EVOLUTION OF WILD AND FERAL VINES FROM THE EGA RIVER GALLERY FOREST ( BASQUE COUNTRY AND NAVARRA , SPAIN ) FROM 1995 TO 2015

1 : Estación Enológica de Haro, Servicio de Investigación Vitivinícola, 26200 Haro (Spain) 2 : Instituto de Recursos Naturales y Agrobiología (CSIC), Av Reina Mercedes 10, 41012 Sevilla (Spain) 3 : Escuela Universitaria de Ingeniería, Universidad del País Vasco, Vitoria-Gasteiz (Spain) 4 : Bureau des subsides, Bâtiment Unicentre-Bureau 331, Université de Lausanne, 1015 Lausanne (Switzerland) 5 : Laboratorio de Entomología Aplicada, Facultad de Biología, Universidad de Sevilla (Spain)


INTRODUCTION
The only Eurasian wild grapevine, Vitis vinifera L. subsp.sylvestris (Gmelin) Hegi, is considered an endangered subspecies in Europe (Di Vecchi-Staraz et al., 2009 ;Bodor et al., 2010).This is due to diverse anthropic impacts on natural habitats, like the exploitation of river-banks, public works, the infestation of downy and powdery mildew as well as the introduction of invasive vines, such as American rootstocks and producer hybrids used as a cultural method to prevent phylloxera infestation from the 19th century (Issler, 1938 ;Arnold, 2002 ;Ocete et al., 2007;Iriarte-Chiapusso et al., 2013).
During the COST Action FA-1003 East-West Collaboration for Grapevine Diversity Exploration and Mobilization of Adaptive Traits for Breeding, developed between 2011 and 2014 (http://www.cost.eu/COST_Actions/fa/FA1003), the objectives were to increase the knowledge on wild grapevine populations by studying its ampelography, genetic pool, sanitary status, and enological characteristics.One of the main targets was to find some vines with potential resistance to pest and diseases, mainly to downy and powdery mildew.
However, current viticulture is affected by an alarming genetic erosion process (This et al., 2001;González-Moreno et al., 2004 ;Esquinas-Alcázar, 2005;Gago et al., 2009) in the framework of climatic warming and markets globalization (Mira de Orduña, 2010; Mozell and Thach, 2014).In modern vineyards there are only a small number of clones with sanitary passport of the cultivars allowed within each Guarantee of Origin, so the genetic pool is declining alarmingly.Global warming will result in loss of acidity and color and together with globalization and open borders will increase the risk of introducing new pests and diseases, which might spread easily in vineyards with low genetic diversity.Due to those facts, Eurasian wild grapevine constitutes a useful phytogenetic resource to be used in breeding of cultivars and rootstocks for those new requirements, previously mentioned (Jones et al., 2005 ;Ocete et al., 2007;Popescu et al., 2013).On the other hand, this liana had important uses in the past (Ocete et al., 2011c) and nowadays it constitutes a relic element of river-bank forests, colluvial positions and coastal cliffs (Arnold, 2002).
According to that idea, our attention was focused on Ega River, a tributary of Ebro River, situated in Northern Spain, because it conserves an extensive riparian vegetation with some disseminated population sites of wild grapevine, as it was shown in a preliminary prospection carried out by Ocete and Pérez (1995).The source of this river is located in Álava province (Basque Country) south of the Cantabrian Mountain-Range, its length is 113 km, the average flow is 13.76 m 3 /s, draining an area of 1,497 km 2 (fig.1).Other plant populations within the Ebro River basin have been described previously (Martínez De Toda and Sancha, 1999;Ocete et al., 2003;Ocete et al., 2011b) The aim of the present paper was to compare the number of wild vines registered in 1995 with those sampled in 2014 and to denounce the lack of a specific legislation for the conservation of this endangered taxon in the Iberian Peninsula.Also, to complete the description of the characteristics of their habitats according to associated plant communities, their sanitary status and their oenological profile.Some information on feral vinifera is also included.

Study area
The coordinates of the relic sites with wild vines were marked by GPS in a new prospection on the gallery forests, from the source to its flows into the Ebro River.Those wild grapevines and feral plants, such as rootstocks, producer hybrids (French hybrids) and cultivars were also counted.

Ampelographic study
The ampelographic study on wild grapevine was based on the list of descriptors published by O.I.V. (2009).
Male and female pollen samples were obtained by brushing mature anthers from both kinds of flowers.Pollen grains were introduced in DPX (Fluka) and observed under optical microscope Olympus BX 61 to study the morphological structure of the grains, according to Gallardo et al. (2009).
Main botanical supporting trees and bushes of the vines as well as the associated vegetation were identified using various botanical keys (Rameau et al., 2008) and the study carried out on several botanical communities from Navarra (Peralta et al., 2013).An evaluation of the active limestone was conducted by a Bernard calcimeter following the procedure of Álvarez-Iglesias et al. (2003).
The observation of symptoms caused by pests and diseases was carried out every two weeks from spring to autumn 2014 on the aerial parts of the plants (shoots, leaves and bunches), from the floor up to 4 m of height.On the other hand, roots of each wild vine were unearthed up to 40-50 cm of depth.Samples of rhizosphere of each vine were assessed under a binocular loupe to detect possible damages caused by phylloxera and root rot fungi.The procedures of Fenwick (1943) and Flegg (1967) were followed in the case of nematodes.
Leaves from two vines from each population were used to test the presence of Grapevine Fan Leaf, Grapevine Leafroll and Grapevine Fleck viruses using the Bioreba ELISA test kit (Martelli, 2000).

Microvinification process
Harvest took place on the 29 th of October 2014.On the 31 st , we began vinification with a volume of 6 kg of grape.Following hand destemming, 1000 berries (for a total weight of 314.6 g) were mashed and color parameters were identified on the juice.
The remaining berries (about 3 kg) were crushed using a pestle, inside a plastic bucket.Then microvinification was conducted at a temperature of 18ºC.As cap management 2 to 3 pumping over were applied daily.The fermentation activity started on the 3rd day.On day 6, a density of 1038 mg/l was measured.That day volatile acidity recorded was 0.87 g/l.For that reason, 0.8 g of metabisulphite were added to the ferment to prevent a further increase of the volatile acidity.
On day 7, the fermentation stopped at a density of 1032 mg/l and a volatile acidity of 0.50 g/l.It was necessary to add commercial yeast (F33 Laffort, 0.3 g/l) and an activator (Nutristar, 0.2 g/l) to resume fermentation.
On day 8, the bucket was warmed.The density dropped to 1030 mg/l.On day 9, biggest fermentation activity was noticed and density decreased down to 1015 mg/l.The process stopped on day 10, at a final density of 1011 mg/l, volatile acidity of 0.67 g/l and 36 g/l of reducing sugars.
Different analytical parameters were obtained following the procedures described by O.I.V. (2015): near infrared for the determination of ethanol concentration; automatic potentiometry in the cases of pH and total acidity ; FCSA autoanalyser for volatile acidity ; and the OIV method for the determination of the color intensity.

RESULTS AND DISCUSSION
The ampelographic characters of the wild grapevine are shown in Table 1.These descriptors are similar to those described in other populations from the Ebro basin (Ocete et al., 2011b).The coordinates and number of vines per location, observed in 1995 and 2014 are shown in Table 2.

Pollen
Pollen grains were tricolporated in male and acolporated in female flowers.

Evolution of the number of wild grapevines
As shown in Table 2, over the last twenty years wild plants were reduced between 57 % and 100 % depending on the site.The main causes were deliberate removal of vines, which was reported earlier in other geographical areas (Martínez de Toda, 1991).Natural lianas climb up bushes and trees seeking maximum sunlight exposure (Carter and Teramura, 1988) and are therefore often considered as weeds by forest managers (Lutz, 1943) mainly due to sunlight competition.
Site by site : in Santa Cruz de Campezo, Marañon and Zubielqui the losses were mainly due to cleaning works at the river-bank forest, whereas in Arquijas, the road enlargement and the subsidiary cleaning of the ditches resulted in the almost complete removal of the vines.The situation was even worse in Estella, where the whole population was destroyed by the construction of a new walkway following Saint James' Way between the church of the Holy Sepulcher and the camping site.

Accompanying vegetation
A phytosociological study performed in the area (Arnold, 2002) revealed that wild grapevines are also present at hillsides in a vegetation type called maquis.This type of vegetation is dominated by Quercus ilex L. and Arbutus unedo L. The understory vegetation is often composed of a dense coverage of Ruscus aculeatus L. Along the Ega River, wild grapevines are present in remnants of gallery forests, at their external edges or in the direct vicinity of water.These forests are woods or groves of alders, poplars and willows (Arozena Concepción and Ferreras Chasco, 2007).Rivas Goday (1964)  In these gallery forests, the most frequent lianae are: Bryonia cretica, Clematis vitalba, Hedera helix and Humulus lupulus.
Having a closer look at the different strata of woods and groves of willows and alders, it becomes obvious that the tree strata is formed by pioneer species (Rameau et al., 2008)  We also observed the presence of xerophile or mesoxerophile trees in the bush and herbaceous strata such as A. monspessulanum, A. unedo, Q. ilex, B. sempervirens and F. carica.
The soil is a typical Fluvisol (Arnold, 2002) with a total lime content of 22.9 % due to its development from calcareous rocks from the Mesozoic Era.
Soil environments and accompanying vegetation provided valuable information about the natural habitats of V. vinifera (Morano and Walker, 1995).

Pests
The incidence of symptoms caused by parasitic species on vines is shown in Table 3.  grapevines in both years.The presence of both monophagous mites was detected in every wine grapevine population from the Iberian Peninsula to the South Caucasian region (Ocete et al., 2011a ;Ocete et al., 2012).
Occasionally, the presence of the leafhopper Empoasca vitis Goethe (Homoptera, Cicadellidae) was observed.Thrips angusticeps Uzel (Thripidae, Thripinae) and their bite marks were found at shoot tips and on very young leaves.It is remarkable that this Thrips was not present during the first sampling.Its presence was detected in nearly 50 % of the vines in 2014.This species was also discovered only in a few wild grapevine populations from the nearby Basque Country (Ocete et al., 2008).

Diseases
Symptoms of powdery mildew, Erysiphe necator (Schwein.)Burriel, were observed on leaves, branches, bunches of flowers of both sexes and also on bunches of berries in female individuals.On leaves "oil spots", together with other damages on shoots and bunches were caused by downy mildew, Plasmopara viticola (Berlese and de Toni).There was a minimal difference of infection in both sampling years.Powdery mildew affected nearly 93 % of the exemplars in both 1995 and 2014.At the same time, downy mildew affected about 78 % of wild grapevine lianas.
No virus infection was detected by ELISA tests.
In all cases it is necessary to emphasize that the level of infestation or infection of the different parasitic species varied from one wild grapevine to another within the same population, probably due to the genetic differences caused by the sexual reproduction by seeds.
No symptoms of infestation or infection attributable to Phylloxera (Daktulosphaira vitifoliae Fitch) (Homoptera, Phylloxeridae), root-knot nematodes or mycelium of root rot fungi were detected on roots, as it was reported previously in wild European grapevine populations from Portugal to Hungary (Ocete et al., 2011a) and also in Transcaucasia (Ocete et al., 2012).The absence of both pests might be prevented by edaphic conditions specific of fluvisols, in spite of some Vitis and Muscadinia species showing resistance to nematodes (Walker et al., 1994).In spite of this, some nematode species, which were reported as possible virus vectors (longidorids), were detected in soil of some Austrian populations of wild grapevines along the Danube River.The most common species strictly linked to grapevines was Longidorus elongatus (Regner et al., 2004).
It is necessary to mention that several roots belonging to Populus nigra and Populus tremula, which sometimes constitute the supporter for wild grapevine, often show clear symptoms of Armillaria mellea (Vahl.).Its mycelium can be observed on poplar roots and occasionally mushrooms of this species can be also found in autumn.Wild grapevines do not seem to be affected by this fungal disease.Thus, it would be necessary to carry out additional tests on the possible resistance or tolerance of wild grapevine roots against this important fungal disease.

Microvinification
Analytical data and procedures are compiled in Table 4.
The process did not develop any sanitary problem.
No strange aromas were detected, and the wine obtained was strongly colored and had a nice fruity flavor.It had low concentration of ethanol, about 7 % (v/v) in spite of a potential alcohol level of 12.4 %.The taste of the wine was unbalanced, due to its low pH (values below the normal range between 3.2-3.8),high concentration of tartaric and malic acids and high index of polyphenols in the taste limit.Due to this, yeasts could not completely metabolize all the sugars in the must, resulting in a lower alcoholic wine than could be expected.Both previous characters had a synergistic effect worsening the wine taste.Additionally, an oxidation of the polyphenols led to a final production of acetaldehyde.Data on anthocyanin and absorbance, at the three wavelengths, and the intensity of color and concentration of polyphenols do not keep an adequate equilibrium.It is due to the fact that wild grapevine bears fruit to reproduce itself not to make good wine.Thus, the story of wine is the story of the taming of vine.Man intervened and diverts the vine from its original purpose, he had a fair bit of work on his hands (Clarke, 2002).Both parameters, intensity of color and acidity, are the two most important characteristics of this microvinification, because they highlight the crescent importance of wild grapevine for breeding of red cultivars in areas under a temperate climatology.In these, the majority of the varieties show low values for both parameters.In consequence, the Syrah cultivar, for example, has been widely introduced to produce red wines in vineyards of Andalusia (Spain).
According to the current process of global warming, wild grapevines constitute a remarkable phytogenetic resource in red cultivar breeding, which could increase the concentration of polyphenols and antioxidant compounds in a new generation of future wines.
The number and kind of the different feral grapevines found is shown in Table 5.
These vitaceae appeared from Santa Cruz de Campezo to San Adrián towns.Their presence is higher in the last stretch of the Ega River, between Zubielqui and the Ebro River.
In 2014, the percentage of rootstocks increased up to 34.8 % compared to 1995.Their ampelographic identification indicated that 17 individuals were phenotypically close to Rupestris of Lot (Vitis rupestris) and 9 were close to Vitis riparia.The percentage of producer hybrids (French Hybrids) grew up to 23.9 % in the same period.Cultivars diminished by 23.1 %, probably due to the presence of both kinds of mildews as well as the cleaning of river-bank forest.Among these cultivars, 6 were Garnacha tinta, 1 Garnacha blanca and 3 Calagraño (Jaén blanco).
The presence of feral vines from American origin is common in river-bank forests of other European rivers (Laguna, 2003 ;Arrigo and Arnold, 2007 ;Zecca et al., 2010).Terpó (1962Terpó ( , 1988) ) gave the alarm in the case of the Danube.Within Spain, in the Valencia region, rootstocks are practically the only current vitaceae existing along the majority of gallery forests of rivers and creeks.A similar situation was found in several Spanish reserves of the Biosphera and Natural parks, such as Montseny (Catalonia), Urdaibai (Basque country) and Cazorla, Segura y Las Villas (Andalusia) and natural ecosystems from Portugal (Lara et al., 2012).
Producer hybrids and rootstocks are particularly competitive and present a high risk of invasion.They are issued from interspecific breeding processes including several different North American Vitis species.They are selected to perfectly fit local soil and climatic conditions.They show high tolerances to downy and powdery mildew, and phylloxera (Laguna, 2003).
The presence of Garnacha, as the most frequent escaped Vitis vinifera cultivar found, is probably due to its relatively high level of tolerance to downy mildew.It was thus imported/introduced from the Aragon region to Navarra, La Rioja and Castilian and Leon vineyards in the end of the 19th century (Ocete et al., 2006).
Due to pollen exchanges and seed dispersal by birds the cultivars and rootstocks were able to invade large territories and often hybridize and introgress with native wild grapevines.In the eighties, rootstocks were already suspected to hybridize with wild grapevines (Terpó, 1988).In addition, rootstocks are able to rapidly establish and spread.They are displacing autochthonous wild grapevine specimens from their natural habitats in Europe (Bodor et al., 2010) including in different sites from the Iberian Peninsula (Lara et al., 2013).Due to these facts and to the lack of suitable sexual reproduction conditions linked to floodings, wild populations decreased in size, leading to severe erosion of the gene-stock (Csepregi, 1992).

CONCLUSION
Our results highlight the importance of conserving, both in situ and ex situ, the few remaining exemplars of these relic populations, in particular as these Iberian wild parents have provided the A chlorotype to autochthonous grapevine cultivars (Arroyo-Garcia et al., 2006).
An essential tool for improving the situation would be to develop and enforce a legal protection and its communication to all organizations involved in the conservation of natural ecosystems, as it is already the case in France, Austria, Germany and Hungary.

Table 2 . Comparison of the number of wild grapevines observed in 1995 and 2014 in each location.
-70 -J.Int.Sci.Vigne Vin, 2016, 50, 2, 65-75 ©Vigne et Vin Publications Internationales (Bordeaux, France) Elena MELÉNDEZ et al.(*) "B" and "E" stand for the two extreme points along the course of the river: Beginning and End.
(Rameau et al., 2008), Salix triandra L., and Populus tremula L. These species are able to colonize bare soils and constitute the first step of colonization by vegetation after flooding.The bush and herbaceous strata are reflecting what Roulier (1998) called a floristic shift of the vegetation.The lack of flooding allows post-pioneer(Rameau et al., 2008)species such as Fraxinus angustifolia Vahl.or F. excelsior L. Ulmus minor Mill. to establish and replace the Salix forests.These species are fast growing and light demanding species.