Density‐Dependent Modulation of the Oxidative Burst by Neutrophils in Response to Planktonic and Biofilm Growing Pseudomonas aeruginosa: When Less Is More

ABSTRACT

The neutrophilic response to chronic Pseudomonas aeruginosa lung infection is strongly associated with collateral tissue damage impairing lung function in people with cystic fibrosis (pwCF). We recently demonstrated the concurrent presence of both biofilm bacteria and planktonic bacteria in chronically infected lungs in pwCF. However, the neutrophilic response to these distinct bacterial forms remains insufficiently characterized. To investigate this, we fractionated P. aeruginosa batch‐cultures to enrich for biofilm or planktonic cells. Confocal microscopy showed that aggregates > 30 μm made up > 92.5% of biofilm biomass, while aggregates < 10 μm comprised > 95% of planktonic biomass. Viable cell numbers were validated by similar correlations between CFU/mL and intracellular DNA content. The intensity of the oxidative burst by neutrophils in response to the fractions was estimated by luminol‐enhanced chemiluminescence. At low density, biofilm fractions triggered stronger responses than planktonic cells, whereas at high density, planktonic fractions induced the highest activation, indicating density‐dependent modulation. These findings suggest that at low densities, biofilms may exacerbate tissue damage by amplifying neutrophil responses. Therefore, the reduced bacterial load in pwCF on ETI therapy could make the balance between biofilm and planktonic P. aeruginosa more critical for lung inflammation, though their distribution under ETI remains unknown.

1. Introduction

Biofilm formation by P. aeruginosa presents a major clinical challenge in people with cystic fibrosis (pwCF) with chronic lung infections, as well as in individuals with infected chronic wounds [1]. These biofilms are highly resistant to both antibiotics treatment and the host immune responses, leading to persistent infections and triggering excessive inflammation that damages surrounding tissue. Neutrophils, recruited to clear the infection, are key contributors to this hyperinflammatory response and the resulting collateral tissue injury. Notably, in pwCF, the neutrophil density in sputum correlates more strongly to reduced lung function than the actual bacterial load of P. aeruginosa [2]. This damage is largely attributed to neutrophil‐derived oxidative and proteolytic activities. In pwCF, the impact of the oxidative activities on lung tissue damage is highlighted by the inverse correlation between oxidated protein in lung specimens and the pulmonary function [3]. Similarly, the concentration of proteolytic neutrophil elastase in lung specimens is inversely correlated with lung function [4]. Recent microscopy‐based studies demonstrated the coexistence of both planktonic and biofilm‐growing P. aeruginosa in chronically infected lungs from pwCF patients [5, 6]. At present the effect of the bacterial biogeographics on the tissue‐damaging response by the neutrophils is largely unknown. It has, however, long been recognized that neutrophils respond with an increased oxidative burst against solitary planktonic P. aeruginosa as compared to the oxidative burst against equivalent numbers of attached P. aeruginosa biofilms [7, 8]. Meanwhile it was recognized that P. aeruginosa aggregates in suspended biofilms rather than surface attached biofilms inchronically infected lungs from pwCF patients [9]. Therefore, we aimed to investigate the significance of biofilm formation in more clinically relevant settings by comparing the neutrophilic oxidative burst induced by planktonic P. aeruginosa with the response to non‐attached biofilms of P. aeruginosa . Given that bacterial density may vary between lung samples, we also determined the impact of the bacterial density on the induced oxidative burst by neutrophils.

An essential part of this study was the establishment of bacterial samples containing equivalent numbers of either planktonic or non‐attached P. aeruginosa biofilms, grown under identical conditions to ensure comparable metabolic state. Liquid batch cultures of P. aeruginosa may contain a mix of single cells and aggregates [10, 11]. Therefore, we grew P. aeruginosa as liquid batch cultures to obtain identical growth conditions and metabolic state before applying filtration to separate the culture into two fractions according to aggregate size. The different distribution of aggregate size was validated by confocal laser scanning microscopy. To ensure samples of equivalent bacterial densities across fractions, we validated CFU‐based quantification by correlating colony counts with the intracellular DNA content. This method enabled us to study the oxidative burst of normal neutrophils in response to samples of biofilm‐growing or planktonic P. aeruginosa of equivalent bacterial density and growth status. To detect activation of the oxidative burst of neutrophils by the fractionated bacteria, we employed a modification of luminol‐enhanced chemiluminescence [12], which is a sensitive method for quantifying neutrophil activation based on the production of intra‐ and extracellular reactive oxygen species (ROS) resulting from the oxidative burst in response to stimulation [13].

2. Materials and Methods

2.1. Bacterial Strains and Growth Conditions

In this study we used the P. aeruginosa isolate PAO1, which was originally isolated from an Australian burn wound patient [14, 15], was obtained from the Pseudomonas Genetic Stock Center as PAO0001 and PAO1 tagged with a green fluorescent protein (GFP) expressed on plasmid pMRP9 [16]. To select bacteria expressing the GFP‐encoded plasmid, a fluorescent microscope was used. PAO1 and GFP‐tagged PAO1 were streaked on Lysogeny‐Broth (LB) agar plates and incubated overnight at 37°C. The following day colonies were collected from the plates and inoculated in 100 mL LB media supplemented with 0.3% glucose. The cultures were incubated at 37°C with shaking (180 rpm) for six days.

2.2. Fractionation of Bacterial Culture Into Planktonic Cells and Aggregates

After six days of incubation 45 mL of the bacterial culture was poured over a stack of cell strainers (PluriSelect, Leipzig, Germany) with mesh sizes of 30 and 10 μm (Figure 1). Aggregates collected in the 30 μm strainer were isolated and collected by carefully scraping and backwashing the filter, and the planktonic cells were collected in the filtrated liquid. The fraction containing aggregates was washed twice with NaCl (0.9%) to reduce the number of planktonic cells. The aggregates were allowed to settle, the upper part of the fraction was removed, and the settled aggregates were washed as done previously [17]. The volume removed during the wash was replaced with NaCl (0.9%). The size fractions were kept at 4°C to prevent further growth until Colony Forming Units, CFU/ml was determined.

FIGURE 1.

Graphical representation of the filtration method used to separate single cells (0–10 μm) from aggregates > 30 μm.

2.3. DNA Extraction of Planktonic Cells and Aggregates

To verify whether CFU/ml was an appropriate measure to compare the number of viable bacteria in aggregates and planktonic cells, DNA extraction of the fractions adjusted to concentrations of 10⁶, 10⁷, and 10⁸ CFU/mL was carried out. The cells were first treated with DNase I (ZYMO Research, Irvine, California, USA) to remove extracellular DNA and DNA from dead cells. DNA from viable bacteria was extracted using the DNeasy Blood and Tissue extraction kit (Qiagen, Hilden, Germany).

One mL of cells with 10⁶, 10⁷, or 10⁸ CFU/mL were centrifuged at 10,000 g for 10 min, and the supernatant was discarded. The pellet was resuspended in 200 μL NaCl (0.9%) with 10 μL of DNase I with a concentration of 1 U/ng and 10 μL DNase buffer, which then was allowed to incubate for 10 min at 37°C on a shaker set to 100 rpm. The enzyme was inactivated heating to 65°C for 10 min. The suspension was centrifuged for 10 min at 10,000 g, the pellet was resuspended in 200 μL saline, and the supernatant was discarded. The cells were washed twice by repeating the following steps: After centrifugation, the pellet was resuspended in 200 μL NaCl (0.9%), and the supernatant was discarded. The suspension was centrifuged for 10 min at 10,000 g.

The rest of the DNA extraction was carried out according to the manufacturer's instructions for the DNeasy Blood and Tissue extraction kit (Qiagen). DNA concentration was determined with a Qubit 4 fluorometer (Thermo Fisher Scientific Inc., Waltham, MA, USA).

2.4. Confocal Laser Scanning Microscopy

A fractionated PAO1 culture was diluted to 10⁶, 10⁷, and 10⁸ CFU/mL for aggregates and planktonic cells, respectively. The fractions were stained with Syto9 (Merck, Darmstadt, Germany) with a final concentration of 2.5 μM and added to channel slides (uncoated μ‐slide VI‐flat, Ibidi, Gräfelfing, Germany). The samples were analyzed with a confocal laser scanning microscope, CLSM 880 (Zeiss, Oberkochen, Germany), using a 20×/0.8 M27 objective and a frame size of the images at 425 × 425 μm. An Argon laser with an excitation wavelength of 488 nm was used to excite the fluorophore. The microscope was set to detect emitted light for wavelengths ranging from 494 to 561 nm. Surface areas of the bacteria in the images were quantified using Imaris (version x64 9.7.2), and areas were grouped into intervals of diameters.

2.5. Isolation of Neutrophils

Human neutrophils were isolated from the peripheral blood of healthy volunteers after obtaining written informed consent and the project was approved by the Regional Committee on Health Research Ethics (Project Nr. H‐22056226). Isolation of neutrophils was performed according to a modification of a previous procedure (15). Briefly, 20 mL blood was drawn into EDTA tubes (BD Vacutainers K2E (EDTA), REF 367525). 5 mL of blood was layered on top of 5 mL of Polymorphprep (ProteoGenix, Schiltigheim, France) in 15 mL centrifuge tubes. The tubes were centrifuged at 500 g for 32 min at 21°C (Sorvall RC 6 Plus Superspeed Centrifuge, Thermo Scientific, Waltham, Massachusetts, USA). The lower band containing neutrophils was transferred to a 50 mL centrifuge tube. The neutrophils were washed by filling the centrifuge tube with Hanks' balanced salt solution (HBSS, H6648, Sigma) [18] at room temperature to the 40 mL mark and centrifuged at 400 g for 10 min at 20°C (Sigma 2‐16P, Buch og Holm, Herlev, Denmark). Remaining unwanted erythrocytes were removed using 10 mL ACK lysing buffer (Gibco, ThermoFisher scientific, A1049201) for 3 min followed by centrifugation for 5 min at 300 g at 20°C (Sigma 2‐16P). The pellet was resuspended in 10 mL HBSS and centrifuged for 10 min at 120 g (Sigma 2‐16P). The supernatant was discarded before resuspending the neutrophil‐enriched pellet in 10 mL HBSS and the concentration of neutrophils was determined by adding 100 μL resuspended neutrophils to 0.9 mL FACS lysing solution (BD Biosciences, REF 249202) diluted 1:10 in MilliQ H₂O with propidium iodide (Sigma, P4170) at a final concentration of 100 μg mL⁻¹ to a TruCount tube (BD Biosciences, REF 663028). The stained neutrophils in the TruCount tubes were analyzed in a flow cytometer (Attune Next, Thermo Fisher Scientific) according to the manufacturer's description. The resuspended neutrophils were centrifuged for 10 min at 120 g at 20°C (Sigma 2‐16P), the supernatant was discarded, and the pellet was resuspended in Krebs‐Ringer buffer (16) with 10 mM d‐(+)‐Glucose (Sigma, G8270) to a final concentration of 4 × 10⁶ neutrophils mL⁻¹.

2.6. Assessment of Neutrophil Oxidative Burst Using Luminol‐Enhanced Chemiluminescence

The assay was performed following a modified protocol based on a previously established method (17). Briefly, to avoid light contamination the assay was carried out in white, 96‐well microtiter plates (Thermo Fisher Scientific). Each well was filled with 175 μL KRB with 140 M luminol (Merck, Darmstadt, Germany) followed by addition of 50 μL bacterial inoculum with appropriate density. For controls, 50 μL of KRB without bacteria was added. Lastly, 75 μL of the neutrophil suspension (4 × 10⁶ neutrophils/mL) were added to reach a final concentration of 10⁶ neutrophils/ml. Immediately after adding the neutrophils, the microtiter plate was placed in a plate reader (Viktor, PerkinElmer, Waltham, Massachusetts, USA) to measure the level of light emitted from the wells. The plate reader was set to a temperature of 37°C while shaking the plate before every measurement. The measurements were repeated 99 times with an interval of 2.5 min between measurements. Light emitted was quantified in counts per second (CPS).

2.7. Statistical Analysis

Graphical presentation and statistical analysis of the data included in the study were conducted using GraphPad Prism 10 (GraphPad Software LLC). The data for the oxidative burst were normalized by log10‐transformation and the normalized data were compared using a Two‐Way repeated‐measures analysis of variance (ANOVA) corrected for multiple comparisons using Tukey's multiple‐comparison test. A p level below 0.05 was considered significant.

3. Results

3.1. Fractionation Into Single Cells and Biofilm Aggregates

A filtration‐based method was developed to separate planktonic and aggregated bacteria. Confocal microscopy images indicated that the planktonic fraction primarily consisted of individual cells, while the aggregate fraction mainly consisted of multicellular aggregates (Figure 2A).

FIGURE 2.

Planktonic cells and aggregates from fractionated bacterial cultures. (A) Visualization of the biomass of the planktonic and aggregate fractions diluted to 10⁶, 10⁷, or 10⁸ CFU/mL. (B) Biomass distribution within non‐diluted fractions (mean ± SD of five technical replicates, from each of three biological replicates of the planktonic fraction and five of the aggregated fraction). (C) DNA (ng) extracted from planktonic cells (blue) and aggregates (purple) adjusted to different CFU/mL. Datapoints represent means ± SD.

Quantification of biomass, based on estimated surface area, of the planktonic (0–10 μm) and aggregate (> 30 μm) fractions showed that 95.5% of the total biomass of the planktonic fraction had a diameter between 0 and 10 μm, with the remaining 4.5% consisting of small aggregates between 10 and 30 μm (Figure 2B). Importantly, no aggregates larger than 30 μm were observed in this fraction. In contrast, 92.5% of the total biomass in the aggregate fraction was in aggregates > 30 μm, 5.7% in the 10–30 μm range, and only 1.8% as aggregates were < 10 μm (Figure 2B). Thus, the 0–10 μm fraction was completely devoid of large aggregates, while the > 30 μm fraction was highly enriched for large aggregates, containing only a minor proportion of smaller aggregates or single cells.

3.2. Validation of Equivalent Bacterial Density

The enumeration of bacteria within non‐attached aggregates and biofilms has traditionally been considered unreliable when using the plate count method, as a single CFU may originate from a multicellular aggregate, potentially leading to substantial underestimation of the true bacterial population. To address this limitation, we validated that the CFU adjusted fractions contained the expected number of viable bacteria. This was achieved by quantifying the intracellular DNA content of viable bacteria by first removing extracellular DNA and DNA from dead cells by treating with DNase I and secondly removing DNase I by washing several times. The remaining DNA was then extracted, quantified for each fraction, and correlated to CFU measurements. We found that the amount of DNA in planktonic cells closely matched that of the aggregate fractions diluted to 10⁶, 10⁷, and 10⁸ CFU/mL, respectively (Figure 2C), indicating that CFU counts provided a reliable estimate of viable bacterial numbers. Based on this correlation, we concluded that CFU counts could be used to construct inocula with equivalent bacterial densities from both the planktonic and the aggregated biofilm fraction.

3.3. Oxidative Burst of Neutrophils in Response to Planktonic Cells or Aggregates

The ability to separate bacterial cultures into planktonic cells and aggregated biofilms—while maintaining comparable bacterial densities—was used to investigate how bacterial density influences the oxidative burst of neutrophils stimulated by either planktonic or biofilm‐forming bacteria. The oxidative burst was quantified as the total ROS production expressed as the area under the curve (AUC). At lower bacterial densities of 10⁵ and 10⁶ CFU/mL, aggregated biofilms induced a significantly stronger oxidative burst in neutrophils compared to planktonic bacteria. No difference in the oxidative burst by neutrophils in response to aggregates or planktonic bacteria was found at bacterial densities of 10⁷ CFU/mL, while the oxidative burst by neutrophils was stronger in response to planktonic bacteria compared to aggregates at bacterial densities of 10⁸ CFU/mL. Compared to exposure to KRB alone, aggregates induced a significantly increased oxidative burst by neutrophils at densities from 10⁵ to 10⁸ CFU/mL while planktonic bacteria stimulated a significantly increased oxidative burst by neutrophils at 10⁷ and 10⁸ CFU/mL (Figure 3).

FIGURE 3.

Density‐dependent activation of the oxidative burst by neutrophils in response to planktonic cells or aggregating P. aeruginosa . Total ROS production by neutrophils upon stimulation for 3 h was expressed as AUC. Blue: Aggregated bacteria, red: Planktonic bacteria, black: Background control without bacteria. *: Significantly higher ROS production in response to aggregated bacteria. **: Significantly higher ROS production in response to planktonic bacteria. Blue straight line: Density interval of significantly higher ROS production in response to aggregated bacteria as compared to the background. Red straight line: Density interval of significantly higher ROS production in response to planktonic bacteria as compared to the background. Statistical test by Two‐Way ANOVA of 4 biological replicates. p < 0.05 was considered significant.

4. Discussion

The recent demonstrations of the simultaneous presence of both planktonic and biofilm‐growing bacteria in chronic infections [5, 19] warrant more detailed knowledge of distinct tissue damaging responses of neutrophils to planktonic and biofilm‐growing bacteria. To experimentally address these observations, we developed a method to separate bacterial cultures into planktonic and aggregated biofilms with clinically relevant sizes. The majority of aggregates in the biofilm‐enriched fraction exceeded 30 μm in diameter, closely resembling aggregate sizes reported in chronic infections [20]. In addition, the relevance of comparing the immunostimulatory capacity of the two fractions depends on our ability to calibrate the numbers of viable bacterial cells. We validated the feasibility of colony enumeration for controlling the density of living bacteria in the fractions by demonstrating similar correlations between CFU/mL and the DNA content in the two fractions. This validation relied upon the assumption that DNA quantification can approximate numbers of living bacterial cells when using monocultures under controlled growth [21] and by applying DNAse treatment to selectively remove DNA from dying bacteria [22] and the extracellular DNA contained in the matrix of biofilm [23].

When exposing biofilm and planktonic fractions at varying bacterial densities to human neutrophils, our findings demonstrate that the oxidative burst of neutrophils is significantly influenced by both the aggregation level of P. aeruginosa (planktonic vs. aggregated in biofilm) and the bacterial density. At lower densities (10⁵ and 10⁶ CFU/mL), aggregates elicited a significantly stronger oxidative burst compared to planktonic cells. At these low bacterial densities, only stimulation with biofilm aggregates resulted in a significant increase in the oxidative burst compared to the background, suggesting that P. aeruginosa aggregates can trigger a neutrophil response at lower density thresholds than their planktonic counterparts. This observation may reflect enhanced immunostimulatory properties of biofilm aggregates at low bacterial concentrations, possibly due to their structural and compositional features. As key effector cells of the innate immune response, neutrophils detect P. aeruginosa through pattern recognition receptors (PRRs) that recognize pathogen‐associated molecular patterns (PAMPs) expressed by the bacteria [24, 25]. It has been suggested that the magnitude of the neutrophil response, including ROS production, depends in part on the numbers of PAMPs interacting with PRRs [26]. Structural differences between aggregating and planktonic bacteria may influence the spatial distribution and local concentration of PAMPs resulting in larger contact areas with expression of high numbers of PAMPs on the aggregates. Thus, within bacterial aggregates, the high cell density could facilitate the clustering or accumulation of PAMPs at levels sufficient to surpass the activation threshold for the neutrophils to respond by initiating the oxidative burst. It has recently been shown that biofilms may contain special compositional features with immunostimulatory properties, and these types of molecular patterns have been termed biofilm‐associated molecular patterns (BAMPs) [26]. In this respect, the immuno‐stimulatory properties of extracellular polysaccharides and DNA in the matrix of the aggregates [27, 28] may have contributed to the enhanced response of neutrophils against aggregates at low bacterial densities. In contrast, at the highest density tested (10⁸ CFU/mL), planktonic bacteria induced a stronger neutrophil response, which is in line with previous observations at similar high bacterial density [7, 8]. The stronger oxidative burst mounted by neutrophils in response to planktonic bacteria at the high densities may reflect increased surface‐to‐volume ratio and accessibility to immune recognition in planktonic bacteria, as well as a saturation effect in aggregate stimulation. In addition, planktonic bacteria may express certain PAMPs at more immunostimulatory levels than bacteria within biofilms. One such example is the flagellum which is recognized by TLR5 [29] and has been shown to play a key role in the heightened oxidative burst observed in neutrophils exposed to planktonic P. aeruginosa , compared to those exposed to biofilm‐associated P. aeruginosa [27]. In addition to differences in the composition of expressed PAMPs, biofilm‐growing bacteria differ from planktonic bacteria in the size of the structures they form. This variation in size may have important consequences for the neutrophil response, as larger objects can induce frustrated phagocytosis, a condition in which the particle is too large to be fully engulfed. This results in incomplete phagosome formation, membrane disruption, and the extracellular release of cellular contents, including inflammatory mediators [30]. Given that the objects in our aggregates fraction were considerably larger than the objects in the planktonic fraction, it is possible that frustrated phagocytosis contributes to the enhanced neutrophil response observed at low bacterial densities. If so, our findings are consistent with previous suggestions that frustrated phagocytosis may drive excessive inflammation during biofilm infections in the lungs of pwCF. However, our measurements of ROS were based on luminol, which detects total ROS production [31] and does not distinguish between intracellular ROS and extracellular ROS resulting from cellular leakage. To further investigate the potential contribution of frustrated phagocytosis, future studies should assess the extracellular release of ROS by isoluminol and neutrophil granule‐derived inflammatory mediators.

Interestingly, at an intermediate bacterial density (10⁷ CFU/mL), there was no significant difference between the responses to the level of aggregation, suggesting a transition point where the contributions of planktonic and biofilm‐associated properties to immune stimulation converge.

Our findings emphasize that bacterial density is a critical determinant in shaping the neutrophil response and bacterial density should be carefully controlled in immunological studies of biofilms. The stronger stimulation observed for aggregates at lower densities may have clinical relevance in early stages of infection or during treatment‐induced biomass reduction, where residual biofilm fragments could provoke substantial neutrophilic activation and associated tissue damage. The demonstration that neutrophils respond more strongly to biofilm‐forming than planktonic P. aeruginosa at low bacterial densities may explain the modest and insignificant improvement in lung function in pwCF after two weeks of antipseudomonal therapy, despite an approximately 2‐log reduction in CFUs [32]. As biofilms remained unchanged, we speculate that their persistence sustains neutrophil‐driven inflammation, thereby limiting functional recovery.

The density‐dependent modulation of the neutrophils' response to planktonic and biofilm‐growing P. aeruginosa may also be of relevance for treating chronic P. aeruginosa lung infection in pwCF who received the most effective CFTR modulator, elexacaftor/tezacaftor/ivacaftor (ETI). Before the introduction of ETI the bacterial densities in sputum samples could typically exceed 10⁷ CFU/mL [33, 34], which is within the range of bacterial densities that stimulated a strong neutrophil response in our setup regardless of planktonic or biofilm mode of bacterial lifestyle. Following ETI, however, bacterial densities may be reduced by 2–3 logs [35] entering the range of low bacterial densities where only biofilm growing bacteria could induce a significant neutrophil response in this study. Sputum samples with bacterial densities below the density required for activation of neutrophils in this study were also observed. However, the distribution of biofilm‐forming and planktonic bacteria in ETI‐treated lungs of pwCF is presently unknown.

A limitation of the current study is that it focuses solely on ROS production as a readout of neutrophil activation. While this is a key effector function linked to tissue damage, other important neutrophil responses—such as degranulation, NETosis, or cytokine release—were not assessed. It is also likely that additional components of the immune system, such as cytokines, complement factors, or opsonizing antibodies, modulate both the activation threshold and the magnitude of neutrophil responses as seen in the chronically infected lungs of pwCF [36].

In conclusion, this study provides new insights into how the physical organization and density of P. aeruginosa influence neutrophil activation. Our results underline the importance of the level of bacterial aggregation and load in modulating host‐pathogen interactions and suggest that biofilm aggregates, even when less abundant, may serve as potent triggers of inflammation in chronic infections such as pwCF with chronic lung infection. Future work should explore the downstream consequences of these differential responses on tissue damage and clinical outcomes, as well as strategies to selectively modulate the immune response to biofilm‐forming bacteria.

Funding

This work was supported by the Novo Nordisk Foundation, Challenge Program to T.B. (Grant NNF19OC0056411).

Ethics Statement

Human neutrophils were isolated from the peripheral blood of healthy volunteers after obtaining written informed consent and the project was approved by the Regional Committee on Health Research Ethics (Project Nr. H‐22056226).

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data available on request from the authors.