Determining thresholds for bunch rot tolerance in wine and detection of unwanted fungal aromas
Abstract
Grapevines are susceptible to several fungal diseases, including grey mould (Botrytis bunch rot), caused by the fungus Botrytis cinerea. This rot attacks grape bunches close to harvest resulting in a loss of yield and grape quality. Different techniques for grey mould estimation were evaluated for their applicability and compared with visual estimations with the naked eye.
These methods included: hyperspectral imaging, quantitative PCR (qPCR), gluconic acid analysis, ergosterol analysis, loop mediated amplification (LAMP) and quantification of Botrytis antigens. Measuring ergosterol allowed for the quantification of fungal biomass present in a grape sample however the method takes two days. Detection of Botrytis antigens using a lateral flow device (LFD) was a more rapid and specific method for quantifying grey mould in a matter of minutes. The performance of each of these analytical methods are discussed, especially in terms of their practical relevance to the wine sector.
Summary
Botrytis cinerea is a filamentous fungal organism that infects grape tissues, most notably grape berries after the onset of ripening. The fungus is responsible for the disease grey mould (Botrytis bunch rot) of grapes. In some vineyards and some growing seasons grey mould infection can be so severe that it can result in total crop failure when none of the vineyard is economically worth harvesting. Early detection is crucial so that a vineyard can be harvested before the level of grey mould reaches a point where it is no longer economically viable to harvest the crop. Detection and ideally quantification of grey mould contamination is not a clear-cut process especially as the disease is less obvious on dark skinned grape varieties.
Grape varieties that produce tightly packed clusters of grapes are more susceptible to grey mould particularly as a result of berry splitting as the individual berries expand. Consequently grey mould is often found within the interior of the grape bunch hidden from view. On first observation an apparently disease-free bunch can subsequently be found to be affected by grey mould once the bunch is pulled apart. Grey mould that is hidden from view within the bunch is difficult to detect using visual estimation with the naked eye. To overcome these issues, a number of experimental techniques to detect and ideally quantify grey mould in grapes were investigated. The methods included: hyperspectral imaging, quantitative PCR (qPCR), gluconic acid analysis, ergosterol analysis, loop mediated amplification (LAMP) and quantification of Botrytis antigens. To investigate the different ways in which grey mould contamination could be evaluated a series of experiments involving field samples of grape bunches affected with grey mould and grape berries and bunches inoculated in the laboratory were utilised. The varieties examined included Chardonnay, Cabernet Sauvignon, Semillon and Shiraz. Grapes were sourced from vineyards in the Riverina (NSW) and also for one experiment, from the Tamar Valley (TAS). The amount of apparent grey mould contamination was estimated by visual observation using the naked eye. Depending upon the experiment, the grapes were analysed for gluconic acid, the fungal sterol ergosterol, amount of Botrytis antigens, the presence of Botrytis by loop mediated isothermal amplification (LAMP) and quantification of Botrytis by the polymerase chain reaction (qPCR). In one experiment grey mould affected grape bunches were analysed by hyperspectral imaging. All of the techniques investigated for grey mould detection were of some value. Yet all had some form of limitation in terms of applicability for a wine industry setting.
As is the case with estimations of grey mould contamination with the naked eye, hyperspectral imaging does not provide information on the presence or absence of grey mould growth within the interior of a tightly packed cluster. Compared to visual estimations, hyperspectral imaging is more accurate for grey mould estimation when the fugal growth is largely on the exterior as opposed to the interior of the bunch. This is particularly the case with dark skinned grape varieties which are harder to assess with the naked eye.
Loop-mediated isothermal amplification (LAMP) for the molecular detection of Botrytis DNA was found to be a sensitive assay but lacked quantification. In some experiments false negative results were obtained with the LAMP assay. The LAMP assay is of value in detecting Botrytis rather than quantifying Botrytis although some inconsistencies were noted in some of the data with regards to false negatives. Unlike the LAMP assay, qPCR was quantitative, but variable results were obtained in some experiments that looked at whole bunches infected with grey mould. There was a weak correlation between the amount of Botrytis detected by qPCR and the perceived amount as determined either by visual estimation or by hyperspectral imaging of Cabernet sauvignon infected bunches. For individual grape berries that were inoculated and mixed with healthy berries, variability in the qPCR data was evident at infection levels of greater than 6 % on a per berry weight basis. The variability in the qPCR data can be explained by the nature of the fungus Botrytis, which is multinuclear and highly variable. This limits the usefulness of qPCR for Botrytis quantification as opposed to Botrytis detection.
Three commercially available enzyme based kits for gluconic acid were evaluated for Botrytis quantification and the results for all three kits were comparable. The kits can be used in a basic laboratory setting and results can be obtained in a matter of hours. Data for all of the kits examined were comparable. Gluconic acid was detected in all un-inoculated grape samples, and in some experiments it was not possible to differentiate between healthy grapes and grapes with low levels of grey mould contamination. Gluconic acid is also not specific for Botrytis, some of the other fungi associated with the rotting of grapes also produce gluconic acid. This limitation has to be weighed up against the fact that gluconic acid assays are relatively simple to use and the time frame to conduct the assays is of the order of minutes or hours as opposed to days. The amount of ergosterol detected in grape samples correlated with the amount of grey mould as determined by either visual estimations or hyperspectral imaging in a linear manner.
The amount of ergosterol present was used to calculate the amount of fungal biomass present on a dry weight basis which can be an advantage in some circumstances. Ergosterol is not unique to Botrytis, yeast species naturally residing on the surface of the grape berry also contain ergosterol and so for accurate estimates of Botrytis, the background ergosterol level on infected grapes should be subtracted. While ergosterol measurements provide an accurate estimate of the amount of grey mould present, the applicability to a field situation during vintage is limited by the time required for analysis, which is typically two days.
Taking less than one hour, detection of grey mould by measuring Botrytis antigens is one of the quickest and genus-specific methods for grey mould estimation. A commercially available kit provided by Mologic Ltd (Bedfordshire, UK), was evaluated and was found to be simple to use and require minimal access to laboratory consumables. The assay is specific for Botrytis and is not subjected to interference from other organisms. The amount of Botrytis antigens detected in experiments positively correlated with both ergosterol estimations and observations with the naked eye. The relationship between the amount of antigens detected and the actual amount of fungus present isn’t strictly linear, and follows a curved relationship. This however, can be overcome by diluting the sample. As a guide, a Signal Intensity (SI) reading using the Mologic Botrytis antigen lateral flow device (LFD) of 50 equated to approximately 0.1g dry mass of Botrytis cinerea per kg fresh weight of grape material when a 1/100 dilution of a grape homogenate was assayed.
In practice wineries may consider a combination of the methods described in this report. It is anticipated that better detection and quantification of grey mould will allow for more informed decisions to be made around harvest dates and a prediction of likely grape and wine quality.