In this study, we examined the temporal dynamics of the fungal microbiome in grapevine rootstocks throughout four stages of the propagation process, using a non-destructive method. To date, research on grapevine microbiome has predominantly focused on the scion cultivar, as it is the visible half-part of the vine and produces the fruit. However, more than 80% of the vineyards worldwide are currently grafted onto rootstocks [41], which have a relevant influence on crop yield, grape quality and give protection against pathogens and pests [2].
The fungal microbiome in both rootstocks analysed in this study, 110 R and 41 B, was dominated by the Ascomycota phylum throughout the propagation process. This result is consistent with previous studies aimed to explore the grapevine endophytic fungal communities by HTAS approach [14–18, 42–49].
A total of 376 OTUs were detected in our study, ranging from 154 to 172 OTUs, and from 241 to 250 OTUs in 110 R and 41 B rootstocks, respectively. The high number of OTUs found in this study differs from results obtained in previous grapevine research conducted by culture dependent approaches. In Switzerland, Casieri et al. [50] identified only 66 OTUs occurring in the wood of 1-year-old V. vinifera grafted plants, whereas Hofstetter et al. [51] isolated 85 fungal species from healthy nursery planting material. In France, Bruez et al. [52] identified 48 OTUs from healthy wood tissues of the trunk disease esca leaf-asymptomatic and symptomatic vines. In this regard, Kraus et al. [11] identified 86 OTUs in healthy grapevine branches with different ages (from 2-month to 8-year-old) in Germany, although it should be noted that the OTU accumulation curves produced for each branch age hardly started to saturate, suggesting that the number of OTUs would probably increase with major number of samples. A comparative study that aimed to investigate changes in the potentially active fungal communities of internal grapevine wood after hot-water treatment (HWT) in nursery material, revealed that HTAS-based procedure was superior to traditional isolation in detection and identification of fungal communities [15].
The core microbiome was composed by the genera Cadophora, Cladosporium, Penicillium and Alternaria in both rootstocks, together with Eucasphaeria, Paraphoma and Fusarium in 110 R, and Acremonium and Aureobasidium in 41 B. These results are in line with recent studies focused on exploring the interior of grapevine wood. The ubiquitous, fast-growing fungi Cladosporium, Alternaria and Aureobasidium were previously found dominating the fungal community in the xylem vessels of healthy grapevine branches for all ages in Germany [11], as well as in the grapevine sap of shoots under high Pierce´s disease pressure throughout the growing season in California [42]. In addition, these fungal genera were frequently found colonizing the grapevine wood after pruning in Spain [18]. Moreover, Cladosporium was found as the dominant fungal taxa inhabiting several biocompartments of the grapevine endosphere in California [43]. The fungal genera Acremonium, Fusarium and Penicillium were also found with high prevalence on grapevine nursery plants, even after HWT, in Spain and Czech Republic [15]. Regarding Eucasphaeria, fungal species belonging to this genus were isolated from grapevine nursey plants in Switzerland [51], and interestingly, the specie E. capensis was isolated from a zone of dark discoloration from wood of rootstock in Germany, although its pathogenic ability on grapevine remained unknown [53]. Recently, Eucasphaeria was found colonizing grapevine pruning wounds in Spain [18]. Also, fungi belonging to Paraphoma genus, a Phoma-like fungi [54], were previously found inhabiting both esca-symptomatic and asymptomatic vines, as well as grapevine nursery plants in Switzerland [51]. The role of Aureobasidium, in particular the species A. pullulans, as a potential biocontrol agent of fungal trunk pathogens of grapevine has been demonstrated in vitro [55] and in planta [56]. Species of this genus have been shown to prevail in the core microbiome of grapevine in recent studies [14, 18, 42, 44]. Several studies have already revealed the presence of Cadophora, associated with Petri disease and esca on young and mature grapevines, respectively, inhabiting the interior of the vine wood in nursery stock [15, 23, 39, 50, 51, 57], during the propagation stages in nurseries [23], and in mature vines [11, 14, 18, 20].
The fungal genus Neofusicoccum was found as a persistent taxon in both rootstocks, which confirms this genus as a primary settler of grapevine vascular tissues. This genus belongs to the Botryosphaeriaceae family and is considered one of the most virulent fungal genera associated with the trunk disease Botryosphaeria dieback [58, 59], being Neofusicoccum parvum the most common Neofusicoccum species isolated from grapevine worldwide [60]. The role of rootstock mother vines as a primary source of Neofusicoccum spp. has been well-documented by Aroca et al. [28]. Neofusicoccum spp. are spread by the dispersion of airborne spores that penetrate into mother plants and mature vines through the pruning wounds [20]. Previous research has found Neofusicoccum spp. inhabiting rootstocks from nursery material at different stages of the propagation process [15, 51, 60, 61].
Results obtained in our study showed that taxa richness and fungal diversity in the vascular system of the grapevine 110 R and 41 B rootstocks was, in general, decreasing throughout the propagation process. This result was surprising since we expected an increase of fungal diversity after the root development stage in the nursery field where multiple interactions between the plant and soil microorganisms can occur. This could be partially attributed to the enhancing microbial interaction with planting material in nursery practices such a hydration and callusing. For instance, hydration stage has been pointed out as a potential source of cross contamination by microorganisms [62].
Fungal functionality analysis showed that the relative abundance of endophytes decreased throughout the propagation process, while the abundance of plant pathogens increased towards the last stage and before selling the plant to the grower. In fact, we identified 8 genera associated with GTDs. Among them, it should be pointed out that the relative abundance of genera associated with black-foot, such as Ilyonectria in 110 R, as well as genera associated with Petri disease, such as Phaeomoniella and Phaeoacremonium in both rootstocks, increased significantly after the root development in the nursery field. The soilborne black-foot pathogens are commonly found in nursery field soils [63], and their capacity to infect grafted grapevine once planted in the field nursery is well-documented [64, 65]. A survey carried out in Spanish nurseries highlighted the relative higher abundance of black-foot fungi after one growing season in field nurseries compared to their abundances in hydration tanks and callusing rooms [65]. Petri disease pathogens can spend part of their life cycle in vineyard soils, which allow them to infect young vines through the roots [66]. Wounds made during the nursery process can also be an important point of entry of GTD fungi, such as Petri disease pathogens [4] or fungi associated with Botryosphaeria [67] and Phomopsis [28] diebacks. This is in line with the results of our study, which resulted in high detection of GTD fungi such as Neofusicoccum spp. (Botryosphaeria dieback) and Diaporthe spp. (Phomopsis dieback) in 110 R, and Neofusicoccum spp. in 41 B, during the early stage of the propagation process. Surprisingly, correlation network analysis resulted in low level of connectivity among GTD fungi in all stages of the propagation process, even though co-infections among these pathogens is common in vascular tissues of young vines [68, 69].
The results obtained from our ddPCR assay indicated that C. luteo-olivacea was present in 97% of the plant material analyzed, with fungal concentrations similar to those found by Maldonado-González et al. [39]. The abundance and DNA quantity of the fungus was higher in sampling moments 2 and 3, after hydration and callusing (following grafting) stages, respectively. Previous research showed that C. luteo-olivacea inoculum is commonly found in hydration tanks and washings from grafting tools in Spanish nurseries [23]. Comparison of both HTAS and ddPCR techniques resulted in positive correlations between sequence reads and DNA quantification in both 110 R and 41 B rootstocks. HTAS is a robust and powerful approach to study microbial populations, but quantitative significance of its results is often debated [70]. In contrast, ddPCR has recently emerged as a powerful, reliable and accurate technique for detection and quantification of microorganisms, and its success for absolute quantification of GTD fungi from environmental samples has been denoted [39, 71, 72]. Recent efforts have been made to validate the results of HTAS data by comparison with relative or absolute quantification approaches in different environments [73, 74]. For instance, a significant positive correlation between sequencing reads and DNA relative abundance of black-foot pathogens was observed in vineyard soil samples in Spain [29].
Given the results of this study and the lack of authorized effective chemicals against GTD fungi in grapevine nurseries, alternative strategies such as biological control with antagonistic microorganisms, HWT, the use of tolerant rootstocks or biofumigation should be implemented to minimize GTD infection throughout the propagation processes. An integrated management strategy is strongly recommended, including HWT at 50 [75] or 53ºC [15, 76] for 30 min, biocontrol with Trichoderma atroviride SC1 in hydration tanks [77], the use of rootstocks with high tolerance to GTD vascular fungi [78], and biofumigation with white mustard in grapevine nursery fields [79].