|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study was to determine the virulence of nontypeable Haemophilus influenzae 2019 (NTHi 2019) and its two lipooligosaccharide (LOS) mutant strains, B29 (gene htrB) and DK1 (gene rfaD), and compare their effect on the middle ear, round window membrane, and inner ear.
Fifteen chinchillas were divided into 3 equal groups and their bullas inoculated bilaterally with 0.5 ml of 102 CFU of parent NTHi 2019, B29 or DK1 mutant strains. Two days after inoculation all animals had otitis media and inflamed middle ear mucosa. There was a trend of greater thickness and infiltration of the round window membrane in animals inoculated with the wild-type NTHi strain compared to the mutant strains and a significant increase in both inflammatory cell infiltration and bacteria presence in the scala tympani area of the inner ear. Strial edema was only observed in the wild type-inoculated group.
LOS mutants of NTHi appear to have a reduced ability to pass through the round window membrane resulting in less inner ear inflammation and pathological changes.
Nontypeable Haemophilus influenzae (NTHi) is a leading cause of otitis media in children . Lipooligosaccharide (LOS), the major immunogen of NTHi, has a highly heterogeneous composition between strains and between bacteria within a strain . LOS glycoforms, containing sialic acid and phosphorylcholine, mimic host structures to allow the organism to evade innate defenses and manipulate host cell metabolism [3,4]. NTHi was shown to adhere to and invade bronchial epithelial cells via an interaction of LOS with the PAF (platelet activating factor) receptor, and binding of LOS to the PAF receptor initiates host cell signaling and bacterial invasion . LOS structure affects the ability of bacteria to adhere to and invade cells effectively . DK1, an rfaD gene mutant, expresses a truncated LOS consisting of only three deoxy-D-manno-octulosonic acid residues, a single heptose and lipid A [6,7]. DK1, an htrB gene mutant, is associated with modifications of lipid A and phosphorylation of LOS . The late acylation of the lipid A encoded by the htrB gene is important in bacterial colonization and represents a key step in LOS biosynthesis . Inner ear damage, which can lead to sensorineural hearing loss, is a serious sequelae of otitis media. The round window membrane is the only soft tissue barrier between the middle ear and the inner ear and has been shown to be permeable to a variety of substances. Passage of bacterial products , inflammatory mediators , and intact bacteria  through the round window membrane are considered the likely cause of inner ear damage. DeMaria et al  showed less virulence of B29 and DK1 mutant NTHi strains in the middle ear compared to the wild-type strain. They found the incidence of labyrinthitis by 5 days postinoculation in only 5% of the B29-inoculated animals, and no labyrinthitis in those inoculated with the DK1 mutant strain.
However, detailed analysis of histopathological changes of middle and inner structures caused by the wild-type and mutant NTHi strains, including the effect of LOS mutations on bacterial invasion and recruitment of inflammatory cells in the middle and inner ears, has not been reported. The purpose of this study was to compare middle and inner ear inflammation and pathology in chinchillas following middle-ear inoculation of either the parent NTHi 2019 strain or its two LOS mutant strains, DK1 (gene rfaD) and B29 (gene htrB).
The parental (wild-type) NTHi 2019, an rfaD gene mutant (DK1) with a truncated LOS, consisting of only three deoxy-D-manno-octulosonic acid residues, a single heptose, and lipid A, and an isogenic htrB mutant (B29) with an altered oligosaccharide core and an altered lipid A, were used in this study [7,9] (all NTHi strains were provided by Dr. Michael A. Apicella, University of Iowa College of Medicine). Overnight cultures of NTHi, on chocolate agar strains, were inoculated into liquid brain-heart infusion (Difco Laboratories) supplemented with hemin/NAD (Sigma) at 37°C with 5% CO2. We added 15 mg/ml of kanamycin to the media for the DK1 mutant that contains a kanamycin resistant cassette, and 1.5 μg/ml of chloramphenicol to the media for the B29 mutant that contains a chloramphenicol resistant gene. Bacteria at mid-log phase (O.D 600nm 0.4-0.5) were harvested by centrifugation, washed with phosphate-buffered saline and diluted to desired cell numbers as determined by colony forming units of each strain.
Animals were housed and fed under standard conditions at our institutional animal care facility. Experiments were performed on young (~ 1-year-old) chinchillas weighing, 250-350 g that had normal external auditory canals and tympanic membranes. The care and use of animals was approved by the Institutional Animal Care and Use Committee of the University of Minnesota. All animals were anesthetized prior to intrabullar inoculations with a combination of ketamine (100 mg/kg) and acepromazine (10 mg/kg). Animals were sacrificed by overdose of sodium pentobarbital.
A total of 15 chinchillas were given bilateral intrabullar inoculations of 0.5 ml of 102 CFU of NTHi 2019, B29, or DK1 strains (5 animals for each strain). Two days after inoculation, animals were euthanized, bullas removed, and the cochlea perfused via the apex and oval window with 2% glutaraldehyde in 0.1M phosphate buffer (pH 7.4). Fixation was continued by emersion for 2 hours. Samples were decalcified in 10% EDTA on a rotator in a cold room for 3 days. EDTA was changed daily. Samples were washed in phosphate buffer and post-fixed in 1% OsO4 in phosphate buffer (pH 7.4) for 1 hour. They were washed again in buffer, dehydrated in a graded series of ethanol, followed by propylene oxide, and embedded in epoxy resin. Samples were cut at a thickness of 1 μm and stained with Toluidine blue for light microscopic assessment.
Thickness of the middle ear mucosa was measured at the promontory. Measurements of the thickness of the middle ear mucosa and round window membrane were made at the midpoint of the sample and midway between the midpoint and the edge of the sample on each slide, using a 10 × 10 unit eye piece grid calibrated in units of 0.16 μm. Measurments were averaged. The areas of greatest inflammatory cell infiltration in the middle ear mucosa, round window membrane, and scala tympani were counted within the 10 × 10 unit grid of for each sample. Thickness and inflammatory cell ilnfiltration were assessed at a magnification of 600×. Multiple slides from each ear were averaged and the data presented per ear. Bacteral infiltration of the scala tympani was determined at a magnification of 1000×. All results were expressed as mean ± SE. Standard software SPSS 11 (SPSS Inc, Chicago, IL, USA) was used for statistical analysis. One-way analysis of variance (ANOVA) with LSD post-hoc test was applied for pair-wize multiple comparisons between the treatment groups. Differences were considered to be significant if P ≤ 0.05.
Fourty-eight hours after inoculation, all of the animals inoculated with wild-type and two mutant NTHi strains, B29 and DK1, had middle ear effusions. The middle ear mucosa of animals inoculated with all three NTHi strains showed infiltration of inflammatory cells (predominantly polymorphonuclear leukocytes), mucosal thickening due to inflammatory cell infiltration in the subepithelial space, vascular dilation, and hemorrhage. Fig. 1,a-c shows examples of the most severe inflammation of the middle ear mucosa observed in the B29, DK1, and wild-type NTHi-inoculated groups. Inflammatory cell infiltration of the round window membranes (Fig. 2,a-c) was observed in all three groups. Fig. 3,a-c represents examples of the most severe inflammatory and pathological changes of the inner ear observed in each group. In some animals, inflammatory cell infiltration and bacterial invasion in the inner ear was accompanied by changes of cochlear structures. There was hair cell loss or damage, and severe strial edema in two ears of the wild-type NTHi-inoculated animals. One ear in the DK1-inoculated group had distortion of both inner and outer hair cells; however, there was no loss of hair cells and no strial edema. No damage to the inner ear was observed in the B29-inoculated group. Bacterial cells of the wild-type and both mutant strains, B29 and DK1, could be seen to pass into the scala tympani area of the inner ear. Figure 4 shows an example of bacterial passage through all three layers of the round window membrane in a DK1-inoculated animal. Bacteria were seen inside cells and outside of cells in wild-type and both mutant groups. We also observed that in some cases, polymorphonuclear leukocytes replaced the epithelial layer of the round window membrane. In some animals inoculated with the wild-type NTHi strain, bacteria and inflammatory cells were seen in the organ of Corti. The number of ears with bacteria in the scala tympani was greater in the wild-type NTHi-inoculated group compared to the mutant-inoculated groups (Table 1).
We performed one-way ANOVA statistical analysis to compare the number of inflammatory cells, predominantly polymorphonuclear leukocytes (PMNs), in the middle ear mucosa, round window membrane, and scala tympani, and the thickness of the middle ear mucosa and round window membrane, as well as bacterial presence in the scala tympani area of the inner ear, adjacent to the round window membrane, for animals infected with all three NTHi strains. These data are summarized in the Table 1. Although the mean numbers (Table 1) and ANOVA showed a trend of a greater number of inflammatory cells and an increased thickness of the round window membrane in the wild-type NTHi-inoculated animals compared to the mutant-inoculated animal groups, P-values were higher than 0.05. There was no significant difference between the number of inflammatory cells, in the scala tympani, in animals inoculated with B29 and DK1 strains (P = 0.873); however, significant differences were found between animal groups inoculated with the parent NTHi strain and each of its isogenic mutants (P = 0.032 for B29 mutant and P = 0.044 for DK1 mutant).
Streptococcus pneumoniae and nontypeable Haemophilus influenzae (NTHi) are two of the major pathogens of acute otitis media. Current polysaccharide-based vaccines against Streptococcus pneumoniae contain the most common serotypes causing pneumococcal infection. While the incidence of pneumococcal otitis media decreased since the introduction of pneumococcal conjugate vaccines, the incidence of otitis media attributed to NTHi has been on the rise [14,15]. Progress in developing a vaccine for NTHi has been hampered by the antigenetic heterogeneity of surface molecules among strains. The pathology from NTHi infection is the result of host response to bacterial factors, including lipooligosaccharide (LOS), a major antigenic bacterial component. LOS of NTHi has been shown to elicit many of the histopathological changes observed during experimental NTHi-induced otitis media .
We studied two LOS mutants of NTHi, DK1 and B29, and compared them to the wild-type strain in an effort to better understand the role of LOS components in the pathology of otitis media and inner ear complications secondary to otitis media. DK1, an rfaD gene mutant, expresses a truncated LOS consisting of only three deoxy-D-manno-octulosonic acid residues, a single heptose and lipid A [6,7]. B29, an isogenic htrB mutant, possesses an altered oligosaccharide core and an altered lipid A . Our findings demonstrate, that both B29 and DK1, like the wild-type strain bacteria, were able to cause otitis media and inflammation of the middle ear mucosa; however, they were significantly less likely to pass through the round window membrane into the inner ear or cause inflammatory cell infiltration in the inner ear.
We cannot comment on the persistence of the mutants to colonize the middle and inner ear, because the survival time of our animals was of short duration. Although, approximately, the same number of colony forming units (CFUs) of each strain were inoculated, the relative fitness of these strains and their ability to multiply within the middle ear space, which could be bacterial intrinsic and/or also a consequence of host immune response, were not known.
Our study showed that NTHi LOS mutants caused less severe inflammatory and pathological changes to the inner ear than the wild-type strain, providing histological evidence that LOS plays a role in inner ear inflammation and pathology. The presence of bacterial and inflammatory cells in the scala tympani was significantly greater and resulted in more pathological damage of the cochlea in animals inoculated with the wild-type strain than in the B29 or DK1 mutant-inoculated animals. Most of the endotoxic activity of LOS is attributed to lipid A, and mutation of the htrB gene in NTHi B29 is associated with modifications of lipid A and phosphorylation of LOS . Interestingly, we did not find any significant differences between the DK1-inoculated and the B29-inoculated animal groups.
Studies of other mutations involved in LOS biosynthesis or their combination may be necessary to determine those bacterial components that can be used for therapeutic targeting of NTHi infection.
This work supported in part by: NIDCD R01 DC006452, The International Hearing Foundation, and The Starkey Foundation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.