Since tumor growth factor beta (TGF- ) and its receptor are ubiquitously expressed and because latent TGF- cannot bind to the cell surface receptor, the ability of a cell to activate latent TGF- upon secretion represents an important regulatory mechanism of TGF- action. In vivo, the protease plasm in is considered to be one of the main enzymes operative in the proteolytic cleavage of the latency-associated peptide moiety from TGF- , which converts it into the biologically active form. The TGF- response was characterized in alveolar macrophages during pulmonary inflammation (d 3) and fibrosis (d 120) induced by a single intratracheal instillation of silica particles ( 5 mg/ mouse). To appreciate the role of urokinase-type plasminogen activator (uPA) in the activation of TGF- , the production of total, active and latent TGF- by explanted alveolar macrophages was compared in uPA-deficient (uPA-/-) mice and their normal counterparts (uPA+/ +) . At d 3 and 120 after silica treatment, a significant increase in cell-associated PA activity was found in uPA+/ + mice compared to that of saline controls. As expected, this response was almost totally absent in uPA-/- mice. Alveolar macrophages from uPA+/ + controls were found to release TGF- mainly expressed in a biologically active form. In response to silica treatment, inflammatory cells were found to upregulate, especially at the fibrotic stage, their secretion of total and bioactive TGF- . No significant difference was found between uPA-/- and uPA+/ + silica-treated animals for the expression of total, active, or latent TGF- . Although it has previously been reported that macrophage surface activation of TGF- is dependent on both plasmin generation and uPA cell surface receptor, no evidence was found to support this hypothesis in the present study.