Interaction of the Human Adenovirus Proteinase with Its 11-Amino Acid Cofactor pVIc

Like many virus-coded proteinases, the human adenovirus serotype 2 proteinase (AVP)1 is required for the synthesis of infectious virus (1). Late in infection, young virions are formed in which six of the 12 different major proteins are present in precursor form. In the young virion, approximately 70 AVP molecules become activated (2), and they cleave multiple copies of the six virion precursor proteins more than 3200 times, thereby rendering the virus particle infectious (3). A temperature-sensitive mutant of adenovirus was shown to lack proteinase activity at the nonpermissive temperature (1); the mutation mapped to the L3 23K gene (4). The L3 23K gene has been cloned and expressed in Escherichia coli, and the resultant 204-amino acid protein, AVP, has been purified (5-7). An unusual aspect of AVP is that it requires two cofactors for maximal activity. One cofactor is pVIc, the 11-amino acid peptide from the C-terminus of the precursor to protein VI, pVI. The amino acid sequence of pVIc is GVQSLKRRRCF (5, 8). The four amino acid residues preceding pVIc in pVI constitute an AVP consensus cleavage sequence; thus, AVP can cleave out its own cofactor. Another cofactor is adenovirus DNA (Ad DNA) to which AVP binds nonspecifically, i.e., independent of any specific DNA sequence (5). The two cofactors increase the specificity constant, kcat/Km, for substrate hydrolysis. For an AVP–pVIc complex, the level of increase in kcat/Km, relative to AVP alone, is 1130; for an AVP–Ad DNA complex, it is 110, and for an AVP–pVIc–Ad DNA complex, it is 34100 (31). Our model for the temporal and spatial regulation of AVP activity illustrates how the virus could make use of the cofactors. AVP is initially synthesized in an inactive form and remains inactive until it enters virions which are in part assembled from precursor proteins. If the proteinase were active before virion assembly, it would prematurely cleave virion precursor proteins. Presumably, cleaved precursor proteins cannot form virus particles. Late in infection, in the nucleus, virion proteins assemble into empty capsids (10). Once formed, the viral DNA and core proteins are encapsidated, generating young virions. We postulate that the proteinase enters the empty capsids along with the viral DNA to which it is bound. The AVP–DNA complexes become positioned next to the proteinase consensus cleavage sites on pVI where cleavage liberates pVIc. Liberated pVIc molecules can then bind to the proteinases that cut them out. The ternary complex, AVP–pVIc–DNA, contains a fully active enzyme capable of cleaving viral precursor proteins. AVP–pVIc complexes encounter and cleave viral precursor proteins by moving along the viral DNA via one-dimensional diffusion, much like RNA polymerase moves along DNA looking for a promoter. The crystal structure of an AVP–pVIc complex has been determined to 2.6 A resolution (11). The AVP–pVIc complex is ovoid (Figure 1). There are two domains in the protein. One domain contains the large β-sheet and two peripheral α-helices. The other domain is composed mostly of α-helices, one from the N-terminus and the remainder from the C-terminus of AVP. These helices form the lower middle to the wide end of the molecule. pVIc appears to act like a strap that holds the two domains together in a configuration that is optimal for efficient catalysis. The crystal structure of the AVP–pVIc complex indicates it is a cysteine proteinase. Figure 1 Structure of the AVP–pVIc complex. The two domains of AVP are colored yellow and green. pVIc is colored cyan. The residues involved in catalysis in the active site are colored red. The disulfide bond between Cys104 of AVP and Cys10′ of ... Here, the binding of pVIc to AVP is characterized by measuring the equilibrium dissociation constant, Kd, in the presence and absence of DNA. For AVP–pVIc complexes, the Michaelis constant (Km) and the catalytic rate constant (kcat) for substrate hydrolysis were determined. To identify amino acid residues in pVIc that are essential for the binding of pVIc to AVP and for the stimulation of AVP activity by pVIc, alanine-scanning mutagenesis was performed. Each amino acid of pVIc, except for Cys10′, was individually replaced with an alanine residue. Characterization of Cys10′ of pVIc will be the subject of a separate communication. The 10 mutant pVIcs were assayed for their ability to bind to AVP in the presence and absence of 12-mer single-stranded (ss) DNA, and the Kd values were determined. With the mutant pVIcs bound to AVP, the Km and kcat for substrate hydrolysis were measured. Changes in proteinase secondary structure on binding pVIc and DNA were investigated by vacuum circular dichroism. Finally, pVIc was shown to act as an antiviral agent when prematurely introduced to AVP early in an infection.