In Vitro Characterization of VPS34 Lipid Kinase Inhibition by Small Molecules.

VPS34 is a class III phosphoinositide 3-kinase that acts on vesicle trafficking. This kinase has recently attracted significant attention because of the function it plays in the machinery involved in the early steps of autophagy. Moreover, because significant progress had been made in the optimization of specific kinase inhibitors, its potential to be targeted by catalytic inhibitors has been investigated by different groups. The aim of this review is to present the key in vitro assays necessary for characterizing inhibitors of the catalytic activity of VPS34. The review covers catalytic (IC50 on purified recombinant protein) and binding assays (KD, ka, kd on purified recombinant protein), and a cell-based assay (IC50 in GFP-FYVE expressing cell line). The methodology for crystallization of VPS34 protein is also presented as it can provide guidance for the design by medicinal chemistry of small molecular mass kinase inhibitor.

Crystal structures of two human pyrophosphorylase isoforms in complexes with UDPGlc(Gal)NAc: role of the alternatively spliced insert in the enzyme oligomeric assembly and active site architecture.

The recently published human genome with its relatively modest number of genes has highlighted the importance of post-transcriptional and post-translational modifications, such as alternative splicing or glycosylation, in generating the complexities of human biology. The human UDP-N-acetylglucosamine (UDPGlcNAc) pyrophosphorylases AGX1 and AGX2, which differ in sequence by an alternatively spliced 17 residue peptide, are key enzymes synthesizing UDPGlcNAc, an essential precursor for protein glycosylation. To better understand the catalytic mechanism of these enzymes and the role of the alternatively spliced segment, we have solved the crystal structures of AGX1 and AGX2 in complexes with UDPGlcNAc (at 1.9 and 2.4 A resolution, respectively) and UDPGalNAc (at 2.2 and 2.3 A resolution, respectively). Comparison with known structures classifies AGX1 and AGX2 as two new members of the SpsA-GnT I Core superfamily and, together with mutagenesis analysis, helps identify residues critical for catalysis. Most importantly, our combined structural and biochemical data provide evidence for a change in the oligomeric assembly accompanied by a significant modification of the active site architecture, a result suggesting that the two isoforms generated by alternative splicing may have distinct catalytic properties.

Crystal structure of Streptococcus pneumoniae N-acetylglucosamine-1-phosphate uridyltransferase bound to acetyl-coenzyme A reveals a novel active site architecture.

The bifunctional bacterial enzyme N-acetyl-glucosamine-1-phosphate uridyltransferase (GlmU) catalyzes the two-step formation of UDP-GlcNAc, a fundamental precursor in bacterial cell wall biosynthesis. With the emergence of new resistance mechanisms against beta-lactam and glycopeptide antibiotics, the biosynthetic pathway of UDP-GlcNAc represents an attractive target for drug design of new antibacterial agents. The crystal structures of Streptococcus pneumoniae GlmU in unbound form, in complex with acetyl-coenzyme A (AcCoA) and in complex with both AcCoA and the end product UDP-GlcNAc, have been determined and refined to 2.3, 2.5, and 1.75 A, respectively. The S. pneumoniae GlmU molecule is organized in two separate domains connected via a long alpha-helical linker and associates as a trimer, with the 50-A-long left-handed beta-helix (LbetaH) C-terminal domains packed against each other in a parallel fashion and the C-terminal region extended far away from the LbetaH core and exchanged with the beta-helix from a neighboring subunit in the trimer. AcCoA binding induces the formation of a long and narrow tunnel, enclosed between two adjacent LbetaH domains and the interchanged C-terminal region of the third subunit, giving rise to an original active site architecture at the junction of three subunits.

Dissection of the bifunctional Escherichia coli N-acetylglucosamine-1-phosphate uridyltransferase enzyme into autonomously functional domains and evidence that trimerization is absolutely required for glucosamine-1-phosphate acetyltransferase activity and cell growth.

The bifunctional N-acetylglucosamine-1-phosphate uridyltransferase (GlmU) enzyme catalyzes both the acetylation of glucosamine 1-phosphate and the uridylation of N-acetylglucosamine 1-phosphate, two subsequent steps in the pathway for UDP-N-acetylglucosamine synthesis in bacteria. In our previous work describing its initial characterization in Escherichia coli, we proposed that the 456-amino acid (50.1 kDa) protein might possess separate uridyltransferase (N-terminal) and acetyltransferase (C-terminal) domains. In the present study, we confirm this hypothesis by expression of the two independently folding and functional domains. A fragment containing the N-terminal 331 amino acids (Tr331, 37.1 kDa) has uridyltransferase activity only, with steady-state kinetic parameters similar to the full-length protein. Further deletion of 80 amino acid residues at the C terminus results in a 250-amino acid fragment (28.6 kDa) still exhibiting significant uridyltransferase activity. Conversely, a fragment containing the 233 C-terminal amino acids (24.7 kDa) exhibits acetyltransferase activity exclusively. None of these individual domains could complement a chromosomal glmU mutation, indicating that each of the two activities is essential for cell viability. Analysis of truncated GlmU proteins by gel filtration further localizes regions of the protein involved in its trimeric organization. Interestingly, overproduction of the truncated Tr331 protein in a wild-type strain results in a rapid depletion of endogenous acetyltransferase activity, an arrest of peptidoglycan synthesis and cell lysis. It is shown that the acetyltransferase activity of the full-length protein is abolished once trapped within heterotrimers formed in presence of the truncated protein, suggesting that this enzyme activity absolutely requires a trimeric organization and that the catalytic site involves regions of contact between adjacent monomers. Data are discussed in connection with the recently obtained crystal structure of the truncated Tr331 protein.

Autophosphorylation of phosphoglucosamine mutase from Escherichia coli.

Phosphoglucosamine mutase (GlmM) catalyzes the formation of glucosamine-1-phosphate from glucosamine-6-phosphate, an essential step in the pathway for UDP-N-acetylglucosamine biosynthesis in bacteria. This enzyme must be phosphorylated to be active and acts according to a ping-pong mechanism involving glucosamine-1, 6-diphosphate as an intermediate (L. Jolly, P. Ferrari, D. Blanot, J. van Heijenoort, F. Fassy, and D. Mengin-Lecreulx, Eur. J. Biochem. 262:202-210, 1999). However, the process by which the initial phosphorylation of the enzyme is achieved in vivo remains unknown. Here we show that the phosphoglucosamine mutase from Escherichia coli autophosphorylates in vitro in the presence of [(32)P]ATP. The same is observed with phosphoglucosamine mutases from other bacterial species, yeast N-acetylglucosamine-phosphate mutase, and rabbit muscle phosphoglucomutase. Labeling of the E. coli GlmM enzyme with [(32)P]ATP requires the presence of a divalent cation, and the label is subsequently lost when the enzyme is incubated with either of its substrates. Analysis of enzyme phosphorylation by high-pressure liquid chromatography and coupled mass spectrometry confirms that only one phosphate has been covalently linked to the enzyme. Only phosphoserine could be detected after acid hydrolysis of the labeled protein, and site-directed mutagenesis of serine residues located in or near the active site identifies the serine residue at position 102 as the site of autophosphorylation of E. coli GlmM.

Reaction mechanism of phosphoglucosamine mutase from Escherichia coli.

The phosphoglucosamine mutase (GlmM) from Escherichia coli, specifically required for the interconversion of glucosamine-6-phosphate and glucosamine-1-phosphate (an essential step in the pathway for cell-wall peptidoglycan and lipopolysaccharide biosyntheses) was purified to homogeneity and its kinetic properties were investigated. The enzyme was active in a phosphorylated form and catalysed its reaction according to a classical ping-pong bi-bi mechanism. The dephosphorylated and phosphorylated forms of GlmM could be separated by HPLC and coupled MS showed that only one phosphate was covalently linked to the active site of the enzyme. The site of phosphorylation was clearly identified as Ser102 in the 445-amino acid polypeptide. GlmM was also capable of catalysing the interconversion of glucose-1-phosphate and glucose-6-phosphate isomers, although at a much lower (1400-fold) rate. Interestingly, the mutational change of the Ser100 to a threonine residue resulted in a 20-fold increase of the nonspecific phosphoglucomutase activity of GlmM, suggesting that the presence of either a serine or a threonine at this position in the consensus sequence of hexosephosphate mutases could be one of the factors that determines the specificity of these enzymes for either sugar-phosphate or amino sugar-phosphate substrates.

Enzymatic activity of two caspases related to interleukin-1beta-converting enzyme.

Interleukin-1beta-converting enzyme is a member of a family of human cysteine proteases with specificity for aspartic acid, which have been named caspases. Within this family of enzymes, transcript X (TX) and transcript Y (TY) (caspases 4 and 5, respectively) are very similar to ICE (caspase 1) and form the ICE subfamily. Given the high degree of conservation in the sequences of these proteases (more than 50% amino acid identity in the mature enzymes), it was of interest to examine whether they shared similar substrate specificities. The three enzymes, ICE, TX and TY, were therefore expressed in baculovirus-infected insect cells, as 30-kDa proteins lacking the propeptide. Automaturation into p20 and p10 subunits occurred within the cells. Active ICE, TX and TY were collected in the cell culture supernatants. In addition, their production induced the activation of an endogenous 32-kDa putative cysteine protease (CPP32) like caspase. T7-tagged ICE, TX and TY were purified by immunoaffinity and tested for their catalytic efficiency on YVAD-containing synthetic substrates and on the ICE natural substrate, pro-interleukin-1beta. TX cleaved the same synthetic substrates as ICE (Km of 90 microM and k(cat) of 0.4 s(-1) for Suc-YVAD-NH-Mec, where Suc represents succinyl and NH-Mec represents amino-4-methylcoumarin) and could cleave pro-interleukin-1beta into the same peptides as ICE but less efficiently. On the other hand, TY showed very little efficacy on the different ICE substrates (Km of 860 microM for Suc-YVAD-NH-Mec). These results show that the ICE/TX/TY subfamily has functional heterogeneity and that ICE remains the preferred enzyme for pro-interleukin-1beta cleavage.

The central executioner of apoptosis: multiple connections between protease activation and mitochondria in Fas/APO-1/CD95- and ceramide-induced apoptosis.

According to current understanding, cytoplasmic events including activation of protease cascades and mitochondrial permeability transition (PT) participate in the control of nuclear apoptosis. However, the relationship between protease activation and PT has remained elusive. When apoptosis is induced by cross-linking of the Fas/APO-1/CD95 receptor, activation of interleukin-1beta converting enzyme (ICE; caspase 1) or ICE-like enzymes precedes the disruption of the mitochondrial inner transmembrane potential (DeltaPsim). In contrast, cytosolic CPP32/ Yama/Apopain/caspase 3 activation, plasma membrane phosphatidyl serine exposure, and nuclear apoptosis only occur in cells in which the DeltaPsim is fully disrupted. Transfection with the cowpox protease inhibitor crmA or culture in the presence of the synthetic ICE-specific inhibitor Ac-YVAD.cmk both prevent the DeltaPsim collapse and subsequent apoptosis. Cytosols from anti-Fas-treated human lymphoma cells accumulate an activity that induces PT in isolated mitochondria in vitro and that is neutralized by crmA or Ac-YVAD.cmk. Recombinant purified ICE suffices to cause isolated mitochondria to undergo PT-like large amplitude swelling and to disrupt their DeltaPsim. In addition, ICE-treated mitochondria release an apoptosis-inducing factor (AIF) that induces apoptotic changes (chromatin condensation and oligonucleosomal DNA fragmentation) in isolated nuclei in vitro. AIF is a protease (or protease activator) that can be inhibited by the broad spectrum apoptosis inhibitor Z-VAD.fmk and that causes the proteolytical activation of CPP32. Although Bcl-2 is a highly efficient inhibitor of mitochondrial alterations (large amplitude swelling + DeltaPsim collapse + release of AIF) induced by prooxidants or cytosols from ceramide-treated cells, it has no effect on the ICE-induced mitochondrial PT and AIF release. These data connect a protease activation pathway with the mitochondrial phase of apoptosis regulation. In addition, they provide a plausible explanation of why Bcl-2 fails to interfere with Fas-triggered apoptosis in most cell types, yet prevents ceramide- and prooxidant-induced apoptosis.

Evidence for CPP32 activation in the absence of apoptosis during T lymphocyte stimulation.

Cysteine proteases of the interleukin-1beta-converting enzyme family have been implicated in the effector process of apoptosis in several systems. Among these, CPP32 has been shown to be processed to active enzyme at the onset of apoptosis. Here, we show that CPP32 precursor is cleaved into its active form during phytohaemaglutinin A activation of T lymphocytes. Maximal processing is observed between day 3 and day 4 following addition of mitogen and is a transient process. Precursor cleavage is associated with the appearance of a CPP32-like enzymatic activity in cell lysates. At this time in the culture, almost no apoptotic cell and no dead cell can be detected, and T lymphocytes are actively proliferating. CPP32 processing also occurs when lymphocytes are stimulated through an allogeneic primary mixed lymphocyte reaction. Our results suggest that proteolytic activation of CPP32 could be a physiological step during T lymphocyte activation. In addition, these data indicate that CPP32 activation can occur independently of programmed cell death in T lymphocytes.

Evidence for an interleukin-1beta converting enzyme (ICE)-like activity distinct from ICE and responsible for ICE and CPP32 cleavage during apoptosis.

IL-1beta converting enzyme (ICE) and ICE-related proteases (IRPs) have been suggested to play a central role in apoptosis. We report the use of peptidic ICE inhibitors to reassess the role of this enzyme in the apoptosis induced by Fas or TNFalpha receptor ligation in Jurkat cells, U937 cells or monocytes. Our results show that inhibition of IL-1beta processing can be dissociated from inhibition of apoptosis. Indeed, two out of three com-pounds active on ICE are not inhibitory for apoptosis. This shows that ICE is not required for progression in the apoptotic pathway, but that one or several IRPs are necessary. In addition, Western blot analysis of cell lysates shows that both ICE and CPP32 precursors disappear rapidly after apoptosis induction, while ICH-1L precursor remains intact. Concomitant appearance of cleavage products can be visualized for CPP32, but not for ICE, suggesting that the former is proteolytically activated. In addition, this precursor cleavage can be blocked by an ICE inhibitor active on apoptosis. Altogether, our data support the hypothesis that one or several IRPs are necessary for apoptosis and are responsible for ICE and CPP32 cleavage during this process.

Characterization of a myxoma virus-encoded serpin-like protein with activity against interleukin-1 beta-converting enzyme.

A genomic library of myxoma virus (MV) DNA, a leporipoxvirus that causes myxomatosis, was constructed and screened by in vitro transcription-translation. A clone was selected on the basis of its strong reactivity with MV antiserum. Analysis of the corresponding DNA sequence and the deduced amino acid sequence revealed an open reading frame coding for a 34-kDa protein with strong homologies to members of the serpin superfamily. The gene encoding this new protein, called serp2, was localized on the MV genome. Interestingly, this gene is deleted in an attenuated strain. We constructed a baculovirus vector to produce recombinant Serp2 protein and raised specific antisera that allowed the characterization of Serp2 expression during the MV cycle. The biological relevance of this new serpin from MV was monitored, and it was shown that Serp2 could inhibit human interleukin-1 beta-converting enzyme activity.

Use of monoclonal antibodies to study interleukin-1 beta-converting enzyme expression: only precursor forms are detected in interleukin-1 beta-secreting cells.

We have generated a series of monoclonal antibodies (mAb) using recombinant interleukin (IL)-1 beta-converting enzyme (ICE) p20 and p10 subunits as immunogens. The mAb have been selected for further study based on their reactivity with ICE in transfected COS cells and their lack of cross-reactivity with TX, the closest ICE homolog known to date. Two anti-p20 and one anti-p10 mAb have been used to study ICE expression by Western blotting and immunodetection. In ICE-transfected COS cells, the mAb recognize the p45 ICE precursor and the maturation products (p20 or p10 subunits) for which they are specific. In monocytes and cell lines expressing ICE, only precursor forms are detected and intracellular immunostaining followed by confocal microscopy shows that they are located in the cytoplasm. Quantification experiments show that THP1 cells express approximately 67,000 molecules of ICE precursor per cell, with an estimated precursor to mature ratio of at least 100. In these cells as well as in monocytes, lipopolysaccharide stimulation did not change the pattern of ICE expression, although efficient secretion of mature IL-1 beta was measured. However, upon cell disruption, precursor maturation was observed. Our results, therefore, show that ICE is present in cells as a large pool of intracytoplasmic precursor, and that very limited amounts of mature ICE protein are present, but nevertheless sufficient to allow efficient IL-1 beta cleavage. Altogether, these observations suggest that post-translational maturation of the precursor protein could represent a specific step in the regulation of ICE enzymatic activity.

An approach to the in vitro study of the UTP/UDPglucose/UDP moiety-conserved cycle.

The kinetic behavior of a moiety-conserved ternary cycle is tested experimentally. This system contains the enzymes UDPglucose pyrophosphorylase, glycogen synthase and nucleoside diphosphokinase, converting respectively UTP into UDPglucose, then into UDP and back to UTP in a cyclic manner. The UDPGlc P2ase and NDPK steps are made irreversible by addition of inorganic pyrophosphatase and phosphocreatine kinase, respectively. In order to predict both the evolution and the steady-state values of the various substrates, a model is derived, which takes into account the actual enzyme rate expressions and parameter values, as determined under our experimental conditions. In that model, the UTP, UDPglucose and UDP are taken as the variables, whereas the total concentration of the substrate pool and the four enzyme maximal activities are chosen as the control parameters. Depending upon the various parameter values, monostability, reversible bistability and irreversible transitions may theoretically occur. However, it turns out that some of these values for which multistability might occur, are not accessible experimentally. Under conditions of monostability, the evolutions of the three substrates as experimentally measured are shown to be in good qualitative and quantitative agreement with the model predictions. The relaxation times between two consecutive steady states when a parameter is varied, are shown to be long-lasting processes (several hours). That such an experimental ternary substrate cycle actually exhibits a low sensitivity to any perturbation, addresses the issue to knowing if the same property is likely to occur in vivo, or, in other words, do large moiety-conserved cycles act as metabolic buffers?

Application of the metabolic control theory to the study of the dynamics of substrate cycles.

Substrate cycles are ubiquitous structures of the cellular metabolism (e.g. Krebs cycle, fatty acids beta-oxydation cycles, etc...). Moiety-conserved cycles (e.g. adenine nucleotides and NADH/NAD, etc...) are also important. The role played by such cycles in the metabolism and its regulation is not clearly understood so far. However, it was shown that these cycles can generate multistationarity (bistability), irreversible transitions, enhancement of sensitivity, temporal oscillations and chaotic motions (Hervagault & Canu, 1987; Hervagault & Cimino, 1989; Reich & Sel'kov, 1981; Ricard & Soulié, 1982). [formula: see text] Fig. 1: Scheme of the open binary substrate cycle under study. The substrate S is converted into P with a net rate v2. Substrate P is converted in turn into S with a net rate v3. Step v2 is inhibited by excess of the substrate, S. In addition, the cycle operates under open conditions, that is zero-order input of S at rates alpha 0(v1) and first order outputs of S and P at rates alpha S and alpha P(v4), respectively. The metabolic control theory (see also Fell, 1990), which shows how a metabolic network reacts to small perturbations in the vicinity of a steady state, and is formulated with the so-called "control coefficients", was applied to such a cycle in order to get a better knowledge on the importance of each step at the regulatory point of view. The behaviour of a binary substrate cycle (fig. 1) in which one of the enzymes may be subjected to inhibition by excess of its substrate (v2) was studied theoretically.(ABSTRACT TRUNCATED AT 250 WORDS)

Steady-state properties of a model ternary substrate cycle: theoretical predictions.

Numerous ternary substrate cycles are metabolically operative in vivo. The relative concentrations of the interconverted substrates are generally correlated with different physiological states. These cycles often include reversible and/or substrate-inhibited enzymic steps. The switch between one steady state (metabolic state) and another may be the consequence of either the effect of an exogeneous metabolite or signal, or the alteration of a cycle internal parameter. The interpretation of results obtained with currently designed experiments on substrate cycles seldom take into account the very dynamic and regulatory properties inherent in the cyclic and often autocatalytic nature of the pathway. In the present report, the various dynamic properties of a model ternary substrate cycle, bounded by moiety conservation, are investigated. Three situations with increasing complexity are considered: (i) the three enzymes are michaelian and catalyse irreversible steps; (ii) one of the enzymic steps is reversible; and (iii) one step is subjected to a destabilizing factor, i.e. inhibition by excess of substrate. The behavior(s) of the whole cycle is mainly controlled by four parameters, that is, ST, the total concentration of the substrate pool, and the three enzyme maximal velocities, VMi (i = 1,2,3). As ST (= S1 + S2 + S3) is constant, the Si steady-state concentrations (stable or not) can be represented in barycentric coordinates in a triangle (simplex). This convenient representation allows us to predict the different states of the system when one enzyme maximal activity is varied. The steady-state concentration dependencies as a function of one or several parameters may be either monostable (possibility of zero-order ultrasensitivity) or bistable (with or without reversible transitions). The physiological and experimental relevances of these observations are emphasized.