Immunohistochemical detection and regulation of α₅nicotinic acetylcholine receptor (nAChR) subunits by FoxA2 during mouse lung organogenesis

C57BL/6 mice were housed and used in accordance with approved IACUC protocols at Brigham Young University. Male and female mice were mated and the discovery of a vaginal plug was identified as embryonic day (E) 0.

A rabbit α₅ polyclonal antibody generated against cytoplasmic epitopes was used at a dilution of 1:800 to identify α₅ nAChR subunits in the lung during development. Immunobotting and ELISAs were used to determine the specificity of the α₅ antibody and it was determined to be effective with tissues embedded in paraffin [18]. A rabbit polyclonal antibody against Clara Cell Secretory Protein (CCSP, Seven Hills Bioreagents, Cincinnati, OH) was used at a dilution of 1:1600. A monoclonal antibody for Fox J1 (Seven Hills BioReagents) was used at a dilution of 1:2000. ATII epithelial cells were specifically identified by staining with a rabbit anti-N-terminal proSP-C polyclonal antibody (1:1000, Seven Hills BioReagents) and ATI cells were localized via staining with a monoclonal hamster anti-mouse antibody raised against T1α at a dilution of 1:2000 (Clone 8.1.1, Developmental Studies Hybridoma Bank, Department of Biology, University of Iowa, Iowa City, IA). Immunohistochemical staining involved six mice per time point and staining for each antibody was conducted on three different slides. Immunostaining for CCSP, proSP-C, T1α, FoxJ1 and α₅ was performed with 5-μm serial sections beginning at E18.5 because this period coincided with elevated α₅ expression and the differentiation status of epithelial cells that express these markers [19, 20]. Staining of serial sections was selected over preferred methods of dual labeling immunofluorescence because specific staining using multiple rabbit polyclonal antibodies in the same slide is not easily reproducible. Sections were deparaffinized, and rehydrated by incubation in 100%, 95%, and 70% ethanol then treated with 3% hydrogen peroxide in methanol for 15 min to quench endogenous peroxidase. Following block in 2.0% normal goat serum in PBS for 2 hr at room temperature, sections were incubated with CCSP, proSP-C, T1α, or α₅ primary antibody at 4°C overnight. Control sections were incubated in blocking serum alone. After overnight incubation with primary antibody, all sections (including controls) were washed and positive staining was detected using biotinylated goat anti-rabbit secondary antibodies and a Vector Elite ABC kit (Vector Laboratories; Burlingame, CA). Development in nickel diaminobenzidine was followed by incubation in Tris-cobalt (which enhances antigen localization), and counterstaining was conducted with nuclear fast red. Sections were dehydrated by incubation in 70%, 95%, and 100% ethanol, washed in three changes of HistoClear (Fisher Scientific, Waltham, MA), and mounted under cover slips with mounting medium. Immunohistochemical staining for FoxJ1 was completed using a "Mouse on Mouse" monoclonal antibody kit (Vector) in accordance with the manufacturer's instructions. Individuals blinded to the antibody used initially imaged the serial sections and co-localization was determined by comparing immunolabeling of α₅ with cells that express CCSP, FoxJ1, proSP-C, or T1α.

0.85-kb of the mouse α₅ promoter was obtained by polymerase chain reaction (PCR), ligated into a pGL4.10 reporter vector (Promega, Madison, WS) and verified by sequencing as described previously [21]. Site-directed mutagenesis of a potential FoxA2 binding site (-488) was performed by using the 0.85-kb reporter construct and the QuickChange™ Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA). The sequence verified mutant reporter contained synthetic oligonucleotides for the desired mutation for FoxA2 (CATTTA→GGGGGG). Functional assays of reporter gene constructs were performed by transient transfection of Beas-2B and A549 cells using FuGENE-6 reagent (Roche, Indianapolis, IN) [21]. Beas-2B is a transformed human bronchiolar epithelial cell line and A549 is a human pulmonary adenocarcinoma cell line characteristic of ATII cells [22]. Transfections included 500 ng pRSV-βgal, 100 ng pGL4.10-0.85-kb α₅, 100-400 ng pCMV-FoxA2 or pCMV-GATA-6 and pcDNA control vector to bring total DNA concentration to 1.4 μg. After 48 hours, plates were scraped and centrifuged, and the cleared supernatant was used for both β-gal and luciferase assays such that assays were normalized for transfection efficiency based on the β-gal activity [19]. Data presented are representative of three different experiments, all performed in triplicate.

Results are presented as the means ± S.D. of six replicate pools per group. Means were assessed by one and two-way analysis of variance (ANOVA). When ANOVA indicated significant differences, student t tests were used with Bonferroni correction for multiple comparisons. Results are representative and those with p values < 0.05 were considered significant.

The distribution of α₅ expression in mouse lung was assessed by immunohistochemistry from E13.5 to PN20. At E13.5 (Figure 1A) and E15.5 (Figure 1B), α₅ was primarily detected in epithelial cells that comprise the primitive conducting airways of the developing lung and only sporadically expressed in mesenchyme. At E18.5 (Figure 1C), and PN1 (Figure 1D), α₅ was predominantly expressed in proximal lung epithelial cells with diminished expression in distal lung epithelium. At PN4 (Figure 1E), α₅ was detected in the distal lung, while staining in the conducting airways was markedly decreased. This shift in α₅ expression from proximal to distal lung epithelium at PN1 and PN4 was also observed at PN10 (Figure 1F). At PN20 (Figure 1G), robust α₅ expression returned to proximal lung epithelium while α₅ localization persisted in the distal lung. No staining was observed in sections stained without primary antibody (Figure 1H).

In order to identify specific cell populations that express α₅, co-localizing immunohistochemistry was performed on serial sections obtained from mice at E18.5 through PN20. During the early saccular period (E18.5), α₅ was co-expressed with FoxA2, a general marker of primitive respiratory and airway epithelium in the proximal and distal lung (Figure 2A, B). Co-expression of α₅ and FoxA2 was also detected in proximal and distal pulmonary epithelium at PN1 (Figure 2C, D), PN4 (Figure 2E, F), and PN20 (Figure 2G, H). Expression by differentiating ATII cells at E18.5 was confirmed by co-localizing α₅ expression with proSP-C (Figure 3A, B). Staining for T1α, an ATI-specific marker, revealed that α₅ was not expressed by ATI cells at E18.5 (Figure 3C, D). Significant co-localization with CCSP, a marker for Clara cells in the proximal lung, was also observed at E18.5 (Figure 3E, F).

At PN1, a period that coincides with the mid-saccular stage, α₅ was detected in only a minority of ATII cells via proSP-C co-localization (Figure 4A, B) and ATI cells stained for T1α (Figure 4C, D). At PN1, significant detection of α₅ in CCSP-positive Clara cells (Figure 4E, F) and cells that express FoxJ1 (Figure 4G, H), a transcription factor vital in ciliogenesis, revealed α₅ expression in both non-ciliated and ciliated bronchiolar epithelium. At the end of the saccular period (PN4), staining for proSP-C (Figure 4I, J) and T1α (Figure 4K, L) revealed that α₅ was expressed by ATII and ATI cells, respectively. Immunostaining with CCSP (Figure 4M, N) and FoxJ1 (Figure 4O, P) reveal that α₅ expression is absent in non-ciliated Clara cells and ciliated epithelial cells in the proximal lung. These data suggest that α₅ expression is chiefly identified on Clara cells in the proximal lung at PN1 and on ATII and ATI cells in the distal lung at PN4.

During the mid-alveolar stage of lung development (PN10), staining performed with proSP-C revealed that most but not all ATII cells express α₅ (Figure 5A, B) and staining for T1α demonstrated that ATI cells express α₅ (Figure 5C, D). As was observed at PN4, CCSP co-immunostaining revealed no detectable α₅ expression in proximal lung epithelium (Figure 5E, F). A significant general observation near the end of the alveolar period (PN20) was that α₅ staining markedly returns to the large airways at the conclusion of alveologenesis. Co-localization with proSP-C-positive ATII cells (Figure 5G, H) and T1α-positive ATI cells (Figure 5I, J) confirmed α₅ expression by alveolar epithelial cells. Staining for CCSP also revealed markedly increased α₅ expression by proximal bronchiolar epithelium (Figure 5K, L).

Because the expression pattern of α₅ nAChR subunits coincided with differentiating pulmonary epithelial cells in both the proximal and distal lung compartments, we sought to determine the regulatory effects of FoxA2 and GATA-6 on α₅ transcription. Reporter gene assays in bronchiolar Beas-2B cells revealed that α₅ transcription is significantly increased by FoxA2 (Figure 6A). While increasing concentrations of GATA-6 alone had no effect on α₅ transcription (not shown), when combined, both FoxA2 and GATA-6 synergistically induced elevated α₅ transcription in Beas-2B cells (Figure 6A). In alveolar type II-like A549 cells, FoxA2 also significantly increased α₅ transcription in a dose dependent manner (Figure 6A); however, GATA-6 had no measurable effect, either individually (not shown) or in combination with FoxA2 (Figure 6A). Mutagenesis of a single putative FoxA2 response element resulted in complete ablation of FoxA2 transcriptional activation of α₅ expression in both Beas-2B and A549 cells (Figure 6B). Furthermore, possible interactions between FoxA2 and GATA-6 in the regulation of the α₅ gene were also inhibited when the possible FoxA2 response element was removed (Figure 6B).