Defective processing of the transforming growth factor-beta1 in azoxymethane-induced mouse colon tumors

High levels of the cell growth inhibitor transforming growth factor-beta1 (TGF-beta1) are often found in a variety of human cancers. However, the physiological significance of this overexpression depends on the availability of the biologically active form of TGF-beta1 within the extracellular matrix of the tumor microenvironment. To determine the expression and activation status of TGF-beta1 in chemically induced tumors, 6-wk-old A/J mice were injected intraperitoneally with either azoxymethane (AOM) (10 mg/kg body weight, once a week for 6 wk) or normal saline solution, and colon tumors were isolated 24 wk following the last injection. An enzyme-linked immunosorbent assay for TGF-beta1 revealed a significant increase (1.7-fold, P < 0.05) in total TGF-beta1 protein in tumors. Interestingly, while 80% of the total TGF-beta1 in the control colon tissues was in the active form, only 50% was found to be active in tumors. Together with our earlier observations that TGF-beta1 mRNA levels are unchanged in A/J tumors, these data further support a mechanism whereby elevated TGF-beta1 levels result from a defective activation and turnover of this protein. Because plasmin is known to be a major activator of TGF-beta1 in vivo, we hypothesized that reduced plasmin activity may be responsible for the observed dysregulation of TGF-beta1 processing in these behaviorally benign tumors. With a fluorogenic peptide substrate for serine proteases, a deficiency in plasmin activity was found in the tumors. Furthermore, semiquantitative reverse transcription (RT)-polymerase chain reaction (PCR) analysis of a panel of genes involved in the plasminogen activation system, including plasminogen activator inhibitor-1 (PAI-1), urokinase-plasminogen activator (u-PA), and urokinase-receptor (u-PAR-1), demonstrated a significant upregulation (approximately fourfold to sixfold, P < 0.05) in the expression of each of these genes in the tumor tissue. In addition, no significant changes were observed in the expression levels of thrombospondin-1 (TSP-1) and insulin-like growth factor type II receptor (IGF-IIR), which also mediate the activation of latent TGF-beta1. To gain further insight into the functionality of the TGF-beta1 pathway, cDNA microarrays were performed and the expression levels of a panel of 21 TGF-beta1-specific target genes were determined in AOM-induced tumors that overexpress the ligand. A significant dysregulation in the expression of each of these targets was observed, providing evidence of aberrant TGF-beta1 signaling in tumors. Overall, the present study demonstrates a very low plasmin activity in A/J colon tumors, possibly as a result of the potent inhibitory effect of PAI-1 on the plasminogen activation cascade. The observed deficiency in plasmin activity may not be sufficiently compensated for by other mechanisms of latent TGF-beta1 activation, including TSP-1 and IGF-IIR, thereby resulting in a decreased fraction of the biologically active form of TGF-beta1 and subsequent aberration in TGF-beta1-specific gene regulation in A/J tumors.