The Journal of Immunology
Multiple Mechanisms Contribute to Impairment of Type 1
Interferon Production during Chronic Lymphocytic
Choriomeningitis Virus Infection of Mice1
Lian Ni Lee,2 Shannon Burke,3 Maria Montoya,4 and Persephone Borrow5
Type 1 IFNs, innate cytokines with important effector and immunomodulatory properties, are rapidly induced in the acute phase
of many virus infections; however, this is generally a transient response that is not sustained during virus persistence. To gain
insight into mechanisms that can contribute to down-regulation of type 1 IFN production during virus persistence, we analyzed
type 1 IFN production during acute and chronic lymphocytic choriomeningitis virus (LCMV) infection. High-level type 1 IFN
production was transiently up-regulated in cells including plasmacytoid and conventional dendritic cells (DCs) following LCMV
infection of mice, but LCMV persistence was associated with only low-level type 1 IFN production. Nonetheless, chronically
infected mice were able to up-regulate type 1 IFN production in response to TLR3, 7, and 9 ligands, albeit less efficiently than
uninfected mice. Splenic DC numbers in mice chronically infected with LCMV were decreased, and the remaining cells exhibited
a reduced response to TLR stimulation. LCMV-infected cell lines efficiently up-regulated type 1 IFN production following TLR
ligation and infection with a DNA virus, but exhibited a defect in type 1 IFN induction following infection with Sendai, an RNA
virus. This block in type 1 IFN production by infected cells, together with abnormalities in DC numbers and functions, likely contribute
to the low-level type 1 IFN production in mice chronically infected with LCMV. Impairment of type 1 IFN production may both promote
virus persistence and impact on host immunocompetence. Understanding the mechanisms involved may assist in development of
strategies for control of virus persistence and superinfection. The Journal of Immunology, 2009, 182: 7178 –7189.
T
ype 1 IFNs, mainly consisting of the IFN-␣ and –␤ subtypes, are innate antiviral cytokines with pleiotropic effects. They are rapidly induced during the early stages of
many viral infections and mediate effector and immunomodulatary
functions that directly help to contain viral replication and induce
and regulate the activation of innate and adaptive responses (1).
The importance of type 1 IFNs in combating viral infections has
been demonstrated in experiments using blocking Abs and in type
1 IFN receptor knock-out mice, where inability to control a number of virus infections was observed (2– 6), and is also indicated by
the fact that many viruses have evolved strategies for blocking
type 1 IFN induction and/or impairing its activity (7–11).
During acute viral infections, there is typically a rapidly induced
transient burst of type 1 IFN production shortly before the peak in
viral replication (1). This type 1 IFN may be derived from multiple
sources. Almost every cell type in the body has the capacity to
produce type 1 IFN when it becomes infected with a virus. This is
The Edward Jenner Institute for Vaccine Research, Compton, Berkshire, United
Kingdom
Received for publication July 31, 2008. Accepted for publication March 25, 2009.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by core funding from the Edward Jenner Institute for
Vaccine Research. P.B. is a Jenner Institute Investigator.
2
Current address: Immunological Memory Group, Nuffield Department of Clinical Medicine, University of Oxford, Peter Medawar Building, South Parks Road,
Oxford, U.K.
3
Current address: Laboratory of Lymphocyte Signaling and Development, Babraham
Institute, Cambridge, Cambridgeshire, U.K.
4
Current address: Centre de Recerca en Sanitat Animal (CReSA), UAB-IRTA, Campus de la Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain.
5
Address correspondence and reprint requests to Dr. Persephone Borrow, Nuffield
Department of Clinical Medicine, University of Oxford, The Jenner Institute, Compton, Newbury, Berkshire, U.K. E-mail address: [email protected]
www.jimmunol.org/cgi/doi/10.4049/jimmunol.0802526
stimulated via pattern recognition receptors (PRR)6 able to detect
the presence of viral components such as dsRNA, DNA, or viral
nucleopcapsids. For example, retinoic acid-inducible gene-1 (RIGI), the prototypic member of the RIG-I-like receptor (RLR) family,
is an RNA-sensing helicase present in the cytoplasm of most cells
that, upon detection of viral ssRNA or dsRNA in the cytoplasm,
induces type 1 IFN production through a signaling pathway dependent upon the mitochondrial protein IFN␤-promoter stimulator-1 (12). The type 1 IFN that is produced can then up-regulate
further type 1 IFN expression via the autologous JAK/STAT pathway to amplify the response made (13). However dendritic cells
(DC), chiefly the plasmacytoid DC (pDC) subset, are thought to
act as the major source of systemic type 1 IFN production in many
acute virus infections (14, 15). pDCs are specialized type 1 IFNproducing cells in which type 1 IFN expression can be induced via
additional pathogen-sensing PRR pathways including those involving TLR7 and 9. These TLRs are expressed in the endocytic
compartment of DCs and can be activated by the presence of viral
components in the external environment, enabling DCs to produce
type 1 IFN in response to a viral infection while not being themselves infected by the virus (16 –19).
During persistent viral infections that are accompanied by ongoing viral replication, there is constant exposure to type 1 IFNinducing stimuli, but this is not associated with continuous production of high levels of type 1 IFN, e.g., no elevation in systemic
type 1 IFN is observed in individuals chronically infected with
6
Abbreviations used in this paper: PRR, pattern recognition receptor; RIG-1, retinoic
acid inducing gene-1; RLR, RIG-I-like receptor; DC, dendritic cell; pDC, plasmacytoid DC; LCMV, lymphocytic choriomeningitis virus; cDC, conventional DC; qRTPCR, real-time quantitative RT-PCR; HPRT, hypoxanthine ribosyl transferase; Ct,
cycle threshold; P/S, penicillin-streptomycin; BM-DC, bone marrow-derived DC;
m.o.i., multiplicity of infection; IRF, interferon regulatory factor; NP, nucleoprotein;
polyIC, polyinosinic:polycytidyl acid; FSDC, fetal mouse skin-derived DC line.
Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00
The Journal of Immunology
HIV, hepatitis C, or hepatitis B virus (8, 20, 21). Possible explanations for this may include down-regulation of type 1 IFN production by host homeostatic mechanisms invoked to prevent detrimental consequences of continuous exposure to high levels of
type 1 IFN (22) and/or impairment of type 1 IFN production as a
consequence of virus persistence, which may be actively induced
by the virus to promote its survival in the host (7, 23, 24). To gain
insight into mechanisms contributing to the regulation of type 1
IFN production in the context of persistent viral replication, we
analyzed the type 1 IFN response in mice chronically infected with
the natural murine pathogen lymphocytic choriomeningitis virus
(LCMV).
LCMV infection of mice is associated with rapid up-regulation of type 1 IFN production (25, 26), which promotes effective innate and adaptive control of the infection (2, 27–29). The
cellular sources of type 1 IFN production during acute LCMV
infection remain controversial, but are thought to include both
pDCs and other cell types (14, 30 –33). LCMV can cause either
an acute infection of mice which is quickly cleared, or establish
a chronic infection during which there is ongoing viral replication accompanied by high systemic viral titers; the outcome
of infection is dependent on both host factors (e.g., age and
immunocompetence) and the infecting virus (isolate, dose, and
route of entry) (reviewed in Ref. 34). Type 1 IFN production in
mice chronically infected with LCMV has been less intensively
studied, although several reports indicate that, as in human persistent virus infections, there is only a low level of type 1 IFN
production during murine chronic LCMV infection (25, 35–37).
In this study, we analyzed type 1 IFN production during the
course of an acute LCMV infection that is rapidly cleared and
an LCMV infection that persists. We found that there was rapid
up-regulation of high levels of type 1 IFN expression following
LCMV infection of mice, but that this was only a transient
response which was rapidly down-regulated, independently of
local/systemic viral titers. LCMV persistence was accompanied
by only low-level type 1 IFN production, despite the presence
of high levels of ongoing viral replication. Nonetheless, chronically infected animals were able to up-regulate type 1 IFN
production in response to exogenous TLR stimuli, although
they did so less efficiently than uninfected animals. Analysis of
mechanisms that may be contributing to the low level of type 1
IFN production in mice chronically infected with LCMV revealed that there was a reduction in the number of splenic pDCs
and conventional DCs (cDCs) in these mice, and that those DCs
which were present exhibited an impaired capacity to up-regulate type 1 IFN production in response to in vitro stimuli. We
also found that LCMV selectively impairs type 1 IFN production in response to itself and to infection with another RNA
virus, Sendai, in cells it infects. Together, these mechanisms
likely help to promote virus persistence, but they also impact on
the immunocompetence of the host, e.g., we showed that chronically infected mice had an impaired capacity to mount a CD8ⴙ
T cell response to an unrelated Ag. The implications of these
findings for the choice of therapeutic strategy that could be used
to control human persistent virus infections and superinfections
that may occur in the context of virus persistence is discussed.
Materials and Methods
Mice
Female C57BL/6 mice 8 –12 wk of age supplied by Charles River Laboratories were used for all in vivo experiments. All animal studies were
conducted in accordance with U.K. Home Office regulations and were approved by the site ethical review committee.
7179
Viral isolates, growth, and titration
Stocks of LCMV Armstrong 53b (LCMV Arm), a clone triple plaque purified from LCMV Arm CA1371 (38) and LCMV Armstrong Clone 13
(LCMV Cl13), a triple plaque purified variant of Arm derived from spleen
cells of an adult BALB/WEHI mouse persistently infected from birth with
LCMV Arm (39), were prepared by growth in BHK cells, and titers were
determined by plaque assay on Vero cells as previously described (38).
Influenza strain A/PR/8/34 was obtained from Charles River SPAFAS.
HSV type 1 strain DM-21 was originally obtained from Dr. Terry Hill
(University of Bristol, Bristol, U.K.) and stocks were prepared and titrated
in BHK cells. Sendai virus Z strain, grown in 10-day old embryonated
chicken eggs, was a kind gift from Dr. John Macauley (Institute for Animal
Health, Compton, Berkshire). Sendai virus RNA levels in infected mice
were determined by real-time quantitative RT-PCR (qRT-PCR) using
primers and a probe specific for Sendai virus and conditions described
previously (40).
LCMV infection of mice
Mice were infected by i.v. inoculation with 5 ⫻ 105 PFU of LCMV Arm
or 5 ⫻ 106 PFU of LCMV Cl13 diluted in 200 ␮l serum-free RPMI 1640
medium (Life Technologies) to generate an acute or chronic LCMV infection, respectively. They were housed in category 3 containment facilities and provided with food and water ad libitum. Animals were sacrificed
at different timepoints postinfection. Spleens were removed, snap-frozen in
liquid nitrogen, and stored at ⫺80°C before analysis. Blood was collected
by cardiac puncture and serum was separated and stored in aliquots at
⫺80°C. Plaque assays were conducted on the spleen and serum of infected
mice to determine infectious virus titers in these compartments.
Quantitation of type 1 IFN subtype mRNA levels by qRT-PCR
Total RNA was purified from snap-frozen spleen samples or cell pellets
using the RNeasy Mini RNA Isolation kit (Qiagen). Snap-frozen spleen
samples were homogenized in 300 ␮l of RLT buffer with 1% 2-ME per 30
mg tissue. Homogenized tissue or cell pellets stored in 300 ␮l of supplemented RLT buffer were loaded onto a Qiashredder column and centrifuged to shear the genomic DNA, and the RNA was isolated using RNeasy
mini column purification, with on-column Dnase 1 cleanup (Qiagen).
Where possible, the purity and concentration of the eluted RNA was determined by UV spectroscopy. Three hundred nanograms of total RNA was
used as template for qRT-PCR with primers (Sigma-Genosys) and probes
(Applied Biosystems) specific for hypoxanthine ribosyl transferase
(HPRT), IFN-␤, IFN-␣4 (41), and all IFN-␣ subtypes except ␣4 (IFN-␣
(non-␣4)) (42) in reverse transcriptase qPCR master mix (Eurogentec).
Samples were run and analyzed in duplicate on an ABI Prism 7700 detection system using the recommended cycling conditions. The average cycle
threshold (Ct) value (i.e., the PCR cycle number at which amplification
products first became detectable) for each IFN type in a given sample was
calculated, and values were normalized by subtraction from the Ct value of
the housekeeping gene (HPRT), to give the ⌬Ct value. Higher ⌬Ct values
thus reflect the presence of increased IFN mRNA levels.
Quantitation of type 1 IFN activity by bioassay
Serum or culture supernatants were assayed for type 1 IFN activity by
measuring their ability to confer resistance to encephalomyocarditis virus
infection upon L929 cells, using a method based on that previously described (43). A standard curve of 2-fold serial dilutions of recombinant
IFN-␣ (HyCult Biotechology) was set up on each plate. The sensitivity of
the assay was ⬃2 log2 (4)U/ml IFN.
Analysis of response of mice chronically infected with LCMV
Cl13 to innate stimuli
Mice infected 3 wk previously with LCMV Cl13 and uninfected control
animals were injected by the i.p. route with one of the following TLR
stimuli in 200 ␮l PBS: 5 ␮g polyinosinic acid: polycytidylic acid (polyIC)
(Sigma-Aldrich), 50 nmoles R848 (InVivoGen), 25 nmoles CpG 1826
(MWG Biotech), or with 800 hemagglutinin units of influenza virus PR8.
Three mice per treatment type from both the LCMV Cl13-infected and
uninfected groups and one untreated uninfected control animal were sacrificed at timepoints between 6 and 24 h poststimulation and blood and
spleens were removed for analysis.
Isolation of splenic DCs
Spleen DCs were obtained using a variation of the method described by
Vremec et al. (44). Nine to ten spleens were pooled and perfused with 200
␮l per spleen of DNase (Sigma-Aldrich) (325KU/ml):Collagenase (1 mg/
7180
ml, type III) (Worthington Biochemical) diluted in RPMI 1640 with 10%
FCS (Invitrogen), 0.1 mM EDTA (pH 7.2), 100U/ml polymyxin B (SigmaAldrich) and 1 ␮g/ml penicillin-streptomycin (P/S) (Institute for Animal
Health Media Supplies). Spleens were digested for 30 min and the homogenate was passed through a 0.45-␮m cell sieve. Cells were resuspended in
Nycoprep (1.07g/ml; Invitrogen), layered on Nycoprep, and spun in a Sorvall RT7plus centrifuge at 2000 ⫻ g for 20 min. DCs were further enriched
using anti-CD11c microbeads (Miltenyi Biotec).
Analysis of splenic DC subsets
Although splenic DCs are frequently identified on the basis of CD11c
expression, during the course of infection with LCMV, this marker is upregulated on many non-DC cell types as a result of activation (45). To enumerate and to isolate total RNA from specific DC populations, positively
selected CD11c⫹ DCs were thus stained with anti-CD11c-allophycocyanin,
anti-B220-FITC, and a mixture of PE-conjugated anti-CD3, CD19, and DX-5
(all Abs from BD Pharmingen), then sorted using a Mo-Flo cell sorter (DakoCytomation) into PE-negative, CD11c⫹ B220⫹ (pDCs), and B220⫺ (cDC)
subsets. Total RNA was isolated from an equivalent number of cells of each
type from each group of mice at each timepoint.
Analysis of the response of CD11c⫹ DCs to in vitro stimulation
with TLR ligands
CD11c⫹ cells isolated from murine spleens and depleted of non-DC
CD11c⫹ lymphocytes as described above were seeded at 1 ⫻ 105 cells per
well into a 96-well U-bottom plate in 100 ␮l of medium containing one of
the following: no stimulus, polyIC (1, 10, or 100 ␮g/ml), R848 (1, 10, or
100 ␮g/ml), or CpG (20, 200, or 2000 pmol/ml). After 24 h stimulation at
37°C, the plate was centrifuged at 1200 rpm for 5 min and the supernatant
collected and stored at ⫺80°C before analysis of its IFN content.
Production of LCMV-infected bone marrow-derived DCs
(BM-DCs)
Bone marrow progenitor cells were isolated from the hind limbs of
C57BL/6 mice and cultured in GM-CSF-supplemented medium to generate
BM-DCs as previously described (46). After 2–3 days in culture, the cells
were infected with LCMV Cl13 at a multiplicity of infection (m.o.i.) of 10
in serum-free medium for 1 h at 37°C. Virus was then removed and the
cells were plated into 10 cm culture dishes at a concentration of 1 ⫻ 106
cells/ml in medium supplemented with 20 ␮g/ml GM-CSF and incubated
for a further 5 days. At different times postinfection, supernatant was collected from the infected cultures and control uninfected cultures and stored
at ⫺80°C for subsequent analysis of type 1 IFN levels. Immunostaining
assays were also performed to determine the proportion of cells infected
with LCMV. At the end of the culture period, both infected and uninfected
cells were stained with Annexin V-FITC (BD Pharmingen) and analyzed
by flow cytometry.
Generation of LCMV-infected cell lines
BALBCl7, a murine fibroblast cell line grown in Eagle’s Minimal Essential
Medium (Life Technologies), 7% FCS, and 1 ␮g/ml P/S, or a fetal mouse
skin-derived DC line (FSDC) (47) grown in RPMI 1640, 10% FCS, 1 ␮g/ml
P/S, and polymyxin B were used. Cells were infected with LCMV Cl13 at a
m.o.i. of either 0.1 or 10 in 500 ␮l serum-free medium for 1 h at 37°C. Virus
was removed and the cells were seeded into 6-well plates at 1 ⫻ 106 cells per
well in 1 ml medium and incubated. At different times postinfection, supernatant was removed and stored at ⫺80°C for analysis. Cell pellets were collected in RNA lysis buffer (Qiagen) and stored at ⫺80°C for RNA extraction.
Replicate wells of infected cells were maintained in culture and transferred into
small flasks when they appeared confluent. They were then passaged twice a
week. Immunostaining assays were performed to confirm that the cells were
chronically infected with LCMV.
Analysis of type 1 IFN induction in LCMV-infected BM-DCs
and cell lines in response to stimulation with TLR ligands or
infection with unrelated viruses
Uninfected BM-DCs, BALBCl7 cells, or FSDC cells and cells previously
infected with LCMV Cl13 were seeded at 1 ⫻ 105 cells per well into
U-bottom 96-well plates (Corning). After an overnight incubation, cells
were stimulated with the TLR ligands polyIC (100 ␮g/ml), R848 (50 ␮g/
ml), CpG 1826 (2 nmol/ml), or medium only for 24 h. Alternatively, cells
were infected with Sendai virus at an m.o.i. of 10 or HSV at an m.o.i. of
5 for 1 h, after which the virus was removed and replaced with growth
medium and the plates incubated until 4 or 18 h postinfection. Following
the incubation period, the plates were centrifuged at 200 ⫻ g for 5 min in
TYPE 1 IFN RESPONSE IN CHRONIC LCMV INFECTION
a Sorvall RT7 centrifuge and cells and supernatant were collected as described above and stored at ⫺80°C before analysis.
In vivo CTL assay
Mice were infected with LCMV Cl13 as described above. Three weeks
later, groups of control uninfected and LCMV Cl13-infected mice were
immunized i.p. with 500 ␮g OVA protein (Sigma-Aldrich) dissolved in
PBS either alone or in the presence of 50 ␮g polyIC. Nine days later,
killing of SIINFEKL-coated target cells was assessed using an in vivo CTL
assay. In brief, splenocytes from naive, syngeneic C57BL/6 mice were
incubated in the presence or absence of 1 nM SIINFEKL peptide for 1 h at
37°C. The cells were then washed extensively with PBS/0.1% BSA. The
target and control non-peptide-pulsed populations were labeled with 2 and
0.2 ␮M CFSE (Molecular Probes), respectively, in PBS/0.1% BSA for 7
min. at 37°C. The reaction was stopped with an excess volume of cold
PBS/5% FCS before washing twice more in serum-free medium. Target and
control populations were counted, mixed at a 1:1 ratio and 2 ⫻ 107 total cells
were injected i.v. into immunized mice. Eighteen hours later, spleens were
removed from recipient mice, cell suspensions prepared and cells were analyzed using a FACSCalibur (Becton Dickinson) using CellQuest software
(Becton Dickinson). The ratios (R) of peptide-treated and control populations
in the spleens were determined. Percentage of cytolysis was calculated using
the formula [1 ⫺ (Runinfected/Rinfected)] ⫻ 100.
Statistical analysis
Statistical analyses were performed on Minitab Release 14 software
(Minitab). Results were analyzed using a 1-sample t test or an ANOVA
using a general linear model with Tukey’s pairwise comparison to determine the statistical significance of differences between groups.
Results
LCMV infection of mice is associated with rapid induction of
type 1 IFNs, but high levels of type 1 IFNs are not sustained in
the context of LCMV persistence
To enable analysis of the type 1 IFN response during acute and
chronic LCMV infection, C57BL/6 mice were infected i.v. with
either a moderate dose of LCMV Armstrong, or a high dose of
LCMV Cl13. As shown in Fig. 1A, both LCMV isolates replicated
in systemic sites including the spleen, where infectious virus was
detectable from day 1 postinfection, but whereas splenic viral titers
peaked at day 2 postinfection in LCMV Arm-infected mice and
infectious viral titers fell below the level of detection between 7
and 21 days postinfection, mice infected with LCMV Cl13 became
chronically infected, with infectious virus remaining detectable in
the spleen and serum for more than 30 days postinfection.
Type 1 IFN production over the course of each of these infections was assessed by quantitation of splenic mRNA levels and
measurement of type 1 IFN activity in the serum. Samples from
uninfected animals were also analyzed in parallel at some timepoints to give insight into baseline levels of type 1 IFN production
in uninfected mice housed in the same animal facility. As shown
in Fig. 1B, very low levels of IFN-␤, IFN-␣4, and non-␣4 subtypes
of IFN-␣ could be detected in the spleen of uninfected mice at all
timepoints analyzed. Infection of mice with either LCMV Arm or
Cl13 was associated with a very rapid (within 6 h) increase in
mRNA levels of each of the type 1 IFN subtypes measured, with
peak type 1 IFN mRNA levels being observed at 12–24 h postinfection. In three independent experiments conducted, peak splenic
type 1 IFN mRNA levels in LCMV Arm-infected mice were found
to be slightly higher than those in LCMV Cl13-infected mice. Notably, splenic type 1 IFN mRNA levels fell sharply between 24 and
48 h postinoculation in mice infected with either LCMV Arm or
Cl13, despite the fact that splenic titers of both viruses were increasing at this time. Between 2 and 7 days postinfection, viral
titers in the mice infected with LCMV Arm and Cl13 diverged
considerably, with LCMV Arm titers dropping markedly while
LCMV Cl13 titers initially continued to increase and then underwent a slight decline, but still remained at ⬎7 log10 pfu/g spleen
The Journal of Immunology
7181
FIGURE 1. Type 1 IFN production is transiently up-regulated in the early stages of LCMV infections that are cleared or go on to persist, but chronic
LCMV infection is associated with only low-level type 1 mRNA up-regulation. A, Titers of infectious virus, as determined by plaque assay, in the spleen
and serum of adult C57BL/6 mice at different timepoints following i.v. infection with 5 ⫻ 105 PFU of LCMV Arm or 5 ⫻ 106 PFU of LCMV Cl13. Each
point represents the mean of results obtained in a group of three animals, with the error bars indicating 1 SD above and below the mean. The horizontal
dotted line represents the limit of detection of the plaque assay; BD indicates that viral titers were below the detection level at the timepoint indicated. B,
Splenic levels of IFN-␤, IFN-␣4, and IFN-␣ (non-␣4) mRNA over time after infection with LCMV Arm or Cl13. Total RNA was isolated from spleens
excised at the indicated timepoints and the levels of mRNAs of different IFN subtype(s) measured by qRT-PCR. The results are expressed as the difference
between the Ct values of the housekeeping gene HPRT and that of the target gene of interest (⌬Ct HPRT-target) in each sample. The values shown at each
timepoint represent the mean of results obtained in three animals tested per group. The error bars indicate 1 SD above and below the mean. n.d. indicates
that no target mRNA was detected in the sample from the LCMV-Arm-infected mice at the timepoint shown. The mean mRNA level in the spleen of
uninfected mice is indicated by the horizontal dashed line, with the shaded area defining 1 SD above and below this. C, Type 1 IFN activity in the serum
of LCMV Arm or Cl13-infected mice or uninfected animals, as determined by bioassay. The values shown are the mean serum IFN-␣/␤ activity detected
in groups of three mice infected with each virus, with the error bars indicating 1 SD above and below the mean. The level of IFN activity in the serum
of a single uninfected animal tested in parallel at each timepoint is also shown. BD indicates that serum IFN activity was below the limit of detection at
the timepoint indicated. The data in all parts of the figure are representative of results from three independent experiments.
on day 7. However splenic type 1 IFN mRNA levels in mice infected with the two viruses were virtually indistinguishable over
this time-frame, undergoing a gradual decline toward the baseline
levels observed in uninfected animals. Thus, although a burst of
type 1 IFN mRNA induction was initially triggered by the presence of infection with either LCMV isolate, splenic type 1 IFN
mRNA levels were subsequently down-modulated with kinetics
quite distinct from those of splenic viral replication.
Analysis of serum type 1 IFN levels also confirmed that the type
1 IFN response induced following infection of mice with LCMV
Arm or LCMV Cl13 exhibited very similar kinetics (Fig. 1C). In
mice infected with either virus, type 1 IFN production was upregulated very rapidly after infection, with levels of activity in the
serum reaching a peak within the first 48 h of infection, before
peak viral titers were reached. Serum type 1 IFN levels fell rapidly
from day 3 onwards in both groups of mice, reaching a very low
level similar to that detected in control uninfected animals by day
7 postinfection.
By day 21 postinfection, when infectious virus was no longer
detectable in the spleen of mice initially infected with LCMV Arm,
type 1 IFN mRNA levels in these animals were lower than those
in control uninfected animals. However, in LCMV Cl13-infected
mice where infection persisted, IFN-␤ and IFN-␣4 mRNA levels
remained slightly elevated (Fig. 1B). In line with this, mice chronically infected with Cl13 were not found to have elevated levels of
type 1 IFN activity in the serum at chronic timepoints (Fig. 1C), or
exhibited a low-level elevation in serum activity (Fig. 2) despite
the presence of high viral titers in the serum.
Together, these observations showed that type 1 IFN production
is rapidly induced following infection of mice with LCMV Arm or
Cl13, but then quickly declines, not due to elimination of the IFNinducing stimulus, but rather as a result of host homeostatic control
7182
TYPE 1 IFN RESPONSE IN CHRONIC LCMV INFECTION
FIGURE 2. Type 1 IFN production by mice chronically infected with LCMV in response to innate stimuli. A, Splenic IFN-␤ mRNA levels in uninfected
mice and mice infected 3 wk previously with LCMV Cl13 12 h after i.p. inoculation with polyIC, CpG DNA, R848, or influenza PR8, as measured by
qRT-PCR. The results are expressed as the difference between the Ct values of the housekeeping gene HPRT and IFN-␤ (⌬Ct HPRT-target). The values
shown are the mean ⌬Ct values from six unstimulated control animals (Unstim) or groups of nine mice inoculated with each stimulus, with the error bars
indicating 1 SE above the mean. B, Type 1 IFN activity in the serum of the same animals at 12 h poststimulation, as measured using an antiviral bioassay.
The values shown are the mean of the results obtained in groups of six unstimulated control animals or nine mice inoculated with each stimulus, with the
error bars indicating 1 SE above the mean.
and/or effects of infection (with either LCMV isolate) on type 1
IFN production. Persistence of LCMV Cl13 was associated with a
slight elevation of IFN-␤ and IFN-␣ mRNA levels in the spleen
although little or no elevation was detected in type 1 IFN activity
in the serum. The mechanism(s) responsible for the initial downmodulation of high level type 1 IFN production in the presence of
ongoing stimulation may thus be sustained during chronic LCMV
infection, or different mechanisms may be involved at later times.
Chronically infected mice retain the capacity to up-regulate type
1 IFN production in response to exogenous stimuli, although the
fold increase in type 1 IFN levels is lower than in uninfected
animals
To provide insight into the nature of the block or defect in type 1
IFN production in mice chronically infected with LCMV Cl13, the
ability of chronically infected mice to respond to a selection of
exogenous type 1 IFN-inducing stimuli was addressed. Uninfected
animals and mice chronically infected with LCMV Cl13 were inoculated i.p. with ligands for TLR3 (polyIC), TLR7 (R848), TLR9
(CpG 1826 DNA), or with influenza PR8, which would activate
TLR7 and/or the intracellular infection sensing pathway (18, 48),
and type 1 IFN mRNA levels in the spleen and type 1 IFN activity
in the serum were monitored.
In uninfected mice, inoculation of the TLR3, 7, and 9 ligands
polyIC, R848 and CpG 1826 DNA induced up-regulation of all
type 1 IFN mRNA subtypes in the spleen, with maximal type 1
IFN mRNA levels being reached by 12 h postinfection and levels
remaining elevated at 24 h poststimulation. Splenic IFN-␤ mRNA
levels in control animals and mice inoculated 12 h previously with
each of the TLR ligands are shown in Fig. 2A; each TLR ligand
elicited a response of approximately equal magnitude, with IFN-␤
levels being ⬃7⌬Ct units higher than those in control mice. Likewise, all three TLR ligands elicited a strong increase in the levels
of bioactive type 1 IFN present in the serum (Fig. 2B). The levels
of type 1 IFN mRNA detected in the spleen and bioactivity detected in the blood of LCMV Cl13-infected mice at 12 h postinoculation with each of the TLR ligands were very similar to those
in uninfected mice responding to the same stimuli (Fig. 2, A and
B). However, the uninfected mice up-regulated type 1 IFN production to a greater extent than the infected animals, as the baseline levels of IFN-␤ transcripts in the spleen of uninfected mice
were lower than those in the spleen of Cl13-infected animals and
low levels of type 1 IFN activity could be detected in the serum of
Cl13-infected but not uninfected mice before stimulation.
Uninfected mice also generated a strong type 1 IFN response
following inoculation of influenza virus, with type 1 IFN mRNA
levels in the spleen and activity levels in the blood at 12 h postinfection (when the response was maximal), being slightly higher
than those reached in mice inoculated with TLR ligands. LCMV
Cl13-infected mice also up-regulated type 1 IFN production in
response to inoculation with influenza virus, although the maximal
levels of splenic type 1 IFN mRNA and serum type 1 IFN activity
recorded in these animals were appreciably lower than those in
uninfected mice (Fig. 2, A and B). Analysis by qRT-PCR using
primers that detected influenza PR8 genomic RNA, i.e., both
virion RNA and genome-sense RNA generated in infected cells,
indicated that less influenza virus was present in the spleens of
LCMV Cl13-infected mice than control influenza-infected animals
at 12 and 24 h postinoculation (data not shown), which may have
been a consequence of reduced virion trapping in the spleen and/or
impaired viral replication in the LCMV Cl13-infected animals. It
was unclear whether the reduced type 1 IFN response elicited in
LCMV Cl13-infected mice following inoculation with influenza
virus was a consequence of impaired responsiveness and/or a
lower magnitude stimulus. Nonetheless, this series of experiments
showed that mice chronically infected with LCMV Cl13 retained
the capacity to up-regulate type 1 IFN production to at least some
extent in response to stimulation with TLR3, 7, or 9 ligands and
inoculation with influenza virus, although the fold-increase in type
1 IFN levels was of a lower magnitude than that in uninfected
animals.
Type 1 IFN is produced by pDCs and cDCs in acute LCMV
infection
DCs have been identified as important producers of type 1 IFNs
during many virus infections, with pDCs being implicated as the
source of much of the type 1 IFN found in serum in a number of
infections (although this remains controversial for LCMV; Ref.
31). To gain insight into the roles of pDCs and cDCs in type 1 IFN
production during the early and late phases of LCMV infection, a
kinetic analysis of type 1 IFN mRNA levels in pDC and cDC
subsets isolated from the spleens of LCMV Arm or Cl13-infected
mice was conducted. The levels of IFN-␤, IFN-␣ 4, and IFN-␣
(non-␣4) mRNA in each subset were determined by qRT-PCR. As
The Journal of Immunology
7183
FIGURE 3. In mice chronically infected with LCMV, splenic DC numbers are reduced, as is the level of type 1 IFN mRNA expressed in the remaining
DCs. A, Type 1 IFN mRNA levels in splenic DC subsets during the acute and chronic phases of LCMV infection. pDCs (CD11c⫹ B220⫹) and cDCs
(CD11c⫹ B220⫺) were isolated by sorting and levels of IFN-␤ mRNA within these DC subpopulations were assessed by qRT-PCR. The values shown at
each timepoint are the mean of results from two independent experiments; the error bars indicate 1 SE above the mean. The type 1 IFN mRNA level at
0 h post infection reflects the maximal ⌬Ct value detected in the uninfected controls from these experiments and n.d. indicates that no target mRNA was
detected in that sample. B, Numbers of pDCs (CD11c⫹ pDCA-1⫹) and cDCs (CD11c⫹ B220⫺) in the spleen during the course of an acute or chronic LCMV
infection. CD11c⫹ cells isolated by positive selection from the spleens of groups of 10 –20 mice infected with LCMV Arm or Cl13 were depleted of non-DC
CD11c⫹ cells, then stained and analyzed by flow cytometry to determine the percentage of pDCs and cDCs within the population. The total number of pDC
and cDC per spleen was calculated. The values shown are the mean of cell numbers obtained in two to six independent experiments, with the error bars
indicating 1 SE above the mean.
levels of all the type 1 IFN transcripts measured changed in parallel with very similar kinetics, only the IFN-␤ data is shown (Fig.
3A). There was no appreciable difference in type 1 IFN mRNA
up-regulation in DCs in the acute phase of infection with LCMV
Arm or Cl13. IFN-␤ transcripts were strongly up-regulated in the
pDC subset within 12 h of infection of mice with either virus. By
7184
day 1 postinfection they had begun to decline; a sharper drop
was observed by day 2, and by day 7 postinfection they had
returned to baseline. The kinetics of type 1 IFN mRNA upregulation in the cDC population were slightly slower, with
some increase in IFN-␤ transcripts being observed by 12 h
postinfection (although not as marked an increase as occurred in
pDCs), maximum levels being reached on day 1 postinfection,
and then transcript levels declining again by day 2 and dropping
further by day 7 postinfection. In mice infected with LCMV
Arm, IFN-␤ mRNA levels in both pDCs and cDCs were below
the limit of detection by 3 wk postinfection, i.e., were lower
than those in DCs from uninfected mice, parallel to the results
of analysis of total splenic mRNA (Fig. 1B). In mice chronically
infected with LCMV Cl13, a low level of IFN-␤ mRNA could
be detected in the pDC subset, while IFN-␤ mRNA levels in
cDCs were below the level of detection. Notably, IFN-␤ mRNA
levels in total splenic tissue of chronically infected mice were
higher than those in the spleen of the uninfected mice, suggesting that a cell type other than pDCs or cDCs may constitute the
main source of type 1 IFN production in mice chronically infected with LCMV Cl13.
These results show that both pDCs and cDCs contribute to
type 1 IFN production during the acute phase of LCMV infection. Type 1 IFN production was induced very rapidly in pDCs
and with somewhat slower kinetics in cDCs following infection
of mice with LCMV Arm or Cl13; however, high levels of type
1 IFN mRNA expression were not sustained in these DC subsets. Instead, the response was quickly down-regulated, and remained low/undetectable in mice chronically infected with
LCMV Cl13, despite constant exposure of the DCs in these
animals to virus.
Changes in splenic DC numbers in LCMV-infected mice
Having demonstrated that type 1 IFNs are produced by DCs in
LCMV-infected mice, additional experiments were conducted
to address changes in splenic pDC and cDC numbers in mice
infected with LCMV Arm and Cl13. As shown in Fig. 3B, the
number of pDCs (CD11c⫹ B220⫹ DCs) did not change appreciably during the first few days following infection with LCMV
Arm, consistent with observations we previously made in a
study of mice infected i.p. with this virus (32). However, at 3
wk postinfection, the number of pDCs in the spleen was reduced compared with that in uninfected mice. By contrast, as
we previously observed in mice infected i.p. with LCMV Arm
(32), the number of cDCs (CD11c⫹ B220⫺) in the spleen was
reduced by day 2 postinfection, but had returned to normal by
3 wk postinfection.
The number of pDCs in the spleens of LCMV Cl13-infected
mice did not change appreciably within the first 24 h postinfection, but was reduced by day 2 postinfection and was still
more profoundly decreased in the chronic phase of infection.
Splenic cDC numbers in Cl13-infected mice were also reduced
by day 2 postinfection and again remained lower than cDC
numbers in uninfected mice during chronic infection, although
their numbers were not as severely depressed at 3 wk postinfection as those of the pDC subset. These results show that
changes in splenic DC numbers took place as early as day 2
following infection with Arm or Cl13 and that the number of
splenic pDCs and cDCs remained depressed in Cl13-infected
mice during the chronic phase. Notably, phenotypic analysis
indicated that the low numbers of pDCs and cDCs that were
present in the spleen of mice chronically infected with Cl13
exhibited an activated phenotype, expressing elevated levels of
MHC class I, CD86, CD80, and CD40 (data not shown).
TYPE 1 IFN RESPONSE IN CHRONIC LCMV INFECTION
FIGURE 4. A reduced type 1 IFN response is observed after ex vivo
stimulation of DCs from mice chronically infected with LCMV Cl13
with TLR ligands. CD11c⫹ DCs isolated by positive selection and depletion of contaminating CD11c⫹ lymphocytes from uninfected mice
and mice chronically infected with LCMV Cl13 were cultured overnight in medium alone (Unstim) or with polyIC (10 ␮g/ml), R848 (10
␮g/ml) or CpG 1826 DNA (200 pmol/ml). Type 1 IFN activity in the
supernatant was measured by bioassay. The data shown are the mean of
results obtained in three independent experiments using cells pooled
from groups of three to ten mice; the error bars indicate 1 SE above the
mean. Statistical analysis using a one-sample t test showed a significant
difference (p ⬍ 0.05) between the levels of type 1 IFN activity produced
by DCs from uninfected and Cl13-infected mice in response to stimulation with each of the TLR ligands tested.
Splenic DCs isolated from mice chronically infected with LCMV
Cl13 exhibit a reduced response to in vitro stimulation with
TLR3, 7, and 9 ligands
The observation that DC numbers in mice chronically infected
with LCMV Cl13 were reduced suggested that this may contribute
to the impairment in type 1 IFN production in these animals. Additional experiments were conducted to investigate whether the
existing DCs also exhibited an impaired capacity to respond to
TLR stimuli. CD11c⫹ cells were isolated from the spleens of uninfected mice and mice chronically infected with LCMV Cl13,
depleted of non-DC CD11c⫹ subsets, and stimulated with polyIC,
R848, or CpG 1826 DNA, then type 1 IFN activity in the supernatant was analyzed after an overnight incubation (Fig. 4).
CD11c⫹ DCs from Cl13-infected mice were found to produce type
1 IFN in response to all the TLR stimuli tested, although the levels
produced by these cells were lower than those made by DCs from
uninfected mice. Statistical analysis using a one-sample t test
showed that there was a significant difference between levels of
type 1 IFN produced by DCs from uninfected and LCMV-infected
mice in response to stimulation with polyIC ( p ⫽ 0.03), R848
( p ⫽ 0.05), and CpG DNA ( p ⫽ 0.03). Notably, the difference
between the ability of DCs from uninfected and LCMV Cl13infected mice to produce type 1 IFN in response to TLR ligation
was most pronounced at suboptimal TLR ligand concentrations
(supplementary Fig. 1).7 Both the reduction in DC numbers and
the impaired capacity of DCs to produce type 1 IFN in response to
TLR stimulation may thus contribute to the impairment in type 1
IFN production in mice chronically infected with LCMV Cl13.
Cell lines chronically infected with LCMV do not continuously
produce type 1 IFN, but chronically infected DC cell lines
produce type 1 IFN efficiently in response to certain stimuli
To gain insight into mechanism(s) that may underlie the impairment observed in the ability of DCs from mice chronically infected
with LCMV to produce type 1 IFN in response to TLR stimuli, the
7
The online version of this article contains supplementary material.
The Journal of Immunology
7185
FIGURE 5. FSDC cells chronically infected with LCMV Cl13 show an impaired capacity to up-regulate type 1 IFN production in response to infection
with Sendai virus. A, Type 1 IFN mRNA up-regulation in FSDC cells persistently infected with LCMV in response to infection with Sendai virus or HSV.
Uninfected FSDC cells and cells infected ⬎30 days previously with LCMV were left unstimulated (Unstim) or were infected with Sendai virus or HSV.
At 4 h postinfection, cells were collected, total RNA was extracted and type 1 IFN mRNA levels were determined by qRT-PCR. The results are expressed
as the difference between the Ct values of the housekeeping gene HPRT and IFN-␤ (⌬Ct HPRT-target). n.d. indicates that target mRNA was not detected
in the sample. The results shown are representative of findings made in one of two independent experiments (hence there are no error bars). B, Sendai virus
RNA levels, as determined by qRT-PCR, in uninfected and Cl13-infected FSDC cells at 4 and 18 h after infection with Sendai virus. The results are
expressed as in A. C, Type 1 IFN activity in the supernatant of uninfected or Cl13-infected FSDC stimulated with TLR ligands or infected with Sendai virus
or HSV. The supernatant was collected at 18 –24 h posttreatment and type 1 IFN activity was measured by encephalomyocarditis virus bioassay. The results
shown in B and C are the mean of values obtained in two independent experiments; the error bars indicate 1 SE above the mean. Statistical analysis using
ANOVA’s general linear model was performed and a statistically significant difference was observed between type 1 IFN levels in the supernatant of
uninfected and Cl13-infected cells after infection with Sendai virus (p ⬍ 0.05).
effects of chronic LCMV infection on the type 1 IFN response
were studied in vitro. One series of experiments was conducted
using BM-DCs. BM-DCs were generated by in vitro culture of
bone marrow progenitor cells in GM-CSF-containing medium. Infection of these cells with LCMV Cl13 on day 2 (or day 5, data not
shown) of the culture period resulted in viral infection of the majority of cells, as evidenced by staining with an Ab against the viral
nucleoprotein (NP) (supplementary Fig. 2A). In line with observations made in vivo, infection of these cultures was associated
with induction of type 1 IFN production, with supernatant levels of
type 1 IFN activity peaking at 2–3 days postinfection (supplementary Fig. 2B). This was accompanied by up-regulation of activation
markers including MHC class II on the BM-DCs, suggesting that
they were undergoing maturation (data not shown). However, by
day 5 postinfection, the infected cells were also strongly positive
for Annexin V (supplementary Fig. 2C), indicating that they were
entering an apoptotic state. The poor in vitro survival of these cells
meant that efforts to use BM-DCs in experiments to address the
ability of LCMV-infected DCs to produce type 1 IFN in response
to TLR ligation or infection with other viruses were unsuccessful.
Further work was thus conducted using an immortalized DC
line, FSDC.
Initial experiments showed that infection of FSDC cells, and
also other cell lines including the murine fibroblast cell line BALBCl7, with LCMV Cl13 resulted in generation of a fully viable
chronically infected population of cells, the majority of which
were infected with LCMV as judged by immunostaining with an
LCMV NP-specific Ab (data not shown). Type 1 IFN production
by both FSDC cells and BALBCl7 cells was assessed at different
times following infection by bioassay. Interpretation of results was
complicated by the fact that type 1 IFN production by uninfected
FSDC cells fluctuated somewhat over time, perhaps reflecting a
response to manipulation or medium change during passage and/or
changes in cell confluence. However, while little elevation in supernatant type 1 IFN levels was observed following LCMV infection of BALBCl7 cells, LCMV infection of FSDC cells was found
to be associated with a transient elevation in supernatant type 1
IFN levels that peaked ⬃2 days postinfection (data not shown).
Additional experiments were conducted to address whether
FSDC cells infected with LCMV were able to produce type 1 IFNs
in response to infection with unrelated viruses or stimulation with
TLR ligands. FSDC cells which had been infected at least 30 days
previously with LCMV were infected with a second virus, either
Sendai (an RNA virus) or HSV (a DNA virus), and the type 1 IFN
7186
response induced by the second infection was measured 18 h later.
Uninfected cells responded to infection with either Sendai virus or
HSV by up-regulating expression of IFN-␤, IFN-␣4, and other
IFN-␣ subtypes, with infection by HSV eliciting a stronger response than infection with Sendai (Fig. 5A). In cells chronically
infected with LCMV, HSV also induced a strong up-regulation of
all type 1 IFN subtypes, although the levels produced were slightly
lower than those made by similarly exposed non-LCMV-infected
cells. However, when cells chronically infected with LCMV were
infected with Sendai virus, although there was some up-regulation
of all subtypes of type 1 IFN, the levels produced were lower at
18 h postinfection than those made by Sendai-infected control cells
for all subtypes measured.
To exclude the possibility that the chronically infected cells had
a pre-existing antiviral state and so were resistant to a further infection with Sendai, qRT-PCR assays were used to quantify Sendai
virus NP RNA transcripts in uninfected and chronically infected
cells at different timepoints following infection with Sendai virus.
The results indicated that both uninfected and LCMV-infected
cells were infected equally well by Sendai virus (Fig. 5B). Equivalent levels of Sendai virus NP RNA were detected in both cell
populations at 4 h postinfection and had increased to a similar
extent by 18 h postinfection, indicative of viral replication occurring in a parallel fashion in both control and LCMV Cl13-infected
cells. The reduced up-regulation of type 1 IFN observed when
LCMV-infected cells were infected with Sendai virus thus reflected an impairment in their ability to respond to this stimulus.
These observations at the IFN transcript level were confirmed
by analysis of type 1 IFN activity in the supernatant of cells at 18 h
postinfection (Fig. 5C). In uninfected cells, high levels of type 1
IFN activity were produced in response to infection with Sendai or
HSV (⬃211 U/ml), indicating that both viruses stimulated strong
type 1 IFN responses. However, whereas the levels of type 1 IFN
secreted by uninfected and LCMV-infected cells upon infection
with HSV were similar, the levels of type 1 IFN detected in the
supernatant of chronically infected cells infected with Sendai virus
were significantly lower than those produced by uninfected control
cells in response to infection with this virus ( p ⫽ 0.0013 assessed
by ANOVA using a general linear model). Similar observations
were also made in experiments conducted using the fibroblast cell
line BALBCl7 (data not shown). Chronically infected cells thus
were able to up-regulate type 1 IFN production in response to
infection with the DNA virus HSV, but their ability to respond to
infection by an RNA virus, Sendai, was impaired.
There are several ways in which cells can sense the presence of
viral RNA, including through TLR3 or TLR7, which recognize
double stranded or single stranded RNA, respectively, and via a
cytoplasmic pathway initiated by activation of helicases such as
RIG-I or MDA5 (49). To assess the ability of cells chronically
infected with LCMV Cl13 to respond to TLR stimuli, LCMVinfected FSDC cells were stimulated with polyIC, R848, or
CpG1826. As shown in Fig. 5C, there was no difference in the
level of type 1 IFN induced by the synthetic TLR ligands in uninfected and chronically infected FSDC cells, indicating that the
TLR 3, 7, and 9 pathways were not impaired in cells chronically
infected with LCMV. Together, these results indicate that LCMV
infection does not directly impair the ability of DCs to up-regulate
type 1 IFN production in response to TLR ligation or infection
with a DNA virus, HSV, but suggest that there could be an impairment in the response to stimuli that trigger type 1 IFN activation of cytoplasmic RNA helicases such as RIG-I. The latter observation is consistent with the recent finding that LCMV impairs
nuclear translocation of IFN regulatory factor (IRF)-3 in the cells
TYPE 1 IFN RESPONSE IN CHRONIC LCMV INFECTION
FIGURE 6. Mice chronically infected with LCMV demonstrate a reduced capacity to cross-prime a CTL response against a soluble Ag inoculated with polyIC. Uninfected mice and mice infected three weeks previously with LCMV Cl13 were given a single i.p. immunization with either
soluble OVA protein or OVA plus polyIC. Nine days later an in vivo CTL
assay was performed to examine the killing of SIINFEKL-coated target
cells. The symbols represent the percent specific lysis of target cells in
individual animals, with the horizontal lines indicating the mean percent
specific lysis observed in each group of mice. The results shown are representative of findings made in two independent experiments.
it infects to block the production of type 1 IFN that its own nucleic
acids would otherwise elicit (50).
Chronically infected mice mount a CTL response of reduced
magnitude following immunization with OVA
Type 1 IFNs possess adjuvant properties, as demonstrated in the
OVA-immunization model, where robust CTL responses against
the soluble OVA Ag only develop when the Ag is co-administered
with adjuvants such as type 1 IFNs or a type 1 IFN-inducing stimulus such as polyIC (51). A final series of experiments were conducted using this model system to determine whether the diminished capacity of mice chronically infected with LCMV to produce
type 1 IFNs in response to type 1 IFN-inducing stimuli affects their
ability to mount an immune response to unrelated Ags. Chronically
infected C57BL/6 mice were immunized with OVA in the presence or absence of polyIC and the OVA-specific effector response
elicited was analyzed 10 days later using an in vivo CTL assay,
measuring the percentage cytolysis of autologous target cells
loaded with the SIINFEKL peptide, which is an immunodominant
epitope in the OVA Ag in H-2b mice. As shown in Fig. 6, immunization with OVA protein alone did not induce SIINFEKL-specific effector responses in uninfected mice or in mice chronically
infected with LCMV Cl13. Administration of polyIC, a type 1
IFN-inducer, in combination with OVA during immunization
greatly enhanced the effector response in uninfected mice, generating a high level of SIINFEKL-specific cytotoxic activity. Some
cytotoxic activity was also generated in chronically infected mice,
although this was not as strong as was observed in their uninfected
counterparts. This indicates that the diminished capacity of
chronically infected mice to produce type 1 IFN upon stimulation
with TLR ligands does not abolish the ability of these animals to
mount an effector CTL response when subsequently exposed to
The Journal of Immunology
unrelated Ags, although the reduction in type 1 IFN production
and/or other immunological abnormalities in these animals do result in an impairment in CTL induction.
Discussion
In this study, we analyzed type 1 IFN production during the acute
and chronic phases of LCMV infection of mice, and addressed
mechanisms that may contribute to the low level of type 1 IFN
production in chronically infected animals. Infection of mice with
either LCMV Arm (which was subsequently cleared) or LCMV
Cl13 (which establishes a chronic infection) was associated with
rapid induction of a strong systemic type 1 IFN response. However, this response was short-lived, with splenic IFN mRNA and
serum IFN bioactivity levels starting to fall while virus titers were
still increasing in both infections. Further, chronic LCMV infection was associated with only a low level of ongoing type 1 IFN
production, despite the presence of high viral titers. Chronically
infected mice nonetheless retained some capacity to produce type
1 IFN in response to TLR3, 7, and 9 stimuli, although the level of
IFN mRNA up-regulation was lower than in uninfected animals.
Splenic pDCs (and to a lesser extent cDCs) up-regulated type 1
IFN mRNA expression in acute LCMV infection with kinetics
consistent with them acting as a source of systemic type 1 IFN
production in early infection, and pDCs were also implicated as
one of the cellular sources of type 1 IFN production in mice chronically infected with LCMV Cl13. Splenic DC numbers were reduced during chronic LCMV infection, which was likely one of the
factors that contributed to the low level of type 1 IFN production
in chronically infected mice. The remaining DCs also displayed an
impaired response to TLR stimulation. LCMV-infected cells exhibited a defective capacity to up-regulate type 1 IFN production
in response to infection with Sendai virus, but remained responsive
to TLR stimuli. The reduced response of DCs from LCMV-infected mice to TLR ligation was thus probably not due to infection
of these cells but may have been a consequence of persistent exposure to activating stimuli.
Rapid induction of type 1 IFN production in acute LCMV infection has been previously documented (e.g., Refs. 26, 32, 33, 52)
and in this study, we showed that a very similar response is induced in the acute phase of LCMV infections that go on to persist.
Type 1 IFNs may be induced by several cellular pathways. These
include the RLR family-dependent pathways present in most cell
types that enable them to up-regulate type 1 IFNs following infection by sensing nonself nucleic acids in the cytoplasm (53) and
the TLR-dependent pathways that enable cells expressing these
PRRs, including pDCs, to up-regulate type 1 IFN production following recognition of viral components without the need to be
infected. Both pathways are amplified in an autocrine fashion by
signaling via the type 1 IFN receptor. LCMV was recently shown
to inhibit production of IFN-␤ in the cells it infects via a mechanism that involves a block in nuclear translocation of IRF-3 mediated by the viral nucleoprotein (50), which is expressed at high
levels in mice chronically infected with LCMV Cl13. This is a
conserved function exhibited by nucleoproteins of most arenaviruses, suggesting it favors virus persistence in vivo (54). Infected
cells are therefore unlikely to be a major source of type 1 IFN
production in vivo, in support of which serum IFN-␤ levels in
LCMV-infected IFN␤-promoter stimulator-1-deficient mice were
found to be very similar to those in LCMV-infected control animals, although there was some decrease in IFN-␣ levels (33). Instead, type 1 IFN is probably mainly derived from uninfected cells
recognizing viral components by means of TLRs and other PRRs.
Consistent with this, some (33) but not all (55) studies have shown
that LCMV infection of mice lacking MyD88, an adaptor molecule
7187
involved in transduction of signals from TLRs including TLR7 and
9 into the nucleus, is associated with a reduced type 1 IFN response. DCs, in particular the pDC subset, act as the major source
of circulating type 1 IFN production in many acute viral infections.
The role of pDCs in type 1 IFN production in acute LCMV infection is controversial, as depletion of cells expressing Ly6C, a
marker expressed by pDCs but not cDCs, did not affect type 1 IFN
production in LCMV-infected mice (14, 15). However, we showed
both in this and a previous study in mice acutely infected with
LCMV Arm by the i.p. route (32), that splenic pDCs are induced
to produce type 1 IFN very rapidly after LCMV infection, with
cDCs contributing to the response with delayed kinetics. Support
for the concept that pDCs do constitute the predominant source of
type 1 IFN production within the spleen after LCMV infection was
also provided by a recent study using novel transgenic Ifn-a6-GFP
reporter mice (33). Therefore, the initial burst of type 1 IFN production observed during infection with both LCMV Arm and
LCMV Cl13 may be mainly generated by stimulation of pDCs
(with a lesser contribution by cDCs and other cell types) by viral
components such as ssRNA (likely recognized via TLR7) or the
viral glycoprotein (which may be recognized by other TLR(s)).
In contrast to the high levels of type 1 IFN produced in the acute
phase of LCMV infection, little type 1 IFN is produced during
chronic LCMV infection. We also showed that chronically infected mice have an impaired ability to up-regulate type 1 IFN in
response to TLR stimuli, although some response was induced.
Consistent with this, mice chronically infected with LCMV from
birth were previously found to respond to polyIC treatment by
up-regulating type 1 IFN production and undergoing NK cell activation, although the response induced was reduced compared
with that observed in uninfected control animals (35). There are a
number of possible explanations for the very low levels of type 1
IFN production in mice chronically infected with LCMV Cl13.
Firstly, a large reduction in the number of pDCs and also cDCs
was observed in chronically infected mice. The mechanisms responsible for the reduction in DC numbers in chronically infected
mice are unclear, but our observation that the remaining DCs in
these animals were phenotypically activated suggests that DC activation and maturation (mediated by the presence of infection
and/or the chronic low-level type 1 IFN production) may reduce
the lifespan of these cells, causing pDCs to mature into other DC
subsets (56) and/or both DC subsets to undergo activation-associated apoptotic cell death (32). Consistent with this, we found that
BM-DCs underwent activation and apoptosis following in vitro
infection with LCMV (supplemental Fig. 2). Enhanced DC loss in
vivo may be compounded by defects in the rate of replacement of
DCs. LCMV infection has been reported to cause profound alterations in hematopoietic differentiation (57– 60), which are thought
to be mediated in large part by cytokines induced in the context of
infection. Notably, it has been shown that continuous type 1 IFN
production during chronic LCMV infection suppresses the development of bone marrow precursor cells into DC (61). However,
the reduction in DC numbers is probably not the only explanation
for the low levels of type 1 IFN produced in chronically infected
mice, as we also found that the remaining DCs in these animals
exhibit a reduced capacity to up-regulate type 1 IFN production in
response to TLR ligation. LCMV Cl13 is a DC-tropic LCMV isolate (50, 62) and during chronic infection, may infect as many as
a third of the DCs in the spleen. We thus addressed whether the
impairment observed in TLR-stimulated type 1 IFN production by
DCs from chronically infected mice may be a consequence of infection of these cells with LCMV. However, although we found
that chronic LCMV infection impaired the capacity of a DC line to
up-regulate type 1 IFN production following infection with Sendai
7188
virus (likely as a consequence of the LCMV nucleoprotein-mediated block in nuclear translocation of IRF-3, which would impede
type 1 IFN induction in response to the presence of cytoplasmic
Sendai viral RNA via RLR-family pathways; Ref. 50), no effect
was observed on the ability of the cells to produce type 1 IFNs in
response to TLR3, 7, or 9 stimuli.
The reduced responsiveness of DCs isolated from LCMV-infected mice to TLR ligation may potentially have been a consequence of impaired in vitro survival of these cells; as mentioned
above, they were phenotypically activated, and hence may have
had a reduced lifespan. Alternatively, their reduced responsiveness
may reflect a homeostatic mechanism to limit type 1 IFN-mediated
immunopathology. Excessive production of type 1 IFNs and also
other proinflammatory cytokines can have detrimental effects on
the host, and a number of mechanisms have been described which
negatively regulate TLR signaling cascades at different levels, in
addition to which there are multiple negative-feedback mechanisms that self-limit TLR- and cytokine-driven responses (22, 63).
The DCs in mice chronically infected with LCMV may have been
rendered hyporesponsive to TLR stimulation by the continuous
presence of viral components, resulting in a muted response to the
TLR stimuli we administered. Although there are high circulating
titers of infectious virus in mice chronically infected with LCMV,
LCMV is a noncytopathic virus, so the level of TLR ligands
present systemically may be relatively low compared with those
achieved after administration of exogenous TLR ligands. This
could explain why some up-regulation of type 1 IFN production
was still elicited following inoculation of chronically infected mice
with TLR ligands.
Not only was there little ongoing production of type 1 IFNs in
mice chronically infected with LCMV, but we also found that during the acute phase of infection with either LCMV Arm or Cl13,
type 1 IFN production occurred only transiently, with levels of
splenic mRNA and systemic activity beginning to decline when
viral titers were still rising. The mechanisms involved in type 1
IFN down-regulation in acute LCMV infection likely include some
or all of those contributing to the low-level IFN production in
chronically infected mice. Type 1 IFN mRNA levels in the spleen
began to decrease as early as 24 h postinfection, but splenic DC
numbers declined with slower kinetics, particularly in mice acutely
infected with LCMV Arm. It is thus possible that the initial drop
in IFN mRNA levels is largely due to homeostatic negative feedback mechanisms, with activation-associated apoptotic death of
DCs (compounded by immunopathogenic destruction of infected
cells in Cl13-infected mice) that are inadequately replaced due to
compromised homeostasis enhancing the down-regulation of type
1 IFN production at slightly later timepoints (and in fact resulting
in type 1 IFN mRNA levels in LCMV Arm-infected animals dropping below the baseline level constitutively present in uninfected
animals). The type 1 IFN response induced in the acute phase of
most virus infections is very transient; and likewise virus persistence is also generally associated with little or no ongoing type 1
IFN production (8, 20, 21). The mechanisms we have implicated in down-regulation of type 1 IFN production in LCMV
infection, or a subset thereof, may thus be common to many
infectious situations.
What are the consequences for the chronically infected host of
low level ongoing type 1 IFN production and an impaired ability
to up-regulate type 1 IFN production on stimulation? As type 1
IFNs mediate antiviral and immunostimulatory effects, pre-existing levels of this cytokine may confer some level of resistance to
other viral infections. For example, the reduced levels of influenza
infection we observed in chronically infected animals when the
virus was inoculated into the peritoneal cavity, a site where influ-
TYPE 1 IFN RESPONSE IN CHRONIC LCMV INFECTION
enza virus replicates poorly, may reflect greater viral resistance as
a result of the pre-existing antiviral state in these mice. Conversely, having an impaired ability to up-regulate high levels of
type 1 IFN may also impair the host’s capacity to combat other
infections. Not only will direct control of viral replication be reduced, but type 1 IFN are also critical for the development of
robust NK, T, and B cell responses (51, 64, 65), hence the antiviral
activity of these cell types may be affected. In support of this, we
observed that mice chronically infected with LCMV Cl13 mounted
a CTL response of reduced magnitude following inoculation with
a protein Ag together with polyIC, a system where T cell induction
occurs in an IFN-dependent fashion. Mice chronically infected
with LCMV have also been found to mount impaired immune
responses to a variety of other viruses (66, 67), although other
mechanisms (such as the reduction in DC numbers in mice chronically infected with LCMV) may also contribute to the immune
deficiency observed in all these systems. These findings highlight
the importance of unraveling the mechanisms by which type 1 IFN
production is impaired in chronic viral infections, as such information may help to direct the design of more effective immunostimulatory strategies for use in settings of chronic viral persistence.
Acknowledgments
We thank Andrew Worth for his excellent technical assistance in the MoFlo sorting experiments.
Disclosures
The authors have no financial conflict of interest.
References
1. Biron, C., and G. C. Sen. 2001. Interferons and Other Cytokines. In Fields Virology 4th edition. D. M. Knipe, and P. M. Howley, eds. Lippincott Willians and
Wilkins, Philadelphia, pp. 321–346.
2. Muller, U., U. Steinhoff, L. Reis, S. Hemmi, J. Pavlovic, R. Zinkernagel, and
M. Aquet. 1994. Functional role of type I and type II interferons in antiviral
defense. Science 264: 1918 –1921.
3. van den Broek, M. F., U. Muller, S. Huang, M. Aguet, and R. M. Zinkernagel.
1995. Antiviral defense in mice lacking both ␣/␤ and ␥ interferon receptors.
J. Virol. 69: 4792– 4796.
4. Moskophidis, D., M. Battegay, M. A. Bruendler, E. Laine, I. Gresser, and
R. M. Zinkernagel. 1994. Resistance of lymphocytic choriomeningitis virus to
␣/␤ interferon and to ␥ interferon. J. Virol. 68: 1951–1955.
5. Steinhoff, U., U. Muller, A. Schertler, H. Hengartner, M. Aguet, and
R. M. Zinkernagel. 1995. Antiviral protection by vesicular stomatitis virus-specific antibodies in ␣/␤ interferon receptor-deficient mice. J. Virol. 69: 2153–2158.
6. Ou, R., S. Zhou, L. Huang, and D. Moskophidis. 2001. Critical role for ␣/␤ and
␥ interferons in persistence of lymphocytic choriomeningitis virus by clonal exhaustion of cytotoxic T cells. J. Virol. 75: 8407– 8423.
7. Haller, O., G. Kochs, and W. Friedemann. 2006. The interferon response circuit:
induction and suppression by pathogenic viruses. Virology 344: 119 –130.
8. Capobianchi, M. R., I. Abbate, G. Cappiello, and M. Solmone. 2003. HCV and
interferon: viral strategies for evading innate defence mechanisms in the virushost battle. Cell Death Differ 10 (Suppl. 1): S22–S24.
9. Horvath, C. M. 2004. Weapons of STAT destruction. Interferon evasion by
paramyxovirus V protein. Eur. J. Biochem. 271: 4621– 4628.
10. Leib, D. A. 2002. Counteraction of interferon-induced antiviral responses by
herpes simplex viruses. Curr. Top. Microbiol. Immunol. 269: 171–185.
11. Schweizer, M., P. Matzener, G. Pfaffen, H. Stalder, and E. Peterhans. 2006.
“Self ” and “nonself ” manipulation of interferon defense during persistent infection: bovine viral diarrhea virus resists ␣/␤ interferon without blocking antiviral
activity against unrelated viruses replicating in its host cells. J. Virol. 80:
6926 – 6935.
12. Takeuchi, O., and S. Akira. 2008. MDA5/RIG-I and virus recognition. Curr.
Opin. Immunol. 20: 17–22.
13. Marie, I., J. E. Durbin, and D. E. Levy. 1998. Differential viral induction of
distinct interferon-␣ genes by positive feedback through interferon regulatory
factor-7. EMBO J. 17: 6660 – 6669.
14. Asselin-Paturel, C., A. Boonstra, M. Dalod, I. Durand, N. Yessaad,
C. Dezutter-Dambuyant, A. Vicari, A. O’Garra, C. Biron, F. Briere, and
G. Trinchieri. 2001. Mouse type I IFN-producing cells are immature APCs with
plasmacytoid morphology. Nat. Immunol. 2: 1144 –1150.
15. Siegal, F. P., N. Kadowaki, M. Shodell, P. A. Fitzgerald-Bocarsly, K. Shah,
S. Ho, S. Antonenko, and Y. J. Liu. 1999. The nature of the principal type 1
interferon-producing cells in human blood. Science 284: 1835–1837.
16. Ito, T., R. Amakawa, and S. Fukuhara. 2002. Roles of toll-like receptors in natural interferon-producing cells as sensors in immune surveillance. Hum. Immunol. 63: 1120 –1125.
The Journal of Immunology
17. Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi,
M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA helicase RIG-I
has an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5: 730 –737.
18. Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innate
antiviral responses by means of TLR7-mediated recognition of single-stranded
RNA. Science 303: 1529 –1531.
19. Akira, S., and H. Hemmi. 2003. Recognition of pathogen-associated molecular
patterns by TLR family. Immunol. Lett. 85: 85–95.
20. Bertoletti, A., and A. J. Gehring. 2006. The immune response during hepatitis B
virus infection. J. Gen. Virol. 87: 1439 –1449.
21. von Sydow, M., A. Sonnerborg, H. Gaines, and O. Strannegard. 1991. Interferon-␣ and tumor necrosis factor-␣ in serum of patients in various stages of HIV-1
infection. AIDS Res. Hum. Retroviruses 7: 375–380.
22. Lang, T., and A. Mansell. 2007. The negative regulation of Toll-like receptor and
associated pathways. Immunol. Cell. Biol. 85: 425– 434.
23. Weber, F., G. Kochs, and O. Haller. 2004. Inverse interference: how viruses fight
the interferon system. Viral Immunol. 17: 498 –515.
24. Katze, M., Y. He, and M. J. Gale. 2002. Viruses and interferon: a fight for
supremacy. Nat. Rev. Immunol. 2: 675– 687.
25. Merigan, T. C., M. B. Oldstone, and R. M. Welsh. 1977. Interferon production
during lymphocytic choriomeningitis virus infection of nude and normal mice.
Nature 268: 67– 68.
26. Malmgaard, L., T. P. Salazar-Mather, C. A. Lewis, and C. A. Biron. 2002. Promotion of ␣/␤ interferon induction during in vivo viral infection through ␣/␤
interferon receptor/STAT1 system-dependent and -independent pathways. J. Virol. 76: 4520 – 4525.
27. Kolumam, G. A., S. Thomas, L. J. Thompson, J. Sprent, and K. Murali-Krishna.
2005. Type I interferons act directly on CD8 T cells to allow clonal expansion
and memory formation in response to viral infection. J. Exp. Med. 202: 637– 650.
28. Havenar-Daughton, C., G. A. Kolumam, and K. Murali-Krishna. 2006. Cutting
edge: the direct action of type I IFN on CD4 T cells is critical for sustaining
clonal expansion in response to a viral but not a bacterial infection. J. Immunol.
176: 3315–3319.
29. Nguyen, K. B., W. T. Watford, R. Salomon, S. R. Hofmann, G. C. Pien,
A. Morinobu, M. Gadina, J. J. O’Shea, and C. A. Biron. 2002. Critical role for
STAT4 activation by type 1 interferons in the interferon-␥ response to viral
infection. Science 297: 2063–2066.
30. Louten, J., N. van Rooijen, and C. Biron. 2006. Type 1 IFN deficiency in the
absence of normal splenic architecture during lymphocytic choriomeningitis virus
infection. J. Immunol. 77: 3266 –3272.
31. Dalod, M., T. P. Salazar-Mather, L. Malmgaard, C. Lewis, C. Asselin-Paturel,
F. Briere, G. Trinchieri, and C. A. Biron. 2002. Interferon ␣/␤ and interleukin 12
responses to viral infections: pathways regulating dendritic cell cytokine expression in vivo. J. Exp. Med. 195: 517–528.
32. Montoya, M., M. J. Edwards, D. M. Reid, and P. Borrow. 2005. Rapid activation
of spleen dendritic cell subsets following lymphocytic choriomeningitis virus
infection of mice: analysis of the involvement of type 1 IFN. J. Immunol. 174:
1851–1861.
33. Jung, A., H. Kato, Y. Kumagai, H. Kumar, T. Kawai, O. Takeuchi, and S. Akira.
2008. Lymphocytoid choriomeningitis virus activates plasmacytoid dendritic
cells and induces a cytotoxic T-cell response via MyD88. J. Virol. 82: 196 –206.
34. Borrow, P., and M. B. Oldstone. 1997. Lymphocytic choriomeningitis virus. In
Viral Pathogenesis. N. Nathanson, ed. Lippincott-Raven Publishers, Philadelphia, pp. 593– 627.
35. Bukowski, J. F., C. A. Biron, and R. M. Welsh. 1983. Elevated natural killer
cell-mediated cytotoxicity, plasma interferon, and tumor cell rejection in mice
persistently infected with lymphocytic choriomeningitis virus. J. Immunol. 131:
991–996.
36. Holtermann, O. A., and E. A. Havell. 1970. Reduced interferon response in mice
congenitally infected with lymphocytic choriomeningitis virus. J. Gen. Virol. 9:
101–103.
37. Saron, M. F., Y. Riviere, A. G. Hovanessian, and J. C. Guillon. 1982. Chronic
production of interferon in carrier mice congenitally infected with lymphocytic
choriomeningitis virus. Virology 117: 253–256.
38. Dutko, F. J., and M. B. Oldstone. 1983. Genomic and biological variation among
commonly used lymphocytic choriomeningitis virus strains. J. Gen. Virol. 64 (Pt
8): 1689 –1698.
39. Ahmed, R., A. Salmi, L. D. Butler, J. M. Chiller, and M. B. Oldstone. 1984.
Selection of genetic variants of lymphocytic choriomeningitis virus in spleens of
persistently infected mice: role in suppression of cytotoxic T lymphocyte response and viral persistence. J. Exp. Med. 160: 521–540.
40. Wagner, A. M., J. K. Loganbill, and D. M. Besselson. 2003. Detection of Sendai
virus and pneumonia virus of mice by use of fluorogenic nucleas reverse transcriptase polymerase chain reaction Comp. Med. 53: 173–177.
41. Karaghiosoff, M., R. Steinborn, P. Kovarik, G. Kriegshauser, M. Baccarini,
B. Donabauer, U. Reichart, T. Kolbe, C. Bogdan, T. Leanderson, D. Levy, et al.
2003. Central role for type I interferons and Tyk2 in lipopolysaccharide-induced
endotoxin shock. Nat. Immunol. 4: 471– 477.
7189
42. Montoya, M., R. Dawes, D. M. Reid, L. N. Lee, J. Piercy, P. Borrow,
E. Z. Tchilian, and P. C. Beverley. 2006. CD45 is required for type 1 IFN production by dendritic cells. Eur. J. Immunol. 36: 2150 –2158.
43. Tovey, M., J. Begon-Lours, and I. Gresser. 1974. A method for large-scale preparation of potent interferon preparations. Proc. Soc. Exp. Biol. Med. 146: 809.
44. Vremec, D., M. Zorbas, R. Scollay, D. J. Saunders, C. F. Ardavin, L. Wu, and
K. Shortman. 1992. The surface phenotype of dendritic cells purified from mouse
thymus and spleen: investigation of the CD8 expression by a subpopulation of
dendritic cells. J. Exp. Med. 176: 47–58.
45. Lin, Y., T. J. Roberts, V. Sriram, S. Cho, and R. R. Brutkiewicz. 2003. Myeloid
marker expression on antiviral CD8⫹ T cells following an acute virus infection.
Eur. J. Immunol. 33: 2736 –2743.
46. Lutz, M. B., N. A. Kukutsch, M. Menges, S. Rossner, and G. Schuler. 2000.
Culture of bone marrow cells in GM-CSF plus high doses of lipopolysaccharide
generates exclusively immature dendritic cells which induce alloantigen-specific
CD4 T cell anergy in vitro. Eur. J. Immunol. 30: 1048 –1052.
47. Girolomoni, G., M. B. Lutz, S. Pastore, C. U. Assmann, A. Cavani, and
P. Ricciardi-Castagnoli. 1995. Establishment of a cell line with features of early
dendritic cell precursors from fetal mouse skin. Eur. J. Immunol. 25: 2163–2169.
48. Prakash, A., E. Smith, C. K. Lee, and D. E. Levy. 2005. Tissue-specific positive
feedback requirements for production of type I interferon following virus infection. J. Biol. Chem. 280: 18651–18657.
49. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate
immunity. Cell 124: 783– 801.
50. Martinez-Sobrido, L., E. I. Zuniga, D. Rosario, A. Garcia-Sastre, and
J. C. de la Torre. 2006. Inhibition of the type I interferon response by the nucleoprotein of the prototypic arenavirus lymphocytic choriomeningitis virus.
J. Virol. 80: 9192–9199.
51. Le Bon, A., V. Durand, E. Kamphuis, C. Thompson, S. Bulfone-Paus,
C. Rossmann, U. Kalinke, and D. F. Tough. 2006. Direct stimulation of T cells
by type I IFN enhances the CD8⫹ T cell response during cross-priming. J. Immunol. 176: 4682– 4689.
52. Jacobson, S., R. M. Friedman, and C. J. Pfau. 1981. Interferon induction by
lymphocytic choriomeningitis viruses correlates with maximum virulence.
J. Gen. Virol. 57: 275–283.
53. Kato, H., S. Sato, M. Yoneyama, M. Yamamoto, S. Uematsu, K. Matsui,
T. Tsujimura, K. Takeda, T. Fujita, O. Takeuchi, and S. Akira. 2005. Cell typespecific involvement of RIG-I in antiviral response. Immunity 23: 19 –28.
54. Martinez-Sobrido, L., P. Giannakas, B. Cubitt, A. Garcia-Sastre, and
J. C. de la Torre. 2007. Differential inhibition of type I interferon induction by
arenavirus nucleoproteins. J. Virol. 81: 12696 –12703.
55. Zhou, S., E. A. Kurt-Jones, L. Mandell, A. Cerny, M. Chan, D. T. Golenbock, and
R. W. Finberg. 2005. MyD88 is critical for the development of innate and adaptive immunity during acute lymphocytic choriomeningitis virus infection. Eur.
J. Immunol. 35: 822– 830.
56. Zuniga, E. I., D. B. McGavern, J. L. Pruneda-Paz, C. Teng, and M. B. Oldstone.
2004. Bone marrow plasmacytoid dendritic cells can differentiate into myeloid
dendritic cells upon virus infection. Nat. Immunol. 5: 1227–1234.
57. Bro-Jorgensen, K., and S. Knudtzon. 1977. Changes in hemopoiesis during the
course of acute LCM virus infection in mice. Blood 49: 47–57.
58. Borrow, P., S. Hou, S. Gloster, M. Ashton, and L. Hyland. 2005. Virus infectionassociated bone marrow B cell depletion and impairment of humoral immunity to
heterologous infection mediated by TNF-␣/LT␣. Eur. J. Immunol. 35: 524 –532.
59. Bro-Jorgensen, K. 1978. The interplay between lymphocytic choriomeningitis
virus, immune function, and hemopoiesis in mice. Adv. Virus Res. 22: 327–369.
60. Binder, D., J. Fehr, H. Hengartner, and R. M. Zinkernagel. 1997. Virus-induced
transient bone marrow aplasia: major role of interferon-␣/␤ during acute infection
with the noncytopathic lymphocytic choriomeningitis virus. J. Exp. Med. 185:
517–530.
61. Hahm, B., M. J. Trifilo, E. I. Zuniga, and M. B. Oldstone. 2005. Viruses evade
the immune system through type I interferon-mediated STAT2-dependent, but
STAT1-independent, signaling. Immunity 22: 247–257.
62. Borrow, P., C. F. Evans, and M. B. Oldstone. 1995. Virus-induced immunosuppression: immune system-mediated destruction of virus-infected dendritic cells
results in generalized immune suppression. J. Virol. 69: 1059 –1070.
63. Rothlin, C. V., S. Ghosh, E. I. Zuniga, M. B. Oldstone, and G. Lemke. 2007.
TAM receptors are pleiotropic inhibitors of the innate immune response. Cell
131: 1124 –1136.
64. Nguyen, K. B., T. P. Salazar-Mather, M. Y. Dalod, J. B. Van Deusen, X. Q. Wei,
F. Y. Liew, M. A. Caligiuri, J. E. Durbin, and C. A. Biron. 2002. Coordinated and
distinct roles for IFN-␣ ␤, IL-12, and IL-15 regulation of NK cell responses to
viral infection. J. Immunol. 169: 4279 – 4287.
65. Le Bon, A., and D. F. Tough. 2002. Links between innate and adaptive immunity
via type I interferon. Curr. Opin. Immunol. 14: 432– 436.
66. Tishon, A., P. Borrow, C. Evans, and M. B. Oldstone. 1993. Virus-induced immunosuppression, 1: age at infection relates to a selective or generalized defect.
Virology 195: 397– 405.
67. Odermatt, B., M. Eppler, T. P. Leist, H. Hengartner, and R. Zinkernagel. 1991.
Virus-triggered acquired immunodeficiency by cytotoxic T-cell-dependent destruction of antigen-presenting cells and lymph follicle structure. Proc. Natl.
Acad. Sci. USA 88: 8252– 8256.