Download REVIEW ARTICLE Regulation of Expression of the Integrated

J. gen. Virol. (1981), 54, 1-8
1
Printed in Great Britain
REVIEW ARTICLE
Regulation of Expression of the Integrated Retrovirus Genome
By R O B E R T
A. W E I N B E R G 1. AND D A V I D
L. S T E F F E N 2
1Centerfor Cancer Research and Department of Biology, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, U.S_4. and 2Worcester Foundation for
Experimental Biology, 222 Maple Ave, Shrewsbury, Massachusetts 01545, U.S,4.
INTRODUCTION
Retroviruses synthesize a DNA copy of their genome by reverse transcription. This
genome is found integrated into the chromosomal DNA of a cell shortly after infection. The
integrated viral DNA is termed a provirus and serves as a template for the synthesis of viral
mRNA and progeny virion RNA. An important control point in this infectious cycle occurs
at the level of the transcriptional regulation of the provirus. The expression of the provirus
may be controlled by a variety of regulatory mechanisms and this control is the subject of the
present review. Since the molecular mechanisms of this regulation are not well understood.
this review is largely confined to surveying the phenomena of regulation.
Regulation of transcription of endogenous proviruses
The most striking and illustrative examples of transcriptional control of proviruses derive
from those retrovirus genomes which are carried as genetic determinants in the germ lines of
many vertebrates. Although several theories have been proposed for the origin of these
proviruses, accumulating evidence suggests that they are the residues of infections of germ
line tissue which occurred in the ancestors of various modern species (Jaenisch, 19 76; Todaro
et al., 1978). Such a germ line infection has been recapitulated in the laboratory (Jaenisch,
1976). Infection of Balb/c mouse embryo with Moloney murine leukaemia virus (Mo-MuLV)
resulted in a line of mice (Balb/Mo) which genetically transmit an Mo-MuLV provirus. These
mice provide experimental support for the concept of germ line infection being the source of
endogenous proviruses, as well as providing powerful reagents for the study of endogenous
provirus regulation.
The modulation of expression of endogenous retroviral genomes is extremely complex.
The great majority of these proviruses are transcriptionally silent; a few are activated under
certain specified conditions, and others are permanently inactive and never able to specify
transcripts. This lack of activity is perhaps not unexpected, since these proviruses exist as
tolerated endo-parasites in the germ line of the host organism. To the extent that transcription
of a provirus is occasionally allowed, such expression is tolerable as long as it does not
endanger the reproductive fitness of the individual carrying this provirus. Conversely, those
individuals carrying highly expressed proviruses which specify pathogenic virus particles are
likely to be lost from the breeding pool of the species if this expression endangers their
reproductive fitness. Therefore, a strong evolutionary pressure exists to limit the expression of
many endogenous viral genomes.
This pressure might manifest itself in mechanisms which force the rapid excision of recently
integrated proviruses. Direct evidence for such a mechanism is currently lacking. Other
mechanisms may also render proviruses relatively innocuous. Thus, study of a series of
endogenous chicken proviruses indicates extensive deletions of portions of different
genetically transmitted viral genomes (Hughes et al., 1980). These deletions ensure partial or
Downloaded from www.microbiologyresearch.org by
0022-1317/81/0000-4500 $02.00 © 1981 SGMIP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
2
R.A.
WEINBERG
AND
D.
L.
STEFFEN
total inactivation of expression of the endogenous proviruses. One such deleted provirus is
present in a chicken cell known to express only the glycoprotein of an endogenous virus, and
this physical alteration of a provirus probably explains the limited repertoire of endogenous
viral proteins present in this cell line (Hughes et al., 1980; Hayward et al., 1979). An
analogous situation may occur in the case of certain mouse tissues which appear to express
only the glycoprotein genes of endogenous xenotropic proviruses (Lerner et al., 1976).
The great majority of endogenous viruses are rarely, if ever, expressed, and it is unclear
whether simple explanations such as structural deletions can adequately explain their lack of
expression. Estimates of the proportion of the mouse genome which are of retrovirus origin
range from 0-04% (Callahan & Todaro, 1978) to 0.3% (Lueders & Kuff, 1977) and it is
highly unlikely that more than a small number of this vast array is ever expressed under any
circumstance.
Among endogenous proviruses which are non-defective and which can be expressed, great
variability in the pattern of expression is observed. Endogenous proviruses coding for the
murine leukaemia virus, AKV, have been extensively investigated in this regard. The AKR,
C3H, Balb/c and C57BL strains of mice all carry AKV proviruses (Chattopadhyay et al.,
1975). The AKV-1 and AKV-2 proviruses of the AKR mouse are expressed early in
development and, in a permissive background, will produce a lifelong viraemia in the animal.
In contrast, the Balb/c and C57BL proviruses are expressed only late in the life of an animal.
The behaviour of the C3H provirus is intermediate. It is expressed early, but apparently
slightly later or at a lower level than the AKR proviruses, such that the animal mounts a
successful immune response against the virus and does not become viraemic (Ihle & Joseph,
1978a, b; Ihle et al., 1979). These results cannot be explained on the basis of provirus copy
number or genetic background, as the level of expression of a provirus has been shown to
segregate with single proviruses in genetic crosses, and to be unlinked to other parts of the
genetic background.
Differential expression of endogenous and exogenous proviruses
The transcription of endogenous proviruses is generally tightly repressed. For example,
fibroblasts prepared from A K R or Balb/Mo embryos carry copies of their endogenous
proviruses, but release no virus under most conditions of in vitro culture. The Balb/Mo mice
which carry the genetically transmitted Mov-1 provirus appear to allow its transcription,
but only in the target tissues of this virus, the spleen and thymus. This conclusion is based on
the technique of partial digestion of chromatin with pancreatic DNase, which has been
shown to be an assay for transcriptionally active sequences (Weintraub & Groudine, 1976).
The sensitivity of the Mov-1 provirus DNA in spleen and thymus, and the contrasting
insensitivity of this DNA sequence in non-target organs, strongly suggests that transcription
is only permitted when this provirus is carried in a chromosome of a spleen or thymus cell
(Breindl et al., 1980).
The relaxation of transcriptional repression in some cells of these target tissues can lead to
release of small numbers of infectious particles early in the lives of these animals. This in turn
appears to lead to a rapid amplification of infectious centres in the mouse, since the released
virus is able to grow well in many cells in the mouse (Nobis & Jaenisch, 1980). Thus, a few
spleen cells might act to seed a rapidly spreading systemic infection. A similar process
presumably occurs in AKR mice.
Cells which are infected by intercellular spread of the virus chronically release high levels
of virus particles. These cells represent interesting experimental models, since they contain in
their DNA two kinds of MLV proviruses. The first kind are the endogenous, genetically
acquired proviruses which are present in all cells of the mouse. The second kind of provirus is
represented by those 'exogenous' genomes which have been introduced into the mouse cells
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
Review: Integrated retrovirus genome
3
by the recent horizontal spread. These two types of proviruses must have identical structural
sequences since the 'exogenous' provirus derives from an infection by a virus particle of
endogenous origin. The ability of the exogenous provirus to initiate a highly productive
infection in these cells shows that these cells are not intrinsically non-permissive for provirus
expression. This in turn raises the question of the mechanism which prevents endogenous
expression.
Cis-acting control of provirus expression
A model can be drawn in which the differential expression of the two types of provirus is
governed by cis-acting genetic factors which are linked to each of the integrated genomes.
Such a model implies that the lack of expression of the endogenous provirus is not limited by
the absence of a generally permissive intracellular environment.
There is some experimental evidence that the same cell may contain endogenous and
exogenous proviruses at greatly different levels of expression. This comes from examining the
extent to which the dinucleotide cytosine-guanosine (CpG) is methylated at the cytosine
residue. Such methylation has been shown to be characteristic of non-transcribed regions of
DNA (Sutter & Doerfler, 1980; Vardimon et al., 1980; Desrosiers et al., 1979; van der Ploeg
& Flavell, 1980). Cells containing only endogenous proviruses of Rous-associated virus
(RAV) (Humphries et al., 1979) or mouse mammary tumour virus (MMTV) (Cohen, 1980)
have no unmethylated viral sequences. When these cells become infected, begin producing
virus and acquire exogenous proviruses, they acquire unmethylated viral sequences. In recent
experiments, we have shown that, in A K R mouse embryo fibroblasts containing both
endogenous and exogenous AKV proviruses, the endogenous proviruses remain methylated,
whereas the exogenous proviruses are not methylated (D. L. Steffen & R. A. Weinberg,
unpublished results).
Analogous experiments have also been performed on chick cells containing both the
endogenous RAV-O provirus and an exogenously introduced avian sarcoma viral genome of
close homology to the endogenous genome. Using chromatin-DNase sensitivity as a probe,
the authors were able to conclude that an active exogenous provirus co-existed in a cell with
an endogenous viral genome which was inactive both before and after the infection event
(Groudine et al., 1978).
The above experiments would imply the existence of cis-acting regulatory elements which
repress expression of endogenous proviruses, but which have no effect on active exogenous
proviruses in the same cell. Other experiments also point to the existence of cis-acting
repressor elements. DNA preparations containing endogenous proviruses are non-infectious
when tested by the Graham & van der Eb (1973) transfection procedure. In contrast, DNA
preparations containing the corresponding exogenous proviruses are readily infectious
(Cooper & Temin, 1976; Copeland & Cooper, 1979; Cooper & Silverman, 1978). Two
conclusions follow from these transfections. First, the trait of expressibility or lack thereof is
carried in naked DNA and does not require the participation of other chromosomal elements.
Second, the sequences which encode levels of expression remain linked to the proviruses
throughout the manipulations of these experiments. This argues for cis-acting functions.
Another indication of this cis-acting control comes from other work of Jaenisch and
colleagues. As mentioned above, the Balb/Mo mouse which they have studied carries
Mo-MuLV proviruses whose DNA is in a nuclease-sensitive configuration only in organs of
target tissues. By this measurement, the Mov-1 provirus, as they have termed it, is seen to be
transcribed only in spleen and thymus. One might conclude that this favoured state might
derive from a general permissiveness, or non-permissiveness, of target or non-target tissues
for MuLV transcription.
These workers have more recently generated other Balb/Mo mice in which other
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
R. A. WEINBERG AND D. L. STEFFEN
endogenous Me-MuLV proviruses are expressed at high levels in a variety of target and
non-target tissues (Jaenisch et al., 1980; Jaenisch, 1980). This newer result indicates that the
non-target tissues are in fact quite permissive for Mo-MuLV provirus expression. This raises
the question as to why the originally studied Mov-1 provirus is expressed in only one set of
tissues, whereas the more recently derived proviruses are expressed in many developing
organs. The essential difference appears to lie in the integration sites associated either with the
Mov-1 provirus or its more recently derived counterparts. The Mov-1 provirus appears to be
integrated into a chromosomal region which permits expression only in spleen and thymus
but prohibits expression in other organs. The other proviruses appear to be located in
chromosomal regions which are hospitable to expression in a variety of developmental
environments.
In addition to providing the beginnings of a proof of cis-acting control, this work opens a
novel and important area of investigation. Namely, proviruses can be used as probes to
monitor the modulation of expression of specific chromosomal regions during different
periods of development.
Mechanisms of interaction between proviruses and neighbouring cellular elements
These various sources of data suggest that provirus expression is not regulated solely by
diffusible trans-acting factors which might coordinately activate or repress a series of
proviruses scattered throughout different chromosomal sites on the genome. Rather, the
evidence suggests that the activity of a provirus is strongly dependent upon its chromosomal
environment. A hospitable chromosomal environment may be a necessary, although perhaps
not a sufficient pre-condition for expression. Moreover, the permissiveness of this
chromosomal environment may be modulated during differentiation.
It is unclear how a chromosomal environment might act to perturb provirus expression.
One set of experiments suggests that neighbouring cell sequences may act negatively on
provirus expression. Cooper and colleagues have, as mentioned above, shown that
DNA-carrying endogenous provirus is non-infectious (Cooper & Temin, 1976; Copeland &
Cooper, 1979). Extending this observation, they found that mechanical shearing of
DNA-carrying endogenous proviruses now imparted infectivity to the DNA. They suggested
that the shearing was able to break the linkage between the endogenous provirus and a
negatively acting cellular control element (Cooper & Temin, 1976; Cooper & Silverman,
1978).
Unrelated work on the proviruses of MMTV has examined a different mode of interrelation
between host and viral sequences. In this case, the DNase sensitivity of two MMTV
proviruses of exogenous infectious origin has been probed. These two proviruses are found in
two different cell lines and integrated into different chromosomal sites. Measurements of viral
RNA indicate that one provirus is transcriptionally active while the other is apparently
transcriptionally inactive. DNase sensitivity studies of the chromatin carrying the viral
sequence support this conclusion (Yamamoto et al., 1980).
The cellular sequences adjacent to these proviruses have been isolated by molecular
cloning and used as sequence probes for measurements of levels of RNA and chromatin
DNase sensitivity. These cellular probes indicate that none of these cellular sequences is
transcribed in these chromosomal regions prior to the integration event. This strongly suggests
that the viral transcription observed with the active provirus after integration is dependent
upon a transcriptional promoter which is brought in with the provirus. Nuclease sensitivity
studies using these cellular sequence probes indicate that the active provirus was integrated
into a chromosomal region whose DNA was nuclease-sensitive prior to integration.
Conversely, the transcriptionally inactive provirus became established in a chromosomal
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
Review: Integrated retrovirus genome
5
region whose sequences were nuclease-resistant both before and after integration (Yamamoto
et al., 1980).
This work is preliminary in that it characterizes the functioning of only two proviruses.
Nevertheless, it suggests an interesting development in the mechanism of eis-acting control:
that the configuration of chromatin prior to integration is a strong determinant of expression
of any subsequently integrated viral genome.
Trans-acting control mechanisms
There are several striking examples of activation of expression which must be mediated by
diffusible intracellular factors. One example comes from the induction of endogenous viruses
by halogenated pyrimidines and the other from the modulation of MMTV expression by
steroids.
Several lines of mouse cells which do not normally release endogenous viruses can be
induced to do so by incubation in the presence of bromo- or iododeoxyuridine (BrdUrd,
IdUrd) (Lowy et al., 1971; Aaronson et al., 1971). These thymidine analogues are efficiently
incorporated into the DNA of these cells. It might be argued that their ability to turn on
otherwise tightly repressed proviruses results directly from the incorporation of these
analogues into the DNA of the provirus or associated cellular sequences (Teich et aL, 1973).
As a consequence, the structure of the DNA might be perturbed in a fashion which disturbed
the normal interactions between DNA sequences and bound, repressor-like molecules.
Lowy has taken the DNA from IdUrd-treated cells and transfected it in an attempt to see
whether the incorporation of analogue into DNA activated an otherwise non-infectious
provirus (Lowy, 1978). The desired activation was not observed. However, a different
protocol gave striking and unexpected results. When the recipient cells in the transfection
were treated with IdUrd prior to transfection, then the introduction of analogue-free DNA to
these cells resulted in release of virus particles of endogenous origin. The choice of cell lines in
this experiment ensured that the virus released was encoded by the transfected donor DNA
and not by the genome of the recipient cells.
This experiment indicates that exposure to analogue induces a physiological state in the
recipient cell which allows the transcription of the introduced, normally quiescent provirus.
The analogue need not be incorporated into the D N A of the provirus itself. The nature of
these IdUrd-induced, diffusible control factors is unknown. It is of some interest that
cycloheximide treatment is also able to induce expression of some endogenous mouse viruses
(Cabradilla et al., 1976). These authors' speculation that this drug inhibits the synthesis of an
unstable, diffusible repressor remains untested.
An unrelated corpus of work derives from the induction of MMTV expression by steroids.
Application of dexamethasone to cells bearing proper steroid receptors results in a 50- to
1000-fold enhancement in virus expression (Yamamoto & Alberts, 1976; Young et al., 1977;
Yamamoto et al., 1980; Varmus et al., 1979). This process appears to amplify established
patterns of transcription. Thus, proviruses which were transcribed at low constitutive levels in
the absence of the hormone are now expressed at high levels. Proviruses which were
previously totally inactive remain so in the presence of dexamethasone.
These data provide strong evidence for trans-acting modulation of provirus transcription.
Cellular proteins associated with steroid receptors must interact with specific DNA sequences
controlling the intensity of MMTV transcription. These sequences appear to be carried in
with the viral genome and thus are unlikely to be present in the chromosomal region prior to
integration. The above-mentioned work on MMTV indicates that while transcription of
MMTV proviruses is highly steroid-responsive, transcription of the neighbouring cellular
sequences is unaffected in either uninfected or infected cells (Yamamoto et al., 1980). Thus, it
would seem that these neighbouring sequences do not provide a steroid-responsive site to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
R.
(a)
A.
17
WEINBERG
AND
D.
L.
STEFFEN
Parental R N A
I DNA synthesis
Host
Provirus
(b)
Host
I sequence
RNA synthesis
(c)
[']
Progeny transcripts
Fig. 1. Interconversion of retrovirus genomic forms. (a) Reverse transcription of parental virion R N A
into D N A results in a series of unintegrated D N A forms, one of which becomes integrated to create the
provirus, depicted in (b). The provirus, in turn, serves as template for synthesis of progeny RNA,
indicated in (c). As shown here, reverse transcription and genomic rearrangement results in redundant
copies of sequences present originally at the left (5') and right (3') ends of the virion RNA. A small
arrow in (b) indicates the site of the putative transcriptional promoter of the provirus.
modulate expression of an MMTV provirus integrated nearby. Detailed structural analysis of
the MMTV provirus should lead to identification of the implicated steroid-responsive control
sequences.
Molecular models of control
As seen in Fig. 1, the complex transpositions of sequences during provirus synthesis (Gilboa
et al., 1979) lead to a redundancy in the terminal genetic sequences of the viral genome.
Importantly, when RNA and DNA genomes are aligned, as in this figure, it is apparent that
the DNA provirus extends beyond the RNA at both ends. Thus, the (forward) transcription
of progeny RNA begins within a region of viral sequence, several hundred nucleotides away
from the nearest cellular sequence. Similarly, transcription ends within a region of viral
sequence.
This scheme leads to the conclusion that the promoter for initiation of viral transcription is
very likely to be encoded by the viral genome. The functioning of this promoter could well be
affected by trans-acfing factors such as those implicated in MMTV induction.
In addition to the adjacent cellular sequences, a potentially important locus of the
cis-acting control may lie within the group of viral sequences indicated by the small arrow in
Fig. 1 (b). This group probably contains the sequences which influence the efficiency of
transcriptional initiation. Some endogenous proviruses may contain defects in this sequence
while exhibiting an otherwise functional sequence in the remainder of the provirus. Such a
defect could arise during mistakes in reverse transcription which led originally to the
establishment of this provirus. If cellular factors were to allow relief of the transcriptional
block caused by this defect, then the resulting RNA transcript will contain sequences which
have no trace of this lesion, leading possibly to the release of particles capable of inducing
highly productive infections. Thus, these cis-acting lesions in this limited section of viral
sequences (Fig. 1 b) are not transmitted genetically to progeny virus.
One may ask about the nature of the signals which transmit regulatory information to the
provirus. Although specific chromatin structure, as assayed by DNase I sensitivity, has been
shown to correlate with transcriptional activity, this chromatin configuration appears to be a
consequence of underlying molecular mechanisms which govern transcriptional activity. Such
mechanisms are associable with purified DNA, as indicated by transfection experiments.
Purified DNA may impose regulation on the activity of a transfected provirus by virtue of
nucleotlde sequence information
in the
or cellular sequences. However,
an additional and
Downloaded
fromviral
www.microbiologyresearch.org
by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
Review: Integrated retrovirus genome
7
important element of this DNA resides in the methyl groups which are attached to some of
the cytidine residues. These groups may determine the transcriptional activity of a provirus in
vivo or in vitro upon transfection. Such an hypothesis is suggested by recent work of
Eisenman and colleagues (personal communication), as well as work from this laboratory in
which application of the methylation antagonist 5-azacytidine causes induction of
endogenous proviral expression. Granting the potential regulatory importance of such methyl
groups, one is still left without explanations of the mechanisms regulating the methylases
which alter these DNAs.
Currently employed techniques of molecular cloning will aid in the rapid identification of
the cellular and viral sequences which are central to these control processes. Such informatior/
will in turn have important implications on the nature of mechanisms governing the
expression of all cellular genes.
REFERENCES
AARONSON, S., TODARO, G. & SCOLNICK, E. (1971). Induction of murine C-type viruses from clonal lines of
virus-free Balb/3T3 cells. Science 174, 157-159.
BREINDL, M., BACHELOR, L., FAN, H. & JAENISCH, R. (1980). Chromatin conformation of integrated Moloney
leukemia virus D N A sequences in tissues of Balb/Mo mice and in virus-infected cells. Journal of Virology 34,
373-382.
CABRADILLA, C. D., ROBBINS, K. C. & AARONSON, S. A. (1976). Induction of mouse type-C virus by translational
inhibitors: evidence for transcriptional derepression of a specific class of endogenous viruses. Proceedings
of the National Academy of Sciences of the United States of America 73, 4541-4545.
CALLAHAN, R. & TODARO, G. J. (1978). Four major endogenous retrovirus classes each genetically transmitted in
various species of Mus. In Origins of Inbred Mice, pp. 689-713. Edited by H. C. Morse. New York:
Academic Press.
CHATFOPADHYAY, $. K., LOWY, D. W., TEICH, N. M., LEVINE, A. S. & ROWE, W. P. (1975). Qualitative and quantitative
studies of AKR-type murine leukemia virus sequences in mouse D N A . Cold Spring Harbor Symposia on
Quantitative Biology 39, 1085-1101.
COHEN, J. C. (1980). Methylation of milk-borne and genetically transmitted mouse m a m m a r y tumor virus proviral
D N A . Cell 19, 653-662.
COOPER, G. M. & SILVERMAN,L. (1978). Linkage of the endogenous avian leukosis virus genome of virus producing
cells to inhibitory cellular D N A sequences. Cell 15, 573-577.
COOPER, G. M. & TEMIN, H. M. (1976). Lack of infectivity of the endogenous avian leukosis virus-related genes in the
D N A of uninfected chicken cells. Journal of Virology 17, 422-430.
COPELAND, N. G. & COOPER, G. M. (1979). Transfection by exogenous and endogenous murine retrovirus D N A s .
Cell 16, 347-356.
DESROSIERS, R., MULDER, C. & FLECKENSTEIN, B. (1979). Methylation of Herpesvirus saimiri D N A in lymphoid
tumor cell lines. Proceedings of the National Academy of Sciences of the United States of America 76,
3839-3843.
GILBOA, E., MITRA, S. W., GOFF, S. & BALTIMORE,D. (1979). A detailed model of reverse transcription and tests of
crucial aspects. Cell 18, 93-100.
GRAHAM, F. L. & VAN DER EB, A. J. (1973). A new technique for the assay of infectivity of h u m a n adenovirus 5
DNA. Virology 52, 4 5 6 4 6 7 .
GROUDINE, M., DAS, S., NEIMAN, V. & WEINTRAUB,n. (1978). Regulation of expression and chromosomal sub-unit
conformation of avian retrovirus genomes. Cell 14, 865-878.
HAYWARD, W. S., BRAVERMAN, S. B. & ASTRIN, S. M. (1979). Transcriptional products and D N A structure of
endogenous avian proviruses. Cold Spring Harbor Symposia on Quantitative Biology 44, l 111-1121.
HUGHES, S. H., TOYOSHIMA,K., BISHOP, J. M. & VARMUS,H. E. (1980). Organization of the endogenous proviruses of
chickens: implications for origin and expression (in press).
HUMPHRIES, E. H., GLOVER, C., WEISS, R. A. & ARRAND, J. R. (1979). Differences between the endogenous and
exogenous D N A sequences of Rous-assoeiated virus-O. Cell 18, 803-815.
IHLE, J. N. & SOSEPn, D. R. (1978a). Serological and virological analysis o f N I H (NIH X A K R ) mice: evidence for
three A K R murine leukemia virus loci. Virology 87, 287-297.
IHLE, J. N. & JOSEPH, D. R. (1978b). Genetic analysis of the endogenous C3H murine leukemia virus genome:
evidence for one locus unlinked to the endogenous murine leukemia virus genome of C 57B 1/6 mice. Virology
87, 298-306.
IHLE, J. N., JOSEPH, D. R. & DOMOTOR, J. J., JR. (1979). Genetic linkage of C 3 H / H e J and Balb/c endogenous
eeotropic C-type viruses to phosphoglucomutase 1 on chromosome 5. Science 204, 71-73.
JAENISCH, R. (1976). Germ line integration and Mendelian transmission of the exogenous Moloney leukemia virus.
Proceedings of the National Academy of Sciences of the United States of America 73, 1260-1264.
JAENISCH, R. (1980). Retroviruses and embryogenesis: microinjection of Moloney leukemia virus into midgestation
from www.microbiologyresearch.org by
mouse embryos. Cell 19,Downloaded
181-188.
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48
8
R.
A.
WEINBERG
AND
D.
L.
STEFFEN
JAENISCH, R., JAHNER, D. & GROTKOPP, D. ( 1 9 8 0 ) . Derivation of three mouse strains carrying Moloney leukemia
virus in their germ line at different genetic loci. In ICN-UCLA Symposium of Animal Virus Genetics (in
press).
LERNER, R. A., WILSON, C. B., DELVILLANO, B. C., McCONAHEY, P. J. & DIXON, F. J. (1976). Endogenous oncoviral
gene expression in adult and fetal mice: quantitative, histologic and physiologic studies on the major viral
glycoprotein. Journal of Experimental Medicine 143, 151-166.
LOWY, D. R. (1978). Infectious murine leukemia virus from D N A of virus-negative A K R mouse embryo cells.
Proceedings of the National Aeademy of Sciences of the United States of America 75, 5539-5543.
LOWY, D., ROWE, W., TEICH, N. & HARTLEY, J. (1971). Murine leukemia virus: high frequency activation in vitro by
5-iododeoxyuridine and 5-brom0deoxyuridine. Science 174, 155-157.
LUEDERS, K. & KUFF, L. (1977). Sequences associated with intracisternal A particles are reiterated in the mouse
genome. Cell 12, 963-972.
NOBIS, P. & JAENISCH, R. ( 1 9 8 0 ) . Passive immunotherapy: antiserum treatment prevents expression of endogenous
Moloney virus and amplification of proviral DNA in Balb/Mo mice. Proceedings of the NationalAcademy of
Sciences of the United States of America (in press).
SUTTER, D. & DOERFLER, W. ( 1 9 8 0 ) . Methylation of integrated adenovirus type 12 DNA sequences in transformed
cells is inversely correlated with viral gene expression. Proceedings of the National Academy of Sciences of
the United States of America 77, 253-256.
TEICH, N., LOWY, D. R., HARTLEY, J. W. & ROWE, W. P. (1973). Studies on the mechanisms of induction of infectious
murine leukemia virus from A K R mouse embryo cell lines by 5-iododeoxyuridine and 5-bromodeoxyuridine.
Virology 51,163-173.
TODARO, G. J., CALLAHAN, R., SHERR, C. J., BENVENISTE, R. E. & DELARCO, J. E. (1978). Genetically transmitted
viral genes of rodents and primates. In Persistent Viruses, pp. 133-146. Edited by J. E. Stevens, G. J. T~daro
and C. F. Fox. New York: Academic Press.
VAN DER PLOEG, L. H. T. & FLAVELL, R. A. (1980). DNA methylation in the human ?3fl-globin locus in erythroid
and nonerythroid tissues. Cell 19, 947-958.
VARDIMON, L., NEUMANN, R., KUHLMANN, I., SUTTER D. & DOERFLER, W. (1980). DNA methylation and viral gene
expression in adenovirus-transformed and infected cells. Nucleic Acids Research 8, 2461-2473.
VARMUS, H. E., RINGOLD, G. & YAMAMOTO, K. R. (1979). Regulation of mouse mammary tumor virus gene
expression by glucocorticoid hormones. In Monographs on Endocrinology, vol. 12, pp. 254-271. Edited by
J. D. Baxter and G. G. Rousseau. Berlin: Springer-Verlag.
WEINTRAUB~ H, & GROUDINE, M. (1976). Chromosomal subunits in active genes have an altered conformation.
Science 193, 848-856.
YAMAMOTO, K. R. & ALBERTS, B. M. (1976). Steroid receptors: elements of modulation of eukaryotic transcription.
Annual Review of Biochemistry 45, 721-746.
YAMAMOTO, K., CHANDLER, V. L., ROSS, S. R.~ UCKER, D. S., RING, J. C. & FEINSTEIN, S. C. (1980). Integration and
activity of mammary tumor virus genes: regulation by hormone receptors and chromosomal proteins. Cold
Spring Harbor Symposia on Quantitative Biology (in press).
YOUNG, H. A., SHIH, T. Y., SCOLNICK, E. M. & PARKS, W. P. (1977). Steroid induction of mouse mammary tumor
virus: effect upon synthesis and degradation of viral RNA. Journal of Virology 21, 139.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 18:12:48