In this report we demonstrate marked strain heterogeneity of respiratory leukocyte subsets in five major inbred mouse strains. To our knowledge this is the first analysis that has quantitated DC subsets, macrophages, classical lymphocyte subsets, Treg subsets, innate lymphocytes including γδ T cells and linneg leukocytes in the respiratory tract. Environmental effects are unlikely to account for these differences, since specific-pathogen free age-matched animals were received from the same provider and housed under specific-pathogen free conditions. The enumerated leukocyte subsets are essential for different immunological tasks and therefore it is possible that the observed strain heterogeneity influences respiratory immunity to different pathogens.
With respect to respiratory DC subsets we found a striking elevation of respiratory pDC in 129SV mice in comparison to all other strains. These results extend the finding of Asselin-Paturel et al., and Nakano et al. who reported significantly elevated pDC frequencies in spleen and blood of 129SV mice
[20, 21]. Since respiratory pDC represent a key DC subset responding to viral infection and additionally play an immunoregulatory role by preventing airway hyperresponsiveness, higher pDC numbers may influence pathogenesis of respiratory allergy and viral infection in 129SV mice versus other strains. With respect to respiratory CD103+ DC our results indicated that C3H mice exhibited significantly higher numbers in comparison to 129Sv and DBA mice. Given the critical role of respiratory CD103+ DC for activation of naive CD8+ T killer cells in respiratory viral infections
[5, 22], different CD103+ DC numbers may impact the strain-dependent capacity to generate viral CD8+ T cell immunity. Moreover, differences have also been reported according to DC endocytosis receptor expression, such as mannose receptor
 and CD207 (langerin, a c-type specific lectin)
 highlighting separate levels of strain heterogeneity. In line with these reports, our analysis of respiratory leukocytes indicated significant strain differences to produce TNF-α and IL-10 that is partially dependent on the selected Toll-like receptor ligand and the activated leukocyte subset. Our results revealed that respiratory leukocytes of BALB/c mice produced the highest TNF-α levels both after TLR4 stimulation through LPS and after T-cell stimulation through CD3/CD28 mAbs. These results support the finding of Gosselin et al. demonstrating that resistance of BALB/c mice to pseudomonas aeruginosa is dependent on increased TNF-α production
In addition to significant differences of DC subsets and cytokine production capacity we found marked differences of respiratory Foxp3+ Treg frequencies across different inbred strains. In general terms, BALB/c and 129SV mice exhibited significantly higher respiratory Treg numbers than C3H and DBA mice. Respiratory Treg elevation was evident with respect to four different Treg subsets, namely CD4+ CD25high, CD4+ CD25high Foxp3+, CD4+ CD25high Foxp3+ CD39 and CD4+ CD25+ CD127low/neg Tregs. Given the important role of Treg in immunoregulation, additional studies are required to dissect the genetic cause und functional relevance of Treg differences in inbred mouse strains.
Both numerical and functional leukocyte strain variabilities are likely to contribute to innate resistance and susceptibility to infection with various pathogens. Examples are the unique susceptibility of DBA/2 mice to pulmonary tuberculosis as well as resistance of 129S6/SVevTac mice to salmonella typhimurium infection
. Genetic dissection of strain heterogeneity promoted the discovery of several highly relevant immune response genes and modifiers underlining the importance of inbred mouse models to understand innate and adaptive immunity
Furthermore, knowledge of significant numerical differences of respiratory leukocyte subsets is also practically relevant for the planning and setup of ex vivo experiments. Several functionally distinct cells, like pDC, represent extremely rare subsets in the lung and therefore ex vivo analysis of sufficient numbers of purified respiratory pDC is technically difficult. Accordingly, our results may help identification of the suitable inbred mouse strain with the highest pDC numbers. Given pDC frequencies in the five major inbred strains ranging from 0.15% (mean of C3H mice) to 0.58% (mean of 129SV mice) of leukocytes while having similar total leukocyte counts, selection of the 129SV mouse would increase the pDC yield 3 to 4-fold and would save both animal numbers and procedural time. Similar conclusions can be drawn with respect to sorting of rare Linneg leukocytes populations from the lungs for ex vivo experiments. Our results indicated that C57BL/6 mice have only few Linneg CD90neg leukocytes (mean 0.34% of leukocytes) in comparison to C3H mice (mean 1.25 %).
In more general terms, our study indicated characteristic respiratory immunophenotypes of inbred mouse strain: BALB/c mice are rich of eosinophils and Treg cells; C57BL/6 mice are rich of B cells and low in eosinophils. In contrast, 129SV mice are high in pDC and Tregs and C3H mice are high in CD103+ DC, NK cells and γδ T cells. These characteristic immunophenotypes may help the selection of strains for respiratory disease models, e.g. if the analysis and purification of respiratory pDC during viral pneumonia is required then 129SV mice would be a suitable model. However, besides different respiratory immunophenotypes, many other functionally relevant strain differences have been reported. Therefore, careful strain selection based on disease model and published strain differences may help to select the best mouse strain representing human phenotypes. In this context, DeVooght et al. recently reported, that choice of mouse strain significantly influenced the outcome in a model of chemical-induced asthma. They reported that the human phenotype of chemical-induced asthma was best reproduced in BALB/c when compared to 6 other strains
Inbred mouse strain differences have been reported with respect to different aspects of inflammation. With respect to the respiratory irritant ozone, C57BL/6 mice have been reported relatively susceptible when compared to relatively resistant C3H mice
. With respect to chronic pseudomonas aeruginosa infection, BALB/c mice are resistant in comparison to highly susceptible DBA/2 mice
. Similarly, in a pneumovirus model, BALB/c and C57BL/6 strains were reported to be relatively resistant in comparison to DBA/2 or 129SV mice
In summary, the present study highlights marked numerical differences of several important leukocytes subsets in the lungs of major inbred mouse strains. Although our results are primarily descriptive, it may provide the basis for additional functional studies under inflammatory in vivo conditions. Variation of respiratory immune responses and cytokine production capacity may partially be related to strain differences in respiratory leukocyte composition and function.