Introduction

Verticillium diseases are among the most devastating plant diseases affecting numerous species worldwide, ranging from herbaceous annuals to woody perennials (Bhat and Subbarao 1999; Pegg and Brady 2002; Smith et al. 1988; Fradin and Thomma 2006). After a recent revision of the genus Verticillium by Inderbitzin et al. (2011), ten Verticillium species are distinguished, namely V. dahliae, V. albo-atrum, V. alfalfae, V. longisporum, V. nonalfalfae, V. tricorpus, V. zaregamsianum, V. isaacii, V. nubilum and V. klebahnii. Among these species V. dahliae Kleb. (initially isolated from Dahlia and described by Klebahn 1913) has the greatest economic impact and is among the most widespread plant diseases worldwide (EFSA PLH Panel 2014; Klosterman et al. 2009; Pegg and Brady 2002; Smith et al. 1988). Moreover, with a few exceptions, e.g. wilt of Ceanothus interregimus Hook. & Arn. and Acer pensylvanicum L. caused by V. nonalfalfae (Hibben 1959; Harrington and Cobb 1984; Kasson et al. 2014), and wilt of Liriodendron tulipifera L. caused by both V. dahliae and V. albo-atrum (Donohue and Morehart 1978), all reports of Verticillium wilt in trees are related to V. dahliae. Although no exact statistics exist on the number of species that are susceptible to Verticillium dahliae, it is estimated that at least 300 (Berlanger and Powelson 2005) to 400 (Klosterman et al. 2009) plant species, ranging from herbaceous annuals to woody perennials, are affected. All woody hosts that are susceptible to Verticillium wilt belong to the Dicotyledonaceae, whereas monocotyledonous trees and Gymnophytes are not affected (Hiemstra 1998a; Sinclair and Lyon 2005). Olive plantations in the Mediterranean Basin and tree nurseries in more temperate regions are the most important agricultural sectors that involve woody species affected by the disease (Goud et al. 2011; Hiemstra and Harris 1998; Jiménez-Díaz et al. 2012). In this review, we discuss Verticillium wilt of major tree hosts as well as symptomatology, genetic diversity and detection of the pathogen, the Verticillium wilt disease cycle, reactions of infected plants and recovery, and management of Verticillium in the main woody hosts. We will concentrate on Verticillium wilts of tree species with emphasis on olive as the most important fruiting species, and on ash and maple as the most important shade trees affected by Verticillium dahliae.

Major tree hosts and symptoms

Olive (Olea europaea L.), a member of the Oleaceae family, is considered as one of the economically, socially and ecologically most important trees within olive producing countries. It originates from the Persian high plateau and coastal Syria, from where it was spread throughout the Mediterranean Basin, at first by the Greeks and Phoenicians, later by the Carthaginians, Romans, and Arabs. Later olive cultivation expanded to the Americas, South Africa, China, Japan and Australia (Blázquez-Martínez 1996; Civantos 2004). Verticillium wilt of olive was first reported from Italy (Ruggieri 1946), and soon thereafter from various other regions, including California, European and Asian countries as well as Australia (López-Escudero and Mercado-Blanco 2011; Navas-Cortés et al. 2008), and Argentina (Docampo et al. 1981). Initially, Verticillium wilt mostly occurred in olive groves that were established in fields that were previously used for cultivation of crops that are susceptible to V. dahliae, especially cotton, or in groves established next to fields with susceptible crops (Jiménez-Díaz et al. 1998, 2012). Currently, Verticillium wilt is considered as the most important disease that threatens olive production, causing serious concern to growers, nursery companies and the olive-oil industry throughout the world (López-Escudero and Mercado-Blanco 2011; Jiménez-Díaz et al. 2012; Tsror 2011). This is particularly relevant since most olive cultivars are susceptible to V. dahliae (Antoniou et al. 2001, 2008; Cirulli et al. 2008; López-Escudero et al. 2004; López-Escudero and Mercado-Blanco 2011), although a number of relatively resistant genotypes have been identified in artificial inoculation assays (García-Ruiz et al. 2014; López-Escudero et al. 2004; Martos-Moreno et al. 2006) as well as in field experiments (López-Escudero and Mercado-Blanco 2011; Trapero et al. 2013). However, most of the agronomically and economically relevant olive cultivars are susceptible or extremely susceptible to highly virulent strains of V. dahliae (López-Escudero and Mercado-Blanco 2011).

In olive, two pathotypes of Verticillium wilt have been distinguished, namely the defoliating (D) and non-defoliating (ND) pathotypes (Navas-Cortés et al. 2008). The syndrome caused by isolates that belong to the D pathotype is characterized by early drop of asymptomatic, green leaves from individual twigs and branches that eventually give rise to complete defoliation and necrosis and death of the tree. These symptoms can develop from late fall through late spring. Conversely, the syndrome caused by isolates that belong to the ND pathotype comprises two symptom complexes: an acute form (‘apoplexy’) and a chronic form (‘slow decline’) (Blanco-López et al. 1984; Jiménez-Díaz et al. 1998; Tosi and Zazzerini 1998). The ‘apoplexy’ form, which mainly occurs in late winter and early spring, is characterized by rapid outbreaks involving severe wilting of main and secondary branches. Leaves first become chlorotic, and then turn light-brown and roll back towards the abaxial side while remaining attached to the branches. Ultimately, a rapid dieback of twigs, shoots and branches takes place, especially in young plants, which may result in death of the entire tree (Jiménez-Díaz et al. 1998; Jiménez-Díaz et al. 2012; López-Escudero and Blanco-López 2001). The ‘slow decline’ syndrome is characterized by necrosis of inflorescences, chlorosis of leaves and heavy defoliation of green or dull green leaves. On infected plants, flowers mummify and remain attached to the shoots. The bark of affected shoots may become reddish-brown, and the inner vascular tissues show a dark-brown discoloration. These symptoms usually begin in spring and slowly progress to early summer (Jiménez-Díaz et al. 1998; Jiménez-Díaz et al. 2012; López-Escudero and Mercado-Blanco 2011).

Verticillium wilt is also a major problem in shade tree nurseries in more temperate regions, and can occur also in landscape plantings, especially at locations where susceptible field crops were grown previously (Hiemstra and Harris 1998; Riffle and Peterson 1989). Maples (Acer spp.) are popular trees for residential and commercial landscapes, but generally very susceptible to Verticillium wilt (Gleason and Hartman 2001; Harris 1998; Frank et al. 2012). Among the most frequently grown maple species, Norway maple (A. platanoides) is known as a highly susceptible species on which most of the investigations on Verticillium wilt of maple have been conducted (Harris 1998; Townsend et al. 1990). Verticillium dahliae can induce a range of symptoms in maple including leaf yellowing, curling, wilting and necrosis. Leaf scorch can also occur at leaf margins. Leaves on one side of the tree or on just an individual branch may suddenly wilt and die. Leaves are yellowish and smaller than normal. Also a dark olive-green discolouration develops in the sapwood that is more likely to be present in the larger branches than in the smaller twigs, and is more common near the bases of larger, symptomatic branches. Infected shoots may die back leading to death of branches, and possibly of the whole tree (Frank et al. 2012; Harris 1998; Pscheidt and Ocamb 2013a).

Ash (Fraxinus spp.), like olive a member of the Oleaceae family, is another widely cultivated genus with tree species that are well-known for their high quality timber and ornamental value. Several species in this genus, especially common ash (F. excelsior) which is the most widely distributed species in Europe (Fraxigen 2005), may be severely affected by Verticillium wilt (Heffer and Regan 1996; Hiemstra 1998a, 1998b; Worf et al. 1994). Wilting and defoliation are the earliest symptoms of this disease on ash trees. Leaves may turn to a lighter greyish green colour before wilting or falling off. In extreme cases complete necrosis of leaves may occur. Symptoms may affect the entire crown or only part of it. Ash rarely produces the wilting and discoloration of sapwood common to other trees such as maple. However, some affected trees show a discoloration in the cambial zone, the wood or the pith of stems or branches. In summer after the first heat stress of the year upper branches may die back in a random or one-sided distribution on the tree (Hiemstra 1998b; Pscheidt and Ocamb 2013b). In young trees, although death of affected trees may occur, most of the affected trees recover. Older trees, however, show more gradual disease progress and may decline over a period of months or even years whereas others may completely recover from infection (Hiemstra 1995b). This disease may occur in all kinds of plantations: nurseries, roadside, amenity and recreational plantations as well as forest stands of ash (Hiemstra 1995a, 1995b).

Apart from the above-mentioned major tree hosts, V. dahliae can attack fruit tree species including stone fruits, pistachio and cocoa, as well as other shade tree species including well known genera as Catalpa, Tilia, Ulmus and Robinia (Hiemstra 1998a; Sinclair and Lyon 2005). Wilt, leaf curling or dying, abnormal red or yellow colour of entire leaves, leaf scorch, defoliation, dieback and death, and sapwood discolouration are common symptoms in most of these woody hosts infected with Verticillium wilt (Hiemstra and harris 1998; Sinclair and Lyon 2005; Stipes and Hansen 2009). Eventually, particularly infected young trees may die slowly over a period of several years or suddenly within a few weeks (Douglas 2008; Dykstra 1997; Heimann and Worf 1997).

Asymptomatic infections with V. dahliae have been reported to occur in olive and several other host plants (Evans and Gleeson 1973; Mathre 1986; Malcolm et al. 2013; Karajeh and Masoud 2006; Karajeh 2006). Colonization of V. dahliae without inducing any symptoms has also been reported on monocotoledonous plant species, such as barley, oat and wheat (Krikun and Bernier 1987; Mol 1995), and also on numerous weeds, including dicotelydonous species such as common blackberry (Rubus allegheniensis Porter ex L. H. Bailey), nettle (Urtica spp.), Pennsylvania smartweed (Polygonum pennsylvanicum L.), lamb’s quarters (Chenopodium album), common purslane (Portulaca oleraceae), and black nightshade (Solanum nigrum) (Malcolm et al. 2013; Pegg and Brady 2002; Vallad et al. 2005). This may be explained by the fact that V. dahliae can colonize plant species as an endophyte without inducing any visible symptoms of disease (Malcolm et al. 2013; Petrini 1991). Endophytic colonization of V. dahliae in plant hosts implies that asymptomatic plants may serve as a reservoir of inoculum and may potentially initiate epidemics of Verticillium wilt disease.

Genetic diversity and detection of V. dahliae

Little information is available about variation of the virulence among V. dahliae isolates causing wilt in trees. An exception to this is the classification of isolates from olive as defoliating (D) and non-defoliating (ND) isolates (Rodríguez-Jurado et al. 1993). This dichotomy was first described by Schnathorst and Mathre (1966) for Verticillium infections on cotton (Gossypium hirsutum L.). Isolates belonging to the D pathotype are highly virulent and cause complete defoliation of affected plants, whereas isolates belonging to the ND pathotype are generally less aggressive and cause milder wilt symptoms that do not include defoliation (Schnathorst and Mathre 1966). Interestingly, although isolates of both types cause defoliation in olive, isolates that belong to the D pathotype on cotton are also highly virulent on olive, while isolates that belong to the ND pathotype on cotton are also less virulent on olive (Rodríguez-Jurado et al. 1993; Schnathorst and Sibbett 1971). However, despite the high virulence of isolates of the D pathotype on cotton and olive, different levels of virulence have been observed on other hosts. Moreover, on particular plant species D isolates can be highly virulent without inducing defoliation (Jiménez-Díaz et al. 2006; Korolev et al. 2008; Schnathorst and Mathre 1966). So far, presence of the D pathotype has been reported in North and South America, Europe, the Middle East, and Asia (Jiménez-Díaz et al. 2012). However, no information is available about the differential effects of these two types on other woody hosts.

Differentiation of V. dahliae pathotypes infecting cotton and olive from diverse regions has been conducted through the use of molecular markers (Mercado-Blanco et al. 2001, 2003; Pérez-Artés et al. 2000). Pérez-Artés et al. (2000) designed PCR primers specific for D and ND isolates of V. dahliae, based on sequences of pathotype-associated RAPD bands, and tested them on 67 V. dahliae isolates from cotton and olive collected from southern Spain, China, Italy and the USA. Subsequently, nested-PCR primers were designed and optimized for specific detection of D and ND pathotypes in planta and in soil samples (Mercado-Blanco et al. 2002; Pérez-Artés et al. 2005). However, although these primers have worked for several isolates tested worldwide, it was found that they do not produce the desired amplicon on all V. dahliae isolates (Collins et al. 2005), which may be explained by the high genetic variability that exists among V. dahliae isolates (De Jonge et al. 2012, 2013; Faino et al. 2015, 2016). Arguably, genetic differences between V. dahliae isolates are underlying their behaviour as D or as ND, and the most reliable molecular marker would be a marker on the gene(s) that is (are) responsible for the differential disease phenotype. Currently, however, the molecular mechanism that explains the occurrence of defoliation by some isolates and non-defoliation by others remains unknown.

PCR based assays for detection of V. dahliae have been developed by several groups for detection in soil (e.g. Pérez-Artés et al. 2005; DeBode and Van Poucke 2011; Bilodeau et al. 2012) as well as for detection in plant samples (e.g. Schena et al. 2004; Karajeh and Masoud 2006; Gayoso et al. 2007). So far, however, these protocols have not been developed into procedures for routine screening of planting stock or fields to be planted with crops susceptible to Verticillium wilt. If field soils are screened before planting this uHiemstra, 1995a sually is done using laborious and time consuming wet or dry sieving and plating techniques (Hiemstra, 2015a; Termorshuizen 1998).

Another way to characterize genetic diversity in fungi is to classify individual isolates in vegetative compatibility groups. According to the ability of individual fungal strains to undergo hyphal anastomosis and form stable heterokaryons they can be classified into vegetative compatibility groups (VCGs), such that compatible isolates are placed in the same VCG group (Joaquim and Rowe 1990; Katan 2000; Leslie 1993). V. dahliae isolates have been classified into six VCGs (VCG1 through VCG6). VCG1, VCG2 and VCG4 were further divided into subgroups A and B based on the frequency and vigour of complementation (Chandelier et al. 2003; Chen 1994; Dervis et al. 2010; Goud and Termorshuizen 2002; Jiménez-Díaz et al. 2006; Jiménez-Díaz et al. 2011; Jiménez-Díaz et al. 2012; Korolev et al. 2000, 2001, 2008; López-Escudero and Mercado-Blanco 2011). So far, vast numbers of V. dahliae isolates from maple, ash, olive and some other woody hosts in USA and Europe have been analysed through complementation tests and classified into VCG1A, VCG1B, VCG2A, VCG2B and VCG4B (Chandelier et al. 2003; Chen 1994; Hiemstra and Rataj-Guranowska 2000; Jiménez-Díaz et al. 2011; Jiménez-Díaz et al. 2012; Neubauer et al. 2009). Recently, however, Papaioannou and Typas (2015) showed that although VCGs may be helpful in characterising different isolates, they are genetically not completely isolated.

Disease cycle of verticillium wilt of trees

The life cycle of V. dahliae consists of a parasitic part, in which the fungus lives in its host, and a non-parasitic part, in which it is dormant. For tree hosts the disease cycle of V. dahliae has been described in detail by Hiemstra (1998a) (Fig. 1). During the non-parasitic phase in the soil, V. dahliae survives as microsclerotia, either as dispersed propagules or embedded within plant debris, mainly in the upper layer of the soil from where it can easily be spread by wind, rain or irrigation water, human and animal activities, and agricultural tools and machines (Pegg and Brady 2002; Schnathorst 1981; Wilhelm 1950). Microsclerotia are very persistent and enable the pathogen to attack new plantings even after a long period without hosts being present (Wilhelm 1955). The infection process of V. dahliae in woody plants is similar to that in herbaceous plants. Microsclerotia are stimulated to germinate by exudates from nearby growing roots (Schreiber and Green 1963). V. dahliae begins its parasitic phase when hyphae from germinating microsclerotia penetrate roots of a susceptible host (Lockwood 1977; Nelson 1990; Schreiber and Green 1963). Subsequently, hyphae grow inter- and intracellular within the root cortex to reach the xylem vessels and enter these (Prieto et al. 2009). Vallad and Subbarao (2008) reported that V. dahliae germ tubes can form appressoria before penetration and colonization of the cortical tissues. Also inflated structures that are thought to be functionally analogous to appressoria, and that are named “hyphopodia”, were observed at the penetration sites of roots of cotton and olive inoculated with GFP-labelled isolates of V. dahliae (Reusche et al. 2014; Zhao et al. 2016; Keykhasaber 2017). It has recently been proposed that expression of the tetraspanin VdPls1 and VdNoxB, a catalytic subunit of membrane-bound NADPH oxidases for reactive oxygen species (ROS) production, in hyphopodia is essential for redirecting fungal growth toward host cells to penetrate roots and to colonize the host vascular system (Zhao et al. 2016).

Fig. 1
figure 1

Disease cycle of V. dahliae in trees (Drawing by P.J. Kostense; reprinted with permission from Hiemstra and Harris 1998)

Full size image

Once inside the vessels, V. dahliae colonizes susceptible plants through conidia that are carried with the flow of xylem fluid (Keykhasaber 2017) until they are trapped at vessel ends or protruding parts of vessel elements. Here they may germinate and the new hyphae penetrate into adjacent vessel elements (Beckman et al. 1976). During this process of colonization of the xylem of infected plants cycles of fungal proliferation and (partial) fungal elimination (probably driven by plant defence responses) may occur (El-Zik 1985; Fradin and Thomma 2006; Klosterman et al. 2009; Heinz et al. 1998; Mercado-Blanco et al. 2003). Accumulation of V. dahliae hyphae, ultrastructural and chemical alterations resulting from defense reactions, and aggregates resulting from degradation of external material of the xylem vessel walls by fungal enzymatic activity may cause occlusion of V. dahliae-infected xylem vessels (Baídez et al. 2007; Hiemstra and Harris 1998; Pegg and Brady 2002). As a result, the water flow through the xylem is hampered and symptoms of water stress may develop. Wilting, defoliation and early senescence comprising chlorosis, necrosis, and stunting are the main symptoms of Verticillium wilt disease (Fig. 2). Moreover, sparse foliage and branch dieback may also occur (Berlanger and Powelson 2005; Hiemstra 1998a; Riffle and Peterson 1989; Sinclair and Lyon 2005). Plants with acute infections may start with symptoms on individual branches or on one side of the plant. This is often called “flagging”, which can be diagnostic for Verticillium wilt disease. Furthermore, one or several branches may suddenly wilt and die and buds may fail to leaf out in spring (Douglas 2008; Himelick 1968; Piearce and Gibbs 1981).

Fig. 2
figure 2

Wilting and leaf necrosis in maple, ash and olive trees affected by Verticillium dahliae. a Wilting and desiccation of leaves in a young maple tree (photograph: M. Keykhasaber). b Necrosis and wilting of leaves in a young ash tree (photograph: M. Keykhasaber). c Partial dieback of shoots and branches in an olive tree (photograph: J.A. Hiemstra)

Full size image

Finally, at late stages of the disease, resting structures are formed in dying tissues. Melanized microsclerotia are the survival structures of V. dahliae (Klosterman et al. 2009). Light and electron microscopy analysis of morphological events during microsclerotia formation has shown that initially hyphae become swollen, vacuolated, and form numerous septa. Subsequently, clusters of hyphal cells form in these swollen hyphae. Finally, melanin particles are extruded into the interhyphal spaces of the microsclerotium, and the individual cells of mature microsclerotia also possessed a thickened cell wall surrounded by melanin (Klimes et al. 2008; Xiong et al. 2014). This melanin enables the mirosclerotia to resist UV irradiation, temperature extremes, enzymatic lysis, and fungicidal activities (Bell and Wheeler 1986). The genes involved in melanin biosynthesis have been identified and their functions have been characterized (Gao et al. 2010; Tzima et al. 2012; Klimes et al. 2008; Xiong et al. 2014). It was also confirmed that VDH1, encoding a class II hydrophobin, is involved in the formation of microsclerotia (Klimes et al. 2008; Klimes and Dobinson 2006). In addition, many genes have been characterized that are involved in signal transduction pathways and regulation of microsclerotia formation of V. dahliae, such as VMK1, encoding a mitogen-activated protein kinase (Rauyaree et al. 2005), VdGARP1, encoding a glutamic acid-rich protein (Gao et al. 2010), the G protein β subunit (named as VGB) (Tzima et al. 2012), Vta2, encoding a nuclear zinc finger protein, (Tran et al. 2014), the MADS-box transcription factor VdMcm1 (Xiong et al. 2016), the mitogen-activated protein kinase (MAPK) Hog1 (Wang et al. 2016), and the MAPK Msb and Pbs2 (Tian et al. 2014; Tian et al. 2016). Unravelling the functional characterizations of these genes provides insight into the genetic control of microsclerotia formation in V. dahliae.

The presence of V. dahliae in petioles of infected trees in the form of microsclerotia has been demonstrated for several tree species including Acer (Hiemstra 1997), Liriodendron tulipifera (Morehart and Melchior 1982), olive (Prieto et al. 2009), and Fraxinus (Rijkers et al. 1992). Recently, formation of microsclerotia was also found inside peduncles and flowers of infected olive trees (Trapero et al. 2011). After incorporation of infected plant debris in the top soil layer and decomposition by the activity of soil-borne organisms, microsclerotia survive in the soil for prolonged times (years) and become available as inoculum for new infections (Hiemstra 1997; Hiemstra and Harris 1998; Morehart and Melchior 1982; Rijkers et al. 1992; Tjamos and Botseas 1987; Tjamos and Tsougriani 1990; Townsend et al. 1990).

Reactions of infected trees and recovery

In some cases trees infected by Verticillium wilt are able to recover from the disease (Hiemstra 1998a). This phenomenon has been reported in olive as re-growth from existing crowns that suffer from limited dieback, or from the stem base after complete dieback (López-Escudero and Blanco-López 2001, 2005; Levin et al. 2003; Markakis et al. 2009; Mercado-Blanco et al. 2001). Recovery has similarly been reported in Catalpa bignonioides and Sassafras albidum (Kasson et al. 2015) as re-growth from the crown, in Acer platanoides with re-growth from stem base after extensive dieback (Goud et al. 2011), and in Fraxinus excelsior as re-growth without dieback of twigs (Hiemstra 1998b). The inherent structure of the xylem and the ability of trees to produce new layers of xylem has a significant impact on the potential of recovery (Banfield 1968; Emechebe et al. 1974; Sinclair et al. 1981; Tippett and Shigo 1981). In ring-porous tree species (like robinia and ash) most of the water transport is in the most recent growth ring. This implicates that as long as these trees are able to produce new xylem vessels every year, they can substitute their blocked vessels with new ones, which enables complete recovery, often even without dieback of the crown. In tree species with a diffuse-porous structure of the xylem, such as maple, xylem vessels in each growth ring remain functional for several years. Hence loss of a major part of the water transport capacity in infected trees often cannot sufficiently be compensated by the wood in a new growth ring. Such trees therefore probably show much more dieback of the aerial parts and recovery starts by regrowth from healthy parts of the stem base or roots (Hiemstra and Harris 1998; Keykhasaber 2017). Compared to the healthy plants, however, recovered plants have higher probability of becoming diseased again (Goud et al. 2011).

Differences in the severity of symptoms and in the percentage of recovery in tree species may be related to differences in the capacity to compartmentalize infected xylem. Compartmentalization is a boundary-setting process that is activated following fungal vascular invasion and tends to limit the spread of infection and the loss of normal functioning of sapwood (Hiemstra 1998a; Nicole and Gianinazzi-Pearson 1996; Shigo 1984; Tippett and Shigo 1981). Compartmentalization was first proposed as a mechanism against spread of decay in trees by isolating the damaged tissues and replacing it by new functional tissues (Shigo and Marx 1977). Later it was reported that this mechanism that causes changes in anatomy and chemistry of xylem cells also has an important role in protecting trees against colonization by vascular pathogens (Bonsen et al. 1985; Shigo 1984; Tippett and Shigo 1981; Manion 2003; Smith 2006). The principle of the compartmentalization lies in the establishment of four types of “walls”. While wall 1 restricts pathogen movement longitudinally, wall 2 consists of the growth ring boundary and restricts pathogen movement centripetally, and wall 3 limits the tangential movement of pathogen and is associated with ray parenchyma. Wall 4 is the strongest and referred to as the “barrier zone” that is produced by cambial activity after injury or infection of the existing xylem, and separates the tissue present at the time of infection from new, uninfected tissue (Shigo 1984). It has been reported that, shifts in enzymes and growth regulators in the vascular cambium result in the formation of barrier zone (Shigo 1984; Shigo and Dudzik 1985; Smith and Shortle 1990; Tippett and Shigo 1981). Also a waxy layer of suberin may be synthesized at the boundary between healthy and infected sapwood (Pearce 1990). In fact, suberization response is a heritable trait under control of genetic elements that can be considered as part of a breeding program for increased compartmentalization (Biggs et al. 1992). Studies on clones of Populus deltoides Bartr. (eastern cottonwood) and Liquidambar styraciflua L. (sweetgum) have shown that different clones vary in their compartmentalization ability, suggesting that this phenomenon is under genetic control, and making it possible to screen species for genotypes that display superior compartmentalization traits (Garrett et al. 1979; Shain and Miller 1988; Smith 2006).

In vascular diseases of annual plants, it has been observed that infection may induce transdifferentiation of bundle sheath cells to novel, functional xylem vessels, or may increase xylem cells within the vascular bundle as a result of prolonged or renewed activity of the vascular cambium (Reusche et al. 2012). This phenomenon allows novel vegetative growth of affected stems and branches and enhances recovery (Tjamos et al. 1991; López-Escudero and Blanco-López 2001). Seven putative NAC (for NAM, ATAF1/2, and CUC2) transcription factors have been identified in Arabidopsis thaliana, which are involved in transdifferentiation and fall into the VND subfamily (Vascular related NAC Domain) (Demura et al. 2002; Kubo et al. 2005; Yamaguchi et al. 2010). Within this subfamily, VND6 and VND7 seem to have specific roles on Verticillium-triggered transdifferentiation of bundle sheath cells, with VND6 regulating metaxylem (xylem tissue that consists of rigid thick-walled cells and occurs in parts of the plant that have finished growing) formation, and VND7 inducing protoxylem (the first-formed xylem tissue, consisting of extensible thin-walled cells thickened with rings or spirals of lignin) development (Kubo et al. 2005; Reusche et al. 2012). Interestingly, homologs of NAC domain protein genes (PtVNS/PtrWND) have been identified in poplar (Populus trichocarpa) and their role in differentiation of the xylem vessel element has been demonstrated (Hu et al. 2010; Ohtani et al. 2011). This implies that similar mechanisms may occur in tree species resulting in increased numbers of vascular elements being formed after vascular infection. Thus, studying the distribution of these genes or their homologs in other trees, and also their impact on Verticillium-triggered changes in differentiation of cells from the cambium or even within existing tissues, may help to design strategies to stimulate recovery of susceptible trees.

Expansion of V. dahliae in xylem vessels of infected plants triggers defense reactions, including ultrastructural and chemical alterations (Adams and Thomas 1985; Hiemstra and Harris 1998; Pegg and Brady 2002). Host plants may deposit coating materials (such as lipid-rich or fibrillar coatings) onto xylem vessel walls and into pit membranes (Robb et al. 1982; Street et al. 1986), and accumulate gums and form tyloses to prevent pathogen spread (Baídez et al. 2007). Infected plants also secrete phytoalexins, terpenoid and phenolic substances that have antimicrobial activity during pre-vascular and vascular phases of infection (Daayf et al. 1997; Laouane et al. 2011; Mace et al. 1989; Mansfield 2000; El-Zik 1985; Rodríguez-Jurado et al. 1993; Ryan and Robards 1998; Treutter 2006). In olive trees infected with V. dahliae it was observed that the level of phenolic components such as verbascoside, quercetin, luteolin aglycons, rutin, oleuropein, luteolin-7-glucoside, tyrosol, p-coumaric acid and catechin increased in vascular tissues during infection and colonization (Baídez et al. 2007; Markakis et al. 2010). Their antifungal activity against V. dahliae was substantiated by in vitro studies, suggesting they are involved in defense (Baídez et al. 2007). Moreover, chemical composition of wood plays a role in the variation of effectiveness of pathogen compartmentalization between tree genotypes (Rolshausen et al. 2008). Plant phenolics contribute substantially to structural lignin as well as many seasonal pigments and defensive compounds in plants (Cheynier et al. 2013). The tannin group of phenolic compounds is widespread in both constitutive protection and induced compartmentalization process. Tannins in foliage and wood are reactive pro-oxidant compounds (i.e., chemicals that induce oxidative stress, either by generating reactive oxygen species or by inhibiting antioxidant systems) that disrupt microbial metabolism (Antilla et al. 2013). Additionally, it has been determined that trees with small xylem vessels have stronger ability than trees with large xylem vessels to compartmentalize infections (Eckstein et al. 1979).

Management of the disease

Control of Verticillium wilt is very difficult due to the characteristics of the pathogen and the nature of the infection. Especially the long survival time of microsclerotia in soil, the long lifetime of a tree with continuous exposure to inoculum present in the soil, and the absence of methods to cure infected trees are important factors. Consequently, the use of an integrated strategy is the best way to deal with this disease. This includes the employment of resistant cultivars or rootstocks, cultural practices (i.e., avoid intercropping with V. dahliae susceptible crops; minimise cultivation practices that damage the roots; avoid contaminated equipment; and avoid irrigation that may disseminate the pathogen) to avoid spreading of the disease, and measures (i.e., disinfestation of V. dahliae-infested soil with fumigants, soil solarisation) to avoid build-up of soil inoculum and to reduce soil inoculum levels wherever possible (Barranco et al. 2010; Hiemstra 2015b; Jiménez-Díaz et al. 1998; López-Escudero and Mercado-Blanco 2011; Tjamos and Jiménez-Díaz 1998). Green amendments or biological soil infestation would be also promising methods for control of Verticillium wilt in tree nurseries, but these methods are costly and the efficacy relies on the soil type (Blok et al. 2000; Hiemstra et al. 2013). Moreover, accurate quantification of inoculum in soil would provide valuable information for disease prediction, since density of inoculum in soil is correlated with final disease incidence values (Goud et al. 2011; López-Escudero and Blanco-López 2007). Replacement of diseased trees with non-host plants might also be an environmentally friendly management solution to control Verticillium wilt. Studies on replacement of dead or severely diseased olive trees with apple trees revealed that this would be an appropriate approach in an integrated disease management to control Verticillium wilt disease (Karajeh and Owais 2012). Use of biological control agents, including beneficial bacteria is another approach to manage Verticillium wilts (Markakis et al. 2015; Mercado-Blanco et al. 2004; Prieto et al. 2009). However, the use of resistant cultivars and the screening of new planting sites and planting stock for infection by V. dahliae are the most efficient tools for control of Verticillium wilt of trees (Hiemstra 2015b; López-Escudero and Mercado-Blanco 2011; Tjamos and Jiménez-Díaz 1998). Measures to be included in integrated strategies for control of Verticillium wilt in trees are summarized in Table 1. More elaborated decision schemes for control of Verticillium wilt especially in olive are provided by Hiemstra (2015b) and López-Escudero and Mercado-Blanco (2011).

Table 1 Elements in integrated strategies to control Verticillium wilt in trees (compiled from Barranco et al. 2010, Hiemstra 2015b; López-Escudero and Mercado-Blanco 2011)
Full size table

To evaluate the level of resistance of particular plant genotypes, the pathogen inoculation method that is used, the virulence of the isolates, and environmental conditions all influence the response of the genotype (Levin et al. 2003; López-Escudero et al. 2010; Blanco-López et al. 1998; López-Escudero and Mercado-Blanco 2011). In all programs for evaluating host resistance to V. dahliae it is important to include isolates with different levels of virulence, since different V. dahliae isolates may be differentially pathogenic to different host genotypes (Barbara et al. 1998). Root-dipping and stem inoculation are the most common methods to inoculate trees (Rodríguez-Jurado et al. 1993; Resende et al. 1995) (Fig. 3). In the root-dipping method, the bare root system is inoculated with conidiospore suspension, microsclerotia, or a semisolid fluid mass of mycelium (García-Ruiz et al. 2014), while in the stem inoculation method the stem of tree is inoculated by drilling a trunk hole or making a horizontal incision of a few millimetres deep through the bark of the stem into the xylem to inject a conidiospore suspension (Antoniou et al. 2008; Hiemstra 1995b). Both of these methods have characteristics that must be taken into account during the evaluation of disease progression. Stem inoculation results in a rapid development of disease due to injecting a dense conidial pathogen suspension directly into the vascular tissue of the trees to enable fast resistance evaluation (Antoniou et al. 2008). Upon this type of inoculation, the pathogen escapes the resistance mechanisms operating in roots (Chen et al. 2004; Gold and Robb 1995; Heinz et al. 1998), while upon root dipping the first fungal elimination that occurs in the roots may affect the distribution of the pathogen and progression of the disease. The type of inoculum used for root dipping may also impact colonization by the pathogen and disease progression (Markakis et al. 2009). Also, there is a correlation between inoculum density and final disease incidence values (López Escudero and Blanco-López 2007; Bejarano-Alcázar et al. 1995). Moreover, it has been observed that for a given inoculum density, disease incidence varies greatly depending on the crop (Berbegal et al. 2007; Xiao and Subbarao 1998; Grogan et al. 1979; Harris and Yang 1996; Atallah et al. 2011). Consequently, validation and standardization of V. dahliae inoculation methods and inoculum density is needed to provide accurate assessment of wilt resistance in tree hosts.

Fig. 3
figure 3

Inoculation methods in trees. a rood-dipping inoculation. b stem-inoculation of a tree by making a horizontal incision through the bark into the xylem of the stem by using a snap-off cutter, and putting the conidiospore suspension on the knife with a disposable transfer pipette. c stem-inoculation of a tree by drilling a trunk hole through the bark of the stem into the xylem and injecting a conidial suspension

Full size image

Conclusion

Over the last decades, a strong increase of Verticillium Wilt of olive was associated with the establishment of new olive orchards on infested soils, the use of infected plant material, and the spread of highly virulent pathogen isolates. Also tree nurseries are often established on former agricultural land where hosts of V. dahliae may have been grown, and therefore may serve as infection sources. Therefore, improving the resistance of cultivars, as well as developing protocols for fast and reliable detection of V. dahliae in planting stocks and at planting sites are of the highest importance for establishing effective integrated disease management strategies. PCR-based methods for sensitive and accurate detection and discrimination of V. dahliae isolates allow for the rapid and reliable assessment of soil contaminations and plant infections by V. dahliae. Furthermore, revealing the genetics and molecular background of resistance mechanisms, and of the recovery phenomenon, may provide essential information that can be used in breeding programs to increase the natural resistance of trees against Verticillium wilt. Collectively, these two approaches will provide essential tools for integrated disease management strategies to control Verticillium wilt in tree plantations and nurseries.