In physiological conditions, collagen degradation by fibroblasts occurs primarily via phagocytosis, an intracellular pathway that is thought to require collagen receptors and actin assembly for fibril internalization and degradation. Currently it is unclear which specific steps of collagen phagocytosis in fibroblasts involve actin filament assembly. As studies of phagocytosis in fibroblasts are complicated by the relatively slow rate of particle internalization compared to professional phagocytes, we have examined the role of collagen receptors and actin only in the initial collagen binding step. Prior to the binding of collagen-coated fluorescent beads by human gingival fibroblasts, a cell type that is avidly phagocytic in vitro, cells were treated with cytochalasin D (actin filament barbed-end capping) or swinholide A (actin dimer sequestering and severing) or latrunculin B (actin monomer sequestering). Bead binding and immunostaining of (alpha)(2)(beta)(1) and (alpha)(3)(beta)(1) integrin collagen receptors were measured by flow cytometry. After 1-3 hours of coincubation with beads, cytochalasin D or swinholide A eliminated actin filaments stained by rhodamine-phalloidin and inhibited collagen bead binding (reductions of 25% and 50%, respectively), possibly because of cell rounding and restricted interactions with beads. In contrast, latrunculin enhanced binding dose-dependently over controls (twofold at 1 microM) and induced the formation of brightly staining aggregates of actin and the retention of long cytoplasmic extensions. Latrunculin also reduced surface (beta)(1), (alpha)(2) and (alpha)(3) integrin staining up to 40% in bead-free and bead-loaded cells, indicating that latrunculin enhanced collagen receptor internalization. As determined by fluorescence recovery after photobleaching, latrunculin increased the mobility of surface-bound (beta)(1) integrin. The stimulatory effect of latrunculin on collagen bead binding was reduced to control levels by treatment with a (beta)(1) integrin inactivating antibody while a (beta)(1) integrin blocking antibody abrogated both bead binding and the latrunculin-induced stimulation. Immunoblotting of bead-associated proteins showed that latrunculin completely eliminated binding of (beta)-actin to collagen beads but did not affect (beta)(1) integrin binding. These data indicate that latrunculin-induced sequestration of actin monomers facilitates the disengagement of actin from (beta)(1) integrin receptors, increases collagen bead binding and enhances collagen receptor mobility. We suggest that these alterations increase the probability of adhesive bead-to-cell interactions.