| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Human Anatomy and Genetics (L.P.C., M.J.E., J.F.M., H.C.C.), University of Oxford, Oxford OX1 3QX, United Kingdom; and Department of Neuroendocrinology (J.C.B.), Division of Neuroscience and Psychological Medicine, Faculty of Medicine, Imperial College of Science Technology and Medicine, Hammersmith Hospital Campus, London W12 0NN, United Kingdom
Address all correspondence and requests for reprints to: Dr. Helen C. Christian, Department of Human Anatomy and Genetics, University of Oxford, South Parks Road, Oxford OX1 3QX, United Kingdom. E-mail: helen.christian{at}anat.ox.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Secretion of ANXA1 was first inferred from its identification in rat peritoneal exudates (12) and subsequently its detection extracellularly in lung lavage fluid, plasma, seminal fluid, and cell culture media (13, 14, 15, 16). Glucocorticoids act as a specific stimulus for ANXA1 externalization in several different cell types (9, 13, 17, 18). However, ANXA1 lacks a cleavable signal sequence at its amino terminus (19), such that it is unlikely to be secreted by the classical secretory pathway that involves targeting of proteins to the plasma membrane via the endoplasmic reticulum and Golgi apparatus. In human blood, mononuclear cells (20) and anterior pituitary cells (21) glucocorticoid-induced externalization of ANXA1 is not influenced by inhibitors of the classical regulated secretory pathway. Furthermore, the ANXA1-rich FS cells in the anterior pituitary contain no classical regulated secretory granules (10). Due to the difficulty in isolating FS cells in sufficient numbers and purity for functional in vitro studies, the mechanism of export of ANXA1 by primary FS cells has not previously been investigated. However, we have recently demonstrated that TtT/GF cells, a well-characterized stable mouse pituitary-derived FS cell line (22, 23), express and externalize ANXA1 in response to dexamethasone specifically at the tips of the cell processes (10, 24). These data clearly suggest that some form of externalization mechanism is localized in these specific areas of the plasma membrane.
There is increasing evidence that some proteins that lack a signal sequence, for example the cytokine IL-1ß, are transported across membranes by ATP-binding cassette (ABC) transporters (25). These ATP-driven pumps form channel-like structures in the membrane of prokaryotic and eukaryotic cells. Members of the ABC transporter family constitute a superfamily of highly conserved proteins involved in the membrane transport of a variety of substrates including ions, amino acids, peptides, sugars, vitamins, and steroid hormones (26, 27). There are over 100 reported members of the group and the best characterized examples include the cystic fibrosis transmembrane conductance regulator (CFTR), the sulfonylurea receptors (SURs) 1 and 2, multidrug resistance (MDR) proteins, and the Tangier disease protein ABCA1 (27). The full-size ABC proteins are defined by the presence of the ABC unit that contains two nucleotide-binding folds with conserved Walker A and B motifs and two transmembrane domains each consisting of six membrane-spanning helices (27). In addition to their structural similarity, SURs, CFTR, MDR, and ABCA1 are also similar in that their activity is inhibited by the sulfonylurea, glyburide (glibenclamide; Refs. 28 and 29). Glyburide inhibits ABC transporter-mediated translocation of IL-1ß in human monocytes and mouse macrophages (30), basic fibroblast growth factor in an osteogenic cell line (31) and IL-12 in macrophages (32). Although ABC transporters are ubiquitous, there are at present little data regarding their expression in anterior pituitary cells. ABCA1 mRNA has been shown to be present in the human pituitary (33), and SUR mRNA is expressed in anterior pituitary cell lines (34) and human pituitary adenomas (35). However, which specific cell types in the gland express these transporters is unclear.
In the present study, we have used the TtT/GF cell line and rat anterior pituitary tissue in vitro to examine the role of ABC transporters in the mechanism by which glucocorticoids induce the translocation of ANXA1 to the cell membrane. We have also investigated the expression of ABC transporters in A549 cells, a human lung adenocarcinoma cell line that has been used extensively to analyze the possible roles and mechanism of action of ANXA1 (18, 36, 37, 38). A549 cells express ANXA1 in large amounts, externalize ANXA1 in response to dexamethasone, and respond to exogenous ANXA1, which inhibits their growth and blocks epidermal growth factor-stimulated proliferation (18, 38). A549 cells are therefore a useful model in which to investigate possible transporters of ANXA1 in parallel with TtT/GF cells. In the present study, we have investigated the hypothesis that a member of the ABC transporter family is responsible for ANXA1 externalization from cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals
Adult normal male Sprague Dawley rats (200–250 g) and C57BL6 mice (8 wk old) bred from colonies in the Department of Human Anatomy and Genetics, Oxford, were used. The Principles of Laboratory Animal Care (NIH Publication No. 85-23) were followed and the study was carried out under license in accordance with the UK Animals (Scientific Procedures) Act 1986. The rats and mice were housed after weaning in groups of five per cage in a quiet room with 14-h light, 10-h darkness and temperature maintained at 20–21 C; food and water were available ad libitum. All experiments were started between 0800 h and 0900 h to avoid changes associated with the circadian rhythm. Rats were killed by stunning followed by decapitation, the pituitary gland removed, and the anterior lobe separated from the posterior lobe. Mice were terminally anesthetized by ip injection of 3 mg sodium pentobarbital (Sagatal, Rhone Merieux, France) and perfused through the heart with heparinized saline (0.9% NaCl and 10 U/ml heparin) until blood was cleared; the pituitary gland was then removed.
Isolation of enriched FS cells from mouse anterior pituitary
Mice (n = 15) were anesthetized and perfused with Earle’s balanced salts solution (EBSS; 37 C, pH 7.4) containing 12.5 U/ml of heparin until all blood was cleared. This ensures that blood cells do not contaminate the final population of FS cells. Anterior pituitaries were removed, cut into approximately 0.5-mm2 pieces, and incubated for 60–90 min in EBSS (37 C, pH 7.4; 95% O2, 5% CO2) containing 4% collagenase and 0.02% deoxyribonuclease. The suspension was agitated regularly using a pipette. Once all large pieces of tissue had disappeared, the suspension was centrifuged (100 x g, 10 min) and the resultant pellet resuspended in 1 ml EBSS. The cell suspension was then gently layered on top of a solution of 4% BSA in EBSS (37 C, pH 7.4) and centrifuged (100 x g, 10 min). Granulated endocrine cells sink to the bottom to form a pellet while FS cells remain in the supernatant. The supernatant was collected and the pellet resuspended, layered onto 4% BSA in EBSS, centrifuged two further times, and in each case the supernatant collected. The supernatant from each gradient was then spun (200 x g, 10 min) to obtain a pellet of enriched FS cells.
Electron microscopy.
To confirm enrichment of the FS cells, a sample of cells was retained and prepared for electron microscopy. Isolated cells were prepared for electron microscopy by standard methods. Briefly, cells were postfixed in osmium tetroxide (1% wt/vol in 0.1 M sodium phosphate buffer) contrasted with uranyl acetate (2% wt/vol in distilled water), dehydrated through increasing concentrations of ethanol (70–100%) and embedded in Spurr’s resin [Agar Scientific (UK), Stansted, UK]. Ultra-thin sections (50–80 nm) were prepared by use of a Reichert Ultracut S microtome (Leica Corp., Milton Keynes, UK), mounted on 200-mesh nickel grids, incubated for 2 h with anti-S100 (DAKO Corp., Carpinteria, CA, dilution 1:1000) for labeling of FS cells and for 1 h with antirabbit IgG 15 nm gold complex (British Biocell, Cardiff, UK), then lightly counterstained with uranyl acetate and lead citrate. All antisera were diluted in 0.1 M phosphate buffer containing 0.1% egg albumin. Sections (50–80 nm) were viewed with a JEM-1010 transmission microscope (JEOL USA, Inc., Peabody, MA). FS cells were identified and counted on a basis of ir-S100 immunogold labeling and morphological appearance, and the percentage FS cells quantified.
A549 and GH3 cell culture
A549 cells were maintained in DMEM/F12 medium containing 10% fetal calf serum, 100 IU/ml penicillin, and 100 µg/ml streptomycin in T-150 flasks in a humidified atmosphere of 5% CO2 in air at 37 C. GH3 cells, a clonal anterior pituitary cell line composed of heterogeneous cells that secrete GH and prolactin, were grown in Ham’s F10 medium supplemented with 15% horse serum, 2.5% fetal bovine serum, and 1% penicillin/streptomycin in a humidified atmosphere of 5% CO2 in air at 37 C. Before harvesting, the cells were washed with PBS and then scraped off with a rubber policeman for either mRNA isolation (A549 cells) or into homogenization buffer for Western blotting of ABCA1 (A549 and GH3 cells; see Western blot analysis of ABCA1).
TtT/GF cell culture
TtT/GF cells were cultured in DMEM-Ham’s F12 medium containing 10% fetal calf serum, 2.5 mM L-glutamine, 100 IU/ml penicillin, and 100 µg/ml streptomycin. Flasks containing approximately 2–5 million TtT/GF cells were allowed to reach confluence and were incubated at 37 C for 3 h or 24 h in either 12 ml incubation medium [1% (vol/vol) aprotonin (Bayer Corp., Saffron Waldon, UK) and 1% (vol/vol) penicillin/streptomycin in oxygenated EBSS (pH 7.4) phenol red-free] or the ABC transporter inhibitor glyburide (100 µM). In addition, cells were incubated in incubation medium alone or with glyburide (100 µM) with or without 0.1 µM dexamethasone sodium phosphate (3 h). At the completion of the incubation periods surface ANXA1 was collected in an EDTA wash for Western blot analysis (see Detection of ANXAI by Western blotting). Additional experiments were performed whereby ANXA1 was detected by immunofluorescence and quantified by a fluorescence plate reader. In this case, TtT/GF cells were plated at 50 cells/mm2 density in nonfluorescent 96-well plates (Microtest, 96-well assay plate Optilux-Plus, Becton Dickinson and Co., Oxford, UK) with eight wells per treatment group. Cells were treated as above with either incubating medium, 56 mM K+ EBSS, or 0.1 µM dexamethasone (500 µl, 30 min or 3 h) either in the presence or absence of 100 µM glyburide. Cell viability was checked at the end of each experiment by trypan blue exclusion (cell viability was always >95%). The cells were then fixed for 10 min in a mixture of freshly prepared 3% paraformaldehyde and 0.05% glutaraldehyde in 0.1 M PBS (pH 7.4) at 37 C (10) for immunostaining of ANXA1.
Externalization of ANXA1 and secretion of ACTH in vitro by anterior pituitary segments
Rat anterior pituitary glands were cut into four roughly equal segments. The segments were distributed randomly (1 segment/well) in the wells of 24-well tissue culture plates (Costar, Cambridge, MA) and incubated at 37 C for 90 min in 1 ml incubation medium under a humidified atmosphere saturated with 95% O2-5% CO2. The medium was changed after 1 h and 1.5 h. The segments were then incubated for a further 1 h in medium containing forskolin (100 µM); controls were exposed to an equal volume (1 ml) of medium alone. Where appropriate, dexamethasone (0.1 µM) and/or glyburide (100 µM) were included in the medium throughout both the preincubation and final incubation periods. Medium from the final incubation was collected and stored in aliquots (300 µl, -20 C) for subsequent measurement of immunoreactive (ir)-ACTH by RIA. Pituitary segments were weighed on a torsion balance and segments retained for subsequent Western blotting of surface and intracellular ANXA1 (see Detection of ANXAI by Western blotting).
Detection of ANXA1 by Western blotting
ANXA1 in TtT/GF cells and anterior pituitary segments was also detected by SDS-PAGE and Western blot analysis. Proteins bound to the outer surface of the TtT/GF cell membranes and anterior pituitary segments by Ca2+-dependent mechanisms (surface ANXA1) were removed by washing the cells for 1 min in PBS (500 µl) containing EDTA (1 mM) and protease inhibitors (1 mM phenylmethylsulfonyl fluoride and 1 µg/ml aprotinin; Ref. 9). In the case of anterior pituitary segments ANXA1 in the remaining tissue (i.e. intracellular annexin) was then extracted into 500 µl PBS containing EDTA (10 mM), Triton (1% vol/vol, BDH Chemicals Ltd., Poole, UK), phenymethylsulfonyl fluoride (1 mM), and aprotinin (1 µg/ml) followed by sonication on ice. The EDTA washes and the extracts were analyzed immediately after the total protein concentration of the extracts had been determined by use of the Pierce Chemical Co. (Chester, UK) BCA protein assay reagent (Pierce Chemical Co.).
SDS-PAGE and Western blot analysis.
Proteins extracted from TtT/GF cells and enriched mouse FS cells were separated by SDS-PAGE [4 µg/channel (EDTA washes) and 20 µg/channel (cell extracts) in a volume of 20 µl by use of a midget gel electrophoresis system and power pack, LKB, Milton Keynes, UK] and transferred electrophoretically (64 mA, 1 h) to nitrocellulose paper (Bio-Rad Ltd., Hemel Hempstead, UK). To concentrate proteins harvested in the EDTA cell washes each wash (500 µl) was precipitated with 10% trichloroacetic acid for 30 min on ice. The protein pellet was washed with a 1:1 mixture of ethanol and ether and resuspended in gel-loading gel buffer (200 nM Tris-HCl, pH 8.8; 1 M sucrose; 5 mM EDTA; 0.01% bromophenol blue; 5 mM dithiothreitol; and 2% sodium dodecyl sulfate). ANXA1 was detected by overnight incubation (4 C) with anti-ANXA1 (diluted 1:5000) and visualized by sequential exposure to a peroxidase-conjugated donkey antisheep antiserum (diluted 1:5000) and diaminobenzidine (50 ml, 0.05% wt/vol) containing H2O2 (20 µl). The molecular weights of immunoreactive bands were determined by comparison with the migration of molecular mass standards (high molecular weight range rainbow labeled, Amersham International, Buckinghamshire, UK). The optical densities of bands were measured semiquantitatively, using a Fujix Bas 1500 imaging system with a low level light sensitive camera and TINA software (Raytek, Sheffield, UK). Intensity values were normalized relative to control values.
RT-PCR detection of ABCA1 transporter mRNA
mRNA was isolated from either TtT/GF cells, A549 cells, or enriched FS cells prepared from perfused mouse anterior pituitary tissue. Cell suspensions were spun at 300 x g for 10 min to generate a pellet that was homogenized in GITC-containing lysis buffer (buffer RLT supplied with the QIAGEN RNeasy kit with 1% ß-mercaptoethanol) and QIAshredder columns (QIAGEN Ltd., Crawley, UK). mRNA was then isolated using the QIAGEN RNeasy kit (QIAGEN Ltd.). For each sample, 1 µg of mRNA was reverse transcribed using the Expand reverse transcriptase kit (Roche, Lewes, UK) to yield cDNA in a final volume of 20 µl. PCR was carried out in a 50-µl reaction volume using the Expand High Fidelity PCR system (Roche) using 1 µl of the cDNA mix as a template. PCR was optimized at 1.5 mM MgCl2, an annealing temperature of 60 C and x 30 cycles (1 min 95 C; 1 min 60 C; 1 min 72 C).
Primers.
These studies employed both degenerate primers that recognize specific sequences found in all ABC transporters sequenced so far (ATP-binding domain and active transport sequence; Ref. 39) and specific primers designed against the mouse and human ABCA1 mRNA sequence (40). The sense primer (5'-GGCTGCAGTGGCKSWGGVAPA-3' where K = G or T, S = G or C, W = A or T, V = G or A or C, and P = A or G) recognizes a conserved sequence on the cDNA, which codes for the Walker motifs A and B that are involved in nucleotide binding (39). The antisense primer (5'-GTGCATGCCCHCCHSWCAGCTG-3' where H = A or T or C, S = G or C, and W = A or T) recognizes a conserved sequence on the cDNA, which codes for the active transport sequence of ABC transporters. All of the primers used spanned at least one intron. In all transporters, these sequences are separated by a conserved 340-bp sequence. Three combinations of primers were used against the cDNA ABCA1 sequence. Primer pair 1: sense (forward 1 F1): 5'-TGCCAACAACCCCTGCTTCC-3', antisense (reverse 1, R1): 5'-GCTGGGTCGGGAGATGAGATGT-3' (sequence amplified predicted 529 bp). Primer pair 2: sense (F2): 5'-CAAGGTCCTGAGAATGTTACGGCA-3'; antisense R2: 5'-TCTGGGTGGAGGAGCTGGAGC-3' (sequence amplified predicted 520 bp). Primer pair 3: sense (F3): 5'-TGGTGATGCGGAGTTTCTGCC-3'; antisense R3: 5'-TTTGTGCTGAACAGCTCTTGGGC-3' (sequence amplified predicted 209 bp). A positive control was run in parallel using specific established primers against the cDNA sequence coding for G3PDH that amplify a sequence of 960 bp. A negative control PCR was also run using the ABCA1 primers but omitting the cDNA in the reaction tube.
Western blot analysis of ABCA1
Immunoblot analysis of ABCA1 was performed as described previously (41). Enriched mouse FS cells, TtT/GF cells, GH3, and A549 cells were homogenized in 10 mM Tris-HCl, pH 7.3; 1 mM MgCl2; and 0.5% Nonidet P-40 in the presence of protease inhibitors (0.5 µg/ml leupeptin, 1 µg/ml aprotinin, 1 µg/ml pepstatin A). Postnuclear supernatants from cell lysates were prepared by centrifugation at 3000 x g for 10 min at 4 C. Samples containing 40 µg protein were reduced with 2-mercaptoethanol in gel-loading buffer, fractionated by 7.5% SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad Ltd.). Immunoblotting was performed using an antihuman ABCA1 antiserum (Novus, Littleton, CO) raised in rabbit (diluted 1:500) and the immunoreactive bands visualized by incubation with sheep antirabbit IgG peroxidase (diluted 1:5000) and diaminobenzidine as above.
RIA of ACTH
ACTH was determined in duplicate by RIA using a primary antibody of defined specificity raised in rabbits against human ACTH-(1–39) (reagent supplied by the National Hormone and Pituitary Program), synthetic human ACTH-(1–39) as a reference preparation (National Institute for Biological Standards and Control, South Mimms, UK) and 125I ACTH-(1–39) as the tracer. The assay sensitivity was 10 pg/ml, and the interassay and intraassay coefficients of variation were 10.0% and 5.2%, respectively. Dilution curves of test samples were all parallel to that of the standard preparations.
Detection of cell surface ABCA1, ANXA1, and S100B by immunofluorescence
A double immunofluorescence reaction was used to study colocalization of ABCA1, ANXA1, and the FS cell marker protein S100B in TtT/GF cells. TtT/GF cells were plated at 50 cells/mm2 on glass culture slides (Nunc, Life Technologies, Inc. Ltd., Paisley, UK) and grown until the cells reached confluence. Nonspecific antibody binding sites were initially blocked with 3% BSA in PBS at room temperature for 60 min before incubation with anti-ABCA1 antibody, diluted 1:300 (Novus). Immunoreacted cells were washed with PBS then incubated with fluorescein-conjugated goat antirabbit secondary antibody (Vector Laboratories, Inc., Burlingame, CA) for 1 h at room temperature. Sections were then washed and incubated with either an anti-ANXA1 antibody (a gift from Dr. Jamie Croxtall, William Harvey Research Institute, London, UK) diluted 1:6000 (42) or anti-S100B monoclonal antibody diluted 1:1000 (Sigma) at 4 C, overnight. The cells were then washed and incubated with either Texas red-conjugated donkey antisheep (ANXA1) or Texas red-conjugated horse antimouse (S100B) secondary antibody (Vector Laboratories, Inc.) for 1 h at room temperature. All sera were diluted in PBS containing 3% BSA. Nuclei were counterstained with TO-PRO-3 (Molecular Probes, Inc., Eugene, OR) and the slides mounted in Vectashield (Vector Laboratories Inc.) before visualization with a Leica TCS confocal microscope (Leica Microsystems, Wetzlar GmBH, Germany). Nonspecific immunostaining and background were assessed by substitution of nonimmune sheep, rabbit and mouse sera for primary antisera and incubation with secondary antibodies alone.
Data analysis
Semiquantitative measures of ANXA1 expression were made by comparisons of Western blot band optical densities (arbitrary units) to give a relative numerical guide to the ratios of the band intensities and their variance. Responses to dexamethasone and/or glyburide were calculated as a percentage of the corresponding drug-free control and expressed as the mean ± SEM (n = 3 gels). Measurements of ANXA1 immunofluorescence by plate reader were expressed as mean ± SEM (n = 8 wells). The experiment was repeated three times, and in each case the data profile was similar. Statistical comparisons between the normally distributed data from groups were made by ANOVA. Differences were considered significant if P < 0.05.
Data from the in vitro anterior pituitary release studies were initially shown to be normally distributed (Shapiro and Wilks test) and were then analyzed by standard parametric tests (ANOVA with post hoc comparisons by Duncan’s multiple range test). Statistical comparisons were made within experiments only and differences were considered significant if P < 0.05. Each of the studies was repeated several times (for specific details, see figure legends), and in all instances the data profile was similar.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Effects of glyburide on the externalization of ANXA1 by TtT/GF cells
Figure 1
demonstrates the time-dependent effect of glyburide (100 µM) on the amount of ANXA1 on the cell surface. Glyburide (100 µM) treatment for 3 h in culture alone did not affect the amount of ANXA1 detected on the surface of TtT/GF cells [Fig. 1, A (glyburide-treated, lanes 3 and 4; vs. control, lanes 1 and 2) and C]. However, 24-h glyburide treatment significantly (P < 0.01) reduced the amount of external ANXA1 [Fig. 1, B (glyburide-treated, lanes 3 and 4; vs. control lanes, 1 and 2) and C]. Figure 2 demonstrates the effects of glyburide (100 µM) on dexamethasone-induced translocation of ANXA1. A typical Western blot is shown in Fig. 2A. Cotreatment of TtT/GF cells with glyburide (3 h) blocked (P < 0.05) the increase in surface ANXA1 induced by dexamethasone (lanes 5 and 6, dexamethasone and glyburide; vs. lanes 3 and 4, dexamethasone alone; Fig. 2, A and B). These data were confirmed by immunofluorescence assay of ANXA1 immunoreactivity in nonpermeabilized cells treated in the same conditions (Fig. 2C) for 30 min or 3 h. However, in contrast, externalization of ANXA1 induced by 56 mM K+ was not affected by the presence of glyburide (Fig. 2C).
|
|
illustrates the effects of glyburide (100 µM) on the amount of ANXA1 on the outer surface of the anterior pituitary cells (i.e. in EDTA washes) and in the remaining tissue extract representing the intracellular ANXA1. Dexamethasone (0.1 µM) induced the expected increase in external ANXA1 [P < 0.01; Fig. 3, A (lanes 3 and 4 vs. control lanes 1 and 2) and C] as described previously (9) and a concomitant decrease (P < 0.05) in the ANXA1 content of the tissue extract [Fig. 3, B (lanes 3 and 4 vs. control lanes 1 and 2) and C]. Glyburide (100 µM) alone did not influence the amount of ANXA1 either in the surface or intracellular compartments (lanes 5 and 6 vs. control lanes 1 and 2, Fig. 3, A–C), but it completely blocked the externalization of ANXA1 induced by dexamethasone.
|
demonstrates the ability of glyburide (100 µM) to abrogate the dexamethasone-induced inhibition of the release of ir-ACTH evoked by forskolin (100 µM). In the absence of dexamethasone glyburide did not influence either the basal release of ir-ACTH or the significant (P < 0.01) increases in peptide release induced by forskolin. However, glyburide (100 µM) reversed significantly the powerful inhibitory effect (P < 0.01) of dexamethasone on the release of ir-ACTH induced by forskolin.
|
, lanes 4 and 5). Because the primers amplify conserved target sequences, this band presumably contained a mixture of sequences derived from the species of ABC transporter expressed by the cells. Specific ABCA1 cDNA was detected in TtT/GF cells (Fig. 5B) as appropriately sized bands produced by PCR using three pairs of ABCA1-specific primers (Fig. 5B, ABC1 PCR using F1 and R1 as predicted, 529 bp lane 2; F2 and R2 as predicted, 520 bp, lane 4; F3 and R3 as predicted, 209 bp, lane 6). ABCA1 cDNA was also detected in isolated primary mouse FS cells (Fig. 5C) as a 529-bp band following RT-PCR with the primer pair F1 and R1 (Fig. 5C, lane 2). In each reaction, the positive control RT-PCR for mouse G3PDH yielded a band at 960 bp (Fig. 5, A, lanes 1–3; B, lane 1; C, lane 1), whereas the negative control reaction yielded no band (Fig. 5, A, lane 6; B, lanes 3 and 5; C, lane 3). In human lung-derived adenocarcinoma A549 cells, a cell type well characterized to translocate ANXA1 in response to dexamethasone (Fig. 5D; Ref. 22), mRNA for ABCA1 was detected at the predicted 671-bp band size (Fig. 5D, lanes 1–3; primers F1 and R2). Figure 6 shows the typical appearance of isolated enriched FS cells viewed by electron microscopy. The nuclei contained condensed chromatin and numerous characteristic microvilli and cell processes were observed. Counting of ir-S100 positive cells revealed that the enriched FS cell preparation contained 92 ± 5% FS cells, 8 ± 3% damaged cell debris. Western blot analysis (Fig. 7) using an antihuman ABCA1-specific antibody that cross-reacts with mouse and rat ABCA1 readily detected the presence of the predicted band at 220 K in A549 cells (lanes 3 and 4), enriched mouse FS cells (lanes 5 and 6), and TtT/GF cells (lanes 7 and 8). However, no immunoreactive bands were detected in extracts of GH3 cells (lanes 1 and 2). Double immunofluorescence labeling of ABCA1 and ANXA1 revealed near total overlap of ANXA1 and ABCA1 immunoreactivity (Fig. 8, A–C). In TtT/GF cells that had been fixed but not permeabilized with detergent, ANXA1 immunofluorescence was concentrated in several peripheral patches of immunoreactivity at the tips of the processes that the TtT/GF cells develop in culture (as previously reported by us; Ref. 24). ABCA1 immunoreactivity was similarly distributed (Fig. 8, B and C). Colocalization (overlay Fig. 8F) of the S100B FS marker protein (Fig. 8E) with ABCA1 (Fig. 8D) was also detected. No immunolabeling was detected on cells incubated with the respective nonimmune sera or secondary antibodies only (data not shown).
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our experiments used both TtT/GF cells and primary static incubates of pituitary segments. TtT/GF cells display the morphological features of FS cells (processes, follicle formation) and express the FS cell marker proteins S100 and glial fibrillary acidic protein (23). Our recent studies demonstrate that TtT/GF cells contain abundant ANXA1 and externalize ANXA1 in response to glucocorticoid exposure (24). The TtT/GF cell line is therefore an excellent model for investigating the mechanism of export of ANXA1 in a pure FS cell population. The advantage of the pituitary segment model however is that the FS cells reside in situ and therefore retain the intimate FS and endocrine cell-cell contacts and three-dimensional tissue arrangements that sustains the paracrine communication inherent to the tissue as a whole (2, 3, 4, 5). Due to their extensive processes and small size, FS cells are notoriously difficult to separate and purify (22). A method based on BSA sedimentation to separate primary FS cells from isolated mouse anterior pituitary after perfusion to remove blood cells was developed. The resulting yield of FS cells was low but a highly enriched preparation was achieved as validated by examination of isolated cell preparations by electron microscopy. Unfortunately, the low yield precluded the use of isolated FS cells in functional experiments, but sufficient numbers were obtained for investigation of ABCA1 transporter expression by RT-PCR and Western blot analysis.
The striking ability of glyburide to reverse both the export of ANXA1 and the inhibitory actions of dexamethasone in the anterior pituitary segment preparation adds further support to our hypothesis that export of ANXA1 plays a key role in effecting the negative feedback actions of glucocorticoids on the hypothalamo-pituitary-adrenal axis in the rat (2, 3, 9). We have previously suggested that the cellular exportation of ANXA1 is critical to its action as it provides a means by which the protein can gain access to receptors on the outer surface of the cells. This concept is supported by evidence that treatments that block the acute production of the protein (ANXA1 antisense or protein synthesis inhibitors) also inhibit the regulatory actions of dexamethasone on ir-ACTH release and by previous observations that the generation and exportation of the protein develops in parallel with the inhibition of hormone release (9). Moreover, antisera to ANXA1 that would not be expected to penetrate cell membranes but could sequester ANXA1 at the cell surface specifically reverse the inhibitory actions of glucocorticoids on ir-ACTH release from pituitary segments in vitro and in vivo (9). Similarly, ANXA1 recombinant protein ANXA1 1–346 and ANXA1 1–188, which would also be unlikely to enter cells easily, readily depress ir-ACTH release (9). We have previously demonstrated the presence of high affinity, saturable ANXA1-binding proteins on the surface of several pituitary cell types including corticotrophs, which are essential for the biological actions of ANXA1 (11). FS cells contain glucocorticoid receptors (44) and are the principal source of ANXA1 within the pituitary gland (10). Dexamethasone may therefore act indirectly via the FS cells to inhibit ACTH release from corticotrophs via stimulation of ABC transporter-mediated externalization of ANXA1. In the absence of dexamethasone, although glyburide treatment of TtT/GF cells for 3 h has no effect, treatment for 24 h caused a significant reduction in the amount of external ANXA1. This effect of long-term glyburide treatment revealed the process of continuous replenishment of extracellular ANXA1 from intracellular ANXA1 stores in the absence of steroid. These data could not be explained by a toxic effect of glyburide because TtTGF cell viability was unaffected. Although the precise roles of FS cells are not well understood increasing evidence suggests that, in addition to ANXA1, FS cells release a number of paracrine factors (e.g. IL-6, basic fibroblast growth factor, S100B, leptin, TGF-ß), which influence the secretory activity of the surrounding endocrine cells (22, 45). The mechanism of secretion of these factors is unknown.
The molecular mechanism underlying the inhibitory effect of ABC transport blockers on ANXA1 secretion is not clear. The results allow us to surmise the implication of an ABC related-transporter in the secretory pathway of ANXA1. Whether ABCA1 is indeed the channel through which ANXA1 translocates the membrane or whether it fulfils a permissive role for the process to take place remains to be investigated. Although ANXA1 may appear very large (37K) for ABC transmembrane transport, ABC transporters can handle a wide variety of substrates and are able to mediate energy-dependent membrane translocation of large molecular weight proteins, such as toxins secreted by gramnegative bacteria (26). The reduced secretion of ANXA1 in response to dexamethasone did not result from altered synthesis or stability of intracellular pools as the drug did not influence the amount and size of intracellular ANXA1 measured in anterior pituitary cells. The inhibition by glyburide appears to be selective via an ABC-linked mechanism as secretion of ir-ACTH via the classical exocytotic pathway was not affected in the presence of glyburide. Although glyburide may be influencing externalization of ANXA1 via inhibitory effects on SUR, such an effect would be expected to result in cell depolarization (the mechanism underlying the therapeutic effect of the drug to stimulate ß-cell insulin release). However, glyburide was not found to have an enhancing effect on either ir-ACTH release or ANXA1 externalization. It is not known whether either FS or TtT/GF cells express SURs. Moreover, ANXA1 externalization in response to the nonspecific depolarizing stimulus 56 mM K+ was not influenced by the presence of glyburide and therefore does not appear to act via an ABC-transporter mechanism. Membrane depolarization in response to K+ may affect a variety of membrane properties including phospholipid reorganization and intracellular messenger systems but as yet the mechanism by which 56 mM K+ stimulates ANXA1 externalization is unknown.
Among the several subclasses of ABC transporters, the ABCA subclass has received considerable attention because mutation of the ABCA1 gene causes a severe high density lipoprotein deficiency syndrome called Tangier disease (46). ABCA1 is proposed to be a transmembrane transporter for intracellular cholesterol and phospholipids and is crucial for the production of high density lipoproteins (47). However, the expression of ABCA1 in tissues not involved in cholesterol transport suggests that ABCA1 is likely to have functional roles in addition to cholesterol/phospholipid efflux (48). ABCA1 was selected as a candidate to investigate as a putative ANXA1 transporter in FS cells for a number of reasons. Firstly, ABCA1 mRNA has been demonstrated to be strongly expressed in tissues that are known to export ANXA1 (lung, adrenal, pituitary, macrophages; Refs. 33 and 48). Secondly, ABCA1 has been shown to mediate translocation of other proteins lacking a signal sequence, including IL-1ß from mouse macrophages (which also express and externalize ANXA1; Refs. 1 and 30). Thirdly, ABCA1 is involved in physiological processes in other systems that have been ascribed to FS cells (49), e.g. engulfment of apoptotic bodies (50, 51) and anion efflux (52). ABC transporter mRNA of the predicted size was initially detected in TtT/GF cells by use of degenerate primers designed to recognize any ABC transporter sequence. FS cells therefore express member(s) of the ABC transporter family, but given the large number of ABC family members that exist this was not surprising (26, 27). Subsequent RT-PCR experiments using primer pairs specific to the ABCA1 transporter demonstrated that TtT/GF cells and isolated FS cells express ABCA1 mRNA. Furthermore, ABCA1 protein expression in these cells was confirmed by Western blotting and immunofluorescence labeling. We do not exclude the possibility that other pituitary cell types express ABCA1 as in the literature the expression of ABCA1 appears ubiquitous (48). However, GH3 cells that secrete GH and prolactin did not contain immunoreactive ABCA1. In TtT/GF cells, our recent studies have shown that ANXA1 is localized to specific foci on the plasma membrane and that glucocorticoid treatment increases the intensity of ANXA1 at these foci (24). The present study confirmed the localization of ANXA1 in TtT/GF cells and demonstrated that ABCA1 was colocalized in the specific areas of the plasma membrane in the processes of FS cells immunoreactive for ANXA1. An identical, specific staining pattern was obtained for surface S100B and ABCA1 in TtT/GF cells. Immunogold localization of ANXA1 in anterior pituitary has shown accumulations of immunoreactivity in the FS cell membrane adjacent to the sites of contact with the endocrine cells (24). Together, these observations suggest the novel possibility that ANXA1 and other intercellular signals are transferred directly from processes of FS cells by an ABCA1 mechanism to surrounding endocrine cells at points of stable anatomical contact.
The finding that ABCA1 mRNA and protein was also detected in lung adenocarcinoma A549 cells adds support to the hypothesis that a conserved mechanism for ANXA1 externalization exists between different ANXA1-secreting cell types. ANXA1 is secreted into broncho-alveolar fluid and lung tissue expresses large amounts of both ABCA1 and ANXA1 localized to alveolar macrophages and small bronchioles (48). Although the ABCA3 transporter is also expressed in human type II cells, it has been localized to the limiting membrane of lamellar bodies (53) and would therefore be an unlikely candidate for the ANXA1 transporter in these cells.
The present studies in TtT/GF cells confirm our previous observations in anterior pituitary tissue that dexamethasone stimulates ANXA1 externalization within 30 min (9). The glucocorticoid is therefore likely to be acting via a novel rapid mechanism, and our more recent work has identified a role for protein kinase C in this respect and shown that the exported protein is serine phosphorylated (54). Increasing evidence has accumulated to support the premise that glucocorticoids (and other steroid hormones) exert rapid effects that cannot be accounted within the classical mechanism through alterations in transcription brought about by interactions of activated ligand-receptor complex with DNA, transcription factors and coregulators (55). A possibility is that serine phosphorylation is required for ABC-dependent translocation of ANXA1. In addition, rapid glucocorticoid effects may influence ABC transporter activity directly. The activity of ABCA1 is stimulated by cAMP when expressed in Xenopus oocytes (52) and, in the long term, transcription of ABCA1 has also been shown to be induced by cAMP (56) as well as by nuclear receptor LXR and RXR pathways in macrophages (57, 58). The effect of glucocorticoids on ABCA transporter expression and activity remains to be investigated.
In conclusion, these studies strongly suggest that ANXA1 externalization, rapidly induced by dexamethasone in a FS cell line and in rat anterior pituitary, is mediated by an ABC transporter mechanism. Consistent with these findings, TtT/GF FS cells and primary anterior pituitary FS cells express ABC transporters and specifically ABCA1, a good candidate for the transporter of ANXA1. It is possible that other ABC transporter isoforms are expressed in FS cells and current studies are underway to prove ANXA1 transport is dependent on ABCA1 in recombinant cell systems.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ABC, ATP-binding cassette; ANXA1, annexin 1; CFTR, cystic fibrosis transmembrane conductance regulator; EBSS, Earle’s balanced salts solution; FS, folliculo-stellate; ir, immunoreactive; MDR, multidrug resistance; SURs, sulfonylurea receptors 1 and 2.
Received June 25, 2002.
Accepted for publication December 2, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Hastie, J. R. Masters, S. E. Moss, and S |
, dexamethasone (0.1 µM)-treated cells, 
, K+ (56 mM)-treated cells. Expressed as mean ± SEM, n = 8 wells. **, P < 0.01, *, P < 0.05 compared with control; 2 + P < 0.01, +P < 0.05 vs. dexamethasone-alone group at relevant time point. NS, Not significant, ANOVA. Data are typical of three experiments.
) and intracellular (
