Structural basis of TMPRSS2 zymogen activation and recognition by the HKU1 seasonal coronavirus.

The human seasonal coronavirus HKU1-CoV, which causes common colds worldwide, relies on the sequential binding to surface glycans and transmembrane serine protease 2 (TMPRSS2) for entry into target cells. TMPRSS2 is synthesized as a zymogen that undergoes autolytic activation to process its substrates. Several respiratory viruses, in particular coronaviruses, use TMPRSS2 for proteolytic priming of their surface spike protein to drive membrane fusion upon receptor binding. We describe the crystal structure of the HKU1-CoV receptor binding domain in complex with TMPRSS2, showing that it recognizes residues lining the catalytic groove. Combined mutagenesis of interface residues and comparison across species highlight positions 417 and 469 as determinants of HKU1-CoV host tropism. The structure of a receptor-blocking nanobody in complex with zymogen or activated TMPRSS2 further provides the structural basis of TMPRSS2 activating conformational change, which alters loops recognized by HKU1-CoV and dramatically increases binding affinity.

A simple method to resolve rate constants when the binding mechanism obeys induced fit or conformational selection.

Many interactions involving a ligand and its molecular target are studied by rapid kinetics using a stopped-flow apparatus. Information obtained from these studies is often limited to a single, saturable relaxation that is insufficient to resolve all independent rate constants even for a two-step mechanism of binding obeying induced fit (IF) or conformational selection (CS). We introduce a simple method of general applicability where this limitation is overcome. The method accurately reproduces the rate constants for ligand binding to the serine protease thrombin determined independently from the analysis of multiple relaxations. Application to the inactive zymogen precursor of thrombin, prethrombin-2, resolves all rate constants for a binding mechanism of IF or CS from a single, saturable relaxation. Comparison with thrombin shows that the prethrombin-2 to thrombin conversion enhances ligand binding to the active site not by improving accessibility through the value of kon but by reducing the rate of dissociation koff. The conclusion holds regardless of whether binding is interpreted in terms of IF or CS and has general relevance for the mechanism of zymogen activation of serine proteases. The method also provides a simple test of the validity of IF and CS and indicates when more complex mechanisms of binding should be considered.

Massively parallel, computationally guided design of a proenzyme.

Confining the activity of a designed protein to a specific microenvironment would have broad-ranging applications, such as enabling cell type-specific therapeutic action by enzymes while avoiding off-target effects. While many natural enzymes are synthesized as inactive zymogens that can be activated by proteolysis, it has been challenging to redesign any chosen enzyme to be similarly stimulus responsive. Here, we develop a massively parallel computational design, screening, and next-generation sequencing-based approach for proenzyme design. For a model system, we employ carboxypeptidase G2 (CPG2), a clinically approved enzyme that has applications in both the treatment of cancer and controlling drug toxicity. Detailed kinetic characterization of the most effectively designed variants shows that they are inhibited by ∼80% compared to the unmodified protein, and their activity is fully restored following incubation with site-specific proteases. Introducing disulfide bonds between the pro- and catalytic domains based on the design models increases the degree of inhibition to 98% but decreases the degree of restoration of activity by proteolysis. A selected disulfide-containing proenzyme exhibits significantly lower activity relative to the fully activated enzyme when evaluated in cell culture. Structural and thermodynamic characterization provides detailed insights into the prodomain binding and inhibition mechanisms. The described methodology is general and could enable the design of a variety of proproteins with precise spatial regulation.

Protease propeptide structures, mechanisms of activation, and functions.

Proteases are a diverse group of hydrolytic enzymes, ranging from single-domain catalytic molecules to sophisticated multi-functional macromolecules. Human proteases are divided into five mechanistic classes: aspartate, cysteine, metallo, serine and threonine proteases, based on the catalytic mechanism of hydrolysis. As a protective mechanism against uncontrolled proteolysis, proteases are often produced and secreted as inactive precursors, called zymogens, containing inhibitory N-terminal propeptides. Protease propeptide structures vary considerably in length, ranging from dipeptides and propeptides of about 10 amino acids to complex multifunctional prodomains with hundreds of residues. Interestingly, sequence analysis of the different protease domains has demonstrated that propeptide sequences present higher heterogeneity compared with their catalytic domains. Therefore, we suggest that protease inhibition targeting propeptides might be more specific and have less off-target effects than classical inhibitors. The roles of propeptides, besides keeping protease latency, include correct folding of proteases, compartmentalization, liganding, and functional modulation. Changes in the propeptide sequence, thus, have a tremendous impact on the cognate enzymes. Small modifications of the propeptide sequences modulate the activity of the enzymes, which may be useful as a therapeutic strategy. This review provides an overview of known human proteases, with a focus on the role of their propeptides. We review propeptide functions, activation mechanisms, and possible therapeutic applications.

Structure, interdomain dynamics, and pH-dependent autoactivation of pro-rhodesain, the main lysosomal cysteine protease from African trypanosomes.

Rhodesain is the lysosomal cathepsin L-like cysteine protease of Trypanosoma brucei rhodesiense, the causative agent of Human African Trypanosomiasis. The enzyme is essential for the proliferation and pathogenicity of the parasite as well as its ability to overcome the blood-brain barrier of the host. Lysosomal cathepsins are expressed as zymogens with an inactivating prodomain that is cleaved under acidic conditions. A structure of the uncleaved maturation intermediate from a trypanosomal cathepsin L-like protease is currently not available. We thus established the heterologous expression of T. brucei rhodesiense pro-rhodesain in Escherichia coli and determined its crystal structure. The trypanosomal prodomain differs from nonparasitic pro-cathepsins by a unique, extended α-helix that blocks the active site and whose side-chain interactions resemble those of the antiprotozoal inhibitor K11777. Interdomain dynamics between pro- and core protease domain as observed by photoinduced electron transfer fluorescence correlation spectroscopy increase at low pH, where pro-rhodesain also undergoes autocleavage. Using the crystal structure, molecular dynamics simulations, and mutagenesis, we identify a conserved interdomain salt bridge that prevents premature intramolecular cleavage at higher pH values and may thus present a control switch for the observed pH sensitivity of proenzyme cleavage in (trypanosomal) CathL-like proteases.

Proteome-wide covalent ligand discovery in native biological systems.

Small molecules are powerful tools for investigating protein function and can serve as leads for new therapeutics. Most human proteins, however, lack small-molecule ligands, and entire protein classes are considered 'undruggable'. Fragment-based ligand discovery can identify small-molecule probes for proteins that have proven difficult to target using high-throughput screening of complex compound libraries. Although reversibly binding ligands are commonly pursued, covalent fragments provide an alternative route to small-molecule probes, including those that can access regions of proteins that are difficult to target through binding affinity alone. Here we report a quantitative analysis of cysteine-reactive small-molecule fragments screened against thousands of proteins in human proteomes and cells. Covalent ligands were identified for >700 cysteines found in both druggable proteins and proteins deficient in chemical probes, including transcription factors, adaptor/scaffolding proteins, and uncharacterized proteins. Among the atypical ligand-protein interactions discovered were compounds that react preferentially with pro- (inactive) caspases. We used these ligands to distinguish extrinsic apoptosis pathways in human cell lines versus primary human T cells, showing that the former is largely mediated by caspase-8 while the latter depends on both caspase-8 and -10. Fragment-based covalent ligand discovery provides a greatly expanded portrait of the ligandable proteome and furnishes compounds that can illuminate protein functions in native biological systems.

Modifying a bacterial tyrosinase zymogen for use in protease activity assays.

Current clinical laboratory assays are not sufficient for determining the activity of many specific human proteases yet. In this study, we developed a general approach that enables the determination of activities of caspase-3 based on the proteolytic activation of the engineered zymogen of the recombinant tyrosinase from Verrucomicrobium spinosum (Vs-tyrosinase) by detecting the diphenolase activity in an increase in absorbance at 475 nm. Here, we designed three different zymogen constructs of Vs-tyrosinase, including RSL-pre-pro-TYR, Pre-pro-TYR, and Pro-TYR. The active domain was fused to the reactive site loop (RSL) of α1-proteinase inhibitor and/or its own signal peptide (pre) and/or its own C-terminal domain (pro) via a linker containing a specific caspase-3 cleavage site. Further studies revealed that both RSL peptide and TAT signal peptide were able to inhibit tyrosinase diphenolase activity, in which RSL-pre-pro-TYR had the lowest background signals. Therefore, a specific protease activity such as caspase-3 could be detected when a suitable zymogen was established. Our results could provide a new way to directly detect the activities of key human proteases, for instance, to monitor the efficacy and safety of tumor therapy by determining the activity of apoptosis-related caspase-3 in patients. KEY POINTS: • RSL inhibited the activity of Verrucomicrobium spinosum tyrosinase. • N-pre and C-terminal domain exerted stronger dual inhibition on the Vs-tyrosinase. • The activity of caspase-3 could be measured by the zymogen activation system.

The role of propeptide-mediated autoinhibition and intermolecular chaperone in the maturation of cognate catalytic domain in leucine aminopeptidase.

Leucyl aminopeptidase A from Aspergillus oryzae RIB40 (AO-LapA) is an exo-acting peptidase, widely utilised in food debittering applications. AO-LapA is secreted as a zymogen by the host and requires enzymatic cleavage of the autoinhibitory propeptide to reveal its full activity. Scarcity of structural data of zymogen aminopeptidases hampers a better understanding of the details of their molecular action of autoinhibition and how this might be utilised to improve the properties of such enzymes by recombinant methods for more effective bioprocessing. To address this gap in the literature, herein we report high-resolution crystal structures of recombinantly expressed AO-LapA precursor (AO-proLapA), mature LapA (AO-mLapA) and AO-mLapA complexed with reaction product l-leucine (AO-mLapA-Leu), all purified from Pichia pastoris culture supernatant. Our structures reveal a plausible molecular mechanism of LapA catalytic domain autoinhibition by propeptide and highlights the role of intramolecular chaperone (IMC). Our data suggest an absolute requirement for IMC in the maturation of cognate catalytic domain of AO-LapA. This observation is reinforced by our expression and refolding data of catalytic domain only (AO-refLapA) from Escherichia coli inclusion bodies, revealing a limited active conformation. Our work supports the notion that known synthetic aminopeptidase inhibitors and substrates mimic key polar contacts between propeptide and corresponding catalytic domain, demonstrated in our AO-proLapA zymogen crystal structure. Furthermore, understanding the atomic details of the autoinhibitory mechanism of cognate catalytic domains by native propeptides has wider reaching implications toward synthetic production of more effective inhibitors of bimetallic aminopeptidases and other dizinc enzymes that share an analogous reaction mechanism.

Zymogen and activated protein C have similar structural architecture.

Activated protein C is a trypsin-like protease with anticoagulant and cytoprotective properties that is generated by thrombin from the zymogen precursor protein C in a reaction greatly accelerated by the cofactor thrombomodulin. The molecular details of this activation remain elusive due to the lack of structural information. We now fill this gap by providing information on the overall structural organization of these proteins using single molecule FRET and small angle X-ray scattering. Under physiological conditions, both zymogen and protease adopt a conformation with all domains vertically aligned along an axis 76 Å long and maximal particle size of 120 Å. This conformation is stabilized by binding of Ca2+ to the Gla domain and is affected minimally by interaction with thrombin. Hence, the zymogen protein C likely interacts with the thrombin-thrombomodulin complex through a rigid body association that produces a protease with essentially the same structural architecture. This scenario stands in contrast to an analogous reaction in the coagulation cascade where conversion of the zymogen prothrombin to the protease meizothrombin by the prothrombinase complex is linked to a large conformational transition of the entire protein. The presence of rigid epidermal growth factor domains in protein C as opposed to kringles in prothrombin likely accounts for the different conformational plasticity of the two zymogens. The new structural features reported here for protein C have general relevance to vitamin K-dependent clotting factors containing epidermal growth factor domains, such as factors VII, IX, and X.

Roles of the clip domains of two protease zymogens in the coagulation cascade in horseshoe crabs.

The lipopolysaccharide (LPS)-triggered coagulation cascade in horseshoe crabs comprises three protease zymogens: prochelicerase C (proC), prochelicerase B (proB), and the proclotting enzyme (proCE). The presence of LPS results in autocatalytic activation of proC to α-chelicerase C, which, in turn, activates proB to chelicerase B, converting proCE to the clotting enzyme (CE). ProB and proCE contain an N-terminal clip domain, but the roles of these domains in this coagulation cascade remain unknown. Here, using recombinant proteins and kinetics and binding assays, we found that five basic residues in the clip domain of proB are required to maintain its LPS-binding activity and activation by α-chelicerase C. Moreover, an amino acid substitution at a potential hydrophobic cavity in proB's clip domain (V55A-proB) reduced both its LPS-binding activity and activation rate. WT proCE exhibited no LPS-binding activity, and the WT chelicerase B-mediated activation of a proCE variant with a substitution at a potential hydrophobic cavity (V53A-proCE) was ∼4-fold slower than that of WT proCE. The kcat/Km value of the interaction of WT chelicerase B with V53A-proCE was 7-fold lower than that of the WT chelicerase B-WT proCE interaction. The enzymatic activities of V55A-chelicerase B and V53A-CE against specific peptide substrates were indistinguishable from those of the corresponding WT proteases. In conclusion, the clip domain of proB recruits it to a reaction center composed of α-chelicerase C and LPS, where α-chelicerase C is ready to activate proB, leading to chelicerase B-mediated activation of proCE via its clip domain.

Saving the horseshoe crab: A synthetic alternative to horseshoe crab blood for endotoxin detection.

Horseshoe crabs have been integral to the safe production of vaccines and injectable medications for the past 40 years. The bleeding of live horseshoe crabs, a process that leaves thousands dead annually, is an ecologically unsustainable practice for all four species of horseshoe crab and the shorebirds that rely on their eggs as a primary food source during spring migration. Populations of both horseshoe crabs and shorebirds are in decline. This study confirms the efficacy of recombinant Factor C (rFC), a synthetic alternative that eliminates the need for animal products in endotoxin detection. Furthermore, our findings confirm that the biomedical industry can achieve a 90% reduction in the use of reagents derived from horseshoe crabs by using the synthetic alternative for the testing of water and other common materials used in the manufacturing process. This represents an extraordinary opportunity for the biomedical and pharmaceutical industries to significantly contribute to the conservation of horseshoe crabs and the birds that depend on them.

Blocking the proteolytic activity of zymogen matriptase with antibody-based inhibitors.

Matriptase is a member of the type-II transmembrane serine protease (TTSP) family and plays a crucial role in the development and maintenance of epithelial tissues. As all chymotrypsin-like serine proteases, matriptase is synthesized as a zymogen (proform), requiring a cleavage event for full activity. Recent studies suggest that the zymogen of matriptase possesses enough catalytic activity to not only facilitate autoactivation, but also carry out its in vivo functions, which include activating several proteolytic and signaling cascades. Inhibition of zymogen matriptase may therefore be a highly effective approach for limiting matriptase activity. To this end, here we sought to characterize the catalytic activity of human zymogen matriptase and to develop mAb inhibitors against this enzyme form. Using a mutated variant of matriptase in which the serine protease domain is locked in the zymogen conformation, we confirmed that the zymogen form of human matriptase has catalytic activity. Moreover, the crystal structure of the catalytic domain of zymogen matriptase was solved to 2.5 Å resolution to characterize specific antibody-based matriptase inhibitors and to further structure-based studies. Finally, we describe the first antibody-based competitive inhibitors that target both the zymogen and activated forms of matriptase. We propose that these antibodies provide a more efficient way to regulate matriptase activity by targeting the protease both before and after its activation and may be of value for both research and preclinical applications.

Occlusion of anion-binding exosite 2 in meizothrombin explains its impaired ability to activate factor V.

The proteolytic conversion of factor V to factor Va is central for amplified flux through the blood coagulation cascade. Heterodimeric factor Va is produced by cleavage at three sites in the middle of factor V by thrombin, yielding an N terminus-derived heavy chain and a C terminus-derived light chain. Here, we show that light chain formation resulting from the C-terminal cleavage is the rate-limiting step in the formation of fully cleaved Va. This rate-limiting step also corresponded to and was sufficient for the ability of cleaved factor V to bind Xa and assemble into the prothrombinase complex. Meizothrombin, the proteinase intermediate in thrombin formation, cleaves factor V more slowly than does thrombin, resulting in a pronounced defect in the formation of the light chain. A ∼100-fold reduced rate of meizothrombin-mediated light chain formation by meizothrombin corresponded to equally slow production of active cofactor and an impaired ability to amplify flux through the coagulation cascade initiated in plasma. We show that this defect arises from the occlusion of anion-binding exosite 2 in the catalytic domain by the covalently retained propiece in meizothrombin. Our findings provide structural insights into the prominent role played by exosite 2 in the rate-limiting step of factor V activation. They also bear on how factor V is converted into a cofactor capable of assembling into prothrombinase.

Recombinant production of active microbial transglutaminase in E. coli by using self-cleavable zymogen with mutated propeptide.

Microbial transglutaminase from Streptomyces mobaraensis (MTG) has been widely used in food industry and also in research and medical applications, since it can site-specifically modify proteins by the cross-linking reaction of glutamine residue and the primary amino group. The recombinant expression system of MTG in E. coli provides better accessibility for the researchers and thus can promote further utilization of MTG. Herein, we report production of active and soluble MTG in E. coli by using a chimeric protein of tobacco etch virus (TEV) protease and MTG zymogen. A chimera of TEV protease and MTG zymogen with native propeptide resulted in active MTG contaminated with cleaved propeptide due to the strong interaction between the propeptide and catalytic domain of MTG. Introduction of mutations of K9R and Y11A to the propeptide facilitated dissociation of the cleaved propeptide from the catalytic domain of MTG and active MTG without any contamination of the propeptide was obtained. The specific activity of the active MTG was 22.7 ± 2.6 U/mg. The successful expression and purification of active MTG by using the chimera protein of TEV protease and MTG zymogen with mutations in the propeptide can advance the use of MTG and the researches using MTG mediated cross-linking reactions.

Multiple Architectures and Mechanisms of Latency in Metallopeptidase Zymogens.

Metallopeptidases cleave polypeptides bound in the active-site cleft of catalytic domains through a general base/acid mechanism. This involves a solvent molecule bound to a catalytic zinc and general regulation of the mechanism through zymogen-based latency. Sixty reported structures from 11 metallopeptidase families reveal that prosegments, mostly N-terminal of the catalytic domain, block the cleft regardless of their size. Prosegments may be peptides (5-14 residues), which are only structured within the zymogens, or large moieties (<227 residues) of one or two folded domains. While some prosegments globally shield the catalytic domain through a few contacts, others specifically run across the cleft in the same or opposite direction as a substrate, making numerous interactions. Some prosegments block the zinc by replacing the solvent with particular side chains, while others use terminal α-amino or carboxylate groups. Overall, metallopeptidase zymogens employ disparate mechanisms that diverge even within families, which supports that latency is less conserved than catalysis.

Chemoproteomics Using Nucleotide Acyl Phosphates Reveals an ATP Binding Site at the Dimer Interface of Procaspase-6.

Acyl phosphates of ATP (ATPAc) and related nucleotides have proven to be useful for the interrogation of known nucleotide binding sites via specific acylation of conserved lysines (K). In addition, occasional K acylations are identified in proteins without such known sites. Here we present a robust and specific acylation of procaspase-6 by ATPAc at K133 in Jurkat cell lysates. The K133 acylation is dependent on π-π stacking interactions between the adenine moiety of ATPAc and a conserved Y198-Y198 site formed at the homodimeric interface of procaspase-6. Significantly, the Y198A mutation in procaspase-6 abolishes K133 acylation but has no effect on the proteolytic activity of the mature, active caspase-6 Y198A variant. Additional in vitro studies show that ATP can inhibit the autoproteolytic activation of procaspase-6. These observations suggest that ATP, and possibly other nucleotides, may serve as the endogenous ligands for the allosteric site at the procaspase-6 dimer interface, a site that has persisted in its "orphan" status for more than a decade.

Intermolecular autocatalytic activation of serine protease zymogen factor C through an active transition state responding to lipopolysaccharide.

Horseshoe crab hemolymph coagulation is believed to be triggered by the autocatalytic activation of serine protease zymogen factor C to the active form, α-factor C, belonging to the trypsin family, through an active transition state of factor C responding to bacterial lipopolysaccharide (LPS), designated factor C*. However, the existence of factor C* is only speculative, and its proteolytic activity has not been validated. In addition, it remains unclear whether the proteolytic cleavage of the Phe737-Ile738 bond (Phe737 site) of factor C required for the conversion to α-factor C occurs intramolecularly or intermolecularly between the factor C molecules. Here we show that the Phe737 site of a catalytic Ser-deficient mutant of factor C is LPS-dependently hydrolyzed by a Phe737 site-uncleavable mutant, clearly indicating the existence of the active transition state of factor C without cleavage of the Phe737 site. Moreover, we found the following facts using several mutants of factor C: the autocatalytic cleavage of factor C occurs intermolecularly between factor C* molecules on the LPS surface; factor C* does not exhibit intrinsic chymotryptic activity against the Phe737 site, but it may recognize a three-dimensional structure around the cleavage site; and LPS is required not only to complete the substrate-binding site and oxyanion hole of factor C* by interacting with the N-terminal region but also to allow the Phe737 site to be cleaved by inducing a conformational change around the Phe737 site or by acting as a scaffold to induce specific protein-protein interactions between factor C* molecules.

A statistical model for activation of Factor C by binding to LPS aggregates.

Published data on Factor C activity at various LPS and Lipid A concentrations (Nakamura et al. in Eur J Biochem 176:89, 1988; Kobayashi et al. in J Biol Chem 37:25987, 2014) were rearranged to show that Factor C exhibited its maximum activity at a specific concentration of LPS. A statistical model was proposed for examining whether a single LPS molecule binding activates Factor C (monomeric activation) or dimerization of Factor C is necessary for the activation (dimeric activation). In the monomeric activation model the plots of the relative activity of Factor C against the molar ratio of LPS to Factor C were different from those in the published data. The plots in the dimeric activation model lie on a bell-shaped curve, whatever the Factor C concentration, matching the published data and indicating the appropriateness of that model. We suggest that Factor C is activated by multiple molecular interactions of Factor C with LPS aggregates on which it dimerises and that this explains why larger aggregates are less effective at activating Factor C than smaller ones.

Comparing sequence and structure of falcipains and human homologs at prodomain and catalytic active site for malarial peptide based inhibitor design.

BACKGROUND: Falcipains are major cysteine proteases of Plasmodium falciparum involved in haemoglobin degradation and remain attractive anti-malarial drug targets. Several inhibitors against these proteases have been identified, yet none of them has been approved for malaria treatment. Other Plasmodium species also possess highly homologous proteins to falcipains. For selective therapeutic targeting, identification of sequence and structure differences with homologous human cathepsins is necessary. The substrate processing activity of these proteins is tightly controlled via a prodomain segment occluding the active site which is chopped under low pH conditions exposing the catalytic site. Current work characterizes these proteases to identify residues mediating the prodomain regulatory function for the design of peptide based anti-malarial inhibitors. METHODS: Sequence and structure variations between prodomain regions of plasmodial proteins and human cathepsins were determined using in silico approaches. Additionally, evolutionary clustering of these proteins was evaluated using phylogenetic analysis. High quality partial zymogen protein structures were modelled using homology modelling and residue interaction analysis performed between the prodomain segment and mature domain to identify key interacting residues between these two domains. The resulting information was used to determine short peptide sequences which could mimic the inherent regulatory function of the prodomain regions. Through flexible docking, the binding affinity of proposed peptides on the proteins studied was evaluated. RESULTS: Sequence, evolutionary and motif analyses showed important differences between plasmodial and human proteins. Residue interaction analysis identified important residues crucial for maintaining prodomain integrity across the different proteins as well as the pro-segment responsible for inhibitory mechanism. Binding affinity of suggested peptides was highly dependent on their residue composition and length. CONCLUSIONS: Despite the conserved structural and catalytic mechanism between human cathepsins and plasmodial proteases, current work revealed significant differences between the two protein groups which may provide valuable information for selective anti-malarial inhibitor development. Part of this study aimed to design peptide inhibitors based on endogenous inhibitory portions of protease prodomains as a novel aspect. Even though peptide inhibitors may not be practical solutions to malaria at this stage, the approach followed and results offer a promising means to find new malarial inhibitors.

A highly sensitive endotoxin sensor based on redox cycling in a nanocavity.

We report a highly sensitive and rapid electrochemical method for the detection of endotoxin, based on a Limulus amebocyte lysate (LAL) assay using redox cycling at a pair of electrodes in a nanocavity for electrochemical signal amplification. We have previously developed Boc-Leu-Gly-Arg-p-aminophenol (LGR-pAP) as a substrate for the amperometric LAL assay, and in this work, Z-Leu-Gly-Arg-aminomethylferrocene (LGR-AMF) was newly prepared. They were examined as substrates for a LAL-based endotoxin assay using a nanocavity device. During the last step of the endotoxin-induced LAL cascade reaction, pAP or AMF is generated from the substrate, which can be detected electrochemically with efficient signal amplification by redox cycling between the two electrodes in the nanocavity. A device with a 190 nm-high nanocavity was fabricated by photolithography. With the fabricated device in model assay solutions prepared by mixing LGR-pAP and pAP, we demonstrated that pAP could be quantitatively detected from the difference in oxidation potentials between LGR-pAP and pAP. For LGR-AMF and AMF, a difference in the formal potential of 0.1 V was obtained which was considered to be insufficient to distinguish AMF from LGR-AMF. However, we showed for the first time that analytes such as AMF can be detected by differences in diffusion coefficients between the analyte and coexisting molecules (such as LGR-AMF) using a device with high redox-cycling efficiency. Next, the endotoxin assay was performed using the fabricated nanocavity device. Using this method, endotoxin was detected at concentrations as low as 0.2 and 0.5 EU L-1 after LAL reaction times of 1 h and 30 min, respectively, using the LGR-pAP substrate. However, the endotoxin assay using LGR-AMF was not successful because the clotting enzyme did not react with LGR-AMF. This problem might be solved by further design of the substrate. Our nanocavity device represents an effective platform for the simple and rapid detection of endotoxin with high sensitivity.