Specific interactions of the adenovirus proteinase with the viral DNA, an 11-amino-acid viral peptide, and the cellular protein actin.

The adenovirus proteinase (AVP) is synthesized in an inactive form that requires cofactors for activation. The interaction of AVP with two viral cofactors and with a cellular cofactor, actin, is characterized by quantitative analyses. The results are consistent with a specific model for the regulation of AVP. Late in adenovirus infection, inside nascent virions, AVP becomes partially activated by binding to the viral DNA, allowing it to cleave out an 11-amino-acid viral peptide, pVIc, that binds to AVP and fully activates it. Then, about 70 AVP-pVIc complexes move along the viral DNA, via one-dimensional diffusion, cleaving virion precursor proteins 3200 times to render a virus particle infectious. Late in adenovirus infection, in the cytoplasm, the cytoskeleton is destroyed. The amino acid sequence of the C terminus of actin is homologous to that of pVIc, and actin, like pVIc, can act as a cofactor for AVP in the cleavage of cytokeratin 18 and of actin itself. Thus, AVP may also play a role in cell lysis.

Roles of two conserved cysteine residues in the activation of human adenovirus proteinase.

The roles of two conserved cysteine residues involved in the activation of the adenovirus proteinase (AVP) were investigated. AVP requires two cofactors for maximal activity, the 11-amino acid peptide pVIc (GVQSLKRRRCF) and the viral DNA. In the AVP-pVIc crystal structure, conserved Cys104 of AVP has formed a disulfide bond with conserved Cys10 of pVIc. In this work, pVIc formed a homodimer via disulfide bond formation with a second-order rate constant of 0.12 M(-1) s(-1), and half of the homodimer could covalently bind to AVP via thiol-disulfide exchange. Alternatively, monomeric pVIc could form a disulfide bond with AVP via oxidation. Regardless of the mechanism by which AVP becomes covalently bound to pVIc, the kinetic constants for substrate hydrolysis were the same. The equilibrium dissociation constant, K(d), for the reversible binding of pVIc to AVP was 4.4 microM. The K(d) for the binding of the mutant C10A-pVIc was at least 100-fold higher. Surprisingly, the K(d) for the binding of the C10A-pVIc mutant to AVP decreased at least 60-fold, to 6.93 microM, in the presence of 12mer ssDNA. Furthermore, once the mutant C10A-pVIc was bound to an AVP-DNA complex, the macroscopic kinetic constants for substrate hydrolysis were the same as those exhibited by wild-type pVIc. Although the cysteine in pVIc is important in the binding of pVIc to AVP, formation of a disulfide bond between pVIc and AVP was not required for maximal stimulation of enzyme activity by pVIc.

Discovery of a new inhibitor lead of adenovirus proteinase: steps toward selective, irreversible inhibitors of cysteine proteinases.

Using the computer docking program EUDOC, in silico screening of a chemical database for inhibitors of human adenovirus cysteine proteinase (hAVCP) identified 2,4,5,7-tetranitro-9-fluorenone that selectively and irreversibly inhibits hAVCP in a two-step reaction: reversible binding (Ki = 3.09 microM) followed by irreversible inhibition (ki = 0.006 s(-1)). The reversible binding is due to molecular complementarity between the inhibitor and the active site of hAVCP, which confers the selectivity of the inhibitor. The irreversible inhibition is due to substitution of a nitro group of the inhibitor by the nearby Cys122 in the active site of hAVCP. These findings suggest a new approach to selective, irreversible inhibitors of cysteine proteinases involved in normal and abnormal physiological processes ranging from embryogenesis to apoptosis and pathogen invasions.

Human adenovirus proteinase: DNA binding and stimulation of proteinase activity by DNA.

The interaction of the human adenovirus proteinase (AVP) with various DNAs was characterized. AVP requires two cofactors for maximal activity, the 11-amino acid residue peptide from the C-terminus of adenovirus precursor protein pVI (pVIc) and the viral DNA. DNA binding was monitored by changes in enzyme activity or by fluorescence anisotropy. The equilibrium dissociation constants for the binding of AVP and AVP-pVIc complexes to 12-mer double-stranded (ds) DNA were 63 and 2.9 nM, respectively. DNA binding was not sequence specific; the stoichiometry of binding was proportional to the length of the DNA. Three molecules of the AVP-pVIc complex bound to 18-mer dsDNA and six molecules to 36-mer dsDNA. When AVP-pVIc complexes bound to 12-mer dsDNA, two sodium ions were displaced from the DNA. A Delta of -4.6 kcal for the nonelectrostatic free energy of binding indicated that a substantial component of the binding free energy results from nonspecific interactions between the AVP-pVIc complex and DNA. The cofactors altered the interaction of the enzyme with the fluorogenic substrate (Leu-Arg-Gly-Gly-NH)2-rhodamine. In the absence of any cofactor, the Km was 94.8 microM and the kcat was 0.002 s(-1). In the presence of adenovirus DNA, the Km decreased 10-fold and the kcat increased 11-fold. In the presence of pVIc, the Km decreased 10-fold and the kcat increased 118-fold. With both cofactors present, the kcat/Km ratio increased 34000-fold, compared to that with AVP alone. Binding to DNA was coincident with stimulation of proteinase activity by DNA. Although other proteinases have been shown to bind to DNA, stimulation of proteinase activity by DNA is unprecedented. A model is presented suggesting that AVP moves along the viral DNA looking for precursor protein cleavage sites much like RNA polymerase moves along DNA looking for a promoter.

Interaction of the human adenovirus proteinase with its 11-amino acid cofactor pVIc.

The interaction of the human adenovirus proteinase (AVP) and AVP-DNA complexes with the 11-amino acid cofactor pVIc was characterized. The equilibrium dissociation constant for the binding of pVIc to AVP was 4.4 microM. The binding of AVP to 12-mer single-stranded DNA decreased the K(d) for the binding of pVIc to AVP to 0.09 microM. The pVIc-AVP complex hydrolyzed the substrate with a Michaelis constant (K(m)) of 3.7 microM and a catalytic rate constant (k(cat)) of 1.1 s(-1). In the presence of DNA, the K(m) increased less than 2-fold, and the k(cat) increased 3-fold. Alanine-scanning mutagenesis was performed to determine the contribution of individual pVIc side chains in the binding and stimulation of AVP. Two amino acid residues, Gly1' and Phe11', were the major determinants in the binding of pVIc to AVP, while Val2' and Phe11' were the major determinants in stimulating enzyme activity. Binding of AVP to DNA greatly suppressed the effects of the alanine substitutions on the binding of mutant pVIcs to AVP. Binding of either or both of the cofactors, pVIc or the viral DNA, to AVP did not dramatically alter its secondary structure as determined by vacuum ultraviolet circular dichroism. pVIc, when added to Hep-2 cells infected with adenovirus serotype 5, inhibited the synthesis of infectious virus, presumably by prematurely activating the proteinase so that it cleaved virion precursor proteins before virion assembly, thereby aborting the infection.

Prevention of viral drug resistance by novel combination therapy.

A new form of antiviral clinical therapy is proposed in which three different drugs are administered against three different targets on the same virus-coded protein. If the physiological functions of the three different target sites are not independent of each other, then a mutation conferring drug resistance at one site may alter the physiological functions at the other sites and further drug resistance may not arise. The adenovirus proteinase, with its two cofactors that act synergistically on enzyme activity, may be a good model system within which to test the efficacy of this form of combination therapy.

A new form of antiviral combination therapy predicted to prevent resistance from arising, and a model system to test it.

Combination therapy in the treatment of viral infections in which, for example, three different drugs against three different targets on three independent proteins are administered, has been highly successful clinically. However, it is only a matter of time before a virus will arise resistant to all three drugs, because the mutations leading to drug resistance are independent of each other. But, what if the mutations leading to drug resistance are not independent of each other, but confer some cost to the virus? If the cost is too great, than resistance may not arise. To impose such a cost in the clinical treatment of viral infections, we propose a new form of combination therapy. Here, three different drugs against three different targets on the same virus-coded protein are administered. If the physiological functions of the three different target sites are not independent of each other, then, a mutation at one site may alter the physiological functions at the other sites. We present a model system in which to test the efficacy of this new form of triple combination therapy. Human adenovirus has a virus-coded proteinase that is essential for the synthesis of infectious virus. It contains an active site and two cofactor binding sites; the functions of the active site are dependent upon the cofactors interacting with their binding sites. We describe how to obtain drugs against the three different sites.

Sensitive method to identify and characterize proteinases in situ after SDS-PAGE.

Cells and body fluids contain numerous, different proteinases; to identify and characterize them are both important and difficult tasks. Especially difficult to identify and characterize are highly specific proteinases. Here, we present an extremely sensitive and quantitative method to characterize proteinases fractionated by SDS-PAGE that cleave specific rhodamine-based fluorogenic substrates. To test the sensitivity of the technique, we used trypsin as our model system. Filter paper impregnated with rhodamine-based fluorogenic substrates was placed on a gel, and bands of fluorescence originating from specific proteinases were visualized in real time. The method is very sensitive; picogram amounts of trypsin can be detected. The method should be very general, in that even proteinases whose substrates require amino acids C-terminal to the cleavage site may be identified and characterized. The results allow one to obtain not only information on the substrate specificity of a specific enzyme but also information about its molecular weight.

Temporal and spatial control of the adenovirus proteinase by both a peptide and the viral DNA.

The adenovirus proteinase (AVP) uses both an 11-amino acid peptide (pVIc) and the viral DNA as cofactors to increase its catalytic rate constant 6000-fold. The crystal structure of an AVP-pVIc complex at 2.6-A resolution reveals a new protein fold of an enzyme that is the first member of a new class of cysteine proteinases, which arose via convergent evolution.

Preparation and crystallization of a complex between human adenovirus serotype 2 proteinase and its 11-amino-acid cofactor pVIc.

Crystals have been obtained of the recombinant human adenovirus serotype 2 proteinase (AVP) in a complex with its 11-amino-acid cofactor pVIc. AVP-pVIc complexes were formed by the incubation of AVP with a 1.2-fold molar excess of pVIc prior to the crystallization trials. Diffraction-quality crystals were obtained at 18 degrees C by the vapor-diffusion method with 5.6 mg/ml AVP-pVIc in 1.4 M sodium acetate and 0.1 M Hepes, pH 7.5. Diffraction data (99% complete to 2.6 A resolution with Rmerge of 0.077) were collected from native crystals at room temperature at beamline X12-C at the National Synchrotron Light Source. The crystals belong to space group P6(1) with unit cell dimensions a = b = 114.2 A, c = 50.1 A; alpha = beta = 90 degrees, gamma = 120 degrees. The unit cell dimensions and likely mass of the molecular species in the crystals were consistent with there being one 25,000-Da complex (1:1) per asymmetric unit. Additionally, one heavy-atom derivative, obtained by the soaking of preformed crystals, was isomorphous to the native crystal. Diffraction data obtained on these crystals were 95% complete to 3.0 A resolution with an Rmerge of 0.076. Difference-Patterson analysis indicates three heavy atom sites in the derivative asymmetric unit.

Different modes of inhibition of human adenovirus proteinase, probably a cysteine proteinase, by bovine pancreatic trypsin inhibitor.

The type of proteinase and the nature of the active site of the human adenovirus proteinase are unknown. For these reasons we produced an inhibitor profile of the enzyme. Enzyme activity in disrupted virions was inhibited by several serine-specific as well as cysteine-specific proteinase inhibitors. Of the inhibitors that worked, the most useful potentially in illuminating the nature of the active site was bovine pancreatic trypsin inhibitor (BPTI), and for this reason we extensively characterized the interaction with BPTI. In disrupted virions, the enzyme is irreversibly inhibited by BPTI with a Ki of 35 nM and a ki of 6.2 x 10(-4) s(-1). One reason enzyme activity is inhibited is that BPTI, a basic protein, precipitates the viral DNA, a cofactor of enzyme activity. In vitro with purified components, BPTI acts as a competitive inhibitor (Ki 2 microM) of the recombinant proteinase complexed with its 11-amino-acid cofactor pVIc. The recombinant endoproteinase is beat labile whereas its 11-amino-acid cofactor is heat stable. We estimate there are about 50 molecules of proteinase per virus particle.

Crystal structure of the human adenovirus proteinase with its 11 amino acid cofactor.

The three-dimensional structure of the human adenovirus-2 proteinase complexed with its 11 amino acid cofactor, pVIc, was determined at 2.6 A resolution by X-ray crystallographic analysis. The fold of this protein has not been seen before. However, it represents an example of either subtly divergent or powerfully convergent evolution, because the active site contains a Cys-His-Glu triplet and oxyanion hole in an arrangement similar to that in papain. Thus, the adenovirus proteinase represents a new, fifth group of enzymes that contain catalytic triads. pVIc, which extends a beta-sheet in the main chain, is distant from the active site, yet its binding increases the catalytic rate constant 300-fold for substrate hydrolysis. The structure reveals several potential targets for antiviral therapy.

Characterization of human adenovirus proteinase activity in disrupted virus particles.

Virus-coded proteinases are attractive targets for antiviral therapy; however, lack of quick, sensitive, quantitative, and selective assays for enzyme activity makes it difficult to characterize these proteinases and to screen large numbers of potential inhibitors. Here we describe new substrates for the adenovirus proteinase, fluorogenic Rhodamine-based substrates containing tetrapeptides corresponding to sequences cleaved in adenovirus precursor proteins. Proteinase activity in as few as 10(9) disrupted virions could be quantitatively detected in a 30-min assay. With the substrate (Leu-Arg-Gly-Gly-NH)2-Rhodamine, the Km was 1.4 microM and the Vmax was 3.24 pmol substrate hydrolyzed/sec/pmol virus. Enzyme activity was stimulated by dithiothreitol and inhibited by several serine-specific as well as cysteine-specific proteinase inhibitors. In a thiol protection experiment, the virion enzyme was shown to have a cysteine residue with an unusually low pKa, a pKa similar to that of the active-site nucleophile of the cysteine proteinase papain. The curve of Vmax as a function of pH is unlike the curve from papain and implied that there are at least three ionizable groups whose protonation state can affect catalysis - one with a pKa of 6.2, another with a pKa of 7.2, and a third with a pKa of 8.3.

Characterization of three components of human adenovirus proteinase activity in vitro.

Human adenovirus contains a virion-associated proteinase activity essential for the development of infectious virus. Maximal proteinase activity in vitro had been shown to require three viral components: the L3 23-kDa protein, an 11-amino acid cofactor (pVIc), and the viral DNA. Here, we present a quantitative purification procedure for a recombinant L3 23-kDa protein (recombinant endoproteinase (rEP)) expressed in Escherichia coli and the procedure that led to the purification and identification of pVIc as a cofactor. The cofactors stimulate proteinase activity not by decreasing Km, which changes by no more than 2-fold, but by increasing kcat. rEP alone had a small amount of activity, the kcat of which increased 355-fold with pVIc and 6072-fold with adenovirus serotype 2 (Ad2) DNA as well. Curves of Vmax of rEP.pVIc complexes with the substrate (Leu-Arg-Gly-NH)2-rhodamine as a function of pH in the absence and presence of Ad2 DNA indicate that the pKa values of amino acids that affect catalysis are quite different from those that affect catalysis by the cysteine proteinase papain. The pKa values in the absence of Ad2 DNA are 5.2, 6.4, 6.9, 7.5, and 9.4, and those in its presence are 5.2, 6.5, 7.4, and 8.8.

Viral DNA and a viral peptide can act as cofactors of adenovirus virion proteinase activity.

Human adenovirus (Ad2), like many other viruses, contains a virion-associated proteinase essential for the synthesis of infectious virus particles. We observed proteinase activity in wild-type virus but not in the ts-1 virus, which contains a mutation in the Ad2 L3 endoprotease gene that confers temperature-sensitive processing of virion precursor proteins. Unexpectedly, we did not observe proteinase activity with purified recombinant endoprotease protein (M(r) 23 K). Purified recombinant endoprotease protein, however, complemented the mutation in ts-1 virions, restoring proteinase activity when mixed together. This implied that cofactors may be required. Here we reconstitute proteinase activity in vitro with three purified viral components: (1) the recombinant endoprotease protein; (2) an 11-amino-acid peptide that originates from the carboxy terminus of pVI, the precursor to virion component VI; and (3) adenovirus DNA. The use of DNA for a proteinase activity is unprecedented.