(1988), 69, 355-374. Printed in Great Britain Differences in Cell Type

J. gen. Virol.
(1988),69, 355-374.
Printed in
Great Britain
355
Key words: CMV/gene expression/DNA replication
Differences in Cell Type-specific Blocks to Immediate Early Gene
Expression and DNA Replication of Human, Simian and Murine
Cytomegalovirus
By R O B E R T L. L A F E M I N A * ~ AND G A R Y S. H A Y W A R D
The Virology Laboratories, Department of Pharmacology and Molecular Sciences, The Johns
Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, Maryland 21205,
U.S.A.
(Accepted 6 November 1987)
SUMMARY
We have previously described blocks to the viral lytic cycle at two different levels in
cytomegalovirus (CMV)-infected non-permissive cells. BALB/c-3T3 cells express only
the predominant immediate early (IE) nuclear phosphoproteins (IE68 or IE94) of
human CMV (HCMV) or simian CMV (SCMV) and do not replicate the input viral
genomes. However, in human teratocarcinoma stem cells and 293 cells, expression of
the HCMV IE68 gene (but not the SCMV IE94 gene) is blocked at the transcriptional
level. Here we report the results of an extensive comparison of the level of
permissiveness for HCMV, SCMV and murine CMV (MCMV) in a variety of
additional cell types of human, monkey and mouse origin. We also describe a subtle
change in the tryptic peptide pattern of the IE68 polypeptide produced in BALB/c-3T3
cells compared to permissive human foreskin fibroblasts. Neither the IE68 nor IE94
proteins could be detected by biochemical labelling procedures in infected mouse Ltkor F9 teratocarcinoma stem cells, although IE94 was synthesized after retinoic acidinduced differentiation of the F9 cells. Synthesis of [35S]methionine-labelled IE94
protein, but not that of HCMV IE68, was detected in infected Vero cells and in human
peripheral blood leukocyte cultures. The failure to synthesize detectable IE68 protein
in infected Vero cells appeared to be unrelated to a lack of entry of viral DNA and to a
lack of appropriate transcription factors. Indeed, immunofluorescence assays showed
that the IE68 antigen was expressed efficiently in DNA-transfected Vero cells and in a
small fraction of infected Vero cells. Overall, two clear host range trends emerged.
First, whilst all three viruses showed a tendency for repression of IE expression in
transformed cell lines, the effect was severe for HCMV and only minimal for SCMV.
Secondly, progression of infection to the viral DNA synthesis level in non-transformed
fibroblast cell types occurred in a much wider range of host species cell types for SCMV
and MCMV than for HCMV.
INTRODUCTION
Human cytomegalovirus (HCMV) displays a strict host cell type and species specificity such
that in tissue culture systems at least, high titres of progeny virions are produced only in human
diploid fibroblasts (Rowe et al., 1956; Smith, 1956; Weller et al., 1957). Other human cell types
tested such as epithelial cells are, at best, only moderately permissive (Knowles, 1976; Vonka et
aL, 1976). However, efficient growth even in fibroblast cultures apparently requires adaptation
by serial passaging in cell culture. Fresh clinical isolates of HCMV have been reported to
produce some degree of viral gene expression and phenotypic change in peripheral blood
i" Present address: Merck Sharp & Dohme, Virus & Cell Biology Research Laboratories, West Point,
Pennsylvania 19486, U.S.A.
0000-7955 © 1988 SGM
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
356
R.L.
L A F E M I N A AND G. S. H A Y W A R D
lymphocyte, T cell and macrophage cell cultures, whereas fibroblast-adapted laboratory isolates
usually do not (Rice et al., 1984; Schrier et al., 1985). Since HCMV is believed to persist in a
latent state in cells of the monocyte/macrophage lineage in the human host, an understanding of
the specific features of different isolates and adapted strains which control host-range, and the
interactions of viral and cellular regulatory elements in general is likely to be of great
importance. Because of the great difficulties at present in carrying out biochemical analyses
with fresh clinical isolates, we have opted to explore and compare further the host range
properties of different mammalian CMV strains that have been adapted for growth in fibroblast
culture.
The HCMV genome is organized into long and short unique regions, each bounded by
inverted repeated sequences, and the packaged virion DNA molecules exist in four different
isomeric forms (Kilpatrick & Huang, 1977; Fleckenstein et al., 1982; LaFemina & Hayward,
1980). In permissive cells, HCMV gene expression is regulated such that under immediate early
(IE) conditions, the bulk of transcription is limited to within the major IE locus between 0.71
and 0-75 map units (DeMarchi, 1981 ; Wathen & Stinski, 1982; Stenberg et al., 1984; Stinski et
al., 1983). At delayed early (DE) times, only a trace amount of the HCMV IE68 protein
continues to be detectable and the remainder of the viral genome becomes transcriptionally
active (Stinski, 1978; DeMarchi, 1981; Wathen et al., 1981; LaFemina & Hayward, 1983).
Following DNA replication, the late structural proteins are synthesized, capsid assembly and
genome packaging occur and progeny virions are released. The well defined major IE promoter/
regulatory region contains several series of repetitive elements between positions - 5 0 and
- 5 5 0 (Thomsen et al., 1984) and can act as a strong transcriptional enhancer (Boshart et al.,
1985). The gene encoding the major IE68 nuclear phosphoprotein consists of four exons and
gives rise to a spliced mRNA species of 1.95 kb (Stinski et al., 1983; Stenberg et al., 1984; Akrigg
et al., 1985). In all CMV species the IE polypeptides probably act as transcriptional
transactivators of subsequent viral gene expression (Everett, 1984; O'Hare & Hayward, 1984;
Spaete & Mocarski, 1985; Koszinowski et al., 1986; M. Pizzorno, P. O'Hare, L. Sha, R. L.
LaFemina & G. S. Hayward, unpublished results).
In non-permissive cells, blocks to progression of the lytic cycle could occur at any of the steps
between adsorption and virion production, although restrictions at the IE transcriptional level
and DE post-transcriptional level are the only ones that have been described in any detail so far.
In rodent fibroblasts, both HCMV (Towne) and simian CMV (SCMV) (Colburn) IE proteins
are overproduced and DE transcripts accumulate in the nucleus, but no significant amounts of
DE and late polypeptides are produced and no viral DNA synthesis occurs (Jeang et al., 1982;
DeMarchi, 1983; Jeang, 1984; Geballe et al., 1986). A second type of block occurs in human
teratocarcinoma stem cells where HCMV (Towne) IE transcripts fail to accumulate (LaFemina
& Hayward, 1986; Nelson & Groudine, 1986), but subsequent retinoic acid-induced cell
differentiation results in complete IE viral gene expression, DNA replication and release of
progeny virions (G6ncz61 et al., 1984, 1985; LaFemina & Hayward, 1986). In contrast, SCMV
(Colburn) infection of human teratocarcinoma cell cultures is fully permissive for both IE gene
expression and DNA replication even in the undifferentiated stem cells (LaFemina & Hayward,
1986). The change in status of the teratocarcinoma cells for HCMV after differentiation bears at
least a superficial resemblance to the situation in vivo with murine CMV (MCMV) in which
differentiation of monocytes to macrophages leads to reactivation of latent infection (Dutko &
Oldstone, 1981).
DNA hybridization studies show that the genomes of HCMV (Towne) and the Colburn strain
of African green monkey SCMV have several colinear homologous loci (Jeang, 1984) and this
similarity is extended to the protein level in that the two species express many counterpart
polypeptides, some of which are antigenically related (Gibson, 1983). However, SCMV and
MCMV DNA molecules do not display the same typical inverting L/S type of genome structure
as HCMV, but possess a single isomer, linear genomic organization (LaFemina & Hayward,
1980; Jeang & Hayward, 1983; Eberling et al., 1983; Mercer et al., 1983). Although the
structural genome arrangement of SCMV and MCMV differs from that of HCMV, IE gene
expression is also confined to a single major colinear locus at 0-71 to 0-73 map units for SCMV
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
C M V expression in non-permissive cells
357
and 0-77 to 0-81 map units for M C M V (Jeang et al., 1982, 1984; Marks et al., 1983; Keil et al.,
1984). The apparent Mr of both the SCMV IE94 and M C M V IE98 polypeptides are
significantly larger than that of H C M V IE68 (Jeang & Gibson, 1980; Keil et al., 1985), but the
genes encoding the H C M V IE68 polypeptide and the SCMV IE94 polypeptide are remarkably
similar in their intron/exon structure and m a i n t a i n homologous protein domains despite the size
differences (Jeang et al., 1984; Jeang, 1984; Stenberg et al., 1984).
The present studies were directed at further understanding the interactions of m a m m a l i a n
CMV species with cultured cells. To examine whether or not repression of CMV major IE gene
expression is a generalized property of transformed cells and to ask in what other cell types the
differences between H C M V and SCMV IE gene expression might also be manifested, we have
carried out additional analyses of the level of HCMV, SCMV and M C M V expression in infected
cultures over a large range of cell types from a variety of host species. We show that H C M V
infection of non-permissive BALB/c-3T3 cells results in expression of an altered H C M V IE68
polypeptide, and that infection of F9, Ltk-, Vero, N B E and HL60 cells even at high m.o.i, fails
to yield significant levels of IE68 synthesis. I n contrast, synthesis of the SCMV IE94 polypeptide
occurs in many cell lines that do not express H C M V IE68. Both SCMV and M C M V also display
a much wider permissive host cell range than H C M V , with viral genome replication being
detectable in some transformed cell lines and across species barriers.
METHODS
Cells and virus. The sources of the cell lines used in these studies were as follows: 293 human embryonic kidney
cell line expressing adenovirus E1A and E1B genes, Dr G. Ketner, Department of Biology, Johns Hopkins
University; NBE human fibroblasts transformed by simian virus 40 (SV40), Dr S. Rhode III, Eppley Institute,
University of Nebraska, Omaha; NT2/D1 human teratocarcinoma cells, Dr P. Andrews, Wistar Institute,
Philadelphia, Pa. ; HL60 human premyelocyticleukaemia cells, Dr C. Civin, Johns Hopkins University Oncology
Center; 142 tk- human osteosarcoma tumour-derived cell line, Dr F. Graham, University of Ontario, Hamilton,
Canada; MG63, human osteosarcoma tumour cell-derived cell line, American Type Culture Collection; FOS
human megakaryocyte-like cells, Dr D. Morgan, Hahnemann University, Philadelphia, Pa.; PBL primary
lymphocytes from HCMV seronegative donors, Johns Hopkins University Hemapheresis Center; BALB/c-3T3
established murine cells, Dr T. Kelly Jr, Department of Molecular Genetics, Johns Hopkins University Schoolof
Medicine; Ltk--transformed murine cells, Dr S. Silverstein, Columbia University School of Medicine, N.Y. ; F9
murine teratocarcinoma cells and PYS murine transformed parietal yolk sac ceils, Dr J. Shaper, Johns Hopkins
University Oncology Center; MEF murine embryonic kidney fibroblasts, Drs W. Burns and S. Sanford, Johns
Hopkins University Oncology Center; AGMKF African green monkey kidney fibroblasts, Whittaker, M.A.
Bioproducts, Walkersville, Md. ; Vero-established AGMK cells, American Type Culture Collection. HF cells are
early passage secondary cultures prepared from diploid human foreskin fibroblasts.
Cells cultured as monolayers were grown in Dulbecco's modified Eagle's medium (DMEM, Gibco)
supplemented with 10~ foetal calf serum (FCS, Hyclone). Suspension cells were cultured in RPMI 1640 (Gibco)
supplemented with 10~ FCS for HL60 and primary lymphocyte cultures. Stock cultures of HCMV Towne strain,
SCMV Colburn strain and MCMV Smith strain, (provided by Drs K. Burns and S. Sanford) were passaged at low
m.o.i, to preclude the appearance of defective viruses. HCMV and SCMV were grown on HF cells and MCMV
was cultured in secondary cultures of MEF cells.
All experiments which required either 32Pi or [35S]methioninewere conducted in phosphate-free or methioninefree DMEM (Gibco) containing 10~ dialysed FCS (DFCS) added at the time of addition of the radiolabel as
previouslydescribed (LaFemina & Hayward, 1983). The m.o.i, for these infections ranged from approximately 20
p.f.u./cell for DNA labelling to 50 p.f.u./cell for protein labelling. Becauseof the difficultiesin maintainingfrozen
CMV stocks at high titres, fresh stocks of supernatant virions were used in each experiment.
Extraction ofintracellular and virion DNA. The infected cell DNA samples used in this study (25 cm2dishes) were
harvested by scraping, and processed followingcell lysis, Pronase digestion and phenol/chloroform extraction as
described by LaFemina & Hayward (1986). Infected cell DNA was precipitated in 0.3 M-sodiumacetate and 70
ethanol without prior dialysis, and subjected to restriction enzyme cleavage analysis. For the detection of newly
synthesized viral DNA, cells infected for 18 to 24 h were further cultured in DMEM containing 10~ DFCS and
100 ~Ci/ml 32p, for an additional 48 to 96 h depending on the times of development of the maximum cytopathic
effect in the control permissive cells. Radiolabelled infected cell DNA samples were analysed by restriction
endonuclease cleavage for the presence of newly synthesized viral DNA fragments by electrophoresis of digested
DNA through 1~ agarose gels and subsequent transfer to nitrocellulose (LaFemina & Hayward, 1983, 1986).
Under these conditions, infected HF cells processed before 24 h post-infection did not contain detectable levels of
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
358
R.L.
LAFEMINA AND G. S. HAYWARD
newly synthesized viral DNA. Samples collected from 48 to 96 h contained replicative forms lacking terminal
restriction fragments and samples extracted beyond 96 h contained L and S terminal restriction fragments,
demonstrating that viral DNA maturation had occurred (LaFemina & Hayward, 1983). SCMV and MCMV
displayed a slightly faster replicative cycle than HCM'V such that mature genomes could be detected by 96 h after
infection.
Protein labelling, SDS-PAGE and tryptic peptide analysis. For the detection of the CMV IE polypeptides, cells
were infected in the presence of 50 ~tg/ml cycloheximide (Sigma, C 6255) for 12 to 18 h followed by incubation for
3 h in methionine-free DMEM supplemented with 10~o DFCS and 25 IxCi/ml [35S]methionine. Infected cell
samples were processed and SDS-PAGE was performed as described in LaFemina & Hayward (1983). The
procedure used for the analysis of tryptic peptide cleavage products of the IE68 protein was that of Weiner &
Gibson (1983). Cell extracts were electrophoresed through a 10~ SDS-polyacrylamide gel, after which the gel was
dried for autoradiography. After locating the radiolabelled IE68 band, the appropriate region was excised from
the gel and rehydrated, then the polypeptide was eluted and digested to completion with trypsin (Worthington).
The resulting tryptic peptides were spotted on to pre-coated cellulose plates (EM Reagents) and were separated
initially by electrophoresis (acetic acid 78 :formic acid 25 :water 897 pH 1.9) and subsequently by chromatography
(butanol 75 :pyridine 50 : acetic acid 15 : water 60).
Retinoic acid-induced cell differentiation. The F9 murine teratocarcinoma stem cell line was cultured in DMEM
containing 10~ FCS and confluent monolayers were subcultured by trypsinization and dilution (1:50). For
retinoic acid-induced cell differentiation, cells were trypsinized, diluted (1 : 25) and cultured in DMEM containing
10~ FCS and 1 ~tM-retinoic acid (Sigma, R 2625) for 4 days prior to infection. In some experiments F9 cells were
differentiated further by the addition of 1 mM-dibutyryl cyclic AMP (dbcAMP, Sigma, D 0627; Strickland et al.,
1980). Retinoic acid and dbcAMP were maintained in the media throughout the course of the infection and
radiolabelling.
RNA dot blot analysis. Cells cultured in 75 cm 2 dishes were infected with HCMV (Towne) under similar
conditions to those used in the polypeptide labelling experiments, using an m.o.i, of 50 p.f.u./cell in the presence of
cycloheximide (50 ~tg/ml for 16 h). RNA was isolated following cell lysis with guanidinium isothiocyanate and
centrifugation through 5-7 M-CsCI as described by Chirgwin et al. (1979). Infected cell RNA was transferred to
nitrocellulose by filtration through a 96-well dot blot apparatus. The procedures described by Thomas (1980) were
used for prehybridization, hybridization and filter washing. The hybridization probe was prepared by nick
translation of pRL42 plasmid DNA, which contains a BamHI (0-740 map unit) to HindIII (0.766 map unit)
subfragment of HCMV (Towne) BamHI J and includes the coding region for the IE68 polypeptide. Confirmation
that equivalent amounts of RNA (2 to 4 ~tg) from the various cell types used in these assays was obtained by
methylene blue staining of the nitrocellulose filters after hybridization.
Immunofluorescence. Confluent monolayers of 2 × 105 cells cultured in two-well slide culture chambers were
transfected with 2-0 ~tg of supercoiled pRL103 plasmid DNA containing HCMV (Towne) HindlII C (0-675 to
0.766 map units) by the standard calcium phosphate technique (Middleton et al., 1982). At 48 h after transfection
or 18 h after infection, cells were fixed in absolute methanol at - 20 °C for 5 rain. Cells were stained at 37 °C for 30
min, with either the E3 monoclonal antibody (Goldstein et al., 1982; Stinski et al., 1983) or a commercial antiCMV IE monoclonal antibody (Biotech), followed by incubation with fluorescein isothiocyanate-labelled goat
anti-mouse antibody, and were then examined under oil immersion with a Leitz epifluorescence u.v. microscope.
Photographs were taken on 125 ASA PLUS-X film using a x 40 objective and narrow band fluorescein
isothiocyanate filters.
Chloramphenicol acetyltransferase (CAT) assays. Confluent monolayers of 5 x 105 cells grown in six-weU dishes
were transfected with 2.0 ktg of supercoiled DNA by the calcium phosphate precipitation procedure as described
by O'Hare & Hayward (1984). After 4 h the ceils were treated with 1 ml of 15 ~ glycerol in DMEM without serum
for 1 rain. The cultures were then washed with calcium- and magnesium-free phosphate-buffered saline (PBS-A)
and further incubated for an additional 44 h in DMEM with 10~o FCS. Cell extracts were prepared and CAT
assays were performed as described by Gorman et al. (1982) and modified by O'Hare & Hayward (1984).
RESULTS
Characterization o f the IE68 polypeptide synthesized in non-permissive B A L B / c - 3 T 3 cells
P r e v i o u s studies s h o w e d t h a t following H C M V ( T o w n e ) infection in the p r e s e n c e of
c y c l o h e x i m i d e and subsequent reversal (i.e. u n d e r I E conditions) n o n - p e r m i s s i v e B A L B / c - 3 T 3
cells o v e r p r o d u c e d a n o v e l p o l y p e p t i d e of Mr 68K. T o d e t e r m i n e w h e t h e r or not the 6 8 K
p o l y p e p t i d e p r o d u c e d in m o u s e cells was p h o s p h o r y l a t e d , B A L B / c - 3 T 3 and H F cells were
infected u n d e r I E c y c l o h e x i m i d e reversal c o n d i t i o n s in the p r e s e n c e of [35S]methionine or 32P i.
Analysis o f extracts of both infected cell types r e v e a l e d an a b u n d a n t m e t h i o n i n e - l a b e l l e d 6 8 K
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
359
C M V expression in non-permissive cells
(a)
(b)
1
3
4
1
2
3
4
IE68
~i~ i "~
M
•
{
Fig. 1. The major HCMV IE68 polypeptide is phosphorylated in non-permissive mouse cells. (a) Nonpermissive BALB/c-3T3 and (b) permissive HF cell cultures were both infected at an m.o.i, of 20 in the
presence of 50 p-g/ml cycloheximide for 12 h. The cells were then washed with phosphate-buffered
saline (reversal) and the IE polypeptides were labelled with either [3sS]methionine (lanes 1 and 2) o r 3 2 p i
(lanes 3 and 4) for 3 h. Mock-infected (lanes 1 and 3) and HCMV-infected (lanes 2 and 4) IE
polypeptides were separated by electrophoresis through 10~ SDS-polyacrylamide gels. The position of
the HCMV IE68 polypeptide is indicated by an arrow in the autoradiographs.
polypeptide that was not detected in mock-infected cells and this species also proved to be the
major phosphate-labelled protein present in both the permissive and non-permissive cell
extracts (Fig. 1).
Further evidence that the 68K polypeptide synthesized in infected mouse cells was indeed
equivalent to the authentic viral IE68 protein from H F cells was obtained by two-dimensional
analysis of tryptic peptide cleavage products. The 35S-labelled infected cell polypeptides were
overproduced by cycloheximide reversal and the major 68K bands were isolated by preparative
gel electrophoresis (Fig. 2). Both 68K bands gave similar tryptic peptide cleavage patterns, with
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
~i~i!~i:••
•~::•~iii~i::iii::? i~ilZ:i~ii~:
: •i i!¸
Mock
Infected
BALB/c-3T3
• :~ i
i~i•~.~ i~!i¸!i~
...... •~:: • ~!~:,~••~ •~
Mock
HF
~i•~!i:: ~i~!i!~!ii~~ i~,,i¸ ii
Infected
T
+
~
BALB/c-3T3
Electrophoresis
~,
HF
Fig. 2. Two-dimensional separation of [35S]methionine-labeUed HCMV IE68 tryptic peptide cleavage products shows the presence of a novel peptide in IE68
synthesized in mouse cells. (a) The [35S]methionine-labelled HCMV polypeptides synthesized in infected BALB/c-3T3 or HF cells after cycloheximide reversal were
separated from other viral and cellular proteins by preparative SDS-polyacrylamide gel electrophoresis. (b) The 68K polypeptide bands were extracted from the gel
as described by Weiner & Gibson (1983) and digested to completion with trypsin. The resulting tryptic peptides were separated by electrophoresis (direction
indicated by - and + ) and chromatography (direction indicated by the thin arrow). The major [35S]methionine_containing peptide that was detected in the mouse
:ell IE68 protein but was barely visible in the human cell IE68 protein is indicated by the thick arrow.
68K
(a)
(b)
>
~7
t~
C M V expression in non-permissive cells
361
the exception that the mouse cell protein band produced at least one major tryptic peptide
(arrowed) that was underrepresented by approximately 20-fold in the control IE68 band
extracted from HCMV-infected human cells. A very similar result was obtained in a second
experiment of this type and all of the more intensely labelled tryptic peptides (including the
variable one) were absent following similar processing of protein extracted from the 68K region
in the gel lanes containing mock-infected cell polypeptides (not shown). Similar separation of
tryptic peptides from the isolated 32p-labelled 68K proteins of infected BALB]c-3T3 and HF
cells resulted in identical but much simpler patterns from both host cell types (not shown).
Therefore, the variable peptide was not included amongst the phosphorylated tryptic peptides of
IE68 and there was no evidence that the phosphorylation patterns differed in permissive and
non-permissive cells.
Differences between H C M V and S C M V IE protein expression in transformed mammalian cell
lines
We have previously demonstrated that synthesis of the HCMV IE68 protein is not detectable
by our assays following infection of the 293 adenovirus DNA-transformed human cell line or the
NT2/D1 human teratocarcinoma stem cell line, although both are permissive for SCMV IE94
protein expression and viral DNA replication (LaFemina & Hayward, 1986). Similar results
were obtained in the current studies after infection of the SV40-transformed, large T antigen
positive NBE human cell line (not shown) and in Vero ceils (Fig. 3). In contrast, we observed
that transformed mouse F9 stem cells and the PYS derivative did not synthesize detectable
amounts of either the HCMV IE68 or SCMV IE94 proteins after high multiplicity infection
(Fig. 4). Like the human teratocarcinoma cells, the permissiveness of F9 stem cells could be
altered after differentiation by retinoic acid treatment, but only for synthesis of the SCMV IE94
protein and not HCMV IE68 (Fig. 4). Further differentiation induced by dbcAMP treatment of
F9 cells still did not result in HCMV IE68 synthesis (Fig. 4). We were unable to detect
replication of either the HCMV or SCMV genomes following infection of any of these cell lines
in the presence or absence of retinoic acid (not shown).
We have previously been unable to detect accumulation of steady-state IE68 mRNA
following infection of human teratocarcinoma stem cells with HCMV in the presence of
cycloheximide (LaFemina & Hayward, 1986), which implied that the block to IE68 expression
occurred at the level of mRNA transcription or stabilization. Several of the non-permissive
transformed cell lines used in these present studies were similarly examined by an RNA dot blot
hybridization procedure for their ability to express IE68 m R N ~ . Using nick-translated plasmid
pRL42 D N A (which contains the coding region for the HCMV IE68 protein) as probe, little if
any IE68-specific RNA was detected in either 293, Vero or F9 cells (Fig. 5). Therefore,
apparently these cells were also blocked at the mRNA level. In contrast, abundant HCMV IE68
RNA was produced in infected permissive HF cells.
Differences between H C M V and S C M V IE protein expression in human lymphocyte cell types
Infection of primary human lymphocytes in culture with clinical isolates of HCMV has been
reported to yield a low fraction of positive nuclei by immunofluorescence assay with monoclonal
antibody against the IE68 protein (Rice et al., 1984). However, laboratory-adapted strains of
HCMV which grow to high titres in fibroblasts are generally negative in these assays. We were
unable to detect HCMV (Towne) IE68 polypeptide synthesis by [3SS]methionine radiolabeiling
and SDS-PAGE in either PBL (Fig. 6a), or in the FOS cell line (Morgan & Brodsky, 1985) or the
human HL60 cell line (Reitsma et al., 1983) either before or after 12-O-tetradecanoylphorbol 13acetate- or retinoic acid-induced differentiation. Nevertheless, low level synthesis of the SCMV
(Colburn) IE94 polypeptide was detected following infection of both PBL (Fig. 6a) and the FOS
cell line (not shown), although not in HL60 cells (not shown). Despite the reports that fresh
clinical isolates of HCMV express the IE68 protein in a small proportion of PBL as assayed by
immunofluorescence, we were also unable to detect a 35S-labelled IE68 protein band under IE
conditions after infection of the PBL culture with a recent clinical HCMV isolate (JHH 431) at
fourth passage in HF cells (lanes 3 in Fig. 6a). We also asked whether SCMV DNA replication
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
362
R. L. LAFEMINA AND G. S. HAYWARD
(a)
1
2
(b)
3
1
(c)
2
3
1
2
3
•
IE94
•
IE68
Fig. 3. HCMV IE68 is not synthesized in infected Vero cells. BALB/c-3T3 (a), human diploid
fibroblasts (b) and Vero (c) cells were mock-infected (lanes 1) or infected with SCMV (lanes 2) or
HCMV (lanes 3). The locations of the IE68 and IE94 polypeptides are indicated by arrows.
could be detected in the PBL cultures which permitted synthesis of the IE94 protein. Parallel H F
and PBL cultures were infected, and newly synthesized 3,.P_labelled D N A was examined for the
presence of virus-specific restriction enzyme fragments. However, we were unable to detect
replication of the SCMV genome following infection of PBL cultures (Fig. 6b), even though the
cells remained viable as evidenced by 32p i incorporation into host cell D N A . Not surprisingly
the lymphocytes were also non-permissive for H C M V D N A replication.
Detection of input H C M V genomes in the nuclei of infected Vero cells
Since the type of non-permissiveness found in transformed cell lines could potentially be
associated with a lack of virion receptors on the cell surface, it was important that we address
this question for H C M V infection in a cell line that did not p e r m i t synthesis of the H C M V IE68
protein. Therefore, we infected BALB/c-3T3, H F and Vero cells with purified virions the D N A
of which had been radiolabelled during replication in H F cells in the presence of 32p i. As a
measure of virus entry, cell nuclei were isolated at 16 h after infection by N P 4 0 lysis and
centrifugation and found to contain the parental 32p-labelled H C M V D N A (Fig. 7a).
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
C M V expression in non-permissive cells
(a)
363
(c)
(b)
RA
1
dbcAMP
I
I
1
IE94
2
3
1
2
3
1
2
3
1
2
3
1
2
J,
IE68
~•:
iI ~:
~
~
~.......
ii
i:~i:
Fig. 4. Differentiation of murine F9 teratocarcinoma cells. Infections of untreated HF cells (a) and the
transformed PYS cells (c) are shown with F9 cells (b) before and after treatment with retinoic acid (RA)
alone or in conjunction with dbcAMP. Polypeptides from uninfected (lanes 1), HCMV (Towne)infected (lanes 2) and SCMV (Colburn)-infected (lanes 3) cells were labelled with [35S]methionine and
analysed after SDS-PAGE and fluorography. The locations of SCMV IE94 and HCMV IE68
polypeptides are indicated.
Furthermore, although the proportion of input 32p-labelled D N A recovered in the nuclear
fraction of infected Vero cells was somewhat less than that in the permissive HF cells it was
equal to that of the input SCMV genomes in all three cases. Thus, although Vero cells do not
express IE68, an adequate amount of HCMV D N A does enter the nuclei. This result is
consistent with the observations of Einhorn et aL (1982) who demonstrated the presence of
HCMV receptors on the surface of various non-permissive cell types including Vero cells.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
364
R. L. L A F E M I N A A N D G. S. H A Y W A R D
RNA
HF Mock
HF HCMV
293 HCMV
Vero HCMV
F9 HCMV
Fig. 5. The block to HCMV IE68 expression in transformed cells occurs at the transcriptional level.
RNA produced in mock-infectedHF- and HCMV-infected HF, 293, F9 and Vero cells in the presence
of a cycloheximideblock was extracted and fixed onto nitrocellulose filters. Duplicate samples of 2 to 4
Ixgof RNA were hybridized with a 32p-labelledpRL42 DNA probe which contains the coding region
for HCMV IE68 as a BarnHI to HindlII fragment (0-740 to 0-766 map units) within this plasmid.
Analysis of the B a m H I restriction digest pattern of input 32p-labelled H C M V D N A at 16 h
after infection of permissive HF cells revealed that the L terminus represented by the B a m H I Z
fragment was decreased in intensity relative to the similarly sized B a m H I Y fragment (Fig. 7 a).
This result suggested that a significant fraction of the input genomes had generated covalently
linked termini prior to D N A replication, similar to an observation reported for MCMV (Marks
& Spector, 1984). In contrast, the parental B a m H I Z fragment present in the digest from infected
BALB/c-3T3 cells was barely reduced in intensity implying that few if any molecules displayed
linked L/S termini. The formation of a small percentage of joined termini in HCMV is difficult
to document clearly because the resulting circular or concatemeric structures merely increase the
molarity of already existing L/S joint restriction fragments. However, this problem was avoided
with SCMV, where linkage of the left and right termini generates a completely new restriction
fragment which was easily detected in the infected HF sample (Fig. 7 b, A*). Therefore, the total
absence of this novel fragment in the recovered input D N A from SCMV-infected BALB/c-3T3
cells provided further support for our earlier conclusion that prereplicative intermediates with
joined termini do not form in non-permissive mouse cells (LaFemina & Hayward, 1983).
S C M V and M C M V have a broader host-range than H C M V
An overall summary of IE gene expression and viral D N A replication in a variety of primary,
established and transformed rodent, simian and human cell cultures infected with HCMV,
SCMV or MCMV is given in Table 1. Sufficient HCMV IE68 synthesis to give a 35S-labelled
band by P A G E after reversal of a cycloheximide block was never observed in transformed cell
lines of human, monkey or mouse origin (except in the 143 and MG63 human osteosarcoma cell
lines). Although IE68 expression did occur in all diploid fibroblasts tested, evidence for H C M V
D N A replication obtained by 32p i incorporation into the characteristic newly synthesized viral
D N A fragments was not detected in secondary diploid kidney fibroblast cell cultures of African
green, cynomolgus or rhesus monkey origin, nor in MEF. In contrast, SCMV displayed a much
wider host cell range, with IE94 expression being undetectable only in transformed mouse cells
such as the Ltk- and F9 stem cell lines and in human HL60 cells. Furthermore, SCMV D N A
replication was observed even in transformed human 293 cells, the NT2/D1 teratocarcinoma
stem cell line and in Vero cells, although not in any of the mouse cell cultures.
Extension of these studies to MCMV revealed that it also displayed an expanded host cell
range in contrast to the tightly restricted host cell range of HCMV. Although we routinely
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
C M V expression in non-permissive cells
(a)
(b)
HF
I
M
365
PBL
2
3
II
M
1
PBL
HF
I
2
1
2
3
II
1
2
3
IE94"--~
IE68---~
:
b
Oq
#
....
is.
Fig. 6. PBL express SCMV IE94 but are non-permissive for SCMV DNA replication. (a) Human PBL
and HF cells were infected with either the 431 clinical isolate of HCMV (lanes 1), the established
Towne strain of HCMV (lanes 2) or SCMV Colburn (lanes 3).or mock-infected (M) in the presence of
cycloheximide, followed by reversal in the presence of [3sS]methionine. Newly synthesized
polypeptides were separated by electrophoresis in 10~o SDS-polyacrylamide gels and visualized by
fluorography. The location of HCMV IE68 and SCMV IE94 polypeptides are indicated by arrows. (b)
Human PBL and HF cells were infected for 16 h followed by 32 P radiolabelling of newly synthesized
DNA as described in Methods. The autoradiograph shows the agarose gel electrophoresis profiles of
EcoRI restriction enzyme cleavage products from newly synthesized total cellular DNA from these
cultures. The arrows denote the locations of mitochondrial DNA fragments.
passaged M C M V (Smith) in secondary cultures of M E F cells, infection of the established
BALB/c-3T3 cell line proved to result in easily detectable synthesis of 32p-labelled M C M V
D N A and expression o f the M C M V major IE98 protein (which under our electrophoresis
conditions a p p e a r e d slightly larger than that reported by Keil et al., 1985). Surprisingly, the
transformed mouse L t k - cell line, which failed to permit expression of either IE68 or IE94, also
proved to contain a block to M C M V IE98 (Fig. 8a). The F9 teratocarcinoma cells proved not to
be directly informative for study of M C M V IE gene expression because they expressed a cellular
98K polypeptide. However, infection of the F9 cells resulted in low but detectable levels o f viral
D N A replication against a high background o f host cell D N A synthesis (data not shown) and we
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
R. L. LAFEMINA AND G. S. HAYWARD
366
(a)
(b)
BALB
1
2
HF
1
Vero
2
1
2
1
2
3
<1
I>
AS
A*
- - 1 2 kb
- - 5 . 5 kb
Z
Fig. 7. Entry and terminal joining of viral DNA in permissive compared to non-permissive cells. (a)
Parimtal input HCMV (Towne) DNA does enter the nucleus in infected non-permissive Vero cells.
BALB/c-3T3, HF and Vero cells were infected with 32p-labelled parental HCMV (lanes 1) or SCMV
(lanes 2) for 16 h at which time nuclei were collected by NP40 lysis at 0 °C followed by centrifugation.
Nuclear DNA was extracted and examined for the presence of input parental radiolabelled virion
DNA by BamHI digestion and electrophoresis through 1~ agarose gels. The presence of the SCMV
BamHI joint restriction fragment is indicated (closed arrows A*) and also the HCMV L terminus
BamHI-Z fragment (closed arrows) which displays decreased relative molarity presumably related to its
joining with the terminal BamHI-L and -N fragments from the S segment. (b) Unambiguous formation
of novel joint fragments during SCMV replication. Newly synthesized 32p-labelled DNA was extracted
from HF cells infected with SCMV for 72 (lane 1) or 120 (lane 2) h and compared with SCMV DNA
extracted from supernatant progeny virions (lane 3). The infected cell samples contain BamHI-A*
species whereas released virions do not. The sizes of the SCMV BamHI terminal fragments are given.
The open arrows denote high Mr cellular DNA.
p r e s u m e that IE98 expression m u s t h a v e occurred. T r e a t m e n t of F9 cells w i t h retinoic acid
reduced the a m o u n t o f host D N A synthesis such that replication o f the input M C M V g e n o m e s
was readily detectable (Fig. 8 b). I n f e c t e d secondary cultures o f diploid A G M K F cells (Fig. 8 b)
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
CMV expression in non-permissive cells
T a b l e 1.
t"
367
Summary of cytomegalovirus IE expression and DNA replication in various cell types
Host cell infected
HCMV (Towne)
SCMV (Colburn)
A
&
A.
MCMV (Smith)
Cell type
Characteristics
IE*
DNAt
IE*
DNAt
Human
HF
Diploid fibroblast
++++ ++++ ++++ ++++
NT2/DI:~
Human teratocarcinoma
+ q- q+ + -]NT2/D1 + RA~: R e t i n o i c a c i d - d i f f e r e n t i a t e d
+ + +
+ + +
+ + +
+ + -{143
Osteosarcoma, TK+ + +
+
ND§
ND
MG63
Osteosarcoma
+ + +
+
+ + +
+
PBL
Peripheral blood lymphocytes
+
FOS
Megakaryocyte-like
+
HL60
Premyelocytic leukaemia
.
.
.
.
NBE
HF, SV40 T Ag +
ND
ND
293:~
HEK, adenovirus E1A +
-+++
++
Simian
AGMKF
African green monkey kidney
fibroblasts
++
++
++
CyMF
Cynomolgus monkey fibroblasts
+ +
+ +
-{- +
RhMF
R h e s u s monkey f i b r o b l a s t s
+ +
+ +
+ +
Vero
AGMK, immortalized
+++
+
CVI
AGMK, immortalized
ND
ND
+++
ND
COS
CVI, SV40 TAg +
ND
ND
+ + +
ND
Murine
MEF
Mouse embryo fibroblast
+ +
ND
ND
+
BALB/c-3T3
Mouse, immortalized
++++
+++
F41A1
BALB/c-3T3, herpes simplex
virus type 2 mtrlI
+
-ND
ND
F9
Mouse teratocarcinoma
.
.
.
.
F9 + RA
Retinoic acid-differentiated
++
PYS
Parietal yolk sac, transformed
.
.
.
.
LtkMouse, transformed TK.
.
.
.
.
A
IE*
DNAI"
+
+
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
+
+
ND
ND
ND
ND
ND
ND
ND
ND
+ + +
+ + + +
++
++
ND
(+)
(+)
ND
ND
+
+
ND
* Relative level of synthesis of [35S]methionine-labelled HCMV IE68, SCMV IE94, or MCMV IE98 proteins
after cycloheximide reversal. HCMV or SCMV infection in fully permissive HF cells and MCMV in MEF cells
defines the + + + + level; 30 % to 70% of that level of synthesis is indicated by + + + ; 10 % to 30 % by + + ; 3 % to
10% by + and less than 3% by - .
t Relative level of newly synthesized 32p-labelled viral DNA. The symbols used represent the same range of
values used in *
:l: Data from LaFemina & Hayward (1986).
§ ND, Not done.
a n d H F cells ( d a t a n o t s h o w n ) b o t h e x p r e s s e d I E 9 8 a n d g a v e low level s y n t h e s i s o f M C M V
DNA.
I n s u m m a r y , H F cells lay at o n e e n d o f t h e s p e c t r u m , p e r m i t t i n g b o t h I E g e n e e x p r e s s i o n a n d
v i r a l D N A r e p l i c a t i o n for all t h r e e v i r u s species ( H C M V , S C M V a n d M C M V ) . A t t h e o t h e r
e x t r e m e , m o u s e L t k - cells w e r e n o t p e r m i s s i v e for I E g e n e e x p r e s s i o n w i t h a n y o f t h e s e v i r u s e s
( a n d h e n c e also failed to s y n t h e s i z e v i r a l D N A ) . D i p l o i d f i b r o b l a s t s f r o m all t h r e e h o s t species
g r o u p s w e r e p e r m i s s i v e for I E g e n e e x p r e s s i o n for all t h r e e C M V types, b u t s h o w e d a c l e a r
h i e r a r c h y for D N A r e p l i c a t i o n , w i t h M C M V D N A s y n t h e s i s o c c u r r i n g in m o u s e , m o n k e y a n d
h u m a n cells, w h e r e a s S C M V D N A s y n t h e s i s w a s l i m i t e d to m o n k e y a n d h u m a n cells, a n d
H C M V D N A s y n t h e s i s w a s r e s t r i c t e d to h u m a n cells only.
HCMV IE gene expression in non-permissive cells following DNA transfection
Considering both the structural similarities and strong constitutive promoters of IE68 and
I E 9 4 , it w a s s o m e w h a t s u r p r i s i n g t h a t we w e r e u n a b l e to d e t e c t I E 6 8 m R N A or p o l y p e p t i d e
s y n t h e s i s in i n f e c t e d N T 2 / D 1 , 2 9 3 or N B E cells, w h e r e a s t h e s e t r a n s f o r m e d h u m a n cell lines d i d
n o t d i s p l a y a n y b l o c k s to I E 9 4 g e n e e x p r e s s i o n . T h e r e f o r e , to a s k w h e t h e r t h e s e t y p e s o f cells, or
V e r o cells, m i g h t s h o w a n i n t r i n s i c a n d s e l e c t i v e b l o c k to H C M V I E 6 8 t r a n s c r i p t i o n , we
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
368
R. L. L A F E M I N A A N D G. S. H A Y W A R D
(a)
BALB
I
1
2
3
Ltk~
4
I[
1
2
3
4
I
(b)
1
2
3
4
5
IE98
IE94
IE68
Fig. 8. MCMV IE98 gene expression and DNA replication occur over an extended host cell range. (a)
Polypeptidessynthesized in uninfected cells (lanes 1) or in MCMV (lanes 2), SCMV (lanes 3) or HCMV
(lanes 4) infected BALB/c-3T3 and Ltk- cells under IE conditions were extracted and analysed by
gel eleetrophoresis and fluorography. The location of the MCMV IE98, SCMV IE94 and HCMV IE68
polypeptides expressed in BALB/c-3T3 but not Ltk- cells are arrowed. (b) 32p-labelled newly
synthesized DNA from uninfected (lane 1) or MCMV-infected cultures of MEF cells (lanes 2 and 4),
mouse teratocarcinoma (F9) cells treated with 1 ~tM-retinoicacid (lane 3) and secondary AGMKF cells
(lane 5), was extracted and analysed by SalI restriction endonuclease digestion, gel electrophoresis and
autoradiography. The arrowheads denote the location of high Mr cellular DNA.
examined transient expression of hybrid genes under the control of the H C M V IE68 promoter/
enhancer sequences after D N A transfection. The plasmids p W T 7 6 0 - C A T (Stinski & Roehr,
1985) and p S V 2 - C A T contain a bacterial C A T gene under the transcriptional control of either
the H C M V IE68 gene promoter or the early promoter/enhancer region o f SV40. These
plasmids were introduced into several representative H C M V permissive and non-permissive
cells by the calcium precipitation procedure and cell extracts were examined for C A T activity
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
CMV expression in non-permissive ceils
(a)
369
(b)
760
SV2
HF
760 SV2
293
760
SV2
Vero
Fig. 9. DNA transfection assays demonstrate that there is no intrinsic block to HCMV IE gene
expression in transformed cells. (a) Expression of CAT enzyme from a hybrid HCMV IE promoter/
enhancer construct. SupercoiledpWT760-CAT or pSV2-CAT DNA (2 p.g)was precipitated by calcium
phosphate and added directly to permissive HF and non-permissive293 or Vero cells. Cell extracts were
made at 48 h and were used in standard CAT assays. (b) Expression of the HCMV intact IE68 gene in
293 cells following transfection of pRL103 plasmid DNA containing the HCMV (Towne), HindlII-C
fragment (0.675 to 0.766 map units). At 48 h cells were fixed in absolute methanol and synthesis of the
IE68 protein was detected in the cell nuclei by immunofluorescence using a monoclonal antibody
against the HCMV IE nuclear antigen.
Table 2. Comparison of H C M V IE68 gene expression after infection or DNA transfection in
different cell types
Cell type
HF
MG63
BALB-c/3T3
293
Vero
I'E68 protein
CHX/Rev*
++++~
+++
++++
---
Virus infection (m.o.i. 20)
x
IE68 mRNA HCMV DNA
Input DNA
CHX
synthesis
in nucleus
+++
++++
+++
ND§
+
NO
ND
-+++
--ND
--++
DNA
~
transfection
Anti-IE68
Anti-IE68
nuclear IF~- nuclear IF
>95%
0.2%
20%
0.2%
>80%
0.2%
0"2%
5-15%
1-2%
1 2%
* Cycloheximide and reversal.
t IF, immunottuorescence.
- , +, + +, + + +, + + + +, relative levelsof efficiencyof protein, RNA and DNA synthesis or DNA entry.
§ ND, Not done.
after 48 h. The pWT760-CAT D N A gave high level basal expression in 293 and Vero cells but
only low level expression in H F cells (Fig. 9a). All three cell types displayed more a b u n d a n t
C A T expression following transfection with p W T 7 6 0 - C A T D N A than with pSV2-CAT D N A .
The known low efficiency of D N A transfection into H F cells presumably accounts for the low
C A T levels obtained in the permissive control. In contrast, 293 cells in particular are known to
be highly efficient for D N A transfection (Alwine, 1985).
We next asked whether we could detect expression of the intact 1E68 polypeptide itself in nonpermissive transformed cells following short-term D N A transfection. For these studies we used
the 20 kb plasmid pRL103 (HindlII C) which includes the complete major IE gene region from
H C M V (Towne). The IE68 nuclear antigen proved to be readily detected in the nucleus of 5 % to
15% of transfected 293 cells, in 1% to 2% of Vero cells and in approximately 0.2% of HF,
BALB/c-3T3 and MG63 cells by immunofluorescence using the E3 monoclonal antibody (Fig.
9b and Table 2). Again, this result appears to reflect only the D N A transfection efficiency
properties of the cells and was independent of their permissive or non-permissive character for
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
370
R.L.
L A F E M I N A AND G. S. H A Y W A R D
expression of the IE68 protein after virus infection. Thus the block to HCMV IE68 gene
expression in transformed cells was not evident following the introduction of plasmid DNA
containing either the intact HCMV IE68 gene or hybrid CAT genes under the control of the
HCMV IE promoter.
A small proportion of cells in the non-permissive transformed cell cultures escape the block to IE
expression
When the E3 monoclonal antibody assay was employed to monitor IE68 expression at 18 h
after HCMV infection in the absence of cycloheximide treatment (at our standard high m.o.i, of
20 p.f.u./cell), strong nuclear fluorescence was observed in l ~ to 2 ~ of Vero cells and in
approximately 0-2~0 of the 293 cells, compared to 20~o in MG63 cells and greater than 95~o in
HF cells (Table 2). At an even higher m.o.i. (greater than 100 p.f.u./cell) the proportion of
positive nuclei was increased to 5 ~ in Vero cells and 1 ~ in 293 cells, whereas at a lower m.o.i.
(i.e. 1 p.f.u./cell), which was still sufficient to give 30~ to 50~ positive nuclei in HF cells, less
than 0.2 ~o positives were obtained in Vero cells. Therefore, these results further support the idea
that there is no absolute block to IE68 expression in the transformed non-permissive cells and
that apparently once some threshold is reached the level of IE polypeptide produced in an
individual cell may be just as high as in a standard permissive cell. However, at a given m.o.i, the
proportion of cells that reach this threshold is extremely small in Vero or 293 cell cultures
compared to HF cells.
DISCUSSION
The additional [3SS]methionine-labelled tryptic peptide present in the HCMV IE68 protein
made in BALB/c-3T3 ceils was almost totally absent from the equivalent protein synthesized in
permissive human fibroblasts. This result could be explained by an additional protein
processing event that occurs only in permissive cells. Alternatively, the 1-95 kb IE68 mRNA
transcript may be spliced differently in mouse cells compared to human fibroblasts, with the
altered mRNA encoding the additional or substituting tryptic peptide. A trace of the extra
peptide was also found in the HF cell IE68 protein sample and therefore we presume that a low
level of the aberrant pi'otein processing or mRNA splicing must also occur in permissive cells.
Since the apparent lack of proper functioning of IE68 in non-permissive BALB/c-3T3 cells
might potentially be related to the presence of this novel form of the IE68 protein, the origin of
the extra peptide and its role in affecting the function of IE68 in non-permissive cells warrants
further investigation.
The outcome of HCMV infection is extremely host cell-dependent. Both rodent and simian
fibroblasts allow abundant expression of the HCMV IE68 polypeptide, but we have been unable
to detect any HCMV DNA replication in these cells. In contrast, IE68 mRNA expression was
blocked at either the transcription initiation or mRNA stabilization level in several established
transformed cell lines including those of human origin (Fig. 6; LaFemina & Hayward, 1986;
Nelson & Groudine, 1986). This tight block was noted only following virus infections and not
after introduction of p!asmid DNA containing the intact IE68 gene or hybrid IE68 promoter/
enhancer CAT constructs by DNA transfection protocols. Therefore, there is no intrinsic block
to transcription of the !E68 gene in the non-permissive transformed cells. Since we know that
HCMV DNA enters these cells efficiently after infection (Fig. 7), several questions arise
regarding the nature of the virus celt interaction. The repression of the HCMV IE68 gene in
transformed cells following virus infection may be the result of autoregulation or of the
introduction of a negative virion factor acting directly or indirectly on the IE promoter itself.
Evidence has been presented for the existence of positive HCMV virion factors (Spaete &
Mocarski, 1985; Stinski & Roehr, 1985) and it may be that these function properly only in
permissive cells with aberrant regulation occurring in non-permissive cells. Alternatively, the
repression may result from a more generalized host cell response to HCMV infections. Clearly,
transcription factors inthe host cell are of considerable importance and these may vary from cell
type to cell type. We cannot conclude as yet whether infected HF cells have positive factors
necessary for HCMV IE gene expression or are lacking in some repressor-like activities found in
the non-permissive transformed cells.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
C M V expression in non-permissive cells
371
In F9 cells, CAT target constructions containing the polyomavirus or simian virus 40 early
promoter/enhancer region or the Moloney murine leukaemia virus long terminal repeat are
expressed poorly even in short-term DNA transfection protocols (Linney et al., 1984; Amati,
1985 ; Sleigh & Lockett, 1985). However, we observed substantial basal IE68-CAT expression in
both Vero and 293 cells in transient assays. Potentially, the DNA transfection protocols may
introduce such high concentrations of DNA that limiting amounts of cellular repressor
molecules are titrated out, allowing the HCMV IE gene to be expressed. In addition, other
considerations such as the high efficiency of 293 cells for stabilizing input plasmid DNA may
also be important (Alwine, 1985). Our preliminary experiments suggest that SCMV
superinfection of non-permissive Ltk- cells that had previously received an IE94-CAT
plasmid by transient DNA transfection results in shut-off of the high basal expression of
IE94-CAT, indicating that a negative virion factor may indeed be a necessary component
of the non-permissive interaction in infected cells (D. Gay & G. S. Hayward, unpublished
data).
Two quite different trends emerged with regard to the cell types that allowed expression of
CMV IE gene products compared to those that permitted viral DNA replication. At both levels,
the highest degree of restriction was exhibited by HCMV, with SCMV displaying the broadest
host range with regard to IE gene expression and MCMV the broadest with regard to DNA
synthesis. In terms of IE gene expression, some unknown feature of ti'ansformed cell lines and
teratocarcinoma stem cells compared to diploid fibroblasts or differentiated teratocarcinoma
cells appeared to predominate over species-specific effects, whereas for viral DNA replication
the species type of the host cell was the most important feature.
The significance of the very small percentage of cells in HCMV-infected non-permissive Vero
and 293 cell cultures that did express IE68 detectable by the more sensitive immunofluorescence
assay is not clear at present. Zerbini et al. (1985) have shown that the level of HCMV early
antigen expression can be increased fourfold after heat shock treatment in infected Vero cells. It
is neither unreasonable nor unprecedented for limiting cellular 'repressor' factors to be
overcome at high m.o.i, or for a small proportion of defective virus particles (that may lack shutoff factors) to be present. We presume that additional criteria, including surface receptor and
entry or adsorption problems, may hinder infection in other cell types, especially those of
lymphoid origin (Einhorn et al., 1982).
We have previously suggested that differences in the organization of the upstream regulatory
elements of the HCMV and SCMV IE promoters may influence the cell type-specific expression
of IE genes (LaFemina & Hayward, 1986). Considerable DNA sequence homology occurs in
the upstream promoter/regulatory sequences and 5' introns of the HCMV IE68 and SCMV IE94
genes (Thomsen et al., 1984; Jeang et al., 1987), although otherwise the protein products have
little amino acid homology. Both HCMV and SCMV IE genes possess several nuclear factor I
(NFI)-binding sites just beyond the enhancer region, but only SCMV contains an additional
cluster of 20 tandemly repeated NFI-binding sites (at - 5 5 0 to - 1300 bp) (Rawlins et al., 1986;
Hennighausen & Fleckenstein, 1986). Potentially, the presence of these tandemly repeated NFIbinding sites, together with multiple consensus CCAAT factor binding sites in the far upstream
5'-regulatory region may help to overcome some form of enhancer repression that occurs in
transformed and teratocarcinoma stem cell lines, thus permitting SCMV IE94 protein synthesis
in a variety of infected cell types in which HCMV IE68 is not expressed.
HCMV and MCMV are both believed to establish latent infections in vivo in lymphoid cells,
most probably of the monocyte/macrophage lineage. MCMV can be reactivated after
stimulation of monocyte differentiation in infected mice (Dutko & Oldstone, 1981) and infection
with clinical isolates has been reported to alter the phenotypic characteristics of monocyte
subpopulations from human PBL in culture (Rice et al., 1984). Our eventual goal is to be able to
characterize these interactions of HCMV biochemically in cell culture. However, until high titre
virus stocks of HCMV isolates with different target cell specificity than the 'fibroblast-adapted'
laboratory strains or cell lines of appropriate susceptibility become available, the ability of
SCMV (Colburn) to express the major IE gene product in human PBL cultures and the
megakaryocyte-like FOS cell line may offer some promise for useful investigations.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
372
R. L. L A F E M I N A AND G. S. H A Y W A R D
We would like to thank the individuals listed in Methods for their gifts of cell lines and virus stocks. W.e thank
Wade Gibson for advice and the use of equipment to perform the tryptic peptide cleavage studies, Alice Irmiere
for assistance with the tryptic peptide studies, Jane Gimigliano for assistance with the immunofluorescence
studies, Pamela Wright for photographic expertise and Phyllis Broughton for assistance in the preparation of this
manuscript. These studies were funded by grants to G.S.H. from the D H E W (CA28473 and CA37314).
REFERENCES
AKRIGG, A., WILKINSON, G. W. G. & DRAM,J. D. (1985). The structure of the major immediate-early genes of human
cytomegalovirus AD169. Virus Research 2, 107-121.
ALWlNE, J. C. (1985). Transient gene expression control: effects of transfected D N A stability and trans-activation
of viral early proteins. Molecular and Cellular Biology 5, 1034-1042.
AMATI, I'. (1985). Polyoma regulatory region: a potential probe for mouse cell differentiation. Cell 43, 561-562.
BOSHART, M., WEBER, F., JAHN, G., DORSCH-HASLER,K., FLECKENSTEIN,F. & SCHAFFNER,W. (1985). A very strong
enhancer is located upstream of an immediate early gene of human cytomegalovirus. Cell 41, 521-530.
CHIRGWIN, J. M., PRZYBYLA, a. E., MACDONALD,R. J. & RUTTER, W. J. (1979). Isolation of biologically active
ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299.
DEMARCHI,J. M. (1981). Human cytomegalovirus D N A : restriction enzyme cleavage maps and map locations for
immediate-early, early and late RNAs. Virology 114, 23-38.
DEMARCHI,J. M. (1983). Nature of the block in the expression of some early virus genes in cells abortively infected
with human cytomegalovirus. Virology 129, 287-297.
DUTKO, F. & OLDSTONE,M. (1981 ). Cytomegalovirus causes latent infection in undifferentiated cells and is activated
by induction of cell differentiation. Journal of Experimental Medicine 154, 1636-1651.
EBERLING, A., KEIL, G. M., KUNST, E. & KOSZINOWSKI,V. R. (1983). Molecular cloning and physical mapping of
murine cytomegalovirus DNA. Journal of Virology 47, 421~433.
EINHORN, L., GADLER, M. & WAHREN, B. (1982). Absorption of purified human cytomegalovirus and induction of
early antigens in different cells. Journal of Medical Virology 10, 225-234.
EVERETT, R. D. (1984). Trans-activation of transcription by herpes virus products: requirement for two HSV-1
immediate-early po!ypeptides for maximum activity. EMBO Journal 3, 3135-3141.
FLECKENSTEIN, B., MULLER, I. & COLLINS, J. (1982). Cloning of the complete human cytomegalovirus genome in
cosmids. Gene 18, 3%46.
GEBALLE,A. P., SPAETE,R. R. & MOCARSKI,E. S. (1986). A cis-acting element within the 5' leader of a cytomegalovirus
fl transcript determines kinetic class. Cell 46, 865-872.
GIBSON,W. (1981). Immediate-early proteins of human cytomegalovirus strains AD169, Davis and Towne differ in
electrophoretic mobility. Virology 112, 350-354.
GmsoN, w. (1983). Protein counterparts of human and simian cytomegaloviruses. Virology 128, 391-406.
GOLDSTEIN, L. C., McDOUGALL,J., HACKMAN,R., MEYERS,J. D., THOMAS,E. D. & NOWlNSKI, R. C. (1982). Monoclonal
antibodies to cytomegalovirus: rapid identification of clinical isolates and preliminary use in diagnosis of
cytomegalovirus pneumonia. Infection and Immunity 38, 273-281.
G6NCZ6L, E., ANDREWS, P. W. & PLOTKIN, S. A. (1984). Cytomegalovirus replicates in differentiated but not in
undifferentiated human embryonal carcinoma cells. Science 224, 159-161.
GONCZOL, E., ANDREWS,P. W. & PLOTKIN,S. A. (1985). Cytomegalovirus infection of human teratocarcinoma cells in
culture. Journal of General Virology 66, 509-515.
GORMAN, C. M., MERLINO, G. T., WILLINGHOUSE,M. D., PASTAN,1. & HOWARD,B. (1982). The Rous carcinoma virus
long terminal repeat is a strong promoter when introduced into a variety of cells by DNA-mediated
transfection. Proceedings of the National Academy of Sciences, U.S.A. 79, 6777-6781.
HErCNIGHAUSEN,L. & FLECKENSTEIN,B. (1986). Nuclear factor I interacts with five D N A elements in the promoter
region of the human cytomegalovirus major immediate-early gene. EMBO Journal 5, 1367-1371.
JEANG, K.-T. (1984). Molecular analysis ofcytomegalovirus (Colburn) gene expression, Ph.D. thesis, Johns Hopkins
University.
JEANG, K.-T. & GIBSON, W. (1980). A cycloheximide-enhanced protein in cytomegalovirus infected cells. Virology
107, 362-374.
JEANG, K.-T. & HAYWARD,G. S. (1983). A cytomegalovirus D N A sequence containing tracts of tandemly repeated
CA dinucleotides hybridizes to highly repetitive dispersed elements in mammalian cell genomes. Molecular
and Cellular Biology 3, 1389-1402.
JEANG, K.-T., CHIN, G. & HAYWARD,G. S. (1982). Characteristics of cytomegalovirus immediate-early genes. I. Nonpermissive rodent cells overproduce the IE94K protein from CMV (Colburn). Virology 121, 393~403.
JEANG, K.-T., CHO, M.-S. & HAYWARD, G. S. (1984). Abundant constitutive expression of the immediate-early
94K protein from CMV (Colburn) in a DNA-transfected mouse cell line. Molecular and Cellular Biology 4,
2214-2223.
JEANG, K.-T., RAWLINS, D. R., ROSENFELD, P. J., SHERO, J. H., KELLY, T. J. & HAYWARD, G. S. (1987). Tandemly
repeated binding sites for cellular nuclear factor I upstream from the strong constitutive immediate early
promoter of cytomegalovirus (Colburn). Journal of Virology 61, 1559-1570.
KEIL, G. M., EBELING-KEIL, A. & KOSZINOWSKI, U. H. (1984). Temporal mapping of murine cytomegalovirus
transcription and mapping of viral R N A synthesized at immediate-early times after infection. Journal of
Virology 50, 784-795.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
C M V expression in non-permissive cells
373
KEIL, G. M., FIBI, M. R. & KOSZINOWSKI,U. H. (1985). Characterization of the major immediate-early polypeptides
encoded by murine cytomegalovirus. Journal of Virology 54, 422-428.
KILPATRICK, B. A. & HUANG, E. S. (1977). Human cytomegalovirus genome: partial denaturation map and
organization of genome sequences. Journal of Virology 24, 261-276.
KNOWLES, W. A. (1976). In vitro cultivation of human cytomegalovirus in thyroid epithelial cells. Archives of
Virology 50, 119-124.
KOSZINOWSKI,U. H., KEIL, G. M., VOLKMER,H., FIBI, M. R., EBELING-KEIL,A. & MUNCH, K. (1986). The 89,000-Mr
murine cytomegalovirus immediate-early protein activates gene transcription. Journal of Virology 58,
59-66.
LAFEMINA, R. & HAYWARD, G. S. (1980). Structural organization of the D N A molecules from human
cytomegalovirus. In Animal Celt Genetics, ICN-UCLA Symposia on Molecular Biology, vol. 18, pp. 39-55.
Edited by R. Jaenisch, B. Fields & C. F. Fox. New York: Academic Press.
LAFEMINA,R. L. & HAYWARD,G. S. (1983). Replicative forms of human cytomegalovirus D N A with joined termini
are found in permissively infected human cells but not in non-permissive Balb/c-3T3 mouse cells. Journal of
General Virology 64, 373-389.
LAFEMINA,R. L. & HAYWARD,G. S. (1986). Constitutive and retinoic acid-inducible expression of cytomegalovirus
immediate-early genes in human teratocarcinoma cells. Journal of Virology 58, 434-440.
LINNEY, E., DAVIS,B., OVERHAUSER,l., CHAO,E. & FAN, H. (1984). Nonfunction o f a Moloney murine leukaemic virus
regulatory sequence in F9 embryonal carcinoma cells. Nature, London 308, 470-472.
MARKS,J. R. & SPECTOR,D. H. (1984). Fusion of the termini of the murine cytomegalovirus genome after infection.
Journal of Virology 52, 24-28.
MARKS,J. R., MERCER, J. A. & SPECTOR,D. H. (1983). Transcription in mouse embryo cells permissively infected by
murine cytomegalovirus. Virology 131, 247-254.
MERCER,J. A., MARKS,J. R. & SPECTOR,D. H. (1983). Molecular cloning and restriction endonuclease mapping of the
murine cytomegalovirus genome (Smith Strain). Virology 129, 94-106.
MIDDLETON, M., REYES, G. R., CIUFO, D. M., BUCHAN,A., MACNAB,J. & HAYWARD,G. S. (1982). Expression of cloned
herpes virus genes. I. Detection of nuclear antigens from herpes simplex virus type 2 inverted repeat regions
in transfected mouse cells. Journal of Virology 43, 1091-1101.
MORGAN, D. A. & BRODSKY, I. (1985). Novel peripheral blood-derived human cell lines with properties of
megakaryocytes. Journal of Cell Biology 100, 565-573.
NELSON, J. A. & GROUDINE,M. (1986). Transcriptional regulation of the human cytomegalovirus major-immediateearly gene is associated with induction of DNase I hypersensitive site. Molecular and Cellular Biology 6, 452461.
O'HARE, P. & HAYWARD,G. S. (1984). Expression of recombinant genes containing herpes simplex virus delayedearly and immediate-early regulatory region and trans-activation of herpesvirus infection. Journal of Virology
52, 522-531.
RAWLINS,D. R., ROSENFELD, P. J., KELLY, T. J., MILMAN, G. R., JEANG, K.-T., HAYWARD,S. D. & HAYWARD,G. S. (1986).
Sequence-specific interactions of cellular nuclear factor 1 and Epstein-Barr virus nuclear antigen with
herpesvirus DNAs. In Cancer Cells, vol. 4, pp. 525-542. Edited by M. Botcham, T. Grodzicker & P. A. Sharp.
New York: Cold Spring Harbor Laboratory.
REITSMA,P. H., ROTHBERG, P. G., ASTRIM,S. N., TRIAL, J., BAR-SHAVIT,Z., HALL, A., TEITELBAUM,S. L. & KAHN, A. J.
(1983). Regulation of myc gene expression in HL60 leukaemia cells by a vitamin D metabolite. Nature,
London 306, 492-494.
RICE, G. P. A., SCHRIER, R. D. & OLDSTONE, i . B. A. (1984). CMV infection of human lymphocytes and monocytes:
virus expression is restricted to immediate-early gene products. Proceedings of the National Academy of
Sciences, U.S.A. 81, 6134-6138.
ROWE, w. P., HARTLEY, J. w., WATERMAN,S. W., TURNER, H. C. & HUEBNER, R. J. (1956). Cytopathogenic agent
resembling human salivary gland virus recovered from tissue culture of human adenoids. Proceedings of the
Society for Experimental Biology and Medicine 92, 418-424.
SCHRIER, R. D., NELSON,J. A. & OLDSTONE,i . B. A. (1985). Detection of human cytomegalovirus in peripheral blood
Iymphocytes in a natural infection. Science 230, i048-i051.
SLEIGH, M. L & LOCKETr, T. J. (1985). SV40 enhancer activation during retinoic acid-induced differentiation o f F 9
embryonal carcinoma cells. EMBO Journal 4, 3831-3837.
SMITH, M. G. (1956). Propagation in tissue cultures of a cytopathogenic virus from human salivary gland virus
(SGV) disease. Proceedings of the Society for Experimental Biology and Medicine 92, 424-430.
SPAETE,R. R. & MOCARSKI,E. S. (1985). Regulation of cytomegalovirus gene expression: ct and fl promoters are transactivated by viral functions in permissive human fibroblasts. Journal of Virology 56, 135-143.
STENBERG, R. i . , THOMSEN,D. R. & STINSKI, M. F. (1984). Structural analysis of the major immediate-early gene of
human cytomegalovirus. Journal of Virology 49, 190-199.
STINSKI, M. F. (1978). Sequence of protein synthesis in cells infected by human cytomegalovirus: early and late
virus-induced polypeptides. Journal of Virology 26, 686-701.
STINSKI,i . F. & ROEHR,T. J. (1985). Activation of the major immediate-early gene of human cytomegalovirus by cisacting elements in the promoter-regulatory sequence and by virus-specific trans-acting components. Journal
of Virology 55, 431-444.
STINSKI, M. F., THOMSEN, D. R., STENBERG, R. M. & GOLDSTEIN, L. C. (1983). Organization and expression of the
immediate-early genes of human cytomegalovirus. Journal of Virology 46, 1-14.
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48
374
R. L. L A F E M I N A AND G. S. H A Y W A R D
STRICKLAND, S., SMITH, K. K. & MAROTrI, K. R. (1980). Hormonal induction of differentiation in teratocarcinoma
stem cells: generation of parietal endoderm by retinoic acid and dibutyryl cAMP. Cell 21, 347-355.
THOMAS, P. S. (1980). Hybridization of denatured R N A and small D N A fragments transferred to nitrocellulose.
Proceedings of the National Academy of Sciences, U.S.A. 77, 5201-5205.
THOMSEN, D. R., STENBERG, R. M., GOINS, W. F. & STINSKI, M. F. (1984). Promoter-regulatory region of the major
immediate-early gene of human cytomegalovirus. Proceedingsof the Notional Academy"of Sciences, U.S.A. 81,
659-663.
VONKA,V., ANISIMOVA,E. & MACEK,i . (1976). Replication of cytomegalovirus in human epithelial diploid cell line.
Archives of Virology 52, 283-296.
WATHEN, M. W. & STINSKI, M. F. (1982). Temporal patterns of human cyt.omegalovirus transcription: mapping the
viral RNAs synthesized at immediate-early, early and late times after infection. Journal of Virology 41,
462-477.
WATHEN, M. W., THOMSEN, D. R. & STINSKI, M. F. (1981). Temporal regulation of human cytomegalovirus
transcription at immediate-early and early times after infection. Journal of Virology 38, 446-459.
WEINER, D. & GIBSON,W. (1983). Phosphorylation, maturational processing and relatedness of CMV strain Colburn
matrix proteins. Virology' 129, 155-169.
WELLER, T. H., MOCAULEY,J. C., CRAIG,J. M. & WIRTH, P. (1957). Isolation ofintranuclear inclusion producing agents
from infants with illnesses resembling cytomegalic inclusion disease. Proceedings of the Society for
Experimental Biology and Medicine 94, 4-12.
ZERBIN1, M., MUSIANI,M. & LAPLACA,i . (1985). Effect of heat shock on Epstein-Barr virus and cytomegalovirus
expression. Journal of General Virology 66, 633-636.
(Received 20 July 1987)
Downloaded from www.microbiologyresearch.org by
IP: 78.47.27.170
On: Tue, 25 Oct 2016 03:31:48