Abstract

1. Introduction

Italy is the fourth largest table grape producer in the world after China, Turkey, and Chile and ahead of the United States and South Africa, with a production of 1,005,552 million tons in 2021, and the main European reference in this sector (OIV, https://www.oiv.int/what-we-do/country-report?oiv, accessed June 19, 2023). In Italy, table grape production is localized in the Southern Italian regions, especially Apulia and Sicily, which are characterized by a hot and dry summer climate, mild winters with rainfall not exceeding 500–600 mm/year, and winter temperatures rarely below 0°C. Almost 60% of the Italian production is concentrated in Apulia, mainly with the varieties ITALIA, REGINA BIANCA, VITTORIA, MICHELE PALIERI, and RED GLOBE, which carry the protected geographical indication “Uva di Puglia I.G.P.” (UE Reg. n. 680/2012, 2012; 2020, ISTAT). Seedlessness and other traits, such as crispness, berry shape, color, and bunch compactness, have become highly valued by consumers (Somogyi et al., 2020). Since the commercialization of the variety Sultanina under the name Thompson Seedless in the 1970s (Anderson, 2013), many other varieties have been successfully developed and cultivated worldwide, such as SUGRAONE, CRIMSON SEEDLESS, and AUTUMN ROYAL. These varieties all carry steno-spermocarpic seedlessness, which is a highly heritable and dominant trait that can be selected very early using the intragenic microsatellite marker, p3_VvAGL11, mapped in the regulatory region of the VviAGL11 gene (Mejía et al., 2011; Bergamini et al., 2013; Royo et al., 2018; di Rienzo et al., 2020). However, these varieties are sold under Plant Breeders’ Rights protection, which limits their availability and increases costs. They also have adaptation problems in Apulia, which lead to browning of the skin and splitting of the berries and require special agronomic procedures (Otto et al., 2022). Another important quality parameter for table grapes is the color variation of the berry skin, which has become highly diversified through hybridization and human selection. Berry skin color is mainly related to the content and composition of anthocyanins, which is determined by the allelic status of an important locus spanning a 200-kb region on chromosome 2 (Azuma et al., 2008; Fournier-Level et al., 2009; Matus et al., 2009). In the last decade, several studies have shown that the R2R3-MYB gene is the major genetic determinant of anthocyanin content and composition in grape berry skin. However, further studies are needed to understand the complex network acting on this trait to fully understand and apply it in inbreeding programs.

The control of diseases by genetic resistance sources, in particular powdery mildew (PM) and downy mildew (DM), caused by the biotrophic fungi Erysiphe necator (Schwein.) Burrill and Plasmopara viticola (Berk. & M.A. Curtis) Berl. and De Toni, respectively, has become necessary to obtain high-quality grape productions. The former usually attacks in warm and dry seasons, while the latter disease occurs in warmer and more humid climates (Miazzi et al., 1997; Hajjeh et al., 2008; Fröbel and Zyprian, 2019). Both diseases require a massive usage of pesticides, which is associated with high economic and environmental costs, including the emergence of resistance in pathogens (Hajjeh et al., 2005; Miazzi et al., 2008; Pedneault and Provost, 2016). The strict regulation of pesticides (Directive 2009/128/EC; Regulation 2002/473/EC) and the increasing demand for organic products have made sustainability an important issue in grape cultivation. Efforts to develop more sustainable viticulture have multiplied prompting private and public organizations to expand the use of resistant varieties introgressing genes from American non-vinifera species (such as Vitis riparia, Vitis rupestris, Muscadinia rotundifolia, Vitis cinerea, Vitis berlandieri, Vitis lincecumii, Vitis labrusca), Asian species (Vitis piasezkii, Vitis amurensis, Vitis romanetii), and Vitis vinifera Kishmish vatkana. These species have been used in breeding programs for decades (Bellin et al., 2009; Venuti et al., 2013; Feechan et al., 2015; Ibáñez et al., 2015; Ruehl et al., 2015; Merdinoglu et al., 2018; Vezzulli et al., 2018; Bove et al., 2019; Pomarici and Vecchio, 2019; Weinmann et al., 2019; Possamai et al., 2022; Schneider et al., 2019; Karn et al., 2021). After the first poor-quality hybrids, new resistant varieties, such as Regent or Solaris, have met the quality standards of the European market [30–32] (Töpfer et al., 2011, Buonassisi et al., 2017) paving the way for a wider use of genotypes coming from interspecific crosses also in the competitive table grape breeding programs, together with additional traits such as seedlessness. According to the VIVC database (https://www.vivc.de/index.php), 35 loci for downy mildew resistance (Rpv loci) and 13 loci for powdery mildew resistance (Run and Ren loci) have been mapped so far and are available for breeding even if only a few of them have been used so far. Associated markers are available for these loci, enabling marker-assisted selection (MAS) of resistant individuals at early stages of plant development (Eibach et al., 2007; Pathania et al., 2017; Cobb et al., 2019; Vezzulli et al., 2019; Zini et al., 2019), thus, greatly facilitating breeding programs.

This is particularly true for table grapes, where fungicides have to be applied repeatedly since early spring and throughout the growing season (Miazzi and Hajjeh, 2011; Chen et al., 2020). Over the last 20 years, developments in biotechnology and molecular biology have provided breeders with useful tools to speed up programs and make them more efficient. Genotyping and traceability methods based on molecular markers (Di Rienzo et al., 2017; Miazzi et al., 2020; Sanzani et al., 2016), and next-generation sequencing (NGS) technologies have led to more innovative and targeted studies in grape breeding (Butiuc and Coste, 2023; Vervalle et al., 2022). The intensive chemical treatments required to control them imposes expensive costs on growers and the environment, contribute significantly to pollution, and ultimately lead to a decline in the effectiveness of pesticides (Reynolds, 2015; Lykogianni et al., 2021).

This paper presents the results of a five-year breeding program carried out since 2015 by the Apulian winegrowers’ association “Italian Variety Club” (IVC, Bari, Italy) to obtain new grape varieties that combine seedlessness with resistance to PM and DM. The analysis of parental combinations/cultivation conditions for the development of new hybrids and an evaluation of the success of pyramidization will provide useful information to improve the efficiency of breeding seedless resistant grape varieties.

2. Material and methods

2.2. Crosses and seeds recovery

From 2015 to 2019, a total of 113 crosses was made. A single variety was used in combination with several different varieties to compare the crossing performances of the different combinations ( Table 1 ). Most of the cultivated grapes are hermaphroditic, and fertilization occurs mainly by self-pollination (Harst et al., 2009). In programmed crosses, therefore, the emasculation of plants to be used as female parents is necessary. Late-seeded varieties were chosen as female. Pollen was collected from fully flowering vines (at least 30%) at stages BBCH-65 and BBCH-68, while the female parent was used at stage BBCH-60 (first flower hoods detached from the receptaculum) (Lorenz et al., 1995). Pollen was collected avoiding humid, windy, and rainy conditions and stored at 18°C. Before use, pollen germinability was tested on paper in Petri dishes as percentage of granules producing germination tubes; samples with a germination of <50% were discarded.

A total of 1,612 bunches were pollinated. The inflorescences were emasculated with sterile tweezers and pollinated twice with a sterile brush within 48 h. The pollinated inflorescences were immediately packed until complete berry development. After first fruit development (10 days), bags were removed to allow regular growth of the grapes. Depending on the combination, fully ripe grapes with completely woody seeds were harvested from the end of August through September. The seeds were extracted from the ripe berries, washed in 1% hypochlorite solution, rinsed three times in sterile distilled water, and dried on paper at 25°C and low humidity for approximately 4–5 days until they were completely dry. Then they were stored in Petri dishes in the dark, under cool and dry conditions (4°C) for vernalization (Wang et al., 2022) ( Figure 1 ).

Open in a separate window

Figure 1

Flowchart of the breeding process followed to obtain the new varieties: (A) emasculation of the flowers of the female parents; (B) pollination and casing of the bunches with paper bags; (C) collection of the seeds; (D) stratification of the seeds in sand for vernalization; (E) germination of the seeds; (F) transfer and acclimatization of the vines in the greenhouse; (G) selection of the vines in the field.

3. Results

3.2. Seed vernalization, germination, and plant development

All collected seeds were sent for vernalization during which some die-offs occurred. During vernalization, the average loss was 12.3%, but very high losses were observed in cross Victoria × AP36 (74.8), Cardinal × AP28 (62%), and Emperor × AP28 (69.3%) ( Supplementary Table 2 ; Figure 2 ).

Open in a separate window

Figure 2

Percentage of seed loss observed during vernalization and germination for each cross; the percentage of viable seedlings transferred to the greenhouse is shown in green. The squares with a solid line indicate some of the best crosses; the dotted squares indicate some of the worst crosses. Overall, the crosses with cultivars Victoria and Red Globe showed losses >50%, while very low losses were observed in the progenies of cultivars Italia (× AP24, 3.3%; × AP23, 3.5%; × AP30, 3.9%; × AP29, 7.2%; × AP25, 2.4%; × AP4, 3.3%) as well as cultivars ALICANTE, BOLGAR RAZY, CORNIOLA, and GUZUN ( Supplementary Table 2 ; Figure 2 ).

A total of 103,119 seeds were brought into the greenhouse for germination and acclimatization. During these stages, a further average loss of 62% was observed reducing the number of seedlings to 32,638 ( Supplementary Table 2 ). The highest losses were observed in the progeny of varieties, such as CARDINAL, BOLGAR RAZY, MOLDOVA, PIZZUTELLA, and Victoria, as female plants. In contrast, the lowest losses were observed in progenies of ITALIA, SOURI, SACRA ROSSA, RED ITALY, RED GLOBE, DATTERO, MOSCATO D’AMBURGO, among others (see Figure 2 ). Progenies from the crosses BOLGAR RAZY × AP19, GUZUN × ITALIA, IGNOTA IBRIDO F23PB × AP18, CORNIOLA ROSA × AP7, and EMPEROR × AP28 were cleared ( Supplementary Table 2 ; Figure 2 ). An effect of the combination “female × pollen donor” was observed in the percentage of progeny surviving in the different developmental stages. For example, a few progenies from crosses of Italia with AP1, AP7, and AP30 survived, while approximately 70% of the progeny from crosses with AP4, AP24, and AP28 survived with a small loss during the germination stage. Losses in the greenhouse and in the field were negligible, as good crop protection and cultivation practices were applied.

3.5. In-field phenotypic evaluations

The progenies selected just for seedlessness were propagated in 2016 and evaluated in a comparison field made in 2017 using 24 plants for each genotype according to the criteria required for registration (see Material and Methods). Overall, seedlessness was confirmed in 79% of the vines, while 21% of the vines carried herbaceous (C1) or woody seeds (C3). In particular, herbaceous seeds (C1) were observed in the progeny of ITALIA × AP29 (14%), MENNAVACCA × AP25 (9%), PIZZUTELLA × AP23 (10%), and PIZZUTELLA × AP35 (15%), and woody seeds (C3) were found in the progeny of BLACK MAGIC × AP33 (31%), BLACK MAGIC × AP36 (6%), CORNIOLA × AP17 (26%), CORNIOLA × AP25 (52%), ITALIA × AP24 (13%), MENNAVACCA × AP25 (26%), MOSCATO D’ADDA × AP25 (37%), and MOSCATO GIALLO × AP13 (36%) ( Supplementary Table 3 ). As for color, the breeding program produced 55.1% white, 3.6% pink, 24.6% red, and 4.1% black vines ( Supplementary Table 5 ), with recombination occurring in 34 out of 56 crosses ( Figure 3 ), particularly in crosses with BLACK MAGIC, MOSCATO, and RED GLOBE.

Open in a separate window

Figure 3

The percentages of genotypes with different berry color observed in F1 progenies obtained in the crosses.

In terms of flavor, 48.2% of the offspring were neutral, 17.9% were aromatic, while the remaining 33.9% were partly aromatic and partly neutral ( Figure 4 ).

Open in a separate window

Figure 4

Pie chart illustrating the percentage of progenies with berry flavor neutral (blue), aromatic (orange), and mixed (gray).

The seedless progenies of the crosses also selected for DM and PM resistance were scored in the field during the peak disease pressure using a scale from 0 to 9. The results generally confirmed the molecular data for both DM and PM and showed medium-high score on the resistance scale, although some contradictory data were also found ( Supplementary Table 4 ). Further investigation is needed both in molecular screening with more markers and markers for additional loci and in the field for symptoms. Indeed, the absence of R loci stands solely for the loci analyzed and leaves open the possibility that other markers may be useful for detecting additional sources of resistance.

4. Discussion

In Apulia, table grapes are of crucial economic importance, and growers demand good table grape varieties that are adapted to the local climate and combine seedlessness with resistance to destructive diseases such as powdery mildew and downy mildew. To meet these expectations, the Apulian winegrower’s association “Italian Variety Club” launched an extensive crossing program in 2015 to combine seedlessness with durable resistance in new grape varieties. The plan comprised 113 crosses, which included seedless and elite varieties as parents, as well as four varieties with resistance to PM and DM diseases. A total of 121,723 seeds were collected with the crossing program, but vernalization resulted in an average loss of 12.1% of the seeds, which proved to be a very critical phase. Nevertheless, the results were encouraging considering that the germination capacity of grapevine seeds is generally approximately 30%–50% (Gao et al., 2014; Wang et al., 2022). Seed germination is one of the most important determinants of reproductive performances and depends on various factors [8], such as stratification time, amount of moisture, mold development, amount of inhibitory phenolic compounds, seed hardness, and cultivation conditions (Lin et al., 2009; Zhang et al., 2009). A complete understanding of dormancy, germination, and seedling formation in grape seeds is lacking, and more comprehensive studies are needed for perennial species (Leida et al., 2010; Graeber et al., 2012). Indeed, large differences were found between crosses, with the progeny of cv Italia showing only minor losses, while the progeny of varieties, such as VICTORIA and RED GLOBE, reached 74.8%. Heavy losses (up to 59%) were also observed in seedlings in the first stages of growth, especially in the progeny of cv. CARDINAL, BOLGAR REZY, MOLDOVA, and PIZZUTELLA, while the progeny of cv. ITALIA, SOURI, and SACRA ROSSA were hardly affected. Here, too, diseases, unsuitable growing conditions, and transplanting methods can have a major influence on the establishment rate of the seedlings, but the variety also seems to play a decisive role (Butiuc and Coste, 2023). Overall, the Italia variety provided very encouraging results. Italia is a late-maturing table grape bred in 1911 from a cross of BICANE × MUSCAT AMBURG. It is the most important Apulian variety and is known for its large bunches and tasty, crunchy berries, as well as its good transportability and storability (Torres et al., 2017). The good performance observed makes this variety a good candidate for table grape breeding programs.

The early application of MAS led to the identification of 7,696 seedless putative plants (49% of the total progeny) of which 79% were confirmed in the field trials indicating a good efficiency of the marker p3_VvAGL11 and its usefulness in long-term, labor-intensive, and expensive grapevine selection programs (Bergamini et al., 2013). However, the lower predictive power of the marker observed in the CORNIOLA and AP25 crosses suggests that alternative/additional modifications at the associated 196-bp locus may be involved in the determinism of seedlessness. A single nucleotide modification in the VvAGL11 gene has been shown to be responsible for seedlessness in V. vinifera (Royo et al., 2018; Ocarez et al., 2020). However, it is very likely that other recombination events and SNPs in the coding region, with small but stable effects, are involved in the complex genetic architecture of apyrenia leading to the failure of the marker (Ocarez et al., 2020; Li et al., 2015). Further characterization of the seedlessness locus and further genetic analyses will help to clarify the reasons for the failure of the marker’s predictive power in crosses with these varieties.

Regarding the berry colors of the progeny, recombination was observed in most crosses. Such variability is not surprising in table grape, a crop in which crosses between different varieties are widespread resulting in extensive gene recombination. In grapevine, differences in berry color are generally due to somatic mutations associated with the VvMybA gene family, which is highly polymorphic and determines the variation in anthocyanin content in berries (Lijavetzky et al., 2006; Fournier-Level et al., 2009; Carbonell-Bejerano et al., 2017; Ferreira et al., 2020; Röckel et al., 2022). Here, too, it is likely that minor loci also play a role, which would explain the controversial results especially for the varieties BLACK MAGIC, MOSCATO, and RED GLOBE. On the other hand, these varieties seem to be a very good resource for increasing color variability in table grape breeding programs. As far as the berry flavor trait is concerned, half of the crosses between neutral female and muscadine male resulted in 49% of neutral offspring, 18% in aromatic offspring, and 34% in partly aromatic and partly neutral progeny. In grapevine, monoterpenes are the key compounds responsible for the Muscat flavor, and a major QTL was co-localized with the 1-deoxy-D-xylulose 5-phosphate synthase (VvDXS) gene, encoding for the 1-deoxy-D-xylulose 5-phosphate synthase enzyme, which is involved in the plastidial pathway of terpene biosynthesis (Battilana et al., 2009; Emanuelli et al., 2010; Li et al., 2023). Our results are difficult to interpret, and further genetic studies of this trait and its heritability will be carried out in the future.

The crosses carried out to obtain the pyramiding of seedlessness and resistance to DM and PM resulted in 230 putative resistant seedless genotypes, 20 of which were resistant to both diseases. The preliminary field observations yielded results that were generally consistent with molecular selection with offspring showing medium-to-high scores of resistance. However, some inconsistencies were also recorded, which makes further evaluation crucial in the coming years. It will also be necessary to extend the analysis to other resistance alleles involved in PM such as Ren3 and Ren9 (Zini et al., 2019). The pedigree of parental varieties, such as PÖLÖSKEI MUSKOTÁLY, TERÉZ, and PALATINA, created in Hungary in 1957 from back-crosses of Seyve-Villard and other French–American hybrids, lack information on the origin of their resistance and should be integrated. This would help to establish an efficient protocol for the early identification of resistance genes in V. vinifera, and would facilitate the lengthy and costly process of grapevine breeding (Hoffmann et al., 2007; Tóth-Lencsés et al., 2015).

5. Conclusions

In recent years, a more sustainable and environmentally friendly management of table grape production has also become possible through the use of resistant varieties. The use of multiple sources of resistance and MAS facilitates the pyramiding of key resistance genes. The breeding program described here has enabled the selection of 10 new selections ( Supplementary Figure 1 ) that combine seedlessness with phenotypically important commercial traits such as color and aroma. These selections are in the process of patent registration in the National Register of Vine Varieties. Others will be the subject of an evaluation plan in the coming years, which will lead to the selection of varieties that meet the ever-changing requests of producers and consumers, also with regard to the new diseases that continue to appear in the Mediterranean area, such as Xylella fastidiosa fastidiosa, the causal agent of Pierce’s disease, which has recently also appeared in Puglia (https://www.osservatoriofitosanitario.regione.puglia.it/). The availability of new tools, such as genomic selection approaches using genome-wide molecular markers, will certainly be better suited to capture the complex genetic architecture of resistance and other quantitative traits. The implementation of these approaches in grape breeding will provide new opportunities to promote more sustainable and environmentally friendly viticulture.

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

Author contributions

MMM: Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. LP: Data curation, Investigation, Writing – original draft. CP: Data curation, Investigation, Writing – review & editing. VR: Data curation, Writing – review & editing. DB: Investigation, Validation, Writing – review & editing. IM: Visualization, Writing – review & editing. PLN: Validation, Visualization, Writing – review & editing. CM: Funding acquisition, Methodology, Validation, Writing – review & editing, Resources.

Funding Statement

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The program of grapevine breeding was funded by the Italian Variety Club (IVC). This study was carried out within the Agritech National Research Center and received funding from the European Union Next-Generation EU (PIANO NAZIONALE DI RIPRESA E RESILIENZA (PNRR)—MISSIONE 4 COMPONENTE 2, INVESTIMENTO 1.4—D.D. 1032 17/06/2022, CN00000022). This study was supported by the BANDO Joint Project 2018 from the University of Verona.

References