Investigation of botrytis bunch rot in wine grapes: the disease cycle and symptom quantification

The research described in this thesis investigated various aspects of botrytis bunch rot (BBR), a disease of grapes caused by the fungus Botrytis cinerea Pers., including susceptibility of various growth stages of grapevines to infection, in planta growth of the fungus and quantification of BBR symptoms. This thesis is presented as a ‘thesis by publication’ and consists of four papers that have either been published or submitted for publication. Nitrate non-utilising (nit) strains were used to track B. cinerea infections throughout the growing season. It was demonstrated that infections occurring at any of three phenological stages of grapevines, flowering, pre-bunch closure (PBC) and veraison, could result in the establishment of latent infections and BBR symptoms at harvest. No single phenological stage led to higher latent incidence or harvest severity than any other stage. Vitis vinifera berries and flowers were inoculated with green fluorescent protein (GFP)-expressing B. cinerea and visualised using fluorescence microscopy, confocal microscopy and brightfield light microscopy. Extensive mycelial growth was observed on flowers, while very little was observed on berries collected and inoculated at PBC. In planta growth in mature berries was highly branched and intercellular, growing between the cuticle and the epidermis, with some enlarged hyphae penetrating deeper into the berry tissue. This type of growth is similar to that seen in other host species. Wounding was found to be important for infection, with visibly less hyphal growth observed in non-wounded, mature berries compared with wounded berries. Infrared spectroscopy was investigated as a potential BBR quantification method and alternative to visual estimation, currently the most commonly used method. Near-infrared (NIR; 800–2690 nm) and mid-infrared (mid-IR; 2510–25770 nm) spectroscopy methods were developed. The spectral range of 1260–1370 nm with Savitzky-Golay smoothing and first derivative pre-processing produced the PLS model with the highest predictive ability in the NIR spectral region. The spectral range of 8760–9520 nm with Savitzky-Golay smoothing and first derivative preprocessing produced the PLS model with the highest predictive ability in the mid-IR spectral region. Both methods demonstrated the potential for spectroscopic quantification of BBR. However, further calibration is required to increase the accuracy of these models, particularly at low BBR severities, if they are to be considered suitable for use in the vineyard. The accuracy of visual estimation was evaluated and compared with four other quantification methods: digital image analysis, NIR and mid-IR spectroscopy and quantitative PCR (qPCR). Visual estimation was found to vary significantly between assessors, suggesting that assessor training or correction is required to ensure adequate accuracy and repeatability. Image analysis software (RotBot) was developed to measure BBR severity from digital images of grape bunches using pixel hue. All quantification methods showed significant relationships with visual estimation. Quantitative PCR was the most accurate method but it is too labour intensive to be considered useful for routine use in the vineyard. RotBot was found to be the most suitable alternative to visual estimation, as it is an objective measure and requires no specialised equipment. This thesis provides new information on the timing of B. cinerea infections in vineyards, the in planta growth of B. cinerea in V. vinifera flowers and berries and describes a number of potential alternatives to visual estimation for BBR quantification. It is hoped that this information can be used to improve the control and management of BBR in commercial vineyards as well as enhancing future BBR research