JOURNAL OF VIROLOGY, Apr. 1979, p. 84-89 0022-538X/79/04-0084/06$02.00/0
Vol. 30, No. 1
Transcription of the Marek's Disease Virus Genome in VirusInduced Tumors SANDRA SILVER, MARY SMITH, AND MEIHAN NONOYAMA* Laboratory of Molecular Virology, Life Sciences, Inc., St. Petersburg, Florida 33710 Received for publication 14 August 1978
Transcription of the Marek's disease virus (MDV) genome in tumor tissues from MDV-infected chickens has been studied by analyzing the hybridization kinetics of 3H-labeled MDV DNA with unlabeled RNA extracted from these tissues. Lymphoid tumors of ovary, spleen, liver, and kidney contained MDV genomes, but the virus-specific RNA sequences were transcribed from less than 15% of the viral DNA. A virus nonproductive lymphoblastoid cell line, designated MKT-1, has been established from a kidney lymphoma and contains 15 MDV genomes per cell. In these cells, 12 to 14% of the viral DNA was transcribed. Thus transcription of the MDV genome was restricted both in tumor tissues and MKT1 cells. A hybridization experiment where RNA extracted from MKT-1 cells and RNA extracted from a spleen tumor were mixed and hybridized to 'H-labeled MDV DNA indicated that the virus-specific RNAs from the two sources were encoded by the same DNA sequences. The polyribosomal fractions of MKT-1 cells and this spleen tumor contained only a portion of the virus-specific RNA sequences found in whole-cell extracts, indicating the existence of a posttranscriptional control mechanism which prevents the transfer of certain viral RNA transcripts to the polyribosomes. The data suggest that the repressed expression of the viral genome in lymphoid tumor tissues and MKT-1 cells may be the result of precise controls within the cell at the transcriptional and posttranscriptional levels.
Using DNA-RNA hybridization techniques (8), the virus-specific RNA in MKT-1 and several tumor tissues were analyzed to compare the extent of transcription of the viral genome and to determine whether a specific portion of the MDV DNA is transcribed. Hybridization experiments using RNA from whole-cell extracts and RNA from polyribosomes can determine the proportion of viral RNA sequences selectively transferred to the polyribosomes. This may offer some insight into the control mechanism at the posttranscriptional level in virus-induced tumors and transformed cells.
Marek's disease (MD) virus (MDV) is a herpesvirus which causes a lymphoproliferative disease in chickens (4, 11). The virus replicates in the feather follicle epithelium and produces infectious cell-free virus (6), whereas in MDVinfected cell cultures the virus is cell associated, usually causing a cytopathic effect (7, 12). Lymphoid tumors from MDV-infected chickens contain copies of the MDV genome but are usually free of MD antigens and virus particles (13); however, certain established cell lines derived from MD lymphomas produce virus particles (or viral antigens) in a small percentage of the cell population (1, 2, 16). In contrast, a lymphoblastoid cell line, designated MKT-1, has been established from an MD kidney lymphoma, is nonproductive of virus and viral antigens, but contains 15 complete MDV genomes per cell
MATERIALS AND METHODS Tumor tissues. Tumors were collected from 5- to 8-week-old, isolator-held, LSI-SPF chickens that were inoculated intraabdominally with virus at 2 days of age. Four of the tumors examined were from chickens inoculated with 100 to 150 focus-forming units of GA strain cell-free MDV, extracted from feather follicle epithelium, per bird. The other three were from chickens concurrently infected with MDV (same dosage as above) and Rous-associated virus-2 (RAV-2) at a dosage of 102 to 104 PFU per bird. Cells. MKT-1 cells were maintained at 41°C in RPMI 1640 medium supplemented with 10% fetal calf
(19). Since MDV induces tumors in its natural host and the MKT-1 cell line was derived from such a tumor, experiments to analyze regulations at the transcriptional and posttranscriptional levels were undertaken to determine, in part, if the mechanisms involved in the repressed expression of the MDV genome are similar in lymphoid tumors and in MKT-1 cells. 84
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
serum and antibiotics. Maximum cell density reached approximately 107 cells per ml. Chicken embryo fibroblast (CEF) cultures infected with MDV were grown in minimal essential medium supplemented with 10% tryptose phosphate broth and 5% fetal calf serum. Cells were harvested when 50% of the culture showed cytopathic effects. Preparation of RNA extracted from whole cells. The RNA extraction procedure used was that described by Hayward (10) with modifications. Each cell pellet or homogenized tumor tissue was suspended in 9 volumes of RNA extraction buffer consisting of 0.01 M Tris-hydrochloride (pH 8.1), 0.02 M NaCl, and 0.002 M CaCl2. Proteinase K (Beckman, Palo Alto, Calif.) and sodium dodecyl sulfate were added to the suspensions at final concentrations of 1.0 mg/ml and 1.0%, respectively, and the lysates were incubated at 37°C for 45 to 60 min. Each lysate was then passed through a 20-gauge needle three to four times to reduce viscosity caused by cellular DNA. NaCl was added to a final concentration of 0.15 M, and EDTA (pH 7.4) was added to 0.002 M. The nucleic acids were extracted three times at room temperature with a mixture of phenol-chloroform-isoamyl alcohol (1:1: 0.01 vol/vol/vol) containing 0.05% 8-hydroxyquinoline. The aqueous phase was precipitated overnight at -20°C after the addition of 2 volumes of ethanol. The nucleic acids were then pelleted, suspended in 5 to 10 ml of TNM buffer (consisting of 0.01 M Tris-hydrochloride, pH 7.4, 0.15 M NaCl, and 0.001 M MgCl2), and digested with 20 jig of DNase 1 (Worthington Biochemical Corp., Freehold, N.J.) per ml for 20 min at 37°C. The protein was removed by extraction with phenol-chloroform-isoamyl alcohol-8-hydroxyquinoline, and the resulting nucleic acids were alcohol precipitated, as described previously. After two additional cycles of DNase treatment, phenol extraction, and alcohol precipitation, the resulting RNA was suspended in less than 1.0 ml of 0.0025 M EDTA (pH 7.4). Approximately 1 mg of RNA per 2.5 x 108 cells was obtained. The purity of RNA was checked by diphenylamine assay, and the preparation showed no detectable DNA. Preparation of polyribosomal RNA. The method used for isolating polyribosomes of MKT-1 cells and tumor tissues was that described by Hayward and Kieff (9). Briefly, cell pellets or homogenized tumor tissue was suspended in a buffer consisting of 0.01 M Tris-hydrochloride (pH 7.4), 0.01 M NaCl, and 0.0015 M MgCl2, treated with 0.5% Nonidet P-40 at 4°C for 15 min, and homogenized with a Dounce homogenizer to ensure cell breakage. Nuclei were removed by centrifugation at 2,000 rpm in a Beckman JA-20 rotor for 10 min at 4°C, and the cytoplasm was then centrifuged at 10,000 rpm in the same rotor for 20 min. Sodium deoxycholate was added to the supernatant to a final concentration of 0.2%. The solution was layered onto a discontinuous gradient consisting of 2.5 ml of 0.05 M sucrose over 2.0 ml of 2 M sucrose, both in a buffer consisting of 0.05 M Tris-hydrochloride (pH 7.4), 0.025 M NaCl, and 0.005 MgCl2, and centrifuged at 38,000 rpm in a Beckman 50 Ti rotor for 2.5 h at 4°C. The polyribosome pellets were suspended in TNE buffer, consisting of 0.01 M Tris-hy-
85
drochloride (pH 7.4), 0.01 M NaCl, and 0.002 M EDTA (pH 7.4). The RNA was extracted and DNase treated by following the method described for preparation of whole cell RNA. Preparation of MDV DNA. MDV was grown in CEF cultures in minimal essential medium supplemented with 10% tryptose phosphate broth and 5% fetal calf serum. Virus was extracted from infected cells with 1.0% Nonidet P-40, and the extract was centrifuged through a 35% sorbitol cushion at 16,000 rpm in a Beckman JA-20 rotor for 1 h at 4°C. The virus pellet was purified further by centrifugation through a 15 to 30% sucrose gradient (wt/vol in TE [0.01 M Tris-hydrochloride, pH 7.4, and 0.001 M EDTA]) at 22,000 rpm in a Beckman SW27 rotor for 30 min at 4°C. The fractions containing the virus peak (located approximately midway down the gradient) were pooled and centrifuged at 22,000 rpm in a Beckman SW27 rotor for 90 min at 4°C. The resulting virus pellet was suspended in 1 to 2 ml of 0.05 M Trishydrochloride (pH 9.0), 0.4 M NaCl, and 0.01 M EDTA. After treatment of the virus with 100 pg of proteinase K per ml and 1.0% sodium dodecyl sulfate, 55S viral DNA was isolated by 10 to 30% glycerol gradient centrifugation at 38,000 rpm in a Beckman SW41 rotor for 4 h at 18°C. Preparation of radioactively labeled MDV DNA in vitro. Purified MDV DNA was labeled in vitro with [3H]TTP (specific activity, 57 Ci/mmol; New England Nuclear Corp., Boston, Mass.) as previously described (14, 17, 19). The specific activity of the labeled DNA was approximately 6 x 106 cpm/pg. DNA-RNA hybridization conditions. RNA samples ranging from 2.5 to 5.0 mg/ml were each mixed with 20,000 cpm of MDV [3H]DNA in 0.01 M Trishydrochloride (pH 7.4) and 0.0025 EDTA (pH 7.4). Each mixture was heat denatured, and NaCl was added to a final concentration of 0.5 M. The reaction was carried out at 66°C, and 50- to 100-,l samples were assayed for the amount of hybridized DNA by digestion of single-stranded DNA with S1 nuclease (Seikagaku Kogyu Co., Tokyo, Japan) (3). The extent of renaturation of labeled DNA reached 10 to 12% after 72 h of incubation. The values obtained from the CEF control hybridization were subtracted at each point from the values obtained for hybridization of MDV [3H]DNA with test RNA. Hybridization kinetics were plotted as percentage MDV [3H]DNA hybridized versus Rot = moles of nucleotide RNA x seconds x liter-' (8). DNA-DNA reassociation kinetics. MDV genomes in tumor tissues and MKT-1 cells were detected by DNA-DNA reassociation kinetics using methods previously described (14, 19). Briefly, unlabeled DNA was extracted from tumor tissues, MKT-1 cells, and CEF and sonically disrupted. Each DNA sample (250 ,ug) was mixed with the equivalent of one genome of MDV [3H]DNA per cell. The mixture was heat denatured and reassociated in 1.5 M NaCl at 66°C. Samples were taken at time intervals extending to 24 h, and the amount of reassociated DNA was measured by digestion of single-stranded DNA with S1 nuclease. The number of genomes was calculated from the observed rate of reassociation, which is expressed ac-
86
SILVER, SMITH, AND NONOYAMA
cording to the equation: (C0/S) 22 = 1 + K Cot, where Co is the initial concentration of MDV [3H]DNA, S is the amount of single-stranded DNA at time t, and K is the kinetic constant (5; J. F. Morrow, Ph.D. thesis, Stanford University, Stanford, Calif., 1974).
RESULTS Tumors tissues. Tumors were taken from LSI-SPF chickens inoculated with MDV (Table 1) at the dosage levels indicated in Materials and Methods. Some groups of chickens were concurrently inoculated with MDV + RAV-2 to test the effect of RAV-2 on the pathogenesis of MD. The number of chickens that developed tumors did not vary significantly from the 30% average incidence shown by groups inoculated with MDV alone; however, dual infection with MDV + RAV-2 enhanced mortality two- to threefold as reported previously (18). The presence of RAV-2 in the inoculum did not appear to affect the biochemical analyses of MDV infection in individual tumors and will not be considered significant to the results that follow. MDV genomes in tumor tissues and lymphoblastoid cells, MKT-1. All tumors contained multiple copies of MDV genome per cell (Table 1), which confirms data reported by Nazerian et al. (13). It should be mentioned that the tumor tissues contained a variety of cell types, therefore the value obtained for the average number of genomes per cell is not absolute TABLE 1. Detection ofMDVgenomes in tumor tissues from chickens inoculated with MDV or MDV + RAV-2 Avg. no. of geTumor tissue or Inoculum" cell line nomes per cell, 12 MDV Liver (no. 4)C MDV 6 Kidney (no. 5) MDV 5 Spleen (no. 6) MDV 3 Spleen (no. 7) 8 MDV + RAV-2 Ovary (no. 1) MDV + RAV-2 3 Kidney (no. 2) MDV + RAV-2 3 Spleen (no. 3) Normal spleen Negatived MKT-le 15 MSB-1f 43 a Chickens were inoculated with MDV (100 to 150 focus-forming units per bird) or MDV (same dosage) + RAV-2 (102 to 104 PFU per bird) as described in the text. b MDV genomes were detected by DNA-DNA reassociation kinetics as described in the text. 'Numbers correspond to tumor tissue numbers in Fig.d 3 (no. 1-6) and 4 (no. 7). The level of sensitivity was 0.1 genome per cell. 'Nonproductive cell line established from MD tumor in the kidney of a chicken (19). f Slightly productive cell line established from MD tumor in the spleen of a chicken (1).
J. VIROL.
but merely an indication of the presence of MDV DNA. In contrast, MKT-1 is an established nonproductive lymphoblastoid cell line carrying MDV genomes, the average number of which remains a constant 15 genomes per cell. MKT-1 cells contain complete virus genomes as evidenced by the recovery of infectious virus from chickens transplanted with these cells (unpublished data). Virus-specific RNA in whole-cell extracts. Total cellular RNA was extracted from tumor tissues, MKT-1 cells, and MDV-infected CEF. CEF infected with MDV were harvested when the culture showed 50% cytopathic effect. Hybridization of denatured MDV [3H]DNA with RNA extracted from these cells reached a plateau at 45% (Fig. 1), indicating that 45% of MDV DNA was transcribed, or 90% of the viral genome if only the coding strand of the doublestranded DNA is copied. RNA extracted from MKT-1 cells hybridized to 12 to 14% of MDV DNA (Fig. 2). The extent of transcription ofthe viral genome in all seven tumors was analyzed by DNA-RNA hybridization. All tumors contained viral RNA homologous to less than 15% of MDV DNA (Fig. 3 and 4), and in all but two tumors (no. 5 kidney and no. 6 spleen) RNA hybridized to MDV [3H]DNA to the same extent as MKT-1, i.e., 12 to 14%. It is possible that the RNA extracted from no. 5 kidney tumor and no. 6 spleen tumor did not contain a sufficient concentration of virus-specific RNA to drive the hybridization
N0 .0
z
a2 1000 2000 3000
Rot FIG. 1. Hybridization of MDV [3HJDNA with unlabeled RNA from CEF infected with MDV. Heatdenatured MDV [3HJDNA (20,000 cpm/ml) was mixed with an excess amount of unlabeled RNA extracted from CEF infected with MDV. Hybridization was carried out at 66"C, and samples were taken at intervals up to 72 h. The percentage of MDV [3HJDNA hybridized at each interval was determined by SI nuclease digestion as described in the text. Rot = moles of nucleotide RNA x seconds x liter-. (- *) Hybridization with whole-ceU RNA; (0-----0) hybridization with polyribosomal RNA.
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
reaction to a plateau at an Rot value of 3,000, but might have eventually hybridized to 12 to 14% of MDV DNA at a higher Rot value. It is concluded from these data that in productively infected CEF the MDV genome was
.20
I10.
.0
X40,
0
1000 2000 3000 Rot FIG. 2. Hybridization of MDV [3H]DNA with unlabeled MKT-1 RNA. Hybridization conditions were the same as described in Fig. I and the text. (@-*) Hybridization with whole-cell RNA; (0-----0) hybridization with polyribosomal RNA. I. Ovary
2. Kidney
20 10
~~~~~~4.
3.
Spleen
Uver
20
l.0 0
-. 2000 3000
000
3000
Rot FIG. 4. Hybridization of MDV [3H]DNA with unlabeled RNA from an MD spleen lymphoma. RNA was extracted from spleen tumor no. 7 (Table 1). Hybridization conditions were the same as described in Fig. 1 and the text. (-*) Hybridization with whole-cell RNA; (0-----0) hybridization with polyribosomal RNA.
20
1000
_
1000 2000
0-v.
87
2000
300O
Rot FIG. 3. Hybridization of MDV [3H]DNA with unlabeled RNA from lymphoid tumors of visceral organs from MDV-infected chickens. Hybridization was carried our using total cellular RNA extracted from homogenized tumor tissues under conditions described in Fig. 1 and the text.
almost completely transcribed, assuming transcription is asymmetric, whereas in tumor tissues from MDV-infected chickens and in MKT1 cells transcription was restricted to only a portion of the viral genome. Virus-specific RNA in polyribosomes. Transcription of the MDV genome in tumor tissues was analyzed further by detection of virus-specific RNA found in the polyribosomal fraction. Polyribosomes in MDV-infected CEF contained RNA sequences homologous to 45% of the viral DNA (Fig. 1), indicating that all viral RNA sequences in MDV-infected CEF were transferred to the polyribosomes. Polyribosomes were isolated from a portion of a lymphoid tumor of the spleen, induced by inoculation with MDV alone (Table 1), and RNA was extracted. Another portion of the tumor was used to extract total cellular RNA. As shown in Fig. 4, the spleen tumor contained RNA sequences homologous to 12% of the MDV DNA, showing the same degree of transcription of the viral genome as MKT-1 (Fig. 2) and other tumors tested (Fig. 3). The polyribosomes from this spleen tumor contained RNA sequences homologous to only 5 to 6% of the viral DNA (Fig. 4). Therefore, the viral RNA transcripts in the polyribosomal fraction represent only a portion of the RNA sequences found in whole cells. RNA extracted from the polyribosomes of MKT-1 cells contained sequences homologous to 8 to 10% of the viral DNA (Fig. 2). Since the viral RNA sequences found in whole-cell extracts of MKT-1 cells were encoded from 12 to 14% of the MDV DNA, then, like the spleen tumor, the viral RNA in the polyribosomal fraction represents only a portion of these sequences. Thus, unlike productively infected CEF, where all the virus-specific RNA sequences found in whole cells also were found in the
88
J. VIROL.
SILVER, SMITH, AND NONOYAMA
polyribosomes, certain viral RNA sequences were excluded from the polyribosomes in MKT1 and tumor tissue. Comparison of virus-specific RNA sequences in MKT-1 and spleen lymphoma. To determine whether the virus-specific RNAs found in tumor tissue and MKT-1 cells were encoded by the same DNA sequences, total cellular RNA extracted from the above spleen tumor (Table 1, Fig. 4) was mixed with RNA extracted from MKT-1 cells and hybridized to MDV [3H]DNA. If the viral RNA transcripts from these two sources were synthesized from different DNA sequences, the hybridization value of the mixture would be greater than 12 to 14%, i.e., greater than, or possibly the sum of, the values observed for the spleen tumor and MKT-1. If the viral RNA sequences from the two sources were identical, hybridization should not exceed 14%. As shown in Fig. 5, hybridization reached 12%, which indicates that the viral RNA transcripts in the spleen tumor were encoded by the same DNA sequences as the viral RNA transcripts in MKT-1. DISCUSSION
Several features concerning transcription of the MDV genome in tumor tissues were revealed through analysis of DNA-RNA hybridization using MDV [3H]DNA as a probe and RNA extracted from tumor tissues and MKT-1 cells. In productively infected CEF, 45% of the MDV DNA template was transcribed or 90% of the viral genome, assuming only the coding strand of the double-stranded DNA was transcribed. In
20 I0
(7 1000
2000 Rot
3000
FIG. 5. Hybridization of MDV [3H]DNA with a mixture of RNA extracted from an MD spleen lymphoma and RNA extracted from MKT-1 cells. A 0.5ml hybridization mixture consisting of 10,000 cpm of heat-denatured MDV [3H]DNA, 1.25 mg of wholecell RNA extracted from spleen tumor no. 7 (Table 1, Fig. 4), and 1.25 mg of RNA extracted from MKT-1 was incubated at 66°C in 0.5 M NaCI. Samples were taken at intervals up to 72 h and assayed by digestion with single-strand-specific SI nuclease to determine the percentage of MDV [3H]DNA hybridized as described in the text.
virus nonproductive MKT-1 cells and in all seven tumors tested, transcription was restricted to a much smaller portion of the viral DNA, i.e., 12 to 14%. (Tumor tissues from chickens inoculated with MDV + RAV-2 appeared to be biochemically the same with regard to transcription of the MDV genome as tumor tissues from chickens inoculated with MDV alone.) Thus, the nonproductive state MD tumors in vivo may be the result of a control mechanism within the cell which restricts the synthesis of virus-specific RNA so that only a small portion of the MDV DNA template is transcribed. It is likely that the virus-specific RNAs transcribed in tumor tissues and MKT-1 are encoded by the same DNA sequences. This became evident, at least in one spleen tumor, when RNA extracted from the spleen tumor was mixed with RNA extracted from MKT-1 cells and hybridized to MDV [3H]DNA. The hybridization value did not exceed 12 to 14%, i.e., the observed value for each of the RNAs, indicating that viral RNA sequences in MKT-1 were the same as the viral RNA sequences in the spleen tumor. This indicates that a similar cellular (or viral) regulation, which restricts transcription of the MDV genome to a specific portion of the viral DNA template, may control the expression of resident viral genomes in both MKT-1 cells and the spleen tumor tissue. All virus-specific RNA sequences transcribed from 45% of the MDV DNA were transferred to the polyribosomes of CEF productively infected with MDV. In contrast, the virus-specific RNA sequences found in the polyribosomes of MKT1 cells were encoded from 8 to 10% of the viral DNA and in the spleen tumor only 5%. This represents only a portion of the virus-specific RNA sequences found in whole cells, suggesting the existence of a posttranscriptional control mechanism within the cell which selectively excludes specific RNA transcripts from stable association with the polyribosomes. The mechanism by which RNA sequences are selected for transfer to polyribosomes for translation may involve polyadenylation of the RNA. Hayward and Kieff (9) have shown in Epstein-Barr virusinfected nonproductive Raji cells that 25% of the viral DNA is transcribed, but only a portion of the viral RNA sequences were found in the polyribosomes. It was found also that the polyribosomes contained only polyadenylated viral RNA sequences (15). This suggests that polyadenylation may determine the fate of viral RNA in nonproductive infections. The data reported here suggest that transcription of the MDV genome in tumor tissues of MDV-infected chickens is restricted to the same sequences as in the established
specific DNA
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
nonproductive lymphoblastoid cells MKT-1, representing 12 to 14% of the viral DNA, and that certain virus-specific RNA sequences are prevented from attachment to the polyribosomes. Thus, the repressed expression of the viral genome in tumor tissues as well as MKT-1 cells may be the result of precise control mechanisms within the cell at the transcriptional and posttranscriptional levels. MKT-1 cells share biochemical similarities with MD lymphoid tumors and, therefore, serve as a model for studying virus-cell interaction involved in MDV transformation. ACKNOWLEDGMENT This work was supported by a contract (NO-I-CP33205) within the Virus Cancer Program of the National Cancer Institute. LrERATURE CITED 1. Akiyama, Y., and S. Kato. 1974. Two cell lines from lymphomas of Marek's disease. Biken J. 17:105-116. 2. Akiyama, Y., S. Kato, and N. Iwa. 1973. Continuous cell culture from lymphoma of Marek's disease. Biken J. 16:177-179. 3. Ando, T. 1966. A nuclease specific for heat denatured DNA isolated from a product of Aspergillus oryzae. Biochem. Biophys. Acta 114:158-168. 4. Biggs, P. ML 1973. Marek's disease, p. 557-594. In A. S. Kaplan (ed.), The herpeviuses. Academic Press Inc., New York. 5. Britten, R. J., and E. H. Davidson. 1976. Studies of nucleic acid reassociation kinetics: emperical equations describing DNA reassociation. Proc. Natl. Acad. Sci. U.S.A. 73:415-419. 6. Calnek, B. W., Hans K. Adlinger, and D. E. Kahn. 1970. Feather follicle epithelium: a source of enveloped and infectious cell-free herpesvirus from Marek's disease. Avian Dis. 14:219-233. 7. Churchill, A. E., R. C. Chubb, and W. Boxendale. 1969. The attenuation, with loss of oncogenicity, of the herpes-type virus of Marek's disease (strain HPRS-16) on passage in cell culture. J. Gen. Virol. 4:557-564.
89
8. Frenkel, M., and R. Roizman. 1972. Ribonucleic acid synthesis in cells infected with herpes simplex virus: controls of transcription and of RNA abundance. Proc. Natl. Acad. Sci. U.S.A. 69:2654-2658. 9. Hayward, S. D., and E. D. Kieff. 1976. Epstein-Barr virus-specific RNA. I. Analysis of viral RNA in cellular extracts and in the polyribosomal fraction of permissive and nonpermissive lymphoblastoid cell lines. J. Virol. 18:518-525. 10. Hayward, W. 1977. Size and genetic content of viral RNAs in avian oncovirus-infected cells. J. Virol. 24:4763. 11. Nazerian, K. 1973. Marek's disease: a neoplastic disease caused by a herpesvirus. Adv. Cancer Res. 17:279-317. 12. Nazerian, K., and B. R. Burmester. 1968. Electron microscopy of a herpesvirus associated with the agent of Marek's disease in cell culture. Cancer Res. 28:24542462. 13. Nazerian, K., T. Lindahl, G. Klein, and L F. Lee. 1973. Deoxyribonucleic acid of Marek's disease virus in virusinduced tumors. J. Virol. 12:841-846. 14. Nonoyama, M*, and J. S. Pagano. 1973. Homology between Epstein-Barr virus DNA and viral DNA from Burkitt's lymphoma and nasopharyngeal carcinoma determined by DNA-DNA reasciation kinetics. Nature (London) 242:44-47. 15. Orellana, T., and E. Kieff. 1977. Epstein-Barr virusspecific RNA. II. Analysis of polyadenylated viral RNA in restringent, abortive, and productive infections. J. Virol. 22:321-330. 16. Powell, P. C., L N. Payne, J. A. Frazier, and M. Rennie. 1974. T lymphoblastoid cell lines from Marek's disease lymphomas. Nature (London) 251:79-80. 17. Rigby, P. W. J., M. Dieckman, C. Rhodes, and P. Berg. 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237-251. 18. Smith, M. E., W. F. Campbell, W. M. Farrow, and J. W. Frankel. 1975. Enhancement and interference in chickens inoculated with Marek's disease herpesvirus and oncornaviruses. Proc. Soc. Exp. Biol. Med. 150: 574-577. 19. Tanaka, A., S. Silver, and M. Nonoyama. 1978. Biochemical evidence of nonintegrated status of Marek's disease virus DNA in virus transformed lymphoblastoid cells of chickens. Virology 88:19-24.
Vol. 30, No. 1
Transcription of the Marek's Disease Virus Genome in VirusInduced Tumors SANDRA SILVER, MARY SMITH, AND MEIHAN NONOYAMA* Laboratory of Molecular Virology, Life Sciences, Inc., St. Petersburg, Florida 33710 Received for publication 14 August 1978
Transcription of the Marek's disease virus (MDV) genome in tumor tissues from MDV-infected chickens has been studied by analyzing the hybridization kinetics of 3H-labeled MDV DNA with unlabeled RNA extracted from these tissues. Lymphoid tumors of ovary, spleen, liver, and kidney contained MDV genomes, but the virus-specific RNA sequences were transcribed from less than 15% of the viral DNA. A virus nonproductive lymphoblastoid cell line, designated MKT-1, has been established from a kidney lymphoma and contains 15 MDV genomes per cell. In these cells, 12 to 14% of the viral DNA was transcribed. Thus transcription of the MDV genome was restricted both in tumor tissues and MKT1 cells. A hybridization experiment where RNA extracted from MKT-1 cells and RNA extracted from a spleen tumor were mixed and hybridized to 'H-labeled MDV DNA indicated that the virus-specific RNAs from the two sources were encoded by the same DNA sequences. The polyribosomal fractions of MKT-1 cells and this spleen tumor contained only a portion of the virus-specific RNA sequences found in whole-cell extracts, indicating the existence of a posttranscriptional control mechanism which prevents the transfer of certain viral RNA transcripts to the polyribosomes. The data suggest that the repressed expression of the viral genome in lymphoid tumor tissues and MKT-1 cells may be the result of precise controls within the cell at the transcriptional and posttranscriptional levels.
Using DNA-RNA hybridization techniques (8), the virus-specific RNA in MKT-1 and several tumor tissues were analyzed to compare the extent of transcription of the viral genome and to determine whether a specific portion of the MDV DNA is transcribed. Hybridization experiments using RNA from whole-cell extracts and RNA from polyribosomes can determine the proportion of viral RNA sequences selectively transferred to the polyribosomes. This may offer some insight into the control mechanism at the posttranscriptional level in virus-induced tumors and transformed cells.
Marek's disease (MD) virus (MDV) is a herpesvirus which causes a lymphoproliferative disease in chickens (4, 11). The virus replicates in the feather follicle epithelium and produces infectious cell-free virus (6), whereas in MDVinfected cell cultures the virus is cell associated, usually causing a cytopathic effect (7, 12). Lymphoid tumors from MDV-infected chickens contain copies of the MDV genome but are usually free of MD antigens and virus particles (13); however, certain established cell lines derived from MD lymphomas produce virus particles (or viral antigens) in a small percentage of the cell population (1, 2, 16). In contrast, a lymphoblastoid cell line, designated MKT-1, has been established from an MD kidney lymphoma, is nonproductive of virus and viral antigens, but contains 15 complete MDV genomes per cell
MATERIALS AND METHODS Tumor tissues. Tumors were collected from 5- to 8-week-old, isolator-held, LSI-SPF chickens that were inoculated intraabdominally with virus at 2 days of age. Four of the tumors examined were from chickens inoculated with 100 to 150 focus-forming units of GA strain cell-free MDV, extracted from feather follicle epithelium, per bird. The other three were from chickens concurrently infected with MDV (same dosage as above) and Rous-associated virus-2 (RAV-2) at a dosage of 102 to 104 PFU per bird. Cells. MKT-1 cells were maintained at 41°C in RPMI 1640 medium supplemented with 10% fetal calf
(19). Since MDV induces tumors in its natural host and the MKT-1 cell line was derived from such a tumor, experiments to analyze regulations at the transcriptional and posttranscriptional levels were undertaken to determine, in part, if the mechanisms involved in the repressed expression of the MDV genome are similar in lymphoid tumors and in MKT-1 cells. 84
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
serum and antibiotics. Maximum cell density reached approximately 107 cells per ml. Chicken embryo fibroblast (CEF) cultures infected with MDV were grown in minimal essential medium supplemented with 10% tryptose phosphate broth and 5% fetal calf serum. Cells were harvested when 50% of the culture showed cytopathic effects. Preparation of RNA extracted from whole cells. The RNA extraction procedure used was that described by Hayward (10) with modifications. Each cell pellet or homogenized tumor tissue was suspended in 9 volumes of RNA extraction buffer consisting of 0.01 M Tris-hydrochloride (pH 8.1), 0.02 M NaCl, and 0.002 M CaCl2. Proteinase K (Beckman, Palo Alto, Calif.) and sodium dodecyl sulfate were added to the suspensions at final concentrations of 1.0 mg/ml and 1.0%, respectively, and the lysates were incubated at 37°C for 45 to 60 min. Each lysate was then passed through a 20-gauge needle three to four times to reduce viscosity caused by cellular DNA. NaCl was added to a final concentration of 0.15 M, and EDTA (pH 7.4) was added to 0.002 M. The nucleic acids were extracted three times at room temperature with a mixture of phenol-chloroform-isoamyl alcohol (1:1: 0.01 vol/vol/vol) containing 0.05% 8-hydroxyquinoline. The aqueous phase was precipitated overnight at -20°C after the addition of 2 volumes of ethanol. The nucleic acids were then pelleted, suspended in 5 to 10 ml of TNM buffer (consisting of 0.01 M Tris-hydrochloride, pH 7.4, 0.15 M NaCl, and 0.001 M MgCl2), and digested with 20 jig of DNase 1 (Worthington Biochemical Corp., Freehold, N.J.) per ml for 20 min at 37°C. The protein was removed by extraction with phenol-chloroform-isoamyl alcohol-8-hydroxyquinoline, and the resulting nucleic acids were alcohol precipitated, as described previously. After two additional cycles of DNase treatment, phenol extraction, and alcohol precipitation, the resulting RNA was suspended in less than 1.0 ml of 0.0025 M EDTA (pH 7.4). Approximately 1 mg of RNA per 2.5 x 108 cells was obtained. The purity of RNA was checked by diphenylamine assay, and the preparation showed no detectable DNA. Preparation of polyribosomal RNA. The method used for isolating polyribosomes of MKT-1 cells and tumor tissues was that described by Hayward and Kieff (9). Briefly, cell pellets or homogenized tumor tissue was suspended in a buffer consisting of 0.01 M Tris-hydrochloride (pH 7.4), 0.01 M NaCl, and 0.0015 M MgCl2, treated with 0.5% Nonidet P-40 at 4°C for 15 min, and homogenized with a Dounce homogenizer to ensure cell breakage. Nuclei were removed by centrifugation at 2,000 rpm in a Beckman JA-20 rotor for 10 min at 4°C, and the cytoplasm was then centrifuged at 10,000 rpm in the same rotor for 20 min. Sodium deoxycholate was added to the supernatant to a final concentration of 0.2%. The solution was layered onto a discontinuous gradient consisting of 2.5 ml of 0.05 M sucrose over 2.0 ml of 2 M sucrose, both in a buffer consisting of 0.05 M Tris-hydrochloride (pH 7.4), 0.025 M NaCl, and 0.005 MgCl2, and centrifuged at 38,000 rpm in a Beckman 50 Ti rotor for 2.5 h at 4°C. The polyribosome pellets were suspended in TNE buffer, consisting of 0.01 M Tris-hy-
85
drochloride (pH 7.4), 0.01 M NaCl, and 0.002 M EDTA (pH 7.4). The RNA was extracted and DNase treated by following the method described for preparation of whole cell RNA. Preparation of MDV DNA. MDV was grown in CEF cultures in minimal essential medium supplemented with 10% tryptose phosphate broth and 5% fetal calf serum. Virus was extracted from infected cells with 1.0% Nonidet P-40, and the extract was centrifuged through a 35% sorbitol cushion at 16,000 rpm in a Beckman JA-20 rotor for 1 h at 4°C. The virus pellet was purified further by centrifugation through a 15 to 30% sucrose gradient (wt/vol in TE [0.01 M Tris-hydrochloride, pH 7.4, and 0.001 M EDTA]) at 22,000 rpm in a Beckman SW27 rotor for 30 min at 4°C. The fractions containing the virus peak (located approximately midway down the gradient) were pooled and centrifuged at 22,000 rpm in a Beckman SW27 rotor for 90 min at 4°C. The resulting virus pellet was suspended in 1 to 2 ml of 0.05 M Trishydrochloride (pH 9.0), 0.4 M NaCl, and 0.01 M EDTA. After treatment of the virus with 100 pg of proteinase K per ml and 1.0% sodium dodecyl sulfate, 55S viral DNA was isolated by 10 to 30% glycerol gradient centrifugation at 38,000 rpm in a Beckman SW41 rotor for 4 h at 18°C. Preparation of radioactively labeled MDV DNA in vitro. Purified MDV DNA was labeled in vitro with [3H]TTP (specific activity, 57 Ci/mmol; New England Nuclear Corp., Boston, Mass.) as previously described (14, 17, 19). The specific activity of the labeled DNA was approximately 6 x 106 cpm/pg. DNA-RNA hybridization conditions. RNA samples ranging from 2.5 to 5.0 mg/ml were each mixed with 20,000 cpm of MDV [3H]DNA in 0.01 M Trishydrochloride (pH 7.4) and 0.0025 EDTA (pH 7.4). Each mixture was heat denatured, and NaCl was added to a final concentration of 0.5 M. The reaction was carried out at 66°C, and 50- to 100-,l samples were assayed for the amount of hybridized DNA by digestion of single-stranded DNA with S1 nuclease (Seikagaku Kogyu Co., Tokyo, Japan) (3). The extent of renaturation of labeled DNA reached 10 to 12% after 72 h of incubation. The values obtained from the CEF control hybridization were subtracted at each point from the values obtained for hybridization of MDV [3H]DNA with test RNA. Hybridization kinetics were plotted as percentage MDV [3H]DNA hybridized versus Rot = moles of nucleotide RNA x seconds x liter-' (8). DNA-DNA reassociation kinetics. MDV genomes in tumor tissues and MKT-1 cells were detected by DNA-DNA reassociation kinetics using methods previously described (14, 19). Briefly, unlabeled DNA was extracted from tumor tissues, MKT-1 cells, and CEF and sonically disrupted. Each DNA sample (250 ,ug) was mixed with the equivalent of one genome of MDV [3H]DNA per cell. The mixture was heat denatured and reassociated in 1.5 M NaCl at 66°C. Samples were taken at time intervals extending to 24 h, and the amount of reassociated DNA was measured by digestion of single-stranded DNA with S1 nuclease. The number of genomes was calculated from the observed rate of reassociation, which is expressed ac-
86
SILVER, SMITH, AND NONOYAMA
cording to the equation: (C0/S) 22 = 1 + K Cot, where Co is the initial concentration of MDV [3H]DNA, S is the amount of single-stranded DNA at time t, and K is the kinetic constant (5; J. F. Morrow, Ph.D. thesis, Stanford University, Stanford, Calif., 1974).
RESULTS Tumors tissues. Tumors were taken from LSI-SPF chickens inoculated with MDV (Table 1) at the dosage levels indicated in Materials and Methods. Some groups of chickens were concurrently inoculated with MDV + RAV-2 to test the effect of RAV-2 on the pathogenesis of MD. The number of chickens that developed tumors did not vary significantly from the 30% average incidence shown by groups inoculated with MDV alone; however, dual infection with MDV + RAV-2 enhanced mortality two- to threefold as reported previously (18). The presence of RAV-2 in the inoculum did not appear to affect the biochemical analyses of MDV infection in individual tumors and will not be considered significant to the results that follow. MDV genomes in tumor tissues and lymphoblastoid cells, MKT-1. All tumors contained multiple copies of MDV genome per cell (Table 1), which confirms data reported by Nazerian et al. (13). It should be mentioned that the tumor tissues contained a variety of cell types, therefore the value obtained for the average number of genomes per cell is not absolute TABLE 1. Detection ofMDVgenomes in tumor tissues from chickens inoculated with MDV or MDV + RAV-2 Avg. no. of geTumor tissue or Inoculum" cell line nomes per cell, 12 MDV Liver (no. 4)C MDV 6 Kidney (no. 5) MDV 5 Spleen (no. 6) MDV 3 Spleen (no. 7) 8 MDV + RAV-2 Ovary (no. 1) MDV + RAV-2 3 Kidney (no. 2) MDV + RAV-2 3 Spleen (no. 3) Normal spleen Negatived MKT-le 15 MSB-1f 43 a Chickens were inoculated with MDV (100 to 150 focus-forming units per bird) or MDV (same dosage) + RAV-2 (102 to 104 PFU per bird) as described in the text. b MDV genomes were detected by DNA-DNA reassociation kinetics as described in the text. 'Numbers correspond to tumor tissue numbers in Fig.d 3 (no. 1-6) and 4 (no. 7). The level of sensitivity was 0.1 genome per cell. 'Nonproductive cell line established from MD tumor in the kidney of a chicken (19). f Slightly productive cell line established from MD tumor in the spleen of a chicken (1).
J. VIROL.
but merely an indication of the presence of MDV DNA. In contrast, MKT-1 is an established nonproductive lymphoblastoid cell line carrying MDV genomes, the average number of which remains a constant 15 genomes per cell. MKT-1 cells contain complete virus genomes as evidenced by the recovery of infectious virus from chickens transplanted with these cells (unpublished data). Virus-specific RNA in whole-cell extracts. Total cellular RNA was extracted from tumor tissues, MKT-1 cells, and MDV-infected CEF. CEF infected with MDV were harvested when the culture showed 50% cytopathic effect. Hybridization of denatured MDV [3H]DNA with RNA extracted from these cells reached a plateau at 45% (Fig. 1), indicating that 45% of MDV DNA was transcribed, or 90% of the viral genome if only the coding strand of the doublestranded DNA is copied. RNA extracted from MKT-1 cells hybridized to 12 to 14% of MDV DNA (Fig. 2). The extent of transcription ofthe viral genome in all seven tumors was analyzed by DNA-RNA hybridization. All tumors contained viral RNA homologous to less than 15% of MDV DNA (Fig. 3 and 4), and in all but two tumors (no. 5 kidney and no. 6 spleen) RNA hybridized to MDV [3H]DNA to the same extent as MKT-1, i.e., 12 to 14%. It is possible that the RNA extracted from no. 5 kidney tumor and no. 6 spleen tumor did not contain a sufficient concentration of virus-specific RNA to drive the hybridization
N0 .0
z
a2 1000 2000 3000
Rot FIG. 1. Hybridization of MDV [3HJDNA with unlabeled RNA from CEF infected with MDV. Heatdenatured MDV [3HJDNA (20,000 cpm/ml) was mixed with an excess amount of unlabeled RNA extracted from CEF infected with MDV. Hybridization was carried out at 66"C, and samples were taken at intervals up to 72 h. The percentage of MDV [3HJDNA hybridized at each interval was determined by SI nuclease digestion as described in the text. Rot = moles of nucleotide RNA x seconds x liter-. (- *) Hybridization with whole-ceU RNA; (0-----0) hybridization with polyribosomal RNA.
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
reaction to a plateau at an Rot value of 3,000, but might have eventually hybridized to 12 to 14% of MDV DNA at a higher Rot value. It is concluded from these data that in productively infected CEF the MDV genome was
.20
I10.
.0
X40,
0
1000 2000 3000 Rot FIG. 2. Hybridization of MDV [3H]DNA with unlabeled MKT-1 RNA. Hybridization conditions were the same as described in Fig. I and the text. (@-*) Hybridization with whole-cell RNA; (0-----0) hybridization with polyribosomal RNA. I. Ovary
2. Kidney
20 10
~~~~~~4.
3.
Spleen
Uver
20
l.0 0
-. 2000 3000
000
3000
Rot FIG. 4. Hybridization of MDV [3H]DNA with unlabeled RNA from an MD spleen lymphoma. RNA was extracted from spleen tumor no. 7 (Table 1). Hybridization conditions were the same as described in Fig. 1 and the text. (-*) Hybridization with whole-cell RNA; (0-----0) hybridization with polyribosomal RNA.
20
1000
_
1000 2000
0-v.
87
2000
300O
Rot FIG. 3. Hybridization of MDV [3H]DNA with unlabeled RNA from lymphoid tumors of visceral organs from MDV-infected chickens. Hybridization was carried our using total cellular RNA extracted from homogenized tumor tissues under conditions described in Fig. 1 and the text.
almost completely transcribed, assuming transcription is asymmetric, whereas in tumor tissues from MDV-infected chickens and in MKT1 cells transcription was restricted to only a portion of the viral genome. Virus-specific RNA in polyribosomes. Transcription of the MDV genome in tumor tissues was analyzed further by detection of virus-specific RNA found in the polyribosomal fraction. Polyribosomes in MDV-infected CEF contained RNA sequences homologous to 45% of the viral DNA (Fig. 1), indicating that all viral RNA sequences in MDV-infected CEF were transferred to the polyribosomes. Polyribosomes were isolated from a portion of a lymphoid tumor of the spleen, induced by inoculation with MDV alone (Table 1), and RNA was extracted. Another portion of the tumor was used to extract total cellular RNA. As shown in Fig. 4, the spleen tumor contained RNA sequences homologous to 12% of the MDV DNA, showing the same degree of transcription of the viral genome as MKT-1 (Fig. 2) and other tumors tested (Fig. 3). The polyribosomes from this spleen tumor contained RNA sequences homologous to only 5 to 6% of the viral DNA (Fig. 4). Therefore, the viral RNA transcripts in the polyribosomal fraction represent only a portion of the RNA sequences found in whole cells. RNA extracted from the polyribosomes of MKT-1 cells contained sequences homologous to 8 to 10% of the viral DNA (Fig. 2). Since the viral RNA sequences found in whole-cell extracts of MKT-1 cells were encoded from 12 to 14% of the MDV DNA, then, like the spleen tumor, the viral RNA in the polyribosomal fraction represents only a portion of these sequences. Thus, unlike productively infected CEF, where all the virus-specific RNA sequences found in whole cells also were found in the
88
J. VIROL.
SILVER, SMITH, AND NONOYAMA
polyribosomes, certain viral RNA sequences were excluded from the polyribosomes in MKT1 and tumor tissue. Comparison of virus-specific RNA sequences in MKT-1 and spleen lymphoma. To determine whether the virus-specific RNAs found in tumor tissue and MKT-1 cells were encoded by the same DNA sequences, total cellular RNA extracted from the above spleen tumor (Table 1, Fig. 4) was mixed with RNA extracted from MKT-1 cells and hybridized to MDV [3H]DNA. If the viral RNA transcripts from these two sources were synthesized from different DNA sequences, the hybridization value of the mixture would be greater than 12 to 14%, i.e., greater than, or possibly the sum of, the values observed for the spleen tumor and MKT-1. If the viral RNA sequences from the two sources were identical, hybridization should not exceed 14%. As shown in Fig. 5, hybridization reached 12%, which indicates that the viral RNA transcripts in the spleen tumor were encoded by the same DNA sequences as the viral RNA transcripts in MKT-1. DISCUSSION
Several features concerning transcription of the MDV genome in tumor tissues were revealed through analysis of DNA-RNA hybridization using MDV [3H]DNA as a probe and RNA extracted from tumor tissues and MKT-1 cells. In productively infected CEF, 45% of the MDV DNA template was transcribed or 90% of the viral genome, assuming only the coding strand of the double-stranded DNA was transcribed. In
20 I0
(7 1000
2000 Rot
3000
FIG. 5. Hybridization of MDV [3H]DNA with a mixture of RNA extracted from an MD spleen lymphoma and RNA extracted from MKT-1 cells. A 0.5ml hybridization mixture consisting of 10,000 cpm of heat-denatured MDV [3H]DNA, 1.25 mg of wholecell RNA extracted from spleen tumor no. 7 (Table 1, Fig. 4), and 1.25 mg of RNA extracted from MKT-1 was incubated at 66°C in 0.5 M NaCI. Samples were taken at intervals up to 72 h and assayed by digestion with single-strand-specific SI nuclease to determine the percentage of MDV [3H]DNA hybridized as described in the text.
virus nonproductive MKT-1 cells and in all seven tumors tested, transcription was restricted to a much smaller portion of the viral DNA, i.e., 12 to 14%. (Tumor tissues from chickens inoculated with MDV + RAV-2 appeared to be biochemically the same with regard to transcription of the MDV genome as tumor tissues from chickens inoculated with MDV alone.) Thus, the nonproductive state MD tumors in vivo may be the result of a control mechanism within the cell which restricts the synthesis of virus-specific RNA so that only a small portion of the MDV DNA template is transcribed. It is likely that the virus-specific RNAs transcribed in tumor tissues and MKT-1 are encoded by the same DNA sequences. This became evident, at least in one spleen tumor, when RNA extracted from the spleen tumor was mixed with RNA extracted from MKT-1 cells and hybridized to MDV [3H]DNA. The hybridization value did not exceed 12 to 14%, i.e., the observed value for each of the RNAs, indicating that viral RNA sequences in MKT-1 were the same as the viral RNA sequences in the spleen tumor. This indicates that a similar cellular (or viral) regulation, which restricts transcription of the MDV genome to a specific portion of the viral DNA template, may control the expression of resident viral genomes in both MKT-1 cells and the spleen tumor tissue. All virus-specific RNA sequences transcribed from 45% of the MDV DNA were transferred to the polyribosomes of CEF productively infected with MDV. In contrast, the virus-specific RNA sequences found in the polyribosomes of MKT1 cells were encoded from 8 to 10% of the viral DNA and in the spleen tumor only 5%. This represents only a portion of the virus-specific RNA sequences found in whole cells, suggesting the existence of a posttranscriptional control mechanism within the cell which selectively excludes specific RNA transcripts from stable association with the polyribosomes. The mechanism by which RNA sequences are selected for transfer to polyribosomes for translation may involve polyadenylation of the RNA. Hayward and Kieff (9) have shown in Epstein-Barr virusinfected nonproductive Raji cells that 25% of the viral DNA is transcribed, but only a portion of the viral RNA sequences were found in the polyribosomes. It was found also that the polyribosomes contained only polyadenylated viral RNA sequences (15). This suggests that polyadenylation may determine the fate of viral RNA in nonproductive infections. The data reported here suggest that transcription of the MDV genome in tumor tissues of MDV-infected chickens is restricted to the same sequences as in the established
specific DNA
VOL. 30, 1979
TRANSCRIPTION OF MDV GENOME IN TUMORS
nonproductive lymphoblastoid cells MKT-1, representing 12 to 14% of the viral DNA, and that certain virus-specific RNA sequences are prevented from attachment to the polyribosomes. Thus, the repressed expression of the viral genome in tumor tissues as well as MKT-1 cells may be the result of precise control mechanisms within the cell at the transcriptional and posttranscriptional levels. MKT-1 cells share biochemical similarities with MD lymphoid tumors and, therefore, serve as a model for studying virus-cell interaction involved in MDV transformation. ACKNOWLEDGMENT This work was supported by a contract (NO-I-CP33205) within the Virus Cancer Program of the National Cancer Institute. LrERATURE CITED 1. Akiyama, Y., and S. Kato. 1974. Two cell lines from lymphomas of Marek's disease. Biken J. 17:105-116. 2. Akiyama, Y., S. Kato, and N. Iwa. 1973. Continuous cell culture from lymphoma of Marek's disease. Biken J. 16:177-179. 3. Ando, T. 1966. A nuclease specific for heat denatured DNA isolated from a product of Aspergillus oryzae. Biochem. Biophys. Acta 114:158-168. 4. Biggs, P. ML 1973. Marek's disease, p. 557-594. In A. S. Kaplan (ed.), The herpeviuses. Academic Press Inc., New York. 5. Britten, R. J., and E. H. Davidson. 1976. Studies of nucleic acid reassociation kinetics: emperical equations describing DNA reassociation. Proc. Natl. Acad. Sci. U.S.A. 73:415-419. 6. Calnek, B. W., Hans K. Adlinger, and D. E. Kahn. 1970. Feather follicle epithelium: a source of enveloped and infectious cell-free herpesvirus from Marek's disease. Avian Dis. 14:219-233. 7. Churchill, A. E., R. C. Chubb, and W. Boxendale. 1969. The attenuation, with loss of oncogenicity, of the herpes-type virus of Marek's disease (strain HPRS-16) on passage in cell culture. J. Gen. Virol. 4:557-564.
89
8. Frenkel, M., and R. Roizman. 1972. Ribonucleic acid synthesis in cells infected with herpes simplex virus: controls of transcription and of RNA abundance. Proc. Natl. Acad. Sci. U.S.A. 69:2654-2658. 9. Hayward, S. D., and E. D. Kieff. 1976. Epstein-Barr virus-specific RNA. I. Analysis of viral RNA in cellular extracts and in the polyribosomal fraction of permissive and nonpermissive lymphoblastoid cell lines. J. Virol. 18:518-525. 10. Hayward, W. 1977. Size and genetic content of viral RNAs in avian oncovirus-infected cells. J. Virol. 24:4763. 11. Nazerian, K. 1973. Marek's disease: a neoplastic disease caused by a herpesvirus. Adv. Cancer Res. 17:279-317. 12. Nazerian, K., and B. R. Burmester. 1968. Electron microscopy of a herpesvirus associated with the agent of Marek's disease in cell culture. Cancer Res. 28:24542462. 13. Nazerian, K., T. Lindahl, G. Klein, and L F. Lee. 1973. Deoxyribonucleic acid of Marek's disease virus in virusinduced tumors. J. Virol. 12:841-846. 14. Nonoyama, M*, and J. S. Pagano. 1973. Homology between Epstein-Barr virus DNA and viral DNA from Burkitt's lymphoma and nasopharyngeal carcinoma determined by DNA-DNA reasciation kinetics. Nature (London) 242:44-47. 15. Orellana, T., and E. Kieff. 1977. Epstein-Barr virusspecific RNA. II. Analysis of polyadenylated viral RNA in restringent, abortive, and productive infections. J. Virol. 22:321-330. 16. Powell, P. C., L N. Payne, J. A. Frazier, and M. Rennie. 1974. T lymphoblastoid cell lines from Marek's disease lymphomas. Nature (London) 251:79-80. 17. Rigby, P. W. J., M. Dieckman, C. Rhodes, and P. Berg. 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237-251. 18. Smith, M. E., W. F. Campbell, W. M. Farrow, and J. W. Frankel. 1975. Enhancement and interference in chickens inoculated with Marek's disease herpesvirus and oncornaviruses. Proc. Soc. Exp. Biol. Med. 150: 574-577. 19. Tanaka, A., S. Silver, and M. Nonoyama. 1978. Biochemical evidence of nonintegrated status of Marek's disease virus DNA in virus transformed lymphoblastoid cells of chickens. Virology 88:19-24.
