Search

P-Rex (PI(3,4,5)P3-dependent Rac exchanger) guanine nucleotide exchange factors potently activate Rho GTPases. P-Rex guanine nucleotide exchange factors are autoinhibited, synergistically activated by Gβγ and PI(3,4,5)P3 binding and dysregulated in cancer. Here, we use X-ray crystallography, cryogenic electron microscopy and crosslinking mass spectrometry to determine the structural basis of human P-Rex1 autoinhibition. P-Rex1 has a bipartite structure of N- and C-terminal modules connected by a C-terminal four-helix bundle that binds the N-terminal Pleckstrin homology (PH) domain. In the N-terminal module, the Dbl homology (DH) domain catalytic surface is occluded by the compact arrangement of the DH-PH-DEP1 domains. Structural analysis reveals a remarkable conformational transition to release autoinhibition, requiring a 126° opening of the DH domain hinge helix. The off-axis position of Gβγ and PI(3,4,5)P3 binding sites further suggests a counter-rotation of the P-Rex1 halves by 90° facilitates PH domain uncoupling from the four-helix bundle, releasing the autoinhibited DH domain to drive Rho GTPase signaling.

To control infections phagocytes can directly kill invading microbes. Macrophage-expressed gene 1 (Mpeg1), a pore-forming protein sometimes known as perforin-2, is reported to be essential for bacterial killing following phagocytosis. Mice homozygous for the mutant allele Mpeg1

Automation in microscopy is the key to success in long and complex experiments. Most microscopy manufacturers provide Application Programming Interfaces (API) to enable communication between a user-defined program and the hardware. Although APIs effectively allow the development of complex routines involving hardware control, the developers need to build the applications from basic commands. Here we present a Software Development Kit (SDK) for easy control of Focussed Ion Beam Scanning Electron Microscopes (FIB/SEM) microscopes. The SDK, which we named OpenFIBSEM consists of a suite of building blocks for easy control that simplify the development of complex automated workflows.

The terminal C5b-9 complement complex (TCC) exists in two forms; the soluble sC5b-9 and the solid-phase inserted Membrane Attack Complex (MAC). The MAC is responsible for forming large β-barrel channels in the membranes of pathogens and can target cells such as gram-negative bacteria. In addition, off-target MAC assembly on endogenous tissue is associated with inflammatory diseases and cancer. Accordingly, a human C5b-9 specific antibody, aE11, has been developed that detects a neoepitope exposed in C9 when it is incorporated into the C5b-9 complex and in polyC9, but not present in the plasma native C9. For nearly four decades aE11 has been routinely used to study complement, MACrelated inflammation, and pathophysiology. However, the identity of this C9 neoepitope remains unknown. Herein, we determined the cryo-EM structure of aE11 in complex with polyC9 at 3.2 Åresolution. The aE11 binding site revealed that the neoepitope was formed by two separate surfaces on the pore periphery, rather than a conformational change in the protein structure, and is therefore a discontinuous quaternary epitope. These surfaces are contributed by portions of the adjacent TSP1, LDLRA and MACPF domains of two neighbouring C9 protomers. By substituting key antibody interacting residues to the murine orthologue, we validated the unusual binding modality of aE11. Furthermore, our observations indicated that aE11 can recognise a partial epitope in purified monomeric C9, albeit binding was two orders of magnitude weaker. Taken together, our results reveal the basis for MAC recognition by aE11 and that the C9 neoepitope can be formed without substantial conformational rearrangements adding insight into how aE11 is used to quantify MAC formation in disease.

Patients with hereditary angioedema (HAE) experience episodes of bradykinin (BK)-induced swelling of skin and mucosal membranes. The most common cause is reduced plasma activity of C1 inhibitor, the main regulator of the proteases plasma kallikrein (PKa) and factor XIIa (FXIIa). Recently, patients with HAE were described with a Lys311 to glutamic acid substitution in plasminogen (Plg), the zymogen of the protease plasmin (Plm). Adding tissue plasminogen activator to plasma containing Plg-Glu311 vs plasma containing wild-type Plg (Plg-Lys311) results in greater BK generation. Similar results were obtained in plasma lacking prekallikrein or FXII (the zymogens of PKa and FXIIa) and in normal plasma treated with a PKa inhibitor, indicating Plg-Glu311 induces BK generation independently of PKa and FXIIa. Plm-Glu311 cleaves high and low molecular weight kininogens (HK and LK, respectively), releasing BK more efficiently than Plm-Lys311. Based on the plasma concentrations of HK and LK, the latter may be the source of most of the BK generated by Plm-Glu311. The lysine analog ε-aminocaproic acid blocks Plm-catalyzed BK generation. The Glu311 substitution introduces a lysine-binding site into the Plg kringle 3 domain, perhaps altering binding to kininogens. Plg residue 311 is glutamic acid in most mammals. Glu311 in patients with HAE, therefore, represents reversion to the ancestral condition. Substantial BK generation occurs during Plm-Glu311 cleavage of human HK, but not mouse HK. Furthermore, mouse Plm, which has Glu311, did not liberate BK from human kininogens more rapidly than human Plg-Lys311. This indicates Glu311 is pathogenic in the context of human Plm when human kininogens are the substrates.

Background: CUB domain-containing protein 1 (CDCP1) is a cell surface receptor regulating key signalling pathways in malignant cells. CDCP1 has been proposed as a molecular target to abrogate oncogenic signalling pathways and specifically deliver anti-cancer agents to tumors. However, the development of CDCP1-targeting agents has been questioned by its frequent proteolytic processing which was thought to result in shedding of the CDCP1 extracellular domain limiting its targetability. In this study, we investigated the relevance of targeting CDCP1 in the context of pancreatic ductal adenocarcinoma (PDAC) and assess the impact of CDCP1 proteolysis on the effectiveness of CDCP1 targeting agents. Methods: The involvement of CDCP1 in PDAC progression was assessed by association analysis in several PDAC cohorts and the proteolytic processing of CDCP1 was evaluated in PDAC cell lines and patient-derived cells. The consequences of CDCP1 proteolysis on its targetability in PDAC cells was assessed using immunoprecipitation, immunostaining and biochemical assays. The involvement of CDCP1 in PDAC progression was examined by loss-of-function in vitro and in vivo experiments employing PDAC cells expressing intact or cleaved CDCP1. Finally, we generated antibody-based imaging and therapeutic agents targeting CDCP1 to demonstrate the feasibility of targeting this receptor for detection and treatment of PDAC tumors. Results: High CDCP1 expression in PDAC is significantly associated with poorer patient survival. In PDAC cells proteolysis of CDCP1 does not always result in the shedding of CDCP1-extracellular domain which can interact with membrane-bound CDCP1 allowing signal transduction between the different CDCP1-fragments. Targeting CDCP1 impairs PDAC cell functions and PDAC tumor growth independently of CDCP1 cleavage status. A CDCP1-targeting antibody is highly effective at delivering imaging radionuclides and cytotoxins to PDAC cells allowing specific detection of tumors by PET/CT imaging and superior anti-tumor effects compared to gemcitabine in in vivo models. Conclusion: Independent of its cleavage status, CDCP1 exerts oncogenic functions in PDAC and has significant potential to be targeted for improved radiological staging and treatment of this cancer. Its elevated expression by most PDAC tumors and lack of expression by normal pancreas and other major organs, suggest that targeting CDCP1 could benefit a significant proportion of PDAC patients. These data support the further development of CDCP1-targeting agents as personalizable tools for effective imaging and treatment of PDAC.

The zinc-dependent metalloprotease meprin α is predominantly expressed in the brush border membrane of proximal tubules in the kidney and enterocytes in the small intestine and colon. In normal tissue homeostasis meprin α performs key roles in inflammation, immunity, and extracellular matrix remodelling. The latter activity is furthermore important for driving aggressive metastasis in the context of certain cancers such as colorectal carcinoma. Accordingly, meprin α is the target of drug discovery programs. In contrast to meprin β, meprin α is secreted into the extracellular space, whereupon it oligomerises to form giant assemblies and is the largest extracellular protease identified to date (~6 MDa). Here, using cryo-electron microscopy, we determine the high-resolution structure of the zymogen and mature form of meprin α, as well as the structure of the active form in complex with a prototype small molecule inhibitor and human fetuin-B. Our data reveal that meprin α forms a giant, flexible, left-handed helical assembly of roughly 22 nm in diameter. We find that oligomerisation improves proteolytic and thermal stability but does not impact substrate specificity or enzymatic activity. Furthermore, structural comparison with meprin β reveal unique features of the active site of meprin α, and helical assembly more broadly.

Macrophages are an ancient blood cell lineage critical for homeostasis and defence against pathogens. Although their numbers were long thought to be sustained solely by haematopoietic organs, it has recently become clear that their proliferation, or self-renewal, also plays a major role. In the Drosophila larva, macrophages undergo a phase of rapid self-renewal, making this an attractive model for elucidating the signals and regulatory mechanisms involved. However, a central self-renewal pathway has not been identified in this system. Here, we show that the PDGF- and VEGF-receptor related (Pvr) pathway fulfils this role. Our data show that two of the three known Pvr ligands, PDGF- and VEGF-related factor 2 (Pvf2) and Pvf3, are major determinants of overall macrophage numbers, yet they each act in a temporally independent manner and via distinct mechanisms. While Pvf3 is needed prior to the self-renewal period, we find that Pvf2 is critical specifically for expanding the larval macrophage population. We further show that Pvf2 is a potent macrophage mitogen that is kept at limiting quantities by its transient expression in a remarkably small number of blood cells. Together, these data support a novel mechanism for the regulation of macrophage self-renewal rates by the dynamic transcriptional control of Pvf2. Given the strong parallels that exist between Drosophila and vertebrate macrophage systems, it is likely that a similar self-renewal control mechanism is at play across animal phyla.

Protein structure fundamentally underpins the function and processes of numerous biological systems. Fold recognition algorithms offer a sensitive and robust tool to detect structural, and thereby functional, similarities between distantly related homologs. In the era of accurate structure prediction owing to advances in machine learning techniques and a wealth of experimentally determined structures, previously curated sequence databases have become a rich source of biological information. Here, we use bioinformatic fold recognition algorithms to scan the entire AlphaFold structure database to identify novel protein family members, infer function and group predicted protein structures. As an example of the utility of this approach, we identify novel, previously unknown members of various pore-forming protein families, including MACPFs, GSDMs and aerolysin-like proteins.

Protein structure fundamentally underpins the function and processes of numerous biological systems. Fold recognition algorithms offer a sensitive and robust tool to detect structural, and thereby functional, similarities between distantly related homologs. In the era of accurate structure prediction owing to advances in machine learning techniques, previously curated sequence databases have become a rich source of biological information. Here, we use bioinformatic fold recognition algorithms to scan the entire AlphaFold structure database to identify novel protein family members, infer function and group predicted protein structures. As an example of the utility of this approach, we identify novel, previously unknown members of various pore-forming protein families, including MACPFs, GSDMs and aerolysin-like proteins. Further, we explore the use of structure-based mining for functional inference.

CUB-domain containing protein 1 (CDCP1) is a cancer associated cell surface protein that amplifies pro-tumorigenic signalling by other receptors including EGFR and HER2. Its potential as a cancer target is supported by studies showing that anti-CDCP1 antibodies inhibit cell migration and survival in vitro, and tumor growth and metastasis in vivo. Here we characterize two anti-CDCP1 antibodies, focusing on immuno-conjugates of one of these as a tool to detect and inhibit ovarian cancer. Methods: A panel of ovarian cancer cell lines was examined for cell surface expression of CDCP1 and loss of expression induced by anti-CDCP1 antibodies 10D7 and 41-2 using flow cytometry and Western blot analysis. Surface plasmon resonance analysis and examination of truncation mutants was used to analyse the binding properties of the antibodies for CDCP1. Live-cell spinning-disk confocal microscopy of GFP-tagged CDCP1 was used to track internalization and intracellular trafficking of CDCP1/antibody complexes. In vivo, zirconium 89-labelled 10D7 was detected by positron-emission tomography imaging, of an ovarian cancer patient-derived xenograft grown intraperitoneally in mice. The efficacy of cytotoxin-conjugated 10D7 was examined against ovarian cancer cells in vitro and in vivo. Results: Our data indicate that each antibody binds with high affinity to the extracellular domain of CDCP1 causing rapid internalization of the receptor/antibody complex and degradation of CDCP1 via processes mediated by the kinase Src. Highlighting the potential clinical utility of CDCP1, positron-emission tomography imaging, using zirconium 89-labelled 10D7, was able to detect subcutaneous and intraperitoneal xenograft ovarian cancers in mice, including small (diameter

VEK50 is a truncated peptide from a Streptococcal pyogenes surface human plasminogen (hPg) binding M-protein (PAM). VEK50 contains the full A-domain of PAM, which is responsible for its low nanomolar binding to hPg. The interaction of VEK50 with kringle 2, the PAM-binding domain in hPg (K2(hPg)), has been studied by high-resolution NMR spectroscopy. The data show that each VEK50 monomer in solution contains two tight binding sites for K2(hPg), one each in the a1- (RH1; R(17)H(18)) and a2- (RH2; R(30)H(31)) repeats within the A-domain of VEK50. Two mutant forms of VEK50, viz., VEK50[RH1/AA] (VEK50(△RH1)) and VEK50[RH2/AA] (VEK50(△RH2)), were designed by replacing each RH with AA, thus eliminating one of the K2(hPg) binding sites within VEK50, and allowing separate study of each binding site. Using (13)C- and (15)N-labeled peptides, NMR-derived solution structures of VEK50 in its complex with K2(hPg) were solved. We conclude that the A-domain of PAM can accommodate two molecules of K2(hPg) docked within a short distance of each other, and the strength of the binding is slightly different for each site. The solution structure of the VEK50/K2(hPg), complex, which is a reductionist model of the PAM/hPg complex, provides insights for the binding mechanism of PAM to a host protein, a process that is critical to S. pyogenes virulence.

Macrophage-expressed gene 1 (MPEG1/Perforin-2) is a perforin-like protein that functions within the phagolysosome to damage engulfed microbes. MPEG1 is thought to form pores in target membranes, however, its mode of action remains unknown. We use cryo-Electron Microscopy (cryo-EM) to determine the 2.4 Å structure of a hexadecameric assembly of MPEG1 that displays the expected features of a soluble prepore complex. We further discover that MPEG1 prepore-like assemblies can be induced to perforate membranes through acidification, such as would occur within maturing phagolysosomes. We next solve the 3.6 Å cryo-EM structure of MPEG1 in complex with liposomes. These data reveal that a multi-vesicular body of 12 kDa (MVB12)-associated β-prism (MABP) domain binds membranes such that the pore-forming machinery of MPEG1 is oriented away from the bound membrane. This unexpected mechanism of membrane interaction suggests that MPEG1 remains bound to the phagolysosome membrane while simultaneously forming pores in engulfed bacterial targets., Macrophage-expressed gene 1 (MPEG1) functions within the phagolysosome to damage engulfed microbes, presumably via forming pores in target membranes. In order to provide insights into the mechanism of MPEG1 function and membrane binding, the authors present structures of hexadecameric MPEG1 prepores both in solution and in complex with liposomes.

Members of the membrane attack complex/perforin-like (MACPF) protein superfamily have long captured interest because of their unique ability to assemble into large oligomeric pores on the surfaces of cells. The best characterised of these act in vertebrate immunity where they function to deliver pro-apoptotic factors or induce the cytolysis and death of targeted cells. Less appreciated, however, is that rather than causing cell death, MACPF proteins have also evolved to control cellular signalling pathways and influence developmental programmes such as pattern formation and neurogenesis. Torso-like (Tsl) from the fruit fly Drosophila, for example, functions to localise the activity of a growth factor for patterning its embryonic termini. It remains unclear whether these developmental proteins employ an attenuated form of the classical MACPF lytic pore, or if they have evolved to function via alternative mechanisms of action. In this minireview, we examine the evidence that links pore-forming MACPF proteins to the control of growth factor and cytokine signalling. We will then attempt to reconcile how the MACPF domain may have been repurposed during evolution for developmental events rather than cell killing.

Plasminogen (Plg) is the zymogen form of the serine protease plasmin (Plm), and it plays a crucial role in fibrinolysis as well as wound healing, immunity, tissue remodeling and inflammation. Binding to the targets via the lysine-binding sites allows for Plg activation by plasminogen activators (PAs) present on the same target. Cellular uptake of fibrin degradation products leads to apoptosis, which represents one of the pathways for cross-talk between fibrinolysis and tissue remodeling. Therapeutic manipulation of Plm activity plays a vital role in the treatments of a range of diseases, whereas Plm inhibitors are used in trauma and surgeries as antifibrinolytic agents. Plm inhibitors are also used in conditions such as angioedema, menorrhagia and melasma. Here, we review the rationale for the further development of new Plm inhibitors, with a particular focus on the structural studies of the active site inhibitors of Plm. We compare the binding mode of different classes of inhibitors and comment on how it relates to their efficacy, as well as possible future developments.

The dual-specificity phosphatase PTEN functions as a tumor suppressor by hydrolyzing PI(3,4,5)P3 to PI(4,5)P2 to inhibit PI3K-AKT signaling and cellular proliferation. P-Rex2 is a guanine nucleotide exchange factor for Rho GTPases and can be activated by Gβγ subunits downstream of G protein-coupled receptor signaling and by PI(3,4,5)P3 downstream of receptor tyrosine kinases. The PTEN:P-Rex2 complex is a commonly mutated signaling node in metastatic cancer. Assembly of the PTEN:P-Rex2 complex inhibits the activity of both proteins, and its dysregulation can drive PI3K-AKT signaling and cellular proliferation. Here, using cross-linking mass spectrometry and functional studies, we gained mechanistic insights into PTEN:P-Rex2 complex assembly and coinhibition. We found that PTEN was anchored to P-Rex2 by interactions between the PDZ-interacting motif in the PTEN C-terminal tail and the second PDZ domain of P-Rex2. This interaction bridged PTEN across the P-Rex2 surface, preventing PI(3,4,5)P3 hydrolysis. Conversely, PTEN both allosterically promoted an autoinhibited conformation of P-Rex2 and blocked its binding to Gβγ. In addition, we observed that the PTEN-deactivating mutations and P-Rex2 truncations combined to drive Rac1 activation to a greater extent than did either single variant alone. These insights enabled us to propose a class of gain-of-function, cancer-associated mutations within the PTEN:P-Rex2 interface that uncouple PTEN from the inhibition of Rac1 signaling.

Neurofibromin (NF1) is a tumour suppressor mutated in neurofibromatosis type 1 (von Recklinghausen disease), one of the most common human genetic diseases(1). NF1 regulates cellular growth through suppressing the Rat Sarcoma (RAS) pathway and, accordingly, mutations in this protein drive numerous cancers, including melanoma, ovarian, breast and brain cancer(2, 3). Currently, however, the molecular basis for NF1 function remains to be understood. Here we address this problem and use cryogenic Electron Microscopy (cryo-EM) to determine the structure of fulllength NF1. The 640 kDa NF1 homodimer forms an extraordinary lemniscate (∞) shaped molecule that is ~30 nm in length and ~ 10 nm wide. Each NF1 monomer comprises an N-terminal HEAT-repeat domain (N-HEAT), a guanosine triphosphatase activating protein (GAP)-related domain (GRD), a Sec14 homologous and pleckstrin homologous module (SEC-PH), and a C-terminal HEAT domain (C-HEAT). The core NF1 scaffold is formed via a head-to-tail dimer of the N- and C-HEAT domains. This platform, which is responsible for interacting with more than 10 regulatory binding partners, comprises an extraordinary array of over 150 α-helices. Analysis of these EM data revealed that the GRD and SEC-PH domain are highly mobile with respect to the core scaffold and could not initially be accurately placed in electron density. Strikingly, however, using 3D variability analysis we were able to identify a significant subpopulation of NF1 particles and determine the complete NF1 structure to 5.6 Å resolution. These data revealed that the catalytic GRD and lipid binding SEC-PH domain are positioned against the core scaffold in a closed, autoinhibited conformation. We postulate that interaction with the plasma membrane may release the closed conformation in order to promote RAS inactivation. Our structural data further allow us to map the location of disease-associated NF1 variants and provide a long sought-after structural explanation for the extreme susceptibility of the molecule to loss-of-function mutations. Finally, it is suggested that approaches to combat NF1-linked diseases may include release of the autoinhibited state in order to improve NF1 catalytic efficiency.

Macrophages are an ancient animal blood cell lineage critical for tissue homeostasis and defence against pathogens. Until recently, their numbers were thought to be sustained solely by specialised hematopoietic organs. It is now clear that many macrophages are instead replenished by self-renewal, yet the signals that regulate this remain poorly understood. InDrosophila melanogaster, macrophages (known as plasmatocytes) undergo a phase of rapid population expansion via self-renewal, makingDrosophilaan attractive model for revealing the signals and regulatory mechanisms involved. However, no central self-renewal pathway has been identified inDrosophila. Here, we investigated the PDGF-/VEGF-receptor pathway as a candidate for playing this role. Analysis of larvae deficient for each of the three PDGF-/VEGF-receptor ligands Pvf1-3 revealed Pvf2 as a major driver of macrophage self-renewal inDrosophila. We further found that only a small proportion of blood cells expressPvf2, and knockdown experiments implicate these cells as a major source ofPvf2in self-renewal. Lineage tracing studies support the idea thatPvf2expression in blood cells occurs transiently throughout the macrophage self-renewal period, and in response to an as yet unidentified cue. These data define the regulation ofPvf2expression in blood cells as a central mechanism by which macrophage self-renewal is controlled. Given the strong parallels that exist betweenDrosophilaand vertebrate macrophage systems, it is likely that similar mechanisms are at play across animal phyla.

The PTEN:P-Rex2 complex is one of the most commonly mutated signaling nodes in metastatic cancer. Assembly of the PTEN:P-Rex2 complex inhibits the activity of both proteins, and its dysregulation can drive PI3K-AKT signaling and cell proliferation. Here, using extensive crosslinking mass spectrometry and functional studies, we provide crucial mechanistic insights into PTEN:P-Rex2 complex assembly and co-inhibition. PTEN is anchored to P-Rex2 by interactions between the PTEN PDZ-BM tail and the second PDZ domain of P-Rex2. This interaction bridges PTEN across the P-Rex2 surface, occluding PTEN membrane-binding and PI(3,4,5)P3hydrolysis. Conversely, PTEN both allosterically promotes an autoinhibited P-Rex2 conformation and occludes Gβγ binding and GPCR activation. These insights allow us to define a new gain-of-function class of cancer mutations within the PTEN:P-Rex2 interface that uncouples PTEN inhibition of Rac1 signaling. These findings provide a mechanistic framework to understand the dysregulation of the PTEN:P-Rex2 signaling node in metastatic cancer.

Harnessing the immunomodulatory activity of cytokines is a focus of therapies targeting inflammatory disease. The interleukin (IL)-1 superfamily contains pro-inflammatory and anti-inflammatory members that help orchestrate the immune response in adaptive and innate immunity. Of these molecules, IL-37 has robust anti-inflammatory activity across a range of disease models through inhibition of pro-inflammatory signaling cascades downstream of tumor necrosis factor, IL-1, and toll-like receptor pathways. We find that IL-37 is unstable with a poor pharmacokinetic and manufacturing profile. Here, we present the engineering of IL-37 from an unstable cytokine into an anti-inflammatory molecule with an excellent therapeutic likeness. We overcame these shortcomings through site-directed mutagenesis, the addition of a non-native disulfide bond, and the engineering of IL-37 as an Fc-fusion protein. Our results provide a platform for preclinical testing of IL-37 Fc-fusion proteins. The engineering approaches undertaken herein will apply to the conversion of similar potent yet short-acting cytokines into therapeutics.

Antifibrinolytic drugs provide important pharmacological interventions to reduce morbidity and mortality from excessive bleeding during surgery and after trauma. Current drugs used for inhibiting the dissolution of fibrin, the main structural component of blood clots, are associated with adverse events due to lack of potency, high doses, and nonselective inhibition mechanisms. These drawbacks warrant the development of a new generation of highly potent and selective fibrinolysis inhibitors. Here, we use the 14-amino acid backbone-cyclic sunflower trypsin inhibitor-1 scaffold to design a highly potent (Ki = 0.05 nM) inhibitor of the primary serine protease in fibrinolysis, plasmin. This compound displays a million-fold selectivity over other serine proteases in blood, inhibits fibrinolysis in plasma more effectively than the gold-standard therapeutic inhibitor aprotinin, and is a promising candidate for development of highly specific fibrinolysis inhibitors with reduced side effects.

The roles of proteolytic cleavage have been intensively investigated and discussed during the past two decades. This irreversible chemical process has been frequently reported to influence a number of crucial biological processes (BPs), such as cell cycle, protein regulation and inflammation. A number of advanced studies have been published aiming at deciphering the mechanisms of proteolytic cleavage. Given its significance and the large number of functionally enriched substrates targeted by specific proteases, many computational approaches have been established for accurate prediction of protease-specific substrates and their cleavage sites. Consequently, there is an urgent need to systematically assess the state-of-the-art computational approaches for protease-specific cleavage site prediction to further advance the existing methodologies and to improve the prediction performance. With this goal in mind, in this article, we carefully evaluated a total of 19 computational methods (including 8 scoring function-based methods and 11 machine learning-based methods) in terms of their underlying algorithm, calculated features, performance evaluation and software usability. Then, extensive independent tests were performed to assess the robustness and scalability of the reviewed methods using our carefully prepared independent test data sets with 3641 cleavage sites (specific to 10 proteases). The comparative experimental results demonstrate that PROSPERous is the most accurate generic method for predicting eight protease-specific cleavage sites, while GPS-CCD and LabCaS outperformed other predictors for calpain-specific cleavage sites. Based on our review, we then outlined some potential ways to improve the prediction performance and ease the computational burden by applying ensemble learning, deep learning, positive unlabeled learning and parallel and distributed computing techniques. We anticipate that our study will serve as a practical and useful guide for interested readers to further advance next-generation bioinformatics tools for protease-specific cleavage site prediction.

Phosphorylation is the most important type of protein post-translational modification. Accordingly, reliable identification of kinase-mediated phosphorylation has important implications for functional annotation of phosphorylated substrates and characterization of cellular signalling pathways. The local sequence context surrounding potential phosphorylation sites is considered to harbour the most relevant information for phosphorylation site prediction models. However, currently there is a lack of condensed vector representation for this important contextual information, despite the presence of varying residue-level features that can be constructed from sequence homology profiles, structural information, and physicochemical properties. To address this issue, we present PhosContext2vec which is a distributed representation of residue-level sequence contexts for potential phosphorylation sites and demonstrate its application in both general and kinase-specific phosphorylation site predictions. Benchmarking experiments indicate that PhosContext2vec could achieve promising predictive performance compared with several other existing methods for phosphorylation site prediction. We envisage that PhosContext2vec, as a new sequence context representation, can be used in combination with other informative residue-level features to improve the classification performance in a number of related bioinformatics tasks that require appropriate residue-level feature vector representation and extraction. The web server of PhosContext2vec is publicly available at http://phoscontext2vec.erc.monash.edu/.

In Drosophila, key developmental transitions are governed by the steroid hormone ecdysone. A number of neuropeptide-activated signaling pathways control ecdysone production in response to environmental signals, including the insulin signaling pathway, which regulates ecdysone production in response to nutrition. Here, we find that the Membrane Attack Complex/Perforin-like protein Torso-like, best characterized for its role in activating the Torso receptor tyrosine kinase in early embryo patterning, also regulates the insulin signaling pathway in Drosophila. We previously reported that the small body size and developmental delay phenotypes of torso-like null mutants resemble those observed when insulin signaling is reduced. Here we report that, in addition to growth defects, torso-like mutants also display metabolic and nutritional plasticity phenotypes characteristic of mutants with impaired insulin signaling. We further find that in the absence of torso-like, the expression of insulin-like peptides is increased, as is their accumulation in insulin-producing cells. Finally, we show that Torso-like is a component of the hemolymph and that it is required in the prothoracic gland to control developmental timing and body size. Taken together, our data suggest that the secretion of Torso-like from the prothoracic gland influences the activity of insulin signaling throughout the body in Drosophila.