A Complex of LaoA and LaoB Acts as a Tat-Dependent Dehydrogenase for Long-Chain Alcohols in Pseudomonas aeruginosa

Abstract

The opportunistic pathogen Pseudomonas aeruginosa can utilize unusual carbon sources, like sodium dodecyl sulfate (SDS) and alkanes. Whereas the initiating enzymatic steps of the corresponding degradation pathways have been characterized in detail, the oxidation of the emerging long-chain alcohols has received little attention. Recently, the genes for the Lao (long-chain-alcohol/aldehyde oxidation) system were discovered to be involved in the oxidation of long-chain alcohols derived from SDS and alkane degradation. In the Lao system, LaoA is predicted to be an alcohol dehydrogenase/oxidase; however, according to genetic studies, efficient long-chain-alcohol oxidation additionally required the Tat-dependent protein LaoB. In the present study, the Lao system was further characterized. In vivo analysis revealed that the Lao system complements the substrate spectrum of the well-described Exa system, which is required for growth with ethanol and other short-chain alcohols. Mutational analysis revealed that the Tat site of LaoB was required for long-chain-alcohol oxidation activity, strongly suggesting a periplasmic localization of the complex. Purified LaoA was fully active only when copurified with LaoB. Interestingly, in vitro activity of the purified LaoAB complex also depended on the presence of the Tat site. The copurified LaoAB complex contained a flavin cofactor and preferentially oxidized a range of saturated, unbranched primary alcohols. Furthermore, the LaoAB complex could reduce cytochrome c₅₅₀-type redox carriers like ExaB, a subunit of the Exa alcohol dehydrogenase system. LaoAB complex activity was stimulated by rhamnolipids in vitro. In summary, LaoAB constitutes an unprecedented protein complex with specific properties apparently required for oxidizing long-chain alcohols. IMPORTANCE Pseudomonas aeruginosa is a major threat to public health. Its ability to thrive in clinical settings, water distribution systems, or even jet fuel tanks is linked to detoxification and degradation of diverse hydrophobic substrates that are metabolized via alcohol intermediates. Our study illustrates a novel flavoprotein long-chain-alcohol dehydrogenase consisting of a facultative two-subunit complex, which is unique among related enzymes, while the homologs of the corresponding genes are found in numerous bacterial genomes. Understanding the catalytic and compartmentalization processes involved is of great interest for biotechnological and hygiene research, as it may be a potential starting point for rationally designing novel antibacterial substances with high specificity against this opportunistic pathogen.

Keywords: Pseudomonas aeruginosa; SDS degradation; Tat system; alcohol dehydrogenase; cytochrome c; flavin-dependent alcohol dehydrogenase; flavoenzymes; hitchhiker mechanism; long-chain-alcohol oxidation; protein complex; rhamnolipids.

FIG 1

Growth of P. aeruginosa PAO1 wild type (squares) and the deletion mutants PAO1ΔexaA (circles) and PAO1ΔlaoA (diamonds) with the primary alcohols (C₂ to C₁₄) indicated in the individual panels. Alcohols were provided at carbon amounts equivalent to 3.5 mM 1-dodecanol as sole carbon and energy sources. Results of one representative of three independent experiments are shown. Error bars indicate standard deviations (n = 3).

FIG 2

Analysis of the Tat motif of LaoB with P. aeruginosa strains PAO1 (filled circles), PAO1ΔlaoB (filled triangles), PAO1ΔlaoB(pUCP18::laoB) (open circles), and PAO1ΔlaoB(pUCP18::laoB*) (open triangles). (A) Growth of strains with 3.5 mM 1-dodecanol as the sole carbon and energy source. (B) SDS quantification of supernatants from cultures grown with 3.5 mM SDS as the sole carbon and energy source. Error bars indicate standard deviations (n = 3).

FIG 3

In vitro characterization of the LaoAB complex. In all experiments, LaoA and LaoB were coexpressed in E. coli Rosetta; in all panels, LaoA-containing signals are indicated by black triangles, while signals containing LaoB only are indicated by white triangles. (A) SDS-PAGE (12.5%) of purified LaoA-Strep coexpressed with untagged LaoB (2). (B) Activity staining. Lane 1, LaoA-Strep catalyzed 1-dodecanol oxidation activity bands in native PAGE (6%). Lanes 2 to 4, Western blot of LaoA-Strep separated in a native PAGE gel (6%) and blotted with HRP-conjugated Strep-tag antibody. (C) Two-dimensional gel electrophoresis of purified LaoA-Strep comprising native PAGE (6%, activity-stained for reference) followed by SDS-PAGE and detection of LaoA-Strep and LaoB by silver staining. (D) Western blot analysis with HRP-conjugated Strep or His tag antibody of purified LaoA-Strep and LaoB-His fractions (10 μg protein each) after separation by SDS-PAGE. Strep, streptavidin-purified fraction; His, Ni-NTA-purified fraction.

FIG 4

Activity staining with purified proteins. Oxidation of 1-dodecanol by 10 μg purified LaoA-Strep (lane 1), LaoA-Strep coexpressed with LaoB* (lane 2), or LaoA-Strep coexpressed with LaoB (lane 3) in native PAGE (6%).

FIG 5

Localization of LaoB in the soluble cell fraction, as determined by Western blotting and immunodetection. Cells of P. aeruginosa PAO1ΔlaoB(pME6032::laoAB-Strep) (expression strain) and PAO1ΔlaoB(pME6032) (vector control) were disrupted (cell extract) and cleared of debris before the membrane fraction was separated from the soluble fraction by ultracentrifugation. Lanes contain membrane fraction (lane 1) and membrane-free soluble fraction (lane 2) of the expression strain without enrichment by affinity chromatography, and cell extract of vector control (lane 3), cell extract (lane 4), and membrane-free cell soluble fraction of the production strain (lane 5) after enrichment by affinity chromatography. The recombinantly expressed and purified protein Strep-PqsD (20 ng, 35 kDa; lane R) of P. aeruginosa PAO1 (our unpublished data) served as a reference for the staining. Complex samples were loaded at 50 μg per lane. The prestained PAGE ruler, which features a StrepII epitope in the 55-kDa band, was used as a marker and transfer control (M).

FIG 6

Verification of nonyl aldehyde as reaction product of LaoAB with 1-nonanol. Recombinant LaoAB (1 μM) was incubated with 50 μM 1-nonanol, the most efficient substrate of the reaction in solution, and an excess of PMS and DCPIP as electron acceptors. (A) Aldehydes formed were derivatized with 2,4-DNPH (2 mM) and analyzed by HPLC (red trace) coupled to electrospray ionization with MS (ESI-MS). Identity of the putative nonyl-2,4-dinitrophenylhydrazone was confirmed with authentic nonyl aldehyde (blue trace) and suitable controls (black traces). (B) The mass spectrum obtained between 18.4 and 18.5 min shows the nonyl adduct (322 m/z), as well as a 2,4-dinitrophenyl azan fragmentation product (182 m/z). (C) The UV spectrum of the compound is characteristic of DNPH adducts, with an absorption maximum at 360 nm.

FIG 7

Evaluation of electron acceptors for the LaoAB complex in vitro. (A) Catalytic activity of purified LaoAB complex with 800 μM 1-nonanol in the presence of different electron acceptors. Horse heart cytochrome c (Eq.d. cyt.c) and ExaB reductions were calculated based on differential extinction coefficient of cytochrome c. Error bars indicate standard deviations (n = 3). (B) Spectroscopic analysis of ExaB reduction by the LaoAB complex. The gray trace corresponds to ExaB as isolated (30 μM), and the green trace shows ExaB after oxidation with ammonium persulfate. Reduction of ExaB with the LaoAB complex and 1-nonanol led to the blue spectrum.

FIG 8

Influence of rhamnolipids on catalytic activity of the LaoAB complex. (A) Influence of 0.8% rhamnolipids. Substrate concentrations used for the assay were kept below the respective solubility limits, eliminating the influence on substrate solubilization by rhamnolipids (1-butanol, 10 mM; 1-hexanol, 5 mM; 1-heptanol, 2 mM; 1-octanol, 1 mM; 1-nonanol, 800 μM; 1-decanol, 100 μM). (B) Reaction rates as a function of rhamnolipid concentration at 800 μM 1-nonanol. A hyperbolic function can be used to fit the data, resulting in an EC₅₀ of 0.16% (wt/vol). (C) Hanes-Woolf representation of LaoAB complex substrate dependency in the absence (squares) and presence (diamonds) of rhamnolipids. Addition of 0.3% rhamnolipids leads to a non-Michaelis-Menten characteristic of the reaction. In the absence of rhamnolipids, LaoAB complex catalysis corresponds to a typical Michaelis-Menten model (solid line). (D) Arrhenius plot illustrating the temperature dependency of the LaoAB complex at different concentrations of rhamnolipids (0% [diamonds], 0.2% [circles], 0.5% [squares], and 0.8% [triangles]). Slopes of the linear fits correspond to activation energies of the reaction under the respective conditions.