Improved pharmacokinetic and biodistribution properties of the selective urokinase inhibitor PAI-2 (SerpinB2) by site-specific PEGylation: implications for drug delivery.

PURPOSE: Overexpression of the serine protease urokinase (uPA) is recognised as an important biomarker of metastatic disease and a druggable anticancer target. Plasminogen activator inhibitor type-2 (PAI-2/SerpinB2) is a specific uPA inhibitor with proven potential for use in targeted therapy. However, PAI-2 is rapidly cleared via the renal system which impairs tumor uptake and efficacy. Here we aimed to improve the pharmacological properties of PAI-2 by site-specific PEGylation. METHODS: Several cysteine to serine substitution mutants were generated for PEGylation with PEG-maleimide (size range 12-30 kDa) and the physico-chemical and biochemical properties of the PEG-PAI-2 conjugates characterised. Radiolabeled proteins were used for evaluation of blood clearance and tissue uptake profiles in an orthotopic breast tumor xenograft mouse model. RESULTS: PEGylation of the PAI-2(C161S) mutant gave a predominant mono-PEGylated-PAI-2 product (~90%) with full uPA inhibitory activity, despite a significant increase in hydrodynamic radius. Compared to un-PEGylated protein the plasma half-life and AUC for PEG20-PAI-2(C161S) were significantly increased. This translated to a 10-fold increase in tumor retention after 24 h compared to PAI-2(C161S), an effect not seen in non-target organs. CONCLUSIONS: Our data underscores the potential for PEG20-PAI-2(C161S) drug conjugates to be further developed as anti-uPA targeted therapeutics with enhanced tumor retention.

Different radiolabelling methods alter the pharmacokinetic and biodistribution properties of plasminogen activator inhibitor type 2 (PAI-2) forms.

INTRODUCTION: Tumour-associated urokinase plasminogen activator (uPA) is a critical marker of invasion and metastasis, and it is recognised as having strong prognostic relevance as well as being a therapeutic target. The specific uPA inhibitor plasminogen activator inhibitor type-2 (PAI-2, SerpinB2) specifically targets cell bound uPA and is internalised. Furthermore, preclinical studies have established the "proof-of-principle" of uPA-targeting by PAI-2-cytotoxin conjugates in human carcinoma models. However, these studies also suggest that PAI-2 is rapidly cleared via the renal system with low total dose reaching the tumour. In this study, a comparative single photon emission computed tomography (SPECT) and biodistribution (BD) analysis of different forms of PAI-2 labelled with the radioisotopes iodine-123 ((123)I) and technetium-99m ((99m)Tc) was undertaken. METHODS: The pharmacokinetic (PK) properties and BD of wild-type, ΔCD-loop and PEGylated ΔCD-loop PAI-2 labelled with the commonly used diagnostic SPECT radioisotopes (99m)Tc or (123)I were compared in mouse models of human prostate carcinoma. Whole body SPECT imaging was also performed. RESULTS: Both wild-type and the shorter but active ΔCD-loop form of PAI-2 (123)I-labelled indirectly via conjugation to free amine groups (termed (123)I-Bn-PAI-2) exhibited low tumour uptake, rapid excretion and similar PK profiles. Preliminary studies with a short branched-chain PEGylated (123)I-Bn-PAI-2 ΔCD-loop indicated an increase in blood retention time and tumour uptake. All (123)I-Bn-labelled radiotracers were largely excreted through the kidneys. By comparison, both wild-type (123)I-PAI-2 (labelled directly via tyrosine residues) and (99m)Tc-PAI-2 displayed different PK/BD patterns compared to (123)I-Bn-PAI-2, suggesting greater liver based catabolism and thus slower elimination. SPECT imaging mimicked the BD results of all radiotracers. CONCLUSION: The different labelling methods gave distinct PAI-2 BD and tumour uptake profiles, with radioiodination resulting in the best non-tumour organ clearance profiles. Preliminary analyses with short branched-chain PEGylated (123)I-Bn-PAI-2 ΔCD-loop suggest that further investigations with other PEGylation reagents are required to optimise this approach for tumour imaging. These findings impact on the use of PAI-2 for drug delivery and/or diagnostic development.

Pharmacokinetics and toxicity of (213)Bi-labeled PAI2 in preclinical targeted alpha therapy for cancer.

OBJECTIVES: The plasminogen activator inhibitor type 2 (PAI2) when labelled with (213)Bi forms the (213)Bi-PAI2 alpha conjugate (AC). This AC has been shown to be efficacious in preclinical studies with breast, ovarian, prostate and pancreatic cancers. The objectives of this study were to investigate the pharmacokinetics and in vivo stability of (213)Bi-PAI2 in mice, its toxicity in mice and rabbits; and to determine whether a prior injection of a metal chelator (Ca-DTPA) or lysine can reduce toxicity by decreasing renal uptake. METHODS: Two chelators (CHX-A"-DTPA and cDTPA) were used for preparation of the (213)Bi-PAI2 conjugate, for intraperitoneal administration in mice and ear vein injection in rabbits. The mice were sacrificed at different time points for pharmacokinetic studies. Blood and organs were collected for toxicity studies for all groups. RESULTS: Both chelators were found to have similar %ID/g in the kidneys over four hours. Mice and rabbits did not show any short term toxicity over 13 weeks at 1420 MBq/kg and 120 MBq/kg (213)Bi-PAI2 respectively. Kidney uptake was decreased three fold by lysine. Radiation nephropathy was observed at 20-30 weeks in mice, leading to severe weight loss, whereas severe and widespread renal tubular necrosis was observed at 13 weeks in rabbits. CONCLUSIONS: Radiation nephropathy is the dose limiting toxicity observed in mice and rabbits. Lysine can reduce kidney uptake by three fold. Based on long-term monitoring, the maximum tolerance doses (MTD) are 350 and 120 MBq/kg for mice and rabbits respectively.

Kinetic analysis of plasminogen activator inhibitor type-2: urokinase complex formation and subsequent internalisation by carcinoma cell lines.

The overexpression of urokinase (uPA), which plays a key role in tumour invasion and metastasis, is an established prognostic marker and potential therapeutic target. Plasminogen activator inhibitor type 2 (PAI-2), an efficient and specific inhibitor of uPA, has been shown to selectively deliver potent cytotoxins to tumour cells. However, a direct quantitative analysis of both the inhibition kinetics and subsequent fate of PAI-2 upon interaction with cell-surface uPA has not been previously undertaken. In this study, we analysed specific PAI-2 binding to receptor-bound uPA on human breast and prostate cancer cell lines to directly measure inhibition kinetics. Cell-surface uPA:PAI-2 complex formation, which is reflective of complete uPA inhibition, was found to be very efficient (inactivation constant [K(I)] = 60-80 pM, depending on cell line used) and rapid (inactivation rate constant [k(inact)] = 0.32-0.47 min(-1) at 37 degrees C, depending on cell line used). To directly quantify and visualise cellular internalisation and localisation, we developed a novel assay based on the use of PAI-2 labelled with Alexa(488) fluorochrome and a polyclonal antibody to quench Alexa(488) fluorescence. The efficient and rapid formation of uPA:PAI-2 complexes was thus shown to be associated with specific and rapid internalisation of PAI-2, which could be localised within endosomes and lysosomes. PAI-2 was subsequently degraded, presumably within lysosomes. This study is the first to provide definitive evidence for uPA/uPAR-mediated PAI-2 endocytosis.

Increased permeability of psoriatic skin to the protein, plasminogen activator inhibitor 2.

The penetration and permeation of the recombinant protein plasminogen activator inhibitor type 2 (PAI-2) in two formulations, one containing a penetration enhancer, into the psoriatic and uninvolved skin of eight patients with plaque-type psoriasis were investigated. Penetration and permeation of PAI-2 were measured by gamma counting and imaging following radiolabelling of a fraction of the applied PAI-2 with (123)I. The feasibility of topical delivery of drug to psoriatic plaques was confirmed by the finding that the permeability of psoriatic plaques to radiolabelled PAI-2 (P=0.007) and free (123)I (P=0.001) was approximately tenfold higher than the permeability of uninvolved skin. The addition of a penetration enhancer improved the permeation of PAI-2 into psoriatic plaques from an average of 35% to 46% (P=0.005). Occlusion decreased the permeation amount of PAI-2 from 46% to 15% due to losses on the occlusive dressing (P=0.001).