rpfF is not required for X. translucens pv. undulosa pathogenesis

Bacterial cells self-coordinate via a mechanism called quorum sensing. In Xanthomonas species the gene rpfF encodes the quorum sensing autoinducer synthase. Xanthomonas species are divided into two main phylogenetic groups called Clade I and Clade II. The rpf quorum sensing system has been well studied in multiple Clade II Xanthomonas species and deletion of rpfF resulted in a major loss of virulence on susceptible hosts. However, the only Clade I Xanthomonas species in which the rpfF system was previously studied was in the sugarcane pathogen X. albilineans. In X. albilineans the rpf cluster plays a relatively small role in pathogenesis. Xanthomonas translucens pv. undulosa (Xtu) is a Clade I Xanthomonas species that causes bacterial leaf streak (BLS) and black chaff of wheat and barley and has increased as a concern in recent decades. Neither major resistance nor chemical treatments are available to prevent disease caused by Xtu. Interference with rpf bacterial quorum sensing systems has demonstrated some success in other systems. It was unknown whether BLS caused by Xtu could be prevented via quorum sensing interference. We found that Xtu encodes an rpfF homolog and we created an rpfF knockout mutant to study the role of the rpf system in Xtu. We found that the rpfF mutant was unaffected in its pathogenesis as it caused BLS symptoms and multiplied within wheat plants to the same levels as the wildtype strain. The Xtu rpfF mutant grew normally in lag and log phases in vitro, however it exhibited a shorter stationary phase and an early death phase in plant-derived media. The importance of RpfF in Xtu’s life cycle is unknown, though it appears to carry out a role in population stability. Our research determined that rpfF is not a major Xtu pathogenicity factor. Therefore, we do not recommend the targeting of the rpf quorum sensing system as a preventative treatment for BLS of wheat.


Introduction
Parasitic bacteria frequently complete complex life cycles including drastic shifts in population behavior as a response to the environment and to the bacterial population size and status (Joshi et al. 2021).During these shifts, bacterial cells self-coordinate via a mechanism called quorum sensing to change gene expression and thus behavior in a cell densitydependent manner (González and Keshavan 2006;Tomasz 1965;Whiteley et al. 2017;Nealson et al. 1970).In quorum sensing, an autoinducer molecule unique to a set of bacteria is produced and perceived by the bacterial community.The amount of autoinducer perceived by a bacterium's cognate receptor is a proxy for the quantity of like individuals.When a high level of autoinducer is present, a signaling cascade leads to changes in gene expression and a shift to high cell density behavior(s) (González and Keshavan 2006).
Phytopathogenic bacteria express diverse virulence factors during disease establishment and progression in their hosts (Benali et al. 2014;Pontes et al. 2020;Timilsina et al. 2020).Quorum sensing often regulates the expression of bacterial virulence factors, because many bacteria are required to express them simultaneously for efficacy (Joshi et al. 2021).
Xanthomonads, including phytopathogenic Xanthomonas species, use cis-2-unsaturated fatty acids called diffusible signal factors (DSFs) as a distinct family of autoinducers.Genes involved in Xanthomonas DSF production, sensing and response are located in the regulation of pathogenicity factors (rpf) cluster (Tang et al. 1991).DSFs are known to be produced via the fatty acid synthesis elongation cycle, with the thioesterase RpfF, encoded by rpfF, completing a crucial final step in the pathway (Zhou et al. 2015;Bi et al. 2014).
The loss of rpfF eliminates DSF production, preventing shifts in transcription related to DSF at high cell density (Figure 1).Therefore, rpfF deletion mutants have been used as a target to study the role of quorum sensing in Xanthomonas (Dow et al. 2003;Singh et al. 2022;Chatterjee and Sonti 2002;Rott et al. 2013;Huang et al. 2013).Xanthomonas species are divided into two main phylogenetic groups called Clade I and Clade II (Parkinson et al. 2007;Ferreira-Tonin et al. 2012;Jacques et al. 2016;Grau et al. 2016).The rpf quorum sensing system has been well studied in multiple Clade II Xanthomonas species.Deletion of rpfF resulted in a major loss of virulence on susceptible hosts for X. campestris, X. oryzae, X. citri and X. axonopodis (Dow et al. 2003;Chatterjee and Sonti 2002;Huang et al. 2013;Thowthampitak et al. 2008).DSF produced via RpfF was also found to have roles in motility and biofilm formation, however these varied depending on the species, and even strain, in question (Dow et al. 2003;Huang et al. 2013;Chatterjee and Sonti 2002).This fits with the broader Xanthomonadacae, as deletion of rpfF from Xylella fastidiosa resulted in hypervirulence in planta because X. fastidiosa has a drastically different lifestyle from characterized Xanthomonas species (Newman et al. 2004).
However, the only Clade I Xanthomonas species in which the rpfF system was previously studied was in the sugarcane pathogen X. albilineans.In X. albilineans the rpf cluster plays a relatively small role in pathogenesis (Rott et al. 2013).Virulence of an rpfF mutant was slightly, but significantly, decreased (Rott et al. 2013).Loss of rpfF also did not have an effect on X. albilineans motility (Rott et al. 2013).We do not know if the reduced importance of rpfF as the DSF synthase is unique to X. albilineans or if this is a characteristic of Clade I Xanthomonas species.X. albilineans also lacks the hrp regulon which is a link between the quorum sensing system and regulation of the Type III secretion system (T3SS) in X. campestris (Jiang et al. 2018;Pieretti et al. 2009).Other Clade I Xanthomonas species such as X. translucens, on the other hand, have an intact T3SS (Goettelmann et al. 2022).
Xanthomonas translucens pv.undulosa (Xtu) is a Clade I Xanthomonas species that causes bacterial leaf streak (BLS) and black chaff of wheat and barley (Sapkota et al. 2020).It can cause yield losses as great as 40% (Forster and Schaad 1988;Sapkota et al. 2020) and in recent decades has increased as a concern, especially in North America (Ledman et al. 2023;Tambong et al. 2021;Curland et al. 2018Curland et al. , 2020;;Hangamaisho et al. 2024).Xtu is thought to spread by wind, rain and on seeds (Sapkota et al. 2020;Ledman et al. 2023).It enters leaves via natural openings such as stomata or damage points (Sapkota et al. 2020;Heiden et al. 2023;Adhikari et al. 2012).Once it gains entry to a wheat plant, Xtu extensively colonizes the apoplastic spaces of the mesophyll tissue (Heiden et al. 2023;Gluck-Thaler et al. 2020;Bragard et al. 1997;Sapkota et al. 2020;Jones et al. 1917).Once it reaches a high population, bacterial leaf streak manifests by macroscopic watersoaking symptoms followed by chlorotic and necrotic vein-delimited symptoms which give bacterial leaf streak its name (Ledman et al. 2023;Bamberg 1936;Jones et al. 1917;Sapkota et al. 2020).Once an Xtu population reaches a high level and encounters humid conditions, it oozes back out of stomata in exudates (Bamberg 1936).These exudates dry and are easily moved via wind or rain, enabling the spread to new leaves on the same or a different plant (Jones et al. 1917).In some cases, Xtu reaches the grain of a wheat plant and causes black chaff (Heiden et al. 2023;Bamberg 1936;Jones et al. 1917).
There is increasing interest in developing treatment options to combat BLS, including efforts to develop resistant germplasm in wheat (Sapkota et al. 2018), but currently no major resistance is available for growers.There are no available chemical treatments, and antibiotics can quickly drive the development of resistance when applied in the field (Sundin and Wang 2018).Plants broadly produce compounds which interfere with pathogen quorum sensing systems as part of their immune system (Joshi et al. 2021).Interference with bacterial quorum sensing systems has demonstrated some success in other systems.For example, plant expression of rpfF causes resistance to the Xanthomonad Xylella fastidiosa (Caserta et al. 2014(Caserta et al. , 2017;;Lindow et al. 2014).However, it is currently unknown what role the rpf cluster plays in Xtu pathogenesis.
In this study, we investigated the role of the Xtu ortholog of rpfF, which encodes the critical DSF synthase in other Xanthomonas species.We tested the hypothesis that rpfF plays a similar major role in pathogenesis and plant colonization in Xtu.Specifically, we compare X. translucens to the well characterized Clade II Xanthomonas species: X. campestris.Our research focused on determining whether the rpf quorum sensing system is a main determinant of plant colonization in Xtu and therefore a desirable target for development of new strategies to prevent BLS caused by Xtu.

Materials and Methods:
Bacterial strains and culture conditions Bacterial strains were grown on nutrient agar (NA; beef extract 3g/L, peptone 5g/L, agar 15g/L), nutrient broth (NB; beef extract 3g/L, peptone 5g/L), Luria-Bertani (LB; 10g/L tryptone, 10g/L NaCl, 5g/L yeast extract) broth or Luria-Bertani agar (LBA; 10g/L tryptone, 10g/L NaCl, 5g/L yeast extract, 15g/L agar), or described media with supplemented 50µg/mL kanamycin, 40μg/mL spectinomycin, or 5μg/mL of gentamycin when applicable.Bacteria were grown for two or three days on NA plates at 28 ˚C and then an overnight growth was resuspended to the desired optical density for the experiments in this study.

Knockout mutant generation
To generate an in-frame deletion of rpfF in Xtu strain UPB513, genomic DNA was first extracted with the DNeasy Blood & Tissue Kit (QIAGEN, Cat.No. 69504) using TRIzol TM (Invitrogen, catalog #15596026) following the manufacturer's protocol with the following modification: 2mL of cell suspension with a concentration of 0.200 optical density at 600nm (O.D. 600nm ) was centrifuged and resuspended with 0.75 mL TRIzol TM .The quality of genomic DNA was ensured via running in a 1.5% agarose gel at 110 V for 40 minutes.Upstream and downstream PCR fragments containing overhang regions with each other and the plasmid vector pk18mobsacB were amplified using the following two pairs of primers: rpfF-del-F1 and rpfF-del-R1 (Downstream fragment #1) and rpfF-del-F2 and rpfF-del-R2 (Upstream fragment #2) designed from the Xtu UPB513 genome (NCBI GenBank assembly GCA_023221635.1).
The PCR products were cleaned using using the QIAquick® PCR Purification Kit (Qiagen, Cat. No. / ID: 28104) according to the manufacturer's instructions.The vector pK18mobsacB (Schäfer et al. 1994) was linearized with HindIII and cleaned using the QIAquick® PCR Purification Kit.The two fragment products of rpfF and the linearized pK18mobsacB (Schäfer et al. 1994) were then isothermally assembled using the Gibson cloning kit (NEB; Gibson et al. 2009) following the manufacturer's instructions in a 5:1 insert to backbone ratio.Primers rpfFko-conf-F and rpfF-ko-conf-R were used to confirm the construct.
E. coli RHO3 (López et al. 2009) cells were grown overnight in LB supplemented with Nsuccinyl-L-diaminopimelic acid desuccinylase and then subcultured for one hour until they reached an approximate 0.3 O.D. 600nm in 50mL of LB broth.The suspension was centrifuged at 4000 x g for five minutes and then washed with 5mL of ice cold CaCl 2 before being resuspended in ice cold 10% glycerol.5µL of Gibson assembly product containing the assembled vector was added to 150uL of the competent E. coli and incubated on ice for 10 minutes and then placed into a 42˚C water bath for 45 seconds.The tubes were returned to ice for 2 minutes and then 1mL of LB was added and the culture was placed in a shaking incubator for 1 hour.All of the transformation mixture was plated onto LB plates supplemented with kanamycin.Resistant colonies were selected and tested for the presence of the assembled fragments with primers rpfF-ko-conf-F and rpfF-ko-conf-R.The colonies were then tested with the primers rpfF-ko-conf-F and rpfF-ko-conf-R for the correct fragment insertion.A positive colony was then grown overnight, and the transformation plasmid was extracted with the QIAprep Spin Miniprep Kit (Cat.No. / ID: 27106X4).Xtu UPB513 cells were grown for two days on NA plates and then transferred into nutrient broth (NB) for overnight growth.Competent cells were generated from an overnight culture that was subjected to 4 washes with 10% glycerol and centrifugation at 4000 x g. 400ng of the transformation plasmid was added to 100uL of competent UPB513 cells along with 0.75 µL of TypeOne™ Restriction Inhibitor (Epicenter) and incubated on ice for 2 minutes.Cells were electroporated at 2.5kV.The transformation mixture was incubated for 3 hours at 28 o C in 1mL of NB medium with 250rpm shaking and then plated onto NA plates supplemented with kanamycin.Resistant colonies were picked and then plated onto both NA with kanamycin and NA with kanamycin and 10% sucrose.Colonies that were growth deficient in the presence of sucrose were pooled and grown overnight in NB before being plated on NA with 10% sucrose.
Colonies that were no longer inhibited by sucrose were selected and plated on NA with kanamycin and 10% sucrose and NA with only 10% sucrose.Those colonies that were now sucrose insensitive and kanamycin sensitive were selected as potential mutants.Mutants were tested with PCR using the primers rpfF-del-F1 and rpdF-del-R2 and amplicons were sequenced with Sanger sequencing to confirm a clean deletion of 672 nucleotides in rpfF to create UPB513ΔrpfF.

Knockout mutant complementation and introduction of gentamycin resistance
NEB Quickload polymerase and the primers rpfF-clone-F and rpfF-clone R were used to clone the entirety of the rpfF promoter, open reading frame and terminator region from UPB513 genomic DNA with regions overlapping into pUC18-miniTn7-T-Gm digested with HindIII.The cloned sequence was inserted into pUC18-miniTn7-T-Gm (Choi and Schweizer 2006) with Gibson assembly and then E. coli dH10b competent cells were transformed with the assembled plasmid using the previously described heat shock protocol.Cells were selected on NA with gentamycin and checked for the presence of rpfF with the primers rpfF-del-F1 and rpfF-del-R2.
The plasmids were extracted with a Qiagen Miniprep kit and then Sanger sequenced using the primers rpfF-clone-F, rpfF-clone-R, walkprimer1, walkprimer2 and walkprimer3 to ensure that there were no mutations present in the inserted sequence.UPB513 and UPB513ΔrpfF were grown for two days on NA and then grown overnight in 10mL NB to an O.D. 600nm of approximately 0.9.Competent cells were generated using the protocol from Choi et al. (2006) Cells were resuspended in 1mL 330mM sucrose at the final step.100µL of the UPB513ΔrpfF competent cells were mixed with 250ng of the pUC18-miniTn7-T-Gm plasmid containing the cloned rpfF and 250ng of pTNS3.The mixtures were electroporated at 2.5kV and 1mL of NB was immediately added.After three hours of outgrowth at 28˚C with 250rpm shaking the transformed cells were plated on NA supplemented with gentamycin.Colonies that had gained resistance to gentamycin were confirmed for the presence of both wildtype and mutant alleles of rpfF with the primers rpfF-del-F1 and rpfF-del-R2 and for genomic insertion at the attTn7 site (Choi et al. 2005) with primers insertion-conf-F and insertion-conf-R to validate the strain UPB513ΔrpfF::miniTn7T-Gm-rpfF.100μL of the UPB513ΔrpfF and UPB513 competent cells were also each transformed with the same protocol using 250ng of pUC18-miniTn7-T-Gm and 250ng of pTNS3 to generate the strains UPB513ΔrpfF::miniTn7T-Gm and UPB513::miniTn7T-Gm.

In vitro growth curve assay
Bacterial strains are grown from a starting concentration of 0.0001 O.D. 600nm .Each replicate was grown in 100uL in a 96 well plate with shaking at 567cpm for 48-96 hours at 28 o C in a Biotek plate reader (Agilent, Santa Clara CA, USA).Measurements of O.D. 600nm were taken every 30 minutes.

Plant inoculations
Three cm sections of the 1 st leaf of 14-day-old wheat cv.Chinese Spring seedlings were infiltrated with a water mock or a 0.0001 O.D. 600nm suspension of UPB513::miniTn7T-Gm or UPB513ΔrpfF::miniTn7T-Gm.Plants were grown in a growth chamber with a 12 hour photoperiod, 70% humidity and 25˚C temperature.Pots were enclosed after inoculation for three days.

In planta growth curves
Pictures were taken of four infiltrated leaf sections from the treatment groups and then a 1.5mm 2 hole punch was taken from each infiltrated section.Hole punches were macerated and serially diluted.Serial dilutions were plated on NA with gentamycin and colony forming units were counted at 5 days.This experiment was repeated twice.Percentage of water soaking symptoms were determined for pictures taken at day six with ImageJ (Schneider et al. 2012).

Mesophyll fluid extraction:
The protocol from Roman-Reyna et al. ( 2024) for barley mesophyll apoplastic fluid (BMAF) extraction was adapted by removing the ribitol spiking step because metabolites were not quantified by normalization to ribitol.

Xtu encodes a homolog of rpfF
The presence of rpfF is required for the pathogenesis of multiple Xanthomonas species.
We therefore hypothesized that Xtu encodes an ortholog of rpfF.The amino acid sequence of X. campestris pv.campestris (NCBI accession 3M6M_A, Supplemental File 1) was used as a query with NCBI's tblastn tool (Gerts et al. 2006) against the genome of Xtu UPB513 (NCBI GenBank assembly GCA_023221635.1) which returned a single significant alignment with 93% coverage and 53.5% identity (E-value = 2e -96 , Max score = 308).This open reading frame was identified as the Xtu ortholog of rpfF (Supplemental File 2).

rpfF is not required for Xtu growth and pathogenesis
Mutants in rpfF in other Xanthomonads have been critically impaired in their ability to colonize and cause disease in host plants.Therefore, we hypothesized that Xtu requires rpfF to coordinate behavior in planta and cause BLS symptoms.To challenge our hypothesis, we conduced multiple in planta experiments.The end stage of disease in which wheat leaf tissue is completely necrotic was achieved equally in plants infected with wildtype UPB513 and with UPB513ΔrpfF (Figure 2A).We observed exudates in both wildtype UPB513 and UPB513ΔrpfF.
We hypothesized that although the end state of disease may be similar, it is possible that loss of rpfF impacts Xtu's colonization of its wheat host in a minor way that our high-inoculum inoculation could not detect.To examine in planta growth, we tracked the population of both UPB513 and UPB513ΔrpfF over eight days.In two independent replicates we found that the populations of both UPB513 and UPB513ΔrpfF were equal every day (Figure 2B).We also measured the percentage of water-soaking in leaves at 6 days and did not find any differences between the mutant and wildtype infected leaves (Figure 2C).Xanthomonas motility is impacted in rpfF mutants in multiple Clade II Xanthomonas species (Chatterjee and Sonti 2002;Dow et al. 2003;Huang et al. 2013), but not in an rpfF mutant in the Clade I species X. albilineans (Rott et al. 2013).We hypothesized that rpfF is required for motility in Xtu.However, we found no evidence that loss of rpfF led to impaired motility (Figure 2D).

rpfF is not required for normal growth in lag and log phases, but is required for maintenance of stationary phase in plant fluid
In X. campestris, rpfF plays important roles in life stage transition (Dow et al. 2003;Tang et al. 1991).We hypothesized that rpfF is required for normal growth and sustainment of Xtu in rich media.To test this, we compared the growth of UPB513 and UPB513ΔrpfF in vitro in both complex artificial media and fluid extracted from a plant environment.There were no differences in the lag, log and stationary phases between UPB513 and UPB513ΔrpfF in nutrient broth (Figure 3A).Growth rates were not significantly different between treatments.We found that UPB513ΔrpfF was also not deficient in lag and log growth phases in barley mesophyll apoplast fluid (BMAF) (Figure 3B).There was a clear difference in stationary phase in BMAF between UPB513 and UPB513ΔrpfF.UPB513ΔrpfF completed a normal shift to stationary phase, however a death phase began earlier in UPB513ΔrpfF than the wildtype UPB513 (Figure 3B).
Wildtype stationary phase behavior was rescued by re-introduction via genomic insertion of the intact rpfF gene (Figure 3B).When run in the same in vitro experiments as previously described for Xtu, X. campestris ΔrpfF had irregular growth in lag and log phase in rich media (Figure 3C) but also exhibited a similar early death phase as demonstrated by X. translucens in BMAF (Figure 3D).

Discussion:
In this study we found that Xtu encodes a homolog of the DSF synthase gene rpfF.We discovered that Xtu was able to cause BLS symptoms and colonize wheat plants equally as well without rpfF.The rpfF mutant also retained wildtype motility.The Xtu rpfF mutant was unaffected in lag and log phases in vitro in both rich media and BMAF.The X. campestris rpfF mutant, in contrast, exhibited irregular growth in log phase in rich media.We found evidence that rpfF does play some role in stationary phase for both Xtu and X. campestris as population decrease occurred earlier for rpfF mutants.
The decreased importance of rpfF for Xtu may be a feature of Clade I Xanthomonas species.The gene rpfF has been well studied in multiple Clade II Xanthomonas species and found to play important roles in plant colonization and virulence in all cases.In stark contrast, Xtu colonized its primary niche, the mesophyll apoplastic space of wheat leaves, as normal despite deletion of rpfF.We speculate that some evolutionary pressures, such as plant recognition of DSF, may have led to changes in its function in Xtu.A rpfF mutant in X. albilineans, another Clade I Xanthomonas species, did have a decreased ability to colonize sugarcane stalks but this phenotype was much less extreme than what has been described in Clade II Xanthomonas species.Examination of the importance of this cluster in other Clade I species will be required to determine whether this pattern holds true.
X. albilineans and Xtu have substantially different lifestyles.Xtu is a nonvascular pathogen of cereal plants (Gluck-Thaler et al. 2020), while X. albilineans colonizes the vascular xylem of sugarcane plants (Rott et al. 2013).It is possible that Clade I rpfF may be relevant to vascular colonization, but not for nonvascular colonization such as that carried out by Xtu.
Future work examining the role of rpfF of vascular X. translucens strains such as those in the pathovars translucens and graminis would test this hypothesis.X. albilineans also lacks the hrp regulon which is a link between the quorum sensing system and regulation of the T3SS in X. campestris (Jiang et al. 2018;Pieretti et al. 2009).Xtu on the other hand, has an intact T3SS (Goettelmann et al. 2022).Therefore, it's probable that X. albilineans and Xtu have evolved distinct strategies to utilize and regulate their T3SS during plant colonization suggesting that the function of rpfF may not be analogous in both cases.
We described a role for rpfF in stationary phase in both Xtu and X. campestris.In summary, Xtu was able to grow normally without rpfF, however in late stationary phase the optical density, which is a proxy for population, decreased.There was a similar population decrease for both Xtu and X. campestris, however this effect was more pronounced and occurred earlier for X. campestris which correlates with the more important role that rpfF plays in X. campestris.Interestingly, this only occurred in plant-extracted mesophyll fluid, and not in artificial rich media.
In X. oryzae, Singh et al. (2022) determined that rpfF plays a role in maintaining membrane stability.The stress experienced by a cell increases in stationary phase relative to log phase (Singh et al. 2022).It is unknown whether there is a link between membrane stability and the phenotype we saw in the stationary phase of UPB513ΔrpfF.The existence of this death phase phenotype in an Xtu rpfF mutant suggests that rpfF does indeed play some role in coordinating cell-to-cell communication in Xtu populations.
The high cell density switch to stationary phase in vitro and to virulence-related behaviors in planta is not fully explained by the rpf system in Xtu.This strongly suggests that additional signaling systems are involved in perception of and response to high cell density.We cannot rule out the additional possibility that this system is still responsible for perception of an autoinducer signal via RpfC, and that rpfF is not required for autoinducer synthesis.The in-vitro loss of planktonic growth phenotype we discovered could be used in a high-throughput screening of mutants to find additional genes important for quorum sensing.
A main question that remains as a result of our research is what role does rpfF play in Xtu's life cycle.Xtu encodes rpfF which implies a selective pressure to retain this gene and loss of rpfF results in changes in stationary phase behavior in vitro.However, there was no discernible importance of rpfF in a susceptible wheat plant.Xtu has a wide host range (Heiden et al. 2023) and rpfF may be required for colonization of other hosts or more resistant wheat varieties which present more of a challenge for Xtu.It is also possible that this gene is important for environmental survival in suboptimum nutrient environments.
Our main objective was to determine whether rpfF was a critical pathogenicity factor that could be interfered with to combat BLS of wheat.Our research determined that rpfF is not a major Xtu pathogenicity factor.Therefore, we do not recommend the targeting of the rpf quorum sensing system as a preventative treatment for BLS of wheat.3cm sections of the 1 st leaf of 14-day-old wheat cv.Chinese Spring seedlings were infiltrated with a water mock or a O.D. 600nm 0.1 suspension of UPB513 and UPB513ΔrpfF.Plants were left covered for three days to create 100% relative humidity, and then were left uncovered for the subsequent 11 days.Exudates were observed in all treatments except for the mock at 3 days past inoculation (dpi).Pictures were taken at 14dpi.(B) X. translucens UPB513.3cm sections of the 1 st leaf of 14-day-old wheat cv.Chinese Spring seedlings were infiltrated with a water mock or a O.D. 600nm 0.0001 suspension of UPB513::miniTn7T-Gm, or UPB513ΔrpfF::miniTn7T-Gm.
1.5mm 2 hole punches were taken from leaves of each treatment (n=4) at two, four, six and eight days after infiltration.Colony forming units were counted on selective media.At six days pictures were taken of samples leaves and percent water area was quantified with ImageJ.This experiment was repeated twice.(C) 5µL of UPB513 or UPB513ΔrpfF were spotted onto nutrient agar media plates with 0.3%, 0.5% or 1% agar.Colony expansion was measured after 21 days of growth at 28 o C.

Figure 2 .
Figure 2. X. translucens does not require rpfF is not required for pathogenesis on wheat.(A)

Figure 3 .
Figure 3. rpfF is required for normal in vitro growth of X. translucens and X. campestris. 4

Table 1 .
Bacterial strains and plasmids used in this study.

Table 2 .
Primers used in this study