PHOTO: Piet Goussard.
A number of viruses have been reported from vines with grapevine leafroll disease (GLD) and are known as grapevine leafroll-associated viruses. Amongst these, grapevine leafroll-associated virus 3 (GLRaV-3) appears to be the most prevalent and widespread virus and this certainly is the case in South Africa.
Commercial grapes are primarily cultivars of Vitis vinifera, V. labrusca, Muscadinia rotundifolia, V. amurensis and several interspecific hybrids. Worldwide, grapevines are prone to various virus diseases of which GLD is the most common and widespread (Figure 1). It is economically important as it reduces plant vigour and longevity, fruit yield and quality.
FIGURE 1. Prevalence of grapevine leafroll disease in many vineyards in South Africa and internationally.
GLD can be successfully controlled using an integrated control strategy which includes producing and planting certified, virus-tested propagation material, vector number and dispersal management, and rogueing of infected vines. Detection of GLD for the purpose of propagation of healthy vines in certification schemes and rogueing infected vines is easily achieved in a number of red cultivars where symptoms of GLD are obvious in autumn and infected plants easily identified (Figure 2). This is more difficult in white cultivars, where symptoms are not obvious (Figure 3) and where laboratory-based virus detection techniques have to be employed to detect the viruses associated with the disease.
FIGURE 2. Obvious symptoms of leafroll disease in a number of red cultivars of Vitis vinifera (Merlot, Cabernet Sauvignon, Shiraz and Pinotage), in this case autumn 2017, on Vergelegen in a vineyard where infected vines are rogued annually.
FIGURE 3. Grapevine leafroll disease is difficult to detect in white cultivars. In the picture above, Chardonnay, which is one of the white cultivars that shows the disease best, displays the symptoms which gave grapevine leafroll disease its name.
Adding complexity to certification schemes, scions of V. vinifera are grafted (Figure 4) onto a number of American Vitis species rootstocks in countries, such as South Africa, where phylloxera (Daktulosphaira vitifoliae) occurs. This is required as V. vinifera vines on their own roots are attacked by this aphid-like insect.
FIGURE 4. Grafting of the Vitis vinifera scion onto a rootstock via omega grafting, as a means of managing the effect of phylloxera.
Many rootstocks are asymptomatic hosts of GLD (Figure 5) and require virus specific laboratory-based detection methods to test for virus infection. The reliability of detection of GLRaV-3 in rootstocks in certification schemes by PCR has, however, not been fully assessed anywhere.
FIGURE 5. Rootstocks do not show symptoms of grapevine leafroll disease.
In this study, we assessed the detection by the very sensitive laboratory technique, reverse-transcriptase polymerase chain reaction (RT-PCR) of GLRaV-3 in individual vines of the most widely utilised rootstocks in South Africa compared with their corresponding scions (Figure 6). Most of the initial work was done on Richter 99 (Vitis berlandieri X Vitis rupestris) and most of the information presented here refers to this rootstock. We also compare the variants of GLRaV-3 found in both Richter 99 rootstocks and scions of selected individual vines and we determine the presence of other leafroll-associated viruses in Richter 99.
FIGURE 6. Vines were selected where the scion (red cultivars) displayed clear symptoms of grapevine leafroll disease, but where vigorous, lignified Richter 99 rootstocks were present.
Cane material was collected separately (Figure 7) from both the scion and rootstock of 69 individual vines grafted on Richter 99 (Vitis berlandieri X Vitis rupestris). These were from two commercial wine estates and two trial sites in Wellington and Stellenbosch during 2014, 2015 and 2016. Specimens were selected based on the occurrence of clear GLD symptoms on the scions. Only vines with sizeable lignified Richter 99 canes were selected for sampling. As rootstock suckers are usually pruned in practice, such plants were relatively rare, but 69 such vines with Richter 99 rootstocks were collected over three seasons. In this regard, we were fortunate to have had access to an abandoned vineyard (Figure 8).
FIGURE 7. Collect lignified cane material separately from the scion and rootstock of leafroll-infected vines, perform RNA extraction on the phloem tissue and test for GLRaV-3 by RT-PCR separately in both components.
FIGURE 8. An abandoned vineyard where the Richter 99 rootstock had not been removed over a number of seasons.
Samples were processed by removing the outer bark and preparing phloem shavings separately of the scion and rootstock material of each vine and then extracting RNA from these. RT-PCR was then performed separately on the scion and rootstock. Clear differences in GLRaV-3 infection status were observed between the Vitis vinifera scion and V. berlandieri X V. rupestris Richter 99 rootstocks of individual leafroll disease-infected vines (Figure 9). The scion material of all 69 vines analysed contained GLRaV-3. GLRaV-3 could not be detected by RT-PCR in 66% of Richter 99 rootstocks from these, despite the fact that the corresponding scions were positive for GLRaV-3, displayed obvious GLD symptoms, and were a constant source of GLRaV-3 inoculum to the rootstock.
FIGURE 9. Results of a RT-PCR reaction for GLRaV-3 on electrophoresis gel. The presence of the white bands above is clear in the scion of any vine, but not in its corresponding Richter 99 rootstock (each red rectangle represents a single vine).
Of the 23 Richter 99 samples that did contain GLRaV-3, only five yielded levels of amplicons in end-point PCR reactions comparable to that of the scions, while the remainder all yielded considerably less, yielding only very faint bands in agarose electrophoresis gels.
Using next generation sequencing, we also only found minor differences in the GLRaV-3 variant composition of Richter 99 and corresponding scions in those instances where GLRaV-3 was found in the rootstock. We also demonstrated that Richter 99 can also be infected with GLRaV-1, GLRaV-2, grapevine virus A (GVA), grapevine virus B (GVB) and grapevine rupestris stem pitting-associated virus (GRSPaV).
It remains unknown whether: 1) GLRaV-3 is generally present in Richter 99 when the scion is infected, but at titers sub-detectable to the detection methods employed, 2) GLRaV-3 has an uneven distribution in Richter 99 resulting in poor detection, 3) if GLRaV-3 has variants or genome components selected for in the Richter 99 rootstock which are less efficiently detected by PCR, or 4) the V. berlandieri X V. rupestris interspecific hybrid produces inhibitors to the PCR reaction. Also, it is not known in instances where GLRaV-3 is at detectable levels, whether it is a) either due to “resistance breaking” variants capable of partially overcoming the Richter 99 defence mechanisms, or b) genetic variation amongst various clones Richter 99, some of which may allow replication of GLRaV-3. Any of these possibilities or combinations of them may account for the differences in GLRaV-3 status observed amongst scion and Richter 99 rootstocks and further studies to assess these possibilities need to be conducted.
– For more information, contact Gerhard Pietersen at gpietersen@sun.ac.za.
