A Novel Method for Isolating Individual Cellular ... - ATS Journals

A Novel Method for Isolating Individual Cellular Components from the Adult Human Distal Lung Naoya Fujino1, Hiroshi Kubo1, Chiharu Ota1, Takaya Suzuki2, Satoshi Suzuki5, Mitsuhiro Yamada3, Toru Takahashi1, Mei He1, Takashi Suzuki4, Takashi Kondo2, and Mutsuo Yamaya1 1 Department of Advanced Preventive Medicine for Infectious Disease, 2Department of Thoracic Surgery, Institute of Development, Aging, and Cancer, 3Department of Infection Control and Laboratory Diagnostics, and 4Department of Pathology and Histotechnology, Tohoku University Graduate School of Medicine, Aobaku, Sendai, Japan; and 5Department of Thoracic Surgery, Japanese Red Cross Ishinomaki Hospital, Hebita, Ishinomaki, Japan

A variety of lung diseases, such as pulmonary emphysema and idiopathic pulmonary fibrosis, develop in the lung alveoli. Multiple cell types are localized in the alveoli, including epithelial, mesenchymal, and endothelial cells. These resident cells participate in the pathogenesis of lung disease in various ways. To elaborate clearly on the mechanisms of these pathologic processes, cell type–specific analyses of lung disease are required. However, no method exists for individually isolating the different types of cells found in the alveoli. We report on the development of a FACS-based method for the direct isolation of individual cell types from the adult human distal lung. We obtained human lung tissue from lung resections, and prepared single-cell suspension. After depleting CD45-positive cells, a combination of antibodies against epithelial cell adhesion molecule (EpCAM), T1a, and vascular endothelial (VE)-cadherin as used to delineate alveolar cell types. Alveolar Type II cells were highly purified in the EpCAMhi/T1a2 subset, whereas the EpCAM1/T1a2/low subset contained a mixed epithelial population consisting of alveolar Type I and bronchiolar epithelial cells. The EpCAM2/T1a2 subset included both microvascular endothelial and mesenchymal cells, and these were separated by immunoreactivity to VE-cadherin. Lymphatic endothelial cells existed in the EpCAM2/T1ahi subset. Isolated cells were viable, and further cell culture studies could be performed. These results suggest that this novel method enables the isolation of different cellular components from normal and diseased lungs, and is capable of elucidating phenotypes specific to certain alveolar cell types indicative of lung disease. Keywords: cell isolation; fluorescence-activated cell sorting; pulmonary alveoli; cell culture

The lung is a complex organ containing more than 30 different cell types within a three-dimensional structure (1). Alveoli reside at the end of this structure, and are thus difficult to approach. A variety of lung diseases develop in the alveoli, including pulmonary emphysema (2), idiopathic pulmonary fibrosis (3), pneumonia caused by microorganisms (4), acute lung injury/acute respiratory distress syndrome (5), and lung cancer (6). In the pathologic processes of (Received in original form May 26, 2011 and in final form October 21, 2011) This work was supported by Japanese Society for the Promotion of Science grant 22390163 (H.K.). Author Contributions: N.F. and H.K. designed and performed the experiments and contributed to writing the manuscript. N.F., C.O., Takaya S., M. Yamada, T.T., M.H. and M. Yamaya performed the experiments. Takaya S., S.S., and T.K. obtained informed consent from the patients and contributed to the analyses of clinical data. Takashi S. performed histological analyses and contributed to the electron microscopic evaluation Correspondence and requests for reprints should be addressed to Hiroshi Kubo, M.D., Ph.D., Department of Advanced Preventive Medicine for Infectious Disease, Tohoku University Graduate School of Medicine, 2-1 Seiryoumachi, Aobaku, Sendai 980-8575, Japan. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Cell Mol Biol Vol 46, Iss. 4, pp 422–430, Apr 2012 Copyright ª 2012 by the American Thoracic Society Originally Published in Press as DOI: 10.1165/rcmb.2011-0172OC on October 27, 2011 Internet address: www.atsjournals.org

CLINICAL RELEVANCE We developed a novel technique of cell isolation from human lung tissue, using a combination of cell surface antigens. With this technique, we are able to isolate component cells in alveoli individually and still alive, and then analyze or culture these cells. Because alveoli are the main targets of many pulmonary diseases, analyzing the component cells in alveoli is useful for understanding the pathophysiology of disease development, epigenetic analyses, and drug discovery.

these lung diseases, alveolar cells may initially be injured by various insults, including noxious gases, oxidants, and inflammatory cells. The repair process after an injury relies on the protective effects of various types of alveolar cells, and the loss of these protective effects can lead to the development of lung disease. Thus, to yield insights into the molecular basis of cellular phenotypes and the pathologic processes in these lung diseases, an investigation of the roles of specific cell types in the alveoli in health and disease is necessary (1). To isolate specific cell types in normal and diseased human lungs, laser capture microdissection (LCM) and FACS have been used. Previous studies reported that LCM can be used to isolate bronchiolar epithelial cells from lung tissue affected by chronic obstructive pulmonary disease (7–9). However, LCM is unsuitable for the isolation of alveolar component cells for two reasons. The first involves technical limitations in isolating individual alveolar component cells using LCM, because alveolar cells are firmly ensconced within the thin alveolar walls (10). Secondly, microarray data derived from samples prepared by LCM showed a significantly higher level of contamination than occurred with the use of FACS when neural populations from the murine brain were isolated (11). In contrast, FACS is thought to be useful for the isolation of specific cell types from mixed cell populations, based on the assumption that specific combinations of cell surface markers are already well characterized. Recently, considerable progress has been made in nanotechnology and genomic and other “-omic” approaches, increasingly facilitating the characterization of cell surface markers among different cell types in the alveoli, and enabling the establishment of methodologies for cell type–specific isolation (1). Here, we report on the development of a FACS-based method for the direct isolation of individual cell types from the adult human distal lung. The combination of three surface markers, epithelial cell adhesion molecule (EpCAM), T1a, and vascular endothelial (VE)-cadherin, allowed for the isolation of alveolar Type II (ATII) cells, as well as microvascular endothelial cells, lymphatic endothelial cells, and mesenchymal cells, in a viable condition. We further cultured isolated cells and characterized the phenotypes of each component. Some of the results of this study were previously presented in abstract form (12).

Fujino, Kubo, Ota, et al.: Isolation of Human Alveolar Cellular Components

423

Figure 1. Epithelial cell adhesion molecule (EpCAM)-delineated and T1a-delineated subpopulations of adult human distal lung cells. (A) A representative FACS dot plot shows the expression of EpCAM and T1a in a live and single-cell–gated CD45-negative fraction from normal lung tissue. The dot plot is a representation of the results from 40 patients. (B) Representative immunofluorescence images of cells isolated from normal lung tissues. (C) An electron micrograph of an EpCAMhi/T1a2 cell. ATII, alveolar epithelial Type II; ATI, alveolar epithelial Type I; pan-CK, pan-cytokeratin; pro-SP-C, prosurfactant protein–C; AQP5, aquaporin 5; CCSP, Clara cell– specific protein. Scale bars: B, 20 mm; C, 2 mm.

MATERIALS AND METHODS Patients and Preparation of Tissue Samples Human lung tissue was obtained from patients who underwent lung resections at the Department of Thoracic Surgery at Tohoku University Hospital (Aobaku, Sendai, Japan) or at the Ishinomaki Red Cross Hospital (Hebita, Ishinomaki, Japan). The indication for lung resection in all patients was primary lung cancer. Lung tissue was obtained at sites distal from the tumors. Through histopathologic study, we confirmed that the harvested tissue did not contain tumor lesions and did not exhibit emphysema, fibrosis, or inflammatory changes. In addition, these patients manifested normal lung function, as determined by spirometry. This study was approved by the Ethics Committees at Tohoku University

School of Medicine and the Ishinomaki Red Cross Hospital. All subjects gave informed consent.

Preparation of Single-Cell Suspensions from Human Lung Tissue Human lung cells were isolated as previously described, with some modifications (13).

Flow Cytometry and Sorting of Lung Component Cells We used phycoerythrin-conjugated anti-human EpCAM antibody (catalogue number 12-9236, clone 1B7; eBioscience, San Diego, CA), Alexa

424

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 46 2012

Fluor 647–conjugated anti-human T1a antibody (catalogue number 337008, clone NC-08; Biolegend, San Diego, CA), and FITC-conjugated anti-human VE-cadherin antibody (catalogue number 560411, clone 557H1; BD Pharmingen, San Diego, CA). To discriminate between live and dead cells, we used 7-amino actinomycin D (catalogue number 00-6993; eBioscience). We sorted live and single-cell–gated subpopulations, based on their staining patterns with EpCAM, T1a, and VE-cadherin, using a FACS Aria II Cell Sorter and FACS Diva, version 6.1 (BD Biosciences, San Jose, CA). FACS analyses were performed using the FlowJo software package (Tree Star, Ashland, OR).

Immunofluorescence Staining and Immunohistochemistry For immunofluorescence staining, the cytospun cells or cultured cells were fixed with 4% paraformaldehyde, blocked, and permeabilized. We stained samples with primary antibodies, as shown in Table E1 in the online supplement. Immunofluorescence images of lung sections were taken using the Nikon C2 system (Nikon, Tokyo, Japan).

Electron Microscopy Analysis Electron microscopy analysis was performed as previously described (13).

Isolation of ATII Cells According to a Density-Gradient Method ATII cells were isolated from lung cell suspensions using a densitygradient method, as previously reported (14).

Culture of Alveolar Epithelial Type II Cells Isolated from Human Lungs Sorted EpCAMhi/T1a2 cells were plated on collagen I–coated culture slides (Thermo Fisher Scientific, Waltham, MA) with a SAGM Bulletkit (Lonza, Basel, Switzerland) containing 1% FBS, as previously described, with some modifications (15).

Culture of Microvascular Endothelial and Mesenchymal Cells Isolated from Human Lungs Sorted VE-cadherin1 and VE-cadherin2 cells were plated on fibronectincoated plates (24 wells; BD Falcon, San Jose, CA) and cultured with two types of media, EGM-2-MV BulletKit (Lonza) and Dulbecco’s Modified Eagle’s Medium/10% FBS/penicillin/streptomycin/amphotericin B.

In Vitro Angiogenesis Assay A tube formation assay was performed according to previously reported methods, with some modifications (16).

Statistical Analyses Statistical analyses were performed with GraphPad Prism, version 5.0b (GraphPad Software, La Jolla, CA). Data were compared using an unpaired t test. Statistical significance was defined as P , 0.05.

TABLE 1. THE YIELD OF DIFFERENT CELL TYPES (PER GRAM OF TISSUE) ISOLATED FROM ADULT HUMAN LUNGS (MEAN 6 SD; n ¼ 4) Surface Antigen Expression hi

2

EpCAM /T1a EpCAM1/T1a2/low EpCAM2/T1a2/VE-cadherin1 EpCAM2/T1a2/VE-cadherin2 EpCAM2/T1ahi

Number of Cells 4.2 1.2 1.4 2.0 3.9

3 3 3 3 3

106 106 106 106 105

6 6 6 6 6

3.8 7.8 8.4 1.0 2.2

3 3 3 3 3

106 105 105 106 105

Definition of abbreviations: EpCAM, epithelial cell adhesion molecule; VEcadherin, vascular endothelial cadherin.

RESULTS Identification of Distinct Subpopulations in the Human Distal Lung

To isolate specific cell types from distal lung tissue, we characterized cell surface markers and sought to develop a FACSbased isolation method. We enzymatically digested subpleural lung tissue, and prepared single-cell suspensions. We collected whole lung cells of 1.3 3 107 6 7.1 3 106 cells/g tissue (mean 6 SD; n ¼ 11). The viability of the freshly collected cells was determined using trypan blue. The percentage of trypan blue–negative cells was 87.7% 6 5.7% (mean 6 SD; n ¼ 11). CD45-expressing hematopoietic cells containing alveolar macrophages (17) were depleted from the single-cell suspensions, using anti-human CD45 antibody–coated microbeads. We confirmed the complete depletion of CD451 cells from the single-cell suspension by flow cytometry, using another clone of an antibody against CD45 (data not shown). The percentage of collected CD452 lung cells among the whole lung cells was 22.7% 6 8.1% (mean 6 SD; n ¼ 24). We fractionated CD452 lung cells into four subpopulations, using antibodies specific for EpCAM and T1a (Figure 1A). The yield of each subset is shown in Table 1. Freshly sorted and cytospun cells were stained with each lineage marker by immunofluorescence. We found that ATII cells were enriched in the EpCAMhi/T1a2 subset, and that the EpCAM1/T1a2/low subset contained alveolar Type I (ATI) cells and bronchiolar epithelial cells, including Clara cells. Immunofluorescence staining demonstrated that the EpCAMhi/T1a2 subset expressed pro– surfactant protein–C (pro–SP-C, an ATII cell marker) and pancytokeratin (an epithelial marker), but did not express Clara cell–specific protein (CCSP, a Clara cell marker) or aquaporin 5 (AQP5, an ATI cell marker) (Figure 1B and Table 2). In addition, electron microscopy showed that the sorted EpCAMhi/T1a2 cells displayed lamellar bodies that contained pulmonary surfactants (18) (Figure 1C). In contrast, the EpCAM1/T1a2/low subset expressed pan-cytokeratin, CCSP, and AQP5, indicating that the EpCAM1/T1a2/low subset is a mixed cell population consisting of ATI cells and bronchiolar epithelial cells, including Clara cells (Figure 1B and Table 2). The EpCAM2/T1a2 and the EpCAM2/T1ahi subset expressed vimentin but not other epithelial markers, suggesting that these subsets contained mesenchymal cells and endothelial cells (Figure 1B). A previous study reported that T1a was expressed by lymphatic endothelial cells but not vascular endothelial cells (19). Therefore, the EpCAM2/T1ahi subset shown in Figure 1A predominately contained lymphatic endothelial cells, and the remaining EpCAM2/T1a2 subset contained vascular endothelial cells and mesenchymal cells such as fibroblasts, pericytes, and smooth muscle cells. To isolate ATII cells, previous studies used a method combined with a density-gradient technique, the depletion of alveolar macrophages using anti-CD14 antibody–coated microbeads, and

TABLE 2. ANALYSIS OF IMMUNOFLUORESCENCE STAINING OF SUBPOPULATIONS DEFINED BY EPCAM AND T1a EpCAMhi T1a2 Pan-cytokeratin Pro–SP-C Aquaporin 5 CCSP Vimentin

98.1 94.0 1.1 0.4 1.8

6 6 6 6 6

0.4 1.6 1.0 0.7 0.7

EpCAM1 T1a2/low 96.8 3.0 7.5 4.3 1.1

6 6 6 6 6

1.2 3.2 6.2 2.9 0.6

EpCAM2 T1a2 0.0 0.2 0.4 0.0 93.9

6 6 6 6 6

0.0 0.4 0.7 0.0 3.1

EpCAM2 T1ahi 0.0 0.3 0.0 0.0 98.6

6 6 6 6 6

0.0 0.5 0.0 0.0 1.0

Values represent mean percentages 6 SD of cells positive for the indicated markers. Three different samples were evaluated for each marker. pro–SP-C, pro–surfactant protein–C; CCSP, Clara cell–specific protein.


425

Figure 2. Cells isolated as EpCAMhi/T1a2 show ATII cell phenotypes in vitro. (A) Phase contrast images of cultured EpCAMhi/T1a2 cells. Left: Cells on Day 2. Right: Cells on Day 7. (B) Immunofluorescence staining of pro-SP-C, AQP5, and T1a on Day 2 (left) and Day 7 (right). (C) Immunofluorescence staining of E-cadherin, zonula occludens–1 (ZO-1), and occludin of cultured cells on Day 7. Scale bars: A, 100 mm; B and C, 50 mm.

the incubation of cells on human IgG–coated dishes (14, 20). We compared the percentages of pro-SP-C1 cells isolated by our FACS-based method and by the density-gradient method. We found that cells isolated via the density-gradient method contained not only pro-SP-Chi but also pro-SP-Clow cells, compared with cells isolated by the FACS-based method (Figure E1). The percentage of pro-SP-C1 cells, including both SP-Chi and SPClow cells, was significantly higher in our FACS-based method (n ¼ 3) than in the density-gradient method (n ¼ 5) (93.7% 6 1.7% versus 34.9% 6 13.5%, respectively, mean 6 SD, P ¼ 0.0003). We further confirmed and characterized the phenotypes of the EpCAMhi/T1a2 subset (an ATII cell population) under culture conditions. After the sorting procedure, the percentage of trypan blue–negative cells in EpCAMhi/T1a2 cells was 84.3% 6 9.9% (mean 6 SD, n ¼ 4). Cultured cells derived from the EpCAMhi/T1a2 subset attached on culture slides. These cells displayed a cuboidal shape on Day 2 (Figure 2A). However, on Day 7, the cultured cells showed a flattened and broad shape,

and grew to confluent monolayers (Figure 2A). To verify the immunophenotypes of cultured cells, we performed immunofluorescence staining using antibodies against pro-SP-C, AQP5, and T1a. We confirmed that the cells on Day 2 expressed pro-SP-C, but not both AQP5 and T1a, indicating that cultured cells on Day 2 still retained the phenotypes of ATII cells. However, cultured cells on Day 7 expressed AQP5 and T1a, but lost their expression of pro- SP-C, suggesting that the cultured cells differentiated into ATI-like cells on collagen I under culture conditions, without any specific growth factors (Figure 2B). These data demonstrated that the sorted EpCAMhi/T1a2 cells had ATII cell phenotypes, with the potential to differentiate spontaneously into ATI-like cells in vitro. In addition, the cultured cells forming confluent monolayers on Day 7 expressed E-cadherin and tight-junction proteins (zonula occludens–1 and occludin) (Figure 2C). Immunostaining examinations of E-cadherin and tight-junction proteins verified that the isolated ATII cells formed epithelial monolayers during the culture period.

426


Figure 3. Immunofluorescence staining of EpCAM and T1a on alveolar walls. EpCAM was expressed by pro-SP-C1 cells (ATII cells; A, arrows) and AQP51 cells (ATI cells; B, arrows). Pro-SP-C1 cells did not express T1a (C, arrows). AQP51 cells expressed T1a (D, arrows). Scale bars, 20 mm. Data are representative of five patients. DAPI, 4’,6-diamidino-2-phenylindole; DIC, differential interference contrast image.

Localization of EpCAM-Expressing or T1a-Expressing Cells in the Human Distal Lung

To verify the localization of EpCAM and T1a in normal adult human lungs, we performed immunohistochemical and immunofluorescence staining, using specific antibodies. As previously described (21), alveolar and bronchiolar epithelial cells stained positive for EpCAM (Figures E2A and E2B), whereas endothelial and mesenchymal cells did not express EpCAM. We confirmed that both ATII and ATI cells expressed EpCAM through the costaining of EpCAM with either pro-SP-C or AQP5 (Figures 3A and 3B). T1a was shown to be expressed by ATI cells in rat lungs (22), and by lymphatic endothelial cells in adult human lungs (19). We found that T1a was expressed not only by ATI cells and lymphatic endothelial cells, but also by bronchiolar epithelial cells, in human lungs (Figures E1C–E1E). In addition, co-immunostaining demonstrated that T1a1 cells expressed AQP5 but not pro–SP-C (Figures 3C and 3D), indicating that T1a1 cells located in the alveolar epithelium were ATI cells. To determine whether T1a1 luminal cells in the interstitium observed via immunohistochemistry (Figure E1E) were lymphatic endothelial cells, we performed immunofluorescence staining, using antibodies against T1a and other markers for lymphatic endothelial cells (Prospero homeobox–1 [Prox-1] and lymphatic endothelial hyaluronan receptor–1 [LYVE-1]). We found that T1a1 luminal cells in the lung interstitium expressed Prox-1 and LYVE-1 (Figure 4). These data demonstrate that the T1a1 luminal cells in the interstitium were pulmonary lymphatic endothelial cells. In addition, immunostaining showed that lymphatic endothelial cells expressed T1a more strongly than did ATI cells (Figures 4A

and 4B). This observation was consistent with results obtained from FACS analyses (Figure 1A). Separation of Microvascular Endothelial Cells from Mesenchymal Cells in the EpCAM2/T1a2 Subset

To separate the vascular endothelial population from the mesenchymal population in the EpCAM2/T1a2 subset, we performed triple staining for EpCAM, T1a, and VE-cadherin. Dot plots showed that the VE-cadherin1 and VE-cadherin2 subsets were distinct subpopulations (Figure 5A). The number of cell types isolated is shown in Table 1. In contrast, we did not observe a distinct subpopulation in FACS scattergrams using anti-platelet/ endothelial cell adhesion molecule 1 (PECAM1), anti–vascular endothelial growth factor receptor–2, anti-CD34, or anti-endoglin antibodies (data not shown). To verify that the EpCAM2/T1a2/ VE-cadherin1 subset and the EpCAM2/T1a2/VE-cadherin2 subset predominately contained vascular endothelial cells and mesenchymal cells, respectively, we cultured each subset and characterized the cell phenotypes (n ¼ 6). Cells sorted from the EpCAM2/T1a2/VE-cadherin1 subset were cultured in endothelial culture medium containing several growth factors (e.g., vascular endothelial growth factor). The cultured VE-cadherin1 cells expressed vascular endothelial markers (PECAM1 and Tie2),but not mesenchymal markers (a-smooth muscle actin [a-SMA] and CD90) (Figures 5B and E3). In contrast, the EpCAM2/T1a2/VE-cadherin2 subpopulation cultured under the same endothelial culture conditions did not express PECAM1 or Tie-2 (data not shown). The VE-cadherin2 subset cultured in mesenchymal culture medium containing FBS, but not specific growth factors, displayed spindle-shaped cells expressing


427

Figure 4. Immunofluorescence staining of T1a in lymphatic endothelial cells. (A) T1a1 luminal cells in lung interstitium expressed Prospero homeobox– 1 (Prox-1) in the nuclei (arrowheads). Prox-1 is a transcription factor known to be expressed by lymphatic endothelial cells. Notably, lymphatic endothelial cells expressed T1a more strongly than did ATI cells (arrows). (B) T1a1 luminal cells co-expressed lymphatic endothelial hyaluronan receptor–1 (LYVE-1, a surface marker for lymphatic endothelial cells) in the perivascular region (asterisks). Alv, alveolus. Scale bars: A, 20 mm; B, 50 mm. Data are representative of five patients.

a-SMA and CD90, but not endothelial markers (Figures 5B and E3). The freshly isolated VE-cadherin1 subpopulation died after several days in the mesenchymal medium. An in vitro angiogenesis assay demonstrated that cells derived from the VE-cadherin1 subset formed capillary-like tubes in Matrigel (Figure 5C). On the other hand, the VE-cadherin2 subset formed corded aggregates, but not capillary-like tubes (Figure 5C), similar to dermal fibroblasts (23), suggesting that the VE-cadherin2 cells which expanded in the endothelial medium did not have the potential to become endothelial progenitor cells.

DISCUSSION We developed a FACS-based method for the direct isolation of individual component cell types from the alveoli of normal or diseased lungs. Most lung diseases develop in the distal lung. However, little cell-based knowledge is available, for example, about intercellular signaling or epigenetic changes in each cell type, because the disease sites are situated deep in the lungs and are thus difficult to approach. This difficulty motivated our development of a method to isolate individual ATII cells, microvascular endothelial cells, mesenchymal cells, and lymphatic endothelial cells from the multiple cell types residing in distal lung tissue. Our method uses a combination of antibodies against

EpCAM, T1a, and VE-cadherin (Figure 6). We also demonstrated a practical application for this method in the primary culture of each cell type. In addition, the approach described here can be extended to cell-specific analyses in health and disease, which will provide insights into pathologic processes and perhaps identify new therapeutic targets. ATII cells play a critical role in lung homeostasis and the repair process after injury (24, 25). Methodologies to separate ATII cells from lung cell suspensions have been devised for more than two decades (14, 20, 26–28). Notably, Demling and colleagues described magnetic cell sorting, using anti-EpCAM antibody–coated microbeads combined with a discontinuous Percoll density gradient to enable ATII cells to be more concentrated (15). We found that EpCAM was expressed by ATII cells more strongly than by other lung epithelial cells (Figures 1 and 3). Taken together, the isolation strategy using EpCAM is an efficient method for the separation of ATII cells. However, the expression of pro-SP-C was diminished in ATII cells isolated by the density-gradient method compared with the FACS method (Figure E1). Mechanical stress during centrifugation within a high-density solution may alter the function of ATII cells. The isolation method using a FACS is based on the expression level of EpCAM and T1a, and does not need a highdensity solution. Therefore, we propose that our approach can

428


Figure 5. The EpCAM2/T1a2 subset was fractionated into a vascular endothelial population and a mesenchymal population, using an antibody against vascular endothelial cadherin (VE-cadherin). (A) Representative dot plots show the expression of VE-cadherin in the EpCAM2/T1a2 subset. (B) Phase-contrast images and immunofluorescence staining for endothelial markers (PECAM1 and Tie-2) and a mesenchymal marker (a-SMA and CD90) in cells derived from VE-cadherin–positive and VE-cadherin–negative subpopulations. (C) An in vitro angiogenesis assay on the basement membrane extract. Data are representative of six patients. Scale bars: B, 100 mm; C, 500 mm. PECAM1, platelet/ endothelial adhesion molecule–1; a-SMA, a–smooth muscle actin.

preserve ATII cell function more rigorously than previous methods. In addition, we showed that ATII cells were isolated in a viable condition, and had a potential to generate ATI-like cells under culture condition (Figure 2). These results indicate that our methods for isolating and culturing ATII cells could apply to a variety of in vitro studies about ATII biology, such as toxicology, drug screening, and microbiology. Not only ATII cells, but other cell types, were isolated using our FACS-based method. We separated microvascular endothelial cells and mesenchymal cells, both of which were identified in

the EpCAM2/T1a2 subset. We tried using several surface antigens to delineate a microvascular endothelial subpopulation. These surface markers included established pulmonary vascular endothelial markers such as VE-cadherin (29), PECAM1 (30), and CD34 (31). We found that only VE-cadherin delineated distinct subpopulations (Figure 5A). A subsequent culture and in vitro angiogenesis assay showed that VE-cadherin distinctly separated microvascular endothelial cells from mesenchymal cells (Figures 5B and 5C). However, the microvascular endothelial cells and mesenchymal cells found in the human distal lung are


429

lung cells from patients can be applied to the recently described triple cell co-culture model (42) by means of reconstruction with diseased epithelial, mesenchymal, and endothelial cells. Our new isolation method is a unique and promising tool for the analysis of individual cells residing in distal lungs, and may shed light on the cellular biology within this borderland. Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgments: The authors are grateful to Mr. Katsuhiko Ono (Department of Pathology and Histotechnology, Tohoku University Graduate School of Medicine) for assistance with transmission electron microscopy, and Professor Ryouichi Nagatomi (Department of Biomedical Engineering, Tohoku University Graduate School of Biomedical Engineering) for his advice on this work. The authors also thank the Biomedical Research Core of Tohoku University Graduate School of Medicine for technical support.

References

Figure 6. A schema of the method for the direct isolation of lung component cells, based on the expression of EpCAM, T1a, and VEcadherin.

composed of heterogeneous subpopulations (24, 32). Further studies will be required to better characterize the cellular phenotypes in microvascular endothelial cells and mesenchymal cells. Lymphatic endothelial cells have received increasing attention, because they are thought to be associated with severe lung diseases such as idiopathic pulmonary fibrosis (33, 34), lymphangiomyomatosis (35), and metastatic cancer (36). In this study, we showed that the T1ahi population located in the lung interstitium consisted of lymphatic endothelial cells, based on immunofluorescence staining using lymphatic endothelial markers (Prox-1 and LYVE-1; Figure 4). Prox-1 was recognized as a master regulator of lymphatic endothelial phenotypes (37– 39). LYVE-1 is a hyaluronan receptor, and is expressed on the surface of lymphatic endothelial cells (40). Our histological data support the concept that the EpCAM2/T1ahi subset in FACS predominately contains lymphatic endothelial cells. However, a previous report demonstrated that pulmonary lymphatic endothelial cells consisted of two types of cells that ware functionally different (41). Endothelial cells in the initial lymphatics were oak leaf–shaped cells with button-like junctions, and were important in fluid uptake and the migration of leukocytes. Endothelial cells in collecting lymphatics, on the other hand, were conventional and continuous cells with zipper-like junctions to allow lymphatic fluid to pass through (41). We speculate that both types of endothelial cells might be contained in the EpCAM2/T1ahi subset. Therefore, further characterization will be needed to separate lymphatic endothelial cells that are basically heterogenous. In this study, we could not separate ATI cells from bronchiolar epithelial cells. ATI cells are thin, flat cells responsible for gas exchange (24). T1a is a specific marker for ATI cells in rat lungs (22), but as shown here, T1a was not a specific marker for ATI cells in human lungs (Figures 3 and E2). Thus, a better characterization of cell surface markers for ATI cell–specific isolation from human lungs is required. In conclusion, we demonstrated a novel method for the direct and individual isolation of lung component cells from distal lung tissue. Applying this method of isolation to bioinformatics or system biology may provide the molecular bases for pathologic processes in the development of lung disease. In addition, isolated

1. Franks TJ, Colby TV, Travis WD, Tuder RM, Reynolds HY, Brody AR, Cardoso WV, Crystal RG, Drake CJ, Engelhardt J, et al. Resident cellular components of the human lung: current knowledge and goals for research on cell phenotyping and function. Proc Am Thorac Soc 2008;5:763–766. 2. Taraseviciene-Stewart L, Voelkel NF. Molecular pathogenesis of emphysema. J Clin Invest 2008;118:394–402. 3. Selman M, King TE, Pardo A. Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann Intern Med 2001;134:136–151. 4. Mizgerd JP. Acute lower respiratory tract infection. N Engl J Med 2008; 358:716–727. 5. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334–1349. 6. Yousem SA, Beasley MB. Bronchioloalveolar carcinoma: a review of current concepts and evolving issues. Arch Pathol Lab Med 2007;131: 1027–1032. 7. Fuke S, Betsuyaku T, Nasuhara Y, Morikawa T, Katoh H, Nishimura M. Chemokines in bronchiolar epithelium in the development of chronic obstructive pulmonary disease. Am J Respir Cell Mol Biol 2004;31: 405–412. 8. Suzuki M, Betsuyaku T, Ito Y, Nagai K, Nasuhara Y, Kaga K, Kondo S, Nishimura M. Down-regulated NF-E2–related factor 2 in pulmonary macrophages of aged smokers and patients with chronic obstructive pulmonary disease. Am J Respir Cell Mol Biol 2008;39:673–682. 9. Nagai K, Betsuyaku T, Suzuki M, Nasuhara Y, Kaga K, Kondo S, Nishimura M. Dual oxidase 1 and 2 expression in airway epithelium of smokers and patients with mild/moderate chronic obstructive pulmonary disease. Antioxid Redox Signal 2008;10:705–714. 10. Sirianni FE, Chu FS, Walker DC. Human alveolar wall fibroblasts directly link epithelial Type 2 cells to capillary endothelium. Am J Respir Crit Care Med 2003;168:1532–1537. 11. Okaty BW, Sugino K, Nelson SB. A quantitative comparison of celltype–specific microarray gene expression profiling methods in the mouse brain. PLoS ONE 2011;6:e16493. 12. Fujino N, Ota C, Suzuki T, He M, Takahashi T, Yamada M, Suzuki S, Kondo T, Yamaya M, Kubo H. A new flow-cytometric method for isolating component cells individually from adult human distal lungs. Am J Respir Cell Mol Biol 2011;183:A5100. 13. Fujino N, Kubo H, Suzuki T, Ota C, Hegab AE, He M, Suzuki S, Yamada M, Kondo T, Kato H, et al. Isolation of alveolar epithelial Type II progenitor cells from adult human lungs. Lab Invest 2011;91: 363–378. 14. Fang X, Song Y, Hirsch J, Galietta LJ, Pedemonte N, Zemans RL, Dolganov G, Verkman AS, Matthay MA. Contribution of CFTR to apical– basolateral fluid transport in cultured human alveolar epithelial Type II cells. Am J Physiol Lung Cell Mol Physiol 2006;290:L242–L249. 15. Demling N, Ehrhardt C, Kasper M, Laue M, Knels L, Rieber EP. Promotion of cell adherence and spreading: a novel function of RAGE, the highly selective differentiation marker of human alveolar epithelial Type I cells. Cell Tissue Res 2006;323:475–488. 16. Arnaoutova I, Kleinman HK. In vitro angiogenesis: endothelial cell tube formation on gelled basement membrane extract. Nat Protoc 2010;5: 628–635.

430


17. Hodge SJ, Hodge GL, Holmes M, Reynolds PN. Flow cytometric characterization of cell populations in bronchoalveolar lavage and bronchial brushings from patients with chronic obstructive pulmonary disease. Cytometry B Clin Cytom 2004;61:27–34. 18. Fehrenbach H. Alveolar epithelial Type II cell: defender of the alveolus revisited. Respir Res 2001;2:33–46. 19. Kambouchner M, Bernaudin JF. Intralobular pulmonary lymphatic distribution in normal human lung using D2–40 antipodoplanin immunostaining. J Histochem Cytochem 2009;57:643–648. 20. Wang J, Edeen K, Manzer R, Chang Y, Wang S, Chen X, Funk CJ, Cosgrove GP, Fang X, Mason RJ. Differentiated human alveolar epithelial cells and reversibility of their phenotype in vitro. Am J Respir Cell Mol Biol 2007;36:661–668. 21. Kasper M, Behrens J, Schuh D, Muller M. Distribution of E-cadherin and Ep-CAM in the human lung during development and after injury. Histochem Cell Biol 1995;103:281–286. 22. Dobbs LG, Williams MC, Gonzalez R. Monoclonal antibodies specific to apical surfaces of rat alveolar Type I cells bind to surfaces of cultured, but not freshly isolated, Type II cells. Biochim Biophys Acta 1988;970: 146–156. 23. Bikfalvi A, Cramer EM, Tenza D, Tobelem G. Phenotypic modulations of human umbilical vein endothelial cells and human dermal fibroblasts using two angiogenic assays. Biol Cell 1991;72:275–278. 24. Herzog EL, Brody AR, Colby TV, Mason R, Williams MC. Knowns and unknowns of the alveolus. Proc Am Thorac Soc 2008;5:778–782. 25. Crosby LM, Waters CM. Epithelial repair mechanisms in the lung. Am J Physiol Lung Cell Mol Physiol 2010;298:L715–L731. 26. Bingle L, Bull TB, Fox B, Guz A, Richards RJ, Tetley TD. Type II pneumocytes in mixed cell culture of human lung: a light and electron microscopic study. Environ Health Perspect 1990;85:71–80. 27. Cunningham AC, Milne DS, Wilkes J, Dark JH, Tetley TD, Kirby JA. Constitutive expression of MHC and adhesion molecules by alveolar epithelial cells (Type II pneumocytes) isolated from human lung and comparison with immunocytochemical findings. J Cell Sci 1994;107: 443–449. 28. Elbert KJ, Schafer UF, Schafers HJ, Kim KJ, Lee VH, Lehr CM. Monolayers of human alveolar epithelial cells in primary culture for pulmonary absorption and transport studies. Pharm Res 1999;16:601–608. 29. Herwig MC, Muller KM, Muller AM. Endothelial VE-cadherin expression in human lungs. Pathol Res Pract 2008;204:725–730. 30. Hewett PW, Murray JC. Human lung microvessel endothelial cells: isolation, culture, and characterization. Microvasc Res 1993;46:89–102.

31. Muller AM, Hermanns MI, Skrzynski C, Nesslinger M, Muller KM, Kirkpatrick CJ. Expression of the endothelial markers PECAM-1, vWF, and CD34 in vivo and in vitro. Exp Mol Pathol 2002;72: 221–229. 32. Stevens T, Phan S, Frid MG, Alvarez D, Herzog E, Stenmark KR. Lung vascular cell heterogeneity: endothelium, smooth muscle, and fibroblasts. Proc Am Thorac Soc 2008;5:783–791. 33. El-Chemaly S, Malide D, Zudaire E, Ikeda Y, Weinberg BA, PachecoRodriguez G, Rosas IO, Aparicio M, Ren P, MacDonald SD, et al. Abnormal lymphangiogenesis in idiopathic pulmonary fibrosis with insights into cellular and molecular mechanisms. Proc Natl Acad Sci USA 2009;106:3958–3963. 34. Ebina M, Shibata N, Ohta H, Hisata S, Tamada T, Ono M, Okaya K, Kondo T, Nukiwa T. The disappearance of subpleural and interlobular lymphatics in idiopathic pulmonary fibrosis. Lymphat Res Biol 2010;8:199–207. 35. Kumasaka T, Seyama K, Mitani K, Souma S, Kashiwagi S, Hebisawa A, Sato T, Kubo H, Gomi K, Shibuya K, et al. Lymphangiogenesismediated shedding of LAM cell clusters as a mechanism for dissemination in lymphangioleiomyomatosis. Am J Surg Pathol 2005;29: 1356–1366. 36. Tammela T, Alitalo K. Lymphangiogenesis: molecular mechanisms and future promise. Cell 2010;140:460–476. 37. Wigle JT, Oliver G. Prox1 function is required for the development of the murine lymphatic system. Cell 1999;98:769–778. 38. Wigle JT, Harvey N, Detmar M, Lagutina I, Grosveld G, Gunn MD, Jackson DG, Oliver G. An essential role for Prox1 in the induction of the lymphatic endothelial cell phenotype. EMBO J 2002;21:1505– 1513. 39. Johnson NC, Dillard ME, Baluk P, McDonald DM, Harvey NL, Frase SL, Oliver G. Lymphatic endothelial cell identity is reversible and its maintenance requires Prox1 activity. Genes Dev 2008;22:3282– 3291. 40. Baluk P, McDonald DM. Markers for microscopic imaging of lymphangiogenesis and angiogenesis. Ann N Y Acad Sci 2008;1131:1–12. 41. Baluk P, Fuxe J, Hashizume H, Romano T, Lashnits E, Butz S, Vestweber D, Corada M, Molendini C, Dejana E, et al. Functionally specialized junctions between endothelial cells of lymphatic vessels. J Exp Med 2007;204:2349–2362. 42. Huh D, Matthews BD, Mammoto A, Montoya-Zavala M, Hsin HY, Ingber DE. Reconstituting organ-level lung functions on a chip. Science 2010;328:1662–1668.