Anti-complement immunofluorescence establishes nuclear localization of human cytomegalovirus matrix protein

A monospecific, polyclonal antiserum to the 69-kDa matrix protein of human cytomegalovirus (HCMV) was prepared in a guinea pig and used to determine the intracellular distribution of this viral antigen. The resulting antiserum was specific for infected cells as tested by immunofluorescence, and specific for the HCMV matrix protein as determined by "nitrocellulose immunoassay" of electrophoretically separated, infected-cell proteins. Antibodies were reacted with fixed, infected human fibroblasts, and visualized by the anti-complement immunofluorescence procedure to avoid complications arising from the strong IgG Fc binding activity of the infected-cell-specific cytoplasmic inclusion. Results establish that the matrix protein is located in the nucleus, and indicate that it is concentrated in the nucleoplasm rather than within the intranuclear inclusions.     

Evidence that HIV-1 gag precursor shares antigenic sites with the major capsid protein of human cytomegalovirus.

A rabbit antiserum prepared against disrupted sucrose-banded HIV-1 virus (strain FRE-3) reacted with antigens present in nuclear inclusions, pathognomonic for human cytomegalovirus (HCMV). This cross-reactivity was observed in autopsy specimens from individuals infected with CMV, in the presence or absence of co-infection with HIV-1. A Towbin immunoassay showed that the serum reacted specifically with the HCMV major capsid protein (MCP, 153 kDa), both in the nuclear fraction of infected cells and in virions. Direct evidence that these proteins share antigenic determinants was provided by the two-way cross-reactivity of affinity-selected antibodies (i.e., anti-MCP with HIV-1 gag precursor Pr55; anti-Pr55 with MCP). All four strains of HCMV tested showed this reactivity, but the counterpart proteins of simian CMV and herpes simplex virus type 1 did not, indicating that the determinant is not common to all herpes group viruses.     

An intact sequence-specific DNA-binding domain is required for human cytomegalovirus-mediated sequestration of p53 and may promote in vivo binding to the viral genome during infection

The p53 protein is stabilized during infection of primary human fibroblasts with human cytomegalovirus (HCMV). However, the p53 in HCMV-infected cells is unable to activate its downstream targets. HCMV accomplishes this inactivation, at least in part, by sequestering p53 into viral replication centers within the cell's nucleus soon after they are established. In order to better understand the interplay between HCMV and p53 and the mechanism of sequestration, we constructed a panel of mutant p53-GFP fusion constructs for use in transfection/infection experiments. These mutants affected several post-translational modification sites and several sites within the central sequence-specific DNA-binding domain of the protein. Two categories of p53 sequestration were observed when the mutant constructs were transfected into primary fibroblasts and then infected at either high or low multiplicity. The first category, including all of the post-translational modification mutants, showed sequestration comparable to a wild-type (wt) control, while the second category, mutants affecting the DNA-binding core, were not specifically sequestered above control GFP levels. This suggested that the DNA-binding ability of the protein was required for sequestration. When the HCMV genome was analyzed for p53 consensus binding sites, 21 matches were found, which localized either to the promoters or the coding regions of viral proteins involved in DNA replication and processing as well as structural proteins. An analysis of in vivo binding to these identified sites via chromatin immunoprecipitation assays revealed differential binding to several of the sites over the course of infection.     

Human antibody response to human cytomegalovirus-specific DNA-binding proteins

Human antibody responses to human cytomegalovirus (HCMV) specific DNA-binding proteins were studied in serum samples by the Western blot technique. The molecular weights of six DNA-binding proteins found in HCMV-infected cells, ranged from 52kD to 18kD. The sera obtained from patients with acute HCMV infections reacted well with the six HCMV specific DNA-binding proteins. The strongest reactivity was observed with the 52kD and 35kD proteins. The sera from healthy HCMV seropositive donors reacted only with the 52kD DNA-binding protein as visualized in Western blots, but 2 out of 8 sera failed to react with any HCMV specific DNA-binding proteins.     

Altered expression of adenovirus 12 DNA-binding protein but not DNA polymerase during abortive infection of hamster cells.

Replication of human adenovirus type 12 DNA is blocked in abortively infected baby hamster kidney cells. The activity and accumulation of adenovirus 12 DNA polymerase is equivalent in infected hamster and human cell extracts. However, the accumulation of adenovirus type 12 DNA-binding protein is approximately 120-fold lower in extracts from infected hamster cells when compared to infected permissive human cells. This difference in accumulation is not due to replication of viral DNA during productive infection, since this difference is observed in the presence of hydroxyurea. The DNA-binding protein from infected hamster cells retains the ability to bind denatured DNA-cellulose. An adenovirus 5 early region 1 transformed hamster cell line competent to complement the adenovirus 12 DNA replication defect also stimulates accumulation of the DNA-binding protein even when the cells are treated with hydroxyurea. Thus, the reduced expression of the viral DNA-binding protein may play a role in the mechanism of abortive infection of hamster cells by adenovirus 12.     

Cytomegalovirus-infected cells contain a DNA-binding protein

Cells infected with cytomegalovirus (CMV, strain Colburn) contain large amounts of an intranuclear 51,000 dalton (51K) protein. This protein was dramatically enriched for by two different procedures used to identify DNA-binding proteins. The first of these demonstrated that the 51K protein was recovered following polyethyleneimine/high salt extraction of infected cells for proteins associated with DNA in vivo. And the second showed that, when analyzed by affinity chromatography using single-strand DNA coupled to Sepharose beads, the 51K protein was bound efficiently in low salt; was partially (40%) eluted at 0.30 M salt, along with “B-capsid” and “matrix” virus structural proteins; and constituted greater than 70% of the protein eluted at 0.45 M salt. Two-dimensional (charge-size) separations in denaturing polyacrylamide gels established that this acid soluble protein has a comparatively strong net positive charge; resolved the infected cell-specific 51K protein from a more neutrally charged host cell protein of about the same size; and verified observations that the virus-specific 51K protein is phosphorylated. In addition, results of biosynthetic radiolabeling experiments showed that expression of the 51K protein begins prior to that of most other viral proteins, but requires preceding viral protein and DNA synthesis.     

Comparative study with monospecific and monoclonal antibodies gainst a 65 K human cytomegalovirus protein

Summary Immunofluorescence assay using monospecific and monoclonal antibodies to the 65K major protein of human cytomegalovirus (HCMV) was carried out to monitor the expression of this protein in infected cells. Regardless of differences in the reactivity of the monoclonal antibodies, as determined by immunoblotting and immunofluorescent staining, all stained cytoplasmic inclusion bodies localized to the site of the HCMV-induced receptor for the Fc portion of IgG, suggesting that most of the 65K major protein of HCMV colocalizes with the HCMV-induced FcR.     

Antibody reactivity to human and vervet cytomegalovirus early antigen(s) in sera from patients with active cytomegalovirus infections and from asymptomatic donors.

Human fibroblasts were infected with vervet cytomegalovirus (VCMV) and cultured in medium containing 50 micrograms of cytosine arabinoside per ml for 72 h. Early antigens (EAg) were detected in the nuclei of infected cells by an indirect fluorescent antibody test with human sera having antibody to EAg of human cytomegalovirus (HCMV). Sera from patients with serological and/or virological evidence of active HCMV infection and from asymptomatic blood donors were examined for antibody activity with the HCMV and VCMV EAg's. The HCMV and VCMV EAg's were comparable in detecting levels of antibody activity and fluctuations in antibody titer of paired sera from patients. A total of 81% of patient sera reactive with HCMV EAg were also reactive with VCMV EAg. In contrast, only 14% of the asymptomatic donor sera reactive with HCMV EAg were also reactive with VCMV EAg. The VCMV EAg, therefore, appeared to differentiate latent from active infections in humans more effectively than did HCMV EAg.     

Monoclonal antibodies which recognize native and denatured forms of the adenovirus DNA-binding protein

Two hybridoma cell lines were obtained, A1 and B6, which produced monoclonal antibodies reacting with the 44,000-MW C-terminal domain of the adenovirus type 5 DNA-binding protein (DBP). Clone A1 antibodies reacted with the native form of the DBP, but failed to recognize this protein after denaturation (by exposure to sodium dodecylsulfate, or production of the DBP at 39.5 degrees by H5ts107, a temperature-sensitive DBP mutant). Clone B6 antibodies bound to both the native and denatured forms of the DBP. Immunofluorescent staining of wild-type-virus infected cells revealed the DBP located in discrete nuclear patches. A1 and B6 antibodies detected this patched localization of the DBP in nuclei of H5ts107-infected cells grown at 32 degrees. However, at the nonpermissive temperature of 39.5 degrees, A1 antibodies failed to detect the DBP, and B6 antibodies gave a uniform nuclear fluorescent distribution of the DBP. Thus the nuclear pattern of localization for the DBP synthesized by H5ts107 was temperature dependent in this mutant.     

Identification and characterization of deoxyribonucleoprotein complexes containing the major DNA-binding protein of herpes simplex virus type 1

The stimulation of host cell DNA synthesis was studied in permissive human embryonic lung (HEL) cells and in nonpermissive rabbit kidney (RK) cells infected with human cytomegalovirus (HCMV). Host cell DNA synthesis was induced by HCMV infection in resting cells of both types. In permissive cultures the stimulation of cellular DNA synthesis was detectable mainly in those cells which had not become productively infected and in which virus antigens were not detectable. In abortively infected RK cells, on the other hand, stimulation of host cell DNA synthesis and the expression of virus antigens were detected in the same cells. Infection of actively growing permissive HEL cells resulted in a shutdown of cellular DNA synthesis beginning approximately 10 hr postinfection. Shutdown of cellular DNA synthesis also occurred when the infected cells were treated with phosphonoacetic acid and was thus classified as an "early" virus function. In actively growing, abortively infected RK cells, on the other hand, host cell DNA synthesis was not affected, indicating that the early virus function(s) responsible for inhibition of cellular DNA synthesis was not expressed in these cells. Virus-encoded DNA polymerase activity, another early virus gene function, was also not detected in these abortively infected cultures. In RK cells the cellular DNA synthesized as a result of infection was capable of undergoing at least one further round of replication, indicating that the HCMV gene expression which occurred in abortively infected RK cells was not lethal for these cells.     

Ultrastructural localization of viral nucleic acid by in situ hybridization

In situ hybridization has become a standard technique in the localization of viral nucleic acids in tissue sections and cytologic preparations at the light microscopic level. We have extended this technique to the electron microscopic level using human cytomegalovirus (CMV) infection in cultured human foreskin fibroblasts, and have shown for the first time that colloidal gold can be used to study intranuclear localization of viral replication. CMV-infected fibroblasts exhibiting early (4-day) and late (18-day) cytopathic effect were fixed in formalin, gently permeabilized with detergent and protease, and hybridized with a biotinylated CMV DNA probe. Hybridized sequences were localized by a pre-embedding technique using streptavidin-conjugated 15 to 20 nm colloidal gold particles. Ultrastructural nuclear and cytoplasmic architecture were well preserved through permeabilization and hybridization steps. Viral DNA was clearly detected in fibroblast nuclei containing nascent and well-formed electron-dense viral inclusions. Gold particles were localized to the periphery of electron-dense nuclear inclusions, occasionally in association with 70 nm nuclear dense bodies, but not with complete viral nucleocapsids. DNA hybridization was abolished by pretreatment of infected cells with DNase. Cross-hybridization of CMV DNA sequences with human DNA or with herpes simplex virus genome was not observed. The ultrastructural findings suggest that CMV DNA replication may occur at the margins of electron-dense regions in maturing viral inclusions, and that viral DNA associated with core dense bodies is available for hybridization with complementary nucleic acid sequences. This technique can be useful in studies of viral pathogenesis.     

Sequence Analysis and Expression of the Murine Cytomegalovirus Phosphoprotein pp50, a Homolog of the Human Cytomegalovirus UL44 Gene Product

The coding sequence for the murine cytomegalovirus (MCMV) nonstructural phosphoprotein pp50 has previously been mapped to the EcoRI-L fragment of the viral genome. Nucleotide sequencing of both genomic DNA and cDNA clones revealed an open reading frame of 411 amino acids that is homologous to both the human cytomegalovirus (HCMV) UL44 (ICP36) and the human herpesvirus type 6 (HHV-6) p41 open reading frames. Four of the five conserved cysteine residues are located within the highly conserved N-termini of the three proteins. Similar to the HCMV and HHV-6 proteins, pp50 is predicted to have a hydrophilic C-terminus which contains a putative nuclear targeting sequence. Immunofluorescence experiments with monoclonal antibodies specific for pp50 demonstrated that recombinant pp50 expressed transiently in COS-1 cells or during infection by recombinant Vaccinia virus was localized to the nucleus. Similarly, during MCMV infection of Balb/3T3 fibroblasts, pp50 was present in the nuclei of infected cells between 8 and 24 hr postinfection, suggesting that the protein contains a functional nuclear targeting sequence. In addition, we have demonstrated that pp50 is a DNA-binding protein with affinities for both single-stranded and double-stranded DNA. Finally, results from our analysis by nonreducing and native gel electrophoresis, as well as fractionation by sucrose gradients suggested that pp50 expressed in the absence of viral infection can form disulfide-linked homopolymers. Such homopolymers were also present in MCMV-infected cells.     

The synthesis of the major DNA-binding protein (ICP8) in cells infected with the strain HSZP or KOS of herpes simplex virus type 1.

Synthesis of the major DNA-binding protein (ICP8) was investigated in primary rabbit kidney (RK) and Vero cells infected with the syncytial (syn) strain HSZP or with the non-syn strain KOS of herpes simplex virus type 1 (HSV-1). Results showed the following: 1. In contrast to strain KOS, the rate of viral polypeptide synthesis was accelerated in Vero cells infected with strain HSZP. The ICP8 could be detected in the nuclei of cells by one hour post-infection (hr p. i.) where it became associated with the viral DNA (DNase sensitive form). Later on (7 hr p.i.), the synthesis of viral polypeptides decreased and no further translocation of ICP8 from the cytoplasm into the nucleus was observed. 2. Strain HSZP was approx. three times more resistant to the action of phosphonoacetic acid (PAA) than strain KOS. In order to block the synthesis of HSZP gamma-2 polypeptides, a concentration of 600 micrograms PAA/ml had to be used. Under this condition, the HSZP ICP8 was translocated into the cell nucleus at later interval only (7 hr p.i.), and it was still possible to release this polypeptide from the nucleus by DNase treatment. The failure of the HSZP ICP8 to associate with the nuclear matrix (DNase resistant form) of infected cells in the absence of viral DNA replication may reflect its predominant affinity for the viral DNA which, in turn, may be responsible for the observed accelerated synthesis of the HSZP polypeptides in infected Vero cells. 3. In primary RK cells infected with strain HSZP the ICP8 did not translocate into the cell nucleus. Therefore, no gamma-2 polypeptides were synthesized.     

The DNA-binding protein P52 of human cytomegalovirus reacts with monoclonal antibody CCH2 and associates with the nuclear membrane at late times after infection.

Monoclonal antibody CCH2 is commonly used for the detection of human cytomegalovirus (HCMV) infected cells in tissue sections as well as in cultured cells. The specificity of CCH2 was determined by screening a recombinant lambda-gt11 cDNA gene bank from HCMV-infected fibroblasts. By sequencing a reactive clone, the antigen was identified to be the non-structural DNA binding protein p52 of HCMV (UL44 reading frame). The viral insert from the lambda clone was recloned in bacterial expression vectors. For this, a new vector, pRos-RS, was constructed. The resulting clones were tested in immunoblot analyses. They were reactive with CCH2 as well as with reconvalescent sera positive for antibodies against HCMV, by this proving the specificity of CCH2. Using this monoclonal antibody in confocal microscopy, the subcellular localization of p52 in infected cells was analyzed. In these analyses, p52 was found to be nuclear and to be associated with the nuclear membrane at late times after infection.     

Genomic locus for a 140 kDa structural protein (pp150) of human cytomegalovirus in strains Towne and AD169

An HCMV specific clone was isolated from a genomic library of human cytomegalovirus (HCMV) DNA cloned into the expression vector lambda gt11. This clone (lambda 111-1) expressed an HCMV/beta-galactosidase fusion protein which was reactive with rabbit antibody prepared against purified HCMV virions and dense bodies as well as human HCMV immune serum. By probing Western blots of HCMV virion proteins or HCMV-infected cells with antibody prepared against the fusion protein, the authentic gene product of clone lambda 111-1 was identified as a high molecular weight polypeptide of 140. Probing the restriction digests of HCMV DNA with insert DNA from the immunoreactive lambda gt11 clone permitted us to localize the coding sequence for the 140 kDa polypeptide to the long unique region (map coordinates of 0.16-0.18) on HCMV Towne and AD169 genomes.     

Monoclonal antibodies to cytomegalovirus: rapid identification of clinical isolates and preliminary use in diagnosis of cytomegalovirus pneumonia.

Two monoclonal antibodies which react specifically with cells infected by cytomegalovirus (CMV) are described. One antibody, 6-E3, reacts with a 72,000-dalton protein that appears early in infection and remains localized in the cell nucleus. The other antibody, 6-C5, reacts with an 80,000-dalton protein that appears late in infection and remains localized in cytoplasmic inclusion bodies. Both monoclonal antibodies react with conventional laboratory strains of CMV and can be used in immunofluorescence assays to identify clinical isolates of CMV in culture. Preliminary tests on lung tissues from patients with CMV pneumonia show that only antibody 6-C5 detects CMV infection in primary clinical specimens. A comparison of culture, histological, and immunological methods demonstrates that the monoclonal antibodies possess sufficient specificity and sensitivity to warrant their continued development as immunodiagnostic tools for the detection of CMV infection in both tissue culture and tissues obtained directly from patients.     

Comparison of different immunostaining techniques and monoclonal antibodies to the lower matrix phosphoprotein (pp65) for optimal quantitation of human cytomegalovirus antigenemia.

The main parameters of immunostaining techniques, i.e., the type of fixative, immunocytochemical reaction, and quality of monoclonal antibodies (MAbs), for quantitation of human cytomegalovirus (HCMV) antigenemia in peripheral blood polymorphonuclear leukocytes (currently performed by the indirect immunofluorescence or immunoperoxidase reaction by using MAbs to HCMV pp65) were investigated in order to optimize procedural steps and reagents. Significantly better results (in terms of the number of positive cells) were obtained on multiple cytospin preparations from heart transplant recipients with HCMV viremia when we used (i) formalin instead of methanol-acetone fixation and (ii) the indirect immunofluorescence reaction instead of the immunoperoxidase reaction, the avidin-biotin complex method, or the alkaline phosphatase antialkaline phosphatase procedure. In addition, comparison of the staining capabilities of three MAbs to pp65, which were developed in the laboratory and which were reactive to different epitopes of the protein, with a commercially available MAb (Clonab CMV) for determination of HCMV antigenemia showed that, while individual MAbs did not provide better results, the pool of MAbs detected a significantly higher number of positive peripheral blood polymorphonuclear leukocytes than Clonab CMV did. In addition, the sensitivity of the pool in detecting patients with low levels of viremia (less than 5/2 x 10(5) cells inoculated) as antigenemia positive was 100%, whereas the sensitivity of Clonab CMV was 47%. No differences in the specificities between the two MAb preparations were observed.