Keywords

1 Introduction

The olive tree (Olea europaea L.) is a subtropical, evergreen oil-producing tree belonging to the family Oleaceae, the subfamily Oleoideae, the tribe Oleeae, the genus Olea and the subgenera Olea (Heywood 1978). The genus Olea comprises more than 40 species including the cultivated, wild and feral forms under the name O. europaea L. (section Olea). However, O. europaea is the only species producing an edible fruit. Enormous confusion prevails around the taxonomical classification in this family. There is lack of consensus over the nomenclature adopted to distinguish the cultivated forms from the wild forms. At least three different ways of naming the olive tree, cultivated in the Mediterranean region, can be found in the literature (O. europaea subspecies sativa, O. europaea subspecies europaea var. sativa and O. europaea subspecies europaea var. europaea). A similar situation can be found in case of the wild olive trees, popularly known as acebuches (O. europaea subespecies sylvestris, O. europaea subespecies europaea var. sylvestris and O. europaea subespecies europaea var. oleaster).

Despite the huge impact of the olive tree cultivation on the economy and culture of Mediterranean countries, the knowledge about the genetics of this species is very limited. High number of chromosomes (2n  =  46) is an indicator of its polyploid (tetraploid) origin (Taylor 1945; Brousse 1987). Stergiou et al. (2002) speculated on the possible role of tropical and subtropical species, like Olea chrysophylla Lam. and Olea excelsa Ait., in its evolution. The hypothesis that domesticated olive comes from an ancient polyploid is supported by molecular data, as some microsatellite markers have been reported to show multi-locus amplification in modern olive cultivars (Cipriani et al. 2002; Diaz et al. 2006a). An alternative explanation is argued by Minelli et al. (2000), who point out the existence of chromosome fusion and rearrangements in a primitive genome, consisting of 48 chromosomes, as the probable origin of the current chromosome set. The range of possibilities (contradictory, in many cases) about the olive genome structure and origin exemplifies the scarcity of knowledge about the species and the need for undertaking a deeper investigation on its genetic behaviour. Actually, microsatellite patterns and flow cytometry analyses have confirmed the hexaploid and tetraploid nature of O. europaea subspp. maroccana and cerasiformis, respectively (Besnard et al. 2008; Brito et al. 2008), and the triploid status of some individuals belonging to O. europaea subsp. laperrinei (Batt. & Trab.) Ciferri (Besnard and Baali-Cherif 2009).

2 Origin and History

Some authors date the beginning of olive cultivation around the Copper Age, 4000–3000 bc (Loukas and Krimbas 1983). This long lapse of time was more delimited by Zohary and Hopf (1994), suggesting that olive domestication took place between 5,500 and 5,700 years ago. In any case, its origin is very ancient and its cultivation goes back to the Prehistory, even if it has not been possible yet to determine ­certainly the course of its progressive or intermittent propagation through the time (Civantos 1999).

Different hypothesis have been proposed regarding the place in which the olive tree was used as a crop for the first time (Chevalier 1948; Moazzo 1994). One of them postulates that it comes from the coasts of Syria, Lebanon and Israel while another one considers the olive tree native to Asia Minor. The historian and man of letters De Candolle is among those supporting Syria as the centre of origin of O. europaea L. (De Candolle 1884). Some other authors locate the beginnings of the olive cultivation in a region around Palestine, Crete and Egypt. There are still others who maintain that the olive tree originated in Ethiopia. However, the most extended idea among the researchers is that the olive tree originated in Eastern Mediterranean, more precisely, to the north of the Dead Sea (Zohary and Spiegel-Roy 1975; Loukas and Krimbas 1983). The first documentary evidences of the olive tree cultivation go back to 4,000 years before the present era when it appeared in the oriental regions of the current Syria and Iran. More recently (2500 bc), the Minoic small clay boards named “Ebla boards” show the olive trees and bear witness to the use of the olive oil at the court of king Minos.

In the last decade, the use of cytoplasmic DNA markers to elucidate migration routes of some plant species has become very common (Tomaru et al. 1998; Sinclair et al. 1999; Gugerli et al. 2001). The study of the mitochondrial DNA polymorphism using amplified fragments length polymorphism (AFLP) markers has helped to conclude that there are, at least, two possible origins for the Mediterranean olive tree (Besnard and Berville 2000; Besnard et al. 2001a). These are, on one hand, the extensive area known as Near East and, on the other hand, Maghreb. So, two different centres of origin are proposed, located to the east and the west of the Mediterranean Sea. This conclusion is based on the discovery of a unique mitotype (ME1) to the populations of wild olive trees from Near East, whereas other two out of the four possible mitotypes (MOM and MCK) could only be found among the wild forms from the west area.

3 Olive Tree Cultivation Spread

At first, it is assumed that olive tree spread from its centre of origin, on one hand, to Cyprus and Anatolia and, on the other hand, to Crete and Egypt. There are some discrepancies in this theory, since some authors support that the olive tree went from Crete to Egypt, Asia Minor, Palestine and Greece, spreading from there to the whole Mediterranean from the second century bc. Some of the researchers consider Palestine being a primary centre of origin of the olive tree. However, Besnard et al. (2001a), who proposed two clearly differentiated areas for its origin, postulated that dispersion of olive tree took place from the oriental region of the Mediterranean toward the western shore, in clear parallel with the migratory movements of men during colonization of the European continent. This would explain why the mitotypes MOM and MCK are only present in the western area.

During the fifteenth, sixteenth and seventeenth centuries, olive cultivation spread all over the Iberian Peninsula. At the beginning of the nineteenth century it was taken to Australia by Italian emigrants. It is grown in many other countries of Latin America and has been spread even up to California. At present the countries around the Mediterranean Basin represent more than 99% of the total hectares dedicated to cultivating olive trees all around the world, Spain being the world’s largest olive oil-producing country (FAOSTAT 2008). World olive oil production is around 2.9 million tonnes, the Mediterranean countries being the major contributors (International Olive Council 2009, http://www.internationaloliveoil.org/downloads/production1_ang.PDF). Up to date, there are 103 germplasm banks disperse around five continents and 25 countries which aim to preserve these resources (Olea databases 2008, http://www.oleadb.it).

4 Breeding Objectives and Procedures

The olive tree is the sixth most important oil crop in the world and in spite of its great economic importance, the most widely planted cultivars are ancient and come from the empirical selection made by growers throughout the centuries (Rallo et al. 2005). Its long juvenile period delays development of new cultivars through breeding methods involving selection among progenies generated from crossing of parents (Rallo 1995). However, the modern olive oil industry requires new and more competitive cultivars that have high oil content, better oil quality, low alternate bearing, increased productivity, suitability for mechanical harvesting and resistance to pests. In the case of table olives, other features to be considered are the shape and size of fruit; uniformity in ripening and a high pulp:stone ratio (Lavee 1994). Some cultivars, such as ‘Barnea’ (Lavee 1986) in Israel and ‘FS 17’ (Fontanazza et al. 1998) and ‘Briscola’ (Roselli and Donini 1982) in Italy, were obtained as a result of breeding programmes. Nevertheless, they are still not widely grown. A new cultivar has been obtained from the crossbreeding programme started in Cordoba (Spain) in 1990 (Rallo 1995). The traits considered for selection were earliness of bearing, high productivity, oil production efficiency, increased oleic acid content, a certain degree of resistance to olive leaf spot and suitability to mechanical harvesting. For this, ‘Arbequina,’ ‘Frantoio’ and ‘Picual’ were chosen as progenitors. More recently, another cultivar called ‘Chiquitita,’ which combines the outstanding characteristics of both parentals (‘Arbequina’ and ‘Picual’) has been developed (Luis Rallo et al. 2008a; Pilar Rallo et al. 2008b). It shows early bearing, high oil content and yield efficiency on one hand and, at the same time, it has low vigour and a compact architecture.

Breeding programmes undertaken have focused on obtaining new cultivars with a combination of superior characteristics. The traits of interest in breeding and improvement of olive are shown below.

4.1 Agronomic Traits

4.1.1 Productivity

Many factors can affect the productivity of an olive tree, ranging from environmental and cultural management to genetic factors. The alternate bearing conditions greatly affect the productivity. Some cultural practices contribute to diminish this phenomenon; and there is a genetic component also as in some cultivars the alternate bearing effects are less pronounced. Another genetic characteristic (though influenced, to some extent, by the environmental conditions) is the rate of bud ­differentiation to reproductive inflorescences or vegetative shoots. The fruit set is also determined by the self- and cross-compatibility of the cultivars (and the nature of the pollen available), apart from other factors having an impact on the tree reproductive ability, like those causing ovary abortion.

4.1.2 Vigour and Plant Architecture

Dwarf olive trees or shrubs are desirable in order to be cultivated in a hedgerow design that would facilitate the mechanical pruning and harvesting. Several studies have evaluated the suitability of some olive cultivars to the modern high-density orchards (De la Rosa et al. 2007; Leon et al. 2007b). As a result, low-vigour cultivars, like ‘Arbosana’ and ‘Arbequina’ showed the highest productivity and oleic acid content (De la Rosa et al. 2007), the latter rending no differences in fruit oil content and moisture when grown at different densities (Leon et al. 2007b).

4.1.3 Earliness of Flowering and Fructification

The reduction of the juvenile period or the lapse of time in which the plant is not able to produce flowers, is one of the most promising olive breeding objectives. As in many other trees, the juvenile period in olive is very long, which is a handicap for growers and breeders. In an attempt to predict the juvenile/adult shift, Moreno-Alias et al. (2009) recorded a number of leaf parameters, finding an organized layer of subepidermal cells in the abaxial mesophyll exclusive to adult trees. So, the authors proposed to use this feature as a criterion of phase change in the olive tree. From a breeding perspective, there is need of an early and easy to measure trait as an indicator of the juvenile period length. In this sense, a correlation between seedling vigour parameters (i.e., height and stem diameter) and the length of the juvenile period has been observed in different studies (De la Rosa et al. 2006; Luis Rallo et al. 2008a; Pilar Rallo et al. 2008b) and is being suggested as precocity selection criterion at early developmental stages. However, Pritsa et al. (2003) observed no such correlation between vigour parameters and earliness of first flowering. Interestingly, in a comparative field assay including 15 selections, a high number of early-bearing genotypes has been reported (Leon et al. 2007a).

4.1.4 Resistance to Pathogens and Pests

Among the most important diseases affecting the olive tree, in terms of economical losses, the olive leaf spot, the Verticilium wilt and the olive knot are the most ­remarkable ones.

Olive leaf spot (also named peacock spot, peacock eye and bird’s eye spot) is caused by the fungus Spilocaea oleaginea and it dramatically decreases the olive productivity. An extensive classification of more than 300 olive cultivars according to their resistance or susceptibility to this disease has been carried out (Trapero and Lopez-Doncel 2005). More precisely, 28 and 20 cultivars were described as resistant and highly resistant, respectively. In Israel, a cultivar resistant to S. oleagina was obtained (Lavee et al. 1999) and a molecular marker linked to the locus conferring the resistance to the disease identified (Mekuria et al. 2002) which could assist in the early selection of resistant individuals in subsequent breeding programmes.

The disease known as Verticilium wilt is caused by the fungal pathogen Verticillium dahliae. An evaluation of 23 olive cultivars carried out by Lopez-Escudero et al. (2004) revealed three cultivars (‘Frantoio,’ ‘Oblonga’ and ‘Empeltre’) being moderately susceptible and resistant to the defoliating (highly virulent) and non-defoliating (mildly virulent) pathotypes, respectively. As expected, ‘Empeltre’ showed a remarkable recovering ability to both fungal isolates, as it is considered a valuable cultivar for inclusion in breeding programmes aimed to develop new genotypes resistant to Verticillium wilt. In this context, new technologies, like the real-time PCR (qPCR), have been optimized to be used as a way to identify the resistant or susceptible nature of individuals in early developmental stages (Mercado-Blanco et al. 2003) and this will undoubtedly facilitate the selection in breeding programmes.

The bacterium Pseudomonas savastonoi is the causal agent of olive knot disease, which infects through wounds. In an attempt to classify olive cultivars into resistant or susceptible types to knot disease Peñalver et al. (2006) inoculated 29 cultivars with two pathogen strains at two different doses and found that none of them was immune to the disease. They categorized olive cultivars as high, medium, or low susceptible to knot disease based on their reaction upon inoculation. According to these authors, some cultivars, widely included in breeding programmes for having outstanding characteristics, like ‘Frantoio’ and ‘Picual’, were among those showing low susceptibility. Nonetheless, these results should be taken cautiously, as ‘Frantoio’ has been reported to be the most susceptible cultivar among those tested by Hassani et al. (2003). The use of potent diagnostic tools, as the qPCR, already employed to detect P. savastonoi in olive samples (Bertolini et al. 2003a, b), will hopefully be useful in such studies.

4.2 Oil Traits

4.2.1 Oil Content

In all selection schemes, oil quantity and quality are the criteria invariably taken into consideration. In this context, the work carried out in the Olive Germplasm Banks of Cordoba and Catalonia (both in Spain), to measure several olive fruit characteristics is very valuable (Del Rio et al. 2005; Tous et al. 2005). The oil content (among other fruit characteristics) was measured in 112 (Del Rio et al. 2005) and 30 (Tous and Romero 2005) olive cultivars. In both cases, the cultivars were classified into five different categories according to the oil content, weight and pulp:stone ratio of their fruits, ranging from “very high” to “very low.”

4.2.2 Oil Quality and Composition

Similarly to the work carried out in the Olive Germplasm Banks of Cordoba and Catalonia (both in Spain) on oil content, the composition of virgin oils coming from 74 (Uceda et al. 2005) and 28 (Tous et al. 2005) olive cultivars has been analysed. Though there were differences among the cultivars with respect to oil quality and composition obtained in both locations (as expected, due to the influence of environment and cultural practices), some important cultivars, like ‘Picual’, show excellent values (i.e., monounsaturated:polyunsaturated fatty acids ratio) in both cases.

Furthermore, in order to identify the relevant genes regulating the fruit metabolism and phenolic content (quality trait) during ripening, Alagna et al. (2009) ­undertook a transcriptomic study of ‘Coratina’ (cultivar with a very high phenolic content) and ‘Tendellone’ (an oleuropein-lacking cultivar), in both cases using material at the beginning and at the end of fruit development. Some of the genes expressed differentially at both stages coded for enzymes involved in the metabolic pathway for terpenoid biosynthesis.

5 Genetic Engineering and Molecular Biology

5.1 Types of Markers and Their Use in Olive

The genetic variability existing in the cultivated olive is enormous. To date, 2,600 different olive cultivars have been described (Rugini and Lavee 1992) and large numbers of mislabelling, homonyms and synonyms have been reported (Barranco and Rallo 2000). The preservation of this valuable genetic patrimony is of paramount importance to avoid its erosion, which would lead to an irreversible narrowing of the genetic background, as it is occurring in many other crops. During the last few decades, considerable exploration, harvesting, characterization and evaluation works of the most outstanding olive cultivars have been performed. Therefore, easy and accurate cultivar identification is an urgent necessity to manage properly the rich olive biodiversity. All the markers listed below have played an important role in the identification and evaluation of the variability present in the species, as well as in the breeding programmes and incompatibility studies.

Besides, even if the juvenile period in olive tree has been considerably shortened by forced growth techniques (Santos Antunes et al. 1999), it is still too long to make a breeding programme viable. The use of molecular markers can speed up this ­process. They are being used not only in identification and compatibility tasks, but also in the selection of individuals with desirable agronomic characteristics in an early stage (marker assisted selection, MAS). For this, obtaining linkage maps in olive is needed, so that markers linked to the traits of interest can be identified.

5.1.1 Morphological Markers

Before the availability and routine use of molecular markers, a great number of morphological traits were studied in an attempt to identify and characterize the enormous variety of forms existing in most of the species of agricultural importance. In olive, the organs most commonly used for this purpose are leaves, inflorescences, whole fruits and seeds (Barranco and Rallo 1984), though pollen grains are also being used as the exine pattern seems to be a valuable discriminating character (Lanza et al. 1996). Important morphological traits used for different studies include the size, the area, the perimeter, or the longitudinal diameter (in leaves, fruits, seeds and pollen), weight (in fruits and seeds), and other organ-specific measurements, like the number of flowers or their density, in the case of the inflorescences. From these data, new parameters can be developed, like the pulp:seed weight ratio or the weight ratio in the whole fruit (pulp and stone) and the seed. Categories based on qualitative characters can also be established, like the leaf or fruit colour and shape. At present, the International Olive Oil Council (IOOC, Madrid) employs the cultivar classification system described by Barranco and Rallo (1984).

The difficulty in evaluation and the influence of general (cultural management and environmental conditions) and plant-specific (age, phenological state, etc.) factors are among the most important handicaps in the use of phenotypic characters as markers. Massei and Hartley (2000) reported a clear example of these effects. Furthermore, these authors observed that the selection aimed to increase the yield and the growth rate in the species O. europaea has led to a decrease in the defence mechanisms deployed by the plant. Another additional problem, considering the high cost of the orchard maintenance, is that the olive tree shows a prolonged unproductive period and most of the phenotypic characters are evaluated when the plant has reached the adult state.

At first, morphological markers were employed to discriminate olive cultivars (Barranco and Rallo 1984, 1985; Leitão 1988; Cimato et al. 1993; Prevost et al. 1993; Tous and Romero 1993; Cantini et al. 1999; Barranco et al. 2000a). Up to date, they are the only markers accepted by the International Plant Genetic Resources Institute (IPGRI, Rome) and the IOOC, though its usefulness is being constantly strengthened by molecular markers (Claros et al. 2000; Sanz-Cotes et al. 2001; Roselli et al. 2002; Rotondi et al. 2003; Corrado et al. 2009; Rao et al. 2009), as sometimes they reveal themselves insufficient to discriminate among olive forms, cultivated and oleaster trees (Hannachi et al. 2008).

5.1.2 Biochemical Markers: Isozymes

Undoubtedly, the biochemical markers most widely used in plant breeding have been the isozymes. They represent the different biochemical forms of an enzyme, easily distinguishable by electrophoresis and coded by different alleles of the same gene (Soltis and Soltis 1989).

The methodology needed to develop and use these codominant markers in the olive tree is relatively easy, quick and affordable (Trujillo et al. 1995). They have been extensively used in olive cultivar identification due to their high level of polymorphism (Pontikis et al. 1980; Loukas and Krimbas 1983; Trujillo et al. 1990; Ouazzani et al. 1993, 1995, 1996; Hilali and El Antari 1994; Trujillo et al. 1995). However, in recent times, breeders are making use of DNA-based markers to acquire genetic knowledge about the olive tree because isozymes have many disadvantages. They are gene products, so their expression is affected by the environment, sometimes is tissue-specific and can be subjected to selective processes. Besides, less than one percent of the genetic variations involve changes in the electrophoretic mobility of the proteins, the number of isozyme systems is limited and many species are monomorphic for these markers, particularly those with a high degree of endogamy or with a narrow genetic base.

5.1.3 Genetic Markers

In last two decades, plant breeding has experimented major advances in the field of genetics with the development of new molecular markers. Genetic markers analyse directly the genotype (DNA), quite the opposite than the previous ones, which were based on the phenotypes and the gene products. For this reason, they are the most extensively used presently. Among them, PCR-based markers, amenable to automation, seem to be the most suitable ones for breeding purposes.

5.1.3.1 Restriction Fragment Length Polymorphisms

This type of markers detects the polymorphism present in the DNA strand when the target site of a restriction enzyme is altered by a mutation or by changes in a segment of the molecule, like deletions and insertions (Botstein et al. 1980). They are codominant and robust markers but their development is time consuming and expensive, and show little polymorphism in species with a small gene pool.

In the last decade, DNA coming from cytoplasmic organelles has been used to unravel origin of the olive tree and the history of its domestication. Mitochondrial (Besnard and Berville 2000; Khadari et al. 2001b; Besnard et al. 2002a; Bronzini de Caraffa et al. 2002) and chloroplastic (Amane et al. 1999; Lumaret et al. 2000; Besnard and Berville 2002; Besnard et al. 2002a, b) RFLPs have been the markers most widely used for this purpose. The discoveries reported by Amane et al. (1999) are especially interesting as they have found a correlation between a concrete chloroplastic genotype (clorotype V) and the olive male sterility. Similar results were reported by Besnard et al. (2000), who associated the male sterility with the CCK chlorotype and the MCK mitotype. The value of these markers linked to male sterility in olive breeding is enormous as they allow establishing the fertility status in early stages of the trees. There are also works in which the polymorphism revealed by nuclear RFLPs is used to elucidate the moment of the genetic divergence of the different taxons included in the species O. europaea (Besnard et al. 2001a). Finally, they have been employed to construct the first olive linkage map, along with other molecular markers (De la Rosa et al. 2003).

5.1.3.2 Random Amplified Polymorphic DNAs

Random amplified polymorphic DNAs (RAPDs) were the first PCR-based DNA markers developed (Welsh and McClelland 1990; Williams et al. 1990). Arbitrary, short primers (around ten nucleotides) are used to amplify genomic DNA, what renders a band profile considered as a dominant marker.

In olive, a massive use of RAPDs has been made to achieve cultivar identification (Bogani et al. 1994; Vergari et al. 1996; Claros et al. 2000; Bandelj et al. 2001; Belaj et al. 2001; Besnard et al. 2001b; Khadari et al. 2001a, b; Sanz-Cotes et al. 2001; and a lot more besides), some cases of homonym and synonym being solved (Wiesman et al. 1998; Mekuria et al. 1999; Barranco et al. 2000a, b). It has been even possible to detect intra-cultivar polymorphism in some of the most important Portuguese cultivars intended for oil production (Gemas et al. 2000, 2004). RAPDs have been also used to estimate the genetic distance among the wild, feral and cultivated forms and within those groups (Besnard and Berville 2000; Belaj et al. 2001, 2002; Besnard et al. 2001a, b, c; Bronzini de Caraffa et al. 2002; Sesli and Yegenoglu 2009), as well as to study the olive propagation in the Macaronesian region (Hess et al. 2000). In this last study, the authors employed ISSR (inter-simple sequence repeat) markers too. The results obtained by Mekuria et al. (2002), in which a clear segregation of the RAPD band patterns in the progeny from crosses among resistant, semi-resistant and sensitive cultivars to olive leaf spot is observed. Thus, markers linked to these disease-resistance gene(s) have been identified are becoming of great interest to the olive breeding programmes. As mentioned above, the data coming from the use of different sorts of molecular markers, RAPDs among them, have been used to construct the first linkage maps of the O. europaea L. genome (De la Rosa et al. 2003; Wu et al. 2004).

5.1.3.3 Amplified Fragment Length Polymorphisms

The methodology necessary to develop amplified fragment length polymorphism (AFLP) markers (Vos et al. 1995) is more complicated than the one used in the case of the RAPD markers. Firstly, the total genomic DNA of the species must be digested and the resulting fragments ligated to adaptors used as priming sites in the following rounds of PCR.

In the olive tree, AFLP markers have been used to study the intra-cultivar (Belaj et al. 2004) and inter-cultivar variation (Sanz-Cortes et al. 2003) showed by different Spanish varieties and to explore the genetic diversity and relationships of Slovene (Bandelj et al. 2004) and southern Italian cultivars (Rao et al. 2009). Owen et al. (2005) extended the study including cultivars from Eastern, Central and Western Mediterranean Basin and found a grouping fashion that supports the East–West divergence of olive. Interestingly, Grati-Kamoun et al. (2006) reported the clustering of olive cultivars grown in Tunisia but coming from different areas of the Mediterranean Basin according to their fruit size but not to their geographical origin when they studied their genetic relationships using AFLPs. Montemurro et al. (2005) observed cultivars from different Mediterranean countries grouped according to their end-use (oil, table or dual purpose) when they analysed data generated from AFLP and microsatellite makers. The associations based on traits of such economic importance could be very useful when it comes to decide crosses in breeding programmes, as sometimes the genitors are chosen not only because of their outstanding characteristics but also because they are not genetically close, minimizing in this way inter-incompatibility issues. Together with RAPDs, AFLPs have been employed to investigate the relationship between the feral and the cultivated olives (Angiolillo et al. 1999; Baldoni et al. 2000). Gallitelli et al. (2001) carried out the same kind of study about the genetic distance among cultivated varieties, including an evaluation of the usefulness of AFLPs compared to RAPDs. Finally, AFLPs are among the markers used for obtaining the first linkage map in olive (De la Rosa et al. 2003).

5.1.3.4 Sequence Characterized Amplified Regions

Any of the DNA markers mentioned above can be transformed into an easier to use and more specific and robust sort of marker named sequence characterized amplified region (SCAR) (Paran and Michelmore 1993). For this purpose, it is necessary to clone and sequence one of the fragments obtained previously and design specific primers to amplify the region by PCR. In the olive tree, the development of several SCAR markers from RAPD (Hernandez et al. 2001a, b; Bautista et al. 2003), AFLP (Busconi et al. 2006; Pafundo et al. 2007) and selective amplification of microsatellite polymorphic loci (SAMPL, Busconi et al. 2006) bands have been reported. Though these markers are less polymorphic than others due to their dominant nature, they can be successfully used for identification purposes. In this way, the ten markers obtained by Bautista et al. (2003) were sufficient to unambiguously discriminate the 22 geographically related cultivars in study.

5.1.3.5 Microsatellites

Microsatellites or “simple sequence repeats” (SSRs) are short (1–6 bp) tandemly repeated DNA motifs (Hamada et al. 1982). They are multi-allelic, hypervariable, codominant and amenable to automation by PCR markers.

Up to date, 120 microsatellites are available in the olive tree (Rallo et al. 2000; Sefc et al. 2000; Carriero et al. 2002; Cipriani et al. 2002; De la Rosa et al. 2002; Wu et al. 2004; Diaz et al. 2006a; Sabino-Gil et al. 2006). Even if a great effort has been made to develop good, reliable molecular tools such as microsatellite markers that could assist breeding programmes, compared to other fruit trees, like the apple tree (Malus domestica B.) or the peach tree (different species of Prunus), it is inevitable to reach to the conclusion that they are still insufficient. All of them have been used for cultivar identification, detecting even intra-cultivar variation in some cases (Cipriani et al. 2002). Their high discrimination power have made possible to solve many cases of homonyms, synonyms and misnamings (La Mantia et al. 2005; Cantini et al. 2008). Additionally, microsatellites have been widely employed for elucidating genetic relationships among olive cultivars (Carriero et al. 2002; Belaj et al. 2003; Bandelj et al. 2004; Diaz et al. 2006a) and among the first 17 selections of an olive breeding programme (Diaz et al. 2007a) that ultimately has rendered a new variety (Luis Rallo et al. 2008a; Pilar Rallo et al. 2008b). Microsatellites have revealed a certain tendency of the ­cultivars to group together according to their geographical origin and routes of propagation (Rallo et al. 2003; Bandelj et al. 2004; Diaz et al. 2006a; Sarri et al. 2006), though some clustering based on their end-use have been reported too (Montemurro et al. 2005; Rekik et al. 2008). Furthermore, it has been verified that microsatellites can be transferred to related species belonging to the genus Olea (Rallo et al. 2003) or even to other genera in the family Oleaceae (De la Rosa et al. 2002), since microsatellites flanking sequences are highly conserved. And, more interestingly, the microsatellites developed in other species belonging to the same family have been employed, together with other markers, for elaborating a linkage map of the olive genome (De la Rosa et al. 2003). Different types of DNA markers, including microsatellites, have been used to construct a new olive linkage map (Wu et al. 2004).

Microsatellites have revealed themselves to be very useful for checking parentage of olive progenies from controlled crossings (De la Rosa et al. 2004; Diaz et al. 2006b, 2007a, b) since their great polymorphism makes it possible to obtain high parentage exclusion probabilities and, in some cases, to assign the paternity to concrete genotypes. De la Rosa et al. (2004) proved the enormous contamination present among the offspring coming from selfings and out-crosses within an olive breeding programme (64.4% of the seedlings had a different pollen donor from the nominal one) using this methodology. Similarly, Diaz et al. (2006b) found that none of the seeds coming from the self-pollination of ‘Picual’ and ‘Arbequina’ olives were really products of self-fecundations. Interestingly, when the offspring from controlled crosses was analysed, the pollen contamination rate was either almost total or almost null depending on the cultivars chosen as genitors (Diaz et al. 2007b). All this supports the idea of an incompatibility system acting in some olive cultivars. Microsatellites have also been used to assign the paternity to olive seeds coming from free-pollination in Australia (Mookerjee et al. 2005). However, these results themselves are not sufficient to affirm that the cultivars chosen as mother trees are self-incompatible, since their flowers were not subjected to self-pollinations; they only corroborate that foreign pollen competes favourably with its own pollen, as it has been extensively reported in the literature (Fernandez-Escobar and Rallo 1981). The same can be argued for the inter-compatibility relationships established. The knowledge of the cross-compatibility ­relationships in olive (Diaz et al. 2006 c, 2007b) is vital to design effective crosses in breeding programmes and the microsatellites seem to be the suitable tools to verify the paternity of the seedlings. Genotyping the individuals at an early developmental stage means time and effort savings since it makes possible to discard the unwanted ones (i.e. those coming from pollen contamination) before reaching the adult phase, when it is feasible to carry out a morphological characterization.

The wild olive germplasm represents a valuable source of variability with a huge potential in breeding programmes. The transfer of both qualitative and quantitative traits from wild into domesticated forms could become an attractive objective in olive breeding programmes. In this sense, an attempt to elucidate the genetic relationships within and between wild and cultivated olives using microsatellites has been made (Erre et al. 2010). This study shows the wild genotypes clustering together in a different gene pool than the cultivated forms, revealing the exotic germplasm as a source of new variability. Regarding the structure of wild populations from north-western Mediterranean, Belaj et al. (2007) observed high and low levels of diversity within and among populations, respectively, using microsatellite markers. They hypothesized that the hybridization with cultivars and the exchange of cultivated genetic resources among different Mediterranean regions could be behind the limited genetic differentiation among populations and the lack of grouping according to their geographical origin.

Actually, microsatellites have shown a higher level of polymorphism when they were compared to other markers, like AFLPs (Belaj et al. 2003; Bandelj et al. 2004; Montemurro et al. 2005) and RAPDs (Belaj et al. 2003). In this context, enterprises like the molecular database, included in the olea databases (http://www.oleadb.it), where the allelic profile of a wide set of olive cultivars for 12 microsatellite markers can be consulted, facilitate access to information continuously being expanded. At the same time, a standardization of some of the microsatellites available is starting to be carried out (Doveri et al. 2008; Baldoni et al. 2009), with the same cultivars being genotyped with a set of markers in different laboratories. This kind of work is aimed to compare the results obtained using diverse methodologies in different laboratories and to rank the markers according to their usefulness in cultivar discrimination.

5.1.3.6 Single Nucleotide Polymorphisms

In recent years, a new generation of molecular markers has entered into the molecular biology field, particularly in the human diseases diagnostic area. These are termed as single nucleotide polymorphisms (SNPs) and consist of single DNA base differences (single base pair changes or deletions) between homologous genomes in which the minor allele is present in 1% of the cases or more (Cooper et al. 1985). The considerable increase in sequences available in databases has revealed the high frequency of these DNA variations in genomes. This abundance turns SNPs into good genome coverage supplier markers, their frequency in several crop species being an order of magnitude higher than that of microsatellites (Kwok et al. 1996). Other desirable characteristic also present in SNPs is their codominant inheritance.

The first SNP-based markers were recently developed in olive (Reale et al. 2006). They were used for cultivar identification purposes, allowing the authors to verify the authenticity of samples coming from the same cultivar but collected in different geographical locations. They made possible to unambiguously discriminate 77% of the cultivars studied. Additionally, an assumed case of synonym between ‘Ottobratica’ and ‘Mirtolia’ was clarified, as both rendered different ­genotypes for the 11 markers tested (nine SNPs among them). This confirms the usefulness of molecular markers in clearing confusion in olive nomenclature. Methodologies to process a high number of samples with a large number of markers (due to their low polymorphism) are complicated. In this context, Consolandi et al. (2007) have used a microarray-based approach to identify 49 olive cultivars using 17 new SNPs. Similarly, Muleo et al. (2009) successfully used a high-resolution melting (HRM) analysis for identification purposes. Nonetheless, the most important drawback in scoring SNPs is the high cost and the necessity of sophisticated equipments for employing the majority of the methodologies developed. For this, the transformation of SNPs into codominant PCR and gel-based markers, like cleaved amplified polymorphic sequences (CAPS), as proposed by Reale et al. (2006), seems to be a good way of getting an easy to use and low-cost method.

Although the development and use of SNPs in olive are still in an early phase, these are markers with an enormous potential in a broad range of applications such as genetic diversity studies, evolutionary and population genetics, mapping, quantifying linkage disequilibrium and marker-assisted plant breeding. In this sense, SNPs located in coding or functional regions of the genomes are especially useful in MAS, since associations between the markers and particular traits allow a more efficient and cost-effective phenotype selection. In short, olive SNPs can be appropriate for the study of genetic diversity and cultivar identification at first, and in future for studies of associations with economically valuable traits.

5.2 Genetic Transformation

Genetic transformation can significantly contribute to plant breeding by generating additional genetic diversity, which can be subsequently subjected to selection through classical and molecular approaches, but also introducing alleles that encode desirable traits into superior cultivars. Successful genetic transformation has been undertaken in some economically important crop species, such as maize, rice, cotton and soybean. Concisely, two methodologies have been developed to transfer an engineered gene into a plant chromosome, the Agrobacterium-mediated transformation and the microprojectile bombardment. The first strategy employed in the case of the olive tree was the agro-transformation (Rugini 1986). This technique has been used with the aim of reducing the olive tree size and enhancing its rooting ability (Rugini and Fedeli 1990). The authors employed Agrobaterium rhizogenes to transform immature zygotic embryos of ‘Morailo’ cultivar. Though the transgenic nature of the calli selected was confirmed molecularly, the regeneration was a ­limitation. Mencuccini et al. (1999) succeeded in obtaining transgenic calli using adult material (leaf petioles) from ‘Dolce Agogia’ cultivar as starting material. However, they failed to regenerate the whole plant. Regarding the biolistic technique, Lambardi et al. (1999) detected GUS transient expression in somatic embryos derived from ‘Canino’ cultivar. Though encouraging results were obtained with cotyledon explants, nowadays, the protocols for olive genetic transformation through biolistic methods are still under optimization (Perez-Barranco et al. 2009).

What becomes clear from above is that the progress in the genetic transformation methodologies in olive must be accompanied by the design of efficient regeneration protocols, via organogenesis and somatic embryogenesis.

5.3 Real-Time Quantitative PCR and Real-Time Quantitative Reverse-Transcription PCR

It is well known that plant diseases impact negatively on yield and fruit quality. For this reason, to develop efficient methods aimed to monitor the sanitary status of the plants is essential to undertake successful breeding programmes. The development of real-time quantitative PCR (qPCR) and real-time quantitative reverse-transcription PCR (qRT-PCR) has allowed the routine and reliable quantification of PCR products with a great specificity and sensitivity, becoming a valuable diagnosis tool (Schaad and Frederick 2002). The qRT-PCR has become a powerful diagnosis tool as in many cases, the viruses can be latent or the symptoms are cultivar-specific, the visual inspections being unreliable. Faggioli et al. (2002) employed one-step qRT-PCR protocol to correlate the infection of olive trees by the strawberry latent ring spot virus (SLRSV) with leaf symptoms. Furthermore, multiplex qRT-PCR has been optimized in olive, allowing to detect a number of different viruses infecting the tree in a single step (Bertolini et al. 2001; Luigi et al. 2009; Varanda et al. 2010). A variant of this technique, the nested qRT-PCR has been successfully used to detect four RNA viruses and the bacterium Pseudomonas savastanoi simultaneously in more than 240 olive samples belonging to 15 different cultivars (Bertolini et al. 2003a; Bautista et al. 2003). The high sensitivity of qRT-PCR made possible to diagnosis infection even when the amount of the viral RNA was below the ­minimum threshold required by other techniques (Grieco et al. 2002; Alabdullah et al. 2009; Varanda et al. 2010). Interestingly, the quantification of DNA coming from highly virulent (defoliating) and mildly virulent (nondefoliating) Verticillium dahliae by qPCR has allowed the establishment of a correlation between those values and the susceptibility of olive cultivars to Verticillium wilt (Mercado-Blanco et al. 2003). This methodology has rendered satisfactory results when combined with others, such as doubled-stranded RNA (dsRNA) analysis and dot blot hybridization, ­complementing and/or improving the data shed by the latter (Montemurro et al. 2008; Alabdullah et al. 2009).

Due to its sensitivity, among many other applications, qRT-PCR can be used to compare the gene expression of samples subject to different treatments. Benitez et al. (2005), coupling this technique to differential display, identified olive genes involved in signalling, transcriptional control and stress response, whose transcript levels were significantly raised after the infection with the fungus Schinia oleagina. Interestingly, the induction of those genes was higher and earlier in the resistant cultivar ‘Lechin de Sevilla’ compared to the susceptible cultivar ‘Picual’. Additionally, the basal expression of some of those genes was increased in the resistant cultivar compared to the susceptible one even when uninfected, suggesting that a constitutive activation of the response pathways could be under its invulnerability. The authors propose the measure of the basal expression of those genes as a way of inferring the resistance or susceptibility level of a particular cultivar, as they observed a correlation between both parameters when different cultivars were analysed. The potential use of this type of assays as a breeding tool to identify and select resistant individuals is obvious.

6 Conclusions

Genetic engineering techniques cannot be considered as substitutes for classical methods in plant breeding. Quite the opposite, the new advances in Genetics and Molecular Biology should be used in combination with conventional breeding, which will facilitate the work of breeders, as it is actually happening in the case of the olive tree. From MAS to expression studies, without forgetting genetic transformation, groups working in olive have known how to incorporate them to their fields of study to generate basic knowledge and to apply it to the breeding of the species.