Stimulated release of fluorescently labeled IgE fragments that efficiently accumulate in secretory granules after endocytosis in RBL-2H3 mast cells

Keli Xu 0 2 Rebecca M. Williams 1 David Holowka 2 Barbara Baird 2 0 Present address: Molecular and Cellular Physiology, Beckman Center, Stanford University School of Medicine , Stanford, CA 94305 , USA 1 Department of Applied and Engineering Physics, Cornell University , Ithaca, NY , USA 2 Department of Chemistry SUMMARY Sensitization of RBL-2H3 mast cells with monomeric fluorescein-5-isothiocyanate (FITC)-labeled immunoglobulin E (IgE) results in slow but highly efficient accumulation of labeled IgE fragments in a pool of acidic peripheral vesicles that are visible by fluorescence microscopy after raising endosomal pH with ammonium chloride. Stimulation of cells containing these FITC-IgE fragments by aggregation of high affinity receptors for IgE (Fce RI) or by Ca2+ ionophore and phorbol 12-myristate 13acetate results in release of FITC fluorescence from the cells, which can be monitored continuously with a spectrofluorometer. The fluorescence release process corresponds to cellular degranulation: it is prevented under conditions that prevent stimulated b -hexosaminidase release, and these two processes exhibit the same antigen dose-dependence and kinetics. Pulse-chase labeling reveals that aggregation of FITC-IgE bound to Fce RI at the cell surface causes internalization and delivery to the regulated secretory vesicles with a high efficiency similar to monomeric IgE-Fce RI, but more rapidly. Binding of Cy3For RBL-2H3 cells, a highly studied rat mast cell line, aggregation of Fce RI causes cellular degranulation and consequent release of preformed mediators, including histamine (Barsumian et al., 1981), serotonin (Taurog et al., 1979) and b -hexosaminidase (Ortega and Pecht, 1988). These cells are of mucosal mast cell lineage, based on morphological and histochemical criteria (Seldin et al., 1985). They lack the typical appearance of serosal mast cells from the rat peritoneum, which contain a large number of tightly packed dense granules in their cytoplasm that undergo explosive compound exocytosis after stimulation (Galli et al., 1984). However, exocytic granules in RBL-2H3 and peritoneal mast cells share an antigenic marker, an 80 kDa transmembrane protein, which is also found in lysosomes and lytic granules of natural killer cells but is absent from the exocytic granules of endocrine pituitary cells and exocrine pancreatic acinar cells (Bonifacino et al., 1986, 1989). This and other evidence suggests that secretory granules in natural killer cells, mast modified IgE to Fce RI results in labeling of the same secretory vesicles as in FITC-IgE-sensitized cells, and these Cy3-labeled vesicles can be observed by fluorescence microscopy without neutralization of intracellular compartments. Simultaneous three-photon microscopy of serotonin fluorescence and two-photon microscopy of Cy3 fluorescence reveals that these Cy3-labeled vesicles coincide with serotonin-labeled secretory granules. After stimulation of the cells via aggregation of IgE-Fce RI or addition of Ca2+ ionophore and phorbol 12-myristate 13acetate, depletion of the Cy3 label from the intracellular vesicles is observed with confocal microscopy. These results provide strong evidence for the lysosomal nature of secretory granules in these cells. In addition, they provide the basis for a direct, real-time method for monitoring single cell degranulation. cells and other hematopoietic cells are actually regulated secretory lysosomes (Griffiths, 1996). A previously untested prediction of this hypothesis is that proteins internalized via the endocytic machinery and destined for degradation should be delivered to these secretory granules and released upon appropriate stimulation. Little is known about the molecular events involved in the terminal steps of Fce RI-mediated exocytosis, although evidence exists for the involvement of GTP binding proteins (Gomperts, 1990). RBL-2H3 cells have a number of attractive features for studying regulated exocytosis, but a limitation of these cells has been the inability to routinely monitor receptormediated exocytosis of individual cells. Previous studies showed that monomeric immunoglobulin E (IgE) binds tightly to Fce RI on RBL-2H3 cells yielding a complex with a very long lifetime at the cell surface (Kulczycki et al., 1974; Isersky et al., 1979). Aggregation of this complex by multivalent antigen at 37C causes rapid IgE-Fce RI endocytosis, such that 50-60% of these complexes are internalized with a half-time of approx. 5 minutes (Furuichi et al., 1984). This process, which occurs via coated pits (Pfeiffer et al., 1985; Mao et al., 1993), delivers the aggregated IgE-Fce RI complexes to endosomal vesicles and lysosomes, where they undergo proteolytic degradation (Isersky et al., 1983). Unlike aggregation-dependent Fce RI-mediated-degranulation, this Fce RI endocytosis does not depend on extracellular Ca2+ (Furuichi et al., 1984) and is inhibited by cytochalasin D, an inhibitor of microfilament polymerization (Ra et al., 1989). We previously characterized the quenching of fluorescein-5isothiocyanate (FITC) fluorescence that occurs when FITC-IgE binds 2,4-dinitrophenyl (DNP)1 ligands (Erickson et al., 1986; Holowka and Baird, 1996). In the course of studying the binding of multivalent 2,4-dinitrophenyl (DNP)-BSA to FITCanti-DNP-IgE on RBL-2H3 cells under stimulating conditions (Xu et al., 1998), we were surprised to observe a timedependent increase in FITC fluorescence that followed an initial quenching phase. Investigation of the basis for these observations revealed that internalized FITC-IgE is efficiently delivered to secretory vesicles that undergo exocytosis after cellular activation. These studies have led to simple fluorescence methods for directly monitoring the kinetics of stimulated exocytosis and for observing stimulated degranulation in individual cells. They also provide new support for a direct relationship between lysosomes in the endocytic pathway and regulated secretory granules in these cells. MATERIALS AND METHODS FITC was purchased from Molecular Probes, Inc. (Eugene, OR). The amino-reactive carbocyanine probe, Cy3, was purchased from Biological Detection Systems, Inc. (Pittsburgh, PA). Cytochalasin D, 5-hydroxytryptamine (serotonin), phorbol 12-myristate 13-acetate (PMA) and the Ca+ ionophore A23187 were purchased from Sigma Chemical Co. (St Louis, MO). Affinity-purified rabbit anti-fluorescein antibody was purchased from East Acre Biologicals (Southbridge, MA). Mouse monoclonal anti-DNP-IgE was purified from ascites cells (Liu et al., 1980) by affinity chromatography (Holowka and Metzger, 1982) and gel-filtration chromatography (Subramanian et al., 1996). Mouse monoclonal IgE specific for 5-(dimethylamino)naphthalene-1-sulfonyl (DNS) was purified from the supernatant of a switch-variant hybridoma cell line 27-74 (Dangl et al., 1988) by affinity chromatography (Weetall et al., 1990). FITC-anti-DNP-IgE was prepared as previously described (Erickson et a (...truncated)