Integrated management of aflatoxin B₁ contamination of groundnut (Arachis hypogaea L.) with Burkholderia sp. and zimmu (Allium sativum L.×Allium cepa L.) intercropping

Abstract

The efficacy of intercropping of the medicinal plant zimmu (Allium sativum L.×Allium cepa L.) and an antagonistic bacterium Burkholderia sp. strain TNAU-1 was evaluated separately and in combination for control of Aspergillus flavus infection and aflatoxin B₁ contamination in groundnut (Arachis hypogaea L.) under field conditions. The Burkholderia sp. strain TNAU-1 inhibited the growth of all the toxigenic and non-toxigenic isolates of A. flavus in vitro and the inhibition zone ranged from 3.1–11.3 mm. A talc-based powder formulation of the Burkholderia sp was developed for field application. When groundnut seeds were treated with the formulation of Burkholderia sp., significant increases in root length, shoot length and seedling vigor were observed. Zimmu intercropping (once in every three rows of groundnut) alone significantly reduced the A. flavus infection and aflatoxin contamination in groundnut. Seed treatment at 10 g/kg or soil application at 2.5 kg/ha on 30, 45, 60 days after sowing with the formulation of Burkholderia sp. also resulted in significant reduction in the infection by A. flavus and aflatoxin B₁ contamination in kernels. The maximum percentage reduction in aflatoxin B₁ content was achieved by the combination of zimmu intercropping, seed treatment and soil application of Burkholderia sp. Seed treatment followed by soil application with the formulation of Burkholderia sp. recorded the highest pod yield in both the field trials. Zimmu intercropping or seed treatment or soil application with formulation of Burkholderia sp. significantly reduced the population of A. flavus in the soil. Application of Burkholderia sp. through seed or soil resulted in high colonization of Burkholderia sp. in rhizosphere soil, and the rhizosphere population increased with increase in the age of the crop. A colony PCR assay was developed using primers made to Burkholderia sp. specific sequence of the ribosomal DNA for accurate detection and confirmation of Burkholderia sp.

Keywords:

• aflatoxin
• Aspergillus flavus
• Arachis hypogaea
• biological control

Introduction

Aflatoxins are toxic secondary metabolites produced by certain strains of Aspergillus flavus Link ex Fries and Aspergillus parasiticus Speare (Woloshuk and Prieto 1998). These compounds are known to be potent animal hepatocarcinogens and are suspected to be carcinogens of humans (Williams et al. 2004). Currently, 18 different types of aflatoxins have been identified, with aflatoxin B₁, B₂, G₁, G₂, M₁, and M₂ being the most common. Of these, Aflatoxin B₁ (AFB₁) and G₁ (AFG₁) occur most frequently, with AFB₁ being the most potent (Mishra and Das 2003). Dietary intake of aflatoxin leads to the disease aflatoxicosis and the risk posed by aflatoxin depends on the level and type of aflatoxin in the diet, the strain of animal and its nutritional status (Williams et al. 2004). Due to concern for the potential effects of aflatoxins on human health even a low level of contamination is important and hence most countries have extensive monitoring programmes and legislation that restrict marketing of aflatoxin-contaminated grains and products (Van Egmond 1989). Most countries have established maximum tolerance levels for total aflatoxin ranging from 4–20 µg/kg. The US Food and Drug Administration has set an aflatoxin limit of 20 ppb for foods and for most feeds and feed ingredients. The European Union has enacted a very stringent aflatoxin tolerance threshold of 2 µg/kg aflatoxin B₁ and 4 µg/kg total aflatoxins for nuts and cereals for human consumption (Bankole and Adebanjo 2003).

Groundnut (Arachis hypogaea L.) with its geocarpic nature of fruit development is highly vulnerable to invasion by aflatoxigenic fungi and subsequent contamination by aflatoxins. The fungi invade groundnut seeds at pre-harvest stage, during post-harvest curing and drying and in storage (Goncalez et al. 2008). Pre-harvest aflatoxin contamination is a major problem in the tropics, which limits marketability and causes economic losses. Drought at the end of the cropping season facilitates aflatoxin contamination of groundnut (Wilson and Stansell 1983). Infection of groundnut by root-knot nematodes (Meloidogyne arenaria) can lead to an increase in aflatoxin kernels when the plants are subjected to drought stress during pod maturation (Timper et al. 2007).

A great deal of effort has been devoted to developing methods for reducing aflatoxin contamination in groundnut. A few A. flavus-resistant genotypes have been developed but the level of resistance in them was not sufficient to protect the kernels from aflatoxin contamination under all conditions (Waliyar et al. 1994; Waliyar et al. 2008; Nigam et al. 2009). The proliferation of A. flavus that occurs in groundnut kernels during storage is predominantly as a result of infection that occurred in the field (Dorner and Cole 2002). Hence pre-harvest elimination of aflatoxin is considered to be the most rational and economic approach to control aflatoxin in groundnut (Dorner 2008). Cultural practices, such as adjustments of sowing and harvesting dates, application of gypsum, maintenance of optimum water relations in the crop through irrigation, control of pod- and seed-feeding insects, optimum timing of digging and harvest, avoidance of mechanical damage during cultivation, harvest and post-harvest handling, and rapid post-harvest drying and maintenance of low temperature and relative humidity (RH) in storage can be effective to a certain extent in preventing aflatoxin contamination in groundnut (Xue et al. 2003). Although some of the fungicides are found effective in preventing the growth of A. flavus (Latif et al. 1985; Badii and Moss 1988; Brenneman et al. 1993), chemical control is neither practical nor economical. Sandosskumar et al. (2007) demonstrated that intercropping of groundnut with the medicinal plant zimmu (Allium sativum L.×Allium cepa L.) is capable of significantly reducing the population of A. flavus in soil and subsequent aflatoxin contamination in groundnut kernels. Several biocontrol agents have been reported to control aflatoxin in groundnut (Cotty 1990; Dorner et al. 2003; Dorner 2004). This approach is based on the use of non-toxigenic strains of A. flavus and/or A. parasiticus to compete with the naturally occurring toxin-producing strains in soil (Dorner et al. 1992; Chourasia and Sinha 1994; Dorner 2004; Dorner and Lamb 2006). However, adequate precautions must be taken while selecting the non-toxigenic strains for biological control. The selected strains must be incapable of producing toxins, and competitive under field conditions. Dorner et al. (1992) demonstrated that application of different non-aflatoxigenic strains of A. parasiticus to soil reduced pre-harvest aflatoxin contamination when peanuts were exposed to late-season drought and temperature stress. However, one of those strains (NRRL 18991) was subsequently shown to produce o-methylsterigmatocystin (OMST), the immediate biosynthetic precursor to aflatoxin B₁ (Brown et al. 1999). An ultraviolet (UV)-induced mutant of the strain NRRL 18786, which was effective in reducing aflatoxin contamination (Dorner et al. 1992), was found to produce versicolorin A. Tran-Dinh et al. (1999) suggested that, if genetic exchange can occur, the introduction of a more aggressive non-toxigenic isolate as a biocontrol agent could result in a recombinant with both increased aggressiveness and toxigenicity. Hence, plant growth-/soil health-promoting native antagonistic microorganisms offer great potential as alternatives to chemical pesticides, which are frequently associated with environmental pollution. Vijayasamundeeswari et al. (2007) isolated a bacterial strain from corn kernels that inhibit the growth of A. flavus and promote plant growth. The antagonistic bacterial strain was identified as Burkholderia sp. based on phenotypic and biochemical characteristics and 16S rRNA gene sequence analysis. In the present study, a talc-based powder formulation of Burkholderia sp. was developed and its efficacy in the control of A. flavus infection and aflatoxin contamination in groundnut either alone or in combination with zimmu intercropping was demonstrated under field conditions.

Materials and methods

Fungal culture

The fungus, A. flavus, was isolated from infected groundnut (Arachis hypogaea L.) kernels and maintained on potato dextrose agar (PDA) (Difco Laboratories, Detroit, MI, USA) medium under laboratory conditions. The Aspergillus cultures were identified up to species level using a taxonomic key and species descriptions (Singh et al. 1991). The isolates were analyzed for their aflatoxin B₁ production potential in potato dextrose broth following the method of Karthikeyan et al. (2009).

Antagonist

Burkholderia sp. strain TNAU-1, which is known to inhibit the growth of Aspergillus flavus and other phytopathogenic fungi in vitro (Vijayasamundeeswari et al. 2007) was obtained from the culture collection of the Department of Plant Pathology, Tamil Nadu Agricultural University, Coimbatore, India. The antagonistic bacterium was Gram-negative, aerobic, straight rod, further characterized by positive reactions in motility, gelatin hydrolysis, catalase reaction, oxidase reaction, and utilization of galactose, glucose, inositol and sorbitol, and negative reactions in production of fluorescent pigments and starch hydrolysis. It was found to belong to the genus Burkholderia on the basis of phenotypic characteristics and phylogenetic analysis of 16S rRNA gene sequences using CLUSTAL-W multiple sequence alignment program (Thompson et al. 1994). The antagonist Burkholderia sp. strain TNAU-1 was compatible with other beneficial rhizobacteria including Bacillus megaterium var. phosphaticum strain PBS, Rhizobium strain TNAU14, Glucanoacetobacter diazotrophicus, Azospirillum brasiliense strain 204, Azospirillum lipoferum strain SP7. The antagonist was cultured in Petri plates on PDA medium under laboratory conditions.

In vitro screening of Burkholderia sp. strain TNAU-1 against A. flavus

In vitro inhibition of mycelial growth of A. flavus was assayed by dual culture technique. Burkholderia sp. strain TNAU-1 was streaked on a Petri plate (1 cm from the edge) containing PDA medium. Simultaneously, an 8-mm-diameter mycelial disc cut from a seven-day-old culture of A. flavus was placed on the center of the Petri plate. PDA medium inoculated with the fungus alone served as control. The plates were incubated at room temperature (28±2°C) for seven days. At the end of incubation period, the zone of inhibition (mm) was recorded by measuring the distance between the edges of the fungal mycelium and the antagonistic bacterium. Three replications were carried out for each treatment. The experiment was repeated once with three replications for each treatment.

Development of talc-based powder formulation of Burkholderia sp. strain TNAU-1

All manipulations were carried out under sterile conditions. Talc-based powder formulation of Burkholderia sp. strain TNAU-1 was developed as described by Vidhyasekaran and Muthamilan (1995). Ten grams of carboxymethylcellulose was mixed with 1 kg of talc (hydrated magnesium silicate) and the pH was adjusted to 7.0 by adding calcium carbonate. The mixture was then autoclaved at 121°C for 30 min for two consecutive days. Burkholderia sp. strain TNAU-1 was grown in potato dextrose broth for 48 h at 28±2°C on a rotary shaker. Four hundred milliliters of bacterial suspension containing 9×10⁸ CFU ml^-1 was added to 1 kg of the talc mixture and thoroughly mixed. The formulation was packed with 35% moisture content in polythene bags, sealed and stored at room temperature (28±2°C). The survival of Burkholderia sp. in the prepared formulation was evaluated at different time intervals by dilution plating.

Seedling vigor test

The plant growth promoting activity of Burkholderia sp. strain TNAU-1 on groundnut was tested by using the roll towel method (International Seed Testing Association 1996). The seeds were treated with the formulation at 10 g/kg and incubated for 6 h at room temperature. The seeds were then placed on coarse blotter paper sheets and covered with a moistened blotter and rolled. The roll was kept on a butter paper sheet and rolled as a bundle and incubated in a growth chamber at 25°C with 80% RH. After 15 d of incubation, percent of germination was recorded and root length and shoot length were measured. Vigor index was calculated by multiplying percent plant stand with sum of shoot and root length.

Field trials

Two field trials were conducted during July–October, 2007, one at the Agricultural Research Station, Bhavanisagar, Tamil Nadu, India, and another trial at the Agricultural Research Station, Vridhachalam, Tamil Nadu, India. The susceptible groundnut cultivars, CO-3 (Bunch type; duration 115–120 days) and VRI-2 (Bunch type; duration 100–105 days) obtained from the Department of Oilseeds, Tamil Nadu Agricultural University, Coimbatore, Tamil Nadu, were used in the field trials. The trials were laid out in a randomized block design with seven treatments and three replications with a plot size of 5×4 m and with a spacing of 30×15 cm. The treatments comprised as follows:

1. Zimmu intercropping (once in every three rows of groundnut, one week prior to sowing of groundnut with a plant to plant spacing of 45 cm);

2. Seed treatment with talc-based formulation of Burkholderia sp. strain TNAU-1 at the rate of 10 g/kg;

3. Soil application of talc-based formulation of Burkholderia sp. strain TNAU-1 at the rate of 2.5 kg/ha on 30, 45, 60 days after sowing (DAS);

4. Seed treatment (10 g/kg) followed by soil application (2.5 kg/ha) with talc-based formulation of Burkholderia sp. on 30, 45, 60 DAS;

5. Zimmu intercropping + seed treatment (10 g/kg) + soil application (2.5 kg/ha) with talc-based formulation of Burkholderia sp. on 30, 45, 60 DAS;

6. Seed treatment with carbendazim at the rate of 2 g/kg;

7. Untreated control.

The population of A. flavus and Burkholderia sp. in rhizosphere soil, kernel infection by A. flavus and aflatoxin B₁ content in groundnut kernels were determined as described below. Pod yield and other growth parameters were also recorded. Oil content in groundnut kernels was estimated using Oxford 4000 NMR analyzer (Oxford Instruments, UK).

Assessment of A. flavus and Burkholderia sp. population in rhizosphere soil

The initial populations of A. flavus as well as population of A. flavus and Burkholderia sp. after 60 and 110 days (prior to harvest) after sowing were determined by dilution plate technique. Soil samples were collected with the help of an auger at 5 cm soil depth (Nahdi 1996). Four grams of soil sample were added to 100 ml of sterile distilled water in a 250 ml conical flask and shaken for 10 min on a rotary shaker at 150 rpm. From this suspension, a series of dilutions were made. One ml of the soil suspension was mixed with 19 ml of sterilized PDA medium and poured in to a sterile 9-mm-diameter Petri dish. The plates were incubated at room temperature (28±2°C) for five to seven days and A. flavus or Burkholderia sp. colonies that developed were counted.

Detection of Burkholderia sp. by colony PCR

Burkholderia sp. specific forward primer (5′ CGAACGGGTGAGTAATAC 3′) and reverse primer (5′ GCTGGCACGTAGTTAGC3′) for PCR assays were designed based on the nucleotide sequence (GenBank accession number EU560426) and synthesized by Operon Company (Operon Biotechnologies, Cologne, Germany) in salt-free status. PCR was under taken in 20 a µl volume consisting of 5 mM each dNTPs, 20 pmol of each primers, 0.5 U of Taq DNA polymerase and single loop of bacterial colony. The reaction was carried out in a Palm-Cycler Model CGI-960 (Corbett Research, Mortlake, NSW, Australia) programmed with initial denaturation at 94°C for 3 min, followed by 35 cycles of 94°C for 1 min, 59°C for 1 min, 72°C for 1 min, and a final extension step at 72°C for 10 min. Following amplification, 10 µl of each PCR product was electrophoresed on 2% agarose gel in TAE buffer. The DNA fragments in the gel were stained with ethidium bromide, visualized under ultraviolet light, and recorded with an AlphaImager 2000 (Alpha Innotech, San Leandro, CA).

Assessment of kernel infection by A. flavus

Plants were harvested from each plot and pods were removed and dried in the shade for three to four days. After drying, pods were hand-shelled and 100 seeds were selected randomly and tested in the laboratory to determine the percentage of infection by A. flavus. Seeds were surface sterilized by soaking in 0.1% aqueous solution of mercuric chloride for 3 min, rinsed with sterile distilled water and placed on filter paper in 10 cm-diameter sterile Petri dishes and incubated at 25°C. To maintain high humidity 1–2 ml of sterile distilled water was added everyday during the first five days. After six days of incubation, the number of seeds contaminated by A. flavus was counted (Waliyar et al. 1994).

Analysis of aflatoxin B₁ content in groundnut

Pod samples of approximately 1 kg were drawn from each plot and the pods were thoroughly mixed. From each 1 kg sample, three 100 g sub-samples were drawn and shelled. Aflatoxin B₁ content in the samples was determined by indirect competitive ELISA following the method of Reddy et al. (2001).

Statistical analysis

The data were statistically analyzed by Duncan's Multiple Range Test (DMRT). The package used for analysis was IRRI STAT version 92-1 developed by the International Rice Research Institute Biometrics Unit, Philippines.

Results

The Burkholderia sp. strain TNAU-1 inhibited the growth of all the toxigenic and non-toxigenic isolates of A. flavus in dual culture on PDA medium and the inhibition zone ranged from 3.1 mm (AFG47) to 11.3 mm (AFG1) (Table 1). A talc-based powder formulation of the Burkholderia sp. strain TNAU-1 was developed for field application.

Table 1 In vitro antagonistic activity of Burkholderia sp. strain TNAU-1 against aflatoxigenic and non-aflatoxigenic isolates of Aspergillus flavus.

S. No.	Isolate	Toxigenicity	Inhibition zone (mm)
1	AFG 1	Toxigenic	11.3^a
2	AFG 13	Non-toxigenic	4.8^jk
3	AFG 15	Non-toxigenic	5.0^ij
4	AFG 17	Non-toxigenic	8.2^d
5	AFG 19	Non-toxigenic	7.4^e
6	AFG 20	Non-toxigenic	9.6^b
7	AFG 21	Toxigenic	6.1^g
8	AFG 23	Non-toxigenic	5.6^h
9	AFG 24	Non-toxigenic	6.7^f
10	AFG 29	Toxigenic	3.8^l
11	AFG 32	Toxigenic	4.8^jk
12	AFG 40	Toxigenic	7.2^e
13	AFG 43	Non-toxigenic	8.9^c
14	AFG 46	Toxigenic	5.3^hi
15	AFG 47	Toxigenic	3.1^m
16	AFG 48	Non-toxigenic	4.4^k
17	AFG 49	Toxigenic	4.9^ij
18	AFG 52	Non-toxigenic	6.2^g
Data are mean of three replications; Means followed by the same letter in a column are not significantly different (p=0.05) by Duncan's Multiple Range Test (DMRT).

The survival of Burkholderia sp. strain TNAU-1 in the prepared formulation was evaluated by dilution plating. The bacterial strain TNAU-1 survived well in the talc-based powder formulation during up to one month of storage under room temperature. The formulation had a bacterial load of 15.9×10⁸ CFU/g at the time of preparation (Table 2). The number of living bacteria in the talc formulation was 8.8×10⁸ CFU/g after one month of storage at room temperature. During the first 30 days after preparation the population of Burkholderia sp. strain TNAU-1 in the formulation dropped approximately 45%.

Table 2. Population of Burkholderia sp. strain TNAU-1 in the talc-based powder formulation.

Days after preparation	Population of Burkholderia sp. strain TNAU-1 (×10^8 CFU/g)
0	15.9^a
30	8.8^b
60	3.2^c
90	0.1^d
Data are mean of four replications; Means followed by the same letter in a column are not significantly different (p=0.05) by DMRT.

The plant growth promoting activity of Burkholderia sp. strain TNAU-1 on groundnut was tested by using the roll towel method. When groundnut seeds were treated with the formulation of Burkholderia sp. strain TNAU-1, significant increases in root length, shoot length and seedling vigor were observed (Table 3).

Table 3. Effect of seed bacterization with the formulation of Burkholderia sp. strain TNAU-1 on seed germination and seedling vigor of groundnut (cv. CO-3).

Treatment	Germination (%)	Shoot length (cm)	Root length (cm)	Vigor index
Treated	96.4 (81.1)^a	9.9^a	18.3^a	2712.0
Untreated control	87.1 (69.0)^b	8.6^b	14.1^b	1856.7
Data are mean of eight replications; Figures in parentheses are arcsine transformed values; Data followed by the same letter in a column are not significantly different (p=0.05) from each other according to DMRT.

Kernel infection by A. flavus

The effect of seed treatment or soil application with Burkholderia sp. strain TNAU-1 or zimmu intercropping in controlling A. flavus infection in groundnut is shown in Tables 4 and 5. The maximum kernel infection (65%) by A. flavus was recorded when untreated groundnut seeds were sown in the A. flavus infested soil. Zimmu intercropping alone significantly reduced the A. flavus infection in groundnut. In the treatment of zimmu intercropping, colonization of groundnut kernels by A. flavus was not observed in Trial I, and 2% infection was observed in Trial II. Seed treatment or soil application with Burkholderia sp. strain TNAU-1 also resulted in significant reduction in A. flavus infection over control.

Table 4. Effect of seed treatment or soil application with the formulation of Burkholderia sp. strain TNAU-1 or intercropping with zimmu on Aspergillus flavus infection, aflatoxin B₁ concentration, oil content and pod yield of groundnut (cv. VRI-2) under field conditions (Trial I).

Treatment	A. flavus infection (%)	Aflatoxin B1 content ( g/kg)	Oil content%	100 seed wt (g)	Pod yield (kg/ha)
Zimmu intercropping	0 (0.6)	1.6^a	42.1^a	41.4^c	1596.6^e
Seed treatment with Burkholderia sp. (10 g/kg)	5 (13.3)	3.8^a	42.5^a	41.4^c	1731.6^b
Soil application with Burkholderia sp. (2.5 kg/ha on 30, 45, 60 DAS)	10 (18.3)	5.6^a	42.3^a	41.3^c	1815.0^a
Seed treatment + soil application with Burkholderia sp.	10 (18.3)	3.2^a	42.9^a	41.8^b	1821.6^a
Seed treatment + soil application with Burkholderia sp.+Zimmu intercropping	0 (0.6)	0.6^a	42.4^a	42.2^a	1660.0^d
Seed treatment with carbendazim (2 g/kg seed)	50 (45.0)	31.5^b	42.3^a	41.8^b	1696.6^c
Untreated control	65 (53.9)	71.4^c	42.3^a	41.4^c	1255.0^f
Zimmu was intercropped once in every three rows of groundnut, one week prior to sowing of groundnut. Powder formulation of Burkholderia sp. strain TNAU-1 was applied as seed treatment (10 g/kg seed) or soil application (2.5 kg/ha on 30, 45, 60 DAS) or as seed treatment followed by soil application. Figures in parentheses are arcsine transformed values. The data are mean of three replications. Means within a column followed by a common letter are not significantly different (p=0.05) by DMRT.

Table 5. Effect of seed treatment or soil application with the formulation of Burkholderia sp. strain TNAU-1 or intercropping with zimmu on Aspergillus flavus infection, aflatoxin B₁ concentration, oil content and pod yield of groundnut (cv. CO-3) under field conditions (Trial II).

Treatment	A. flavus infection (%)	Aflatoxin B1 content ( g/kg)	Oil content%	100 seed wt (g)	Pod yield (kg/ha)
Zimmu intercropping	2 (7.9)	3.0^a	42.4^a	39.2^a	2625.0^c
Seed treatment with Burkholderia sp. (10 g/kg)	10 (18.4)	4.2^a	42.2^a	39.2^a	3128.3^ab
Soil application with Burkholderia sp. (2.5 kg/ha on 30, 45, 60 DAS)	20 (26.5)	8.7^a	43.6^a	39.2^a	3150.0^ab
Seed treatment + soil application with Burkholderia sp.	15 (23.0)	5.8^a	41.4^a	39.0^a	3223.3^a
Seed treatment + soil application with Burkholderia sp.+Zimmu intercropping	2 (7.9)	2.9^a	43.2^a	39.3^a	2801.6^bc
Seed treatment with carbendazim (2 g/kg seed)	55 (48.1)	29.1^b	43.6^a	39.1^a	3015.0^ab
Untreated control	65 (53.9)	92.4^c	41.5^a	38.9^a	2975.0^ab
Zimmu was intercropped once in every three rows of groundnut, one week prior to sowing of groundnut. Powder formulation of Burkholderia sp. strain TNAU-1 was applied as seed treatment (10 g/kg seed) or soil application (2.5 kg/ha on 30, 45, 60 DAS) or as seed treatment followed by soil application. Figures in parentheses are arcsine transformed values. The data are mean of three replications. Means within a column followed by a common letter are not significantly different (p=0.05) by DMRT.

Aflatoxin contamination

The maximum AFB₁ content of 71.4 and 92.4 g/kg in Trial I and Trial II, respectively, was recorded when untreated groundnut seeds were sown in the A. flavus infested soil (Tables 4 and 5). All the treatments showed significant reduction in AFB₁ content in groundnut over control. Zimmu intercropping alone significantly reduced the AFB₁ content (97% reduction) in groundnut compared to untreated control in both the trials. Seed treatment with Burkholderia sp. strain TNAU–1 resulted in 95% reduction in aflatoxin contamination over control in both the trials. Soil application with Burkholderia sp. strain TNAU-1 powder formulation also exhibited significant reduction (90–92% reduction) in AFB₁ content in groundnut kernels over control. AFB₁ content was significantly reduced in the treatment of zimmu intercropping + seed treatment + soil application of Burkholderia sp. strain TNAU-1 by 99% in Trial I and 97% in Trial II.

Pod yield

Control plots recorded pod yields of 1255 kg/ha and 2975 kg/ha in Trial I and Trial II, respectively (Tables 4 and 5). Seed treatment followed by soil application with the powder formulation of Burkholderia sp. strain TNAU-1 recorded the highest pod yield of 1821 kg/ha and 3223 kg/ha in Trial I and Trial II, respectively. The pod yield was less in zimmu intercropped plots compared to that in control in Trial II, probably because of reduction in population size of groundnut plants. No significant differences in oil content and 100 seed weight were observed between the treatments (Tables 4 and 5).

Population of A. flavus in rhizosphere soil

Treatment of groundnut with Burkholderia sp. strain TNAU-1 resulted in significant reduction in the population density of A. flavus in rhizosphere soil. At the time of sowing, untreated control soil contained A. flavus population of 2×10³ and 1.1×10³ in Trial I and Trial II, respectively (Tables 6 and 7). Zimmu intercropping or seed treatment or soil application or seed treatment followed by soil application with Burkholderia sp. strain TNAU-1 completely suppressed the population of A. flavus 60 DAS and at the time of harvest (110 DAS).

Table 6. Population of A. flavus and Burkholderia sp. in rhizosphere soils sampled at different time intervals after treatment (Trial I).

	Population (×10^3 CFU/g soil)
	60 DAS		110 DAS
Treatment	Burkholderia sp.	A. flavus	Burkholderia sp.	A. flavus
Zimmu intercropping	0^a	0^a	0^a	0^a
Seed treatment with Burkholderia sp.	109^d	0^a	138^d	0^a
Soil application with Burkholderia sp.	143^c	0^a	151^c	0^a
Seed + soil application with Burkholderia sp.	156^b	0^a	158^bc	0^a
Seed treatment + soil application with Burkholderia sp.+Zimmu intercropping	154^b	0^a	162^b	0^a
Seed treatment with carbendazim	0^a	5^b	0^a	6^c
Untreated control	0^a	13^c	0^a	15^c
Initial population of A. flavus at the time of planting was 2×10^3/g soil. Data are mean of three replications. Data followed by the same letter in a column are not significantly different (p=0.05) by DMRT.

Table 7. Population of A. flavus and Burkholderia sp. in rhizosphere soils sampled at different time intervals after treatment (Trial II).

	Population (×10^3 CFU/g soil)
	60 DAS		110 DAS
Treatment	Burkholderia sp.	A. flavus	Burkholderia sp.	A. flavus
Zimmu intercropping	0^a	0^a	0^a	0^a
Seed treatment with Burkholderia sp.	81^d	0^a	101^d	0^a
Soil application with Burkholderia sp.	105^c	0^a	134^c	0^a
Seed + soil application with Burkholderia sp.	113^b	0^a	154^b	0^a
Seed treatment + soil application with Burkholderia sp.+Zimmu intercropping	103^c	0^a	153^b	0^a
Seed treatment with carbendazim	0^a	2^b	0^a	3^b
Untreated control	0^a	6^c	0^a	6^c
Initial population of A. flavus at the time of planting was 1.1×10^3/g soil. Data are mean of three replications. Data followed by the same letter in a column are not significantly different (p=0.05) by DMRT.

Population of Burkholderia sp. in rhizosphere soil

Application of Burkholderia sp. strain TNAU-1 through seed or soil resulted in high colonization of Burkholderia sp. in rhizosphere soil (Tables 6 and 7). A colony PCR assay was developed for accurate detection and confirmation of Burkholderia sp. Primers made to Burkholderia sp. specific sequence of the ribosomal DNA were used to amplify the internal transcribed Spacer (ITS) regions from randomly picked bacterial colonies. The Burkholderia sp. specific primers resulted in the amplification of a 417-bp fragment from all the Burkholderia sp. strain TNAU-1 colonies in colony PCR (Figure 1). There was no cross-reactivity with purified genomic DNA from any other bacterial isolates that were tested: Xanthomonas oryzae pv. oryzae, Pseudomonas fluorescens, Bacillus subtilis and Azospirillum brasiliense. The high sensitivity of PCR enables quick and accurate detection of Burkholderia sp. in soil.

Figure 1.  Detection of Burkholderia sp. by polymerase chain reaction. Lane 1, DNA ladder; Lanes 2–7, Burkholderia sp. strain TNAU-1 colonies from the experimental field; Lane 8, Xanthomonas oryzae pv. oryzae; Lane 9, Pseudomonas fluorescens; Lane 10, Bacillus subtilis; Lane 11, Azospirillum brasiliense; Lane 12, Purified DNA from Burkholderia sp. strain TNAU-1.

Discussion

The process of infection of peanut kernels by A. flavus and subsequent production of aflatoxin is quite complex and different from other soil-borne diseases (Thakur et al. 2003). The level of A. flavus population in soil is important in relation to infection of groundnut pods and seeds. These levels vary greatly from one region to the other (Griffin 1970) and are influenced by soil type and crop rotation (Joffee and Lisker 1970; Nahdi 1996). Nahdi (1996) reported that populations of A. flavus were significantly higher in the pod-zone than in the field soil and increased with maturation of the crop, and significantly high populations of A. flavus were recorded at 0–5 cm depth than at 5–10 cm. The results of the present study indicated that zimmu intercropping alone significantly reduced A. flavus infection and aflatoxin contamination in groundnut kernels. There was a sharp decline in the population of A. flavus in the rhizosphere soil collected from groundnut plants intercropped with zimmu. The root exudates released from zimmu in the rhizosphere might have resulted in inhibition of development of A. flavus population and subsequent afaltoxin contamination (Sandosskumar et al. 2007). Such allelopathic effects of root exudates of plants have also been reported by other authors (Michel et al. 1996; Yu 1999; Agrawal and Tripathi 2003). Michel et al. (1996) demonstrated that planting cowpea between tomato reduced the bacterial wilt in tomato in Taiwan. Adhikari and Bansyat (1998) observed significant reduction in bacterial wilt in intercropping/rotation with maize, okra, cowpea and resistant line of tomato (CL1131) in Nepal. Agrawal and Tripathi (2003) reported that the population of Fusarium udum decreased in plots of pigeonpea intercropped with sorghum, and the reduction in wilt incidence in pigeonpea was attributed to fungitoxic exudates secreted by sorghum roots. Yu (1999) demonstrated that intercropping tomato with Chinese Chive (Allium tuberosum) plants significantly reduced the population of Pseudomonas solanacearum and suppressed the occurrence of bacterial wilt of tomato.

Significant reduction in the population of A. flavus in soil was also recorded in groundnut when the formulation of Burkholderia sp. strain TNAU-1 was applied through seed or soil. Species belonging to the genus Burkholderia have been widely used as biocontrol agents against many phytopathogenic fungi, such as Pythium aphanidermatum, Pythium ultimum, Fusarium sp., Phytophthora capsici, Botrytis cinerea and Rhizoctonia solani (Hebbar et al. 1998; Heydari and Misaghi 1998; Cain et al. 2000; Li et al. 2002). Watanabe et al. (2000) reported that Burkholderia sp. isolate 87-11 obtained from basidiospore of Lentinus lepideus showed antagonistic activity against Pythium aphanidermatum and Rhizoctonia solani. In the present study, seed treatment or soil application with Burkholderia sp. strain TNAU-1 completely suppressed the population of A. flavus in rhizosphere soil 60 DAS and at the time of harvest. The maximum percentage reduction in A. flavus and AFB₁ content in groundnut kernels was achieved by the combination of zimmu intercropping, seed treatment and soil application of Burkholderia sp. strain TNAU-1. The reduction in A. flavus infection and subsequent aflatoxin production in groundnut by application of Burkholderia sp. strain TNAU-1 may be attributable to direct antagonistic effect of Burkholderia sp. strain TNAU-1 on A. flavus. PCR technique has been widely used to detect and monitor a variety of microorganisms in the environment (Steffan and Atlas 1991). In the present study, a colony PCR assay was developed for accurate detection and confirmation of Burkholderia sp. The Burkholderia sp. specific primers resulted in the amplification of a 417-bp fragment from all the Burkholderia sp. colonies from the experimental plot in colony PCR. DNA from other bacterial species tested did not produce amplification products with the Burkholderia sp. specific primers. The DNA primers used in the present study are highly specific for detection of Burkholderia sp.

The results of the present study also indicated that seed treatment followed by soil application with the powder formulation of Burkholderia sp. strain TNAU-1 recorded the highest pod yield. Several strains of plant growth-promoting rhizobacteria (PGPR) are known to stimulate plant growth by secreting auxins, gibberellins and cytokinins (Dubeikovsky et al. 1993) and by suppression of deleterious microorganisms (Gamliel and Katan 1993). Several authors reported increased crop yield due to application of Burkholderia sp. even in the absence of the pathogen (Hebbar et al. 1998; Heydari and Misaghi 1998; Li et al. 2002). The increase in groundnut pod yield due to application of Burkholderia sp. strain TNAU-1 in the present study may be due to plant growth promoting characteristics of Burkholderia sp. strain TNAU-1. The pod yield was less in the zimmu intercropped plots compared to that in control in Trial II, probably because of reduction in population size of groundnut plants.

Due to the increasingly popular trends towards environment friendly ‘organic farming’ the necessity of finding acceptable ‘natural’ effective alternatives to ‘artificial’ chemical pesticides against plant pathogens is of great interest. Biological control of aflatoxin contamination is a promising component of the integrated management of aflatoxin contamination in groundnut. The present study clearly indicates the usefulness of a talc-based powder formulation of Burkholderia sp. strain TNAU-1 for control of A. flavus infection and aflatoxin contamination in groundnut. The results of the present study also showed that intercropping with zimmu alone is capable of significantly reducing the population of A. flavus in soil and subsequent infection and aflatoxin production in groundnut kernels. The biocontrol agent Burkholderia sp. strain TNAU-1 could be used in combination with intercropping with zimmu to provide greater protection of groundnut to kernel infection by A. flavus and aflatoxin contamination. Since zimmu is much used for its medicinal importance and also in certain culinary preparations in India, its use as intercrop may fetch additional income for the farmers.

Acknowledgements

This study was supported by the Indian Council of Agricultural Research, New Delhi.