Journal of Medical Virology 34:241-247 (1991)
Activation of Cellular Oncogenes by Clinical Isolates and Laboratory Strains of Human Cytomegalovirus I. Boldogh, S. AbuBakar, M.P. Fons, C.Z. Deng, and T. Albrecht Department of Microbiology, University o f Texas Medical Branch, Galveston, Texas 77550 (S.AB., M.P.F., C.Z.D., T A . ) ; Department of Microbiology, Medical Uniuersity of Debrecen, Debrecen, Hungary 4012 f1.B.)
latent infection, and may be reactivated under conditions of immunosuppression [Cohen and Corey, 19851. HCMV is well recognized as a major opportunistic pathogen in patients with defective cell-mediated immunity, in particular those suffering from acquired immune deficiency syndrome (AIDS),and is associated with retinitis, adrenelitis, colitis, pneumonitis, esophageal ulceration, and encephalitis [reviewed in Schooley, 19901.Infections by this virus are related also to severe diseases after therapeutic immunosuppression associated with bone marrow [reviewed in Winston et al., 19901,heart, heart-lung [Wreghitt, 19891,kidney [reviewed in Metselaar and Weimar, 19891, or liver [reviewed in Dummer, 19901transplantation, either as a result of primary or reactivated infection. HCMV is frequently transmitted from mother to infant, either transplacentaly or during the perinatal period [reviewed in Alford et al., 19901. In utero transmission often results in severe disturbance of development and disease that may become manifest at birth, such as microcephaly, splenomegaly, hepatitis, thrombocytopaenia [Stagno and Whitley, 19851or later in life as mental and somatic retardation [Harrison and Myers, 19901. Several types of human malignancy have been linked by molecular hybridization andlor seroepidemiological evidence to HCMV, including cervical [Huang et al., 19841, testicular [Mueller et al., 19881 or prostate [Boldogh et al., 1983; Huang et al., 19841cancer, colon carcinoma [Roche et al., 1981; Huang et al., 19841, Kaposi’s sarcoma [Boldogh et al., 1981; Huang et al., 19841, as well as neuroblastoma and Wilms’ tumor [Wertheim and Voute, 19761.A number of studies have KEY WORDS: HCMV, cellular oncogene acticonfirmed the oncogenic potential of HCMV particles vation, clinical isolates [Albrecht and Rapp, 1973; Geder et al., 1976; Boldogh et al., 19781 and HCMV-DNA fragments [Boldogh et al., 1985; El-Beik et al., 1986; Nelson et al., 1982; INTRODUCTION 19841 in vitro. The active role of HCMV-DNA seHuman cytomegalovirus (HCMV) is a common infec- quences in the maintenance of an oncogenic phenotype, tious agent of worldwide occurrence and important clinical significance. The incidence of HCMV infection varies between 50 and 90% in the adult population of North America or Central Europe [Onorato et al., Accepted for publication April 29, 1991. 19851. In most cases, primary infection of children or Address reprint requests to Thomas Albrecht, PhD, Departadults occurs without overt disease. Subsequent to ment of Microbiology, University of Texas Medical Branch, primary infection, the virus persists, presumably as a Galveston, TX 77550. The effect on cellular (c) oncogene RNA levels was investigated after infection of permissive cells with cell culture adapted strains (AD-169, C-87, Davis) and unadapted clinical isolates (82-1, 84-2, 85-1) of human cytomegalovirus (HCMV). The results indicate that both adapted and unadapted strains of HCMV induce substantial increases i n c-oncogene RNA levels for fos, jun, and myc measured by Northern blot hybridization. Elimination of immediate early (IE) protein synthesis between 0 and 3 hrs or reduction of virus infectivity (99.99%) by UV-irradiation did not reduce the increase in c-oncogene RNA levels. Inhibition of viral and cellular protein synthesis by cycloheximide resulted in a high abundance (superinduction) of specific RNAs which hybridized to c-oncogene probes after infection with either adapted or unadapted strains of HCMV. These data suggest that IE viral gene expression is not essential for activation of c-oncogenes. Inhibition of DNA-dependent RNA synthesis by blocking RNA elongation with actinomycin-D or by inhibiting the activity of RNA polymerase II with alpha-amanitin significantly reduced the increase in c-oncogene RNA levels, suggesting that activation of cellular genes by HCMV is controlled at the level of transcription. Activation of c-oncogenes by HCMV may be particularly important because their protein products appear to be involved in initiation and regulation of viral and cellular gene expression.
0 1991 WILEY-LISS, INC.
242 however, could not be established IBoldogh et al., 1985; reviewed in Macnab, 19871. Unfortunately, the molecular and cellular pathogenesis of HCMV infections is poorly understood. In a recent study, the authors demonstrated tha t a cell culture (cc-) adapted, laboratory strain of HCMV, AD169, induced increased levels of RNAs specific for the immediate early (IE) cellular (c)-oncogenes fos, j u n , and myc [Boldogh et al., 19901. A number of biological differences have been noted, between (cc-adapted HCMV strains and unadapted clinical isolates including the ability to infect leukocytes [Einhorn and Ost, 1984; Rice et al., 19841, the expression of IE antigens, the suppression of lymphocyte proliferation, the abrogation of interleukin-1 activity [Dudding and Garnett, 19871, the development and progression of cytopathology, and the length of the replication cycle [Fons et al., 19861. Accordingly, we investigated the possibility that the ability to activate oncogenes might be related to the cell culture adaptation of HCMV strains. Our findings indicate th at both cc-adapted and unadapted HCMVs stimulate increases in c-fos, c-jun, and c-myc RNA levels. Furthermore, activation of these cellular genes by either unadapted or cc-adapted HCMVs did not require de novo synthesis of viral IE or cellular proteins and was not dependent on virus infectivity. These findings suggest th at activation of cellular oncogenes may be a n intrinsic feature of HCMV infection. MATERIALS AND METHODS Tissue Cu ltu res Human embryonic lung (LU) cells were propagated in Eagle’s minimum essential medium (MEM) supplemented with 8% fetal calf serum (FCS), 100 IU/ml penicillin and 100 pg/ml streptomycin. The cultures were routinely tested for mycoplasma contamination, using Hoecsht stain 33258.
Virus Strains The laboratory adapted strains of HCMV used in these experiments were AD-169, passage 86 [Rowe et al., 19561, C87, passage 35 [Benyesh-Melnick et al., 19641, and Davis, passage 51 [Weller et al., 19571. Clinical isolates were recovered and identified in the laboratory of Dr. Richard Pollard, Department of Internal Medicine, the University of Texas Medical Branch. Isolate 82-1 was recovered from liver tissue of a patient who suffered from disseminated HCMV infection after kidney transplantation. Isolates 84-2 and 85-1 were isolated from urine of congenitally infected children. Identification, ultrastructural and biological characterization of cells infected by these strains has been reported previously by us [Fons et al., 19861. Each clinical isolate was subcultured twice in LU cells to obtain sufficient material for this study and virus stock was stored at -80°C. The infectivity of virus stock was determined by plaque assay prior to each experiment.
Boldogh et al. V iru s Purification The virus suspension obtained after sonication of virus-infected cells was pooled and stored a t -80°C until use. Prior to each experiment the virus suspensions were clarified by low speed (2800 x g for 10 min) and high speed (10,000 x g for 15 min) sedimentation. Virus particles were collected from the clarified fluids by ultracentrifugation at 100,000 x g for 60 min. The pellets were resuspended in serum-free MEM, layered on a linear (20 to 70% wtivol) D-sorbitol gradient, and sedimented at 90,000 x g for 60 min as previously described [Stinski, 1976; Taylor and Cooper, 19891. A dense band of viral particles was recovered and diluted in MEM and sedimented at 100,000 x g for 60 min. The virus pellets were collected, resuspended in MEM, and used for experiments as purified HCMV inocula. A rre s t of Cells and Exposure to HCMV LU cells were subcultured in 150 cm2 plastic flasks and maintained in growth medium for 7 days at 37°C. To arrest cells, 72 h r s prior to beginning each experiment, the culture medium was decanted, the cells were washed twice with PBS, and were refed with serum-free MEM. Prior to mock or virus infection the cell cultures were chilled to 4°C for 30 min. The virus suspension or mock-infecting fluids were added a s a small inoculum (1 ml) to the medium of serum-arrested cells and incubated for 30 min at 4°C to allow for virus adsorption. In control experiments the serum-arrested cells were exposed to 10% FCS for 30 min a t 4°C. At the end of the adsorption period, the temperature of the cell cultures was quickly increased to (0 hr postinfection) and maintained at 37°C. Preparation of UV-Irradiated Viru s Stock Purified, virus inoculum in serum-free MEM was placed in 60 mm tissue culture dishes on ice and exposed to UV-light (254 nm) a t a dose rate of 80 ergs/sec/mm2 for selected lengths of time.
Isolation of Total R NA and N o r t h e r n Blot Analysis Cells were washed twice with ice cold PBS and the total RNA was isolated by the acid guanidium isothiocyanate-phenol-chloroform method [Chomczynski and Sacchi, 19871. Fourteen pg of RNA per lane was fractionated on a 1.4% agarose gel after glyoxal denaturation [Maniatis e t al., 19821 and transferred onto a Zeta probe nylon membrane using 10 mM NaOH [Li et al., 19871. The c-oncogene probes were prepared in this laboratory [Boldogh et al., 19901 from constructs obtained from ATCC (c-fos, c-myc) or kindly provided by I.M. Verma, The Salk Institute, San Diego, CA (c-jun).The oncogene sequences were labelled with 32P-dCTP(specific act: 3,000 Ci/mMol, ICN Biomedicals, CAI, using a random primed DNA labelling kit following the manufacturer’s procedure (Boehringer Mannheim Bio-
Activation of Cellular Oncogenes chemical, Indianapolis, Indiana, USA). Prehybridization, hybridization, and washing conditions were done according to protocols described by Schleiher and Schuell. The hybridizations were visualized by autoradiography and quantified by densitometry.
243
-u 0
c
V
INFECTED
al
c
U 0,
Anticomplement Immunofluorescene (ACIF)Test LU cells grown on coverslips were serum-arrested, precooled and infected as described above. ACIF staining was accomplished using the method described by Reedman and Klein [19731. Polyclonal antibody (purified IgG) to HCMV was kindly provided by Dr. J. Middeldorp (Organon Technika BV., Boxtel, NL). Fluorescein conjugated goat anti-human complement (C3) was purchased from Cappel Corporation. RESULTS Increase in c-Oncogene RNA Levels After Infection With Cell Culture Adapted and Unadapted Strains In a preliminary study, we observed that the kinetics of c-oncogene RNA appearance after infection with either a n adapted (AD-169) or unadapted (82-1) strain of HCMV were similar with maximum increases detected at 40 min postinfection (PI) for c-fos or c j u n , or 60 min PI for c-myc. These results were consistent with recently reported findings [Boldogh et al., 19901. Accordingly, total cellular RNA was isolated at 40 or 60 min PI from HCMV (5 to 6 plaque forming units per cell)- or mock-infected, serum-arrested cells and analyzed by Northern blot hybridization. Both cc-adapted (AD-169, C-87, Davis) and unadapted clinical isolates (82-1,84-2, 85-1) induced significant increases in RNA levels for c-fos, c-jun, and c-rnyc (Fig. 1). Increases in RNA levels for adapted strains were between 4 to 7-fold for c-fos, 4 to 6-fold for c-jun, and 3 to 4-fold for c-myc, relative to the changes measured in cells treated with mock-infecting fluids (Fig. 1). Increases in RNA levels after infection with unadapted clinical isolates of HCMV were consistently lower than those for laboratory adapted strains, but substantially exceeded the increases in c-oncogene RNA levels for cells treated with mock-infecting fluids. In the representative experiment illustrated in Figure 1 increases of 2 to 3-, 2-, and 2.5 to 3-fold were observed for c-fos, c-jun, and c-myc, respectively. Cell culture adapted and unadapted strains of HCMV demonstrated consistent quantitative differences in their ability to induce a n increase in RNA levels of c-oncogenes. Both the cell culture adapted strains and clinical isolates demonstrated a slight delay in the induction of concogene RNA levels relative to serum (data not shown), which is consistent with our previous findings [Boldogh et al., 19901. Expression of IE Viral Genes Is Not Essential for Induction of Increased c-Oncogene RNA Levels In the following series of experiments the possible involvement of HCMV IE gene expression in the stim-
Fig. 1. Northern blot analysis of RNA levels for cellular oncogenes (fos,jun,m y )and actin after infection ofserum-arrested LU cells with cell culture ada ted (AD-169,Davis, C-87) and unadapted (82-1,84-2, 85-1) strains of%uman cytomegalovirus. FCS: fetal calf serum (10%) was used as a positive control. The total RNA was analysed as described in Materials and Methods.
ulation of c-oncogene expression was investigated using UV-irradiated HCMV or inhibitors of protein synthesis. Virus suspensions were inactivated by UV-light (1.92 x lo5 ergs/mm2), which resulted in a substantial reduction in virus infectivity and the inability to induce detectable synthesis of HCMV IE antigens (
Activation of Cellular Oncogenes by Clinical Isolates and Laboratory Strains of Human Cytomegalovirus I. Boldogh, S. AbuBakar, M.P. Fons, C.Z. Deng, and T. Albrecht Department of Microbiology, University o f Texas Medical Branch, Galveston, Texas 77550 (S.AB., M.P.F., C.Z.D., T A . ) ; Department of Microbiology, Medical Uniuersity of Debrecen, Debrecen, Hungary 4012 f1.B.)
latent infection, and may be reactivated under conditions of immunosuppression [Cohen and Corey, 19851. HCMV is well recognized as a major opportunistic pathogen in patients with defective cell-mediated immunity, in particular those suffering from acquired immune deficiency syndrome (AIDS),and is associated with retinitis, adrenelitis, colitis, pneumonitis, esophageal ulceration, and encephalitis [reviewed in Schooley, 19901.Infections by this virus are related also to severe diseases after therapeutic immunosuppression associated with bone marrow [reviewed in Winston et al., 19901,heart, heart-lung [Wreghitt, 19891,kidney [reviewed in Metselaar and Weimar, 19891, or liver [reviewed in Dummer, 19901transplantation, either as a result of primary or reactivated infection. HCMV is frequently transmitted from mother to infant, either transplacentaly or during the perinatal period [reviewed in Alford et al., 19901. In utero transmission often results in severe disturbance of development and disease that may become manifest at birth, such as microcephaly, splenomegaly, hepatitis, thrombocytopaenia [Stagno and Whitley, 19851or later in life as mental and somatic retardation [Harrison and Myers, 19901. Several types of human malignancy have been linked by molecular hybridization andlor seroepidemiological evidence to HCMV, including cervical [Huang et al., 19841, testicular [Mueller et al., 19881 or prostate [Boldogh et al., 1983; Huang et al., 19841cancer, colon carcinoma [Roche et al., 1981; Huang et al., 19841, Kaposi’s sarcoma [Boldogh et al., 1981; Huang et al., 19841, as well as neuroblastoma and Wilms’ tumor [Wertheim and Voute, 19761.A number of studies have KEY WORDS: HCMV, cellular oncogene acticonfirmed the oncogenic potential of HCMV particles vation, clinical isolates [Albrecht and Rapp, 1973; Geder et al., 1976; Boldogh et al., 19781 and HCMV-DNA fragments [Boldogh et al., 1985; El-Beik et al., 1986; Nelson et al., 1982; INTRODUCTION 19841 in vitro. The active role of HCMV-DNA seHuman cytomegalovirus (HCMV) is a common infec- quences in the maintenance of an oncogenic phenotype, tious agent of worldwide occurrence and important clinical significance. The incidence of HCMV infection varies between 50 and 90% in the adult population of North America or Central Europe [Onorato et al., Accepted for publication April 29, 1991. 19851. In most cases, primary infection of children or Address reprint requests to Thomas Albrecht, PhD, Departadults occurs without overt disease. Subsequent to ment of Microbiology, University of Texas Medical Branch, primary infection, the virus persists, presumably as a Galveston, TX 77550. The effect on cellular (c) oncogene RNA levels was investigated after infection of permissive cells with cell culture adapted strains (AD-169, C-87, Davis) and unadapted clinical isolates (82-1, 84-2, 85-1) of human cytomegalovirus (HCMV). The results indicate that both adapted and unadapted strains of HCMV induce substantial increases i n c-oncogene RNA levels for fos, jun, and myc measured by Northern blot hybridization. Elimination of immediate early (IE) protein synthesis between 0 and 3 hrs or reduction of virus infectivity (99.99%) by UV-irradiation did not reduce the increase in c-oncogene RNA levels. Inhibition of viral and cellular protein synthesis by cycloheximide resulted in a high abundance (superinduction) of specific RNAs which hybridized to c-oncogene probes after infection with either adapted or unadapted strains of HCMV. These data suggest that IE viral gene expression is not essential for activation of c-oncogenes. Inhibition of DNA-dependent RNA synthesis by blocking RNA elongation with actinomycin-D or by inhibiting the activity of RNA polymerase II with alpha-amanitin significantly reduced the increase in c-oncogene RNA levels, suggesting that activation of cellular genes by HCMV is controlled at the level of transcription. Activation of c-oncogenes by HCMV may be particularly important because their protein products appear to be involved in initiation and regulation of viral and cellular gene expression.
0 1991 WILEY-LISS, INC.
242 however, could not be established IBoldogh et al., 1985; reviewed in Macnab, 19871. Unfortunately, the molecular and cellular pathogenesis of HCMV infections is poorly understood. In a recent study, the authors demonstrated tha t a cell culture (cc-) adapted, laboratory strain of HCMV, AD169, induced increased levels of RNAs specific for the immediate early (IE) cellular (c)-oncogenes fos, j u n , and myc [Boldogh et al., 19901. A number of biological differences have been noted, between (cc-adapted HCMV strains and unadapted clinical isolates including the ability to infect leukocytes [Einhorn and Ost, 1984; Rice et al., 19841, the expression of IE antigens, the suppression of lymphocyte proliferation, the abrogation of interleukin-1 activity [Dudding and Garnett, 19871, the development and progression of cytopathology, and the length of the replication cycle [Fons et al., 19861. Accordingly, we investigated the possibility that the ability to activate oncogenes might be related to the cell culture adaptation of HCMV strains. Our findings indicate th at both cc-adapted and unadapted HCMVs stimulate increases in c-fos, c-jun, and c-myc RNA levels. Furthermore, activation of these cellular genes by either unadapted or cc-adapted HCMVs did not require de novo synthesis of viral IE or cellular proteins and was not dependent on virus infectivity. These findings suggest th at activation of cellular oncogenes may be a n intrinsic feature of HCMV infection. MATERIALS AND METHODS Tissue Cu ltu res Human embryonic lung (LU) cells were propagated in Eagle’s minimum essential medium (MEM) supplemented with 8% fetal calf serum (FCS), 100 IU/ml penicillin and 100 pg/ml streptomycin. The cultures were routinely tested for mycoplasma contamination, using Hoecsht stain 33258.
Virus Strains The laboratory adapted strains of HCMV used in these experiments were AD-169, passage 86 [Rowe et al., 19561, C87, passage 35 [Benyesh-Melnick et al., 19641, and Davis, passage 51 [Weller et al., 19571. Clinical isolates were recovered and identified in the laboratory of Dr. Richard Pollard, Department of Internal Medicine, the University of Texas Medical Branch. Isolate 82-1 was recovered from liver tissue of a patient who suffered from disseminated HCMV infection after kidney transplantation. Isolates 84-2 and 85-1 were isolated from urine of congenitally infected children. Identification, ultrastructural and biological characterization of cells infected by these strains has been reported previously by us [Fons et al., 19861. Each clinical isolate was subcultured twice in LU cells to obtain sufficient material for this study and virus stock was stored at -80°C. The infectivity of virus stock was determined by plaque assay prior to each experiment.
Boldogh et al. V iru s Purification The virus suspension obtained after sonication of virus-infected cells was pooled and stored a t -80°C until use. Prior to each experiment the virus suspensions were clarified by low speed (2800 x g for 10 min) and high speed (10,000 x g for 15 min) sedimentation. Virus particles were collected from the clarified fluids by ultracentrifugation at 100,000 x g for 60 min. The pellets were resuspended in serum-free MEM, layered on a linear (20 to 70% wtivol) D-sorbitol gradient, and sedimented at 90,000 x g for 60 min as previously described [Stinski, 1976; Taylor and Cooper, 19891. A dense band of viral particles was recovered and diluted in MEM and sedimented at 100,000 x g for 60 min. The virus pellets were collected, resuspended in MEM, and used for experiments as purified HCMV inocula. A rre s t of Cells and Exposure to HCMV LU cells were subcultured in 150 cm2 plastic flasks and maintained in growth medium for 7 days at 37°C. To arrest cells, 72 h r s prior to beginning each experiment, the culture medium was decanted, the cells were washed twice with PBS, and were refed with serum-free MEM. Prior to mock or virus infection the cell cultures were chilled to 4°C for 30 min. The virus suspension or mock-infecting fluids were added a s a small inoculum (1 ml) to the medium of serum-arrested cells and incubated for 30 min at 4°C to allow for virus adsorption. In control experiments the serum-arrested cells were exposed to 10% FCS for 30 min a t 4°C. At the end of the adsorption period, the temperature of the cell cultures was quickly increased to (0 hr postinfection) and maintained at 37°C. Preparation of UV-Irradiated Viru s Stock Purified, virus inoculum in serum-free MEM was placed in 60 mm tissue culture dishes on ice and exposed to UV-light (254 nm) a t a dose rate of 80 ergs/sec/mm2 for selected lengths of time.
Isolation of Total R NA and N o r t h e r n Blot Analysis Cells were washed twice with ice cold PBS and the total RNA was isolated by the acid guanidium isothiocyanate-phenol-chloroform method [Chomczynski and Sacchi, 19871. Fourteen pg of RNA per lane was fractionated on a 1.4% agarose gel after glyoxal denaturation [Maniatis e t al., 19821 and transferred onto a Zeta probe nylon membrane using 10 mM NaOH [Li et al., 19871. The c-oncogene probes were prepared in this laboratory [Boldogh et al., 19901 from constructs obtained from ATCC (c-fos, c-myc) or kindly provided by I.M. Verma, The Salk Institute, San Diego, CA (c-jun).The oncogene sequences were labelled with 32P-dCTP(specific act: 3,000 Ci/mMol, ICN Biomedicals, CAI, using a random primed DNA labelling kit following the manufacturer’s procedure (Boehringer Mannheim Bio-
Activation of Cellular Oncogenes chemical, Indianapolis, Indiana, USA). Prehybridization, hybridization, and washing conditions were done according to protocols described by Schleiher and Schuell. The hybridizations were visualized by autoradiography and quantified by densitometry.
243
-u 0
c
V
INFECTED
al
c
U 0,
Anticomplement Immunofluorescene (ACIF)Test LU cells grown on coverslips were serum-arrested, precooled and infected as described above. ACIF staining was accomplished using the method described by Reedman and Klein [19731. Polyclonal antibody (purified IgG) to HCMV was kindly provided by Dr. J. Middeldorp (Organon Technika BV., Boxtel, NL). Fluorescein conjugated goat anti-human complement (C3) was purchased from Cappel Corporation. RESULTS Increase in c-Oncogene RNA Levels After Infection With Cell Culture Adapted and Unadapted Strains In a preliminary study, we observed that the kinetics of c-oncogene RNA appearance after infection with either a n adapted (AD-169) or unadapted (82-1) strain of HCMV were similar with maximum increases detected at 40 min postinfection (PI) for c-fos or c j u n , or 60 min PI for c-myc. These results were consistent with recently reported findings [Boldogh et al., 19901. Accordingly, total cellular RNA was isolated at 40 or 60 min PI from HCMV (5 to 6 plaque forming units per cell)- or mock-infected, serum-arrested cells and analyzed by Northern blot hybridization. Both cc-adapted (AD-169, C-87, Davis) and unadapted clinical isolates (82-1,84-2, 85-1) induced significant increases in RNA levels for c-fos, c-jun, and c-rnyc (Fig. 1). Increases in RNA levels for adapted strains were between 4 to 7-fold for c-fos, 4 to 6-fold for c-jun, and 3 to 4-fold for c-myc, relative to the changes measured in cells treated with mock-infecting fluids (Fig. 1). Increases in RNA levels after infection with unadapted clinical isolates of HCMV were consistently lower than those for laboratory adapted strains, but substantially exceeded the increases in c-oncogene RNA levels for cells treated with mock-infecting fluids. In the representative experiment illustrated in Figure 1 increases of 2 to 3-, 2-, and 2.5 to 3-fold were observed for c-fos, c-jun, and c-myc, respectively. Cell culture adapted and unadapted strains of HCMV demonstrated consistent quantitative differences in their ability to induce a n increase in RNA levels of c-oncogenes. Both the cell culture adapted strains and clinical isolates demonstrated a slight delay in the induction of concogene RNA levels relative to serum (data not shown), which is consistent with our previous findings [Boldogh et al., 19901. Expression of IE Viral Genes Is Not Essential for Induction of Increased c-Oncogene RNA Levels In the following series of experiments the possible involvement of HCMV IE gene expression in the stim-
Fig. 1. Northern blot analysis of RNA levels for cellular oncogenes (fos,jun,m y )and actin after infection ofserum-arrested LU cells with cell culture ada ted (AD-169,Davis, C-87) and unadapted (82-1,84-2, 85-1) strains of%uman cytomegalovirus. FCS: fetal calf serum (10%) was used as a positive control. The total RNA was analysed as described in Materials and Methods.
ulation of c-oncogene expression was investigated using UV-irradiated HCMV or inhibitors of protein synthesis. Virus suspensions were inactivated by UV-light (1.92 x lo5 ergs/mm2), which resulted in a substantial reduction in virus infectivity and the inability to induce detectable synthesis of HCMV IE antigens (
