The Crystal Structure of Burkholderia cenocepacia DfsA Provides

May 19, 2016 - B. cenocepacia, the quorum sensing cell−cell communication system is involved in different processes that are important for bacterial...
0 downloads 3 Views 2MB Size
Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

The crystal structure of Burkholderia cenocepacia DfsA provides insights into substrate recognition and quorum sensing fatty acid biosynthesis Francesca Spadaro, Viola C. Scoffone, Laurent R. Chiarelli, Marco Fumagalli, Silvia Buroni, Giovanna Riccardi, and Federico Forneris Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b00178 • Publication Date (Web): 19 May 2016 Downloaded from http://pubs.acs.org on May 25, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

The crystal structure of Burkholderia cenocepacia DfsA provides insights into substrate recognition and quorum sensing fatty acid biosynthesis Francesca Spadaro‡, Viola C. Scoffone‡, Laurent R. Chiarelli*, Marco Fumagalli, Silvia Buroni, Giovanna Riccardi, Federico Forneris*

Department of Biology and Biotechnology "Lazzaro Spallanzani", University of Pavia, Via Ferrata 9/A, 27100 Pavia (Italy).

KEYWORDS: Quorum sensing; Burkholderia cenocepacia; Crotonase; bifunctional enzyme.

Abstract

Burkholderia cenocepacia is a major concern among respiratory tract infections in cystic fibrosis patients. This pathogen is particularly difficult to treat because of its high level of resistance to the clinically relevant antimicrobial agents. In B. cenocepacia, the quorum sensing cell-cell communication system is involved in different processes important for the bacterial virulence, such as biofilm formation or protease and siderophore production. Targeting the enzymes involved in this process represents a promising therapeutic approach. With the aim of finding effective quorum sensing inhibitors, we have determined the three-dimensional structure of B. cenocepacia diffusible 1 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

factor synthase A, DfsA. This bifunctional crotonase (dehydratase/thioesterase) produces the characteristic quorum sensing molecule of B. cenocepacia, cis-2-dodecenoic acid or BDSF, starting from 3-hydroxydodecanoyl-acyl carrier protein. Unexpectedly, the crystal structure revealed the presence of a lipid molecule into the catalytic site of the enzyme, which was identified as dodecanoic acid. Our biochemical characterization shows that DfsA is able to use dodecanoyl-acyl carrier protein as a substrate, demonstrating that dodecanoic acid, the product of this reaction, is released very slowly from the DfsA active site, therefore acting as a DfsA inhibitor. This molecule shows an unprecedented conformational arrangement inside the DfsA active site. In contrast with previous hypotheses, our data illustrate how DfsA and closely-related homologous enzymes can recognize long hydrophobic substrates without large conformational changes or assistance by additional regulator molecules. The elucidation of the substrate binding mode in DfsA provides the starting point for structure-based drug discovery studies targeting B. cenocepacia quorum sensingassisted virulence.

Introduction The ability of Burkholderia cepacia complex (Bcc) bacteria to spread among Cystic Fibrosis (CF) patients is particularly dangerous, as strains like Burkholderia cenocepacia are able to cause a fatal necrotizing pneumonia (Cepacia syndrome). The high-level multidrug resistance of B. cenocepacia causes a chronic persistence in the host, associated with inflammatory responses leading to rapid deterioration of lung function in CF patients (1). Two approaches can be pursued to fight these infections: one is to search for new antimicrobials, while the other one is to deepen the knowledge on virulence factors which are involved in the pathogenicity of this bacterium in order to reduce them (2). Quorum sensing (QS) is a process developed by many bacterial pathogens to coordinate the expression of genes involved in different pathogenic processes (3). These include biofilm 2 ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

formation, the production of proteases, siderophores and secondary metabolites (4, 5). B. cenocepacia strains growing as a biofilm show high resistance to antibiotic treatment and disinfectants. It is well established that in QS knock-out strains, biofilm formation is reduced, antibiotic treatment is more effective and virulence is reduced (6, 7). Combinations of QS inhibitors and tobramycin greatly increase the efficacy of the antibiotic against biofilm formation, and reduce virulence in Caenorhabditis elegans and Galleria mellonella (7, 8), indicating that quenching of QS is a very promising strategy for therapeutic approaches. In B. cenocepacia, the main QS molecules are N-acylhomoserine lactones (AHLs) (9) and cis-2dodecenoic acid, also called Burkholderia Diffusible Signal Factor (BDSF) (10). BDSF belongs to the Diffusible Signal Factor (DSF) QS system, recently emerging as a conserved cell-cell communication mechanism in bacteria (11).

It has been demonstrated that in B. cenocepacia J2325 the level of AHL in the absence of BDSF synthesis is significantly reduced, suggesting that the two QS systems interact, with a dominant role for the BDSF system (6). The BDSF molecule is produced by DfsA (diffusible factor synthase A, also known as BCAM0581), a bifunctional enzyme which catalyzes the dehydration of the 3hydroxydodecanoyl-acyl carrier protein (ACP) to cis-2-dodecenoyl-ACP, followed by cleavage of the bound acyl thioester, releasing free cis-2-dodecenoic acid (Figure 1A) (12). Although DfsA shares only 37% identity with the more structurally characterized Xhantomonas campestris pv. campestris (Xcc) DSF synthase RpfF, the amino acid network surrounding the active sites of the two enzymes is fully conserved (Figure 1B) (12, 13), as confirmed by the ability of DfsA to replace RpfF in Xcc (12). Both enzymes show low specificity for the thioesterase activity, being active on a broad range of acyl-ACPs substrates in vitro (12, 14). In Xcc, it was proposed that the dehydratase and thioesterase activities might be coupled in vivo, avoiding detrimental cleavage of acyl-ACPs destined to membrane synthesis (12). This was initially supported by the identification of RpfB, an acyl-CoA ligase involved in fatty acid recycling (14) that regulates RpfF function in Xcc. 3 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Interaction of RpfF with RpfB modulates enzymatic activity and, possibly, induces conformational changes affecting substrate specificity (13, 15). However, no homologs of RpfB have been reported in B. cenocepacia until now, suggesting that the activity of these two enzymes might not be necessarily coupled for the regulation of BDSF production in vivo. Furthermore, the accurate description of the molecular mechanisms of BDSF biosynthesis in RpfF and DfsA is still incomplete; the three-dimensional structures of these enzymes in complex with their substrates/products may also provide useful templates for structure-based drug design. To understand the molecular mechanisms of substrate specificity and molecular recognition in QS thioesterases and hydratases, we have determined the three-dimensional structure of B. cenocepacia DfsA and performed biochemical characterization of the enzyme.

Experimental Procedures Recombinant expression and purification. The DfsA coding region was amplified by PCR using B. cenocepacia J2315 genome as template. The gene was transferred into the pET SUMO expression vector (Invitrogen) for recombinant expression. Transformed E. coli BL21 (DE3) One Shot® (Invitrogen) were grown at 37 ºC in Luria-Bertani (LB) broth (DIFCO) until reached OD600 = 0.4-0.6. Induction was performed by adding 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) to the culture, and the temperature was lowered to 25 ºC. Induced culture was gently shaken for 18 hours before harvesting. The cells were harvested and resuspended in Lysis buffer (30 mM NaH2PO4 pH 8, 300 mM NaCl, 10 mM imidazole, 5% v/v glycerol) supplemented with protease inhibitors cocktail (Sigma) and 1.5 µM DNase I (Sigma). Cells were lysed by sonication and the lysate was clarified by centrifugation at 12000 g for 1 hour at 4 °C. The supernatant was applied to a 1 mL column packed with Ni-NTA beads (Qiagen), previously equilibrated with Lysis buffer. The column was then washed with 15 mL of Wash buffer (Lysis buffer supplemented with 20 mM imidazole) to remove non-specifically bound contaminants. His-tagged DfsA was eluted using 4 ACS Paragon Plus Environment

Page 4 of 34

Page 5 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Elution buffer (Lysis buffer supplemented with 250 mM imidazole). Pooled fractions were analyzed using reducing SDS-PAGE on a 12% gel. The pool containing recombinant DfsA was dialyzed against a solution of 30 mM NaH2PO4 pH 8, 150 mM NaCl, 1 mM 1,4-dithio-DL-threitol (DTT), 5% v/v glycerol for 18 hours, and then subject to treatment with 10 U mg-1 of SUMO protease (Invitrogen) at 4 ºC, for 12 hours. The sample was dialyzed again against a buffer containing 25 mM 2-(N-morpholino) ethanesulfonic acid (MES) pH 6.5, 250 mM NaCl, 1 mM DTT, and further purified by size exclusion chromatography using HiLoad 16/60 Superdex 75 (GE Healthcare). The fractions containing DfsA were pooled and concentrated by ultrafiltration (10 kDa cutoff) with Vivacon 2 (Sartorius), reaching a final concentration of 0.5 mg mL-1.

Crystallization and Data Collection. Purified DfsA at 0.5 mg mL-1 was used in crystallization screening experiments performed using the vapor diffusion sitting drop method at 20 ºC, by making droplets composed of an equal volume (1 µL) of protein and reservoir. Crystals suitable for diffraction experiments appeared after 15 days in a reservoir solution containing 50 mM Tris-HCl, pH 8.5, 50 mM Li2SO4, 50 mM Na2SO4, and 30% PEG 400. The crystals were harvested using nylon cryoloops (Hampton Research) and flash-frozen in liquid nitrogen for diffraction testing without the need for additional cryo protection additives. X-ray diffraction data were collected from single crystals of DfsA at beamlines ID23 EH1 at the European Synchrotron Radiation Facility (ESRF, Grenoble, France) and X06SA (PX) at the Swiss Light Source (SLS, Villigen, Switzerland). The diffraction images were processed and scaled using the softwares iMosflm (16) for indexing and integration and Aimless (17) for space group assignment and scaling. The observed resolution limits of the diffraction data were selected by evaluating the correlations between half datasets (CC1/2) (18). Data collection statistics are summarized in Table 1.

Structure Determination. The DfsA structure was determined by molecular replacement using Phaser (19) using the structure of RpfF protein from X. campestris (PDB entry 3M6M) (13) as 5 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

search model. The overall sequence identity is 37% according to pBLAST. After visual inspection of the electron density map after molecular replacement, automated model building in the electron density was performed using ARP/wARP (20). Extensive manual correction of the model and insertion of solvent and ligand molecules were performed using PHENIX suite (21) and COOT (22). Structural refinement was carried out manually with phenix.refine, using three-fold noncrystallographic symmetry restraints, isotropic B-factors and TLS groups after their automated identification in phenix.refine (21). Model quality and validation were assessed using the MolProbity Server (23) and the wwPDB validation server (24). Final refinement statistics are summarized in Table 1. Structural figures were created using PyMol (25). Coordinates and structure factors have been deposited at the Protein Data Bank (PDB) with accession number 5FUS.

Lipid extraction. Extraction of protein-bound lipid was performed as described (26). The extracted sample was dialyzed against a solution of 50 mM ammonium acetate pH 6.8 and 1 mM DTT at 4 ºC for 18 hours. 13.0 mL of protein solution (0.42 mg mL-1) were mixed with 34.7 mL of chloroform and 17.3 mL of methanol in a separating funnel. Once phase separation was reached, the organic phase was recovered and collected in a glass tube. Evaporation to dryness of the organic phase was obtained under reduced pressure.

Sample preparation for GC/MS analysis. The dried, extracted lipid material was dissolved in 1 mL of a 3 N HCl/methanol solution, reacted at 110 °C for 20 min and then cooled to room temperature. Two milliliters of ethyl acetate were added to the sample and the preparation was subsequently shaken for 5 min, and centrifuged at 14000 g for 15 minutes. The ethyl acetate layer was transferred to another tube, and the solvent was evaporated to dryness under nitrogen atmosphere. The methylated residue was dissolved in 0.1 mL of ethyl acetate. One microliter of this sample was subject to gas chromatography and mass spectrometry (GC/MS) analysis.

6 ACS Paragon Plus Environment

Page 6 of 34

Page 7 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Lipid identification using GC/MS. Lipid identification was carried out using a Thermo Scientific GC/MS system (TraceDSQII mass spectrometer, TraceGCUltra gas cromatograph, TriPlus autosampler), Xcalibur MS Software Version 2.1 (including NIST Mass Spectral Library (NIST 08) and Wiley Registry of Mass Spectral Data 8th Edition), with a Restek capillary column, Rtx-5MS 30 m x 0.25 mm x 0.25 µm film thickness. Helium was used as the carrier gas. The column oven temperature was initially maintained at 45 °C for 4 min and then increased linearly to 175 °C, at a rate of 13 °C min-1, and finally to 215 °C, at a rate of 4 °C min-1. Lipid identification was established by comparing the mass spectra of the peaks with those available in the NIST mass spectral library (Wiley registry NIST 08) (http://www.nist.gov/srd/).

Acyl-ACP

preparation

Acyl-ACP,

were

prepared

using

acyl-CoA,

Bacillus

subtilis

phosphopantetheinyl transferase (Sfp) and the E. coli acyl carrier protein (both obtained in recombinant form), as previously reported (27, 28). Briefly, for C12-ACP and C14-ACP, the reaction mixture contained: 50 µM holo-ACP, 150 µM Lauroyl-(C12)-CoA or myristoyl-(C14)CoA, Lithium salts (C12) (Sigma-Aldrich), 10 µM Sfp (Bacillus subtilis phosphopantetheinyl transferase), 2.5 mM MgCl2, 0.5 mM Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) and 75 mM MES buffer, pH 6.0, in a final volume of 1 mL. Reaction mixture was incubated at 37 °C for 3 hours. C8-ACP and C4-ACP were obtained incubating 1 mM ACP with 10 µM Sfp, 10 mM octanoyl-(C8)-CoA or butyryl-(C4)-CoA, 10 µM MgCl2, 1 mM DTT in 100 mM Tris-HCl pH 8.0, at 37 °C for 16 h. The production of acyl-ACP was checked by electrophoresis on 20% polyacrylamide gels containing 0.8 M urea, a conformationally sensitive gel electrophoresis with a concentration of urea optimized for the separation (29). Spf-protein precipitation was achieved by adding to the reaction mixture 50 mg of ammonium sulfate, followed by incubation for 1 hour at 4 °C. Then, the sample was centrifuged at 18000 g for 10 min. at 4 °C, the supernatant was collected, mixed with 2 volumes of cold acetone and placed at -20 °C for 18 hours. The mixture was centrifuged at 18000 g at 4° C for 1 hour. The supernatant was discarded and the corresponding 7 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 34

pellet containing C12-ACP, was dried under reduced pressure. The pellet was resuspended in 50 µL of 100 mM Tris-HCl pH 7.5 and lipid quantitation was assessed by spectrophotometric analysis at 280 nm with a calculated extinction coefficient for C12-ACP of 1490 M-1 cm-1.

DfsA enzymatic activity. DfsA thioesterase activity was assessed in a reaction mixture containing 50 µM C12-ACP, 1 mM DTT, 100 mM 4-(2-Hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES) pH 7.5, and 15.6 µM DfsA, in a final volume of 100 µl. The reaction was incubated at 37 °C for 10 min. The enzyme was inactivated by placing the sample at 95 °C for 2 min, and the release of holo-ACP checked by conformational sensitive electrophoresis on 20% polyacrylamide gel containing 0.8 M urea. Alternatively, DfsA enzyme activity was determined by measuring the holo-ACP formation through the

titration

of

the

free

thiol

group

formed

upon

thioesterase

reaction

with

dichlorophenylindophenol (DCPIP; ε= 19100 M-1 cm-1) (30). The reaction mixture typically contained 50 mM HEPES pH 7.5, 0.005% Nonidet P-40, 0.075 mM DCPIP, 2 µM DfsA and was started by addition of C12-ACP. Steady-state kinetic parameters were determined in triplicate, assaying the enzyme at different concentrations of substrate, and the kinetic constants Km and kcat determined fitting the data to the Michaelis-Menten equation with Origin 8 software. To determine the inhibitory effects of dodecenoic and dodecanoic acids (Sigma-Aldrich, dissolved in dimethyl sulfoxide), the reaction was incubated in the presence of different concentrations of the carboxylic acid for three minutes, and products analysed by conformational sensitive electrophoresis. IC50 values were then determined by the DCPIP assay, measuring the enzyme activity at different dodecenoic and dodecanoic acid concentrations, and using the following equation:

8 ACS Paragon Plus Environment

Page 9 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Propagation of statistical error value was carried out as described (31).

Fatty acid identification using ESI-MS. The release of dodecanoic acid from C12-ACP during the DfsA reaction was assessed by electron spray ionization mass spectrometry (ESI-MS) analysis, using a Thermo LTQ-XL mass spectrometer. The reaction mixtures (in a volume of 200 µl), previously inactivated at 95 °C for 2 min, were partitioned between water and ethyl acetate, and the aqueous layer was extracted with ethyl acetate. Combined organic layer was washed with brine and dried over Na2SO4. Solvent was removed under reduced pressure, and residues were dissolved in methanol for ESI-MS. Mass spectra were generated in positive ion mode and MS/MS spectra, obtained by collision-induced dissociation in the linear ion trap, were recorded with a resolution of 1 Da (m/z).

Results Biochemical and structural characterization of DfsA. We successfully recombinantly expressed and purified functional DfsA from B. cenocepacia in E. coli. To assess the functionality of the recombinant enzyme, we designed an assay to detect DfsA thioesterase activity in vitro, using C12-ACP as substrate. As shown by the conformationally sensitive gel electrophoresis in Figure 2A, holo-ACP was released from the substrate, indicating that recombinant DfsA efficiently cleaves the C12-ACP thioester bond. The reaction products were analyzed using ESI-MS for the identification of the released lipid moiety, confirming the presence of the expected dodecanoic acid (Figure 2B and 2C). Moreover, using a DCPIP based enzyme activity assay, we performed a steady state kinetic analysis of DfsA, which kinetic constants comparable with previously characterized Burkholderia QS synthases that catalyze the thioester cleavage of acyl-ACP (30, 32) (Figure 2D, Table 2).

9 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Even though DfsA has been previously reported to be able to hydrolyze different C12-ACP substrates, such as dodecanoyl-ACP, 3-hydroxydodecanoyl-ACP, cis-2-dodecenoyl-ACP and trans2-dodecenoyl-ACP (12), no data are available on enzymatic activity on different length acyl-ACP substrates. Therefore, we performed biochemical characterization of DfsA enzymatic activity using other acyl-ACP substrates, bearing shorter and longer acyl chains than C12 (i.e., C4-ACP, C8-ACP and C14-ACP, Supplementary Figure 1). Besides C12-ACP, the enzyme showed activity only using C8-ACP as substrate, being unable to hydrolyse the longer C14-ACP or shorter C4-ACP substrates. Furthermore, the enzyme showed low efficiency against C8-ACP, with kcat/Km reduced about 4-fold compared to C12-ACP, mainly due to a lower substrate affinity, as demonstrated by the higher Km value (Table 2, Supplementary Figure 1). Next, we crystallized DfsA and solved its structure at 1.9-Å resolution by molecular replacement (19) using the crystal structure of RpfF trimer from X. campestris as search model, given the 37% sequence identity (13). The electron density maps obtained from molecular replacement using PHASER (19) were used for automated model building using ARP/wARP (20), yielding an almost complete structural model for DfsA, which was further improved by alternated cycles of manual model building and ligand modeling using COOT (22) and refinement using phenix.refine (21). Data collection and refinement statistics are summarized in Table 1. Each DfsA monomer shows the typical ββα crotonase superfamily fold, composed of two almost perpendicular β-sheets surrounded by α-helices (Figure 1C) (33). The asymmetric unit shows three DfsA molecules, arranged to form a compact, flat homotrimer, extended by 75 Å in diameter and 35 Å in width, with each monomer rotated by approximately 120 degrees relative to the others, and its substrate-binding pocket pointing towards the solvent (Figure 1D). In the crystal packing, two DfsA trimers were found stacking onto each other, however this hexameric assembly could not be observed in solution and was indicated by PISA analysis (34) as purely induced by crystal contacts. The DfsA quaternary arrangement shows the typical crotonase monomer self-association fold, with two Cterminal helices folding over the catalytic core of the enzyme, generating an integral catalytic cavity 10 ACS Paragon Plus Environment

Page 10 of 34

Page 11 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

within a single monomer and constituting part of the trimeric interface (33). Formation of the DfsA trimer results in a total buried surface area of 2514 Å2, comprising 41 hydrogen bounds and 17 salt bridges.

A DALI search (35) was performed to identify protein structures sharing similarity to DfsA. The outcome showed that over 250 protein structures share the same structural fold, with r.m.s.d. values ranging from 1.6 to 3.8 Å. Notably, numerous structural homologs of DfsA identified by DALI analysis were co-crystallized in presence of their substrate or product analogs. As expected, all structures belong to the crotonase superfamily. Crotonases are enzymes that catalyze the hydratation of a trans-2-acyl-CoA to the 3-hydroxy species in the β-oxidative cycle of fatty acid degradation (33). The catalytic activity of crotonases requires the activation of a water molecule for the nucleophilic attack at the carbon atom in position 3 of the unsaturated fatty acid substrate and the protonation of its C atom in position 2. These reaction steps are mediated by a structurally conserved amino acid network composed of two glutamate residues facing opposite sides of the catalytic site and two amino groups that form an oxyanion hole for the stabilization of the enolate anion transition state (33). In DfsA, the two catalytic glutamate residues required for activity are Glu138 and Glu158, whereas the oxyanion hole is defined by two glycine residues, Gly83 and Gly135 (Figure 1B and 1C) (12). DALI analysis revealed that DfsA is structurally very similar to crotonases displaying monomer self-association as well as domain swapping (Figure 3a), supporting the notion that these two architectures do not seem to influence the crotonase function or quaternary structure (33). Analysis of the superposition with its closest structural homolog Xcc DFS synthase RpfF (r.m.s.d. 1.12 Å) also shows nearly complete conservation of the architecture of the catalytic site (Figure 1B). At the quaternary structure level, the DfsA homotrimeric arrangement is nearly indistinguishable from homologous crotonase enzymes in superpositions, regardless the presence or the absence of domain swapping elements (Supplementary Figure 2). RpfF is the closest known sequence homolog of DfsA (37% sequence identity), and has been recently 11 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

demonstrated to possess the two same dehydratase and thioesterase activities in vitro (36). Among structural homologs identified by DALI, RpfF displays the closest structural similarity to DfsA also in its quaternary structure. (Supplementary Figure 2).

The DfsA substrate binding pocket is characterized by an elongated cavity that connects the substrate access site near the enzyme's C-terminus, the catalytic residues and a deep tunnel, extended for approximately 20 Å (Figure 1C), reaching residue Ile114 inside the enzyme core. The side chain conformations of aromatic residues Phe88, Tyr103, Phe161, Phe44, Tpr255, and hydrophobic residues Met167, Cys107, Val49, and Val111 create a regular cylindrical shape inside the tunnel, resulting in a strongly hydrophobic cavity facing the two catalytic glutamates immediately preceding the substrate access site. Comparison with known enoyl-CoA hydratase structures shows that the DfsA hydrophobic tunnel extends further, reaching deeper inside the enzyme. Interestingly, despite only 37% sequence identity between DfsA and RpfF, all the residues shaping the hydrophobic tunnel are identical in the two enzymes (Figure 1B).

Identification and characterization of a lipid molecule in DfsA catalytic site. The experimental electron density map showed the presence of a molecule trapped inside the substrate binding cavity, deeply buried in the hydrophobic tunnel proximate to the catalytic site, but at least 4 Å away from the oxyanion hole and the catalytic glutamate residues (Figure 4A, Supplementary Figure 3). The elongated shape of the electron density, and the strongly hydrophobic environment of the tunnel surrounding the ligand suggest the presence of a lipid-like molecule. Despite the high resolution and quality of the experimental data, the exact molecular identity for this ligand molecule could not be derived directly from the electron density map. Therefore, GCMS analysis of the low molecular weight component extracted from denatured protein preparations was performed. After clearance of the GC/MS results from known contaminations due to plasticware (37, 38), the most prominent GC peak displays a retention time of 14.54 minutes. 12 ACS Paragon Plus Environment

Page 12 of 34

Page 13 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Comparison of the mass spectra of the peak with those available in the NIST mass spectral library unambiguously identified this peak as dodecanoic (lauric) acid (Figure 4B).

The fatty acid molecule shows an elongated conformation, occupying the entire DfsA substrate tunnel and interacting with virtually all the hydrophobic residues responsible for shaping the substrate cavity (Figure 4C). By comparing the dodecanoic acid conformation in DfsA with the three-dimensional structures of enoyl-CoA hydratases in complex with their substrates identified by DALI search, we observed that the ligand does not adopt the canonical substrate conformation displayed by the shorter lipid moieties of enoyl-CoA hydratase substrates (Figure 5A) (33, 39), providing a possible explanation on how longer substrates such as BDSF can be accommodated inside the enzyme's catalytic pocket.

To further confirm that dodecanoic acid is able to interact with the enzyme, we further assayed DfsA activity in the presence of dodecanoic and dodecenoic acid (Figure 4D). As shown in Fig. 4E, at subsaturating C12-ACP concentrations (20 µM) the dodecanoic acid has proven to inhibit DfsA, showing an IC50 of 21.1 ± 2.8 µM. On the contrary, the DfsA physiological reaction product dodecenoic acid showed inhibitory effects only at very high concentrations (IC50 599 ± 57 µM) (Figure 4E). These results suggest that the dodecanoic acid molecule identified into the active site pocket is conceivably an end product of the enzyme reaction, which is slowly released. Attempts to remove the lipid molecule from DfsA for further biochemical and structural studies were not successful, as the electron density of DfsA crystal structures after lipid removal treatment consistently displayed the presence of the lauric acid molecule in the same position in the catalytic site as for the untreated enzyme.

Discussion 13 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

We have determined the crystal structure of DfsA, a bifunctional enzyme involved in the synthesis of the QS-signaling molecule BDSF in B. cenocepacia. The three-dimensional structure of DfsA shows that a fatty acid molecule, identified as lauric acid by GC/MS, is stably trapped inside the enzyme's catalytic site. The DfsA samples were not treated with any lipid compounds prior to crystallization. We therefore conclude that the molecule identified in the experimental electron density was probably an endogenous lipid molecule commonly present in E. coli (40), trapped inside the enzyme during the recombinant expression procedures. Despite the presence of such endogenous inhibitory molecule, our biochemical analysis shows that recombinant DfsA is an active enzyme, as illustrated by its capability to convert the substrate analog dodecanoyl-ACP into dodecanoic acid. Interestingly, dodecanoic acid is the product of this DfsA reaction, suggesting that this molecule may display an inhibitory effect due to very slow release of the extremely hydrophobic product from the active site after catalysis. This hypothesis is strongly supported by the relatively low IC50 values displayed by dodecanoic in DfsA enzymatic assays (Figure 4E). Conversely, the physiological product BDSF inhibits the enzyme only at very high concentrations, indicating lower affinity for the enzyme, conceivably due to the reduced flexibility of this molecule. It has been proposed that the very broad specificity for thioesterase activity of DFS synthases would be detrimental to membrane lipid synthesis of bacteria (12), but, unlike Xcc, B. cenocepacia does not possess an analogue of counteracting enzyme RpfB (14). From this point of view, a low rate of product release could avoid unnecessary cleavage of C12 acyl-ACP. Similarly, the detrimental cleavage of others acyl-ACP of different length could be prevented by the high selectivity of the enzyme for 12-carbon atom chain acyl-ACP substrates, as highlighted by DfsA inactivity on substrates longer than C12 and its lower affinity for the shorter C8 acyl chain. Such physiologically-related role of lauric acid in DfsA provides insights on how long lipid chains can be hosted inside the enzyme's substrate binding site. Moreover, in the three-dimensional structure of ligand-free RpfF, Cheng et al. could not rule out how the BSDF precursor molecule can be accommodated inside the catalytic site, expecting conformational changes in the helical 14 ACS Paragon Plus Environment

Page 14 of 34

Page 15 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

arrangement of the enzyme to allow catalysis (13). Given the almost complete conservation of the hydrophobic environment of DfsA and RpfF (Figure 1B), the 12-mer lipid molecule observed in the DfsA structure illustrates how the long hydrophobic BDSF precursor (sterically very similar to dodecanoic acid) can fit into both DfsA and RpfF, simply by altering the side chain positions of Phe91 (Phe88 in DfsA) and Met170 (Met167 in DfsA), without additional conformational changes (Figure 5B). Furthermore, by taking advantage of the extended structural repertoire of enoyl-CoA hydratases in complex with their substrates, we could easily model the endogenous DfsA substrate molecule. Such model was obtained by merging the phosphopantheine fragment found on the external side of an enoyl-CoA hydratase catalytic cavity with the lipid tail of dodecanoic acid, almost without changing the position of the aliphatic chain of the fatty acid and without structural clashes in the catalytic site (Figure 5C). This model illustrates how the BDSF precursor substrate may interact with DfsA and RpfF enzymes during catalysis, with its C2 and C3 aliphatic atoms approaching to the catalytic glutamates and the oxyanion hole. Upon breakage of the BDSF-ACP thioester bond, the hydrophobic cavity hosting the BDSF is exposed, allowing slow release of the signaling molecule outside of the substrate binding pocket. Given the striking sequence similarity between DfsA and RpfF catalytic sites, a structure/function explanation for the different catalytic functions of these two enzymes relies almost exclusively on the specific RpfF binding partners. Interactions with RpfB/RpfC may influence substrate accessibility, catalysis and product release of the core RpfF module (13, 36), likely regulating the catalytic spectrum of this enzyme in vivo when compared to DfsA. In conclusion, our results contribute to the knowledge on QS pathway of B. cenocepacia, an important target for the development of new molecules able to attenuate the virulence of this dangerous opportunistic pathogen and favoring the therapeutic effect in vivo (2). Our crystal structure of DfsA in complex with its reaction product, lauric acid, provides unprecedented insights on the molecular mechanisms of substrate processing and product release in QS synthases, and provides a novel structural template for future in silico drug discovery studies for the identification 15 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of new lead candidates (41) that may effectively target virulence as an alternative and a complementary strategy to the discovery of new antimicrobials (42).

16 ACS Paragon Plus Environment

Page 16 of 34

Page 17 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

FIGURE LEGENDS Figure 1. Biochemical and structural features of the DfsA enzyme. (A) schematic summary of the reactions catalyzed by DfsA, enabling firstly the conversion from 3-hydroxydodecanoyl-S-ACP to dodecenoyl-S-ACP, then its hydrolysis to dodecenoic acid (BDSF) with release of free ACP. (B) Primary sequence alignment of DfsA with homologous previously characterized enoyl-CoA hydratases. Stars indicate the conserved catalytic glutamate (red) and the oxyanion hole (blue) residues responsible for enzymatic activity. Blue triangles highlight amino acid residues shaping the hydrophobic substrate cavity, as identified in the three-dimensional structure of DfsA. Sequences displayed in the alignment: DfsA from B. cenocepacia J2315 (UniProt B4EKM5); RpfF from X. campestris (UniProt Q7CLS3); Enoyl-CoA-hydratase GK1992 from G. kaustophilus (UniProt Q5KYF9); Enoyl-CoA-hydratase GBAA_4761 from B. antracis (UniProt Q81L70); Mitochondrial Enoyl-CoA-hydratase Echs1 from R. norvegicus (UniProt P14604); Enoyl-CoA-hydratase YngF from B. antracis (UniProt Q81Q82); Mitochondrial Methylglutaconyl-CoA hydratase from H. sapiens (UniProt Q13825). Image created using ESPRIPT3 (43). (C) Crystal structure of DfsA. One enzyme monomer is shown as cartoon, with gradient coloring from blue (N-terminus) to red (Cterminus). Catalytic residues are shown as sticks, whereas the substrate-binding pocket, determined using the cavity tool options in PyMol (25), is displayed as grey mesh. (D) Cartoon representation of the DfsA trimer. Each monomer is shown with a different color. The fatty acid molecule identified near the DfsA catalytic site in each monomer is shown with yellow spheres. Arrows indicate the substrate access points to the catalytic site of each DfsA monomer. Figure 2. Thioesterase enzymatic activity of DfsA was assayed using dodecanoyl-ACP as substrate monitoring the release of holo-ACP and the release of dodecanoic acid. (A) Conformationally sensitive gel electrophoresis on 20% polyacrylamide gel containing 0.8 M urea, showing the conversion of C-12-ACP into holo-ACP in the presence of DfsA. (B) ESI-MS analysis of the ethylacetate extracted reaction mixture of the blank control without DfsA. (C) ESI-MS analysis of 17 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the ethylacetate extracted reaction mixture shows an extra peak, absent in the blank, with m/z of 202.75. The fragmentation pattern (inset) shows that this peak corresponds to dodecanoic acid. (D) Steady state kinetic analysis of DfsA using dodecanoic-ACP as substrate, and performed by DCPIP based assay. Figure 3. DALI analysis of DfsA structural homology. The superposition shows cartoon representation of the DfsA monomer (orange) superimposed with its closest structural homologs (left). The enzyme catalytic site is indicated with an asterisk. The hinge of the C-terminal domain swap site observed in various members of the crotonase family is shown with an arrow. The table (right) provides a summary of the DALI analysis. Figure 4. Identification and characterization of a lipid molecule in DfsA crystal structure. (A) Overview of the unbiased experimental Fo-Fc electron density map (contour level 3.0 σ) showing the presence of an elongated molecule trapped in the substrate binding cavity of DfsA, surrounded by hydrophobic amino acid residues. (B) Identification of the lipid molecule bound to the DfsA catalytic site. Total ion current spectra from gas chromatography analysis. In the inset the mass fragmentation profile of molecule eluted at retention time 14.45 min identified the compound as methyl-dodecanoate. (C) Schematic of the amino acid environment surrounding the lauric acid (DAO) molecule in DfsA catalytic site. Image created using LigPlot+ (44). (D) Effect of dodecanoic and dodecenoic acids against DfsA enzyme activity, determined by conformationally sensitive gel electrophoresis. Lane 1: no addition; lane 2: 0.1 mM dodecenoic acid; lane 3: 0.5 mM dodecenoic acid; lane 4: 0.1 mM dodecanoic acid; lane 5: 0.5 mM dodecanoic acid; the last two lanes were loaded with C12-ACP and holo-ACP respectively. (E) IC50 determination of dodecenoic acid (white symbols) and dodecanoic acid (black symbols) towards DfsA activity. Figure 5. Comparison of lipid substrate conformations in DfsA and structurally-related crotonase enzymes (A) Comparison of the conformation of the fatty acid chain of lauric acid as identified in DfsA catalytic site (green ball-and-stick) with ligand molecules identified in DfsA structurally18 ACS Paragon Plus Environment

Page 18 of 34

Page 19 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

related crotonase enzymes (3(S)-hydroxy-hexanoyl-CoA, 2-trans-hexenoyl-CoA and 3(S)-hydroxydecanoyl-CoA in rat RPMFE1 (PDB ID 3ZWA, 3ZWB and 3ZWC, light blue, dark blue and cyan, respectively (45)), acetoacetyl-CoA, octanoyl-CoA and hexadienoyl-CoA in rat Echs1 (PDB ID 1DUB, 2DUB and 1MJ3, magenta, dark red and yellow, respectively (39, 46)), octanoyl-CoA in human ECI1 (PDB ID 1SG4, dark green (47)). As reference for structural orientation, the catalytic glutamate residues of DfsA are shown as sticks, and the DfsA substrate cavity, determined as in figure 1c, is shown as grey mesh grid. (B) Superposition of the catalytic sites of DfsA (orange) with RpfF (blue), showing the side-chain conformations of RpfF residues Phe91 and Met170 (Phe88 and Met168 in DfsA, respectively) obstructing the hydrophobic cavity hosting the lauric acid molecule in DfsA (green sticks). (C) Theoretical model of the BDSF precursor cis-2-dodecenoyl-ACP (green sticks) inside the DfsA catalytic cavity (semi-transparent surface, with catalytic glutamates and oxyanion hole residues shown as orange sticks) compared with the conformation of the lauric acid product observed in the DfsA crystal structure (black).

19 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table 1. Crystallographic data collection and refinement statistics for DfsA structure. Data quality statistics1 Beamline

ESRF ID23-EH1

Wavelength (Å)

0.873

Space group

P 31 2 1

Cell dimensions a, b, c (Å)

127.9 127.9 128.8

α, β, γ (°)

90.0 90.0 120.0

Resolution range (Å)

55.69–1.90 (1.93–1.90)

Reflections

496484 (23012)

Unique reflections

95772 (4702)

Completeness (%)

99.8 (99.8)

Multiplicity

5.2 (4.9)

Rsym2

0.15 (0.77)



6.7 (1.8)

CC(1/2)

0.99 (0.35)

Refinement statistics Rwork (%)

0.166

Rfree (%)

0.198

Number of amino acid residues

6387

Number of solvent atoms

487

Number of ligand atoms

124

RMS bonds (Å)

0.016

RMS angles

1.46

Average B factors Macromolecules (Å2)

31

Ligands (Å2)

58

Solvent (Å2)

44 20 ACS Paragon Plus Environment

Page 20 of 34

Page 21 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Ramachandran favored (%)

98.0

Ramachandran outliers (%)

0.0

1 Values in parentheses are for reflections in the highest resolution shell. 2 Rsym = Σ | I – | / Σ I, where I is the observed intensity for a reflection and is the average intensity obtained from multiple observations of symmetry-related reflections.

21 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 34

Table 2. Kinetic analysis of DfsA towards acyl-ACP substrates Km (µM)

kcat/Km (min-1µM-1)

n.a.

n.a.

n.a.

octanoyl-ACP (C8)

1.53 ± 0.16

73.0 ± 8.2

0.021 ± 0.004

lauroyl-ACP (C12)

1.56 ± 0.15

19.0 ± 1.2

0.082 ± 0.080

n.a.

n.a.

n.a.

Substrate butyryl-ACP (C4)

myristoyl-ACP (C14)

kcat (min-1) a

n.a.= no activity detected

22 ACS Paragon Plus Environment

Page 23 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Corresponding Authors * Laurent R. Chiarelli - [email protected]; Federico Forneris - [email protected]

Author Contributions G.R. and S.B. designed the research project. F.S. and V.C.S. recombinantly expressed, purified and crystallized BCAM0581. F.S. and L.R.C. designed and performed enzymatic assays. M.F. collected and analyzed mass spectrometry data. F.F. collected diffraction data, solved and refined DfsA structure. F.S. and F.F. analyzed the structural data. F.F. prepared the figures. G.R., S.B., L.R.C. and F.F. supervised the research work. All authors contributed to data analysis and interpretation, and participated to the writing of the manuscript.



These authors contributed equally.

*

Corresponding authors.

ACKNOWLEDGMENT We thank Dr. J. Rubén Gómez Castellanos for critical reading of the manuscript.

FUNDING INFORMATION This work was supported by the Italian Cystic Fibrosis Research Foundation (FFC project #19/2015 to G.R.). The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under BioStruct-X (grant agreements n. 7551 and 10205). F.F. is supported by a Career Development Award from the Armenise-Harvard Foundation and by the "Rita Levi-Montalcini" award from the Italian Ministry of University and Education (MIUR).

23 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABBREVIATIONS ACP, acyl carrier protein; BDSF, Burkholderia diffusible signal factor; C-12ACP, dodecanoyl-ACP; CF, cystic fibrosis; DFS diffusible signal factor; DfsA, diffusible factor synthase A; DTT, 1,4-dithio-DL-threitol; HEPES, 42-Hydroxyethyl piperazine-1-ethanesulfonic acid; IPTG, isopropyl β-D-1-thiogalactopyranoside; MES, 2- Nmorpholino ethanesulfonic acid; QS, quorum sensing.

24 ACS Paragon Plus Environment

Page 24 of 34

Page 25 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

REFERENCES 1.

Valvano, M. A. (2015) Intracellular survival of Burkholderia cepacia complex in phagocytic cells, Can J Microbiol 61, 607-615.

2.

Rasko, D. A., and Sperandio, V. (2010) Anti-virulence strategies to combat bacteriamediated disease, Nat Rev Drug Discov 9, 117-128.

3.

Goo, E., An, J. H., Kang, Y., and Hwang, I. (2015) Control of bacterial metabolism by quorum sensing, Trends Microbiol 23, 567-576.

4.

Waters, C. M., and Bassler, B. L. (2005) Quorum sensing: cell-to-cell communication in bacteria, Annu Rev Cell Dev Biol 21, 319-346.

5.

Galloway, W. R., Hodgkinson, J. T., Bowden, S. D., Welch, M., and Spring, D. R. (2011) Quorum sensing in Gram-negative bacteria: small-molecule modulation of AHL and AI-2 quorum sensing pathways, Chem Rev 111, 28-67.

6.

Udine, C., Brackman, G., Bazzini, S., Buroni, S., Van Acker, H., Pasca, M. R., Riccardi, G., and Coenye, T. (2013) Phenotypic and genotypic characterisation of Burkholderia cenocepacia J2315 mutants affected in homoserine lactone and diffusible signal factorbased quorum sensing systems suggests interplay between both types of systems, PLoS One 8, e55112.

7.

Brackman, G., and Coenye, T. (2015) Quorum sensing inhibitors as anti-biofilm agents, Curr Pharm Des 21, 5-11.

8.

Brackman, G., Cos, P., Maes, L., Nelis, H. J., and Coenye, T. (2011) Quorum sensing inhibitors increase the susceptibility of bacterial biofilms to antibiotics in vitro and in vivo, Antimicrob Agents Chemother 55, 2655-2661.

9.

Suppiger, A., Schmid, N., Aguilar, C., Pessi, G., and Eberl, L. (2013) Two quorum sensing systems control biofilm formation and virulence in members of the Burkholderia cepacia complex, Virulence 4, 400-409.

10.

Boon, C., Deng, Y., Wang, L. H., He, Y., Xu, J. L., Fan, Y., Pan, S. Q., and Zhang, L. H. (2008) A novel DSF-like signal from Burkholderia cenocepacia interferes with Candida albicans morphological transition, ISME J 2, 27-36.

11.

Ryan, R. P., An, S. Q., Allan, J. H., McCarthy, Y., and Dow, J. M. (2015) The DSF Family of Cell-Cell Signals: An Expanding Class of Bacterial Virulence Regulators, PLoS Pathog 11, e1004986.

12.

Bi, H., Christensen, Q. H., Feng, Y., Wang, H., and Cronan, J. E. (2012) The Burkholderia cenocepacia BDSF quorum sensing fatty acid is synthesized by a bifunctional crotonase homologue having both dehydratase and thioesterase activities, Mol Microbiol 83, 840-855. 25 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

13.

Cheng, Z., He, Y. W., Lim, S. C., Qamra, R., Walsh, M. A., Zhang, L. H., and Song, H. (2010) Structural basis of the sensor-synthase interaction in autoinduction of the quorum sensing signal DSF biosynthesis, Structure 18, 1199-1209.

14.

Bi, H., Yu, Y., Dong, H., Wang, H., and Cronan, J. E. (2014) Xanthomonas campestris RpfB is a fatty Acyl-CoA ligase required to counteract the thioesterase activity of the RpfF diffusible signal factor (DSF) synthase, Mol Microbiol 93, 262-275.

15.

Almeida, R. P., Killiny, N., Newman, K. L., Chatterjee, S., Ionescu, M., and Lindow, S. E. (2012) Contribution of rpfB to cell-to-cell signal synthesis, virulence, and vector transmission of Xylella fastidiosa, Mol Plant Microbe Interact 25, 453-462.

16.

Leslie, A. G. W., and Powell, H. R. (2007) Processing diffraction data with MOSFLM, Vol. 245.

17.

Evans, P. R. (2011) An introduction to data reduction: space-group determination, scaling and intensity statistics, Acta Crystallogr D Biol Crystallogr 67, 282-292.

18.

Karplus, P. A., and Diederichs, K. (2012) Linking crystallographic model and data quality, Science 336, 1030-1033.

19.

McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and Read, R. J. (2007) Phaser crystallographic software, J Appl Crystallogr 40, 658-674.

20.

Langer, G., Cohen, S. X., Lamzin, V. S., and Perrakis, A. (2008) Automated macromolecular model building for X-ray crystallography using ARP/wARP version 7, Nat Protoc 3, 1171-1179.

21.

Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P. H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution, Acta Crystallogr D Biol Crystallogr 66, 213-221.

22.

Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K. (2010) Features and development of Coot, Acta Crystallographica Section D 66, 486-501.

23.

Chen, V. B., Arendall, W. B., 3rd, Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G. J., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2010) MolProbity: all-atom structure validation for macromolecular crystallography, Acta Crystallogr D Biol Crystallogr 66, 12-21.

24.

Read, R. J., Adams, P. D., Arendall, W. B., 3rd, Brunger, A. T., Emsley, P., Joosten, R. P., Kleywegt, G. J., Krissinel, E. B., Lutteke, T., Otwinowski, Z., Perrakis, A., Richardson, J. S., Sheffler, W. H., Smith, J. L., Tickle, I. J., Vriend, G., and Zwart, P. H. (2011) A new 26 ACS Paragon Plus Environment

Page 26 of 34

Page 27 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

generation of crystallographic validation tools for the protein data bank, Structure 19, 13951412. 25.

Schrodinger, LLC. (2010) The PyMOL Molecular Graphics System, Version 1.3r1.

26.

Montersino, S., Orru, R., Barendregt, A., Westphal, A. H., van Duijn, E., Mattevi, A., and van Berkel, W. J. (2013) Crystal structure of 3-hydroxybenzoate 6-hydroxylase uncovers lipid-assisted flavoprotein strategy for regioselective aromatic hydroxylation, J Biol Chem 288, 26235-26245.

27.

Quadri, L. E., Weinreb, P. H., Lei, M., Nakano, M. M., Zuber, P., and Walsh, C. T. (1998) Characterization of Sfp, a Bacillus subtilis phosphopantetheinyl transferase for peptidyl carrier protein domains in peptide synthetases, Biochemistry 37, 1585-1595.

28.

Cronan, J. E., and Thomas, J. (2009) Bacterial fatty acid synthesis and its relationships with polyketide synthetic pathways, Methods Enzymol 459, 395-433.

29.

Post-Beittenmiller, D., Jaworski, J. G., and Ohlrogge, J. B. (1991) In vivo pools of free and acylated acyl carrier proteins in spinach. Evidence for sites of regulation of fatty acid biosynthesis, J Biol Chem 266, 1858-1865.

30.

Christensen, Q. H., Grove, T. L., Booker, S. J., and Greenberg, E. P. (2013) A highthroughput screen for quorum-sensing inhibitors that target acyl-homoserine lactone synthases, Proc Natl Acad Sci U S A 110, 13815-13820.

31.

Bevington, P. H. (1969) Data Reduction and Error Analysis for the Physical Sciences, , McGraw-Hill, New York.

32.

Montebello, A. N., Brecht, R. M., Turner, R. D., Ghali, M., Pu, X., and Nagarajan, R. (2014) Acyl-ACP substrate recognition in Burkholderia mallei BmaI1 acyl-homoserine lactone synthase, Biochemistry 53, 6231-6242.

33.

Hamed, R. B., Batchelar, E. T., Clifton, I. J., and Schofield, C. J. (2008) Mechanisms and structures of crotonase superfamily enzymes--how nature controls enolate and oxyanion reactivity, Cell Mol Life Sci 65, 2507-2527.

34.

Krissinel, E., and Henrick, K. (2007) Inference of macromolecular assemblies from crystalline state, J Mol Biol 372, 774-797.

35.

Holm, L., and Rosenstrom, P. (2010) Dali server: conservation mapping in 3D, Nucleic Acids Res 38, W545-549.

36.

Zhou, L., Yu, Y., Chen, X., Diab, A. A., Ruan, L., He, J., Wang, H., and He, Y. W. (2015) The Multiple DSF-family QS Signals are Synthesized from Carbohydrate and Branchedchain Amino Acids via the FAS Elongation Cycle, Sci Rep 5, 13294.

27 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

37.

Jenke, D., Castner, J., Egert, T., Feinberg, T., Hendricker, A., Houston, C., Hunt, D. G., Lynch, M., Shaw, A., Nicholas, K., Norwood, D. L., Paskiet, D., Ruberto, M., Smith, E. J., and Holcomb, F. (2013) Extractables characterization for five materials of construction representative of packaging systems used for parenteral and ophthalmic drug products, PDA J Pharm Sci Technol 67, 448-511.

38.

Reid, A. M., Brougham, C. A., Fogarty, A. M., and Roche, J. J. (2007) An investigation into possible sources of phthalate contamination in the environmental analytical laboratory, International Journal of Environmental Analytical Chemistry 87, 125-133.

39.

Engel, C. K., Kiema, T. R., Hiltunen, J. K., and Wierenga, R. K. (1998) The Crystal Structure of Enoyl-CoA Hydratase Complexed with Octanoyl-CoA Reveals the Structural Adaptations Required for Binding of a Long Chain Fatty Acid-CoA Molecule, J Mol Biol 275, 859-847.

40.

Marr, A. G., and Ingraham, J. L. (1962) Effect of Temperature on the Composition of Fatty Acids in Escherichia Coli, J Bacteriol 84, 1260-1267.

41.

Lyne, P. D. (2002) Structure-based virtual screening: an overview, Drug Discov Today 7, 1047-1055.

42.

Keyser, P., Elofsson, M., Rosell, S., and Wolf-Watz, H. (2008) Virulence blockers as alternatives to antibiotics: type III secretion inhibitors against Gram-negative bacteria, J Intern Med 264, 17-29.

43.

Robert, X., and Gouet, P. (2014) Deciphering key features in protein structures with the new ENDscript server, Nucleic Acids Res 42, W320-324.

44.

Laskowski, R. A., and Swindells, M. B. (2011) LigPlot+: multiple ligand-protein interaction diagrams for drug discovery, J Chem Inf Model 51, 2778-2786.

45.

Kasaragod, P., Schmitz, W., Hiltunen, J. K., and Wierenga, R. K. (2013) The isomerase and hydratase reaction mechanism of the crotonase active site of the multifunctional enzyme (type-1), as deduced from structures of complexes with 3S-hydroxy-acyl-CoA, FEBS J 280, 3160-3175.

46.

Bell, A. F., Feng, Y., Hofstein, H. A., Parikh, S., Wu, J., Rudolph, M. J., Kisker, C., Whitty, A., and Tonge, P. J. (2002) Stereoselectivity of enoyl-CoA hydratase results from preferential activation of one of two bound substrate conformers, Chem Biol 9, 1247-1255.

47.

Partanen, S. T., Novikov, D. K., Popov, A. N., Mursula, A. M., Hiltunen, J. K., and Wierenga, R. K. (2004) The 1.3 A crystal structure of human mitochondrial Delta3-Delta2enoyl-CoA isomerase shows a novel mode of binding for the fatty acyl group, J Mol Biol 342, 1197-1208. 28 ACS Paragon Plus Environment

Page 28 of 34

Page 29 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 1 406x297mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 257x157mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 30 of 34

Page 31 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 3 254x87mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4 249x224mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 32 of 34

Page 33 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 5 227x159mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents Graphics 330x317mm (72 x 72 DPI)

ACS Paragon Plus Environment

Page 34 of 34