The 3 h cultures were pelleted by centrifugation, washed in phosphate buffered saline (PBS) containing 0.1% w/v gelatin, and resuspended to an optical density of 0.5 at 605 nm in the same buffer. The bacterial suspension was diluted by adding 1.0 mL into 5.0 mL of PBS containing 0.1% gelatin and was used to inoculate media for growth curves (approximate initial concentration of 200,000 cfu signaling pathway ml-1). In vitro competition studies were performed by mixing equal numbers of the wild type and mutant strains (starting total of approximately

2 × 105 cfu ml-1) in 50 ml of either sBHI or hdBHI supplemented with limiting concentrations of hemoglobin (5 μg ml-1. Bacterial counts were determined for the duration

of the 28 hour experiment by plating samples using the track find more dilution method, as previously described [38], on sBHI or sBHI containing spectinomycin to allow enumeration of both strains. Chinchilla model of otitis media Adult chinchillas (Chinchilla lanigera) with no signs of middle ear infection by either otoscopy or tympanometry at the beginning of the study were used. Animals were allowed to acclimate to the vivarium for at least 14 days prior to transbullar challenge. Animal procedures have been previously described in detail [39–41]. Two separate experiments, one to assess virulence and a second to assess competitive fitness, were performed in the chinchillas. In the first experiment to compare virulence, two groups of 5 animals were challenged heptaminol in both ears by transbullar injection with approximately 2,000 cfu of either strain 86-028NP or its hfq deletion mutant HI2207. Transbullar inocula were delivered in 300 μl 0.1% gelatin in PBS by direct injection into the superior bullae. Actual bacterial doses were confirmed by plate count.

On days 4, 7, 11, and 14 post-challenge middle ear effusions (MEE) were collected by epitympanic tap as previously described [29]. Bacterial titers in recovered MEE were determined using the track dilution method. In the second experiment, to assess competitive fitness, five animals were challenged in both ears transbullarly with a mixture containing equal numbers of 86-028NP and its hfq deletion mutant HI2207 (total of approximately 2,000 cfu). Epitympanic taps were performed on all ears on days 4, 7, 11, and 14 after nontypeable H. influenzae challenge. Recovered MEE were plated on sBHI and sBHI containing spectinomycin in order to determine the total bacterial titer and the titer of the mutant strain respectively. Rat model of bacteremia The infant rat model for hematogeneous meningitis following intraperitoneal infection with H. influenzae[42] was used to compare the abilities of strains R2866 and the ∆hfq mutant, HI2206, to cause bacteremia. Again two experiments were performed, one to assess virulence and a second to assess competitive fitness.