RESEARCH ARTICLE

Increased susceptibility to otitis media in a Splunc1-deficientmouse modelJennifer A. Bartlett1,*, David K.Meyerholz2,*, Christine L.Wohlford-Lenane1, PaulW. Naumann2, Nita H. Salzman3

and Paul B. McCray, Jr1,‡

ABSTRACTOtitis media (inflammation of the middle ear) is one of the mostcommon diseases of early childhood. Susceptibility to otitis isinfluenced by a number of factors, including the actions of innateimmune molecules secreted by the epithelia lining the nasopharynx,middle ear and Eustachian tube. The SPLUNC1 (short palate, lung,nasal epithelial clone 1) protein is a highly abundant secretory productof the mammalian nasal, oral and respiratory mucosa that is thoughtto play a multifunctional role in host defense. In this study weinvestigated Splunc1 expression in the ear of the mouse, andexamined whether this protein contributes to overall host defense inthe middle ear and/or Eustachian tube. We found that Splunc1 ishighly expressed in both the surface epithelium and in submucosalglands in these regions in wild-typemice. In mice lacking Splunc1, wenoted histologically an increased frequency of otitis media,characterized by the accumulation of leukocytes (neutrophils withscattered macrophages), proteinaceous fluid and mucus in themiddle ear lumens. Furthermore, many of these mice had extensiveremodeling of the middle ear wall, suggesting a chronic course ofdisease. From these observations, we conclude that loss of Splunc1predisposes mice to the development of otitis media. The Splunc1−/−

mouse model should help investigators to better understand both thebiological role of Splunc1 as well as host defense mechanisms in themiddle ear.

KEY WORDS: SPLUNC1, Otitis media, Mouse model, Middle ear,Eustachian tube, Host defense

INTRODUCTIONOtitis media, or inflammation of the middle ear, is one of the mostfamiliar illnesses of early childhood. By the age of 3, over 80% ofchildren have had at least one episode of otitis media (Teele et al.,1989), making it one of the most common reasons for physicianvisits and accounting for an estimated US$2.88 billion to US$3.8billion in direct and indirect healthcare costs each year in the UnitedStates (O’Brien et al., 2009; Ahmed et al., 2014). The inflammationassociated with otitis media can be infectious or non-infectious.Although acute otitis media often resolves with antibiotic treatment,the condition can progress to prolonged or recurrent forms known as

chronic otitis media with effusion or chronic suppurative otitismedia. Over time, chronic otitis media can lead to complicationssuch as conductive hearing loss (Teele et al., 1990; Klausen et al.,2000), and problems with balance and motor coordination(Casselbrant et al., 1995; Gawron et al., 2004). Otitis media is amultifactorial condition, with a variety of environmental risk factorsincluding daycare outside the home, parental smoking, the presenceof siblings and lack of breastfeeding (Uhari et al., 1996). Studieswith twins and triplets indicate that there is also a significant geneticcomponent to otitis susceptibility (Casselbrant et al., 2004).

Todefend themiddle ear fromenvironmental insults, the epithelia ofthe nasopharynx, middle ear and Eustachian tube secrete an array ofantimicrobial and immune modulatory molecules. In the chinchilla,these secretions include the proposed innate immune protein Splunc1(short palate, lung, nasal epithelial clone 1; also known as PLUNC,LUNX, NASG, SPURT and BPIFA1) (McGillivary and Bakaletz,2010). In addition to its expression in the ear, Splunc1 expressionis widespread throughout the conducting airways. In humans andother mammals, Splunc1 is expressed by the surface epitheliumand submucosal glands and ducts of the trachea, bronchus, nasalepithelium, nasopharynx and salivary glands (Weston et al., 1999;Bingle and Bingle, 2000; LeClair et al., 2001; Sung et al., 2002;Wheeler et al., 2002; Di et al., 2003; Campos et al., 2004; Bingle et al.,2005; Larsen et al., 2005; Kim et al., 2006; Vargas et al., 2008;McGillivary and Bakaletz, 2010; Musa et al., 2012). Accordingly, theprotein is present in human nasal secretions (Ghafouri et al., 2002,2003b, 2004; Casado et al., 2005; Kim et al., 2006), tracheal aspirates(Campos et al., 2004) and saliva (Ghafouri et al., 2003a), as well as inporcine bronchoalveolar lavage (Bartlett et al., 2013). The emergingpicture suggests that Splunc1 is a highly abundant secretory product ofthemammalian respiratory system, whose expression extends from theupper respiratory tract into the oral cavity, nasopharynx andmiddle ear.

Although the precise biological function of the Splunc1 proteinis not yet well-defined, there is growing evidence that it participatesin host defense in the conducting airways. In humans, alteredSPLUNC1 levels are associated with exposure to allergens(Seshadri et al., 2012), airway irritants (Fornander et al., 2013),and chronic bacterial colonization and airway inflammation dueto cystic fibrosis (Bingle et al., 2007). Animal studies suggestthat Splunc1 has antibacterial and/or immunomodulatory effects.Although Splunc1-deficient mice do not develop spontaneouslung disease in the absence of a bacterial insult, they do exhibitimpaired responses to intrapulmonary challenge with Mycoplasmapneumoniae (Gally et al., 2011), Pseudomonas aeruginosa(Liu et al., 2013a) and Klebsiella pneumoniae (Liu et al., 2013b),and overexpression of Splunc1 protects mice from inhaledP. aeruginosa (Lukinskiene et al., 2011). Although directmicrobicidal action is the mechanism most frequently invoked toexplain this Splunc1-mediated protective effect, Splunc1 is alsoproposed to possess immunomodulatory (Thaikoottathil et al., 2012),Received 2 January 2015; Accepted 7 March 2015

1Department of Pediatrics, Carver College of Medicine, University of Iowa, IowaCity, IA 52242, USA. 2Department of Pathology, Carver College of Medicine,University of Iowa, Iowa City, IA 52242, USA. 3Department of Pediatrics, Division ofGastroenterology, Medical College of Wisconsin, Milwaukee, WI 53226, USA.*These authors contributed equally to this work

‡Author for correspondence (paul-mccray@uiowa.edu)

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

501

© 2015. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

anti-biofilm (Gakhar et al., 2010) and/or chemotactic (Sayeed et al.,2013) properties.Given its distribution throughout the conducting airways, most

studies of Splunc1−/− mice have centered on identifying defectsin lung host defense. The lack of spontaneous lung diseasein Splunc1−/− mice suggests that this protein plays a host defenserole that is redundant with other immune molecules found in theairways, or that its function is most relevant in specific contexts (e.g.early infection). In this study, we examined the consequences ofSplunc1 deficiency in tissues outside the respiratory tract, focusingon the middle ear and Eustachian tube. In contrast to the lungphenotype in these animals, we discovered spontaneous middle eardisease in the Splunc1−/− mice. Here, we characterize Splunc1expression in the murine middle ear and Eustachian tube, andinvestigate the incidence of otitis media in Splunc1−/− animals.

RESULTSSplunc1 mRNA and protein is reportedly expressed in the middleear and Eustachian tube of the chinchilla (McGillivary andBakaletz, 2010). Using immunohistochemistry, we observed avery similar expression pattern for Splunc1 in the wild-type mouse.

In these animals, Splunc1 protein is abundantly expressed in boththe middle ears and Eustachian tubes (Fig. 1), with a pattern verysimilar to that in the conducting airways. In the Eustachiantube, which possesses a pseudostratified epithelium continuouswith the respiratory epithelium of the nasopharynx, Splunc1expression localizes to ciliated and nonciliated columnar epithelialcells (Fig. 1A). A similar distribution was observed for the surfaceepithelium of the middle ear (Fig. 1B). Additionally, substantialSplunc1 expression was noted in serous cells of the submucosalglands throughout both the middle ear and Eustachian tube regions(Fig. 1B,C). Splunc1 protein was completely absent from thesetissues in Splunc1−/− mice (Fig. 1D-F).

Considering the abundance of Splunc1 in these regions in wild-type mice, we hypothesized that loss of Splunc1 would increasesusceptibility to infection and/or inflammation in the middle ear.To explore this hypothesis, we performed a histological analysis onhead sections from a total of 26 Splunc1−/− mice and 18 age-matched Splunc1+/+ littermate controls to look for evidence of acuteand/or chronic otitis. Reasoning that the characteristic changes to theepithelium associated with recurrent infection and/or inflammationwould be most apparent in older mice, we focused on relatively agedmice (10-18 months of age; median age for both groups was10 months).

We found that, as a group, Splunc1−/− mice exhibited anincreased incidence of otitis relative to the wild-type control mice atthe time of necropsy. This phenotype was incompletely penetrant,with Splunc1−/− mice exhibiting a range of severity with respect toinflammatory markers and epithelial changes. Several Splunc1−/−

mice (8 out of 26; 30.8%) exhibited overt otitis media, associatedwith mucopurulent material completely or partially filling themiddle ear lumen (Figs 2 and 3). Otitis was generally unilateral(Fig. 2), although a bilateral presentation was observed in one case.In contrast, overt otitis was noted in only 1 out of 18 (5.5%) wild-type mice. In Splunc1−/− mice with more moderate phenotypes,modest numbers of polymorphonuclear neutrophils (PMNs) couldbe seen within the middle ear lumen (Figs 3 and 4).

Interestingly, our studies suggest that many of the Splunc1−/−

mice had likely experienced recurrent episodes of otitis.Histological findings indicative of chronic otitis media includedfrequent remodeling of the middle ear epithelium, characterized byepithelial hyperplasia, polyps and thickening of the middle earmucosa (Figs 2 and 3). Less frequently, we observed cholesterolclefts admixed in cellular debris in the middle ear lumen (Fig. 3B).We additionally noted that, relative to their wild-type counterparts,Splunc1−/− mice tended to exhibit greater amounts of cell debris inthe fluid of the middle ear, including eosinophilic ‘ghost’ cells(Fig. 4). Although the origin of these ghost cells is not well-established, we speculate that they represent dead PMNs possiblyassociated with past inflammatory events, because ghost cells andscattered PMN infiltration were sometimes seen in conjunction.

The inciting event for the observed inflammation in Splunc1−/−

middle ears was unclear because histological stains of middle earsections from affected mice were negative for bacterial and fungalpathogens (Fig. 5). When histological methods failed to uncoverevidence for microbial infections in these middle ears, we usedfluorescence in situ hybridization (FISH) to search for bacterialgenetic material in tissue sections from affected mice. Middle earsections from Splunc1−/− mice exhibiting otitis, as well asunaffected wild-type controls, were hybridized with a ‘universal’bacterial probe recognizing a conserved region of the 16S rRNA(Amann et al., 1990). This approach also failed to detect bacteria inthe middle ears of the mice (not shown).

TRANSLATIONAL IMPACT

Clinical issueOtitis media, or inflammation of the middle ear due to infectious or non-infectious causes, is a very frequent medical problem in infants andyoung children. The economic impact of this problem is substantialbecause it affects the majority of children at some point in their lives andoften necessitates doctor visits. Susceptibility to otitis media isinfluenced by a complex interplay of environmental and genetic riskfactors, some of which have been explored using knockout mousemodels. In many of these models, otitis arises spontaneously owing toanatomical defects, such as craniofacial malformations that impairEustachian tube function or cilia in the middle ear and/or Eustachiantube. Less is known, however, about how loss of innate immune factorsmight also predispose to the development of middle ear disease.

ResultsHere, we describe spontaneous development of otitis media in micelacking the innate immune molecule Splunc1 (short palate, lung, nasalepithelial clone 1). Splunc1 is best known as a highly expressedsecretory product of the conducting airways, where it participates in hostdefense against a variety of airway pathogens. We found that, in additionto its expression in airway tissues, Splunc1 is also expressed by theepithelia of the middle ear and Eustachian tube in the mouse. Ourhistopathological analysis indicates that Splunc1–/– mice develop otitismedia at a higher frequency than their wild-type littermates, with anincreased incidence of inflammatory markers such as leukocytes andcell debris in the middle ear lumen. We also observed remodeling of themiddle ear epithelium, suggesting that otitis media is a chronic problem inthese mice.

Implications and future directionsThe Splunc1–/– mouse represents a new mouse model with which tostudy the complex biological origins of otitis media. The Splunc1 proteinis thought to play a multifunctional role in host defense, with proposedantibacterial, anti-biofilm, immunomodulatory and chemotacticproperties. Therefore, our findings suggest the interesting possibilitythat loss of Splunc1 function might render Splunc1–/– mice vulnerable tootitis throughmultiple overlappingmechanisms. Additionally, the insightsprovided by studies of the Splunc1–/– mouse have potential therapeuticimplications because this study points to the SPLUNC1 gene as apossible contributor to otitis media susceptibility in the human populationand might suggest possibilities for therapeutic interventions to treat orprevent this illness.

502

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

We also looked for pathology of the Eustachian tube in theSplunc1−/− mice. A common contributing factor in otitis media isloss of patency in the Eustachian tube, which can prevent drainageand clearance of pathogens from the middle ear. To help maintainpatency, the Eustachian tube epithelium normally secretessurfactant molecules that lower the opening pressure between themiddle ear and the nasopharynx. Of note, Splunc1 has previouslybeen demonstrated to exhibit surfactant-like properties (Gakharet al., 2010). Therefore, we hypothesized that loss of Splunc1expression might result in persistent loss of Eustachian tube patencyin Splunc1−/− mice. To address this, we examined Eustachian tubesin coronal head sections from Splunc1+/+ and Splunc1−/− mice(aged 11-17 months) to assess the frequency of Eustachian tubecollapse at the time of necropsy. As shown in Fig. 6, we were unableto observe evidence for collapsed Eustachian tubes in eitherSplunc1+/+ or Splunc1−/− mice (Fig. 6A,B). Furthermore, averageEustachian tubewidthmeasurements were similar between thewild-type and Splunc1−/− mice (Fig. 6C,D).To compare the overall frequencyof otitis media in the Splunc1−/−

mice and controls, a Fisher exact test was performed for ‘overt otitis’

in the two populations. Using this approach, we observed a trendtoward increased otitis frequency in the Splunc1−/−mice that did notquite reach statistical significance (P=0.0603). In support of this,histological scoring indicated a trend toward increased PMNrecruitment to the middle ear lumen, accompanied by significantlygreater numbers of macrophages and increased cell debris andremodeling in the middle ears of Splunc1−/− mice relative to wild-type controls (summarized in Fig. 7).

DISCUSSIONOur findings indicate that Splunc1 is highly expressed in theepithelium and glandular tissue of the middle ear and Eustachiantube of the mouse, and that loss of Splunc1 from these regionspredisposes animals to develop otitis media. Thus, we can add theSplunc1−/− mouse to the list of models that feature spontaneouschronic otitis media as a component of their phenotype. A review ofother otitis-prone mouse models highlights the diverse biologicalprocesses that can influence susceptibility to this condition. In theeyes absent homolog 4 (Eya4−/−) mouse and the Tail-short (Ts)mouse (which harbors a deletion in the ribosomal-subunit gene

Fig. 1. Splunc1 protein is expressed inthe mouse middle ear and Eustachiantube. Splunc1 (brown stain) is present inthe ciliated and non-ciliated cells of thesurface epithelium lining the Eustachiantube (A) and middle ear (B) in the mouse.Splunc1 is also abundant in the serouscells of the submucosal glands in thesetissues (A,C). Splunc1 expression iscompletely absent in the middle ears andEustachian tubes ofSplunc1−/−mice (D-F).L, lumen of the Eustachian tube; SMG,submucosal glands; M, mucous tubules;S, serous acini. Scale bar: 200 µm.

Fig. 2. Unilateral otitis media in aSplunc1−/− mouse. H&E-stainedcoronal sections through the heads of10- to 18-month-old Splunc1+/+ andSplunc1−/− mice were inspected forgross abnormalities. The top paneldepicts a representative coronal sectionfrom a Splunc1+/+ mouse. The bottompanel is from a Splunc1−/− mouseexhibiting unilateral otitis media,characterized by purulent material in themiddle ear lumen (white asterisk). In thismouse, thickening of the tympanicmembrane and middle ear epithelium(black arrowhead) are indicative ofchronic otitis media. E, external earcanal; M, middle ear; in both panels,black arrows indicate the tympanicmembranes. Scale bar: 0.7 mm.

503

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

Rpl38), craniofacial and Eustachian tube dysmorphologies result inabnormal clearance and/or drainage of the middle ear andEustachian tube (Depreux et al., 2008; Noben-Trauth andLatoche, 2011). Mutations leading to ciliary defects (such asthose seen in the Chibby- and Spag6-knockout mice) alsopredispose to otitis, owing to reduced mucociliary clearance inthese regions (Voronina et al., 2009; Li et al., 2014). Chronic otitis isalso associated with loss of the transcription factors Fbxo11(mutated in the Jeff mouse model), Evi1 (deleted in the Junbomouse model) and transforming growth factor 1 (Tgif ), which affecta broad range of developmental and other phenomena through theirinvolvement in the TGF-β signaling pathway (Hardisty-Hugheset al., 2006; Parkinson et al., 2006; Tateossian et al., 2013).Susceptibility to otitis might also be impacted by deficiencies ininnate immune molecules, such as the microbial pattern recognitionreceptors Toll-like receptor 4 (TLR4) (MacArthur et al., 2006) andToll-like receptor 2 (TLR2) (Leichtle et al., 2009), and the immunesignaling molecule MyD88 (Hernandez et al., 2008).We have yet to determine how loss of Splunc1 leads to increased

otitis susceptibility. However, the multifunctional nature of thisprotein suggests that there could be multiple mechanismscontributing to the Splunc1−/− middle ear phenotype. There issome evidence to support the notion that Splunc1−/− mice arepredisposed to develop otitis media owing to loss of antibacterialactivity by the Splunc1 protein. Several groups have reported thatSplunc1 has direct bactericidal and/or bacteriostatic effects (Chu

et al., 2007; Zhou et al., 2008; Sayeed et al., 2013), although it seemsthat these effects are limited to certain pathogens. Of interest,McGillivary and colleagues demonstrated that Splunc1 proteininhibited the growth in vitro of nontypeable Haemophilusinfluenzae (NTHi), one of the organisms most frequently associatedwith otitis media (McGillivary and Bakaletz, 2010). However, in vivoexperiments suggested that Splunc1 knockdown in the middle ear ofthe chinchilla did not significantly impair the animal’s ability tohandle an acute NTHi challenge, making these findings somewhatdifficult to interpret (McGillivary and Bakaletz, 2010).

Previous reports indicate that, as a surfactant, Splunc1 inhibitsbacterial growth indirectly by interfering with biofilm formation(Gakhar et al., 2010; Liu et al., 2013b). Biofilms contribute to thepersistence of otitis-causing microbes in the middle ear (Post, 2001;Ehrlich et al., 2002; Hall-Stoodley et al., 2006; Hoa et al., 2009;Thornton et al., 2011); it is possible that loss of Splunc1 results inovergrowth of bacterial biofilms and a reduced ability to clearbacteria from the ear. It is important to point out that wewere unableto find evidence for bacteria in the middle ears of the Splunc1−/−

mice, by histological methods or by looking for bacterial geneticmaterial, which suggests that we might be observing a sterileinflammation process in the Splunc1−/− middle ear. As such, ourfindings might reflect a necessity for Splunc1 in maintaininghomeostasis in the middle ear space, with loss of Splunc1 resultingin increased cell death and debris and a consequent influx ofinflammatory cells. Additionally, we cannot rule out the possibility

Fig. 3. Otitis media in Splunc1−/− mice is incompletelypenetrant. Panel A depicts the lumen of the middle earfrom a representative Splunc1+/+ mouse, which is largelyfree of PMNs or epithelial hyperplasia. In contrast, middleears from differentSplunc1−/−mice (B-F) exhibited variableseverity with respect to inflammation and epithelialremodeling at the time of necropsy. The abundance ofaccumulated PMNs and cell debris in the middle earlumens of these mice ranged from moderate (B,C) to verysignificant (D-F). Occasionally, small spicule-shapedspaces known as cholesterol clefts (indicative ofcholesterol crystals in the original tissue) were observedalong with cellular debris in the middle ears of Splunc1−/−

mice (B). Some Splunc1−/− mice also presented withvarying amounts of epithelial proliferation and mucosalthickening in the middle ear (D, yellow arrow) and tympanicmembrane (D, black arrow), suggesting a history ofrepeated otitis. In the mouse depicted in panels E and F,remodeling manifests as thickening of the tympanicmembrane (black arrows) by hyperkeratosis (asterisks),inflammation and increased connective tissue. The regionof middle ear adjacent to inflammation often had increasedconnective tissue, epithelial proliferation and/or polyploidchanges (arrowheads). E, external ear canal; M, middleear. Scale bar: 180 µm (A,B,D-F); 90 µm (C).

504

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

that the Splunc1−/− middle ears could have harbored bacteria atvarious time points throughout the lifespans of the mice, and that theinflammation seen in aged mice could represent responses toinfections that failed to resolve normally.The SPLUNC1 protein is thought to share evolutionary origins

with lipopolysaccharide-binding protein (LBP) and bactericidal/permeability increasing protein (BPI). LBP and BPI both act assensors for the bacterial cell-wall component lipopolysaccharide(LPS), coordinately regulating the presentation of LPS to itsreceptor, TLR4, and thus directing TLR4-mediated inflammatoryresponses. The evolutionary relationship between SPLUNC1 andboth LBP and BPI has long invited speculation that SPLUNC1

might have either pro- or anti-inflammatory effects in the context ofa bacterial infection or other inflammatory stimulus. It has beenreported that SPLUNC1 binds LPS (Ghafouri et al., 2004; Sayeedet al., 2013), and SPLUNC1 binds directly to P. aeruginosa (Sayeedet al., 2013), suggesting that, like LBP and BPI, SPLUNC1 mightact as a bacterial sensor molecule. Interestingly, mice lacking Tlr4also spontaneously develop chronic otitis media (MacArthur et al.,2006), owing to their inability to mount a sufficient inflammatoryresponse to bacterial pathogens; the observation of a similarphenotype in the Splunc1−/− mouse suggests that these mice sufferfrom a deficiency in Tlr4-mediated inflammation owing to alteredresponsiveness to LPS. It has also been observed that Tlr2 signalingin response to M. pneumoniae-derived lipoproteins is inhibited inthe presence of Splunc1, possibly because Splunc1 directly bindsthe lipoproteins and prevents them from engaging the receptor Tlr2(Chu et al., 2007).

In addition to possible roles in modulation of Tlr4- and Tlr2-mediated inflammation, Splunc1 has also been implicated inregulating allergic inflammation (Thaikoottathil et al., 2012) andin the response to a ‘sterile’ inflammatory stimulus, carbonnanotubes (Di et al., 2013). One measure of the acuteinflammatory response – influx of PMNs and alveolarmacrophages – was enhanced in the lungs of Splunc1-overexpressing mice after receiving carbon nanotubes (Di et al.,2013). This observation suggests that Splunc1 might havechemotactic properties, a finding that was confirmed directly inin vitro cell migration assays (Sayeed et al., 2013). Thus, a reducedcapacity for leukocyte recruitment could also contribute todysregulated inflammation in the Splunc1−/− middle ear.

A final consideration is the possibility of Eustachian tubedysfunction in the Splunc1−/− mouse. As described in the Results,we found no histological evidence for an increased frequency ofEustachian tube collapse in the Splunc1−/− mice. Although thismost likely indicates that the Eustachian tube is not affected by loss

Fig. 4. Themiddle ears ofSplunc1−/−mice harbor increased inflammatorycells and cellular debris relative to wild-type littermate controls. The toppanel depicts a representative image of a middle ear from aSplunc1+/+mouse,in which the middle ear epithelium is covered by a layer of fluid containingglobular fluid material. In the Splunc1−/− middle ear image (bottom panel) thisfluid contains multiple punctate eosinophilic ‘ghost’ cells (black arrowheads)along with a small number of solitary PMNs (black arrows). Scale bar: 43 µm.

Fig. 5. Lack of evidence for bacterial or fungal organisms in Splunc1−/−

middle ears. Middle ear sections from Splunc1−/− mice with otitis wereexamined for the presence of microorganisms. (A) A modified Gram stain wasnegative for bacteria. (B) Gomori methenamine silver (GMS) staining wasperformed to survey for fungal organisms. In this representative image, GMSstaining produced typical black staining of both goblet cell mucus and luminalsecreted mucus, but was negative for fungal organisms in luminal material.Scale bar: 57 µm (A); 38 µm (B).

Fig. 6. Assessment of Eustachian tube patency in Splunc1+/+ andSplunc1−/− mice. The nasopharynx and Eustachian tubes were visualized incoronal sections from Splunc1+/+ and Splunc1−/− mice (representative imagesin panels A and B, respectively). NP, nasopharynx; asterisks indicate thelumens of Eustachian tubes. Scale bar: 1.1 mm. To assess patency at the timeof necropsy, Eustachian tubes were measured to determine width at theirwidest point (summarized in C), as well as at their widest point on thecontralateral side (D). In these graphs, bars depict the mean and s.e.m. for n=4Splunc1+/+ and n=4 Splunc1−/− mice. Unpaired t-tests were performed to testfor statistically significant differences.

505

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

of Splunc1 surfactant activity, it should be noted that analysis ofhistological sections might not be the most effective approach forcapturing a dynamic process such as the opening and closing of theEustachian tube. It is also possible that loss of Splunc1 surfactantactivity reduces mucociliary clearance in the middle ears andEustachian tubes of the Splunc1−/− mice. Several studies havedocumented the ability of surfactants to enhance mucociliarytransport rates (Allegra et al., 1985; Kakuta et al., 1991; Ballardet al., 2006), and it has been proposed that one role of the Splunc1protein could be to promote mucociliary clearance throughout theconducting airways (Gakhar et al., 2010; McGillivary and Bakaletz,2010). McGillivary and Bakaletz (2010) measured reducedmucociliary clearance rates in the chinchilla Eustachian tubesubsequent to Splunc1 knockdown, supporting the idea thatreduced clearance of bacteria and other inflammatory stimuli fromthe middle ear might contribute to middle ear infections inSplunc1−/− mice.In conclusion, our results indicate that the phenotype of the

Splunc1−/− mouse should now be broadened to include increasedrisk of spontaneous development of middle ear disease. Futurestudies with this animal model will likely include a more detailedanalysis of the molecular processes that lead from Splunc1deficiency to the development of otitis. In addition to providingadditional insight into the unique, and possibly varied, roles that theSplunc1 protein plays in mucosal host defense, this mouse shouldalso serve as a useful model with which to explore the biologicalpathways that influence otitis susceptibility. In particular, theseresults suggest that genetic variation at the SPLUNC1 locus couldimpact vulnerability to otitis media in the human population, and

that measurement of SPLUNC1 levels in the middle ear might havepredictive value in identifying patients at increased risk ofdeveloping chronic otitis. Along these lines, we note that Saferaliand colleagues recently reported a single-nucleotide polymorphism(SNP; rs1078761) that is associated with a decreased abundance ofSPLUNC1 protein, and worsened lung function, in the airways ofindividuals with cystic fibrosis (Saferali et al., 2015). These resultssupport a role for SPLUNC1 in mucosal host defense, and implythat this protein might have therapeutic potential in the treatmentand/or prevention of otitis media.

MATERIALS AND METHODSAnimalsThis study was approved by the University of Iowa Animal Care and UseCommittee (IACUC). All mice were housed under specific pathogen-freeconditions, in accordancewith the specifications of the National Institutes ofHealth. Splunc1−/− mice used in this study harbored a nonsense mutation inexon 50 of the Splunc1 gene (L50X) (Liu et al., 2013b), on the C3HeB/FeJstrain background. This mutation is associated with complete ablation ofSplunc1 expression in the lungs. Heterozygous (Splunc1+/−) breeder micewere crossed to generate Splunc1−/−mice and Splunc1+/+ littermate controlsfor all experiments. For this analysis, all mice were necropsied at10-18 months of age.

Tissue processingMice were euthanized by carbon dioxide exposure with confirmationof euthanasia by bilateral thoracotomy according to IACUC approvedguidelines, and tissues immediately harvested. Skulls were dissectedfree of soft tissues, the lower jaw and brain removed, and skulls wereplaced in 10% neutral buffered formalin for 4 days. Following fixation,skulls were washed with tap water for 2 h and decalcified in 14%EDTA (Sigma-Aldrich, St Louis, MO, USA), pH 7.3, for 7 days withagitation. The skulls were washed for 2-3 h with tap water, placed back into10% neutral buffered formalin and then routinely processed, paraffinembedded, and sectioned to view the middle ears or Eustachian tubes.Tissue sections were then either processed for immunohistochemicalanalysis or stained with hematoxylin and eosin (H&E) for histopathology.Additional stains used to survey for microorganisms included modifiedGram staining (Stoltz et al., 2010) and Gomori methenamine silver (GMS)staining (Grocott, 1955).

ImmunohistochemistryHead sections (6 µm) were heated in a 55°C oven for 15 min anddeparaffinized by incubating with xylenes for two consecutive 15-minintervals. Sections were then washed twice with 100% ethanol (2 min perwash), followed by incubations for 2 min each in 95% ethanol, 75% ethanol,50% ethanol, and double-distilled water. Antigen retrieval was achievedby incubating with Antigen Unmasking Solution (Vector Laboratories,Burlingame, CA, USA), diluted 1:100 in water and heated in a microwave.Endogenous peroxidase activity was quenched by washing sections twicewith PBS and incubating for 30 min with a 0.3% H2O2 solution. The M.O.M.™ (‘Mouse on mouse’) Immunodetection Kit (Vector Laboratories) wasused to detect Splunc1 protein expression in mouse head sections. To dothis, sections were blocked for 1 h using the M.O.M.™Mouse Ig Blockingreagent, washed twice with PBS, and incubated in the M.O.M.™ Diluent.Sections were then incubated overnight at 4°C with SPLUNC1 monoclonalantibody (R&D Systems, Minneapolis, MN, USA; catalog numberMAB1897), diluted 1:1500 in M.O.M.™ Diluent. After washing withPBS, sections were incubated withM.O.M.™Biotinylated Anti-Mouse IgGReagent for 10 min at room temperature. Sections were washed again withPBS, followed by a 30-min incubation with VECTASTAIN® ABC Reagent(Vector Laboratories). Following another PBS wash, color detection wasperformed by incubating sections with Vector® DAB peroxidase substratesolution (Vector Laboratories). Sections were counterstained withhematoxylin and mounted in Permount mounting medium (ThermoFisher Scientific, Waltham, MA, USA).

Fig. 7. Summary of histopathological analysis ofSplunc1−/− andwild-typemouse middle ears. Middle ear sections from Splunc1−/− mice and age-matched wild-type littermate controls were examined for histological evidenceof otitis or other significant changes in the middle ear, using scoring rubricsdescribed in the Materials and Methods. Sections were scored for severalparameters including the infiltration of PMNs and macrophages in the middleear lumen (panels A and B, respectively), accumulated cell debris (C), and thedegree of hyperplasia and remodeling exhibited by the epithelium (D). Scoresare summarized as box and whiskers plots. In these plots, the median score isdenoted by a thick line, the box indicates the 25th and 75th percentiles, and theminimum and maximum values are depicted by the outer bars. Mann–Whitneytests were used to test for statistically significant differences (n=18 Splunc1+/+

and n=26 Splunc1−/− mice).

506

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

Fluorescence in situ hybridization (FISH)Mouse heads were fixed, processed and paraffin embedded as describedabove.Head sections (3 µm)weremounted on slides andFISHwas performedas described by Canny et al. (2006). Briefly, slides were deparaffinized, dried,and hybridized with the Texas-red-labeled probe (Bact338 TR-GCTGCCTCCGTAGGAGT) (Operon Technologies, Huntsville, AL, USA)for 90 min at 50°C in hybridization buffer (0.9 MNaCl, 20 mMTris-HCl pH7.4, 0.05%SDS). Slideswerewashed for 5 min at 50°C inwash buffer (0.9 MNaCl, 20 mM Tris-HCl pH 7.4, 0.01% SDS), rinsed in water and allowed toair dry. Tissue sections were mounted with coverslips using Vectashieldmounting medium (Vector Laboratories, Burlingame, CA, USA) forfluorescence. Slides were viewed by fluorescence microscopy using a NikonE400 upright microscope. Images were captured using a PhotometricsCoolSnap ES CCD camera (Photometrics, USA), and analyzed usingMetaview software (Universal Imaging Corporation, Molecular Devices,USA). For these studies,mouse intestinal tissue served as a positive control forthe probe.

Histological analysisHistological analysis was performed according to principles described byGibson-Corley et al. (2013). H&E-stained head sections were examined by apathologist masked to groups and scored as one batch for signs of infectionand inflammation. In this analysis, ‘overt otitis’ refers to readily observabledisease (inflammation/remodeling) that is apparent at low magnification(20× magnification) at histological examination. To assess the degree ofPMN infiltration into the middle ear lumen, each middle ear section wasscored according to the following ordinal scale: 0=no PMN in lumen; 1=1-5PMN, isolated or scattered infiltrates in middle ear; 2=6-20 PMN, scatteredor in very small loose aggregates in middle ear; 3=PMN cellular aggregatesfilling 25% of the middle ear; 4=PMN cellular aggregates filling 26-50% ofthe middle ear; 5=PMN cellular aggregates filling >50% of the middle ear.Middle ear lumens were scored for the presence of macrophages as follows:0=no macrophages in lumen; 1=1-5 macrophages, isolated or scattered inmiddle ear; 2=6-20 macrophages, scattered or in very small loose aggregatesin middle ear; 3=21-40 macrophages, cellular aggregates; 4=>40macrophages, aggregates. Cellular debris in the middle ear was assessedaccording to this scale: 0=no debris in middle ear; 1=1-5, isolated orscattered in middle ear; 2=6-20, scattered or in very loose aggregates inmiddle ear; 3=21-40, cellular aggregates; 4=>40, aggregates. Epithelialremodeling was scored as follows: 0=normal; 1=attenuation of epithelium;2=focal hyperplasia; 3=widespread epithelial hyperplasia and thickening,expansion of lamina propria, inflammatory cell exocytosis.

For all parameters, scores are presented as box and whisker plotsrepresenting a total of 18 Splunc1+/+ and 26 Splunc1−/− mice. Mann–Whitney tests were performed to test for statistically significant differencesbetween Splunc1+/+ and Splunc1−/− mice for each parameter, assuming anα=0.05. Reported P-values represent two-tailed values in each case. Theoverall incidence of overt otitis media in the Splunc1+/+ and Splunc1−/−

groups was compared using a Fisher exact test, in which ‘overt otitis’ wasdefined as a score of 3 or greater in the PMN scale described above. Usingthis criterion, there was a total of 1 out of 18 wild-type mice and 8 out of 26Splunc1−/− mice that were considered to exhibit overt otitis.

Eustachian tube width measurements were made using ImageJ, and wereplotted as bar graphs representing the mean and s.e.m. For this data set,unpaired t-tests were performed to test for statistically significantdifferences, with an α=0.05. All statistical analyses were performed usingGraphPad Prism, version 6.02.

AcknowledgementsWe thank Eugene Chang, Janice Staber and Sateesh Krishnamurthy for criticalreview of the manuscript.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsJ.A.B., D.K.M. and P.B.M. designed the experiments; D.K.M., C.L.W.-L. and P.W.N.processed tissues and performed immunohistochemistry; D.K.M. performed the

histopathological analysis; N.H.S. performed fluorescence in situ hybridizationexperiments; J.A.B., D.K.M. and P.B.M. wrote the manuscript.

FundingWe acknowledge support from the National Institutes of Health P50 HL-61234(P.B.M.) and P01 HL-091842 (P.B.M.), as well as the Roy J. Carver Charitable Trust(P.B.M.).

ReferencesAhmed, S., Shapiro, N. L. and Bhattacharyya, N. (2014). Incremental health care

utilization and costs for acute otitis media in children. Laryngoscope 124, 301-305.Allegra, L., Bossi, R. and Braga, P. (1985). Influence of surfactant on mucociliary

transport. Eur. J. Respir. Dis. Suppl. 142, 71-76.Amann, R. I., Binder, B. J., Olson, R. J., Chisholm, S. W., Devereux, R. and

Stahl, D. A. (1990). Combination of 16S rRNA-targeted oligonucleotide probeswith flow cytometry for analyzing mixed microbial populations. Appl. Environ.Microbiol. 56, 1919-1925.

Ballard, S. T., Parker, J. C. and Hamm, C. R. (2006). Restoration of mucociliarytransport in the fluid-depleted trachea by surface-active instillates. Am. J. Respir.Cell Mol. Biol. 34, 500-504.

Bartlett, J. A., Albertolle, M. E., Wohlford-Lenane, C., Pezzulo, A. A., Zabner, J.,Niles, R. K., Fisher, S. J., McCray, P. B., Jr and Williams, K. E. (2013). Proteincomposition of bronchoalveolar lavage fluid and airway surface liquid fromnewborn pigs. Am. J. Physiol. Lung Cell. Mol. Physiol. 305, L256-L266.

Bingle, C. D. and Bingle, L. (2000). Characterisation of the human plunc gene, agene product with an upper airways and nasopharyngeal restricted expressionpattern. Biochim. Biophys. Acta 1493, 363-367.

Bingle, L., Cross, S. S., High, A. S., Wallace, W. A., Devine, D. A., Havard, S.,Campos, M. A. and Bingle, C. D. (2005). SPLUNC1 (PLUNC) is expressed inglandular tissues of the respiratory tract and in lung tumours with a glandularphenotype. J. Pathol. 205, 491-497.

Bingle, L., Barnes, F. A., Cross, S. S., Rassl, D., Wallace, W. A., Campos, M. A.and Bingle, C. D. (2007). Differential epithelial expression of the putative innateimmune molecule SPLUNC1 in cystic fibrosis. Respir. Res. 8, 79.

Campos, M. A., Abreu, A. R., Nlend, M. C., Cobas, M. A., Conner, G. E. andWhitney, P. L. (2004). Purification and characterization of PLUNC from humantracheobronchial secretions. Am. J. Respir. Cell Mol. Biol. 30, 184-192.

Canny, G., Swidsinski, A. and McCormick, B. A. (2006). Interactions of intestinalepithelial cells with bacteria and immune cells: methods to characterize microfloraand functional consequences. Methods Mol. Biol. 341, 17-35.

Casado, B., Pannell, L. K., Iadarola, P. andBaraniuk, J. N. (2005). Identification ofhuman nasal mucous proteins using proteomics. Proteomics 5, 2949-2959.

Casselbrant, M. L., Rubenstein, E., Furman, J. M. and Mandel, E. M. (1995).Effect of otitis media on the vestibular system in children. Ann. Otol. Rhinol.Laryngol. 104, 620-624.

Casselbrant, M. L., Mandel, E. M., Rockette, H. E., Kurs-Lasky, M., Fall, P. A.,Bluestone, C. D. and Ferrell, R. E. (2004). The genetic component of middle eardisease in the first 5 years of life. Arch. Otolaryngol. Head Neck Surg. 130,273-278.

Chu, H.W., Thaikoottathil, J., Rino, J. G., Zhang,G.,Wu, Q., Moss, T., Refaeli, Y.,Bowler, R., Wenzel, S. E., Chen, Z. et al. (2007). Function and regulation ofSPLUNC1 protein in Mycoplasma infection and allergic inflammation. J. Immunol.179, 3995-4002.

Depreux, F. F., Darrow, K., Conner, D. A., Eavey, R. D., Liberman, M. C.,Seidman, C. E. and Seidman, J. G. (2008). Eya4-deficient mice are a model forheritable otitis media. J. Clin. Invest. 118, 651-658.

Di, Y.-P., Harper, R., Zhao, Y., Pahlavan, N., Finkbeiner, W. and Wu, R. (2003).Molecular cloning and characterization of spurt, a human novel gene that isretinoic acid-inducible and encodes a secretory protein specific in upperrespiratory tracts. J. Biol. Chem. 278, 1165-1173.

Di, Y. P., Tkach, A. V., Yanamala, N., Stanley, S., Gao, S., Shurin, M. R., Kisin,E. R., Kagan, V. E. and Shvedova, A. (2013). Dual acute proinflammatory andantifibrotic pulmonary effects of short palate, lung, and nasal epithelium clone-1after exposure to carbon nanotubes. Am. J. Respir. Cell Mol. Biol. 49, 759-767.

Ehrlich, G. D., Veeh, R., Wang, X., Costerton, J. W., Hayes, J. D., Hu, F. Z.,Daigle, B. J., Ehrlich, M. D. and Post, J. C. (2002). Mucosal biofilm formation onmiddle-ear mucosa in the chinchilla model of otitis media. JAMA 287, 1710-1715.

Fornander, L., Graff, P., Wåhlen, K., Ydreborg, K., Flodin, U., Leanderson, P.,Lindahl, M. and Ghafouri, B. (2013). Airway symptoms and biological markers innasal lavage fluid in subjects exposed to metalworking fluids. PLoS ONE 8,e83089.

Gakhar, L., Bartlett, J. A., Penterman, J., Mizrachi, D., Singh, P. K., Mallampalli,R. K., Ramaswamy, S. and McCray, P. B.Jr. (2010). PLUNC is a novel airwaysurfactant protein with anti-biofilm activity. PLoS ONE 5, e9098.

Gally, F., Di, Y. P., Smith, S. K., Minor, M. N., Liu, Y., Bratton, D. L., Frasch, S. C.,Michels, N. M., Case, S. R. and Chu, H. W. (2011). SPLUNC1 promotes lunginnate defense against Mycoplasma pneumoniae infection in mice. Am. J. Pathol.178, 2159-2167.

507

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms

Gawron, W., Pospiech, L. and Orendorz-Fraczkowska, K. (2004). An evaluationof postural stability and the effects of middle-ear drainage on vestibulo-spinalreflexes of children with chronic otitis media with effusion. Int. J. Pediatr.Otorhinolaryngol. 68, 1175-1179.

Ghafouri, B., Ståhlbom, B., Tagesson, C. and Lindahl, M. (2002). Newly identifiedproteins in human nasal lavage fluid from non-smokers and smokers using two-dimensional gel electrophoresis and peptide mass fingerprinting. Proteomics 2,112-120.

Ghafouri, B., Tagesson, C. and Lindahl, M. (2003a). Mapping of proteins in humansaliva using two-dimensional gel electrophoresis and peptide mass fingerprinting.Proteomics 3, 1003-1015.

Ghafouri, B., Kihlstrom, E., Ståhlbom, B., Tagesson, C. and Lindahl, M. (2003b).PLUNC (palate, lung and nasal epithelial clone) proteins in human nasal lavagefluid. Biochem. Soc. Trans. 31, 810-814.

Ghafouri, B., Kihlstrom, E., Tagesson, C. and Lindahl, M. (2004). PLUNC inhuman nasal lavage fluid: multiple isoforms that bind to lipopolysaccharide.Biochim. Biophys. Acta 1699, 57-63.

Gibson-Corley, K. N., Olivier, A. K. and Meyerholz, D. K. (2013). Principles forvalid histopathologic scoring in research. Vet. Pathol. 50, 1007-1015.

Grocott, R. G. (1955). A stain for fungi in tissue sections and smears using Gomori’smethenamine-silver nitrate technic. Am. J. Clin. Pathol. 25, 975-979.

Hall-Stoodley, L., Hu, F. Z., Gieseke, A., Nistico, L., Nguyen, D., Hayes, J.,Forbes, M., Greenberg, D. P., Dice, B., Burrows, A. et al. (2006). Directdetection of bacterial biofilms on the middle-ear mucosa of children with chronicotitis media. JAMA 296, 202-211.

Hardisty-Hughes, R. E., Tateossian,H., Morse, S.A., Romero,M. R., Middleton,A.,Tymowska-Lalanne, Z., Hunter, A. J., Cheeseman, M. and Brown, S. D. M.(2006). A mutation in the F-box gene, Fbxo11, causes otitis media in the Jeffmouse. Hum. Mol. Genet. 15, 3273-3279.

Hernandez, M., Leichtle, A., Pak, K., Ebmeyer, J., Euteneuer, S., Obonyo, M.,Guiney, D. G., Webster, N. J., Broide, D. H., Ryan, A. F. et al. (2008). Myeloiddifferentiation primary response gene 88 is required for the resolution of otitismedia. J. Infect. Dis. 198, 1862-1869.

Hoa, M., Syamal, M., Sachdeva, L., Berk, R. and Coticchia, J. (2009).Demonstration of nasopharyngeal and middle ear mucosal biofilms in an animalmodel of acute otitis media. Ann. Otol. Rhinol. Laryngol. 118, 292-298.

Kakuta, Y., Sasaki, H. and Takishima, T. (1991). Effect of artificial surfactant onciliary beat frequency in guinea pig trachea. Respir. Physiol. 83, 313-321.

Kim, C.-H., Kim, K., Jik Kim, H., Kook Kim, J., Lee, J.-G. and Yoon, J.-H. (2006).Expression and regulation of PLUNC in human nasal epithelium. ActaOtolaryngol. 126, 1073-1078.

Klausen, O., Moller, P., Holmefjord, A., Reisaeter, S. and Asbjornsen, A. (2000).Lasting effects of otitis media with effusion on language skills and listeningperformance. Acta Otolaryngol. 120, 73-76.

Larsen, K., Madsen, L. B. and Bendixen, C. (2005). Porcine SPLUNC1: molecularcloning, characterization and expression analysis. Biochim. Biophys. Acta 1727,220-226.

LeClair, E. E., Nguyen, L., Bingle, L., MacGowan, A., Singleton, V., Ward, S. J.and Bingle, C. D. (2001). Genomic organization of the mouse plunc gene andexpression in the developing airways and thymus. Biochem. Biophys. Res.Commun. 284, 792-797.

Leichtle, A., Hernandez, M., Pak, K., Yamasaki, K., Cheng, C.-F., Webster, N. J.,Ryan, A. F. and Wasserman, S. I. (2009). TLR4-mediated induction of TLR2signaling is critical in the pathogenesis and resolution of otitis media. InnateImmun. 15, 205-215.

Li, X., Xu, L., Li, J., Li, B., Bai, X., Strauss, J. F., III, Zhang, Z. andWang, H. (2014).Otitis media in sperm-associated antigen 6 (spag6)-deficient mice. PLoS ONE 9,e112879.

Liu, Y., Di, M. E., Chu, H. W., Liu, X., Wang, L., Wenzel, S. and Di, Y. P. (2013a).Increased susceptibility to pulmonary Pseudomonas infection in Splunc1knockout mice. J. Immunol. 191, 4259-4268.

Liu, Y., Bartlett, J. A., Di, M. E., Bomberger, J. M., Chan, Y. R., Gakhar, L.,Mallampalli, R. K., McCray, P. B., Jr and Di, Y. P. (2013b). SPLUNC1/BPIFA1contributes to pulmonary host defense against Klebsiella pneumoniae respiratoryinfection. Am. J. Pathol. 182, 1519-1531.

Lukinskiene, L., Liu, Y., Reynolds, S. D., Steele, C., Stripp, B. R., Leikauf, G. D.,Kolls, J. K. and Di, Y. P. (2011). Antimicrobial activity of PLUNC protects againstPseudomonas aeruginosa infection. J. Immunol. 187, 382-390.

MacArthur, C. J., Hefeneider, S. H., Kempton, J. B. and Trune, D. R. (2006). C3H/HeJ mouse model for spontaneous chronic otitis media. Laryngoscope 116,1071-1079.

McGillivary, G. and Bakaletz, L. O. (2010). The multifunctional host defensepeptide SPLUNC1 is critical for homeostasis of the mammalian upper airway.PLoS ONE 5, e13224.

Musa,M.,Wilson,K.,Sun,L.,Mulay,A.,Bingle,L.,Marriott,H.M., LeClair,E.E. andBingle, C. D. (2012). Differential localisation of BPIFA1 (SPLUNC1) and BPIFB1(LPLUNC1) in the nasal and oral cavities of mice. Cell Tissue Res. 350, 455-464.

Noben-Trauth, K. and Latoche, J. R. (2011). Ectopic mineralization in the middleear and chronic otitis media with effusion caused by RPL38 deficiency in the Tail-short (Ts) mouse. J. Biol. Chem. 286, 3079-3093.

O’Brien, M. A., Prosser, L. A., Paradise, J. L., Ray, G. T., Kulldorff, M., Kurs-Lasky, M., Hinrichsen, V. L., Mehta, J., Colborn, D. K. and Lieu, T. A. (2009).New vaccines against otitis media: projected benefits and cost-effectiveness.Pediatrics 123, 1452-1463.

Parkinson, N., Hardisty-Hughes, R. E., Tateossian, H., Tsai, H.-T., Brooker, D.,Morse, S., Lalane, Z., MacKenzie, F., Fray, M., Glenister, P. et al. (2006).Mutation at the Evi1 locus in Junbo mice causes susceptibility to otitis media.PLoS Genet. 2, e149.

Post, J. C. (2001). Direct evidence of bacterial biofilms in otitis media. Laryngoscope111, 2083-2094.

Saferali, A., Obeidat, M., Berube, J.-C., Lamontagne, M., Bosse, Y., Laviolette, M.,Hao, K., Nickle, D. C., Timens, W., Sin, D. D. et al. (2015). Polymorphismsassociated with expression of BPIFA1/BPIFB1 and lung disease severity in cysticfibrosis. Am. J. Respir. Cell Mol. Biol. [Epub ahead of print] DOI: 10.1165/rcmb.2014-0182OC.

Sayeed, S., Nistico, L., St Croix, C. and Di, Y. P. (2013). Multifunctional role ofhuman SPLUNC1 in Pseudomonas aeruginosa infection. Infect. Immun. 81,285-291.

Seshadri, S., Lin, D. C., Rosati, M., Carter, R. G., Norton, J. E., Suh, L., Kato, A.,Chandra, R. K., Harris, K. E., Chu, H. W. et al. (2012). Reduced expression ofantimicrobial PLUNC proteins in nasal polyp tissues of patients with chronicrhinosinusitis. Allergy 67, 920-928.

Stoltz, D. A., Meyerholz, D. K., Pezzulo, A. A., Ramachandran, S., Rogan, M. P.,Davis, G. J., Hanfland, R. A., Wohlford-Lenane, C., Dohrn, C. L., Bartlett, J. A.et al. (2010). Cystic fibrosis pigs develop lung disease and exhibit defectivebacterial eradication at birth. Sci. Transl. Med. 2, 29ra31.

Sung, Y. K., Moon, C., Yoo, J.-Y., Moon, C., Pearse, D., Pevsner, J. and Ronnett,G. V. (2002). Plunc, a member of the secretory gland protein family, is up-regulated in nasal respiratory epithelium after olfactory bulbectomy. J. Biol. Chem.277, 12762-12769.

Tateossian, H., Morse, S., Parker, A., Mburu, P., Warr, N., Acevedo-Arozena, A.,Cheeseman, M., Wells, S. and Brown, S. D. (2013). Otitis media in the Tgifknockout mouse implicates TGFbeta signalling in chronic middle ear inflammatorydisease. Hum. Mol. Genet. 22, 2553-2565.

Teele, D. W., Klein, J. O. and Rosner, B. (1989). Epidemiology of otitis mediaduring the first seven years of life in children in greater Boston: a prospective,cohort study. J. Infect. Dis. 160, 83-94.

Teele, D. W., Klein, J. O., Chase, C., Menyuk, P. and Rosner, B. A. (1990). Otitismedia in infancy and intellectual ability, school achievement, speech, andlanguage at age 7 years. Greater Boston Otitis Media Study Group. J. Infect. Dis.162, 685-694.

Thaikoottathil, J. V., Martin, R. J., Di, P. Y., Minor, M., Case, S., Zhang, B., Zhang,G., Huang, H. and Chu, H. W. (2012). SPLUNC1 deficiency enhances airwayeosinophilic inflammation in mice. Am. J. Respir. Cell Mol. Biol. 47, 253-260.

Thornton, R. B., Rigby, P. J., Wiertsema, S. P., Filion, P., Langlands, J., Coates,H. L., Vijayasekaran, S., Keil, A. D. and Richmond, P. C. (2011). Multi-speciesbacterial biofilm and intracellular infection in otitis media. BMC Pediatr. 11, 94.

Uhari, M., Mantysaari, K. and Niemela, M. (1996). A meta-analytic review of therisk factors for acute otitis media. Clin. Infect. Dis. 22, 1079-1083.

Vargas, P. A., Speight, P. M., Bingle, C. D., Barrett, A. W. and Bingle, L. (2008).Expression of PLUNC family members in benign and malignant salivary glandtumours. Oral Dis. 14, 613-619.

Voronina, V. A., Takemaru, K.-I., Treuting, P., Love, D., Grubb, B. R., Hajjar,A. M., Adams, A., Li, F.-Q. and Moon, R. T. (2009). Inactivation of Chibby affectsfunction of motile airway cilia. J. Cell Biol. 185, 225-233.

Weston, W. M., LeClair, E. E., Trzyna, W., McHugh, K. M., Nugent, P., Lafferty,C. M., Ma, L., Tuan, R. S. and Greene, R. M. (1999). Differential displayidentification of plunc, a novel gene expressed in embryonic palate, nasalepithelium, and adult lung. J. Biol. Chem. 274, 13698-13703.

Wheeler, T. T., Haigh, B. J., McCracken, J. Y., Wilkins, R. J., Morris, C. A. andGrigor, M. R. (2002). The BSP30 salivary proteins from cattle, LUNX/PLUNC andvon Ebner’s minor salivary gland protein are members of the PSP/LBPsuperfamily of proteins. Biochim. Biophys. Acta 1579, 92-100.

Zhou, H.-D., Li, X.-L., Li, G.-Y., Zhou, M., Liu, H.-Y., Yang, Y.-X., Deng, T., Ma, J.and Sheng, S.-R. (2008). Effect of SPLUNC1 protein on the Pseudomonasaeruginosa and Epstein-Barr virus. Mol. Cell. Biochem. 309, 191-197.

508

RESEARCH ARTICLE Disease Models & Mechanisms (2015) 8, 501-508 doi:10.1242/dmm.019646

Disea

seModels&Mechan

isms