| Journal of General Virology |
| SUMMARY | MAIN TEXT | FOOTNOTES | REFERENCES |
| First posted online 9 May 2001 | SHORT COMMUNICATION |
| Rec 24 January 2001; Acc 27 April 2001 | DOI: 10.1099/vir.0.17649-0 |
Laurent Deleu, Aurora Pujol, Jürg P. F. Nüesch and Jean Rommelaere
Applied Tumor Virology Programme, Abteilung
F0100 and Institut National de la Santé et de la Recherche
Médicale U 375, Deutsches Krebsforschungszentrum, 69120 Heidelberg,
Germany
Nonstructural protein 1 (NS1) of minute virus of mice is involved in viral DNA replication, transcriptional regulation and cytotoxic action in the host cell. Viral DNA replication is dependent on the ability of NS1 to form a homo-oligomer. To investigate whether oligomerization is required for NS1 transcriptional activities, a functionally impaired mutant derivative of NS1 that was able to interact with the wild-type (wt) protein and inhibit its activity in a dominant-negative manner was designed. This mutant provided evidence that transactivation of the parvoviral P38 promoter and transinhibition of a heterologous promoter by NS1 were both affected by the co-expression of the wt and the dominant-negative mutant form of NS1. These results indicate that additional functions of NS1, involved in promoter regulation, require oligomer formation.
Main Text |
Autonomous parvoviruses are small non-enveloped
viruses that infect a variety of animal species, including humans. Their 5
kb genome is a linear single-stranded DNA molecule flanked by palindromic
sequences. The coding sequences of the prototype strain of minute virus of
mice (MVMp) used in the present study is divided into two overlapping
transcription units (Pintel et al., 1983
) that encode two capsid
proteins (VP) and at least four non-structural (NS) proteins (Cotmore
& Tattersall, 1987
). Of these latter polypeptides, NS1, an 83 kDa
nuclear phosphoprotein, is the only polypeptide essential for a productive
infection in all cell types (Cater & Pintel, 1992
; Naeger et
al., 1990
). NS1 is endowed with intrinsic ATPase, site-specific
endonuclease, helicase and DNA-binding activities (Baldauf et al.,
1997
; Christensen et al., 1997
; Cotmore et al., 1995
;
Cotmore & Tattersall, 1998
; Nüesch et al., 1995
; Wilson
et al., 1991
). These various functions are involved in viral DNA
replication and transcription, which allows NS1 to serve both as an
initiator of viral DNA amplification and as a regulator of viral gene
expression. NS1 strongly induces the viral P38 promoter (Rhode &
Richard, 1987
) and activates or inhibits gene expression from heterologous
promoters (Faisst et al., 1993
; Legendre & Rommelaere, 1992
;
Vanacker et al., 1993
). Although the molecular mechanisms involved
in these various effects are still elusive, genetic analysis has shown
that the transcription-regulating domains of NS1 are confined to the
amino- and carboxy-terminal portions of the protein and can be dissociated
from its replicative function (Legendre & Rommelaere, 1992
).
To be fully functional, a number of proteins
involved in DNA metabolism require self-association. This is the case of
most helicases (West, 1996
), many replication and transcription factors
(Baler et al., 1993
; Lazazzera et al., 1996
; Mastrangelo
et al., 1991
), and multifunctional proteins, such as the large tumour
antigen of simian virus 40 (SV40) (San Martin et al., 1997
).
Transcription factors forming oligomers can be inactivated in a
dominant-negative manner by co-expressing a non-functional derivative
capable of associating with the wild-type (wt) protein and forming
inactive oligomers (Smith & Birrer, 1996
). Recently, it was
demonstrated that NS1 is able to form homo-oligomers in an ATP-dependent
manner (Nüesch & Tattersall, 1993
). This self-association is
required for NS1 to fulfill its helicase function and to support viral DNA
replication (Pujol et al., 1997
). In order to study promoter
regulation by NS1, we used this observation to design an approach
involving a dominant-negative mutant. We engineered a mutant form of NS1
that is non-functional with regard to transcriptional activities and
unable to bind to its DNA recognition motif, but can still oligomerize
with full-length NS1. Should such a mutant act in a dominant-negative
manner, this would argue for a role of self-association in the
transcription-regulating functions of NS1. A putative dominant-negative
mutant NS1 derivative,
DT, which is unable to bind to its DNA cognate
motif (deletion of aa 95254) and lacks the acidic
activation domain (deletion of 67 aa from the C terminus), but contains
the oligomerization domain, was designed. A control mutant,
DOT,
which is identical to the
DT mutant but impaired for oligomerization due to a
deletion within the domain required for self-association
(VETTVT-X9-IQT) was also constructed (Fig.
1 A) (Pujol et al., 1997
). All constructs were engineered by
PCR using the wt NS1 DNA sequence and cloned into the cytomegalovirus
(CMV)-driven expression vector pX (Pujol et al., 1997
). The two
different mutant derivatives were expressed at similar levels, as shown by
immuno-blotting of extracts prepared from A9 cells transfected with a
construct harbouring NS1 under the control of the bacteriophage T7
promoter and infected with a recombinant vaccinia virus, vTF7-3,
expressing the bacteriophage T7 RNA polymerase (Fig. 1
B) (Nüesch et al., 1992
). By two-hybrid analysis, the
O
deletion of aa 277309 has been shown to prevent NS1 from
homo-oligomerization (Pujol et al., 1997
).
Fig. 1. Schematic representation and expression
of the wt and mutant forms of NS1. (A) Hatched boxes indicate the
positions of NS1 DNA-binding, oligomerization and C-terminal acidic
activation domains. The regions of NS1 that are deleted in the mutant
derivatives are indicated by open spaces. The vertical line shows the
position of the NLS. Numbers correspond to amino acid residues. (B)
Protein extracts from A9 cells transfected with the indicated constructs
were subjected to immunoblotting using an NS1-specific antiserum.
NS1 exhibits a mainly nuclear localization during
natural MVM infection (Cotmore et al., 1986
) and when produced
independently of other parvoviral proteins or viral DNA in transient
expression systems (Nüesch et al., 1992
; Nüesch &
Tattersall, 1993
). Previously, it has been shown that mutant NS1
polypeptides with an impaired nuclear localization signal (NLS) display
nuclear localization when co-expressed with wt NS1. Plasmids expressing wt
NS1, FLAG-NS1
DT or FLAG-NS1
DOT, under the control of the CMV early promoter
were transfected separately into A9 cells. At 36 h post-transfection,
proteins were detected by double-immunofluorescence using an M2 monoclonal
antibody (Kodak), which recognizes the FLAG epitope, together with the
polyclonal antisera SP8 (Brockhaus et al., 1996
), which recognizes
the C terminus of NS1 absent in the
DT and
DOT
proteins. Immunofluorescence staining was analysed by confocal laser
microscopy. As presented in Fig. 2(A), wt NS was found
mainly in the nucleus of A9 cells, as described previously (Nüesch
et al., 1992
). As expected from the lack of the NLS in the mutant
forms (Fig. 1), the
DT
derivative only displayed cytoplasmic localization, while the
DOT
mutant exhibited uniform distribution between the cytoplasm and the
nucleus (Fig. 2 C). The migration of
DOT
into the nucleus is most likely due to both the small size of this mutant
(55 kDa) and its inability to form higher order oligomers, which may allow
the polypeptide to enter the nucleus independently of an NLS. When A9
cells were co-transfected with pX-NS1wt and pX-FLAG-NS1
DT at a
ratio of 1:1, the
DT mutant co-localized with wt NS1 in the nuclear
compartment (Fig. 2 D). In contrast, A9 cells
co-transfected with
DOT- and wt NS1-expressing vectors showed no
co-localization of mutant
DOT with wt NS1 (Fig. 2 E).
These findings clearly indicate that the ability of
DT to
interact with NS1 is due to the self-association domain, as
DOT,
which lacks this domain, fails to associate with wt NS1.
Fig. 2. Subcellular localization of wt and
mutant forms of NS1 expressed separately or in combination. The
subcellular localization of NS1 polypeptides was determined by double
immunofluorescence staining at 36 h post-transfection of A9 cells with
plasmids expressing wt NS1 (A), NS1
DT (B) or NS1
DOT
(C), wt NS1 and
DT (D), or wt NS and
DOT
(E). In co-transfection experiments (D, E), wt and mutant plasmids were
transfected at a 1:1 ratio. Representative confocal optical sections
through transfected cells are shown. (i) Expression of NS1
DT or
NS1
DOT, as detected with M2 monoclonal antibodies and revealed
with Texas red-conjugated goat anti-mouse antibodies (red). (ii)
Expression of wt NS1, as detected with SP8 antiserum and revealed with
FITC-conjugated goat anti-rabbit antibodies (green). (iii) Merged images
of (i) and (ii) indicate when proteins co-localize (yellow). Pictures were
obtained with a Zeiss 310 confocal laser scanning microscope, with
readings at 488 and 534 nm wavelengths for FITC and Texas red
fluorescence, respectively. Images were obtained with a Mitsubishi
sublimation printer. An oil immersion X63 Plan Apochromat objective was
used.
The parvoviral P38 promoter is activated in the
presence of wt NS1. Transcriptional activation mediated by NS1 is
dependent on site-specific interaction of NS1 with its cognate DNA-binding
motif, which is located within the tar region of the promoter
(Christensen et al., 1995
), as well as NS1 interaction with the
transcription factor SP1 (Krady & Ward, 1995
; Lorson et al.,
1998
). In order to investigate whether oligomerization of NS1 is indeed a
prerequisite for its ability to transactivate the P38 promoter, we
attempted to inhibit the activity of wt NS1 by co-expression with
DT that
was C-terminally fused to the SV40 large tumour antigen NLS allowing migration
to the nucleus as confirmed by immunofluorescence (data not shown). The
plasmid P38-Luc, which contains the luciferase gene under the control of
the MVMp P38 promoter, was used as a reporter to reveal the
transactivation capacity of NS1 in transient expression assays. A9 mouse
fibroblasts were co-transfected with 50 ng of P38-Luc and various amounts
of pX, pX-NS1, pX-
DT-NLS or pX-
DOT-NLS. Luciferase activity was
measured 36 h post-transfection. To exclude any effect of the pX plasmid
on P38 promoter function, luciferase activities from extracts of cells
transfected with P38-Luc in the presence or absence of 400 ng of pX were
compared. Since no differences were found between the two conditions (data
not shown), the total amount of the effector plasmid was maintained at 400
ng per transfection. As illustrated in Fig. 3(A), wt
NS1 was able to transactivate the P38 promoter up to 100-fold above that
of the background level detected with the pX control vector. In contrast,
no significant transactivation was observed with NS1-
DT-NLS
or NS1-
DOT-NLS. Using co-expression of either mutant polypeptide with
wt NS1, we determined whether interaction with
DT
would impair NS1 capacity for transactivation. Mouse A9 cells were
co-transfected with P38-Luc (reporter), pX-NS1wt
(transactivator) and the potential transinhibitor pX-
DT-NLS
(or pX-
DOT-NLS as a negative control) in wt versus mutant ratios of
1:1, 1:3 or 1:6. As illustrated in Fig. 3(B), P38
transactivation by wt NS1 was reduced in cells co-expressing
DT-NLS
in a dose-dependent manner. In contrast, the oligomerization-negative
DOT-NLS
protein failed to suppress wt NS1-induced transactivation, even when
supplied in a sixfold excess over pX-NS1wt. The reduction of wt
NS1-induced P38 activity in the presence of
DT, but
not
DOT, under conditions in which target cells accumulate similar
amount of both types of mutant proteins (Fig. 2),
argues for oligomerization as the mechanism by which the
DT form
exerts its dominant-negative effect. This interpretation is supported by
recent work that shows that NS1 directly binds, in an ATP-dependent
manner, to a promoter P38 region that mediates the NS1 transactivation
response in cis (Christensen et al., 1995
). The
ATP-dependence of NS1-specific binding to DNA can be circumvented in the
presence of antibodies that are likely to cross-link NS1 molecules. These
data, together with the footprinting analysis of NS1DNA interactions
(Christensen et al., 1995
), point to the fact that NS1
homo-oligomerization is involved in specific DNA binding and ensuing
promoter transactivation. This is in keeping with the proposed mode of
action whereby the dominant-negative
DT mutant forms inactive oligomers with
wt NS1.
Fig. 3. Effect of NS1, NS1
DT and
NS1
DOT on reporter gene expression driven by either the P38
promoter (A, B) or the CMV early promoter (C, D). (A, B) Cultures of
1.5x105 A9 cells were co-transfected with 50 ng of the reporter
plasmid P38-Luc and indicated amounts of effector plasmid pX-NS1wt,
pX-
DT-NLS or pX-
DOT-NLS. Effector plasmids were given either
separately (A) or in combination (B). The DNA inoculum was adjusted to 400
ng with the control plasmid pX. Transfected cultures were incubated for
48h and assayed for luciferase activity. The relative light units (RLU)
measured for the P38 promoter-driven reporter plasmid co-transfected with
each effector are shown (A) and the fold induction of the P38 promoter is
shown for the different effectors (B). (C, D) The CMV early
promoter-driven reporter construct pCMV SEAP was used to measure the
transinhibiting effects of wt or mutant NS1-expressing plasmids that were
inoculated either separately (C) or in combination (D). All transfection
mixtures were adjusted to 600 ng of effector plasmid with the empty vector
pX. SEAP activity was measured with the Phospha-Light kit (Tropix),
according to the manufacturer's instructions. The RLU measured for the CMV
promoter-driven reporter plasmid co-transfected with each effector are
shown (C) and the fold inhibition of the CMV promoter is shown for the
different effectors (D). Data are average values from three independent
transfections, each carried out in triplicate.
Besides regulating the viral P38 promoter, NS1 is also able to suppress
transcription from a number of cellular and heterologous viral promoters
(Legendre & Rommelaere, 1992
; Rhode & Richard, 1987
). At present,
it is not clear whether this inhibition results from a specific repression
by NS1 or a mere squelching of transcription factors that interact with
the N- or C-terminal domains of the polypeptides (Legendre &
Rommelaere, 1992
). The dominant-negative mutant approach was used to
investigate whether homo-oligomerization is involved in the
transinhibiting activity of NS1. The early promoter of CMV was chosen as a
target for NS1-dependent transinhibition. A fixed amount (50 ng) of an
expression vector expressing secreted alkaline phosphatase (SEAP) from the
CMV promoter (pCMV SEAP) (Tropix) was co-transfected with increasing
amounts of plasmids expressing either wt NS1 or the mutant derivatives
DT-NLS or
DOT-NLS. The total amount of the effector plasmid
was maintained at 400 ng per transfection for the same reason as that
described for the P38 promoter. As illustrated in Fig.
3(C), wt NS1 suppressed CMV promoter activity by more than fourfold,
whereas both of the deletion mutants were deficient in this respect. When
plasmids expressing wt NS1 and
DT were co-transfected, the mutant
polypeptide exerted a dominant-negative effect on the wt protein, in that
it prevented the latter from transrepressing the CMV early promoter (Fig. 3 D). In contrast, the oligomerization-deficient NS1
DOT mutant form was ineffective in suppressing the
transinhibitory activity of wt NS1. These results indicate that
NS1-mediated transinhibition of the CMV promoter is achieved by a
multimeric form of the polypeptide requiring functionally active subunits.
In conclusion, the NS1
DT
mutant protein proved able to negatively interfere with the capacity of wt
NS1 for both activation and repression of target promoters in a way that
is dependent on the self-association of the NS1 polypeptide. These data
indicate that the transcription-regulating function of NS1 is likely to
require NS1 homo-oligomerization. Further investigations are necessary to
determine whether homo-oligomerization controls the interaction of NS1
with the transcription machinery and/or other transcription
factors.
L. Deleu and A. Pujol contributed equally to this work. We are grateful to Dr Johanna Bridges for the confocal laser scanning microscopy pictures and to Dr Romuald Corbau for the kind gift of P38-Luc plasmid. We acknowledge the help of Dr Jean-Claude Jauniaux and Celina Cziepluch in the establishment of the yeast two-hybrid system. We also thank Dr Jean-Marc Vanacker, Nathalie Salome, François Fuks and Jan Cornelis for helpful discussions. This work was supported by the Commission of the European Communities, the Deutsche KrebshilfeDr Mildred Scheel Stiftung für Krebsforschung and the GermanIsraeli Foundation for Scientific Research and Development. A.P. and L.D. are fellows of the Commission of the European Community.
References |
© 2001 SGM
This article is now available in the August 2001 print issue of JGV (vol. 82, 19291934). The complete issue of the journal may be seen in electronic form on JGV Online.