Influenza virus induces bacterial and nonbacterial otitis media

Abstract

Otitis media (OM) is one of the most common childhood diseases. OM can arise when a viral infection enables bacteria to disseminate from the nasopharynx to the middle ear. Here, we provide the first infant murine model for disease. Mice coinfected with Streptococcus pneumoniae and influenza virus had high bacterial load in the middle ear, middle ear inflammation, and hearing loss. In contrast, mice colonized with S. pneumoniae alone had significantly less bacteria in the ear, minimal hearing loss, and no inflammation. Of interest, infection with influenza virus alone also caused some middle ear inflammation and hearing loss. Overall, this study provides a clinically relevant and easily accessible animal model to study the pathogenesis and prevention of OM. Moreover, we provide, to our knowledge, the first evidence that influenza virus alone causes middle ear inflammation in infant mice. This inflammation may then play an important role in the development of bacterial OM.

Figure 1.

Influenza virus facilitates pneumococcal otitis media (OM). A, Representative (n ≥ 21 per treatment group) in vivo images of Streptococcus pneumoniae (SP) EF3030^Lux in the middle ears of mice 6 days after intranasal (i.n.) infection with influenza A virus (IAV) or mock. B, Titers of S. pneumoniae EF3030^Lux (SP) in the middle ears of mice after i.n. infection with IAV or mock. Each symbol indicates the bacterial titers for a specific mouse shown in A. C, Titers of S. pneumoniae ML-2302 (SP) in the middle ears of mice after i.n. infection with IAV or mock. Data are pooled from at least 2 independent experiments, and bacterial counts are represented as the mean titer derived from the left and right ear of each mouse. Triple asterisks indicate a highly statistically significant difference (P < .001) between treatment groups using a 2-tailed Mann–Whitney U test. Double asterisks indicate a strongly statistically significant difference (P < .01) between treatment groups using a 2-tailed Mann–Whitney U test. The geometric mean is represented by a horizontal line, and dashed lines indicate the detection limit of the assay.

Figure 2.

Pneumococcal-influenza otitis media (OM) resolves overtime. Representative in vivo images (n ≥ 14 per time point) of mice colonized with Streptococcus pneumoniae EF3030^Lux at various days (d) after IAV infection. The arbitrary letter assigned to each mouse (Ms) is shown.

Figure 3.

Middle ear inflammation in infected mice. Middle ear inflammation in mock- or Streptococcus pneumoniae (SP)–colonized mice 6 days after infection with influenza A virus (IAV) or mock. A, Representative images of middle ear paraffin sections stained with hematoxylin and eosin. Images were taken at 40× magnification. The Eustachian tube (ET), tympanic membrane (TM), middle ear cavity (MEC), and cochlear (Co) are labeled. Insert panels show the inflammatory cell infiltrate of ”SP+ IAV” and ”Mock+ IAV” mice at increased magnification (100×). B, Scoring of middle ear inflammation observed by histology 6 days after infection with IAV or mock. Data are pooled from at least 2 independent experiments, and each data point represents a single ear from an infected mouse (12 mice treated with S. pneumoniae alone, 11 mock-treated mice, 17 mice treated with IAV alone, and 16 mice treated with S. pneumoniae and IAV). Triple asterisks indicate a highly statistically significant difference (P < .001) between treatment groups using a Kruskall Wallis test accompanied by Dunn's posttest. The arithmetic mean is represented by a horizontal line. ”Mock + Mock” and ”SP + Mock” images shown in A were given a score of 0. The ”SP + IAV” image in A was given a score of 75, and the ”Mock + IAV” image was given a score of 20.

Figure 4.

Infection with influenza virus alone causes middle ear inflammation. A, Scoring of middle ear inflammation in germ-free mice 6 days after infection with influenza A virus (IAV) or mock. Each data point represents either the left or right ear of an infected mouse (6 mock-treated mice and 8 mice treated with IAV alone). The asterisk indicates a statistically significant difference (P < .05) between treatment groups using a 2-tailed Mann–Whitney U test. The arithmetic mean is represented by a horizontal line. B, Fluorescent in situ hybridization (FISH) analysis of 16S ribosomal RNA (red) in the middle ear. Sections were obtained from SPF-housed C57BL/6 mice infected with IAV or mock. Images were taken at 40× magnification, and cell nuclei are shown in blue. Arrows indicate the presence of 16S ribosomal RNA. In each image, the middle ear cavity (MEC) is labeled. C, Representative immunofluorescent images (n = 5) of the middle ear of a mouse infected with S. pneumoniae (SP) EF3030^lux and IAV. Sections were stained with universal rabbit negative control (Control-Ab) or with S. pneumoniae serotype 19F rabbit antisera (αSP-Ab), followed by a FITC-conjugated secondary anti-rabbit antibody. The MEC is shown. Pneumococci in the MEC are predominantly localized around the inflammatory cell infiltrate. Images were taken at 60× magnification.

Figure 5.

Coinfected and IAV-infected mice display hearing loss. Hearing loss in mock- or Streptococcus pneumoniae (SP)–colonised mice 6 days after infection with influenza A virus (IAV) or mock. Data are pooled from at least 2 independent experiments, and each data point represents the hearing threshold increase of either the left or right ear of an infected mouse (10 mice infected with S. pneumoniae alone, 12 mice infected with S. pneumoniae and IAV, and 17 mice infected with IAV alone). The threshold increase was calculated by subtracting the mean hearing loss of mock-infected, age-matched mice from each data point. Triple asterisks indicate a highly statistically significant difference (P < .001) between treatment groups using a Kruskall-Wallis test accompanied by Dunn's posttest. Single asterisks indicate a statistically significant difference (P < .05) between treatment groups. The arithmetic mean is represented by a horizontal line.