The role of 2,4-dihydroxyquinoline (DHQ) in Pseudomonas aeruginosa pathogenicity

Bacterial strains, plasmids, and media

P. aeruginosa wild-type strain PAO1 and its derived mutants, and E. coli strains were grown in Luria–Bertani (LB) medium at 37 °C in a shaker incubator. Cystic fibrosis mimic media was prepared according to Palmer et al. (2005). Hypoxic conditions were generated by flushing media with N₂ for 10 min and incubating cultures in a screw-cap vial sealed with a Tephlon^© insert and nitrogen-filled head space. Antibiotic concentrations for bacterial selection were: 50 µg/ml and 200 µg/ml carbenicillin (CBC) for E. coli and P. aeruginosa, respectively, 30 µg/ml kanamycin (Kan), 34 µg/ml chloramphenicol (Cam), 30 µg/ml gentamicin (Gm). Culture density was assessed by measuring absorbance at 600 nm. DHQ (2,4-dihydroxyquinoline) and farnesol were purchased from Sigma-Aldrich and PQS (2-heptyl-3-hydroxy-4-quinolone) and HHQ (2-heptly-4-quinolone) were purchased from Qingdao Vochem Co.

Generation of mutants

P. aeruginosa mutant strains ΔpqsAB, ΔpqsB, and ΔpqsC were generated by homologous recombination using a protocol described previously (Choi & Schweizer, 2005). The mutant alleles were constructed by overlapping PCR to contain a gentamicin-resistance cassette flanked by 5′ and 3′ fragments of the gene to be deleted. The mutant fragments were inserted into pEX18ApGW, a suicide vector, to produce gene knockout plasmids. Each knockout plasmid was transformed into E. coli strain SM10 and conjugally transferred from SM10 to PAO1. The resultant integrants were selected on PIA medium containing 30 µg/ml gentamicin (Gm30). To resolve merodiploids, Gm-resistant colonies were streaked for single colonies on LB+Gm30 plates containing 5% sucrose. The unmarked deletion mutants were generated by Flp-mediated marker excision utilizing pFLP2. Potential mutants were screened by PCR using corresponding flanking primers and were confirmed by sequencing.

Generation of protein-expressing plasmids and protein purification

Coding sequences of PqsR and PqsR-C⁸⁷, a truncated version of PqsR containing the ligand-binding domain starting from Cys87, were amplified by PCR and cloned into a modified pET-28a vector to express recombinant PqsR and PqsR-C⁸⁷ as His-tagged SUMO fusion proteins for increased solubility. E. coli strain BL21(DE3) or Rosetta containing the expression plasmids was grown in LB with antibiotics (Kan for BL21(DE3), Kan and Cam for Rosetta) to OD₆₀₀ of 0.6 and heat-shocked at 45 °C for 45 min. Overexpression of recombinant protein was induced by 0.1 mM IPTG at 16 °C for 16 h. The cells were harvested by centrifugation, resuspended in 20 mM Tris–HCl (pH 8.0) with 500 mM NaCl, and lysed by sonication. His-tagged protein in the cell-free extract was purified by nickel affinity chromatography. Fractions containing pure protein were pooled, concentrated, and stored at −80 °C. Protein concentration was determined using the Bradford method with γ-globulin as the protein standard.

Caenorhabditis elegans survival and imaging

Bristol N2 strain of C. elegans was maintained and synchronized on nematode growth media (NGM) containing E. coli OP50 according to the WormBook (Girard et al., 2007). Synchronized L4 nematodes (around 30 per plate) were transferred to 60 mm petri dishes containing a lawn of the tested P. aeruginosa strain. The antimetabolite 5-fluorouracil (25 µM) was added to the NGM agar to inhibit worm reproduction during the course of the experiment. The plates were incubated at 25 °C and worms were monitored daily for survival. Worms were determined dead when no movement was observed following stimulation with a platinum wire. To assess in vivo bacterial colonization, P. aeruginosa strains PAO1, ΔpqsAB, and ΔpqsB carrying the GFP-expressing plasmid pSMC2 (Bloemberg et al., 1997) were used for infection. At the indicated time-points, nematodes were transferred onto fresh agar to remove free-living bacteria, placed into 10 µl of water on a glass slide, and viewed using a Nikon TE2000-S Epifluorescent microscope with a CRI-Nuance imaging system. Images were color-enhanced and analyzed with the ImageJ software.

Pyocyanin quantification

Pyocyanin concentration was measured using a modified protocol from previous work (Das & Manefield, 2012). Briefly, 400 µl of cell-free supernatant from an overnight culture was mixed vigorously with 240 µl chloroform. After organic-phase separation, 200 µl of the organic phase was transferred to a new tube and mixed with 60 µl 0.2 N HCl. Pyocyanin was measured spectrophotometrically at 520 nm using a NanoDrop ND-1000 spectrophotometer.

Quantitative real-time PCR (qRT-PCR)

P. aeruginosa strains were grown in LB until early stationary phase. Total RNA was extracted, treated with DNaseI, and precipitated overnight at −20 °C. cDNA was generated using the Bioline Tetro cDNA synthesis kit. Real-time quantitative PCR was performed on a Bio-Rad MyiQ single-color real-time PCR detection system. The primers used for each gene were as follows: for pqsA, forward primer (TTCTGTTCCGCCTCGATTTC) and reverse primer (AGTCGTTCAACGCCAGCAC); for rpoD, forward primer (GGGCGAAGAAGGAAATGGTC) and reverse primer (CAGGTGGCGTAGGTGGAGAAC). PCR reactions were performed in a 96-well plate with a reaction mixture containing 10 µl SensiFAST SYBR Green mix (Bioline Reagents, London, UK), 200 ng cDNA, and 200 nM each of forward and reverse primers in a total volume of 20 µl. All reactions were performed in triplicate. The thermal cycling conditions were set at 95 °C for 3 min, followed by 40 cycles of 2-step amplification (10 s at 95 °C and 45 s at 57 °C). Data were analyzed with MyiQ software. The abundance of pqsA gene was normalized against the house-keeping rpoD gene using the ΔΔCt method.

pqsA’-LacZ fusion reporter assay

LacZ reporter assay was performed in both E. coli and in P. aeruginosa strains as previously described (Cugini et al., 2007). Briefly, overnight culture of E. coli containing the pEAL08-2 reporter construct was diluted to OD_600nm of 0.05, grown in the presence of quinolones for 2 h, and harvested by centrifugation. P. aeruginosa strains with pEAL08-2 were grown overnight before harvest. Cells were lysed and the LacZ activity was measured using a β-galactosidase enzyme assay kit (Promega, Madison, WI, USA). The LacZ product was measured at 420 nm on a plate reader (BioTek Synergy HT; BioTek, Winooski, VT, USA).

Electrophoretic mobility shift assay (EMSA)

EMSA assays were performed using cell lysates from PAO1, pqs mutants, and Rosetta strains induced for His-SUMO or His-SUMO-PqsR production as described (Cugini et al., 2007). For each sample, 10 µg of cell-lysate was incubated with 0.15 fmol of biotinylated 248-bp pqsA promoter DNA (pqsA’), which was generated by PCR using biotinylated forward primer (TTCTTGCTTGGTTGCCG) and reverse primer (GACAGAACGTTCCCTCTT). Unlabeled pqsA’ probe was generated using the same primers as the biotinylated-pqsA’ without the biotin modification. The reaction mixture contained 10 mM Tris–HCl (pH 8.0), 1 mM EDTA, 50 mM NaCl, 1 mM DTT, 1 µg/µl Poly(dI-dC) and was incubated at room temperature (24 °C) for 20 min. Samples were separated on a 5% polyacrylamide gel in 0.5×Tris Borate EDTA (TBE) run at 100 V for 50 min and transferred onto a positively-charged nylon membrane at 40 V for 1 h. Biotinylated DNA on the nylon membrane was probed by streptavidin-HRP conjugate, detected by a chemiluminescent substrate (Pierce LightShift Chemiluminescent EMSA kit; Thermo Fisher Scientific, Waltham, MA, USA), and visualized by exposing to either an X-ray film or GE ImageQuant RT-ECL.

Saturation transfer difference (STD) NMR

STD NMR experiments were prepared with 1 µM SUMO-PqsR-C⁸⁷ with increasing concentrations of DHQ (10 nm to 100 µM). Data were collected at 298 K on a Bruker Avance III 600 MHz NMR spectrometer equipped with a 5 mm cryogenically-cooled QC-Inverse probe and using a standard STD pulse sequence with 30 ms 8.4 kHz spin lock to minimize background protein resonances (Kemper et al., 2010). Solvent suppression was achieved using the excitation sculpting scheme. Saturation of the protein signals was performed using a train of 10, 20, or 59 selective 56 dB Gaussian pulses of 50 ms duration with total saturation times of 0.25 s to 5.0 s. The on-resonance frequency was set up at −0.5 ppm. STD spectra were acquired from 64 scans, 2,050 receiver gain, and 14 ppm sweep width.

Quantification of quorum-sensing molecules

Supernatants of overnight cultures of PAO1 grown aerobically or anaerobically were collected and acidified with 0.1% formic acid. RP-HPLC coupled with a mass detector was used to analyze and quantitate QS molecules in culture supernatants. The HPLC-MS system used was a Waters HPLC system equipped with a Waters ZQ Single Quadrapole Mass Detector outfitted with MASS LYNX software. The HPLC/MS method was water/acetonitrile (ACN) gradient with 0.1% formic acid in both solvents. Samples (10 µl) were loaded onto an Ascentis Express C18 column (Sulpeco Analytical, 5 µm particle size, 150mm × 2.1mm) and eluted using an acetonitrile (ACN) gradient at a flow rate of 0.4 ml/min: 30% ACN for 2 min, a linear gradient from 30 to 100% ACN over 30 min, 100% ACN for 5 min. Column temperature was held at 30 °C. The concentrations of QS molecules were determined using calibration curves generated with commercial chemical standards at concentrations ranging from 0.1 µM to 10 mM.

Patient recruitment

This study protocol was approved by the Institutional Review Board of the Medical University of South Carolina, and the study was carried out according to the approved guidelines (IRB number: Pro00012798). Written informed consent was obtained from each patient at the time of recruitment. Patients were selected using the following criteria: over 18 years old, had a positive culture history of P. aeruginosa but negative for Burkholderia species, non-smoker, readily produced sputum, and received routine treatment at the Adult CF Center of the Medical University of South Carolina. Sputum samples were collected during patients’ routine visits at the CF Center clinic and stored at −80 °C until the samples were analyzed.

Quantification of QS molecules in sputum samples

Sputum samples were thawed and digested with 10% sputolysin (Calbiochem, San Diego, CA, USA) in PBS at 37 °C for 15 min. Digested samples were extracted with 3 ml acidified ethyl acetate (0.1% formic acid) twice. The organic layers were pooled and dried under nitrogen. Samples were dissolved in 500 µl methanol and 10 µl sample was analyzed by the HPLC-MS as described above to determine the concentrations of QS molecules in the sputum.

Statistical analysis

The Log-Rank (Mantel-Cox Test) test was used to analyze the C. elegans survival data. A p-value of less than or equal to 0.05 was considered as a significant difference. Other data was analyzed using the Student’s t-test (DF = 1), significant differences were identified with a p value of less than or equal to 0.05.

DHQ maintained P. aeruginosa virulence in C. elegans

C. elegans is an established infection model to study bacterial pathogenicity in research labs (Korgaonkar et al., 2013; Kwak, Jacoby & Hooper, 2013). P. aeruginosa infections are lethal to the worms due to toxic virulence factors such as pyocyanin and hydrogen cyanide, chelation of iron by PQS, and colonization of the intestinal lumen. To determine the effect of DHQ on P. aeruginosa pathogenicity, we monitored the survival of C. elegans infected with wild-type P. aeruginosa strain PAO1 and pqs mutants (Fig. 1). E. coli OP50 is a food source to maintain the nematodes and served as a non-pathogenic control in our study. PAO1 requires both virulence factor production and colonization of the nematodes to kill C. elegans following several days of incubation (>200 h) (Mahajan-Miklos et al., 1999; Tan, Mahajan-Miklos & Ausubel, 1999; Tan et al., 1999). The Pqs system regulates virulence factor production and plays a role in biofilm formation, both of which play a role in C. elegans killing. Using the DHQ-only producing mutant, ΔpqsB and ΔpqsC strain (Zhang et al., 2008), we were able to examine the effect of DHQ production on virulence compared to a quinolone-null mutant, ΔpqsAB.

None of the nematodes survived infections from PAO1, ΔpqsB or ΔpqsC, which only produces DHQ, after 220 h of incubation, while 50% of the nematodes infected with ΔpqsAB remained viable (Fig. 1). The quinolone-null mutant ΔpqsAB used in this study was also used in our previous work (Rella et al., 2012), which lacks an internal promoter of the pqs operon. The reduction in virulence of the ΔpqsAB mutant was attributed to the complete loss of quinolone production, while production of DHQ in the ΔpqsB and ΔpqsC mutant maintained pathogenicity towards the nematodes. PqsB and PqsC are both essential for alkylquinolone synthesis, but are not required to produce DHQ (Zhang et al., 2008). C. elegans killing by the ΔpqsB and ΔpqsC mutants showed that DHQ production maintained the same level of virulence in both mutants, which were more virulent against the worms than the quinolone-null mutant ΔpqsAB. Thus, the production of DHQ increased pathogenicity of P. aeruginosa compared to a quinolone-null strain.

Fluorescence-producing bacteria have been used to infect C. elegans to study the internalization of bacteria and their progress through the intestines (Sifri et al., 2003). To visualize colonization of C. elegans by different P. aeruginosa strains, we monitored nematodes infected with PAO1 and the pqs mutants carrying a plasmid that stably expressed GFP (Bloemberg et al., 1997). After 4 h of co-incubation, all nematodes examined contained GFP-expressing bacteria (Fig. 2). After 24 h, GFP-expressing bacteria were distributed throughout the intestinal tract of the nematodes. At 144 h, C. elegans infected with PAO1 and ΔpqsB showed higher GFP fluorescence than those infected with the ΔpqsAB mutant (Fig. 2). We determined the difference in bacterial colonization by comparing fluorescence of the three strains following 1 s of excitation versus 5 s of excitation. A 5 s excitation led to oversaturated GFP signal in nematodes infected with PAO1 and the ΔpqsB mutant, while minimal fluorescence was detected within the ΔpqsAB mutant infected worms. A 1 s excitation resolved the oversaturation in PAO1 and ΔpqsB- infected nematodes and resulted in only background fluorescence from nematodes infected with the ΔpqsAB mutant (Fig. 2). The 144 h was an important time point as it coincided with a sharp decrease in C. elegans survival among the nematodes infected PAO1 and ΔpqsB (Fig. 1). The GFP-fluorescent signals from the infected worms were quantitated using the software ImageJ (Fig. S1). Consistent with the observations on the fluorescent images, no significant difference was detected at 4 h and 24 h post infection. At 144 h time point, worms infected with the pqsAB mutant exhibited significant lower GFP signal than PAO1-infected (p < 0.05) and pqsB-infected worms (p < 0.05). Altogether, our results indicated that production of DHQ increased the ability of P. aeruginosa to colonize and infect C. elegans, which may point to a role in causing infection of a human host.

DHQ supported pyocyanin production in vitro

Virulence factor production is also important for PAO1 infection of C. elegans (Eriksson et al., 2009). Pyocyanin, a blue–green redox-active pigment, is a potent virulence factor that is toxic towards other microorganisms and eukaryotic cells (Lau et al., 2004; Papaioannou, Utari & Quax, 2013; Rella et al., 2012). Both PqsR and PqsE are important for Pyocyanin production in P. aeruginosa: PqsR activates transcription of the pqs operon (Williams & Camara, 2009), PqsE function remains enigmatic until recently being shown to have pathway-specific thioesterase activity and promote alkylquinolone signaling (Drees & Fetzner, 2015).

In order to demonstrate that changes in virulence of the pqs mutants were not due to defects in growth, the growth of the pqs mutants were monitored in LB and synthetic cystic fibrosis medium (SCFM) that was developed based on the nutrient composition of the CF sputum (Palmer, Aye & Whiteley, 2007). The growth curves of the pqs mutants were similar to that of the wild-type PAO1 in both media, demonstrating that these mutants were not defective in replication (Figs. 3A and 3C). Next, we quantified pyocyanin from overnight cultures of PAO1 and pqs mutants grown in LB media. PAO1 produced the highest amount of pyocyanin (about 2.5 µg/ml) followed by the ΔpqsB mutant, which produced 50% of the wild-type level (Fig. 3B). The ΔpqsAB mutants produced the least amount of pyocyanin (Fig. 3B). Nutrient conditions can alter bacterial phenotypes. The SCFM mimics the natural environments of P. aeruginosa and provides less nutrients than LB. Although reduced, a similar trend in pyocyanin production by the strains was observed compared to LB-grown cultures (Fig. 3D). These results support that DHQ production partially maintained pqs signaling and pyocyanin production, whilst alkylquinolones were required for full virulence production.

DHQ activated PqsR for transcription of the pqs operon

PQS and HHQ activate PqsR for transcription of the pqs operon (Diggle et al., 2007; Dubern & Diggle, 2008). To determine if DHQ affected pqs operon transcription, we used qRT-PCR to monitor the expression of pqsA from PAO1 and pqs mutants using RNA polymerase sigma factor rpoD as the house-keeping gene (Fig. 4). ΔpqsB mutant expressed pqsA at 47% of the wild-type level, while ΔpqsAB mutant showed no detectable pqsA transcript (Fig. 4). Deletion of pqsR also reduced pqsA transcription to basal level, which was consistent with PqsR being an activator of the pqs operon. Deletion of both pqsR and pqsB also resulted in basal level of pqsA transcript (Fig. 4), suggesting that the activation of pqs operon transcription in ΔpqsB was dependent on PqsR.

To investigate the effect of exogenous and endogenous DHQ on PqsR activity, we performed promoter-fusion assays in E. coli and P. aeruginosa strains (Fig. 5). The promoter-fusion construct contained the upstream regulatory element (−500 bp) of pqsA fused with the coding sequence of LacZ (Cugini et al., 2007). The reporter construct also contained the pqsR gene, which was controlled by a tac promoter. DHQ supplemented to E. coli containing the reporter construct displayed a concentration-dependent increase in LacZ activity (Fig. 5A). 100 µM DHQ increased the pqsA promoter activity by 60% over the DMSO-treated control, compared with 110% activation by 1 µM PQS (Fig. 5A), suggesting that PQS was more potent in activating PqsR. The effect of endogenous quinolones on PqsR activity was tested in PAO1 and pqs mutants carrying the reporter construct. The pqsA promoter activity from the DHQ-only ΔpqsB mutant was similar to that of PAO1 while the ΔpqsAB strain displayed basal activity (Fig. 5B), indicating endogenous produced DHQ can activate PqsR binding to pqsA promoter.

To test if DHQ can compete with alkylquinolones for PqsR, DHQ was supplemented in the presence of HHQ or PQS in the E. coli reporter system. DHQ supplemented with either HHQ or PQS did not significantly alter the pqsA promoter activity compared to HHQ and PQS alone (Fig. 5C). Various fungal species produce farnesol as a quorum-sensing molecule, which is also an antagonist of PqsR (Cugini et al., 2007). In the E. coli reporter system, farnesol reduced LacZ activity below the background level (Fig. 5D), consistent with farnesol as an inhibitor of PqsR. Interestingly, co-treatment of farnesol and DHQ alleviated the inhibition of PqsR by farnesol, albeit at a lower level compared with PQS (Fig. 5D). Taken together, endogenous transcription of pqsA from PAO1 and pqs mutants, as well as activity from exogenous DHQ supplemented to the reporter system, show that DHQ can function as a ligand of PqsR to activate the transcription of the pqs operon.

In vitro studies on the interaction of DHQ with PqsR

The interactions between alkylquinolones with PqsR are demonstrated by a recently solved crystal structure of PqsR complexed with its ligand, which shows how ligand binding affects PqsR affinity for its DNA-binding site (Ilangovan et al., 2013). To test how DHQ binding affects PqsR interaction to its DNA binding site, we performed electrophoretic mobility shift assays (EMSA) with biotinylated-pqsA promoter (Bio-pqsA) and cell lysates of PAO1 and the pqs mutants (Fig. 6). Only PAO1 and ΔpqsB cell lysates shifted Bio-pqsA migration, while the ΔpqsAB and ΔpqsR mutants did not show any effect (Fig. 6A). Addition of unlabeled pqsA probe to PAO1 and ΔpqsB cell lysates decreased the amount of shifted Bio-pqsA probe, demonstrating the interaction was specific for pqsA promoter sequence (Fig. 6B). Recombinant PqsR as a fusion protein with N-terminal His-tagged SUMO also showed increased shifted pqsA probe in the presence of PQS or DHQ (Fig. 6C). No pqsA probe was shifted in the control lysate containing His-SUMO tag, demonstrating the shifted pqsA probe was due to PqsR activity specifically (Fig. 6C). The affinity of PqsR for DHQ and PQS was determined semi-quantitatively by EMSA by measuring free Bio-pqsA in response to quinolone supplemented at different concentrations. The calculated Kd of PqsR for DHQ was 150 nM and 33 nM for PQS (Fig. 6D).

Saturation transfer difference NMR (STD-NMR) detects transient binding of ligand to a protein in solution and was used to study the interaction between DHQ and PqsR. The ligand binding domain of PqsR starting from residue Cys87 (PqsR-C87) was purified as a His-tagged SUMO fusion protein and used in the STD-NMR experiments (Fig. 7). Novel peaks on the PqsR-C87 spectra were quantified for intensity and modeled for single-site binding kinetics (Fig. 7A). The Kd was estimated to be 450 nM (Fig. 7B). Interestingly, the saturation time showed a strong interaction after 3 s, while exhibiting fast-exchange interactions from 0.5 to 2.5 s. The fast-exchange may be explained by hydrophobic interactions within the hydrophobic pocket of PqsR until DHQ was coordinated through hydrogen binding. Using the D-COSY assignments, we mapped the 1-D hydrogen spectra to show that the meta-hydrogen of DHQ participated in hydrogen-binding with PqsR (Fig. 7C). The ortho- and para-hydroxyl groups were not resolved from the water peak. However, we predict those groups were also participating in hydrogen binding based on preliminary docking analyses.

DHQ production under anaerobic conditions

Our results suggested DHQ, similar to PQS and HHQ, promotes P. aeruginosa pathogenesis by activating PqsR; however, conditions found in CF lungs may limit PQS production due to the lack of oxygen (Schobert & Jahn, 2010; Worlitzsch et al., 2002). Further complicating alkylquinolone syntheses is the reduced metabolism in P. aeruginosa biofilms limiting the availability of fatty acid substrates from β-oxidation. As DHQ formation requires smaller, less energy intensive precursors and does not rely on oxygen in the environment, DHQ may play a role apart from the alkylquinolones during chronic colonization of the CF lung.

Strict anaerobic cultures do not produce PQS and express lower pqs operon activity compared to aerobic cultures (Lee et al., 2011). To determine how DHQ concentration is affected during anaerobic growth, we quantified QS molecules from PAO1 cultures grown aerobically and anaerobically (Fig. 8). Concentration of C₄-HSL was higher than 3-oxo-C₁₂-HSL in the anaerobic samples, consistent with the relative abundance of the two HSL signals under anaerobic conditions as found in the CF lungs (Hassett et al., 2009). High levels of extracellular DHQ were detected from cultures grown in LB both in the presence and absence of oxygen (Fig. 8A). Interestingly, anaerobic culture supernatants contained higher concentrations of DHQ compared to aerobic cultures in LB. As expected, PQS was not detected under anaerobic conditions. Similar trends of QS molecules were observed from PAO1 cultured in SCFM aerobically and anaerobically (Fig. 8B). Comparison of the DHQ concentrations between LB and SCFM showed that DHQ level was higher in the SCFM, but the difference was not statistically significant (Fig. 8). Together, these results showed DHQ was the predominant quinolone produced in both rich and SCFM medium regardless of the oxygen level. The production of DHQ in anaerobic condition suggest a specific function of DHQ under low oxygen environments where PQS production is limited; however, further research is needed to understand P. aeruginosa transcriptional regulation in hypoxic conditions and the dynamics of QS molecules within the anaerobic mucous.

Detection of DHQ in CF patient sputa

Alkylquinolones have been detected in the CF patient samples (Collier et al., 2002). We analyzed the sputum samples from stable CF patients to determine whether DHQ was detectable in the CF lungs. Sputum samples were collected during CF patient routine clinic visits and were analyzed by HPLC-MS for DHQ concentrations. More than 80 sputum samples from 45 patients were collected and analyzed. DHQ was readily detected in 34 sputum samples, and the concentrations covered a wide range. The median of DHQ concentration in stable CF sputa was 123 µM with an interquartile range between 25 and 464 µM. The wide range of DHQ concentrations in the patient samples reflected the non-uniform nature of the sputum samples. Nonetheless, the presence of DHQ in the patient samples demonstrated that DHQ was produced during an infection in the CF lungs, and the effects of long-term exposure to high concentration of DHQ on host cells remains to be determined.