Introduction

The Gram-negative opportunistic pathogen P. aeruginosa is a major cause of hospital-acquired infections and is a significant antibiotic resistance threat. To survive in a broad range of conditions, this bacterium has evolved a variety of signaling pathways that detect changes in nutrients, chemical gradients, and flow, and activates responses that promote its survival (Fang et al., 2016; Sanfilippo et al., 2019). Understanding how extracellular-sensing pathways detect and respond to host cues is critical for improving the treatment of bacterial infections and minimizing the generation of antibiotic-resistant bacteria.

Mammalian immune systems respond to the presence of bacterial pathogens by recruiting immune cells that migrate rapidly through the circulatory system to sites of infection (Aymonnier et al., 2023). Neutrophils are the first and most abundant immune cells recruited to inflamed or damaged tissues (Lavoie et al., 2011) and are stimulated into an “activated” state by bacterial components, such as the membrane constituent lipopolysaccharide. Activated neutrophils produce multiple products that are toxic to bacteria including reactive oxygen species (ROS) and reactive chlorine species (RCS) (Hampton et al., 1998; Segal, 2005; Sultana et al., 2020; Ulfig and Leichert, 2020).

While the toxic effects of ROS on bacteria are understood, far less is known about how bacteria are affected by RCS (Fang, 2011; Gray et al., 2013). Hypochlorous acid (HOCl) is a major RCS product of neutrophils that is formed through myeloperoxidase enzymes from hydrogen peroxide (H₂O₂) and chloride ions, and is strongly bactericidal (Albrich and Hurst, 1982; Harrison and Schultz, 1976; Segal, 2005; Winterbourn et al., 2016). P. aeruginosa is a potent activator of neutrophil respiratory bursts (Jensen et al., 1990) but whether and how the bacteria detect and respond to neutrophil components is not understood.

P. aeruginosa responds to shear generated by the flow of fluids via the fro operon, which is upregulated in the bacterium during infection in humans (Cornforth et al., 2018; Padron et al., 2023; Sanfilippo et al., 2019) (Figure 1A). In addition, fro is necessary for successful colonization of the gastrointestinal tract and lung infections (Potvin et al., 2003; Skurnik et al., 2013), although the reasons for this is unclear. Given its important role in infection, it is possible that Fro could respond to immune-associated cues, such as the presence of neutrophils. The operon consists of PA14_21570, PA14_21580, PA14_21590, and PA14_21600, which was termed froABCD (Sanfilippo et al., 2019), and the extracytoplasmic function (ECF) sigma (σ) and anti-σ factors PA14_21550 and PA14_21560 (Boechat et al., 2013), which were termed froR and froI respectively, by (Sanfilippo et al., 2019). Expression of froABCD is regulated by FroR and FroI: overexpression of froR upregulates fro, whereas overexpression of froI downregulates fro (Sanfilippo et al., 2019). Flow was reported to activate FroR-dependent transcription of froABCD through a mechanism that involves chemical transport (Padron et al., 2023).

Figure 1.

Here, we show that the fro operon responds directly to HOCl, which is produced robustly by neutrophils during immune response. Using a reporter system in live bacteria and through quantitative measurements of transcription, we show that fro expression is upregulated by activated neutrophils and in a murine model of corneal infection. Through transcriptional profiling, we determine that bacteria mitigate the toxic effects of HOCl through a FroR-dependent mechanism that activates methionine sulfoxide reductases to relieve oxidative stress. These data suggest a novel mechanism in which bacterial defense against host innate immunity is primed by a single system that detects both flow and HOCl.

Results

Macrophages and neutrophils activate fro expression in P. aeruginosa under non-flow conditions

To investigate whether immune cells directly activate the fro operon, we quantified fro expression using a P. aeruginosa strain that co-expresses the yellow fluorescent protein (YFP) under the transcriptional control of the fro promoter and mCherry under the control of a constitutive promoter (Figure 1B). While fro expression increased under flow (Figure 1—figure supplement 1A), experiments in this study were performed in the absence of flow in order to deconvolve the effects of flow from potential activation of fro by immune cells. Because neutrophils and macrophages are the first immune cells recruited to sites of infection, we co-incubated P. aeruginosa with human neutrophils or mouse macrophages for 30 minutes. Relative to P. aeruginosa that were incubated in host cell-free media, P. aeruginosa co-incubated with either neutrophils or macrophages increased YFP/mCherry ratios by comparable levels (Figure 1C-D), suggesting that close proximity to or direct interaction with these host immune cells increases fro expression in P. aeruginosa.

Activated neutrophils induce fro expression

To assess whether fro upregulation in P. aeruginosa could be part of a response to respiratory bursts from neutrophils and macrophages, we focused on the fro response to neutrophils (Figure 2A) because these cells produce larger respiratory bursts and greater levels of ROS than macrophages (Colombo et al., 2017; Forman and Torres, 2002; Thekkan et al., 2019). P. aeruginosa were incubated in conditioned medium from human neutrophils that were stimulated using phorbol 12-myristate 13-acetate (PMA), which triggers respiratory bursts (Karlsson et al., 2000). The expression of fro was quantified by averaging the ratios of YFP to mCherry fluorescence for at least one hundred individual P. aeruginosa cells. In support of our hypothesis, conditioned medium from PMA-stimulated neutrophils increased fro expression in bacteria by 30-fold compared to conditioned medium from unstimulated neutrophils (Figure 2B-C, Figure 2—source data 2). We confirmed that this fro response was not due to PMA alone, as PMA-treated medium without neutrophils had no effect on fro expression (Figure 2B-C). Similarly, conditioned medium from unstimulated neutrophils had no significant effect on fro expression compared to medium alone, suggesting that the upregulation of fro is due to respiratory burst products.

Figure 2.

HOCl and other RCS are produced from ROS during respiratory bursts via myeloperoxidase (MPO). To investigate whether P. aeruginosa fro expression is induced by MPO-derived RCS, we pre-treated neutrophils prior to PMA stimulation with 4-aminobenzoic acid hydrazide (4-ABAH) (Figure 2D), a specific and irreversible MPO inhibitor (Kettle et al., 1997). Indeed, the activation of fro expression by PMA-stimulated neutrophil medium was abrogated in neutrophils that were pre-treated with the MPO inhibitor. The degree of fro inhibition by 4-ABAH followed a dose-dependent trend (Figure 2—figure supplement 1), consistent with a graded fro response to different RCS concentrations. HOCl production in neutrophils due to PMA stimulation and its inhibition by 4-ABAH was confirmed using a fluorometric HOCl-specific assay (Figure 2—figure supplement 2). These results indicate that fro upregulation depends on MPO activity and suggest that fro is activated specifically in response to HOCl.

HOCl but neither H₂O₂ nor HNO₃ induces fro expression

ROS and RCS have inhibitory effects on bacterial growth. To explore the hypothesis that these classes of molecules induce fro expression in P. aeruginosa, we measured the effects of H₂O₂, HNO₃, and NaOCl on fro expression at concentrations approaching their minimum inhibitory concentrations (MICs) (Figure 3A). NaOCl increased fro expression 74-fold at 1 µM relative to untreated (Figure 3B, Figure 3B–Source Data). To validate that the increase in YFP intensity is due to increased transcription, reverse transcription quantitative PCR (RT-qPCR) was also performed, which yielded a 200-fold increase in fro transcription in response to 1 µM NaOCl treatment (Figure 3C). While 2 µM NaOCl caused only a 3-fold increase in fro expression, we attribute this to the effects of NaOCl on transcription and cell viability, as this concentration also decreased the constitutive mCherry reporter intensity and cell size (Figure 3—figure supplement 1).

Figure 3.

To rule out that the individual sodium or chloride ion components of NaOCl were responsible for the increase in fro expression, we repeated experiments using NaCl, which yielded no change in fro expression (Figure 3D). To assess whether the fro response may be affected by pH, we tested NaOH at the same concentration as NaOCl, and at a concentration at which a change in pH could be discerned. Due to the robust buffering capacity of the medium, a pH change of ∼0.1 required 6 mM NaOH, which is a concentration that had no obvious effect on growth but is significantly higher than the concentration of NaOCl needed to activate fro expression. Regardless, neither concentration of NaOH had any significant impact on fro expression (Figure 3D). As fro activation is not due to sodium or chloride ions nor changes in pH, we attributed the effects to hypochlorite (OCl^-), which is in equilibrium with hypochlorous acid (HOCl) at the pH of 7.1 in the medium used in our experiments.

To address the possibility that fro expression is activated by oxidation alone, we quantified fro expression in response to H₂O₂ and HNO₃, at concentrations up to their respective MICs (Figure 3A). While both are strong oxidizers, HNO₃ is not produced by neutrophils. Neither H₂O₂ nor HNO₃ induced any significant changes in fro expression up to near-MIC concentrations (Figure 3E). The lack of fro expression due to H₂O₂ treatment is not due to defects in fluorescent protein production, as H₂O₂ did not induce any change in fro transcript abundance, as assayed using RT-qPCR (Figure 3C).

Together, these indicate that strong oxidation alone is not sufficient to upregulate fro. We assessed whether other antimicrobial components of neutrophils or other bacterial stressors could be responsible for the fro response. Stimulated neutrophils release the antimicrobial peptide LL-37 and histones, which together constitute a synergistic antimicrobial attack (Doolin et al., 2020; Duong et al., 2025). Tobramycin is an aminoglycoside, a class of antibiotics that induces oxidative stress within bacteria (Kohanski et al., 2008). Neither the combination of LL-37 and histones nor tobramycin at near-MIC concentrations induced fro expression (Figure 3F). Together, these results indicate that fro expression is activated specifically by the neutrophil effector HOCl and not by intracellular oxidative stress or other neutrophil defense factors including LL-37, histones, or H₂O₂.

Corneal infection activates fro expression in bacteria

P. aeruginosa is a major cause of corneal ulcers and keratitis, which can cause visual impairment and blindness. Neutrophils are recruited within hours to corneal epithelial abrasions, with neutrophil cell density peaking at 18 hours (Li et al., 2006). We tested the hypothesis that fro expression is upregulated during corneal infection using a murine bacterial keratitis model in which P. aeruginosa strain PAO1F is inoculated onto a corneal abrasion, which we previously demonstrated results in significant neutrophil recruitment to the cornea (Minns et al., 2023; Ratitong et al., 2022). Mice corneas were harvested 2 or 20 hours post-infection and fro transcripts were quantified using RT-qPCR. Transcript abundance was normalized to the number of P. aeruginosa cells using 5S ribosomal RNA. In support of the hypothesis, fro expression increased significantly by 20 hours post-infection (Figure 4A). Expression levels showed no significant change within the first two hours following infection, which we attribute to a lower level of immune cell recruitment during this period. The increase in fro expression 20 hours post-infection was not due to differences between the P. aeruginosa strains, because fro transcription in PAO1F also increased in response to NaOCl but not H₂O₂ (Figure 4—figure supplement 1), consistent with previous results.

Figure 4.

Methionine sulfoxide reductase upregulation requires FroR

FroR is the σ factor required for activation of the fro operon in response to flow (Sanfilippo et al., 2019). To understand the response to HOCl, we performed transcriptional profiling of wild-type (WT) and ΔfroR P. aeruginosa strains. The genes that were most upregulated by NaOCl in the WT strain compared to the ΔfroR mutant were those in the fro operon, the phosphate and pyrophosphate-specific outer membrane porin genes oprO and oprP, and the methionine sulfoxide reductase genes msrB, msrP, and msrQ (Figure 5A and Appendix 1—table 1). Methionine sulfoxide reductases (MSRs) have a critical role in bacterial survival and defense against RCS stress. HOCl oxidizes methionine and cysteine residues 100-fold more rapidly than other amino acids (Pattison and Davies, 2006; Winterbourn, 1985) and oxidation of methionine in bacterial membrane proteins by neutrophil-produced HOCl correlates with increased bactericidal killing of E. coli and S. aureus (Rosen et al., 2009). HOCl oxidation of methionine forms methionine sulfoxide, which impairs protein function.

Figure 5.

MSRs restore proper protein function by reducing methionine sulfoxide back to methionine (Chandra et al., 2023; Gray et al., 2013). Our data thus suggest that FroR is required for the upregulation of MSRs in response to HOCl stress. We did not observe upregulation in PA14_27570, which is homologous to MsrC, which reduces strictly free methionine sulfoxide, or in PA14_66330 (MsrA), which reduces both free and protein-bound methionine sulfoxide (Appendix 1—table 1). No significant changes in the expression of enzymes that repair cysteine oxidation, such as thioredoxin, glutathione redoxin, or disulfide repair, were observed, suggesting that the fro response is directed towards repairing methionine oxidation.

We reasoned that HOCl could oxidize methionine residues within a P. aeruginosa sensor, which could in-turn upregulate fro expression. A number of anti-σ factors contain sensory domains that transduce sensory signals to their cytoplasmic σ factor (Paget, 2015). FroI is an anti-σ factor for the FroR σ factor (Boechat et al., 2013), which raises the possibility that HOCl or its secondary products could oxidize FroI directly, resulting in the activation of FroR. Consistent with this possibility, we found that FroI contains the highest concentration of methionine and cysteine residues of all identified P. aeruginosa PA14 anti-σ factors (Appendix 2—table 1).

We hypothesized that fro activation would be suppressed in the presence of excess of free methionine, as it would reduce available HOCl and thus oxidation of the FroI sensor. Indeed, co-treatment of NaOCl with methionine suppressed fro expression (Figure 5B). We reasoned that other antioxidants such as cysteine and β-mercaptoethanol (βME) could also function as alternative oxidation targets for HOCl, and therefore also suppress fro. Consistent with this hypothesis, co-treatment of NaOCl with cysteine or βME significantly suppressed fro expression compared to cysteine or βME alone (Figure 5B and Figure 5—figure supplement 1A-B).

We considered how fro could be upregulated by stimulated neutrophils in the context of these findings. HOCl and other reactive chlorine species produced by neutrophils could oxidize the P. aeruginosa sensor, resulting in fro upregulation. Under this interpretation, supplying excess methionine, which would reduce HOCl, should also inhibit activation by neutrophils. Indeed, supplementing conditioned media from PMA-stimulated neutrophils with methionine completely suppressed fro expression (Figure 5C). Together, these data suggest a model in which fro responds to HOCl-induced oxidation of a sensor by upregulating MSRs.

FroR protects P. aeruginosa against HOCl

The upregulation of MSRs is expected to improve bacterial growth against HOCl stress. We measured the growth profiles of WT and ΔfroR strains of P. aeruginosa. Treatment with 4 µM NaOCl significantly inhibited the growth of the wild-type strain (Figure 6A) but resulted in even greater growth inhibition in the ΔfroR strain. This observation suggests that the fro operon improves P. aeruginosa tolerance to HOCl stress.

Figure 6.

We considered other genes which may improve WT P. aeruginosa tolerance to HOCl. HOCl increased the expression of iron uptake genes (pvdA and pvdQ) and downregulated the expression of oxidative phosphorylation genes (cyoCDE) (Figure 6B and Appendix 1—table 2). These changes suggest that HOCl causes an iron deficiency and decreases oxidative phosphorylation. In order to compensate for decreased oxidative phosphorylation, P. aeruginosa may utilize amino acid catabolism as an alternative energy source, as evidenced by a concomitant upregulation of amino acid catabolism genes (hpd, bkdA1, bkdA2, and bkdB) by HOCl (Figure 6B and Appendix 1—table 2).

While the ΔfroR mutant was more sensitive to HOCl than WT, its growth was not fully inhibited (Figure 6A), suggesting that additional mechanisms relieve HOCl stress. In the ΔfroR mutant, HOCl upregulated genes encoding multidrug efflux pumps (mexEF-oprN and mexXY) (Figure 6C and Appendix 1—table 3). These pumps can relieve oxidative stress (Fargier et al., 2012; Fraud and Poole, 2011) but were not activated in the WT strain. This observation suggests that bacterial defense against HOCl may be hierarchical, in which Fro-regulated and Mex pathways could be first-line and second-line responses to HOCl stress, respectively. Supporting this interpretation, it has been reported that treatment of the WT strain using significantly higher (millimolar) HOCl concentrations upregulates mexEF-oprN (Farrant et al., 2020) and the mexXY regulator mexT (Groitl et al., 2017).

Discussion

Neutrophils are activated by the presence of bacteria that infect host tissue. In their activated state, neutrophils generate respiratory bursts that produce an abundance of HOCl. We have demonstrated that P. aeruginosa responds to activated neutrophils by upregulating the transcriptional expression of the fro operon, which relieves HOCl stress. Our data suggest that the upregulation of fro could function as a bacterial defense mechanism against neutrophil attack, thus improving P. aeruginosa survival and colonization of human tissue during infection.

Detection and response to HOCl

Based on (i) our observation that the expression of multi-drug efflux pumps is activated in the absence of FroR (Figure 6C and Appendix 1—table 3), (ii) that this expression occurs at high HOCl concentrations (Farrant et al., 2020), and (iii) that multiple systems repair HOCl damage, the HOCl response could follow a hierarchy of activation. Our data suggest a model in which the fro system is the first-line response to HOCl. FroR activates two distinct classes of reductases that repair HOCl-oxidized methionine in proteins: MsrB reduces methionine sulfoxide in the cytoplasm using electrons provided by the thioredoxin system (Ezraty et al., 2005; Lourenço Dos Santos et al., 2018), while MsrPQ works in the periplasm and inner membrane using electrons from the respiratory chain (Gennaris et al., 2015). The activation of these two reductase types enables P. aeruginosa to limit HOCl stress in separate cellular spaces by utilizing distinct electron donor sources.

How does P. aeruginosa detect HOCl? As an anti-σ factor, FroI could immobilize FroR, inhibiting FroR from reaching transcriptional targets. Given the particularly high density of methionine and cysteine residues in the FroI, this protein could serve as an HOCl sensor by releasing the sigma factor FroR in response to oxidation of these residues (Figure 6D). While our data shows that HOCl can directly activate fro expression, HOCl reacts with neutrophil components, forming secondary RCS, including monochloramines, protein-derived chloramines, and taurine chloramine (Kim and Cha, 2014; Klebanoff et al., 2013; Ulfig and Leichert, 2020). Taurine chloramine is an expected major RCS released from neutrophils due to the high abundance of taurine in these cells (Kim and Cha, 2014; Marcinkiewicz and Kontny, 2014). Because it retains oxidizing potential towards cysteine and methionine (Peskin and Winterbourn, 2001), taurine chloramine could contribute to the activation of fro expression alongside HOCl. Future work will need to assess the relative efficacies of HOCl and other secondary RCS towards activating fro expression.

Relationship with neutrophil-mediated killing and flow

One of the striking features of the Fro system is that it is tuned for maximal activation at micromolar HOCl concentrations (1 µM) that are close to the MIC. While our experiments measured the effects of HOCl at steady-state using medium that does not contain amino acids, host proteins, and other host factors, HOCl dynamics in host environments are much more complex, where HOCl production is estimated to be in quantities that exceed the sensitivity of Fro. For example, HOCl production is predicted in excess of 100 mM per minute in the phagosome (Winterbourn et al., 2006). In addition, stimulated neutrophils at densities comparable to those found in blood (Tahir and Zahra, 2026) using PMA produces extracellular HOCl at 80 µM per hour, as measured through taurine chloramine absorbance (Weiss et al., 1982). Additional measurements using slight changes in stimulation time, neutrophil density, or PMA concentration yielded comparable HOCl production rates within a factor of 2 (Dypbukt et al., 2005; Test et al., 1984). At sites of inflammation, HOCl production in interstitial fluid has been predicted to be in the millimolar range due to the greater density of neutrophils there (Weiss, 1989).

Despite the high production rates of HOCl, the steady-state concentration in neutrophils and in vivo is expected to be significantly lower due to its rapid conversion into secondary RCS through reaction with neutrophil or plasma components. For example, the lifetime of HOCl is estimated to be 0.1 µs and have an action radius under 0.1 µm in the phagosome (Schürmann et al., 2017), which would mitigate the antimicrobial effects of HOCl but retain oxidizing potential. The secondary RCS taurine chloramine is long-lived, well-tolerated in the body, and retains significant bactericidal activity in the micromolar range (Gottardi and Nagl, 2010; Kim and Cha, 2014; Nagl et al., 2000). It is plausible that diminished HOCl or longer-lived secondary RCS such as taurine chloramine fall within the activating range of Fro, which would in turn mitigate their antimicrobial effects through upregulation of the MSRs. Importantly, our observation that fro expression is activated during corneal infection, which involves significant neutrophil recruitment, supports a role for the Fro response in a neutrophil-dense environment.

An additional layer of complexity of understanding the role of Fro during infection is that flow exposes bacteria to higher concentrations of molecules in a fluid. The bacterial response to flow, which was initially referred to as rheosensing (Sanfilippo et al., 2019), was more recently attributed to the effects of molecular transport by flow (Padron et al., 2023). In particular, flow stimulated a response to H₂O₂ in cells that were otherwise unresponsive to it in the absence of flow (Padron et al., 2023). Flow could similarly amplify the stimulating effects of RCS, causing P. aeruginosa to respond to circulating neutrophils that produce RCS at much lower concentrations. At the same time, flow could amplify the bactericidal effects against P. aeruginosa. Future work will need to determine the relative steady-state abundances of different secondary RCS and address their impact in combination with flow on bactericidal activity and the activation of the Fro response.

At the fundamental level, Fro is a system that increases antimicrobial tolerance, which can promote antimicrobial resistance. For example, bacteria that are exposed to low concentrations of HOCl have higher rates of horizontal gene transfer and can acquire antibiotic resistance genes more rapidly (Hou et al., 2019; Jin et al., 2020). The Fro system could thus facilitate antibiotic resistance through increasing P. aeruginosa tolerance to HOCl. Our findings suggest that the bacterial HOCl response could thus be an important bacterial defense mechanism to target in the development of novel antibiotic therapeutics. Inhibition of this pathway could increase P. aeruginosa susceptibility to neutrophil-mediated killing while minimizing antibiotic resistance.

Materials and methods

Bacterial Culture Conditions

Strains were struck out from frozen stocks maintained at -80°C onto plates containing LB-Miller broth and 2% Bacto agar and incubated overnight at 37°C. Single colonies were inoculated into 2mL of modified MinA minimal medium (60.3mM K₂HPO₄, 33.0mM KH₂PO4, 7.6mM (NH₄)₂SO_4, 1.0mM MgSO₄; (Malke, 1993) containing 0.2% (w/v) sodium citrate (hereafter referred to as MinA minimal medium), and cultured for 18 hours in a roller drum at 17 rpm and 37 °C.

Imaging P. aeruginosa under flow

Microfluidic devices were fabricated using standard soft lithography techniques as established previously (Siryaporn et al., 2015) with channel dimensions of 100 x 50 μm (width x height). P. aeruginosa was cultured for 18 hours in LB-miller medium, diluted 1:100, cultured to mid-exponential phase, and injected into the microfluidic device. Fresh LB was injected at a constant rate of 10μL/min. Images were acquired as described in the fluorescence microscopy section.

Neutrophil Isolation

Neutrophils were isolated freshly from whole blood as described in (Clark et al., 2018). Briefly, 3% dextran in PBS was used to separate red blood cells (RBCs) from whole blood. Neutrophils were purified from the remaining cells by overlaying on a Ficoll density gradient and centrifugation in for 30 minutes at 500 × g. RBCs were lysed using RBC Lysis Buffer and centrifugation for 5 mins at 300 × g. Supernatant was aspirated and the remaining neutrophil pellets were resuspended in RPMI 1640 medium prior to normalizing cell counts. Neutrophils were used immediately for stimulation, preparation of conditioned media, or co-incubation experiments with P. aeruginosa as described below, and purity was assessed in parallel using the fluorescent antibodies APC-CD11b, FITC-CD16 and PE-CD66b and a ACEA NovoCyte Flow Cytometer. This procedure routinely yielded greater than 95% CD11b-positive neutrophil populations.

Co-incubation of Mammalian Cells with P. aeruginosa

Neutrophils were prepared as described above. J774A.1 mouse macrophages were cultured in 10 mL DMEM in T75 flasks at 37 °C with 5% CO₂. Macrophages were passaged every three days through scraping and were passaged every 5 days through trypsinization. Mammalian cells were washed three times by centrifugation at 300 × g for 5 minutes and resuspension in DPBS, and 1 mL aliquots of 4x10⁵ cells were transferred into flat-bottom dishes. P. aeruginosa overnight (∼18h) cultures were centrifuged at 4600 × g for 2 minutes, resuspended in DPBS three times, diluted to a final OD₆₀₀ of 0.2, added to the mammalian cells at a multiplicity of infection of 10, and incubated for 30 minutes at 37°C. Dishes were aspirated until approximately 20 µL of medium remained. 1% agarose pads containing DPBS were placed on top of the samples in the dishes to immobilize bacteria and cells. Samples were imaged as described in the fluorescence microscopy section.

Conditioned Medium from Neutrophils

To activate neutrophils, 2 million freshly harvested neutrophils per mL in 10mM phosphate buffer (pH 7.4; 2.46 mM monobasic phosphate, 7.54 mM dibasic sodium phosphate) containing 140 mM sodium chloride, 10 mM potassium chloride, 0.5 mM magnesium chloride, 1 mM calcium chloride, 1 mg/mL glucose, and 5 mM taurine (Dypbukt et al., 2005), were treated with 100 ng/mL of PMA in 14mL polystyrene round-bottom Falcon tubes for 1 hour at 37°C and 17 rpm. For MPO inhibition studies, neutrophils were pre-treated with 2 mM of 4-ABAH for 30 mins prior to addition of PMA. This concentration of 4-ABAH was scaled up based on its usage of 500 μM of 4-ABAH to treat one million human neutrophils per mL (Tseng et al., 2018). Supernatant was collected after centrifugation for 5 minutes at 4,600 × g and 24 °C and used as conditioned medium immediately for HOCl detection assay as described below. In parallel, P. aeruginosa that were cultured for 18 hours in MinA minimal medium were diluted 1:1000 into conditioned media and were incubated for 3 hours at 37 °C in a rolling drum at 17 rpm and imaged as described in the fluorescence microscopy section.

HOCl Fluorometric Detection Assay

Conditioned medium was assessed immediately using the Amplite Fluorometric Hypochlorite Assay Kit according to manufacturer protocol. Samples were incubated at room temperature in the dark for 10 minutes and measured for fluorescence using a BioTek Synergy HTX plate with 540/25 nm and 585/10 nm filters for excitation and emission, respectively, and using the autogain set to the 1 μM NaOCl condition. HOCl concentrations were determined by fitting the calibration values to the Hill equation using a three-parameter fit and the lowest value in the data series as the minimum value. HOCl concentrations were background-subtracted by the neutrophil buffer-only sample.

Treatments of P. aeruginosa with oxidizers, antioxidants, and stressors

P. aeruginosa were cultured 18 hours in MinA minimal medium and diluted 1:1000 into the same medium. HOCl, H₂O₂, HNO₃, NaOH, NaCl, methionine, L-cysteine, or β-mercaptoethanol (βME) were added to 50 mL cultures at the concentrations indicated in the main text and were incubated in 250 mL flasks for 3 hours at 37 °C in an orbital shaker at 225 rpm. LL-37, histones and tobramycin were added following the 1:1000 diluted and incubated in 14 mL falcon tubes in a roller drum at 17 rpm. Cultures were imaged and analyzed as described in the fluorescence microscopy section.

Fluorescence and Phase Contrast Microscopy

P. aeruginosa were immobilized on 1% agarose pads containing MinA minimal medium and imaged immediately. Cultures containing densities below 10 P. aeruginosa per frame were concentrated using a syringe filter and imaged immediately. Experiments in Figures 2D and 3F were performed without the syringe filter in order to confirm that the P. aeruginosa response was not affected by the filtering procedure.

Imaging

Images were acquired using Nikon NIS-Elements software and a Nikon Eclipse Ti-E microscope containing a Nikon 100X Plan Apo (1.45 N.A.) objective, Nikon Ph3 phase contrast condenser annulus, Sola light engine, Hamamatsu Orca Flash 4.0 V2 camera, 459/526/596 dichroic, 575/25 excitation and 641/75 emission filters for mCherry, and 509/22 excitation and 544/24 emission filters for YFP (Semrock, Rochester, NY).

Image Analysis

Images were analyzed in MATLAB using custom built software written previously (see the Key Resources Table to download). Briefly, P. aeruginosa masks were determined from phase contrast images using an edge-detection algorithm. For size analysis, the total pixel area of each individual P. aeruginosa cell was determined by computing the mask area and converting from pixels to μm² by multiplying the mask area by a factor of 0.004225 μm²/pixel to account for the microscope camera pixel size and objective magnification. Fro expression was quantified by averaging computed ratios of YFP to mCherry fluorescence from the masked areas for at least one hundred individual P. aeruginosa cells per condition for each experiment.

Bacterial Growth Curves

Overnight cultures of bacteria were diluted into MinA minimal medium to an optical density at 600 nm (OD₆₀₀) of 0.01, given treatments, and incubated in sterile 96-well microplates at 37°C in a Synergy HTX multi-mode plate reader with continuous orbital shaking at 180 rpm and a 3mm amplitude, with OD₆₀₀ readings taken every 30 minutes.

Reverse Transcription Quantitative PCR (RT-qPCR)

For non-murine model samples, bacterial overnight cultures were diluted 1:1000 into fresh MinA medium and incubated for 2.5 hours at 37°C and 225 rpm preceding a 30 minute treatment with 1 µM NaOCl or vehicle control. Samples were concentrated using syringe filters and centrifuged at 13,000 × g for 4 minutes. Pellets were snap-frozen in liquid nitrogen and stored at -80C before performing RNA extraction with the NucleoSpin RNA kit. cDNA was prepared from the using the High-Capacity cDNA Reverse Transcriptase Kit, and quantitative PCR was performed using the SsoAdvanced Universal SYBR Green Supermix using the C1000 Thermal Cycler with a CFX96 real-time detection system. froA transcripts were quantified using the primers PA14_froA_v2_F and PA14_froA_v2_R that target PA14_21570 in PA14 and PA3284 in PAO1 (see Key Resources Table). Transcript abundances for each sample were normalized by 5S rRNA abundance using the primers PAO1_PA14_5S_F and PAO1_PA14_5S_R (see Key Resource Table), yielding normalized count thresholds (Cts). The average fold change in transcript abundance was determined by computing the ratio of the averaged normalized Cts and raising by the power of 2, as performed previously (Bru et al., 2019). All primers were validated using a standard calibrating curve to assess Cts and transcript abundance.

Murine Model of Corneal Infection

Expression of froA was assessed in a murine model of corneal infection as described previously (Minns et al., 2023; Sun et al., 2012). The work received prior approval from the Institutional Animal Care and Use Committee at the University of California Irvine under protocol #AUP-21-123. Briefly, P. aeruginosa strain PAO1F was cultured to mid-exponential phase in MinA minimal medium that was modified to contain only 0.05% citrate and supplemented with 0.1% casamino acids and 0.2% glucose as the carbon sources, centrifuged, and resuspended in PBS. Corneal epithelia of C57BL/6J mice aged 6-8 weeks were abraded with 3x5 mm scratches using a 25G needle, and 2 µL of PBS containing 5x10⁴ of P. aeruginosa strain PAO1F was applied topically. Mice were selected at random and those displaying signs of illness or distress were excluded from the study. Comparable numbers of male and female mice were used in each experimental group. The number of animals used for each experiment was determined based on a power analysis estimate. Mouse eyeballs were collected and homogenized in 1 mL of PBS 2 or 20 hours post-infection and centrifuged for 5 minutes at 150 × g to remove corneal materials. Supernatant was collected and centrifuged for 4 minutes at 13,000 × g. Pellets containing bacteria were suspended in lysis solution (10 mM Tris-HCl, 1 mM EDTA pH 8.0, 0.5 mg/mL lysozyme, 1% SDS) (Bru et al., 2019) RNA was prepared using the NucleoSpin RNA kit and assessed for mRNA expression using RT-qPCR. Pellets containing bacteria were suspended in lysis solution (10 mM Tris-HCl, 1 mM EDTA pH 8.0, 0.5 mg/mL lysozyme, 1% SDS) (Bru et al., 2019).

Samples were not blinded because they were processed as they were collected at different timepoints, which revealed their identity.

RNA-seq Library Preparation and Data Analysis

P. aeruginosa was cultured and RNA was prepared as described in the RT-qPCR section. RNA yield was measured using a Nanodrop 2000. Samples were depleted of ribosomal RNA using the NEBNext rRNA Depletion kit, from which cDNA libraries were constructed using the NEBNext Ultra II Directional Library kit, which were sequenced by the UC Irvine Genomics Research & Technology Hub (Irvine, CA) using an Illumina NovaSeq X Plus using paired-end 150 bp reads at a depth of approximately 10 million reads per sample. Raw reads were checked with fastQC and trimmed and filtered into paired and unpaired reads using Trimmomatic in Java using the ‘PE’ setting. Paired and unpaired reads were aligned to the PA14 genome (NCBI accession NC_008463.1) using Bowtie2 using the default ‘sensitive’ mode, were counted using the featureCounts program in Subread with the ‘countReadPairs’ option enabled for paired-end reads, and were fit into a negative binomial model to compute fold-change in gene expression using DeSeq2. The statistical significance of the changes were computed using the Wald test in DeSeq2. Gene Ontology enrichment analysis for Figure 1A was performed using GOEnrichment.

Statistical Analyses

Statistical analyses and figures were generated using Excel, R, and RStudio.

Data availability

Raw data for all figures that contain statistical analyses are available in the supplemental source data files and appendices. All other datasets and software are described in the Key Resources Table.

Acknowledgements

The authors thank Michael Z. Zulu and Stephanie Matsuno for their feedback on prior drafts of this article, Hew Yeng Lai for imaging of bacteria in microfluidic devices, and Hanjuan Shao for assistance with RNA preparation.

Additional information

Author Contributions

IPF, SJK, RS and AS conceived and designed the experiments. IPF, SJK, RS, SA, MEM, LAU, LD, and APH performed experiments and analyzed data. IPF and LAU wrote the initial draft of the manuscript. SJK and AS revised it. All authors edited the paper.

Funding

This work was supported by the Stanley Behrens Fellows in Medicine Award to L.U., and grants from the National Institutes of Health (NIH) National Institute of Biomedical Imaging and Bioengineering (NIBIB) (R21EB027840-01), the National Cancer Institute (NCI) (DP2CA250382-01), the National Science Foundation (NSF) (DMS1763272), and the Simons Foundation (594598QN) to TLD. IPF received funding from the Horizon Europe Research and Innovation Program through the Marie Skłodowska-Curie grant #847413 and from the Program of the Ministry of Science and Higher Education #5005/H2020-MSCA-COFUND/2019/2. Experiments in the revised manuscript were supported by a grant from the National Institute of Allergy and Infectious Disease (R01AI163196) to AS and EP.

Funding

HHS | NIH | National Institute of Biomedical Imaging and Bioengineering (NIBIB) (R21EB027840)

• Timothy Downing

HHS | NIH | National Cancer Institute (NCI) (DP2CA250382)

• Timothy Downing

National Science Foundation (NSF) (DMS1763272)

• Timothy Downing

Simons Foundation (SF) (594598QN)

• Timothy Downing

EC | Horizon Europe | Excellent Science | HORIZON EUROPE Marie Sklodowska-Curie Actions (MSCA)

https://doi.org/10.3030/847413

• Ilona P Foik

Ministry of Science and Higher Education (5005/H2020-MSCA-COFUND/2019/2)

• Ilona P Foik

HHS | NIH | National Institute of Allergy and Infectious Diseases (NIAID) (R01AI163196)

• Albert Siryaporn

Additional files

Supplementary Materials

Source Data