A1

EMERGING INSECT-TRANSMITTED PLANT DISEASES: THE BACTERIUM XYLELLA FASTIDIOSA AS A CASE STUDY¹

Rodrigo P. P. Almeida^2*and L. Nunney³

Introduction

Emerging vector-borne plant diseases may have severe economic, social, environmental, and cultural impacts. Factors driving the emergence of these diseases include vector and/or pathogen introductions into new areas where susceptible plant host species occur, the adaptation of pathogens and their vectors to management strategies such as pesticides or pathogen resistant plant varietal selections, the emergence of novel pathogens, as well as human-mediated environmental changes such deforestation and climate. Unlike animal and human emerging diseases, however, there is no recent large scale analysis of global trends of the types of emerging diseases affecting plants, or what are the main factors driving the emergence of these diseases (the last analysis being Anderson et al., 2004). For this reason, we chose to address the issue of the emergence of plant diseases by choosing one representative pathogen, and exploring some of the factors responsible for its rise from relative obscurity one or two decades ago (Hopkins and Purcell, 2002; Purcell, 2013). Thus, in this Chapter we address in detail the factors affecting the emergence of one important insect-transmitted plant pathogen, the bacterium Xylella fastidiosa.

Biology of a Plant and Insect Colonizer

Xylella fastidiosa is a bacterium that colonizes two distinct habitats, the xylem network of host plants and the foregut of xylem-sap feeding insects (Chatterjee et al., 2008). Processes leading to plant colonization are yet to be fully understood. Movement from vessel to vessel occurs through intact and damaged pit membranes, and is a necessary process for successful X. fastidiosa movement within plants (Baccari and Lindow, 2011; Chatelet et al., 2006; Newman et al., 2003). The specific mechanisms leading to disease remain poorly understood, but recently studies addressing this question from a host plant perspective suggest

___________________

¹ This article has been accepted for publication in the peer-reviews journal Plant Disease and was slightly modified for inclusion in this Workshop Summary.

² Department of Environmental Science, Policy and Management, University of California, Berkeley.

³ Department of Biology, University of California, Riverside.

^* Corresponding author: e-mail: rodrigoalmeida@berkeley.edu.

that symptom development is, initially, a consequence of physiological responses commenced by water deficit responses (Choi et al., 2013; Daugherty et al., 2010b; Sun et al., 2013).

Research on X. fastidiosa focuses on its role as a plant pathogen; however, to understand its ecology and evolution we propose that a broader view is necessary, recognizing that disease is the outcome of interactions between specific pathogen genotypes and host species (Casadevall and Pirofski, 2014). Infection dynamics of X. fastidiosa will be influenced by the extensive list of host plants species that can be infected (at least temporarily), the plant-host specificity of different genotypes, and the wide range of potential insect vectors. In 1995 Hill and Purcell (1995) compiled published data and concluded that plants in 29 families were hosts of this bacterium. More recently, a report listed 309 plant species in 63 families as hosts of X. fastidiosa (EFSA PLH Panel, 2015). This bacterium is capable of persisting at the inoculation site in many plant species under greenhouse and field conditions when either insect or mechanically inoculated (Purcell and Saunders, 1999). It can also be recovered from a wide range of weedy plants in infected agricultural areas (e.g., Lopes et al., 2003). X. fastidiosa does not appear to cause disease in most of these species; however the available data suggest that these asymptomatic infections typically declined over time (Purcell and Saunders, 1999). Thus, X. fastidiosa colonization of plants does not equal disease development.

Even though X. fastidiosa is a plant pathogen of considerable economic importance, mechanisms of host plant-pathogen specificity remain unknown and a major question in the field. The limited genomic structural variability within X. fastidiosa suggests phylogenetic groups colonizing different host plants have similar pathogenicity machineries (van Sluys et al., 2003). The only study to address the mechanisms of host plant specificity experimentally showed that an isolate could expand its host range if a cell-cell signaling-based gene regulation system was disrupted, suggesting that alleles or gene regulation, but not loci, were associated with specificity (Killiny and Almeida, 2011). But the approach used did not allow for the identification of candidate loci for future testing, and therefore the question remains largely open.

Disease is the outcome of complex X. fastidiosa-plant interactions and is obviously important, but it is not known what proportion of the interactions X. fastidiosa engages in within natural ecosystems result in plant disease. This leaves open the possibility that disease may represent a relatively small proportion of these interactions, leading to the suggestion that X. fastidiosa may be considered primarily an endophyte rather than a pathogen (Chatterjee et al., 2008). It may be notable that in X. fastidiosa cell-cell signaling regulates limited virulence to plants while promoting vector plant-to-plant transmission (Newman et al., 2004). Transmission rates can also be directly affected by vectors responding to disease symptoms. In two X. fastidiosa disease systems studied, sharpshooter leafhoppers do not avoid infected yet asymptomatic plants, but discriminate against infected and symptomatic plants, or healthy plants painted to simulate disease

symptoms (Daugherty et al., 2011; Marucci et al., 2005). This behavior may be advantageous for these insects: water stressed- and X. fastidiosa-infected plants have some shared physiological characteristics, of which xylem sap under high tension is of paramount relevance. Increased tension in the water column leads to a food source that is energetically expensive, resulting in the ingestion of less xylem sap (Andersen et al., 1992; Miranda et al., 2013) and possibly promoting the movement of vectors to another host. Since symptomatic plants are heavily colonized by X. fastidiosa (Newman et al., 2003), vector avoidance may act to reduce transmission rates (Zeilinger and Daugherty, 2014) and select for decreased bacterial virulence. This effect could be important when transmission rates are low; however, if vectors are common and transmission rates high, rapid bacterial growth leading to increased virulence may be favored, a pattern often observed in diseases that are transmitted between hosts (e.g., malaria; de Roode et al., 2005).

Experimentally identified insect vectors of X. fastidiosa belong to two insect groups, the sharpshooter leafhoppers (Cicadellidae, Cicadellinae) and spittlebugs (superfamily Cercopoidea, with five species of Aphrophoridae and two species of Clastopteridae identified) (Almeida et al., 2005; EFSA PLH Panel, 2015). In addition, there are two reports of cicadas (Cicadidae) transmitting X. fastidiosa (Krell et al., 2007; Paião et al., 1996), which need to be confirmed through larger experiments. Colonization of these insects by X. fastidiosa occurs in a noncirculative yet persistent manner (Purcell and Finlay, 1979), with the bacterium colonizing the foregut on insect vectors (Purcell et al., 1979). Consequently, there is no transovarial or transtadial transmission (Almeida and Purcell, 2003; Freitag, 1951; Purcell and Finlay, 1979). Colonization of regions in the foregut called cibarium and precibarium were first shown microscopically (Brlansky et al., 1983; Purcell et al., 1979), and later correlated with insect inoculation of plant hosts during feeding (Almeida and Purcell, 2006). So far, no other plant pathogen is known to be transmitted in a similar manner, with the possible exception of Ralstonia syzigii, which is transmitted by spittlebugs in the Machaerotidae (Eden-Green et al., 1992).

Transmission efficiency of X. fastidiosa increases with both the time an insect feeds on an infected host plant (acquisition) and the subsequent time it feeds on an uninfected host (inoculation), up to 48-96 hours (Almeida and Purcell, 2003; Purcell and Finlay, 1979). Presumably a longer feeding time increases the likelihood of insect vectors reaching colonized xylem vessels in the case of acquisition, and performing specific probing behaviors in the case of inoculation. The colonization of a vector by the bacteria is a critical part of acquisition and is a complex process, similar to biofilm formation on surfaces, which has been explored in some detail (e.g., Killiny and Almeida, 2009a,b, 2014). Specific probing behaviors involved in inoculation are yet to be determined; however the inoculation of X. fastidiosa into dormant grapevines with positive xylem sap pressure (positive root pressure) indicates that vector behaviors are required for the inoculation of bacterial cells into plants (Almeida et al., 2005).

One important aspect of X. fastidiosa transmission relevant to the emergence of new diseases is that it lacks vector specificity (Almeida et al., 2005). The insect groups that transmit X. fastidiosa are distributed worldwide in tropical and temperate climates, and all insect species belonging to the above-mentioned groups should be considered as potential vectors until proven otherwise. For example, one vector species has been shown to transmit X. fastidiosa isolates belonging to four different X. fastidiosa subspecies (Almeida and Purcell, 2003; Purcell et al., 1999; Sanderlin and Melanson, 2010; Saponari et al., 2014). And a X. fastidiosa subspecies originally from South America has been transmitted by various vectors in South America, one in North America, and another in Europe (Brlansky et al., 2002; Damsteegt et al., 2006; Marucci et al., 2008; Saponari et al., 2014). This lack of specificity increases the likelihood that newly introduced X. fastidiosa isolates, when reaching a novel environment, will be transmitted by an endemic vector species. However, while the ability to transmit X. fastidiosa is widespread, transmission efficiency is highly variable and dependent on a range of vector-plant-pathogen interactions (Lopes et al., 2009). Transmission efficiency may vary for different vector species on the same host plant (Daugherty and Almeida, 2009; Lopes et al., 2009), or the same vector species feeding on different tissues of the same plant (Daugherty et al., 2010a); however, observations suggest that the general mechanisms of transmission are conserved. The one caveat is that most of the research on X. fastidiosa transmission has been conducted with two vector species (Graphocephala atropunctata and Homalodisca vitripennis) and one X. fastidiosa subspecies (subsp. fastidiosa), so it is important that a broader range of taxa be studied to confirm these results. Until that is done, the effectiveness of individual sharpshooter leafhopper species in transmitting X. fastidiosa should not be extrapolated from epidemic to epidemic without considering the novel ecological context.

A Plant Generalist or Not: Revisiting Xylella fastidiosa Systematics

Xylella fastidiosa currently is the sole species in the genus Xylella; Xanthomonas spp. are sister taxa to X. fastidiosa (Retchless et al., 2014). As noted earlier, X. fastidiosa has traditionally been referred to as a having a “wide host range” or as a “generalist”. This is accurate in the sense that a very large number of plant species have been demonstrated to sustain X. fastidiosa infections; however, there is mounting evidence suggesting that while this description is accurate it is misleading. Specifically, very few of these plants sustain long-term infections and become symptomatic. Furthermore, it is now clear that specific symptomatic hosts are only susceptible to isolates in one or a limited number of X. fastidiosa phylogenetic clades, with the result that specific clades of X. fastidiosa have a small number of symptomatic host plant species (Nunney et al., 2013). Such insights are of great relevance in understanding disease outbreaks.

We revisit X. fastidiosa taxonomy in face of new findings and, consequently, novel questions, with two important caveats. First, it is fully expected that new clades of X. fastidiosa will be reported in the future (e.g., Nunney et al., 2014a). Second, many of the plant species listed as hosts of X. fastidiosa should in fact be considered putative hosts since most associations studied so far are derived from symptomatic plant tissue, but without experimental work to confirm the pathogenicity of isolates. Although associations are relevant, the fulfillment of Koch’s postulates is a requirement to demonstrate that individual genotypes are pathogenic to specific host plant species. The importance of experimental work to determine the host range of pathogens remains paramount. It is possible that ecological conditions limit the host range and/or virulence of pathogens, which may be ‘released’ in new environments where other vector species and host plants are present, or abiotic factors such as climate and precipitation vary. In summary, we emphasize the importance of experimentally determining plant species susceptibility to X. fastidiosa, as there are plant species-pathogen genotype associations that do not lead to disease. Moreover, even when symptoms eventually develop, a delay of several months following infection is not uncommon. These issues are especially relevant given the economic and quarantine importance of this bacterial species. In this context, it is important to note that so far no native plant hosts of X. fastidiosa in South and Central America have been identified. In contrast, a number of native hosts (primarily trees) of the North American subsp. multiplex have been identified, including several oak species (Quercus spp.), American elm (Ulmus Americana), American sycamore (Platanus occidentalis), sweetgum (Liquidambar styraciflua), and pecan (Carya illinoinensis) (for a more complete list see Table 2 in Nunney et al. [2013]).

Studies of X. fastidiosa genetic and phenotypic diversity have historically been confusing and inconclusive, despite the efforts of a small and dedicated group of scientists. Purcell (2013) made a case for the importance of researchers naïve to a new field of science being able to address old questions, and the advent of DNA sequencing has facilitated studies of X. fastidiosa diversity and host range that were until recently technically intractable. As a result, we can argue that the main drivers of conflicting results can be summarized under four headings. First, isolates from a small range of host plants and geographical distribution have been used for studies, over-representing a limited and narrow sampling of genetic diversity. Second, procedures for typing have relied on within-study comparisons of the above-mentioned small number of available isolates with methods that provided inadequate phylogenetic resolution. Third, methodological differences in the typing of isolates limited comparisons among studies, a problem that is disappearing now that sequencing technology is easily available worldwide. Lastly, X. fastidiosa is naturally competent (Kung and Almeida, 2011); gene flow deeply impacts the systematics and evolution of bacteria (Polz et al., 2013).

The current view of X. fastidiosa genetic diversity has overcome most of these limitations, largely through the use of a portable multi-locus sequence

typing (MLST) approach (Maiden et al., 1998). MLST for X. fastidiosa was first introduced by Scally in 2005 (Scally et al., 2005) and refined by five years later into the form currently employed (Yuan et al., 2010). MLST has been successfully used to study X. fastidiosa diversity at the species/subspecies level, and to infer the phylogenetic placement of newly identified isolates. These data have resulted in a robust taxonomy for the species. Furthermore, the MLST classification of isolates into sequence types (STs) (unique genotypes based on the 7 loci used in MLST) has provided insights about X. fastidiosa evolution and host specificity. For example, comparing subsp. pauca STs found on coffee and citrus, it has been shown that in general they are reciprocally host specific (Almeida et al., 2008; Nunney et al., 2012).

Based on current knowledge, X. fastidiosa is primarily a species of the Americas. A distant relative is found in Taiwan (Su et al., 2014), but should probably be classified as a separate species. Two other exceptions that must yet be confirmed and for which no genetic information is available, are reports from Iran (Amanifar et al., 2014) and Turkey (Guldur et al., 2005). Lastly, the recent introduction of X. fastidiosa into Italy is an important change to its geographical distribution (Saponari et al., 2013). The American representatives were initially divided into three subspecies subsp. fastidiosa, multiplex and pauca based on DNA-DNA hybridization data (Schaad et al., 2004). MLST sequence data confirmed the status of these subspecies, and suggested a fourth, subsp. sandyi, which was not present among the earlier strains that were tested (Scally et al., 2005). Subsequent sampling and analysis based on MLST has indicated that these subspecies evolved in geographical isolation with subsp. pauca native to South America (Nunney et al., 2012), subsp. multiplex native to temperate and subtropical North America (Nunney et al., 2012, 2014b), subsp. fastidiosa is found in Costa Rica and is presumed to be native to southern Central America (Nunney et al., 2010), and subsp. sandyi has only been detected in southern regions of the USA (Yuan et al., 2010). Subspecies morus represents a new proposal and is discussed below. Historical geographical isolation of the original four subspecies is consistent with the known biology of X. fastidiosa: this bacterium can only invade a new region by long-distance dispersal of infected insects or infected plants. In the absence of human intervention, the former is very unlikely and the latter is close to impossible. However, it has become apparent that in the recent past human-mediated invasion is the primary driver of economically costly X. fastidiosa introductions. We discuss three main pathways leading to the emergence of X. fastidiosa diseases, following examples available in the literature.

Introduction of Exotic Genotypes

The most common pathway leading to X. fastidiosa epidemics is the introduction of exotic genotypes into environments that are ecologically prone to the maintenance of the bacterium in the plant community. Although the introduction

of insect vectors carrying X. fastidiosa represents a potential pathway, only one vector species is considered invasive (Homalodisca vitripennis, Cicadellidae, a sharpshooter leafhopper), and another is distributed beyond its region of origin (Philaneus spumarius, Aphrophoridae, a spittlebug). The expansion in the geographical range of these species has not been associated with the spread of X. fastidiosa, therefore we consider this an unlikely route. The main dispersal pathway would then be the movement of infected, and potentially asymptomatic, plant material from areas where the pathogen occurs. A recent report by the European Food Safety Authority evaluation on the risk of X. fastidiosa introductions into the European Union reached similar conclusions (EFSA Panel on Plant Health, 2015) with a much more detailed and systematic analysis of potential pathways. Here we discuss examples with conclusive evidence from the available literature.

The most recent case of an introduction is the outbreak of rapid olive decline in the Apuglia region in southern Italy, first reported in October 2013 (Saponari et al., 2013). While the distribution and consequences of this introduction are yet to be determined, it is known that this outbreak is associated with a strain of X. fastidiosa subsp. pauca classified as ST53 (Elbeaino et al., 2014). Subspecies pauca is of South American origin but this sequence type so far has not been found in South America; however, it has been detected in Costa Rica infecting primarily oleander (Nunney et al., 2014a). Thus, this particular ST of subsp. pauca has now been introduced into two new regions, and infecting novel hosts. While olive is currently considered the primary host in the Italian outbreak, infection of oleander has also been observed, illustrating a common feature of X. fastidiosa: oleander and olive are hosts of the same strain, and yet they are in different Orders (Gentianales vs. Lamiales). As a result, given our current knowledge, it is not possible to predict potential hosts following an invasion.

Yet another introduction involved the best studied X. fastidiosa disease, Pierce’s disease of grapevines. It had been first proposed that the Gulf Coast Plain area of the USA was the center of origin of the etiological agent of the disease based on the fact that species of grapevines (Vitis spp.) native to the USA were tolerant to infection, while the exotic European grapevine (Vitis vinifera) was susceptible (Hewitt, 1958). With the recent availability of larger datasets on the genetic diversity of X. fastidiosa, we now know that the genotype causing disease in grapevines in the USA originated from Central America (Nunney et al., 2010). The lack of genetic diversity among isolates belonging to this clade in the USA is evidence of a relatively recent introduction (Yuan et al., 2010), and it has been proposed that the introduction into the USA of a single genotype was via an infected coffee plant, a known host of X. fastidiosa in Central America (Nunney et al., 2010). Isolates derived from this single genotype are now widely distributed through grape-growing regions of the USA, from Florida to California. Interestingly, an isolate from this same almost monomorphic clade found in the USA has now been reported causing Pierce’s disease of grapevines in Taiwan (Su et al., 2013), suggesting that X. fastidiosa infected-plant material originating from

the USA was inadvertently introduced into the country, eventually leading to an epidemic.

A similar scenario appears to have occurred with the emergence of plum leaf scald in Argentina, Paraguay, and Brazil (French and Kitajima, 1978; Kitajima et al., 1975). The disease in plum and other Prunus species were known in the southeast USA, but the origin of the X. fastidiosa genotype(s) causing plum leaf scald in South America remained unidentified until Nunes et al. (2003) studied the gene content of several isolates. They determined that the tested plum isolate from Brazil grouped with North instead of South American isolates (i.e., belonging to subsp. multiplex), demonstrating yet another introduction, this time from the USA to South America. These examples illustrate the challenges of limiting the inadvertent transportation of X. fastidiosa-infected plant material from one country, or continent, to another. For a detailed risk assessment analysis of X. fastidiosa introductions we direct readers to a recent review by the European Food Safety Authority (EFSA Panel on Plant Health, 2015).

Introduction of an Invasive Vector

To our knowledge there is only one example of a X. fastidiosa vector being considered invasive, spreading over vast geographical distances and reaching large populations at various environmental conditions (Grandgirard et al., 2006; Petit et al., 2008). Homalodisca vitripennis (Cicadellidae, Cicadellinae) is native to the southeastern USA (Turner and Pollard 1959; Young, 1958); in 1989 it was first detected in California (Sorensen and Gill, 1996), but only in the late 1990s it became a problem as the vector of X. fastidiosa driving the oleander leaf scorch epidemic in Southern California (Purcell et al., 1999), and at the same time a Pierce’s disease epidemic in the grape-growing region of Temecula, also in Southern California (Hopkins and Purcell, 2002). Estimates suggested 1−2 million insects per hectare in the region (Coviella et al., 2006), populations which are thought to have allowed for its transportation to French Polynesia (Grandgirard et al., 2006), where biological control successfully controlled very large populations that developed in those tropical regions (Grandgirard et al., 2008). It is notable that these invasions were not associated with the introduction of X. fastidiosa, supporting our view that the primary mechanism of X. fastidiosa invasions is the movement of infected live plants. In the case of California, the introduction of H. vitripennis had several important consequences; we focus here on the emergence of X. fastidiosa diseases alone. Newton Pierce, after whom Pierce’s disease was later named, studied the first known outbreak of the disease that was in Southern California (Pierce, 1892). Since that time, X. fastidiosa has been regularly reported in grapevines, almonds, and alfalfa, indicating it has been continuously present. However, disease outbreaks were primarily limited to small areas, apparently due to habitat specific of endemic vectors and the broader ecological context.

There were two main consequences associated with the extremely large populations of H. vitripennis in southern California in the two decades subsequent to its introduction (see Almeida, 2008, for further discussion). The first was the development of a Pierce’s disease epidemic, where very large populations of a relatively inefficient vector (H. vitripennis is not an efficient vector on grapevines when compared to other species), (Daugherty and Almeida, 2009) led to the effective spread of the pathogen to a focal crop under new ecological conditions, decimating the vineyards of the Temecula region (Hopkins and Purcell, 2002). Chemical control of H. vitripennis populations in the region has led to the restoration of the local wine industry to economically profitable levels (M. Daugherty personal communication). The second consequence is based on associations rather than conclusive epidemiological data; yet, the contention is well supported by field observations. We contend that the introduction of the highly polyphagous H. vitripennis, led to the establishment of various X. fastidiosa diseases in Southern California, notably oleander leaf scorch (Purcell et al., 1999) and scorch diseases of a range of ornamental trees (Hernandez-Martinez et al., 2007, 2009). A large list of diseases associated with X. fastidiosa has been generated, albeit Koch’s postulates have only been fulfilled for a few of them (e.g., Hernandez-Martinez et al., 2009; Purcell et al., 1999). We suggest that X. fastidiosa genotypes had been widely established in Southern California ahead of the H. vitripennis invasion, albeit restricted to disease cycles with endemic vectors and asymptomatic hosts or associated with species where it caused disease rarely enough to be overlooked. The presence of H. vitripennis resulted in increments of such rare events due to its large populations, or in the displacement of genotypes from endemic cycles to disease cycles that incorporated hosts of this invasive vector. The lack of vector-pathogen specificity is the trait most responsible for this outcome. In fact, H. vitripennis is the only vector species shown to transmit X. fastidiosa belonging to all currently accepted subspecies (fastidiosa, multiplex, sandyi, and pauca), although this should be expected from all known and potential X. fastidiosa vector species.

Recombination and Adaptation to New Plant Hosts

The anthropogenic introduction of X. fastidiosa subspecies into new regions can have two effects: the emergence of a known disease in a new area, and/or the emergence of a new disease involving a new plant host. In this section we focus on the second of these possibilities. One example of X. fastidiosa invading a new host is the case of mulberry leaf scorch. It was first noted in the early 1980s in Washington DC, and subsequent sampling revealed infected trees (the native Morus rubra) along the eastern seaboard as far north as southern New York (Kostka et al., 1986). Within a few years the disease was found on the west coast with infected trees (the introduced Morus alba) observed in California (Hernandez-Martinez et al., 2007). Initial genetic typing showed that the mulberry

isolates always grouped together, but their relationship to the other subspecies was marker dependent. The reason for this ambiguity was revealed using MLST: the genome is a roughly equal mix of genetic material from subsp. fastidiosa and multiplex, such that an examination of the 7 MLST loci revealed 3 alleles from subsp. fastidiosa, 3 from subsp. multiplex, and one chimeric allele containing sequence from both subspecies and consequently a recombination breakpoint (Nunney et al., 2014c). All other forms of X. fastidiosa are genetically very distinct from the mulberry type, which themselves show almost no genetic variability. Since they do not group with any pre-existing subspecies, and since they appear to be unique in naturally infecting mulberry, it’s been proposed that they define a new subspecies (subsp. morus), that was created by one or more massive genetic exchanges between subsp. fastidiosa and multiplex that created a chimeric genome via intersubspecific homologous recombination (IHR) (Nunney et al., in preparation).

The genetic exchange that created subsp. morus has also resulted in a group of genotypes (recombinant multiplex) that cluster with subsp. multiplex, presumably due to repeated backcross exchanges with the native subspecies (multiplex) (Nunney et al., 2014b). Of interest is that the isolates from diseased blueberry plants (from Georgia and Florida) were all of only two sequence types, both of which were recombinant subsp. multiplex. No non-recombinant subsp. multiplex have yet been isolated from blueberry strongly suggesting that we have a second example of genetic mixing between an introduced and native subspecies resulting in the infection of a new host. The involvement of IHR in the genesis of subsp. morus, and the subsequent formation of the group of recombinant subsp. multiplex, might seem like a special event unlikely to be repeated; however, we now have evidence that a similar genetic exchange occurred in South America. Studies of citrus and coffee X. fastidiosa isolates from Brazil have provided evidence of IHR from subsp. multiplex to subsp. pauca (Almeida et al., 2008; Nunney et al., 2012). Based on MLST data, Nunney et al. (2012) estimated that about half of the genome was polymorphic for subsp. multiplex sequence, suggesting that, as in the case of subsp. morus, one or more major genetic exchanges had occurred. However, non-recombinant subsp. pauca has not been found, although it seems probable that it will eventually be isolated by more thorough sampling away from agricultural areas. These examples highlight the important question of the consequences of gene flow on the emergence of X. fastidiosa diseases. We propose that the introduction of novel allelic diversity into countries/regions where X. fastidiosa is already present poses a significant risk and should be a major concern to regulatory bodies around the world.

In addition to host species switches induced by IHR, genetic exchange within subspecies occurs (Almeida et al., 2008; Nunney et al., 2013). This, together with IHR, may be highly relevant in determining the ability of X. fastidiosa to adapt to resistant plant genotypes. Specifically, the breeding programs that are developing resistant plant material for various X. fastidiosa hosts (notably wine grapes) should take into account the potential for X. fastidiosa to adapt. The groups of bacterial

genes that are frequently exchanged and maintained in a population and those that are quickly purged have not been identified. Similarly, general patterns of short- and long-term genome evolution have so far not been analyzed. These are essential components for the robust deployment of resistant plant material, transgenic or not, as the strong selective pressure on X. fastidiosa populations due to the usage of new technologies will eventually lead to the selection of novel pathogen variants that are capable of breaking down resistance. This process is equivalent to antibiotic resistance strains of human pathogens, such as tuberculosis, or loss of Bacillus thuringiensis derived plant resistance to pests. Our argument is not that new technologies will not be successful; our argument is that the evolution of X. fastidiosa needs to be considered and incorporated into management practices aimed at prolonging the utilization of such plant lines. That, however, cannot be done with the very superficial and limited knowledge currently available.

Last Thoughts

Xylella fastidiosa is no longer a plant pathogen limited to a few countries in the Americas, where its geographical distribution ranges from Canada to Argentina. The long-term presence of X. fastidiosa in Taiwan raises questions about its potential distribution in Asia, and its introduction into Europe and recent report from Iran will dramatically and permanently change its geographic range. Is this bacterium present elsewhere, or where is it not present? And, as shown recently in Central America (Nunney et al., 2014a), how much of the genetic diversity of X. fastidiosa remains to be described? Old and unaddressed questions are now more relevant than ever, especially for Europe and the Mediterranean basin, where the plant community has, as far as we know, not been exposed to X. fastidiosa. Among those is what drives host specificity in this pathogen, in other words, why do genotypes cause disease in one plant species and not another, while still being able to colonize various plant species with different degrees of success without inducing symptom expression. Finally, we still know very little about X. fastidiosa outside of its crop hosts. We are strong believers that much would be gained from studies of X. fastidiosa in natural environments, no only in regards to its biology, ecology, and evolution, but also on how to better manage diseases it causes in crops of economic importance.

Acknowledgments

We thank collaborators and colleagues with whom interactions helped shape our views on many of the topics discussed here. However, we are solely responsible for omissions and opinions expressed in this article. Xylella fastidiosa research in our groups has been funded primarily by the United States Department of Agriculture and California Department of Food and Agriculture Pierce’s Disease Program.

References

Daugherty, M. P., and R. P. P. Almeida. 2009. Estimating Xylella fastidiosa transmission parameters: Decoupling sharpshooter number and feeding period. Entomologia Experimentalis et Applicata 132:84–92.

Daugherty, M. P., J. R. S. Lopes, and R. P. P. Almeida. 2010a. Vector within-host feeding preference mediates transmission of a heterogeneously distributed pathogen. Ecological Entomology 35:360–366.

Daugherty, M. P., J. R. S. Lopes, and R. P. P. Almeida. 2010b. Strain-specific alfalfa water stress-induced by Xylella fastidiosa. European Journal of Plant Pathology 127:333–340.

Daugherty, M. P., A. Rashed, R. P. P. Almeida, and T. M. Perring. 2011. Vector preference for hosts differing in infection status: Sharpshooter movement and Xylella fastidiosa transmission. Ecological Entomology 36:654–662.

de Roode, J. C., R. Pansini, S. J. Cheesman, M. E. H. Helinski, S. Huijben, A. R. Wargo, A. S. Bell, B. H. K. Chan, D. Walliker, and A. F. Read. 2005. Virulence and competitive ability in genetically diverse malaria infections. Proceedings of the National Academy of Sciences of the United States of America 102:7624–7628.

Eden-Green, S. J., R. Balfas, and T. Sutarjo. 1992. Characteristics of the transmission of Sumatra disease of cloves by tube-building cercopoids, Hindola spp. Plant Pathology 41:702–712.

EFSA PLH Panel (EFSA Panel on Plant Health). 2015. Scientific opinion on the risks to plant health posed by Xylella fastidiosa in the EU territory, with the identification and evaluation of risk reduction options. EFSA Journal 13:2989.

Elbeaino, T., F. Valentini, A. K. Kubaa, P. Moubarak, T. Yaseen, and M. Digiaro. 2014. Multilocus sequence typing of Xylella fastidiosa isolated from olive affected by ‘Olive Quick Decline Syndrome’ in Italy. Phytopathologia Mediterranea 53:533–542.

Freitag, J. H. 1951. Host range of the Pierce’s disease virus of grapes as determined by insect transmission. Phytopathology 41:920–633.

French, W. J., and E. W. Kitajima. 1978. Occurrence of plum leaf scald in Brazil and Paraguay. Plant Disease Reporter 62:1035–1038.

Grandgirard, J., M. S. Hoddle, G. K. Roderick, J. N. Petit, D. Percy, R. Putoa, C. L. Garnier, and N. Davies. 2006. Invasion of French Polynesia by the glassy-winged sharpshooter, Homalodisca coagulata (Hemiptera: Cicadellidae): A new threat to the South Pacific. Pacific Science 60:429–438.

Grandgirard, J., M. S. Hoddle, J. N. Petit, G. K. Roderick, and N. Davies. 2008. Engineering an invasion: Classical biological control of the glassy-winged sharpshooter, Homalodisca vitripennis, by the egg parasitoid Gonatocerus ashmeadi in Tahiti and Moorea, French Polynesia. Biological Invasions 10:135–148.

Guldur, M. E., B. K. Caglar, M. A. Castellano, L. Ulnu, S. Guran, M. A. Yilmaz, and G. P. Martelli. 2005. First report of almond leaf scorch in Turkey. Journal of Plant Pathology 87:246.

Hernandez-Martinez, R., K. A. de la Cerda, H. S. Costa, D. A. Cooksey, and F. P. Wong. 2007. Phylogenetic relationships of Xylella fastidiosa strains isolated from landscape ornamentals in Southern California. Phytopathology 97:857–864.

Hernandez-Martinez, R., D. A. Cooksey, and F. P. Wong. 2009. Leaf scorch of purple-leafed plum and sweetgum dieback: Two new diseases in Southern California caused by Xylella fastidiosa strains with different host ranges. Plant Disease 93:1131–1138.

Hewitt, W. B. 1958. The probable home of Pierce’s disease virus. Plant Disease Reporter 42:211–215.

Hill, B. L., and A. H. Purcell. 1995. Multiplication and movement of Xylella fastidiosa within grapevine and four other plants. Phytopathology 85:1368–1372.

Hopkins, D. L., and A. H. Purcell. 2002. Xylella fastidiosa: Cause of Pierce’s disease of grapevine and other emergent diseases. Plant Disease 86:1056–1066.

Killiny, N., and R. P. P. Almeida. 2009a. Xylella fastidiosa afimbrial adhesins mediate cell transmission to plants by leafhopper vectors. Applied and Environmental Microbiology 75:521–528.

Killiny, N., and R. P. P. Almeida. 2009b. Host structural carbohydrate induces vector transmission of a bacterial plant pathogen. Proceedings of the National Academy of Sciences of the United States of America 106:22416–22420.

Killiny, N., and R. P. P. Almeida. 2011. Gene regulation mediates host specificity of a bacterial pathogen. Environmental Microbiology Reports 3:791–797.

Killiny, N., and R. P. P. Almeida. 2014. Factors affecting the initial adhesion and retention of the plant pathogen Xylella fastidiosa in the foregut of an insect vector. Applied and Environmental Microbiology 80:420–426.

Kitajima, E. W., M. Bakarcic, and M. V. Fernandez-Valiela. 1975. Association of rickettsialike bacteria with plum leaf scald disease. Phytopathology 65:476–479.

Kostka, S. J., T. A. Tattar, J. L. Sherald, and S. S. Hurtt. 1986. Mulberry leaf scorch, new disease caused by a fastidious, xylem-inhabiting bacterium. Plant Disease 70:690–693.

Krell, R. K., E. A. Boyd, J. E. Nay, Y.-L. Park, and T. M. Perring. 2007. Mechanical and insect transmission of Xylella fastidiosa to Vitis vinifera. American Journal of Enology and Viticulture 58:211–216.

Kung, S. H., and R. P. P. Almeida. 2011. Natural competence and recombination in the plant pathogen Xylella fastidiosa. Applied and Environmental Microbiology 77:5278–5284.

Lopes, J. R. S., M. P. Daugherty, and R. P. P. Almeida. 2009. Context-dependent transmission of a generalist plant pathogen: Host species and pathogen strain mediate insect vector competence. Entomologia Experimentalis et Applicata 131:216–224.

Lopes, S. A., S. Marcussi, S. C. Z. Torres, V. Souza, C. Fagan, S. C. França, N. G. Fernandes, and J. R. S. Lopes. 2003. Weeds as alternative hosts of the citrus, coffee, and plum strains of Xylella fastidiosa in Brazil. Plant Disease 87:544–549.

Maiden, M. C. J., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, et al., 1998. Multilocus sequence typing: A portable approach to the identification of clones within populations of pathogenic microorganisms. Proceedings of the National Academy of Sciences of the United States of America 95:3140–3145.

Marucci, R. C., J. R. S. Lopes, J. D. Vendramim, and J. E. Corrente. 2005. Influence of Xylella fastidiosa infection of citrus on host selection by leafhopper vectors. Entomologia Experimentalis et Applicata 117:95–103.

Marucci, R. C., J. R. S. Lopes, and R. R. Cavichioli. 2008. Transmission efficiency of Xylella fastidiosa by sharpshooters (Hemiptera: Cicadellidae) in coffee and citrus. Journal of Economic Entomology 101:1114–1121.

Miranda, M. P., E. S. Villada, S. A. Lopes, A. Fereres, and J. R. S. Lopes. 2013. Influence of citrus plants infected with Xylella fastidiosa on stylet penetration activities of Bucephalogonia xanthophis (Hemiptera: Cicadellidae). Annals of the Entomological Society of America 106:610–618.

Newman, K. L., R. P. P. Almeida, A. H. Purcell, and S. E. Lindow. 2003. Use of a green fluorescent strain for analysis of Xylella fastidiosa colonization of Vitis vinifera. Applied and Environmental Microbiology 69:7319–7327.

Newman, K. L., R. P. P. Almeida, A. H. Purcell, and S. E. Lindow. 2004. Cell-cell signaling controls Xylella fastidiosa interactions with both insects and plants. Proceedings of the National Academy of Sciences of the United States of America 101:1737–1742.

Nunes, L. R. 2003. Microarray analyses of Xylella fastidiosa provide evidence of coordinated transcription control of laterally transferred elements. Genome Research 13:570–578.

Nunney, L., X. Yuan, R. Bromley, J. Hartung, M. Montero-Astúa, L. Moreira, B. Ortiz, and R. Stouthamer. 2010. Population genomic analysis of a bacterial plant pathogen: Novel insight into the origin of Pierce’s disease of grapevine in the U.S. PLoS ONE 5:e15488.

Nunney, L., X. Yuan, R. E. Bromley, and R. Stouthamer. 2012. Detecting genetic introgression: High levels of intersubspecific recombination found in Xylella fastidiosa in Brazil. Applied and Environmental Microbiology 78:4702–4714.

Nunney, L., D. B. Vickerman, R. E. Bromley, S. A. Russell, J. R. Hartman, L. D. Morano, and R. Stouthamer. 2013. Recent evolutionary radiation and host plant specialization in the Xylella fastidiosa subspecies native to the United States. Applied and Environmental Microbiology 79:2189–2200.

Nunney, L., B. Ortiz, S. A. Russell, R. R. Sánchez, and R. Stouthamer. 2014a. The complex biogeography of the plant pathogen Xylella fastidiosa: Genetic evidence of introductions and subspecific introgression in Central America. PLoS ONE 9:e112463.

Nunney, L., D. L. Hopkins, L. D. Morano, S. E. Russell, and R. Stouthamer. 2014b. Intersubspecific recombination in Xylella fastidiosa strains native to the United States: Infection of novel hosts associated with an unsuccessful invasion. Applied and Environmental Microbiology 80:1159–1169.

Nunney, L., E. L. Schuenzel, M. Scally, R. E. Bromley, and R. Stouthamer. 2014c. Large-scale intersubspecific recombination in the plant-pathogenic bacterium Xylella fastidiosa is associated with the host shift to mulberry. Applied and Environmental Microbiology 80:3025–3033.

Paião, F. G., A. M. Meneguim, E. C. Casagrande, and R. P. Leite, Jr. 1996. Envolvimento de cigarras (Homoptera, Cicadidae) na transmissão de Xylella fastidiosa em cafeeiro. Fitopatologia Brasileira 27:S67.

Petit, J. N., M. S. Hoddle, J. Grandgirard, G. K. Roderick, and N. Davies. 2008. Invasion dynamics of the glassy-winged sharpshooter Homalodisca vitripennis (Germar) (Hemiptera: Cicadellidae) in French Polynesia. Biological Invasions 10:955–967.

Pierce, N. B. 1892. The California vine disease. U.S. Department of Agriculture, Division of Vegetable Pathology Bulletin 2.

Polz, M. F., E. J. Alm, and W. P. Hanage. 2013. Horizontal gene transfer and the evolution of bacterial and archaeal population structure. Trends in Genetics 29:170–175.

Purcell, A. H., and A. H. Finlay. 1979. Evidence for noncirculative transmission of Pierce’s disease bacterium by sharpshooter leafhoppers. Phytopathology 69:393–395.

Purcell, A. H., A. H. Finlay, and D. L. McLean. 1979. Pierce’s disease bacterium: Mechanism of transmission by leafhopper vectors. Science 206:839–841.

Purcell, A. H., and S. R. Saunders. 1999. Fate of Pierce’s disease strains of Xylella fastidiosa in common riparian plants in California. Plant Disease 83:825–830.

Purcell, A. H., S. R. Saunders, M. Hendson, M. E. Grebus, and M. J. Henry. 1999. Causal role of Xylella fastidiosa in oleander leaf scorch disease. Phytopathology 89:53–58.

Purcell, A. H. 2013. Paradigms: Examples from the bacterium Xylella fastidiosa. Annual Review of Phytopathology 51:339–356.

Retchless, A. C., F. Labroussaa, L. Shapiro, D. C. Stenger, S. E. Lindow, and R. P. P. Almeida. 2014. Genomic insights into Xylella fastidiosa interactions with plant and insect hosts. In Genomics of Plant-Associated Bacteria, edited by D. C. Gross, A. Lichens-Park, and C. Kole. Berlin, Germany: Springer Berlin Heidelberg. Pp. 177–202.

Sanderlin, R. S., and R. A. Melanson. 2010. Insect transmission of Xylella fastidiosa to pecan. Plant Disease 94:465–470.

Saponari, M., D. Boscia, F. Nigro, and G. P. Martelli. 2013. Identification of DNA sequences related to Xylella fastidiosa in oleander, almond, and olive trees exhibiting leaf scorch symptoms in Apulia (southern Italy). Journal of Plant Pathology 95:668.

Saponari, M., G. Loconsole, D. Cornara, R. K. Yokomi, A. De Stradis, D. Boscia, D. Bosco, G. P. Martelli, R. Krugner, and F. Porcelli. 2014. Infectivity and transmission of Xylella fastidiosa by Philaenus spumarius (Hemiptera: Aphrophoridae) in Apulia, Italy. Journal of Economic Entomology 107(4):1316–1319.

Scally, M., E. L. Schuenzel, R. Stouthamer, and L. Nunney. 2005. Multilocus sequence type system for the plant pathogen Xylella fastidiosa and relative contributions of recombination and point mutation to clonal diversity. Applied and Environmental Microbiology 71:8491–8499.