Cleavage of the Glycoprotein of Arenaviruses
- 340 Downloads
The arenaviruses are a large family of emerging negative-stranded RNA viruses that include several severe human pathogens causing hemorrhagic fevers with high mortality. During the arenavirus life cycle, processing of the viral envelope glycoprotein precursor (GPC) by the cellular subtilisin kexin isozyme-1 (SKI-1)/site-1 protease (S1P) is crucial for productive infection. The ability of newly emerging arenaviruses to hijack human SKI-1/S1P is a key factor for zoonotic transmission and human disease potential. Apart from being an essential host factor for arenavirus infection, SKI-1/S1P is involved in the regulation of important physiological processes and linked to major human diseases. This chapter provides an overview of the mechanisms of arenavirus GPC processing by SKI-1/S1P including recent findings. We will highlight to what extent the molecular mechanisms of SKI-1/S1P cleavage of viral GPC differ from processing of SKI-1/S1P’s cellular substrates and discuss the implications for virus-host interaction and coevolution. Moreover, we will show how the use of the viral GPC as a “molecular probe” uncovered novel and unusual aspects of SKI-1/S1P biosynthesis and maturation. The crucial role of SKI-1/S1P in arenavirus infection and other major human diseases combined with its nature as an enzyme makes SKI-1/S1P further an attractive target for therapeutic intervention. In the last part, we will therefore cover past and present efforts to identify specific SKI-1/S1P inhibitors.
KeywordsOld World arenaviruses New World arenaviruses Lassa virus Machupo virus Junin virus Glycoprotein Proprotein convertases Proteolytic processing Maturation SKI-1/S1P Drug screening SKI-1/S1P sensor
3.1 Arenavirus Structure, Genome Organization, and Basic Virology
The arenaviruses are a large and diverse family of emerging enveloped negative-stranded viruses that include several severe human pathogens (Buchmeier et al. 2007). The Arenaviridae family has been recently separated by the International Committee on Taxonomy of Viruses into the genus Mammarenavirus and the genus Reptarenavirus (Radoshitzky et al. 2015). Based on phylogenetic and serological data, the mammarenaviruses are divided into two major groups: the Old World and the New World complex. For simplicity, we will henceforth use the term “arenaviruses” synonymous for the entire family and members of the Mammarenavirus genus , whereas viruses of the Reptarenavirus genus will be specifically referred to. The Old World arenavirus lineage contains the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV) with worldwide distribution. The infection of LCMV in the mouse represents one of the most powerful models in experimental virology and immunology (Oldstone 2002). LCMV is further a relevant human pathogen in pediatric and transplantation medicine (Bonthius 2009; Palacios et al. 2008). The highly pathogenic Lassa virus (LASV) is endemic in Western Africa (McCormick and Fisher-Hoch 2002), and Lujo virus (LUJV) recently emerged in Southern Africa associated with a cluster of fatal infections (Briese et al. 2009). The African arenaviruses Mopeia, Mobala, and Ippy virus have so far not been associated with human disease. The New World arenaviruses are divided into Clades, A, B, C, and D, the latter corresponding to former Clade A/B or A/rec (Radoshitzky et al. 2015). Clade B contains the human pathogenic Junin (JUNV), Machupo (MACV), Guanarito (GTOV), Sabia (SABV), and Chapare (CHAV) virus, together with the nonpathogenic Tacaribe (TCRV), Amapari, and Cupixi virus. In nature each arenavirus species has one or a limited number of closely related rodent species as reservoirs that are persistently infected, with the exception of TCRV that was isolated from bats (Buchmeier et al. 2007) and recently detected in host-seeking Amblyomma americanum ticks (Sayler et al. 2014). The current phylogenetic diversity of arenaviruses is likely the result of long-term coevolution between viruses and their host species, involving vertical and horizontal transfer of viruses within and between populations (Emonet et al. 2009).
Several excellent reviews cover different steps of the arenavirus life cycle (Fehling et al. 2012; Grant et al. 2012; Urata and de la Torre 2011; Emonet et al. 2011; Torriani et al. 2017; Nunberg and York 2012; Wolff et al. 2013; Loureiro et al. 2012), and only a short summary will be given here. The first step of arenavirus infection requires attachment of the viral particle to cellular receptor(s). Most Old World and Clade C New World arenaviruses use dystroglycan, a ubiquitously expressed receptor for proteins of the extracellular matrix (ECM) as a high-affinity receptor (Cao et al. 1998; Oldstone and Campbell 2011). The cellular receptor for the pathogenic Clade B New World arenaviruses was identified as human transferrin receptor 1 (TfR1) (Radoshitzky et al. 2007), a highly conserved cargo receptor involved in iron metabolism. The ability of a Clade B New World arenavirus to use human TfR1 is crucial for its potential to cause zoonotic infection and hemorrhagic fever in man, whereas nonpathogenic viruses use TfR1 orthologues from other species (Helguera et al. 2012; Radoshitzky et al. 2011). More recently, the Tyro3/Axl/Mer (TAM) receptor tyrosine kinases Axl and Tyro3/Dtk, T cell immunoglobulin mucin (TIM) proteins 1 and 4, as well as the C-type lectins DC-specific ICAM-3-grabbing nonintegrin (DC-SIGN) and LSECtin have been identified as novel candidate receptors for arenaviruses (Shimojima and Kawaoka 2012; Shimojima et al. 2012; Jemielity et al. 2013; Goncalves et al. 2013; Martinez et al. 2013). Upon initial attachment to the target cell, arenavirus particles are taken up by receptor-mediated endocytosis. Consistent with the use of TfR1 as a receptor, Clade B New World viruses enter via clathrin-mediated endocytosis (Martinez et al. 2007), whereas Old World arenaviruses use a pathway resembling macropinocytosis (Iwasaki et al. 2014; Oppliger et al. 2016; Torriani et al. 2017). The virus passes through the multivesicular endosome and reaches the late endosome (Pasqual et al. 2011b), where low pH triggers fusion of the viral membrane with the limiting membrane of the late endosome by the fusion-active GP2, creating a “fusion pore.” At the late endosome, LASV GP1 undergoes a unique “receptor switch” and engages the late endosomal/lysosomal resident protein LAMP1 for efficient fusion (Jae et al. 2014). The dependence of LASV, but not other arenaviruses including the closely related LCMV, on LAMP1 as a late endosomal entry factor represents an interesting analogy to the filoviruses Ebola virus, whose fusion depends on the late endosomal protein Niemann-Pick C1 (Jae and Brummelkamp 2015).
By an unknown mechanism of “uncoating,” the arenavirus ribonucleoprotein (RNP) comprised of viral RNA, NP, and L is released into the cytosol. Viral transcription is initiated at the incoming polymerase complex, resulting in expression of NP and L. As NP accumulates, the viral polymerase shifts to a replicase mode, generating full-length antigenomic RNAs serving as templates for the transcription of GPC and Z as well as synthesis of genomic RNA. Newly synthesized NP assembles the viral replication-transcription complexes that are membrane-associated structures that contain cellular lipids and proteins (Baird et al. 2012; Knopp et al. 2015). In the final stages of the arenavirus life cycle, progeny particles assemble and are released by budding from the plasma membrane. The key factor in the budding process is the small RING finger Z protein that functions as a bona fide matrix protein in arenavirus particle assembly (Urata and De La Torre 2011; Perez et al. 2003). As with other matrix proteins of enveloped viruses, arenavirus Z interacts with the cytosolic tail of GP2 (Capul et al. 2007) and specific cellular factors of the endosomal/multiple vesicle body pathway to drive the budding of viral particles from “budding zones” (Wolff et al. 2013; Urata and De La Torre 2011; Fehling et al. 2012; Perez et al. 2003).
Recent studies isolated, identified, and characterized novel and highly divergent arenaviruses from snakes associated with boid inclusion body disease (Bodewes et al. 2013; Hetzel et al. 2013; Stenglein et al. 2012). The genome organization of these viruses corresponds to arenaviruses, and they show a high degree of divergence. Notably, the GPC of the viral envelope seems more related to filoviruses. No cases of infections in other species have been reported so far, though reptarenaviruses are capable of infecting mammalian and arthropod cells in vitro (Hepojoki et al. 2015). Interestingly, reptarenavirus infection of mammalian cells occurred efficiently at 30 °C but was markedly reduced at mammalian body temperature, likely highlighting adaptation to their reptile hosts (Hepojoki et al. 2015).
The most prevalent human pathogen among the arenaviruses is the Old World arenavirus LASV that causes a severe viral hemorrhagic fever with high mortality in humans. Every year LASV causes over 300,000 infections in Western Africa (Mccormick and Fisher-Hoch 2002) and has been declared one of the eight top emerging pathogens by WHO in 2015 (Sweileh 2017). There is currently neither an efficient cure nor a licensed vaccine, resulting in case fatality rates of 15–30% (Yun and Walker 2012). In the USA, JUNV, MACV, GTOV, and SABV have emerged as causative agents of hemorrhagic fevers with high case-fatality rates with JUNV representing the most important public health problem (Grant et al. 2012). Novel arenaviruses emerge on the average every 2 years and can be associated with severe diseases (Briese et al. 2009; Delgado et al. 2008). Highly pathogenic arenaviruses have been included in the list of Category A pathogens by the Centers for Disease Control and Prevention. Global climate changes may influence population dynamics of natural rodent host populations, likely increasing exposure of humans to these pathogens.
The pathophysiology of fatal arenavirus infection is not well understood and involves viral and host immune factors (Yun and Walker 2012; Prescott et al. 2017). A highly predictive parameter for disease outcome is the viral load, indicating a close competition between viral spread and replication and the patient’s immune system (Prescott et al. 2017). Drugs targeting specific steps of the viral life cycle can reduce multiplication and spread of the virus. This may provide the patient’s immune system a window of opportunity to develop an antiviral immune response. An in-depth understanding of the molecular mechanisms underlying arenavirus multiplication and virus-host cell interaction is therefore of great importance to develop novel and efficacious strategies for antiviral therapeutic intervention.
3.2 Arenavirus GP Structure and Function
The arenavirus GPC is synthesized initially as a single polypeptide precursor that is sequentially cleaved by cellular signal peptidases and then by SKI-1/S1P (Lenz et al. 2001; Rojek et al. 2008; Beyer et al. 2003) (Fig. 3.1a, b). Processing of GPC by SKI-1/S1P yields the N-terminal GP1, which is implicated in binding to the cellular receptors (Borrow and Oldstone 1992) and the transmembrane GP2 that mediates fusion and resembles class I viral fusion proteins (Eschli et al. 2006; Igonet et al. 2011; Parsy et al. 2013). Arenavir us GPC contains a remarkably stable signal peptide (SSP) of 58 amino acids that contains two hydrophobic domains and undergoes myristoylation at its N-terminus (Eichler et al. 2003a, b; York et al. 2004; Froeschke et al. 2003). The SSP becomes part of a mature tripartite complex SSP/GP1/GP2 where it interacts with the GP2 subunit (Fig. 3.1b). Recent electron cryomicroscopy combined with tomography revealed that SSP/GP1/GP2 complexes of LASV assemble into a trimeric spike that is 9 nm high and 10 nm wide and undergoes significant changes when exposed to low pH (Li et al. 2016). Structural studies on the GP1 of LASV, MACV, and JUNV revealed a similar compact α/β fold, despite significant sequence deviation (Bowden et al. 2009; Cohen-Dvashi et al. 2015; Mahmutovic et al. 2015). Structural studies on the complex of MACV GP1 with its cellular receptor hTfR1 revealed that the GP1 monomer represents the functional unit of receptor recognition and that trimerization is not required for receptor binding (Abraham et al. 2010; Radoshitzky et al. 2011). Notably, MACV GP1 binds to the apical surface of hTfR1 without competing with transferrin binding. More recent crystall ographic studies resolved the structure of the pre-fusion conformation of the mature envelope GP of the prototypic Old World arenavirus LCMV (Hastie et al. 2016). Within the pre-fusion trimer, LCMV GP1 and GP2 undergo extensive interactions, involving ionic bonds. In contrast to the New World arenaviruses, monomeric LCMV GP1 is unable to bind the receptor dystroglycan with high affinity, suggesting that either avidity or the quaternary structure of the pre-fusion trimer is required. Once delivered to the late endosome, low pH sets off a series of conformational changes leading to shedding of GP1 and triggering of fusion of the viral and cellular membrane mediated by GP2 (Fig. 3.1c). The post-fusion conformation of arenavirus GP2 is similar to the six-helix bundle conformation common to a number of class I fusion proteins of enveloped viruses (Igonet et al. 2011; Parsy et al. 2013). The SSP is crucial for transport and processing of arenavirus GPC (Messina et al. 2012; York and Nunberg 2007; York et al. 2004; Eichler et al. 2003a, b). Both N- and C-termini of SSP are located in the cytosol (Agnihothram et al. 2007), and SSP associates non-covalently with a zinc-binding domain within the c ytoplasmic tail of GP2 (Agnihothram et al. 2006; Briknarova et al. 2011). The SSP-GP2 interactions critically modulate pH-induced activation of membrane fusion (York and Nunberg 2006, 2009) and are targeted by a range of potent arenavirus fusion inhibitors (Shankar et al. 2016; York et al. 2008), pinpointing this unique feature of arenavirus fusion as a target for the development of antiviral therapeutics.
3.3 The Proprotein Convertase SKI-1/S1P Cleaves Arenavirus GPC
A crucial step of arenavirus infection is the maturation of the envelope glycoprotein precursor GPC. With the exception of the Crimean-Congo hemorrhagic fever virus which belongs to the Bunyavirus family, mammarenaviridae are the only viral pathogens known to hijack the proprotein convertase SKI-1/S1P to process their envelope GP. Proprotein convertases (PC) are a family of nine conserved calcium-dependent serine endoproteases and include the basic PCs PC1/3, PC2, furin, PC4, PACE4, PC5/6, and PC7, as well as the n onbasic PCs SKI-1/S1P and PCSK9 (Seidah and Prat 2007, 2012). The PCs share homology to the kexin subfamily of subtilases with a distinctive “Ser/His/Asp” catalytic triad that mediates peptide bond scission (Seidah and Prat 2002). Furin, PC5/6B, PC7, and SKI-1/S1P are membrane-anchored, while the remaining enzymes are secreted (PC4, PC5/6A, PACE4, and PCSK9) or retained in granules (PC1/3, PC2) (Seidah 2011). Basic PCs have similar but not identical consensus sequences K/RXnR↓ that may result in overlapping patterns of substrate cleavage. In contrast, SKI- 1/S1P and PCSK9 cleave after hydrophobic or small residues, BX(hydrophobic)X↓ (Pasquato et al. 2006) and VFAQ↓, respectively (Benjannet et al. 2004). Processing by PC is essential for the proper function of a plethora of cellular proteins, including prohormones, growth factor precursors, transcription factors, proteases, and adhesion molecules.
3.4 The Mechanism of SKI-1/S1P Processing of Arenavirus GPC Differs from Cellular Substrates
Amino acid sequences at the GP1/GP2 cleavage sites of arenavirus GPC
Maturation of GPC by SKI-1/S1P is strictly required for the production of infectious particles and viral cell-to-cell spread (Beyer et al. 2003; Lenz et al. 2001; Rojek et al. 2008). A crucial role for SKI-1/S1P for arenavirus dissemination in vivo is further suggested by the observation that mice bearing the “wood rat” mutation Y496C in SKI-1/S1P show enhanced resistance to infection with LCMV due to impairment of GPC processing (Popkin et al. 2011). Accordingly, proof-of-concept studi es with protein- and peptide-based SKI-1/S1P inhibitors revealed that targeting GPC maturation represents a novel and promising antiviral strategy (Maisa et al. 2009; Rojek et al. 2010) as will be further developed below. Inhibition of SKI-1/S1P in infected cells results in the formation of noninfectious “naked” particles that contain viral RNP but lack GP (Lenz et al. 2001; Kunz et al. 2003; Rojek et al. 2008), indicating specific incorporation of fully mature, processed GP. How the arenavirus budding machinery is capable to achieve this specificity is currently unknown. Notably, SKI-1/S1P processing is not required for cell-surface transport of arenavirus GPC (Kunz et al. 2003; Schlie et al. 2010a), and small amounts of uncleaved GPC can be detected at the surface of infected cells (Kunz et al. 2003). However, in contrast to trimeric mature SSP/GP1/GP2 complexes, the uncleaved GPC forms monomers and oligomers spanning a wide size range, indicating that SKI-1/S1P processing is critical for the correct oligomeric state (Schlie et al. 2010a). Moreover, mutations in the cytosolic tail of LCMV and LASV GP2 affect SKI-1/S1P processing of the ectodomain (Schlie et al. 2010b; Kunz et al. 2003), suggesting some sort of transmission of structural information through the membrane. Since viral budding requires interactions of the cytosolic domain of GP2 with the matrix protein Z (Capul et al. 2007), processing by SKI-1/S1P may be required for targeting mature GP to putative “budding domains” and/or unmasking GP2 binding domains to Z.
The cleavage sites of arenavirus GPC differ from cellular substrates and resemble the B and C autoprocessing motifs of SKI-1/S1P (Fig. 3.2c). The GPC of JUNV contains the sequence RSLK↓ (B site), whereas LASV and LCMV GPCs are cleaved at the motifs RRLL↓ and RRLA↓ (C site). LASV GPC with the recognition sequence RRLL undergoes SKI-1/S1P processing early in the secretory pathway (Lenz et al. 2001), whereas LCMV GPC containing RRLA is processed in late Golgi or post-Golgi compartments (Wright et al. 1990; Beyer et al. 2003). Membrane-associated SKI-1/S1P is found predominantly in the early Golgi where cellular SKI-1/S1P substrates are cleaved (Pullikotil et al. 2007). Thus, the data at hand indicate that SKI-1/S1P is active in at least three different sub-compartments of the secretory pathway, ER/cis-Golgi (LASV GPC), median Golgi (SREBPs, ATF6, CREBs, GNPTAB), and late Golgi (LCMV GPC) (Fig. 3.3b). How arenaviruses selected specific subcellular compartments for SKI-1/S1P-mediated GPC maturation is still not fully understood, but recent studies gave some hints. Subtle changes of the sequence at the cleavage site can have drastic effects on the location and efficiency of GPC maturation, despite maintaining the RXLX↓ consensus motif. As an example, processing of an LCMV GPC mutant containing the cleavage site RRLL derived from LASV GPC is redirected from late Golgi to the ER/cis-Golgi. In contrast, introduction of the LCMV GPC cleavage motif RRLA into the LASV GPC backbone results in an uncleavable protein (Burri et al. 2012). The unusual mechanism of zymogen activation and maturation of SKI-1/S1P described above (Fig. 3.2b) may contribute to this phenomenon. As mentioned above, autocatalytic processing is necessary but not sufficient to remove the SKI-1/S1P prosegment, resulting in different already active forms of the enzyme still bearing prodomain fragments of distinct lengths (Fig. 3.2b). It is conceivable that such enzyme/prodomain complexes located in defined sub-compartments of the secretory pathway may show differential specificity for viral and cellular substrates. Further evidence for differential recognition of viral and cellular substrates by SKI-1/S1P comes from the observation that the mutations R130E and R134E within the B′/B autoprocessing site result in selective impairment of viral GPC processing, but not cleavage of cellular substrates (Burri et al. 2012).
Due to their non-lytic strategy of replication, arenaviruses can establish persistent infections in vitro and in vivo without causing overt signs of pathology. Considering the multiple roles of SKI-1/S1P in maintaining cellular functions, the largely nonoverlapping subcellular localization of viral vs. cellular substrates (Fig. 3.3b) may be a consequence of the extensive coevolution of the viruses with their reservoir hosts. Accordingly, high expression levels of GPC during acute arenavirus infection do not interfere with the SKI-1/S1P-mediated processing of ATF6 involved in the host cell’s ER stress response (Pasqual et al. 2011a). As a consequence, arenavirus infection results in specific and transient activation of the ATF6-regulated branch of the cellular ER stress response that includes upregulation of chaperones, adj usting the folding capacity to the increased demand. The differential subcellular location of SKI-1/S1P processing of viral and cellular substrates may therefore allow extensive viral replication and gene expression without causing overt cytopathic effects (Oldstone 2002).
3.5 Optimized Recognition of Arenavirus GPC by SKI-1/S1P: Viral Advantage and Achilles’ Heel
Comparison of the currently known sequences of arenavirus GPCs revealed the presence of a highly conserved aromatic residue at position P7 relative to the SKI-1/S1P recognition sites in Old World and Clade C New World arenaviruses, but not in New World viruses of Clades A and B or cellular substrates (Burri et al. 2013) (Table 3.1). Early experimental evidence already supported the notion that an aromatic amino acid at P7 somehow promotes SKI-1/S1P cleavage of both LASV (Pasquato et al. 2006) and LCMV GPC (Beyer et al. 2003). Subsequent molecular modeling allowed docking of the LASV GPC-derived peptides into the putative catalytic pocket of SKI-1/S1P (Burri et al. 2013). These studies revealed that the aromatic “signature residue” in position P7 of some viral GPC recognition sequences interacts with residue Y285 located in the extended substrate-binding pocket of SKI-1/S1P (Burri et al. 2013). Indeed, introduction of the mutation Y285A into SKI-1/S1P gives an enzyme that is markedly impaired in processing of LASV GPC, but not GPC of New World arenaviruses or cellular substrates. During coevolution with their mammalian hosts, the GPCs of Old World and Clade C New World viruses apparently expanded the molecular contacts with SKI-1/S1P beyond the classical four amino acid recognition sequences, resulting in an enlarged binding surface. The concept that critical residues flanking the classical recognition sites can modulate PC processing is also supported by findings with basic PCs and their cellular substrates. A comparative study showed the influence of surrounding amino acids on the relative PC cleavage efficiency (Remacle et al. 2008). The pres ence of an N at position P1′ after the scissile bond in the substrate growth differentiation factor-11 markedly reduces cleavage by PACE4, furin, and PC7 while being selectively permissive to PC5/6 (Essalmani et al. 2008). Moreover, Constam and colleagues demonstrated that L and G at P2′ and P3′ positions in nodal, a regulator of the fate of pluripotent cells, dramatically enhance basic PC processing, further supporting this concept (Constam and Robertson 1999). The specificity of the interaction between Y285 of SKI-1/S1P and aromatic P7 residues for LASV GPC processing, but not cleavage of cellular substrates, makes this interaction a promising target for the development of specific antiviral drugs against this important human pathogen. Perturbation of the interaction of the highly conserved aromatic side chain in position P7 of LASV GPC with the contact residue Y285 of SKI-1/S1P, e.g., by a small molecule, is not expected to affect the SKI-1/S1P catalytic triad, limiting unwanted side effects. The markedly reduced processing of LASV GPC bearing a Y to A mutation in position P7 (Burri et al. 2013) suggests that viral escape variants lacking the aromatic signature residue at P7 m ay have impaired fitness, making the P7/Y285 interaction a true “Achilles’ heel” of the virus.
3.6 The Processing of Reptarenavirus GPC Is Largely Unknown
For the mammarenaviridae, several lines of evidence suggest that the maturation of the viral GPC by SKI-1/S1P is a crucial step in the virus life cycle. In contrast, little is known about the biosynthesis, maturation, and processing of reptarenavirus GPC. The sequences of reptarenavirus GPC seem to deviate from the canonical SSP/GP1/GP2 composition of mammarenaviruses and bear some resemblance to filovirus GP (Li et al. 2016; Stenglein et al. 2012). In contrast to arenavirus GPC, whose processing critically depends on SKI-1/S1P, the GP of the filovirus Ebola undergoes processing by furin, the prototypic member of the basic PC family, although this cleavage seems dispensable for virus infection and propagation (Neumann et al. 2002) (see also Chap. 5). Further studies are required to identify the specific cellular protease(s) involved in reptarenav irus GPC processing and to see if this step is crucial for functional maturation. A very different mechanism cannot be ruled out, including the use of another class of host-derived proteases that may function in the secretory pathway during biosynthesis, at the cell surface, or during the entry process, as illustrated by the use of endosomal cathepsins by filoviruses (Hunt et al. 2012) or coronaviruses that use multiple proteases, including cathepsins, cell surface transmembrane protease/serine proteases, furin, and trypsin (Millet and Whittaker 2015, see also Chap. 4).
3.7 Targeting SKI-1/S1P-Mediated GPC Processing as an Anti-arenaviral Strategy
A major challenge for the development of drugs against human pathogenic arenaviruses is the limited structural information available on the pathogens. As all viruses, arenaviruses critically depend on the molecular machinery of the host cell for their multiplication. This is particularly true for the biosynthesis of the envelope GP that involves a complex interplay with cellular factors, including SKI-1/S1P. So far, all human pathogenic arenaviruses seem to be SKI-1/S1P-dependent, and SKI-1/S1P activity is crucial for productive infection. In the absence of SKI-1/S1P activity, infected cells produce noninfectious naked particles devoid of GP, due to a yet-unknown mec hanism of selective incorporation of cleaved GP. Studies have been carried out to investigate the ability of arenaviruses to escape SKI-1/S1P. Recombinant LCMV containing the furin cleavage site RRRR↓ instead of RRLA↓ was found to be replication competent and behaved similarly to the wild-type virus in cell culture (Rojek et al. 2010). Although this suggested that arenaviruses may, at least in principle, use other proteases than SKI-1/S1P for GPC processing, persistent virus infection of SKI-1/S1P null cells or inhibitor treatment so far never resulted in the emergence of SKI-1/S1P-independent viral escape variants (Rojek et al. 2008, 2010; Pasquato et al. 2012b). Moreover, complete inhibition of SKI-1/S1P seems not required to restrict arenavirus infection in vivo, since the partially active “wood rat” variant of the enzyme conferred significant protection and prevented persistent viral infection with LCMV (Popkin et al. 2011). In sum, inhibition of SKI-1/S1P appears as a promising therapeutic approach to combat arenavirus infection (Pasquato et al. 2012a).
Direct inhibition of the mature protease is a widely used approach to interfere with the normal enzymatic activity. Although this approach often gives excellent results, a catalytically dead enzyme results in a general loss of activity toward all substrates, cellular and viral alike. Unwanted side effects must be carefu lly taken into consideration. Considering the crucial role of SKI-1/S1P in major physiological processes and disorders, including infection with highly pathogenic arena- and bunyaviruses, efforts have been made to develop a panel of protein-based and small molecule inhibitors that will be covered below.
3.7.1 Protein-Based Strategies
The naturally occurring serpin α1-antitrypsin (AT) had been long known as a potent suicide inhibitor for trypsin-like proteases, entrapping the enzyme in a stable complex following cleavage at the reactive site loop. Sub sequently, mutations have been inserted into the reactive site loop of α1-AT to introduce the B(X)nB↓ motif of basic PCs, yielding a potent protein-based furin inhibitor (α1-PDX) (Anderson et al. 1993). Based on the homology of the catalytic sites of basic and nonbasic P Cs, the reactive site loop of α1-AT was further mutated to introduce the SKI-1/S1P BX(hydrophobic)X↓ motif RRVL. In cell culture experiments, α1-AT RRVL efficiently blocked SKI-1/S1P-mediated processing of SREBPs. Overexpression of α1-AT RRVL inhibited LASV GPC maturation and had a strong antiviral effect by markedly reducing cell-to-cell spread and infectious viral particle production (Maisa et al. 2009), providing pr oof of concept.
3.7.2 Peptide-Based Compounds and Small Molecules
Analogous to peptide chloromethyl ketones (CMK) developed to inhibit furin-like proteases, two CMK peptides were designed, containing the IYISRRLL and RRLL motifs derived from LASV GPC. Both CMK peptid es act as irreversible inhibitors of SKI-1/S1P at the level of substrate maturation (Pasquato et al. 2006). Proof-of-principle studies showed that these small molecules potently block infection of LCMV (Rojek et al. 2010). However, due to their toxicity, CMK inhibitors are not suitable as therapeutic agents.
The aminopyrrolidine amide compound PF-429242 is a reversible, competitive inhibitor of SKI-1/S1P discovered by Pfizer Inc. that efficiently blocks processing of endogenous cellular substrates (Hay et al. 2007). Following PF-429242 administration in vivo, SREBP-2 activation in mice is dramatically reduced resulting in a marked drop of plasma cholesterol levels (Hawkins et al. 2008). Pharmacological inhibition of SKI-1/S1P activity by PF-429242 also blocks LASV and LCMV GPC maturation reducing cell-to-cell propagation with only mild off-target effects (Urata et al. 2011; Pasquato et al. 2012b). Cells persistently infected with LCMV were efficiently cleared by treatment with PF-429242 without emergence of drug-resistant viral escape variants (Pasquato et al. 2012b). Initial evaluation of PF-429242 in vivo in a murine model raised concerns about applications against chronic diseases, such as familial hypercholesterolemia (Hawkins et al. 2008). However, human pathogenic arenaviruses cause acute diseases, limiting antiviral treatment to a time window of a few weeks. Considering these relatively short periods of treatment, the toxicological and pharmacokinetic profile of PF-429242 makes it still an interesting experimental drug candidate (Hawkins et al. 2008).
The current standard of care fo r treatment of human arenavirus infection is an off-label use of the nucleoside analogue ribavirin (1-β-d-ribofuranosyl-1,2,4-triazole-3-carboxamide) (Parker 2005). Early administration of ribavirin reduces the fatality in human Lassa fever (Mccormick et al. 1986) and experimental infections with MACV (Kilgore et al. 1995) and JUNV (Weissenbacher et al. 1987) in animals. However, to achieve high efficacy, ribavirin needs to be administered early during infection intravenously and is often associated with side effects. Novel anti-arenaviral drugs may be used individually or in combination with therapy and ribavirin to combat human pathogenic arenaviruses, allowing lower doses of ribavirin. Indeed, the combination of PF-429242 with ribavirin revealed stronger than additive effect of the two drugs (Pasquato et al. 2012b). The basis for this apparent synergism may lie in the distinct underlying antiviral mechanisms of the two drugs. Depending on the concentration used, ribavirin inhibits arenavirus infection at the level of replication (Ruiz-Jarabo et al. 2003) and shows drug action a s a mutagen (Moreno et al. 2011). In contrast, PF-429242 affects the biosynthesis of the viral GP, blocking the formation of infectious progen y virus from infected cells.
3.7.3 A Novel Cell-Based Sensor for SKI-1/S1P as a Platform for High-Throughput Drug Screening
As outlined above, the currently available SKI-1/S1P inhibitors are potent and specific and represent invaluable experimental drugs for proof-of-concept studies. However, they face considerable restrictions regarding therapeutic use in clinical medicine. The protein nature of α1-AT RRVL makes cell permeability and in vivo drug delivery challenging. The smaller SKI-1/S1P-specific decanoylated CMK peptides show good cell permeability, but due to their toxicity profile and short half-life, their use is restricted. PF-429242 has low cytotoxicity and shows in vitro IC50 in the low micromolar range. However, the compound has an unfavorable pharmacokinetic profile in vivo and has to the best of our knowledge not yet entered the clinical test phase. Considering the promise of SKI-1/S1P as a drug target and the limitations of the candidate inhibitors at hand, the identification of novel small molecule inhibitors for SKI-1/S1P is of high priority.
PCs are essential for normal cell functions during development and in adults (Seidah and Prat 2012); nonetheless several studies showed that specific inhibition of a definite PC is not detrimental in case of in vivo short-term treatment (Hay et al. 2007; Shiryaev et al. 2007). Thus, targeting viral GP cleavage is a novel promising therapeutic approach to fight against virulent pathogens such as hemorr hagic a renaviruses. Our modern globalized world increasingly faces the threat of emerging viruses due to human migration, rapidly progressing urbanization, almost free global trade, exposure to animals, and climatic changes. The development of novel broadly specific antivirals is therefore important to help meeting these unmatched medical problems.
3.7.4 Use of the SKI-1/S1P Sensor to Predict Protease Use of Newly Emerging Arenaviruses
New arenaviruses are rapidly emerging and are in some cases associated with severe human diseases. With the advent of powerful next-generation sequencing approaches, we expect to see accelerated discovery of many new arenavirus species, pathogenic or not, in the years to come. However, in many cases viruse s may not be isolated, and only genetic information will become available. Considering the extensive variation at the known and putative SKI-1/S1P recognition sequences in known arenavirus GPC, defined consensus sequences cannot easily be found (Burri et al. 2013; Pasquato et al. 2011). Data at hand indicate that knowledge of the sequence P1–8 and residue P1′ of the putative GP1/GP2 cleavage site of a novel arenavirus GPC would be necessary and sufficient to test if the new virus can hijack human SKI-1/S1P, which is a prerequisite for productive infection and hence disease potential in man. In a recent study, the cell-based SKI-1/S1P sensor was applied to make a first prediction if the recently emerged LUJV GPC is processed by human SKI-1/S1P. Phylogenetic analysis identified LUJV as an outlier within the Old World arenaviruses as the sequence RKLM↓K at the putative GP1/GP2 border differed significantly from the known consensus (Burri et al. 2013). Despite these important differences, a sensor containing a 9mer peptide derived from LUJV GPC underwent efficient processing by human SKI-1/S1P, which was then validated with authent ic full-length GPC (Oppliger et al. 2015).
Considering the promise of SKI-1/S1P as therapeutic target for novel antiviral drugs to combat human pathogenic arenaviruses, SKI-1/S1P dependence of a newly emerging pathogenic arenavirus assessed by this new sensor may open the possibility for rapid intervention. The SKI-1/S1P sensor can further be used to assess the processing of arenavirus GPCs by SKI-1/S1P orthologues derived from other species, shedding light on the complex ecology of arenaviruses. Conceptually, the sensor platform developed for SKI-1/S1P based on lessons learned from arenavirus GPC cleavage may be applicable for other human proteases that are responsible for processing of a plethora of viral envelope GPs in a wide range of species. For the virology research community such a platform may serve as a rapid and cost-effective evaluation of viral GP processing by human proteases that may contribute to our prepare dness against the threat of emerging viruses.
The authors would like to apologize to all those colleagues whose excellent work could not be covered due to space limitations. This work was supported by Swiss National Science Foundation grant 310030_170108 to S.K. and funds to S.K. from the University of Lausanne.
- Achilleos A, Huffman NT, Marcinkiewicyz E, Seidah NG, Chen Q, Dallas SL, Trainor PA, Gorski JP. Mbtps1/Ski-1/S1p proprotein convertase is required for Ecm signaling and axial elongation during somitogenesis and vertebral development. Hum Mol Genet. 2015;24:2884–98.PubMedPubMedCentralCrossRefGoogle Scholar
- Benjannet S, Rhainds D, Essalmani R, Mayne J, Wickham L, Jin W, Asselin MC, Hamelin J, Varret M, Allard D, Trillard M, Abifadel M, Tebon A, Attie AD, Rader DJ, Boileau C, Brissette L, Chretien M, Prat A, Seidah NG. Narc-1/Pcsk9 and its natural mutants: zymogen cleavage and effects on the low density lipoprotein (Ldl) receptor and Ldl cholesterol. J Biol Chem. 2004;279:48865–75.CrossRefPubMedGoogle Scholar
- Briese T, Paweska JT, Mcmullan LK, Hutchison SK, Street C, Palacios G, Khristova ML, Weyer J, Swanepoel R, Egholm M, Nichol ST, Lipkin WI. Genetic detection and characterization of lujo virus, a new hemorrhagic fever-associated arenavirus from Southern Africa. PLoS Pathog. 2009;5:E1000455.PubMedPubMedCentralCrossRefGoogle Scholar
- Buchmeier MJ, De La Torre JC, Peters CJ. Arenaviridae: the viruses and their replication. In: Knipe DL, Howley PM, editors. Fields Virology. 4th ed. Philadelphia: Lippincott-Raven; 2007.Google Scholar
- Delgado S, Erickson BR, Agudo R, Blair PJ, Vallejo E, Albarino CG, Vargas J, Comer JA, Rollin PE, Ksiazek TG, Olson JG, Nichol ST. Chapare virus, a newly discovered arenavirus isolated from a fatal hemorrhagic fever case In Bolivia. PLoS Pathog. 2008;4:E1000047.PubMedPubMedCentralCrossRefGoogle Scholar
- Gorski JP, Huffman NT, Vallejo J, Brotto L, Chittur SV, Breggia A, Stern A, Huang J, Mo C, Seidah NG, Bonewald L, Brotto M. Deletion of Mbtps1 (Pcsk8, S1p, Ski-1) gene in osteocytes stimulates soleus muscle regeneration and increased size and contractile force with age. J Biol Chem. 2016;291:4308–22.PubMedPubMedCentralCrossRefGoogle Scholar
- Hawkins JL, Robbins MD, Warren LC, Xia D, Petras SF, Valentine JJ, Varghese AH, Wang IK, Subashi TA, Shelly LD, Hay BA, Landschulz KT, Geoghegan KF, Harwood HJ Jr. Pharmacologic inhibition of site 1 protease activity inhibits sterol regulatory element-binding protein processing and reduces lipogenic enzyme gene expression and lipid synthesis in cultured cells and experimental animals. J Pharmacol Exp Ther. 2008;326:801–8.CrossRefPubMedGoogle Scholar
- Helguera G, Jemielity S, Abraham J, Cordo SM, Martinez MG, Rodriguez JA, Bregni C, Wang JJ, Farzan M, Penichet ML, Candurra NA, Choe H. An antibody recognizing the apical domain of human transferrin receptor 1 efficiently inhibits the entry of all New World hemorrhagic fever arenaviruses. J Virol. 2012;86:4024–8.PubMedPubMedCentralCrossRefGoogle Scholar
- Hetzel U, Sironen T, Laurinmaki P, Liljeroos L, Patjas A, Henttonen H, Vaheri A, Artelt A, Kipar A, Butcher SJ, Vapalahti O, Hepojoki J. Isolation, identification, and characterization of novel arenaviruses, the etiological agents of boid inclusion body disease. J Virol. 2013;87:10918–35.PubMedPubMedCentralCrossRefGoogle Scholar
- Jemielity S, Wang JJ, Chan YK, Ahmed AA, Li W, Monahan S, Bu X, Farzan M, Freeman GJ, Umetsu DT, Dekruyff RH, Choe H. Tim-family proteins promote infection of multiple enveloped viruses through virion-associated phosphatidylserine. PLoS Pathog. 2013;9:E1003232.PubMedPubMedCentralCrossRefGoogle Scholar
- Palacios G, Druce J, Du L, Tran T, Birch C, Briese T, Conlan S, Quan PL, Hui J, Marshall J, Simons JF, Egholm M, Paddock CD, Shieh WJ, Goldsmith CS, Zaki SR, Catton M, Lipkin WI. A new arenavirus in a cluster of fatal transplant-associated diseases. N Engl J Med. 2008;358:991–8.CrossRefPubMedGoogle Scholar
- Pasquato A, Pullikotil P, Asselin MC, Vacatello M, Paolillo L, Ghezzo F, Basso F, Di Bello C, Dettin M, Seidah NG. The proprotein convertase Ski-1/S1p. In vitro analysis of Lassa virus glycoprotein-derived substrates and ex vivo validation of irreversible peptide inhibitors. J Biol Chem. 2006;281:23471–81.CrossRefPubMedGoogle Scholar
- Popkin DL, Teijaro JR, Sullivan BM, Urata S, Rutschmann S, De La Torre JC, Kunz S, Beutler B, Oldstone M. Hypomorphic mutation in the site-1 protease Mbtps1 endows resistance to persistent viral infection in a cell-specific manner. Cell Host Microbe. 2011;9:212–22.PubMedPubMedCentralCrossRefGoogle Scholar
- Radoshitzky SR, Abraham J, Spiropoulou CF, Kuhn JH, Nguyen D, Li W, Nagel J, Schmidt PJ, Nunberg JH, Andrews NC, Farzan M, Choe H. Transferrin receptor 1 is a cellular receptor for New World haemorrhagic fever arenaviruses. Nature. 2007;446:92–6. Epub 2007 Feb 7PubMedPubMedCentralCrossRefGoogle Scholar
- Radoshitzky SR, Bao Y, Buchmeier MJ, Charrel RN, Clawson AN, Clegg CS, Derisi JL, Emonet S, Gonzalez JP, Kuhn JH, Lukashevich IS, Peters CJ, Romanowski V, Salvato MS, Stenglein MD, De La Torre JC. Past, present, and future of arenavirus taxonomy. Arch Virol. 2015;160:1851–74.CrossRefPubMedGoogle Scholar
- Remacle AG, Shiryaev SA, Oh ES, Cieplak P, Srinivasan A, Wei G, Liddington RC, Ratnikov BI, Parent A, Desjardins R, Day R, Smith JW, Lebl M, Strongin AY. Substrate cleavage analysis of furin and related proprotein convertases. A Comparative Study. J Biol Chem. 2008;283:20897–906.PubMedPubMedCentralCrossRefGoogle Scholar
- Seidah NG, Mowla SJ, Hamelin J, Mamarbachi AM, Benjannet S, Toure BB, Basak A, Munzer JS, Marcinkiewicz J, Zhong M, Barale JC, Lazure C, Murphy RA, Chretien M, Marcinkiewicz M. Mammalian subtilisin/kexin isozyme Ski-1: a widely expressed proprotein convertase with a unique cleavage specificity and cellular localization. Proc Natl Acad Sci U S A. 1999;96:1321–6.PubMedPubMedCentralCrossRefGoogle Scholar
- Shiryaev SA, Remacle AG, Ratnikov BI, Nelson NA, Savinov AY, Wei G, Bottini M, Rega MF, Parent A, Desjardins R, Fugere M, Day R, Sabet M, Pellecchia M, Liddington RC, Smith JW, Mustelin T, Guiney DG, Lebl M, Strongin AY. Targeting host cell furin proprotein convertases as a therapeutic strategy against bacterial toxins and viral pathogens. J Biol Chem. 2007;282:20847–53.PubMedPubMedCentralCrossRefGoogle Scholar
- Stenglein MD, Sanders C, Kistler AL, Ruby JG, Franco JY, Reavill DR, Dunker F, Derisi JL. Identification, characterization, and in vitro culture of highly divergent arenaviruses from boa constrictors and annulated tree boas: candidate etiological agents for snake inclusion body disease. MBio. 2012;3:E00180-12.PubMedPubMedCentralCrossRefGoogle Scholar