Animals
Mice bearing the PDGF-Rα-GFP construct have been described [
7,
14,
15]. Production and nuclear localization of green fluorescent protein (GFP) is under the control of the endogenous
pdgf-rα promoter. GFP expression in the PDGF-Rα-GFP mice spatially and temporally recapitulated endogenous
pdgf-rα expression, and the intensity of GFP-fluorescence correlated with the abundance of immunoreactive PDGFRα [
14]. The mice used in this study carried one
pdgf-rα
-GFP allele (which does not encode for active PDGF-Rα) and one functional
pdgf-rα allele. The heterozygous mice are phenotypically identical to wildtype (GFP
-) mice, except for nuclear GFP, which enables their identification [
14]. Mice expressing Cre-recombinase, which was targeted to the endogenous transgelin gene, B6.129S6-Tagln
tm2(cre)Yec/J, stock number 006878 (Jackson Laboratories, Bar Harbor, ME) were bred with mice with LoxP-flanking exons 1 and 4 of the
pdgfRα gene B6.Cg-Pdgfra
tm8Sor/EiJ Jackson Laboratories stock number 006492 to obtain mice which contained at least one transgelin allele (TGCre +/−) and two LoxP-flanked
pdgfRα alleles (PDGFRαF/F). Mice were genotyped using DNA isolated from tail biopsies and PCR, as recommended by Jackson Laboratories and were euthanized at postnatal day 8 or 12. Mice containing one transgelin (TG) Cre allele are denoted as TGCre+/−, one or two LoxP-flanked PDGFRα-allele(s) are denoted as PDGFRαF/- or PDGFRαF/F, respectively. Mice bearing the Cre-mediated deletion of PDGFRα are designated as TGCre+/−;PDGFRαF/F. The right lungs were uniformly inflated using 50 μl of fixative containing agarose per gram body weight, and fixed for 6 hours for laser scanning confocal microscopy (LSCM) [
3]. The left lungs were inflated with fixative and then fixed overnight in 4% paraformaldehyde at 4
o and embedded in paraffin. Mice were housed in Thoren cages with access to food and water
ad libitum in a thermally regulated environment and a 12-hour light/dark cycle. Protocols for animal use were approved by the Iowa City Veterans Affairs Medical Center animal use committee [
1]. Lung fibroblasts were isolated from heterozygous mice on postnatal day 12 or 14 day old mice using a previously reported method [
1].
Antibodies
Immunofluorescence and laser scanning confocal microscopy (LSCM) p27Kip: Cell Signaling #3686, 75D8, rabbit monoclonal antibody (Mab) IgG 1:800 dilution. αSMA: Sigma-Aldrich (St. Louis MO), clone 1A4, mouse IgG2a, 1:400 dilution. FoxO3a: Millipore #07-1719, 1 μg/ml. Ki67: Dako rat anti-mouse Ki67 Mab clone TEC-3, 1:200 dilution. Immunofluorescence of cultured primary mouse LF FoxO3a: Cell signaling # 2497, 75D8 rabbit MAb. Western immunoblotting FoxO3a: Millipore #07-1719 1 μg/ml. anti- p(S253) FoxO3a Millipore rabbit polyclonal IgG, 06–953 1 μg/ml, directed at the N-terminus of FoxO3a. Secondary antibody was horse radish peroxidase-labeled goat anti rabbit IgG diluted 1:2000. Mouse anti-cleaved Asp214 poly-ADP ribose polymerase (PARP-1, 0.5 μg/ml, BD Biosciences, San Jose, CA).
Analysis of DNA in mouse LF using flow cytometry
Our procedure for isolation LF has been published [
7]. Using flow cytometry we have not found cytokeratin 18+ alveolar epithelial cells in the GFP+subpopulation and by this criterion, epithelial cells comprised only 2.7 ± 3.0 or 8.1 ± 6.0% (mean±SD) of the GFP- population at P4 and P12. Fewer than 6% of the GFP + cells stained for any of the markers for either for epithelial, endothelial, or macrophages. For analyzing the cell cycle distribution, freshly isolated mouse LF were fixed for 20 minutes at 4° with 0.5% paraformaldehyde, washed and then fixed for 1 h at 4° with 70% methanol that had been cooled to -20°. After washing, the LF were resuspended in 0.145 M NaCl, 0.0015 M KH2PO4, 0.0027 M KCl, 0.0086 M Na2HPO4, pH 7.4 (PBS) and incubated for 30 minutes with 20 Kunitz units of RNase A at 25°, and then with 35 μg/ml propidium iodide for 1 h at 4°. The cells were subjected to flow cytometry using a LSR II instrument (BD Biosciences, San Jose, CA) [
7]. A minimum of 20,000 gated events were analyzed. The background fluorescence from the IgG isotype controls was subtracted prior to calculating the proportions of the different fibroblast populations. The data were analyzed using CellQuest Software (BD Biosciences) [
7]. The proportions of LF in either the GFP + or the GFP- residing in G
0/G
1, S, or G
2/M were calculated based on the intensity of PI-fluorescence using ModFitLT (Verity Software, Topsham ME). The phases of the cell cycle were distinguished from a plot of forward scatter vs. PI area, after correcting for PI-width to exclude aggregates of LF.
Analysis of foxO3agene expression using real-time quantitative PCR (RT-qPCR)
Total RNA was isolated using Tri-Reagent (Sigma-Aldrich, St. Louis, MO), subjected to reverse transcription, and
foxO3a (Mm01185722.m1), and
β-2-microglobulin (Mm00437762.m1) mRNA were quantified using Taqman Gene Expression Assays [
7]. Four independent experiments were performed. Values for
foxO3a expression were normalized to β-2-microglobulin using the 2
(−ΔΔCT) relative quantification method [
16] and expressed relative to the 2
(−ΔΔCT) value at P4. The β-2-microglobulin Taq-man probe was used for normalization because we have not observed variation in β2 M mRNA, which appeared to be related experimental treatment conditions or developmental age. The mean±SEM Ct values for β2 M for the mice at P4, P8, and P12 were 19.66 ± 0.23, 19.18 ± 0.30, and 18.94 ± 0.38, respectively.
Phospho FoxO3a Western Immunoblotting
After washing with 25 mM HEPES, pH 7.4, 150 mM NaCl, 2 mM Na
3VO
4 at 4°, the cell layers were lysed with 10 mM Tris–HCl, pH 7.4, 5 mM EDTA, 50 mM NaCl, 50 mM sodium fluoride, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 2 mM Na
3VO
4, and 20 μg/ml aprotinin [
1]. Equal quantities of protein were subjected to SDS-PAGE using a 6% acrylamide gel, and the proteins were transferred to nitrocellulose. Anti-phospho (S253) FoxO3a and anti-FoxO3a were diluted 1:1000. After washing, the primary antibodies were detected using goat anti-rabbit peroxidase and enhanced chemiluminescence (ECL, GE Healthcare, Piscataway, NJ). Fluorography was performed and the film was imaged and analyzed using a Quantity-One Imaging system (BioRad, Hercules, CA). The densities of phospho-FoxO3a were normalized to the density of FoxO3a for each lane, to account for differences in the amounts of protein which were loaded.
Effects of PDGF-A on the distribution FoxO3a in cultured LF
LF, which had been isolated from PDGFRα-GFP+mice, were cultured on cleaned glass coverslips which had been coated with 8 μg/ml vitronectin to promote cell adhesion. The mouse LF were allowed to adhere to the coverslips for 1 hour, non-adherent cells were removed by washing and the adherent LF were cultured overnight in Ham’s F-12 medium containing 10% FBS. The next morning the LF were washed and the medium was changed to Opti-MEM for 12 hours prior to adding 50 ng/ml PDGF-A to some of the coverslips. The LF were cultured for an additional 12 h, washed with PBS and fixed for 20 min at 4° with 2% paraformaldehyde. After washing with PBS and water the coverslips were air dried and stored at 4° until they were immunostained. After permeabilization with 0.3% Triton X-100 and blocking with 2% normal goat serum, the coverslips were incubated overnight with rabbit anti-FoxO3a (Cell Signaling #2497, 1:200 dilution), anti-Ki67 (1:200 dilution), washed and then incubated with goat anti-rabbit IgG AlexaFluor 350 and goat anti-rat AlexaFluor 568 respectively, both at 1:2000 dilutions. The stained coverslips were mounted and imaged using an Olympus IX-81 microscope using the appropriate filter sets for A350, A568, and GFP. Nuclei and the cytoplasmic boundaries were identified using phase contrast optics and images using phase contrast and the three filters were captured with an Olympus XM-10 camera at a 1024 X 1024 pixel density. The images were merged using Cell-Sens® Software (Olympus, Center Valley, PA) and nuclei containing Ki67 or FoxO3a were enumerated in the GFP+and GFP- sub-populations.
Analysis of images acquired using LSCM
General considerations: Because of elevated non-specific staining with the anti-FoxO3a and anti-p27 antibodies using 50 or 100 μm sections, 7 micron sections were used to enumerate nuclei containing one of these antigens. For the analysis of p27kip1, nuclei were counterstained with PoPo3 and for the analysis of FoxO3a, nuclei were stained with 4’,6-diamidino-2-phenylindole, diacetate (DAPI). Because the LSM710 did not have a UV laser, DAPI fluorescence was excited by a metal halide lamp (Exfo X-cite120XL Exfo Mississauga, ON, Canada) metal halide lamp, and emitted light was collected without pin-hole optics. The Zeiss czi-images were converted to 8-bit TIF images and the colors were merged into RGB TIF files using image J (National Institutes of Health). IP-Lab for Windows (BD-Scanalytics, San Jose, CA) was used to identify nuclei which contained p27kip1 or FoxO3a, using uniform segmentation criteria.
Evaluation of nuclear p27
kip
in PDGFRα-GFP+mice: For the analysis of p27, cells in the alveolar walls were classified as follows: nuclear GFP+(pdgf-rα expressing, green), PoPo3+ (all nuclei, red), p27kip1(AlexaFluor 647, pseudocolor blue) in nuclei without GFP (PoPo3, but without GFP) and p27kip1 in nuclei containing GFP (and stained with PoPo3). Five fields were analyzed per mouse using 3 mice at each age. Mice for each of the ages were included, and the LSM710 acquisition parameters were held constant for all ages within a staining and imaging cohort. Using uniform segmentation criteria and IP-Lab, the pixels-regions that contained (a) both red and blue, but not green, (nuclei which were p27kip1 + and GFP-), (b) red, blue, and green (nuclei which were p27kip1+ and GFP+), and (c) all nuclei (red). The nuclei of cells which were p27kip1 + GFP-, or p27kip1 +, GFP+nuclei were enumerated and expressed relative to the total of PDGFRα-expressing alveolar cells (GFP+) or other, non-expressing alveolar cells (GFP-). The percentages of nuclei in p27kip1+, GFP+or p27kip1+, GFP- groups were calculated for each field and a mean and SEM were determined for each mouse. The data from four mice were combined to obtain a mean and SEM for each age.
Evaluation of nuclear FoxO3a in PDGFRα-GFP+mice : Four mice and a total of 21 fields were analyzed at each age: P4, P8 and P12 and one mouse from every age was included in each of the four staining sessions. Rabbit anti-FoxO3a was identified using goat-anti rabbit IgG AlexaFluor 647 (blue pseudocolor), nuclei were counterstained with DAPI (and assigned either a green or blue pseudocolor), and Ki67 was identified using goat anti-rat AlexaFluor 568 (red). Images, which had been captured using the LSM710, were exported in an 8-bit TIF format and converted to RGB images using ImageJ. The RGB images were analyzed using IP-Lab using uniform segmentation criteria. To identify Ki67 or FoxO3a in GFP- nuclei DAPI was pseudocolored green and GFP was not included in the color map. The pixel areas containing green and red (Ki67) or green and blue (FoxO3a) above the specified thresholds were considered Ki67+, GFP- nuclei. To analyze Ki67 and FoxO3a within GFP+nuclei DAPI was excluded and the pixel areas containing GFP (green), FoxO3a (blue) and Ki67 (red) were ascertained. To determine all pixels overlying nuclei, GFP, Ki67 and FoxO3a were excluded and the pixel area containing DAPI was ascertained. Likewise to ascertain the GFP+nuclear area, after excluding FoxO3a and Ki67, the pixel area containing both GFP (green) and DAPI (blue pseudocolor) were ascertained. The pixel area of alveolar nuclei containing Ki67 or FoxO3a but not GFP were expressed relative to the pixel area of DAPI-stained nuclei lacking GFP. Likewise, the pixel area of alveolar nuclei containing Ki67 or FoxO3a, as well as GFP, was expressed relative to the pixel area of DAPI-stained nuclei also containing GFP.
Evaluation of nuclear p27
kip
in TGCre; PDGFRαFlox mice: Because these mice do not contain the GFP insertion into pdgfRα, we used αSMA-FITC to identify regions where PDGFRα-expressing would be expected to reside. p27kip1 was stained as described for lungs from pdgfRα -GFP mice and PoPo3 was used to stain all nuclei. The mice were euthanized at P12 and 5 fields were analyzed per mouse using 4 mice TGCre+/−;PDGFRαF/F or littermate controls (bearing only one LoxP-flanked allele and or no Cre-alleles). Mice from both control and LoxP-flanked groups were included in every staining cohort, and the LSM710 acquisition parameters were held constant within a staining and imaging cohort. Uniform segmentation criteria were used to uniformly identify collections of pixels that contained both p27kip1 and PoPo3 (nuclei containing p27kip1). The cytoplasm around all nuclei was examined to determine whether it contained filamentous structures that stained with anti-αSMA. Nuclei which contained p27kip1 and were adjacent to were classified as p27kip1+, αSMA+, and those which were not adjacent to αSMA were classified as p27kip1+ and αSMA -. The enumerated nuclei of p27kip1+, αSMA+or p27kip1+, αSMA– cells were expressed relative to the total nuclei which were (αSMA +) or were not adjacent to αSMA, respectively.
Analysis of cleaved poly-ADP ribose polymerase (PARP) in TGCre; PDGFRαFlox mice: Lungs from TGCre+/−;PDGFRαF/F and littermate control mice ages P8 or P12 were uniformly inflated, fixed, and 100 μm cryosections were used for immunostaining and confocal microscopy. The sections were permeabilized with 0.3% Triton X-100 and 0.5 mg/ml 2.4 g2 rat-anti mouse Fc-receptor was added to block mouse Fc receptors. The primary antibody was mouse anti-cleaved PARP, the secondary antibody was goat anti-mouse IgG-AlexaFluor 647, and nuclei were counterstained with PoPo3. Anti-αSMA-FITC was used to locate regions where mesenchymal cells would be expected to reside. Three Z-stacks for each of 3 different portions of the tissue were imaged 1.5 at μm intervals for each mouse and sections were analyzed at 3 different z-levels. The starting z-level was in the center of the tissue and images were acquired in the 9 μm above and below this level. Comparisons were made at the same z-level for all lungs to normalize for differences in laser-penetration at different tissue depths. Uniform segmentation criteria were applied to identify nuclei which contained PARP-staining nuclei (more than 75% of the pixels in the nuclei met the blue segmentation criterion). Data were expressed as the ratio of cleaved PARP-containing nuclei to the total nuclei in the field. Three mice were used at each age for each genotype and the mean±SEM proportions of nuclei were calculated.
Analysis using the proliferation marker Ki67 Lungs from TGCre+/−;PDGFRαF/F and littermate control mice at P8 were inflated, fixed and 100 μm sections were prepared. The tissues were incubated with rat anti-mouse Ki67 (DAKO, clone TEC-3, 18.5 μg/ml) for 1.5 h at 25°, followed by washing, and then with goat anti-rat IgG-A647 and anti-αSMA-FITC. Nuclei were counterstained with PoPo3. Z-stacks were acquired at 2.5 μm intervals over 42 μm and nuclei were enumerated using the optical fractionator. Nuclei were classified as staining for Ki67 or only PoPo3 and as adjacent or not adjacent to αSMA. The data from four mice of the same genotype were pooled.