Protein kinase activity associated with the large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10).

The large subunit of the herpes simplex virus type 2 (HSV-2) ribonucleotide reductase (RR1) is demonstrated to possess serine/threonine-specific kinase activity. Computer-assisted sequence analysis identified regions within the amino terminus of ICP10 that are homologous to the catalytic domain of known protein kinases (PKs). An in vitro kinase assay confirmed the ability of ICP10, immunoprecipitated from either HSV-2-infected cells or from cells transfected with an ICP10 expression vector, to autophosphorylate and transphosphorylate exogenous substrates in the presence of ATP and Mg2+. The HSV-1 RR1 was shown to be negative for PK activity under these conditions. PK activity was localized to a 57-kDa amino-terminal region within ICP10 that lacked RR activity.     

The transmembrane helical segment but not the invariant lysine is required for the kinase activity of the large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10).

The large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10) is a chimera consisting, at the amino terminus, of a Ser/Thr protein kinase (PK) with features of a signal peptide and a transmembrane (TM) helical segment, and at the carboxy-terminus, of the ribonucleotide reductase (Chung et al., 1989, 1990). Membrane immunofluorescence of ICP10 transformed cells with antibodies to synthetic peptides located upstream or downstream of the TM indicates that ICP10 is a membrane-spanning protein. Site-directed and deletion mutants were used to further characterize ICP10-PK. Mutation of Gly106 in catalytic motif I or of the invariant Lys in catalytic motif II, and deletion of both motifs (amino acids 106-178) did not eliminate kinase activity. PK activity was retained by the invariant Lys mutant expressed in bacteria and following protein separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transfer to membrane filters. Both ICP10 and the invariant Lys mutant bound 14C-labeled rho-fluorosulfonylbenzoyl 5'-adenosine, an ATP affinity analog. The deletion mutant had 4-fold lower kinase activity than ICP10-PK, and it was insensitive to Mn2+, suggesting that these motifs are involved in Mn2+ activation of kinase activity. PK activity was lost by deletion of the TM segment (amino acid residues 85-106).     

Affinity purification of active subunit 1 of herpes simplex virus type 1 ribonucleotide reductase exhibiting a protein kinase activity

Herpes simplex virus (HSV) ribonucleotide reductase is formed by the association of two distinct dimeric subunits, R1 and R2. Attempts to purify either the HSV holoenzyme or its R1 subunit in their active form have been unsuccessful until now. The C terminus of the R2 protein being involved in the association with R1, the synthetic nonapeptide corresponding to this terminus, impedes the formation of the holoenzyme by competing with R2 for a critical site on R1. Based upon these observations, we developed an affinity chromatographic procedure to purify the R1 protein from HSV-1-infected baby hamster kidney cells. Specific binding of R1 to an affinity column made by linking the peptide HSV R2-(326-337) to Affi-Gel 10, followed by specific elution with an excess of an analogous peptide exhibiting a higher affinity for R1 yielded, in a single step, highly purified R1 protein. The purified R1 preparations contained approximately 95% of intact R1, the remaining 5% consisting of two R1 copurifying proteolytic breakdown products. The purified R1 protein exhibited a high reductase specific activity when mixed with an excess of the R2 subunit. Moreover, in vitro kinase assays revealed that the purified R1 protein of HSV-1 possesses an autophosphorylating activity also able to phosphorylate alpha-casein and histone II-S. The intrinsic protein kinase activity of HSV R1 is associated with its unique N-terminal domain which is absent from all other reductase subunits 1 and contains consensus motifs found in Ser/Thr protein kinases. A preliminary characterization of the kinase activity of the R1 protein of HSV-1 ribonucleotide reductase is presented.     

The role of herpes simplex virus ribonucleotide reductase small subunit carboxyl terminus in subunit interaction and formation of iron-tyrosyl center structure.

Herpes simplex virus ribonucleotide reductase consists of two nonidentical subunits, proteins R1 and R2, which are required together for activity. Active R2 protein contains a tyrosyl free radical and a binuclear iron center. A truncated form of the R2 subunit, lacking 7 amino acid residues in the carboxyl terminus, was constructed, overexpressed in Escherichia coli and purified to homogeneity. In the presence of ferrous iron and oxygen, the truncated protein readily generated similar amounts of tyrosyl free radical as the intact protein. However, the radical showed differences in EPR characteristics in the truncated protein compared with the normal one, indicating an altered structural arrangement of the radical relative to the iron center. The truncated R2* protein was completely devoid of binding affinity to the R1 protein, demonstrating that the subunit interaction is totally dependent on the 7 outermost carboxyl-terminal amino acids of protein R2.     

Characterization of the novel protein kinase activity present in the R1 subunit of herpes simplex virus ribonucleotide reductase.

We have compared the protein kinase activities of the R1 subunits from herpes simplex virus types 1 (HSV-1) and 2 (HSV-2) ribonucleotide reductase following expression in Escherichia coli. Autophosphorylation activity was observed when kinase assays were performed with immunoprecipitated R1 or proteins purified to homogeneity, and the activity was stimulated by the basic protein protamine. Transphosphorylation of histones or calmodulin by purified or immunoprecipitated HSV-1 and HSV-2 R1 was not observed, and our results suggest that the activities of these two proteins are similar. We further characterized the protein kinase activity of HSV-1 R1 by producing insertion and deletion mutants constructed with a plasmid expressing R1 amino acids 1 to 449. C-terminal deletion analysis identified the catalytic core of the enzyme as comprising residues 1 to 292, and this polypeptide will be useful for structural determinations by X-ray crystallography. Insertion of a 4-amino-acid sequence at sites within the protein kinase domain identified regions essential for activity; insertions at residues 22 and 112 completely inactivated activity, and an insertion at residue 136 reduced activity sixfold. Similar insertions at residues 257, 262, 292, and 343 had no effect on activity. The ATP analog 5'-fluorosulfonylbenzoyladenosine, which covalently modifies conventional eukaryotic kinases at an essential lysine residue within the active site, did label HSV R1, but this labelling occurred outside the N-terminal domain. These data indicate that the HSV R1 kinase is novel and distinct from other eukaryotic protein kinases.     

The R1 subunit of herpes simplex virus ribonucleotide reductase has chaperone-like activity similar to Hsp27

HSV-2 R1, the R1 subunit of herpes simplex virus (HSV) ribonucleotide reductase, protects cells against apoptosis. Here, we report the presence in HSV-2 R1 of a stretch exhibiting similarity to the alpha-crystallin domain of the small heat shock proteins, a domain known to be important for oligomerization and cytoprotective activities of these proteins. Also, the HSV-2 R1 protein, which forms multimeric structures in the absence of nucleotide, displayed chaperone ability as good as Hsp27 in a thermal denaturation assay using citrate synthase. In contrast, mammalian R1, which does not contain an alpha-crystallin domain, has neither chaperone nor anti-apoptotic activity. Thus, we propose that the chaperone activity of HSV-2 R1 could play an important role in viral pathogenesis.     

Neutralization of herpes simplex virus ribonucleotide reductase activity by an oligopeptide-induced antiserum directed against subunit H2.

Herpes simplex virus type 1 ribonucleotide reductase is associated with two polypeptides of apparent molecular weights 136,000 and 38,000. The two polypeptides form a tight complex and, therefore, are often coprecipitated by monoclonal antibodies. We report here that immunoglobulins G purified from polyclonal rabbit antisera (P9) raised against a nonapeptide corresponding to the carboxy terminus of the 38,000-dalton polypeptide specifically neutralize the herpes simplex virus ribonucleotide reductase activity. We suggest that the P9 immunoglobulin G neutralizes the reductase activity by impairing the association of the two subunits (H1 and H2) of the enzyme.     

Ribonucleotide reductase of herpes simplex virus type 2 resembles that of herpes simplex virus type 1.

The ribonucleotide reductase (ribonucleoside-diphosphate reductase; EC 1.17.4.1) induced by herpes simplex virus type 2 infection of serum-starved BHK-21 cells was purified to provide a preparation practically free of both eucaryotic ribonucleotide reductase and contaminating enzymes that could significantly deplete the substrates. Certain key properties of the herpes simplex virus type 2 ribonucleotide reductase were examined to define the extent to which it resembled the herpes simplex virus type 1 ribonucleotide reductase. The herpes simplex virus type 2 ribonucleotide reductase was inhibited by ATP and MgCl2 but only weakly inhibited by the ATP X Mg complex. Deoxynucleoside triphosphates were at best only weak inhibitors of this enzyme. ADP was a competitive inhibitor (K'i, 11 microM) of CDP reduction (K'm, 0.5 microM), and CDP was a competitive inhibitor (K'i, 0.4 microM) of ADP reduction (K'm, 8 microM). These key properties closely resemble those observed for similarly purified herpes simplex virus type 1 ribonucleotide reductase and serve to distinguish these virally induced enzymes from other ribonucleotide reductases.     

Purification and characterization of recombinant mouse and herpes simplex virus ribonucleotide reductase R2 subunit.

Overexpression of recombinant mouse and herpes simplex virus ribonucleotide reductase small subunit (protein R2) has been obtained by using the T7 RNA polymerase expression system. Both proteins, which constitute about 30% of the soluble Escherichia coli proteins, have been purified to homogeneity by a rapid and simple procedure. At this stage, few of the molecules contain the iron-tyrosyl free-radical center necessary for activity; however, addition of ferrous iron and oxygen under controlled conditions resulted in a mouse R2 protein containing 0.8 radical and 2 irons per polypeptide chain. In this reaction, one oxygen molecule was needed to generate each tyrosyl radical. Both proteins had full enzymatic activity. EPR spectroscopy showed that iron-center/radical interactions are considerably stronger in both mouse and viral proteins than in E. coli protein R2. CD spectra showed that the bacterial protein contains 70% alpha-helical structure compared to only about 50% in the mouse and viral proteins. Light absorption spectra between 310 and 600 nm indicate close similarity of the mu-oxo-bridged binuclear iron centers in all three R2 proteins. Furthermore, the paramagnetically shifted iron ligand proton NMR resonances show that the antiferromagnetic coupling and ligand arrangement in the iron center are nearly identical in all three species.     

A truncated protein kinase domain of the large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10) expressed in Escherichia coli.

The amino-terminal domain of the large subunit of herpes simplex virus type 2 ribonucleotide reductase (ICP10) contains a serine/threonine-specific protein kinase that has characteristics of a growth factor receptor (Chung, T. D., Wymer, J. P., Smith, C. C., Kulka, M., and Aurelian, L. (1989) J. Virol. 63, 3389-3398; Chung, T. D., Wymer, J. P., Kulka, M. Smith, C. C., and Aurelian, L. (1990) Virology 179, 168-178). To characterize this protein kinase (PK) domain further we constructed a bacterial expression vector (pJL11) containing DNA sequences encoding ICP10 amino acid residues 1-445. Bacteria containing pJL11 were induced to express a 29-kDa protein (designated pp29la1) that represents a truncated portion of the ICP10-PK domain (includes PK catalytic motifs I-V) as demonstrated by immunoprecipitation with antibodies that recognize different antigenic domains, competition studies with extracts of ICP10-positive eukaryotic cells, and peptide mapping.pp29la1 has autophosphorylating and transphosphorylating activity for calmodulin. The enzyme is activated by Mn2+ but not by Mg2+ ions, and autophosphorylation is inhibited by histone. It differs from the authentic ICP10-PK in that phosphorylation is specific only for threonine.     

Cloning of a ribonucleotide reductase gene of the herpes simplex virus type 2 strain G

The ribonucleotide reductase (RR) 2 gene of the HSV-2 strain G was cloned, sequenced, and expressed in an E. coli cell. The RR2 gene was located on the PstI 2.4 kb fragment, which was cloned and sequenced. The ORF of the gene was 1,011 bp and its termination codon was TAG; also, the CATATAA sequence was present in the promoter of the RR2 gene. A Poly A signal sequence (AATAAA) was found in the 3'-noncoding region. The RR2 proteins that were produced in the E. coli and Vero cells were confirmed using a Western blot analysis. SDS-PAGE revealed that the molecular weights of the fusion-RR2 that was produced in the E. coli cells were approximately 24 kDa and 38 kDa in the Vero cells. The RR2 proteins were soluble. The differences in the molecular weights might be due to modifications in the Vero cells.     

Immunological characterization of herpes simplex virus type 1 and 2 polypeptide(s) involved in viral ribonucleotide reductase activity

Mammalian cells infected with herpes simplex virus (HSV) express a novel ribonucleotide reductase which is biochemically and immunologically distinct from the uninfected-cell enzyme. Using polyvalent rabbit antiserum raised against partially purified HSV type 2 reductase as well as monoclonal antibodies to HSV type 1 and HSV type 2 early antigens, we have been able to show that in both serotypes reductase activity is associated with phosphoproteins of molecular weights 144,000 and 38,000 encoded between map units 0.566 and 0.602 in the viral genomes. The major antigenic species (144,000) have been tentatively identified as HSV type 1 ICP6 and HSV type 2 ICP10.     

Herpes simplex type 1 ribonucleotide reductase. Mechanism studies with inhibitors

Several known inhibitors of mammalian ribonucleotide reductase were studied for their interactions with herpes simplex virus type 1 (HSV-1) ribonucleotide reductase. MAIQ (4-methyl-5-amino-1-formylisoquinoline thiosemicarbazone) produced apparent inactivation of HSV-1 ribonucleotide reductase. Only catalytically cycling, not resting, enzyme could be inactivated. Double reciprocal replots of the rates of inactivation versus the concentration of MAIQ indicated that a reversible complex with the enzyme was formed prior to inactivation. In the presence of 10 microM CDP, the maximum rate of inactivation was 20 per h (t1/2 = 3 min). The half-maximum rate was achieved at about 15 microM MAIQ. INOX (periodate-oxidized inosine) also appeared to inactivate HSV-1 ribonucleotide reductase. In contrast to MAIQ, it readily inactivated resting as well as cycling enzyme. CDP retarded the rates of inactivation by INOX. An initial reversible complex between INOX and enzyme was not detectable under the conditions used. IMPY (2,3-dihydro-1H-pyrazolo(2,3-a)imidazole) and guanazole (3,5-diamino-1,2,4-triazole) produced reversible inhibition. Although the data with both inhibitors were most consistent with the noncompetitive inhibition model (versus CDP), the data with guanazole were also marginally consistent with the uncompetitive model.     

Resistance of herpes simplex virus type 1 to peptidomimetic ribonucleotide reductase inhibitors: selection and characterization of mutant isolates.

Herpes simplex virus (HSV) encodes its own ribonucleotide reductase (RR), which provides the high levels of deoxynucleoside triphosphates required for viral DNA replication in infected cells. HSV RR is composed of two distinct subunits, R1 and R2, whose association is required for enzymatic activity. Peptidomimetic inhibitors that mimic the C-terminal amino acids of R2 inhibit HSV RR by preventing the association of R1 and R2. These compounds are candidate antiviral therapeutic agents. Here we describe the in vitro selection of HSV type 1 KOS variants with three- to ninefold-decreased sensitivity to the RR inhibitor BILD 733. The resistant isolates have growth properties in vitro similar to those of wild-type KOS but are more sensitive to acyclovir, possibly as a consequence of functional impairment of their RRs. A single amino acid substitution in R1 (Ala-1091 to Ser) was associated with threefold resistance to BILD 733, whereas an additional substitution (Pro-1090 to Leu) was required for higher levels of resistance. These mutations were reintroduced into HSV type 1 KOS and shown to be sufficient to confer the resistance phenotype. Studies in vitro with RRs isolated from cells infected with these mutant viruses demonstrated that these RRs bind BILD 733 more weakly than the wild-type enzyme and are also functionally impaired, exhibiting an elevated dissociation constant (Kd) for R1-R2 subunit association and/or reduced activity (kcat). This work provides evidence that the C-terminal end of HSV R1 (residues 1090 and 1091) is involved in R2 binding interactions and demonstrates that resistance to subunit association inhibitors may be associated with compromised activity of the target enzyme.     

Identification and separation of the two subunits of the herpes simplex virus ribonucleotide reductase.

The herpes simplex virus ribonucleotide reductase is associated with two viral proteins which are both immunoprecipitated by monoclonal antibodies specific for the enzyme. We separated the two proteins and showed that individual antibodies react solely with one or the other. In addition, antibodies to either protein can neutralize enzymatic activity. Our data demonstrate that the proteins are associated in a complex and constitute the subunits of the enzyme.     

Temperature-sensitive mutants of herpes simplex virus type 1 defective in lysis but not in transformation.

Twelve temperature-sensitive (ts) mutants of herpes simplex virus type 1 (HSV-1), representing seven complementation groups, were isolated subsequent to 5-bromodeoxyuridine mutagenesis. These mutants were identified by their inability to replicate in a line of monkey (CV-1) cells at 39 C. Seven of these mutants, representing six complementation groups, induced thymidine kinase (tk) and transformed Ltk- cells, a line of mouse L cells lacking tk, to a tk+ phenotype at both the permissive (34 C) and nonpermissive (39 C) temperatures. Thus, the defective cistrons in these six complementation groups, although necessary for lysis, have no essential function in this transformation system. Transformation by these 12 mutants was dependent on prior UV irradiation. Infection of cells with unirradiated virus under conditions which did not permit virus replication was not sufficient to allow cell transformation. Five mutants, representing two complementation groups, were tk- and were incapable of causing the tk--to-tk+ transformation at either 34 C of 39 C. The tk defects in these mutants are probably unrelated to the ts defects, since one of these complementation groups contains a tk+ member. Therefore, transformation of Ltk- cells to a tk+ phenotype by HSV-1 requires an active viral tk gene. One complementation group was represented by a single tk- member. The role of this cistron in transformation remains undetermined since the primary block to transformation is presumed to be the tk- phenotype. Mutants representing the seven complementation groups were unable to replicate at 39 C in two lines of HSV-1-transformed cells, indicating that the activities of resident wild-type copies of the defective cistrons, if present, could not be detected by complementation.     

Association of ICP0 but not ICP27 with purified virions of herpes simplex virus type 1.

Recent studies have shown that ICP4, one of the major immediate-early proteins of herpes simplex virus type 1 is present within the tegument region of the virion (F. Yao and R. J. Courtney, J. Virol. 63:3338-3344, 1989). With monoclonal antibodies to two additional immediate-early proteins, ICP0 and ICP27, and Western blot (immunoblot) analysis, ICP0, but not ICP27, was also found to be associated with purified virus particles. In an effort to localize the ICP0 within the virion, purified virions were treated with trypsin in the presence and absence of detergent. The data suggest that ICP0 is located within the tegument region of the virion and is not localized in the envelope or within the nucleocapsid. The number of molecules of ICP0 per virion was estimated to be approximately 150.     

The ribonucleotide reductase R1 homolog of murine cytomegalovirus is not a functional enzyme subunit but is required for pathogenesis

Ribonucleotide reductase (RNR) is the key enzyme in the biosynthesis of deoxyribonucleotides. Alpha- and gammaherpesviruses express a functional enzyme, since they code for both the R1 and the R2 subunits. By contrast, betaherpesviruses contain an open reading frame (ORF) with homology to R1, but an ORF for R2 is absent, suggesting that they do not express a functional RNR. The M45 protein of murine cytomegalovirus (MCMV) exhibits the sequence features of a class Ia RNR R1 subunit but lacks certain amino acid residues believed to be critical for enzymatic function. It starts to be expressed independently upon the onset of viral DNA synthesis at 12 h after infection and accumulates at later times in the cytoplasm of the infected cells. Moreover, it is associated with the virion particle. To investigate direct involvement of the virally encoded R1 subunit in ribonucleotide reduction, recombinant M45 was tested in enzyme activity assays together with cellular R1 and R2. The results indicate that M45 neither is a functional equivalent of an R1 subunit nor affects the activity or the allosteric control of the mouse enzyme. To replicate in quiescent cells, MCMV induces the expression and activity of the cellular RNR. Mutant viruses in which the M45 gene has been inactivated are avirulent in immunodeficient SCID mice and fail to replicate in their target organs. These results suggest that M45 has evolved a new function that is indispensable for virus replication and pathogenesis in vivo.     

The fibroblast growth factor receptor is not required for herpes simplex virus type 1 infection.

The early events mediating herpes simplex virus type 1 (HSV-1) infection include virion attachment to cell surface heparan sulfates and subsequent penetration. Recent evidence has suggested that the high-affinity fibroblast growth factor (FGF) receptor mediates HSV-1 entry. This report presents three lines of experimental evidence showing that the high-affinity FGF receptor is not required for HSV-1 infection. First, rat L6 myoblasts lacking FGF receptors were as susceptible to HSV-1 infection as L6 cells genetically engineered to express the FGF receptor. Second, a soluble FGF receptor fragment that inhibited FGF binding and receptor activation did not inhibit HSV-1 infection. Finally, basic FGF (but not acidic FGF) inhibited HSV-1 infection in L6 cells lacking FGF receptors, presumably by blocking cell surface heparan sulfates also required for HSV-1 infection. These results show that the high-affinity FGF receptor is not required for HSV-1 infection but instead that specific low-affinity basic FGF binding sites are used for HSV-1 infection.