Xylella fastidiosa
(Pierce's disease of grapevines)

The distribution given is for the grapevine, peach, coffee and citrus pathogens. Apparently some strains of X. fastidiosa sensu lato (oak strains) occur farther north in North America (Hartman et al., 1996; Goodwin and Zhang, 1997).

Grapevines with symptoms of Pierce's disease in Kosovo, former Yugoslavia, were confirmed as harbouring grape-virulent stains of X. fastidiosa (Berisha et al., 1996). These were tolerant grape varieties capable of harbouring populations of X. fastidiosa with mild symptoms. This record is considered dubious and to date there are no reports of natural spread of Pierce's disease to other vines.

An isolated record from almonds in India (Jindal and Sharma, 1987) is based on a test described in the 1970s; it requires confirmation by more modern methods. In Taiwan pear leaf scorch is caused by a strain of X. fastidiosa (Leu et al., 1993, 1998). Multi-locus sequence typing suggests that this strain is genetically dissimilar to described X. fastidiosa genotypes and would be assigned a new subspecies or even a new species (Almeida and Nunney, 2015).

Güldr et al. (2005) suggested the occurrence of almond leaf scorch in Turkey, however this has not been confirmed by any other studies or surveys. According to the NPPO of Turkey (2014), this record should be considered as invalid.

The citrus disease has been reported from Argentina (Brlansky et al., 1991) and spread within Brazil, from Sao Paulo and Minas Gerais, as first reported in 1987, to north-eastern Brazil by 1996 (Laranjeira et al., 1996). The grapevine disease has been reported from Venezuela (Jimenez, 1985; Hernandez and Ochoa, 1994).

Pierce's disease has been confirmed in all states along the Gulf of Mexico from Florida to Texas; New Mexico, Arizona and California; northern Mexico and Costa Rica, but probably throughout Central America; Venezuela.

Phony peach has been confirmed in south-eastern USA from Florida north to North Carolina and west as far as Missouri and eastern Texas. Records for Montana (CABI/EPPO, 2006) and Nebraska (Sherald, 1993b) require further confirmation.

Plum leaf scorch is reported in Paraguay, Argentina, southern Brazil (Leite et al., 1997); and in the USA along the Gulf coast (Raju et al., 1982).

Coffee leaf scorch is widespread in Brazil (de Lima et al., 1996, 1998).

Tree leaf scorches are reported in the USA from Florida to New York and the Ontario Peninsula of Canada (rare, unconfirmed) and eastward to the central midwest (Kentucky, Kansas, Nebraska) and south-west as far as Texas.

See also CABI/EPPO (1998, No. 290).

The greatest threats are to mild winter regions with one or more of the following crops: grapevines, citrus, almond, stone fruits (Prunus spp.), coffee, oleander, lucerne or where the tree species that can be potentially affected by leaf scorch diseases occur.

EPPO considers X. fastidiosa as an A1 quarantine pest (EPPO, 1989) and it is also a quarantine pest for COSAVE. In the EPPO region, it is clear that the grapevine strain of X. fastidiosa has the potential to kill large numbers of grapevines and to make areas unfit for growing Vitis vinifera. Its vectors in North America do not occur in the EPPO region, but vector capacity is so non-specific that one could certainly expect European Cicadellinae (e.g., Cicadella viridis) or Cercopidae to transmit the bacterium if introduced. After the recent introduction and spread of the vector Homalodisca vitripennis in California, USA, this leafhopper has become extremely abundant on citrus and other crops in southern California (Sorensen and Gill, 1996). This illustrates that this exotic vector species can become abundant in Mediterranean climates, despite originating in a humid subtropical region (south-eastern USA). The main danger in the long term in Europe is that X. fastidiosa could become established in natural vegetation which would then act as a reservoir for infection of vineyards or other susceptible crops or forest trees. It is less likely that Pierce’s disease would become a problem in the production of planting material, for it is easily detected, can be prophylactically eliminated by hot-water treatment, and is rapidly self-eliminating if introduced into fields. Nevertheless, infected planting material could introduce the disease to new areas. Hoddle (2004) modelled the distribution of X. fastidiosa in California and projected the model to predict world distribution. Cold winter temperatures were seen as the limiting factor for X. fastidiosa establishment and X. fastidiosa is only likely to establish as a serious problem in the warmer parts of the EPPO region, which have winter temperatures approaching those of the southern USA. Species distribution models employed by Bosso et al. (2016a) suggest that X. fastidiosa in Italy has the potential to spread beyond the current boundaries of distribution and affect wide areas of Italy outside of Apulia. Mediterranean climates are particularly favourable for X. fastidiosa establishment and X. fastidiosa may establish within Portugal, Spain, Corsica, Albania, Montenegro, Greece and Turkey as well as all countries of northern Africa and the Middle East (Bosso et al., 2016b). The persistence of X. fastidiosa in the cooler climate of Kosovo, former Yugoslavia, over several years suggests Pierce's disease could also become established in coastal parts of southern France, and northern Italy. Information on the incidence and severity of the disease in the former Yugoslavia was not available as of early 1998. The potential natural range and severity of Pierce’s disease in Europe may depend on the distribution and biology of potential vectors and is accordingly rather difficult to assess.

The South American strain on citrus also presents a major risk, for the climatic conditions of Mediterranean countries seem favourable to its development. The reported damage in Brazil suggests potential damage even greater than for the grapevine disease. No critical evaluation of possible European vectors has yet been made, but the arguments developed above for grapevine should certainly also apply to citrus.

The peach strain is relatively less important, but X. fastidiosa still presents a definite danger for peaches, plums, almonds, oaks and, sensu lato, for other fruit and amenity trees (see Hosts) (Dunez, 1981).

X. fastidiosa occupies habitats of insect vectors where suitable host plants occur. For citrus, habitats are chiefly commercial orchards. For Pierce's disease strains, habitats may be riparian vegetation, irrigated pastures, hay fields, or ornamental landscapes.

No grapevine (Vitis spp.) species are known to be immune to Pierce’s disease strains of X. fastidiosa, but American species used as rootstocks (V. aestivalis, V. berlandieri, V. candicans, V. rupestris) and hybrids derived from them are tolerant and some may be resistant, as is V. rotundifolia (Goheen and Hopkins, 1988). Almonds and lucerne can be hosts of the grapevine strains, but the diseases caused by X. fastidiosa in these three crop species are independent within California, USA, suggesting as yet unidentified biological differences (Purcell, 1980b). A very high percentage (>75% of those tested) of crop, wild plant and weed species can carry Pierce’s disease strains of the bacterium without symptoms (e.g. wild grasses, sedges, lilies, various bushes and trees) (Raju et al., 1983; Hopkins and Adlerz, 1988; Hill and Purcell, 1995b). It is likely that in most symptomless host species, X. fastidiosa multiplies to lower populations and moves systemically less often than in pathological hosts. For example, blackberry (Rubus spp.) can be a systemic host, but the bacterium multiplies in mugwort (Artemisia douglasiana) without systemic movement (Hill and Purcell, 1995b). Hosts can be classified as propagative or non-propagative, systemic or non-systemic, and symptomatic or non-symptomatic (Purcell and Saunders, 1999b). Propagative, systemic hosts are the best hosts for efficient vector acquisition of bacteria, but vectors can acquire the bacterium from non-systemic hosts. Acquisition efficiency is proportional to the populations of live bacterial cells within plant tissues (Hill and Purcell, 1997).

Peach (Prunus persica) strains of X. fastidiosa cause peach phony disease (Wells et al., 1983), which also attacks Prunus salicina (causing leaf scald). All cultivars, forms and hybrids of peach are attacked, whether on their own roots or other rootstocks. Plums (Prunus domestica), almonds (P. dulcis), apricots (P. armeniaca) and the wild P. angustifolia were reported susceptible to phony disease before the association with X. fastidiosa was established. This reported range partly overlaps that of the grapevine-infecting strains. Various perennial weeds of orchards, such as Sorghum halepense, may act as reservoirs for the peach-infecting strain (Yonce, 1983; Yonce and Chang, 1987), but the plant host range of Prunus strains from the south-eastern USA has not been investigated extensively. Pierce's disease strains also cause almond leaf scorch disease (Davis et al., 1980), but the almond strains infect grape in low populations and without causing disease (Almeida and Purcell, 2003).

X. fastidiosa in the wide sense also causes leaf scorch in Acer rubrum (Sherald et al., 1987), Morus rubra (Kostka et al., 1986), Platanus occidentalis (Sherald, 1993a,b) (wilt and leaf scorch), Quercus rubra (Chang and Walker, 1988), Ulmus americana and Vinca minor (stunt). Strains from Ulmus and from P. occidentalis are not reciprocally infectious (Sherald, 1993a). The bacteria involved are not known to be transmissible to grapevine. Diseases of numerous woody ornamental plants in southern California, USA, including olive, date palm and rosemary, have been associated with X. fastidiosa but a causal relationship is still unproven (Wong and Cooksey, 2004). Until their relationships and pest significance have been clarified, they can all be regarded as potentially dangerous for Europe and the Mediterranean region.

X. fastidiosa causes citrus variegated chlorosis in Brazil (Lee et al., 1991; Chang et al., 1993; Hartung et al., 1994) and Argentina (Brlanksy et al., 1993). The disease affects mostly sweet oranges (Citrus sinensis); it has been observed especially on cultivars Pera, Hamlin, Natal and Valencia. It occurs on trees propagated on all commonly used rootstocks in Brazil: C. limonia, C. reshni and C. volkameriana. The disease has not been observed on C. latifolia or mandarins (C. reticulata), even when the trees were planted in severely affected orange groves (Li et al., 2000). The effectiveness of removing diseased citrus trees to prevent further spread of variegated chlorosis in citrus (Rodas, 1994) strongly suggests that most spread of this disease is from tree to tree within the crop. Control measures require the production of disease-free nursery trees in protected environments.

Citrus blight in Florida, USA, has been associated with X. fastidiosa (Adlerz et al., 1989; Hopkins et al., 1996); however the preponderance of evidence suggests that it is not the cause of blight (Derrick and Timmer, 2000).

Plum leaf scald is an important crop-limiting disease caused by X. fastidiosa from Brazil through Argentina. The South American plum leaf scald strains appear to differ from those in North America, as there are no reports of phony disease of peach in South America. The plum leaf scald strains in Brazil may have wide plant host ranges (Leite et al., 1997). A leaf scorching disease of coffee (De Lima et al., 1998) is caused by strains of X. fastidiosa that appear to be closely related to the citrus variegated chlorosis strains (Rosato et al., 1998), but its ability to cause disease in citrus (Li et al., 2001) is controversial.

In Europe and the Mediterranean region, grapevine and citrus are clearly the most significant potential crop hosts, although peach and plum are also important. Strains that cause leaf scorch diseases in oak, elm, sycamore (plane) (Hearon et al., 1980), mulberry (Kostka et al., 1986) and other tree species are also potentially damaging. Many other hosts could carry the bacterium, without necessarily being significantly affected.

X. fastidiosa has been implicated as the causal agent of olive quick decline syndrome in Europe. In 2013, X. fastidiosa subsp. pauca was associated with quick decline syndrome on olive, almond and oleander in Europe (southern Italy, Apulia region) (European Food Safety Authority, 2015). Symptomatic olive trees were often affected by multiple pests, including X. fastidiosa, several fungal species, and Zeuzera pyrina (leopard moth) (Nigro et al., 2013). Recent experimental evidence (Saponari et al., 2016) has confirmed X. fastidiosa as the causal agent of olive quick decline syndrome in Italy (European Food Safety Authority, 2016). In the USA a study evaluating olive as a host for X. fastidiosa concluded that subsp. multiplex was present but was not the cause of the leaf scorch and dieback symptoms observed on olive trees in California (Krugner et al., 2014). However, X. fastidiosa subsp. pauca has been implicated as a causal agent of olive plant dieback and leaf desiccation in Argentina (Haelterman et al., 2015). More recently, leaf scorch symptoms on olive trees in Brazil have been associated with X. fastidiosa subsp. pauca (Coletta-Filho et al., 2016).

The host range of X. fastidiosa based on the available peer-reviewed literature is presented in European Food Safety Authority (2015).
According to the European Food Safety Authority (2016), the current list of host plant species for X. fastidiosa consists of 359 plant species (including hybrids) from 204 genera and 75 different botanical families.

X. fastidiosa proliferates only in xylem vessels, in roots, stems and leaves. The vessels are ultimately blocked by bacterial aggregates and by tyloses and gums formed by the plant, although disease induction may not be due simply to water blockage (Goodwin et al., 1988). Pierce's disease strains of the bacterium are efficiently acquired by vector insects, with no latent period, and persist in infective adult insects indefinitely (Severin, 1949). In tests made during the 1940s in California, USA, 75 of 100 tested plant species proved to be hosts from which vectors could acquire the pathogen by feeding (Freitag, 1951). Additional studies with culturing and targeted gene amplification (PCR) of X. fastidiosa revealed additional symptomless hosts (Purcell and Saunders, 1999b; Costa et al., 2004; Wistrom and Purcell, 2005). Most of these hosts have only very mild (slight stunting) or no symptoms when infected.

Winter climate is a key factor in delimiting the areas where X. fastidiosa can persist from one season to the next. Pierce's disease and phony disease only occur in areas with a mild winter, presumably in relation to survival of the bacterium in dormant plants (Purcell, 1980a). Experimental cold therapy of diseased grapevines suggests that freezing temperatures can eliminate the bacterium directly from plants (Purcell, 1980a). Wet winters promote survival of high vector populations and favour disease spread in regions with dry summers. In temperate climates with regular freezing winter temperatures, infections of X. fastidiosa established in grapevines during the early growing season are most likely to persist until the following year (Purcell, 1981; Feil et al., 2003a). In California, USA, there is no evidence of vine to vine spread of chronic infections, probably because when bacteria first appear in new foliage (Smart et al. 1998), the chances of establishing infections that will survive the subsequent winter are small. For this reason, vectors that survive the winter may be especially important in establishing early season infections. The lack of potential vectors that overwinter as adults and thus can introduce new infections into woody plants may explain the absence of natural spread or establishment in temperate climate regions such as Europe.

X. fastidiosa is reported to be transmitted to seedlings from infected fresh seed in citrus (Li et al., 2003). This should be taken into account when importing citrus seed for propagation but is not likely to present a risk for introducing the bacterium via commercial fruit. The bacterium is sensitive to dry conditions commonly found in most seeds, but tests of seed transmission have only been reported for citrus. It tends to multiply rapidly in infected grapevines. Infected vines are conspicuous the year after infection and are easily destroyed or die within one to several years, depending on vine cultivar and age. Newly planted vines that are infected by grafting die within a year. In non-systemic hosts such as willow, X. fastidiosa dies within weeks of infecting a xylem cell via vector inoculation (Purcell and Saunders, 1999). Bacterial populations are maintained in natural vegetation by vector transmission to and latent multiplication in wild plants.

On citrus, the variegated chlorosis disease appears to spread from tree to tree within citrus groves (Laranjeira, 1997; Laranjeira et al., 1998). The 9-12 months incubation before symptoms appear in orange trees facilitates the spread of infected nursery trees and probably explains the rapid movements of the disease across Brazil within a decade (Coletta-Filho et al., 2000). Regulations in Brazil now require that nursery trees are produced under screens to prevent infection. The vectors of citrus strains appear to have the same vector group specificity (suctorial xylem feeders) as grapevine strains, but transmission trials indicate that vector transmission is much less efficient in citrus compared with grape (Lopes et al., 1996; Roberto et al., 1996; Yamamoto et al., 2002).

Vectors

Virtually all sucking insects that feed predominantly on xylem fluid are potential vectors (Purcell, 1989). Leafhoppers (Cicadellidae) in the subfamily Cicadellinae (sharpshooters) and spittle bugs or froghoppers (Cercopidae) are by far the most common species of known vectors within the natural range of X. fastidiosa in North America. As previously predicted, cicadas (Cicadidae) are vectors as well (Paião et al., 2002). Cicadella viridis (Cicadellinae) and the meadow spittle bug, Philaenus spumarius (Cercopidae), are common and widespread in central and southern Europe (Pavan and Gambon, 2003) and reported to occur on grape (Hering, 1966). In California, USA, all tested members of the subfamily Cicadellinae (including Carneocephala fulgida, Draeculacephala minerva and Graphocephala atropunctata) were vectors of the grapevine strain (Hewitt et al., 1949). Homalodisca vitripennis, H. insolita, Oncometopia orbona, Graphocephala versuta and Cuerna costalis are reported vectors of the peach strain (Turner and Pollard, 1959; Yonce, 1983). All these are xylem-feeding, suctorial insects which acquire the bacterium rapidly when feeding (less than 2 h). The bacterium adheres to the mouthparts and is released directly from them when the insect feeds again (Purcell et al., 1979). It multiplies in the vector but does not circulate in its haemolymph, nor does it require a latent period before transmission (unlike a phytoplasma). Very few live cells in the vector's head are needed for efficient transmission (Hill and Purcell 1995a). Transmission is usually from wild, generally symptomless, hosts to cultivated hosts (grapevines, peaches) rather than between cultivated hosts, though the latter can occur. Feeding preferences of G. atropunctata for different grapevine cultivars have been noted, but these seem to not have a role in field susceptibility (Purcell, 1981). A study of X. fastidiosa vectors on olive trees in Italy suggests that potential vectors may include the phloem-feeder Euscelis lineolatus (Elbeaino et al., 2014b). Further studies are needed to confirm transmission.

The biology of the vectors is important in understanding the epidemiology of the disease. In California, USA, species such as D. minerva and C. fulgida inhabit permanent pastures alongside vineyards, or live on weeds within them. Irrigation and weed control practices which produce foci of preferred host plants, including Cynodon dactylon and Echinochloa crus-galli, increase vector populations and the spread of the bacterium (Purcell and Frazier, 1985). Other species, such as G. atropunctata, multiply on grapevine but overwinter on a variety of other wild hosts, chiefly in riparian vegetation.

The most sensitive and reliable detection method is polymerase chain reaction (Minsavage et al., 1994). Positive results have been obtained with lucerne, peach, oak, citrus, oleander, plum and numerous grape and almond strains with the Minsavage et al. primer. ELISA is less sensitive and problems have arisen with false positives. Culturing on PW medium (Davis et al., 1983) provides definitive proof of living bacteria in a plant, estimates of populations of viable bacteria (Hill and Purcell, 1995a), and facilitates later strain identification.

Work by Pooler et al. (1997) allows a much larger sized sample of plant tissue to be processed for PCR and avoids plant inhibitors, which are probably the biggest obstacle to using PCR to detect X. fastidiosa in plant materials, in addition to the irregular distribution of the bacterium within plants making it very easy to miss in the small tissue samples used for PCR. Some plants have inhibitors of PCR that a typical DNA extraction procedure would not completely avoid.

Li et al. (2013) have recently developed two TaqMan-based assays, one targeting the 16S rDNA signature region for the identification of X. fastidiosa at the species level and the other specific identification of the citrus variegated chlorosis (CVC) strains (Li et al., 2013).

Symptoms are not reliable for detection of infected plants in transit.

X. fastidiosa can be detected microscopically (light or electron) in vessels in cross-sections of petioles (French et al., 1977) or by examining xylem sap squeezed from symptomatic stems or petioles or flushed from stems or petioles onto microscope slides (De Lima et al., 1998). Flushing of xylem sap from shoots with a pressure chamber allows the testing of larger sample sizes and avoids inhibitors for PCR (Bextine and Miller, 2004). Methods such as grafting to susceptible indicator plants or vector tests (Hutchins et al., 1953) are still available, and may have their place in certification schemes in which woody indicators are routinely used. X. fastidiosa can also be isolated onto suitable selective media (Davis et al., 1978, 1983; Raju et al., 1982; Wells et al., 1983). The identity of cultured bacteria can be confirmed by SDS-PAGE (Bazzi et al., 1994). Serological methods are less sensitive (10- to 100-fold) than culture but are the easiest means of detecting and identifying the bacterium, by ELISA or use of fluorescent antibodies (French et al., 1978; Walter, 1987; Hopkins and Adlerz, 1988; Sherald and Lei, 1991). Strains differ in quantitative reaction to antisera and in ease and efficiency of culture. DNA hybridization probes and PCR primers specific to X. fastidiosa have been developed (Firrao and Bazzi, 1994; Minsavage et al., 1994). X. fastidiosa can also be detected in its insect vectors (Yonce and Chang, 1987). The characterization and identification of strains chiefly employs molecular genetic methods (e.g., Chen et al., 1992; Hendson et al., 2001; Coletta-Filho et al., 2003), and can be expected to remain indefinitely in a state of change.

Different diagnostic methods used or developed for the detection and identification of X. fastidiosa are detailed in Janse (2009). Recent advances in detection include on-site molecular detection using real-time loop-mediated isothermal amplification (Yaseen et al., 2015).

Zinc deficiency may cause similar patterns of chlorosis in grape and citrus but are distinctively different in pattern to an experienced observer. Diseased plants do not respond to zinc fertilization. In grape, esca (France, Italy, Spain) or measles (USA) has leaf scorching symptoms similar to Pierce's disease, but esca normally occurs early in the summer (June, July in Europe) following at least several days of hot weather, whereas Pierce's disease symptoms begin to appear after fruits begin to colour (verasion).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Phytosanitary Measures

Grapevine-growing countries should prohibit or severely restrict importation of grapevine planting material from countries where X. fastidiosa occurs. As recommended by EPPO (1990), if planting material is imported under licence, it should be maintained in post-entry quarantine for 2 years and shown to be free from the pest. Imported plants and fruits should be free from vectors, possibly by use of an appropriate treatment. A temperature treatment has been shown to eliminate the bacterium (45°C for at least 3 h) (Goheen et al., 1973), and may have potential as a phytosanitary measure.

Citrus-growing countries should similarly prohibit or severely restrict importation of citrus planting material from South America. Peach and other Prunus material from a country where the peach or plum strain occurs should come from a reliable certification scheme, with particular emphasis on preventing re-infection of healthy material via the vectors.

While the hazard presented by X. fastidiosa in other hosts (oak, plane, maple and others) still has to be evaluated, inspection services should be aware that these hosts and the many symptomless hosts also present a certain risk.

Chemical Control

Chemical control of Xylella diseases has not been successful in the field. Tetracycline drenches cause a temporary remission of symptoms in potted grapevines (Hopkins and Mortenson, 1971).
 

19/07/16 Reviewed by:

Rebekah Robinson, Royal Horticultural Society, RHS Garden Wisley, Woking, UK