Tubulin Folding Cofactor TBCB is a Target of the Salmonella Effector Protein SseK1

International Journal of Molecular Sciences Article Tubulin Folding Cofactor TBCB is a Target of the Salmonella Effector Protein SseK1 Juan Luis Araujo-Garrido 1, Fernando Baisón-Olmo 1,2, Joaquín Bernal-Bayard 1, Francisco Romero 3and Francisco Ramos-Morales 1,* 1Departamento de Genética, Facultad de Biología, Universidad de Sevilla, 41012 Sevilla, Spain; [email protected] (J.L.A.-G.); [email protected] (F.B.-O.); [email protected] (J.B.-B.) 2Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas y Farmacéuticas, Universidad de Chile, Santiago 1058, Chile 3Departamento de Microbiología, Facultad de Biología, Universidad de Sevilla, 41012 Sevilla, Spain; [email protected] *Correspondence: [email protected] Received: 8 April 2020; Accepted: 28 April 2020; Published: 30 April 2020   Abstract: Salmonella enterica serovar Typhimurium is a human and animal pathogen that uses type III secretion system effectors to manipulate the host cell and fulfill infection. SseK1 is a Salmonella effector with glycosyltransferase activity. We carried out a yeast two-hybrid screen and have identified tubulin-binding cofactor B (TBCB) as a new binding partner for this effector. SseK1 catalyzed the addition of N-acetylglucosamine to arginine on TBCB, and its expression promoted the stabilization of the microtubule cytoskeleton of HEK293T cells. The conserved Asp-x-Asp (DxD) motif that is essential for the activity of SseK1 was required for the binding and modification of TBCB and for the effect on the cytoskeleton. Our study has identified a novel target for SseK1 and suggests that this effector may have a role in the manipulation of the host cell microtubule network to provide a safe niche for this pathogen. Keywords: TBCB; SseK1; tubulin; type III secretion; Salmonella; arginine N-glycosyltransferase 1. Introduction Salmonella is Gram-negative rod-shaped motile bacteria belonging to the family Enterobacteriaceae. Salmonella and many other Gram-negative symbionts and pathogens of animals and plants rely on type III secretion systems (T3SS) for their interactions with the host [ 1 , 2 ]. These systems usually work when bacteria are in contact with host cells and are able to translocate proteins, known as effectors, into the host cell cytoplasm [ 3 ]. S. enterica expresses two distinct T3SS for the manipulation of host cells, T3SS1 and T3SS2, that are encoded by genes located in the, Salmonella pathogenicity island (SPI) 1 and SPI2, respectively, and secrete more than 30 effectors. The T3SS1 provides a mechanism for host cell invasion that depends on the localized reorganization of actin filaments and the formation of membrane ruffles on the surface of host cells [ 4 ]. Once inside the Salmonella containing vacuole (SCV), the acidic pH and limitation of nutrients characteristic of this niche induce the expression of the T3SS2. This is a multifunctional virulence system that translocates effectors through the SCV to manipulate trafficking and maturation of this phagosome, providing a suitable niche for Salmonella intracellular survival and replication [5]. S. enterica serovar Typhimurium encodes three effectors, SseK1, SseK2 [ 6 ], and SseK3 [ 7 ], that belong to the NleB family of Asp-x-Asp (DxD)-containing glycosyltransferases [ 8 ]. This family of effectors includes NleB from Citrobacter rodentium, and NleB1, and NleB2 from enterohemorrhagic and enteropathogenic Escherichia coli. NleB1 adds N-acetylglucosamine (GlcNAc) to arginine in Int. J. Mol. Sci. 2020,21, 3193; doi:10.3390/ijms21093193 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2020,21, 3193 2 of 18 TNFR1-associated death domain protein (TRADD), FAS-associated death domain protein (FADD), and other death domain-containing proteins with an N-glycosidic linkage. This modification blocks the interaction between death domains, thereby disrupting inflammatory nuclear factor kappa B (NF- κ B) signaling, caspase 8-dependent apoptosis, and necroptosis [ 9 , 10 ]. Similarly, during macrophage infection, both SseK1 and SseK3 regulate the activation of NF- κ B negatively, and therefore, inhibit host cell death [ 11 ]. SseK1 promotes the GlcNAcylation of TRADD [ 9 ] and FADD [ 11 ], whereas SseK3 induces weaker arginine-GlcNAcylation of TRADD and no detectable modification of FADD, and there are conflicting reports about the biochemical activity of SseK2 [ 11 , 12 ]. SseK3 binds to the E3-ubiquitin ligase TRIM32 [ 13 ], but the significance of this interaction is unknown since this host protein is not modified by SseK3 and is not necessary for SseK3-dependent inhibition of NF- κ B activation during the infection of macrophages [ 11 ]. In addition, a recent report using a mass spectrometry-based approach to enrich GlcNAcylated peptides from infected host cells showed that SseK3 modifies the death domains of receptors TNFR1 and TRAILR [14]. This work was undertaken to identify new host proteins targeted by SseK1. We show here that this effector binds tubulin-binding cofactor B (TBCB), transfers GlcNAc to this target, and modulates the stability of the microtubule cytoskeleton in a manner dependent on its glycosyltransferase activity. 2. Results 2.1. Salmonella SseK1 Interacts with Human TBCB in the Yeast Two-Hybrid System The initial hypothesis of this work was that, after translocation through a T3SS, SseK1 would interact with host proteins that could also be substrates for the arginine glycosyltransferase activity of this Salmonella effector. In order to identify putative host interacting partners, SseK1 from the S. enterica serovar Typhimurium strain 14028 was fused to the DNA-binding domain of the bacterial LexA transcriptional factor using plasmid pLEX10 as a vector to yield plasmid pIZ2203. A yeast two-hybrid screen was carried out by transforming the Saccharomyces cerevisiae strain L40/pIZ2203 with a Jurkat (human lymphocyte cell line) cDNA library fused to the activation domain of the yeast transcriptional factor Gal4 using the vector pGAD1318. A total of 878 clones out of 7 × 10 5 transformants were able to grow in synthetic medium lacking histidine, that was used to select for the interactions. The expression of a second reporter gene, lacZ, was also tested, and 180 clones demonstrated β -galactosidase activity (blue color in the presence of X-Gal). Plasmid derivatives of pGAD1318 were isolated from these positive clones, and a subset of them was reintroduced in yeasts L40/pIZ2203 and L40/pLEX10 and tested for reporter gene activation. Sequencing of the specific clones revealed that most of them carried cDNA from human TBCB. The interaction of SseK1 with TBCB in the yeast two-hybrid system is shown in Figure 1. Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 2 of 17 and other death domain-containing proteins with an N-glycosidic linkage. This modification blocks the interaction between death domains, thereby disrupting inflammatory nuclear factor kappa B (NF- B) signaling, caspase 8-dependent apoptosis, and necroptosis [9,10]. Similarly, during macrophage infection, both SseK1 and SseK3 regulate the activation of NF-B negatively, and therefore, inhibit host cell death [11]. SseK1 promotes the GlcNAcylation of TRADD [9] and FADD [11], whereas SseK3 induces weaker arginine-GlcNAcylation of TRADD and no detectable modification of FADD, and there are conflicting reports about the biochemical activity of SseK2 [11,12]. SseK3 binds to the E3- ubiquitin ligase TRIM32 [13], but the significance of this interaction is unknown since this host protein is not modified by SseK3 and is not necessary for SseK3-dependent inhibition of NF-B activation during the infection of macrophages [11]. In addition, a recent report using a mass spectrometry-based approach to enrich GlcNAcylated peptides from infected host cells showed that SseK3 modifies the death domains of receptors TNFR1 and TRAILR [14]. This work was undertaken to identify new host proteins targeted by SseK1. We show here that this effector binds tubulin-binding cofactor B (TBCB), transfers GlcNAc to this target, and modulates the stability of the microtubule cytoskeleton in a manner dependent on its glycosyltransferase activity. 2. Results 2.1. Salmonella SseK1 Interacts with Human TBCB in the Yeast Two-Hybrid System The initial hypothesis of this work was that, after translocation through a T3SS, SseK1 would interact with host proteins that could also be substrates for the arginine glycosyltransferase activity of this Salmonella effector. In order to identify putative host interacting partners, SseK1 from the S. enterica serovar Typhimurium strain 14028 was fused to the DNA-binding domain of the bacterial LexA transcriptional factor using plasmid pLEX10 as a vector to yield plasmid pIZ2203. A yeast two-hybrid screen was carried out by transforming the Saccharomyces cerevisiae strain L40/ pIZ2203 with a Jurkat (human lymphocyte cell line) cDNA library fused to the activation domain of the yeast transcriptional factor Gal4 using the vector pGAD1318. A total of 878 clones out of 7 × 105 transformants were able to grow in synthetic medium lacking histidine, that was used to select for the interactions. The expression of a second reporter gene, lacZ, was also tested, and 180 clones demonstrated Figure 1. Human tubulin-binding cofactor B (TBCB) interacts with SseK1 in the yeast two-hybrid system. Strain L40 of S. cerevisiae was cotransformed with derivatives of plasmids pLEX10, to generate fusions with the DNA-binding domain of LexA, and pGAD1318, to generate fusions with the activation domain of Gal4, as indicated. The interaction between the two hybrid proteins is shown by the growth in the absence of histidine (-His), and the detection of blue color in the presence of X-Gal after a -galactosidase filter assay. Empty vectors were used as negative controls. Figure 1. Human tubulin-binding cofactor B (TBCB) interacts with SseK1 in the yeast two-hybrid system. Strain L40 of S. cerevisiae was cotransformed with derivatives of plasmids pLEX10, to generate fusions with the DNA-binding domain of LexA, and pGAD1318, to generate fusions with the activation domain of Gal4, as indicated. The interaction between the two hybrid proteins is shown by the growth in the absence of histidine (-His), and the detection of blue color in the presence of X-Gal after a β-galactosidase filter assay. Empty vectors were used as negative controls. Int. J. Mol. Sci. 2020,21, 3193 3 of 18 Additional clones were identified by PCR amplification with specific primers for TBCB and sequencing. Finally, a total of 130 clones corresponding to seven different genes demonstrated a specific interaction with SseK1 in the yeast two-hybrid system. A summary of the candidates identified is shown in Table 1. Table 1. Candidate host partners of SseK1 identified in a yeast two-hybrid screen. Gene Number of Clones Description of the Product Amino Acids Included in the Identified Clones TBCB 116 Tubulin-binding cofactor B 1–244 4–244 29–244 ZBTB16 7Zing finger and BTB domain-containing protein 16 420–673 448–673 552–673 H3F3B 3 Histone H3.3 1–136 40–136 H3F3A 1 Histone H3.3 18–136 CENPA 1Histone H3-like centromeric protein A 1–140 HIRIP3 1 HIRA-interacting protein 3 340–556 XRCC6 1X-ray repair cross-complementing protein 6 335–609 2.2. SseK1 Copurifies and Coimmunoprecipitates with TBCB Since most of the isolated clones corresponded to TBCB, we focused our study on the interaction of SseK1 with this host protein. To confirm the results obtained with the yeast two-hybrid approach, the bacterial glutathione S-transferase (GST) expression system was used to produce GST or GST-TBCB in a strain of Salmonella expressing a chromosomal SseK1-3xFLAG fusion (strain SV7071) [ 15 ]. Then, we performed affinity purification using glutathione-agarose beads to isolate GST proteins, and we further analyzed the presence of SseK1 by western blot. SseK1-3xFLAG copurified with GST-TBCB but not with GST (Figure 2A). Next, we performed coimmunoprecipitation assays to investigate if both proteins were able to interact in the more physiological context of the host cell. Human epithelial HeLa cells were cotransfected with pIZ2047, a derivative of plasmid pBABEpuro expressing SseK1-3xFLAG, and pIZ3423, a derivative of plasmid pCS2 expressing 3xHA-TBCB, or transfected with the later plasmid alone. SseK1-3xFLAG was immunoprecipitated from lysates of these cells and the copurification of 3xHA-TBCB was analyzed by immunoblot with anti-HA antibodies (Figure 2B). TBCB was coimmunoprecipitated with SseK1, but it was not precipitated in the absence of SseK1 (cells transfected with 3xHA-TBCB alone), showing the specificity of the interaction. 2.3. SseK1 Glycosylates TBCB in Vitro Next, we wondered whether TBCB, being a new interacting partner of SseK1, could also be a new substrate for the glycosyltransferase activity of this Salmonella effector. To test this hypothesis, we incubated TBCB, purified as GST-fusion protein, with GST-SseK1 in the presence of UDP-GlcNAc, and detected substrate glycosylation by western blotting using a specific antibody against Arg-GlcNAc [ 16 ]. SseK1 glycosylated TBCB but not the N-terminal (amino acids 1–125) or C-terminal (amino acids 126–244) fragments of this protein, suggesting that the full-length protein is necessary to get a proper interaction and modification of this substrate (Figure 3). Self Arg-GlcNAcylation of SseK1 was also observed in these experiments, as previously reported for SseK3 [ 14 ]. SseK1 contains the conserved DxD catalytic motif found in glycosyltransferases. As expected, mutation of this motif to AAA caused the loss of both self-glycosylation and glycosylation of the host substrate TBCB (Figure 3). Int. J. Mol. Sci. 2020,21, 3193 4 of 18 Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 4 of 17 Figure 2. TBCB interacts with SseK1 in bacteria and host cells. (A) Bacterial lysates from strain SV7071 (14028 sseK1::3xFLAG) expressing glutathione S-transferase (GST) or GST-TBCB were prepared in NP40 buffer. Then, purified GST fusion proteins were blotted on nitrocellulose membranes, stained with Ponceau S red (left panel), and developed with monoclonal anti-FLAG (right panel). Aliquots of the original lysates were also included. The molecular weights of the marker bands (M) are indicated. (B) HeLa cells were transiently transfected with a derivative of pBABEpuro expressing SseK1- 3xFLAG, a derivative of pCS2 expressing 3xHA-TBCB or cotransfected with the same plasmid and a derivative of pBABEpuro expressing SseK1-3xFLAG. NP40 lysates from 6 × 106 transfected cells were subjected to immunoprecipitation (IP) with anti-FLAG monoclonal antibodies and, after washing, resolved by SDS-PAGE, transferred to nitrocellulose membranes, and developed with monoclonal anti-HA. Total lysates from 5 × 105 transfected cells are included in the blot. M, molecular weight marker. Molecular weights in kDa are indicated on the left. Figure 2. TBCB interacts with SseK1 in bacteria and host cells. ( A ) Bacterial lysates from strain SV7071 (14028 sseK1::3xFLAG) expressing glutathione S-transferase (GST) or GST-TBCB were prepared in NP40 buffer. Then, purified GST fusion proteins were blotted on nitrocellulose membranes, stained with Ponceau S red (left panel), and developed with monoclonal anti-FLAG (right panel). Aliquots of the original lysates were also included. The molecular weights of the marker bands (M) are indicated. ( B ) HeLa cells were transiently transfected with a derivative of pBABEpuro expressing SseK1-3xFLAG, a derivative of pCS2 expressing 3xHA-TBCB or cotransfected with the same plasmid and a derivative of pBABEpuro expressing SseK1-3xFLAG. NP40 lysates from 6 × 10 6 transfected cells were subjected to immunoprecipitation (IP) with anti-FLAG monoclonal antibodies and, after washing, resolved by SDS-PAGE, transferred to nitrocellulose membranes, and developed with monoclonal anti-HA. Total lysates from 5 × 10 5 transfected cells are included in the blot. M, molecular weight marker. Molecular weights in kDa are indicated on the left. Int. J. Mol. Sci. 2020,21, 3193 5 of 18 Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 5 of 17 2.3. SseK1 Glycosylates TBCB in Vitro Next, we wondered whether TBCB, being a new interacting partner of SseK1, could also be a new substrate for the glycosyltransferase activity of this Salmonella effector. To test this hypothesis, we incubated TBCB, purified as GST-fusion protein, with GST-SseK1 in the presence of UDP-GlcNAc, and detected substrate glycosylation by western blotting using a specific antibody against Arg- GlcNAc [16]. SseK1 glycosylated TBCB but not the N-terminal (amino acids 1–125) or C-terminal (amino acids 126–244) fragments of this protein, suggesting that the full-length protein is necessary to get a proper interaction and modification of this substrate (Figure 3). Self Arg-GlcNAcylation of SseK1 was also observed in these experiments, as previously reported for SseK3 [14]. SseK1 contains the conserved DxD catalytic motif found in glycosyltransferases. As expected, mutation of this motif to AAA caused the loss of both self-glycosylation and glycosylation of the host substrate TBCB (Figure 3). Figure 3. SseK1 glycosylates TBCB in vitro. Assay for SseK1 GlcNAc modification of TBCB using recombinant GST fusion proteins in the indicated combinations and 1 mM UDP-GlcNAc. SseK1AAA: SseK1(D223A/D225A). TBCBN: N-terminal fragment of TBCB (amino acids 1–125). TBCBC: C-terminal fragment of TBCB (amino acids 126–244). A representative immunoblot is shown incubated with anti- Arg-GlcNAc (upper panel). The same blot reincubated with anti-GST was used as the loading control (lower panel). Molecular weights in kDa are indicated on the left. Figure 3. SseK1 glycosylates TBCB in vitro . Assay for SseK1 GlcNAc modification of TBCB using recombinant GST fusion proteins in the indicated combinations and 1 mM UDP-GlcNAc. SseK1 AAA : SseK1(D223A/D225A). TBCB N : N-terminal fragment of TBCB (amino acids 1–125). TBCB C : C-terminal fragment of TBCB (amino acids 126–244). A representative immunoblot is shown incubated with anti-Arg-GlcNAc (upper panel). The same blot reincubated with anti-GST was used as the loading control (lower panel). Molecular weights in kDa are indicated on the left. 2.4. SseK1 Glycosylates TBCB In Vivo The results obtained in the previous section prompted us to examine the glycosylation of TBCB by SseK1 in the context of the mammalian host cell. Human HEK293T cells were transiently transfected with pIZ2336 (pEGFPC1-SseK1) or cotransfected with the same plasmid and pIZ3423 (pCS2-3xHA-TBCB), lysates were prepared in NP40 buffer 48 h after transfection and an immunoblot was carried out to detect Arg glycosylation. Several specific glycosylated proteins were detected in cells expressing GFP-SseK1 alone (Figure 4A). Interestingly an additional glycosylated band of the size expected for TBCB was detected in cells co-transfected with plasmids expressing GFP-SseK1 and 3xHA-TBCB. To ascertain that this band corresponded to glycosylated TBCB, lysates from transfected Int. J. Mol. Sci. 2020,21, 3193 6 of 18 cells were further immunoprecipitated with an anti-HA antibody and analyzed by western blot with the anti-ArgGlcNAc antibody. A specific band was detected in lysates from cotransfected cells but not from cells expressing SseK1 only, confirming that TBCB was specifically glycosylated by SseK1 (Figure 4B). Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 6 of 17 2.4. SseK1 Glycosylates TBCB in Vivo The results obtained in the previous section prompted us to examine the glycosylation of TBCB by SseK1 in the context of the mammalian host cell. Human HEK293T cells were transiently transfected with pIZ2336 (pEGFPC1-SseK1) or cotransfected with the same plasmid and pIZ3423 (pCS2-3xHA-TBCB), lysates were prepared in NP40 buffer 48 h after transfection and an immunoblot was carried out to detect Arg glycosylation. Several specific glycosylated proteins were detected in cells expressing GFP-SseK1 alone (Figure 4A). Interestingly an additional glycosylated band of the size expected for TBCB was detected in cells co-transfected with plasmids expressing GFP-SseK1 and 3xHA-TBCB. To ascertain that this band corresponded to glycosylated TBCB, lysates from transfected cells were further immunoprecipitated with an anti-HA antibody and analyzed by western blot with the anti-ArgGlcNAc antibody. A specific band was detected in lysates from cotransfected cells but not from cells expressing SseK1 only, confirming that TBCB was specifically glycosylated by SseK1 (Figure 4B). Figure 4. SseK1 glycosylates TBCB in vivo. (A) NP40 lysates from HEK293T cells transiently transfected with plasmids expressing GFP-SseK1 and/or 3xHA-TBCB, as indicated, were analyzed by immunoblot using anti-Arg-GlcNAc antibodies. (B) Lysates from HEK293T transiently transfected with a plasmid expressing GFP-SseK1 or with this plasmid and another plasmid expressing 3xHA- TBCB were subjected to immunoprecipitation (IP) with an anti-HA antibody. Precipitates were analyzed by western blot (W) with an anti-Arg-Glc-NAc antibody. Molecular weights in kDa are indicated on the left. The arrow indicates the band corresponding to TBCB. 2.5. SseK3 Interacts with TBCB Since SseK2 and SseK3 share similar sequences and belong to the same family of effectors as SseK1, we decided to test the interaction of these two Salmonella effectors with human TBCB using the yeast two-hybrid system. The results were positive for SseK3 but not for SseK2 (Figure 5). Figure 4. SseK1 glycosylates TBCB in vivo . ( A ) NP40 lysates from HEK293T cells transiently transfected with plasmids expressing GFP-SseK1 and/or 3xHA-TBCB, as indicated, were analyzed by immunoblot using anti-Arg-GlcNAc antibodies. (B) Lysates from HEK293T transiently transfected with a plasmid expressing GFP-SseK1 or with this plasmid and another plasmid expressing 3xHA-TBCB were subjected to immunoprecipitation (IP) with an anti-HA antibody. Precipitates were analyzed by western blot (W) with an anti-Arg-Glc-NAc antibody. Molecular weights in kDa are indicated on the left. The arrow indicates the band corresponding to TBCB. 2.5. SseK3 Interacts with TBCB Since SseK2 and SseK3 share similar sequences and belong to the same family of effectors as SseK1, we decided to test the interaction of these two Salmonella effectors with human TBCB using the yeast two-hybrid system. The results were positive for SseK3 but not for SseK2 (Figure 5). Int. J. Mol. Sci. 2020,21, 3193 7 of 18 Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 7 of 17 Figure 5. SseK3 interacts with TBCB in the yeast two-hybrid system. Strain L40 of S. cerevisiae was cotransformed with derivatives of plasmids pLEX10, to generate fusions of SseK1, SseK2, or SseK3 with the DNA-binding domain of LexA, and pGAD1318, to generate a fusion of TBCB with the activation domain of Gal4, as indicated. The interaction between the two hybrid proteins is shown by the growth in the absence of histidine (-His), and detection of blue color in the presence of X-Gal after a -galactosidase filter assay. 2.6. TBCB is not Glycosylated by SseK2 or SseK3 Next, we tested whether SseK2 or SseK3 were able to catalyze the glycosylation of TBCB. Western blot with anti-ArgGlcNAc antibody after the in vitro assay showed that TBCB was not glycosylated by SseK2 or SseK3, and only self-modification of SseK3 (but not SseK2) was detected (Figure 6). These results suggest that TBCB is specifically glycosylated by SseK1. Figure 6. SseK2 and SseK3 are unable to glycosylate TBCB in vitro. An assay for GlcNAc modification of TBCB in the presence of SseK2 or SseK3 using recombinant GST fusion proteins in the indicated combinations and 1 mM UDP-GlcNAc. SseK3AAA: SseK3(D226A/D228A). TBCBN: N-terminal fragment of TBCB (amino acids 1–125). TBCBC: C-terminal fragment of TBCB (amino acids 126–244). An immunoblot incubated with anti-Arg-GlcNAc is shown (upper panel). The same blot was reincubated with anti-GST (lower panel). Molecular weights in kDa are indicated on the left. 2.7. SseK1 Promotes Microtubule Stability TBCB, together with TBCE, plays a role in microtubule biosynthesis [17], and overexpression of these cofactors leads to microtubule depolymerization [18]. The results presented above indicate that TBCB may interact with and be modified by SseK1, opening the possibility that this Salmonella effector may have an impact on microtubule dynamics. To test this hypothesis, we transfected HEK293T cells Figure 5. SseK3 interacts with TBCB in the yeast two-hybrid system. Strain L40 of S. cerevisiae was cotransformed with derivatives of plasmids pLEX10, to generate fusions of SseK1, SseK2, or SseK3 with the DNA-binding domain of LexA, and pGAD1318, to generate a fusion of TBCB with the activation domain of Gal4, as indicated. The interaction between the two hybrid proteins is shown by the growth in the absence of histidine (-His), and detection of blue color in the presence of X-Gal after a β-galactosidase filter assay. 2.6. TBCB is not Glycosylated by SseK2 or SseK3 Next, we tested whether SseK2 or SseK3 were able to catalyze the glycosylation of TBCB. Western blot with anti-ArgGlcNAc antibody after the in vitro assay showed that TBCB was not glycosylated by SseK2 or SseK3, and only self-modification of SseK3 (but not SseK2) was detected (Figure 6). These results suggest that TBCB is specifically glycosylated by SseK1. Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 7 of 17 Figure 5. SseK3 interacts with TBCB in the yeast two-hybrid system. Strain L40 of S. cerevisiae was cotransformed with derivatives of plasmids pLEX10, to generate fusions of SseK1, SseK2, or SseK3 with the DNA-binding domain of LexA, and pGAD1318, to generate a fusion of TBCB with the activation domain of Gal4, as indicated. The interaction between the two hybrid proteins is shown by the growth in the absence of histidine (-His), and detection of blue color in the presence of X-Gal after a -galactosidase filter assay. 2.6. TBCB is not Glycosylated by SseK2 or SseK3 Next, we tested whether SseK2 or SseK3 were able to catalyze the glycosylation of TBCB. Western blot with anti-ArgGlcNAc antibody after the in vitro assay showed that TBCB was not glycosylated by SseK2 or SseK3, and only self-modification of SseK3 (but not SseK2) was detected (Figure 6). These results suggest that TBCB is specifically glycosylated by SseK1. Figure 6. SseK2 and SseK3 are unable to glycosylate TBCB in vitro. An assay for GlcNAc modification of TBCB in the presence of SseK2 or SseK3 using recombinant GST fusion proteins in the indicated combinations and 1 mM UDP-GlcNAc. SseK3AAA: SseK3(D226A/D228A). TBCBN: N-terminal fragment of TBCB (amino acids 1–125). TBCBC: C-terminal fragment of TBCB (amino acids 126–244). An immunoblot incubated with anti-Arg-GlcNAc is shown (upper panel). The same blot was reincubated with anti-GST (lower panel). Molecular weights in kDa are indicated on the left. 2.7. SseK1 Promotes Microtubule Stability TBCB, together with TBCE, plays a role in microtubule biosynthesis [17], and overexpression of these cofactors leads to microtubule depolymerization [18]. The results presented above indicate that TBCB may interact with and be modified by SseK1, opening the possibility that this Salmonella effector may have an impact on microtubule dynamics. To test this hypothesis, we transfected HEK293T cells Figure 6. SseK2 and SseK3 are unable to glycosylate TBCB in vitro . An assay for GlcNAc modification of TBCB in the presence of SseK2 or SseK3 using recombinant GST fusion proteins in the indicated combinations and 1 mM UDP-GlcNAc. SseK3 AAA : SseK3(D226A/D228A). TBCB N : N-terminal fragment of TBCB (amino acids 1–125). TBCB C : C-terminal fragment of TBCB (amino acids 126–244). An immunoblot incubated with anti-Arg-GlcNAc is shown (upper panel). The same blot was reincubated with anti-GST (lower panel). Molecular weights in kDa are indicated on the left. 2.7. SseK1 Promotes Microtubule Stability TBCB, together with TBCE, plays a role in microtubule biosynthesis [ 17 ], and overexpression of these cofactors leads to microtubule depolymerization [ 18 ]. The results presented above indicate that Int. J. Mol. Sci. 2020,21, 3193 8 of 18 TBCB may interact with and be modified by SseK1, opening the possibility that this Salmonella effector may have an impact on microtubule dynamics. To test this hypothesis, we transfected HEK293T cells with plasmids expressing GFP-SseK1 or GFP alone, and we measured the acetylation level of α -tubulin by fluorescence microscopy since acetylation of α -tubulin correlates with microtubule stability [ 19 – 21 ]. The expression of SseK1 significantly increased the level of acetylated microtubules (Figure 7). Importantly, this effect was significantly reduced when cells were transfected with a plasmid expressing GFP-SseK1 AAA , the catalytic mutant of this effector, suggesting that its GlcNAc transferase activity was involved in the alteration of the microtubule dynamics of the host cell. Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 8 of 17 with plasmids expressing GFP-SseK1 or GFP alone, and we measured the acetylation level of - tubulin by fluorescence microscopy since acetylation of -tubulin correlates with microtubule stability [19–21]. The expression of SseK1 significantly increased the level of acetylated microtubules (Figure 7). Importantly, this effect was significantly reduced when cells were transfected with a plasmid expressing GFP-SseK1AAA, the catalytic mutant of this effector, suggesting that its GlcNAc transferase activity was involved in the alteration of the microtubule dynamics of the host cell. Figure 7. SseK1 increases the acetylation level of tubulin in HEK293T cells. (A) Representative fluorescence microscopy images showing GFP (Control) or GFP fusion proteins (GFP-SseK1 or GFP- SseK1AAA,) in green, and acetyl--tubulin in red. HEK293T cells were transfected with derivatives of plasmid pEGFPC1, and 48 h after transfection cells were fixed with 4% paraformaldehyde and stained with anti-acetyl--tubulin rabbit monoclonal antibody and Alexa Fluor Plus 594-conjugated antirabbit secondary antibody. (B) Quantification of acetyl--tubulin from immunofluorescence images of cells transfected with GFP (Control) or GFP fusion proteins (GFP-SseK1, GFP-SseK1AAA). The mean and standard deviation from 100 cells are represented for each condition. RFU, relative fluorescence units. ***, P < 0.001; ****, P < 0.0001, for the indicated comparisons using a Kolmogorov–Smirnov test. Figure 7. SseK1 increases the acetylation level of tubulin in HEK293T cells. ( A ) Representative fluorescence microscopy images showing GFP (Control) or GFP fusion proteins (GFP-SseK1 or GFP-SseK1 AAA ,) in green, and acetylα -tubulin in red. HEK293T cells were transfected with derivatives of plasmid pEGFPC1, and 48 h after transfection cells were fixed with 4% paraformaldehyde and stained with anti-acetyl- α -tubulin rabbit monoclonal antibody and Alexa Fluor Plus 594-conjugated anti-rabbit secondary antibody. ( B ) Quantification of acetylα -tubulin from immunofluorescence images of cells transfected with GFP (Control) or GFP fusion proteins (GFP-SseK1, GFP-SseK1 AAA ). The mean and standard deviation from 100 cells are represented for each condition. RFU, relative fluorescence units. ***, P<0.001; ****, P<0.0001, for the indicated comparisons using a Kolmogorov–Smirnov test. Int. J. Mol. Sci. 2020,21, 3193 9 of 18 3. Discussion As mentioned in the Introduction, there are conflicting reports about the binding partners and preferred substrates of effectors belonging to the NleB/SseK family, suggesting that different methodologies are necessary to identify all possible host targets. Our approach to finding a new host substrate for the catalytic activity of SseK1 was based on a yeast two-hybrid screen, a useful technique used to identify protein interactions in vivo . Using that approach, we report here the identification of the tubulin-binding cofactor TBCB as a new binding partner for the Salmonella type III secretion effector SseK1. In addition, our in vitro and in vivo experiments demonstrated that TBCB is a substrate of the arginine-glycosylation activity of SseK1. Besides TBCB, our screen also identified other putative interacting partners of SseK1. However, these interactions need additional confirmation using independent techniques. Identification of binding partners for an effector, even if they are not substrates of its catalytic activity, as in the case of TRIM32 and SseK3, may also be important to understand as fully as possible the function of these virulence factors. SseK1, SseK2, and SseK3 constitute a family of highly similar Salmonella effectors. SseK1 is 60% identical in amino acid sequence to SseK2 and SseK3, whereas SseK2 and SseK3 are 75% identical. The three proteins share a catalytic domain containing a DxD motif that is essential for the enzymatic function of most glycosyltransferases of the GT-A family [ 22 ]. Therefore, we tested if TBCB was also a substrate for SseK2 and SseK3. We detected a stable interaction between TBCB and SseK3. However, TBCB was not Arg-GlcNAcylated in the presence of this effector. This is in line with previous reports showing that there is some overlap, but also differential substrate specificity for these effectors [ 11 , 14 ] and that stable binding and catalytic activity are distinct aspects of the NleB/SseK family of effectors that are not always coupled. For instance, SseK1 is able to bind to and glycosylate glyceraldehyde 3-phosphate dehydrogenase (GAPDH) [ 12 ]; in contrast, SseK3 binds to TRIM32 [ 13 ] but is unable to promote its arginine GlcNAcylation [11]. Interestingly, SseK2 was unable to bind or modify TBCB. Although there is a report showing that SseK2 glycosylates FADD [ 12 ], other studies suggested that this effector lacks Arg-GlcNAc transferase activity [ 11 , 14 ]. Our results support the second possibility since we did not detect the apparent self-modification that was observed for SseK1 and SseK3. SseK2 might catalyze a different modification that is not detected by the anti-Arg-GlcNAc antibody, as suggested previously [11]. The best-characterized function of effectors of the NleB/SseK family is the inhibition of NF- κ B signaling and of host cell death through the modification of death domain-containing proteins [ 9 – 11 , 14 , 23 , 24 ]. However, other ligands of NleB/SseK effectors have been reported [ 8 ]. For instance, C. rodentium NleB [ 25 ], enterohemorrhagic E. coli NleB1, and S. enterica SseK1 [ 12 ] are able to bind and glycosylate GAPDH. Glycosylation of GAPDH would disrupt its interaction with TRAF2 and TRAF3, leading to inhibition of NF- κ B and type I IFN signaling [ 25 , 26 ]. A totally different role has been suggested for NleB1 from enteropathogenic E. coli through its interaction and modification of the α subunit of the hypoxia-inducible factor (HIF-1 α ) [ 27 ]. HIF-1 is a master regulator of cellular and systemic homeostatic response to hypoxia that activates transcription of genes involved in energy metabolism, angiogenesis, and apoptosis [ 28 ]. GlcNAcylation of HIF-1 α enhances its transcriptional activity, leading to changes in host glucose metabolism. NleB1 also GlcNAcylated arginine on HIF-2 α but did not enhance HIF-2 α transcriptional activity [27]. NleB/SseK effectors also have a number of host binding partners whose GlcNAcylation has not been demonstrated. For instance, as mentioned above, TRIM32, the first binding partner described for SseK3 [ 13 ], is an E3 ubiquitin ligase that is not glycosylated by SseK3 [ 13 ] and is not required for the inhibition of NF- κ B signaling during the infection of macrophages [ 11 ]. Human proteins DRG2, LRRC18, and POLR2E were found as partners of enterohemorrhagic E. coli NleB1 in a yeast two-hybrid screen [ 29 ], but these interactions were not confirmed by independent methods. A novel binding partner of this effector is MAP7 (ensconsin), which was identified by a quantitative proteomic approach and whose interaction was confirmed by coimmunoprecipitation [30]. Int. J. Mol. Sci. 2020,21, 3193 16 of 18 14. Newson, J.P.; Scott, N.E.; Yeuk Wah Chung, I.; Wong Fok Lung, T.; Giogha, C.; Gan, J.; Wang, N.; Strugnell, R.; Brown, N.F.; Cygler, M.; et al. Salmonella effectors SseK1 and SseK3 target death domain proteins in the TNF and TRAIL signaling pathways. Mol. Cell. Proteom. 2019,18, 1138–1156. [CrossRef] 15. Bais ó n-Olmo, F.; Galindo-Moreno, M.; Ramos-Morales, F. Host cell type-dependent translocation and PhoP-mediated positive regulation of the effector SseK1 of Salmonella enterica.Front. Microbiol. 2015,6, 396. [CrossRef] [PubMed] 16. Pan, M.; Li, S.; Li, X.; Shao, F.; Liu, L.; Hu, H.-G. Synthesis of and specific antibody generation for glycopeptides with arginine N-GlcNAcylation. Angew. Chem. Int. Ed. Engl. 2014,53, 14517–14521. [CrossRef] [PubMed] 17. Carranza, G.; Castaño, R.; Fanarraga, M.L.; Villegas, J.C.; Gonçalves, J.; Soares, H.; Avila, J.; Marenchino, M.; Campos-Olivas, R.; Montoya, G.; et al. Autoinhibition of TBCB regulates EB1-mediated microtubule dynamics. Cell. Mol. Life Sci. 2013,70, 357–371. [CrossRef] 18. Kortazar, D.; Fanarraga, M.L.; Carranza, G.; Bellido, J.; Villegas, J.C.; Avila, J.; Zabala, J.C. Role of cofactors B (TBCB) and E (TBCE) in tubulin heterodimer dissociation. Exp. Cell Res. 2007,313, 425–436. [CrossRef] 19. Schulze, E.; Asai, D.J.; Bulinski, J.C.; Kirschner, M. Posttranslational modification and microtubule stability. J. Cell Biol. 1987,105, 2167–2177. [CrossRef] 20. Webster, D.R.; Borisy, G.G. Microtubules are acetylated in domains that turn over slowly. J. Cell Sci. 1989 ,92 (Pt 1), 57–65. 21. Tan, H.; Liao, H.; Zhao, L.; Lu, Y.; Jiang, S.; Tao, D.; Liu, Y.; Ma, Y. HILI destabilizes microtubules by suppressing phosphorylation and Gigaxonin-mediated degradation of TBCB. Sci. Rep. 2017 ,7, 46376. [CrossRef] [PubMed] 22. Lairson, L.L.; Henrissat, B.; Davies, G.J.; Withers, S.G. Glycosyltransferases: Structures, Functions, and Mechanisms. Annu. Rev. Biochem. 2008,77, 521–555. [CrossRef] 23. Pearson, J.S.; Hartland, E.L. A surprising sweetener from enteropathogenic Escherichia coli.Gut Microbes 2014,5, 766–769. [CrossRef] 24. Scott, N.E.; Giogha, C.; Pollock, G.L.; Kennedy, C.L.; Webb, A.I.; Williamson, N.A.; Pearson, J.S.; Hartland, E.L. The bacterial arginine glycosyltransferase effector NleB preferentially modifies Fas-associated death domain protein (FADD). J. Biol. Chem. 2017,292, 17337–17350. [CrossRef] [PubMed] 25. Gao, X.; Wang, X.; Pham, T.H.; Feuerbacher, L.A.; Lubos, M.-L.; Huang, M.; Olsen, R.; Mushegian, A.; Slawson, C.; Hardwidge, P.R. NleB, a bacterial effector with glycosyltransferase activity, targets GAPDH function to inhibit NF-κB activation. Cell Host Microbe 2013,13, 87–99. [CrossRef] [PubMed] 26. Gao, X.; Pham, T.H.; Feuerbacher, L.A.; Chen, K.; Hays, M.P.; Singh, G.; Rueter, C.; Hurtado-Guerrero, R.; Hardwidge, P.R. Citrobacter rodentium NleB Protein Inhibits Tumor Necrosis Factor (TNF) Receptor-associated Factor 3 (TRAF3) Ubiquitination to Reduce Host Type I Interferon Production. J. Biol. Chem. 2016 ,291, 18232–18238. [CrossRef] [PubMed] 27. Xu, C.; Liu, X.; Zha, H.; Fan, S.; Zhang, D.; Li, S.; Xiao, W. A pathogen-derived effector modulates host glucose metabolism by arginine GlcNAcylation of HIF-1αprotein. PLoS Pathog. 2018,14, e1007259. [CrossRef] 28. Hirota, K. Basic biology of hypoxic responses mediated by the transcription factor HIFs and its implication for medicine. Biomedicines 2020,8, 32. [CrossRef] 29. Blasche, S.; Arens, S.; Ceol, A.; Siszler, G.; Schmid, M.A.; Häuser, R.; Schwarz, F.; Wuchty, S.; Aloy, P.; Uetz, P.; et al. The EHEC-host interactome reveals novel targets for the translocated intimin receptor. Sci. Rep. 2014 ,4, 7531. [CrossRef] 30. Law, R.J.; Law, H.T.; Scurll, J.M.; Scholz, R.; Santos, A.S.; Shames, S.R.; Deng, W.; Croxen, M.A.; Li, Y.; de Hoog, C.L.; et al. Quantitative Mass Spectrometry Identifies Novel Host Binding Partners for Pathogenic Escherichia coli Type III Secretion System Effectors. J. Proteome Res. 2016,15, 1613–1622. [CrossRef] 31. El Qaidi, S.; Scott, N.E.; Hays, M.P.; Geisbrecht, B.V.; Watkins, S.; Hardwidge, P.R. An intra-bacterial activity for a T3SS effector. Sci. Rep. 2020,10, 1073. [CrossRef] [PubMed] 32. Lopez-Fanarraga, M.; Avila, J.; Guasch, A.; Coll, M.; Zabala, J.C. Review: Postchaperonin tubulin folding cofactors and their role in microtubule dynamics. J. Struct. Biol. 2001,135, 219–229. [CrossRef] [PubMed] 33. Lundin, V.F.; Leroux, M.R.; Stirling, P.C. Quality control of cytoskeletal proteins and human disease. Trends Biochem. Sci. 2010,35, 288–297. [CrossRef] [PubMed] 34. Fontalba, A.; Paciucci, R.; Avila, J.; Zabala, J.C. Incorporation of tubulin subunits into dimers requires GTP hydrolysis. J. Cell Sci. 1993,106, 627–632. Int. J. Mol. Sci. 2020,21, 3193 17 of 18 35. Tian, G.; Huang, Y.; Rommelaere, H.; Vandekerckhove, J.; Ampe, C.; Cowan, N.J. Pathway leading to correctly folded β-tubulin. Cell 1996,86, 287–296. [CrossRef] 36. Fanarraga, M.L.; Villegas, J.C.; Carranza, G.; Castaño, R.; Zabala, J.C. Tubulin cofactor B regulates microtubule densities during microglia transition to the reactive states. Exp. Cell Res. 2009,315, 535–541. [CrossRef] 37. Wang, W.; Ding, J.; Allen, E.; Zhu, P.; Zhang, L.; Vogel, H.; Yang, Y. Gigaxonin interacts with tubulin folding cofactor B and controls its degradation through the ubiquitin-proteasome pathway. Curr. Biol. 2005 ,15, 2050–2055. [CrossRef] 38. Allen, E.; Ding, J.; Wang, W.; Pramanik, S.; Chou, J.; Yau, V.; Yang, Y. Gigaxonin-controlled degradation of MAP1B light chain is critical to neuronal survival. Nature 2005,438, 224–228. [CrossRef] 39. Ding, J.; Liu, J.J.; Kowal, A.S.; Nardine, T.; Bhattacharya, P.; Lee, A.; Yang, Y. Microtubule-associated protein 1B: A neuronal binding partner for gigaxonin. J. Cell Biol. 2002,158, 427–433. [CrossRef] 40. Zhang, D.D.; Lo, S.C.; Sun, Z.; Habib, G.M.; Lieberman, M.W.; Hannink, M. Ubiquitination of Keap1, a BTB-Kelch substrate adaptor protein for Cul3, targets Keap1 for degradation by a proteasome-independent pathway. J. Biol. Chem. 2005,280, 30091–30099. [CrossRef] 41. Vadlamudi, R.K.; Barnes, C.J.; Rayala, S.; Li, F.; Balasenthil, S.; Marcus, S.; Goodson, H.V.; Sahin, A.A.; Kumar, R. p21-Activated Kinase 1 Regulates Microtubule Dynamics by Phosphorylating Tubulin Cofactor B. Mol. Cell. Biol. 2005,25, 3726–3736. [CrossRef] [PubMed] 42. Serna, M.; Carranza, G.; Mart í n-Benito, J.; Janowski, R.; Canals, A.; Coll, M.; Zabala, J.C.; Valpuesta, J.M. The structure of the complex between α -tubulin, TBCE and TBCB reveals a tubulin dimer dissociation mechanism. J. Cell Sci. 2015,128, 1824–1834. [CrossRef] [PubMed] 43. Gadadhar, S.; Bodakuntla, S.; Natarajan, K.; Janke, C. The tubulin code at a glance. J. Cell Sci. 2017 ,130, 1347–1353. [CrossRef] [PubMed] 44. de Souza Santos, M.; Orth, K. Subversion of the cytoskeleton by intracellular bacteria: Lessons from Listeria, Salmonella and Vibrio.Cell. Microbiol. 2015,17, 164–173. [CrossRef] [PubMed] 45. Van Nhieu, G.T.; Romero, S. Common Themes in Cytoskeletal Remodeling by Intracellular Bacterial Effectors. Handb. Exp. Pharmacol. 2017,235, 207–235. 46. Ohlson, M.B.; Huang, Z.; Alto, N.M.; Blanc, M.; Dixon, J.E.; Chai, J.; Miller, S.I. Structure and function of Salmonella SifA indicate that its interactions with SKIP, SseJ, and RhoA family GTPases induce endosomal tubulation. Cell Host Microbe 2008,4, 434–446. [CrossRef] 47. Raines, S.A.; Hodgkinson, M.R.; Dowle, A.A.; Pryor, P.R. The Salmonella effector SseJ disrupts microtubule dynamics when ectopically expressed in normal rat kidney cells. PLoS ONE 2017,12, e0172588. [CrossRef] 48. Palazzo, A.F.; Cook, T.A.; Alberts, A.S.; Gundersen, G.G. mDia mediates Rho-regulated formation and orientation of stable microtubules. Nat. Cell Biol. 2001,3, 723–729. [CrossRef] [PubMed] 49. Schmieger, H. Phage P22-mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 1972 , 119, 75–88. [CrossRef] 50. Maloy, S.R. Experimental Techniques in Bacterial Genetics; Jones and Bartlett: Burlington, MA, USA, 1990; ISBN 0867201185. 51. Hanahan, D. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 1983 ,166, 557–580. [CrossRef] 52. Boyer, H.W.; Roulland-Dussoix, D. A complementation analysis of the restriction and modification of DNA in Escherichia coli.J. Mol. Biol. 1969,41, 459–472. [CrossRef] 53. Vojtek, A.B.; Hollenberg, S.M.; Cooper, J.A. Mammalian Ras interacts directly with the serine/threonine kinase raf. Cell 1993,74, 205–214. [CrossRef] 54. Van Aelst, L. Two-hybrid analysis of Ras-Raf interactions. Methods Mol. Biol. 1998,84, 201–222. [PubMed] 55. Selig, L.; Benichou, S.; Rogel, M.E.; Wu, L.I.; Vodicka, M.A.; Sire, J.; Benarous, R.; Emerman, M. Uracil DNA glycosylase specifically interacts with Vpr of both human immunodeficiency virus type 1 and simian immunodeficiency virus of sooty mangabeys, but binding does not correlate with cell cycle arrest. J. Virol. 1997,71, 4842–4846. [CrossRef] [PubMed] 56. Sherman, F.; Fink, G.R.; Hicks, J.B. Laboratory Course Manual for Methods in Yeast Genetics; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, USA, 1987; ISBN 0879691972. Int. J. Mol. Sci. 2020,21, 3193 18 of 18 57. Fuxman Bass, J.I.; Reece-Hoyes, J.S.; Walhout, A.J.M. Colony Lift Colorimetric Assay for β -Galactosidase Activity. Cold Spring Harb. Protoc. 2016. [CrossRef] 58. Schneider, C.A.; Rasband, W.S.; Eliceiri, K.W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012,9, 671–675. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).