Grapevine rhizosphere bacteria: influence of diversity and function on two root diseases

The overall goal of this research was to determine what, if any, role grapevine rhizosphere bacteria play in the differing susceptibilities of New Zealand grown rootstocks to Cylindrocarpon black foot disease. The size and diversity of bacterial populations associated with the rhizospheres of grapevine rootstocks: 101-14, 5C, Schwarzmann and Riparia Gloire were evaluated. Dilution plating showed that total bacterial (P=0.012, P=0.005 for NA and KB, respectively) and fluorescent Pseudomonad (P=0.035) rhizosphere counts differed between rhizosphere and bulk soils but did not correlate with the differing susceptibilities of the rootstock varieties to black foot. No varietal differences were found for spore forming bacteria (P=0.201). SSCP banding patterns showed that species diversity was similar for most rootstocks, but that there were some differences in the composition of bacterial populations, probably attributable to vigour. Some functional characteristics of the bacteria isolated from the rhizospheres of the most and least susceptible rootstock varieties were assessed to investigate their potential to suppress the pathogen. In dual culture, bacteria from Riparia Gloire, 101-14 and the control soil all had little ability to antagonise Cylindrocarpon destructans. However, they differed in their degrees of activity for glucanase (P=0.000), protease (P=0.001) and siderophores (P=0.000). In all tests, bacterial isolates from the rhizosphere of 101-14 had the largest number of active isolates (P≤0.002); however, those from Riparia Gloire had the greatest degree of positive responses for the glucanase and siderophore assays. Bacterial isolates from the control soil produced few glucanases and no siderophores, but had the highest degree of protease activity. Bands excised and sequenced from SSCP gels frequently matched to other ‘uncultured bacteria’ in GenBank, as well as to other bacterial phyla, classes and genera commonly isolated from soil and sediment samples. These included members of the Firmicutes, Proteobacteria (α, δ, γ), Verrucomicrobia, Acidobacteria and Chromatiales. The pathogenicity of C. destructans and Fusarium oxysporum was investigated by inoculating soil containing wounded ungrafted rootstocks of 101-14, 5C, Schwarzmann and Riparia Gloire. Results indicated that F. oxysporum might be a more aggressive pathogen than C. destructans. Inoculation with F. oxysporum or C. destructans increased disease severity, P=0.018 and P=0.056, respectively at 0 cm. Rootstock variety influenced disease severity caused by C. destructans (P<0.001) and F. oxysporum (P=0.090), with rootstocks 101-14 and 5C being most susceptible to C. destructans, and Riparia Gloire and Schwarzmann most susceptible to F. oxysporum. There was also an indication that inoculation with one pathogen increased plant susceptibility to the other, with increased F. oxysporum infection in the C. destructans inoculated treatments of Riparia Gloire and Schwarzmann (P<0.05). The effect of carbohydrate stress (leaf trimming) and inoculation on C. destructans disease severity, incidence, and rootstock rhizosphere bacterial populations was evaluated by inoculating the soil containing one year old plants of Sauvignon Blanc scion wood grafted to rootstocks 101-14 and Schwarzmann. Disease severity and incidence was similar for both Schwarzmann (8.4% and 29.3%, respectively) and 101-14 (14.9% and 31.0%, respectively). When data for the moderate and no stress treatments were combined, because their effects were similar, the disease severity was significantly higher for the highly stressed plants(P=0.043). Stress did not influence disease incidence (P=0.551). Infection occurred in the non-inoculated plants, but disease severity was higher in the plants inoculated with C. destructans than those that were not. Root dry weight of highly stressed plants was lower than in both the moderately stressed (P=0.000) and unstressed plants (P=0.003). An interaction between inoculation and stress (P=0.031) showed that inoculated and highly stressed plants had the lowest root dry weight but there was no effect of rootstocks (P=0.062). There was no significant effect of carbohydrate stress (P=0.259) or inoculation (P=0.885) on shoot dry weight. SSCP banding patterns showed that bacterial diversity was generally similar between rootstocks, but stress and inoculation altered rhizosphere bacterial communities. This study has demonstrated that functionality of grapevine rhizosphere bacteria do differ between grapevine rootstock varieties that have different susceptibilities to black foot disease, but that this role needs to be further investigated if more accurate and practically relevant conclusions are to be drawn.