PCSK-generated cleavages can either activate the cognate precursor by releasing bioactive products or inactivate it by removing bioactive moieties. Substrates of proprotein convertases (PCs) are precursors of hormones, growth factors, receptors, transcription factors, surface glycoproteins, etc.

The kexin-like basic amino acid (aa)-specific proprotein convertases, pyrolysin-like subtilisin kexin isozyme 1 (SKI 1; encoded by the MBTPS1 gene) and proteinase K like proprotein convertase subtilisin kexin 9 (PCSK9) are individually grouped to emphasize their distinct subclasses. The various domains and N glycosylation positions are emphasized, along with the primary (depicted using light grey arrows, and a light grey double arrow for SKI 1) as well as the secondary autocatalytic processing sites (depicted using dark grey arrows). Note that proprotein convertase 4 (PC4), PC7 and PCSK9 have only a primary site and that only PC5 exhibits two validated alternatively spliced forms, namely PC5A and PC5B. The membrane-bound human proprotein convertases include furin, PC4, PC5B, PC7 and SKI 1. The presence of a signal peptide, a prosegment and catalytic domain is common to all convertases that exhibit the typical catalytic triad residues Asp, His and Ser, as well as the Asn residue comprising the oxyanion hole (Asp for PC2). Downstream of the catalytic domain, all the basic amino acid-specific convertases, except for SKI 1 and PCSK9, exhibit a β barrel P domain that apparently stabilizes the catalytic pocket. The carboxy terminal domain of each convertase contains unique sequences regulating their cellular localization and trafficking. PC5 and paired basic amino acid cleaving enzyme 4 (PACE4) contain a specific Cys-rich domain, which binds to heparan sulphate proteoglycans at the cell surface and the extracellular matrix. By contrast, PCSK9 exhibits a Cys-His-rich domain (CHRD) that is required for the trafficking of the PCSK9–LDLR (low-density lipoprotein receptor) complex to endosomes and lysosomes. 

Discovery of proprotein convertases (1990-2003).  Nabil G. Seidah identified & cloned 7/9 members of the PC family. PC1, PC2 (DNA 1990; Mol Endo 1991); PC4 (Mol Endo 1992); PC5 (PNAS 1993); PC7 (PNAS 1996); SKI-1 (PNAS 1999) & PCSK9 (PNAS 2003). Furin (1990) & PACE4 (1991) were identified by Dutch and American groups, respectively. This was a truly enormous contribution in defining the molecular mechanisms underlying the cellular generation of active peptides/proteins from secretory precursors

Upon exiting the endoplasmic reticulum (ER), most of the basic amino acid-specific proprotein convertases traverse the Golgi apparatus towards the trans-Golgi network (TGN). At this branching point the proprotein convertase 1 (PC1) and PC2 are sorted to immature secretory granules where they are activated, cleave polypeptide hormones and are stored in dense-core secretory granules (SGs) awaiting signals for regulated secretion. Conversely, the constitutively secreted furin, PC5 and paired basic amino acid cleaving enzyme 4 (PACE4) reach the cell surface from the TGN. PC7 is able to reach the same destination from both the TGN and by an unconventional secretion pathway directly from the ER. Upon binding to heparan sulphate proteoglycans (HSPGs) and tissue inhibitors of metalloproteinases, PC5A and PACE4 are retained at the cell surface and/or extracellular matrix. The membrane-bound furin, PC5B and PC7 are recycled through endosomes back to the TGN. The activated membrane-bound subtilisin kexin isozyme 1 (SKI 1) is mostly concentrated in the cis- and medial-Golgi, from where it is then sent to lysosomes for degradation, and does not normally reach the cell surface. Proprotein convertase subtilisin kexin 9 (PCSK9) is secreted from the TGN directly into the medium as an enzymatically inactive non-covalent complex of the protease and its prosegment. Upon binding to the low-density lipoprotein receptor (LDLR) at the cell surface, the PCSK9–LDLR complex is internalized into endosomes and then sent to lysosomes for degradation. 

The unexpected identification of 7B2 (FEBS Lett & Arch Biochem Biophys 1982/83), a 186 aa protein found in dense core granules of neural & endocrine cells, was made 13 years before the discovery of its major function, i.e., to chaperone the functional activity of PC2 (FEBS Lett 1995). Cloning of 7B2 mRNA & the KO of its gene revealed PC2-related functions (Biochem J 2001) & novel ones important in the regulation of aggressive behavior (JBC 2013). Exit of pro7B2 from the endoplasmic reticulum (ER) and its processing in the Trans Golgi Network (TGN) by furin to generate a C-terminal peptide (CT) that inhibits the activity of PC2.  The CT-peptide is then inactivated by PC2 in immature secretory granules (ISG). 

The seminal discovery of β-endorphin, its sequence & morphinomimetic activity (Lancet, 1976; Can J Biochem & BBRC 1976), & the processing of the precursors POMC & proInsulin were landmarks in peptide hormones & the generation of multiple functional peptides (ACTH, α,β,γ-MSHs, β-endorphin; Insulin) via limited proteolysis at specific basic residues (PNAS 1978,1979; Int J Pept Protein Res 1984) & a harbinger for the future discovery of the cognate proteinases, the proprotein convertases (PCs).

Autoradiograms show the expression of PC7 in brain regions, pituitary, digestive system, reproductive tissues, and various peripheral organs. Total RNA (5 mg) was loaded into each lane. The major form of PC7 observed is 4.2 kb. A band of lesser intensity is also sometime observed migrating at 6.5 kb. The rPC7 complementary RNA probe used was 285 bp containing nt 982–1266. The x-ray films were exposed for 3 days.

In situ hybridization histochemistry of PC7 mRNA distribution in the rat brain (A), pituitary (B), thymus (C), cerebellum (D), testis (E), and spleen (F). X-ray film exposures were for 4–7 days. The antisense 742-nt rPC7 complementary RNA probe represents nt 2170–2911. For controls, adjacent sections for each of these tissues were hybridized with the sense-strand PC7 probe of equal length and specific activity. No hybridization signal was observed in any of these controls (data not shown). Arc, arcuate nucleus of the hypothalamus; DMH, dorsomedial hypothalamic nucleus; VMH, ventromedial hypothalamic nucleus; DG, dentate gyrus; Hb, habenular nucleus; AL, anterior lobe of the pituitary; NL, neural lobe of the pituitary; IL, intermediate lobe of the pituitary; GL, granular layer of the cerebellum; C, cortex of the thymus; M, medulla of the thymus; WP, white pulp of the spleen. A, 32.7; B, 36.5; C, 32.5; D, 33.0; E, 32.9; F, 32.5. 

The V5-tagged PCSK9 or its active site mutant H226A were transiently expressed in HEK293 cells. The next day the cells were pulse-labeled with [35S]Met/Cys for 4 hours. Cell extracts (C) and media (M) were immunoprecipitated with a V5 antibody and the precipitates were & by SDS/PAGE on an 8% tricine gel. The migration positions of molecular mass standards (kDa), proPCSK9, PCSK9 and the prosegment are emphasized, together with the furin cleaved product observed. The active site mutant is neither cleaved not secreted, emphasizing the necessity of prodomain autocatalytic cleavage for PCSK9 to exit the ER and be secreted. 

Schematic representation of proPCSK9 (75 kDa, 692 aa) and its processed form PCSK9 (62 kDa) that is non-covalently bound to the autocatalytically processed prodomain /    (pro; 15 kDa) at VFAQ152-SIP. Notice the active site residues Asp, His, and Ser and the oxyanion hole Asn, as well as the single N-glycosylation site (in blue).

High PCSK9 levels (left part of the graph) or GOF forms enhance the degradation of the LDLR following the endocytosis of the PCSK9–LDLR complex in clathrin-coated vesicles, together with the LDL, leading to the degradation of the PCSK9–LDLR complex in lysosomes. This results in low levels of the LDLR at the cell surface and increased levels of circulating LDL-c. Low levels of PCSK9 or LOF forms lead to increased levels of cell surface LDLR because the latter can be recycled back to the surface after delivery of LDL particles to acidic endosomes/lysosomes. PCSK9 mAbs target the extracellular pathway by sequestering extracellular PCSK9 and hence preventing its binding to the LDLR. 

Primary hepatocytes isolated from PCSK9 12-week-old male knockout mice were incubated with ~5 μg/ml of human purified PCSK9 for 5 h at 37 °C in the absence (top ) or presence (bottom) of 0.35 μM evolocumab, a PCSK9 mAb inhibitor. After the first 3 h of incubation, 5 μg/ml of the fluorescent DiI-LDL (red) was added to measure LDL internalization for the next 2 hours. After the 5 hours of incubation, the cells were fixed for analyses. The level of cell-surface LDLR (green) was measured by immunocytochemistry with an anti-mouse LDLR antibody. Nuclei are stained with DAPI (blue). Notice the increase in cell-surface LDLR immunolabeling (green) and DiI-LDL internalization (red) in the presence of the mAb. Scale bar, 25 μm.

The clinical fields relevant to each PC are indicated. To date, only PCSK9 is clinically targeted using mAbs. These mAbs are used for the treatment of hypercholesterolemia and are very promising for the prevention of septic shock. In this schematic, overlaps between PC clinical fields are only qualitative. Diabetes is related to both PC1/3 and PC2 through insulin processing. Cancers are mainly regulated by furin, PC5/6, and PACE4 (prostate) via processing of growth factors and their receptors. Most viral surface glycoproteins are processed by furin and, in some cases, by SKI-1/S1P (arenaviruses). Cholesterol and TG synthesis are regulated by SKI-1/S1P and LDL-c uptake by PCSK9. The convertase PC4 has a unique role in fertility, and PC7 in iron metabolism and anxiety/depression. Finally, in humans, PC1/3, PC5/6, and PACE4 were genetically linked to obesity, developmental defects, and osteoarthritis pain, respectively.