Entomology and Plant Pathology Department, University of Tennessee, Knoxville, TN 37996, USA

Background and objectives
Seed treatment with Pseudomonas sp. MFl 03 or Bacillus sp. MB105 can suppress take-all of wheat, caused by Gaeumannomyces graminis var. tritici [1]. Carriers serve to attach the bacteria to seed, provide a carbon source, and protect the bacteria from dessication. The carrier may influence biocontrol activity by affecting population levels of the bacteria, and by the effect of the carbon source supplied by the carrier on antibiotic production by the biocontrol bacteria. This is important since antibiosis often plays a key role in take-all suppression. The objectives of this study were (i) to determine the effect of type and concentration of carrier on populations of seed-applied Pseudomonas sp. MF1 03, stored at 25C and at 4C over time, and (ii) to determine the effect of carrier on take-all suppression by Bacillus sp. MB105 and Pseudomonas sp. MF103.

Materials and methods
Cells of Pseudomonas sp. MF103, cultured on King's B medium, were combined (1:1, v/v) with four concentrations (0.5, 1, 2, 4%) of five carders (Polysurf 67, cellulose gum, gum arabic, methyl cellulose, and polyvinyl alcohol). 'Carrier-only' control solutions were prepared also. Wheat seeds were mixed with the suspensions, air-dried, and MF103 populations were determined. Treated seeds were stored at either 25 or 4C. MF103 populations were determined after storage of seed for 3 and 7 weeks at 25C, and after 5 and 12 weeks at 4C. To determine biocontrol efficacy, seeds were treated with five carriers (2% cellulose gum, 1% gum arabic, 4% methyl cellulose, 2% Polysurf 67, 1% polyvinyl alcohol) or with carders and either Bacillus sp. MB105 or Pseudomonas sp. MF103. Seedling disease assays were conducted as described previously [2].

Results and conclusions
The 4% rates of Polysurf 67 and cellulose gum were too viscous and were not used. Although MF103 populations declined over time at both storage temperatures with all carriers, populations remained significantly higher on seed stored at 4 than at 25C (log 6.18 vs log 5.37). Overall, populations were highest on seeds with methyl cellulose (log 5.92), than with polyvinyl alcohol (log 5.77), gum arabic (log 5.73), Polysurf 67 (log 5.73), or cellulose gum (log 5.72). There were significant interactions between storage time and temperature, and carrier type, and between carrier type and concentration, and storage time. On seeds stored at 25C, over time, MF103 populations were positively correlated with concentration of gum arabic and polyvinyl alcohol, and negatively correlated with concentration of cellulose gum. With methyl cellulose and Polysurf 67, MF103 populations declined rapidly at all concentrations. On seeds stored at 4C, MF103 populations remained higher over time with higher concentrations of Polysurf 67 and methyl cellulose, with intermediate concentrations of polyvinylalcohol and with lower concentrations of cellulose gum. MF103 populations declined uniformly with all concentrations of gum arabic at 4C. In the first seedling assay, disease pressure was low and there was a significant interaction between carrier and seed treatment. With MB105, disease severity was lower with gum arabic, methyl cellulose, and polyvinyl alcohol than with Polysurf 67 or cellulose gum, suggesting that carrier selection influenced the biocontrol activity of MB105. With the 'carrier only' control, disease severity was lower with cellulose gum and methyl cellulose than with Polysurf 67 or polyvinyl alcohol, while gum arabic was intermediate. This suggests that with low disease pressure, the carrier plays a role in disease suppression, perhaps by providing a carbon source for unidentified microorganisms in the rhizosphere and enabling them to out-compete the pathogen for sites on roots. With MF103, carrier selection had no effect on disease severity. In the second seedling assay, disease pressure was greater and there were significant differences in the effect of seed treatment on disease severity and shoot height. Regardless of carrier, treatment with MF103 resulted in less disease and greater shoot height than with the 'carrier only' control or MB105, suggesting that with greater disease pressure, the biocontrol agent is more important than the carrier. These results illustrate the importance of carrier selection in biocontrol studies.

1. Clark BL, Reeder RB, Ownley BH, 1995. Phytopathology 85, 11-91.
2. Ownley BH, Welter DM, Thomashow LS, 1992. Phytopathology 82, 178-184.