Inhibition of papain-like cysteine proteases and legumain by caspase-specific inhibitors: when reaction mechanism is more important than specificity.

We report here that a number of commonly used small peptide caspase inhibitors consisting of a caspase recognition sequence linked to chloromethylketone, fluoromethylketone or aldehyde reactive group efficiently inhibit other cysteine proteases than caspases. The in vitro studies included cathepsins B, H, L, S, K, F, V, X and C, papain and legumain. Z-DEVD-cmk was shown to be the preferred irreversible inhibitor of most of the cathepsins in vitro, followed by Z-DEVD-fmk, Ac-YVAD-cmk, Z-YVAD-fmk and Z-VAD-fmk. Inactivation of legumain by all the inhibitors investigated was moderate, whereas cathepsins H and C were poorly inhibited or not inhibited at all. Inhibition by aldehydes was not very potent. All the three fluoromethylketones efficiently inhibited cathepsins in Jurkat and human embryonic kidney 293 cells at concentrations of 100 microM. Furthermore, they completely inhibited cathepsins B and X activity in tissue extracts at concentrations as low as 1 microM. These results suggest that data based on the use of these inhibitors should be taken with caution and that other proteases may be implicated in the processes previously ascribed solely to caspases.

Human recombinant pro-dipeptidyl peptidase I (cathepsin C) can be activated by cathepsins L and S but not by autocatalytic processing.

Human dipeptidyl peptidase I was expressed in the insect cell/baculovirus system and purified in its active (rhDPPI) and precursor (pro-rhDPPI) forms. RhDPPI was very similar to the purified enzyme (hDPPI) with respect to glycosylation, enzymatic processing, oligomeric structure, CD spectra, and catalytic activity. The precursor, which was a dimer, could be activated approximately 2000-fold with papain. Cathepsin L efficiently activated pro-rhDPPI in vitro at pH 4.5 (k(app) approximately 2 x 10(3) min(-)(1) M(-)(1)), and two cleavage pathways were characterized. The initial cleavage was within the pro region between the residual pro part and the activation peptide. Subsequently, the activation peptide was cleaved from the catalytic region, and the latter was cleaved into the heavy and light chains. Alternatively, the pro region was first separated from the catalytic region. Cathepsin S was a less efficient activating enzyme. Cathepsin B and rhDPPI did not activate pro-rhDPPI, and the proenzyme was incapable of autoactivation. Incubation of both pro-rhDPPI and rhDPPI with cathepsin D resulted in degradation. Cystatin C and stefins A and B inhibited rhDPPI with K(i) values in the nanomolar range (K(i) = 0.5-1.1 nM). The results suggest that cathepsin L could be an important activator of DPPI in vivo and that cathepsin D and possibly the cystatins may contribute to DPPI downregulation.

Interaction of cystatin C variants with papain and human cathepsins B, H and L.

Recombinant human cystatin C and two of its mutants were expressed in Escherichia coli. The recombinant inhibitor was found to be identical to authentic cystatin C as judged by isoelectric focusing (pI 9.2) and kinetics of inhibition of papain and human cathepsins B, H and L. N-terminal truncation of 8 residues resulted in a decrease of isoelectric point (pI 7.8), but the inhibitory properties were similar to those of recombinant cystatin C, suggesting that Leu9 is a critical residue for the inhibition. The mutation of Trp106 to Ser, however, resulted in a decreased affinity of the inhibitor for the enzymes tested, with the largest effect on cathepsin B inhibition (approximately 100-fold increase in Ki).

Cathepsin L is capable of truncating cystatin C of 11 N-terminal amino acids.

Cystatin C with the 11 N-terminal amino acids truncated shows a much lower affinity for cysteine proteinases than the intact inhibitor. Such truncation of cystatin C is recorded after action of glycyl endopeptidase and cathepsin L. Incubation of cystatin C with papain, cathepsin B or cathepsin H led to no changes in the cystatin C molecule. Isoelectric focusing of the cathepsin L and cystatin C mixture showed the formation of two new bands. One of them appeared whether E-64 or PMSF was added or not, evidently representing a cystatin C/cathepsin L complex. The other band is the truncated cystatin C molecule. N-terminal sequencing after separation by HPLC showed that cystatin C is cleaved by cathepsin L at the Gly11-Gly12 bond. The action of cathepsin L on cystatin C may be explained by the cleavage of the scissile bond in an inappropriate complex.

Acidic pH as a physiological regulator of human cathepsin L activity.

Human cysteine protease cathepsin L was inactivated at acid pH by a first-order process. The inactivation rate decreased with increasing concentrations of a small synthetic substrate, suggesting that substrates stabilize the active conformation. The substrate-independent inactivation rate constant increased with organic solvent content of the buffer, consistent with internal hydrophobic interactions, disrupted by the organic solvent, also stabilizing the enzyme. Circular dichroism showed that the inactivation is accompanied by large structural changes, a decrease in alpha-helix content being especially pronounced. The high activation energy of the reaction at pH 3.0 (200 kJ.mol-1) supported such a major conformational change occurring. The acid inactivation of cathepsin L was irreversible, consistent with the propeptide being needed for proper folding of the enzyme. Aspartic protease cathepsin D was shown to cleave denatured, but not active cathepsin L, suggesting a potential mechanism for in-vivo regulation and turnover of cathepsin L inside lysosomes.

Decelerated degradation of short peptides by the 20S proteasome.

Based on a twelve residue master peptide comprising all five specific cleavage sites defined for the proteasome, a set of variant peptides was generated in order to probe specificity and to elucidate the mechanism which determines product size. It is shown that the rate of degradation by the 20S proteasome from Thermoplasma acidophilum depends critically on the length of the peptide substrate. Peptides of 14 residues and longer are degraded much faster than shorter peptides although the sites of cleavage remain unchanged. The decelerated degradation of peptides shorter than 14 residues explains the accumulation of products with an average length of seven to nine residues.

Subunit topology of the Rhodococcus proteasome.

The 20S proteasome, isolated from the nocardioform actinomycete Rhodococcus erythropolis strain NI86/21, is built from two alpha-type and two beta-type subunits. In order to probe the subunit topology, we have set up an expression system which allows coexpression of the genes encoding the alpha- and beta-subunits in all possible combinations. The four respective constructs obtained yielded fully assembled and proteolytically active proteasomes. Biochemical, kinetic and electron microscopy analysis allow us to rule out several of the models which were originally envisaged for the subunit topology of the Rhodococcus proteasome. The experiments further indicate that the assembly pathways of the Rhodococcus and of the Thermoplasma proteasome differ in some important details.

Regenerated rat fast muscle transplanted to the slow muscle bed and innervated by the slow nerve, exhibits an identical myosin heavy chain repertoire to that of the slow muscle.

The hypothesis that the limited adaptive range observed in fast rat muscles in regard to expression of the slow myosin is due to intrinsic properties of their myogenic stem cells was tested by examining myosin heavy chain (MHC) expression in regenerated rat extensor digitorum longus (EDL) and soleus (SOL) muscles. The muscles were injured by bupivacaine, transplanted to the SOL muscle bed and innervated by the SOL nerve. Three months later, muscle fibre types were determined. MHC expression in muscle fibres was demonstrated immunohistochemically and analysed by SDS-glycerol gel electrophoresis. Regenerated EDL transplants became very similar to the control SOL muscles and indistinguishable from the SOL transplants. Slow type 1 fibres predominated and the slow MHC-1 isoform was present in more than 90% of all muscle fibres. It contributed more than 80% of total MHC content in the EDL transplants. About 7% of fibres exhibited MHC-2a and about 7% of fibres coexpressed MHC-1 and MHC-2a. MHC-2x/d contributed about 5-10% of the whole MHCs in regenerated EDL and SOL transplants. The restricted adaptive range of adult rat EDL muscle in regard to the synthesis of MHC-1 is not rooted in muscle progenitor cells; it is probably due to an irreversible maturation-related change switching off the gene for the slow MHC isoform.

Interaction of human cathepsin C with chicken cystatin.

Cathepsin C was purified from human spleen by a rapid procedure, which included homogenization, ammonium sulfate precipitation, gel filtration on Sephacryl S-200 and finally affinity chromatography on chicken cystatin-Sepharose. The interaction between cathepsin C and chicken cystatin was further characterized. It was found to be accompanied by a maximum decrease in fluorescence emission intensity at 336 nm. Fluorescence titration showed that human cathepsin C can bind four chicken cystatin molecules. The 4:1 binding stoichiometry was confirmed by titration monitored by the loss of enzyme activity. A non-competitive-competitive type of inhibition was determined from a double-reciprocal Lineweaver-Burk plot with a Ki value of 0.22 nM for the non-competitive inhibition.

Adaptive range of myosin heavy chain expression in regenerating soleus is broader than in mature muscle.

In adult rat muscles experimentally exposed to various patterns of activation, expression of myosin heavy chain isoforms changes, but only within a certain adaptive range. It is characteristic and different in fast or slow muscles. This may be due either to different intrinsic properties of the myogenic cells of the two types of muscles or to extrinsic factors. To test these assumptions, either rat soleus or extensor digitorum longus muscles were injured and transplanted to the bed of the extensor digitorum longus muscle. They regenerated and were reinnervated by the extensor digitorum longus nerve. Expression of myosin heavy chain isoforms was demonstrated immunohistochemically and by in situ hybridization, and analysed by SDS-gel electrophoresis. Three months after cross-transplantation, regenerated soleus expressed all adult myosin heavy chain isoforms, including the myosin heavy chain-2B. The latter was detected in about 50% of muscle fibres and contributed about 10-20% of all myosin heavy chains. The same percentage of myosin heavy chain-2B was found in regenerated extensor digitorum longus. In this regard therefore, the adaptive range of the regenerated soleus muscle was not significantly different from that of the extensor digitorum longus regenerating under the same conditions. This indicates that restriction of the adaptive range in a mature soleus muscle is not due to intrinsic properties of its myogenic cells. It is probably imposed by an extrinsic factor leading to irreversible shut-down of individual myosin heavy chain genes. On the other hand, myosin heavy chain-1 expression was significantly greater in the regenerated soleus than in the extensor digitorum longus innervated by the same nerve. Myosin heavy chain-1 and myosin heavy chain-2B were co-expressed in some regenerated soleus muscle fibres.

Major histocompatibility complex class II-associated p41 invariant chain fragment is a strong inhibitor of lysosomal cathepsin L.

The invariant chain (Ii) is associated with major histocompatibility complex class II molecules during early stages of their intracellular transport. In an acidic endosomal/lysosomal compartment, it is proteolytically cleaved and removed from class II heterodimers. Participation of aspartic and cysteine proteases has been observed in in vitro degradation of Ii, but the specific enzymes responsible for its in vivo processing are as yet undefined. We have previously isolated a noncovalent complex of the lysosomal cysteine protease cathepsin L with a peptide fragment derived from the p41 form of Ii from human kidney. Here we show that this Ii fragment, which is identical to the alternatively spliced segment of p41, is a very potent competitive inhibitor of cathepsin L (equilibrium inhibition constant Ki = 1.7 X 10(-12) M). It inhibits two other cysteine proteases, cathepsin H and papain, but to much lesser extent. Cysteine proteases cathepsins B, C, and S, as well as representatives of serine, aspartic, and metalloproteases, are not inhibited at all. These findings suggest a novel role for p41 in the regulation of various proteolytic activities during antigen processing and presentation. The Ii inhibitory fragment shows no sequence homology with the known cysteine protease inhibitors, and may, therefore, represent a new class.

Oligomeric structure and substrate induced inhibition of human cathepsin C.

Cathepsin C has been purified from human kidney by a modified procedure. Human cathepsin C was isolated as pure protein with a pI close to 6.0. The enzyme was shown to have a molecular mass of 200 kDa and to consist of four identical subunits, each composed of three different polypeptide chains, two of them disulfide-bound. Their NH2-terminal amino acid sequences were determined. Two chains showed pronounced similarity with the heavy and light chains of other papain-like cysteine proteinases, whereas the third one corresponded to the prosequence of the enzyme, thus showing that a substantial part of the proregion remains bound in the mature enzyme. The kinetics of substrate hydrolysis deviated substantially from standard Michaelis-Menten kinetics, demonstrating substrate inhibition at higher substrate concentrations. These data are explained by a sequential cooperative interaction model, where an enzyme molecule can bind up to four substrate molecules but where only the binary enzyme-substrate complex is catalytically active. Substrate inhibition was observed over the whole range of pH activity. From the pH activity profile it can be concluded that at least three ionizable groups with pKa values 4.2, 6.8, and 7.7 are involved in substrate hydrolysis. Human cathepsin C thus appears to differ qualitatively from other cysteine proteinases of different origin.

High-affinity binding of two molecules of cysteine proteinases to low-molecular-weight kininogen.

Human low-molecular-weight kininogen (LK) was shown by fluorescence titration to bind two molecules of cathepsins L and S and papain with high affinity. By contrast, binding of a second molecule of cathepsin H was much weaker. The 2:1 binding stoichiometry was confirmed by titration monitored by loss of enzyme activity and by sedimentation velocity experiments. The kinetics of binding of cathepsins L and S and papain showed the two proteinase binding sites to have association rate constants kass,1 = 10.7-24.5 x 10(6) M-1 s-1 and kass,2 = 0.83-1.4 x 10(6) M-1 s-1. Comparison of these kinetic constants with previous data for intact LK and its separated domains indicate that the faster-binding site is also the tighter-binding site and is present on domain 3, whereas the slower-binding, lower-affinity site is on domain 2. These results also indicate that there is no appreciable steric hindrance for the binding of proteinases between the two binding sites or from the kininogen light chain.

Molecular cloning and sequence analysis of human preprocathepsin C.

A cDNA clone (C1) coding for human preprocathepsin C was isolated from a human ileum cDNA library using a rat kidney-derived RT-PCR probe and its complete nucleotide sequence determined. The full-length 1857 bp sequence codes for a protein of 463 amino acid residues with a calculated molecular mass of 51848 Da. Comparison of the deduced amino acid sequence with that of rat preprocathepsin C indicates an 87.5% identity. A multiple alignment of the deduced cathepsin C sequence of 233 residues which, by analogy to other cystein proteinases, corresponds to the mature protein, confirms that human cathepsin C belongs to the papain superfamily.

Regulation of the activity of lysosomal cysteine proteinases by pH-induced inactivation and/or endogenous protein inhibitors, cystatins.

The kinetics of pH-induced inactivation of human cathepsins B and L was studied by conventional and stopped-flow methods. The inactivation of both enzymes was found to be an irreversible, first-order process. The inactivation rate constants increased exponentially with pH for both enzymes. From log kinac vs pH plots, 3.0 and 1.7 protons were calculated to be desorbed for pH-induced inactivation of cathepsins L and B. Cathepsin B was thus substantially more stable than cathepsin L (approximately 15-fold at pH 7.0 and 37 degrees C). Cathepsin B was efficiently inhibited by cystatin C at pH 7.4, whereas the inhibition by stefin B and high molecular weight kininogen was only moderate. In contrast, cathepsin L was efficiently inhibited by both chicken cystatin and stefin B at this pH kass approximately 3.3 x 10(7) m-1 s-1).

Identification of bovine stefin A, a novel protein inhibitor of cysteine proteinases.

For the first time, three different stefins, A, B and C, have been isolated from a single species. The complete amino acid sequence of bovine stefin A was determined. The inhibitor, with a calculated M(r) of 11,123, consists of 98 amino acid residues. Although it exhibits considerable similarity to human and rat stefin A, some significant differences in inhibition kinetics were found. Bovine stefin A bound tightly and rapidly to cathepsin L (kass = 9.6 x 10(6) M-1.s-1, Ki = 29 pM). The binding to cathepsin H was also rapid (kass = 2.1 x 10(6) M-1.s-1), but weaker (Ki = 0.4 nM) due to a higher dissociation rate constant. In contrast, the binding to cathepsin B was much slower (kass = 1.4 x 10(5) M-1.s-1), but still tight (Ki = 1.9 nM).

Human cathepsin B is a metastable enzyme stabilized by specific ionic interactions associated with the active site.

The effect of neutral or alkaline pH on cathepsin B activity and structure was investigated. An irreversible loss of activity, accompanied by large structural changes, was observed at pH > or = 7.0. The high activation energy of 183.5 kJ mol-1, calculated for the inactivation process, is in good agreement with structural changes observed by circular dichroism. Both the pH-induced inactivation and the pH-induced unfolding of cathepsin B were found to be first-order processes, exponentially increasing with increasing pH of the solution. The good agreement of the rate constants of inactivation and unfolding of the enzyme indicates an important structure-function relationship. Cathepsin B was also found to be destabilized both by increasing ionic strength and organic solvent content.

Satellite cells in slow and fast rat muscles differ in respect to acetylcholinesterase regulation mechanisms they convey to their descendant myofibers during regeneration.

The hypothesis of satellite cell diversity in slow and fast mammalian muscles was tested by examining acetylcholinesterase (AChE) regulation in muscles regenerating 1) under conditions of muscle disuse (tenotomy, leg immobilization) in which the pattern of neural stimulation is changed, and 2) after cross-transplantation when the regenerating muscle develops under a foreign neural stimulation pattern. Soleus (SOL) and extensor digitorum longus (EDL) muscles of the rat were allowed to regenerate after ischemic-toxic injury either in their own sites or had been cross-transplanted to the site of the other muscle. Molecular forms of AChE in regenerating muscles were analyzed by velocity sedimentation in linear sucrose gradients. Neither tenotomy nor limb immobilization significantly affected the characteristic pattern of AChE molecular forms in regenerating SOL muscles, suggesting that the neural stimulation pattern is probably not decisive for its induction. During an early phase of regeneration, the general pattern of AChE molecular forms in the cross-transplanted regenerating muscle was predominantly determined by the type of its muscle of origin, and much less by the innervating nerve which exerted only a modest modifying effect. However, alkali-resistant myofibrillar ATPase activity on which the separation of muscle fibers into type I and type II is based, was determined predominantly by the motor nerve innervating the regenerating muscle. Mature regenerated EDL muscles (13 weeks after injury) which had been innervated by the SOL nerve became virtually indistinguishable from the SOL muscles in regard to their pattern of AChE molecular forms. However, AChE patterns of mature regenerated SOL muscles that had been innervated by the EDL nerve still displayed some features of the SOL pattern. In regard to AChE regulation, muscle satellite cells from slow or fast rat muscles convey to their descendant myotubes the information shifting their initial development in the direction of either slow or fast muscle, respectively. The satellite cells in fast or slow muscles are, therefore, intrinsically different. Intrinsic information is expressed mostly during an early phase of regeneration whereas later on the regulatory influence of the motor nerve more or less predominates.

Purification of the complex of cathepsin L and the MHC class II-associated invariant chain fragment from human kidney.

The complex of cathepsin L and the fragment of the MHC class II-associated invariant chain was purified from human kidney. M(r) of the complex, as determined by gel filtration, is about 40,000. Both components were identified by amino acid and sequence analyses. The bound invariant chain fragment is almost identical to the additional segment found in p41, but not in the p31 form of the invariant chain. The complex has significantly enhanced stability at neutral and slightly alkaline pH, and reduced proteolytic activity against the synthetic substrate Z-Phe-Arg-MCA compared to free cathepsin L. The complex exhibits no enzymatic activity against the protein substrate azocasein. For the first time, the invariant chain was found in a complex with a protein, which was not an MHC molecule.

Pig leukocyte cysteine proteinase inhibitor (PLCPI), a new member of the stefin family.

A new stefin type low-M(r) cysteine proteinase inhibitor (PLCPI) was isolated from pig polymorphonuclear leukocytes as a contaminant of the cathelin sample. The inhibitor consists of 103 amino acids, and its M(r) was calculated to be 11,768. The inhibitor exhibits considerable sequence identity with inhibitors from the stefin family, particularly with human stefin A. The PLCPI is a fast acting inhibitor of papain and cathepsins L and S (k(ass) > or = 1 x 10(6) M-1 x s-1) and forms very tight complexes with these enzymes (Ki < or = 190 pM). The affinity for cathepsins B and H (Ki > or = 125 nM) was lower. These results also show that the inhibitory activity previously ascribed to cathelin was due to the presence of PLCPI.