Maitotoxin induces a concentration-dependent 45Ca uptake in primary cultures of rabbit tracheal epithelial cells. This response is insensitive to the calcium channel antagonists nifedipine, diltiazem and verapamil up to 20 microM. However, verapamil at 200 microM completely prevents 45Ca uptake. Measurements of indo-1 fluorescence show that MTX induces a very sustained (> or = 2 h) [Ca]i rise, which is completely inhibited by 200 microM of verapamil. Genistein (110 microM) (an inhibitor of tyrosine kinases) also strongly inhibits it. The inhibitory effect of 50 microM miconazole (an inhibitor of cytochrome P450) is only partial. Okadaic acid (inhibitor of protein-phosphatases) primarily delays the response to the toxin without decreasing its magnitude. MTX induces the formation of (1,4,5) inositol trisphosphate (IP3). The MTX response curve is biphasic. Stimulation is transient (< or = 10 min) and is not inhibited by chelation of intracellular Cai with BAPTA, nor by verapamil (200 microM) or U73122 (10 microM) (an inhibitor of activation of PLC beta 1 through a trimeric G protein). Results suggest that MTX independently activates a calcium transport process (which might imply phosphorylation on tyrosine residues) and a PLC not linked to a trimeric G protein.

A quantitative analysis by flow cytometry of the rate and extent of endocytosis of ligands bound to the membrane immunoglobulins of mouse B-splenocytes is reported. The temperature dependence and the response to inhibitors of oxidative metabolism are described. Inhibitors of the cytoskeleton (cytochalasin B, vinblastine and colchicine) and of calmodulin (trifluoperazine) do not interfere with endocytosis at non-lethal doses. Similarly fetal calf serum does not modify the rate and extent of the process. Endocytosis occurs in a similar time range, but to a lesser extent, when the ligand is monovalent than when it cross-links the membrane immunoglobulins. Finally, it is shown that, within minutes after its internalization, the divalent ligand is found in an acidic environment, while the monovalent ligand is not. These results, in agreement with the model of receptor recycling, suggest that the divalent ligand-receptor complex is indeed internalzed and captured in an acidic environment while the monovalent ligand-receptor complex is internalized and probably rapidly recycled back to the cell surface.

No abstract

Pemphigus is an intraepidermal autoimmune blistering disease of humans caused by circulating IgGs. We have investigated the binding mode and the fate of bound antibodies from Pemphigus sera (P-IgG) on guinea pig keratinocytes in suspension in order to find clues to the loss of cell adhesion in vivo (acantholysis). Flow cytometry, following indirect immunofluorescent labeling of the keratinocytes, and dead cells' staining with ethidium bromide, demonstrated the specific surface binding of P-IgG onto living keratinocytes only. This was shown with several Pemphigus sera or purified P-IgG. This technique, used with various Pemphigus sera, showed that the specific binding is increased when the serum titer is higher, and "Km" values for P-IgG were roughly and inversely correlated to the titers. Upon saturation the same average number of Pemphigus IgG sites per cell were found for the sera of different patients. Analysis of the specific binding of [125I]-P-IgG onto Percoll-separated (living) keratinocytes showed the existence of two classes of sites: 2 x 10(6) sites/cell high-affinity sites (Kd = 1.5 x 10(-6) M total IgG) and 25 x 10(6) sites/cell low-affinity sites (Kd = 6 x 10(-5) M total IgG). Cell sorting and flow cytometry of individual cells allowed us to correlate the light-scattering signal, the RNA content, the size and morphology, and the P-IgG binding to the cells. The results indicated that P-IgG binding is homogeneous within the living keratinocytes and increases with cell size (cell maturity). Cell-sorter analysis of cells with membrane-bound P-IgG, coupled to direct determination of P-IgG released in the medium, revealed the fate of bound P-IgG: 40-60% of the P-IgGs were released in the medium within 30 minutes at 37 degrees C. This was accompanied and followed by a much slower, metabolic energy-dependent, internalization process of the membrane-bound P-IgG. The internalization has been confirmed by electron microscopy of bound P-IgG labeled with protein A-gold. Internalized IgGs were seen in the cells in coated membranous vesicles and other endocytic compartments. Similar behavior was also observed with two other membrane ligands: i.e., concanavalin A and multispecific rabbit "antisurface" antibodies.