The plant microbiome represents an area of active ongoing research efforts and represents a promising approach to increase crop productivity and optimise agricultural management strategies (Li et al., 2020; Taye et al., 2019). The productivity of agricultural systems is greatly dependent on the functional processes of soil microbial communities, especially within the rhizosphere (Bakker et al., 2012). The quest for higher yields leads to an intensification in crop production, which results in a loss of soil fertility and quality (Gattullo et al., 2020). However, soil quality and soil health are important factors for sustainable farming (Karimi et al., 2020). Sustainability is a frequently used term (Lewandowski et al., 1999), especially nowadays of increasing awareness of ongoing climate change, loss of biodiversity, and environmental pollution. Due to this increased awareness, and also in response to rising restrictions on synthetic pesticide usage, there has been a movement toward the adoption of more sustainable agricultural and viticultural practices (Andreolli et al., 2021; Úrbez-Torres et al., 2020; Zucca et al., 2009).

Sustainability in general, or sustainable agriculture, respectively, is characterised by a systems perspective of natural and human resource management; it encompasses three objectives: environmental health, economic viability, and social and economic equity (Zucca et al., 2009).

There are different approaches to achieve sustainable farming. On the one hand, it is possible to genetically modify plants in such a way that they become resistant to the pathogen. The study of Yamamoto et al. (2000) was the first report of grapevine being genetically modified for resistance against fungal diseases. They used transgenic grapevine plants harbouring a rice chitinase gene, which enhanced resistance to powdery mildew and anthracnose (Yamamoto et al., 2000). As it takes many years to breed disease-resistant grapevines, the identification of resistance genes is a good starting point for improving grapevine cultivars (Yamamoto et al., 2000). However, since genetically modified organisms (GMOs) lack social acceptance, two new research approaches are increasingly coming into focus for viticulture: plant engineering and rhizosphere engineering. Plant engineering looks at the extent to which the plant genotype can be optimised through breeding programs. In this way, the newly used fungus-resistant grape varieties were developed. In rhizosphere engineering, the rhizosphere of the plant is studied and optimised. For instance, specific fertilisers or microbial inoculants can alter the composition of the microbial community in the rhizosphere. With these changes, better nutrient uptake or the induction of plant defence responses are aimed for. Both approaches are commonly used in agriculture (Taye et al., 2019; Mendes et al., 2018; Figueroa-López et al., 2016), but with regard to viticulture, there is little experience. Therefore, this review aims to discuss whether these two options, especially rhizosphere engineering, offer an advantage for sustainable viticulture as well.

Grapevines (Vitis vinifera L.) are one of the economically most important fruit crops worldwide (Álvarez-Pérez et al., 2017; Andreolli et al., 2016) mainly used for wine production (Andreolli et al., 2016; Compant et al., 2013). During the growing season, vineyard soils are subject to a number of management practices intended to support the grapevines growth and health such as tillage, weed management, fertilisation and application of pesticides or fungicides (Cordero et al., 2020; Hendgen et al., 2018). Since grapevines are susceptible to various pathogens (Lakkis et al., 2019; Trotel-Aziz et al., 2008) and biotic and abiotic stresses, it is essential to protect the vines and grapes by applying pesticides and fungicides (Carro-Huerga et al., 2020). Climate change further increases the need for protection (Karimi et al., 2020) due to pathogen pressure and water limitation. In total, around 35 % of pesticides are used in viticulture (Compant et al., 2013), which accounts for only 0.005 % of the world's arable land (Klein Goldewijk et al., 2017; OIV, 2016). However, increased environmental awareness of consumers and producers promotes on the other hand the desire for sustainable viticulture (Giri et al., 2019; Compant et al.,2013) and a drastic reduction of all types of chemical inputs (Salmon et al., 2020). To make viticulture more sustainable, alternative strategies are required to control serious grapevine diseases like Plasmopara viticola and Botrytis cinerea (Lakkis et al., 2019; Yamamoto et al., 2000). Thus, alternative methods for protecting the grapevines such as biocontrol (Compant et al., 2013) or microbe-assisted crop production are inevitable. The use of such methods is particularly promising on the plant rhizosphere (Figure 1).

Soil inhabits multiple biotic and abiotic processes, which are linked to key ecosystem functions like the cycling of carbon and other nutrients (Köberl et al., 2020). In terms of plant nutrition, plant health and crop quality, the most important is the soil that directly surrounds the plant's root space – the rhizosphere (Berg and Smalla, 2009). The rhizosphere differs from bulk soil and is defined as the area of the soil that is physically, chemically and biologically influenced by the living root of the plant (Ryan et al., 2009). It is characterised by high concentrations of plant-derived organic exudates, released root cap mucilage and root border cells (Vieira et al., 2020). Due to these released substances, the rhizosphere contains the highest concentration and diversity of bacteria in the soil (Giri et al., 2019).

In particular, the microbiome of the rhizosphere can positively influence plant development, vitality and growth by providing and acquiring nutrients and has frequently been targeted to identify positive plant–microbial relationships (Taye et al., 2019).

It is known that certain beneficial bacteria influence the production of different growth regulators in plants (Giri et al., 2019). They are therefore called “plant growth-promoting bacteria (PGPB)” (Compant et al., 2010) or “plant growth-promoting rhizobacteria (PGPR)” (Trotel-Aziz et al., 2008). PGPBs can promote plant growth by direct or indirect mechanisms (Compant et al., 2019), which are linked to microbial metabolisms such as nitrogen fixation, phytohormone production, phosphate solubilisation, and pathogen suppression (Giri et al., 2019; Pacifico et al., 2019). In addition, by the stimulation of the host plant, these beneficial organisms also lead to increased enzymatic catalysis, enhanced water and nutrient uptake or defence responses (Pacifico et al., 2019). Some of them can produce e.g., auxin, cytokinin, or gibberellin, which affect plant growth as phytohormones (Compant et al., 2019). In addition, PGPR plays an important role in the industry of sustainable agriculture (Giri et al., 2019) as they can act as efficient biocontrol agents. Bacterial biocontrol agents protect their host plants against pathogens either by direct interaction with the pathogen or by induction of the host's resistance (Trotel-Aziz et al., 2008). This so-called induced resistance (IR) occurs when plants develop their defensive capacity when they are triggered by pathogens (Trotel-Aziz et al., 2008). By contrast, the rhizosphere can also form the path of infection for soil-borne pathogens (Berlanas et al., 2019).

Hence, the microorganisms living in and on the roots can influence plant fitness and provide effective protection against abiotic stress by induction of defence reactions (Li et al., 2020). To provide such soil functions, a diverse and active soil microbiome is required (Di Giacinto et al., 2020).

It has been shown, that the (endophytic) microbiome of grapevines has a major impact on the regional terroir of wines (Pacifico et al., 2019; Zarraonaindia et al., 2015). Furthermore, Bokulich et al. (2014) note that regional factors could influence the "microbial terroir" of grape surfaces within global viticulture. However, over the past ten years, the microbiome of vineyard soils has gained raising attention as well (Novello et al., 2017). Although, much progress has already been made in describing the grapevine microbiome and its effects on growth, yield and product quality, little is known about how different factors such as cultivars influence the grapevine microbiome (Awad et al., 2020). The investigation of the soil microbial community could be a new strategy to develop a biological indicator for vineyard soil quality and health (Burns et al., 2016). The soil microbial community of the rhizosphere can be investigated by cultivation-dependent and cultivation-independent approaches. Silva-Valderrama et al. (2020) isolated 378 different fungi and yeasts from multiple grapevine tissues using a cultivation-dependent approach. Additionally, Compant et al. (2011) cultivated and isolated bacteria and fungi from different grapevine tissues. Especially in the rhizosphere, they examined different species such as Pseudomonas spp. and Bacillus spp. (Compant et al., 2011), which are also known to act as biocontrol agents (Figueroa-López et al., 2016; Trotel-Aziz et al., 2008). To examine the microbiome more closely, molecular biological methods like metagenomics and next-generation sequencing (NGS) have become essential (Berlanas et al., 2019). Such techniques may also help to identify beneficial organisms to develop bio-inoculants that can be used not only to improve plant health but also to eliminate unwanted microbes (Giri et al., 2019). Deyett and Rolshausen (2019) used a culture-independent approach and examined 2875 bacterial amplicon sequencing variants (ASVs) and 2694 fungal ASVs in 68 samples of the xylem sap of grapevine. As the main bacterial phyla, they identified Proteobacteria, Firmicutes, Actinobacteria, and Bacteroidetes (Deyett and Rolshausen, 2019). In a study conducted by Liang et al. (2019) Proteobacteria, Acidobacteria, Actinobacteria, Chloroflexi, and Firmicutes were found in vineyard soil.

Furthermore, in the past few years, molecular markers for several agronomically relevant traits such as berry colour and weight or disease resistance against mildews have been revealed in grapevine (Di Gaspero and Cattonaro, 2010). This could help in the development of a biomarker strategy for manipulating plant microbiome ecosystems (Giri et al., 2019).

Continuing advances in biotechnology and bioinformatics promote researchers to evaluate the microbiome to a greater depth, to include more replications and to account for variables such as genotype, time, and space (Taye et al., 2019). In agricultural systems, the plant genotype is carefully controlled through breeding and cultivar selection (Bakker et al., 2012). However, traditional approaches of plant breeding do not take the plant microbiome into account, although the importance of the rhizosphere microbiome for the plant ecosystem functioning has been widely recognised (Mendes et al., 2018). Knowledge about plant–microbial interactions could be added to plant breeding programs to create and maintain healthy and beneficial microbial communities in the rhizosphere (Figure 1B). Integrating the knowledge on multifunctional interactions between crop plants and microbes in future agricultural systems and plant breeding will eventually lead to sustainable solutions to reduce the threat imposed by soil‐borne pathogens (Wille et al., 2019). By an analysis of the rhizosphere microbiomes of common bean cultivars with different degrees of Fusarium oxysporum (Fox) resistance, Mendes et al. (2018) showed that Fox resistance is based on plant genetic traits. Their data support the hypothesis that breeding for resistance may have unintentionally altered the rhizosphere microbiome composition, altering the frequency of beneficial microorganisms and traits that may contribute to plant growth or assist in protection against the pathogen. Considering that Fox resistance is based on genetic and chemical alterations in the plant, their findings suggest that the observed changes in the rhizosphere microbiome may enforce the first line of defence, limiting pathogen invasion through a higher abundance of specific microbial groups and functions, high microbial diversity, abundance and a more complex network structure (Mendes et al., 2018).

Based on their results, Taye et al. (2019) suggested that different Brassica napus genotypes have an extensive and selective control on associated rhizosphere bacterial genera. Given these controls are genetically based, they may represent potential breeding targets if the associated bacteria show to be positively correlated with yield or other positive traits in subsequent work (Taye et al., 2019).

Traditional plant breeding approaches and advanced plant genome editing-based methods are promising ways to accumulate favourable alleles associated with stress tolerance in a plant genome (Ryan et al., 2009). Given the wide range of genotypes that can be collected and/or generated per a specific plant species, genetic diversity is a potentially important asset in maintaining or increasing plant ecosystem values, e.g., in controlling stability and stress resilience in native and cultivated ecosystems, productivity in cultivated ecosystems, and ecosystem functioning (Ahkami et al., 2017).

It is of considerable interest to characterise the structure and composition of rhizosphere microbial communities as a first step towards its manipulation to improve crop performance (Shi et al., 2019). Farmers influence the environment around the roots of their crops and pasture species every time they irrigate their fields or apply fertilisers (Ryan et al., 2009). Progress toward sustainability is offered through the development of crop varieties that selectively enhance beneficial functions within the soil microbiome (Bakker et al., 2012). Identifying genetically controlled positive plant–microbial interactions by comparing lines within breeding programs and across diversity panels is the first step in enabling plant breeders to develop varieties by selecting for genetic factors controlling beneficial plant–microbial interrelationships (Taye et al., 2019).

Although each plant species is thought to select its specific microorganisms (Berg and Smalla, 2009), the understanding of the impact of the genetic variation between the microorganisms and the host plant is still incomplete, especially in grapevine species (Berlanas et al., 2019). However, Berlanas et al. (2019) determined the grapevine genotype as the most important factor for shaping the microbiome in the grapevine rhizosphere. The investigation of the interactions between the grapevines and their microbiome could therefore provide a new tool to improve plant breeding programs for grapevines, leading to an improvement in grape production and vineyard management practices (Bokulich et al., 2014).

However, as Vitis vinifera grape varieties are usually grafted on Phylloxera-resistant Vitis sp. rootstocks, future research in rootstock selection is inevitable with regard to their microbial community as well (Zarraonaindia et al., 2015). Thus, for example, rootstocks displaying a better performance regarding mycorrhiza development may be better suited for viticulture under water stress conditions (Schreiner, 2003). Nonetheless, the study of rootstock resistance remains difficult, since the plant genomes of rootstock cultivars could differ depending on the rootstock (Stempien et al., 2020).

Besides the increasing intensity of agricultural land use, climate change also creates a need for improving rhizosphere ecosystems. Rhizosphere ecology and ecosystem function will be concerned by global climate change, including rising temperatures and disruptive weather patterns due to increasing levels of atmospheric CO₂ (Ahkami et al., 2017). For instance, this will lead to more abiotic stresses like drought stress or flooding. This notion suggests that selection of both genotypes and species should be considered in ecosystem design and breeding programs (Ahkami et al., 2017). Plant-specific microbiomes play an indisputable role in supporting plant health and adaptation to changing environmental conditions (Marasco et al., 2018). Since viticulture already takes place in diverse environments, a variety of grapevine genotypes exists that can help in breeding to adapt vines to the changing climatic conditions (Brunori et al., 2016).

It is important to note that impacting the rhizosphere via plant engineering can be a very complex process due to degradation or inactivation of the engineered compound in the soil, insufficient rate of exudation to influence the rhizosphere, limited knowledge about root exudates composition, and changes in exudate releasing time and level with plant development and external stimuli (Ryan et al., 2009). Bioengineering of synthetic microbial communities for plant growth promotion, disease resistance, and stress tolerance or regulation presents a unique opportunity (Ahkami et al., 2017). While hundreds of bacterial strains have been identified to have beneficial effects; engineering a sustainable synthetic microbial community represents a significant challenge (Ahkami et al., 2017). Microorganisms are a vital component of the rhizosphere, and the total biomass and composition of rhizosphere microbial populations markedly affect interactions between plants and the soil environment. There is considerable interest in developing methods for encouraging the proliferation of beneficial introduced or indigenous microbial populations that facilitate nutrient uptake (e.g., rhizobia and mycorrhiza), promote plant growth directly, or suppress plant pathogens. Once the complexity of the rhizosphere is unravelled, one can attempt to create conditions most beneficial to plant growth by amending the soil, breeding or engineering better plants, and manipulating plant/microorganism interactions (Ryan et al., 2009).

It has been shown that the grapevine microbiome can be shaped by various factors, like seasonality, plant genotype, age, pedo-climatic features, surrounding wild plants or presence of pathogens (Pacifico et al., 2019). Furthermore, soil organisms can be impacted by management systems like organic, integrated or biodynamic farming (Hendgen et al., 2018). The microbial community of vineyard soil is affected by various factors, like cover crop use, tillage, compost application, and management practices (Burns et al., 2016, Figure 1C). It was shown that the bacterial community had a higher diversity in vineyard soils that were tilled less, biodynamically farmed, and had received compost application (Di Giacinto et al., 2020; Burns et al., 2016).

However, organic farming, in general, is considered to be a more sustainable form of agriculture (Brunori et al., 2016). A study by Gattullo et al. (2020) showed that fescue (Festuca arundinacea Schreb.) cover crop strongly improves the soil microbiological and soil chemical properties of a table grape vineyard. Another study showed that an intercropping with subterranean clover (Trifolium subterraneum L.) improved grapevine plant performance (Contreras et al., 2019).

However, there are more ways to engineer the rhizosphere. It is also possible to applicate specific fertiliser, nutrients and bioinoculants with bacteria and/or fungi (Figure 1A). In a study by Bach et al. (2016), the authors examined three bacteria (Bacillus mycoides B38V, Burkholderia cepacia 89, and Paenibacillus riograndensis SBR5) regarding their competitive characteristics in vitro. They showed that all three bacteria showed good rhizosphere competence through hydrolytic enzyme production or antagonistic activities, so they might be tested as biocontrol agents against different plant diseases in future studies (Bach et al., 2016). Figueroa-López et al. (2016) investigated three Bacillus strains from the rhizosphere of maize that showed antagonistic activity against Fusarium verticillioides. Furthermore, they suggest possible control mechanisms against these fungi by inoculating the Bacillus strains into the rhizosphere (Figueroa-López et al., 2016).

Regarding grapevine, it has been discovered by Yacoub et al. (2016) that young vines can be protected by various inoculants of the oomycete Pythium oligandrum against a pathogen that is involved in the Esca disease complex. P. oligandrum is known as a biocontrol agent (reviewed by Brožová, 2002) and is available in different products for crop plants. It is also known that Azotobacter species including A. vinelandii, A. beijerinckii, A. nigricans, A. salinestri and A. chroococcum are used in various inoculants against plant pathogens and for nitrogen fixation (Giri et al., 2019). Even in viticulture, a study by Trotel-Aziz et al. (2008) could show that some grapevine associated bacteria like Acinetobacter lwoffii, Pseudomonas fluorescens, Pantoea agglomerans and Bacillus subtilis can act as new biocontrol agents against Botrytis cinerea.

In a review by Compant et al. (2013) some beneficial bacteria and their secondary metabolites for controlling grapevine pathogen diseases are summed up. However, they concluded that there were not enough examples of biocontrol agents used for grapevine yet (Compant et al., 2013). Yacoub et al. (2018) inoculated roots of Cabernet Sauvignon with Pythium oligandrum, a biocontrol agent. They concluded by transcriptomic analysis that several transcripts of vine induced defence systems against P. oligandrum. This study shows that inoculation with biocontrol agents could lead to changes in the whole plant root transcriptome due to a complex plant response (Yacoub et al., 2018). In a further study, these authors showed that grapevine root treatment with P. oligandrum reduced wood necrosis (about 60 %) resulting from inoculation with Neofusicoccum parvum (Yacoub et al., 2020). However, it is important to investigate the pathways within the plant to use it beneficially. In another study, Carro-Huerga et al. (2020) showed that inoculation with a Trichoderma strain biocontrol agent can protect the vine from Phaeoacremonium minimum, known as a pioneer fungus involved in Grapevine Trunk Diseases (GTD) like Esca. Trichoderma spp. is known to have direct effects on GTD pathogens as a biocontrol agent, by activation of host defence genes, therefore, many formulations based on Trichoderma spp. strains have been already assessed for protection (Stempien et al., 2020; Yacoub et al., 2020). Stempien et al. (2020) showed the colonisation of Trichoderma spp. to three different tested rootstock cultivars, but activation of host defence was cultivar dependent. Thus, the use of Trichoderma spp. for grapevine root application needs to be studied for more rootstocks under field conditions (Stempien et al., 2020). González-García et al. (2019) showed that Streptomyces sp. strains isolated from grapevine roots can be readily introduced into young grapevine plants by both immersion and injection into the rootstock as potential biocontrol agents to control fungal pathogens causing young grapevine decline (YGD). Additionally, Andreolli et al. (2021) examined inhibitory effects from the rhizospheric and plant growth-promoting strain Pseudomonas protegens MP12 toward different grapevine phytopathogens via soil inoculation when the vines were planted in the nursery. However, further studies are required to evaluate these results under field conditions (Andreolli et al., 2021).

Furthermore, it was already shown by different studies that arbuscular mycorrhizas (AM) fungal inoculation enhanced the growth of grapevine rootstocks (Aguín et al., 2004; Schreiner, 2003). In a study carried out by Velásquez et al. (2020), it was shown that the arbuscular mycorrhizal fungus Funneliformis mosseae IN101 and the plant growth-promoting rhizobacterium Ensifer meliloti TSA41 promoted the growth of vine plants, as both single inoculants and co-inoculants, increasing plant height and total dry weight. However, it would be more effective to produce mycorrhizal plants at the nursery than inoculate them in the field (Aguín et al., 2004), as root tissues already colonised with AM could directly promote vine growth in the vineyard after planting out ( Giri et al., 2019; Aguín et al., 2004). This could be a promising new technology to reduce the incidence of grapevine fungal infections through the root system (Álvarez-Pérez et al., 2017).

However, it is not only important to have knowledge about the interaction between the plant and the beneficial microorganism but also to understand the colonisation and modulation of the resident microbiome (Compant et al., 2019; Compant et al., 2010). Only after successful colonisation of added inoculants, the plant health or growth will be improved (Compant et al., 2010). However, applications of bioinoculants or biocontrol agents in the field often failed. In addition, the success of such organisms depends on the target crop, product availability, costs, application options and environmental conditions (Giri et al., 2019). Therefore it is important to investigate in future studies how the sole usage of bacterial inoculants can improve soil quality and enhance plant health (Giri et al., 2019).

Table

A) Increasing beneficial microorganisms via microbial inoculants. B) Enhancing beneficial microorganisms through plant breeding or plant engineering. C) Stimulating beneficial microorganisms through management practices like cover cropping or tillage. Created with BioRender.com. Adapted from Hohmann et al. (2020).

We thank Forschungsring des Deutschen Weinbaus (FDW) for funding this project.