Figure 1.
Schematic representation showing the peptidoglycan cleavage specificities of the type-6-secretion amidase (Tae) effector families.
Cleavage specificity of the Tae1–4 effector proteins on Gram-negative tetrapeptide stems (left) or Gram-positive pentapeptide stems (right) is depicted based on previous reports [3], [4]. Tae2 and Tae3 effector proteins hydrolyze the amide bond between the mDAP-D-Ala cross-link in Gram-negative as well as Gram-positive peptidoglycan. In contrast, members of the Tae1 (including Tse1 from P. aeruginosa) and the Tae4 effector family degrade the peptidoglycan scaffold by cleaving the γ-D-glutamyl-mDAP bond. Whereas, Tae4 exclusively hydrolyzes the acceptor stem of cross-linked as well as non-cross-linked Gram-negative peptidoglycan, Tse1 specifically cleaves the cross-linked donor peptide stem. Furthermore, Gram-positive peptidoglycan is a poor substrate for Tae4 family members.
Table 1.
Data collection and refinement statistics.
Figure 2.
rystal structure and conservation of the Tae4 effector from S. typhimurium.
(A) Ribbon representation of S. typhimurium Tae4. Tae4 adopts the classical papain-like fold of cysteine peptidases of the NlpC/P60 family. The core of the Tae4 effector protein is formed by a central anti-parallel β-sheet which is flanked by α-helices. Cys44, His126 and Asp137 are shown as a stick model. α-helices are shown in orange and β-strands are colored in green. The loop region between residue Asn143 and Asn147 which is not included in the model is indicated in gray. (B) Amino acid sequence conservation of Tae4 effector proteins mapped on the molecular surface colored according to (E). The crevice harboring the conserved active site residues is highlighted with a red dashed circle. (C) Crystal structure of the effector protein Tae4 with the conserved NlpC/P60 catalytic core in gray (same orientation as in (A)). The additional α-helices B, C, G and F are colored in orange. The Tae4-specific β-strands 1 and 2 forming the β-hairpin as well as the β-strand 3 are shown in green. Cys44, His126 and Asp137 are depicted as stick model. (D) Crystal structure of the effector protein Tse1 from P. aeruginosa [15] with the conserved NlpC/P60 catalytic core in gray. Tse1 specific elements are colored in blue. The Cys30, His91 and Cys110 residues similar to Tae4’s Cys44, His126 and Asp137 are depicted as stick model. (E) Amino acid sequence alignment of Tae4 effectors from Salmonella typhimurium (NP_459275.1), Escherichia coli B354 (ZP_06652154.1), Enterobacter cloacae (YP_006478116.1) and Pantoea sp. Sc1 (ZP_09929551.1) going from dark green (identical residues) to light green, orange and yellow (with decreasing conservation). α-helices and β-strands are colored in orange and green according to (A). Residues of the Tae4 catalytic triad are marked with a red star. Green pentagons below the sequence alignment indicate surface exposed residues that interact with the Tai4 immunity proteins.
Figure 3.
Growth of Escherichia coli expressing wild-type and mutated variants of Tae4 and Tai4 proteins.
Inoculi were prepared by serial dilutions from 100 to 10−6 of overnight cultures and spotted with decreasing optical density from left to the right onto LB-agar plates containing IPTG to induce protein expression. The effector protein Tae4 leads to a significant reduction of bacterial growth upon periplasmic localization. Expression of variants mutated in the active site residues Cys44, H126 and D137 in Tae4 to alanine residues did no more interfere with bacterial growth. A replacement of Cys135 and Cys139 by serine residues did not affect bacterial growth compared to wild-type Tae4. Wild-type Tai4 could rescue the growth defect induced by periplasmic localization of the Tae4. In contrast, a Tai4 variant missing its periplasmic leader sequence (Tai4ΔN26) could no more counteract the growth defect. Additionally, mutations in the effector/immunity interface in Tai4(E71A_S98A) did not rescue the growth phenotype as efficient as the wild-type protein. Proteins which were expressed from pET22b vector constructs contained the pelB leader sequence for artificial periplasmic localization and are labeled in black. Proteins which were expressed from pET28b vector constructs are labeled in red. Vector controls can be found at the bottom of the panel.
Figure 4.
Crystal structure and conservation of the Tai4 immunity protein from S. typhimurium.
(A) Ribbon representation of Tai4. Tai4 is a purely α-helical protein forming a head-to-tail dimer in the crystal structure. The disulfide bond (DSB) formed by Cys48 and Cys108 that links the α-helices b and e with each other is shown as stick representation. Ser98 which is important for effector inhibition is depicted as stick representation as well. (B) Dimeric arrangement of the Tai4 inhibitor showing the molecular surface of one of the immunity proteins. Amino acid sequence conservation (C) is mapped on the molecular surface. (C) Amino acid sequence alignment of Tai4 immunity proteins from Salmonella typhimurium (NP_459276.1), Escherichia coli B354 (ZP_06652155.1), Enterobacter cloacae (YP_006478115.1) and Pantoea sp. Sc1 (ZP_09929550.1) going from dark green (identical residues) to light green, orange and yellow (with decreasing conservation). Secondary structure elements above the sequence alignment are colored according to (A). Ser98, located in the inhibition loop between the α-helices d and e as well as the conserved Glu71 residue which makes extensive contacts to Tae4 are marked with a red star. Residues which interact with Tae4 are marked with a green pentagon. Residues that are involved in Tai4 dimer formation by either hydrophobic interactions (blue rectangles) or hydrogen bonds (black dots) are also indicated below the sequence alignment. DSB formation between Cys48 and Cys108 is represented as a dashed line. Cleavage site of the periplasmic leader sequence in Tai4 is indicated with a black arrow above the sequence.
Figure 5.
Biological assembly of Tai4 and the Tae4/Tai4 complex.
(A) Cross-linking experiments of wild-type Tai4 using glutaraldehyde. Aliquots of cross-linked Tai4 have been taken at different time points (0 to 60 min) and separated on Coomassie stained SDS-PAGE. Bands of monomeric Tai4 with an electrophoretic mobility corresponding to 13 kDa as well as increasing amounts of cross-linked dimeric species of Tai4 (26 kDa) were observed. Note, that minor traces of unspecific intermolecular cross-links are observed (marked with a star) which result in apparent trimeric and tetrameric Tai4 species. LMW: low molecular weight marker in kDa. (B) Gel-filtration and static light scattering of the Tae4/Tai4 effector/immunity complex (solid line) as well as of the wild-type Tai4 immunity protein (dotted line). According to gel-filtration experiments using a gel-filtration standard from BioRad the proteins elute at a corresponding molecular weight of approximately 25 kDa for Tai4 and 46 kDa for the Tae4/Tai4 complex. However, using a multi-angle static light detector averaged molecular masses of 25.26±0.07 kDa for the Tai4 protein and 60.38±0.12 kDa for the Tae4/Tai4 protein complex could be determined. Note that the discrepancy between the masses calculated from the gel-filtration and the static light scattering experiments most likely are caused by the elongated shape of the Tae4/Tai4 complex as observed in the crystal structure. (C) Cross-linking experiments of mutated Tai4(E71A_S98A) using glutaraldehyde have been performed similar to wild-type Tai4 (A).
Figure 6.
Cyrstal structure of the Tae4/Tai4 heteroteramer.
(A) The two symmetry related Tai4 molecules (red transparence and opaque) form a heterotetramer with two Tae4 molecules (orange and green) in our crystal structure. (B) Close up of the interaction between Tae4 and Tai4 at the Tae4 active site. Tai4 inserts the loop region between α-helix d and e (d-e-loop) into the active site of Tae4. This loop harbors Ser98 that forms a hydrogen bond to the catalytic important His126 residue and thereby prevents deprotonation of Cys44. (C) Molecular surface representation of the Tai4 dimer (one monomer colored in white and the second in gray) bound to Tae4 with amino acid sequence conservation mapped onto the molecular surface and colored according to 2E and 4C. Ser98 in Tai4 which has been mutated in this study and its interacting residue in Tae4, His126, are colored in red. The second mutated residue in Tai4, Glu71, and its interacting residues in Tae4, Val80 and Asn81, are highlighted in the red as well. These residues are part of the hydrogen bond network between Tae4 and Tai4. (D) “Open-Book view” of the Tae4/Tai4 heterotetramer interface (as indicated by a dashed line in (C)) with amino acid sequence conservation mapped onto the molecular surface.
Figure 7.
Affinity and association/dissociation kinetics of the Tae4/Tai4 complex.
(A) Isothermal titration calorimetry experiment. The binding constant (KD), as well as the stoichiometric ratio (N) are determined by fitting a single binding-site model to the data. Original titrations are shown above the fit. (B) Kinetic traces of Tae4-ATTO 488 and wild-type as well as mutated Tai4(E71A_S98A) binding have been recorded at different Tai4 concentration (0.3–1.5 µM from light gray to black). The time has been plotted against the relative fluorescence signal and single exponential functions have been fitted to the data (red). The time traces and the fits are exemplarily shown for binding of the Tai4(E71A_S98A) protein (C) The observed rate constants kobs from (B) have been plotted against the concentrations of Tai4wt (filled black circles) and Tai4(E71A_S98A) (empty gray circles) and fitted using a linear regression (Tai4wt:black, Tai4(E71A_S98A):gray). The association rate constants of Tai4 binding could be extracted from the slopes of the linear functions. Although the slope of the linear functions is just slightly increased for Tai4(E71A_S98A), the y-axis intercept (corresponding to the off-rate) is significantly lower for wild-type Tai4. (D) The dissociation rate constants were determined separately by dissociation experiments using unlabelled Tae4 protein to chase off the Tae4-ATTO 488/Tai4 complex. Shown are the time traces recorded from 1 µM Tae4 to chase off the effector/immunity complex. The time trace for wild-type Tae4/Tai4 complex is shown in blue. The traces for the mutated Tae4/Tai4 protein complex is shown in black with the corresponding exponential fit in red. Simulated time traces using the exponential fit values of the Tai4(E71A_S98A) experiment with the dissociation rate constant divided by up to a factor of 104 are depicted in gray. Simulated time traces suggested an off-rate between 0.7 ×10−3 and 0.7 ×10−4 s−1 for wild-type Tai4; although, it can not be excluded to be lower. All values obtained from the stopped-flow experiments are shown in Table 2.
Table 2.
Affinity constants for wild-type and mutant Tai4 protein binding to Tae4.