Tissue-specific Microvascular Endothelial Cell Lines from H-2K

[CANCER RESEARCH 63, 2971–2976, June 1, 2003]

Tissue-specific Microvascular Endothelial Cell Lines from H-2Kb-tsA58 Mice for Studies of Angiogenesis and Metastasis1 Robert R. Langley, Karen M. Ramirez, Rachel Z. Tsan, Melissa Van Arsdall, Monique B. Nilsson, and Isaiah J. Fidler2 Department of Cancer Biology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

ABSTRACT Microvascular endothelial cells play a critical role in tumor progression and metastasis by forming capillary networks that encourage tumor growth and by promoting the attachment of circulating tumor cells to the vascular wall of distant tissues. Efforts to study the molecular mechanisms that mediate these complex processes in different anatomical compartments have been impeded by difficulties in the isolation and propagation of endothelial cells from different organs. To overcome these limitations, we used two-color flow cytometry to identify and select microvascular endothelial cells from primary cultures obtained from different organs of mice whose tissues harbor a temperature-sensitive SV40 large T antigen (H-2Kb-tsA58 mice; ImmortoMice). The selection strategy targeted cell populations expressing the inducible endothelial cell adhesion molecules, E-selectin and VCAM-1, and proved successful in generating microvascular endothelial cell lines from a number of different organs. When cultured under permissive temperatures (33°C), individual cell lines displayed doubling times consistent with endothelial cells possessing an angiogenic phenotype. The transfer of endothelial cells to nonpermissive temperatures (37°C) resulted in cell differentiation and the induction of a quiescent state. Established cell lines exhibited several inherent endothelial properties, including the expression of constitutive and inducible levels of endothelial cell adhesion molecules E-selectin, intercellular adhesion molecule-1, and vascular cell adhesion molecule-1, internalization of acetylated low-density lipoprotein, and formation of loop structures on Matrigel surfaces. The immortalized endothelial cell lines established from H-2Kb-tsA58 mice provide, for the first time, a cell culture system to examine factors regulating angiogenesis and tumor cell arrest in different organ systems.

INTRODUCTION It is now firmly established that the growth and metastasis of the vast majority of neoplasms are dependent on the formation of adequate vasculature, i.e., angiogenesis (1–3). The onset of angiogenesis appears to be the result of an imbalance between stimulatory and inhibitory factors that leads to the activation of previously quiescent endothelial cells (4, 5). Activation of endothelial cells elicits a complex series of responses that include the elaboration of proteolytic enzymes, migration, and proliferation. Endothelial cells begin to branch off from preexisting microvessels and form capillary sprout structures that will ultimately coalesce and perfuse the tumor tissue. In addition to providing metabolic support to the expanding tumor population, the newly formed vessels also facilitate tumor cell entry into the systemic circulation (6). Circulating tumor cells then exploit endothelial cell receptors in distal vascular beds to promote their arrest and retention in target organs (7). Given that endothelial cells play a central role in several rate-limiting steps of metastasis, the Received 11/27/02; accepted 4/2/03. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported in part by Cancer Center Support Core Grant CA16672, SPORE in Prostate Cancer Grant CA90270, SPORE in Ovarian Cancer Grant CA93639, and SPORE in Head and Neck Cancer Grant CA97007 from the National Cancer Institute, NIH. 2 To whom requests for reprints should be addressed, at the Department of Cancer Biology (Unit 173), The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: (713) 792-8577; Fax: (713) 792-8747; E-mail: [email protected].

interactions that take place between tumor cells and endothelial cells have been the focus of much investigation. However, efforts to examine the in vitro behavior of tumor cells with endothelial cells from different anatomical regions have been prohibited because, in large part, of the limited number of organ-derived endothelial cell lines available for study. One of the major impediments to obtaining a large number of endothelial cells from different tissues has been the inability to purify and propagate these cells in culture. Obtaining pure populations of endothelial cells has been difficult because many of the markers used to distinguish endothelial cells are also expressed by a variety of other vascular cells (8). Once endothelial cells are removed from a heterogeneous population of cells, it is often difficult to obtain a sufficient number of cells for experimental analysis. In fact, estimations of the normal capillary proliferation rate predict that only 0.01% of endothelial cells are actively engaged in replication at any given time (9). Commercially available endothelial cell lines may alleviate some of the purification concerns, but, at present, cells from only a very few tissues are available. For these reasons, many of the examinations between tumor and endothelial cells have been performed on HUVEC3 lines. Because HUVECs originate from a region that is not associated with metastasis, it remains unclear whether these cells provide a relevant model system. To overcome the above-mentioned limitations, we established a number of organ-derived cultures from transgenic mice whose tissues harbor a temperature-sensitive SV40 large T antigen (H-2Kb-tsA58 mice; Ref. 10). When primary cultures from these mice are stimulated with proinflammatory cytokines, the endothelial cell fraction responds by up-regulating the inducible endothelial cell adhesion molecules, E-selectin and VCAM-1. This subpopulation of cells is then targeted for selection through the application of a series of stringent, FACS protocols. The resulting endothelial cell lines described in this study retain their phenotype after repeated passage, require minimal nutritional support and, importantly, may be useful for identifying factors that regulate endothelial cell proliferation and tumor cell adhesion in different anatomical regions.

MATERIALS AND METHODS Antibodies. The following antibodies were titrated and used in the isolation and/or characterization of microvascular endothelial cell lines: PE-conjugated anti-E-selectin (10E9.6), FITC-conjugated anti-E-selectin (10E9.6), FITC-conjugated anti-VCAM-1 (429), anti-ICAM-1 (3E2), and two isotype standards, FITC rat IgG2a and PE rat IgG2a, all obtained from PharMingen (San Diego, CA); anti-flg (FGFR-1; C-15), anti-Flk-1 (C1158), anti-E-selectin (Y-18), and Tie-2 (C20) from Santa Cruz Biotechnology (Santa Cruz, CA); and anti-VCAM (MCA1229) from Serotech (Oxford, United Kingdom). Secondary reagents included goat antirat horseradish peroxidase and goat antirabbit peroxidase-conjugated F(ab⬘)2 fragments both from Jackson ImmunoRe3 The abbreviations used are: HUVEC, human umbilical vein endothelial cell; DiIAc-LDL, 1,1⬘-dioctadecyl-3,3,3⬘,3⬘-tetramethyl-indocarbocyanine acetylated low-density lipoprotein; FACS, fluorescence-activated cell sorting; FGFR, fibroblast growth factor receptor; Flk, fetal liver kinase; ICAM, intercellular adhesion molecule; mAb, monoclonal antibody; PE, phycoerythrin; TNF, tumor necrosis factor; VCAM, vascular cell adhesion molecule.

2971

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

ORGAN-DERIVED ENDOTHELIAL CELL LINES

search (West Grove, PA) and Alexa fluorescent 594 antigoat IgG and Alexa fluorescent 594 goat antirabbit complex from Molecular Probes (Eugene, OR). A tyramide signal amplification reagent (NEN Life Science Products, Boston, MA) was also used in the study according to the manufacturer’s instructions. Primary Culture Conditions. Five male and five female mice, homozygous for a temperature-sensitive SV40 large T antigen (ImmortoMice; CBA/ca X C57Bl/10 hybrid; Charles River Laboratories), were killed by cervical dislocation, and multiple organs were harvested aseptically under a laminar flow hood. Organs were placed into 100-mm plates containing ice-cold HBSS and cut into small (1 mm) fragments. The fragments were collected into 50-ml polypropylene centrifuge tubes containing 0.2% type IV collagenase (Sigma Chemical Co., St. Louis, MO) and then immersed in a 37°C water bath for 1 h. Tubes were centrifuged at 200 ⫻ g for 10 min at 10°C. The supernatants were discarded, and the tissue was resuspended in DMEM supplemented with 10% fetal bovine serum, 2-mM L-glutamine, sodium pyruvate, nonessential amino acids, and a vitamin solution (Life Technologies, Inc., Rockville, MD). The suspensions were passed through a sterile 100-␮m stainless steel tissue sieve, and the filtered fractions were plated into 75-cm2 flasks (coated previously with 1% gelatin). Twenty-four hours later, the medium was discarded, and fresh 10% DMEM containing 10 units/ml IFN-␥ (PharMingen) was added. The addition of IFN-␥ was used to enhance the expression of the MHC H-2Kb class I promoter, which regulates the level of large T antigen protein in ImmortoMouse-derived cells (10). Primary cultures were supported at 33°C in a mixture of 5% carbon dioxide and 95% oxygen, and the media was replaced as needed. Establishment of Endothelial Cell Lines. Primary isolates were grown to confluence (⬃10 –20 days), at which time the medium was replaced with 10% DMEM containing 10 ng/ml recombinant murine TNF-␣ (R&D Systems, Minneapolis, MN). After a 5-h incubation period, the cells were detached from the flask with a 0.25% trypsin-0.02% EDTA solution (volume for volume). Harvested cells were centrifuged for 5 min at 200 ⫻ g and then prepared for cell sorting. The endothelial cell fraction was labeled by resuspending the pellet in 2% DMEM containing 4 ␮g/ml PE-conjugated rat antimouse Eselectin mAb and 2 ␮g/ml FITC-conjugated rat antimouse VCAM-1 mAb. Cells were incubated in this antibody-containing solution for 45 min at 4°C, washed twice, and then resuspended in 2% DMEM. Murine endothelioma cells that express several endothelial markers (Ref. 11 and our observations), stimulated with 10 ng/ml TNF-␣, served as positive controls. Primary isolates that were incubated with identical concentrations of FITC- and PE-conjugated isotype standards were used to assess the level of background intensity. Cell staining was evaluated with a Beckman Epics Elite flow cytometer (Beckman Coulter, Miami, FL) equipped with an air-cooled argon ion laser. The emission wavelengths used for recognition of FITC and PE labeling were 520 and 575 nm, respectively. Gating parameters were adjusted based on the fluorescence histograms for the positive and negative controls. Cells that expressed both VCAM-1 and E-selectin were collected in sterile tubes containing 10% DMEM and plated on gelatin-coated T-25 flasks. After the cultures reached confluence (⬃10 days), the cells were subjected to a second FACS-based selection using the methodology described above or a slight modification applied to bone cells, which are known to possess stromal cell populations that express VCAM-1 (12). Specifically, cell cultures from bone tissue were incubated in 10% DMEM containing 10 ng/ml TNF-␣ and 10 ␮g/ml fluorescent probe of acetylated LDL, DiI-Ac-LDL (Biochemical Technologies, Cambridge, MA), for 4 h. Cells were harvested in the manner described above with the exception that during this labeling period, the rat anti-E-selectin mAb (10E9.6) conjugated to FITC replaced PE-conjugated 10E9.6. Individual cultures of 3T3 (a murine fibroblast line) and endothelioma cells subjected to identical treatment were used for negative and positive controls, respectively. The emission wavelength used for cell sorting of DiIAc-LDL-labeled cells was 550 nm. Positive cells were identified and sorted based on upper and lower intensity values. Cells that had undergone two rounds of selection were maintained in 10% DMEM without IFN-␥. Immunohistochemical Analysis. Endothelial cells plated onto two-chamber slides at a density of 2 ⫻ 105 cells/well in 10% DMEM were incubated overnight. The cells were then washed once, and the medium was replaced with fresh 10% DMEM. To determine inducible endothelial cell adhesion molecule expression (E-selectin, VCAM-1, and ICAM-1), some of the chambers were given medium containing 10 ng/ml TNF-␣ for 4 – 6 h (these time points were selected to correlate with the known kinetics of endothelial cell

adhesion molecule expression; Refs. 13 and 14). Tissue fixation for E-selectin, VCAM-1, ICAM-1, and the basic fibroblast growth factor, flg, was carried out using a protocol consisting of three sequential immersions in ice-cold solutions containing acetone, acetone-chloroform (50:50 volume for volume), and acetone (5 min each). Fixation using 4% paraformaldehyde at room temperature for 5 min was used to determine expression of Tie-2 and Flk-1. Slides were rinsed in PBS three times (5 min each) and blocked in PBS containing 5% normal horse serum and 1% normal goat serum. The slides were incubated with the primary antibody overnight at 4°C, rinsed three times with PBS, incubated for 10 min in protein blocking solution, and incubated with the appropriate secondary reagent. To stain endothelial cells for adhesion molecules, a peroxidase-conjugated goat antirat IgG F(ab⬘)2 fragment was added for 45 min at room temperature and then removed by washing with PBS. The slides were then incubated for 10 min in a biotin tyramide solution, washed twice, and incubated for 45 min in streptavidin-conjugated Alexa fluorescent 594. Staining for all other determinants was accomplished using a standard two-step procedure with Alexa 594 serving as the visualization reagent. After the slides containing the fluorescent label were washed twice in PBS, cell nuclei were stained with Hoescht 33342 (Polysciences, Inc., Warrington, PA) for 2 min followed by washing in PBS. Fluorescence bleaching was minimized by mounting the slides with glycerol/ PBS medium containing 0.1 M propyl gallate (Sigma). Immunofluorescence microscopy was performed using a Zeiss Axioplan fluorescence microscope (Carl Zeiss, Inc., Thornwood, NY) equipped with a 100-W Hg lamp and narrow band pass excitation filters (Chroma Technology Corp., Brattleboro, VT). Images were captured with a cooled charged coupled device Hamamatsu C5810 camera (Hamamatsu Photonics K.K., Bridgewater, NJ) and Optimas software (Media Cybernetics, Silver Spring, MD) on a Dell computer (Round Rock, TX). Composite photographs were made using PhotoShop software (Adobe Systems, Mountain View, CA). Determination of Doubling Time. Endothelial cells (passage 5) from each line were plated onto 96-well plates at a density of 1000 cells/well in 10% DMEM. Cell growth was evaluated under both permissive (33°C) and nonpermissive (37°C) temperatures. The proliferative activity was determined every 24 h by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenol-tetrazolium bromide assay using a Dynatech MR-5000 96-well microtiter plate reader set at 570 nm. Increase in absorbance was considered a measure of cell proliferation (15, 16). Analysis of Endothelial Cell Tube Formation on Matrigel. The ability of the endothelial cells to form capillary-like structures was assessed by placing them on the solubilized basement membrane preparation Matrigel (Becton Dickinson, San Jose, CA), according to the manufacturer’s instructions. In brief, Matrigel (1 mg/ml) was placed on ice and thawed overnight at 4°C, and then 120 ␮l were applied to each well of a precooled 48-well plate using chilled pipette tips. Matrigel-containing plates were allowed to incubate at 37°C for 30 min and then seeded with endothelial cells at a density of 4 ⫻ 104 cells/well in 10% DMEM. Tube formation was monitored with a Leica DMIL inverted microscope (Deerfield, IL) equipped with a VI-470 camera (Optronics Engineering, Goleta, CA). Images were evaluated using the Optimas software package. Internalization of Acetylated LDL. Endothelial cells that had undergone two sessions of FACS-based selection were evaluated for their ability to metabolize fluorescence-labeled acetylated LDL (DiI-Ac-LDL). Cells from each organ-specific line were plated onto two-chamber slides at a density of 1 ⫻ 105 cells/chamber and allowed to incubate overnight. Cells were washed once with serum-free DMEM and then incubated in 10% DMEM containing 10 ␮g/ml DiI-Ac-LDL for 4 h. Endothelioma cells and 3T3 fibroblast cells were used as a positive and negative reference, respectively. Cells were washed twice with label-free medium, and the slides were fixed in 4% paraformaldehyde for 10 min. DiI-Ac-LDL internalization was evaluated on an Axioplan fluorescence microscope. Captured images were evaluated with Optimas software.

RESULTS Endothelial Cell Selection Using Flow Cytometry. One of the primary objectives of this study was to establish a protocol to facilitate the identification and selection of microvascular endothelial cells

2972

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

ORGAN-DERIVED ENDOTHELIAL CELL LINES

Fig. 1. Fluorescence histograms obtained from primary culture of colon cells (A) and at the time of second sorting cycle (B). Cells were exposed to TNF-␣ for 5 h (10 ng/ml) and labeled with a PE-conjugated anti-E-selectin mAb and FITC-conjugated antiVCAM-1 mAb.

from primary organ cultures with a high degree of precision. Pilot studies using a variety of strategies suggested that this goal could be best achieved by stimulating primary organ cell cultures with TNF-␣ (10 ng/ml) and targeting cells expressing the endothelial cell adhesion molecules E-selectin and VCAM-1 through a FACS-driven strategy.

Fig. 1, A and B show the fluorescent histograms obtained from colon cell cultures during an initial and second session of cell sorting. The patterns displayed in these two histograms are representative of those obtained for all other tissues. On average, the number of cells expressing both E-selectin and VCAM-1 found in primary cultures ranged from 3 to 10% of the total cell population. The highest level of expression was found in heart and bladder endothelial cells. Efforts to promote a selective increase in the number of endothelial cells present in primary cultures by using different media preparations (MCDB131, D-Valine, etc.) were largely unsuccessful. The percentage of cells expressing both E-selectin and VCAM-1 (or DiI-Ac-LDL for bone-derived cultures) after 5 h of TNF-␣ stimulation in the second series of selection increased significantly to reach values of 30 – 40% for most cultures. It was at this stage that highly purified populations of endothelial cells could be obtained by adjusting the gating parameters to select only the most intensely labeled cells. Characterization of Endothelial Cell Morphology. Three types of structural variations were visible in the tissue-specific endothelial cell lines (Fig. 2). Cultured endothelial cells from lung, colon, kidney, bladder, liver, and ovary displayed a cobblestone appearance after

Fig. 2. Phase-contrast images of organ-derived endothelial cells (37°C culture conditions) that had undergone two successive selection processes by flow cytometry. Scale bar, 200 ␮m.

2973

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

ORGAN-DERIVED ENDOTHELIAL CELL LINES

reaching confluence. Endothelial cells from bone, heart, prostate, and brain tended to form a slightly more rounded appearance at confluence. Endothelial cells from uterus and pancreas displayed a more spindle-shaped appearance that was most apparent in proliferating cultures. Growth Characteristics. The rate of cell proliferation varied only slightly between the different endothelial cell lines (Fig. 3). The majority of cell lines growing at the permissive temperature (33°C) underwent an initial population doubling within 48 h. The tapering off of cell growth after this period may be a reflection of nutrient depletion or cell– cell contact inhibition of growth. Similar to cells growing at 33°C, endothelial cells plated in the 37°C environment also engaged in cell division during the first 24 h. However, during the next 72 h, the growth of most endothelial cells in 37°C conditions dropped to 50% of that of their counterparts growing at 33°C (data not shown). Endothelial cells from heart, bladder, and colon provided notable exceptions as they continued to proliferate despite the transition to 37°C. However, after one to two additional passages, these endothelial cells’ proliferation was significantly reduced. Immunohistochemical Staining. The constitutive and inducible expression of the endothelial cell adhesion molecules E-selectin, ICAM-1, and VCAM-1 was assessed by immunohistochemistry (Figs. 4, A–C). As expected, all of the endothelial cell lines showed diminished constitutive expression of E-selectin (data not shown) and VCAM-1 (Fig. 4C) in comparison with surface levels of ICAM-1 (Fig. 4B). However, after 5 h of stimulation with the cytokine TNF-␣, a marked increase in labeling of each of these inducible glycoproteins was found. The ability to up-regulate endothelial cell adhesion molecules after exposure to cytokines was unaffected by repeated subculturing of cells, because we noted identical staining patterns for passage 30 cells (we did attempt to extend this observation to include later passage cells). We also noted constitutive surface expression of the tyrosine kinase receptors Tie-2, FGFR-1, and Flk-1 on proliferating and resting endothelial cells (data not shown). Formation of Loop Structures on Matrigel. Each of the endothelial cell lines demonstrated an ability to orient into capillary-like structures when placed onto a Matrigel surface (Fig. 5A). The onset of tube structure formation could be visualized as early as 4 h after plating for most endothelial cell lines, with extensive networks becoming visible within 12 h after plating. Internalization of Acetylated LDL. With only one exception, all of the endothelial cells (at 33°C or 37°C) from the different cell lines demonstrated endocytosis of DiI-Ac-LDL (Fig. 5B). Endothelial cells derived from the brain demonstrated a complete absence of uptake after 4 h of incubation (at 33°C or 37°C) with this fluorescent reagent (data not shown). At both 33°C and 37°C, the reference cell lines used

Fig. 3. Cell proliferation analysis of endothelial cells growing in 10% DMEM at 33°C (permissive) conditions. Endothelial cells were plated at 1 ⫻ 103 cells/well in a series of 96-well plates, and proliferation was assessed every 24 h by 3-(4,5-dimethylthiazol-2-yl)2,5-diphenol-tetrazolium bromide analysis. SE of 10 determinations per cell line is shown.

in this assay, endothelioma and 3T3 fibroblast cells, showed positive and negative staining, respectively (data not shown). DISCUSSION We capitalized on one of the innate responses of microvascular endothelial cells, i.e., the up-regulation of cell adhesion molecules in response to inflammatory agents, to generate endothelial cell lines from different organs. The inducible cell adhesion molecules E-selectin and VCAM-1 are differentially expressed along the surface of the vascular endothelium and particularly enriched at the level of the postcapillary venule (17), where E-selectin mediates leukocyte and VCAM-1, lymphocyte recruitment processes (14). Thus, our selection strategies, which were based on targeting endothelial populations that expressed both of these inducible glycoproteins, suggest that the endothelial cell lines generated in this study are most likely from this microcirculatory region. The mouse line (H-2Kb-tsA58; ImmortoMouse) used in this study has been exploited previously to provide a number of conditionally immortalized epithelial lines for experimental study (18 –20). This murine model system offers a superior source of immortalized cell populations as many of the difficulties associated with the in vitro transfection process are eliminated (e.g., initial requirement for many cells, different sites of gene integration, multiple copy number; Ref. 10). One of the most attractive features of this model is the presence of the thermolabile large T antigen, which allows the user to regulate the level of cell differentiation. The large T antigen of the ImmortoMouse is under the control of the MHC H-2Kb class I promoter and, thus, cell proliferation can be augmented by exposing cultures to agents such as IFN-␥. Indeed, we found that the addition of 10 units/ml IFN-␥ to primary cultures and first-round sorted cells significantly reduced the time period necessary to obtain purified endothelial cell populations. However, because IFN-␥ is widely known to influence several endothelial properties, such as junctional integrity (21) and adhesion molecule expression (22), we elected to remove this cytokine from cultures immediately after the second session of cell selection. Despite the omission of IFN-␥ from the culture system, the doubling times of endothelial cells growing in a 33°C environment averaged ⬃48 h, a rate of cell division which agreed with reports published previously of endothelial cell proliferation found in pathologic tissues (23, 24). This activated phenotype of endothelial cells may be regulated rather easily and cells directed toward a more differentiated state by making slight adjustments in the incubation temperature. We have observed on several occasions that, for the majority of endothelial cell lines, the transfer to 37°C conditions for 72 h correlated with a 50% reduction in cell proliferation, a value that continued to decrease with time. Indeed, we found that in cell lines maintained at the nonpermissive temperature (37°C) for two to three additional passages, cells began to assume characteristics associated with a senescent phenotype (formation of giant cells, absence of cell division). Previous reports examining the kinetics of cell division in H-2Kb-tsA58-derived cell lines have determined that the cessation of cell division seen at nonpermissive conditions directly correlates with the steady-state decline in large T antigen protein (25). Once endothelial cells from different tissues were obtained, we evaluated the effects of prolonged culture conditions on several physiological properties and found that in all of the cultures that had been subjected to two sessions of selection with flow cytometry, TNF-␣ could elicit the characteristic up-regulation of the endothelial cell adhesion molecules E-selectin, ICAM-1, and VCAM-1. This finding was not surprising given that the cell lines were established based on the inducible expression of two of these glycoproteins (E-selectin and VCAM-1). However, unlike HUVECs or other commercially available endothelial cell lines, the endothelial cells generated from the

2974

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

ORGAN-DERIVED ENDOTHELIAL CELL LINES

Fig. 4. Representative examples of immunohistochemical staining of endothelial cell adhesion molecules under constitutive and cytokine-stimulated (TNF-␣: 10 ng/ml, 37°C) conditions. A, comparison of E-selectin expression in control colon (top left panel) to TNF-␣-stimulated expression in colon (center panel) and ovarian (right panel) microvascular endothelial cells. Staining was evaluated 4 h after TNF-␣ exposure. Scale bar, 100 ␮m. B, lung microvascular ICAM-1 staining in (left to right) control, constitutive, and TNF-␣ (5 h)-stimulated conditions. Scale bar, 200 ␮m. C, evaluation of distribution of VCAM-1 on bladder endothelial cells in (left to right) control, constitutive, and TNF-␣ (5 h)-stimulated conditions. Scale bar, 200 ␮m.

H-2Kb-tsA58 mice retained their ability to mobilize these adhesion molecules despite undergoing numerous population doublings (currently 30 passages). The constitutive expression of these endothelial cell adhesion molecules was very similar for most of the lines derived from different organs, i.e., very low expression levels of E-selectin and VCAM-1 and more pronounced expression of ICAM-1. We also examined several of the cell lines for the presence of additional receptors that are generally considered characteristic of the microvascular phenotype. Using immunohistochemistry, we detected basal levels of the tyrosine kinase receptors Flk-1, FGFR-1, and Tie-2.4 The downstream signaling pathways of some growth factor receptors was also intact because many of the cell lines exhibited proliferative responses to certain of the known endothelial cell mitogens.5 Two additional features frequently associated with, but not exclusive to, endothelial cells are the ability to orient into tube-like structures when placed on a surface containing Matrigel extract and a 4 5

Unpublished observations. Manuscript in preparation.

propensity to internalize and degrade chemically modified LDL. Each of the endothelial cell lines established from the H-2Kb-tsA58 transgenic mice was capable of generating vascular-like channels on Matrigel within 12 h. We chose to evaluate the ability of each endothelial cell line to internalize DiI-Ac-LDL because this reagent has been used in the past as a selection agent to establish different endothelial cell lines (26), and the internalization of DiI-Ac-LDL is one of the listed specifications of commercially available endothelial cell lines. Only endothelial cells derived from the brain showed a complete lack of internalization of the fluorescent probe. This latter observation adds to the expanding body of evidence indicating that brain microvascular endothelial cells lack receptors for chemically modified LDL (27, 28). The cell culture system described here provides, for the first time, a means with which to perform controlled detailed examinations on both the activated and differentiated phenotype of organ-specific microvascular endothelial cells. Many of the endothelial cell lines described here have been considered previously inaccessible for study of tumor angiogenesis, progression, and metastasis. Whether these

2975

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

ORGAN-DERIVED ENDOTHELIAL CELL LINES

Fig. 5. Phase-contrast images of organ-derived endothelial cells (37°C) on Matrigel surface (A) and after 4 h of incubation with DiI-Ac-LDL (B). In A, endothelial cells were plated at a density of 4 ⫻ 104 cells/well on 48-well plates that had been coated previously with Matrigel (1 mg/ml) and visualized 8 h later. Scale bar, 200 ␮m. In B, endothelial cells from different tissues were seeded at a density of 1 ⫻ 105 cells/chamber in two-chamber slides and incubated with 10 mg/ml DiI-Ac-LDL in 10% DMEM for 4 h. Cells were washed with label-free medium and fixed in 4% paraformaldehyde. Scale bar, 50 ␮m.

endothelial cell lines retain specialized tissue-specific phenotype and whether the phenotype of these endothelial cells can be maintained under in vivo conditions is now under active investigation. ACKNOWLEDGMENTS We thank Dominic Fan and Ewa Kruzel for sharing their expertise in cell culture systems, Walter Pagel for critical editorial review, and Lola Lo´ pez for expert assistance in the preparation of this manuscript.

REFERENCES 1. Folkman, J., and Shing, Y. Angiogenesis. J. Cell Biol., 267: 10931–10934, 1992. 2. Folkman, J. What is the evidence that tumors are angiogenesis dependent? J. Natl. Cancer Inst. (Bethesda), 82: 4 – 6, 1990. 3. Fidler, I. J. Critical determinants of metastasis. Semin. Cancer Biol., 12: 89 –96, 2002. 4. Hanahan, D., and Folkman, J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell, 86: 353–364, 1996. 5. Bouck, N., Stellmach, V., and Hsu, S. How tumors become angiogenic. Adv. Cancer Res., 69: 135–174, 1996. 6. Liotta, L. A., Kleinerman, J., and Saidel, G. M. Quantitative relationships of intravascular tumor cells, tumor vessels, and pulmonary metastases following tumor implantation. Cancer Res., 34: 997–1004, 1974. 7. Orr, F. W., Wang, H. W., Lafrenie, R. M., Scherbarth, S., and Nance, D. M. Interactions between cancer cells and the endothelium in metastasis. J. Pathol., 190: 310 –329, 2000. 8. Garlanda, C., and Dejana, E. Heterogeneity of endothelial cells-specific markers. Arterioscler. Thromb. Vasc. Biol., 17: 1193–1202, 1997. 9. Engerman, R. L., Pfaffenbacvh, D., and Davis, M. D. Cell turnover of capillaries. Lab. Investig., 17: 738 –743, 1967. 10. Jat, P. S., Noble, M. D., Ataliotis, P., Tanaka, Y., Yannoutsos, N., Larsen, L., and Kioussis, D. Direct derivation of conditionally immortal cell lines from an H-2KbtsA58 transgenic mouse. Proc. Natl. Acad. Sci. USA, 88: 5096 –5100, 1991. 11. Obeso, J., Weber, J., and Auerbach, R. A hemangioendothelioma-derived cell line: its use as a model for the study of endothelial cell biology. Lab. Investig., 63: 259 –269, 1990. 12. Simmons, P. J., Masinovsky, B., Longenecker, B. M., Berenson, R., Torok-Storb, B., and Gallatin, W. M. Vascular cell adhesion molecule-1 expressed by bone marrow stromal cells mediates the binding of hematopoietic progenitor cells. Blood, 80: 388 –395, 1992. 13. Henninger, D. D., Pane´ s, J., Eppihimer, M. J., Russell, J., Gerritsen, M., Anderson, D. C., and Granger, D. N. Cytokine-induced VCAM-1 and ICAM-1 expression in different organs of the mouse. J. Immunol., 158: 1825–1832, 1997.

14. Bevilacqua, M. Endothelial-leukocyte adhesion molecules. Annu. Rev. Immunol., 11: 767– 804, 1993. 15. Fan, D., Chakrabarty, S., Scid, C., Bell, C. W., Schackert, H., Morikawa, K., and Fidler, I. J. Clonal stimulation or inhibition of human colon carcinomas and human renal carcinomas mediated by transforming growth factor-␤-1. Cancer Commun., 1: 117–125, 1989. 16. Fan, D., Bucana, C. D., O’Brian, C. A., Zwelling, L., Scid, C., and Fidler, I. J. Enhancement of murine tumor cell sensitivity to adriamycin by presentation of the drug in phosphatidylcholine-phosphatidylserine in liposomes. Cancer Res., 50: 3619 – 3626, 1990. 17. Ley, K., and Gaehtgens, P. Endothelial, not hemodynamic, differences are responsible for preferential leukocyte rolling in rat mesenteric venules. Circ. Res., 69: 1034 –1041, 1991. 18. Chambers, T., Owens, J., Hattersly, G., Jat, P., and Noble, M. Generation of osteoclast-inductive and osteoclastogeneic cell lines from the H-2Kb-tsA58 transgenic mouse. Proc. Natl. Acad. Sci. USA, 90: 5578 –5582, 1993. 19. Jagger, D. J., Griesinger, C. B., Rivolta, M. N., Holly, M. C., and Ashmore, J. F. Calcium signaling mediated by the 9 acetylcholine receptor in a cochlear cell line from the ImmortoMouse. J. Physiol., 527: 49 –54, 2000. 20. Whitehead, R. H., Van Eeden, P. E., Noble, M. D., Ataliotis, P., and Jat, P. S. Establishment of conditionally immortalized epithelial cell lines from both colon and small intestine of adult H-2Kb-tsA58 transgenic mice. Proc. Natl. Acad. Sci. USA, 90: 587–591, 1993. 21. Beynon, H. L., Haskard, D. O., Davies, K. A., Haroutunain, R., and Walport, M. J. Combinations of low concentrations of cytokines and acute agonists synergize in increasing the permeability of endothelial monolayers. Clin. Exp. Immunol., 91: 314 –319, 1993. 22. Groves, R. W., Ross, E. L., Barker, J. N., and MacDonald, D. M. Vascular cell adhesion molecule-1: expression in normal and diseased skin and regulation in vivo by interferon gamma. J. Am. Acad. Dermatol., 29: 67–72, 1993. 23. Denekamp, J., and Hobson, B. Endothelial cell proliferation in experimental tumors. Br. J. Cancer, 46: 711–720, 1982. 24. Hobson, B., and Denekamp, J. Endothelial proliferation in tumors and normal tissues: continuous labeling studies. Br. J. Cancer, 49: 405– 413, 1984. 25. Takacs-Jarrett, M., Sweeney, W. E., Avner, E. D., and Cotton, C. U. Generation and phenotype of cell lines derived from CF and non-CF mice that carry the H-2Kb-tsA58 transgene. Am. J. Physiol. Cell Physiol., 280: C228 –C236, 2001. 26. Voyta, J. C., Via, D. P., Butterfield, E., and Zetter, B. R. Identification and isolation of endothelial cells based on their increased uptake of acetylated-low density lipoprotein. J. Cell Biol., 99: 2034 –2040, 1984. 27. Gaffney, J., West, D., Arnold, A., Sattar, A., and Kumar, S. Differences in the uptake of modified low density lipoproteins by tissue cultured endothelial cells. J. Cell Sci., 79: 317–324, 1985. 28. Ghazanfari, F. A., and Stewart, R. R. Characteristics of endothelial cells derived from the blood-brain barrier and of astrocytes in culture. Brain Res., 890: 49 – 65, 2001.

2976

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.

Tissue-specific Microvascular Endothelial Cell Lines from H-2Kb-tsA58 Mice for Studies of Angiogenesis and Metastasis Robert R. Langley, Karen M. Ramirez, Rachel Z. Tsan, et al. Cancer Res 2003;63:2971-2976.

Updated version

Cited articles Citing articles

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/63/11/2971

This article cites 28 articles, 13 of which you can access for free at: http://cancerres.aacrjournals.org/content/63/11/2971.full.html#ref-list-1 This article has been cited by 46 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/63/11/2971.full.html#related-urls

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].

Downloaded from cancerres.aacrjournals.org on August 25, 2015. © 2003 American Association for Cancer Research.