Pseudomonas syringae is the most ubiquitous bacterial plant pathogen on Earth. Numerous pathogenic varieties (pathovars) of P. syringae infect a plethora of different host plants worldwide. The life cycle of P. syringae is characterized by an epiphytic and an endophytic phase (Xin et al., 2018). The epiphytic phase encompasses bacterial survival and proliferation in fluctuating environmental conditions at the leaf surface and is characterized by the effort to gain access to the apoplast, an intercellular space filled with air and water, through leaf openings (stomata, hydathodes or wounds). By entering the apoplast, P. syringae starts the endophytic phase, establishing a parasitic lifestyle and proliferating in this nutrient-accessible environment (Baltrus et al., 2017).

Like many other Gram-negative bacterial plant pathogens (Büttner and He, 2009), P. syringae uses its type III secretion system (T3SS) for infection and suppression of plant immunity (Dillon et al., 2019). The T3SS is a syringe-like structure that injects virulence effector proteins directly into the cytosol of eukaryotic target cells (Galán and Wolf-Watz, 2006; Büttner, 2012; Liu et al., 2016; Deng et al., 2017). In P. syringae, the extracellular pilus, which functions as a conduit for effector transfer (Loquet et al., 2012), is helically assembled from hundreds of copies of small HrpA proteins (around 11 kDa), which form a long narrow channel (2 μm in length, 6–8 nm in diameter) extruding from the bacterial outer membrane (Xin et al., 2018). While the HrpA sequence varies among pathovars of P. syringae, possibly as an adaptation to different cuticular and cell wall barriers in host plants, HrpA proteins in different pathovars share similar physicochemical and structural properties (Guttman et al., 2006). P. syringae strains have an exceptionally large variety of T3SS effector proteins, classified into 70 different families (Bundalovic-Torma et al., 2022), of which individual strains express 15–30 effectors (Wei et al., 2018). The four effectors encoded in the conserved Hrp¹ pathogenicity island effector locus (AvrE, HopI, HopAA and HopM)² are considered as core effectors, whereas the majority of effectors are encoded elsewhere in the genome (Lindeberg et al., 2012). Besides the virulence effectors, plant pathogens secrete T3SS helper proteins (harpins) involved in effector translocation into target cells. The diversity in the T3SS arsenal indicates that the strain-specific effector proteins repertoire plays an important role in host specificity and overall severity of P. syringae infections. However, effectors and harpins can also activate plant-triggered defense mechanisms (effector-triggered immunity, ETI) (Laflamme et al., 2020), such as the hypersensitive response (Choi et al., 2013). Effectors which could elicit ETI are widespread in P. syringae populations. A recent study showed that 97.2% of the 494 genomes of P. syringae strains organized in the P. syringae Type III Effector Compendium (PsyTEC) possess at least one orthologue of putative ETI-eliciting alleles, with 74.3% of strains encoding more than one (up to eight) putative ETI-elicitors (Martel et al., 2022).

In addition to the T3SS, type VI secretion systems (T6SS) have been attributed a major role in virulence and plant colonization of P. syringae; their importance in inter-bacterial competition has also recently been highlighted (Chien et al., 2020). The T6SS resembles a phage tail-like device formed by a rigid tube of hexameric Hcp (TssD) rings wrapped in a contractile sheath. Upon contraction of the sheath, the Hcp tube, which is sharpened by a tip of the VgrG (TssI) and PAAR proteins, is propelled outside the cell, and into a target cell (Cherrak et al., 2019). Hcp and VgrG proteins are detectable in the extracellular medium of T6SS-active bacteria, which allows monitoring of T6SS activity in vitro (Pukatzki et al., 2006). Hcp and VgrG can carry additional C-terminal extension regions acting as effector domains (Pukatzki et al., 2007; Blondel et al., 2009). Such so-called specialized Hcp proteins with effector domains are common among Enterobacteriaceae (Ma et al., 2017), while specialized VgrG proteins are widespread in Betaproteobacteria and Gammaproteobacteria (Lien and Lai, 2017). Other T6SS effectors are so-called cargo effectors, which can be secreted by association with the exported T6SS components. Some cargo effectors, such as rearrangement hotspot (Rhs) and related YD-peptide repeat proteins, are often associated with an N-terminal PAAR domain or downstream of VgrG and PAAR-encoding genes (Koskiniemi et al., 2013; Pei et al., 2020). In contrast, most cargo effectors are encoded dispersed in the genome, usually next to a gene encoding their cognate immunity protein, which prevents self-harm. Accordingly, many T6SS effectors still await their identification. Bacteria, including plant pathogens such as P. syringae, Agrobacterium tumefaciens, and Pantoea anantis, can harbor multiple T6SS clusters in single strains (Sarris et al., 2010; Chien et al., 2020); however, other P. syringae pathovars contain only one T6SS cluster (Bernal et al., 2018).

Besides manipulation of the host by secretion systems, motility is a critical factor for bacterial plant colonization and the successful development of both life stages, epiphytic and endophytic (Colin et al., 2021). The flagellum is a long helical structure composed of numerous flagellin subunits, attached to a transmembrane motor complex (Rossez et al., 2015), the activity of which influences various pathogenic properties of P. syringae (Taguchi et al., 2010; Morris et al., 2019). Moreover, flagellin-derived peptides are among the first recognized elicitors of plant immunity (Sanguankiattichai et al., 2022), and the secretion of flagellin proteins in apoplast-like condition could be one indicators for assessing the pathogenic potential of P. syringae strains.

P. syringae pv. aptata strains are causative agents of leaf spot disease, which affects several hosts, such as sugar beet, Swiss chard, cantaloupe and squash (Morris et al., 2000; Koike et al., 2003; Sedighian et al., 2014; Nikolić et al., 2018). Symptoms of the disease include circular or irregular necrotic spots on leaves, with a light-brown center and glassy black margins in the early stage (Stojšin et al., 2015). Mounting evidence for disease occurrence in sugar beets fields worldwide indicates the robust pathogenic potential of this particular P. syringae pathovar (Stojšin et al., 2015; Arabiat et al., 2016; Rotondo et al., 2019; Nampijja et al., 2021). A previous comparative characterization of individual strains from a collection of P. syringae pv. aptata found high diversity in terms of pathogenicity and host range (Nikolić et al., 2018; Morris et al., 2019). In particular, pathogenicity assays and measurements of disease severity conducted in greenhouse experiments revealed distinctive groups of strains, classified as low, intermediate and high virulence. The high virulence strains display a broad host range (10–16 plant species), whereas low virulence strains caused disease in one or few tested host plants, indicating a narrow host range (Nikolić et al., 2018; Morris et al., 2019). In this report, we analyze the secretome of six selected P. syringae pv. aptata strains covering a wide range of distinctive pathogenic characteristics (low/high virulence and narrow/broad host range) under in vitro conditions that mimic the apoplast environment, in order to link the repertoire of secreted effector proteins and the disease outcome. To this aim, we performed the first in-depth secretome analysis of the emerging sugar beet pathogen Pseudomonas syringae pv. aptata and focused on the identification and quantification of T3SS and T6SS effectors and motility factors. Notably, we found that the investigated low virulence strains of P. syringae pv. aptata secreted large amounts of a defined spectrum of type III secretion effectors, whereas high virulence strains have a broader spectrum of effectors generally secreted in lower abundance. While all strains showed high T6SS activity, medium and high virulence strains secreted an additional subset of known T6SS effectors and previously uncharacterized co-secreted proteins that are candidates for novel P. syringae T6SS effectors. This identification of distinct protein secretion profiles may reveal defined infection strategies, which correlate with the infection severity and host range of P. syringae pv. aptata.

The establishment of a bacterial plant infection involves multiple virulence factors and corresponding host defense mechanisms, all of which are modulated by environmental factors (Morris et al., 2019). Understanding the pathogenic potential of P. syringae requires appreciating the structure, functionality, and mode of action of these virulence factors, especially at the crucial step when the pathogen enters into the plant apoplast environment. The main aim of this study was to evaluate the hypothesis that differences in pathogenicity between P. syringae pv. aptata strains correspond to variations in the repertoire and level of secreted virulence factors. To this aim, we performed functional assays and quantitative proteomics for six P. syringae pv. aptata strains with known pathogenic potential, focusing on proteins exported by the T3SS and T6SS, as well as flagellar proteins, which all are known to contribute to the pathogenic potential of bacterial plant pathogens (Pontes et al., 2020).

A recent genome-wide analysis in P. syringae pv. aptata P16 and P21 showed that these strains have 16 and 25 ORFs for T3SS effectors, respectively (Ranković et al., 2023), in line with a previous study finding 12 to 40 Hop/Avr open reading frames in various P. syringae pathovars (Baltrus et al., 2011). Here, using label-free quantification LC/MS–MS, we found that 23 different T3SS effectors were secreted into the extracellular fraction across the six tested P. syringae pv. aptata strains, confirming the secretion of most effectors detected in the genome-wide analysis (Ranković et al., 2023). Among the most abundant effectors were AvrE1 and HopM1, which have a crucial role in modulating humidity conditions after bacteria have entered the leaf apoplast (Xin et al., 2016), providing water-soaked spots, which were observed as an early disease symptom in previous in planta testing (Nikolić et al., 2018; Morris et al., 2019). Another highly abundant effector is HopBF1, involved in the phosphorylation of the molecular chaperone HSP90, which plays a role in pathogen-associated molecular pattern-triggered immunity, preventing activation of plant immunity response leading to the development of disease symptoms (Lopez et al., 2019).

The plant immune system can recognize the secretion of T3SS effectors by plant pathogens, which may result in effector-triggered immunity (ETI). As a response, plant pathogens have evolved mechanisms to avoid ETI, such as (i) mutations, recombination or loss of effector genes (Bundalovic-Torma et al., 2022), (ii) fine-tuning of secretion dosage of the core effectors (Laflamme et al., 2020), and (iii) metaeffector (effector-effector) interactions, where certain effectors suppress ETI triggered by other effectors (Martel et al., 2022). The low pathogenicity strains in our study (P16 and P17) show a significantly higher abundance of harpins, which could have a role in inducing a hypersensitive reaction (Choi et al., 2013) and well-known ETI-eliciting effectors (AvrE1, AvrRpm1, HopI1, HopZ3, HopAZ1 HopAA1 and HopBA1) (Laflamme et al., 2020), in comparison to high pathogenic strains. This suggests that the absence of fine-tuning in the secretion of core effectors (AvrE1, HopI1, HopAA1) and abundant secretion of effectors, which could lead to ETI, results in reduced pathogenicity. This hypothesis aligns with a recent finding that ETI in Arabidopsis was elicited upon overexpression of the P. syringae pv. tomato DC3000 effector AvrE1 (Laflamme et al., 2020). Furthermore, AvrE, HopI, HopZ, HopAZ, HopAA, and HopBA, which were abundantly secreted by the low pathogenicity strains P16 and P17 in our study, are members of effector families that elicit ETI in Arabidopsis.

Our study identified three T3SS effectors (AvrRpm1, HopAW1, and HopAU1) that were not detected in P. syringae pv. aptata so far, further broadening the effector spectrum of the pathovar. AvrRpm1 contributes to the suppression of plant defensive mechanisms by phosphorylation of the RIN4 protein, which represents a regulator of basal defense response in plants (Geng et al., 2016). HopAW1, also found in P. syringae pv. phaseolicola, is an effector protein containing a catalytic triad CHD motif, indicating cysteine protease activity (Mansfield, 2009). HopAU1 plays a role in the late stage of infection in P. syringae pv. phaseolicola, when it creates a replication-permissive niche in the apoplast, as ΔhopAU1 strains display reduced growth in the plant apoplast (Macho et al., 2012). Quite suggestively, we found these effectors to be abundant in the secretome of strains P21 and P26, which have medium and high pathogenic features, indicating a possible role of these effectors in pathogenicity (aggressive virulence and broader host range) in a strain-specific manner.

In terms of pathogenic properties, strain P21 displays very high virulence with more than 70% leaf plate necrosis on sugar beet plants and with a wide host range, including the same virulence level on 16 different plant species (Morris et al., 2019), while strain P26 showed intermediate virulence on sugar beet with 35% leaf plate necrosis (Nikolić et al., 2018) and a medium host range, causing disease on pea, tomato, onion, and parsley (Morris et al., 2019). Here, our results showed that these two strains exhibited low secretion of T3SS effectors, which are among the recognizable effectors of the plant immune system, suggesting that fine-tuning the ETI-triggering effectors in combination with these three effectors could be lethal to different plant species. On the other hand, strains P16 and P17 showed extremely low virulence on sugar beet with less than 10% necrosis and a narrow host range (parsley only for P16; onion and parsley for P17) (Nikolić et al., 2018; Morris et al., 2019). Our secretome analysis results showed a surprisingly high abundance of T3SS effectors secreted in strains P16 and P17, suggesting that a high abundance of effectors could lead to a rapid and pronounced response of the plant immune system and consequently to a weak development of disease symptoms.

In addition to T3SS-secreted effectors, we observed significant amounts of secreted T6SS components in the culture supernatant under apoplast-like conditions. T6SS in Gram-negative bacteria can be involved both in virulence and inter-bacterial competition (Gallegos-Monterrosa and Coulthurst, 2021; Unni et al., 2022). The number of T6SS clusters varies within the P. syringae species complex. While only one T6SS cluster is present in P. syringae pv. actinidae, P. syringae pv. syringae and P. syringae pv. aesculi (Bernal et al., 2018), the presence of two T6SS clusters was confirmed in P. syringae pv. tomato, P. syringae pv. oryze and P. syringae pv. tabaci (Sarris et al., 2010). Genomic mapping of strain P21 showed the existence of two different T6SS genetic islands in this genome (Figure 2). The Hcp1 and VgrG proteins of cluster 1 were found to be secreted in high abundance by all strains, whereas another Hcp1, VgrG and a Rhs protein were distributed distinctly differently. By comparing the co-regulation of protein secretion in the tested strains, we identified several proteins as putative novel T6SS effectors, expanding the repertoire of putative effectors within the P. syringae species complex: Serralysin-like alkaline metalloproteases have also been detected to be co-regulated with T6SS proteins in P. syringae (Ali et al., 2022). The family of YD repeat proteins includes many bacterial toxins and proteins involved in cell–cell contact and communication, some of which are T6SS effectors (Koskiniemi et al., 2013). TonB-dependent siderophore receptors mediate substrate-specific transport across the outer membrane, mostly involved in the uptake of iron (Lin et al., 2017), in line with the recent finding that some T6SS effectors have a role in metal uptake when secreted into the extracellular environment and interact with TonB receptors allowing active transport of metal ions (Hernandez et al., 2020). Although P. syringae T6SS target both eukaryotic and prokaryotic cells, the used apoplast-like conditions without any additional cues mimicking interbacterial competition, conceivably induce the secretion of effectors targeted for plant cells rather than other microbes. In line with this, while we did not test for antimicrobial effects of the T6SS effector candidates detected in this study, bioinformatic analysis did not yield any evidence for such antimicrobial activity.

Our study shows a complex relationship between the function of T3SS and T6SS in P. syringae pv. aptata in apoplast-like conditions and suggests that the interplay between these two important systems shapes pathogenic strategies in a strain-specific manner. Strains P23 and P96 showed strong secretion of T6SS-related proteins and low secretion of most T3SS effectors, suggesting that in apoplast-like conditions, these strains rely more on the activity of the T6SS. On the other hand, the low-virulence strains P16 and P17 secreted higher amounts of most T3SS effectors, and lower amounts of T6SS-related proteins. Wang and colleagues found that T6SS mutants of P. syringae pv. actinidae had no pathogenicity phenotype in planta, but showed a strong down-regulation of several T3SS genes (Wang et al., 2021), highlighting the interplay of both secretion systems during infection. While further research is clearly needed to understand the role of the T6SS in P. syringae pathogenicity and its interaction with the T3SS, our results indicate that effector repertoire and secretion are correlated between the two secretion systems, contributing to distinct infection strategies.

Motility greatly impacts several aspects of P. syringae life cycle, especially in the transition from epiphytic to endophytic phase, and consequently to plant tissue invasion, pathogenicity, and disease symptoms development (Venieraki et al., 2016). In contrast to the results for T3SS-and T6SS-secreted proteins, we did not find a clear correlation between the abundance of flagellar proteins and swarming or swimming motility among tested strains, with the possible exception of strain P16, which showed low levels of exported flagellar components and the lowest motility (Supplementary Figure S1).

This study was initially motivated by previous findings of large intra-pathovar diversity in pathogenicity within a strain collection. We reasoned that this divergence could be based on the differential secretion of effector proteins. Our secretome profiling of six P. syringae pv. aptata strains in apoplast-like conditions defines the T3SS effector repertoire and shows clear differences in the abundance of secreted effectors among strains with different pathogenic phenotypes. Based on the higher secretion level of conserved T3SS effectors in the low pathogenic strains, we hypothesize that P. syringae infection outcome depends on the fine-tuning of effector secretion, especially for ETI elicitors. In addition, we have expanded the spectrum of T3SS and T6SS effectors in P. syringae strains pathogenic on sugar beet, leading us to conclude that P. syringae pv. aptata may use protein secretion in a more diverse and flexible manner during infection than was previously known. Our analysis of the repertoire and abundance of secreted virulence factors between strains of different pathogenicity allows to distinguish three strategies in P. syringae strains pathogenic on sugar beet: (i) strong secretion of the majority of T3SS effectors/low secretion of T6SS effectors, (ii) strong secretion of differently regulated T3SS effectors/low secretion of T6SS effectors, and (iii) strong secretion of T6SS effectors/low secretion of all T3SS effectors. These patterns of secreted major virulence factors could contribute to a better understanding of overall virulence strategies of P. syringae and allow for tailored treatment of high-pathogenicity strains and pathovars of this important plant pathogen.

The datasets presented in this study can be found at: http://www.ebi.ac.uk/pride/archive/projects/PXD040715.

IN and AD: experimental work. IN, TR, TG, and AD: methodology and data analysis. IN and AD: writing the manuscript. SS, TB, and AD: review and editing. All authors contributed to the article and approved the submitted version.

Work in the research group of Andreas Diepold is supported by the Max Planck Society. This work was supported by the Ministry of Science, Technological Development and Innovations of Serbia, Grant no. 451–03-47/2023–01/200178 and Postdoc Fellowship no. 451–03-854/2019–14. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.