| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Subtypes In Vivo: Intramuscular Injection of IFN Expression Constructs Reduces Cytomegalovirus Replication1
Department of Microbiology, University of Western Australia, Queen Elizabeth II Medical Center, Nedlands, Perth, Western Australia, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cytokines belong to a multigene family. However, the in
vivo biological functions of each of the IFN- subtypes is unknown.
Recently, we developed an experimental model in which the tibialis
anterior muscles of mice were transfected in situ with naked DNA
plasmids encoding an IFN transgene. Here we use this model to
investigate the in vivo effect of the expression of three murine
IFN- subtypes (A1, A4, and
A9) on murine CMV replication in C57BL/6, BALB/c,
and A/J mice. CMV was shown to replicate in the tibialis anterior
muscles of mice for at least 6 days and induced an inflammatory
infiltrate. However, mice expressing the IFN- transgenes showed a
marked reduction in the peak titers of virus replication, with less
severe inflammation in the muscles compared with control mice that were
inoculated with blank vectors. Moreover, mice expressing the IFN-1
transgene had significantly lower CMV titers in the inoculated muscle
than mice expressing either the IFN-4 or the IFN-9 transgenes.
Furthermore, IFN-/ß receptor knockout mice had markedly higher
levels of CMV replication in the tibialis anterior muscles than the
wild-type parental strain (129/Sv/Ev) following IFN-1 transgene
inoculation, suggesting that the protection observed is due to host
cell-mediated IFN signaling. These data provide the first evidence
indicating that there are in vivo differences in the antiviral efficacy
of the IFN- subtypes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subtypes, a single IFN-ß, IFN-
, and IFN-
(1). Humans express
>15 IFN- (IFNA) subtypes, a single ß
(IFNB), and one
(IFNW) located together on chromosome 9 (2).
Similarly in the mouse, there are >10 IFN- subtypes and one IFN-ß
subtype (3, 4), all of which are located on chromosome 4 (5).
Collectively, these IFNs have diverse biologic functions, including
antiviral (6, 7), antiproliferative (8), and immunomodulatory
activities (9, 10, 11) mediated via signal transduction pathways (12).
Although evidence from in vitro studies suggests differences in the
biologic effects of the subtypes (13, 14, 15, 16), the in vivo significance of
these findings remains obscure. Some researchers have hypothesized that
the multiple IFN- subtypes are a series of evolutionary duplications
of an original gene. It may therefore be the case that these subspecies
have an equivalent function in vivo. Alternatively, it has been
hypothesized that each subtype may have evolved specific functions
(17). Increased knowledge of the activities of the individual subtypes
in vivo will enhance utilization of such subtypes for improved clinical
efficacy over the current usage of either a mixture of IFN subspecies
or a single rIFN subtype.
One approach to study the functions of individual IFN subtypes in vivo
was made using transgenic mice (18, 19, 20, 21). Transgenic mice were created
with either murine IFNA1 or IFNB under the
control of the metallothionein I promoter and unexpectedly expressed
the gene only in the testes (IFN- and -ß) and liver (IFN-ß) upon
induction with cadmium. IFN was only detected in the sera of the
IFN-ß transgenic animals. However, both transgenic lines showed
inhibition of spermatogenesis, resulting in sterility of male mice. No
further experimental work with these animals has been reported. We have
also unsuccessfully attempted to produce transgenic mice expressing the
murine IFN-1 and IFN-4 genes under the control of a mutated
(non-leaky) metallothionein IIA promoter (W.-S. Yeow and M. W.
Beilharz, Department of Microbiology, University of Western Australia,
Perth, Australia, unpublished observations). It is likely that even low
levels of expression of the IFN transgene in the developing embryo are
conditionally lethal.
The advent of naked DNA transfer technology provided an alternative
approach to the manipulation of gene expression in vivo (22, 23). The
persistence of the plasmid allows ongoing expression of the transgene,
and this finding has been pursued for gene therapy (24). Immunization
with plasmids expressing viral genes has induced protective immune
responses, and these plasmids are being developed for improved vaccines
(25, 26). Recently, we have developed a mouse experimental model in
which the tibialis anterior
(TA)3 muscles were
transfected in situ with naked DNA plasmids encoding an IFN transgene
(27). In this model, a mammalian expression vector carrying
MuIFNA9 was injected into the mouse skeletal TA muscle.
Dot-blot analysis showed that the IFNA9 transcript was
present in the DNA-inoculated muscle. Biologic IFN activities were
found in both muscle homogenates and sera of inoculated animals for up
to 2 mo post-DNA injection. This model allows for the first time
investigations into the comparative biologic efficacies of individual
IFN- subtypes. With this experimental model, we have chosen in the
present study to compare the protective levels of a panel of three
mouse IFN- subtypes (namely, MuIFN-1, -4, and -9) expressed
in mice challenged with murine CMV (MCMV).
Type I IFNs play an important role in early defense mechanisms against MCMV infection. The administration of antiserum specific for type I IFN increased the susceptibility of adult mice to MCMV (28). The mouse genotype (H-2 and non-H-2 loci) determines the level of resistance to MCMV infection, with C57BL/6 (H-2b) mice being two- to fourfold more resistant to lethal challenge of virus, and A/J (H-2a) mice being twofold more susceptible to virus infection compared with BALB/c (H-2d) mice (29, 30).
In this paper, we report the in vivo effects of three individual
subtypes of IFN- on MCMV replication in skeletal muscle. Three
murine IFN- genes (A1, A4, and
A9) were individually subcloned into a mammalian
expression vector and were used to inoculate the TA muscles of C57BL/6,
BALB/c, and A/J mice, which were subsequently challenged with MCMV i.m.
(TA). Expression of MuIFNA1, -A4, and
-A9 reduced MCMV replication in vivo to different extents in
each of the mouse strains. The greatest antiviral effects were observed
for MuIFNA1. The effects of IFN were presumably mediated by
host cell signaling events, since IFN-/ß receptor (IFNAR1)
knockout mice could not use the plasmid-expressed IFNs for protection
against virus infection, in contrast to the parental wild-type mouse
strain. Together, these findings provide supporting evidence for the
hypothesis that individual IFN subtypes may have differential biologic
functions in vivo.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Specific pathogen-free female C57BL/6 (B6), A/J, and BALB/c
female mice were purchased from Animal Resources Center (Murdoch,
Western Australia) and housed under minimal disease conditions. All
mice used were between 6 and 8 wk of age. Specific pathogen-free male
IFN-/ß receptor (IFNAR1) knockout mice (derived from M. Aguet,
University of Zurich, Zurich, Switzerland) and the parental wild-type
strain (129/Sv/Ev) were obtained from Drs. Alistair Ramsay and Malcolm
Lawson (University of Western Australia).
Virus
The K181 strain of MCMV was prepared as a salivary gland homogenate from infected 3-wk-old female BALB/c mice and stored in the gas phase of liquid nitrogen. Virus titers in infected mice were quantitated by plaque assay of 20% tissue homogenates (30) and expressed as mean pfu/ml of 100% tissue homogenates.
Anesthetized mice were inoculated with a 25-µl injection volume of MCMV diluted in mouse osmolarity-buffered saline (MOBS) i.m. into each TA muscle. Both muscles were injected with half the specified dose of MCMV per mouse.
Plasmid constructs
The mammalian expression vector, pkCMVint.blank (Fig. 1
A), was donated by
VICAL (San Diego, CA). This vector contains the human CMV
immediate-early (IE-1) gene enhancer/promoter and intron A for
transcription initiation coupled with the SV40 polyadenylation signal.
The construction of pkCMVint.IFNA9 (Fig. 1B), encoding the full-length MuIFNA9
gene, has previously been described in detail (27). The cloning of the
full-length MuIFNA4 gene cassette into pkCMVint involved an
initial subcloning of the EcoRI/HindIII
IFNA4 cassette present in pBR322 (3) into the shuttle vector
pBluescript II KS+ (Stratagene, La Jolla, CA) and then from
the shuttle vector to the PstI/SalI site of
pkCMVint. This construct was designated pkCMVint.IFNA4 (Fig. 1C). The mammalian expression vector construct
containing the full-length MuIFNA1 gene also driven by CMV
IE-1 enhancer/promoter (pCMVint.IFNA1; Fig. 1D) was provided by Dr. Alistair Ramsay (John Curtin
School, Canberra, Australia). This expression vector was obtained from
Drs. James Arthos and James Mullins (31) (Stanford University School of
Medicine, Stanford, CA).
|
Muscle regeneration
To induce muscle regeneration by crush injury (27), mice were anesthetized with 2.5% avertin, fur was removed with depilatory cream from the hind legs, and TA muscles were exposed by surgical incision of the skin. The TA muscles were partially sheared away from the underlying bones with forceps and crushed in three equally spaced locations along the muscles for 3 s using artery forceps. The incisions were sutured, and the mice allowed to recover for 5 days before inoculation with the DNA constructs.
In separate experiments, muscle regeneration was induced by bupivacaine. Anesthetized mice were injected bilaterally in the TA muscles with 25 µl of 0.5% bupivacaine 5 days before inoculation of the DNA constructs.
Mouse injection of DNA constructs
Anesthetized (2.5% avertin) mice were inoculated with 200 µg of DNA plasmids in a 25-µl volume of saline bilaterally in the TA muscles by inserting the needle longitudinally into the muscle. Similarly, control groups of mice were given the blank expression vectors. The expression ability of these plasmids in vivo was confirmed using a similar vector that contained the lacZ gene cassette driven by the CMV IE-1 enhancer/promoter and intron A (pVR-1412, VICAL), and positive staining for ß-galactosidase was found in the DNA-injected TA muscle (9 days post-DNA; data not shown).
Dot-blot analysis
Dot-blot analysis was performed to determine the presence of
plasmid in the transfected muscles. The human CMV IE-1
enhancer/promoter probe (654 bp) was obtained as a SpeI- and
SacII-cut fragment from pkCMVint.blank (nucleotides
272–925), purified from agarose with Geneclean (BIO 101, Inc., La
Jolla, CA), and labeled with [-32P]dCTP using random
primers (Megaprime DNA labeling kit, Amersham International, Aylesbury,
U.K.). The sequence of the probe has no homology with viral sequences
of MCMV (GenBank U68299). DNA extracts were prepared from TA muscles of
mice on day 7 post-DNA injection and dot-blotted onto
Hybond-N+ membranes (Amersham International). Hybridization
to the labeled probe was performed at 60°C for 5 h (6 x
SSC/7% SDS and 100 µg/ml salmon sperm DNA).
IFN bioassay
The TA muscles were collected from mice at 7 days post-DNA
inoculation and treated at pH 2 to remove acid-labile IFNs, and IFN
content was titrated in an in vitro bioassay using encephalomyocarditis
virus-infected L929 cells as previously described (27). The standard
murine IFN-/ß (Lee Biomolecular, San Diego, CA; 1000 IU/ml) was
also titrated in the bioassay as a control.
Histology
Muscles were removed, fixed in 10% formalin-buffered saline, and paraffin embedded. Sections (5 µm) were stained by hematoxylin and eosin and observed microscopically for evidence of pathology. Sections were also stained by the immunoperoxidase method using hyperimmune sera from MCMV-infected mice to reveal virus-infected cells (32). Whole TA muscles from mice injected with the plasmid containing the lacZ gene cassette were fixed with 4% paraformaldehyde and stained for the presence of ß-galactosidase using X-gal, frozen in liquid nitrogen, and then sectioned and counterstained with eosin as previously described (33).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To establish the kinetics of MCMV replication in a skeletal muscle
compartment, we inoculated mice with MCMV bilaterally in the TA
muscles. B6 mice were given a sublethal dose of MCMV (4 x
104 pfu/mouse), and the amount of virus present in the
muscles at 4 h and 1, 2, 3, 4, and 6 days postinfection (p.i.) was
titrated by plaque assay (three mice per time point; Fig. 2). Within 24 h of MCMV inoculation,
the virus titers in the muscles had decreased significantly
(p < 0.05) before increasing steadily with
time. Virus replication peaked on day 3 and was almost cleared by day
6 p.i. The kinetics of MCMV replication in the muscles of a more
susceptible mouse strain (A/J) were also investigated following
inoculation of the same dose of MCMV as that used to infect B6 mice.
There was also a significant decrease (p <
0.005) in the MCMV titer in A/J muscles within 24 h of infection
followed by an increase in the MCMV titer (Fig. 2), such that by day 3
the titer was approximately fourfold higher than that obtained from B6
muscles. Thus, MCMV replicated to markedly higher titers in the
susceptible A/J mice compared with those in the resistant B6 mice, and
no evidence of clearance was seen by day 4 p.i. Immunoperoxidase
staining of TA muscle sections from the above MCMV-infected mice
revealed further evidence of virus-infected cells (data not shown),
providing confirmation of virus replication in the skeletal muscle
tissue.
|
Previously, we have described the construction, delivery,
and expression of IFN-9 in our mouse model (27). Preliminary
experiments were conducted to assess the antiviral activity of
MuIFNA9 in B6 mice following MCMV infection in vivo. B6 mice
(five mice per group) were injected with 200 µg of plasmid constructs
carrying the full-length gene of MuIFNA9 driven by the human
cytomegalovirus (HCMV) IE enhancer/promoter
(pkCMVint.IFNA9; Fig. 1
B).
Control groups of five mice received either no DNA or the blank
expression vector, pkCMVint.blank (Fig. 1A).
Regenerating TA muscles were induced by crush injury 5 days before DNA
injection. Seven days post-gene transfer, MCMV (4 x
104 pfu/mouse) was injected bilaterally into the TA
muscles. These muscles were sampled on both day 3 and day 4 p.i.,
around the time of peak virus replication in B6 mice (see Fig. 2), and
the levels of MCMV present in the muscle homogenates were determined by
plaque assay.
In control mice, in which no DNA was injected into the crush-injured
muscles, MCMV titers obtained on both day 3 and day 4 p.i. were
similar to the MCMV titers determined earlier for the kinetics
experiment (Fig. 2). Since there was a small degree of protection
observed in mice immunized with the blank vector compared with control
mice that were not injected with plasmid DNA, all comparisons of
plasmid constructs were performed with mice receiving the blank
expression vector as the baseline. On day 3 p.i., the time when
peak MCMV titers were found, the mean level of MCMV titer determined in
muscles expressing the MUIFNA9 transgene (3.58 x
103 pfu/ml of TA ± 8.78 x 102) was
significantly lower (p 
0.04) than that
determined in muscles injected with the blank vector (6.43 x
103 pfu/ml of TA ± 1.00 x 103).
Furthermore, on day 4 p.i., virus titers were significantly lower
in mice inoculated with pkCMVint.IFNA9 (2.14 x
103 pfu/ml of TA ± 6.24 x 102;
p 0.02) than in mice inoculated with blank vector
(7.95 x 103 pfu/ml of TA ± 1.95 x
103).
Antiviral activities of MuIFNA1, MuIFNA4, and MuIFNA9 in vivo
We next wished to compare individual murine IFN- subtypes for
their relative antiviral activities in vivo using our mouse model of
MCMV i.m. infection. Mammalian expression plasmids encoding the genes
for murine IFN-4 and IFN-1 were constructed (see Materials
and Methods; Fig. 1, C and D). Three
mouse strains (B6, A/J, and BALB/c, eight mice per group) were injected
with 200 µg of plasmid constructs carrying the full-length gene of
MuIFNA1, MuIFNA4, or MuIFNA9 driven by
the HCMV IE enhancer/promoter (pCMVint.IFNA1,
pkCMVint.IFNA4, and pkCMVint.IFNA9,
respectively). Control groups of mice received the blank expression
vector, pkCMVint.blank (Fig. 1A). The regenerating TA
muscles were induced by bupivacaine inoculation 5 days before DNA
injection. This alternative method to crush injury for the induction of
muscle regeneration was used in all additional experiments. Seven days
after DNA injection, half of the TA pools from three mice (of the
eight) were collected and used for the determination of plasmid in the
transfected muscle by dot-blot analysis and for the estimation of IFN
protein titers by bioassay. All the plasmids (pkCMVint.blank,
pCMVint.IFNA1, pkCMVint.IFNA4, and
pkCMVint.IFNA9) from muscle extracts of DNA-injected
A/J mice showed similar hybridization signal intensities (the
sensitivity limit was 50 pg detected in the 1/10 dilution of DNA
samples; data not shown), suggesting that the plasmids were maintained
at comparable levels in the muscles. DNA extracts were hybridized to
the human CMV IE promoter/enhancer-labeled probe a fragment (657 bp) of
the pkCMVint. vector. A further DNA sample of murine salivary
gland-derived MCMV was included in the dot blot and resulted in no
specific hybridization with the human CMV IE probe. Production of IFN
proteins in the TA muscle from B6 mice injected with the different IFN
constructs was evidenced by bioassay of muscle homogenates; results
ranged from 24 to 150 IU/TA (24, IFNA9; 75,
IFNA4; 150, IFNA1). Whereas B6 mice
injected with the blank vector did not have detectable levels of IFN
titer (1.7 IU/TA).
The remaining five mice (of the eight) were challenged with virus 7 days after DNA injection. MCMV was injected bilaterally into the TA muscles of the mice. B6 mice received 4 x 104 pfu/mouse, while the more susceptible strains (A/J and BALB/c) received a lower virus dose (2 x 104 pfu/mouse). The levels of MCMV present on day 3 p.i. in the muscle homogenates of all the mouse strains investigated were determined by plaque assay.
The bupivacaine-injured muscles of B6 mice that were transfected with
an IFN transgene (A1, A4, or
A9) had significantly reduced
(p 0.0001) MCMV titers compared with the
muscles of control B6 mice given the blank vector (Fig. 3). Histologically, this is supported by
reduced inflammation of the muscle from MuIFNA1 (Fig. 4B), MuIFNA4
(Fig. 4C), and MuIFNA9 (Fig. 4D) transfected B6 mice compared with muscle from
mice given the blank vector (Fig. 4A). Furthermore,
B6 mice transfected with the MuIFNA1 transgene showed a
greater reduction (8.0-fold) in virus replication than the mice
transfected with MuIFNA4 or MuIFNA9 (2.4- and
3.5-fold, respectively).
|
|
0.006)
were determined in mice given the MuIFNA1 transgene than in
mice given blank vector (data not shown). The hematoxylin- and
eosin-stained transverse muscle sections from BALB/c mice showed a more
severe inflammatory response in mice given the blank vector (Fig. 4E) than in mice given pCMVint.IFNA1 (Fig. 4F). Moreover, the MCMV titers from the liver and
spleen of BALB/c mice given pkCMVint.IFNA1 were markedly
lower (13- and 20-fold, respectively) than those in mice transfected
with the blank vector (data not shown).
We were also interested to test whether our model of IFN transgene
delivery was effective when virus infection was further delayed to 2 wk
post-DNA inoculation. MCMV was injected (2 x 104
pfu/mouse) bilaterally into the TA muscles of A/J mice 14 days post-DNA
injection (five mice per group). The levels of MCMV present in the
muscle homogenates, as determined by plaque assay, are shown in Figure 5. Again, each of the MuIFN
subtypes significantly reduced MCMV replication, and
MuIFNA1-transfected mice showed the lowest MCMV titer
compared with that in mice transfected with the other DNA
constructs.
|
We next compared the effect of MuIFNA1 expression on
MCMV replication in mice depleted of functional IFN-/ß receptors
(IFNAR1 knockouts) and in the wild-type parental strain
(129/Sv/Ev). MCMV was injected (2 x 104 pfu/mouse)
bilaterally into the TA muscles of the IFNAR1 knockout and wild-type
mice 7 days post-MuIFNA1 injection (six mice per group). The
mean titers of MCMV present in the muscle homogenates are shown in
Table I. IFNAR1 knockout mice had
19.4-fold higher levels of MCMV replication in the muscles than
wild-type mice following MuIFNA1 inoculation. Moreover, the
MCMV titers from liver and spleen of wild-type mice were markedly lower
(8.2- and 265.3-fold, respectively) than those in the IFNAR1
knockout mice.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subtypes (A1, A4, and
A9). Each of the three IFN- subtypes investigated
could independently reduce MCMV replication in the TA muscle
compartment, where both the transgene and virus were administered. The
primary observation from this study suggests, for the first time, that
there are differences in the biologic efficacies of the IFN-
subtypes in vivo. Indeed, the MCMV titers in TA muscles of B6, A/J, and
BALB/c mice transfected with the MuIFNA1 transgene were
consistently lower than the MCMV titers in TA muscles transfected with
the MuIFNA4 or MuIFNA9 transgene. Furthermore,
MuIFNA1-transfected mice had lower MCMV titers in liver and
spleen, indicating partial protection from virus dissemination to these
organs following i.m. virus challenge. Thus, mice allowed to express
the MuIFNA1 transgene for 1 wk before virus infection
appeared better protected against MCMV challenge than mice given either
MuIFNA4 or MuIFNA9 transgene. In some cases partial protection was afforded by the blank vector via a non-IFN-specific mechanism. This has also been noted in other models of DNA vaccines including studies of influenza virus (C. M. Lawson and S. L. Epstein, National Institute of Allergy and Infectious Diseases and Food and Drug Administration, National Institutes of Health, unpublished observations). DNA sequences in the backbone of the plasmid containing cytosine guanine (CG) dinucleotides in a particular base context may have immunostimulatory effects on the immune system (34). Since we used the data obtained from the control group of mice injected with the blank vector as the baseline for comparison with other groups of mice injected with the vector constructs containing the IFN insert genes, we may be underestimating the degree of protection.
In considering the potential in vivo interactions of several genes and
their effects on IFN transcription, we observed no hybridization of
MCMV genes with the human CMV IE promoter/enhancer. Therefore, we
believe that trans-activation of IFN expression through the
virus would be unlikely and would be expected to occur equally with
each of the individual plasmid constructs, since they have identical
promoters. It is known that a rIFN (human/mouse hybrid) can turn off
transcription of mouse CMV IE genes due to the negative regulation of
IE enhancer sequences (35). However, since IFN does not appear to alter
the mRNA levels of human CMV IE genes (36), putative IFN--responsive
elements might be missing in human CMV IE genes. Our studies suggest
that there is unlikely to be a strong negative regulation of murine IFN
on the HCMV IE genes resulting in the termination of IFN transgene
expression, since newly synthesized IFN from the plasmid does not turn
off detectable plasmid IFN production in the TA, as assessed
kinetically over a 2-mo period (27). Furthermore, the regulatory effect
of murine IFN subtypes (endogenous and plasmid coded) on the human CMV
IE promoter (if any) would be expected to occur equally with each of
the individual plasmid constructs.
Many in vitro studies have shown differences in the biologic activities
with either human IFN- subtypes (37, 38, 39, 40, 41) or mouse IFN- subtypes
(15, 16). However, it has not been possible until now to conduct
comparative studies of the different IFN- subtypes in vivo. Early
attempts to develop an in vivo model for such studies of the biologic
activity of type 1 IFN subtypes have involved making transgenic mice.
The failure of these mice to reproduce and/or to survive embryologic
life makes this an ineffective approach (18, 19, 20) (W.-S. Yeow and
M. W. Beilharz, unpublished observations; B. Williams, unpublished
observation). With the advent of naked DNA technology, we were able to
successfully develop an in vivo model in which the expression of a
specific IFN subtype predominates (27).
The finding of this study, that the MuIFN-1 subtype appeared to be
more protective than the MuIFN-4 or the MuIFN-9 subtype in vivo,
was unexpected. Bioassay of the muscle homogenates taken on day 7 after
plasmid inoculation for type I IFN titers varied from 24 to 150 IU in
C57BL/6 mice. However, it must be emphasized that this read-out of
titers reflects a wide array of in vivo interactions of the IFN subtype
with its environment. The sp. act. of these subtypes in vivo are not
known. However, the specific antiviral activities have been determined
from in vitro experiments using MuIFN-1, -4, and -9 (15, 16)
(S. J. Boyer and M. W. Beilharz, unpublished observations).
The in vitro findings indicated that the antiviral activity of
MuIFN-4 was 2-fold higher than that of MuIFN-1 or MuIFN-9,
that the antiproliferative activity of MuIFN-4 was approximately
10-fold higher than that of MuIFN-1 or MuIFN-9, and that the NK
cell activity of MuIFN-9 was higher than MuIFN-4 and MuIFN-1.
Thus, one would not have predicted that the MuIFN-1 subtype would be
more protective in vivo against a viral infection than the other two
subtypes, especially the MuIFN-4 subtype. We emphasize that the in
vitro data were obtained with the use of a different challenge virus
(encephalomyocarditis virus) and mouse L cells, and is thus limited and
cannot be used for comparisons in the present in vivo study. The
importance of determining functional differences in vivo is highlighted
by the above contrasting data. Indeed, further detailed molecular
analysis beyond our general assumptions, including the idea that the
rate of production of each IFN subtype in situ at a molar level is
equivalent, need to be undertaken, which would allow direct comparisons
and strengthen our claim that there are differential antiviral
differences in vivo among the IFN subtypes.
Our present in vivo study probably reflects the pleiotropic nature of
IFNs in their capacities at inducing multiple biologic effects and
affecting the host immune response (9, 10). Firstly, the principal
antiviral effect of IFN against MCMV is at the level of inhibiting the
transcription of the MCMV IE gene (42). Secondly, IFN was known to be a
negative growth factor for skeletal muscle by inhibiting myogenesis in
vivo (43, 44, 45). Thirdly, type 1 IFNs provide an early and rapid
nonspecific immune response (for example, by augmenting the
proliferation and activation of NK cells) to limit the extent of viral
spread before Ag-specific responses can more fully control the
infection. It is not possible from the present data to establish
whether the protective effect of IFN was mediated by each of the
mechanisms listed above or by a combination of the mechanisms. However,
the MuIFN-1 subtype was more efficient at triggering such a
protective response to MCMV challenge than were the other two subtypes
in vivo. The relationship between the different subtypes and the immune
responses that they trigger is presently under further study in our
laboratory. For example, the importance of augmenting the NK cell
responses is being investigated in NK cell-deficient (beige) mice and
in congenic strains of mice depleted of NK cells.
While the mechanism(s) by which one IFN- subtype can trigger a
better protective response than another subtype against the same viral
challenge is not defined at present, a potential mechanism may involve
priming (46, 47). This phenomenon, defined as an enhanced response to
an IFN-inducing agent following pretreatment with IFN, has been
described both in vitro (46, 47, 48, 49, 50) and in vivo (51). This effect was
shown to be IFN dose dependent (48, 51) and was determined to be a
nonantiviral function of IFN (49). Priming of human leukocytes with a
mixture of human IFN- subtypes was shown to selectively increase the
levels of some IFN- subtypes but not others (50). Although the
priming ability of human IFN-1 was shown to be equivalent to that of
IFN-2 using human peripheral blood leukocytes (52), the priming
abilities of the other individual IFN- subtypes are not known. In
the present study the MuIFN-1 transgene could be a superior primer
over the other IFN- transgenes at enhancing the expression of a
panel of IFN subtypes that act synergistically to induce a protective
response to the subsequent MCMV infection. We are currently in the
process of determining the panel of IFN subtypes that were expressed in
the TA muscles in response to MCMV challenge to determine whether its
composition and/or magnitude has been changed.
In our in vivo study, mice depleted of functional IFN-/ß receptors
(knockout mice) were shown to be poor responders to the protective
effects of MuIFNA1 compared with mice with such receptors.
We expected the parental wild-type strain (129/Sv/Ev), which has a
similar susceptibility level to MCMV infection as A/J mice, to display
virus titers in the same order of magnitude as A/J mice (9 x
104 ± 1 x 104 pfu/ml of TA). Thus, the
titers obtained from 129/Sv/Ev mice injected with the IFNA1
plasmid demonstrate a strong protective effect. On the other hand, the
level of MCMV replication in the knockout mice was significantly higher
than that in all other mouse strains tested with IFN-/ß receptors.
These data clearly establish that the protective effect of IFN is
receptor mediated.
Overall, our study demonstrates for the first time that there are in
vivo differences in the relative biologic efficacies of the different
MuIFN- subtypes during a virus infection. Although many more
questions have now been raised by these data, the mouse model we
described should be capable of allowing experimental resolution of
these issues. We are currently investigating whether the same MuIFN
subtype (1) can induce a better protective immune response in
animals challenged with different viruses. Collectively, these studies
will ultimately provide an improved rationale for effective clinical
use of the IFNs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Cassandra Lawson, Department of Microbiology, University of Western Australia, Queen Elizabeth II Medical Center, Nedlands, Perth, Western Australia 6009, Australia. E-mail address:
3 Abbreviations used in this paper: TA, tibialis anterior; MCMV, murine cytomegalovirus; IFNAR1, the alpha chain of the interferon-/ß receptor; B6, C57BL/6; pfu, plaque-forming unit; MOBS, mouse osmolarity-buffered saline; IE, immediate early; p.i., postinfection; HCMV, human cytomegalovirus.
Received for publication August 25, 1997. Accepted for publication November 25, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
interferon genes. Nucleic Acids Res. 13:805. gene. J. Virol. 56:216. genes to chromosome 4. Gene 26:181.[Medline]
, ß and 
: kinetics of establishment of their antiproliferative effects and the rapid development of resistance to IFNs and ß. J. Biol. Regul. Homeostatic Agents 2:173.
-interferon subtypes. J. Interferon Res. 6:323.[Medline]
4 and -1. J. Interferon Res. 11:9.[Medline]
1 and 4: large difference in antiproliferative effect. Antiviral Res. 19:149.[Medline]
-interferons?. J. Interferon Res. (Special Issue) 11:185.
subtypes in mammalian cell cultures. J. Gen. Virol. 57:233.2 biologically active on mouse cells. EMBO J. 6:591.[Medline]
subtypes and not others. Biochim. Biophys. Acta 887:80.[Medline]
1 or interferon-2 enhances the production of both subtypes simultaneously. J. Interferon Cytokine Res. 15:637.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. N. van den Pol, M. D. Robek, P. K. Ghosh, K. Ozduman, P. Bandi, M. D. Whim, and G. Wollmann Cytomegalovirus InducesInterferon-Stimulated Gene Expression and Is Attenuated by Interferon in the Developing Brain J. Virol., January 1, 2007; 81(1): 332 - 348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Lubyova, M. J. Kellum, A. J. Frisancho, and P. M. Pitha Kaposi's Sarcoma-associated Herpesvirus-encoded vIRF-3 Stimulates the Transcriptional Activity of Cellular IRF-3 and IRF-7 J. Biol. Chem., February 27, 2004; 279(9): 7643 - 7654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Izaguirre, B. J. Barnes, S. Amrute, W.-S. Yeow, N. Megjugorac, J. Dai, D. Feng, E. Chung, P. M. Pitha, and P. Fitzgerald-Bocarsly Comparative analysis of IRF and IFN-alpha expression in human plasmacytoid and monocyte-derived dendritic cells J. Leukoc. Biol., December 1, 2003; 74(6): 1125 - 1138. [Abstract] [Full Text]
|
|
|
|
 |
|
 C. M. U. Hilkens, J. F. Schlaak, and I. M. Kerr Differential Responses to IFN-{alpha} Subtypes in Human T Cells and Dendritic Cells J. Immunol., November 15, 2003; 171(10): 5255 - 5263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Harle, V. Cull, M.-P. Agbaga, R. Silverman, B. R. G. Williams, C. James, and D. J. J. Carr Differential Effect of Murine Alpha/Beta Interferon Transgenes on Antagonization of Herpes Simplex Virus Type 1 Replication J. Virol., June 5, 2002; 76(13): 6558 - 6567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-C. Chin and P. Cresswell Inaugural Article: Viperin (cig5), an IFN-inducible antiviral protein directly induced by human cytomegalovirus PNAS, December 18, 2001; 98(26): 15125 - 15130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lehner, T. Felzmann, K. Clodi, and W. Holter Type I interferons in combination with bacterial stimuli induce apoptosis of monocyte-derived dendritic cells Blood, August 1, 2001; 98(3): 736 - 742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J.J. Carr, P. Härle, and B. M. Gebhardt The Immune Response to Ocular Herpes Simplex Virus Type 1 Infection Experimental Biology and Medicine, May 1, 2001; 226(5): 353 - 366. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Goodbourn, L. Didcock, and R. E. Randall Interferons: cell signalling, immune modulation, antiviral response and virus countermeasures J. Gen. Virol., October 1, 2000; 81(10): 2341 - 2364. [Full Text]
|
|
|
|
|
|
L. Aurisicchio, P. Delmastro, V. Salucci, O. G. Paz, P. Rovere, G. Ciliberto, N. La Monica, and F. Palombo Liver-Specific Alpha 2 Interferon Gene Expression Results in Protection from Induced Hepatitis J. Virol., May 15, 2000; 74(10): 4816 - 4823. [Abstract] [Full Text]
|
|
|
|
|
|
D. M. Miller, Y. Zhang, B. M. Rahill, W. J. Waldman, and D. D. Sedmak Human Cytomegalovirus Inhibits IFN-{alpha}-Stimulated Antiviral and Immunoregulatory Responses by Blocking Multiple Levels of IFN-{alpha} Signal Transduction J. Immunol., May 15, 1999; 162(10): 6107 - 6113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Monini, F. Carlini, M. Stürzl, P. Rimessi, F. Superti, M. Franco, G. Melucci-Vigo, A. Cafaro, D. Goletti, C. Sgadari, et al. Alpha Interferon Inhibits Human Herpesvirus 8 (HHV-8) Reactivation in Primary Effusion Lymphoma Cells and Reduces HHV-8 Load in Cultured Peripheral Blood Mononuclear Cells J. Virol., May 1, 1999; 73(5): 4029 - 4041. [Abstract] [Full Text]
|
|
|
|
|
|
S. Noisakran, I. L. Campbell, and D. J. J. Carr Ectopic Expression of DNA Encoding IFN-{alpha}1 in the Cornea Protects Mice from Herpes Simplex Virus Type 1-Induced Encephalitis J. Immunol., April 1, 1999; 162(7): 4184 - 4190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. M. Horton, D. Anderson, P. Hernandez, K. M. Barnhart, J. A. Norman, and S. E. Parker A gene therapy for cancer using intramuscular injection of plasmid DNA encoding interferon alpha PNAS, February 16, 1999; 96(4): 1553 - 1558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. J. Carr, L. A. Veress, S. Noisakran, and I. L. Campbell Astrocyte-Targeted Expression of IFN-{alpha}1 Protects Mice from Acute Ocular Herpes Simplex Virus Type 1 Infection J. Immunol., November 1, 1998; 161(9): 4859 - 4865. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
