Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Arch Otolaryngol Head Neck Surg. Author manuscript; available in PMC 2010 July 29.
Published in final edited form as:
PMCID: PMC2912141

Gene Expression Differences in Infected and Noninfected Middle Ear Complementary DNA Libraries



To investigate genetic differences in middle ear mucosa (MEM) with nontypeable Haemophilus influenzae (NTHi) infection. Genetic upregulation and downregulation occurs in MEM during otitis media (OM) pathogenesis. A comprehensive assessment of these genetic differences using the techniques of complementary DNA (cDNA) library creation has not been performed.


The cDNA libraries were constructed from NTHi-infected and noninfected chinchilla MEM. Random clones were picked, sequenced bidirectionally, and submitted to the National Center for Biotechnology Information (NCBI) Expressed Sequence Tags database, where they were assigned accession numbers. These numbers were used with the basic local alignment search tool (BLAST) to align clones against the nonredundant nucleotide database at NCBI.


Analysis with the Web-based statistical program FatiGO identified several biological processes with significant differences in numbers of represented genes. Processes involved in immune, stress, and wound responses were more prevalent in the NTHi-infected library. S100 calcium-binding protein A9 (S100A9); secretory leukoprotease inhibitor (SLPI); β2-microglobulin (B2M); ferritin, heavy-chain polypeptide 1 (FTH1); and S100 calcium-binding protein A8 (S100A8) were expressed at significantly higher levels in the NTHi-infected library. Calcium-binding proteins S100A9 and S100A8 serve as markers for inflammation and have antibacterial effects. Secretory leukoprotease inhibitor is an antibacterial protein that inhibits stimuli-induced MUC1, MUC2, and MUC5AC production.


A number of genes demonstrate changes during the pathogenesis of OM, including SLPI, which has an impact on mucin gene expression; this expression is known to be an important regulator in OM. The techniques described herein provide a framework for future investigations to more thoroughly understand molecular changes in the middle ear, which will likely be important in developing new therapeutic and intervention strategies.

Otitis Media (OM) is a common pediatric illness that requires substantial health care expenditures in treatment. There are an estimated 5 million annual episodes at a cost of $3 to $6 billion per year in the United States.1,2 Approximately 5% to 10% of acute OM cases progress to chronic OM with effusion (OME), which is a leading cause of hearing loss in children.3

Despite the considerable potential for morbidity from OM and increasing challenges with antimicrobial resistance, there still exist major knowledge deficits with respect to the pathogenesis of this disease, particularly on a molecular and genetic level. Many genes that are upregulated and downregulated in the middle ear mucosa (MEM) during OM have yet to be examined in detail. Such gene expression profiles will likely provide insights into the molecular and cellular changes that occur during OM and may provide avenues to novel therapeutic treatments.

Cytokines and their role in OM have been studied extensively.46 Despite these advances, less is known about their expression on a molecular scale. Tumor necrosis factor (TNF), interleukin 1β (IL-1β), and other proinflammatory cytokines are detected at high levels in middle ear (ME) effusions and likely play an important role in ME inflammation and the pathologic development of OM.7,8 Tumor necrosis factor is a proinflammatory cytokine, produced primarily by macrophages in response to bacterial or viral infection, that has antimicrobial effects, stimulates additional cytokine production, and enhances fibroblast proliferation.9,10 Interleukin 1β is produced by activated macrophages, is also an important mediator of the inflammatory response, and is involved in a variety of cellular activities, including cell proliferation, differentiation, and apoptosis.7

Mucins are a family of large, heavily glycosylated proteins that have been shown to be important in the pathophysiologic development of OM.11,12 Nineteen human mucin genes have been described, and their expression pattern in MEM has recently been characterized by our laboratory.13 Middle ear epithelial (MEE) mucins can result in ME effusions that are highly viscous, hindering normal mucociliary clearance1416 and subsequently resulting in OME and hearing loss. However, mucins also play an important protective role in MEM by providing a protective barrier to reduce pathogen invasion, as well as aiding in ME clearance of pathogens and antigen presentation to host immune cells.1722 Tumor necrosis factor and IL-1β have been shown to mediate a differential expression and upregulation of mucin gene 1 (MUC1), mucin gene 2 (MUC2), mucin gene 4 (MUC4), and mucin gene 5AC (MUC5AC) in chinchilla MEE.23

Nontypeable Haemophilus influenzae (NTHi) is one of the leading causative agents of bacterial OM, and in this investigation we examined the impact of this pathogen on gene expression in the chinchilla MEM using complementary DNA (cDNA) library techniques. The chinchilla has historically been the animal model of choice for studying OM because it is not naturally susceptible to the disease and has favorable anatomy and genetic similarity to humans.3,2325 However, gene sequence information for this species is notably lacking owing to relatively limited applications of this animal model for other areas of research.

With the explosion of sequence data in the past decade, the need for common terminology has become a necessity. The Gene Ontology (GO) project was created in an effort to construct a consistent vocabulary that describes genes and gene products across a wide range of different databases ( Three structured vocabularies (ontologies) are used to describe gene products in terms of their associated biological processes, cellular components, and molecular functions in a species-independent manner. FatiGO is a free, downloadable Web-based application ( that is used to examine relevant GO terms for a group of genes with respect to a set of genes of reference. Relevance is determined by the application of a Fisher exact test that takes into consideration the multiple-testing nature of the statistical contrast performed.27 Using FatiGO, an investigator is able to compare differential expression of genes, categorized by biological processes, from 2 or more different cDNA libraries, and these techniques were employed in this investigation.



The chinchilla model of acute OM has been well described in numerous laboratories, including those of the current investigative team, and was used for all of the current investigations.3,2325,28,29 The MEM tissue specimens used in this study were obtained from young adult (6–10 months old), mixed-breed chinchillas weighing 400 to 600 g that were obtained free of ME disease as culls from the fur industry (McClenahan Chinchilla Ranch, New Wilmington, Pennsylvania). Animals were treated in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act; the animal use protocol was approved by the Institutional Animal Care and Use Committee of the Allegheny-Singer Research Institute. A total of 12 chinchillas, 6 for each of 2 clinical conditions, were used to perform the experiments described.

Prior to chinchilla ME inoculation, or MEM harvest in the case of the uninfected cohort, all ears were examined by pneumatic otoscopy. Abnormal findings on evaluation would have required elimination of the animal from the study. One cohort of animals was used to harvest healthy MEMs, and the second cohort was infected with a low-passage clinical isolate (PittDD) of NTHi. The animals in the normal group were evaluated and killed immediately after their appropriate acclimation period. Prior to tissue harvest, the chinchillas were anesthetized via intramuscular injection of 0.1 mL of a solution of ketamine hydrochloride, 100 mg/mL, xylazine hydrochloride, 30 mg/mL, and acepromazine acetate, 5 mg/mL. Deep anesthesia was confirmed by the abolishment of the eye-blink reflex, followed by euthanasia via intracardiac injection of pentobarbital sodium, 2 g (Abbott Laboratories, North Chicago, Illinois) as approved by the Panel on Euthanasia of the American Veterinary Medical Association. The temporal bone, including tympanic membrane and ME cavity, was removed bilaterally. The tympanic membrane and ME cavity were examined closely to ensure there was no evidence of inflammation or infection. Animals in the NTHi group were inoculated 3 days prior to MEM harvest. This 3-day time frame has been employed in previous investigations and was used to allow for a robust episode of OM to develop prior to harvest. Anesthesia was induced as described followed by bilateral transpolar injection of 0.1 mL of a 105 colony-forming units/mL–NTHi suspension using a 0.5-in, 27-gauge needle attached to a 1-mL syringe.


PittDD is a low-passage, ampicillin-sensitive NTHi isolate that was obtained from a child with OME at Children’s Hospital of Pittsburgh (Pennsylvania). Initially, a culture of PittDD was grown on chocolate agar and then subcultured once in brain heart infusion broth (Becton Dickinson, Sparks, Maryland) supplemented with hemin, 10 µg/mL (Fisher Scientific, Pittsburgh), nicotinamide adenine dinucleotide, 2 µg/mL (Sigma, St Louis, Missouri), and thiamine hydrochloride, 20 µg/mL (Sigma) and grown at 37°C in a humidified 5% carbon dioxide atmosphere prior to the preparation of frozen stock. For all subsequent studies, 1 part of the initial frozen stock was thawed and used for culture.


For both the NTHi-infected and noninfected chinchilla cDNA libraries, RNA was extracted from pooled tissue. Blunt-ended cDNA was made using the Clone tech Super Smart PCR (polymerase chain reaction)cDNA Synthesis Kit (Mountain View, California) and amplified by long-distance PCR. The double-stranded cDNA samples were treated with Taq polymerase to add A overhangs and the vector TA cloned into the pCR II vector using the Dinitrogen TA cloning kit (Carlsbad, California). The library was then transformed into competent bacteria and propagated under antibiotic selection. The culture was used to prepare frozen stock that was in 10% glycerol and stored at −80°C.

Library stocks from the NTHi-infected and noninfected chinchilla cDNA libraries were plated at a density of approximately 100 to 200 well-isolated colonies per plate. To characterize these libraries, 1100 to 1300 colonies were picked at random and transferred to 96-well plates containing 200 L of lauria broth (100 µg per l mL of ampicillin) in each well and incubated, shaking at 37°C, overnight. Two microliters of the overnight culture was transferred to a PCR plate to be sequenced. The remaining 198 µL was mixed with 50 µL of 50% glycerol solution, sealed, and stored at −80°C.

Clones were PCR amplified and prepared for sequencing using the BigDye Primer Cycle Sequencing Kit (Applied Biosystems, Foster City, California). Each clone was sequenced bidirectionally with vector primers (T7 and either SP6 or M13 reverse) on a 3730 DNA Analyzer (Applied Biosystems). The sequence was returned and was defined as a usable sequence if it was without double peaks, low signal intensity, or vector-only derivatives. A usable sequence was then submitted for further analysis.


Raw sequence data were imported into Contig Express (Vector NTI Advance 9; Invitrogen) and trimmed according to the following algorithm. Vector sequence was removed by identifying the 5′ EcoRI site and insert tag and trimming all bases upstream of the tag. The 3′ end of the sequence was trimmed just prior to the poly(A) tail. If no poly(A) tail was found, the sequence was trimmed manually at the first 25 bases containing 6 or more ambiguities. For each clone, only the direct sequence was used for analysis.

Trimmed sequences were submitted to the Expressed Sequence Tags (EST) database at GenBank (, and each sequence was assigned a unique GenBank accession number. These numbers were used to BLAST each sequence against the nonredundant nucleotide database at the National Center for Biotechnology Information (NCBI) ( The highest basic logical alignment search tool (BLAST) hit having more than 70% homology over more than 100 base pairs (bp) was chosen, and the NCBI Unigene identification numbers and gene symbols, if available, were recorded. Hits with Unigene annotations in human, mouse, and rat orthologs, respectively, were preferentially chosen if the hit scores were within 10%. The gene names obtained from the BLAST analysis were used to do a FatiGO analysis of the biological processes represented in the 2 libraries (


Both PCR and real-time PCR were used to verify the relative expression of a subset of differentially expressed genes from the cDNA libraries as well as to study the expression of 2 cytokines in the noninfected and NTHi-infected libraries. This was necessary given that the libraries used were not normalized and, therefore, only the most commonly expressed cDNAs were identified with the initial library investigation presented herein.

The genes MUC19 (OMIM 612170), TNFA (OMIM 191160), and IL1B (OMIM 147720) were of specific interest to us and were therefore chosen for real-time PCR investigation. S100 calcium-binding protein A9 (S100A9) was investigated given the enormous upregulation demonstrated and to ensure with quantitative PCR that the results found in our cDNA libraries were corroborated.

For the purpose of designing primers, highly conserved sequences in IL1B and TNFA were identified by aligning sequences for the human, rat, mouse, and pig orthologs. Primers used in the chinchilla were then designed in regions of high homology. Primers for S100A9 were designed based on clones sequenced from the NTHi-infected library. TaqMan real-time PCR primers (Table 1) for chinchilla MUC19 and HPRT (OMIM 308000) were custom designed using the manufacturer’s software (Applied Biosystems). Primer pairs used for regular reverse transcriptase–polymerase chain reaction (RT-PCR) are listed in Table 1. All primers were tested on 25 ng of cDNA from the NTHi-infected chinchilla MEM library using Platinum Blue PCR Supermix (Invitrogen) with the following parameters: 34 cycles of 94°C for 30 seconds, 61°C for 30 seconds, 72°C for 30 seconds, preceded by a 5-minute denature at 94°C and followed by 72°C elongation. Products were run on a 3% agarose gel stained with GelStar (Lonza, Basel, Switzerland) and visualized with a UV light. The PCR products were purified and sequenced using the ABI 3100 (Applied Biosystems). Sequence results were performed using the BLAST process against the nonredundant nucleotide database at NCBI ( Both IL-1β and TNF PCR products from the NTHi-infected chinchilla MEM cDNA library had high homology (>90% homology over >100 bp) to IL-1β and TNF orthologs in the guinea pig.

Table 1
Primers and Probes Used for RT-PCR and Real-Time RT-PCR

Real-time PCR was performed on cDNA from the noninfected and NTHi-infected chinchilla MEM cDNA libraries to test qualitatively and quantitatively for the expression of IL1B and TNF. A 20-µL reaction was set up in triplicate for each gene consisting of the following: 1X Platinum SYBR Green quantitative PCR Supermix-uracil DNA glycosylase (UDG) (Invitrogen), 0.2µM forward and reverse primers, 150-ng cDNA template, and 50nM fluorescein (Invitrogen). Thermal cycling conditions were as follows: UDG decontamination at 50°C for 2 minutes, initial activation at 95°C for 10 minutes, followed by 40 cycles of denaturation at 95°C for 15 seconds, and annealing and extension at 60°C for 1 minute. Reactions were run on the BioRad iCycler iQ (BioRad, Hercules, California) and a melt-curve analysis was performed at the end of each run to check the reactions for primer dimmer artifacts. An aliquot (2 µL) from each reaction was run on a 2% agarose gel to verify the amplicon size. Threshold cycle (Ct) values of HPRT and the gene of interest were then used to calculate the relative mean normalized expression in each gene.30

Real-time PCR, using TaqMan technology (Applied Biosystems) was performed on noninfected and NTHi-infected libraries to test for the presence of MUC19. Twenty-microliter reactions were set up in triplicate for each gene. Each triplicate reaction contained 30 µL of 2X TaqMan Universal PCR Mastermix, 1 µL of 20X primer/probe mix, 150 ng cDNA, and water to a total of 60 µL. Thermal cycling conditions were as follows: UDG decontamination at 50°C for 2 minutes, initial activation at 95°C for 10 minutes, followed by 50 cycles of denaturation at 95°C for 15 seconds, and annealing and extension at 60°C for 1 minute, followed by a final 4°C hold. Reactions were run on the iCycler iQ (BioRad). The Ct values of HPRT and MUC19 were then used to calculate the relative mean normalized expression.30


None of the animals in the uninfected group demonstrated evidence of MEM inflammation at the time of harvest. Each of the animals in the infected group demonstrated mucosal edema and purulent secretions.

As illustrated in Table 2, 1251 clones were sequenced from the noninfected chinchilla cDNA library, of which 724 (58%) yielded usable sequence data that could be further analyzed. The mean insert size was 340 bp. A total of 1155 clones from the NTHi-infected chinchilla cDNA library were sequenced, of which 506 (44%) were of usable quality with a mean insert size of 375 bp. Both the noninfected and infected libraries had similar numbers with respect to clones with useful annotations (71% in the noninfected group and 74% in the infected group). However, the 2 libraries differed in the number of unique transcripts represented, with the noninfected library having 66% and the infected library 43%. The relative abundance of transcripts in the noninfected chinchilla MEM library and NTHi-infected chinchilla MEM library is shown in the Figure. In the noninfected library, the most abundant transcripts (having >5 copies) accounted for 8% of the annotated transcripts. These included matrix gla-protein precursor, β-actin, secreted phosphoprotein 1, Cas-Br-M (murine) ecotropic retroviral transforming sequence b, Finkel-Biskis-Reilly murine sarcoma virus, and ubiquitously expressed FAU. The most abundant transcripts in the NTHi-infected library accounted for almost 50% of the annotated transcripts. S100A9 was the most abundant, accounting for 35% of annotated transcripts, followed by secretory leukocyte peptidase inhibitor (SLPI); β2-microglobulin (B2M); ferritin, heavy-chain polypeptide 1 (FTH1); and calcium-binding protein A8 (S100A8). Each of the transcripts with good sequence data has been submitted to GenBank at the NCBI, and a complete catalog of those transcripts is beyond the scope of this article. The consecutive GenBank accession numbers which have been assigned to the submitted sequences from the libraries are as follows: noninfected sequence; EV780884 through EV781607; and NTHi-infected sequence, EX149816 through EX150319. Additional information on these genes and sequences utilizing these accession numbers can be retrieved online at

Relative distribution of the most frequent transcripts in the noninfected vs nontypeable Haemophilus influenzae infection–infected (NTHi) chinchilla middle ear mucosa complementary DNA libraries. A, The 10 genes most frequently isolated from the ...
Table 2
Overview of Clone Sequences and Annotations From the Noninfected and NTHi-Infected Chinchilla cDNA Libraries

The FatiGO analysis was performed on the annotated sequences from the noninfected and NTHi-infected chinchilla MEM cDNA libraries to identify level 3 biological process GO terms that were differentially represented (P<.05) between the 2 libraries (Table 3). This differential representation is based on the percentage of annotated genes assigned to a particular biological process, independent of the frequency with which the gene occurs in the particular library. Therefore, in Table 3, 5 genes (2.23%) of the total annotated genes from the noninfected library fall into the biological process of defense response, whereas 13 genes (14.13%) of the total annotated genes from the infected library fall into the biological process of defense response. The biological processes with the most notable differences (2-fold or higher) are shown in Table 3 and include defense response, response to external stimulus, response to stress, and response to biotic stimulus and immune system processes. For all of these groups, the NTHi-infected library had a 2-fold or higher increase in the number of annotated genes assigned to that GO term.

Table 3
Results of FatiGO Analysis on Noninfected vs NTHi-Infected Chinchilla MEM cDNA Libraries

We tested our libraries for the presence of MUC19, IL1B, TNFA, and S100A9 using real-time PCR as described in the “Methods” section. The gene MUC19 was expressed at levels 2-fold higher in the NTHi-infected library compared with the noninfected library. The gene TNFA was expressed at levels 1.2-fold higher in the NTHi-infected library compared with the noninfected library. The gene IL1B was present at levels 23-fold higher in the NTHi-infected library compared with the noninfected library. To verify the expression of the most highly upregulated transcript from the NTHi-infected library, we assayed the original RNA from both the noninfected and NTHi-infected tissue for S100A9 calcium-binding protein. Quantitative RT-PCR for S100A9 showed a 500-fold increase in expression in the NTHi-infected library compared with the noninfected library.


Complementary DNA libraries provide the possibility for unbiased investigation of gene expression. No a priori design is necessary to decide which genes might be important or should be investigated. The results from this investigation demonstrate the usefulness of this avenue of discovery for characterization of processes that are important in the pathogenesis of OM. A number of the genes discussed in this section have not been specifically investigated in NTHi OM pathogenesis and provide considerable promise for future investigations. The use of a nonnormalized library, as in this study, allows for comparison of relative gene expression with an intervention, in this case, NTHi infection. However, it is more limited in the ability to identify more rarely expressed, yet important, genes, given its concentration on the most highly expressed transcripts. Quantitative PCR is another technique to assess comparative expression of selected genes as described in this section for TNFA, IL1B, and MUC19.

Although not initially cloned from either the NTHi-infected or noninfected libraries, TNFA, IL1B, and MUC19 were easily detected by RT-PCR from the RNA used to construct our cDNA libraries. On the one hand, because our libraries are nonnormalized, one could expect only the most common transcripts to be represented in smaller samplings of clones. On the other hand, RT-PCR gives qualitative as well as quantitative expression on a much more sensitive scale. By this method we were able to detect TNFA, IL1B, and MUC19 at higher levels in the NTHi-infected library compared with the noninfected library. These data support previous findings from our laboratory that demonstrate that mucins are upregulated in response to TNFA and IL1B.23 To our knowledge, this is the first investigation to examine the response of MUC19 in OM, which has only recently been described in our laboratory as expressed in the MEE.13 The protein products of MUC2, MUC5AC, MUC5B, and MUC6 in humans have been previously identified as gel-forming mucins. The recently identified MUC19 has also been characterized as producing a gel-forming protein product, and this glycoprotein is the largest mucin protein identified to date. The genes MUC2, MUC5AC, MUC5B, and MUC19 are gel-forming mucins found in the human ME in both in vivo and in vitro models.13 With their ability to increase ME fluid viscosity and inhibit mucociliary clearance, gel-forming mucin products should be a focus of future investigations to understand their functions, regulation, and interactions. To our knowledge, this study is the first to examine the regulation of the MUC19 mucin gene in MEM and demonstrates that it also is upregulated in NTHi-mediated OM.

We also assayed the original RNA to verify expression of one of the genes shown to be upregulated in the NTHi-infected library. The gene S100A9 was highly abundant in clones from the NTHi-infected library, representing 35% of the annotated transcripts. This important upregulation is the main factor in the difference in unique transcripts noted in this initial investigation of these libraries: 66% in the noninfected library and 43% in the infected library. This upregulation, in essence, overwhelms the ability to see other genes. In the ongoing characterization of these libraries, the number of unique transcripts will be expected to become more even. S100A9 is a calcium-binding protein that belongs to the S100 family. S100A9 and S100A8, also highly expressed in our NTHi-infected library, are capable of forming a heterodimeric complex, termed calprotectin, which has antimicrobial properties as well as the ability to stimulate neutrophils.31 Overexpression of S100A9 and S100A8 has been associated with many human inflammatory diseases, including cystic fibrosis, dermatitis, and rheumatoid arthritis.3234 To our knowledge, linkage of S100 proteins, specifically S100A9 and S100A8, to NTHi infection has not been described. Investigations into these proteins and NTHi OM pathogenesis are currently under way.

Other genes that were highly represented by the clones in our NTHi-infected chinchilla cDNA MEM library included SLPI, FTH1, and B2M. Secretory leukoprotease inhibitor is an antibacterial protein secreted by cells at mucosal surfaces. It binds tightly to mucins, protecting them against proteolysis by neutrophil elastase.35 It has been demonstrated36 that mucin gene expression, induced by cystic fibrosis respiratory stimuli, can be inhibited by SLP1, indicating its potential importance as an antimucin agent. Given these interactions and the importance of mucin in OM pathogenesis, this particular gene and its protein product certainly warrant further investigation. A central question in these investigations, given that mucins are involved in both MEM protective functions and pathologic characteristics associated with overproduction, will be a clearer understanding of temporal relationships between gene expression of SLP1 and specific mucin genes during the process of NTHi infection.

Ferritin, heavy chain 1, is the major intracellular iron storage protein in both prokaryotes and eukaryotes and has been shown to have a role in mitogen-activated protein kinase 8 signaling by inhibiting TNF-induced apoptosis. The affinity of Haemophilus (its name, meaning “blood loving,” is of Greek origin) species for iron has been well described. Residence and propagation of NTHi within the human ME, which is a highly iron-restricted environment, requires that this organism have an efficient means by which to acquire iron from its host. Recently, Harrison et al37 suggested that careful balancing of levels of intracellular iron was important for minimizing the effects of oxidative stress during NTHi colonization and infection. To our knowledge, the specific linkage of FTH1 to NTHi infection in the ME has not been described and warrants further investigation.

Many genes were shown to be differentially expressed in our noninfected and NTHi-infected chinchilla MEM cDNA libraries. Although the upregulation of S100A9 and SLPI were not surprising considering their previously described roles in inflammatory processes, the use of a nonnormalized cDNA library in this investigation allowed their identification as excellent candidates for additional study in NTHi OM pathogenesis and, specifically, with respect to their interactions with mucin gene regulation. Likewise, one would also expect the proinflammatory cytokines Il-1β and TNF to be expressed at higher levels in infected tissue. The presence of these transcripts at higher levels in the NTHi-infected tissue speaks to the ability of these libraries to represent the OM disease process. Further characterization of these libraries, which is ongoing, is expected to yield additional fruitful insights into this disease process.


Funding/Support: The investigations detailed in this article were supported by National Institutes of Health grants through the National Institute on Deafness and Other Communication Disorders: K08DC00192 and DC007903 (Dr Kerschner); DC02148, DC04173, and Health Resources and Services Administration (Dr Ehrlich); and DC05659 (Dr Post). This work was also supported, in part, through funding by the Department of Otolaryngology and Communication Sciences, Medical College of Wisconsin and Allegheny Singer Research Institute.


Author Contributions: Drs Kerschner, Horsey, Ahmed, Cioffi, Hu, and Ehrlich and Mss Erbe and Khampang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kerschner, Hu, Post, and Ehrlich. Acquisition of data: Kerschner, Horsey, Ahmed, Erbe, Khampang, and Hu. Analysis and interpretation of data: Kerschner, Erbe, Cioffi, and Ehrlich. Drafting of the manuscript: Kerschner, Ahmed, and Erbe. Critical revision of the manuscript for important intellectual content: Kerschner, Horsey, Khampang, Cioffi, Hu, Post, and Ehrlich. Statistical analysis: Kerschner. Obtained funding: Kerschner, Post, and Ehrlich. Administrative, technical, and material support: Kerschner, Horsey, Ahmed, Erbe, Khampang, Cioffi, Hu, and Ehrlich. Study supervision: Kerschner, Cioffi, Hu, and Ehrlich.

Financial Disclosure: Dr Ehrlich serves as a consultant to Medtronics Ear, Nose, and Throat (ENT) Division and also serves on the Medtronics ENT Division infectious disease–biofilm panel.

Previous Presentation: This study was presented at the 2008 American Society of Pediatric Otolaryngology Scientific Program; May 4, 2008; Orlando, Florida.


1. Bluestone CD, Klein JO. Otitis media, atelectasis, and eustachian tube dysfunction. In: Bluestone CD, Stool SE, Kenna MA, editors. Pediatric Otolaryngology. Philadelphia, PA: WB Saunders; 1996. pp. 388–582.
2. Rosenfeld RM, Casselbrant ML, Hannley MT. Implications of the AHRQ evidence report on acute otitis media. Otolaryngol Head Neck Surg. 2001;125(5):440–448. [PubMed]
3. Kerschner JE, Meyer TK, Burrows A, Ehrlich G, Post JC. Mucin gene cDNA sequence characterization in chinchilla middle ear epithelium. Int J Pediatr Otorhinolaryngol. 2006;70(8):1449–1456. [PubMed]
4. Ophir D, Hahn T, Schattner A, Wallach D, Aviel A. Tumor necrosis factor in middle ear effusions. Arch Otolaryngol Head Neck Surg. 1988;114(11):1256–1258. [PubMed]
5. Yellon RF, Leonard G, Marucha PT, et al. Characterization of cytokines present in middle ear effusions. Laryngoscope. 1991;101(2):165–169. [PubMed]
6. Juhn SK, Tolan CT, Garvis WJ, Cross DS, Giebink GS. The levels of IL-1 beta in human middle ear effusions. Acta Otolaryngol Suppl. 1992;493:37–42. [PubMed]
7. Smirnova MG, Kiselev SL, Gnuchev NV, Birchall JP, Pearson JP. Role of the proinflammatory cytokines tumor necrosis factor-alpha, interleukin-1 beta, interleukin-6 and interleukin-8 in the pathogenesis of the otitis media with effusion. Eur Cytokine Netw. 2002;13(2):161–172. [PubMed]
8. Samuel EA, Burrows A, Kerschner JE. Cytokine regulation of mucin secretion in a human middle ear epithelial model. Cytokine. 2008;41(1):38–43. [PMC free article] [PubMed]
9. Tracey KJ, Cerami A. Tumor necrosis factor and regulation of metabolism in infection: role of systemic versus tissue levels. Proc Soc Exp Biol Med. 1992;200(2):233–239. [PubMed]
10. Tracey KJ, Cerami A. Tumor necrosis factor: a pleiotropic cytokine and therapeutic target. Annu Rev Med. 1994;45:491–503. doi:10.1146/ [PubMed]
11. Brown DT, Litt M, Potsic WP. A study of mucus glycoproteins in secretory otitis media. Arch Otolaryngol. 1985;111(10):688–695. [PubMed]
12. Carrie S, Hutton DA, Birchall JP, Green GG, Pearson JP. Otitis media with effusion: components which contribute to the viscous properties. Acta Otolaryngol. 1992;112(3):504–511. [PubMed]
13. Kerschner JE. Mucin gene expression in human middle ear epithelium. Laryngoscope. 2007;117(9):1666–1676. [PubMed]
14. Hutton DA, Guo L, Birchall JP, Severn TL, Pearson JP. MUC5B expression in middle ear mucosal glands. Biochem Soc Trans. 1998;26(2):S117. [PubMed]
15. FitzGerald JE, Green GG, Stafford FW, Birchall JP, Pearson JP. Characterization of human middle ear mucus glycoprotein in chronic secretory otitis media (CSOM) Clin Chim Acta. 1987;169(2–3):281–297. [PubMed]
16. Hutton DA, Fogg FJ, Murty G, Birchall JP, Pearson JP. Preliminary characterization of mucin from effusions of cleft palate patients. Otolaryngol Head Neck Surg. 1993;109(6):1000–1006. [PubMed]
17. Zeiher BG, Hornick DB. Pathogenesis of respiratory infections and host defenses. Curr Opin Pulm Med. 1996;2(3):166–173. [PubMed]
18. Velden VH, Versnel HF. Bronchial epithelium: morphology, function and pathophysiology in asthma. Eur Cytokine Netw. 1998;9(4):585–597. [PubMed]
19. Obregon D, Hou H, Bai Y, et al. CD40L disruption enhances Abeta vaccine-mediated reduction of cerebral amyloidosis while minimizing cerebral amyloid angiopathy and inflammation. Neurobiol Dis. 2008;29(2):336–353. [PMC free article] [PubMed]
20. Macfarlane GT, Macfarlane S. Growth of mucin degrading bacteria in biofilms. Methods Mol Biol. 2000;125:439–452. [PubMed]
21. Macfarlane S, McBain AJ, Macfarlane GT. Consequences of biofilm and sessile growth in the large intestine. Adv Dent Res. 1997;11(1):59–68. [PubMed]
22. Nikawa H, Nishimura H, Hamada T, Yamashiro H, Samaranayake LP. Effects of modified pellicles on Candida biofilm formation on acrylic surfaces. Mycoses. 1999;42(1–2):37–40. [PubMed]
23. Kerschner JE, Meyer TK, Burrows A. Chinchilla middle ear epithelial mucin gene expression in response to inflammatory cytokines. Arch Otolaryngol Head Neck Surg. 2004;130(10):1163–1167. doi:10.1001/archotol.130.10.1163. [PubMed]
24. DeMaria TF, Yamaguchi T, Bakaletz LO, Lim DJ. Serum and middle ear antibody response in the chinchilla during otitis media with effusion induced by nonviable nontypeable Haemophilus influenzae. J Infect Dis. 1992;165 suppl 1:S196–S197. [PubMed]
25. Post JC, Aul JJ, White GJ, et al. PCR-based detection of bacterial DNA after antimicrobial treatment is indicative of persistent, viable bacteria in the chinchilla model of otitis media. Am J Otolaryngol. 1996;17(2):106–111. [PubMed]
26. Ashburner M, Ball CA, Blake JA, et al. Gene Ontology Consortium. Gene ontology: tool for the unification of biology. Nat Genet. 2000;25(1):25–29. [PMC free article] [PubMed]
27. Al-Shahrour F, Diaz-Uriarte R, Dopazo J. FatiGO: a Web tool for finding significant associations of gene ontology terms with groups of genes. Bioinformatics. 2004;20(4):578–580. doi:10.1093/bioinformatics/btg455. [PubMed]
28. Canafax DM, Nonomura N, Erdman GR, Le CT, Juhn SK, Giebink GS. Experimental animal models for studying antimicrobial pharmacokinetics in otitis media. Pharm Res. 1989;6(4):279–328. [PubMed]
29. Mason KM, Munson RS, Jr, Bakaletz LO. Nontypeable Haemophilus influenzae gene expression induced in vivo in a chinchilla model of otitis media. Infect Immun. 2003;71(6):3454–3462. [PMC free article] [PubMed]
30. Muller PY, Janovjak H, Miserez AR, Dobbie Z. Processing of gene expression data generated by quantitative real-time RT-PCR. Biotechniques. 2002;32(6):1372–1374. 1376, 1378–1379. [PubMed]
31. Kerkhoff C, Klempt M, Sorg C. Novel insights into structure and function of MRP8 (S100A8) and MRP14 (S100A9) Biochimica Biophysic. 1998;1448(2):200–211. [PubMed]
32. Roth J, Sunderkotter C, Goebeler M, Gutwald J, Sorg C. Expression of the calcium-binding proteins MRP8 and MRP14 by early infiltrating cells in experimental contact dermatitis. Int Arch Allergy Immunol. 1992;98(2):140–145. [PubMed]
33. Frosch M, Strey A, Vogl T, et al. Myeloid-related proteins 8 and 14 are specifically secreted during interaction of phagocytes and activated endothelium and are useful markers for monitoring disease activity in pauciarticular-onset juvenile rheumatoid arthritis. Arthritis Rheum. 2000;43(3):628–637. [PubMed]
34. Wilkinson MM, Busuttil A, Hayward C, Brock DJ, Dorin JR, Van Heyningen V. Expression pattern of two related cystic fibrosis-associated calcium-binding proteins in normal and abnormal tissues. J Cell Sci. 1988;91(pt 2):221–230. [PubMed]
35. Klingmann KL, Murphy TF. Genetics of bacteria: role in pathogenesis of infection of the respiratory tract. In: Senior RM, editor. Molecular Biology of the Lung. Vol 1. Berlin, Germany: Birkhäuser Verlag; 1998. pp. 149–162.
36. Griffin S, Carroll TP, Greene CM, O’Neill SJ, Taggart CC, McElvaney NG. Effect of pro-inflammatory stimuli on mucin expression and inhibition by secretory leucoprotease inhibitor. Cell Microbiol. 2007;9(3):670–679. doi:10.1111/j.1462-5822.2006.00819.x. [PubMed]
37. Harrison A, Ray WC, Baker BD, Armbruster DW, Bakaletz LO, Munson RS., Jr The OxyR regulon in nontypeable Haemophilus influenzae. J Bacteriol. 2007;189(3):1004–1012. [PMC free article] [PubMed]