Transforming growth factor-beta1 (TGFbeta1) is a widely-expressed cytokine that has major effects on most cell types. TGFbeta1 is anti- inflammatory, pro-fibrotic, and casually linked to fibrotic diseases, e.g. pulmonary fibrosis. TGFbeta1 is secreted in an inactive complex (FTGFbeta1) with its pro-peptide dimer, which is called latency- associated (LAP). Activation of LTGFbeta1 is a key control point in TGFbeta1 biology, but is poorly understood. Only thrombospondin-1 (TSP1) has been shown previously to active PTGFbeta1 in normal animals. We found that LAP is a ligand for the epithelium-specific integrin alphavbeta6, and that cells expressing alphavbeta6 bind and activate latent TGFbeta1. This mechanism can explain the heretofore puzzling phenotype of beta6 integrin knock-out mice: inflammation in lung and skin, and protection from bleomycin-induced pulmonary fibrosis. Our results provide the first evidence that dysregulated TGFbeta1 activation causes fibrosis. Our goals are to understand quantitatively the interactions between LTGFbeta and alphavbeta6 that lead to activation, and to develop an animal model an animal model and knowledge to explore fully the biological role of a of avbeta6-mediated LTGFbeta1 activation.
In Aim 1 we will analyze the activation mechanism by focusing on alphavbeta6- LTGFbeta1 interactions. We will make TGFbeta1-mull alphavbeta6- expressing cells to which specifically engineered forms of LTGFbeta1 will be added (either by transfection or as recombinant protein). In this system we will then determine the relative activatability of two major forms of LTGFbeta1 (the so-called small and large latent complexes), the relative effects of LTGFbeta1 concentration and alphavbeta6 expression levels of activation, the activatability of soluble and matrix-bound latent TGFbeta1, and the integrin: LTGFbeta1 stoichiometry required for activation. Also, we will assess the influence of integrin/LTGFbeta1 binding affinity on activation. The results will be incorporated into an activation model and related to activation in vivo.
In Aim 2, we will create a mouse expressing a mutant form of LTGFbeta1 that cannot be activated by integrins (the RGD integrin binding sites in LAP will be mutated to RGE). The phenotypes of these mice and beta6 integrin null mice will be compared to confirm that the beta6 null phenotype is due specifically to loss of TGFbeta1 activation. To interpret the phenotype will test the ability of other RGD-binding integrins to activate LTGFbeta1. Finally, we will cross RGE-TGFbeta1 mice with TSP1 null mice to assess the summed effects of the two currently known TGFbeta1 activation mechanisms, namely alphavbeta6 and TSP1. The results of these aims will lead to better understanding of alphavbeta6-mediated TGFbeta1 activation in disease.
