The extremely restricted tropism of parvovirus B19 to human erythroid progenitors cells has been suggested to be mediated by blood group P antigen, which the virus uses as a cellular receptor for binding. We reported the generation of recombinant parvovirus B19 vectors (J. Virol., 72: 5224-5230, 1998), and concluded that P antigen is necessary, but not sufficient for parvovirus B19 infection (J. Virol., 75: 4110-4116, 2001). We subsequently identified activated α5β1 integrin as a co-receptor for viral entry (Blood, 102:3927-3933, 2003). To further elucidate the mechanisms that elicit the co-receptor activity of α5β1 integrins, we investigated the mediators involved in the proposed two step-model of integrin activation (release from cytoskeletal restraints and re-engagement of the cytoskeleton and assembly of a signaling complex upon ligand binding). Low-dose cytochalasin D treatment (10 nM cytoD), which disrupts the cortical cytoskeleton and releases integrins from their cytoskeletal restraints (step 1), increased parvovirus B19 transduction in PMA-differentiated K562 cells, whereas higher concentrations of cytoD (> 2 mM) inhibited parvovirus B19 transduction, suggesting that disruption of the cytoskeleton/signaling complex (step 2) impaired viral entry. The low-dose cytoD-induced increase in viral entry was abrogated by combined cell-surface immobilization of β1 and β3, but not β2, integrins. We also investigated the role of small GTPase Rap1, which facilitates cytoskeletal re-engagement and assembly of integrin signaling complex (step 2), in parvovirus B19 entry stably transfecting K562 cells with either constitutively active (Rap63E) or dominant negative (Rap17N) Rap1 expression cassettes. Transduction of PMA-differentiated Rap63E and Rap17N transfectants was reduced by 50% and 30%, respectively, compared to PMA-differentiated control cells, and correlated with 50% and 30% reduction, respectively, in cell surface expression of β1 integrins, suggesting that GDP/GTP cycling is involved in PMA-induced integrin expression. Outside-in signaling of β1 integrins was similar in PMA-differentiated Rap63E transfectants and control cells but impaired in Rap17N transfectants, suggesting that activated Rap1 proteins are crucial for the induction of β1 integrin co-receptor activity. Interestingly, over-expression of constitutively active Rap1 protein in undifferentiated K562 cells was sufficient to render these cells permissive for parvovirus B19 infection. Transduction of undifferentiated Rap63E transfectants was sensitive to outside-in signaling through β1 integrins, as determined by antibody-mediated stabilization of the high affinity β1 integrin conformation and β1 integrin cross-linking, and was completely abrogated in the presence of function-blocking β1 integrin antibodies and cytoD. Thus, our results suggest that Rap1 protein is crucial for β1 integrin activation observed after prolonged PMA-treatment, and that active Rap1 is sufficient to convert 'default inactive' β1 integrins expressed on non-adherent erythroleukemic K562 cells into functionally active co-receptors for parvovirus B19 entry. These studies have implications in the potential use of recombinant parvovirus B19 vectors in human hematopoietic cell gene therapy.