Author contribution: KC was the primary investigator; K.C., B.C.G., G.S.L., M.L., and D.C.S performed the experiments and analyzed the data. K.C., M.A.K., A.T., and S.J.M contributed to data analysis and interpretation as well as manuscript development.
Correspondence to: Dr. Steven J. Mentzer, Room 259, Brigham and Women's Hospital, Boston, MA 02115. Fax: 617-249-0821. E-mail: email@example.com
Postpneumonectomy lung growth is a remarkable example of tissue morphogenesis. In most mammals studied, the removal of one lung results in the compensatory growth of the remaining lung (Hsia et al., 2004). The compensatory lung growth is associated with not only significant cell proliferation, but dramatic capillary angiogenesis as well. Design-based stereology (Muhlfeld et al., 2010) and corrosion cast studies (Konerding et al., 2012) indicate that postpneumonectomy lung growth is associated with significant postpneumonectomy angiogenesis; perhaps, between 1 and 3 km of new blood vessels within 14–21 days of surgery (Chamoto et al., 2012b). The intimate anatomic and functional relationship between epithelial cells and endothelial cells within the alveolus suggests the likelihood of a coordinated growth response to pneumonectomy.
Previous attempts to characterize the epithelial cell dynamics in the postpneumonectomy lung have demonstrated significant changes in both the Type I and II cell populations. In the normal lung, the cell turnover of both types of alveolar epithelial cells is low; the cell turnover time has been estimated at 4–10 weeks (Blenkinsopp, 1967; Bowden et al., 1968). Most studies using histologic and autoradiographic techniques have shown a dramatic increase in epithelial turnover with lung injury from oxygen (Bowden et al., 1968; Adamson et al., 1970), nitric dioxide (Evans et al., 1973), cadmium (Strauss et al., 1976), bleomycin (Aso et al., 1976), and bacterial pneumonia (Pine et al., 1973); however, similar observations have been made after pneumonectomy (Brody et al., 1978). Postpneumonectomy compensatory growth has been associated with an increase in Type II cell size (Uhal et al., 1989b; Uhal and Etter, 1993), proliferation (Cagle et al., 1990; Uhal and Etter, 1993; Kaza et al., 2002; Li et al., 2005) and volume density (Hsia et al., 1994; Takeda et al., 1999); postpneumonectomy Type II cells have also demonstrated an increase in metabolic activity (Uhal et al., 1989a) and gene expression (Li et al., 2005).
Recent studies of postpneumonectomy angiogenesis have demonstrated a similar transition; that is, pneumonectomy triggers a coincidental change in endothelial cells from a quiescent to rapidly proliferating cell population (Lin et al., 2011). Morphologic evidence indicates that this cell proliferation occurs in discrete regions in the growing lung (Konerding et al., 2012)—a finding suggesting local control. In addition to local proliferation, a blood-derived population of endothelial progenitor cells have been identified (Chamoto et al., 2012b). These blood-derived endothelial cells appear to be rapidly incorporated into the vessel lining (Chamoto et al., 2012b). The similar proliferative time course, as well as their complementary function in gas exchange, suggests that epithelial and endothelial cell dynamics are linked; however, the potential regulatory contribution of Type II cells to alveolar neoangiogenesis is unknown.
In this report, we investigated the hypothesis that alveolar Type II cells regulate postpneumonectomy angiogenesis. Our objectives were to 1) define the population dynamics of alveolar Type II cells after pneumonectomy—including the potential contribution of blood-derived Type II cells—and 2) define the alveolar Type II transcriptional profile relevant to capillary angiogenesis.
Male mice, 8- to 10-week-old wild type C57BL/6 (Jackson Laboratory, Bar Harbor, ME), were used for all nonparabiotic experiments. Wild-type (WT) and green fluorescence protein (GFP)+ C57BL/6-Tg (under the direction of the ubiqutin C promoter, UBC) 30Scha/J mice (Jackson Laboratory) with similar age and weights were selected for parabiosis. The care of the animals was consistent with guidelines of the American Association for Accreditation of Laboratory Animal Care (Bethesda, MD) and approved by our Institutional Animal Care and Use Committee.
After the induction of general anesthesia (Gibney et al., 2011b), the animal was ventilated on a Flexivent (SciReq, Montreal, QC Canada) at ventilator settings of 200/min, 10 mL kg−1, and PEEP of 2 cmH2O with a pressure limited constant flow profile. A thoracotomy was created in the left fifth intercostal space and a left pneumonectomy was performed by hilar ligation (Gibney et al., 2011a). At the completion of the procedure, the animal was removed from the ventilator and observed for spontaneous respirations. Once spontaneous muscle activity was noted, the animal was extubated and observed in a warming cage. Sham thoracotomy involved an identical incision and thoracotomy. The pleural cavity was entered, but there was no surgical manipulation of the left lung. The thoracotomy closure was identical to the pneumonectomy condition.
Light and transmission electron microscopy
Lungs designated for microscopy were harvested after cannulation of the trachea. The tissue was fixed by instillation of 2.5% buffered glutaraldehyde into the bronchial system followed by the instillation of 50% OCT Tissue Tec in saline. After postfixation, samples of the cardiac lobe were cut out and processed according to standard protocols and embedded in Epon (Serva, Heidelberg, Germany). Semithin sections (0.5 μm) were stained with tolouidine blue and analyzed with a Zeiss Axiophot microscope (Zeiss, Oberkochen, Germany). About 700 -Å ultrathin sections were analyzed using a Leo 906 digital transmission electron microscope (Leo, Oberkochen, Germany).
The animals were paired using a modification of the technique described by Bunster (Bunster and Meyer, 1933). The animals were anesthetized with an intraperitoneal injection of ketamine 100 mg kg−1 (Fort Dodge Animal Health, Fort Dodge, IA) and xylazine 10 mg kg−1 (Phoenix Scientific, St. Joseph, MO). The right side of the wild type mouse and the left side of the GFP+ mouse were surgically joined as previously described (Gibney et al., 2012). Postoperatively, each parabiont was given a subcutaneous 1 mL bolus of warmed 0.9% sodium chloride (Abbott Laboratories, North Chicago, IL) twice daily for 48 hr. Analgesia (Buprenorphine, 0.05 mg kg−1, Webster Generics, Sterling, MA) was given twice daily for 48 hr and supplemented as clinically indicated. Nearly 1% Sulfatrim water (Webster Generics, Sterling, MA) was provided in the cage for 28 days. Left pneumonectomy was performed 28 days after parabiosis.
BAL cells were counted using a Neubauer hemacytometer (Fisher, Pittsburgh, PA). Dead cells were excluded by trypan blue (Sigma, St Louis, MO). The numbers of CD11b+ cells were calculated by using flow cytometric analysis: (CD11b+ number)=(total BAL cell number) × (% of CD11b+ cells among total cells)/100.
For flow cytometric lung cell analyses, fluorescein isothiocynate (FITC), phycoerythrin (PE), allophycocyanin (APC), PE-Cy7-conjugated monoclonal antibodies (mAb) were used: anti-CD45 mAb (FITC, rat IgG2b, clone 30-F11, eBioScience), anti-major histocompatibility complex (MHC) class II mAb (PE-Cy7, Rat IgG2b, clone M5/114.15.2, ebioscience), anti-pan cytokeratin mAb (PE, mouse IgG1, clone C-11, abcam), anti-T1α mAb (PE, hamster IgG, clone eBio8.1.1, eBioscience), anti-CD11b mAb (PE-Cy7, rat IgG2b, clone M1/70, BD Bioscience), anti-CD31 mAb (APC, IgG2a, clone 390, eBioScience), isotype control rat IgG2a (PE, clone eBR2a, eBioScience), isotype control rat IgG2b (PE-Cy7, clone RTK4530, Biolegend), isotype control hamster IgG (PE, clone HTK888, Biolegend) and isotype control mouse IgG1 (PE, clone P184.108.40.206.1, eBioscience).
Phosphine, a lamellar body-staining fluorescent dye, was used to identify Type II epithelial cells (Pfaltz & Bauer, Waterbury, CT). For optimizing the working concentration of phosphine, digested lung cells (5 × 105) were stained at the final concentration of 1, 0.1, or 0.025 μg mL−1 for 10 min at 4 °C. Phosphine staining was performed concurrently with cell surface staining (CD45, MHC class II). We decided the appropriate concentration at 0.1 μg mL−1 for FACS analysis and Type II epithelial sorting by Aria (BD, Franklin Lakes, NJ).
For standard phenotyping, the cells were incubated with a fivefold excess of anti-mouse antibodies directly conjugated with FITC, PE, APC, PE-Cy7, or biotin. The cells stained with biotinylated antibodies were washed by FACS buffer (BD Biosciences) and subsequently stained with streptavidin-RPE (Sigma, St Louis, MO). The cells were analyzed using a FACSCanto II (BD, Franklin Lakes, NJ) with tri excitation laser (407, 488, and 633 nm ex). The data were analyzed by FCS Express 4 software (De Novo Software, Los Angeles, CA). In all analyses, debris were eliminated by gating the alive cell population of side scatter (SSC) and forward scatter (FSC), and further by gating the 7AAD (BD Biosciences)-negative population. For Type II cell isolation, CD45− MHC class II+ phosphine+ population was gated and sorted by FACSAria (BD, Franklin Lakes, NJ). Statistical criteria were used for gating the CD45− population, effectively excluding any detectable leukocyte contamination.
The commercially available angiogenesis array (catalog PAMM-024) obtained from SABiosciences (Frederick, MD) was used for all polymerase chain reaction (PCR) array experiments. Real-time PCR was performed with SYBR green qPCR master mixes that include a chemically modified hot start Taq DNA polymerase (SABiosciences). PCR was performed on ABI 7300 real-time PCR system (Applied Biosystems). For all reactions, the thermal cycling conditions were 95°C for 10 min followed by 40 cycles of denaturation at 95°C for 15 sec and simultaneous annealing and extension at 60°C for 1 min. The two sets of triplicate control wells were also examined for interwell and intraplate consistency; standard deviations of the triplicate wells were uniformly <1 Ct. To reduce variance and improve inferences per array (Kendziorski et al., 2005), we used a design strategy that used three to four samples/group (typically three mice/sample). Previous comparison of the angiogenesis array in nonsurgical controls and sham thoracotomy controls demonstrated no difference (Lin et al., 2011).
RNA was extracted using Pico Pure RNA isolation kit (Applied Biosystems Arcturus Products, Carlsbad, CA) according to the instruction. For the rapid analysis of RNA quantity and quality, all samples were analyzed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), microfluidics-based nucleic acid separation system. In most samples, the RNA Pico 6000 LabChip kit (Agilent Technologies) was used. RNA integrity numbers (RIN) of the RNA samples were uniformly >7.0 (Schroeder et al., 2006).
Our quantitative PCR assumed that DNA template and/or sampling errors were the same for all amplifications; our internal control replicates indicated that our sample size was sufficiently large that sampling errors were statistically negligible (Stolovitzky and Cecchi, 1996). The exponential phase of the reaction was determined by a statistical threshold (10 standard deviations). Flow cytometry statistical analysis was based on measurements in at least three different mice. The unpaired Student's t test for samples of unequal variances was used to calculate statistical significance. The data was expressed as mean±one standard deviation. The significance level for the sample distribution was defined as P<0.05.
Epithelial Cell Anatomy
In the evaluation of the alveolar epithelium 9 days after murine pneumonectomy, light microscopy of the cardiac lobe demonstrated thickened subpleural alveolar septae and an apparent increase in alveolar Type II cells compared to nonsurgical controls (Fig. 1A,B). Transmission electron microscopy (TEM) of the peripheral cardiac lobe at 3 days (Fig. 1C,D) and 9 days (Fig. 1E,F) after pneumonectomy demonstrated alveolar Type II cells intimately associated with septal blood vessels. In serial TEM sections, regions were commonly observed with no basal lamina visibly separating the pneumocytes from endothelial cells (not shown).
Epithelial Population Dynamics
To provide a quantitative assessment of epithelial population dynamics after pneumonectomy, alveolar epithelial cells were studied by flow cytometry. To define alveolar Type II cells, enzymatically digested control (no surgery) lungs were studied by flow cytometry. Based on isotype controls (Fig. 2A), 4.7% of the isolated CD45− lung cells were MHC class II+ (Fig. 2B). After gating on the CD45− population to exclude leukocyte contamination, 82% of the class II+ population was also cytokeratin+ (Fig. 2D). To confirm that the CD45−, class II+, cytokeratin+ cells were predominantly Type II cells, the CD45− cells were exposed to varying concentrations of the lamellar body-staining fluorescent dye phosphine. Based on flow cytometry dose-response studies, alveolar Type II cells were defined as CD45−, class II+, phosphine+ when staining was performed at a phosphine concentration of 0.1 μg mL−1.
Using this empiric definition of alveolar Type II cells, the post-pneumonectomy right lungs were enzymatically digested and studied by flow cytometry at various times after surgery. The number of nonleukocyte (CD45−) MHC class II+ cells—a phenotype broadly including alveolar Type II cells—demonstrated a relative decrease in cell number on Day 3 (Fig. 3A). The number of CD45−, MHC class II+ cells increased on Days 7 and 14, then decreased on Day 21. These results were confirmed by similar studies using phosphine (Fig. 3B). In contrast to the decreased number of Type II cells on Day 21, cells expressing the alveolar Type I cell marker T1α increased until Day 14 (Fig. 3C).
To test the possibility that the postpneumonectomy increase in alveolar Type II cells reflected a contribution of blood-borne progenitor cells, a parabiotic postpneumonectomy model was studied. After establishing a complete crosscirculation between WT and GFP parabionts (28 days after parabiotic surgery) (Gibney et al., 2012), a left pneumonectomy was performed. The remaining lung was harvested 14 days after pneumonectomy and the cells were analyzed by flow cytometry. After gating on CD45− cells, the cell populations positive for the T1α (Type I cells) and MHC class II (Type II cells) were studied (Fig. 4). Less than 3% of the Type II cells were positive for GFP—a finding consistent with the absence of a blood-borne contribution to alveolar epithelial cells. Compared to other cell types—reflecting previously published data (Chamoto et al., 2012a a,b)—alveolar epithelial cells demonstrated significantly less evidence for blood-derived migration than leukocytes or endothelial cells (P<0.001) (Fig. 5).
Transcriptional Profile of Type II Cells
To better characterize the potential regulation of capillary growth by mature alveolar Type II cells, nonsurgical control CD45−, MHC class II+, phosphine+ cells were isolated by cell sorting. The angiogenic transcriptional profile of these normal adult mouse alveolar Type II cells was studied by angiogenesis PCR arrays. When compared to a larger group of epithelial and smooth muscle cells (CD11b−, CD31−), the Type II cells demonstrated significantly enhanced expression of 14 angiogenesis-associated genes (P<0.01; Fig. 6B,C). The gene expression profile included the enhanced expression of the extracellular matrix genes laminin alpha 5 (Lama5) and collagen type XVIIIα1 (Col18a1). Also enhanced was the expression of cell cycle genes controlling of epithelial (Efna1, Jag1, and Tgfa) and mesenchymal (Pdgfa and Vegfa) cells.
To determine the potential contribution of the flow cytometry-defined Type II cells to compensatory lung growth, the transcriptional activity of the alveolar Type II cells was studied when endothelial gene transcription and cell cycle activity was greatest (Lin et al., 2011); that is, 7 days after pneumonectomy (Fig. 7). When the Type II cells on Day 7 after pneumonectomy were compared to nonsurgical controls, 10 genes demonstrated significantly increased expression (P<0.05; Fig. 7). In contrast to the normal adult Type II cells, there was notable expression of inflammation-associated genes (Ccl2, Cxcl2, Ifng) as well as genes associated with epithelial growth (Ereg, Lep). Additionally, the enhanced transcription of Mmp9 and Fgf6 indicated an active contribution to structural remodeling and capillary growth.
In this report, we studied the population dynamics and transcriptional activity of flow cytometry-defined alveolar Type II cells after murine pneumonectomy. Our data indicated that 1) alveolar Type II cells, empirically defined as a CD45−, MHC class II+, phosphine+ phenotype, demonstrated an increase in cell number after pneumonectomy; 2) the increase in cell number preceded the increase in Type I (T1α+) cells, and 3) did not appear to involve the contribution of blood-borne Type II (or Type I) cells. 4) The CD45−, MHC class II+, phosphine+ cells demonstrated the active transcription of angiogenesis-related genes both before and after pneumonectomy. Together, the data suggest the local contribution of alveolar Type II cells to alveolar growth.
Our definition of alveolar Type II cells was based on cytologic and morphologic features; that is, cuboidal morphology and ultrastructural lamellar bodies. The cuboidal morphology produced a distinctive “optical phenotype” (Wilson et al., 1986) detected by flow cytometry light scatter analysis. The lamellar bodies, subcellular structures containing the lipid–protein complex of the surfactant system (Ochs, 2010), were detected using the lipid-soluble fluorescent dye phosphine (Uhal and Etter, 1993; Harrison et al., 1995) and flow cytometry. The selectivity of phosphine binding to lamellar bodies has been demonstrated by confocal microscopy (Bakewell et al., 1991). The strength of the phosphine-associated fluorescence signal was attributable to the density of lamellar bodies: Type II cells can contain>100 lamellar bodies—collectively comprising nearly 10% of the pneumocyte cell volume (Young et al., 1991).
An intriguing, but poorly understood, phenotypic characteristic of alveolar Type II cells is the high constitutive expression of MHC class II molecules (Cunningham et al., 1997). MHC class II molecules, prominently linked to CD4 T cell antigen presentation, is notably expressed on “professional” antigen presenting cells such as dendritic cells, mononuclear phagocytes and B lymphocytes. Although alveolar Type II cells express some of the important processing enzymes linked to the classic MHC class II antigen presentation pathway (Watts, 2004), Type II cells are not potent antigen presenting cells (Cunningham et al., 1997; Corbiere et al., 2011). Although the biological role of the molecule is unclear, the MHC class II molecule was a useful marker for alveolar Type II cell isolation by flow cytometry cell sorting.
Recent interest in the therapeutic potential of bone marrow-derived progenitor cells (Kotton et al., 2001; Theise et al., 2002; MacPherson et al., 2005) has led to >40 reports of blood-borne epithelial progenitor cells (Kassmer and Krause, 2010) and several notably negative reports (Wagers et al., 2002; Kotton et al., 2005). The controversy in the field is due, in part, to the difficulty in identifying progenitor cells in the lung by fluorescence microscopy (Kassmer and Krause, 2010). Here, we used a parabiotic crosscirculation (WT/GFP) model to identify potential blood-borne progenitor cells. Compensatory growth after pneumonectomy in the WT parabiont produced an ∼30% increase in lung weight and volume without the infiltration of confounding blood-borne inflammation (Chamoto et al., 2012a a,b). Because of stable GFP expression, migrating cells provided a “fate map” of blood-borne cells—a marker that was independent of migratory path, differentiation history, or surface phenotype. Thus, a blood-derived Type II progenitor cell could be expected to express GFP whether its fate contributed to Type II cells, intermediate epithelial forms, or mature postmitotic Type I cells. The near-absence of GFP+ Type II or Type I cells in the 21 days after pneumonectomy provided convincing evidence that Type I and II lung epithelial cells are not derived from the peripheral blood, but are locally renewing.
Although our findings seem to conclusively demonstrate the local renewal of Type II epithelial cells, there are several potential limitations. First, it is possible that the putative blood-borne epithelial progenitor cell did not express GFP. Because of the prominent GFP expression of Type II cells in the GFP parabiont, and the uniformly positive fluorescence of peripheral blood cells in the GFP mouse (Gibney et al., 2012), we consider this possibility unlikely. Second, the release of the presumptive progenitor cell from the GFP+ bone marrow might be brief relative to the cross-circulation kinetics; that is, the epithelial progenitor cells in the GFP parabiont might not have sufficient time to equilibrate with the circulation in the WT parabiont. Since we have been able to detect other bone marrow-derived progenitor cells in the WT parabiont after pneumonectomy (Chamoto et al., 2012b), we also consider this possibility unlikely.
Flow cytometry cell sorting provided an opportunity to define the expression of angiogenesis-related genes in normal adult mouse Type II cells. The “constitutive” expression of extracellular genes such as laminin alpha 5 (Lama5) and collagen type XVIIIα1 (Col18a1) is consistent with ongoing maintenance of the lung extracellular matrix. Also notable was the enhanced expression of genes associated with the control of epithelial cell (Efna1, Jag1, and Tgfa) and mesenchymal (Pdgfa and Vegfa) mitogenicity. We interpret this transcriptional activity as reflecting the Type II cells' control and regulation of the surrounding microenvironment; that is, the adjacent epithelial cells, endothelial cells and extracellular matrix apparent on both light and electron microscopy.
Perhaps more revealing was the change in the Type II cell transcriptional profile after pneumonectomy. Expression of the inflammation-associated chemokine Ccl2 (monocyte chemotactic protein-1; MCP-1) suggests that Type II cells participate in the recruitment of CD11b+ blood-borne cells into the postpneumonectomy lung. This observation is supported by increased transcription of Cxcl2 (macrophage inflammatory protein 2-alpha; MIP2-alpha), a molecule chemotactic for hematopoietic stem cells (Pelus et al., 2002).
Potentially contributing to epithelial growth was the increased transcription of epiregulin (Ereg) and leptin (Lep). Epiregulin, an epidermal growth factor receptor ligand (EGFR, erbB1), is also a ligand of most members of the ERBB (v-erb-b2 oncogene homolog) family of tyrosine-kinase receptors. The expression of Ereg in epithelial cells (Type II) that give rise to other epithelial cells (Type I) suggests the possibility of both an autocrine and paracrine regulation of epithelial cell growth. The active stimulation of epithelial growth is consistent with the population dynamics we observed after pneumonectomy.
More unexpected was the enhanced transcription of leptin (Lep). Leptin, also known as the product of the OB gene, is synthesized and secreted mainly by white adipose tissue, but it is also expressed in fetal (Bergen et al., 2002) and injured lung tissue (Malli et al., 2010). In addition, there is growing evidence that leptin stimulates endothelial growth and angiogenesis (Bouloumie et al., 1998; Sierra-Honigmann et al., 1998). Finally, data indicates that vascular endothelium in rodents expresses both the short and long forms of the leptin receptor (Margetic et al., 2002). Based on these findings, it is likely that leptin produced by the Type II cell plays an active regulatory role in postpneumonectomy angiogenesis.