Species and glycosylation dependency of PAI-1 inhibition

Cardiovascular diseases can result from an imbalance between plasminogen activators (mainly t-PA) and plasminogen activator inhibitors (mainly PAI-1). Restoration of this imbalance in the plasminogen activator system is possible via two approaches i.e. either by increasing the levels of circulating t-PA or by decreasing the levels of active PAI-1 through PAI-1 inhibition. Current thrombolytic therapy is mainly focused on increasing the circulating t-PA concentration but it remains associated with severe side effects such as bleeding complications. Co-administration of a PAI-1 inhibitor together with current thrombolytic drugs might help to reduce these complications and improve the thrombolytic efficacy and treatment outcome.Besides its role in thrombotic disorders plasma PAI-1 levels are also indicative for poor prognosis in cancer and are highly elevated in some non-thrombotic conditions, like obesity and the metabolic syndrome. Also for these pathologies, interference with PAI-1 activity might have a positive impact on disease progression and prognosis.A lot of research has focused on the development of PAI-1 inhibitors, yet to date no commercial PAI-1 inhibitor is available. Although the role of PAI-1 in cardiovascular events is well documented, its role in non-thrombotic conditions such as cancer definitely requires further validation before potential (pre-)clinical studies with a PAI-1 inhibitor. Preclinical evaluation of putative therapeutic targets in vivo is often started in rodent models since disease pathogenesis in mice or rats tends to approximate disease progression in humans. However, most animal proteins differ substantially from their human counterparts. As a result, PAI-1 inhibitory antibodies generated toward human PAI-1 often do not cross-react with PAI-1 from other species. In this study we describe the generation and in vitro and in vivo characterization of a new panel of monoclonal antibodies reacting with vitronectin-bound glycosylated mouse PAI-1. These antibodies provide interesting tools to further study the complex role of PAI-1 in different thrombotic and non-thrombotic mouse models. Comparative evaluation of the reactivity of PAI-1 inhibiting monoclonal antibodies toward rat and mouse PAI-1 wild-type and glycosylation knock-out mutants revealed a clear glycosylation dependent reactivity profile. Electrophoretic mobility of mouse PAI-1 glycosylation knock-out mutants before and after deglycosylation indicates the presence of glycan chains at position N265, whereas N209 and N329 do not seem to be glycosylated. For rat PAI-1, N-glycosylation is observed at asparagine N65 and at asparagine N265. These data stress the importance of selecting the proper source of PAI-1 when evaluating PAI-1 inhibitors in vitro, since reactivity of inhibitors toward putative therapeutic targets is species and glycosylation dependent. In the last part of the study we compared the functional properties of stabilized glycosylated versus non-glycosylated human PAI-1. A stabilized form of human PAI-1, comprising five point mutations, was expressed in eukaryotic cells and the stability of this purified protein was compared with the stability of the non-glycosylated human PAI-1 mutant. In conclusion, this study provides good immunological tools to further unravel the complex role of PAI-1 in various physiological processes using mouse models and adds to our knowledge about the biochemical and functional properties of glycosylated versus non-glycosylated PAI-1 from different species.status: publishe