Pseudomonas avellanae : From Specific Pathogens to a Species Complex

ABSTRACT

Taxonomy

Bacteria; Proteobacteria, gamma subdivision; Order Pseudomonadales; Family Pseudomonadaceae; Genus Pseudomonas; Pseudomonas avellanae species complex, which includes different pathogenic lineages previously classified within the P. syringae complex, namely the pathovars avellanae, theae, actinidiae, and morsprunorum race 2.

Microbiological Properties

As a whole, the P. avellanae species complex genome comprises 19.4% conserved genes and 80.6% flexible genome. Mobile genetic elements are present within the complex. Strains associated with bacterial canker of European hazelnut possess two plasmids, whereas multiple plasmids and megaplasmids can occur in the lineages associated with bacterial canker of kiwifruit and cherry. Fourteen type III secretion system effectors are shared by all lineages of the species, and each lineage apparently possesses specific effector proteins. Phytotoxins, such as phaseolotoxin, coronatine, and tabtoxin, can be produced by some lineages of P. avellanae . Siderophores are present in the P. avellanae species complex.

Host Range

Corylus avellana , Camellia sinensis var. sinensis , Actinidia chinensis var. chinensis, A. chinensis var. deliciosa, Prunus avium , P. cerasus , P. domestica . A strain of P. avellanae has also been isolated from river epilithon.

Disease Symptoms

Pseudomonas avellanae can cause severe symptoms, including bud necrosis (kiwifruit), leaf necrosis and wilting (hazelnut, tea, kiwifruit, cherry/plum), shoot and twig blight (hazelnut, tea, kiwifruit, cherry/plum), flower necrosis (kiwifruit), fruit necrosis (cherry/plum), trunk canker (hazelnut, kiwifruit, cherry/plum), and plant death (hazelnut, tea, kiwifruit).

Epidemiology

Some P. avellanae lineages are favoured by air temperatures below 0°C. Penetration mainly occurs through stomata, leaf scars, and wounds induced by frost events. P. avellanae species complex exhibits both epiphytic and endophytic phases. During summer, very high temperatures can reduce the multiplication and dispersal of the pathogen. The global circulation of latently infected propagative plant material largely favours the spread of P. avellanae lineages across different cultivation areas.

Detection

Pathogens belonging to this complex can be detected in both symptomatic and asymptomatic plant tissues (e.g., twigs, buds, and cankers) using various conventional techniques and molecular tools (standard‐, duplex‐, TaqMan‐, real‐time PCR, LAMP, MLSA). Depending on the targeted symptoms, pathogenicity tests may be conducted on leaf, leaf scars or woody tissues.

Disease Control

Disease management is based on spray treatments to the crown using copper compounds applied in spring and autumn, as well as after harvest, during leaf fall, and following hail, frost events, and pruning. Balanced plant nutrition is important to reduce the risk of severe infection. Alternative promising strategies include the application of biocontrol agents, natural products, and bacteriophages.

The Pseudomonas avellanae species complex includes pathogens formerly classified within Pseudomonas syringae, namely the pathovars avellanae, actinidiae, theae, and morsprunorum race 2.

1. Introduction

In 1915, a severe disease of tea plants ( Camellia sinensis var. sinensis ) was reported in Japan, and Bacillus theae was indicated as the causal agent (Hori 1915). In 1955, the disease was named bacterial shoot blight of tea, with Pseudomonas theae proposed as the causal agent (Okabe and Goto 1955). In 1978, following the introduction of the pathovar concept, P. syringae pv. theae was used to refer to this disease (Young et al. 1978). In 1975, a variant of P. morsprunorum affecting sweet cherry was described in England (Freigoun and Crosse 1975). In 1976, extensive twig and branch dieback, accompanied by longitudinal cankers along the trunk, was observed in European hazelnut ( Corylus avellana ) orchards in northern Greece. The disease was attributed to P. syringae sensu lato (Psallidas and Panagopoulos 1979), and in 1993, P. syringae pv. avellanae was formally described as the causal agent of bacterial canker of European hazelnut (Psallidas 1993). In the same period, a very similar disease was observed in several European hazelnut orchards in central Italy (Viterbo province), where P. syringae pv. avellanae was isolated and characterised in 1994 (Scortichini and Tropiano 1994). In 1980, a severe disease of green‐fleshed kiwifruit orchards ( Actinidia chinensis var. deliciosa) was observed in Japan; in 1989, the bacterial canker of kiwifruit was formally described, and P. syringae pv. actinidiae was indicated as the causal agent (Takikawa et al. 1989).

In 1996, P. syringae pv. avellanae was formally elevated to species level as P. avellanae (Janse et al. 1996). Furthermore, based on DNA–DNA relatedness among strains belonging to the P. syringae complex, the seminal paper of Gardan et al. (1999) placed P. avellanae and P. syringae pv. theae within genomospecies 8, with P. avellanae as the reference species. Strains of P. syringae pv. actinidiae were not included in this study. The relevant taxonomic similarity of P. syringae pv. actinidiae strains with genomospecies 8 was pointed out in subsequent studies (Scortichini et al. 2002; Manceau and Brin 2003). Extensive genomic analyses have confirmed the close taxonomic relationships among the causal agents of bacterial shoot blight of tea, bacterial canker of European hazelnut, and bacterial canker of kiwifruit (Scortichini et al. 2013; Marcelletti and Scortichini 2014; Gomila et al. 2017). Interestingly, this species also includes P. syringae pv. morsprunorum race 2 (Scortichini et al. 2013; Marcelletti and Scortichini 2014, 2019; Nowell et al. 2016; Gomila et al. 2017), one of the causal agents of bacterial canker of sweet and sour cherry, and European plum ( Prunus avium , P. cerasus , P. domestica ) (Freigoun and Crosse 1975; Hulin, Jackson, et al. 2020). This pathovar belongs to phylogroup PG01 of the P. syringae complex, according to Berge et al. (2014). According to these studies, four distinct P. syringae pathovars can now be included in a taxonomically delineated species complex, defined as “a single species composed of distinct populations capable of infecting a limited number of host plants” (Stolp et al. 1965; Doudoroff and Palleroni 1974).

The distinct genomic lineages (i.e., populations sensu Stolp et al.) of the P. avellanae species complex cause severe economic losses in different host plants across various geographic areas. Significant examples include more than 1000 ha of hazelnut orchards initially destroyed by bacterial canker of European hazelnut in central Italy (Scortichini 2002). Economic losses exceeded €2 million in Italy in 2009 at the onset of the epidemic and later reached more than €400 million in New Zealand in 2014 due to bacterial canker of kiwifruit (Narouei‐Khandan et al. 2022). Notably, this disease is considered pandemic, as it has spread to all major kiwifruit‐growing regions worldwide (Nunes da Silva et al. 2022). Bacterial canker of cherry also accounts for substantial economic losses, with reports of up to 75% yield loss in Oregon, United States (Spotts et al. 2010). Interestingly, a strain isolated in the United States from river epilithon (i.e., rock‐attached biofilms composed of algae, diatoms, rotifers, bacteria, and nematodes) also belongs to P. avellanae (Morris et al. 2007; Turco et al. 2022).

This review concisely addresses the taxonomy, biology, host range, disease symptoms, epidemiology, detection, and disease control of the P. avellanae species complex. In this review, we use the term “ Pseudomonas avellanae species complex” to describe the genetically related, biologically and genomically differentiated lineages assigned to the genomic species P. avellanae . Such lineages belong to the formally recognised former pathovars: P. syringae pv. avellanae, P. syringae pv. theae, P. syringae pv. actinidiae and P. syringae pv. morsprunorum race 2. This should also facilitate reading and comparison with the published literature.

2. Taxonomy

Pseudomonas avellanae as a distinct species appears quite robust. DNA–DNA relatedness, average nucleotide identity (ANI) coupled with TETRA analysis, genome‐to‐genome distance calculator (GGDC), 16S rRNA gene sequencing, multilocus sequence type analysis (MLSA), as well as whole‐genome sequencing (WGS) and comparative analyses all provide clear evidence that this species can be distinguished from the P. syringae complex (Janse et al. 1996; Gardan et al. 1999; Scortichini et al. 2002; Scortichini et al. 2013; Gomila et al. 2017). The inclusion of three P. syringae pathovars, namely theae, actinidiae and morsprunorum race 2, within the P. avellanae cluster is also taxonomically coherent. Phylogenetic analyses based on the concatenation of more than 2000 orthologous genes, as well as the MLSA, support this inclusion (Nowell et al. 2016; Gomila et al. 2017; Hulin, Mansfield, et al. 2018). In addition, ANI and TETRA analyses indicated similarity values among strains of the P. avellanae cluster, including the three pathovars, ranging from 97.59% to 99.92%, with P. syringae pv. tomato being the most closely related pathogen (Scortichini et al. 2013). A split consensus network based on concatenated nucleotide sequences of seven housekeeping genes also indicated a close relationship between P. syringae pv. tomato strains and P. avellanae (Marcelletti and Scortichini 2014). Similarity between this pathovar and P. avellanae was also revealed by DNA–DNA analysis (Gardan et al. 1999). The same overall results were obtained by Gomila et al. (2017), who, using a phylogenomic approach based on ANI and UPGMA analyses, clearly demonstrated the distinctiveness of the P. avellanae cluster, which also includes the three P. syringae pathovars mentioned above (Figure 1). These results were also obtained from the comparison of concatenated amino acid sequences of 219 monocopy proteins from the core genome of 127 genomes of the P. syringae complex, which clearly showed the separation of the P. avellanae species complex from all the other phytopathogenic pseudomonads (Gomila et al. 2017). These analyses also indicate P. syringae pv. tomato as the most closely related pathogen. Additional analyses further confirmed the robustness of this species complex, showing a minimal intracluster MLSA similarity of 99.0% and a minimal intracluster core genome similarity of 99.2% (Gomila et al. 2017). The strain isolated from the river epilithon also belongs to the P. avellanae species complex and is closely related, but not identical, to strains obtained from hazelnut based on ANI analysis (Turco et al. 2022). The strains of the P. avellanae species complex can also be phenotypically circumscribed using MALDI‐TOF mass spectrometry (Mulet et al. 2012).

FIGURE 1.

Dendrogram based on ANI values and UPGMA analysis of 139 genomes of pseudomonads belonging to the Pseudomonas syringae species complex. The Pseudomonas avellanae species complex includes the P. syringae pathovars theae, actinidiae, and morsprunorum race 2, with 'Pseudomonas tomato' as the most closely related species (from Gomila et al. 2017).

3. Microbiological Properties

Pseudomonas avellanae appears to be widely distributed worldwide, as it has been isolated from different sources, including natural ones (i.e., river epilithon) and wild plants of the same species as cultivated hosts, in Asia, Europe and North America. As a whole, the P. avellanae species complex genome comprises 19.4% conserved genes and 80.6% flexible genome. Notably, the proportion of the flexible genome in this species is among the highest within the P. syringae complex, together with P. amygdali (i.e., 82.3%) (Gomila et al. 2017). Within this species complex, the different lineages (i.e., the four former P. syringae pathovars avellanae, theae, actinidiae and morsprunorum race 2) exhibit distinct microbiological characteristics. An interesting feature distinguishing the different P. avellanae lineages is the occurrence of the WHOP (woody host Pseudomonas syringae ) genomic region, which is involved in virulence and is present in lineages causing bacterial canker of kiwifruit and cherry, but absent in the lineage causing bacterial canker of European hazelnut (Turco et al. 2022).

Strains isolated from symptomatic European hazelnut in Greece and Italy can be differentiated according to their geographical origin using molecular typing methods such as repetitive sequence PCR and MLSA (Scortichini et al. 1998; Popović Milovanović et al. 2026). Within the former P. syringae pv. actinidiae, five different biovars are recognised (1, 2, 3, 5 and 6), which are differentiated based on virulence, single‐nucleotide polymorphisms, the occurrence of phytotoxins, or specific type III secretion system (T3SS) effector proteins (Scortichini et al. 2023). The former P. syringae pv. morsprunorum race 2 is traditionally differentiated from race 1 and from P. syringae pv. syringae based on specific biochemical tests such as gelatin and aesculin hydrolysis, tyrosinase activity, utilisation of tartaric and lactic acid, as well as specific pathogenicity assays (i.e., lesions on Syringa vulgaris leaves, colonisation of sweet cherry leaf scars, necrosis on sour cherry fruits, necrotic lesions on sour cherry and P. domestica leaves, and lesions on lemon fruits) and the production of syringomycin (Bultreys and Kaluzna 2010).

A basic effector repertoire of the T3SS of the P. avellanae species complex is shown in Figure 2. Fourteen effectors are shared by all lineages of the species, and each lineage apparently possesses a specific effector protein. Further analyses including additional strains identified specific T3SS effectors in lineages infecting European hazelnut (i.e., HopAA1, Hop AH1, HopAI1, HopB2 and HopM) (Turco et al. 2022). The T3SS effector repertoire has been studied in detail in the lineage corresponding to P. syringae pv. actinidiae. A list of T3SS effectors present in virulent and less virulent strains associated with bacterial canker of kiwifruit is shown in Table 1 (Choi et al. 2017). This lineage is capable of modulating and coordinating gene expression during the early phases of host colonisation. Nutrient acquisition (mainly phosphate) in the first 3 h following contact with the host plant is mediated by the type VI secretion system (T4SS). Subsequently, within 24 h, T3SS effectors are involved in subverting plant defence mechanisms and initiating infection. This is followed by alginate production and increased nutrient acquisition (McAtee et al. 2018). Upon colonisation of the apoplast, the host plant rapidly activates the production of peroxidase, β‐galactosidase, and heat shock proteins, followed by chitinases and pathogenesis‐related proteins (Petriccione et al. 2014). A substantial number of T3SS effectors (i.e., 24–34) is also present in the lineage morsprunorum race 2, located either on the chromosome or on plasmids (Hulin, Armitage, et al. 2018).

FIGURE 2.

Venn diagram of the basic type III secretion system (T3SS) effector gene complements of the Pseudomonas avellanae species complex. The diagram is based on a comparison with the corresponding complements of other sequenced plant‐pathogenic pseudomonads and shows complete identity in reciprocal comparisons. Pave: Pseudomonas avellanae ; Pth: Pseudomonas syringae pv. theae; Pan: P. syringae pv. actinidiae; Pmp: P. syringae pv. morsprunorum race 2 (adapted from Scortichini et al. 2013).

TABLE 1.

Repertoire of type III secretion system (T3SS) effectors in virulent (V13) and less virulent (LV5) strains of Pseudomonas syringae pv. actinidiae (adapted from Choi et al. 2017).

Effectors	Psa‐V13	Psa‐LV5	Amino acid identity (%)	Predicted protein size (kDa)
AvrB4‐1	+		NA	36.0
AvrD1	+			34.6
AvrE1	+	+	97.1	195.3
AvrPto5	+			17.2
AvrRpm1	+			24.9
HopA1	+ a	+		22 (V13)/42.2 (LV5)
HopD1	+			34.6
HopE1		+		24.1
HopF1		+		21.9
HopF4b (HopF2)	+			22.0
HopH1	+			24.3
HopI1	+			49.3
HopM1	+			75.9
HopN1	+	+	99.6	38.7
HopO1		+		32.6
HopQ1	+			48.8
HopR1	+	+	99.0	210.3
HopS1		+		19.9
HopS2	+	+	94.9	18.7
HopT1		+		41.5
HopW1	+ a	+		40.8 (V13)/82.8 (LV5)
HopX1		+		40.6
HopX2		+		38.0
HopX3	+			40.9
HopY1	+	+		30.9
HopZ3	+			45.1
HopAA1	+ a	+		21.8 (V13)/50.4 (LV5)
HopAB3		+		62.6
HopAE1	+	+	99.8	126.6
HopAF1	+	+		23.4
HopAF1‐2		+		30.9
HopAG1		+		77.6
HopAH1	+	+	91.2	42.8
HopAI1 b	+	+	81.6	29.7
HopAM1‐1	+			31.3
HopAO2	+			38.1
HopAR1		+		28.5
HopAS1	+	+	98.9	149.0
HopAU1	+			88.2
HopAV1	+			123.5
HopAW1	+			24.9
HopAY1 c	+	+	91.9	27.1 (V13)/35.4 (LV5)
HopAZ1	+	+	99.5	24.7
HopBB1‐1 (2)	+			30.7
HopBN1	+			56.8

Note: “+” means present; “NA” means not analysed.

^a Effectors highly truncated in Psa V13 indicating a pseudogene.

^b Amino acid identity between alleles from Psa V13 and LV5 lower than 90%.

^c Psa V13 version has shorter amino acid sequence.

Mobile genetic elements are present in the P. avellanae species complex. Strains associated with bacterial canker of European hazelnut possess two plasmids (Turco et al. 2022), whereas multiple plasmids and megaplasmids can occur in lineages associated with bacterial canker of kiwifruit and cherry (McCann et al. 2017; Hulin, Armitage, et al. 2018). Mobile genetic elements and horizontal gene transfer play an important role in determining pathogenicity in the P. avellanae species complex.

An exchange of T3SS effectors located on plasmids has been reported between different pseudomonads affecting cherry trees, likely contributing to convergent pathogenicity towards the same host plant among taxonomically divergent Pseudomonas spp., such as P. avellanae and P. amygdali , the latter including the former named as P. syringae pv. morsprunorum race 1 (Hulin, Armitage, et al. 2018). Exchange of variable genomic regions also occurs between the closely related species P. avellanae and P. syringae pv. tomato (Figure 3). Different integrative conjugative elements (ICEs) of genomic islands, identified in strains of the P. syringae species complex, largely contribute to differentiating P. syringae pv. actinidiae populations from different geographical areas, all belonging to biovar 3 (Butler et al. 2013). Some phytotoxins are also present within the species complex, but their occurrence varies among lineages. The presence or absence of phytotoxins is a key factor in discriminating biovars of strains associated with bacterial canker of kiwifruit. In particular, phaseolotoxin and coronatine may be present or absent, being encoded in the chromosome or in a megaplasmid, respectively (Fujikawa and Sawada 2019). Tabtoxin occurs in strains associated with bacterial canker of hazelnut, cherry and kiwifruit, whereas coronatine, phaseolotoxin, syringomycin and syringopeptin are absent in these lineages (Marcelletti and Scortichini 2015; Turco et al. 2022). Siderophores are present in the P. avellanae species complex. Bacterioferritin, ferritin, pyoverdine, pyridine and yersiniabactin occur in strains associated with bacterial canker of hazelnut (Marcelletti and Scortichini 2015), whereas pyoverdine, enterobactin and yersiniabactin occur in strains associated with bacterial canker of kiwifruit (Marcelletti et al. 2011). The ice nucleation gene cluster is only partially present in the P. avellanae species complex (Turco et al. 2022).

FIGURE 3.

Partial genome alignment between strains associated with bacterial canker of kiwifruit (i.e., Pseudomonas syringae pv. actinidiae) (I‐Psa: Isolated in Italy in 1992; I2‐Psa: Isolated in Italy in 2008; J‐Psa: Isolated in Japan in 1984) and P. syringae pv. tomato DC3000. The violet segments (right) indicate variable region 3, present in all three P. syringae pv. actinidiae strains but absent in P. syringae pv. tomato DC3000, whereas the dark blue segments (left) correspond to variable region 2, present in J‐Psa and I‐Psa but absent in I2‐Psa and P. syringae pv. tomato DC3000. The figure also shows several shorter regions (i.e., light green segments), which likely represent examples of horizontal gene transfer as well as a region in common between P. syringae pv. tomato DC3000 and I2‐Psa (blue green segments) (adapted from Marcelletti et al. 2011).

4. Host Range and Disease Symptoms

The host range of the P. avellanae species complex is shown in Table 2. The different lineages induce various disease symptoms in woody and shrub plant species such as Corylus avellana , Camellia sinensis var. sinensis , Actinidia chinensis var. deliciosa, A. chinensis var. chinensis, Prunus avium , P. cerasus , and P. domestica .

TABLE 2.

Host range, symptoms induced by different lineages of the Pseudomonas avellanae species complex, and primers used for their detection. Lineage names refer to the former P. syringae pathovar names.

Species	Host	Symptom	PCR	Molecular identification
				Primers	Sequence	References
P. avellanae	Corylus avellana	Leaf necrosis and wilting, shoot and twig blight, trunk canker, plant death	Standard	PAV 1 PAV 22	5′‐GGCGACGATCCGTAACTGGTCTGAGA‐3′ 5′‐TTCCCGAAGGCACTCCTCTATCTCTAAAG‐3′	Scortichini and Marchesi (2001)
			Standard	WA WC	5′‐TCCACAGGACGCCAGCAAGA‐3′ 5′‐TGCGGTGTTACGCCACCATC‐3′	Loreti and Gallelli (2002)
			TaqMan	Pavel‐F Pavel‐R Pavel‐MGB	5′‐GCACTTTAAGTTGGGAGGAA‐3′ 5′‐CAGAGTTAGCCGGTGCTTA‐3′ 5′‐AATACGTATCTGTTTTG‐3′‐MGB	Gervasi and Scortichini (2009)
			Real‐time	Pav‐hsdF Pav‐hsdR	5′‐GAGGCGCAGGGGCTGATGTT‐3′ 5′‐TGGAGGACATCTTGCTTCTTCT‐3′	Turco et al. (2023)
P. syringae pv. theae	Camellia sinensis	Leaf necrosis and wilting, shoot and twig blight, plant death
P. syringae pv. actinidiae	Actinidia chinensis var. chinensis Actinidia chinensis var. deliciosa	Bud necrosis, leaf necrosis and wilting, shoot and twig blight, flower necrosis, trunk canker, plant death	Standard	PsaF1 PsaR2	5′‐TTTTGCTTTGCACACCCGATTTT‐3′ 5′‐CACGCACCCTTCAATCAGGATG‐3′	Rees‐George et al. (2010)
			Standard	PsaF3 PsaR4	5′‐ACCTGGTGAAGTTGGTCAGAGC‐3′ 5′‐CGCACCCTTCAATCAGGATG‐3′
			Duplex	KN‐F KN‐R AvrPtoB‐LF AvrPtoB‐LR	5′‐CACGATACATGGGCTTATGC‐3′ 5′‐CTTTTCATCCACACACTCCG‐3′ 5′‐GGAGAGGATCAGCATATG‐3′ 5′‐TCAGGGGACTATTCTAAAAG‐3′	Gallelli et al. (2011)
			Simplex	P0F P6R	5′‐CTGCAACAGGCGACGGCGAGGC‐3′ 5′‐CATAGGCTTCTGGTTTTCTTCCTGATCC‐3′	Gallelli et al. (2014)
			Real‐time	P3F P5R1	5′‐GGTTTCGGACACCGCAGGTTCTACCGAG‐3′ 5′‐CTTCCTGATCCCCGTTACCCATCGAC‐3′	Gallelli et al. (2014)
			Real‐time	hopA1 Psa_A1F2 Psa_A1 R1	5′‐GCCTCGATGTCGGCGC‐3′ 5′‐ATTCGATAGAAGAACTTCTTTGCGTTT‐3′	Andersen et al. (2018)
			Real‐time	hrpZ5 Psa_Z5 F2 Psa_Z5 R2	5′‐ACAACTTCAGGCTACAATACTTACGC‐3′ 5′‐CTCAGGATGCGTTTCGGTTAC‐3′	Andersen et al. (2018)
			LAMP	F3 B3 FIP BIP loopF loopR	5′‐GGCTCTCCTAGCAAGCATAC‐3′ 5′‐TGAGAAGGGACGCAACCA‐3′ 5′‐CGGATTCGCAACGCTCCAAGATCTGCTGA GCACGTTGGTC‐3′ 5′‐ACTCTTCCGCAACGAGTTTGGGTCCCCAC ATGGAGTTGTCT‐3′ 5′‐TGCCGATCGAGTATCCATTTCCT‐3′ 5′‐AGCCTTTTCCGAACGGTCT‐3′	Ruinelli et al. (2017)
			LAMP (biovar 3)	F3 B3 FIP BIP loopF loopR	5′‐GCTATGGAATCCATTGCGGT‐3′ 5′‐CGCATCTGCTGGATCATCC‐3′ 5′‐ATCCCTTGCCCAGCACGAACATGAGGTCG AGGTGTCTGA‐3′ 5′‐ATCCACAGTGGGTACACGGACGGGGGCA CCTTCTTTCTTGG‐3′ 5′‐GGCCTTTTCAATGCGGTCAATATCC‐3′ 5′‐GGTGTCAATCTGGTGGTGCAT‐3′	Ruinelli et al. (2017)
P. syringae pv. morsprunorum race 2	Prunus avium Prunus cerasus Prunus domestica	Leaf necrosis and wilting, shoot and twig blight, fruit necrosis, trunk canker	Standard	Psm2‐8F/ Psm2‐8R	5′‐CTTTTTAGATGGTGAGGTTTTGTA‐3′ 5′‐ACTTTCGGATCATCGTTTTCTA‐3′	Kałużna et al. (2016)
			Real‐time	Psm2‐1F‐RT/ Psm2‐1R‐RT	5′‐GGTTTGCCTTTTCCTCAG‐3′ 5′‐ATTGCATTACTTCTTTGTTGC‐3′	Kałużna et al. (2016)

The main symptoms are as follows: bud necrosis (kiwifruit), leaf necrosis and wilting (hazelnut, tea, kiwifruit, cherry/plum), shoot and twig blight (hazelnut, tea, kiwifruit, cherry/plum), flower necrosis (kiwifruit), fruit necrosis (cherry/plum), trunk canker (hazelnut, kiwifruit, cherry/plum) and plant death (hazelnut, tea, kiwifruit) (Table 2 and Figure 4). Notably, plants infected with bacterial canker of kiwifruit may exhibit bacterial exudates during early spring. For all the lineages of the P. avellanae species complex, the symptoms are more evident in spring, autumn and late winter, whereas summer is not conducive to either pathogen multiplication or disease progression. Under artificial inoculation, different lineages of the P. avellanae species complex exhibit differential pathogenicity towards selected host plant organs. Strains associated with hazelnut do not cause symptoms when inoculated onto cherry fruitlets and peach leaves. However, they can induce water‐soaked lesions on almond leaves. A strain associated with tea can induce water‐soaked lesions on peach and almond leaves. Strains associated with kiwifruit can induce water‐soaked lesions on cherry fruitlets, whereas strains from cherry can induce such lesions on cherry fruitlets, as well as on peach and almond leaves (Ruinelli et al. 2019). A strain resembling P. avellanae has been isolated from Cornelian cherry (Cornus mas) that caused leaf spotting (Kałużna 2019).

FIGURE 4.

Representative disease symptoms caused by different lineages of Pseudomonas avellanae species complex in (a) European hazelnut, (b) sweet cherry, (c) yellow‐fleshed kiwifruit, and (d) tea. In sweet cherry, canker lesions on branches may also harbour strains of Pseudomonas syringae pv. syringae and Pseudomonas amygdali pv. morsprunorum race 1. Photo credit: (d) M. Goto, University of Shizuoka, Japan.

5. Epidemiology

Detailed information is available on the epidemiology of the P. avellanae species complex. Despite the absence of a complete gene cluster encoding ice nucleation activity, some P. avellanae lineages are favoured by air temperatures below 0°C. The pathogenicity of strains associated with kiwifruit and tea is strongly promoted by frost events (Tomihama et al. 2009; Ferrante and Scortichini 2014), whereas frost damage to plants strongly favours colonisation by the strains associated with hazelnut (Scortichini 2002). The lineage associated with cherry/plum is favoured by air temperatures around 15°C (Bultreys and Kaluzna 2010). The lineage associated with kiwifruit displays an epiphytic phase in spring on buds, leaves, twigs, and flowers. However, the buds colonised during the previous season may fail to develop leaves (Donati et al. 2020). On leaves, optimal conditions for growth are found at the base of glandular trichomes, as these structures retain water and provide nutrients such as sugars, proteins, and secondary metabolites. Leaf wetness is an essential prerequisite for further plant colonisation, which occurs through stomata and floral organs (i.e., stigma and anthers). Upon colonisation of the leaf apoplast, this lineage enters its endophytic phase and, through leaf veins and the petiole, reaches the twig. Visible symptoms (i.e., leaf and shoot blight) occur after the pathogen reaches cell densities ranging from 10³ to 10⁴ CFU g/fresh weight, with optimal air temperatures of 21°C–25°C (Donati et al. 2020). During summer, increased temperatures and reduced leaf wetness limit pathogen multiplication. In autumn, additional systemic invasion may occur through leaf and fruit abscission scars and pruning cuts, a process strongly favoured by rain and wind. This lineage can migrate within the plant both acropetally and basipetally and may also colonise the root system. It overwinters in infected plant organs, fallen leaves and pruning debris, and can also survive on leaves of wild or ornamental plants such as Bellis perennis , Chenopodium album , Taraxacum officinale , Pinus radiata , Cryptomeria japonica and Chamaerops humilis (Donati et al. 2020). At the end of dormancy, high root pressure may displace pathogen cells from the roots to the trunk and buds, leading to systemic infection of shoots. During this period, the pathogen can be released from infected plants through exudates containing bacterial cells, facilitating disease spread to neighbouring plants via wind. In spring, the pathogen may also colonise the anthers internally, leading to pollen infection, which presents another important pathway for environmental dissemination (Donati et al. 2020).

For strains associated with hazelnut, acidic soils, often combined with high aluminium content (i.e., approximately 20%), present important predisposing factors that increase pathogen virulence. In both the Viterbo area and northern Greece, the most severe symptoms (i.e., rapid decline and plant death) were observed in hazelnut orchards with soil pH below 5.0 (Scortichini 2002). Penetration through leaf scars, favoured by rain and wind, presents another key factor in pathogen dissemination in autumn. Pathogenicity assays confirmed this route of entry by showing a high success rate in reproducing field symptoms (Scortichini and Tropiano 1994; Popović Milovanović et al. 2026). As mentioned above, winter and spring frost, by causing wounds on trunks and branches, favours subsequent colonisation by bacterial cells carried by rain and wind. Once inside the tree, similarly to strains infecting kiwifruit, the pathogen can move acropetally and basipetally, reaching either the apical parts or the root system. Consequently, large‐scale disease spread occurs through the circulation of latently infected propagative material derived from apparently healthy mother plants (Scortichini and Tropiano 1994). This dissemination pathway was also confirmed by the occurrence of bacterial canker of hazelnut in Serbia (Popović Milovanović et al. 2026).

Early autumn frost also promotes symptom development in strains associated with bacterial shoot blight of tea (Tomihama et al. 2009). Interestingly, in this pathosystem, a clear association was observed between ice nucleation‐active strains of Xanthomonas campestris on the leaf surface (particularly around stomata) and pseudomonads. These X. campestris strains, by causing damage to plant tissues, facilitate colonisation and systemic invasion by pseudomonads already present on the phylloplane (Tomihama et al. 2009). The disease is favoured by air temperatures ranging from 10°C to 15°C, and similarly to other pathosystems, high summer temperatures are not conducive to disease progression (Tomihama et al. 2009).

For the lineage associated with cherry and plum, tree wounds caused by frost or mechanical damage present the primary entry points, whereas leaf scars are of secondary importance (Hulin, Mansfield, et al. 2018). Strains of this lineage are also capable of systemic movement from the inoculation site and can induce gummosis from cankers (Hulin, Mansfield, et al. 2018). It is noteworthy that C. avellana , Actinidia spp. and Prunus spp. are host plants for taxonomically distinct phytopathogenic pseudomonads belonging to different species complexes, namely P. avellanae , P. amygdali and P. syringae , representing cases of convergent evolution and adaptation to the same host. European hazelnut and kiwifruit can be infected by strains of P. avellanae and P. syringae , whereas cherry and plum can be infected by strains of all three species (Gomila et al. 2017).

6. Detection

Pseudomonads causing cankers, as well as twig or leaf blight and dieback, can be identified using selective media, pathogenicity tests, and genetic methods (Schaad et al. 2001). Despite the growing amount of genomic information available for these pathogens, standardised diagnostic procedures remain limited. Currently, within the P. avellanae species complex, an official diagnostic protocol has been established only for P. syringae pv. actinidiae, with guidelines for detection and identification in different types of plant material (EPPO 2021).

All lineages within the P. avellanae complex can be detected in both symptomatic and asymptomatic plant tissues. Isolation is recommended as an initial step for identification and can be performed using tissues such as twigs, buds, and cankers. Bacteria can be cultured on general or semiselective media, including nutrient agar (NA), NA supplemented with 5% sucrose (NSA), NSA supplemented with antibiotics, King's B medium (KB), modified KB, as well as peptone sucrose agar (PSA) (Scortichini 1994). Colonies are levan‐positive and typically appear smooth, circular, and convex with entire margins. After 3 days of incubation on NSA, they develop as pearly‐white, mucoid, dome‐shaped colonies. On modified KB medium, colonies are generally smooth and flat, with entire to slightly lobed margins, and show a pearly‐white to yellowish coloration. In P. syringae pv. morsprunorum race 2, colony morphology may initially differ from typical strains, forming levan‐type “fried‐egg” colonies on NSA after 40–84 h. However, after 3–4 days, they develop the typical morphology of standard strains (Freigoun and Crosse 1975). Although isolation is recommended, it shows lower sensitivity than molecular methods (Janse et al. 1996; Scortichini 2012; EPPO 2021; Loreti et al. 2018).

Different phenotypic traits and molecular diagnostic tools have been proposed to characterise and differentiate P. avellanae populations (Janse et al. 1996; Scortichini et al. 1998, 2002, 1979, 2005, 2013; Loreti et al. 2009; Popović Milovanović et al. 2026). Strains of P. avellanae , P. syringae pv. theae, P. syringae pv. actinidiae, and P. syringae pv. morsprunorum race 2 share several key phenotypic characteristics, including levan production, oxidase negativity, absence of potato soft rot and arginine dehydrolase activity, together with the induction of a hypersensitive response (HR) in tobacco, which places them within LOPAT group Ia (Scortichini et al. 2002; Kałużna et al. 2016). Differentiation among P. syringae pathogens originating from stone fruits can be performed using GATTa test (Freigoun and Crosse 1975; Latorre and Jones 1979). However, variable results have been reported for P. syringae pv. morsprunorum race 2 (Gilbert et al. 2009). Scortichini (1994) reported that nutritional and protein profile analyses revealed slight differences between Japanese and Italian strains of P. syringae pv. actinidiae. Reliable protocols for pathogenicity testing of pseudomonads of woody origin are available. Artificial inoculation of 1‐year‐old hazelnut twigs at leaf scars in early autumn with P. avellanae strains reliably induces twig dieback in the following spring (Scortichini and Lazzari 1996; Scortichini 2012). The pathogenicity of P. syringae pv. actinidiae can be assessed on plantlets of A. chinensis (cultivars Erica, Hort16A) and A. deliciosa (cultivars Hayward, Soreli) (EPPO 2021), as well as by in vitro leaf‐disc assays (Prencipe et al. 2018; Brunetti et al. 2020). In plantlet assays, typical symptoms appear 6–10 days after inoculation and are monitored for up to 3 weeks, whereas in leaf‐disc assays, necrotic spots develop after 6 days and spread over 14–20 days to cover the entire disc. According to early studies (Freigoun and Crosse 1975), P. syringae pv. morsprunorum variant isolates infected leaf scars of 4‐year‐old cv. Roundel trees, mostly producing localised lesions, whereas typical strains caused progressive branch lesions and bud death. For inoculation purposes, wound inoculation is recommended, as variant isolates were generally more virulent. They produced longer cankers and more severe branch dieback with specificity for cv. Roundel (Freigoun and Crosse 1975). Kałużna (2019) described a method based on infiltration of bacterial suspensions into leaves using a hypodermic syringe when testing pathogenic strains isolated from Cornelian cherry, which are among the closest relatives of hazelnut‐associated P. avellanae strains.

Serological methods, including ELISA, slide agglutination, and immunofluorescence (IF), are still widely used in plant pathogen diagnostics and have been applied to various species of genus Pseudomonas. However, their usefulness for the P. avellanae and P. syringae complexes from stone fruits is limited due to cross‐reactions with nonpathogenic bacteria and poor discrimination among strains (Fiori et al. 2003; Vicente et al. 2004). A rapid serological test (immunochromatographic strips), based on the use of a polyclonal antiserum against the effector HopZ5, was recently developed for P. syringae pv. actinidiae detection (Chen et al. 2018), but no interlaboratory validation data are currently available.

The majority of methods used for detecting lineages of the P. avellanae species complex rely on molecular tools, particularly PCR, due to their higher sensitivity and specificity (Table 2). Two populations of P. avellanae both isolated from hazelnut can be differentiated using BOX‐PCR and ERIC‐PCR (Scortichini 2012; Popović Milovanović et al. 2026). A specific primer set (PAV 1/PAV 22; Table 2) for conventional PCR detection of P. avellanae was developed by Scortichini and Marchesi (2001), based on regions of the 16S rRNA gene of strain PD 2390 that exhibited variability compared with the corresponding regions in the related species P. syringae pv. syringae. Soon after, Loreti and Gallelli (2002) developed a primer set (WA/WC; Table 2) for the detection of virulent P. avellanae strains, selected by comparing nucleotide sequences of the hrpW gene in P. avellanae , P. syringae pv. tomato and P. syringae pv. syringae. A TaqMan real‐time PCR assay was developed by Gervasi and Scortichini (2009) using two primers and a probe designed based on regions of the 16S rRNA gene of P. avellanae strain ISF 2059: the forward primer Pavel‐F, the reverse primer Pavel‐R, and the probe Pavel‐MGB (Table 2). Turco et al. (2023) developed a specific BRYT Green dye‐based qPCR assay, a sensitive tool for the detection and quantification of P. avellanae in field samples, based on a pangenomic analysis of the bacterial genome (Table 2).

Several molecular techniques have been developed for the detection and identification of P. syringae pv. actinidiae, including LAMP, PCR‐restriction fragment length polymorphism (PCR‐RFLP), multiplex, real‐time quantitative, and nested PCR (Rees‐George et al. 2010; Balestra et al. 2013; Biondi et al. 2013; Gallelli et al. 2014; Ruinelli et al. 2017). Rees‐George et al. (2010) developed two sets of PCR primers for conventional PCR, PsaF1/PsaR2 and PsaF3/PsaR4, designed to target a portion of the 16S–23S rDNA intertranscribed spacer (ITS) region (Table 2). Balestra et al. (2013) designed a primer pair, P. syringae pv. actinidiae F/ P. syringae pv. actinidiae R (Table 2), for a multiplex PCR assay for the detection of P. syringae pv. actinidiae. As some of these methods may yield false‐positive results, several alternative assays have been developed and validated. These include conventional duplex PCR (Gallelli et al. 2011), followed by conventional simplex and real‐time PCR (Gallelli et al. 2014), LAMP (Ruinelli et al. 2017), and real‐time PCR (Andersen et al. 2018). These methods were validated by EPPO (2021) and confirmed to detect P. syringae pv. actinidiae biovars 1, 2, and 3 (Table 2). Recently, Luo et al. (2025) developed a propidium monoazide (PMA)‐PCR assay for the detection of viable cells of P. syringae pv. actinidiae. Strain‐specific primers targeting the hopZ3 effector gene were combined with the nucleic acid‐binding dye PMA to exclude DNA from dead cells. The method showed high sensitivity (detection limit: 10 CFU/mL or 1 pg/μL) and was successfully applied to infected kiwifruit tissues, providing a rapid and reliable alternative to conventional PCR for detecting viable P. syringae pv. actinidiae. Kałużna et al. (2016) developed primer sets for the detection of P. syringae pv. morsprunorum race 2: Psm2‐8F/Psm2‐8R for conventional PCR and Psm2‐1F‐RT/Psm2‐1R‐RT for real‐time PCR (Table 2).

Phylogenetic analysis of lineages within the P. avellanae species complex can be performed using MLSA based on four housekeeping genes, following the MLSA scheme recommended for P. syringae: gapA (glyceraldehyde‐3‐phosphate dehydrogenase), gyrB (DNA gyrase subunit B), gltA (citrate synthase), and rpoD (RNA polymerase sigma factor RpoD) (Sarkar and Guttman 2004; Hwang et al. 2005; Ferrante and Scortichini 2010, 2011; Kałużna et al. 2010). More recently, Iličić et al. (2022) recommended the use of recG (ATP‐dependent DNA helicase RecG) for the identification and characterisation of the P. syringae species complex. Subsequently, Popović Milovanović et al. (2026) demonstrated that this gene can also be used in phylogenetic analyses of hazelnut‐associated P. avellanae . Assignment of biovars of P. syringae pv. actinidiae is possible using a scheme based on seven genes (acnB, cts, gapA, gyrB, pfk, pgi, and rpoD) (Sarkar and Guttman 2004; Sawada et al. 2014).

For many years, identification of canker‐causing pseudomonads has largely relied on the detection of genes involved in the production of specific virulence factors, including the toxins coronatine and syringomycin, as well as the siderophore yersiniabactin. Within the P. avellanae species complex, the siderophore yersiniabactin (irp1 gene) has been detected in P. avellanae (Marcelletti and Scortichini 2015) and P. syringae pv. morsprunorum race 2 (Bultreys et al. 2006). The cfl gene responsible for coronatine synthesis has been reported in P. syringae pv. actinidiae. In addition, P. syringae pv. actinidiae possesses the tox–argK gene encoding phaseolotoxin (Templeton et al. 2005). Within four genetically distinct populations of P. syringae pv. actinidiae, Psa1 and Psa2 produce phaseolotoxin and coronatine, respectively, whereas Psa3 and Psa4 do not produce these toxins (Scortichini 2012; Vanneste et al. 2013; Abelleira et al. 2014).

Scortichini et al. (2002) found that ARDRA, repetitive PCR using ERIC, BOX and REP primer sets, whole‐cell protein analysis, biochemical and nutritional tests, and pathogenicity tests can be used for the identification and comparison of P. avellanae , P. syringae pv. theae and P. syringae pv. actinidiae. The rapid progress of high‐throughput sequencing technologies, particularly next‐generation and third‐generation sequencing (NGS and TGS), has greatly expanded the possibilities for comparative genomic studies. Earlier research relied mainly on molecular markers derived from a limited number of conserved housekeeping genes or protein profiles, providing only partial insight into genetic diversity. In contrast, the increasing availability of complete genome sequences now enables genome‐wide comparisons among strains. Such genome‐scale analyses, including pangenome studies, have significantly improved our understanding of the genetic basis of pathogenicity and host adaptation, revealing factors that may facilitate specialisation on the same host or, in some cases, enable pathogens to colonise novel host plants (Marcelletti and Scortichini 2015, 2019).

7. Disease Control

In general, sites characterised by frequent winter and spring frost should be avoided for the cultivation of host plants susceptible to the P. avellanae species complex. In addition, certain agronomic practices may strongly favour disease progression, such as excessive nitrogen fertilisation, iron deficiency, and water stress (Mauri et al. 2016). Disease management should primarily focus on protecting the main points of entry into the plant (i.e., wounds, leaf scars, stomata) during periods conducive to disease development (i.e., late winter, spring, and autumn), as well as on the use of healthy propagative plant material. Although several models based on climatic parameters and pathogen biology have been developed to assess the potential global suitability of geographical areas for future epidemics (Narouei‐Khandan et al. 2022), fine‐tuned disease prediction at a local scale, based on weather forecasts, is still not available for these different lineages. This presents a limitation for the effective management of diseases causing significant economic losses. It should also be noted that this species complex has an endophytic phase during a substantial part of its life cycle, which further complicates the achievement of effective crop protection. Nevertheless, several strategies, when combined, may contribute to symptom mitigation. For bacterial canker of kiwifruit, management is mainly based on the application of copper‐based compounds in early spring to reduce epiphytic colonisation of new leaves. These compounds are also widely applied after harvest, during leaf fall, and following hail events, heavy rain, and winter frost, in order to protect petioles, leaf scars, and wounds (Cameron and Sarojini 2014). Pruning cuts should be protected with copper as well.

However, due to their phytotoxicity, pathogen penetration through flowers cannot be effectively prevented. In addition, a major concern is the increasing occurrence of copper‐resistant strains, which have frequently been isolated from kiwifruit orchards subjected to repeated copper applications (Colombi et al. 2017; Cesari et al. 2023; Thomidis et al. 2025). The use of chitosan‐based compounds may present an alternative to replace or complement copper applications (Scortichini 2014). Biocontrol agents also appear promising, either for flower protection or for limiting epiphytic colonisation by the pathogen. Some strains of Bacillus spp., namely B. amyloliquefaciens D747 and B. subtilis QST703, are already used as commercial products. An interesting strategy involves the selection of protective strains from the native kiwifruit microflora, such as P. azotoformans and P. putida isolated from Actinidia spp., which produce effective siderophores (Correia et al. 2025). The advantage of this approach lies in the use of microorganisms already adapted to common field management practices, which are capable of efficient host colonisation. Lactobacillus plantarum is another promising biocontrol agent, capable of preventing the epiphytic colonisation through pH reduction associated with lactic acid production (Daranas et al. 2019).

Natural products, endophytic microflora, and bacteriophages also offer promising opportunities for managing bacterial canker of kiwifruit. Adesmia balsamica and Leptospermum scoparium may provide plant extracts and endophytic microflora (i.e., Pseudomonas spp.), respectively, capable of reducing in vivo growth of the pathogenic strains associated with bacterial canker of hazelnut (Wicaksono et al. 2018; Díaz et al. 2025). Lytic phages and phage cocktails, including dual and triple combinations, are also being extensively investigated for the development of commercial products for the management of bacterial canker of kiwifruit, with promising results reported for in vivo symptom mitigation (Yu et al. 2016; Flores et al. 2020; Fiorillo et al. 2023).

Similarly to bacterial canker of kiwifruit, copper‐based compounds are widely used for the management of bacterial canker of hazelnut and cherry/plum. Their application occurs mainly in autumn to protect leaf scars (hazelnut) and in early spring to protect frost wounds. In addition, for bacterial canker of cherry, bacteriophages are currently being investigated as alternative management strategies (Rabiey et al. 2020). The search for resistant plant germplasm within wild species presents another strategy to replace highly susceptible cherry cultivars (Hulin, Vadillo‐Díeguez, et al. 2020).

8. Concluding Remarks and Perspective

In‐depth taxonomic studies have revealed previously unrecognised relationships among different phytopathogenic pseudomonads belonging to the P. syringae complex. Within this group, the P. avellanae species complex comprises taxonomically related yet distinct lineages capable of infecting different, botanically unrelated host plants. In addition, its occurrence is not limited to cultivated plant species, as a strain of this species complex has also been isolated from a natural habitat, namely river epilithon. Strains associated with the lineage causing bacterial canker of kiwifruit have also been detected in natural habitats linked to wild Actinidia spp. (McCann et al. 2017). Strains of this species complex can readily colonise plant organs and thus spread to different cultivation areas through the movement of propagative plant material, as observed for the pandemic strain of bacterial canker of kiwifruit (Butler et al. 2013; McCann et al. 2017), as well as for outbreaks of bacterial canker of European hazelnut in Italy and Serbia (Scortichini and Tropiano 1994; Popović Milovanović et al. 2026).

The global circulation of plant material presents a major pathway for the worldwide spread of these diseases. Effective and specific detection techniques are therefore essential for the rapid identification of latently infected plant material at borders as well as for detecting initial infection foci in newly affected areas. The ability to colonise host plants through endophytic spread is another key feature shared by lineages within the P. avellanae species complex. Lineages associated with bacterial canker of European hazelnut and kiwifruit can extensively colonise the entire plant, whereas the lineage causing bacterial shoot blight of tea can systemically colonise leaves and twigs. This capability may lead to rapid host plant decline and death. The lineage associated with bacterial canker of cherry is characterised by a strong capacity to colonise branch parenchyma following penetration through wounds (Freigoun and Crosse 1975). These characteristics present a major constraint for effective field management and highlight the need to develop alternative strategies capable of replacing copper‐based compounds, particularly those targeting the pathogen within plant tissues. Modern disease management approaches, including biocontrol agents, bacteriophages, nanoparticles, and disease forecasting models, may help reduce the impact of the P. avellanae species complex.

Author Contributions

Marco Scortichini: conceptualization, supervision, visualization, resources, writing – original draft, writing – review and editing, data curation. Tatjana Popović Milovanović: conceptualization, writing – original draft, writing – review and editing, supervision. Renata Iličić: writing – original draft, writing – review and editing, supervision.

Funding

The authors acknowledge support from the Ministry of Education, Science and Technological Development, Republic of Serbia (contract numbers 451‐03‐33/2026‐03/200010 and 451‐03‐33/2026‐03/200117).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing is not applicable as no new data were generated.