Identification of new quorum sensing autoinducer binding partners in Pseudomonas aeruginosa using photoaffinity probes

Design, synthesis and application of PQS and HHQ probes for investigating quinolone quorum sensing pathways using photoaffinity labeling.

Optimisation of labelling conditions. The labelling protocol was investigated as follows.
The ligand binding domain of PqsR was overexpressed as a maltose binding protein fusion (PqsR LBD -MBP) using an E. coli strain with the plasmid pMAL-c2X containing a pqsR ligand binding domain (LBD) insert. 1 The cells were harvested, lysed in PBS containing a protease inhibitor cocktail tablet (Roche) by sonication and clarified by centrifugation. The cell pellet was discarded and the supernatant used as the cell-lysate samples. The Bradford Assay was then used to determine the protein concentration and the sample was diluted with PBS containing a protease inhibitor cocktail tablet (Roche) to a final concentration of 1 mg/mL.

SDS PAGE analysis of pull-downs
Late exponential phase pull-downs. Cells were harvested at late exponential phase by centrifugation (Time point A, Supplementary Fig. 3). Late exponential phase was chosen initially as this lies just before the point at which the QS genes are "switched" on. A second reason for this is that Cao et al. reported that PqsR is at maximal levels during late exponential phase growth and is then cleaved and secreted into the supernatant in the stationary phase. 2 The harvested cells were lysed and centrifuged at low speed, keeping both the pellet and the supernatant.
PQS probe 10, HHQ probe 11 and control probe 12 were used to prepare samples for MS identification. Cell-lysates were incubated with the probes at 10 μM for 30 min before irradiation with UV light (302 nm, 15 min). Copper catalysed click chemistry was used it tag a biotin alkyne to the labelled proteins. The protein samples were then incubated with streptavidin-coated magnetic beads. The beads were then washed stringently and the captured proteins eluted by boiling in gel loading buffer containing SDS and free biotin. The proteins were then separated by SDS-PAGE and silver-stained ( Supplementary Fig. 4).
In addition competitive binding experiments were also performed experiment. In this case, 100 μM of PQS was added to a second sample with 10 μM of PQS probe 10 and 100 μM of HHQ was added to a second experiments using 10 μM of HHQ probe 11. The pull-downs were performed identically to those described above.
Supplementary Figure 4. Silver stain of proteins isolated from late exponential phase labelled using probes 10, 11 and 12. Bands marked with an * were analysed by tryptic digest followed by LC-MS. + indicates the competitive control lane. WT = wild type cell-lysate; M = pqsR mutant cell-lysate.
For analysis, the proteins identified in the control lane of the gel at the same migration distance and proteins found in on-bead digest negative control experiments (discussed in further detail later) were removed. Ribosomal proteins were also disregarded, as they are known to be highly abundant and stick to the beads used.
The band at around 40 kDa had evidence of FtsZ (Mascot score of 80) and PqsC (Mascot score 39), a protein required for the synthesis of HHQ and PQS. Unfortunately, the LCMS trace for this sample had low levels of material. Despite this, PqsC had the highest scoring peptide (Supplementary Table 9).
The second band analysed at approximately 17 kDa was found to contain 27 potential proteins, many of which are presumably fragments of larger proteins. For this reason only the proteins that were also identified in the later on-bead digest were considered. The only protein from this list with convincing peptide scores was peptidoglycan-associated lipoprotein pal (also referred to as OprL in the literature) with a Mascot score of 85 and the correct mass for the migration by SDS-PAGE.
The evidence for the pull-down of OprL using the PQS probe 10 is strong, with three highscoring peptides and the correct migration distance for the protein's molecular weight. The band is also intense when compared with the corresponding competitive binding assay. The band is not seen for probe 11 suggesting the protein interacts with PQS and not HHQ, again supporting the validity of this result. As the band can be seen with probe 12 it is also possible that it is protein that is under QS control which binds lipophilic compounds. Whilst further biochemical studies are required to confirm that PQS does in fact bind OprL, the pull-down of this protein is promising; one possible hypothesis is that the interaction between PQS and OprL is linked to membrane vesicle (MV) formation.
The role of PQS in membrane vesicle formation has been well documented. It has previously been shown that PQS stimulates membrane vesicle formation and it has been proposed that this is through direct association with the lipopolysaccharide in the outer membrane. 3 Figure 5. Effect of introducing ligation handles groups on quinolone activity. A) PqsR stimulation determined using the method reported by Cugini et al. 6 These compounds were tested. B) Pyoverdine production determined using the method reported by Welsh et al. 7 Dark bars show PQS analogues which were tested at 60 nM and light bars show HHQ analogues which were tested at 1 μM. Error bars represent the standard deviation of three independent repeats.    were added to 5 mL cultures and the cultures grown for a further 2.5 h before measuring the OD 600 and β galactosidase activity quantified.
Pyoverdine assay. Pyoverdine was quantified as previously described. 12 Overnight cultures of PAO1 were diluted 1:100 into LB medium and quinolone analogues in DMSO (25 μL

Pull-down of biotin tagged PqsR.
Biotin tagged proteins were precipitated using the CHCl 3 /MeOH procedure. 13 Pull-downs were then performed as outlined previously. 14 Briefly, SDS (10 μL, 10% w/v in H 2 O) was added to the biotin labelled protein pellets the sample heated to 100 °C for 5 min. This was diluted with water (4 x 50 μL) with vortexing after each addition. The proteins were precipitated using the CHCl 3 /MeOH procedure 13  Cells were allowed to thaw on ice before being resuspended in ice cold PBS containing a cOmplete, mini, EDTA free protease inhibitor cocktail tablet and lysed by ultrasonication (10 x 30 s) taking care to keep the samples below 4 °C. The protein samples were then clarified using low speed centrifugation (7400 xg, 10 min, 4 °C) and the protein concentration determined using the Bradford protein assay before being snap frozen using liquid nitrogen and stored at -80 °C until use.
Cell-lysate photocrosslinking and labelling with biotin. Cell-lysates were adjusted to a protein concentration of 5 mg/mL using PBS containing a protease inhibitor cocktail tablet (Roche). The probes were then added (10 μM final concentration from 1 mM stock solution in DMSO) to 0.2 mL of protein sample, which were then incubated at room temperature for 30 min with gentle shaking. The samples were then irradiated with UV light (λ = 302 nm) for 15 min. Click chemistry with biotin alkyne was used to install biotin (as described earlier) and the proteins were precipitated using a Chloroform/Methanol precipitation. 13

Pull-down experiments for protein identification
Pull-down of biotin tagged proteins. Biotin tagged proteins were precipitated using the CHCl 3 /MeOH procedure. 13 Pull-downs were then performed as outlined previously. 14   Fragmented precursor ions that were measured twice within 10 s were dynamically excluded for 120 s and ions with z < 2 or unassigned were not analysed. For peptide identification, mascot generic files (mgf) were generated from the .raw data files and matched with the
The reaction was stirred at room temperature for 1 h, then heated to 50 °C for a further 2 h and cooled to room temperature to give a ~0.5 M solution of the Grignard reagent in THF.
The Grignard solution was then added to a solution of 2-chloro-N-methoxy-Nmethylacetamide (500 mg, 3.6 mmol, 0.5 eq) in THF (25 mL) and the mixture was stirred overnight at room temperature. The reaction was diluted two-fold with toluene, cooled to 0 °C and quenched with an equal volume of 0.1 M HCl. The organic phase was then washed with brine, dried over MgSO 4 , filtered and the solvent removed under reduced pressure to give the resulting crude α-chloroketone S1 as a pale yellow oil in 95% yield which was then used without further purification. 1

Synthesis of HHQ analogue 6
Methyl 3-oxodec-9-ynoate S15 The title compound was prepared based on a published procedure by Oikawa et al. 24 Oct-7ynoic acid S14 has was prepared as reported by Garner and Janda. 25 Oxalyl chloride (1 mL, 2 M in CH 2 Cl 2 , 2 mmol, 2.0 eq) was added dropwise to a solution of acid S14 (140 mg, 1.0 mmol, 1.0 eq) and DMF (1 drop) in CH 2 Cl 2 at 0 °C and the reaction was stirred for 1 h. Then the reaction was allowed to warm to room temperature and stirred for a further 2 h. After this the solvent was removed under reduced pressure and the acid chloride was dissolved in CH -2 Cl 2 (100 µL). A solution of Meldrum's acid (131 mg, 0.91 mmol, 1.0 eq) in dry CH 2 Cl 2 (2 mL) was cooled to 0 °C and pyridine (150 µL) was added drop-wise over 20 min. The acid chloride in CH 2 Cl 2 was then added and the mixture was stirred at 0 °C for a further 2 h. The reaction mixture was allowed to warm to room temperature, diluted with CH 2 Cl 2 (x 2) and poured into an equal volume of cold HCl (2 N). The organic phase was washed with saturated NaCl, dried over MgSO 4 , filtered and evaporated to dryness. The resulting orange/brown oil was heated to reflux in anhydrous MeOH (5 mL) for 5 h, evaporated to dryness and purified by (stepwise gradient of 5%-10% Et 2 O in 40/60 PE) to give S15 (78 mg, 0.4 mmol) as a colourless oil in 40% yield.
The mixture was then heated under microwave irradiation at 200 °C for 30 min. After cooling to room temperature the reaction was quenched by pouring into ice/water. The resulting mixture was stirred for 5 min before being allowed to settle for 30 min and the precipitate was collected and dried under vacuum to give S17 (421 mg, 1.58 mmol) as a beige amorphous solid in 72% yield.

6-Amino-2-heptyl-3-hydroxyquinolin-4(1H)-one S18
Intermediate S16 (400 mg, 1.3 mmol, 1.0 eq) was dissolved in HBr (33% aqueous solution, 4 mL) and heated to 120 °C for 30 min with stirring. The reaction was cooled to room temperature and a crystalline solid formed. The solid was collected by filtration and partitioned between EtOAc/NaHCO 3 . The organic layer was dried over Na 2 SO 4 , filtered and the solvent removed under reduced pressure to give S18 (308 mg,

2-(5-Bromopentyl)-4-chloroquinoline S26
A solution of compound S25 (50 mg, 0.17 mmol, 1.0 eq) in phosphorus oxychloride (1 mL) was heated to reflux for 1 h. The excess phosphorus oxychloride was removed under reduced pressure and the resulting residue was dissolved in CH 2 Cl 2 (5 mL) and washed with saturated NaHCO 3 (3 x 5 mL). The organic layer was the dried over MgSO 4 , filtered and evaporated to dryness to give the title compound S26 (37 mg, 0.13 mmol) as a colourless oil. The mixture was used in the next step without further purification.
amorphous solid (100% yield assumed, contaminated with free acid chlorgide) which was used without further purification.  PqsD was judged pure by SDS-PAGE analysis. The protein was separated into aliquots, flash-frozen in liquid nitrogen, and stored at -80⁰C.
Fluorimetry analysis. All fluorescence spectra were generated on a Perkin Elmer LS 55 luminescence spectrometer. PqsD protein (10 µM final concentration) was added to 50 mM Tris-HCl, pH 8.0, 100 mM NaCl, and 10% (v/v) glycerol to a final volume of 1 ml in a quartz, optical cuvette with a 1 cm path length. Spectra illustrated (Fig. 1 below) represent an average of three traces, with no significant variation between each of the three replicates. The excitation wavelength used was 292 nm. At a rate of 10 nm/min, emission spectra were recorded between 300 and 450 nm with the excitation and emission slit widths set at 2.5 nm.