The PTI‐suppressing Avr2 effector from Fusarium oxysporum suppresses mono‐ubiquitination and plasma membrane dissociation of BIK1

Abstract Plant pathogens use effector proteins to target host processes involved in pathogen perception, immune signalling, or defence outputs. Unlike foliar pathogens, it is poorly understood how root‐invading pathogens suppress immunity. The Avr2 effector from the tomato root‐ and xylem‐colonizing pathogen Fusarium oxysporum suppresses immune signalling induced by various pathogen‐associated molecular patterns (PAMPs). It is unknown how Avr2 targets the immune system. Transgenic AVR2 Arabidopsis thaliana phenocopies mutants in which the pattern recognition receptor (PRR) co‐receptor BRI1‐ASSOCIATED RECEPTOR KINASE (BAK1) or its downstream signalling kinase BOTRYTIS‐INDUCED KINASE 1 (BIK1) are knocked out. We therefore tested whether these kinases are Avr2 targets. Flg22‐induced complex formation of the PRR FLAGELLIN SENSITIVE 2 and BAK1 occurred in the presence and absence of Avr2, indicating that Avr2 does not affect BAK1 function or PRR complex formation. Bimolecular fluorescence complementation assays showed that Avr2 and BIK1 co‐localize in planta. Although Avr2 did not affect flg22‐induced BIK1 phosphorylation, mono‐ubiquitination was compromised. Furthermore, Avr2 affected BIK1 abundance and shifted its localization from nucleocytoplasmic to the cell periphery/plasma membrane. Together, these data imply that Avr2 may retain BIK1 at the plasma membrane, thereby suppressing its ability to activate immune signalling. Because mono‐ubiquitination of BIK1 is required for its internalization, interference with this process by Avr2 could provide a mechanistic explanation for the compromised BIK1 mobility upon flg22 treatment. The identification of BIK1 as an effector target of a root‐invading vascular pathogen identifies this kinase as a conserved signalling component for both root and shoot immunity.


| INTRODUC TI ON
Plant roots lack the outermost cuticle layer that is present in aboveground structures, that is, leaves or stems.Without this protective barrier, the outer cell layers of the roots are more easily accessible to microbes (De Coninck et al., 2015).Because the microbe-rich soil environment of roots is composed of both beneficial and pathogenic microbes, the activation of root defence responses must be able to distinguish friend from foe (Hacquard et al., 2017;Zhou et al., 2020).
The immune pathways in aboveground structures are more accessible for research and therefore more thoroughly researched than the defences present in roots (De Coninck et al., 2015).It is for this reason that our current understanding of the plant immune system is mostly based on the recognition of foliar pathogens (Couto & Zipfel, 2016;Saijo et al., 2018;Tang et al., 2017).Therefore, it is important to expand our knowledge on resistance mechanisms acting in root tissues and study how root-infecting pathogens manipulate root-based immunity.
Plants have evolved a two-layered immune system to detect biotic threats (e.g., bacteria, fungi, oomycetes) and prevent disease.
The second layer of plant immunity relies on recognition of pathogen-produced virulence factors, called effectors, by either intracellular nucleotide-binding leucine-rich repeat (NB-LRR) or cell-surface receptor-like kinase (RLK)/receptor-like protein (RLP)-type receptors.
Effector recognition by these resistance (R) proteins induces a strong defence response, often culminating in localized cell death and hence referred to as the hypersensitive response (HR) (Jones & Dangl, 2006).
Elucidating the host processes targeted by effector proteins is important to understand the virulence strategy of a pathogen and the basis of host susceptibility.Bacterial effectors are known to manipulate PTI at different signalling levels, that is, at PRRs, co-receptors, RLCKs, or further downstream, targeting the MAPK pathways or WRKY transcription factors (Dou & Zhou, 2012).Pseudomonas syringae effector AvrPto directly targets the PRR complexes FLS2, EFR, and BAK1, blocking PTI probably through inhibition of the PRR kinase activity and complex formation (Shan et al., 2008;Xiang et al., 2008;Xing et al., 2007).Another P. syringae effector, AvrPtoB, interacts with co-receptor BAK1 through its kinase binding domain (Cheng et al., 2011).Multiple members of the PBL family of RLCKs (including BIK1) are targeted by P. syringae effector AvrPphB via protease cleavage (Zhang et al., 2010).The MAPK pathway is a more downstream signalling process that is often targeted.For example, P. syringae effector HopAI1 is a phosphothreonine lyase that dephosphorylates MAP KINASE KINASE 5 (MKK5), disrupting the MAPK cascade (Wang et al., 2010).Fungal effectors tend to employ different immune disruption tactics (e.g., interference with PAMP perception, manipulation of metabolic processes or transcriptional regulators) (Buscaill & van der Hoorn, 2021;Djamei et al., 2011;Lo Presti et al., 2015;Tanaka et al., 2014).Magnaporthe oryzae effector NIS1 is one of the few examples of a fungal effector suppressing PTI by targeting both BAK1 and BIK1 (Irieda et al., 2019).Currently, it is poorly understood whether root-invading pathogens use similar mechanisms to suppress plant defence as those employed by their foliar counterparts.
The soilborne pathogen Fusarium oxysporum is a widespread root colonizer.Colonization of the root surface and cortex is often symptomless, but pathogenic isolates can invade the vasculature.The latter eventually leads to blockage of the xylem vessels, causing vascular wilt disease in a wide variety of plant species (Michielse & Rep, 2009).
Each F. oxysporum forma specialis (f.sp.) has its unique host.F. oxysporum f. sp.lycopersici (Fol) infects tomato; it directs itself toward the root surface via peroxidase sensing, where it enters the root system through cracks and wounds (Nordzieke et al., 2019).From the point of entry, fungal hyphae spread throughout the apoplast to eventually colonize the vasculature (de Lamo & Takken, 2020).Fol effector proteins are secreted in apoplastic spaces of the root cortex and in the xylem sap during infection.Fol-secreted effector proteins were originally identified as Secreted In Xylem (Six) proteins.Four out of 14 of these Six proteins (Six1, Avr2 [Six3], Six5, and Six6) have been shown to be required for full Fol pathogenicity (Gawehns et al., 2014;Houterman et al., 2009;Ma et al., 2015;Rep et al., 2004).Six3 was renamed Avr2 as the effector is recognized by the NB-LRR R protein I-2 (Houterman et al., 2009).As a virulence factor, Avr2 suppresses several PTI responses, including ROS accumulation, callose deposition and MAPK activation, on flg22, chitin, chitosan and NLP (necrosis and ethylene-inducing protein; nlp24) treatment (Coleman et al., 2021;de Lamo et al., 2021;Di, Cao, et al., 2017;Tintor et al., 2020).However, it is currently unknown which PTI signalling protein(s) are targeted by Avr2 to suppress these defence responses.The crystal structure of Avr2 shows a β-barrel conformation that shares structural homology with ToxA from Pyrenophora tritici-repentis and with Tumour necrosis factor Receptor Associated Factor (TRAF) domain-containing proteins (Di, Cao, et al., 2017).TRAF proteins, often acting as cytosolic adaptor proteins in mammals, regulate NLR turnover in A. thaliana by interacting with E3 ligase complexes modulating substrate ubiquitination (Huang et al., 2016).Because several PTI outputs, triggered by highly diverse PAMPs, are affected by Avr2, it seems unlikely that a (single) PRR would be an Avr2 target.Of note, an Arabidopsis thaliana bak1-5 mutant shows a reduction in flg22-and elf18-induced ROS accumulation, MAPK activity and defence gene expression, while retaining wild-type-like development and morphology (Schwessinger et al., 2011).The A. thaliana loss-of-function bik1 mutant shows a reduction in flg22-, elf18-, and chitin-induced ROS accumulation, callose deposition, and a reduced growth phenotype (Zhang et al., 2010), mimicking the phenotype of transgenic Avr2 plants (Di, Cao, et al., 2017).The overlap between defence outputs altered by Avr2 and bak1 and bik1 mutants suggests BAK1 and/or BIK1 as candidates for Avr2-mediated PTI suppression.
We therefore investigated whether co-receptor BAK1 and/or RLCK BIK1 are affected by Avr2, by studying (i) formation of the FLS2-BAK1-BIK1 complex on flg22 application in the absence and presence of Avr2, (ii) the protein accumulation, phosphorylation, and ubiquitination state of these PTI signalling proteins, (iii) the proximity and potential interaction between Avr2 and BIK1, and (iv) the intracellular localization pattern of BIK1 in the presence of Avr2 wild-type or loss-of-virulence mutants.We conclude that Avr2 interferes with PAMP-induced mono-ubiquitination of BIK1 and observe retention of this RLCK at the PM corresponding with its compromised ability to induce PTI responses.

| Avr2 does not affect heterodimerization of the FLS2-BAK1 immune signalling complex on flg22 application
Flg22-triggered FLS2 signalling is suppressed by Avr2 (Di, Cao, et al., 2017).Because the binding of flg22 by FLS2 prompts the recruitment of BAK1 (Chinchilla et al., 2007;Heese et al., 2007;Schulze et al., 2010;Sun et al., 2022), we investigated whether this recruitment process is disrupted in the presence of Avr2.Immunoprecipitation assays were performed on protein extracts isolated from seedlings of wild-type (Col-0) or transgenic ΔspAvr2 A. thaliana.To ensure a cytosolic localization of the effector protein, the region encoding its signal peptide was omitted (Δsp).Accumulation of the endogenous FLS2 and BAK1 proteins was confirmed in protein extracts from both Col-0 and ΔspAvr2 plants using anti-FLS2 and anti-BAK1 antibodies, respectively (Figure 1, input).As expected, HA-tagged Avr2 was detected solely in ΔspAvr2 plants.On anti-FLS2 immunoprecipitation, BAK1 co-precipitation was observed on flg22 treatment in both Col-0 and ΔspAvr2 plants.No co-precipitation was observed in the mock treatment (Figure 1, IP).These data show that heterodimerization of FLS2 and BAK1 was not impaired by Avr2.Avr2-HA did not co-precipitate with the FLS2-BAK1 complex, suggesting that it does not directly interact with these proteins.
To study whether Avr2 co-localizes with BIK1, bimolecular fluorescence complementation (BiFC) assays were performed to assess, in planta, the potential proximity of the two proteins.Constructs were generated encoding A. thaliana BIK1 C-terminally fused to either the N-terminal half or the C-terminal half of the VENUS fluorescent protein (VYN or VYC) (Gehl et al., 2009).In addition, two constructs were generated encoding ΔspAvr2 N-terminally fused to either the N-or C-terminal half of the SUPERCYAN fluorophore (SCYN or SCYC) (Gehl et al., 2009) Although both proteins were readily detectable in the input material and Avr2 could be successfully pulled down using the HA tag, no co-immunoprecipitation of HA-BIK1 was observed (Figure S2).
Taken together, the fluorescence complementation indicates proximity of the Avr2 and BIK1 proteins mostly at the cell periphery.

| Avr2 mutants that do not suppress PTI remain co-localized with BIK1
Two mutations (T53R and T145K) in Avr2 have been identified that compromise the ability of the effector protein to suppress PTI but retain its I-2-mediated recognition (Di, Cao, et al., 2017).We tested whether the inability of the double and single Avr2 mutants to suppress PTI correlates with an altered co-localization with BIK1.To study the interaction between BIK1 and the Avr2 mutants (Avr2T53R, Avr2T145K, and Avr2T145K/T53R) BiFC assays were performed.The three Avr2 variants were fused N-terminally to SCYC (Gehl et al., 2009)

| Avr2 differentially affects BIK1 accumulation in N. benthamiana and A. thaliana
Because Avr2 co-localizes with BIK1, it is conceivable that Avr2, like other effectors (He et al., 2020), affects BIK1 accumulation.To investigate whether Avr2 targets BIK1 for degradation, we assessed the effect of Avr2 on BIK1 protein accumulation in N. benthamiana.
Binary constructs containing either ΔspAVR2 or BIK1-GFP were (co-) expressed via agro-infiltration in N. benthamiana.Proteins were isolated from leaf disks harvested from infiltrated sectors.Immunoblots probed with anti-BIK1 or anti-GFP showed bands corresponding to BIK1-GFP in both the absence and the presence of Avr2 (Figure 4a, lanes 2 and 3).Quantification of the intensity of the BIK1 bands was determined for three to five independent experiments using an ImageJ analysis tool.Notably, in the presence of Avr2, accumulation of BIK1 increased by approximately 4-fold as compared to the mock control (Figure 4b, bars 1 and 2).On flg22 recognition, the FLS2-BAK1 complex activates BIK1 through phosphorylation and ubiquitination, facilitating its release from the membrane-localized complex into the cytosol.To study whether Avr2 also affects accumulation of activated BIK1, a 10-min flg22 treatment was performed prior to leaf disk harvesting of AVR2 and/or BIK1-GFP agro-infiltrated leaves.Using anti-BIK1 and anti-GFP immunoblots, an approximately 3-fold increase in BIK1-GFP protein abundance was observed in agro-infiltrated leaf disks after flg22 treatment (Figure 4a, lanes 2 and 5; Figure 4b, bars 1 and 3).
As the increased signal was detectable within 10 min after flg22 application, it is unlikely to be explained by up-regulation of BIK1 transcription and/or a reduced turnover of the protein.Of note, in the presence of Avr2, irrespective of flg22 treatment, no major differences in BIK1 signals on the immunoblot were observed (Figure 4a, lanes 3 and 6; Figure 4b, bars 2 and 4).
The effect of Avr2 on steady-state BIK1 protein accumulation in A. thaliana leaves was assessed in wild-type Col-0 and transgenic ΔspAVR2 lines.As expected, the immunoblots probed with anti-Avr2 only revealed an Avr2-specific band (c.12 kDa) in the ΔspAVR2 lines (Figure 4c).Immunoblots probed with anti-BIK1 showed bands corresponding to endogenous BIK1 (c.44 kDa) in both Col-0 and Avr2 lines.However, a reduction in BIK1 accumulation of 0.67, 0.64, and 0.66 of wild-type Bik1 was observed in the three ΔspAVR2 A. thaliana lines as compared to the wild-type progenitor (Figure 4c).
In conclusion, Avr2 differentially affects BIK1 abundance in N. benthamiana and A. thaliana.

| Mono-ubiquitination of BIK1 is affected by Avr2
As protein accumulation of BIK1 was affected by the presence of Avr2, we set out to investigate whether Avr2 affects phosphorylation of BIK1.Transient protoplast expression assays were performed followed by immunoblot detection.HA-tagged BIK1 (BIK1-HA), and FLAG-tagged ΔspAvr2 (ΔspAvr2-FLAG) were expressed in A. thaliana Col-0 protoplasts, followed by flg22 treatment for 0, 10, 20 or 30 min.At t = 0 BIK1-HA was solely detected in an unphosphorylated form, as a single band of approximately 44 kDa was observed on an anti-HA probed immunoblot (Figure 5a).On flg22 treatment, a second band of slightly higher molecular weight was observed at time points 10, 20, and 30 min, indicating an equilibrium shift toward phosphorylated BIK1 (pBIK1-HA) (Figure 5a).In the presence of Avr2-FLAG, which was detected using an anti-FLAG immunoblot, a similar shift from unphosphorylated BIK1 toward phosphorylated BIK1 on flg22 treatment was observed, demonstrating that phosphorylation of BIK1 was unaffected by Avr2.
To investigate whether mono-ubiquitination of BIK1 was affected by Avr2, in vivo ubiquitination assays were performed followed by co-immunoprecipitation.Quantification of the intensity of the BIK1 bands was determined of three independent experiments using an ImageJ analysis tool (Figure 5c).A FLAG-tagged ubiquitin construct (FLAG-UBQ) and BIK1-HA were expressed with or without ΔspAVR2-GFP in A. thaliana Col-0 protoplasts.On flg22 treatment, ubiquitinated proteins were pulled down with anti-FLAG beads and the ubiquitinated BIK1-HA (c.52 kDa) was detected on an immunoblot probed with an anti-HA antibody (Figure 5b,c).As reported previously the amount of mono-ubiquitinylated BIK1 in the Col-0 protoplasts increased on flg22 treatment (Ma et al., 2020; Figure 5c).In the presence of ΔspAvr2-GFP, mono-ubiquitinated BIK1 could still be detected, but the bands were of lower intensity than those in the empty vector control and did not increase on flg22 treatment (Figure 5b,c).We conclude that Avr2 does not affect flg22-induced phosphorylation of BIK1 but appears to interfere with its mono-ubiquitination.

| Avr2 alters the subcellular localization pattern of BIK1
The localization of BIK1 in different subcellular compartments (i.e., the cytosol and nucleus) is important for the activation of defence responses (Kadota et al., 2014;Lal et al., 2018;Li et al., 2014).To Total protein was isolated from leaf disks and visualized using immunoblot analysis.Avr2 protein levels were visualized using an anti-Avr2 antibody (middle panel).BIK1 proteins levels were visualized using an anti-BIK1 and anti-GFP antibody (top and middle panels).Equal protein loading was verified by Coomassie brilliant blue (CBB) staining of the blot (bottom panel).Molecular weight markers are indicated on the left.(b).Quantification of BIK1 protein accumulation (depicted as ratios) by measuring band intensity of anti-BIK1 immunoblot as area under the curve for three to five independent experiments using ImageJ.(c).Immunoblot depicting BIK1 protein accumulation in Arabidopsis thaliana Col-0 and transgenic ΔspAVR2.Proteins were isolated from leaves and detected using immunoblot analysis.Avr2 protein levels were visualized using an anti-Avr2 antibody (middle panel).BIK1 proteins levels were visualized using an anti-BIK1 (top panel).Equal protein loading was verified by CBB staining of the blot (bottom panel).Molecular weight markers are indicated on the left.monitor whether Avr2 affects the subcellular distribution of BIK1, a binary vector construct was generated encoding the A. thaliana BIK1 protein C-terminally fused to GFP. (Co-)expression of binary vectors carrying either BIK1-GFP and/or ΔspAvr2 through agro-infiltration in N. benthamiana leaves was performed.The expected size of BIK1-GFP (c.71 kDa) as well as the integrity of the fusion protein was confirmed using immunoblot analysis.Only minimal degradation (i.e., free GFP) was apparent, and a stronger BIK1 signal was observed in the presence of Avr2 (Figure S3).The fluorescence signal, depicting the subcellular localization pattern of BIK1-GFP, was analysed using confocal microscopy.In the absence of Avr2, green fluorescence was observed in the nucleus, cytosolic strands, and at the cell periphery (Figure 6, top panels).However, in the presence of Avr2, green fluorescence was found predominantly at the cell periphery and not in the nucleus nor in cytosolic strands (Figure 6, bottom panels).We conclude that the subcellular localization pattern of BIK1-GFP is altered in the presence of Avr2 such that the BIK1 protein is mostly confined to the cell periphery in the presence of the effector.
To assess the effect of the loss-of-virulence Avr2 mutants

| DISCUSS ION
This study investigates how Avr2 from the tomato root-and xylemcolonizing strain of F. oxysporum suppresses immune signalling induced by a multitude of PAMPs.We show that in the presence of flg22 and Avr2 (i) the PRR FLS2 still dimerizes with its co-receptor BAK1, implying that PAMP perception and the initial signalling steps are unaffected by Avr2, (ii) BIK1 is phosphorylated, demonstrating that the FLS2-BAK1 complex is still functional, but (iii) mono-ubiquitination of BIK1 is reduced and its accumulation and subcellular localization altered, suggesting retention of BIK1 at the PM as a possible explanation for the impaired signalling ability of this RLCK.
RLCKs are common targets for bacterial effectors to disrupt PTI signalling.For example, P. syringae effector AvrPphB, a  (Di, Cao, et al., 2017) that are known to interact with E3 ubiquitin ligases important for the addition of ubiquitin.Ubiquitination targets proteins for degradation or shifts their subcellular localization (Gao et al., 2022).Given this structural homology between Avr2 and TRAF domain-containing proteins, we hypothesize that Avr2 could target BIK1 function by interfering with its mono-ubiquitination process to retain BIK1 at the PM.
The immune-regulatory function of BIK1 depends on both its nuclear and cytosolic localization.Nuclear localization of BIK1 is required for activation of defence gene expression, for example in A. thaliana by phosphorylation of WRKY transcription factors (WRKY 33, WRKY50, and WKRY57) that are involved in regulation of the defence hormones salicylic acid and jasmonic acid (Lal et al., 2018).A cytoplasmic localization of BIK1, on dissociation from the immune receptor complex at the PM, is required, for example, for initiation of ROS signalling through phosphorylation of the PM-localized A. thaliana RBOHD complex (Kadota et al., 2014;Li et al., 2014).In Avr2expressing tomato and A. thaliana plants the ROS burst is decreased on treatment with flg22, chitosan, or nlp24 (Di, Cao, et al., 2017;Tintor et al., 2020), consistent with the hypothesis that BIK1 can no longer activate the RBOHD complex.Notably, in the presence of Avr2 the localization pattern of BIK1-GFP shifted from a nucleocytoplasmic localization to the PM/cell periphery.In the presence of Avr2 the BIK1-GFP signal was no longer observed in the nucleus nor in cytosolic strands.An impaired ability of BIK1 to translocate to the nucleus to activate transcription factors corresponds with the reduced induction of PTI responsive genes in an Avr2 transgenic tomato plant (Di, Gomila, et al., 2017).The Avr2 T53R and T53R/T145K mutants did permit BIK1 to enter the nucleus, consistent with the inability of these Avr2 variants to suppress PTI responses on PAMP treatment.Surprisingly, in the presence of the Avr2 T145K mutant, no nuclear entry of BIK1 was observed while this effector mutant is compromised in its PTI suppressing activity.It is tempting to speculate that a minor fraction of BIK1, below the detection threshold of our experimental setup, regained mobility sufficient to restore its signalling functions in the cytoplasm and nucleus.Taken together, these findings support a mechanism by which Avr2 prevents dissociation of (activated) BIK1 from the PM to allow activation of RBHOD to induce ROS production and defence gene expression.
Release of BIK1 from the PM requires several actions, starting with PAMP recognition by the FLS2-BAK1 complex followed by trans-phosphorylation of BIK1 by BAK1 and subsequent monoubiquitination of BIK1 by E3 ligases RHA3A/B (Ma et al., 2020).
As heterodimerization of the FLS2-BAK1 complex was unaffected in Avr2-overexpressing A. thaliana, the posttranslational modification of BIK1 seems a more likely process to be targeted by Avr2.Phosphorylation of BIK1 was still observed in the presence of Avr2, indicating that phosphorylation sites of BIK1 were readily accessible and functional.Mono-ubiquitination of BIK1, however, was found to be reduced by Avr2, although not completely abolished.
A pool of mono-ubiquitinated BIK1 was still detected in the presence of Avr2, which is in line with Avr2 not being able to completely block PTI defence responses (Di, Cao, et al., 2017).Because monoubiquitination of BIK1 is required for its release from the membrane complex, interference with this process by Avr2 would provide an explanation for the observed BIK1 retention at the cell periphery/ PM.The mechanism by which Avr2 differentially affects BIK1 accumulation in A. thaliana and N. benthamiana is unclear; possibly, retention of the kinase at the PM affects endocytosis and protein turnover.Whether the observed difference is dependent on the plant species or the expression method awaits generation of stable transgenic N. benthamiana plants, and thus remains a question for future research.
How Avr2 may interfere with mono-ubiquitination of BIK1 is unknown.Xanthomonas campestris pv.campestris type III effector AvrAC is a uridylyl transferase that targets two RLCKs, BIK1 and RIPK.By adding uridine 5′ monophosphate, conserved phosphorylation sites in the activation loops of BIK1 and RIPK are concealed, thereby reducing the kinase activity (Feng et al., 2012).Perhaps Avr2, which shares structural homology with TRAFdomain-containing proteins, masks the ubiquitination site of BIK1, thereby concealing it from the RHA3A/B E3 ligases.A possible competition of Avr2 with RHA3A/B for this binding site agrees with the incomplete repression of BIK1 mono-ubiquitination and partial PTI suppression by Avr2.
Avr2 point mutations T145K or T53R (located on opposite interfaces of Avr2 protein) confer loss of Avr2-mediated PTI suppression (Di, Cao, et al., 2017).Interestingly, these Avr2 variants were still able to reconstitute fluorescence in BiFC assays with BIK1, implying their proximity to BIK1.Together with the compromised ability of the mutants to dimerize, it is tempting to speculate that the Avr2 dimerization interface is required for its PTI suppressing activity.Possibly this Avr2 interface is involved in recruitment of additional proteins such as the E3 ligases RHA3A/B.X. campestris pv.vesicatoria effector AvrBs3 homodimerizes prior to nuclear import.A specific repeat region in this effector is essential for virulence and self-interaction (Gürlebeck et al., 2005).Phytophthora sojae effector PsCRN63 also forms homo-and heterodimers and dimerization is required for PTI suppression activities (Li et al., 2016).Likewise, dimerization of Avr2 seems to be important for the activity of Avr2 regarding PTI suppression.
Hence, it is conceivable that Avr2 targets the function of multiple RLCKs involved in PTI signalling.
In conclusion, the effector Avr2 of the tomato root-and xylemcolonizing strain of F. oxysporum interferes with PAMP-induced mono-ubiquitination of BIK1 and causes retention of this RLCK at the PM, compromising its ability to induce PTI responses.Although effector-mediated PTI suppression has been well established for foliar pathogens, this report of an effector of a root-colonizing pathogen that suppresses PTI through targeting a conserved signal transduction component is a novel find.The manipulation of immunity through PTI suppression is seemingly of importance to rootinfecting pathogens as well.Root-based PTI defence responses are spatially restricted to tissues surrounding the vasculature (Emonet et al., 2021).It is therefore noteworthy that Avr2 can move via the symplast (Blekemolen et al., 2022;Cao et al., 2018), allowing it to suppress PTI in distal uninfected cells prior to fungal colonization.It will be interesting to investigate whether this is a shared feature for PTI-suppressing effectors from root-colonizing pathogens.

| Generation of BIK1, Avr2 point mutations, and BiFC constructs
To generate pDONR:BIK1 without a stop codon, PCR was performed on the A. thaliana coding sequence of BIK1 (geneID: AT2G39660) using the following primer set: FP9732 (5′-ACAAG TTT GTA CA A A A A AGC AGG C TC CAT GGG T TC T TGC T TCAGT TC-3′)/ FP9734 (5′-ACTTT GTA CAA GAG AAA GCT GGG TCC ACA AGG TGC CTGCCAAAAG-3′).pDONR:BIK1 without a stop codon was subsequently used for Gateway cloning (Invitrogen) together with pGWB451 to generate pGWB451:BIK1, via an LR reaction.

| Confocal microscopy
Confocal microscopy was performed with a Nikon A1 microscope (Nikon Instruments Inc.).Excitation of GFP and split SUPERCYAN/ VENUS fluorophores was carried out at 488 nm with an Ar-ion laser and the signal emitted was selected using a 500-525/550 nm bandpass filter.SUPERCYAN and DAPI were excited at 405 nm with a diode laser and light signal emitted was selected with a 445/480-515 nm bandpass filter.DAPI staining was performed 1 h before imaging by syringe-infiltration of the DAPI solution into the previously agro-infiltrated leaf sectors.
Seven-week-old snap-frozen A. thaliana leaves were ground using a TissueLyser II (Qiagen) and 25 mg of powder was resuspended in 100 μL of extraction buffer (50 mM Tris-HCl pH 7.5, 2% SDS, 5 mM DTT and 1× protease inhibitor cocktail; Roche), and centrifuged at 4°C at 16,000 g for 25 min.
Proteins were separated by SDS-PAGE (Bio-Rad) using either 10% or 14% acrylamide gels and subsequently blotted on polyvinylidenedifluoride (PVDF) membranes using the semidry blotting method (Thermo Scientific Owl HEP-1 system).Blots were probed with rabbit or rat monoclonal Avr2- (Ma et al., 2015), GFP-, or BIK1-antibodies (Chromotek; Agrisera) at a dilution of 1:2500 or 1:5000.Secondary goat-anti-rabbit or goat-anti-rat antibody (Pierce) was used at a dilution of 1:5000.The signal was visualized with an ECL kit (GE Healthcare, ECL prime of Thermo Scientific, Super Signal West Pico) according to the manufacturer's instructions and detected using a ChemiDoc MP imaging system (Bio-Rad).
For testing FLS2-BAK1 complex formation in different A. thaliana For co-immunoprepcipitation assays, BIK-HA was co-transfected with the empty vector as control or with ΔspAVR2-FLAG construct into protoplasts and then incubated at room temperature for 12 h.
After treatment with 200 nM flg22 for the indicated time, protoplasts were collected by centrifugation at 100 g for 2 min and lysed in 300 μL of IP buffer (20 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, 10% vol/vol glycerol, 0.5% vol/vol Triton X-100, and protease inhibitor cocktail; Sigma) by vortexing.After centrifugation at . Co-expression of the BiFC constructs in Nicotiana benthamiana leaves was performed via agro-infiltration and blue/green fluorescence, indicating potential protein-protein interactions, was analysed using confocal microscopy.Given the ability of Avr2 to form homodimers, SCYN-ΔspAvr2 and SCYC-ΔspAvr2 were co-expressed as positive controls.As expected, a strong blue nucleocytoplasmic fluorescence was observed for the latter constructs, confirming Avr2 dimerization (Figure 2, left panel).In cells co-expressing SCYN-ΔspAvr2 and BIK1-VYC no green fluorescence could be detected (Figure 2, middle panel).However, the SCYC-ΔspAvr2 and BIK1VYN combination did result in a brightgreen fluorescent signal.This signal was mostly located at the cell periphery but was also weakly visible in the nucleus (Figure 2, right panel).Notably, fluorescence originating from the proposed SCYC-ΔspAvr2 and BIK1-VYN interaction was not observed in cytosolic strands, as opposed to the SCYN-ΔspAvr2 and SCYC-ΔspAvr2 interaction, where cytosolic strands were clearly detectible (Figure 2, left and right panels, indicated by arrows; Figure S1).To investigate whether the proteins associate, co-immunoprecipitation assays were performed in A. thaliana Col-0 protoplast transfected with constructs encoding HA-tagged BIK1 and FLAG-tagged ΔspAvr2.
to create SCYC-ΔspAvr2T53, SCYC-ΔspAvr2T145K, and SCYC-ΔspAvr2T145K/T53R, respectively.The SCYC-ΔspAvr2 mutant and SCYC-ΔspAvr2 wild-type BiFC constructs were co-expressed with BIK1-VYN or SCYN-ΔspAvr2 in N. benthamiana leaves via agro-infiltration.Fluorescence, indicating potential protein-protein interactions, was analysed using confocal microscopy.As before, strong green fluorescence at the cell periphery and a weak nuclear signal was observed in cells expressing BIK1-VYN and SCYC-ΔspAvr2 (Figure 3, upper left panel), reaffirming the proximity between Avr2 and BIK1.Of note, co-expression of BIK1-VYN with SCYC-ΔspAvr2T53, SCYC-ΔspAvr2T145K, or SCYC-ΔspAvr2T145K/ T53R similarly resulted in green fluorescence at the cell periphery and a weak nuclear signal (Figure 3, upper middle/right panels).These data indicate proximity between the Avr2 mutants and BIK1.Additionally, we monitored the ability of wild-type Avr2 to dimerize with Avr2 mutants.In cells co-expressing wild-type SCYC-ΔspAvr2 and SCYN-ΔspAvr2, strong blue fluorescence was observed in the nucleus, in cytosolic strands (indicated by arrows), and at the cell periphery (Figure 3, lower left panel), confirming Avr2 dimerization.However, on co-expression of SCYN-ΔspAvr2 with SCYC-ΔspAvr2T53, SCYC-ΔspAvr2T145K, or SCYC-ΔspAvr2T145K/T53R a F I G U R E 2 Avr2 and BIK1 co-localize in planta.Confocal microscopy images of agro-infiltrated Nicotiana benthamiana leaves transiently expressing bimolecular fluorescence complementation constructs (SCYC-ΔspAVR2 and SCYN-ΔspAVR2, BIK1-VYN and SCYC-ΔspAVR2, or BIK1-VYC and SCYN-ΔspAVR2).Cyan and green fluorescence represent protein-protein interactions visualized by complementation of the fluorescent protein halves.Green and blue fluorescence was analysed 48 h after agro-infiltration by confocal microscopy.Three biological replicates were performed.Scale bars represent 40 or 99 μm.much weaker cytosolic fluorescence was observed, in addition to relatively large fluorescent punctate structures.These structures were located mostly at the cell periphery and their heterogenous size implies that they are aggregates (Figure 3, lower middle/right panels).Altogether, the point mutations in Avr2 did not result in loss of co-localization with BIK1 but they did disrupt Avr2 homodimerization, resulting in the formation of protein aggregates.

F
I G U R E 4 BOTRYTIS-INDUCED KINASE 1 (BIK1) accumulation is affected by the presence of Avr2.(a) Immunoblot depicting BIK1 protein accumulation in the presence or absence of Avr2 in Nicotiana benthamiana.BIK1-GFP and/or ΔspAVR2 were (co-)expressed in N. benthamiana via agro-infiltration and after 48 h the infiltrated leaf sectors were treated with either water or flg22 (100 nM) for 10 min.

(
Avr2T53R, Avr2T145K, and Avr2T145K/T53R) on the subcellular localization pattern of BIK1, we co-infiltrated binary vectors containing either BIK1-GFP and/or ΔspAvr2, ΔspAvr2T53R, ΔspAvr2T145K, or ΔspAvr2T145K/T53R in N. benthamiana leaves.A DAPI staining was included to accentuate the nuclei.As previously observed, in the absence of Avr2, green fluorescence originating from BIK1-GFP was observed in the nucleus, cytosolic strands, and at the cell periphery (Figure 7, first row, indicated by stars and arrowheads, respectively).In the presence of Avr2, green fluorescence was found predominantly at the cell periphery and not in the nucleus nor in cytosolic strands (Figure 7, second row).Avr2T145K showed a BIK1-GFP cell peripheral localization pattern like that of wild-type Avr2 (Figure 7, third row).Interestingly, Avr2T53R and Avr2T145K/T53R permitted BIK1 mobility based on the weak green fluorescent signal in the nucleus (Figure 7, fourth and fifth rows).

F
I G U R E 5 Mono-ubiquitination of BOTRYTIS-INDUCED KINASE 1 (BIK1) is reduced by Avr2.(a) Immunoblot depicting unphosphorylated and phosphorylated BIK1 accumulation in the presence or absence of Avr2 following flg22 treatment in Arabidopsis thaliana Col-0 protoplasts.BIK1-HA and/or ΔspAVR2 were (co-)expressed in A. thaliana Col-0 protoplasts, followed by treatment with flg22 (200 nM) for 0, 5, 10, 20, or 30 min.Avr2 and BIK1 protein levels were visualized using an anti-FLAG and anti-HA antibodies, respectively.Equal protein loading was verified by Coomassie brilliant blue (CBB) staining of the blot.Molecular weight markers are indicated on the left.(b) Immunoblot depicting mono-ubiquitinated BIK1 accumulation in presence or absence of Avr2 following flg22 treatment in A. thaliana Col-0 protoplasts.BIK1-HA, FLAG-UBQ, and/or ΔspAVR2-GFP were co-expressed in wild-type protoplasts, followed by treatment with flg22 (200 nM) for 0, 5, 10, 20, or 30 min.After immunoprecipitation (IP) with anti-FLAG agarose beads, ubiquitinated BIK1 was detected by immunoblot using anti-HA antibodies.Equal protein loading was verified by CBB staining of the blot.Molecular weight markers are indicated on the left.(c) Quantification of Ub-BIK1 protein accumulation (depicted as ratios) by measuring band intensity of anti-HA immunoblot as area under the curve for three independent experiments using ImageJ.cysteine protease, cleaves several RLCKs (PBS1, BIK1, PBL1, and PBL2) (Zhang et al., 2010).The structure of Avr2 resembles that of the TRAFdomain-containing proteins

F
Avr2 alters the subcellular localization pattern of BOTRYTIS-INDUCED KINASE 1 (BIK1).Confocal microscopy images of agro-infiltrated Nicotiana benthamiana leaves transiently expressing BIK1-GFP in the absence or presence of ΔspAVR2.Asterisks mark green (GFP) fluorescent signals located in the nucleus; arrowheads indicate green fluorescent signal located in cytosolic strands.Fluorescence was visualized 48 h after agro-infiltration using confocal microscopy.Maximum projections of z-stack images are depicted.Three biological replicates were performed, representative examples are shown.Scale bars represent 66 μm.F I G U R E 7 The subcellular localization of BOTRYTIS-INDUCED KINASE 1 (BIK1) is altered by the Avr2 variants.Confocal microscopy images of agro-infiltrated Nicotiana benthamiana leaves transiently expressing BIK1-GFP in the absence or presence of ΔspAVR2or ΔspAVR2T145K, ΔspAVR2T53R, ΔspAVR2T145K/T53R variants.Asterisks mark green (GFP) or blue (DAPI)-fluorescent signals located in the nucleus; arrowheads indicate green fluorescent signal located in cytosolic strands.Fluorescence was visualized 48 h after agro-infiltration using confocal microscopy.Maximum projections of z-stack images are depicted.Three biological replicates were performed, representative figures are shown.Scale bars represent 28 μm.
genotypes, seedlings were sterilized and sown on ½ × MS (Duchefa Biochemie) agar plates and grown for 4 days (16 h light, 8 h dark).Subsequently, 15-20 seedlings were transferred per well of a sixwell plate containing ½ × MS liquid medium and grown for another 10 days.One day before treatment the seedlings were transferred into a glass beaker containing sterile water.The next day, flg22 was added to a final concentration of 100 nM and incubated for 10 min before harvesting seedlings by flash freezing in liquid nitrogen.Proteins were isolated in 50 mM Tris-HCl pH 7.5, 150 mM NaCl, 10% glycerol, 5 mM DTT, 1% protease inhibitor cocktail (Sigma Aldrich), 2 mM Na 2 MoO 4 , 2.5 mM NaF, 1.5 mM activated Na 3 VO 4 , 1 mM phenylmethanesulfonyl fluoride [PMSF] and 1% IGEPAL.For immunoprecipitations α-rabbit Trueblot agarose beads (eBioscience) coupled with α-FLS2 antibodies were used and incubated with the crude extract for 2-3 h at 4°C.Subsequently, beads were washed three times with wash buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM PMSF, 0.5% IGEPAL) before adding Laemmli sample buffer and incubating for 10 min at 95°C.Analysis was carried out by SDS-PAGE and western blots using α-FLS2 α-BAK1(Chinchilla et al., 2007) and α-HA (clone 3F10; Roche) antibodies.BIK1-HA and empty vector or ΔspAVR2-FLAG were transfected into protoplasts that were subsequently incubated at room temperature for 12 h.After treatment with 200 nM flg22 for the indicated time, protoplasts were collected by centrifugation at 100 g for 2 min and lysed in 4 × SDS loading buffer (250 mM Tris-HCl, pH 6.8, 40% vol/vol glycerol, 4% wt/vol SDS, 0.1% wt/vol bromophenol, and 4% vol/vol β-mercaptoethanol) by vortexing.BIK1 phosphorylation was detected by α-HA antibody (1:2000 dilution; Roche) after protein separation on a 10% SDS-PAGE gel.For ubiquitination assays, FLAG-tagged UBQ (FLAG-UBQ) and BIK-HA were co-transfected with the empty vector as control or with ΔspAVR2-GFP construct into protoplasts and then incubated at room temperature for 12 h.