mutants was made on Minimal Media (MM) (Puhalla, 1985) and amended with kalium chlorate concentration ranging from 1.5% (w/v) to 5.0% (w/v). The MM plates were incubated at 25°C. Vegetatively compatible isolates produced abundant aerial mycelium at the interface of two colonies and were placed in the same group.

DNA extraction/PCR amplification and electrophoresis: Total genomic DNA was extracted using the method of Doyle and Doyle (1990) from five days old fungal mycelium grown on PDA. The quality of extracted DNA after RNase treatment was assessed on 0.8% agarose gel and finally the DNA was quantified using spectrophotometer (Optigen 2020 plus). The DNA samples were diluted to 25 ng μL^-1 for PCR amplification.

For RAPD analysis 40 primers from set No. 1 and 10 primers from set No. 2 obtained from the University of British Columbia, Vancouver, Canada were screened. Out of 50 primers only 9 primers were amplified and eight of these primers were removed because of their monomorphic nature and poor reproducibility. The PCR reactions were performed in a 25 μL reaction mixture containing 1x assay buffer, 0.5 units of Taq DNA polymerase, 200 μM of each dNTPs (Bangalore Genei), 0.2 μM primers and 50 ng of template DNA. The PCR reactions were carried out in DNA thermal cycler (Model-CGI-96, Corbett Research, Australia) using a single primer in each reaction. The PCR reactions were repeated thrice for each primer to ensure the reproducibility of RAPD results. Only highly reproducible and polymorphic primers were included in the study.

The PCR amplification conditions for RAPD consisted of initial extended step of denaturation at 94°C for 4 min followed by 44 cycles of denaturation at 94°C for 1 min, primer annealing at 37°C for 1 min and elongation at 72°C for 2 min followed by a final step of extension at 72°C for 4 min. The PCR reaction products were mixed with 4 μL of 6x DNA loading buffer and fractionated on 1.2% agarose for RAPD containing 0.5 μg μL^-1 ethidium bromide. Gels were electrophoresed until the indicator dye reached 10 cm from the well at 55 mA for 4 h. After separation gels were documented using Avigene Gel Doc syastem (Koria).

Scoring and data analysis: Only clear and reproducible bands were scored for the data analysis, but a major band corresponding to a faint band in repetition was also included in the study. RAPD data were scored for the presence (1) or absence (0) and each band was regarded as a locus. λ DNA EcoRI/HindIII double digest marker was used as a standard for the estimation of molecular weight of the RAPD products. Bands with same molecular weight and mobility were considered as a single locus. The total number of alleles, polymorphic alleles, average number of alleles per primer and polymorphism percentage were calculated.

Similarity matrix and cluster analysis: The statistical calculations were done using Ntsys pc 2.02e. Similarity matrix for RAPD primers was constructed using the Jaccard’s similarity coefficient values to find out genotypic relationship. The average distance of a single variety from rest of the genotypes was also calculated. The 0/1 matrix data obtained from RAPD were arranged to get separate similarity matrices which were subjected to UPGMA (unweighted pair-group method with arithmetic averages) analysis to generate dendrogram and compared using the Mantel matrix correspondence test (MxComp module of NTSYSpc).

RESULTS AND DISCUSSION

Total ten isolates of F. oxysporum were recovered from root rotted parts of cumin plants collected from the major producing areas of Rajasthan (Table 1). Colonies of isolates were woolly to cottony with cream to white aerial mycelium and purple pigment. Conidiophores had simple or branched monophialides, microconidia are generally one celled, very seldom two celled, hyaline measuring 2-3.5x5-12 μm. Macroconidia are mostly 3- septate but some have 4-5 septa. They are boat shaped to oblong and 3-4x25-45 μm in size.

Pathogenecity and VCG grouping: Isolates of F. oxysporum were examined for their pathogenecity on seedlings of cumin. Results of pathogenecity tests revealed that all isolates were pathogenic to cumin and produce interveinal yellowing of leaves followed by wilting (Bowden and Leslie, 1992).

F. oxysporum isolates produced chlorate-resistant sectors on media complemented with chlorate. One hundred and twelve nit mutants were generated from 10 isolates using PDA amended with 3 and 5% potassium chlorate. The number of sectors was observed at 3.0% (w/v) of chlorate concentration. The nit mutants were divided into three classes; nit1 (a mutation of nitrate reductase structural locus), nit3 (a mutation of nitrate-assimilation pathway specific locus) and nitM (mutations that affect the assembly of a molybdenum-containing cofactor necessary for nitrate reductase activity). The most frequent phenotype was nit1 (50%), followed by nit3 (20%) and nitM (30%) among total 112 mutants. According to the literature data the frequency of mutant nit1 is higher than the frequency of other types of nit mutants. Similarly, Klittich and Leslie (1988) examined twelve strains of F. moniliforme for their ability to sector spontaneously on toxic chlorate medium and concluded that genetic control of nitrate reduction in F. moniliforme is similar to that in Aspergillus and Neurospora, but that the overall regulation of nitrogen metabolism may be different.

Based on pairing complementary nit mutants of all isolates, mainly with nit1 and nitM mutants, the 10 isolates were grouped into six VCGs that B, and C VCGs have two members. VCGs A had 3 members and D, F and E had only single member (Table 2).

VCG diversity can be calculated by dividing the number of total VCG by the total number of isolates (Smith-White et al., 2001). In the present study, the overall VCG diversity for F. oxysporum was 50% which suggest that VCG analysis showed considerable variations among the isolates. In VCG analysis, the variation could be cause by a single base changes with compatible loci which may divide two almost identical isolates into separate groups. Isolates in a same VCG often share pathological and physiological traits as well as geographical origins (Swift et al., 2002; Sharifi et al., 2008). In this study, this attributes can be seen in F. oxysporum isolates in which the VCGs were grouped according to the symptom and locations. This study was also revealed by various previously carried research works (Hawthorne and Rees-George, 1996; Kistler et al., 1998; Latiffah et al., 2009).

Table 1:	Localities and their attributes selected for collection and study of cumin

Table 2:	Vegetative compatibility groups of F. oxysporum used in the study

Table 3:	List of single arbitrary primers showing total and polymorphic amplicons generated along with PIC of each pattern for 10 fungal genotypes

RAPD analysis: VCG analysis has its limitations especially some isolates were unable to form mutants on chlorate medium and to form heterokaryon in complementation test that why molecular diversity analysis using RAPD analysis was also include in this study for better characterization of isolates from more diverse locations and different cumin species.

RAPD analysis was performed using 40 primers from set No. 1 and 10 primers from set No. 2 obtained from the UBC. Out of 40 primers only 9 primers were amplified and eight of these primers were removed because of their monomorphic nature and poor reproducibility. The total number of bands generated by eleven amplifying primers was 65 with an average amplification of 7.20 bands per primer. The average polymorphism generated by these bands was 63.07%. The size of the amplicons generated varied from 272 to 3615 bp (Hyun and Clark, 1998; White et al., 1990; Cooley, 1992). Polymorphic Information Content (PIC) ranges from 0.070 (GCC 184) to 0.425 (GCC 193) (Table 3, Fig. 2a, b).

Genetic relationship among the accessions and cluster analysis: Genetic similarity estimates based on RAPD banding patterns were calculated using method of Jaccard’s coefficient analysis (Table 4). The similarity coefficient matrix generated was subjected to algorithm Unweighted Pair Group Method for Arithmetic Average (UPGMA) to generate clusters using NTSYS 2.02 pc program (Rohlf, 1998, 2000). The dendrogram showing relationship among various genotypes was constructed using these clusters (Fig. 1).

Table 4:	Jaccard’s similarity coefficient

Fig. 1:	Dendrogram showing relationship among ten fungal genotypes generated by UPGMA analysis based on single primers

Fig. 2(a-b):	RAPD analysis using primer No. (a) 137 and (b) 116

The Jaccard’s pairwise similarity coefficient values ranged from 0.58 (A2 and Jod-1, J2 and Jod-1) to 0.93 (J-1 and A-2) with an average of 0.711, for single primer based RAPD patterns (Table 4). The clusters constructed through NTSYS (2.02 pc) presented in the form of dendrogram has been shown in Fig. 1. The dendrogram has put all the genotypes in two major groups (group A and B) at 66.25% between group similarity. Group A consists A-1, A-2, J-2, J-1 and J-3 total 5 genotype having 79.0% within group similarity and group B consists Jod-1, Jod-2, Jod-3, Job-1 and Job-2, total 5 genotype with 73.5%. Zakaria et al. (2009) and Kubelik and Szabo (1995) also used RAPD PCR for characterization and diversification of fungal isolates and found that this technique efficiently divide isolates into clear groups according to their genetic differences.

CONCLUSION

The results suggested phylogenetic relationships among fungal isolates, collected from different regions of Rajasthan correlates with variation exist at pathogenicity.