Early IFN-?? Production and Innate Immunity During Listeria monocytogenes Infection in the Absence of NK Cells1

Abstract

NK cells are believed to play a mandatory role during the early phases of Listeria monocytogenes infection by producing IFN-γ, which is required for the activation of macrophage effector functions. Mice deficient in the common cytokine receptor γ-chain (γc−/−), which completely lack NK cells, were used to examine whether NK cells were essential for resistance to Listeria infection in vivo. Surprisingly, infected γc−/− mice showed normal innate immunity and macrophage responses against sublethal Listeria infection 2 days postinfection. At this time point, γc−/− mice showed increased blood IFN-γ levels compared with those in noninfected controls, demonstrating an NK-independent source of IFN-γ, which explains early resistance. Listeria-infected γc−/− × recombinase-activating gene-2−/− double-deficient mice were unable to produce IFN-γ and were highly susceptible to L. monocytogenes. Since T cells, but not B cells, are major IFN-γ producers, and γc−/− T cells were found to be efficient IFN-γ producers in vitro, we conclude from these results that T cells functionally replace NK cells for the early IFN-γ production that is necessary for activating the innate immune system following infection with L. monocytogenes. This novel observation in listeriosis underscores how the adaptive immune response can maintain and influence innate immunity.

Full text

Early IFN-
g
Production and Innate Immunity During Listeria
monocytogenes Infection in the Absence of NK Cells
1
Åsa Andersson,* Wen Juan Dai,
†
James P. Di Santo,* and Frank Brombacher
2
*
†‡
NK cells are believed to play a mandatory role during the early phases of Listeria monocytogenes infection by producing IFN-
g
,
which is required for the activation of macrophage effector functions. Mice deficient in the common cytokine receptor
g
-chain
(
g
c2/2
), which completely lack NK cells, were used to examine whether NK cells were essential for resistance to Listeria infection
in vivo. Surprisingly, infected
g
c2/2
mice showed normal innate immunity and macrophage responses against sublethal Listeria
infection 2 days postinfection. At this time point,
g
c2/2
mice showed increased blood IFN-
g
levels compared with those in
noninfected controls, demonstrating an NK-independent source of IFN-
g
, which explains early resistance. Listeria-infected
g
c2/2
3recombinase-activating gene-2
2/2
double-deficient mice were unable to produce IFN-
g
and were highly susceptible to L.
monocytogenes. Since T cells, but not B cells, are major IFN-
g
producers, and
g
c2/2
T cells were found to be efficient IFN-
g
producers in vitro, we conclude from these results that T cells functionally replace NK cells for the early IFN-
g
production that
is necessary for activating the innate immune system following infection with L. monocytogenes. This novel observation in liste-
riosis underscores how the adaptive immune response can maintain and influence innate immunity. The Journal of Immunology,
1998, 161: 5600–5606.
Listeria monocytogenes is a Gram-positive, facultative in-
tracellular bacterium causing disseminated infections in
immunocompromised individuals and in pregnant woman
(1), leading mainly to septicemia and meningitis. Because listeri-
osis provokes a similar pathology in humans and rodents, the
mouse model of infection has been widely used to study cell-me-
diated immunity, which is the main protective host response mech-
anism. Immunity to Listeria proceeds in two stages: 1) an early
innate immune response requiring macrophages, NK cells, and
neutrophils that limit growth of the organism; and 2) the later
“sterilizing” adaptive immune response, which involves Ag-spe-
cific T cells that clear the infection. Still, the essential roles that
cellular and soluble components play during Listeria infection are
not fully elucidated.
Listeria induces its own internalization into nonactivated mac-
rophages and other nonprofessional phagocytes, where the bacteria
survive and replicate. Listeria-infected macrophages respond by
producing IL-12, which, in turn, activates NK cells to release
IFN-
g
(reviewed in Refs. 2–4). IFN-
g
then synergizes with bac-
terial products to maximally activate macrophage effector func-
tions and their secretion of inflammatory cytokines. An important
role of the NK cell/macrophage-IL-12/IFN-
g
loop in listeriosis has
been demonstrated by the increased susceptibility of mice deficient
for IL-12,
3
for IFN-
g
(6) or its receptor (7, 8), or for STAT-1 (9)
or ICSBP/IRF2 (10), with the latter two gene products crucial for
the IFN-
g
-mediated intracellular signal transduction. However, IL-
12
2/2
mice survive low dose infection, implicating an IL-12-in-
dependent induction of IFN-
g
.
3
In contrast, macrophage activation
by IFN-
g
is crucial to survive the first week of Listeria infection as
evidenced by IFN-
g
neutralization studies in Listeria-infected
wild-type (WT) mice (11). Moreover, IFN-
g
2/2
or IFN-
g
R
2/2
mice are highly susceptible even to very low doses of Listeria and
die during the first week following infection (6, 8). We have re-
cently shown that IFN-
g
R
2/2
macrophages are impaired to limit
the exit of Listeria from the phagosome to the cytoplasm. These
listericidal defects eventually culminate in unchecked growth and
dissemination of the organism, resulting in extensive organ necro-
sis and the rapid death of the infected animals (8). The crucial
IFN-
g
-dependent macrophage effector mechanisms that participate
in the elimination of Listeria are not completely elucidated, but
involve activation of TNF-
a
and NF-IL-6, since mice deficient for
these genes are also highly susceptible to Listeria infection (re-
viewed in Ref. 12).
The inflammatory response that leads to granulomatous forma-
tion appears necessary for the subsequent adaptive immune re-
sponse to Listeria. The cellular components at this stage include
extravasative macrophages, NK cells, and neutrophils that serve to
limit bacterial spread in immunocompetent mice. The activation
and proliferation of Listeria-specific T cells during this later adap-
tive immunity eventually clear the invader and lead to acquired
immunity. It is important to note that the presence of
ab
T cells,
gd
T cells (13), or MHC class I-positive T cells (14) is sufficient
for Listeria clearance. However, the presence and coordinated ac-
tion of all T cell subpopulations appear to provide optimal protec-
tion and therefore the most efficient elimination (15).
The common
g
-chain (
g
c
)
4
is a shared cytokine receptor chain
that plays a critical functional role in the receptors to IL-2, IL-4,
*Institut National de la Sante´ et de la Recherche Me´dicale Unit 429, Hoˆpital Necker-
Enfants Malades, Paris, France;
†
Max Planck Institute for Immunobiology, Freiburg,
Germany; and
‡
University of Cape Town, Cape Town, South Africa
Received for publication January 26, 1998. Accepted for publication July 17, 1998.
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 a fellowship from the Wenner-Gren Foundation (to
A.A.) and by Institut National de la Sante´ et de la Recherche Me´dicale, Association
pour la Recherche sur le Cancer, and Ligue contre la Cancer.
2
Address correspondence and reprint requests to Dr. Frank Brombacher, Groote
Schuur Hospital, Immunology Department, H47, Observatory 7925, University of
Cape Town, Cape Town, South Africa. E-mail address: fbrombac@uctgsh1.uct.ac.za
3
F. Brombacher, A. Dorfmu¨ller, J. Magram, J. Ferrante, G. Ko¨hler, A. Wunderlin,
K. Palmer, M. K. Gately, and G. Alber. 1998. Interleukin-12 is dispensable for pro-
tective immunity against low doses of Listeria monocytogenes.Submitted for publi-
cation.
4
Abbreviations used in this paper:
g
c
, common
g
-chain; RAG-2, recombinase-acti-
vating gene-2; HKLM, heat-killed Listeria monocytogenes; WT, wild type.
Copyright © 1998 by The American Association of Immunologists 0022-1767/98/$02.00

IL-7, IL-9, and IL-15 (16).
g
c
deficiency in humans results in X-
linked SCID, characterized by a complete block in T cell and NK
cell differentiation and a dramatic susceptibility to various types of
infection agents (17, 18). Animal models of X-linked SCID have
been generated in mice that share many features of the human
disease phenotype (19–21).
g
c
2/2
mice show a selective absence
of NK cells,
gd
T cells, and gut-associated lymphoid tissue. In
contrast, some mature
ab
T cells and B cells are able to develop,
although the immunologic competence of these residual lymphoid
cells remains untested. It is presumed that
g
c
2/2
mice will dem-
onstrate a marked susceptibility to infection due to the lymphoid
developmental defects present in these mice. Moreover, because of
the differential effects of
g
c
deficiency on the development of the
various lymphoid subsets, infection of
g
c
2/2
mice should reveal
whether NK cells,
gd
T cells, or gut lymphoid cells are essential
for in vivo immune responses to certain pathogens. In this report,
g
c
-deficient mice have been used to define the functional impor-
tance of NK cells for innate immunity to L. monocytogenes.
Materials and Methods
Mice
Mice with a null mutation in the
g
c
have been described previously (21)
and were backcrossed four generations on the C57BL/6 background. Re-
combinase activating gene-2 (RAG-2) mice (22) from the 10th generation
backcross to C57BL/6 were provided by Dr. B. Rocha (Institut National de
la Sante´ et de la Recherche Me´dicale Unit 345, Paris, France). Mice doubly
deficient in
g
c
and RAG-2 were obtained by intercrossing, and genotypes
were determined by PCR on DNA derived from tail snips (primer se-
quences available from the authors). Breeding pairs of IFN-
g
R
2/2
mice
and their WT controls (7) were provided by M. Aguet (Zurich, Switzer-
land). All mice were raised in a specific pathogen-free animal facility (Cen-
tre Nationale de la Recherche Scientifique/CDTA, Orleans, France; or Max
Planck Institute for Immunobiology). Seven- to 12-wk-old mice were in-
fected with L. monocytogenes and housed in filter-top cages.
Bacteria and infection of mice
Virulent L. monocytogenes were grown in trypticase-soy broth (Difco, De-
troit, MI) as described previously (23). Aliquots of log-phase growing cul-
tures were stored at 270°C until use. For each experiment, a vial was
thawed, and bacteria were washed once in saline and diluted in endotoxin-
free PBS before injection. Mice were injected i.v. into the tail vein or i.p.
with 200
m
l of PBS with or without bacteria. The number of viable bacteria
in the inoculum and in organ homogenates was determined by plating
10-fold serial dilutions on trypticase-soy broth-agar plates. Plates were
incubated at 37°C, and numbers of CFU were counted after 24 h. Heat-
killed L. monocytogenes (HKLM) were prepared by incubating bacteria at
60°C for 60 min. Bacteria were pelleted and stored in PBS at 220°C until use.
Histology
Mice were killed by cervical dislocation; their organs were removed, cut in
pieces, and fixed in 5% formalin solution. Tissues were dehydrated in
ethanol and embedded in paraffin. The 5-
m
sections were cut and stained
with hematoxylin and eosin or with naphthol-AS-D-chloroacetate-ester for
special visualization of neutrophils (24). Listeria were identified by silver
staining. Silver stain was used to visualize Listeria (25). All studies were
performed using a Zeiss microscope fitted with image analysis software
(SIS, Munster, Germany) for the computerized morphometry.
Cell preparation and culture
Splenocytes were isolated aseptically, and E were lysed using hypotonic
saline solution. In some experiments, T cells were enriched using Mini-
Macs isolation columns (Tebu, Paris, France) and anti-CD5-coupled para-
magnetic beads according to the manufacturer’s protocol. Cells were cul-
tured in Iscove’s medium (Life Technologies, Paisley, Scotland),
supplemented with 10% (v/v) heat-inactivated FCS, 5 310
25
M 2-ME,
100 U/ml penicillin, and 100
m
g/
m
l streptomycin. Cells (1–2 310
6
cells/
ml) were stimulated in 48- or 96-well plates with anti-CD3 (5
m
g/
m
l; clone
2C11), IL-12 (100 U/ml), or HKLM (2 310
7
CFU/ml). Cell-free super-
natants were harvested at different times following incubation at 37°C.
Determination of cytokine levels in blood and supernatants
The production of cytokines was measured using sandwich ELISAs. Sera
or plasma and culture supernatants were used in three- or fivefold serial
dilutions. Appropriate standards (PharMingen, San Diego, CA) were used
in threefold serial dilutions. The coating and biotinylated detection Abs for
IL-1
a
, IL-6, and IL-10 were purchased from PharMingen, and those for
IFN-
g
were obtained from Genzyme (Cambridge, MA). Alkaline phospha-
tase coupled to streptavidin (Southern Biotechnology Associates, Birming-
ham, AL) was used to detect biotinylated Abs.
Results
Normal innate immunity to Listeria infection in the absence of
IFN-
g
-producing NK cells
NK cells are believed to provide the initial burst of IFN-
g
that is
critical for the macrophage activation that leads to a protective
innate response following infection with L. monocytogenes (re-
viewed in Refs. 3 and 12). In the absence of IFN-
g
or its receptor,
Listeria replicates unabated, and mice rapidly succumb to infection
within days (6–8). Since the
g
c
is required for NK development
(19, 21),
g
c
-deficient mice provided us the means to address the
role of NK cells (and the cytokines they produce (e.g., IFN-
g
)in
the innate response to Listeria. Unexpectedly, NK-deficient
g
c
2/2
mice were resistant to sublethal Listeria infection and survived the
early course of infection (beyond day 7) without mortality (Fig. 1),
indicating an efficient innate response. Listeria-infected mice were
subsequently analyzed on day 2 postinfection, where the innate
responses are at their maximum and where in the absence of IFN-
g
stimulation mice show a significantly higher bacterial burden (8).
The liver and spleen of infected
g
c
2/2
mice showed a similar
bacterial burden as infected organs of control mice (Fig. 2) and a
similar histopathology, concerning the size and number of micro-
abscesses in the infected organs (Fig. 3, aand b), thus failing to
demonstrate any defects in the day 2 response. In contrast, IFN-
g
R
2/2
mice show a dramatic inability to control Listeria at this
time point and succumbed to infection within 4 days (Fig. 1), sug-
gesting that some IFN-
g
had to be present in infected
g
c
-deficient
mice to prime the innate effector cells, such as macrophages. In-
deed, circulating levels of IFN-
g
were detected in the sera of List-
eria-infected
g
c
2/2
mice at 2 days postinfection, although the total
amount was considerable reduced (13-fold) compared with that in
infected controls (Table I). We have demonstrated previously that
the absolute numbers and phenotype of splenic and peritoneal
FIGURE 1. Survival of
g
c
-deficient mice following infection with L.
monocytogenes. Five mice deficient for
g
c
(open circle) or IFN-
g
R (open
rectangle) and C57BL/6 control mice (closed circle) were i.v. infected with
9310
3
CFU L. monocytogenes, and their survival was followed during the
first week of infection. Similar results were obtained in three independent
experiments.
5601The Journal of Immunology

macrophages in
g
c
2/2
mice were normal, and that
g
c
-deficient
macrophages were not impaired in their abilities to produce in-
flammatory mediators or to phagocytose micro-organisms (20). In
addition, peritoneal macrophages from uninfected
g
c
2/2
mice
could produce considerable amounts of nitric oxide after in vitro
culturing (
g
c
2/2
macrophages, 70.0 625.8
m
M; WT macro-
phages, 12.4 68.7
m
M). Moreover, macrophages from day 2-in-
fected
g
c
2/2
mice showed a similar inflammatory cytokine re-
sponse after restimulation with HKLM as WT macrophages (Table
II). Because macrophage activation by IFN-
g
is crucial for some of
FIGURE 2. L. monocytogenes burden during the early course of infection.
Four
g
c
-deficient mice (open symbols) and C57BL/6 control mice (closed
symbols) were infected i.p. with 9 310
3
CFU of L. monocytogenes on day 0.
The number of bacteria recovered from livers (circles) and spleens (triangles)
of individual mice were determined by CFU on trypticase-soy agar on day 2
postinfection. The mean value is indicated by a horizontal line. A representa-
tive of three independent experiments is shown.
FIGURE 3. Comparison of the liver histopathology of L. monocytogenes-infected mice. During the early phase of infection (day 2; see Fig. 2), the liver
of infected C57BL/6 control mice (a)or
g
c
-deficient mice (b) contained small microabscesses with infiltrating neutrophils (stained red with naphthol-AS-
D-chloroacetate-ester). The microabscesses of
g
c2/2
mice were slightly larger than those of control mice. Liver pathology from moribund (day 4; see Figs.
1 and 4)
g
c
3RAG-2 double-deficient mice (c) or IFN-
g
R-deficient mice (d) were comparable, with considerable tissue damage (.60%), multiple foci
of hepatocellular necrosis, and inflammatory cell infiltration. The liver pathologies of chronically infected (day 17; see Fig. 7) RAG-2-deficient mice (e)
and
g
c
-deficient mice (f) were comparable and showed many granulomatous lesions with neutrophil infiltration. The small insets demonstrate silver-stained
bacteria from a granulomatous lesion. Most bacteria are within the inflammatory foci, with some dissemination of bacteria. One representative liver section
from five mice per group analyzed is shown. Bars 50.25 mm.
Table I. IFN-
g
blood levels during early infection with L.
monocytogenes
a
Mouse
IFN-
g
Levels (pg/ml)
Day 0 Day 2
WT ,25 2325 6172
g
c2/2
,25 172 699
RAG2
2/2
,25 600 6264
g
c2/2
3RAG2
2/2
,25 ,25
a
Mice were prebled and then infected with 9 310
3
CFU of L. monocytogenes.
Mice were rebled after 2 days, and IFN-
g
levels were determined by ELISA. No
IFN-
g
induction was observed in PBS-infected mice alone. Results are mean and SD
of four mice from each group.
5602 L. monocytogenes INFECTION IN
g
c-DEFICIENT MICE

these inflammatory responses (8), these results further indicate an
alternative source of IFN-
g
-producing cells in
g
c
2/2
mice.
T cells replace NK cells for priming innate immunity in
g
c
2/2
mice
During the early stages following Listeria infection, NK cells are
the primary IFN-
g
producers. In contrast, primed T cells are potent
in vivo IFN-
g
producers during the late phase (or adaptive) re-
sponse to Listeria infection (26). To investigate the potential pro-
tective role of
g
c
2/2
T and B cells in the early immune response
to Listeria, we generated
g
c
2/2
3RAG-2
2/2
double-mutant mice
and determined their response to sublethal Listeria infection. In
striking contrast to Listeria-infected RAG-2- or
g
c
-deficient mice,
all
g
c
3RAG-2 double-deficient mice succumbed to by day 4
postinfection (Fig. 4), with kinetics of mortality similar to those of
IFN-
g
R
2/2
mice (see Fig. 1). Circulating levels of IFN-
g
were at
the limit of detection in the double-mutant mice (,25 pg/ml) on
day 2 postinfection, whereas
g
c
2/2
or RAG-2
2/2
mice showed a
clear increase in serum IFN-
g
(Table I). Moreover, Listeria-in-
fected
g
c
2/2
3RAG-2
2/2
mice demonstrated a liver pathology
identical with that found following infection of IFN-
g
R
2/2
mice
(Fig. 3, cand d). Both mutant mouse strains showed extensive
necrotic lesions and liver parenchymal destruction with dissemi-
nation of bacteria. These results 1) point to a critical role of IFN-
g
in protective innate responses to L. monocytogenes, 2) rule out the
possibility that immature NK cells were source of IFN-
g
in
g
c
2/2
mice, and 3) strongly suggest that
g
c
2/2
T cells are the early
IFN-
g
producers that prime innate responses following Listeria
infection. To address this last point directly, T cells of naive
g
c
2/2
mice were purified and stimulated with anti-CD3, IL-12, or both,
and IFN-
g
release into the culture supernatant was measured. Both
normal and
g
c
2/2
T cells responded by IFN-
g
production after
stimulation with IL-12 and anti-CD3 (Fig. 5).
Chronic listeriosis in
g
c
-deficient mice
Having shown that T cells play an important role in maintaining
innate immunity to Listeria in
g
c
2/2
mice, we next examined
whether these T cells were also competent to sterilize a Listeria-
infected
g
c
-deficient mouse. It is well established that elimination
of Listeria during the later course of infection is mediated by Ag-
specific T cells (reviewed in Ref. 12). WT and
g
c
2
mice were
sublethally infected, and on day 12 postinfection the bacterial bur-
den in liver and spleen was measured (Fig. 6). Immunocompetent
mice showed low residual bacterial levels in these organs, with
nearly complete sterilization of the Listeria. In contrast,
g
c
2/2
mice were clearly defective in Listeria elimination, harboring up to
10
6
CFU in the infected organs, and thus demonstrated a chronic
state of infection. A comparative infection study of RAG-2
2/2
mice (with no mature T cells) vs
g
c
2/2
mice was next performed.
As expected, RAG-2
2/2
mice survived the early course of infec-
tion, but succumbed during the later infection period. Interestingly,
FIGURE 4. Survival of
g
c
-deficient mice during L. monocytogenes in-
fection in the absence of lymphocytes. Five
g
c2/2
mice (open circle),
g
c2/2
3RAG-2
2/2
mice (open square), RAG-2
2/2
mice (half filled
square), or C57BL/6 control mice (closed circle) were i.v. infected with
3310
3
CFU L. monocytogenes, and their survival was followed during the
first week of infection. Similar results were obtained in a second
experiment.
FIGURE 5.
g
c2/2
T cells produce IFN-
g
. Splenic T cells were isolated
from
g
c
-deficient or C57BL/6 control mice and cultured in vitro as indi-
cated. After 48 h, the IFN-
g
released into the culture supernatant was
measured by ELISA. Similar results were obtained in a second experiment.
FIGURE 6. L. monocytogenes burden late in infection. Five
g
c
-deficient
mice (open symbols) and C57BL/6 control mice (closed symbols) were
infected i.v. with 2 310
3
CFU of L. monocytogenes on day 0 and analyzed
on day 12 postinfection. Mutant mice were unable to clear Listeria, with an
average CFU number of 2 310
5
for liver (circles) and 1 310
4
for spleen
(triangles), whereas control mice showed an ;400-fold reduced CFU num-
ber in liver, and most mice had eliminated Listeria in spleen (,10 CFU).
The mean value is indicated by a horizontal line.
Table II. Cytokine levels after HKLM restimulation of peritoneal
exudate cells
a
Mouse
Cytokine Levels (ng/ml)
IL-1
a
IL-6 IL-10
WT 2.6 60.5 89.4 624.7 17.0 62.6
g
c2/2
1.7 60.4 130.0 623.1 33.3 62.3
a
Five mice per group were infected with 9 310
3
CFU of L. monocytogenes.
Pooled peritoneal exudate cells were restimulated with HKLM in vitro after 2 days,
and cytokine production in the supernatant was determined by ELISA. Results are
mean and SD of triplicate wells.
5603The Journal of Immunology

g
c
-deficient mice showed mortality kinetics similar to those ob-
served in RAG-2
2/2
mice (Fig. 7). Moreover, moribund mice of
both mutant strains showed similar histopathology, with many
granulomatous lesions and neutrophil infiltration, harboring Liste-
ria in the inflammatory foci (Fig. 3, eand f). To determine the
relative degree of T cell dysfunction in
g
c
2/2
mice, T cell re-
sponses to alloantigens were examined.
g
c
2/2
T cells completely
failed to proliferate in response to MHC-disparate stimulator cells,
although control T cells responded normally (data not shown).
Taken together, these results indicate that both alloantigen- and
Ag-specific T cell responses are completely defective in
g
c
2/2
mice.
Discussion
The production of IFN-
g
and TNF-
a
in the early phase of infection
by L. monocytogenes is crucial for the activation of macrophage
effector functions that are required to limit bacterial growth and
control infection (6–8, 27–29). The initial source of IFN-
g
appears
to be NK cell derived, and its production is stimulated by IL-12
produced by resident macrophages following penetration of the
invading micro-organism. This NK/macrophage feedback activa-
tion model of T cell-independent innate resistance (see Fig. 8) has
been mainly developed from extensive Listeria infection studies in
SCID mice (2, 3, 30) and remains the current model to explain
innate immunity against pathogenic infections (31). Despite this
apparently important function of NK cells, a critical question re-
mained concerning alternative sources of initial IFN-
g
production
that may be involved in the early resistance to Listeria.Toun-
equivocally answer this question we used
g
c
2/2
mice, which com-
pletely lack NK cells (19, 21), and showed that these mice were
able to survive the early course of Listeria infection. The observed
innate immunity could be explained by the finding of considerable
circulating levels of IFN-
g
(albeit reduced compared with those in
controls) in the mutant mice during this time. These results dem-
onstrated that 1) NK cells are not absolutely required for innate
resistance to Listeria; and 2) NK-independent sources of early
IFN-
g
production exist in
g
c
2/2
mice.
Which cells might produce early IFN-
g
during the anti-Listeria
response of
g
c
-deficient mice? Activated T cells are efficient IFN-
g
producers during the latter specific immune response to Listeria
infection, and we have shown that unprimed
g
c
2/2
-derived splenic
T cells can be stimulated to produce IFN-
g
. Moreover, Listeria-
infected
g
c
2/2
3RAG-2
2/2
double-deficient mice (lacking NK,
B, and T cells) were unable to induce circulating levels of IFN-
g
and succumbed to Listeria infection with the same pathophysiol-
ogy as IFN-
g
R-deficient mice. These results demonstrated that
FIGURE 7. Chronically infected
g
c2/2
and RAG-2
2/2
mice succumb
to L. monocytogenes infection.
g
c2/2
mice (open circle), RAG-2
2/2
mice
(half filled square), or C57BL/6 control mice (closed circle) were i.v. in-
fected with 2 310
3
CFU, and their survival was followed during infection.
In contrast to control mice, all mutant mice died within 3 wk postinfection
with similar mortality kinetics. Similar results were obtained in a second
experiment.
FIGURE 8. Innate immunity to L. monocytogenes.Left panel, T cell-independent resistance. Resident macrophages or monocytes infected with Listeria
produce a number of cytokines, including IL-12, which stimulates NK cells to produce IFN-
g
, as found in T cell-deficient mice (such as SCID or RAG-2
mutants). IFN-
g
in combination with TNF-
a
activates macrophages for their listericidal effector functions. Center panel, T cell-dependent resistance. In
the absence of NK cells (and
gd
T cells), early IFN-
g
production is supplied by
ab
T cells, as found in
g
c
-deficient mice that demonstrate innate immunity
to L. monocytogenes.Right panel, No resistance. Mice deficient for NK and T cells (RAG-2 3
g
c
double mutants) lack IFN-
g
production. Consequently,
listericidal macrophage functions are absent, and mice demonstrate an absence of innate immune responses.
5604 L. monocytogenes INFECTION IN
g
c-DEFICIENT MICE

g
c
2/2
lymphocytes were indeed the source of early IFN-
g
pro-
duction during innate immune responses to L. monocytogenes in-
fection. Since B cells do not play a protective role during Listeria
infection, T cells seem to be the IFN-
g
producers in Listeria-in-
fected
g
c
2/2
mice.
It appears that alternative mechanisms exist to prime effector
cells of the innate immune system, which is independent of the
classical pathway that requires macrophage-derived IL-12 acting
on NK cells to produce IFN-
g
.In
g
c
2/2
mice, activated T cells
produced sufficient amounts of IFN-
g
to maintain macrophage ef-
fector functions, such that early responses to invading micro-or-
ganisms (in this case Listeria) are effective. Although this T cell-
dependent resistance in
g
c
2/2
mice appears unrelated to the
specific infection, the adaptive immune response in this way may
provide additional nonspecific support of the innate immune
arm. This alternative pathway of T cell-dependent resistance in
innate immunity to L. monocytogenes (see Fig. 8) may have the
evolutionary advantage of protecting the host against secondary
infections.
Which T cell subset(s) is able to mediate the observed IFN-
g
responses, and what mechanisms control the IFN-
g
production? A
variety of T cell subsets are capable of IFN-
g
production, includ-
ing Ag-primed
ab
T cells,
gd
T cells, and the NK1.1
1
ab
T cells
(32, 33). These last two T cell types have also been implicated in
the initial responses to pathogens due to their restricted TCR vari-
ability, which has been hypothesized to interact with nonpolymor-
phic antigenic determinants on the invading micro-organism (34).
However,
gd
T cells and NK-T cells are absent from the peripheral
lymphoid organs of
g
c
2/2
mice (19, 21, 35, 36), thereby ruling out
any essential role of these subsets in the anti-listerial innate re-
sponse. Because anti-CD3 treatment resulted in IFN-
g
production
from
g
c
2/2
T cells stimulated with IL-12 in vitro, TCR cross-
linking is probably also required for optimal innate responses to
Listeria in vivo. However, this hypothesis would predict 1) that
Listeria Ag-specific
ab
T cells are already present in
g
c
2/2
mice
and are rapidly reactivated; 2) that
ab
T cells become activated
oligoclonally by Listeria-produced superantigens; or 3) that
ab
T
cells in
g
c
2/2
mice are constitutively activated through their TCR
by some other mechanism. As we failed to detect cytokine release
(IFN-
g
) from
g
c
2/2
spleen cell cultures in response to HKLM
(data not shown), it is unlikely that preexisting Listeria-specific T
cells or Listeria superantigens are responsible for TCR stimulation
in vivo.
How, then, might residual
g
c
2/2
ab
T cells become nonspe-
cifically activated? Two probable mechanisms involve the crucial
role of
g
c
in responses to IL-2 and IL-4. The IL-2 signaling path-
way appears necessary for peripheral T cell homeostasis, as mice
deficient in IL-2, CD25 (IL-2R
a
), or CD122 (IL-2R
b
) have pe-
ripheral T cell activation and proliferation, autoimmune hemolytic
anemia, and colitis (37–39).
g
c
2/2
mice also develop extramed-
ullary hemopoiesis and colitis (40) (J. P. Di Santo, unpublished
observations), and
g
c
2/2
ab
T cells have an activated/memory
phenotype (40, 41). Therefore, the inability of
g
c
2/2
mice to signal
via IL-2 may result in nonspecific T cell activation, potentially
toward self Ags. In addition, because IL-4 polarizes T cells to the
Th2 phenotype, the absence of IL-4 signaling would result in a
default Th1 differentiation with an IFN-
g
-producing potential. To-
gether, these two mechanisms may explain the ability of naive T
cells from
g
c
2/2
mice to produce IFN-
g
. In a very recent infection
study with an independent
g
c
2/2
mouse strain it has been dem-
onstrated that mutant mice survived Toxoplasma gondii infection
due to sufficient IFN-
g
production. CD4
1
T cells were a source of
the crucial early IFN-
g
production (42), consistent with our pro-
posed mechanism of T cell activation in
g
c
-deficient mice.
In earlier studies, we and others have demonstrated that periph-
eral T cells from
g
c
2/2
mice are poorly functional in response to
mitogen stimulation (19, 21, 43), and we anticipated that
g
c
2/2
mice might be unable to completely eradicate a Listeria infection.
Here we demonstrate that
g
c
2/2
mice establish chronic listeriosis
following infection, which ultimately overwhelms and kills the
mice. These results confirm the requirement for functional T cells
in the specific anti-listerial immune responses that sterilize the
host. Interestingly, the T cell impairment in
g
c
2/2
mice seems to
be rather severe, since the mortality kinetics and histopathology of
Listeria in
g
c
2/2
mice were similar to those in RAG-2-deficient
mice, which completely lack mature T lymphocytes. Thus, the
mature
ab
T cells that develop in the absence of
g
c
appear unable
to respond in an Ag-specific fashion to Listeria or alloantigens
(this report) or to proliferate in response to MHC class II-restricted
hemagglutinin peptides in the context of
g
c
2/2
HNT-TCR trans-
genic mice (5) (Å.A. and J.P.D.S., unpublished observations).
Nevertheless,
g
c
2/2
T cells have an activated phenotype and ac-
cumulate with age (40), suggesting an active, Ag-driven process.
The nature of the stimulatory signals for
g
c
2/2
T cells remains to
be elucidated.
Acknowledgments
We thank M. Held, A. Dorfmu¨ller, and K.-H. Widmann for excellent tech-
nical assistance, Dr. H. Mossmann for organization of the animal facility,
and Drs. D. Guy-Grand and P. Vassalli for rIL-12 and for stimulating
discussions. We are also grateful to Dr. M. Aguet for breeding pairs of
mutant mice.
References
1. Cellin, B. G., and C. V. Broome. 1989. Listeriosis. J. Am. Med. Assoc. 261:1313.
2. Bancroft, G. J. 1993. The role of natural killer cells in innate resistance to in-
fection. Curr. Opin. Immunol. 5:503.
3. Rogers, H. W., C. S. Tripp, and E. R. Unanue. 1995. Different stages in the
natural and acquired resistance to an intracellular pathogen. Immunologist 3/4:
152.
4. Trinchieri, G. 1995. Interleukin-12: a proinflammatory cytokine with immuno-
regulatory functions that bridge innate resistance and antigen-specific adaptive
immunity. Annu. Rev. Immunol. 13:251.
5. Scott, B., R. Liblau, S. Degerman, A. L. Marconi, A. J. Canton, H. O. McDevitt,
and D. Lo. 1997. A role for non-MHC genetic polymorphism in susceptibility to
spontaneous autoimmunity. Immunity 1:72.
6. Harty, J. T., and M. J. Bevan. 1995. Specific immunity to Listeria monocytogenes
in the absence of IFN
g
.Immunity 3:109.
7. Huang, S., W. Hendriks, A. Althage, S. Hemmi, H. Bluethmann, R. Kamijo,
J. Vilcek, R. M. Zinkernagel, and M. Aguet. 1993. Immune response in mice that
lack the interferon-
g
receptor. Science 259:1742.
8. Dai, W. J., W. Bartens, G. Ko¨hler, M. Hufnagel, M. Kopf, and F. Brombacher.
1997. Impaired macrophage listericidal and cytokine activities are responsible for
the rapid death of Listeria monocytogenes-infected IFN-
g
receptor-deficient
mice. J. Immunol. 158:5297.
9. Meraz, M. A., J. M. White, K. C. Sheehan, E. A. Bach, S. J. Rodig, A. S. Dighe,
D. H. Kaplan, J. K. Riley, A. C. Greenlund, D. Campbell, et al. 1996. Targeted
disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in
the JAK-STAT signaling pathway. Cell 84:431.
10. Fehr, T., G. Schoedon, B. Odermatt, T. Holtschke, M. Schneemann,
M. F. Bachmann, T. W. Mak, I. Horak, and R. M. Zinkernagel. 1997. Crucial role
of interferon consensus sequence binding protein, but neither of interferon reg-
ulatory factor 1 nor of nitric oxide synthesis for protection against murine liste-
riosis. J. Exp. Med. 185:921.
11. Buchmeier, N. A., and R. D. Schreiber. 1985. Requirement of endogenous in-
terferon-
g
production for resolution of Listeria monocytogenes infection. Proc.
Natl. Acad. Sci. USA 82:7404.
12. Brombacher, F., and M. Kopf. 1996. Innate versus acquired immunity in listeri-
osis. Res. Immunol. 147:505.
13. Mombaerts, P., J. Arnoldi, F. Russ, S. Tonegawa, and S. H. E. Kaufmann. 1993.
Different roles of
ab
and
gd
T cells in immunity against an intracellular bacterial
pathogen. Nature 365:53.
14. Roberts, A. D., D. J. Ordway, and I. M. Orme. 1993. Listeria monocytogenes
infection in
b
2
-microglobulin-deficient mice. Infect. Immun. 61:1113.
15. Ladel, C. H., I. E. Flesch, J. Arnoldi, and S. H. Kaufmann. 1994. Studies with
MHC-deficient knock-out mice reveal impact of both MHC I- and MHC II-
dependent T cell responses on Listeria monocytogenes infection. J. Immunol.
153:3116.
5605The Journal of Immunology

16. Sugamura, K., H. Asao, M. Kondo, N. Tanaka, N. Ishii, M. Nakamura, and
T. Takeshita. 1995. The common
g
-chain for multiple cytokine receptors. Adv.
Immunol. 59:225.
17. Noguchi, M., H. Yi, H. M. Rosenblatt, A. H. Filipovich, S. Adelstein,
W. S. Modi, O. W. McBride, and W. J. Leonard. 1993. Interleukin-2 receptor
g
chain mutation results in X-linked severe combined immunodeficiency in hu-
mans. Cell 73:147.
18. Fischer, A., M. Cavazzana-Calvo, G. de Saint Basile, J. P. deVillartay, J. P.
DiSanto, C. Hivroz, F. Rieux-Laucat, and F. Le Deist. 1997. Naturally occurring
primary deficiencies of the immune system. Ann. Rev. Immunol. 15:93.
19. Cao, X., E. W. Shores, L. J. Hu, M. R. Anver, B. L. Kelsall, S. M. Russell,
J. Drago, M. Noguchi, A. Grinberg, and E. T. Bloom. 1995. Defective lymphoid
development in mice lacking expression of the common cytokine receptor
g
chain. Immunity 2:223.
20. Andersson, A., S. Grunewald, A. Duschl, A. Fischer, and J. P. DiSanto. 1997.
Murine macrophage development in the absence of the common
g
chain: defining
complexes responsible for IL-4 and IL-13 signaling. Eur. J. Immunol. 27:1762.
21. DiSanto, J. P., W. Muller, D. Guy-Grand, A. Fischer, and K. Rajewsky. 1995.
Lymphoid development in mice with a targeted deletion of the interleukin 2
receptor
g
chain. Proc. Natl. Acad. Sci. USA 92:377.
22. Shinkai, Y., G. Rathbun, K. P. Lam, E. M. Oltz, V. Stewart, M. Mendelsohn,
J. Charron, M. Datta, F. Young, A. M. Stall et al. 1992. RAG-2-deficient mice
lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell
68:855.
23. Dai, W. J., G. Ko¨hler, and F. Brombacher. 1997. Both innate and acquired im-
munity to Listeria monocytogenes infection are increased in IL-10 deficient mice.
J. Immunol. 158:2259.
24. Moloney, W. C., K. McPherson, and L. Fliegermanz. 1960. Esterase activity in
leucocytes demonstrated by the use of naphthol-AS-D-chloroacetase substrate.
J. Histochem. Cytochem. 8:200.
25. Dieterle, R. R. 1924. Method for demonstration of Spirochaeta pallida in single
microscopic sections. Arch. Neurol. Psychol. 73.
26. Zhan, Y., and C. Cheers. 1995. Differential induction of macrophage-derived
cytokines by live and dead intracellular bacteria in vitro. Infect. Immun. 63:720.
27. Rothe, J., W. Lesslauer, H. Lotscher, Y. Lang, P. Koebel, F. Kontgen, A. Althage,
R. Zinkernagel, M. Steinmetz, and H. Bluethmann. 1993. Mice lacking the tu-
mour necrosis factor receptor 1 are resistant to TNF-mediated toxicity but highly
susceptible to infection by Listeria monocytogenes. Nature 364:798.
28. Pfeffer, K., T. Matsuyama, T. M. Kundig, A. Wakeham, K. Kishihara,
A. Shahinian, K. Wlegmann, P. S. Ohashi, M. Kronke, and T. W. Mak. 1993.
Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to en-
dotoxic shock, yet succumb to L. monocytogenes infection. Cell 73:457.
29. Endres, R., A. Luz, H. Schulze, H. Neubauer, A. Futterer, S. M. Holland,
H. Wagner, and K. Pfeffer. 1997. Listeriosis in p47
phox2/2
and TRp55
2/2
mice:
protection despite absence of ROI and susceptibility despite presence of RNI.
Immunity 7:419.
30. Bancroft, G. J., R. D. Schreiber, and E. R. Unanue. 1991. Natural immunity: a
T-cell-independent pathway of macrophage activation, defined in the SCID
mouse. Immunol. Rev. 124:5.
31. Trinchieri, G. 1995. Interleukin-12 and interferon-
g
: do they always go together?
Am. J. Pathol. 147:1534.
32. Billiau, A. 1996. Interferon-
g
: biology and role in pathogenesis. Adv. Immunol.
62:61.
33. Hayakawa, K., B. T. Lin, and R. R. Hardy. 1992. Murine thymic CD4
1
T cell
subsets: a subset (Thy0) that secretes diverse cytokines and overexpresses the
V
b
8 T cell receptor gene family. J. Exp. Med. 176:269.
34. Janeway, C. J. 1989. Approaching the asymptote? Evolution and revolution in
immunology. Cold Spring Harb. Symp. Quant. Biol. 1:1.
35. Lantz, O., L. I. Sharara, F. Tilloy, A. Andersson, and J. P. DiSanto. 1997. Lineage
relationships and differentiation of natural killer (NK) T cells: intrathymic selec-
tion and interleukin (IL)-4 production in the absence of NKR-P1 and Ly49 mol-
ecules. J. Exp. Med. 185:1395.
36. Malissen, M., P. Periera, D. Gerber, B. Malissen, and J. P. DiSanto. 1997. The
common cytokine receptor gamma chain controls survival of
gd
T cells. J. Exp.
Med. 186:1277.
37. Sadlack, B., R. Kuhn, H. Schorle, K. Rajewsky, W. Muller, and I. Horak. 1994.
Development and proliferation of lymphocytes in mice deficient for both inter-
leukins-2 and -4. Eur. J. Immunol. 24:281.
38. Suzuki, H., T. M. Kundig, C. Furlonger, A. Wakeham, E. Timms, T. Matsuyama,
R. Schmits, J. J. Simard, P. S. Ohashi, and H. Griesser. 1995. Deregulated T cell
activation and autoimmunity in mice lacking interleukin-2 receptor
b
.Science
268:1472.
39. Willerford, D. M., J. Chen, J. A. Ferry, L. Davidson, A. Ma, and F. W. Alt. 1995.
Interleukin-2 receptor
a
chain regulates the size and content of the peripheral
lymphoid compartment. Immunity 3:521.
40. Sharara, L. I., A. Andersson, G. D. Guy, A. Fischer, and J. P. DiSanto. 1997.
Deregulated TCR
ab
T cell population provokes extramedullary hematopoiesis
in mice deficient in the common
g
chain. Eur. J. Immunol. 27:990.
41. Nakajima, H., E. W. Shores, M. Noguchi, and W. J. Leonard. 1997. The common
cytokine receptor
g
chain plays an essential role in regulating lymphoid ho-
meostasis. J. Exp. Med. 185:189.
42. Scharton-Kerston, T., H. Nakajima, G. Yap, A. Sher, and W. J. Leonhard. 1998.
Infection of mice lacking the common cytokine receptor
g
-chain (
g
c) reveals an
unexpected role for CD4
1
T lymphocytes in early IFN-
g
-dependent resistance to
Toxoplasma gondii.J. Immunol. 160:2565.
43. DiSanto, J. P., G. D. Guy, A. Fisher, and A. Tarakhovsky. 1996. Critical role for
the common cytokine receptor
g
chain in intrathymic and peripheral T cell se-
lection. J. Exp. Med. 183:1111.
5606 L. monocytogenes INFECTION IN
g
c-DEFICIENT MICE