Proprotein convertases in homeostasis and pathophysiology; perspectives of targeting furin-mediated proprotein processing to affect tumor development involving IGF axis signalling

Zhu, Jingjing [UCL] Vermorken, Alphons JM [KUL] Van de Ven, Wim JM [KUL]

The family of mammalian subtilisin-like serine endoproteases or proprotein convertases (PCs) has gained increasing interest because of their critical role in mammalian biology and their involvement in a wide array of pathophysiological conditions. Consequently, therapeutic targeting of these enzymes for treatment is becoming a focus of intensive research. Post-translational processing of precursor proteins and proproteins into biochemically active molecules is essential for appropriate functioning of cells and constitutes an important mechanism to preserve homeostasis in complex organisms. An important family of processing enzymes is the family of PCs, which in mammals consists of seven structurally related enzymes. They possess the capability to bioactivate inactive precursor proteins into their physiologically active products, often components of critical cellular pathways, via cleavage at sites consisting of multiple basic amino acid residues. The prototype of this family is furin, which is ubiquitously expressed and processes constitutively secreted proteins. The (neuro)endocrine-specific PCSK1 (also known as PC1/3) and PCSK2 (PC2) are involved in the generation of peptide hormones and neuropeptides in the regulated secretory pathway. PCSK4 (PC4) is implicated in testicular and ovarian physiology. PCSK5 (PC5/6) and PCSK6 (PACE4) bind heparin sulphate proteoglycans and are involved in regulating body axis and polarity determinants during biological development. PCSK7 (PC7) exhibits critical functions in the brain. Collectively, the PCs bioactivate a wide variety of growth and differentiation factors, receptors, a broad spectrum of neuropeptides and peptide hormones, blood coagulation factors, adhesion proteins, enzymes, bacterial toxins and viral coat proteins. Bioactivation occurs in various cellular compartments such as the endoplasmic reticulum, the trans-Golgi network (TGN), secretory vesicles, and endosomes. Since some PCs are also shed from cells, it is tempting to speculate that these PCs also exhibit extensive paracrine processing and signalling functions. Because of the nature of their substrate spectrum, these enzymes are involved in numerous physiologically critical processes and they have also been associated with a wide and evolving array of pathological conditions including neurodegenerative diseases, infectious diseases, obesity, osteoarthritis, inflammatory diseases such as atherosclerosis, endocrine pathologies, breakdown in peripheral immune tolerance, and neoplastic diseases. In light of this all, it is not surprising that PCs are being considered as potential targets for future therapeutic interventions in clinical settings. Studies have clearly linked PCs to the processing of a variety of proproteins involved in various stages of neoplastic diseases, such as tumor cell proliferation, blockage of apoptosis, invasion, angiogenesis, phenotypic transition into malignancy, and acquisition of metastatic capabilities. They even have been referred to as “master switches” in the regulation of tumor growth and progression. Consequently, therapeutic targeting of PCs is likely to affect the tumorigenic process at various levels simultaneously. Such synergistic effects might enhance the effectiveness of such a therapeutic approach. However, many unresolved questions remain about the feasibility of specific and effective therapeutic targeting of PCs. Moreover, in light of the involvement of the PCs in a wide array of physiologically critical molecular mechanisms and in the preservation of delicate balances of homeostatic processes in complex organisms, severe side effects might be anticipated. On the other hand, the distinct but in some cases also the overlapping spatiotemporal expression patterns of PCs point towards the phenomenon of redundancy and this may attenuate possible side effects when one member of the family is specifically targeted. In any case, PC targeting should be explored and validated in animal models since spatiotemporally complex proprotein processing profiles and also the cellular microenvironment constitute important parameters in this context. In that perspective, an appropriate and versatile tumor model for studying therapeutic targeting of furin has been developed, i.e. the PLAG1 mouse tumor model system. The PLAG1 oncogene was first found to cause human pleomorphic adenomas of the salivary glands and, later on, it has also been implicated in various other human tumors, which points towards its broader relevance in tumorigenesis. It encodes a zinc finger transcription factor which transactivates numerous genes among them several growth factors and receptors that require PC-mediated cleavage for their bioactivation. Signalling via the IGF and WNT axes has been implicated in PLAG1-induced tumorigenesis. The fact that IGF-1R, which is involved in several human oncogenic transformation processes and which is unquestionable the predominating single factor in the IGF signalling axis, is bioactivated via furin-mediated processing of the corresponding proprotein pro-IGF-1R. This suggests that furin could be critically involved in PLAG1-induced tumorigenesis. Indeed, IGF-1R-deficient cells cannot be transformed by oncogenic PLAG1 and also subsequent genetic studies with the above-cited PLAG1 mouse tumor model seem to provide further substantiation for this possibility. Altogether, these observations indicate that furin is a potential target for affecting PLAG1-induced tumor development in this tumor model system, and presumably also in the many human tumors involving IGF-1R. The PLAG1 mouse tumor model seems, therefore, to constitute a suitable animal model to test this. The challenge is now to select potent inhibitory agents that efficiently and specifically affect the processing activity of furin and to develop concepts and strategies that enable such inhibitory agents to specifically target tumor cells only.