Somatic embryogenesis from seeds in a broad range of Vitis vinifera L. varieties: rescue of true-to-type virus-free plants

Virus infections are commonly found under field conditions in grapevine worldwide and multiple infections are routinely detected. Pathogenic agents, including 65 viruses, five viroids and eight phytoplasmas have been detected in a large number of infected grapevines [1]. Among these viruses, the most important are Grapevine fanleaf virus (GFLV), Arabis mosaic virus (ArMV), Grapevine fleck virus (GFkV), Grapevine leafroll associated virus (GLRaV) species which cause the ‘grapevine leafroll disease’ (GLD), and the rugose wood (RW) complex. Virus infections cause large economic losses in grapevine because of reductions in plant vigour, yield and fruit quality [2]. The long-distance spread of grapevine viruses occurs primarily by the propagation of infected plant material. Therefore, to produce certified material, plants free of the most dangerous viral pathogens are needed. It is essential the selection of virus-free plants or the sanitation of virus-infected plants to produce certified material and conserve germplasm for future necessities. Some viruses are particularly recalcitrant to elimination, so therefore different approaches has been tested for virus sanitation [3]: meristem culture [4, 5]; meristem culture combined with thermotherapy [6, 7], chemotherapy [8, 9], and somatic embryogenesis [3, 10–12].

Somatic embryogenesis consists in the induction of somatic embryos from cells of the explants cultured in vitro and is the preferred method for cell to plant regeneration in Vitis vinifera L. and its intraspecific or interspecific hybrids. It has been used for micropropagation [13, 14], generation of transgenic plants [15] and virus sanitation [3, 12]. Even though many studies have been carried out on embryogenesis in grapevine, the standardization of the conditions for embryogenesis induction and in vitro plant regeneration is still an empirical process because it depends on the genotype, type of explant, composition of the culture medium, physiological status of the donor plant, and culture conditions [16–18]. The most limiting factor in grapevine regeneration through somatic embryogenesis is the low rate of embryo to plant conversion, that is, a high percentage of embryos are not able to develop into normal plants [19–21]. In the work of Peiró et al. [12], additionally to plant sanitation, direct somatic embryogenesis (without callus formation) was achieved with high rates of embryo plant-conversion (starting from cut seeds). The time required to recover plants was also shortened with respect to other protocols.

In the present work, we report the results of two experiments aimed to obtain somatic embryo plants (SE-plants) from a total of 16 grapevine varieties (Airén, Cabernet Franc, Cabernet Sauvignon, Mencía, Merlot, Monastrell, Petit Verdot, Pinot Blanc, Pinot Gris, Pinot Meunier, Pinot Noir, Syrah, Tempranillo, Verdil, Godello, and Valencí Blanc), and rescue virus-free plants from grapevine varieties infected with one, two or three viruses commonly found in grapevine (GFLV, GFkV and GLRaV-3). In the first experiment, the putative influence of thidiazuron (TDZ) and genotype was also studied on both, somatic embryogenesis induction and embryo to plant conversion. Among the SE-virus-free regenerated plants, those regenerated from mother tissue (true-to-type) were selected via SSR genotyping.

The high rates of direct regeneration and plant conversion starting from cut seeds described in our previous work aiming to sanitize the cultivar Valencí Negre [12] led us to evaluate the procedure in other varieties to induce regeneration and estimate the efficiency in virus sanitation. This later implicated the analysis of the regenerated virus-free plants genotypes using SSRs (useful plant were those regenerated from mother tissue).

In the first experiment, seed germination (zygotic) after 1 month of culture occurred from some cut-seeds. This effect that was not observed by Peiró et al. [12] and may be consequence of the germination of undamaged embryos after cutting (germination was not observed in non-cut seeds cultured as controls). Despite germination was not the purpose of this work, this result could be interesting in breeding programs to reduce the intervals in the progenies evaluations as reported in grapevine by Ramming et al. [22].

Embryogenesis induction and direct regeneration as reported in Peiró et al. [12], was observed in all the 16 assayed varieties, with high rates in 12 of them. This kind of regeneration is convenient since it may reduce somaclonal variation [23–25]. The most common growth regulators used to induce embryogenesis in grapevine are dichlorophenoxyacetic acid (2,4-D) combined with 6-benzyladenine (6-BA) [16, 26] which generally produce high callus formation. The effect of the concentration of TDZ (reduced by half in EIM2/2) was also evaluated in this experiment to increase the possibilities of embryo induction and/or plant conversion. The concentration of the growth regulator is a key factor in both processes and TDZ, a powerful growth regulator, may interfere in shoot development in excess [27, 28]. The obtained results showed that a concentration of 0.9 μM of TDZ is adequate for embryogenesis induction and plant development in grapevine. However, similar or higher response was obtained in EIM2 medium with respect to EIM2/2 medium. Therefore, EIM2 medium was used for the second experiment where induction of embryogenesis and embryo to plant conversion was achieved in the grapevine varieties Cabernet Franc, Merlot, Godello, and Valencí Blanc.

The fact that high percentages of embryogenic explants were produced after only 2 months of culture from different parts of cut-seeds in all the varieties indicates that regarding embryogenesis induction this protocol is faster than others, where five [3, 10, 29] or 7 months [18] are required. In addition, the lowest E values obtained in the present work (in Airén or Valencí Blanc; E > 19%) were higher than those reported in other works that used anthers and ovaries as starting explants. For instance, Martinelli et al. [30] reported 2.0% of embryogenic calli in anther and 14.0% in ovary cultures in the Chardonnay cultivar, while E values from 2 to 7% were reported by Oláh et al. [31] - who used anthers of seven interspecific hybrids (V. berlandieri x V. rupestris or V. riparia). The percentages of embryogenesis obtained in the present work were also higher than those reported from stamens and pistils by Prado et al. [18], who achieved an E value of 23.0% in the Mencía cultivar in the best treatment, and Dhekney et al. [17], who reported E values in the range from 0.4 to 35.0% for the best treatments in Cabernet Franc, Cabernet Sauvignon, Merlot, Pinot Gris, Pinot Noir, and Syrah explants. On the other hand, Oláh et al. [32] did not obtained embryogenesis in anthers from the varieties Cabernet Sauvignon, Merlot, Pinot Gris, or Pinot Noir when they used medium containing TDZ (0.22 μM) combined with 2,4-D (4.97 μM) and low efficiencies (E < 5%) were achieved in a medium with 2,4-D (5 μM) and 6-BA (0.44 μM). In the present work, an E value of 27% was achieved in the variety Syrah and E values in a range from 40 to 59% in Cabernet Franc, Cabernet Sauvignon, Mencía, Merlot, Pinot Gris, and Pinot Noir. Besides, this is the first report of successful embryogenesis induction for the varieties Airén, Monastrell, Petit Verdot, Pinot Blanc, Verdil, Godello, and Valencí Blanc. Among the four Pinot varieties, Pinot Meunier had a higher response than Pinot Blanc, Pinot Gris, and Pinot Noir regarding both E and DE. Data from SSR analysis showed higher variability in this cultivar with respect to the other Pinot varieties, which are considered as mutations at the berry colour locus of the variety Pinot Noir [33, 34].

The final step of the embryogenesis procedure is the conversion of embryos into plants. Few reports have focused on this aspect even though a high percentage of grapevine embryos are not able to develop into normal plants [19–21]. In our work, germination of SE occurred directly in the induction medium and embryos were transferred to growth regulator-free medium to follow their development. Noteworthy, several other works reported the requirement of additional labour-intensive steps to go from embryo to plant. For instance, Martinelli et al. [30] first separated clusters of embryos from the embryogenic calli and placed them in a liquid medium; then, the embryos - with the radicals facing downward - were transferred to a medium with 6-BA and IBA or to hormone-free medium for germination. López-Pérez et al. [26] transferred table-grape embryos to a medium with indoleacetic acid, gibberellic acid, and activated charcoal (AC) and then, after germination took place, transferred them to a half-strength MS medium. Comparing to data reported in the literature, the percentages of EC obtained in the present work were high (in average 80.25% after 40 days of culture). EC values around 48%, 55%, and 73% were reported by Dhekney et al. [17] in the varieties Cabernet Franc, Syrah and Merlot, respectively, and an EC of 13.6% was reported for Cabernet Sauvignon by Ben Amar et al. [21]. In varieties different from those used in this work, Goebel-Tourand et al. [19] failed to exceed 20% conversion whereas for López-Pérez et al. [20] conversion ranged from 42.7% to 63.8%. The fact that all the tested varieties in this study produced plants that were well rooted and developed indicates that the growth medium used was good enough for rooting all of them.

Virus sanitation by induction of embryos from stamens or pistils was reported in grapevine to cure plants from GFLV [29], GLRaV [3, 11] or ArMV [10] viruses. In the present work, SE-free plants were obtained from plants infected with at least one of the following viruses: GFkV, GFLV or GLRaV-3. The presence of GFkV was not observed in any of the analyzed SE-plants. This was expected as GFkV was reported as not seed transmitted [35]. However, in our study GFLV and GLRaV-3 were detected in some SE-derived plants by RT-PCR indicating that these viruses were in the seed explants. Whereas GFLV was described to be transmitted by seeds in grapevine [36], to the best of our knowledge this is not reported for GLRaV-3 [35, 37]. Despite this result, the sanitation percentages obtained in this work were high (68–100%) for all tested virus and varieties. This result confirms the usefulness of embryogenesis for virus sanitation and the use of seed as starting explants. Sexual embryos are also common explants for somatic induction in other woody species like oak [38], fraser fir [39] or elm [40]. Since grapevine is highly heterozygous, it is propagated vegetatively to keep the same genetic make-up as the parental material. For this reason, we had to select from among the SE-plants, those regenerated from the coat tissue (mother tissue). In grapevine, egg and central embryo sac cells each fuse with sperm cells giving rise to embryo and endosperm development, respectively, whereas seed coats develop from ovule sporophytic tissue [41]. With this aim, the SSRs developed for identification of grape cultivars [42] and recommended by the OIV were used in the first experiment showing that ca 10% of the SE virus-free plant, for each cultivar, were true-to-type. In the second experiment, we increased the number of SSRs (from six to nine), obtaining similar rates of true-to-type sanitized-plants. The use of SSR is common to identify grapevine varieties as well as determine relationships [43, 44]. In conclusion, around 10% of SE plants were suitable for germplasm storage and/or plant multiplication (sanitized and true-to-type). We cannot compare our efficiency respect to that obtained in works that used ovaries or anthers for grapevine embryogenesis induction and virus sanitation [3, 10, 11, 29, 45] because the true-to-type character of SE-plants was not analysed in any of them. In comparison with other sanitation procedures, like meristem culture, the procedure described in the presented study offers the possibility to collect many starting explants which are easier to manipulate than the meristem. Despite meristem culture is the preferred technology for virus cleansing, embryogenesis may be an alternative to consider. Very small meristems are needed for efficient virus cleansing but sometimes the rescue of virus-free plants is not easy. High amount of meristem from previously in vitro micropropagated plants are also needed. Although good results have been obtained with chemotherapy for some virus and specific varieties, its success depends on the virus, the toxicity produced by anti-viral chemicals, and the variety [8, 46, 47].

The results obtained in two independent experiments showed that the culture of cut-seeds on media containing TDZ is appropriated for embryogenesis induction and plant conversion in a broad sense of grapevine varieties. Cut seeds have the advantage - with respect to other commonly used explants (ovaries and anthers) - that they are easy to collect and use. In comparison with other protocols of grapevine embryogenesis, this procedure reduces the time needed to regenerate plants. In addition, it was confirmed using molecular markers that some embryos were regenerated from mother tissue, which indicates the suitability of this procedure to be applied for different biotechnological purposes such as virus sanitation validated in this work. The high rates of sanitized regenerated plants facilitate the selection of true-to-type ones. In this work, virus-free and true-to-type plants of Tempranillo, Pinot Blanc, Cabernet Franc, Godello, Merlot and Valencí Blanc were obtained from grapevines with GLRaV-3, GFLV and/or GFkV in single or multiple infection. These results open up the possibility of using this technique for the sanitation of grapevine plants infected with other viruses. With respect to virus transmission, GFkV was not found in any of the analyzed plants which was expected because this virus is not transmitted by seed. However, the presence of GLRaV-3 and GFLV in some of the regenerated plants pointed out that both viruses are seed-transmitted.