Contribution of Moraxella catarrhalis Type IV Pili to Nasopharyngeal Colonization and Biofilm Formation

Bacterial strains and culture methods. M. catarrhalis strains 7169 and 7169pilAK4, the isogenic pilA-deficient mutant of 7169 lacking surface TFP expression, have been described previously (25, 26). M. catarrhalis strains were cultured at 37°C on brain heart infusion (BHI) agar or Mueller-Hinton (MH) agar in a humidified 5% CO₂ atmosphere; 7169pilAK4 was cultured with the addition of kanamycin (25 μg per ml) as necessary. BHI and MH broth cultures were grown in a 37°C shaking water bath with constant agitation.

Tissue culture adherence assays.Chang epithelial cells (human conjunctiva clone 1-5c-4, Wong-Kilbourne derivative; ATCC CCL20.2) were cultured in RPMI 1640 medium with l-glutamine plus 10% heat-inactivated fetal bovine serum at 37°C in a humidified 5% CO₂-95% air atmosphere essentially as described previously (31). The cells were seeded into 96-well plates at 10⁴ cells per well and grown to approximately 90% confluence. M. catarrhalis cells grown to early log phase (optical density at 600 nm = 0.2 to 0.3) in MH broth were resuspended in preequilibrated cell culture medium and inoculated in triplicate onto confluent monolayers at a multiplicity of infection (MOI) of 100:1. After the addition of the bacteria, the plates were centrifuged at 500 × g for 5 min and placed at 37°C for 30 min. For enumeration of cell-associated bacteria, each well was subjected to three washes with Dulbecco's phosphate-buffered saline (pH 7.2) to remove nonadherent bacteria, and the monolayers were detached from the wells by treatment with 0.25% trypsin-1 mM EDTA. The contents of each set of triplicate wells were pooled, serially diluted, and plated in triplicate onto MH plates for the determination of colony counts at 24 h. For each adherence assay, at least two sets of triplicate wells were inoculated per strain. Serial dilutions of controls, including wells inoculated with bacteria in the absence of eukaryotic cells to assess whether the organisms exhibited any appreciable growth or adhered to the plastic during the assay, as well as the inoculum stock, were also plated in triplicate and cultured overnight as described above. The number of recovered bacteria was compared to the number of inoculated organisms per well, and the percentage (± standard error) of adherent bacteria was determined. For the duration of the assay, no growth or adherence to the plastic well was observed. Statistical significance was determined using Student's unpaired t test on three independent experiments performed on separate days.

Animal colonization model.Healthy adult chinchillas (Chinchilla lanigera) were purchased from Rauscher's Chinchilla Ranch (LaRue, OH) and provided with food and water ad libitum. All chinchilla studies were performed using protocols approved by the Columbus Children's Research Institute Animal Care and Use Committee. Cohorts of chinchillas were challenged intranasally by passive inhalation of approximately 10⁷ CFU per animal of M. catarrhalis strain 7169 or its pilA mutant, 7169pilAK4, suspended in sterile pyrogen-free saline. The exact inoculum density was confirmed by plate counts. Nasopharyngeal lavage was performed on days 1, 4, 7, 10, 14, and 18 postchallenge to determine the relative abilities of these isolates to colonize the nasopharynx (NP) of this host. Nasopharyngeal lavage was performed by passive inhalation of 500 μl sterile saline followed by recovery of the lavage fluid as described previously (4, 27). To evaluate the adherent population of bacteria, the chinchillas were sacrificed on day 18 postinoculation, and the nasopharyngeal tissues were harvested, weighed, and homogenized. Lavage samples and homogenates were dilution plated on chocolate agar supplemented with vancomycin (10 mg/liter) and trimethoprim (5 mg/liter) to suppress normal chinchilla flora; kanamycin was added to the culture medium when isolating the pilA mutant. Assays were performed using cohorts of three chinchillas per strain on at least two separate occasions. Statistical analyses were conducted by the Biometrics Laboratory at The Ohio State University College of Medicine, using the exact Wilcoxon rank sum test to compare relevant groups. Differences were considered significant if the P value was <0.05.

Biofilm growth in continuous-flow chambers. M. catarrhalis biofilms were grown in continuous-flow chambers, using BHI broth diluted 1:10 in sterile phosphate-buffered saline (13, 17). To compare the ability of the wild-type strain to that of its pilA mutant to form a biofilm in a flow chamber in vitro, 18-hour plate-grown colonies were suspended in sterile saline to obtain an optical density at 490 nm of 0.60. Five hundred microliters of this bacterial suspension was inoculated into a flow chamber (Stovall Flow Cell, Greensboro, NC) and allowed to incubate at 37°C for 1 h (with the coverslip down) to promote adherence. Following this incubation period, the chamber was inverted (coverslip up), and the diluted BHI broth was pumped through the chamber at a flow rate of 180 μl per min for 72 h. The flow cell chambers and medium were maintained at 37°C until the end of the experiment, at which time the biofilms were stained with a Live/Dead BacLight bacterial viability kit (Molecular Probes, Eugene, OR) by injecting 0.5 ml of stain solution into each chamber. Following a 15-minute incubation in the dark prior to being viewed, z-stack composite fluorescent images were captured using a 63× objective on a confocal microscope (510 Meta; Zeiss, Thornwood, NY). The digital three-dimensional z-stack images were generated using Zeiss LSM image browser software.

Contribution of TFP to in vitro adherence.We previously demonstrated that M. catarrhalis clinical isolates constitutively express TFP and that TFP expression by this organism is absolutely essential for natural transformation (26). In this study, we evaluated the contribution of TFP expression to the pathogenesis of M. catarrhalis. A comparative adherence assay using wild-type M. catarrhalis strain 7169 and the PilA-deficient isogenic mutant 7169pilAK4, which lacks surface pili, was performed to evaluate if expression of TFP by M. catarrhalis contributes to the ability of the organism to adhere to Chang conjunctival epithelial cells. 7169pilAK4 was significantly impaired in adherence to Chang cells (P = 0.0054), as only 11.15% ± 1.59% of the bacteria with the pilA mutation were able to adhere to the monolayers, compared to 23.54% ± 3.12% of the wild-type bacteria (Fig. 1). These data suggest a role for TFP in the adherence of M. catarrhalis in vitro.

• Open in new tab
• Download powerpoint

FIG. 1.

Comparison of adherence levels of strain 7169 (WT 7169) and the isogenic pilA-deficient mutant 7169pilAK4 to monolayers of Chang cells. The percent adherence represents the portion of the inoculum that was cell associated at 30 min. Values represent means ± standard deviations.

M. catarrhalis colonizes the chinchilla NP.In hopes of identifying a model system that could be used to evaluate bacterial colonization and pathogenesis under conditions that more closely mimic human disease, we investigated the chinchilla model of nasopharyngeal colonization. The chinchilla model has been effective in evaluating the respiratory mucosal pathogens nontypeable Haemophilus influenzae and Streptococcus pneumoniae, which share a similar ecological niche with M. catarrhalis. We first performed a pilot study to investigate the ability of strain 7169 to colonize and persist in the chinchilla NP. Following intranasal inoculation with approximately 10⁷ CFU of M. catarrhalis, the numbers of recoverable bacteria diminished through day 4, and then a stable level of colonization (approximately 10² CFU per ml of NP lavage fluid) was achieved and maintained for a minimum of 2 weeks (Fig. 2).

• Open in new tab
• Download powerpoint

FIG. 2.

Dynamics of nasopharyngeal colonization of the chinchilla following intranasal inoculation with M. catarrhalis 7169. The mean numbers of CFU per milliliter of lavage fluid ± standard deviations (error bars) for five chinchillas are depicted.

Contribution of TFP to nasopharyngeal colonization.Once we had established that M. catarrhalis readily colonizes the NP of the chinchilla, we investigated the contribution of TFP expression to this process. M. catarrhalis strains 7169 and 7169pilAK4 were introduced into separate cohorts of chinchillas via passive inhalation. The animals in each group were subjected to nasopharyngeal lavage at defined time intervals and were sacrificed on day 18. The nasopharyngeal tissues from these chinchillas were harvested, and homogenates were analyzed for the presence of M. catarrhalis.

On days 1, 4, and 7 following intranasal challenge, more of the pilA-deficient mutant than the parent was recovered via nasopharyngeal lavage (Fig. 3A). Importantly, this difference was highly significant (P = 0.004) 1 day after challenge, a time point when M. catarrhalis predominated as a nearly pure culture in the lavage fluids from chinchillas (data not shown). In fact, 7169pilAK4 was more readily harvested from the NP than the wild type was until day 4 postchallenge; after this time point, no difference in colonization levels was observed, indicating that the lack of TFP did not affect the ability of this organism to survive in the chinchilla host. Conversely, the homogenized mucosae from the nasoturbinates, ethmoid turbinates, and NP of the challenged chinchillas showed that significantly more wild-type than pilA mutant cells were recovered (P = 0.017, 0.026, and 0.004, respectively) (Fig. 3B). Thus, the pilA mutant predominated in terms of the population of M. catarrhalis in the chinchilla NP obtained by lavage, whereas 7169 predominated in terms of the adherent population. Collectively, these data suggest that expression of TFP by M. catarrhalis confers an advantage for the colonization of airway mucosa in this host and perhaps also stabilizes interbacterial interactions, thus reducing the ability of bacteria to be recovered from the NP by gentle lavage.

• Open in new tab
• Download powerpoint

FIG. 3.

Evaluation of NP colonization dynamics for strains 7169 (open squares) and 7169pilAK4 (closed squares) by culture of NP lavage fluids (A) and mucosal tissue homogenates (B). Each datum point represents CFU obtained from an individual chinchilla, and bars represent means. Comparative P values of <0.05 were considered significant.

The wild-type strain that expressed TFP exhibited 7.67- ± 2.5-fold and 9.50- ± 3.4-fold (means ± standard deviations) increased recovery from the NP and the nasoturbinates, respectively, in these animals compared to the mutant lacking PilA expression, whereas for the ethmoid turbinates the increased recovery of the parental isolate was only 2.56- ± 0.1-fold. These results are likely due to the specific nature of the mucosal epithelium present within each of these three anatomical microenvironments. Within the NP of the chinchilla host, the mucosa consists primarily of ciliated columnar epithelium with some goblet cells, and these cell types are also present in the nasoturbinates, which contain predominantly squamous or nonciliated columnar epithelium, with some ciliated columnar cells and a few goblet cells. In contrast, the mucosa of the ethmoid turbinates is principally olfactory epithelium with a few ciliated columnar cells (18). Collectively, these data suggested that M. catarrhalis TFP likely adheres to a receptor(s) expressed by the classic respiratory epithelium, as opposed to those present on olfactory epithelial cells. Moreover, our data suggested that the chinchilla may be an effective rodent host for studying the early steps in M. catarrhalis colonization and pathogenesis and indicated that expression of TFP contributes to bacterial persistence in the upper respiratory tract mucosa in vivo.

Role of M. catarrhalis TFP in biofilm formation.It has been documented that bacterial biofilms formed during chronic human infections, including recurrent otitis media, are often less susceptible to eradication by antimicrobial agents (9, 11, 22). Moreover, TFP have been shown to be important for the initial attachment and microcolony formation at the onset of biofilm formation by several gram-negative bacteria (20, 32-34, 43). To investigate if M. catarrhalis TFP contribute to biofilm formation, we compared the abilities of 7169 and 7169pilAK4 to form biofilms in continuous-flow chambers.

We directly visualized the biofilm formation phenotype of the TFP-defective strain 7169pilAK4 compared to that of its parent strain, 7169, by using light microscopy. As shown in Fig. 4, the difference between the ability of the parent to adhere to the glass slide and that of the pilA mutant was evident even at the earliest time point (1 h). Throughout the monitored time course, aggregates of 7169 formed into characteristic microcolonies that became increasingly dense and exhibited three-dimensional pillar formation. In contrast, 7169pilAK4 appeared to have defects in initial cell attachment and subsequently exhibited delayed microcolony formation and diminished three-dimensional expansion.

• Open in new tab
• Download powerpoint

FIG. 4.

Direct visualization of biofilm formation in a continuous-flow cell chamber over time by strains 7169 (A) and 7169pilAK (B). Representative images of each strain were taken from the center of the flow cell placed on a heated stage for nondestructive image acquisition by phase-contrast microscopy at a magnification of ×40.

In 3-day-old M. catarrhalis biofilms, multicellular mushroom-shaped structures separated by well-demarcated water-filled channels were clearly visible (Fig. 5). It was previously shown that in Pseudomonas aeruginosa biofilms, the development of the prototypical biofilm structures of mushroom-like stalks, subsequently followed by the formation of the mushroom caps, is a process dependent on functional TFP (19, 20). 7169pilAK4 exhibited less pronounced biofilms with an average depth of 72 μm, in contrast to the wild type, which produced more abundant biofilms with an average thickness of 104 μm. These data suggest a contributory role for M. catarrhalis TFP in biofilm development and maturation.

• Open in new tab
• Download powerpoint

FIG. 5.

Representative biofilms formed by strains 7169 (A and C) and 7169pilAK4 (B and D) grown in flow chambers for 72 h and examined by confocal laser scanning microscopy after being stained with a BacLight Live/Dead kit. Panels A and B show three-dimensional z-stacked composites, while panels C and D illustrate the comparative height difference of each biofilm. Average biofilm depths are shown.