The actin cytoskeleton is a key superstructure required for cellular motility and cell adhesion to the extracellular matrix. The actin network and its associated proteins also influences signaling pathways, the cell cycle and cellular metabolism. Hence, the characterization of the interaction between actin and its binding proteins is fundamental to cell biology. Profilin is an ubiquitous 15 kDa protein which has four established direct ligands in vitro: (1) profilin binds to monomeric actin (stoichiometry; one to one): (ii) profilin binds to poly-L-proline (PLP,one profilin binds to a stretch of greater than prolines); (iii) profilin binds membranes containing polyphosphoinositides (one profilin binds to a cluster of 5 phospholipid molecules); (iv) profilin binds VASP (vasodilator stimulated phosphoprotein, unknown stoichiometry), possibly through its interaction wit a proline-rich stretch of the VASP molecule. Although it is clear that profilin regulates the formation of actin polymers in vivo, data from cultured cells and various organisms are confounded by the observation that, depending on the system, profilin can either promote actin assembly, or inhibit actin polymerization. Despite these results, profilin is clearly essential for life in multicellular organisms such as Drosophila and mice. Thus, whatever the role of profilin may be, cells and organisms lacking profilin are not viable. In particular, mice null for profilin I do not survive beyond the earliest organisms lacking profilin are not viable. In particular, mice, null for profilin I do not survive beyond the earliest embryonic stages (few cells). Which function of profilin is critical for its effect on survival remains unknown. To characterize profilin's essential function(s), we propose to combine powerful genetic techniques with biochemical experiment on profilin mutants generated by site directed mutagenesis. We have used a replication incompetent adenovirus as vector, the backbone of which contains the human profilin I cDNA cloned by homologous recombination, to induce rapid overexpression of profilin in aortic endothelial cells. We observed that the effect of profilin overexpression is to promote the adhesion of endothelial cell to extracellular matrix proteins and in particular, to fibronectin. We have engineered purified and studied biochemically fifteen profilin mutants generated by single amino acid substitutions. We have identified mutants with abnormal function in vitro. In particular, mutant 88-R/L (arginine at amino acid 88 substituted for a leucine) does not bind actin nor increase actin nucleotide exchange has reduced interaction with phosphoinositides but normal binding to PLP. Mutant 125-H/D has reduced interaction with phosphoinositides but normal binding to actin, normal nucleotide exchange function, and normal binding to PLP. Mutant 119-H/D has normal binding to phosphoinositides and normal binding to PLP, but reduced binding to actin, and lacks the ability to increase nucleotide exchange. Other mutants have reduced PLP binding activity, but normal interaction with actin and phosphoinositides, or increased affinity for phosphoinositides but normal binding to actin and PLP. We have generated transgenic mice lacking one allele for profilin I, and shown reduced profilin I concentration in the cells of these animals. For this project, we propose to use in vitro assays of profilin to identify nutants displaying selective deficiencies of profilin functions. Next, we will examine the ability of selected profilin mutants to rescue transgenic mice lacking wild type profilin I. With a combination of biochemical studies on recombinant mutants in vitro, overexpression experiments in endothelial cells using replication incompetent adenoviruses and in vivo analysis in transgenic mice, we will gain much greater insight into the in vivo functions of profilin in mammalian cells and organisms.
