De-identified human blood donations provided by healthy adult subjects that span the demographic spectrum of central Ohio were made under the auspices of the Research Institute Blood Donor Services of Nationwide Children’s Hospital after informed written consent was obtained. PMNs were isolated from these blood specimens for use in studies conducted within our laboratory in conformity with, and as approved under, Nationwide Children’s Hospital Institutional Biosafety Committee (IBC) protocol #IBS-00000449. All experiments were conducted in accordance with relevant guidelines and regulations of the Nationwide Children’s Hospital Research Institute Blood Donor Services and Institutional Biosafety Committee.

Nontypeable Haemophilus influenzae (NTHI) strain 86-028NP (Sirakova et al., 1994) was maintained frozen in LN₂ at passage #4 on artificial medium since its original isolation from the nasopharynx of a child who underwent tympanostomy tube insertion due to chronic otitis media. For fluorescent timelapse microscopy and flow cytometry, GFP-expressing NTHI 86-028NP/pRSM2211 (Mason et al., 2003) was utilized. NTHI strains 1128 and 1728, also isolated from children with chronic otitis media (Bakaletz et al., 1988) had been maintained in LN₂ at passage #4 or #7, respectively. All NTHI strains were grown in Brain Heart Infusion (Thermo Fisher Scientific, Waltham, MA) broth supplemented with hemin (2 μg/mL) (Sigma-Aldrich, St. Louis, MO) and β-NAD (2 μg/mL) (Sigma-Aldrich) (e.g., sBHI) or on chocolate agar (Thermo Fisher Scientific) at 37 °C with 5% CO₂ in a humidified atmosphere.

Human neutrophils were isolated from whole blood via magnetic negative selection using the EasySep™ Direct Human Neutrophil Isolation Kit (StemCell Technologies, Inc., Vancouver, Canada). The final PMN pellet was resuspended in 1 mL HBSS (Thermo Fisher Scientific) and total number PMNs/mL HBSS enumerated before making separate solutions of 1e6 PMNs/mL HBSS for timelapse microscopy or 2e6 PMNs/mL HBSS for flow cytometry.

A murine monoclonal antibody of the IgG isotype against a 44-mer tip-chimer peptide designed to mimic protective epitopes of the DNA-binding “tips” of the alpha and beta subunits of a bacterial DNABII protein were prepared for us under contract to Rockland Immunochemicals, Inc., (Philadelphia, PA) and has been previously characterized (Novotny et al., 2016). This monoclonal antibody is referred herein as “MsTipMab.”

For NTHI 86-028NP/pRSM2211 or 1728, biofilms were established as previously described (Wilbanks et al., 2023) after seeding 10 cm² flat tissue culture tubes (TPP, Trasadingen, Switzerland) with 2e5 CFU/mL solution for 16 h. As NTHI 1128 exhibited growth approximately two times that of either NTHI 86-028NP/pRSM2211 or 1728 (as evidenced by growth curve analysis), NTHI 1128 was seeded as previously described (Wilbanks et al., 2023) but at 1e5 CFU/mL. Regardless of NTHI strain, after 16 h biofilms were gently washed to remove non-adherent bacterial cells prior to incubation for 2 h with 5 μg MsTipMab diluted in sBHI/0⋅8 cm² at 37 °C with 5% CO₂ in a humidified atmosphere for collection of α-DNABII NTHI NRel. Planktonically grown NTHI 86-028NP/pRSM2211 or 1728 were allowed to grow statically in sBHI broth at 37 °C with 5% CO₂ in a humidified atmosphere for 3 h which was pre-determined to be mid-log phase. NTHI 1128 was allowed to grow statically at 37 °C with 5% CO₂ in a humidified atmosphere for 1.5 h to mid-log phase due to the faster rate of growth of this isolate. To prepare lag phase planktonically grown bacteria, NTHI 86-028NP/pRSM2211 or 1728 were allowed to grow statically in sBHI at 37 °C with 5% CO₂ in a humidified atmosphere for 1 h to mid-lag phase. NTHI 1128 was allowed to grow statically in sBHI at 37 °C with 5% CO₂ in a humidified atmosphere for 30 m to mid-lag phase. Growth phase status was pre-determined for all three NTHI isolates by conducting overnight growth curves. To generate NBfA, NTHI biofilms were established in flat tissue culture tubes for 16 h, gently washed to remove non-adherent bacteria, then incubated for 2 h with 1.25 mL sBHI only before collecting NBfA bacteria present within the fresh media above the biofilm, as previously described (Boisvert et al., 2016; Kragh et al., 2023; Zhao et al., 2023).

For fluorescent timelapse microscopy and flow cytometry, all NTHI bacterial populations were collected and centrifuged for 3 m at 14,000 rpm at 4 °C. Bacterial pellets were resuspended in 400 μL 0.1M sodium bicarbonate buffer at pH 8.4, centrifuged as before, resuspended in 0.1M sodium bicarbonate buffer, then sonicated for 2 m in a water bath sonicator (Branson Ultrasonics Corporation, Brookfield, CT). All NTHI strains were then labeled with the red fluorescing pH-sensitive dye, pHrodo™ Red, SE (Thermo Fisher Scientific), per manufacturer’s instructions for 30 m at room temperature, protected from light. pHrodo™ Red SE is a pH-sensitive dye that only fluoresces in acidic environments, such as within a phagolysosome where the pH can reach ≤5.0 (Lee et al., 2003), and thus was utilized as an indicator of successful PMN engulfment. After 30 m, for NTHI strains 1128 and 1728 that do not express GFP, these bacterial populations were then also labeled with green fluorescent membrane stain FilmTracer FM™ 1-43 (Invitrogen, Waltham, MA) per manufacturer’s instructions for 15 m at room temperature and protected from light. For strain 86-028NP, we used a constitutive GFP-expressing reporter, (NTHI 86-028NP/pRSM2211), hence referred to as “NTHI 86-028NP” for simplicity (Bakaletz et al., 1988, 1997; Sirakova et al., 1994; DeMaria et al., 1996; Miyamoto and Bakaletz, 1996; Novotny and Bakaletz, 2003). Dual labeling of NTHI strains 1128 and 1728 with FilmTracer FM™ 1-43 was used to ensure that engineering of NTHI 86-028NP/pRSM211 to report GFP was not responsible for the NRel phenotypes observed.

After the labeling periods, all NTHI bacterial populations were centrifuged for 3m at 14,000 rpm at 4 °C. Bacterial pellets were resuspended in 400 μL 0.1M sodium bicarbonate buffer at pH 8.4, centrifuged as before, and the process repeated for a total of two buffer washings. Bacterial pellets were finally resuspended in 0.1M sodium bicarbonate buffer to 1e6 CFU/mL for microscopy or to 4e6 CFU/mL in 0.1M sodium bicarbonate buffer for flow cytometry. Green fluorescent labeling intensity did not differ between NTHI that reported GFP (strain 86-028NP) and those that were labeled with green FilmTracerFM™ (strains 1128 and 1728) during the duration of any of the studies conducted herein, however equivalent photo bleaching would occur for either with prolonged imaging. To ensure similar CFU NTHI/mL were added into each assay, and to ensure bacterial viability, all fluorescently labeled bacterial populations were diluted and plated for CFU determination.

Nontypeable Haemophilus influenzae (NTHI) bacterial solutions at 1e6 CFU/mL were sonicated for 2m in a water bath sonicator before spotting 10 μL of each bacterial population into separate quadrants of a 4-chambered glass bottom dish (Cellvis, Mountain View, CA). α-DNABII NTHI NRel, NBfA, or planktonically grown NTHI were allowed to adhere to the glass for 30 m at room temperature in a humidified atmosphere before imaging. Immediately before imaging, fluids of the 10 μL droplet were carefully pipetted away, then 10 μL PMN solution at 1e6 PMNs/mL was added. Imaging began as soon as PMNs were added to bacterial spots at 100x magnification on a Nikon Ti2-E White Light inverted microscope (Nikon Instruments Inc., Melville, NY). Images were captured every 30 s until we were able to visualize over-saturation of pHrodo™ red fluorescence. The imaging periods were restricted to a constant period of time for each of the three NTHI strains to allow us to visually detect fluorescence differences amongst the three bacterial populations for each strain before we lost the ability to discriminate relative fluorescence by eye. Due to the inherent heterogeneity of NTHI (Duell et al., 2016), this translated to a total of 10 m for NTHI 86-028NP, 20 m for NTHI 1728, and 35 m for NTHI 1128. All assays were conducted at four separate times and on four separate days for each NTHI strain. Images and videos shown are representative of the independent assays. Microscopy data were processed and analyzed via NIS Elements software and arbitrary red fluorescence intensity in units (AFU) was analyzed with an NIS Elements General Analysis program (v. 5.42.03) (Nikon Instruments Inc.) and reported as arbitrary fluorescence units (AFU). To determine and normalize fluorescence, any background fluorescence at time 0 was removed for each individual bacterial population of each of the 3 NTHI isolates tested, and each strain was imaged separately for a set amount of time, e.g., 10, 20, or 35 min. Mean influx values in Table 1 were calculated from all raw data collected from replicate runs for each bacterial population and each strain.

Bacterial solutions at 2e6 CFU/mL were sonicated for 2 m in a water bath sonicator to disrupt any aggregates, then 500 μL each bacterial solution was carefully added into wells of the 24-well plate such that 2e6 CFU/mL was added to wells that contained 500 μL of a 2e6 PMNs/mL solution. Samples were analyzed with a BD LSRFortessa flow cytometer (BD Biosciences, Franklin Lakes, NJ). Flow cytometry data were analyzed with FloJo software (v. 10.10.0) (FloJo, LLC, Ashland, OR). All flow cytometry was performed at least three separate times on at least three separate days for all populations of each NTHI strain.

After NTHI populations were prepared and fluorescently labeled, the sodium bicarbonate buffer-bacterial suspensions were filter sterilized with a 0.22 μm syringe filter and the bacterial cell-free solutions were immediately subjected to endotoxin quantification via the Thermo Scientific™ Pierce™ Chromogenic Endotoxin Quant Kit (Thermo Fisher Scientific). Endotoxin quantification was performed per manufacturer’s instructions. Sterility was confirmed by spread plating on chocolate agar. Data represent the mean ± SEM of three biological replicates that were repeated on three separate days for each NTHI strain.

Quantification of relative extracellular DNA concentration within bacterial cell-free filter sterilized solutions in which each bacterial population had been suspended for incubation with human PMNs was performed via the Qubit dsDNA HS Assay Kit per manufacturer’s instructions (Thermo Fisher Scientific).

Cytokine and chemokine assessment was performed via BD™ Cytometric Bead Array (CBA) Human Inflammatory Cytokine Cytometric Bead Array (CBA) – I Kit (BD Biosciences) or with BD™ Cytometric Bead Array (CBA) Human Flex Sets (BD Biosciences). To generate samples, the methodology we used to prepare NTHI populations for flow cytometry experiments was repeated. The 1 mL PMN-NTHI solution was collected after 10 m of co-incubation. Cytokine/chemokine analysis was performed immediately according to manufacturer’s instructions on an Accuri C6 flow cytometer (BD Biosciences) and data analyzed with FCAP Array™ Software v3.0.19.2091 (BD Biosciences). Data represent the mean ± SEM of three biological replicates that were repeated on three separate days for each NTHI strain.

Statistical tests were performed with GraphPad Prism 10 for all in vitro assays. Due to small sample sizes, a Shapiro-Wilk test for normality was performed on all data as well as a nonparametric Kruskal-Wallis test [with an uncorrected (i.e., non-parametric) Dunn’s test for multiple comparisons] to ensure that any stated statistical significance with parametric one-way ANOVAs (with Dunnett’s multiple comparison correction) (or lack thereof) were retained. Results from parametric testing are retained when data were normally distributed, and non-parametric test outcomes are reported for data that were not normally distributed. Data are presented as mean ± SEM and significance is shown as *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, and ****p ≤ 0.0001. The number of biological replicates is defined within respective figure legends.

When incubated with mid-log phase planktonically grown NTHI 86-028NP, all PMNs in the field of view (FOV) exhibited uptake activity throughout the 10 min imaging period as evidenced by PMNs that turned a light red color in addition to observing that a few PMNs began to exhibit NETosis activity, as assessed by visual loss of cellular membrane integrity (Figures 1A–C). When average arbitrary fluorescence intensity was quantified at the end of the imaging period, PMNs incubated with planktonic NTHI 86-028NP exhibited 98 arbitrary fluorescence units (AFU) (Table 2). When incubated with NBfA NTHI 86-028NP, all PMNs in the FOV engulfed NTHI and, in contrast to what we observed with planktonic NTHI 86-028NP, there was rapid (within 1–2 m) NETosis activity by the majority of PMNs that was evidenced by bright red fluorescence as demonstrated by approximately half of the PMNs in the FOV. This outcome yielded an average of 142 AFU (Figures 1D–F and Table 2). Intriguingly, PMNs incubated with α-DNABII NTHI 86-028NP NRel exhibited the most rapid uptake and NETosis activity, as well as the most striking red fluorescence with an average of 197 AFU which was significantly greater than that observed with either planktonic or NBfA populations (p ≤ 0.05–0.01) (Figures 1G–I and Table 2). In addition to images presented within Figure 1, relative PMN activity was documented by video microscopy (Supplementary Videos 1–3).

When assayed with the second NTHI isolate, all PMNs in the FOV incubated with log phase planktonically grown NTHI 1128 exhibited rapid uptake activity throughout the 35 min imaging period wherein most PMNs exhibited a light red color. In addition, a few PMNs exhibited NETosis activity, which was similar to PMN activity with planktonic NTHI 86-028NP as reported above. PMNs incubated with NTHI 1128 that had been planktonically grown yielded an average of 104 AFU (Supplementary Figure 1A, panels a–c and Table 2). When incubated with NBfA NTHI 1128, all PMNs exhibited both uptake and rapid NETosis activity, with bright visual red fluorescence detected in most PMNs in the FOV for an average of 168 AFU (Supplementary Figure 1A, panels d–f and Table 2). When incubated with α-DNABII NTHI NRel of strain 1128, we again observed rapid uptake and NETosis activity within 2.5–3 min, as well as the most robust and striking visual red fluorescence with an average of 247 AFU which was significantly greater than that observed with either planktonic or NBfA populations (p ≤ 0.05–0.001) (Supplementary Figure 1A, panels g–i and Table 2). In addition to images presented within Supplementary Figure 1A, relative PMN activity was documented by video microscopy (Supplementary Videos 4–6).

When incubated with the third assessed NTHI strain, 1728, we observed engulfment and NETosis activity by PMNs incubated with mid-log phase planktonically grown NTHI 1728, and a few PMNs exhibited a light red color with an average of 109 AFU by the end of the imaging period (Supplementary Figure 1B, panels a–c and Table 2). When incubated with NBfA NTHI 1728, we observed both rapid PMN uptake and NETosis activity, and bright red fluorescence detected in approximately half of the PMNs in the FOV with an average of 143 AFU (Supplementary Figure 1B, panels d–f and Table 2). As seen with the other two NTHI isolates, PMNs incubated with α-DNABII NTHI 1728 NRel exhibited the most rapid uptake and NETosis activities as well as the greatest visual red fluorescence in approximately half the PMNs in the FOV, with an average of 287 AFU which was significantly greater than that observed with either planktonic or NBfA populations (p ≤ 0.05–0.0001) (Supplementary Figure 1B, panels g–i and Table 2). In addition to images presented within Supplementary Figure 1B, relative PMN activity was documented by video microscopy (Supplementary Videos 7–9).

As our observations were primarily subjective to this point, to objectively detect and quantify relative red fluorescence by an additional methodology, we assessed the relative red fluorescence of PMNs collected after co-incubation with pHrodo™ Red SE-labeled planktonic, NBfA, or α-DNABII NRel NTHI populations via flow cytometry. In all cases, PMNs without bacteria served as the negative control and exhibited peak average red fluorescence of approximately 70 (Figure 2, black histograms). PMNs incubated with planktonically grown NTHI 86-028NP exhibited a slight right shift in peak red fluorescence, with a mean emission of approximately 5 × 10² (Figure 2A, brown histograms), whereas those incubated with NBfA NTHI 86-028NP exhibited a greater right shift with a peak mean of approximately 2 × 10³ (Figure 2A, orange histograms). In keeping with data obtained by timelapse microscopy, the greatest and rightmost shift in red fluorescence was exhibited by PMNs incubated with α-DNABII NTHI NRel, wherein peak average red fluorescence was ∼1x10⁴ (Figure 2A, yellow histograms).

When incubated with planktonically grown NTHI 1128, PMNs exhibited a slight right shift in red fluorescence compared to PMNs without bacteria, with the mean peak at approximately 7 × 10² (Figure 2B, brown histograms). When incubated with NBfA NTHI 1128, PMNs exhibited a mean red fluorescence of approximately 2 × 10⁴ (Figure 2B, orange histograms), and again the greatest right shift in peak average red fluorescence of approximately 3x10⁴ when incubated with α-DNABII NRel NTHI (Figure 2B, yellow histograms).

With the last of three clinical isolates assessed, PMNs incubated with planktonically grown NTHI 1728 exhibited a slight increase in red fluorescence relative to PMNs without bacteria, with an average peak of 2 × 10³ (Figure 2C, brown histograms). When incubated with NBfA NTHI 1728 the average red fluorescence occurred at ∼2 × 10⁴ (Figure 2C, orange histograms). The greatest right shift again occurred with PMNs incubated with α-DNABII NTHI 1728 NRel, wherein a peak average of approximately 4 × 10⁴ was obtained (Figure 2C, yellow histograms). While there was variability in relative placement of red fluorescence peaks, the rightmost shift in red fluorescence was consistently observed when PMNs were co-cultured with α-DNABII NTHI NRel of all three strains tested, which aligned well with what we observed by timelapse microscopy.

From our microscopy assays, in addition to PMN engulfment and NETosis activity, we observed that the relative movement of additional PMNs into the initial FOV, which were selected to ensure approximately the same number of PMNs at time 0 min for each NTHI strain and population tested, appeared distinct amongst the three bacterial populations. To investigate this further, we quantified the number of additional PMNs that migrated into a FOV when incubated with all three populations of the three NTHI strains.

When incubated with NTHI 86-028NP, the number of additional PMNs that entered the FOV within 10 min was: 15 ± 4 for planktonic NTHI, 34 ± 3 with NBfA which were statistically significant compared to planktonic (p = 0.02), and 50 ± 9 with α-DNABII NRel which was also statistically significant (p = 0.0008) (Table 1). For NTHI strain 1128, we observed similar results with the following number of additional PMNs having migrated into the FOV within 10 min when incubated with planktonic, NBfA, or NRel: 6 ± 2, 20 ± 5, and 31 ± 1, respectively, with the latter two values being statistically significant (p = 0.09, p = 0.004, respectively) compared to PMNs incubated with planktonic NTHI 1128 (Table 1). For NTHI 1728, 10 ± 2 PMNs entered the FOV within 10 min with planktonically grown, 22 ± 2 with NBfA, and 31 ± 1 with α-DNABII NRel. Again, these latter two values were statistically significant compared to planktonic (p = 0.003, p = 0.0002, respectively) (Table 1).

We recently reported that α-DNABII NTHI NRel are in lag phase as evidenced by their significantly upregulated expression of three genes canonically associated with this phase of bacterial growth (Wilbanks et al., 2023). As such, we wondered if perhaps characteristics of NTHI in this particular stage of growth might have played a role in the observed differences in relative PMN activity with this population of NTHI. To test this hypothesis, we recapitulated our timelapse microscopy assay and assessed relative PMN activity when incubated with mid-log phase planktonically grown NTHI 86-028NP or when incubated with lag phase planktonically grown NTHI 86-028NP. We observed no differences in PMN uptake or NETosis activity (Figure 3A), nor was there a difference in the average number of additional PMNs that migrated into the FOV (Figure 3B). These findings suggested that it was not only that α-DNABII NTHI NRel were specifically in lag phase which stimulated the greatest observed PMN activity.

We next wondered whether either PMN uptake and NETosis activities, or the differences observed in terms of movement of additional PMNs into the FOV when PMNs were specifically incubated with α-DNABII NTHI NRel, might be due to differences in relative amounts of either bacterial endotoxin or DNA within preparations of the three NTHI populations. Therefore, we assayed for both of these NTHI-associated pathogen-associated molecular patterns (PAMPs) within the fluids in which the PMNs had been exposed, as each has been shown to induce PMN NETosis, as reported by Juneau et al. (2011). When we assessed relative levels of endotoxin, we found an average concentration of approximately 30–40 endotoxin units/mL within each of the planktonic, NBfA, and α-DNABII NTHI NRel populations for all three tested NTHI isolates which suggested that differences in relative endotoxin concentration were not responsible for notable differences observed in PMN activity (Figure 4A). We further determined the relative eDNA levels within the three population preparations for the three tested NTHI strains and found an average concentration of 0.3 ng DNA/μL in each (Figure 4B), which suggested that relative eDNA concentration was also not responsible for differences in observed PMN activity.

We next considered whether the PMNs themselves might be signaling for the observed migration of additional PMNs into the FOV when they were incubated with α-DNABII NTHI NRel. To test this hypothesis, we assessed the media in which PMNs and NTHI populations were incubated for the presence of an array of human cytokines or chemokines after 10 min of co-incubation. We found no notable differences or trends in levels of IL-12p70, TNF, IL-10, or IL-6, regardless of the NTHI strain or population (Figures 5A–C). We also assessed levels of additional cytokines known to be both pro-inflammatory and neutrophil recruiters or activators: IL-1α, IL-1β, IFN-γ, and IFN-α (Chan et al., 2021). We observed increased levels of these latter cytokines, however, there were no observed trends in relative concentrations of these additional cytokines amongst the three NTHI populations of any of the three NTHI strains (Figures 5A–C). In contrast, levels of IL-8 were significantly increased in cultures of PMNs incubated with NBfA (average of 2.0–3.5 pg IL-8/mL) relative to those incubated with planktonic NTHI (average of 1.0–2.0 pg IL-8/mL), regardless of the NTHI strain assessed (p < 0.05–0.001) (Figures 5A–C). These results helped to validate our earlier timelapse microscopy-based observations in which we reported that significantly more PMNs entered the FOV when incubated with NBfA relative to those incubated with planktonically grown NTHI of all three strains assessed (see Table 1). Further, levels of IL-8 were also significantly increased and, in fact the greatest overall, when PMNs were incubated with α-DNABII NTHI NRel (average of 2.5–4.0 pg IL-8/mL) relative to those incubated with planktonic NTHI (p < 0.01–0.0001), regardless of the NTHI strain, which also aligned well with our microscopy and quantification of additional PMNs that migrated into the FOV datasets. Further, significantly more PMNs entered the FOV when incubated with α-DNABII NTHI 86-028NP or 1728 NRel relative to when incubated with NBfA of either strain (p < 0.01 and 0.05, respectively) (Figure 5A–C). We did not observe similar significant differences in IL-8 levels between α-DNABII NTHI 1128 NRel and NBfA (p = 0.36) (Figure 5B), which was similarly corroborated by our video microscopy data and quantification of additional migrated PMNs (Table 1).