Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The endothelin family comprises three distinct peptides encoded by separate genes located on different chromosomes (1). Endothelin-1 (ET-1) is the most extensively investigated family member. The ET-1 gene encodes a precursor peptide, preproendothelin-1, which is cleaved by a neutral endopeptidase to form proendothelin-1 or big ET-1. This is further cleaved by endothelin converting enzyme-1 (ECE-1) to ET-1 (2). This proteolytic cleavage of Trp-Val bond by ECE-1 is critical to the function of ET-1 because the precursors have negligible biologic activity. Three isoforms of ECE-1 exist: ECE-1a, ECE-1b (3), and ECE-1c (4). They differ only in their N-terminal regions and are derived from a single gene through the use of alternative promoters and differential splicing. The main difference between the isoforms is their subcellular distribution. ECE-1b is predominantly intracellular, whereas ECE-1a and ECE-1c are mainly extracellular (4). Intracellular ECE-1b converts endogenously produced ET-1 and may therefore be important for an autocrine growth loop to function. The extracellular forms, ECE-1a and ECE-1c, act as ectoenzymes expressed on the cell surface, capable of cleaving big ET-1 supplied from outside the cell. The abundance of each isoform varies between cell types investigated and, because the three isoforms have similar enzymatic properties and cleave big ET-1 with comparable efficiencies, their differential subcellular localization and expression within different cell types may be of regulatory importance in the production and subsequent actions of ET-1.

Since its isolation from porcine aortic endothelial cells (5), ET-1 has been found in epithelial, mesangial, neuronal, and glial cells. Its physiologic actions, especially vasoconstriction, have been well characterized. The mitogenic action of ET-1 has been shown in several cell types, including fibroblasts (6), vascular smooth muscle (7), renal mesangial cells (8), and airway smooth muscle (9).

ET-1 is hydrophilic and unable to cross the plasma membrane; it binds to specific cell-surface receptors, which regulate its effect within the cell. There are at least two receptor subtypes, endothelin A (10) and endothelin B (11) receptors. These belong to the family of G-protein-linked receptors with seven transmembrane-spanning domains. The vasoconstrictor effects of ET-1 are mediated via endothelin A receptors (ETAR) (12), which are typically found on vascular smooth muscle. Stimulation of endothelin B receptors (ETBR), mainly found in vascular endothelium, causes vasodilatation mediated through the production of nitric oxide (13). Both these receptors are present on many cell types other than vascular endothelium and smooth muscle (14). Human lung is particularly rich in both ETAR and ETBR (15), suggesting that ET-1 plays an influential role in the normal physiology of the lung.

Growth promotion by ET-1 involves activation of several signal transduction pathways. The mitogenic effects are mediated through the ETAR by activation of phospholipase C, hydrolysis of phosphatidylinositol (16), and mobilization of the second messengers inositol triphosphate and diacylglycerol (17, 18). These second messengers then cause a biphasic increase in intracellular calcium. There is an initial transient peak derived from intracellular storage sites, followed by a sustained plateau resulting from extracellular calcium influx (19).

It is of interest that ET-1 is produced by several human cancer cell lines, including breast, colon, pancreas (20), cervix and larynx (21), endometrium (22), prostate (23), and ovary (24). Production of ET-1 by these cancer cells suggests that it may be involved in their development and progression. For example, Nelson and coworkers (25) demonstrated that an ETAR-specific antagonist could block the mitogenic actions of ET-1 on prostate cancer cell lines in culture, thus implying that the ETAR mediates the mitogenic effects of ET-1 in prostate cancer. The same investigators proposed that the ETBR subtype was downregulated in prostate cancer because there was little expression of ETBR in the prostate cell lines investigated by reverse transcription/polymerase chain reaction (RT-PCR) and Southern blot analyses.

Lung tumors, particularly small cell lung cancer (SCLC), produce a host of peptides, which act as autocrine growth factors for these cells (26). Of these, the bombesin-like peptides (gastrin-releasing peptide) and vasopressin are the most extensively studied, and their role as an autocrine growth factor for SCLC is well established (27). An autocrine growth loop requires active peptide production together with expression and activation of its receptor by the peptide in the same cell. Involvement of ET-1 in the biology of lung cancer was first suggested when ET-1 messenger RNA (mRNA) was demonstrated in a variety of lung cancers by in situ hybridization (34). More recently, Cohen and colleagues (35, 36) demonstrated expression of ET-1, its receptors, and ECE-1 in several lung cancer cell lines and proposed that ET-1 may act as an autocrine growth factor for these cells.

In this study we investigate the expression of ET-1, ETAR, ETBR, and the ECE-1 isoforms in a panel of lung cancer cell lines and discuss the possible role of ET-1 in the progression of these tumors.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Culture of Cell Lines

All cancer cell lines were maintained in RPMI medium (GIBCO, Paisley, UK) supplemented with 10% bovine calf serum (Sigma, Dorset, UK), at 37°C in a humidified atmosphere and 5% CO₂. SCLC cell lines were cultured as suspension aggregates and passaged by disaggregation every 3 to 5 d. All cell lines were acquired from the American Type Culture Collection (Rockville, MD), unless stated otherwise. SCLC cell lines used were GLC-19, NCI-H69 and NCI-H345, NCI-H711 (generous gift from Dr. B. Johnson, National Cancer Institute, Bethesda, MD), Lu-165 (generous gift from Dr. T. Terasaki, National Cancer Centre, Tokyo, Japan), and COR-L88 and COR-L24, both gifts from Dr. P. Twentyman (established at the MRC Clinical Oncology Research Unit, Cambridge, UK). Non-small cell lung cancer (NSCLC) cell lines were cultured as adherent monolayers and passaged by trypsinization every 4 to 5 d. NSCLC cell lines used were NCI-H460, COR-L23, MOR/P (CORL23 and MOR/P cell lines were a gift from Dr. P. Twentyman as above), and A549. Other cell lines included a promyelocytic leukemic cell line (HL-60), primary normal human bronchial epithelial (NHBE) cells (Clonetics, Hampshire, UK), and the SV40 transformed cell line SV40-HBE (a generous gift from Dr. W. Franklin, University of Colorado Health Sciences Center, Denver, CO), which were cultured as described (37). Human umbilical vein endothelial cells (HUVEC) (a generous gift from Dr. P. Hewitt, CRC Department of Clinical Oncology, Nottingham, UK) were cultured as described elsewhere (38).

RT-PCR, Blotting, and Hybridization

Total RNA was extracted from cell lines using a Purescript RNA isolation kit (Gentra Systems Inc., Minneapolis, MN) according to the manufacturer's protocol. The quantity of RNA was determined by ultraviolet spectrophotometry. Complementary DNA (cDNA) was made using an oligo dT primer (Reverse Transcription System; Promega, Southampton, UK). A total of 3 µl of diluted cDNA template was used for PCR, which was optimized using the Hybaid Touchdown thermal cycler. A total of 35 cycles of PCR was used for reactions unless otherwise stated. Primers were designed to amplify across exon/intron boundaries to distinguish any contaminating genomic DNA amplification. The position of the primers used, in relation to gene structure, is shown in Figure 1. Forward primers for ECE-1 were designed to amplify regions exclusive to each isoform. The same reverse primer was used for all ECE-1 PCR. The 5' region of exon 3 is only specific for ECE-1a; exons 1b and 2 are specific for ECE-1b mRNA, and the ECE-1c isoform shares the ECE-1b-specific region in exon 2 but differs in sequence upstream from this, within exon 1 (Figure 1). RT-PCR for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (39) was used to semiquantify the technique (28 cycles) and test integrity of cDNA. The same cDNA template was used in all PCRs; representative amplification of GAPDH is shown in Figure 2B. PCR products were analyzed on 2% agarose gel electrophoresis. The identity of RT-PCR products was confirmed by automated sequencing (DNA sequencing facility, Queens Medical Centre, Nottingham, UK) using an ABI Prism 373 sequencer (PE Biosystems, Warrington, UK).

View larger version (19K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Figure 1.   Primer maps showing cDNA and relative positions of primers used for RT-PCR. References refer to cDNA sequence source. Exons are not to scale and shown at the top, with amplified PCR products below. (A) Location of ET-1for and ET-1rev primers on preproendothelin-1 cDNA (54). (B) Location of ETARfor and ETARrev primers on endothelin receptor type A cDNA (15). (C) Location of ETBRfor and ETBRrev primers on endothelin receptor type B cDNA (15). (D) Locations of ECE-1 primers on the three ECE-1 isoform cDNAs, exons 1-4 of 20 shown (3, 4). The 5' region of exon 3 is specific to ECE-1a. Exon 2 is common to both ECE-1b and ECE-1c; ECE-1b- and ECE-1c-specific sequences are denoted exons 1b and 1c, respectively.

View larger version (61K): [in this window] [in a new window]   Figure 2.   ET-1 mRNA expression detected by RT-PCR in a panel of cell lines. Ethidium bromide-stained agarose gels showing RT-PCR products for (A) ET-1 and (B) GAPDH. Counting from left to right, lanes representing SCLC cell lines: lane 1, GLC-19; lane 2, NCI-H69; lane 3, NCI-H345; lane 4, NCI-H711; lane 5, Lu-165; lane 6, COR-L88; lane 7, COR-L24; NSCLC cell lines: lane 8, NCI-H460; lane 9, COR-L23; lane 10, MOR/P; lane 11, A549; control cell lines: lane 12, HL60; lane 13, NHBE; lane 14, SV40-HBE; lane 15, HUVEC; lane 16, no cDNA.

  Blotting and hybridization of the PCR products from the ECE-1b reaction was performed by transferring the DNA, electrophoresed on a 2% agarose gel, to a nylon membrane by capillary blotting. The membrane was hybridized at 65°C, overnight with an [-³²P]deoxycytidine triphosphate-labeled DNA probe, sequence-specific for ECE-1b. This was then washed with increasing stringency washes and subjected to an autoradiograph.

Immunocytochemistry

For SCLC cell lines, cytospins were produced by centrifugation at a concentration of 5 × 10⁴ cells/ml. For the adherent NSCLC cell lines, cells were seeded at 1 × 10⁵/ml in four-well chamber slides and incubated overnight at 37°C. Cells were fixed in 1% paraformaldehyde and, where required, permeabilized in 70% ethanol for 10 min. Cells were incubated with primary antibody at room temperature for 1 h. Primary antibodies used were B61 (detecting all isoforms of ECE-1) at a 1:50 dilution (40), anti-ECE-1a (detecting only the ECE-1a isoform), and anti-ECE-1bc (detecting both ECE-1b and ECE-1c isoforms), both used at a 1:100 dilution. All ECE-1 antibodies were a generous gift from Dr. T. Subkowski (BASF, Ludwigshafen, Germany). An antibody to actin (Sigma) was used as a control for intracellular staining (1:100). Secondary antibodies (Sigma) were used at 1:50 dilution: fluorescein isothiocyanate (FITC)-conjugated antimouse to detect B61 antibody and FITC-conjugated antirabbit to detect anti-ECE-1a, anti-ECE-1bc, and antiactin primary antibodies. These antibodies were incubated at room temperature for 1 h. 4,6-Diamidino-2-phenylindole (DAPI) (Sigma) nuclear stain was mixed with mountant at 0.1 µg/ ml, and slides were viewed under fluorescence with green filter (DM400) for DAPI and blue filter (DM510) for FITC.

Enzyme-Linked Immunosorbent Assay (ELISA) to Detect Active ET-1

The Biotrak ET-1 ELISA system (Amersham Pharmacia Biotech, Buckinghamshire, UK) was used to detect ET-1 production by cultured cells. ET-1 in the samples tested was captured by microtiter plates precoated with ET-1 antibody and detected by a peroxidase-labeled Fab' fragment of ET-1 antibody conjugate. HUVECs, A549, and NCI-H69 were cultured in the same medium for 48 h and the supernatants from these cells were processed according to the manufacturer's instructions.

Calcium Mobilization Assay

Cells were harvested at a concentration of 2 × 10⁷ /ml and resuspended in serum-free RPMI. The method was adapted from Woll and Rozengurt (41) as described by Coulson and coworkers (33). The cells were loaded with Fluo-3 acetoxymethyl ester (Alexis, Nottingham, UK) (42, 43), a fluorescent Ca²⁺ chelator, and incubated at 37°C in the dark for 30 min. Cells were then resuspended in a solution containing 1 mg/ml dextrose, 3.6 mM CaCl₂, and 0.9 mM MgCl₂ in phosphate-buffered saline-A (phosphate-buffered saline with 1 mg/ml bovine serum albumin). Flow cytometric assays were performed using a fluorescence-activated cell sorter (FACS) scan (Becton Dickenson, Plymouth, UK). Cells were stimulated with ET-1 at concentrations between 10¹⁰ to 10² M, and fluorescence (FL1 channel) measured over time. An increase in fluorescence indicated intracellular calcium mobilization and was calculated relative to the fluorescence of unstimulated cells. Serum was used as a nonspecific stimulus for calcium mobilization to validate the assay for each sample.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In this study we investigated the expression of ET-1, ECE-1, and ET-1 receptors (ETAR and ETBR) by RT-PCR in lung cancer cell lines and bronchial epithelial cells. The presence and cellular location of ECE-1 and its isoforms were investigated by immunocytochemistry, and the activation of functional receptor was evaluated by flow cytometry. HUVECs were used as a positive control because ET-1 (1) and ECE-1 (3) were both identified and characterized from these cells. Figure 2A shows ET-1 mRNA expression by RT-PCR in five of seven SCLC and four of four NSCLC cell lines, and in the human bronchial epithelial cells NHBE and SV40-HBE. RT-PCR with GAPDH primers amplified comparable products from the cDNAs used, thus semiquantifying the PCR for other templates (Figure 2B). The expression of ET-1 mRNA (Figure 2A) was high in the NHBE and SV40-HBE cells (lanes 13 and 14) and comparable to that in HUVECs (lane 15). The NSCLC adenocarcinoma line A549 (lane 11) and the SCLC cell line COR-L88 (lane 6) also showed high expression comparable to the HUVECs, whereas lower levels of expression were seen in the other NSCLC cell lines. Multiple PCR products were seen in GLC-19 SCLC cell line (lane 1) and the HL-60 leukemic cell line (lane 12). The latter cell type represents a nonepithelial malignancy and was used as a negative control. In the absence or low abundance of the transcript, nonspecific amplification can occur, producing multiple products. NCI-H69 cell line (lane 2) was negative for ET-1.

RT-PCR designed to detect all three ECE-1 isoforms (Figure 3A) and that designed to detect the ECE-1b and ECE-1c isoforms (Figure 3B) were positive throughout the cell lines tested. The ECE-1 isoforms individually, however, had more heterogenous expression. The extracellular ECE-1a isoform (Figure 3C) was seen by RT-PCR in the HUVEC (lane 15), which was used as a positive control, the NSCLC cell line NCI-H460 (lane 8), and the NHBE and SV40-HBE cells (lanes 13 and 14), but the bands were less intense in these lines as compared with that in HUVECs. Similarly, the intracellular ECE-1b isoform was only clearly present by RT-PCR in the HUVEC (Figure 3D, lane 15). The products of this PCR were therefore Southern blotted and hybridized using a probe for ECE-1b (confirmed by direct sequencing). This increases the sensitivity of detection for ECE-1b as compared with the other isoforms, showing it to be present in low levels. Thus, differential expression of ECE-1b between the cell lines is likely to be more important. Figure 3E shows that the only cell line negative for the ECE-1b isoform was the NCI-H69 cell line (lane 2). The other extracellular isoform, ECE-1c, was clearly present throughout the cell lines (Figure 3F), showing that ECE-1c is the most abundant isoform. All PCR products were purified and sequenced to confirm specific amplification of transcript.

View larger version (54K): [in this window] [in a new window]   Figure 3.   ECE-1 isoform mRNA expression detected by RT-PCR in a panel of cell lines. Ethidium bromide-stained agarose gels showing (A) ECE-1 mRNA using ECE-1abc/ECE-1rev primers, detecting all isoforms of ECE-1. (B) ECE-1bc mRNA using ECE-1bc/ECE-1rev primers, detecting the ECE-1b and c isoforms only. (C ) ECE-1a mRNA using ECE-1a/ECE-1rev primers, detecting the ECE-1a isoform only. (D) ECE-1b mRNA using ECE-1b/ECE-1rev primers, detecting the ECE-1b isoform only. (E ) Blotting and hybridization of PCR products with probe for ECE-1b. (F ) ECE-1c mRNA using ECE-1c/ECE-1rev primers, detecting the ECE-1c isoform only. Counting from left to right, lanes representing SCLC cell lines: lane 1, GLC-19; lane 2, NCI-H69; lane 3, NCI-H345; lane 4, NCI-H711; lane 5, Lu-165; lane 6, CORL88; lane 7, CORL24. NSCLC cell lines: lane 8, NCI-H460; lane 9, CORL23; lane 10, MOR/P; lane 11, A549. Control cell lines: lane 12, HL60; lane 13, NHBE; lane 14, SV40-HBE; lane 15, HUVEC; lane 16, no cDNA.

Immunocytochemistry using the pan-ECE-1 isoform antibody B61 confirmed the RT-PCR results and was positive throughout the cell lines tested (Table 1). Positive immunofluorescence was denoted if the intensity of staining was stronger than the negative controls (incubated with only the secondary antibody) and compared with the intensity of staining in the HUVEC, the positive control. The pattern of staining is shown in Figure 4. NCI-H345, an example of an SCLC cell line positive for the pan-ECE-1 isoform antibody, is shown in Figure 4A. These cells are characteristically small in size and have relatively large nuclei with prominent nucleoli and scant cytoplasm. Staining with the FITC-conjugated antibody to detect the pan-ECE-1 isoform antibody shows a bright ring staining the cell membrane, which can be seen separate from the nuclei (Figure 4A, top panel ). Nuclear staining with DAPI for the same field of view is shown to delineate position of cells and to distinguish location of immunofluorescence (Figure 4A, bottom panel ). For comparison, the NCI-H345 SCLC cell line (Figure 4A, top and bottom), the NCI-H460 NSCLC cell line (Figure 4B, top and bottom), and the SV40-HBE cells (Figure 4C, top and bottom) are shown with positive immunofluorescence for the ECE-1 pan-isoform antibody and with DAPI nuclear staining. In contrast, the ECE-1a antibody resulted in positive immunofluorescence only in the NCI-H460 cell line (Figure 4D, top and bottom) and in the SV40-HBE (Figure 4E, top and bottom). The HUVECs again were used as positive control (Table 1). The other lung cancer cell lines tested were negative for the ECE-1a isoform. An example of this is shown in the NCI-H345 cell line (Figure 4F, top and bottom).

View larger version (67K): [in this window] [in a new window]   Figure 4.   Expression and subcellular distribution of ECE-1 isoforms by immunocytochemistry. (A-C) Pan-isoform staining (top panels) and corresponding nuclear staining with DAPI (lower panels). (A) NCI-H345 SCLC cell line, (B) NCI-H460 NSCLC cell line, (C ) SV40-HBE bronchial epithelial cell line. (D-F ) ECE-1a isoform staining (top panels) and corresponding nuclear staining with DAPI (lower panels). (D) NCI-H460 NSCLC cell line, showing pattern of extracellular staining. (E ) SV40-HBE bronchial epithelial cell line. (F ) Negative staining in NCI-H345 SCLC cell line. (G) ECE-1bc staining in permeabilized HUVECs. Intense staining in the permeabilized cells suggests a contribution by the intracellular ECE-1b isoform because only the permeabilized cells would also be able to stain for both the intra- and extracellular isoforms. (H ) Nonpermeabilized NCI-H460. (I ) Permeabilized NCI-H460. No difference is seen in the differential cell preparation, suggesting the main isoform is the ectoenzyme ICE-1c.

Positive immunofluorescence with the ECE-1bc antibody was seen in all of the lung cancer cell lines tested, including the SV40-HBE cells (Table 1). Positive staining is shown in the HUVECs at a higher magnification so that the cellular staining pattern can be appreciated (Figure 4G). An example of positive ECE-1bc staining in the tumor lines is shown in the NCI-H460 cell line with differential cell preparations (Figures 4H and 4I). During immunostaining, the cells were processed with or without a permeabilization step before incubation with the primary antibodies (as described in MATERIALS AND METHODS). This was used to indicate that any fluorescent labeling without permeabilization represents extracellular location of antigen because unpermeabilized cells would not allow antibodies to cross the cell membrane. Permeabilized cells would exhibit staining of both intracellular and extracellular antigen. An antibody against intracellular actin was used to validate the method; only permeabilized cells were positive for this antigen (data not shown).

The pattern of fluorescence, seen as a bright ring at the periphery of the cells or uniformly across the cell surface, is also an indication of extracellular antigen localization. The pattern of staining with the ECE-1bc isoform antibody, in the NCI-H460 cells, was the same for both the nonpermeabilized (Figure 4H) and the permeabilized cells (Figure 4I). This would indicate that the ectoenzyme ECE-1c was the predominant isoform present because the intracellular form ECE-1b would otherwise contribute to fluorescence in the permeabilized cells and produce differential staining pattern in the two cell preparations. This concurs with RT-PCR data because it required blotting and hybridization of the PCR products to detect ECE-1b in the lung cancer cell lines, reflecting its lower level of expression. The HUVECs exhibited brighter fluorescence in the permeabilized cells with the ECE-1bc antibody, indicating higher levels of intracellular ECE-1b isoform contributing to increased fluorescence in the endothelial cells (Figure 4G, top and bottom). The positive ECE-1bc antibody staining seen throughout the cell lines is again likely to be due to high levels of the ECE-1c isoform present. This corresponds to expression of ECE-1c throughout the cell lines by RT-PCR and comparatively reduced expression of ECE-1a and ECE-1b isoforms in the lung cancer cell lines.

Having demonstrated that lung cancer cells have the machinery to activate proendothelin, it was important to show that these cells actually produced active ET-1. An ELISA system with negligible cross-reactivity with proendothelin was used to detect active ET-1 in the supernatants of cultured cells. HUVECs were used as positive controls. The A549 NSCLC cell line showed high expression of preproendothelin by RT-PCR (Figure 2A), comparable to the HUVECs, and thus was selected for this assay. The NCI-H69 SCLC cell line was negative for preproendothelin and thus was selected. Figure 5 shows the levels of ET-1 detected in the supernatants of these cells. The A549 cells showed high levels of ET-1 production, comparable to the HUVECs as shown by the RT-PCR data. The NCI-H69 cells produced negligible amounts of ET-1, also concurring with the RT-PCR data. This would suggest that cells co-expressing the precursor and activating enzyme transcripts by RT-PCR all have the potential of producing active ET-1.

View larger version (11K): [in this window] [in a new window]   Figure 5.   Levels of ET-1 detected by ELISA in the supernatants of lung cancer cells compared with HUVECs. The high levels of ET-1 production in the A549 cells correlate with coexpression of high levels of preproendothelin and ECE-1b by RT-PCR. The NCI-H69 cells show negligible ET-1 production correlating with being negative for both preproendothelin and ECE-1b by RT-PCR, indicating that expression of the precursor and the activating enzyme results in active ET-1 production by these cells.

Expression of the ET-1 receptors (ETAR and ETBR) was investigated by RT-PCR. This showed that ETAR mRNA was most abundant in the bronchial epithelial cells, with little detected in the lung cancer cell lines (Figure 6A). Only five of 11 of the lung cancer cell lines were positive for ETAR, and the PCR product was less intense in each instance compared with that present in the NHBE and SV40-HBE cells. In contrast, the ETBR was expressed throughout the cell panel (Figure 6B), with the most product amplified in the GLC-19 (lane 1) and COR-L24 (lane 7), both SCLC cell lines. It is of note that two PCR products are seen close together for the COR-L23, NSCLC cell line (lane 9). This may represent a variant or aberrant form of the ETBR. The same cDNAs were used as in the previous experiments (Figure 2), and RT-PCR for GAPDH was as represented in Figure 2B.

View larger version (69K): [in this window] [in a new window]   Figure 6.   Expression of the endothelin receptor subtypes mRNA detected by RT-PCR. Ethidium bromide-stained agarose gels showing (A) ETAR and (B) ETBR. Lanes representing SCLC cell lines: (counting from left to right) lane 1, GLC-19; lane 2, NCI-69; lane 3, NCI-H345; lane 4, NCI-H711; lane 5, Lu-165; lane 6, COR-L88; lane 7, COR-L24. NSCLC cell lines: lane 8, NCI-H460; lane 9, COR-L23; lane 10, MOR/P; lane 11, A549. Control cell lines: lane 12, HL60; lane 13, NHBE; lane 14, SV40-HBE; lane 15, HUVEC; lane 16, no cDNA.

Flow cytometric assay of calcium mobilization was used to investigate the activation of functional receptors after ET-1 binding. No response to stimulation by ET-1 at concentrations between 10¹⁰ to 10² M (experiments performed in triplicate) was seen in any of the lung cancer cell lines (three SCLC and three NSCLC) or the human bronchial cells representing normal lung. Serum, acting as a nonspecific stimulus and an internal control for our assay, produced a peak in fluorescence after stimulation, indicating calcium mobilization and confirming functionality of our assay (data not shown). These experiments thus support the finding that the ETAR is downregulated in the lung cancer cell lines, as shown by RT-PCR.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

SCLC is characterized by the production of numerous peptides that can act as autocrine growth factors for these tumors. Vasopressin was the first neuropeptide found to be mitogenic (44); since then many neuropeptides have been shown to be growth factors for a wide range of cell types (45). ET-1 is one such peptide known to act as a mitogen in several tissues. The finding of ET-1 and its receptors in several tumor cell lines has led investigators to propose that it may act as a growth factor for these tumors (24, 25). In the present study, we have shown by RT-PCR that ET-1 was expressed in over 80% of the lung cancer cell lines tested, including 100% of NSCLC cell lines, as well as bronchial epithelial cells. This suggests that it plays a role in the maintenance of both the tumor and normal epithelial cells in the lung.

However, because active ET-1 is synthesized by cleavage from a precursor peptide, big ET-1, it was important to investigate the expression of ET-1 together with its activating enzyme, ECE-1. The primary structure of ECE-1 shows that it is a member of the zinc metalloprotease family (46, 47) and is a key site of regulation of ET-1 activity. To understand the importance of this enzyme and its isoforms in lung cancer, it was important to determine the differential abundance and location of the ECE-1 isoforms in lung tumor cells because this may be of regulatory importance. To our knowledge, this is the first study investigating the distribution of ECE-1 isoforms in lung cancer.

The intracellular localization of the ECE-1b isoform makes its presence important for a proposed autocrine growth loop to function, so that big ET-1, produced by the cell, can be processed to its active form before its release. Expression of the intracellular ECE-1b isoform was downregulated compared with the extracellular ECE-1c and required blotting and hybridization of the PCR products to detect its expression accurately. This was interesting in that levels of expression of ECE-1b correlated with differential expression of preproendothelin in the lung cancer cell lines, by RT-PCR. This was most notable in the A549 NSCLC cell line, which exhibited high levels of preproendothelin by RT-PCR and ET-1 by ELISA, together with high levels of expression of ECE-1b. Interestingly, the NCI-H69 SCLC cell line was negative for both ET-1 (by RT-PCR and ELISA) and ECE-1b. This would indicate that coexpression of preproendothelin with the activating enzyme suggests that lung cancer cells produce active ET-1.

However, the ectoenzyme ECE-1c was the most commonly expressed isoform in lung cancer and normal bronchial epithelial cells, being present in all the cell lines investigated. This is in agreement with the findings of other investigators for a range of tissues types (4). Because this is the most abundant isoform, its involvement in lung cancer is likely to be the most important. Therefore, perhaps proendothelin produced by tumor cells could be activated by ECE-1c expressed on the surface of that cell, causing autocrine growth stimulation. Alternatively, ET-1 produced by tumor cells could be activated by the abundant ECE-1c, and active ET-1 could then act on nearby cells, for example, endothelial cells, to promote angiogenesis. The effects of ET-1 on endothelial cells are well established. Its motogenic and angiogenic activities are mediated via ETBR in the presence of functional endothelial nitric oxide synthase (48, 49). These potential models are illustrated diagrammatically in Figure 7.

View larger version (24K): [in this window] [in a new window]   Figure 7.   Diagrammatic representation of ET-1 autocrine and paracrine models in lung cancer. (A) Autocrine growth loop for lung cancer similar to that proposed in ovarian cancer involving ETAR (24), but lung cancer cells have downregulated ETAR and the intracellular converting enzyme ECE-1b. (B) Proposed paracrine loop, with ET-1 from lung cancer cells affecting endothelial motogenesis and angiogenesis, involving ETBR and ECE-1c. Possible autocrine loop via ETBR is shown, but signaling pathways have not been clearly characterized.

The ETAR, thought to be the receptor involved in promoting growth in other tumor types (25, 50), was expressed in 45% of the lung cancer cell lines investigated. This corresponds to the findings of Cohen and colleagues (35), who found 57% of lung cancer cell lines expressing ETAR. Flow cytometry was used to demonstrate intracellular calcium mobilization after receptor activation by ET-1. We have previously validated this technique to show intracellular Ca²⁺ mobilization in SCLC cell lines in response to stimulation by the peptide vasopressin, shown to act as an autocrine growth factor for SCLC (33). In a similar manner, the ETAR is thought to use Ca²⁺ as an intracellular secondary messenger in mediating its mitogenic actions in a variety of cell lines. ET-1 binding to ETAR is reported to cause a biphasic response, an initial rapid transient phase followed by a sustained plateau phase (51). In our study, this assay for ET-1 receptor activation did not demonstrate a functional transduction pathway involving Ca²⁺ mobilization, following stimulation by ET-1. However, alternative signaling pathways may exist in lung cancer cells because coupling of receptors to differing signal pathways has been noted between cell types (52). In particular, the transduction pathway mediating the actions of the ETBR subtype, which was found to be expressed throughout the lung cancer cell lines, is less well characterized than pathways mediating effects of ETAR. Receptor-signal transduction coupling may be tissue specific and influence subsequent actions following receptor stimulation in different cells (52).

ETAR was downregulated in lung cancer relative to normal bronchial epithelial cells by RT-PCR. ETBR was present throughout the panel of cell lines and may be the more important receptor subtype in lung cancer. As has been found in rat mesangial cells (53), stimulation of the ETBR may induce further production of ET-1 by lung cancer cells. Because it has a short half-life, autoinduction of further ET-1 release from the same or surrounding cells via the ETBR may be responsible for the profound and sustained effects of ET-1. Mitogenic actions mediated by the ETBR via an alternative pathway in lung tumors requires further study. Thus, our findings, showing a lack of active ETAR, an abundance of ETBR, and the relative excess of extracellular ECE-1c, would suggest that instead of an autocrine growth loop, paracrine effects of ET-1 are likely to be more important in lung cancer.

In summary, we have demonstrated by RT-PCR that ET-1, ETBR, and the extracellular ECE-1c isoform mRNA expression are high in the lung cancer cell lines investigated. Coexpression of preproendothelin and its activating enzyme results in active ET-1 production in these cells. In contrast, the ETAR was downregulated, and no stimulation of ETAR by ET-1, resulting in Ca²⁺ mobilization, could be detected. Autocrine growth mediated through the ETAR has been shown in ovarian tumors (50). ET-1 has also been shown to influence growth of prostate cancer cells via the ETAR (25). In both tumor types, ETBR was shown to be downregulated. In addition, ET-1 was proposed as an autocrine growth factor for lung cancer by Cohen and coworkers (35, 36) when they found expression of ET-1 and both its receptors in a number of lung tumors and cell lines derived from lung cancers and bronchial epithelium. Our more detailed study into the ECE-1 isoforms and receptor subtypes would indicate that the ETAR is not involved in an autocrine growth loop in lung cancer as in the other tumor types. Instead, we propose that ET-1 may be involved in complex paracrine circuits involving stimulation of epithelial and endothelial cells surrounding the tumor, influencing its growth perhaps through promoting angiogenesis, and these effects of ET-1 on the vasculature have been well established (48, 49). However, an autocrine loop involving the ETBR may be a line of further investigation in view of its abundance in lung cancer cells and normal lung, and because its signal transduction pathways are less well identified.