The translucens group of Xanthomonas translucens: Complicated and important pathogens causing bacterial leaf streak on cereals

Abstract Xanthomonas translucens is a group of gram‐negative bacteria that can cause important diseases in cereal crops and forage grasses. Different pathovars have been defined according to their host ranges, and molecular and biochemical characteristics. Pathovars have been placed into two major groups: translucens and graminis. The translucens group contains the pathovars causing bacterial leaf streak (BLS) on cereal crops such as wheat, barley, triticale, rye, and oat. In recent years, BLS has re‐emerged as a major problem for many wheat‐ and barley‐producing areas worldwide. The biology of the pathogens and the host–pathogen interactions in cereal BLS diseases were poorly understood. However, recent genome sequence data have provided an insight into the bacterial phylogeny and identification and pathogenicity/virulence. Furthermore, identification of sources of resistance to BLS and mapping of the resistance genes have been initiated. Taxonomy Kingdom Bacteria; Phylum Proteobacteria; Class Gammaproteobacteria; Order Xanthomonadales; Family Xanthomonadacea e; Genus Xanthomonas; Species X. translucens; translucens group pathovars: undulosa, translucens, cerealis, hordei, and secalis; graminis group pathovars: arrhenatheri, graminis, poae, phlei; newly established pathovar: pistaciae. Host range X. translucens mainly infects plant species in the Poaceae with the translucens group on cereal crop species and the graminis group on forage grass species. However, some strains have been isolated from, and are able to infect, ornamental asparagus and pistachio trees. Most pathovars have a narrow host range, while a few can infect a broad range of hosts. Genome The complete genome sequence is available for two X. translucens pv. undulosa strains and one pv. translucens strain. A draft genome sequence is also available for at least one strain from each pathovar. The X. translucens pv. undulosa strain Xt4699 was the first to have its complete genome sequenced, which consists of 4,561,137 bp with total GC content approximately at 68% and 3,528 predicted genes. Virulence mechanisms Like most xanthomonads, X. translucens utilizes a type III secretion system (T3SS) to deliver a suite of T3SS effectors (T3Es) inside plant cells. Transcription activator‐like effectors, a special group of T3Es, have been identified in most of the X. translucens genomes, some of which have been implicated in virulence. Genetic factors determining host range virulence have also been identified.


| INTRODUC TI ON
According to Bamberg (1936), the occurrence of leaf streak-like disease on wheat and barley had been observed and recorded as early as 1893. However, the formal description of the disease and the causal bacterium was not made until 1917, when Jones and coworkers published the identification of bacterial blight on barley (Jones et al., 1917). The causal bacterium was named Bacterium translucens because of the translucent lesions on diseased leaves. Shortly after this the disease on wheat was described, but was named as black chaff, which referred to the disease symptoms on the spikes (Smith et al., 1919). Since then, leaf streak disease has been reported on other small grains and some grasses (Reddy et al., 1924;Hagborg, 1942;Wallin and Reddy, 1945;Fang et al., 1950;Cunfer and Scolari, 1982). On forage grasses, a group of genetically related bacteria cause bacterial wilt (Egli et al., 1975;Egli and Schmidt, 1982).
The classification and nomenclature of bacteria causing bacterial leaf streak (BLS) diseases has been very confusing and has undergone many changes, largely due to their morphological and biochemical similarity and overlapping host range. For a long time, these bacteria were classified as different pathovars (pv.) under the species of Xanthomonas campestris (Dye and Lelliott, 1974). Vauterin et al. (1992Vauterin et al. ( , 1995 proposed the re-establishment of the species Xanthomonas translucens including strains that cause leaf streak on small grains and some grasses (the "translucens" group) and strains that cause bacterial wilt on forage grasses (the "graminis" group). This classification and nomenclature system has been supported by recent molecular and whole genome sequence data (Peng et al., 2016;Langlois et al., 2017;Hersemann et al., 2017). Genomic sequence data also suggest that the pathovars cerealis could be genetically separated from other translucens and graminis group pathovars.
Wheat and barley BLS diseases have been found in almost all wheat-or barley-growing areas worldwide (Bamberg, 1936;Duveiller, 1990) and they can cause substantial yield losses and poor grain quality (Waldron, 1929;Shane et al., 1987;Duveiller and Maraite, 1993). Although BLS epidemics have been sporadic and usually occur in warm and humid subtropics regions, in the last decade the incidence of BLS has dramatically increased in the Midwestern United States, where the majority of the hard red spring wheat and durum wheat are produced .
Most of the cultivars in this region, as well as other places, are highly susceptible and no chemical methods are available for BLS control in the field (McMullen and . Furthermore, breeding for resistant wheat and barley cultivars is difficult due to the lack of sources of resistance and our knowledge regarding host-pathogen interactions. Although research has been done on some aspects of the disease system, in particular the recent genome sequencing projects, there has been no review for this group of bacterial pathogens as well as the diseases they cause except a handbook published in 1997 by the International Maize and Wheat Improvement Center (CIMMYT) (Duveiller et al., 1997). In this work, we provide a comprehensive review mainly of the "translucens" group of X. translucens pathogens that cause BLS on small grains. This review summarizes the current knowledge regarding the diseases, virulence, and genomics of the bacterial pathogens, and the genetics of host resistance. We also provide some thoughts on the future direction of the research aiming to solve this disease problem.

| DIS E A S E SYMP TOMS , DIS TRIBUTION , AND IMP ORTAN CE
The symptoms of BLS disease are mainly observed on leaves and spikes. On the leaf, initially water-soaked streaks develop that subsequently become translucent necrotic lesions (Figure 1a,b). Under warm and humid conditions, bacterial ooze (yellow exudates) can be seen on the leaf surface ( Figure 1a). Under high disease pressure, the whole leaf area may be severely affected by the pathogen ( Figure 1c). The disease, black chaff, refers to the dark-purple streaks on the glumes (Figure 1d). The diagnosis of BLS is sometime difficult in the field mainly because the symptoms resemble those caused by fungal pathogens or genetic or environmental factors (Duveiller et al., 1997;McMullen and Adhikari, 2011). For example, wheat cultivars that possess stem rust resistance gene Sr2 develop a melanic reaction that mimics the black chaff symptoms caused by the X. translucens pathogens . BLS disease has been reported from many geographical regions worldwide where wheat is grown. According to Duveiller et al. (1997) ). It appears that BLS disease has not been reported from western Europe, which is probably due to unfavourable environmental conditions and extensive quarantine efforts (Paul and Smith, 1989; most of the X. translucens genomes, some of which have been implicated in virulence.
Genetic factors determining host range virulence have also been identified.

K E Y W O R D S
bacterial diseases, host resistance, pathogen virulence, wheat, Xanthomonas Duveiller et al., 1997). In the United States, BLS was initially reported in barley and wheat fields in the Midwest (Jones et al., 1917;Smith et al., 1919). Since then, BLS has been found in many places in the United States with outbreaks and epidemics mostly having occurred in the southeastern regions (Milus and Mirlohi, 1994;Tubajika et al., 1999). In recent years, the incidence of BLS has increased in the Upper Midwest region of the United States, including North Dakota, Minnesota, and South Dakota (Adhikari et al., 2012a;Kandel et al., 2012;Curland et al., 2018).
Although BLS is considered as a potential threat for wheat production worldwide, there are no recent reports on yield losses caused by BLS. Furthermore, the importance of BLS varies in different wheat-growing regions and depends mainly on the level of resistance/ susceptibility of wheat cultivars grown and the prevalent environmental conditions. Yield losses due to BLS are generally reported to be 10% or less, but severe infection can cause up to 40% yield losses on highly susceptible cultivars (Waldron, 1929;Forster and Schaad, 1988). Studies have indicated that yield loss due to BLS is generally negatively correlated with BLS severity on flag leaves, and up to 20% yield reduction is possible if 50% leaf area of the flag leaves is infected (Shane et al., 1987;Duveiller and Maraite, 1993). Yield losses are usually due to the reduction in grain weight and the number of kernels per spikes; however, severe infection can cause sterile spikes leading to a complete yield loss (Forster and Schaad, 1988;Tubajika et al., 1998).
In addition, BLS infection can alter the protein content of the grains, resulting in quality reduction (Shane et al., 1987).

| D IS E A S E C YCLE , EPIDEMI OLOGY, AND MANAG EMENT
Many aspects of the BLS aetiology have not been experimentally tested, but a general disease cycle has been proposed (Duveiller et al., 1997, Figure 2). Seed is thought to be an important source of primary inoculum for BLS (Boosalis, 1952;Tsilosani et al., 1977;Timmer et al., 1987;Forster and Schaad, 1988;Milus and Mirlohi, 1995;Rashid et al., 2013). The survival rate of bacteria on seeds and the possibility of transmission to seedlings were shown to be largely dependent on the storage conditions, the length of storage, and the level of susceptibility of genotypes (Boosalis, 1952;Forster and Schaad, 1990;Milus and Mirlohi, 1995). It has been reported that if the number of bacteria is less than 1,000 cfu per gram in the seed lots, no BLS symptoms could be visible in the emerged plants (Klykov, 1945;Duveiller et al., 1997). Studies have shown that weeds and grasses may serve as overwintering hosts or green bridges for the bacterium to spread from one season to other (Wallin, 1946;Fang et al., 1950;Boosalis, 1952;Thompson et al., 1989). The bacteria are also able to survive in soil and crop debris for a short period of time (Milus and Mirlohi, 1995;Duveiller et al., 1997;Stromberg et al., 2000).
Although X. translucens pathogens enter plant tissues through natural openings, some of them have been reported to possess ice nucleation ability that can cause frost damage leading to the formation of wounding sites for direct entry (Kim et al., 1987;Azad and Schaad, 1988;Gurion-Sherman and Lindow, 1993). Warm and humid conditions are thought to be important for BLS development because the disease has been found more in wet seasons or in sprinkler-irrigated fields with warm temperature; however, the exact conditions conducive to BLS development are not well known. BLS epidemics have been reported to be sporadic and vary from year to year (Bamberg, 1936;Duveiller et al., 1991;Tubajika et al., 1998). Artificial introduction of disease in field plots is challenging which makes it difficult to determine the relationship between environmental conditions and BLS epidemics. Duveiller and Maraite (1995) showed that temperature is more important than other factors to initiate epidemics. Other factors implicated in BLS epidemics include damage from wind, hail and frost, dew period, and host genotypes (Duveiller et al., 1997). Several cultural practices may help reduce BLS incidence; however, limited information is available on their efficacy (Duveiller et al., 1997;Adhikari et al., 2011). Crop rotation is not considered as a major control strategy because the bacteria cannot survive in debris for a long time (Milus and Mirlohi, 1995;Duveiller et al., 1997). Seed is considered an important source of primary inoculum, thus use of clean seed might be a way of reducing BLS incidence. Different methods have been developed to detect the bacterium in seeds, including dilution plating with the use of selective media, seedling infection assays, serodiagnostic assays, PCR amplification, and loop-mediated isothermal amplification (Forster and Schaad, 1985;Bragard and Verhoyen, 1993;Maes et al., 1996;Langlois et al., 2017). Seed disinfection methods have also been developed to eliminate the bacterium by using chemical or physical methods (Atanasoff and Johnson, 1920;. However, the effectiveness of seed treatment is still questionable because of contradictory results obtained from different studies (Braun, 1920;Duveiller et al., 1997). Using clean seeds or applying seed treatments may reduce disease incidence, but does not stop the spread of BLS inoculum between fields (Duveiller et al., 1997). Very limited studies have been conducted to test chemicals for controlling cereal BLS in the field. Silva et al. (2010)

| CL A SS IFIC ATI ON , NOMEN CL ATURE , AND IDENTIFI C ATI ON OF X . TR ANSLUCENS
The early classification and nomenclature for this group of bacteria were mainly based on pathogenicity tests and host range.
The early taxonomy for X. translucens had been very confusing because strains varied greatly in host range and levels of host specificity. Jones et al. (1917) first reported BLS disease on barley and named the pathogen B. translucens. Later, BLS was reported on wheat by Smith et al. (1919) where the pathogen was named B. translucens var. undulosum because it morphologically resembled the barley pathogen and was able to infect barley through artificial inoculation. Dowson (1939) created the genus Xanthomonas to include the species X. translucens that caused cereal BLS diseases.
Based on the natural host as well as the ability to infect hosts using artificial inoculation, Hagborg (1942) Duveiller et al. (1997) (barley and oat), and f. sp. cerealis (wheat, barley, rye, and oat). Fang et al. (1950) argued that f. sp. cerealis and hordei-avenae should be combined with f. sp. undulosa and hordei, respectively, and given the name of f. sp. cerealis for strains that naturally occur on smooth bromegrass and quack grass but can infect wheat, barley, rye, and oat using artificial inoculations. Classification by Fang et al. (1950) also included f. sp. phleipratensis, which was originally identified by Wallin and Reddy (1945) from timothy grass.
During the 1970s and 1980s, bacterial wilt was reported on many forage grasses and the causal bacteria were shown to be closely related to X. campestris identified from cereals (Egli et al., 1975;Wilkins and Exley, 1977;Roberts et al., 1981;Egli and Schmidt, 1982;Channon and Hisset, 1984). Those bacteria were named as different pathovars under X. campestris, including pv.
Beside cereals and forage grasses, X. translucens was also found to infect ornamental asparagus and pistachio trees, which are in the genetically distant plant families Liliaceae and Anacardiaceae, respectively (Rademaker et al., 2006;Marefat et al., 2006aMarefat et al., , 2006b.
Very surprisingly, the bacterial pathogens on asparagus were identified as pv. undulosa and were able to infect wheat by crossinoculation and vice versa (Rademaker et al., 2006). The bacterium on pistachio trees was designated a new pathovar, X. translucens pv. Further studies have been conducted to examine the genetic similarity and diversity of the pathovars in translucens group by using DNA markers, biochemical profiling, and pathogenicity tests. The studies conducted by Bragard et al. (1995Bragard et al. ( , 1997 suggested that X. translucens pv. translucens is a synonym of X. translucens pv. hordei, and the translucens group contains three true biological entities, cerealis, translucens and undulosa, with pv. translucens pathogenic on barley, pv. undulosa pathogenic to both barley and wheat, and pv. cerealis pathogenic to barley, wheat, oat, and bromegrass. Not many X. translucens pv. secalis strains were available for analyses, but they were shown to be clustered with pv. undulosa Giblot-Ducray et al., 2009;Curland et al., 2018). Using a few strains isolated from wheat and barley, we showed that wheat strains (undulosa) were capable of causing lesions and water-soaking symptoms on wheat, barley, and triticale (×Triticosecale), while barley strains (translucens) were only able to cause lesions and water-soaking symptom on barley ; Figure 3). However, Curland et al. (2018) reported that most X. translucens pv. undulosa and pv. translucens strains collected from F I G U R E 3 Reaction of wheat or barley to Xanthomonas translucens pv. undulosa (Xtu) and X. translucens pv. translucens (Xtt) strains. The strains were collected from North Dakota, USA the Upper Midwestern United States cause disease symptoms on both wheat and barley by infiltration, even though their virulence was generally correlated with their original host.
In recent years, X. translucens taxonomy has been performed using multilocus sequence analysis with data from the housekeeping gene and/or genome sequences (Wichmann et al., 2013;Gardiner et al., 2014;Peng et al., 2016;Langosis et al. 2017;Curland et al., 2018). The analyses with the DNA sequence data basically agreed with the previous classifications for X. translucens at both species and pathovar levels determined by DNA-DNA hybridization and DNA marker analysis and pathogenicity tests . Evidence has also suggested that cerealis strains could be genetically separated from other translucens and graminis pathovars (Langlois et al., 2017). However, more cerealis strains need to be analysed in order to draw a solid conclusion.

| VIRULEN CE AND PATHOG ENI CIT Y OF X . TR ANSLUCENS
Similar to other xanthomonads, X. translucens bacteria are gramnegative, rod-shaped, non-spore forming, 0.5-0.8 × 1.0-2.5 µm in size, contain a single polar flagellum, and form pale yellow colonies on nutrient agar medium (Figure 4a,b). Using a transmission electron microscope and the diseased wheat leaf samples collected at the fifth day after spray inoculation with a X. translucens pv. undulosa strain, we observed that bacterial cells were mainly distributed in the mesophyll tissue ( Figure 4b). Thus, the bacterial pathogen probably enters plant tissue through stomata and mainly colonizes mesophyll tissues. However, using the same spray inoculation method, we observed little or no disease on barley leaves for X. translucens pv. translucens, suggesting that it may have a different tissue specificity (Liu et al. unpublished data). Leaf clipping and the dip inoculation method have been successfully used in the study by Pesce et al. (2017) to induce disease on barley. This suggests that X. translucens pv. translucens probably resides in the vascular tissue. Tissue specificity differences have been reported for other Xanthomonas species (Bogdanove et al., 2011). For example, X. oryzae pv. oryzae is a vascular bacterial pathogen causing rice leaf blight, while X. oryzae pv. oryzicola, the cause of bacterial leaf streak, is a nonvascular bacterial pathogen (Bogdanove et al., 2011).
Research in the last few decades has substantially advanced our understanding of the molecular mechanisms of pathogenicity and virulence for X. translucens. As mentioned above, X. translucens strains differ in their levels of host specificity, with some capable of infecting multiple host species (broad host range) but others only infecting one host species (narrow host range). The broad host range bacteria may harbour factors controlling virulence rather than narrow-host range bacteria carrying avirulence factors (Mellano and Cooksey, 1988). Mutants with a narrow host range have been obtained from a wide host range in X. translucens by using Tn5 insertion mutagenesis, and host-specific virulence (Hsv) genes have been identified that can restore the ability of the mutants to infect specific hosts (Mellano and Cooksey, 1988;Waney et al., 1991). However, the sequences for these Hsv genes have not been determined.
Virulence differences were also observed among strains within a pathovar (Cunfer and Scolari, 1982;Milus and Chalkly, 1994;Adhikari et al., 2011;Sapkota et al., 2018). Using pathogenicity tests on several wheat genotypes, Adhikari et al. (2011) found that X. translucens pv. undulosa strains collected from North Dakota are relatively diverse in virulence and there is a significant interaction of wheat-X. translucens pv. undulosa. Previous studies coupled with our recent studies showed that several triticale lines harbour dominant resistance genes to some X. translucens pv. undulosa strains, suggesting the presence of a gene-for-gene interaction in triticale (Johnson et al., 1987;Wen et al., 2018). However, it remains unknown if the wheat-X. translucens pathosystem involves a gene-forgene interaction.  Bogdanove et al., 2011;Pesce et al., 2017). X. translucens phylogenetically belongs to clade 1, which has four other species: X. albilineans, X. hyacinthi, X. theicola, and X. sacchari (Young et al., 2008;Gardiner et al., 2014). Among the clade 1 species, X. albilineans and X. sacchari have no hrp cluster, but all the sequenced X. translucens strains have it (Pesce et al., 2017). The genetic structure and organization of the hrp locus in X. translucens genomes is highly conserved among sequenced strains and it is similar to that in clade 1 Xanthomonas species and β-protebacteria (Wichmann et al., 2013;Pesce et al., 2017). Functional analyses have shown that X. translucens T3SS is essential for pathogenicity, hypersensitive response induction and effector delivery (Waney et al., 1991;Gardiner et al., 2014;Peng et al, 2016;Pesce et al., 2017). However, the hrp system in X. graminis strain Xtg29 has been shown not to be required for pathogenicity but to modulate virulence (Wichmann et al., 2013). This indicates that strains of graminis and translucens may have different biology and virulence mechanisms (Hersemann et al., 2017).
Like other Xanthamonas spp., genome sequence data indicate that X. translucens strains harbour a suite of T3Es (Whichmann et al., 2013;Gardiner et al., 2014;Pesce et al., 2015;Peng et al., 2016;Charkhabi et al., 2017;Hersemann et al., 2017). Sequence alignment and comparison showed that a core set of T3Es identified from other Xanthomonas spp. was also present in the sequenced X. translucens genomes. However, variations in the presence/absence and copy number of individual T3Es were detected among different pathovars and among different strains within a pathovar, and also frame-shift and loss-of-function mutations of certain T3Es occur among different strains (Wichmann et al., 2013;Peng et al., 2016;Charkhabi et al., 2017). The unique repertoire of T3Es in different strains and pathovars may reflect the adaptation of strains to various hosts or virulence to specific genotypes of a host (Jacques et al., 2016). However, the function of the majority of T3Es in X. translucens in virulence or colonization have not been tested and determined.
Transcription activator-like effectors (TALEs) are a special type of T3Es that are only present in Xanthomonas and some strains of Ralstonia solanacearum (Bogdanove et al., 2010). TALE genes have been identified from three X. translucens strains with a complete genome sequence as well as a few strains with a draft genome sequence (Pesce et al., 2015;Jaenicke et al., 2016;Peng et al., 2016;Charkhabi et al., 2017). X. translucens TALEs that have been well characterized were from two X. translucens pv. undulosa strains, Xt4699 and ICMP11055, which have eight and seven genes, respectively. Based on the sequences of repeat variable di-residue (RVD) amino acids, it was found that four TALEs are common between the two strains and the rest are either partially conserved or have no obvious similarity (Charkhabi et al., 2017). Sequence comparison revealed that the two strains had several unique RVDs, including KG, QD, Y*, YK and YD, which have not been identified from other Xanthomas spp. (Peng et al., 2016;Charkhabi et al., 2017). It was also found that some TALEs that are widely distributed in other xanthomonads are less frequently present in X. translucens (Charkhabi et al., 2017).
TALEs function to induce host gene expression after binding to the promoter region of a gene. Using microarray analysis and TALE and Ta.14164.1S1_x_at (bHLH family transcription factor). Very recently, it was found that Tal8 is associated with enhanced virulence in XT4699 by modulating ABA biosynthesis and production (Peng et al., 2019). Two TALE genes in ICMP11055 were also shown by site-directed mutagensis to contribute to bacterial virulence based on lesion length (Charkhabi et al., 2017). More work is needed to investigate the roles of TALEs in virulence and their underlying molecular mechanisms.

| G ENOMI C S OF X . TR ANSLUCENS
As mentioned above, genome sequencing and comparative genomics is a powerful tool to reveal candidate genes' underlying virulence/pathogenicity as well as bacterial classification of the bacterial pathogens, and whole-genome sequencing has been performed for X. translucens. The first genome sequence of X. translucens was obtained from X. translucens pv. graminis strain Xtg29 (Wichmann et al., 2013).  (Jaenicke et al., 2016;Peng et al., 2016;Charkhabi et al., 2017). The main features for the three complete genomes are listed in Table 1. The remaining genome sequences were fragmented containing variable numbers of scaffolds or contigs (see the above web link) and the genome size of these strains ranged from 4.1 to 4.8 Mb.
Size difference was observed among different pathovars and also among different strains within a pathovar (Charkhabi et al., 2017;Hersemann et al., 2017). The annotated protein-encoding genes in published X. translucens genomes ranged from 3,160 to 4,413, with pv. graminis relatively having fewer. The available genome sequence data and comparative genomics will greatly facilitate our progress in the understanding host, cultivar, and tissue-specificity in X. translucens species.

| G ENE TI C S AND MAPPING OF HOS T RE S IS TAN CE
Because no chemical control is available, host resistance appears to be the only way to control wheat BLS. Several sources of BLS resistance have been identified in diverse wheat or barley germplasm and related species (Hagborg, 1974;Akhtar and Aslam, 1986;Alizadeh et al., 1994;El Attari et al., 1996;Milus et al., 1996;Adhikari et al., 2011;Kandel et al., 2012;Sapkota et al., 2018). These studies revealed a low percentage of resistant lines and no immune or highly resistant materials were found in the wheat germplasm. Nevertheless, several triticale lines were reported to possess high levels of BLS resistance (Cunfer and Scolari, 1982;Johnson et al., 1987;Sapkota et al., 2018).
Using partial and high levels of BLS resistance in small grain crops, the heritability and genetics of BLS resistance were investigated El Attari et al., 1996;Adhikari et al., 2012b;Kandel et al., 2015;Wen et al., 2018). The heritability of BLS resistance in wheat was reported to vary from low to high depending on the resistant lines used (El Attari et al., 1996;. Classic genetic analysis showed that BLS resistance could be quantitative or qualitative.     also detected the presence of multiple genes controlling BLS resistance in these wheat cultivars. However, resistance to BLS in a few triticale lines appears to be qualitative and dominant (Johnson et al., 1987;Wen et al., 2018).

| FUTURE RE S E ARCH D IREC TI ON S
BLS is a common disease of wheat and barley that has the ability to cause a significant reduction in yield and quality. In the last decade, BLS has become an increasingly important problem in all major wheat-and barley-growing areas around the world. Because there is a lack of effective management tools and the understanding of the disease system is limited, the impact of BLS disease on the world's wheat and barley production will continue to rise. It is imperative to have a global research initiative aiming to solve or mitigate the BLS disease problem. We believe that future research should focus on, but not be limited to, the following areas.
First, an immediate need would be to identify chemicals that can be used at the field scale for disease control while waiting for resistant cultivars to be developed. There are a number of copper-based commercial bactericides that have been recommended for other foliar diseases caused by xanthomonads. These products can be evaluated directly in the field for their efficacy and profitability. At the same time, new types of chemicals and novel delivery systems should be sought for the development of more efficient and profitable management methods.
Second, it is common knowledge that disease control using genetic resistance is the most desirable method as it is sustainable and environmentally friendly. Therefore, breeding programmes should place more emphasis on using current resistant materials and continue to search for novel sources of resistance. Research is needed to investigate the genetics of resistance in the identified wheat cultivars and breeding lines, and to identify DNA markers linked to the resistance genes or QTLs, which help quickly incorporate partial resistance into local cultivars. Although it could take a long time to move the resistance genes from triticale to wheat cultivars, it is worth doing because these genes confer high levels of resistance to BLS.
Third, a standardized, quick protocol is needed to evaluate pathogen virulence and host resistance. Previous studies have used different inoculation methods, for example direct spraying, injection, and infiltration, with a corresponding rating scale that makes the results incomparable. These inoculations and scoring methods should be further investigated in order to establish a good evaluation protocol in both the greenhouse and the field. This is in particularly important for breeding programmes. above has been done for the winter wheat regions, but the results need to be confirmed and compared for spring wheat regions.
Lastly, a better understanding of host-pathogen interactions in X. translucens-cereals is needed for the development of wheat or barley cultivars with durable resistance. Genome sequences and comparative genomics can serve as a powerful tool to identify bacterial genes underlying bacterial virulence/avirulence and host range determinants. The host susceptibility genes that either directly or indirectly interact with these bacterial virulence genes can be subsequently identified, and then can be mutated through gene-editing technologies (Li et al., 2012).

ACK N OWLED G EM ENTS
We would like to thank Dr James Buck for the critical review of the manuscript. This material is based on work supported, in part, by the National Institute of Food and Agriculture, United States Department of Agriculture (USDA), under Hatch project number ND02234.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.