The strains and plasmids used in this study are listed in Table 1. E. coli strains Nova blue or DH5α were used for general cloning. The E. coli strains were propagated at 37°C in Luria Broth (LB) with agitation or on LB agar plates. When appropriate, the growth medium was supplemented with ampicillin (100 μg/ml). S. cerevisiae strains were grown at 30°C in YPD [1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) glucose] broth or agar (2%). Media ingredients were purchased from Himedia and Difco. Identity of all BY4741 deletion strains were verified by UP TAG (barcode) confirmation.

The vopE gene was amplified from the genomic DNA of V. cholerae strain SC110 using the gene specific primers and used to generate recombinant plasmids (Table 1). The conserved residues or regions of VopE were either deleted or substituted with alanine or glycine using Stratagene one step mutagenesis kit. The recombinant plasmid pESCLEU-VopE was used as a template for all mutagenesis reactions. The desired constructs were further transformed into S. cerevisiae strains.

For localization studies, all VopE constructs were C-terminally tagged with mCherry (from pmCherry-C1, Clonetech) by an overlapping PCR as described earlier (Tripathi et al., 2010). For mitochondrial visualization, COX4-GFP fragment was constructed by fusing a fragment encoding first 63 bases of COX4 to a gene encoding GFP (from pGREG575) by overlapping PCR method. This construct was further inserted into pRS413. This recombinant plasmid pCOX4-GFP was transformed into BY4741 to generate reporter strain BY4741-COX4-GFP. For Gem1 over expression, Gem1 gene was amplified from genomic DNA of BY4741 with NotIF/PacIR primers, and further inserted into pESCLEU-VopE to generate dual expression plasmid pESCLEU-VopE-Gem1. For VopX expression, vopX gene was amplified from the genomic DNA of V.cholerae strain SC110 using the gene specific primers, and subsequently inserted into NotI/PacI sites of pESCLEU or pESCLEU-VopE to generate recombinant plasmids. For protein detection, VopE constructs were tagged at C-terminus with 3XFLAG. All recombinant plasmids constructed in this study (Table 1) were confirmed by sequencing and maintained in E. coli strains.

Yeast growth assays were performed as described earlier (Tripathi et al., 2010; Bankapalli et al., 2015). Briefly, overnight cultures of all recombinant strains in selective SC^Raf medium were diluted and grown again in same media at 30°C until exponential phase (OD₆₀₀ ~ 0.8–1.0). Effect of the expression of VopE, its derivatives and VopX on the growth of S. cerevisiae strains was examined by spotting equal number of cells onto SC and SC^Gal plates lacking the corresponding auxotrophic markers to maintain the plasmids. When necessary, liquid growth induction was done for 6 h by adding galactose (2%) to cultures before spotting. Growth was monitored after 60–70 h at 30°C and photographed accordingly. Liquid growth assay was done as described earlier (Bankapalli et al., 2015). For yeast viability plating assay (cfu), recombinant stains carrying VopE or its variants were grown overnight in selective SC^Raf media. Cultures were again diluted to OD₆₀₀ 0.1 in fresh media and incubated up to OD₆₀₀ ~1.0 where cultures were resuspended in selective SC^Gal media. At different times after induction, aliquots were removed, and serial dilutions were plated on selective SC solid media. Plates were incubated for 2–3 days at 30°C before viable colonies were counted (Salomon et al., 2011).

S. cerevisiae strain BY4741 harboring pESCLEU-VopE (BY4741-VopE) was transformed with various lacZ reporter plasmids as listed in Table 1 (2XRlm1-lacZ, P_FUS1-lacZ, and UPRE-lacZ). The HOG pathway related lacZ reporter plasmid (HOG 8XCRE-lacZ) was transformed into W303-1A strain harboring pESCLEU-VopE (W303-VopE). All the plasmids were obtained as gift from Dr. Guido Sessa, Tel Aviv University, Tel Aviv, Israel (Salomon et al., 2012). For β-galactosidase assay, we followed the published protocol (Salomon et al., 2012). Briefly, overnight cultures of recombinant strains were diluted in selective medium containing raffinose (2%) and grown to OD₆₀₀ 0.6–0.9. Galactose (2%) was added and incubated for 6 h. Equal amount of cultures were then distributed into two tubes. In one set, stressors were added and kept for stipulated period as given. To activate the HOG pathway, 0.5 M NaCl for 1 h; to activate the CWI pathway, 3 mM caffeine for 4 h; to stimulate the UPR, 2 mM DTT for 4 h; to induce the pheromone signaling pathway, 5 μM α-factor for 2 h. Cells were finally harvested and subjected to rapid permeabilization procedure as described earlier (Kippert, 1995). Activity is reported as percentage of Miller units.

The immunoblot analysis was carried out based on previous work (Tripathi et al., 2013; Bankapalli et al., 2015). All the recombinant yeast strains carrying wild type VopE and its congeners were grown in selective SC^Raf medium at 30°C until mid-log phase. The cultures were then diluted in selective induction media (SC^Gal). After 6 h of induction, cultures were pelleted and whole cell protein extracts were prepared by adopting a published protocol (Zhang et al., 2011). Equal amount of protein samples were fractionated by SDS-polyacrylamide gel electrophoresis using 12% polyacrylamide gel and transferred to Immobilon–P Transfer Membrane (Millipore). Membrane was probed with either Monoclonal anti-FLAG M2-Peroxidase (HRP) Clone M2 (Sigma) antibody or anti-Porin (yeast mitochondria) Monoclonal Antibody (Invitrogen, Mitosciences). The primary antibody was detected using a horseradish peroxidase conjugated anti-mouse antibody (Millipore) and blot was developed with Luminata™ Forte Western HRP substrate.

To investigate the localization of VopE and its mutagenized congeners, recombinant yeast strains harboring these constructs (Table 1) were harvested after galactose induction. Samples were observed by Confocal Laser Scanning Microscope (Nikon A1R). For isolation of mitochondria, yeast cells containing vector (pESCLEU), or VopE were grown in selective SC media and grown again in 500 ml of fresh media till OD₆₀₀ was 1.5. Subsequently, cells were harvested and expression of VopE was induced by re-suspending the pellet in 500 ml of selective SC^Gal media for 6 h. Purification of yeast mitochondria was performed according to standard procedure (Gregg et al., 2009).

Several V. cholerae strains belonging to non-O1, non-O139 serogroups (laboratory collection) were examined for the presence of the vopE gene. The gene fragment was PCR amplified and cloned under the control of GAL1 promoter in the high copy number vector pESCLEU to generate recombinant plasmid pESCLEU-VopE. Next, VopE (A33_1662) was ectopically expressed in a S. cerevisiae strain W303-1A. Under inducing condition, spotting on solid agar media exhibited no growth inhibition, where as in liquid media strong growth inhibition was observed (Figure 1A, left and right panels). As reported, certain effectors exhibit an altered toxicity in solid and liquid growth assay conditions in yeast model system (Kramer et al., 2007; Siggers and Lesser, 2008; Antic et al., 2014). Interestingly, it has been documented that effector driven growth inhibition is dependent on different yeast genetic backgrounds (Rodríguez-Pachón et al., 2002; Bénit et al., 2010; Salomon et al., 2011). To ascertain the function of VopE in another genetic background, we transformed the pESCLEU-VopE recombinant construct into strain BY4741. An increase in lethality of VopE in BY4741 genetic background over that in W303-1A was observed (Figure 1B left and right panels).

Activity of bacterial effectors carrying GAP domain largely depend on arginine residue, referred as the arginine finger. VopE is a member of bacterial GAP effector family and also carries a conserved arginine along with additional GAP domain features. To ascertain the contribution of arginine at position 125 in VopE mediated toxicity in yeast system, we exchanged arginine with alanine and evaluated the lethality of recombinant VopE^R125A variant. The data clearly demonstrated significant loss in lethality of VopE^R125A variant, thereby underscoring the importance of arginine residue in the functionality of VopE in yeast system (Figure 1B left panel). To investigate in vivo protein stability, wild type and VopE^R125A were C-terminally tagged with FLAG epitope and western blot analysis using anti-FLAG antibody was carried out (Zhang et al., 2011; Tripathi et al., 2013). The results indicated that VopE and VopE^R125A were found to be stable (Figure 2C). The viability of yeast carrying VopE or VopE^R125A was also determined by CFU assay at different time periods up to 24 h. A gradual decrease in number of viable cells suggesting that growth inhibition caused by VopE is due to loss of viability (Figure 1D).

Although, bacteria encode different T3SS effector proteins, quantity of each effector translocated into host varies in amount during infection. Earlier studies have clearly demonstrated that low level expression of effector proteins increases the specificity while high level expression promotes sensitivity of inhibition in yeast model system (Slagowski et al., 2008). In some cases, it is also surmised that high level expression may result in non-specific toxic effects (Salomon et al., 2012). Therefore, we wanted to ascertain whether low level expression of VopE maintains similar toxicity as presented in the preceding section, the gene encoding wild type VopE was cloned in pGML10, a low copy number vector (Table 1). Unlike yeast harboring pESCLEU-VopE construct, recombinant yeast cells transformed with pGML10-VopE show mild growth defect, further indicating that VopE toxicity depends on the level of expression (Figures 1C and 1B right panel). As evidenced, several effector proteins exhibit toxicity only when expressed at high level (Slagowski et al., 2008). During analyzing the function of various Legionella pneumophila effectors in yeast system, it appears that toxicity of some effectors manifests only at high level while others are independent of the level of expression (Havey and Roy, 2015).

VopE contains predicted MTS spanning N-terminal 23 amino acids and removal of this stretch has been reported to affect VopE localization to mitochondria (Suzuki et al., 2014). To investigate the cellular fate of VopE variant lacking MTS (henceforth known as VopE^ΔMTS), growth of recombinant strain BY4741-VopE^ΔMTS carries a VopE^ΔMTS in high copy number vector was examined under inducing condition. Solid agar spotting data indicate that (VopE^ΔMTS) retains some ability to intoxicate yeast cells albeit less than wild type protein VopE (Figure 2B, left panel). The spotting data were further corroborated by liquid growth assay where a similar trend of growth inhibition was noticed (Figure 2B, right panel). To ensure in vivo stability of VopE^ΔMTS, western blot was performed by using anti-FLAG antibody (Figure 2C). In growth viability assay, number of growing cells increased for first 12 h, and then maintained as constant number of viable cells (Figure 1D).

Structural and sequence analysis reveal certain unique features essential for optimal performance of different bacterial GTPase-activating proteins (GAPs) (Litvak and Selinger, 2003). For example, there are several conserved residues and two bulge regions (I & II) in bacterial GAPs whose mutual interactions maintain the precise orientation of the critical arginine and catalytic efficiency of the GAPs. The catalytic arginine residue within the bulge I interacts with the nucleotide phosphate and with the conserved residues of bulge II (Stebbins and Galan, 2000; Wurtele et al., 2001; Litvak and Selinger, 2003). Sequence alignment of VopE with other bacterial GAPs identified bulge II region (Figure 3A). To examine the importance of this region in VopE function, the bulge II region (amino acids 159–167) was deleted and the activity of VopE^Δ159−167 variant was examined in yeast strain BY4741. A loss of lethality was observed in the yeast strain carrying the VopE^Δ159−167 variant (Figure 3B) further corroborating the importance of bulge II region in VopE function. Next, we turned our attention on other amino acids (V¹¹⁹, A¹²⁰, N¹²¹, T¹²⁹, Q¹⁶¹, G¹⁶³, T¹⁶⁴) predicted to involve in network of multiple interactions (Figure 3A). A series of VopE variants carrying alanine substitutions (T129A, Q161A, G163A, and T164A), glycine substitution (A120G) or VAN deletion (Δ119–121) were constructed and their effect was examined in yeast. Significant loss in lethality of VopE alanine (T129A, Q161A, G163A, T164A) and deleted (ΔVAN) variants (Figure 3B), thereby indicating the significance of these interacting residues in maintaining VopE function in addition to arginine¹²⁵ residue. It should be noted that toxicity of VopE variant harboring glycine at position 120 (VopE^A120G) remains unperturbed.

In addition to the residues posited in bulge I and II regions, there are residues participating in the formation of hydrophobic core in the GAP domain of SptP (Stebbins and Galan, 2000). Based on Clustal Omega alignment of VopE with SptP, ExoS, and YopE, three amino acids such as F¹¹⁰, F¹⁶², and V¹⁷³ were selected that might contribute to the formation of hydrophobic core of VopE (Figure 3A). Subsequent replacement of these amino acids with alanine resulted in loss of functionality of VopE (Figure 3B). Interestingly, non-conserved region might also contribute to the function of any given protein (Sampath et al., 2003). Indeed, when a non-conserved region spanning 24–80 amino acids (Figure 2A) was deleted, the resultant VopE^{Δ 24–80} variant was found to be non-lethal in yeast model system (Figure 3B). Collectively, our effort led to the identification of several GAP domain associated residues and non-conserved region contributing to the functionality of VopE. To examine in vivo stability, western blot analysis of recombinant yeast strains harboring wild type and different VopE variants was carried out by using anti-FLAG antibody. We found that all the VopE derivatives were stable under in vivo condition except VopE^Δ24–80 and VopE^{Δ 1–80} (Figure 3C).

To examine mitochondrial localization of VopE in yeast model, a yeast reporter strain BY4741-COX4-GFP was first constructed by transforming pCOX4-GFP in S. cerevisiae BY4741. COX4-GFP fusion protein (targeted to mitochondrial matrix) has been used to visualize mitochondrial morphology in yeast (Otsuga et al., 1998). Next, pVopE-mCherry was transformed into BY4741- COX4-GFP to generate BY4741-COX4-GFP-VopE-mCherry. In addition, VopE variants fused with mCherry (Table 1) were also transformed into BY4741-COX4-GFP. Upon induction, VopE-mCherry conferred growth inhibition in BY4741-COX4-GFP-VopE-mCherry (Figure S1), thus indicating that the fusion of mCherry at the C-terminus of VopE did not alter its activity. To visualize localization of mCherry tagged variants, the recombinant strains were harvested at 6 h following the addition of galactose and visualized by confocal microscopy. Similar to previous reports using mammalian cell lines, in yeast system also VopE and VopE^R125A localize to mitochondria whereas VopE^ΔMTS was detected in the cytosol (Figure 4A). Localization of VopE in mitochondria was further confirmed by sub cellular fractionation where presence of VopE-3XFLAG signal in yeast mitochondrial fraction was detected by immunoblot with anti-FLAG antibody (Figure 4B).

Certain P. syringae pv. tomato DC3000 effectors (HopAA1-1, HopAM1, HopAD1 etc.) when expressed in yeast have been shown to impair respiration and their toxicity enhanced on ethanol glycerol (EG) selective media (Munkvold et al., 2008). As VopE targets to mitochondria, it was therefore conceivable that VopE may alter respiration in yeast and show enhance killing of yeast cells on glycerol. But spotting of recombinant yeast strain harboring VopE cloned under constitutive or inducible promoters did not show any enhanced killing of yeast cells on selective media containing glycerol (2%; Figure 4C), thus indicating VopE does not impair yeast respiration.

Inhibition of mitochondrial clustering by VopE is mediated through the inactivation of Miro GTPases (Suzuki et al., 2014). The mammalian miro Rho GTPase ortholog in the budding yeast is Gem1p (Frederick et al., 2004). Like other miro proteins, Gem1p also contains two GTPase domains, a pair of calcium binding EF-hand motifs and C-terminal transmembrane (TM) domain. Both GTPase domains and EF-hand motifs are essential for Gem1p to maintain mitochondrial dynamics (Frederick et al., 2004). To examine whether the VopE dependent lethality in yeast system was also mediated by the Gem1p (yeast Miro GTPase), VopE was transformed into the yeast strain lacking Gem1p, BY4741Δgem1 (Table 1) and growth was assayed after VopE expression. In addition, Gem1 was also over expressed in wild type yeast harboring VopE. Unexpectedly, VopE expression caused lethality in the gem1 deleted yeast strain and gem1 over expressed strain (Figures 5A,B). Conversely, Gem1, the yeast Miro GTPase is not targeted by VopE.

A battery of cellular components including several GTPases namely mitochondrial Rho GTPases (Miro I and Miro II), Fzo/mitofunsins (Mfn1 and Mfn2), Dnm1/Drp1/Dlp1, Mgm1/Opa1/Msp1 contributing to mitochondrial dynamics have been identified in yeast, flies, worms, and mammals (Shaw and Nunnari, 2002; Frederick and Shaw, 2007). VopE expression in a number of mitochondrial GTPase yeast deletion strains (Δdnm1, Δfzo1, Δmgm1, Δmtg1, Δmtg2) produced lethality similar to that produced by VopE expression in the wild type yeast strain (Figure 5A).

Many translocated bacterial effector proteins do not exhibit lethality in yeast model system either due to non-conservation of targets between yeast and mammalian system or because target (s) is conserved but it is dispensable and not rate limiting for yeast growth under normal conditions (Slagowski et al., 2008; Salomon et al., 2011). In latter cases, certain compounds, or environmental conditions known as “stressors” have been shown to increase the sensitivity of yeast strains toward different effectors and aid in the identification of effector mediated growth-retardation traits (Siggers and Lesser, 2008). The commonly employed stressors are (i) caffeine, a purine analog exerts pleiotropic effects in yeast and activates the CWI MAP kinase pathway. The molecule is highly toxic to the mutants of CWI pathway and Tor pathway (Parsons et al., 2004; Kuranda et al., 2006); (ii) Sorbitol, an osmotic stressor activates HOG (high osmolarity glycerol) MAP kinase pathway and toxic to mutants of the HOG MAPK pathway (Gustin et al., 1998; Hohmann, 2002); (iii) NaCl, an osmotic and ionic stressor that exhibits high level toxicity to the mutants of HOG MAPK pathway and ion homeostasis (Gustin et al., 1998; Hohmann, 2002; Auesukaree et al., 2009); (iv) tunicamycin, an inhibitor of protein glycosylation shows extreme toxicity to the mutants of unfolded protein response (UPR) pathway and mutants with cell wall defects (Parsons et al., 2004); (v) DTT, a reducing agent, is also known to induce ER stress (Papa et al., 2003; Bosis et al., 2011); (vi) 37°C heat stress, causes pleiotropic effects in yeast (Auesukaree et al., 2009). It has also been demonstrated that stressors also suppress the toxic activity of certain effectors. For example, lethality of Coxiella burnetii effector CBU0388-CetCb2 has been shown to be repressed by caffeine (Lifshitz et al., 2014).

Since VopE showed no toxicity in S.cerevisiae strain W303-1A as compared to BY4741 in spotting assay (Figures 1A,B), the effect of various stressors on both S. cerevisiae strains W303-1A and BY4741 harboring VopE was next examined. For this purpose, the yeast strains were grown in the presence of some established stressors like caffeine (3 mM), NaCl (0.5 M), sorbitol (1 M), tunicamycin (0.3 μg/ml), and 37°C. Interestingly, VopE mediated lethality was observed in W303-1A in the presence of the stressors (caffeine, NaCl, sorbitol, and tunicamycin; Figure 6A). To further examine the cellular targets of VopE lacking MTS variant (VopE^ΔMTS), BY4741-VopE^ΔMTS recombinant strain was grown in the presence of various stressors. We observed an increase in lethality of VopE^ΔMTS only in the presence of caffeine and tunicamycin (Figure 6B). Since yeast growth in the presence of caffeine and tunicamycin mediated stress condition require functional CWI and UPR signaling pathways respectively (Parsons et al., 2004; Kuranda et al., 2006), these results suggested that VopE and its mutant form (VopE^ΔMTS) strongly perturbed the activation of CWI and UPR pathways. VopE also caused yeast growth failure in osmotic conditions reflecting attenuation of high osmolarity glycerol MAPK pathway. It may be noted that YopE, the Yersinia homolog of VopE, also targets CWI pathway (Kramer et al., 2007).

There are six MAPK cascades have been characterized in yeast mediating response to different stimuli namely (i) pheromone responsive pathway; (ii) filamentous growth pathway; (iii) high-osmolarity glycerol pathway; (iv) STE vegetative pathway; (v) nutrient starvation pathway; (vi) CWI pathway (Chen and Thorner, 2007; Arias et al., 2011). In order to identify the specific MAP kinase pathways which are majorly attenuated by VopE and its truncated form (VopE^ΔMTS), we performed lacZ-reporter assay with recombinant yeast strains harboring VopE and VopE^ΔMTS along with lacZ-reporter plasmids of various MAPK signaling pathways (CWI, HOG, Pheromone) and ER stress pathways (UPR). The results clearly showed that VopE inhibited the activation of (i) pheromone signaling pathway (32% inhibition); (ii) CWI pathway (73% inhibition); (iii) unfolded protein response pathway (UPR; 70% inhibition). Interestingly, VopE^ΔMTS also significantly suppressed the activation of UPR and CWI pathways (Figures 7A–C).

Targeting multiple MAPK pathways has been reported for other effector proteins e.g., YopJ, HopX1 (Yoon et al., 2003; Salomon et al., 2012). It may be noted that no inhibition of lacZ reporter activity corresponding to HOG pathway was observed (Figure 7D), but toxicity of VopE was observed in the presence of NaCl and sorbitol (Figure 6A). Such discrepancy has also been reported earlier, caffeine increases the toxicity of XopE2, a T3SS effector of Xanthomonas campestris, but no inhibition of lacZ reporter activity corresponding to CWI pathway in the presence of caffeine was observed (Bosis et al., 2011).

To test whether yeast MAPK pathway components modulate the toxicity, we transformed VopE and VopE^ΔMTS in yeast strains deleted for MAPK components of different signaling pathways namely CWI (Δslt2), High osmolarity glycerol (Δhog1), Pheromone (Δkss1), and sporulation (Δsmk1), and examined their growth under galactose inducing conditions. Of these, slt2 deletion strain expressing VopE^ΔMTS showed increased yeast growth inhibition than wild type (Figure 8A, Figure S2).

The CWI pathway maintains cell wall integrity and gets activated in the presence of diverse arrays of external stimuli such as heat stress, mating pheromones, oxidative stress, pH stress, cell wall mutations, cell wall stressors, and endoplasmic reticulum stress all of which are known to cause cell wall stress (Levin, 2005; Arias et al., 2011). We further examined the effect of VopE and its truncated version on various yeast deletion mutants of CWI pathway. Accordingly, VopE and VopE^ΔMTS were transformed into yeast deletion mutants (Δbck1, Δmkk1, Δmkk2, Δrlm1, Δkdx1) and the recombinant yeast strains were examined by spotting on solid agar plate. Unlike VopE, Δbck1 and Δslt2 mutants expressing VopE^ΔMTS exhibit increased yeast growth inhibition as compared to wild type strain (Figure 8A, Figure S2). An increased toxicity in absence of bck1 and slt2 further indicates that VopE and VopE^ΔMTS may activate a cellular response opposing the function of Bck1p and Slt2p. Such cases of hypersensitivity have been recorded with other effectors (e.g., OspF, XopE2, CBU1676, and CBU0885) as well (Kramer et al., 2007; Bosis et al., 2011; Lifshitz et al., 2014). It should be noted that lethality of VopE and VopE^ΔMTS was observed to be enhanced by tunicamycin, an UPR pathway activator. If VopE targets UPR pathway, then it should remain non-lethal in yeast deletion mutant of UPR pathway (Δire1). Instead, we observed maintenance of toxicity which further emphasizes that VopE does not target UPR pathway (Figure S3). It has been pointed out by Sessa and colleagues that tunicamycin is highly toxic to mutants with cell wall defects (Salomon et al., 2012). It is therefore possible toxicity of effectors targeting CWI such as VopE is exacerbated in the presence of tunicamycin.

Next, we wanted to investigate whether VopE mediated growth inhibition was due to initiation of apoptosis. Thus, we transformed VopE in yeast mutant Δyca1, carrying a deletion in the metacaspase gene, YCA1 and subsequently examined the viability of recombinant Δyca1 strain by solid agar spotting assay. If VopE targets YCA1, Δyca1 strain should remain viable in the presence of VopE. But we observed VopE maintained its lethality in YCA1 deleted strain which is comparable to wild type strain background (Figure 8B). These observations further reinforce that VopE and VopE^ΔMTS interact with the components of cell wall maintenance related pathways.

Co-expression of two effectors in yeast model has proven valuable in identification of suppressor function of certain effectors over other. Previously, it has been shown that L. pneumophila effectors, SidD, and SidJ were able to counteract the activity of SidM and SdeA respectively (Tan and Luo, 2011; Havey and Roy, 2015). We investigated the outcome of co-expression of VopE^ΔMTS with another V. cholerae effector VopX targeting similar cellular pathway, the CWI pathway in yeast (Alam et al., 2011; Seward et al., 2015). The dual expression constructs (Table 1) were transformed to S. cerevisiae strain BY4741. The recombinant yeast strains were subjected to solid agar spotting under galactose inducing condition. We observed a partial suppression of VopX mediated toxicity in dual expression condition by VopE^ΔMTS only (Figure 9). To further confirm the specificity, we co-expressed the functionally inert variant of VopE^ΔMTS harboring the substitution of catalytic arginine at position 125 with alanine (VopE^{ΔMTS R125A}) along with VopX. We observed that VopE^{ΔMTS R125A} did not inhibit VopX mediated toxicity further supporting the hypothesis that VopE^ΔMTS mediates suppression of VopX toxicity (Figure 9).

To summarize, the present work unfolds many additional features on functional aspects of VopE by exploiting yeast model system. These are as follows: (i) identification of several critical residues in GAP domain and N-terminal region of VopE conferring lethality in yeast system; (ii) VopE variant lacking mitochondrial localization signal possesses partial lethality; (iii) screening of yeast deletion library against VopE and its derivative VopE^ΔMTS results in the identification of hypersensitive strains; (iv) VopE and VopE^ΔMTS modulates CWI-MAPK pathway in yeast; (v) co-expression of VopE^ΔMTS and VopX causes functional perturbation of the latter.

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.