Colony-stimulating factor-1 induces rapid behavioural responses in the mouse macrophage cell line, BAC1.2F5

C. A. BOOCOCK 0 1 G. E. JONES 0 1 E. R. STANLEY 0 1 J. W. POLLARD 0 1 0 Department of Developmental Biology and Cancer, Albert Einstein College of Medicine , 1300 Morris Park Avenue, Bronx, .YV 10461 , USA 1 Department of Anatomy & Human Biology, King's College London , Strand, London WC2R 2LS , UK - The cloned, SV40-immortalized mouse macrophage cell line, BAC1.2F5, resembles primary macrophages in its dependence on colony-stimulating factor-1 (CSF-1) for both viability and proliferation. Re-addition of CSF-1 stimulates rapid, transient behavioural changes in starved cells, which are rounded, with diffusely organized F-actin and few intracellular vesicles. Within 1 min, cells begin to spread, forming prominent, F-actin-rich ruffles. Small vesicles (O'S-l-O^im), formed throughout extending lamellar processes, move centripetally and, after 3-5 min, fuse to form larger vesicles (2-0-4'Ojwm), clustered around the nucleus. Immunofluorescence demonstrates that CSF-1, bound to cell-surface receptors, is internalized via these vesicles. Cell spreading and ruffling peak about 5 min after restimulation. Interference reflection microscopy indicates no corresponding change in the mode of cell-substratum adhesion: a single area of close adhesion underlies most of the cell and simply broadens during spreading. Analysis of cell aggreColony-stimulating factor-1 (CSF-1) stimulates the proliferation and differentiation of macrophages and their precursors (Stanley, 1986). Its cell surface receptor, a 165 000 My protein tyrosine kinase expressed by cells of the monocyte-macrophage lineage, is the product of the proto-oncogene c-fms (Sherr et al. 1985). On binding of CSF-1, this receptor mediates the phosphorylation of several cellular proteins as well as autophosphorylation at tyrosine residues (Sherr et al. 1985; Yeung et al. 1987; Downing et al. 1988; Jubinsky et al. 1988). Uncontrolled cell proliferation can result from expression of CSF-1 and its receptor by the same cells (Baumbach et al. 1987; gation kinetics shows no effect of CSF-1 on intercellular adhesiveness. Measurement of cell areas after starvation and restimulation demonstrates quantitatively the time-course and concentrationdependence of cell spreading. Mean area doubles within 5 min and, after a transient peak, decreases within 30 min to the value measured before starvation. This time-course corresponds to that of CSF1 internalization and of the phosphorylation and subsequent degradation of CSF-1 receptors. The concentration-dependence of the spreading response resembles that of CSF-1-dependent survival and proliferation. The minimum detectable stimulation of spreading occurs at the concentration (22 pM) that supports survival without proliferation. Increasing stimulation of spreading occurs over the range of concentrations that elicit increasing proliferation. Roussel et al. 1987) or from expression in CSF-1 dependent cells of a modified and constitutively active receptor kinase, encoded by the v-fnts oncogene of McDonough strain feline sarcoma virus (Wheeler et al. 1986). As a model system in which to study the relation between early changes induced by growth factor binding to cells and long-term effects associated with the control of cell growth, we are using the simian virus 40 (SV40)immortalized mouse macrophage cell line BAC1.2F5 (Morgan et al. 1987). BAC1.2F5 cells retain many characteristics of primary macrophages, including dependence on CSF-1 for both survival and proliferation in culture (Tushinski et al. 1982; Morgan et al. 1987). In the absence of CSF-1, they cease to divide and eventually die. On re-addition of CSF-1, cells show a decrease in protein degradation and, after a 16 h lag, reinitiate DNA synthesis. Much earlier changes include the phosphorylation of cytoplasmic and membrane-associated proteins, including the CSF-1 receptor itself, the rapid, receptormediated internalization of CSF-1, degradation of both ligand and receptor, dramatic changes in cell shape and spreading and elevated transcription of the growthassociated genes, c-myc, c-fos, KC and JE (Morgan et al. 1987, Orlofsky & Stanley, 1987; Downing et al. 1988; Jubinsky et al. 1988; Sengupta et al. 1988). Since cell spreading is among the earliest observed responses to re-addition of CSF-1 to either BAC1.2F5 cells (Morgan et al. 1987) or to bone-marrow-derived macrophages (Tushinski et al. 1982), we have characterized this response in detail. Using phase-contrast microscopy, scanning electron microscopy, interference reflection microscopy (IRM), fluorescence cytochemistry and indirect immunofluorescence, we correlate the cell spreading and morphological changes rapidly induced by CSF-1 with the pattern of cell-substratum adhesion, the organization of the actin cytoskeleton and the internalization of cell-surface-bound CSF-1. In addition, by measuring cell areas during this response, we relate the time-course of CSF-1-induced cell spreading to that of the internalization of CSF-1 and the phosphorylation and subsequent degradation of its receptor. We also compare the concentration-dependence of the spreading response with the requirement for BAC1.2F5 cell survival and proliferation. Materials and methods Cell culture Cells used in these experiments were a subclone of the cloned mouse macrophage cell line BAC1.2F5 (Morgan et al. 1987). These were maintained in growth medium consisting of aMEM (Gibco Ltd, Paisley, Scotland) supplemented with 10% (v/v) foetal calf serum (FCS; Imperial Laboratories), 1-32 nMCSF-1 (Cetus Corporation), 2mM-L-glutamine, 0-lSmM-Lasparagine, 15 nM-/?-mercaptoethanol, 77-5 units ml~ streptomycin and 25 units ml"1 penicillin. Cells were grown at 37C in 100 mm tissue culture dishes (Becton Dickinson & Co.) kept in a humid atmosphere of 5 % CO2. Growth medium was changed twice weekly and cells were subcultured weekly at 2X106 cells/dish by dissociating with a Teflon scraper and resuspending in growth medium at 2 x 10s cells ml~'. CSF-1 For immunofluorescent labelling with antibodies against murine CSF-1, CSF-1 was purified from mouse L-cell conditioned medium as described by Stanley (1985). In all other experiments and in growth medium used for maintenance of cells, human recombinant CSF-1 (Cetus Corporation) was used. For comparability with previous papers, we have sometimes given concentrations of CSF-1 in units ml~ . One unit measured by radioimmunoassay corresponds to 0-44 fmol of CSF-1 (Stanley, 1985). Microscopy of living cells Cells precultured 1-4 days on acid-washed glass coverslips were rinsed and incubated for a further 24-36 h in medium lacking CSF-1. Starved cultures mounted in transparent chambers were transferred to the heated stage of a Zeiss IM35 inverted microscope equipped for phase-contrast microscopy and IRM. Cultures were photographed at intervals during perfusion of chambers with prewarmed growth medium containing l-32nMCSF-1. Cells precultured and starved, as described above, on 16 mm diame (...truncated)