Adjusting grape berry ripening to suit a changing climate: plant growth regulator-based solutions

Abstract

This project aimed to investigate the use of plant growth regulators to manipulate the timing of berry harvest to alleviate climate change-associated compressed harvest seasons with, benefits for winery intake scheduling and fruit quality. Trials in commercial vineyards were used to optimise spray treatments that significantly delayed veraison/ripening without negative effects on berry and wine composition and sensory character. These data will also be used to assist product registration for commercial use. Non-volatile developmental stage-specific berry metabolites were identified that could be used to develop an in-vineyard test to define the optimal timing of ripening delaying treatments.

Summary

The main aim of this project was to find a reliable method to delay grape harvest in a selective manner to combat some of the deleterious effects of climate change. A particular target was the hastening of ripening/harvest which has led to harvest season compression and associated problems with berry and wine quality. We know from previous work that some auxinic compounds can delay the onset of ripening and therefore harvest. Here we describe work conducted to develop the use of the auxin, 1-naphthalene acetic acid (NAA) as a commercially viable tool to delay ripening in a targeted manner. To broaden the scope, we tested two reagents liable to advance ripening/harvest and two to delay it. The other reagent that showed promise as a delayer of ripening, Ethrel, was shown to delay ripening but timing the spray treatment to be effective was particularly difficult and whole vine sprays showed that it hastened leaf senescence. NAA was the most effective in delaying harvest and did not damage leaves. A single, well-timed spraying of the fruiting zone is all that is required, reducing spray volume and reagent input.

Apart from delaying harvest, with the benefit of stretching the harvest season and alleviating winery intakes pressures, NAA-treatment has other positive benefits. NAA-treated berries tend to be larger at harvest as they undergo less shrinkage and appear more structurally sound. The rachis of NAA-treated bunches stays green and apparently healthier much longer than that of the untreated bunches. Although not formally proven, this could have an influence in mitigating bunch stem necrosis. Our experiments showed that the delaying effect on ripening/harvest may be through changes in gene expression and enzyme activity of cell wall modifying enzymes. These changes indicate that cellulose synthesis, required for the primary cell wall, was increased by NAA while the activity of enzymes that modify and degrade pectin, which is important for tissue structure, was reduced. The increased rigidity of the walls may contribute to the delaying of ripening by slowing berry expansion. 

NAA-induced harvest delay causes only modest changes in berry composition and wine flavour/aroma. We showed through full and partial peduncle girdling experiments that slowing or halting berry sugar accumulation does not, in many cases, greatly alter berry composition and wine character. The observed increase in rotundone levels in NAA ripening-delayed Shiraz fruit may be a direct effect of NAA on sesquiterpene synthesis, leading to increased peppery notes in the wine. Some of the changes in wine composition may be due to longer hang times e.g. the slightly lower IBMP concentration in NAA harvest delayed wine that may be a result of increased hang time. Lower levels of IBMP, which imparts an undesirable green character to Cabernet Sauvignon wines, would be seen as a positive outcome. Another interesting effect of the NAA harvest delaying treatment, possibly related to changes in berry cell walls, is that the treatment tends to increase Brix synchronicity (i.e. the individual berries tend to have more similar Brix compared to the untreated) and to decrease berry weight synchronicity. As the development of flavour/aroma compounds tends to be linked to sugar concentration, this might be an advantage in terms of wine flavour/aroma with fewer ‘too green’ and ‘too ripe’ flavours.

Apart from these effects on berries and wine, we spent considerable effort attending to the practical issues surrounding NAA use in industry. We have shown that the timing of the application is important to the treatment efficacy, with spraying as close as possible to veraison resulting in the longest delay in harvest. However, if the ripening in the vineyard is very asynchronous, this has to be taken into account, as once a berry commences ripening, NAA treatment cannot delay its harvest. Spraying too late in this situation may make the ripening even less synchronous. Timing of application is very important and in a changing climate, predicting veraison can be difficult. To this end we have performed an extensive analysis on the non-volatile berry metabolites that change in concentration during the late pre-veraison stage. A number of compounds were discovered that increase or decrease in concentration during this period. These markers of developmental stage could be developed into a test for the timing of the just pre-veraison spray application.

The compatibility of NAA with a selection of fungicides, an insecticide and different wetting agents was tested by combining each of these with NAA in a spray treatment, to assess its performance. There was no decrease in the effectiveness of NAA due to the presence of any of the fungicides or the insecticide and no particular wetting agent was markedly more efficacious, but this could be tested further on a larger scale trial.

It would be expected that the length of delay in harvest would be affected not only by the timing of application, but also by the concentration of NAA applied. Our initial studies were completed using 100 ppm NAA, however, we subsequently showed that this can be reduced to 50 ppm and still produce a significant and practically useful delay in harvest timing. The dose response curve shows that the delaying effect of NAA on ripening does not increase linearly at higher dosage levels. Reducing the amount of NAA applied reduces the quantity and cost of reagent applied and the concentration of any residues in the berries and wine. NAA is used to control various aspects of fruit development in apples and pears and is considered to be a soft/safe option. However, residues are always an issue in registering an agent for use in a different crop. Reducing the amount applied is therefore an important consideration. We have shown that the residue levels in grapes are below the level of detection and are low in wines. We have shown that the residues are not located on the berry surface. Studies of NAA residue in bottled wine have shown that it decreases quite rapidly with time. A registration pre-application advice has been commissioned from the APVMA, residues have been tested by an accredited laboratory, and the effort to progress this is ongoing.

In this project we have done extensive field and laboratory studies to develop a cheap, adaptable and reliable method to delay berry ripening and hence harvest time. Apart from expanding the harvest window and relieving pressure on winery intake  caused by compressed harvest seasons, and improving berry and wine quality through harvesting at the optimal time, there are a number of other potential economic benefits, as listed above.

There are other ways to tackle some of the agronomic issues faced by the grape and wine industries. Classical breeding is not currently popular, but genetic methods offer powerful tools to discover target genes for the new technique of gene editing. Targets for editing that would address events such as the timing of budburst, flowering, veraison, the rate of sugar accumulation during the ripening phase etc. will be immensely valuable in the near future and could be used to manipulate not just phenological stages but other parameters as described in this report. Identifying the gene targets for editing the desired traits is an essential part of the process. As a first step we phenotyped a wide range of characters in a breeding population over two years and have demonstrated that there is sufficient natural variation to make large adjustments to traits such as harvest timing. The phenotypic data can be used in future work to define the key genes in these processes and the precise changes in the genes that lead to the desired trait, for example, late budburst leading to a later harvest.