Bacterial Rhs proteins containing toxic domains are often secreted by type VI secretion systems (T6SSs) through unclear mechanisms. Here, we show that the T6SS Rhs-family effector TseI of Aeromonas dhakensis is subject to self-cleavage at both the N- and the C-terminus, releasing the middle Rhs core and two VgrG-interacting domains (which we name VIRN and VIRC). VIRC is an endonuclease, and the immunity protein TsiI protects against VIRC toxicity through direct interaction. Proteolytic release of VIRC and VIRN is mediated, respectively, by an internal aspartic protease activity and by two conserved glutamic residues in the Rhs core. Mutations abolishing self-cleavage do not block secretion, but reduce TseI toxicity. Deletion of VIRN or the Rhs core abolishes secretion. TseI homologs from Pseudomonas syringae, P. aeruginosa, and Vibrio parahaemolyticus are also self-cleaved. VIRN and VIRC interact with protein VgrG1, while the Rhs core interacts with protein TecI. We propose that VIRN and the Rhs core act as T6SS intramolecular chaperones to facilitate toxin secretion and function.
It is fascinating that some proteins manage to keep similar sequences and structural folding in both prokaryotic and eukaryotic cells despite billions of years of evolution1,2,3,4. One such example is the YD-repeat (tyrosine–aspartate) protein family whose members include many toxins in both Gram-positive and Gram-negative bacteria as well as eukaryotic teneurins that are conserved transmembrane adhesion proteins with critical functions in embryogenesis and neural development5,6,7,8,9. In bacteria, YD-repeat signatures are often found within the core domain of Rhs (rearrangement hot spot) proteins followed by a divergent C-terminal toxin domain; some Rhs also possess an additional N-terminal domain with unknown function7,10,11,12. Members of Rhs proteins include the antibacterial WapA in Gram-positive Bacillus subtilis13, the ABC insecticidal Tc-toxins7,14,15, and secreted effectors by the type VI secretion system (T6SS) in Gram-negative species10,11. Known T6SS Rhs effectors have been shown to be frequently associated with an N-terminal PAAR domain or downstream of VgrG and PAAR encoding genes8,10,11,16,17,18,19.
The T6SS plays a critical role in interspecies interaction and during infection by translocating toxic effectors to bacterial and eukaryotic host cells20,21,22,23. The needle-like T6SS resembles a contractile bacteriophage-like tail, consisting of a VipA/B outer sheath, an Hcp inner tube, and a transmembrane-baseplate complex24,25. The tip of the tube is “sharpened” by a spike complex made of a VgrG trimer and a PAAR protein26. The inner Hcp tube carries effectors out of the cell upon sheath contraction. T6SS effectors exhibit diverse toxicities against essential cellular targets in eukaryotic27,28 and prokaryotic cells29,30,31,32,33. Each antibacterial effector has a cognate immunity protein that confers self-protection29,30,33. Some effectors are Hcp, VgrG, and PAAR structural proteins with evolved C-terminal functional domains18,19,26,28,33. Non-structural effectors can be secreted by binding to the inner Hcp tube or to the tip VgrG/PAAR proteins10,19,33; the latter often involves chaperone proteins that are required for stabilization and delivery of effectors34,35,36,37,38.
We previously predicted a chaperone-dependent T6SS effector TseI in Aeromonas dhakensis, a waterborne pathogen associated with skin and soft-tissue infection, gastroenteritis, and bacteremia34,39,40. TseI, made of 1545 amino acids, represents one of the largest T6SS effectors. Here we show that TseI belongs to the Rhs/YD-repeat family and possesses two self-cleavage sites, between residues C420 and P421 and between residues L1433 and S1434, resulting in three fragments, VIRN (VgrG-interacting Rhs N terminus), Rhs core, and VIRC (VgrG-interacting Rhs C-terminus). Importantly, the cleaved products remain in complex through noncovalent interactions. VIRC encodes an endonuclease that degrades DNA and confers the bacterium-killing toxicity of TseI. Secretion of TseI also requires upstream encoded VgrG1 and a conserved DUF4123-domain chaperone TecI34. The Rhs core but not the VIRN/C domains binds to the TecI chaperone. The VIRN domain and the Rhs core are required for VIRC delivery and co-translocated by T6SS, suggesting they may function as secreted chaperones for VIRC. Our results demonstrate the T6SS secretion of self-cleaved Rhs effectors, intramolecular chaperones, and add important new insights in understanding the physiological roles of widespread Rhs homologs in bacteria.
TseI–TsiI as an effector–immunity pair
The tseI gene is located in a multi-gene operon containing three upstream genes encoding two T6SS structural proteins Hcp1 and VgrG1, and a chaperone TecI (Fig. 1a). We also predicted a previously unannotated gene that we name tsiI, whose start codon overlaps with the last amino acid codon and the stop codon of TseI. To determine the function of TseI, we first analyzed its protein sequence using Phyre2 and blastp41,42. TseI comprises three distinct regions, an N terminus of unknown function, a middle Tc-toxin domain, and a C-terminal Tox-HNH-EHHH domain (Fig. 1a). The Phyre-predicted Tc-toxin domain contains 838 residues with 100% confidence and 20% identity with TccC3 in Photorhabuds luminescens (PDB entry 4O9X)14. The Tc-toxin domain also contains multiple Rhs/YD-repeat signatures. The C-terminal domain of TseI belongs to the predicted HNH Endonuclease VII toxin superfamily (pfam15657) with conserved [ED]H motif and two histidine residues (Supplementary Fig. 1).
To test if tseI and tsiI encode a T6SS-dependent effector–immunity pair23,29, we constructed a deletion mutant, ∆tseIctsiI, lacking the predicted tseI functional region including the toxin-coding sequence and the tsiI gene. Competition assay shows that the ∆tseIctsiI mutant was efficiently outcompeted by wild type but not by the T6SS-null ∆vasK mutant (Fig. 1b). Survival of the ∆tseIctsiI was restored when complemented with a plasmid-borne tsiI (Fig. 1b).
We next determined whether the C-terminal domain confers the observed toxicity. Sequence alignment shows that it contains several highly conserved predicted catalytic residues (Fig. 1a). We mutated two predicted catalytic sites, histidine 1497 and residues 1507–1509, histidine, phenylalanine, and histidine, to alanine residues (H1497A and HFH-AAA). Wild-type TseI and its two mutants were expressed using an arabinose-inducible vector pBAD in E. coli43. Survival of E. coli was severely reduced in wild-type samples compared with the mutants, indicating that H1497A and HFH-AAA abolished TseI toxicity (Fig. 1c). Western blot analysis confirmed that the nontoxic mutants were expressed (Supplementary Fig. 2A). In addition, these two corresponding chromosomal tseI mutants failed to outcompete the ∆tseIctsiI mutant (Fig. 1d; Supplementary Fig. 2B).
We then sought to test whether the predicted HNH Endonuclease VII domain possesses endonuclease activities in vitro. To obtain sufficient amount of wild-type TseI and overcome its toxicity during expression, we co-expressed His-tagged TseI and untagged immunity TsiI and then purified TseI under denaturing conditions, followed by regeneration (Fig. 1e; Supplementary Fig. 2C). Nontoxic mutants H1497A and HFH-AAA were purified under the same condition. Results show that only wild-type TseI degraded plasmid DNA, confirming that the conserved H1497 and HFH (1507–1509) are critical residues for the endonuclease activity (Fig. 1e). We used H1497A and HFH-AAA interchangeably hereafter in this study.
The nontoxic H1497A mutation also enabled us to test the interaction of TseI and TsiI using the bacterial two-hybrid assay. We constructed C-terminal fusions of TseI H1497A and its immunity protein TsiI with the two split fragments T25 (224 amino acids) and T18 (175 amino acids) of the Bordetella pertussis adenylate cyclase (CyaA)44, respectively. Results show that chimeric TseI and TsiI but not the T6SS transcriptional regulator VasH could reconstitute CyaA activity (Fig. 1f). Collectively, these results demonstrate that TseI is a T6SS endonuclease effector and TsiI is the cognate immunity protein that confers protection against TseI through direct interaction.
TseI is cleaved into three fragments
While purifying His-tagged wild type and mutant TseI, we noticed three cleaved products (Fig. 2a; Supplementary Figs. 2C and 3A). Using N-terminal amino acid sequencing, we determined that TseI is cleaved after residues cysteine 420 and leucine 1433 (Fig. 2a). These three resulting fragments correspond to the predicted Rhs sequence (116 kDa), the N-terminal domain (45 kDa), and the C-terminal toxicity domain (observed size 18 kDa, predicted size 13 kDa, pI 9.57). We also analyzed each excised protein band using LC-MS analysis and the results confirmed the identity of each fragment (Supplementary Fig. 3A). Interestingly, the three cleaved fragments were co-eluted regardless of the His-tag position, suggesting they remain in complex post cleavage (Supplementary Fig. 3B).
Mutating conserved residues blocks N- and C-terminal cleavage
To understand the cleavage mechanism, we first used a bioinformatics approach to identify conserved residues. Using blastp, we found over >1000 TseI highly similar homologs (E-value = 0, identity > 45%) in the NCBI non-redundant protein database, primarily in the two genera Aeromonas and Pseudomonas (Supplementary Fig. 4). To reduce redundancy, we selected a set of 48 representative sequences with one from each species from the top 1000 hits (Supplementary Data 1). Sequence alignment shows that a highly conserved signature “PVSMVTGEELL”, located right after the N-terminal cleavage site and at the beginning of the Rhs core, is present in all 48 representative sequences (Fig. 2b). The C-terminal sequence also contains a conserved sequence “TPDPxxLAGGxNxYxYxPNPTGWVDPLGL” right before the C-terminal cleavage site. Comparison of TseI with the known self-cleavable Tc-toxin TccC3 shows that the C-terminal cleavage sites are identical and the consensus signature resembles the internal aspartyl protease found in TccC3 (Fig. 2b; Supplementary Fig. 3C)14.
To test the effect of these identified conserved residues on TseI cleavage, we constructed a series of single and combinatorial point mutations changing the two conserved C-terminal protease catalytic residues D1407 and D1429, and the conserved N-terminal residues E428, E429, and D435 to alanine in the nontoxic mutants HFH-AAA or H1497A backgrounds. We then examined the cleavage of those mutants expressed in cells by western blot analysis as well as purified proteins by SDS-PAGE analysis (Fig. 2c; Supplementary Fig. 5A, B). Results collectively show that mutations D1407A and D1429A abolished C-terminal cleavage while E428A and E429A but not D435A abolished N-terminal cleavage. Double mutations of D1407A with E428A or E429A resulted in full-length non-cleaved TseI (Fig. 2c). To eliminate the remote possibility that the C-terminal endonuclease inactivation by the HFH-AAA mutation is responsible for cleavage, we also constructed cleavage mutants of wild-type TseI and expressed them in the ∆tseI mutant in which the chromosomal-encoded immunity TsiI confers protection from TseI toxicity. Western blot analysis of expressed proteins shows results consistent with the mutant cleavage phenotypes (Supplementary Fig. 5C, D).
Next, we tested whether TseI cleavage is dependent on any external protease. We used a commercial in vitro protein synthesis kit with defined purified components to express TseI variants flanked by N-FLAG and C-3V5 epitope tags. Western blot analysis shows that TseI was cleaved similarly in vitro, and mutations of D1407A and E428A also abolished cleavage, suggesting that TseI is self-cleaved (Fig. 2d). Using the anti-FLAG antibody, we did not detect a distinct band corresponding to the cleaved N terminus or full-length TseI in parental and D1407A samples. Instead, we found multiple nonspecific signals across all samples which likely result from lower stability of TseI N terminus under the in vitro expression condition (Fig. 2c, d and source data).
Notably, although N-terminal cleavage is still detectable in the C-terminal cleavage-defective mutants, the amount of full-length TseI was substantially enriched in comparison with wild type (Fig. 2c, d; Supplementary Fig. 5A, C), suggesting that the C-terminal protease-inactivating mutations reduced the efficiency of N-terminal cleavage.
Cleavage is critical for TseI-mediated competition
We next tested if self-cleavage is important for T6SS-dependent delivery of TseI. Using the immunity-defective mutant ∆tseIctsiI as prey and the ∆tseI mutant complemented with plasmid-borne TseI and its N- or C-cleavage-defective mutants as killer, we found that cleavage is critical for the killing of ∆tseIctsiI (Fig. 3a). Control experiments testing protein expression and survival of killer strains ruled out the possibilities that the increased prey survival is due to poor expression of those plasmid-borne constructs (Supplementary Fig. 6A) or impaired growth of cleavage-defective killers during the competition assays (Supplementary Fig. 6B). In fact, despite that the E429A mutant killer showed a log more growth than the wild type or other mutant killer strains (Supplementary Fig. 6B), it was still impaired in killing the ∆tseIctsiI prey.
To provide further evidence, we also constructed chromosomal mutants defective in cleavage, D1407A, D1429A, E428A, and E429A. Competition assays against the ∆tseIctsiI prey confirm that cleavage is critical for TseI-mediated cell-to-cell competition (Fig. 3b). There was no difference in the survival of killer strains (Supplementary Fig. 6C).
Non-cleaved TseI mutants are secreted to the extracellular medium
Next, we tested the effect of cleavage on TseI secretion by expressing C-terminal 3V5-tagged wild type and cleavage-defective mutant TseI in the ∆tseI and the T6SS-null ∆vasK mutants. Using antisera to the C-terminal V5 tag, Rhs, and N terminus respectively, we detected full-length TseI and N-cleaved TseI (Rhs+C) in the secreted samples of the D1407A and the D1429A mutants, while we detected the cleaved C and the Rhs fragments in wild-type samples (Fig. 3c). Secretion of N was ambiguous using the custom α-N antibody due to nonspecific signals but was later confirmed using a plasmid-borne 3V5-tagged N (Supplementary Fig. 6D). Similarly, T6SS-dependent secretion of the cleaved C terminus was detected in all samples expressing wild-type TseI, the E429A, and the D435A mutant TseI (Fig. 3d). Secretion of the N terminus and the Rhs was found in wild type and D435A samples while secretion of the N+Rhs fragment was detected in the E429A mutant sample. Full-length non-cleaved TseI mutants carrying double mutations of E429A D1407A or D435A D1407A were also secreted (Supplementary Fig. 6E). Detection of Hcp, the inner tube of the T6SS, and RpoB, the RNA polymerase subunit B, serves as indicators for T6SS functions and cell lysis, respectively.
To eliminate the possibility that plasmid-borne expression might have unexpected effects on TseI secretion, we also tested the effect of chromosomal mutations D1407A, D1429A, and constructed the ΔRL40 mutant by deleting the whole C-terminal internal protease lacking 40 amino acids from arginine 1394 to leucine 1433, respectively. Using the antiserum to the middle Rhs domain, we detected secretion of full-length TseI and the fragment corresponding to TseI lacking the N terminus (Rhs + C) in mutants D1407A and D1429A, further supporting that full-length TseI is secreted regardless of cleavage (Fig. 3e). Surprisingly, the ΔRL40 deletion abolished not only C-terminal cleavage but also secretion of the mutant TseI and reduced the amount of the Rhs+C fragment (Fig. 3e). The latter two effects might result from the loss of protease activities or indirectly from nonspecific effects of deletion. To confirm that the internal C-terminal protease is not required for N-terminal cleavage, we performed SDS-PAGE analysis of purified ΔRL40 TseIHFH-AAA in comparison with TseIHFH-AAA. Results show that the ΔRL40 mutant seemed to be less stable than its parental but the Rhs+C fragment lacking N was readily detectable in the ΔRL40, indicative of N-terminal cleavage (Supplementary Fig. 6F).
Cleavage is important for TseI toxicity
To understand how non-cleaved TseI mutants were secreted but severely impaired in outcompeting the ∆tseIctsiI mutant, we next tested whether cleavage affects TseI toxicity by comparing the survival of E. coli expressing arabinose-inducible wild type and cleavage-defective TseI mutants (Fig. 3f). The relative survival of cells between induced and uninduced conditions show that mutations D1407A and E429A attenuated toxicity in comparison with wild-type TseI, although both TseI D1407A and E429A mutants also exhibited moderate toxicities in comparison with the nontoxic HFH-AAA mutant (Fig. 3f). Considering that the physiological level of T6SS-delivered TseI is likely much lower than that of intracellular induction, such attenuated toxicity might account for the impaired killing of prey cell by cleavage mutants during competition.
TseI secretion requires VgrG1 and chaperone TecI through direct interaction
The genes upstream of tseI encode VgrG1 and a chaperone protein TecI (Fig. 1a). To test if TseI secretion is dependent on VgrG1 and TecI, we made deletion mutants of these two genes. Both mutants failed to secrete TseI or outcompete the ∆tseIctsiI mutant, suggesting that VgrG1 and TecI are required for TseI secretion (Fig. 4a, b; Supplementary Fig. 7A). We then tested if VgrG1 and TecI directly interact with TseI. Pull-down analysis shows that VgrG1 could pull down both TecI and TseI separately or together (Fig. 4c). Because TseI was detected against its C-terminal 3V5 tag, we could only detect its cleaved C-terminus but not full-length TseI probably due to efficient self-cleavage. As protein purification assay shows that the cleaved products remain in complex (Supplementary Fig. 3B), we then tested the interaction of each TseI fragment with VgrG1 and TecI, respectively. VgrG1 could interact with the N-terminal and C-terminal of TseI but not the middle Rhs core (Fig. 4d). We thus name the N and C terminus VIRN and VIRC, respectively. In contrast, TecI was found to interact with the middle Rhs core and full-length TseI but not the VIRN or the VIRC of TseI (Fig. 4e). Interaction of TecI with the full-length TseI might result from the intramolecular interaction of TseI domains since both VIRN (N) and Rhs (M) fragments can interact with the VIRC (C) toxin directly (Fig. 4f, g). This is consistent with the observation that all three fragments of TseI were co-eluted when either an N-terminal or a C-terminal 6His-tag was used (Supplementary Fig. 3B).
VIRN and Rhs are required for toxin secretion
Next, we tested whether the N-terminal VIRN and the Rhs domains are important for the secretion of the C-terminal VIRC toxin by expressing plasmid-borne full-length TseI, the N terminus deletion mutant (MC), and the VIRC domain (C) only mutant in the ΔtseI and the T6SS-null ΔvasK mutants. Western blot analysis of whole cell and secreted samples revealed that MC was cleaved but not secreted suggesting VIRN is required for TseI secretion but not for its cleavage (Fig. 4h). VIRC alone was not secreted either (Fig. 4h). Consistent with this result, only full-length TseI but not its truncated mutants could functionally complement the ΔtseI mutant in a competition assay against the ∆tseIctsiI immunity defective mutant (Fig. 4i; Supplementary Fig. 7B). Notably, the VIRN and the Rhs domain are not required for toxicity since intercellularly expressed MC and C were toxic to E. coli (Supplementary Fig. 7C). Unlike the VIRN domain (Supplementary Fig. 6D), Rhs was not secreted when it was expressed alone (Supplementary Fig. 7D). Expression of the nontoxic H1497A VIRC domain alone was barely detected in the cell lysate suggesting VIRC is unstable and prone to degradation or insoluble in the absence of VIRN and Rhs (Supplementary Fig. 7E).
N- and C-terminal cleavage of TseI homologs
Because Rhs/YD-repeat family contains over 120,000 members, we next asked whether other Rhs homologs are also subject to N- or C-terminal cleavage. We first used sequence alignment to compare TseI with several published homologs including the Tc-toxin TccC3, T6SS Rhs effectors in Dickeya dadantii, Pseudomonas aeruginosa, Serratia marcescens, Vibrio parahaemolyticus, WapA in Bacillus subtilis7,10,11,45,46, as well as newly predicted ones in Pseudomonas syringae and Vibrio aerogenes (Supplementary Data 2). TseI is clustered with P. syringae homolog PSPTO_5438 and divergent from the previously published Rhs effectors (Fig. 5a). Domain analysis shows that TseI homologs are featured with a conserved Rhs core flanked by divergent N- and C-terminal sequences (Supplementary Fig. 8). The N terminus sequences show variable lengths, with some possessing the T6SS-associated PAAR motif26. The N terminus cleavage site including the residues that affect cleavage is conserved in some but not all homologs while the C-terminal cleavage-activity residues are highly conserved except for RhsA in Escherichia coli strain EDL933 (Fig. 5b).
We next tested the cleavage of selected homologs, P. syringae PSPTO_5438, P. aeruginosa PA2684, and V. parahaemolyticus VP1517. Expression of PSPTO_5438 was highly toxic to E. coli unless its downstream gene PSPTO_5439 was co-expressed, indicative of an effector–immunity pair (Fig. 5c). Purified PSPTO_5438 exhibited three bands corresponding to the predicted size of cleaved N- and C-terminal products (Fig. 5d; Supplementary Fig. 9A). Mutating the N-terminal conserved residue E412 to alanine reduced the N-terminal cleavage and the E413A mutation abolished it (Fig. 5d). The D1385A and D1407A mutations of the C-terminal protease abolished C-terminal cleavage (Fig. 5d). Lastly, we tested expression of PA2684 in the cytosol and in the periplasm using a twin-arginine signal (Tat) since it was previously proposed that PA2684 targets the periplasm10. We found that PA2684 was also cleaved releasing a C-terminal fragment corresponding to the predicted cleavage size regardless of its localization (Fig. 5e). Similar cleavage and toxicity were also observed for VP1517 (Fig. 5f; Supplementary Fig. 9B–D).
Rhs proteins (InterPro IPR001826) form a large ancient protein family with members existing in all kingdoms. Although T6SS-dependent Rhs effectors have been previously reported10,11,26,47,48, we show for the first time self-cleaved Rhs effectors that are secreted by the T6SS. The Rhs effector TseI is self-cleaved into three interacting fragments, the N-terminal VIRN domain, the conserved Rhs core, and the C-terminal VIRC toxin domain (Fig. 6). Blocking cleavage has little effect on effector secretion but impairs toxicity and effector-mediated competition. We demonstrate that the N terminus and the C-terminal toxin of TseI interact with VgrG1, and the Rhs middle core interacts with the chaperone TecI (Fig. 6). Unlike the C-terminal toxin, neither the Rhs core or the N terminus possess any toxic domain nor exhibit any toxicity when cytosolically expressed. In addition, even the nontoxic C-terminus mutant could not be easily expressed and purified in comparison with the Rhs core and the N terminus due to instability or poor solubility (Supplementary Fig. 7E). Because the biological function of TseI is dependent on the C-terminal toxin whose secretion requires not only VgrG1, chaperone TecI, but also the N-terminal domain and the Rhs core, we propose that the Rhs core and the N terminus function as intramolecular chaperones for the C-terminal toxin.
Secreted T6SS chaperones have not been previously reported. Notably, some of the type VII secretion system (T7SS) secreted substrates display chaperone functions49. For example, the Mycobacterium T7SS secretes heterodimeric substrates EsxAB, of which EsxA displays membrane pore formation activities while its binding partner EsxB prevents EsxA aggregation50,51. Although it remains unclear why TseI requires cleavage for its cell-to-cell competition, the VIRC toxin might require VIRN and Rhs core for stability at the physiological levels delivered to recipient cells. Importantly, we found that VIRN could interact with the VIRC toxin in the absence of Rhs, suggesting the chaperoning role of VIRN and Rhs could be synergistic.
Toxins play a key role in bacterial pathogenesis and are likely under continuous selection through evolution. It is possible that the large number of Rhs-associated toxins serve as a reservoir accessible by delivery systems8,9. By contrast to the internal aspartic protease-mediated C-terminal cleavage, the mechanism of N-terminal cleavage is unclear and structural prediction fails to identify any protease signature. We identified two key glutamic acid residues for N-terminal cleavage. Although we cannot determine whether they are catalytic residues of a novel protease fold in this study, identification of these key residues enables us to construct full-length non-cleaved mutants of TseI and its homologs, thereby paving the way for structural characterization that would in turn reveal key information about the cleavage mechanism, as well as the overall structural arrangement of Rhs effectors.
Known structures of Rhs/YD-repeat homologs, including the bacterial Tc-toxin C-subunit TcC and the eukaryotic teneurin extracellular domain, show that the Rhs/YD-repeat core assembles a beta-barrel shell7,14,15,52,53,54; the key structural difference between them is the position of the C-terminus (Supplementary Fig. 10A). While the C-terminus of teneurin protrudes out of its shell52,53, the Tc-toxin TcC shell and the TcB subunit encapsulate the C-terminal toxin that is released through the TcA-formed translocation channel when the holotoxin enters target cells7,15,55. In comparison, expression of wild type and cleavage-defective TseI is toxic to cells suggesting the C-terminal toxin can reach its DNA substrate. Therefore, despite the predicted similarity of TseI to Tc-toxin TcC, TseI might not encapsulate its C-terminal VIRC nuclease toxin. This is further supported by the bacterial two-hybrid assay that showed direct interaction of non-cleaved TseI mutants, the protease-inactive D1407A-D1429A mutant and the protease deletion ΔRL40 mutant, with the immunity protein TsiI when they were fused to the split fragments of CyaA (Supplementary Fig. 10B). These results suggest the arrangement of Rhs-VIRC complex likely resembles that of the eukaryotic teneurin extracellular domain. This prediction warrants validation by structural characterization of TseI in future studies.
TseI self-cleavage is also reminiscent of the common auto-processing of proteases. Some proteases, e.g. human procathepsin L and carboxypeptidase Y, are synthesized as an inactive precursor to prevent from unwanted toxicity and activated upon specific environmental signals by cleaving off the inhibitory peptide56,57,58. Other proteases including subtilisin and alpha-lytic protease employ self-cleaved peptides to guide the folding and stability of the protease domain59,60,61. These cleaved fragments are named intramolecular chaperones as they are encoded in the primary gene sequences62,63. The self-cleaved VIRN and Rhs core thus distinguish themselves from the known T6SS chaperones34,35 and represent a new class of intramolecular chaperone for toxin secretion. These findings, together with the self-cleavage of Rhs proteins as T6SS effectors, further highlight the versatility of T6SS delivery38.
Bacterial strains and growth conditions
Strains, plasmids, and primers used in this study are listed in Supplementary Table 1 and Supplementary Data 4 and 5. Cultures were grown in LB ([w/v] 1% tryptone, 0.5% yeast extract, 0.5% NaCl) aerobically at 37 °C unless otherwise stated. Antibiotics were used at the following concentrations: streptomycin (100 µg/ml), ampicillin (100 µg/ml), kanamycin (50 µg/ml), irgasan (25 µg/ml), gentamicin (20 µg/ml), chloramphenicol (25 µg/ml for E. coli, 2.5 µg/ml for SSU).
Protein secretion assay
Cultures were grown aerobically in LB with appropriate antibiotics at 30 °C to OD600 = 1 and collected by centrifugation at 2500 × g for 8 min. Pellets were resuspended in fresh LB and induced at 30 °C for 1 h. Expression of genes on pBAD vectors was induced with 0.01% [w/v] l-arabinose. Cells were centrifuged at 2500 × g for 8 min at room temperature. Pellets were resuspended in SDS-loading dye and used as whole-cell samples. Supernatants were centrifuged again and then precipitated in 20% [v/v] TCA (trichloroacetic acid) at −20 °C for 20 min. Samples were centrifuged at 15,000 × g for 30 min at 4 °C and pellets were washed twice with acetone, air-dried, and resuspended in SDS-loading dye. Whole cell and secretion samples were boiled for 10 min before SDS-PAGE analysis.
Western blotting analysis
Proteins were resolved on an SDS-PAGE gel and transferred to a PVDF membrane (Bio-Rad) by electrophoresis. The membrane was blocked with 5% [w/v] non-fat milk in Tris-buffered saline with Tween-20 (TBST) buffer (50 mM Tris, 150 mM NaCl, 0.05% [v/v] Tween-20, pH 7.6) for 1 h at room temperature, incubated sequentially with primary antibodies and secondary HRP-conjugated antibodies in TBST with 1% [w/v] milk. Signals were detected using the Clarity ECL solution (Bio-Rad). Monoclonal antibodies to epitope tags were purchased from Sigma Aldrich (FLAG, Product # F1804 and 6His, Product # SAB4600386), Thermo Scientific (V5, Product # 37-7500), and Biolegend (RpoB, Product # 663905). RpoB, the beta subunit of RNA polymerase, was used as a control for equal loading and cell lysis in western blotting analysis. The polyclonal antibodies to Hcp, the VIRN, and the Rhs domains of TseI were custom-made by Shanghai Youlong Biotech (Supplementary Fig. 11). The secondary antibodies (anti-mouse or anti-rabbit IgG HRP linked) were purchased from Cell Signaling Technology (CST, Product # 7076S and # 7074S, respectively). RpoB and secondary antibodies were used at 1:20,000 dilution, while others at 1:10,000 dilution.
Bacterial competition assay
Killer and prey strains were grown in liquid culture to exponential phase (OD600 = 1) and stationary phase (OD600 = 2) respectively. Cells were centrifuged at 4500 × g for 3 min, resuspended in fresh LB, mixed together at a ratio of 5:1 (killer:prey) and spotted on LB-agar plates. After co-incubation for 3 h at 37 °C, cells were resuspended in 500 µl LB and a series of 10-fold dilutions were plated on LB plates with antibiotics. The mean Log10 c.f.u. of recovered preys was plotted and error bars show mean ± standard deviation between three biological replicates. A two-tailed Student’s t-test was used to determine p values.
Protein expression and purification in bacteria
Genes of interest were cloned into pETDUET1, pET28a, or pET22b vectors that were transformed to E. coli BL21(DE3). Cells were grown in LB with appropriate antibiotics to OD600 ~0.6 at 37 °C.
Protein expression was induced with 1 mM IPTG at 20 °C for 18 h. The cells were centrifuged at 10,000 × g for 30 min. The pellets were resuspended in lysis buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 10 mM imidazole) and lysed by sonication. Lysates were centrifuged at 10,000 × g for 40 min and the supernatants were transferred onto Ni-NTA or Cobalt-NTA resin (Smart-lifesciences). Proteins were eluted in elution buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, and variable concentrations of imidazole). Eluted samples were analyzed by SDS-PAGE and western blotting.
In vitro protein expression was performed with a PURExpress® In Vitro Protein Synthesis Kit (NEB) following the instruction of the manufacturer.
For TseI purification under denaturing conditions, His-tagged proteins were first purified with Ni-NTA resin, eluted with elution buffer A (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 250 mM imidazole), and dialyzed with dialysis buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl) three times to remove imidazole. Protein samples were then mixed with the denature buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 8 M guanidine hydrochloride) to a final concentration of 6 M guanidine hydrochloride, incubated at room temperature for 1 h, and mixed with Ni-NTA resin. Proteins were then eluted with elution buffer B (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 6 M guanidine hydrochloride, 250 mM imidazole) and dialyzed in a series of buffers with a decreasing gradient of guanidine hydrochloride at 4M, 2M, 1M, and no guanidine hydrochloride, respectively. All dialysis steps were performed at 4 °C.
Protein pull-down assays
Genes were cloned into pBAD, pETDUET1, and pET22b vectors for expression with either His or 3V5 tags. Cultures were grown in 10 ml LB with appropriate antibiotics to OD600 of 0.6, and induced with 0.1% [w/v] l-arabinose at 30 °C for 3 h or 1 mM IPTG overnight at 20 °C. Cells were centrifuged at 4500 r.p.m. for 10 min and resuspended in 1 ml of lysis buffer (20 mM Tris pH 8.0, 500 mM NaCl, 50 mM imidazole with protease inhibitor (Thermo Scientific)). Resuspended cells were sonicated (20 × 5 s) and cell debris was removed by centrifugation (15,000 × g for 15 min). Clarified supernatants were mixed and loaded to Ni-NTA resin (Smart-lifesciences), washed five times with wash buffer (20 mM Tris pH 8.0, 500 mM NaCl, 50 mM imidazole), and eluted in 100 µl elution buffer (20 mM Tris pH 8.0, 500 mM NaCl, 500 mM imidazole). Cell lysates and eluted samples were analyzed by western blotting.
Bacterial two-hybrid assay
Proteins of interest were fused to the T18 and T25 split domains of the Bordetella adenylate cyclase as previously described34,44. The two plasmids encoding the fusion proteins were co-transformed into the reporter strain BTH101. Three independent colonies for each transformation were inoculated into 500 μl of LB medium. After 5 h growth at 30 °C, 5 μl of each culture were spotted onto LB plates supplemented with ampicillin, kanamycin, IPTG, and X-Gal and incubated for 10 h at 30 °C and then 10 h at room temperature. The experiments were done in triplicate and a representative result is shown.
Bioinformatics analysis and homology modeling
All gene sequences of A. dhakensis SSU are retrieved from the draft genome assembly (GenBank NZ_JH815591.1). The gene sequences of hcp1-vgrG1-tecI-tseI-tsiI are provided in Supplementary Data 3. Benchling was used to manage and analyze DNA and protein sequences and to predict tsiI as a putative open reading frame64. TseI sequence was analyzed with Phyre2 and blastp to identify homologs and species distribution41,42. A representative set of 48 sequences were manually selected from the top 1000 hits and downloaded from the NCBI database. Homologs were aligned using Clustal Omega65. The resulting alignment was used to generate a WebLogo66. Structural models of TseI were constructed using Phyre2 (ref. 41) and the resultant models were compared with known structures teneurin and Tc-toxin Tcc3 using Chimera67. Maximum-likelihood phylogeny was constructed using IQ-TREE web server with bootstrap 1000 times68. The iTOL server was used to visualize the phylogenetic tree69.
Protein toxicity assay
Cells expressing different plasmid constructs were grown in LB supplemented with 0.2% [w/v] glucose at 37 °C overnight. Cells were then collected and resuspended in fresh LB and grown to OD600 = 1. A series of 10-fold dilutions were plated on LB plates containing 0.01% [w/v] l-arabinose (for pBAD vectors), 0.1 mM IPTG (for pET vectors), or 0.2% [w/v] glucose, respectively. Each experiment was repeated three times, with one representative experiment shown.
N-terminal Edman sequencing and LC-MS/MS analysis
Purified TseI was resolved on an SDS-PAGE gel and protein bands were excised individually. The sequence of the first 10 amino acids for each band was determined by Edman sequencing and performed at the BiotechPack Scientific company. LC-MS/MS analysis of excised bands was performed at the Southern Alberta Mass Spectrometry core facility. Original data will be provided upon request.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Breton, C., Oriol, R. & Imberty, A. Conserved structural features in eukaryotic and prokaryotic fucosyltransferases. Glycobiology 8, 87–94 (1998).
Kasho, V. N., Smirnova, I. N. & Kaback, H. R. Sequence alignment and homology threading reveals prokaryotic and eukaryotic proteins similar to lactose permease. J. Mol. Biol. 358, 1060–1070 (2006).
Ito, K., Ebihara, K., Uno, M. & Nakamura, Y. Conserved motifs in prokaryotic and eukaryotic polypeptide release factors: tRNA-protein mimicry hypothesis. Proc. Natl. Acad. Sci. USA 93, 5443–5448 (1996).
Wolf, Y. I., Brenner, S. E., Bash, P. A. & Koonin, E. V. Distribution of protein folds in the three superkingdoms of life. Genome Res. 9, 17–26 (1999).
Tucker, R. P. et al. Phylogenetic analysis of the teneurins: conserved features and premetazoan ancestry. Mol. Biol. Evol. 29, 1019–1029 (2012).
Berns, D. S., DeNardo, L. A., Pederick, D. T. & Luo, L. Teneurin-3 controls topographic circuit assembly in the hippocampus. Nature 554, 328–333 (2018).
Busby, J. N., Panjikar, S., Landsberg, M. J., Hurst, M. R. H. & Lott, J. S. The BC component of ABC toxins is an RHS-repeat-containing protein encapsulation device. Nature 501, 547–550 (2013).
Zhang, D., de Souza, R. F., Anantharaman, V., Iyer, L. M. & Aravind, L. Polymorphic toxin systems: comprehensive characterization of trafficking modes, processing, mechanisms of action, immunity and ecology using comparative genomics. Biol. Direct 7, 18 (2012).
Jamet, A. & Nassif, X. New players in the toxin field: polymorphic toxin systems in bacteria. MBio 6, e00285-15 (2015).
Whitney, J. C. et al. Genetically distinct pathways guide effector export through the type VI secretion system. Mol. Microbiol. 92, 529–542 (2014).
Koskiniemi, S. et al. Rhs proteins from diverse bacteria mediate intercellular competition. Proc. Natl. Acad. Sci. USA 110, 7032–7037 (2013).
Jackson, A. P., Thomas, G. H., Parkhill, J. & Thomson, N. R. Evolutionary diversification of an ancient gene family (rhs) through C-terminal displacement. BMC Genomics 10, 584 (2009).
Foster, S. J. Molecular analysis of three major wall‐associated proteins of Bacillus subtilis 168: evidence for processing of the product of a gene encoding a 258 kDa precursor two‐domain ligand‐binding protein. Mol. Microbiol. 8, 299–310 (1993).
Meusch, D. et al. Mechanism of Tc toxin action revealed in molecular detail. Nature 508, 61–65 (2014).
Gatsogiannis, C. et al. Tc toxin activation requires unfolding and refolding of a β-propeller. Nature 563, 209–233 (2018).
Fritsch, M. J. et al. Proteomic identification of novel secreted antibacterial toxins of the Serratia marcescens type VI secretion system. Mol. Cell. Proteomics 12, 2735–2749 (2013).
Altindis, E., Dong, T., Catalano, C. & Mekalanos, J. Secretome analysis of Vibrio cholerae type VI secretion system reveals a new effector-immunity pair. MBio 6, e00075-15 (2015).
Ma, J. et al. The Hcp proteins fused with diverse extended-toxin domains represent a novel pattern of antibacterial effectors in type VI secretion systems. Virulence 8, 1189–1202 (2017).
Hachani, A., Allsopp, L. P., Oduko, Y. & Filloux, A. The VgrG proteins are “à la Carte” delivery systems for bacterial type VI effectors. J. Biol. Chem. 289, 17872–17884 (2014).
Pukatzki, S. et al. Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system. Proc. Natl. Acad. Sci. 103, 1528–1533 (2006).
Mougous, J. D. et al. A virulence locus of Pseudomonas aeruginosa encodes a protein secretion apparatus. Science 312, 1526–1530 (2006).
Russell, A. B., Peterson, S. B. & Mougous, J. D. Type VI secretion system effectors: poisons with a purpose. Nat. Rev. Microbiol. 12, 137–148 (2014).
Ho, B. T., Dong, T. G. & Mekalanos, J. J. A view to a kill: the bacterialtype VI secretion system. Cell Host Microbe 15, 9–21 (2014).
Wang, J., Brodmann, M. & Basler, M. Assembly and subcellular localization of bacterial type VI secretion systems. Annu. Rev. Microbiol. 73, 621–638 (2019).
Leiman, P. G. et al. Type VI secretion apparatus and phage tail-associated protein complexes share a common evolutionary origin. Proc. Natl. Acad. Sci. USA 106, 4154–4159 (2009).
Shneider, M. M. et al. PAAR-repeat proteins sharpen and diversify the type VI secretion system spike. Nature 500, 350–353 (2013).
Suarez, G. et al. A type VI secretion system effector protein, VgrG1, from Aeromonas hydrophila that induces host cell toxicity by ADP ribosylation of actin. J. Bacteriol. 192, 155–168 (2010).
Pukatzki, S., Ma, A. T., Revel, A. T., Sturtevant, D. & Mekalanos, J. J. Type VI secretion system translocates a phage tail spike-like protein into target cells where it cross-links actin. Proc. Natl. Acad. Sci. USA 104, 15508–15513 (2007).
Hood, R. D. et al. A type VI secretion system of Pseudomonas aeruginosa targets a toxin to bacteria. Cell Host Microbe 7, 25–37 (2010).
Russell, A. B. et al. Type VI secretion delivers bacteriolytic effectors to target cells. Nature 475, 343–347 (2011).
English, G. et al. New secreted toxins and immunity proteins encoded within the Type VI secretion system gene cluster of Serratia marcescens. Mol. Microbiol. 86, 921–936 (2012).
Brooks, T. M., Unterweger, D., Bachmann, V., Kostiuk, B. & Pukatzki, S. Lytic activity of the Vibrio cholerae type VI secretion toxin VgrG-3 is inhibited by the antitoxin TsaB. J. Biol. Chem. 288, 7618–7625 (2013).
Dong, T. G., Ho, B. T., Yoder-Himes, D. R. & Mekalanos, J. J. Identification of T6SS-dependent effector and immunity proteins by Tn-seq in Vibrio cholerae. Proc. Natl. Acad. Sci. USA 110, 2623–2628 (2013).
Liang, X. et al. Identification of divergent type VI secretion effectors using a conserved chaperone domain. Proc. Natl. Acad. Sci. USA 112, 9106–9111 (2015).
Unterweger, D. et al. Chimeric adaptor proteins translocate diverse type VI secretion system effectors in Vibrio cholerae. EMBO J. 34, 2198–2210 (2015).
Bondage, D. D., Lin, J.-S., Ma, L.-S., Kuo, C.-H. & Lai, E.-M. VgrG C terminus confers the type VI effector transport specificity and is required for binding with PAAR and adaptor–effector complex. Proc. Natl Acad. Sci. USA 113, E3931–E3940 (2016).
Cianfanelli, F. R. et al. VgrG and PAAR proteins define distinct versions of a functional type VI secretion system. PLoS Pathog. 12, e1005735 (2016).
Burkinshaw, B. J. et al. A type VI secretion system effector delivery mechanism dependent on PAAR and a chaperone–co-chaperone complex. Nat. Microbiol. 3, 632–640 (2018).
Chen, P.-L., Lamy, B. & Ko, W.-C. Aeromonas dhakensis, an increasingly recognized human pathogen. Front. Microbiol. 7, 793 (2016).
Colston, S. M. et al. Bioinformatic genome comparisons for taxonomic and phylogenetic assignments using Aeromonas as a test case. MBio 5, e02136–14 (2014).
Kelley, L. A., Mezulis, S., Yates, C. M., Wass, M. N. & Sternberg, M. J. E. The Phyre2 web portal for protein modeling, prediction and analysis. Nat. Protoc. 10, 845–858 (2015).
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local alignment search tool. J. Mol. Biol. 215, 403–410 (1990).
Guzman, L. M., Belin, D., Carson, M. J. & Beckwith, J. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177, 4121–4130 (1995).
Karimova, G., Pidoux, J., Ullmann, A. & Ladant, D. A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc. Natl. Acad. Sci. USA 95, 5752–5756 (1998).
Alcoforado Diniz, J. & Coulthurst, S. J. Intraspecies competition in Serratia marcescens is mediated by type VI-secreted Rhs effectors and a conserved effector-associated accessory protein. J. Bacteriol. 197, 2350–2360 (2015).
Salomon, D. et al. Type VI secretion system toxins horizontally shared between marine bacteria. PLoS Pathog. 11, e1005128 (2015).
Ma, J. et al. PAAR-Rhs proteins harbor various C-terminal toxins to diversify the antibacterial pathways of type VI secretion systems. Environ. Microbiol. 19, 345–360 (2017).
Kung, V. L. et al. An rhs gene of Pseudomonas aeruginosa encodes a virulence protein that activates the inflammasome. Proc. Natl. Acad. Sci. USA 109, 1275–1280 (2012).
Phan, T. H. & Houben, E. N. G. Bacterial secretion chaperones: the mycobacterial type VII case. FEMS Microbiol. Lett. 365, 1–8 (2018).
Smith, J. et al. Evidence for pore formation in host cell membranes by ESX-1-secreted ESAT-6 and its role in Mycobacterium marinum escape from the vacuole. Infect. Immun. 76, 5478–5487 (2008).
De Jonge, M. I. et al. ESAT-6 from Mycobacterium tuberculosis dissociates from its putative chaperone CFP-10 under acidic conditions and exhibits membrane-lysing activity. J. Bacteriol. 189, 6028–6034 (2007).
Li, J. et al. Structural basis for teneurin function in circuit-wiring: a toxin motif at the synapse. Cell 173, 735–748.e15 (2018).
Jackson, V. A. et al. Structures of Teneurin adhesion receptors reveal an ancient fold for cell-cell interaction. Nat. Commun. 9, 1079 (2018).
Roderer, D. & Raunser, S. Tc toxin complexes: assembly, membrane permeation, and protein translocation. Annu. Rev. Microbiol. 73, 247–265 (2019).
Roderer, D., Hofnagel, O., Benz, R. & Raunser, S. Structure of a Tc holotoxin pore provides insights into the translocation mechanism. Proc. Natl Acad. Sci. USA 116, 23083–23090 (2019).
Verma, S., Dixit, R. & Pandey, K. C. Cysteine proteases: modes of activation and future prospects as pharmacological targets. Front. Pharmacol. 7, 107 (2016).
Coulombe, R. et al. Structure of human procathepsin L reveals the molecular basis of inhibition by the prosegment. EMBO J. 15, 5492–5503 (1996).
Winther, J. R. & Sørensen, P. Propeptide of carboxypeptidase Y provides a chaperone-like function as well as inhibition of the enzymatic activity. Proc. Natl. Acad. Sci. USA 88, 9330–9334 (1991).
Ikemura, H., Takagi, H. & Inouye, M. Requirement of pro-sequence for the production of active subtilisin E in Escherichia coli. J. Biol. Chem. 262, 7859–7864 (1987).
Silen, J. L., Frank, D., Fujishige, A., Bone, R. & Agard, D. A. Analysis of prepro-alpha-lytic protease expression in Escherichia coli reveals that the pro region is required for activity. J. Bacteriol. 171, 1320–1325 (1989).
Silen, J. L. & Agard, D. A. The alpha-lytic protease pro-region does not require a physical linkage to activate the protease domain in vivo. Nature 341, 462–464 (1989).
Inouye, M. Intramolecular chaperone: the role of the pro-peptide in protein folding. Enzyme 45, 314–321 (1991).
Chen, Y. J. & Inouye, M. The intramolecular chaperone-mediated protein folding. Curr. Opin. Struct. Biol. 18, 765–770 (2008).
Benchling [Biology Software]. https://benchling.com (2019).
Sievers, F. et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 1–6 (2011).
Crooks, G. E., Hon, G., Chandonia, J.-M. & Brenner, S. E. WebLogo: a sequence logo generator. Genome Res. 14, 1188–1190 (2004).
Pettersen, E. F. et al. UCSF chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).
Trifinopoulos, J., Nguyen, L. T., von Haeseler, A. & Minh, B. Q. W-IQ-TREE: a fast online phylogenetic tool for maximum likelihood analysis. Nucleic Acids Res. 44, W232–W235 (2016).
Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v4: recent updates and new developments. Nucleic Acids Res. 47, W256–W259 (2019).
This work was supported by funding from National Key R&D Program of China (2018YFA0901200), National Natural Science Foundation of China (31770082), the Canadian Institutes of Health Research, and the Natural Sciences and Engineering Research Council of Canada. We thank Le Tang, Brianne Burkinshaw, and Steve Hersch for technical assistance. We thank Dr. Jun Zheng for sharing strains and data on protein VP1517. We thank Laurent Brechenmacher and the Southern Alberta Mass Spectrometry Centre for LC-MS/MS analysis. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
The authors declare no competing interests.
Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Pei, TT., Li, H., Liang, X. et al. Intramolecular chaperone-mediated secretion of an Rhs effector toxin by a type VI secretion system. Nat Commun 11, 1865 (2020). https://doi.org/10.1038/s41467-020-15774-z
Journal of Microbiology (2021)