Molecular Basis of the Versatile Regulatory Mechanism of HtrA-Type Protease AlgW from Pseudomonas aeruginosa

Abstract

AlgW, a membrane-bound periplasmic serine protease belonging to the HtrA protein family, is a key regulator of the regulated intramembrane proteolysis (RIP) pathway and is responsible for transmitting the envelope stress signals in Pseudomonas aeruginosa The AlgW PDZ domain senses and binds the C-terminal of mis-localized outer membrane proteins (OMPs) or periplasmic protein MucE, leading to catalytic activation of the protease domain. While AlgW is functionally well studied, its exact activation mechanism remains to be elucidated. Here, we show that AlgW is a novel HtrA protease that can be biochemically activated by both peptide and lipid signals. Compared with the corresponding homologue DegS in Escherichia coli, AlgW exhibits a distinct substrate specificity and regulation mechanism. Structural, biochemical, and mutagenic analyses revealed that, by specifically binding to the C-terminal decapeptide of MucE, AlgW could adopt more relaxed conformation and obtain higher activity than with tripeptide activation. We also investigated the regulatory mechanism of the L_A loop in AlgW and proved that the unique structural feature of this region was responsible for the distinct enzymatic property of AlgW. These results demonstrate the unique and diverse activation mechanism of AlgW, which P. aeruginosa may utilize to enhance its adaptability to environmental stress.IMPORTANCE HtrA-family proteases are commonly employed to sense the protein folding stress and activate the regulated intramembrane proteolysis (RIP) cascade in Gram-negative bacteria. Here, we reveal the unique dual-signal activation and dynamic regulation properties of AlgW, an HtrA-type protease triggering the AlgU stress-response pathway, which controls alginate production and mucoid conversion in Pseudomonas aeruginosa The structural and functional data offer insights into the molecular basis underlying the transition of different activation states of AlgW in response to different effectors. Probing these unique features provides an opportunity to correlate the diverse regulation mechanism of AlgW with the high adaptability of P. aeruginosa to environmental changes during infection.

Keywords: AlgW; HtrA; Pseudomonas aeruginosa; alginate; crystal structure; mucoid phenotype; regulated intramembrane proteolysis.

FIG 1

Degradation of MucA and MucB by AlgW in response to the peptides and lipid signals. (A and B) MucB/RseB (130 μM) was cleaved by AlgW/DegS (25 μM) in the presence of different agonist combinations, including lipid-A (100 μM), lipid-like detergent DDM (400 μM), decapeptide SVRDELRWVF (100 μM), and decapeptide combined with lipid-A or DDM. (C and D) MucB/RseB-protected MucA^peri/RseA^peri degradation by AlgW/DegS in the presence of different lengths of the activating peptides in a time-dependent manner. MucA^peri (125 μM), AlgW (25 μM), MucB (130 μM), activation peptides (tripeptide WVF or decapeptide SVRDELRWVF, 100 μM), and lipid-A (100 μM) were incubated at 37°C in a time-dependent manner in buffer (25 mM Tris-HCl pH 7.5, 150 mM NaCl). Similarly, RseA^peri (125 μM) was cleaved by DegS (25 μM) in the presences of RseB (130 μM), activation peptides (tripeptide YYF or decapeptide DNRDGNVYYF, 100 μM), and lipid-A (100 μM) at 37°C in a time-dependent manner in a phosphate buffer containing 50 mM NaHPO₄ (pH 7.5), 200 mM NaCl, 10% glycerol, and 4 mM EDTA. (E and F) MucA^peri/RseA^peri was cleaved by AlgW/DegS in the presence of different agonist combinations in a time-dependent manner. MucA^peri (125 μM), AlgW (25 μM), activation decapeptide SVRDELRWVF (100 μM), and lipid-A (100 μM) were incubated at 37°C in a time-dependent manner in buffer. Similarly, RseA^peri (125 μM) degradation by DegS (25 μM) took place in the presence of RseB (130 μM), activation decapeptide DNRDGNVYYF (100 μM), and lipid-A (100 μM) at 37°C for a time-dependent manner in a phosphate buffer.

FIG 2

Determination of enzyme apparent kinetic parameters of dual-signal-activated AlgW. (A) Activity assay of tripeptide or decapeptide-activated AlgW using the quenched fluorescent substrates Abz-VLAG-pNA, Abz-TVAW-pNA, and Abz-AAAA-pNA. All assays were placed in a buffer containing 25 mM Tris (pH 7.5) and 150 mM NaCl at 37°C for 1 h by adding 100 μM substrate, 200 nM AlgW, and 10 mM agonist peptides (tripeptide WVF or decapeptide SVRDELRWVF). (B) The apparent Michaelis-Menten parameters of AlgW cleavage quenched fluorescent substrates Abz-VLAG-pNA (100 μM) under the condition of peptide or lipid activation. (C) The activity assay of peptide or DDM activated AlgW using the saturated quenched fluorescent substrates Abz-VLAG-pNA. Under the gradient of peptide or DDM concentration,100 μM substrate and 200 nM AlgW were coincubated with or without 400 μM DDM or 10 mM agonist peptides in a buffer containing 25 mM Tris (pH 7.5) and 150 mM NaCl at 37°C for 1 h. The final assay volume of all experiments was 100 μl, and the appearance of fluorescent products was monitored at 60-s intervals at fluorescence wavelengths of λex = 310 nm and λem = 420 nm. All data points were representative of three independent determinations and were simulated with solid lines through the formula of Y = V_max · X^h/(K_m^h + X^h) or Y = V_max · X^h/(K_half^h + X^h). The K_cat for the AlgW trimer was calculated by K_cat = V_max/(E_total), (E_total) = 200/3 nM trimers. Error bars represent the standard deviation.

FIG 3

Peptide-bound AlgW structures. (A) Structures and funnel-like trimeric organization of peptide-bound AlgW. Side view of the trimer illustrating the relative orientation of AlgW in the periplasm. (B) The oligomeric structure of AlgW-tripeptide/AlgW-S227A-tripeptide, AlgWS227A-decapeptide, was generated by symmetry operation. Each monomer in all structures was displayed in magenta, blue, and cyan. (C) The overall structures of decapeptide-bound AlgW. The decapeptide bound in AlgW was displayed as red cartoon, and the catalytic triad (H123, D153 and S227) sites surrounded by activation loops (L₁-L₃, L_D and L_A) were displayed as red sticks. (D) Structural comparison of the protease domains of peptide-bound AlgW structures. The L_A loop in tripeptide-bound AlgW structures cannot be modeled. The joint angle of the L_A loop in two sets of decapeptide-bound AlgW structures was calculated using UCSF ChimeraX (http://www.cgl.ucsf.edu/chimerax/index.html), which was 7.65°. (E) Structural comparison of the protease domains of AlgW-decapeptide and DegS structures. L₂ and L₃ of AlgW adopt inward conformations compared with those of DegS structures. The catalytic triad sites are displayed as sticks. (F) Structural variations of peptide-bound AlgW structures. The joint angle between the PDZ and the protease domain was defined between the longest inertial vectors of each domain, which was calculated using UCSF ChimeraX. The calculated joint angles in the structures of AlgW-tripeptide, AlgWS227A-tripeptide, AlgW-decapeptide and AlgWS227A-decapeptide were 25.62°, 25.74°, 26.13°, and 26.65°.

FIG 4

The peptide-binding pocket and domain interactions of AlgW and DegS. (A and B) The interactions of agonist peptides with AlgW/DegS. The key residues in the peptide-binding pockets are displayed as sticks. Peptides are displayed as thick sticks. (C and D) The domain-domain interactions in which the peptide binds to PDZ and mediates the activation of the catalytic triad. The key residues mediating the domain-domain interactions are displayed as sticks. E. Multisequence alignment (generated using Espript 3.0; http://espript.ibcp.fr/ESPript/ESPript/) of AlgW/DegS homologues. The residues in the peptide-binding pocket of AlgW and DegS are shaded in cyan, and the catalytic triad (H123, D153, and S227 in AlgW, H96, and D126 and S201 in DegS) sites are shaded in red. The conserved residues involved in domain-domain interactions are shaded in yellow.

FIG 5

Conformational transition and activity regulation mediated by the L_A loop in AlgW. (A and B) Conformational movement of the L_A loop in peptide-bound AlgW structures; the key residues involved in the conformational transition of the L_A loop are displayed as sticks. AlgWS227A-decapeptide- and tripeptide-bound structures adopted a uniform conformation, which represents the major state of the L_A loop. (C) The residue composition of the L_A loop hinge region in AlgW and DegS.

FIG 6

Effect of algW knockout and mutants on alginate production and biofilm formation. (A) Alginate production of the ΔalgW; complementation transformed with plasmid pME6032-algW and variants strain compared with that of the wild-type PAO1. Alginate was measured, and the amount of uronic acid in comparison with a standard curve made with d-mannuronic acid lactone was determined. Each bar represents the mean of three independent measurements (± standard error of the mean [SEM]). (B) Biofilm formation of the ΔalgW and a ΔalgW+pME6032-algW complementation strain compared with that of the wild-type PAO1. Quantification of biofilm biomass via crystal-violet staining and A₅₇₀ was measured using a microplate reader. Data are shown as the change relative to PAO1 and represent three independent experiments. A one-way ANOVA statistical test with equal variances was conducted. The following results were considered significant: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.