TGFbeta, which exists in 3 isoforms, is a pleiotropic cytokine involved in a variety of lung processes and diseases, e.g., development, pulmonary fibrosis, tuberculosis, asthmatic airway remodeling and cancer. TGFbeta bioavailability is tightly regulated in the extracellular space: TGFbeta is secreted in a latent form, which is sequestered in matrix binding sites by the action of an adapter protein (LTBP), and must be activated before it can signal via TGFbeta receptors. TGFbeta latency occurs as a result of binding to the TGFbeta propeptide Latency-Associated Peptide (LAP), and activation involves releasing TGFbeta from LAP. Because all TGFbeta signaling occurs subsequent to TGFbeta activation, the activation mechanisms are critical control points. Mechanisms of TGFbeta activation have been proposed (e.g., thrombospondin-1, proteases and oxidants), but their roles in vivo are unclear. We showed that an epithelium-specific integrin, alphaVbeta6, binds latent TGFbeta1 at an RGD sequence within LAP. Cells expressing alphaVbeta6 activate TGFbeta1, and mice lacking alphaVbeta6 do not develop bleomycin-induced pulmonary fibrosis. We have also shown that alphaVbeta6 can activate TGFbeta3, and that another integrin expressed in lung epithelium (alphaVbeta8) can activate TGFbeta1/3. The goal of this work is to define the role of integrin-mediated TGFbeta1 activation in vivo. To that end, we have generated knockin mice with a mutation in the TGFbeta1 gene that encodes a nonfunctional integrin-binding site in the LAP propeptide. The phenotype of these mice (designated TGFbeta1-RGE) reproduces the most prominent aspect of the TGFbeta1-/- phenotype: severe, multiorgan inflammation that causes death at 3-4 weeks of age. This important finding suggests the possibility that integrin-mediated TGFbeta1 activation occurs upstream of most TGFbeta1 effects.
The specific aims of this work are to (1) characterize the TGFbeta1-RGE mice in comparison to TGFbeta1-/- mice, (2) test the hypothesis that the only TGFbeta1-activating integrins in vivo are alphaVbeta6 and alphaVbeta8, and (3) test the hypothesis that the phenotype of certain integrin KO mice (alphaV-/- and (beta8-/-) is due to combined lack of TGFbeta1 and TGFbeta3 activation. These studies will further delineate the role of RGD-binding integrins in TGFbeta biology and provide a basis for important studies of their role in lung pathology.
