Enhanced response to pulmonary Streptococcus pneumoniae infection is associated with primary ciliary dyskinesia in mice lacking Pcdp1 and Spef2

Background Lower airway abnormalities are common in patients with primary ciliary dyskinesia (PCD), a pediatric syndrome that results from structural or functional defects in motile cilia. Patients can suffer from recurrent bacterial infection in the lung, bronchiectasis, and respiratory distress in addition to chronic sinusitis, otitis media, infertility, and laterality defects. However, surprisingly little is known about the pulmonary phenotype of mouse models of this disorder. Results The pulmonary phenotype of two mouse models of PCD, nm1054 and bgh, which lack Pcdp1 and Spef2, respectively, was investigated by histological and immunohistochemical analysis. In addition, both models were challenged with Streptococcus pneumoniae, a common respiratory pathogen found in the lungs of PCD patients. Histopathological analyses reveal no detectable cellular, developmental, or inflammatory abnormalities in the lower airway of either PCD model. However, exposure to S. pneumoniae results in a markedly enhanced inflammatory response in both models. Based on analysis of inflammatory cells in bronchoalveolar lavage fluid and flow cytometric analysis of cytokines in the lung, the bgh model shows a particularly dramatic lymphocytic response by 3 days post-infection compared to the nm1054 model or wild type animals. Conclusions Defects in ciliary motility result in a severe response to pulmonary infection. The PCD models nm1054 and bgh are distinct and clinically relevant models for future studies investigating the role of mucociliary clearance in host defense.


Background
Streptococcus pneumoniae, an anaerobic, gram-positive bacterium, is the primary cause of bacterial pneumonia and a major cause of sinusitis, otitis media, and bacterial meningitis [1][2][3]. Symptoms of pneumococcal pneumonia commonly include a persistent cough, excess production of sputum, shortness of breath, and fever, with most patients also showing evidence of a pulmonary infiltrate [1,3]. Despite the effectiveness of antibiotics and immunizations, pneumococcal pneumonia continues to result in respiratory failure, hospitalization, and mortality in many patients, particularly young children, the elderly, and individuals with additional underlying medical conditions [1,2,4]. It is estimated that over one million childhood deaths are associated with pneumococcal pneumonia each year [3].
The response to respiratory pathogens involves mucus secretion, clearance by ciliary motility, and an inflammatory response [5][6][7]. Because of the role of cilia in this response, pulmonary disease is a major component of primary ciliary dyskinesia (PCD), an autosomal recessive, pediatric syndrome that results from defects in ciliary and flagellar motility and is commonly characterized by chronic upper and lower respiratory infection, otitis media, male infertility, and situs inversus, as well as less frequent associations of hydrocephalus and female infertility [8][9][10][11][12][13]. Because mucociliary clearance is critical for lung maintenance and host defense, PCD patients typically have a variety of pulmonary defects as early as the neonatal period [5,[14][15][16][17][18]. Several large or longitudinal clinical studies analyzing pulmonary function and disease in diagnosed PCD patients found a variety of common conditions [19][20][21][22][23][24][25][26][27][28][29][30][31][32][33]. Patients typically suffered from recurrent pneumonia, a chronic cough, excess sputum production, and severe bronchiectasis. Neonatal respiratory distress was frequently reported in infants, while airway obstruction, atelectases, and a progressive, irreversible decline in pulmonary function were commonly seen in older patients. Bacteria were identified in the sputum from most patients, with Haemophilus influenzae, Streptococcus pneumoniae, and Staphylococcus aureus being the most common. Pseudomonas aeruginosa, a pathogen commonly found in the lungs of cystic fibrosis patients, was only rarely identified in PCD patients.
Despite the strong correlation of lower airway disease with PCD in patients, there is a surprising paucity of information about lung disease in mouse models of PCD or the usefulness of those models for developing improved therapies for lung disease. Although several mouse models with PCD or PCD-related phenotypes have been described [11,12], pulmonary defects have only been reported in three. Studies involving lower airway disease are complicated in some models that exhibit early lethality due to hydrocephalus or laterality defects, but several other models have been described on genetic backgrounds that are less susceptible to severe hydrocephalus and therefore live a longer lifespan [12]. Mice lacking both sperm associated antigens 6 (Spag6) and 16 L (Spag16L), which have been shown to interact in the cilium, have severe lower airway disease characterized by pneumonia, atelectases, and hemorrhage [34,35]. Despite the presence of other PCD phenotypes, respiratory abnormalities were not reported in mutants lacking either individual protein [36,37], possibly due to the functional relationship between the two proteins. Loss of Spag17 also results in PCD, with homozygous mutants exhibiting an accumulation of fluid in their lungs, damage to the alveolar cells, and a failure to thrive [38]. Severe lower airway disease was also observed in mice lacking the Wnt inhibitor Chibby and is characterized by abnormal lung pathology, reduced proliferation and differentiation of respiratory epithelial cells, and an inability to clear P. aeruginosa after infection [39,40]. While these studies indicate that mouse models may be valuable tools for better understanding and treating the array of pulmonary phenotypes associated with PCD, they are limited in number and have not yet explored the response to the respiratory pathogens most commonly observed in human PCD patients.
PCD has been previously reported in the nm1054 and big giant head (bgh) mouse models and is characterized, in both cases, by hydrocephalus, male infertility, and upper respiratory abnormalities [41,42]. The phenotypes in the nm1054 and bgh models are caused by loss of ciliary proteins primary ciliary dyskinesia protein 1 (Pcdp1) and sperm flagellar protein 2 (Spef2), respectively [41,42]. The PCD phenotypes demonstrate that both genes are required for proper ciliary function, although mutations have not yet been identified in human PCD patients. In the nm1054 mutant, there is a marked accumulation of mucus in the upper airway [41], while the bgh mutant has a similar accumulation of mucus accompanied by an infiltration of neutrophils [42], which is indicative of an inflammatory response in the airway [17,43]. The respiratory epithelial cilia have a normal ultrastructure in both models [41,42]. However, the ciliary beat frequency (CBF) is decreased by approximately 25% in the nm1054 mouse and approximately 17% in the bgh mouse [41,42], likely resulting in the inability to clear mucus in the airway. Given these findings, it is expected that a defect in mucociliary clearance would also affect the lower airway and predispose these models to severe respiratory infection.
In this study, we conducted a comprehensive histopathological analysis of the lungs of both the nm1054 and the bgh model. Despite the pathological findings in the upper airway and the pulmonary abnormalities observed in other PCD models, there are no major cellular, developmental, or inflammatory abnormalities in the lower airway of either the nm1054 or the bgh mutant. In contrast, there is a markedly enhanced acute inflammatory response in both models after infection with S. pneumoniae, with the bgh model showing a particularly dramatic lymphocytic response within 3 days of infection. These findings demonstrate that defects in ciliary motility in both PCD mutants result in severe pulmonary infection and establish clinically relevant models for future studies involving the role of mucociliary clearance in host defense.

Mice
The nm1054 [41] and bgh [42] lines were maintained on both the C57BL/6 J (B6) and 129S6/SvEvTac (129) backgrounds as previously described. All phenotypic analyses were performed on (B6x129)F1 animals at least 8 weeks old. All procedures were approved by either the Sanford Research/USD or the University of South Dakota Institutional Animal Care and Use Committee.

Histology
Three wild type, three nm1054, and three bgh lungs were inflated with 10% phosphate-buffered formalin using an 18-gauge catheter inserted into the trachea. The lungs were removed from the body cavity and immersion fixed in 10% phosphate-buffered formalin. The fixed tissue was embedded in paraffin wax on a Leica 300 ASP tissue processor, serially sectioned through the airway, and stained with hematoxylin and eosin on a Sakura Tissue-Tek automated stainer. Sections were analyzed qualitatively by light microscopy on an Olympus IX71 microscope.

Immunohistochemistry
Sections from the three wild type, three nm1054, and three bgh lungs used for histology were stained with primary and secondary antibodies using the BenchMark XT automated slide staining system (Ventana Medical Systems, Inc.). The mouse acetylated tubulin antibody (Sigma Aldrich) was used at 1:6,000 and incubated for 30 minutes at 37°C. The goat Clara cell secretory protein (CCSP) (courtesy of Barry Stripp, Duke University Medical Center, Durham, NC, USA) and rabbit prosurfactant protein B (Chemicon) antibodies were used at 1:250 and incubated for 1 hour at 37°C. The Syrian hamster T1α glycoprotein 38 kD mucin-like antibody (Hybridoma Bank) was used at 1:1,000 and incubated for 1 hour at 37°C. All secondary antibodies were from Jackson ImmunoResearch. Biotin SP-conjugated AffiniPure goat anti-rabbit and goat anti-Syrian hamster secondary antibodies were used at 1:1,000, and the Biotin SPconjugated AffiniPure donkey anti-mouse and donkey anti-goat secondary antibodies were used at 1:500. Staining was detected using the Ventana iView DAB kit, and the slides were counterstained with hematoxylin and qualitatively analyzed by light microscopy on an Olympus IX71 microscope.

Alveolar chord length measurement
Images of lung sections from the three wild type, three nm1054, and three bgh mice analyzed by histology above were taken on a Nikon 90i microscope using a 10X objective and at a 1,600 × 1,200 resolution. Alveolar cord length was measured as previously described using the Image J software [44,45], and statistical significance was determined using the Student's t-test.

Bacterial infection
Mice anesthetized with isoflurane were infected with the TIGR4 strain of S. pneumoniae (courtesy of Carlos Orihuela, University of Texas Health Science Center, San Antonio, TX, USA) by nasal instillation with a total of 5 × 10 4 CFU in 10 μL inoculated into the left nostril. Body weights were collected for all mice prior to inoculation and at the time of death. Nasal carriage was assessed in surviving mice prior to euthanasia at 1, 2, or 3 days post-infection by instilling 25 μL of phosphate buffered saline into the left nostril, recovering, serially diluting, and plating on 10% sheep red blood cell agar [46]. Lung titer was assessed following euthanasia at 1, 2, or 3 days post-infection by homogenizing the left lung with an IKA T10 Basic S1 disperser, serially diluting the homogenate, and plating on 10% sheep red blood cell agar as previously described. Statistical significance was determined using the Student's t-test. The following numbers of mice were analyzed for nasal carriage and lung titer: on day 1, four wild type, four nm1054, and four bgh; at day 2, four wild type, four nm1054, and eight bgh; and at day 3, four wild type, four nm1054, and eight bgh.

Bronchoalveolar lavage
Bronchoalveolar lavage fluid was isolated from four uninfected wild type, four uninfected nm1054, and three uninfected bgh mice, as well as three wild type, three nm1054, and three bgh mice 2 and 3 days after infection with S. pneumoniae. Lungs were inflated with 1 mL of Hank's Balanced Salt Solution (HBSS) media containing 5 mM EDTA using the inflation procedure described above, and the fluid was collected through the same catheter using a separate syringe. Inflation and collection was repeated six times. Lavage fluid was centrifuged at 400 × g for ten minutes at 4°C. The supernatant was removed, and the cells were pooled and resuspended in 200 μL of HBSS media containing 5 mM EDTA. The cells were loaded onto slides and spun in a StatSpin Cytofuge 2 cytocentrifuge (Iris Sample Processing, Inc.) at 600 rpm for 6 minutes. The slides were fixed with methanol, stained with the Camco differential stain kit, and visualized on an Olympus IX71 microscope.

Cytokine profiling
Cytokines were analyzed in lungs from uninfected mice (day 0) and 1, 2, or 3 days after infection with S. pneumoniae. Right lungs were homogenized as described above and centrifuged at 5,000 rpm for 1 minute. Cytokines in the supernatant were analyzed using the Cytometric Bead Array Mouse Th1/Th2/Th17 Cytokine Kit (BD Biosciences) and an Accuri C6 flow cytometer (BD Biosciences) following manufacturer's instructions. Cytokine levels were normalized to internal standards, and a standard curve was generated using the GraphPad Prism 4 software. Statistical significance was determined using the Student's t-test. The following numbers of mice were analyzed for cytokines: on day 0, four wild type, four nm1054, and four bgh; on day 1, four wild type, four nm1054, and four bgh; on day 2, four wild type, four nm1054, and eight bgh; and on day 3, four wild type, four nm1054, and eight bgh.

Results
Absence of lower airway pathology in nm1054 and bgh lungs Because pulmonary abnormalities are common in patients with PCD, we investigated the lung phenotype in both the nm1054 and bgh mutants. Histological analysis of lung architecture revealed no significant abnormalities or any major damage to tissue structure in either model ( Figure 1A-C). Higher magnification images of the epithelia lining the airway demonstrate that the epithelial cilia are intact in both mutants ( Figure 1D-F). To confirm these histological findings, a more detailed analysis of tissue architecture was enabled by immunohistochemical analysis of markers specific to airway cell types. Acetylated tubulin, which is expressed in the ciliated epithelial cells lining the airway, is properly expressed and localized to the cilia in both mutants (Figure 2A-C), confirming the normal appearance of respiratory epithelial cilia. Expression of CCSP, a marker of the Clara cells that also line the airway, shows a normal appearance and distribution of this epithelial cell type ( Figure 2D-F). The alveolar epithelial cells were detected with antibodies to T1α and Prosurfactant protein B, which are markers for the type I and type II alveolar cells, respectively. Proper expression was detected for both T1α ( Figure 2G-I) and Prosurfactant protein B ( Figure 2J-L) in both PCD mutants, indicating that there are no significant abnormalities in the alveolar cell types. In addition, measurements of alveolar chord length showed no significant differences in the alveolar airspace in either the nm1054 or the bgh lung (Additional file 1). Prosurfactant protein B is also expressed in the Clara cells, and normal expression in both mutants confirms the presence and normal distribution of this cell type. Proper expression of these cellspecific markers indicates that there are no developmental defects or secondary damage to the airway surface due to defective mucociliary clearance.
Because previous histological analysis of the maxillary sinus cavity of bgh mutants indicated the presence of an inflammatory response [42], we investigated whether an inflammatory response was active in the lower airway of both models. Presence or absence of inflammatory cells in the lower airway of nm1054 and bgh mutants was assessed in bronchoalveolar lavage (BAL) fluid. The inflammatory cells in wild type lungs consist primarily of resident and circulatory macrophages ( Figure 3A; Table 1). This is indicative of a healthy respiratory system, where macrophages typically comprise 95% of inflammatory cells [43]. Consistent with an absence of abnormal pathology in the mutant lungs, only resident alveolar macrophages were detected in BAL fluid from nm1054 ( Figure 3B; Table 1) and bgh ( Figure 3C; Table 1) animals. There was no evidence of polymorphonuclear leukocytes or lymphocytes in the airway of either mutant. These findings suggest that, despite evidence of mucus accumulation and an inflammatory response in the upper airway, a defect in respiratory epithelial ciliary motility does not result in a detectable phenotype in the lungs of either the nm1054 or the bgh mutant.

Response to Streptococcus pneumoniae infection in nm1054 and bgh mutants
The surprising absence of a pulmonary phenotype in nm1054 and bgh mutants could be due to a milder defect in mucociliary clearance in the lower airway. It is also possible that the lower airway of these models may be impervious to any significant damage in a clean environment without a respiratory challenge. To investigate whether nm1054 and bgh mutants might exhibit a lower airway phenotype in response to a challenge, we infected both models with S. pneumoniae, a common respiratory pathogen seen in the airway of human PCD patients. Infection appeared to result in increased morbidity in both mutants, with one nm1054 and one bgh mouse dying approximately 3 days after infection. Body weights for infected wild type, nm1054, and bgh mice remained largely unchanged for mice euthanized for analysis of inflammation up to 3 days post-infection (Additional file 2). However, those mice that died as a result of infection exhibited weight loss and wasting prior to death, with the nm1054 mouse weighing approximately 76% of its initial weight prior to infection and the bgh mouse weighing approximately 88% of its initial weight, suggesting a substantial decline in health after infection. In contrast, none of the infected wild type controls exhibited any sign of abnormal health or expired prior to euthanasia.
To investigate the effect of ciliary dysfunction, wild type, nm1054, and bgh mice were analyzed for an acute inflammatory response during the 3 days after infection.  Table 1). In contrast to the wild type response, the nm1054 BAL fluid shows a substantial increase in the number of activated macrophages and T lymphocytes on day 2 ( Figure 4E; Table 1). By day 3, there is almost an equivalent presence of activated macrophages (53%) and T lymphocytes (42%) ( Figure 4H; Table 1), demonstrating an enhanced lymphocytic response to S. pneumoniae infection relative to wild type. The most dramatic response to bacterial challenge was observed in bgh lungs. As in the nm1054 mutant, day 2 shows a shift to activated macrophage and infiltrating T lymphocyte populations ( Figure 4F; Table 1). By day 3, however, T lymphocytes comprise approximately 82% of the inflammatory cell population ( Figure 4I; Table 1). This strong lymphocytic presence demonstrates a particularly aggressive immune response in bgh lungs compared to wild type or nm1054 mice. Several other cell types, including neutrophils, eosinophils, and dendritic cells, were observed in infected lungs at low frequency, but these numbers were negligible compared to the infiltration of macrophages and T lymphocytes.
The acute inflammatory response observed in BAL fluid is supported by flow cytometric analyses of cytokines in lung lysates, which were performed in uninfected lungs and at days 1, 2, and 3 after infection with S. pneumoniae ( Figure 4). The cytokine profiles show a marked response to infection in the bgh mutant. Early infection cytokines interleukin 6 (IL-6) and tumor necrosis factor (TNF),  Cell types were counted in pooled representative fields from three mice in each group in a blinded manner, and approximate percentages of the total number are shown. Percentages less than five are indicated by a dash, and only cell types representing at least 5% of the total cells are included.
which are both secreted by macrophages, showed a statistically significant increase (P <0.05 by Student's t-test) in the bgh lung as early as day 1 ( Figure 4A,B), indicating the onset of an inflammatory response. Both cytokines remained at either significantly higher or elevated levels through day 3 relative to wild type. Lack of statistical significance is likely due to typical variation in immune response within the mouse sampling. In addition, there is either an upward trend or a significant increase in the level of IL-2, interferon gamma (IFNγ), IL-4, and IL-17, all of which are indicators of T lymphocyte activation, by days 2 and 3 ( Figure 4C-F). The elevated levels of lymphocytic cytokines are consistent with the aggressive infiltration of lymphocytes observed in bgh BAL fluid ( Figure 3, Table 1). Finally, an increase in the level of antiinflammatory cytokine IL-10, which is produced by activated T regulatory cells, is observed on days 1 through 3 ( Figure 4G). In contrast to the cytokine profile of bgh lungs, a less striking response is observed in nm1054 lungs. By day 3, a significant increase was only observed for IL-6 ( Figure 4A), which is indicative of an early macrophagic response, and IL-2 ( Figure 4C), which is consistent with the mild lymphocytic response observed in the nm1054 BAL fluid ( Figure 3, Table 1).

Discussion
Although chronic lung disease is commonly associated with PCD in human patients, there are no significant histopathological defects in the lungs of mice lacking either ciliary protein Pcdp1 or Spef2, despite the presence of upper airway abnormalities, male infertility, hydrocephalus, and a decrease in ciliary motility [41,42]. However, infection with S. pneumoniae results in an enhanced acute inflammatory response in both PCD models. It is likely that the defects in ciliary motility result in perturbed mucociliary clearance of the pathogen, which leads to the enhanced immune response. Persistent bacterial infection is indicated by unchanged levels in nasal carriage and lung titer over 3 days after inoculation with S. pneumoniae. The absence of a detectable phenotype in uninfected mutant lungs may in part be due to the housing of these animals in a clean, pathogen-free environment void of any pulmonary challenges. In this environment, the inflammation previously observed in the upper airway of the bgh mutant may be sufficient to protect the lower airway from any significant pathology [42]. Alternatively, absence of an unchallenged pulmonary phenotype could result from a milder defect in mucociliary clearance in the lower airway that is related to the functions of Pcdp1 and Spef2, which is supported by the presence of lung disease in unchallenged mice lacking other ciliary proteins, such as Spag17 or both Spag6 and Spag16L [35,38]. Relatively little is known about the function of either Pcdp1 or Spef2. The Chlamydomonas reinhardtii homolog of Pcdp1, FAP221, was shown to associate with a protein complex that localizes to the central pair apparatus of the cilium and regulates motility in a calciumdependent pathway [47]. Spef2 was shown to associate with intraflagellar transport protein IFT20 in the testis [48], and its likely C. reinhardtii homolog also associates with the central pair apparatus [49]. Absence of laterality defects in both models is likely due to the localization and function of these proteins at the central pair apparatus, which is absent in the nodal cilia required for embryonic left-right patterning [41,42]. It is unknown whether Pcdp1 and Spef2 have related functions in the central pair apparatus, and the mechanisms by which these two proteins regulate mammalian ciliary function remains elusive.
Microscopic analysis of inflammatory cells in the airway and flow cytometric analysis of cytokine levels are consistent with an enhanced acute inflammatory response, particularly in the bgh mutant. The typical response to S. pneumoniae infection involves initial clearance by activated resident macrophages, followed by recruitment of neutrophils, monocytes, and lymphocytes from the bloodstream [2,4,43,50]. Kadioglu et al. showed that, in mice, the number of neutrophils and infiltrating macrophages increases within the first 12 to 24 hours after S. pneumoniae infection, with lymphocytes infiltrating by 48 hours post-infection [51]. Both the nm1054 and the bgh mutant show a shift toward an activated macrophage and infiltrating lymphocytic response by the second day after infection. A milder response is seen in infected wild type animals, where the shift to macrophage activation is not evident until day 3, and the lymphocytic recruitment remains relatively low. By day 3, the bgh animals exhibit a dramatic lymphocytic response, with T lymphocytes comprising more than 80% of the cells isolated from the BAL fluid. A small number of neutrophils were observed in the BAL fluid from each mutant, but this was lower than 5% at all time points. While it is possible that the response in these PCD mutants is primarily macrophagic and lymphocytic, it is more likely that there is an early and brief (See figure on previous page.) Figure 4 Enhanced cytokine levels in response to S. pneumoniae infection. Levels of IL-6 (A), TNF (B), IL-2 (C), IFNγ (D), IL-4 (E), IL-17 (F), and IL-10 (G) were measured in wild type, nm1054, and bgh lung homogenates by flow cytometry using a cytometric bead array mouse Th1/ Th2/Th17 cytokine kit. Mice were infected with S. pneumoniae on day 0, and cytokine levels were measured on days 0, 1, 2, and 3. Cytokine levels are normalized to internal standards. Statistical significance relative to wild type (P <0.05) is indicated by an asterisk. neutrophilic response, after which the number of neutrophils drops dramatically and is replaced by lymphocytes and macrophages by day 2 post-infection.
The acute inflammatory response observed in BAL fluid after infection is supported by quantification of cytokine levels in the lung. There is a statistically significant increase in IL-6 by day 1 in the bgh mutant and by day 3 in the nm1054 mutant compared to wild type. IL-6, which is secreted by resident macrophages and plays a role in both neutrophil recruitment and T cell differentiation, is upregulated in multiple pulmonary inflammatory diseases and is a consistent marker of early inflammation [4,52,53]. In addition, mice lacking IL-6 have an impaired immune response to S. pneumoniae infection, with mutants displaying a higher lung bacterial titer, increased levels of pro-inflammatory and anti-inflammatory cytokines, and increased morbidity [54], further demonstrating that IL-6 plays an important role in early host defense. Therefore, the high levels of IL-6 in nm1054 and bgh lungs indicate the onset of a more persistent inflammatory response in both PCD mutants than infected wild type animals. Increased secretion of TNF, which is secreted by macrophages and, along with IL-6, stimulates production of several other cytokines, is also associated with an early inflammatory response [55]. TNF levels following infection are only increased in the bgh mutant, further suggesting that the immune response is more severe in the bgh mutant than the nm1054 mutant. The increase in TNF levels in the bgh lung is statistically significant on day 1 but loses significance thereafter, likely due to variability observed in TNF levels in the sampling of mutant mice.
IL-2 plays a critical role in driving the lymphocytic response [56]. Consistent with an increased number of lymphocytes in BAL fluid from both PCD mutants, there is a statistically significant increase in IL-2 levels in the bgh mutant by day 2 and in the nm1054 mutant by day 3 relative to infected wild type animals. IL-2 is secreted by naïve T cells and promotes their differentiation into Th1 helper cells, Th2 helper cells, Th17 helper cells, and T regulatory cells, which in turn secrete IFNγ, IL-4, IL-17, and the antiinflammatory cytokine IL-10, respectively [56,57]. A statistically significant increase in IFNγ, IL-4, IL-17, and IL-10 levels was observed in bgh mutants. The increase in IL-10 lost its statistical significance after day 1, but like TNF, this is likely due to variability in the sample population. In contrast, there is no statistically significant change in the levels of any of these T helper or T regulatory cytokines in the nm1054 lung relative to the wild type, providing further evidence that the bgh mice have a more severe lymphocytic response to bacterial infection.
Increases in lymphocytic cytokines in the bgh lung are particularly important given their critical roles in the lymphocytic response. IL-17, which has a pro-inflammatory function in further recruitment of circulatory macrophages and neutrophils, is typically elevated in patients with a variety of pulmonary inflammatory diseases [58,59], and mice lacking the IL-17 receptor A show an enhanced susceptibility to Klebsiella pneumoniae infection and increased morbidity [60]. To prevent excessive damage to the lung, the anti-inflammatory cytokine IL-10, which is secreted by T regulatory cells, plays a critical role in mediating the lymphocytic response by inhibiting naïve T cells, Th1 cells, and Th2 cells [4,[61][62][63]. This role was demonstrated in mice inoculated with both S. pneumoniae and IL-10, which resulted in a higher bacterial titer, decreased levels of proinflammatory cytokines, and increased morbidity [64], while infection with Aspergillus fumigatus in mice lacking IL-10 results in an enhanced inflammatory response and higher levels of pro-inflammatory cytokines in the airway [63]. There is an upward trend in IL-10 levels in bgh mice relative to wild type controls, and despite a lack of statistical significance, these increases are greater than those observed for the Th1, Th2, and Th17 cytokines. This finding suggests that there may be an activation of the anti-inflammatory response to mitigate damage to the lung. Further analyses are required to fully determine the extent of tissue damage in mutant lungs over the course of infection and the potential interventions that can be used to limit this damage in PCD patients.

Conclusions
The findings in this study establish that mice lacking either Pcdp1 or Spef2 are distinct, clinically relevant models with an enhanced lymphocytic response to pulmonary infection with S. pneumoniae, a common respiratory pathogen in patients with PCD. This is particularly significant given that the pulmonary immune response to S. pneumoniae has not been previously investigated in murine models of PCD, and it strongly suggests that further research may be necessary with additional bacterial, viral, and environmental challenges. Future analyses of inflammatory response beyond the acute phase of infection may provide additional evidence of compromised pathogen clearance in these models. The effect of this compromise on pulmonary function, mortality, and susceptibility to infection also warrants further study. Differences in the severity of response to infection in the nm1054 and bgh mutants may be due to differences in Pcdp1 and Spef2 protein function that could have differential effects on mucociliary clearance. Elucidating the biochemical mechanism by which these proteins regulate ciliary motility will ultimately uncover their roles in pathogen clearance and host defense.

Additional files
: Mean alveolar chord length in mutant lungs. Alveolar chord lengths were measured in sections of wild type, nm1054,