| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Immunology, University of Glasgow, Western Infirmary, Glasgow, United Kingdom; and
Department of Microbiology and Immunology, University of Gothenburg, Gothenburg, Sweden
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Of the factors that we
investigated further only IL-12 appeared to be essential for the
immunogenicity of ISCOMS, as IL-6- and inducible nitric oxide synthase
knockout (KO) mice developed normal immune responses to OVA in ISCOMS,
whereas these responses were markedly reduced in IL-12KO mice. The
recruitment of peritoneal exudate cells following an injection of
ISCOMS was impaired in IL-12KO mice, indicating a role for IL-12 in
establishing the proinflammatory cascade. Thus, ISCOMS prime
Ag-specific immune responses at least in part by activating
IL-12-dependent aspects of the innate immune
system. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have used immune-stimulating complexes (ISCOMS)3 containing the saponin adjuvant Quil A for this purpose. ISCOMS are rigid, cage-like structures that form spontaneously in the presence of cholesterol, phosphatidylcholine, and Quil-A (8). We and others have shown that proteins incorporated in ISCOMS become highly immunogenic in vivo, inducing a wide range of immune effector responses, including delayed-type hypersensitivity (DTH), class I MHC-restricted CTL activity, serum Ab production, T cell proliferation, and secretion of both Th1- and Th2-dependent cytokines (2, 9, 10, 11, 12, 13). In addition, administration of ISCOMS by the oral route stimulates a full range of these systemic and immune responses as well as local secretory IgA- and cell-mediated immunity (14).
Little is known about the properties of ISCOMS that underlie their adjuvant effects. Most adjuvants appear to act by recruiting different components of the innate immune response, and previous studies have shown that neutrophil recruitment (15) as well as production of IL-1 from splenocytes cultured in vitro (16), serum IL-6 (17), and serum IL-12 (18), are all induced by ISCOMS. However, the evolution of these responses has not been studied in detail, and the effects of ISCOMS on other aspects of the innate immune system, such as macrophages, are not known. More significantly, it is not known whether activation of any of these innate responses is essential for the induction of a specific adaptive immune response. Here we detail the consequences of local administration of ISCOMS on the innate immune system and have examined these effects on the generation of Ag-specific immunity using gene-targeted knockout (KO) mice. Our results show that ISCOMS recruit a wide range of inflammatory cells and mediators and that IL-12 derived from adherent cells is of central importance for the immunogenicity of ISCOMS.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SPF female C57BL/6 mice were purchased from Harlan Olac (Bicester, U.K.) and female BALB/c mice were bred at the University of Glasgow (Glasgow, Scotland). The p40 IL-12-deficient (IL-12KO) BALB/c and C57BL/6 mice were obtained from Dr. J. Magram (Roche Pharmaceuticals, Piscataway, NJ). Inducible nitric oxide (NO) synthase deficient (iNOSKO) MF1 mice were a gift from Prof. F. Y. Liew, Department of Immunology, University of Glasgow, Glasgow, Scotland. Female IL-6KO SVJ mice were obtained from Prof. J. Alexander (University of Strathclyde, Glasgow, Scotland). All these mice were maintained at the University of Glasgow except the IL-6KO mice, which were bred and maintained at the University of Strathclyde and were first used at 6–8 wk of age. Control mice were age-matched mice of the appropriate wild-type strain. Regular screening of the mouse strains showed these to be negative for all pathogens, including MHV.
Preparation and administration of OVA ISCOMS
ISCOMS containing palmitified OVA were prepared as described
previously (2) using phosphatidylcholine, cholesterol, and Quil-A
(Spikoside, a gift from Prof. B. Morein (Swedish University of
Agricultural Sciences, Uppsala, Sweden)). The integrity of the OVA
ISCOMS was checked by electron microscopy, and the protein content was
assessed by Bradford staining (Bio-Rad, Hemel Hempstead, U.K.). The
endotoxin content of ISCOMS was quantified using the E-Toxate Reagents
Kit (Sigma) and was <0.5 µg/ml (equivalent to 2 ng/injection dose).
The OVA ISCOMS used in this study were 
40 nm in diameter and
contained protein and Quil-A at a ratio of 10:1. ISCOMS contained the
equivalent of 0.5 µg of Quil A and 5 µg of protein and were
administered either i.p. in a volume of 200 µl or s.c. in a
volume of 50 µl. Control mice received saline alone, and peritoneal
exudate cells (PEC) were removed 3.5 h later.
Isolation of PEC
PEC were obtained after sacrifice of mice by injection of 6 ml of prewarmed lavage medium consisting of RPMI 1640 (Life Technologies, Paisley, U.K.) containing 5% heat-inactivated FCS (Life Technologies). The abdomen was massaged gently, and the lavage fluid was aspirated with a 19-gauge needle. PEC were washed in RPMI, counted, and stored on ice until required. Individual cell counts were performed on PEC removed from three mice per group, but PEC phenotyping and culture were performed using pooled PEC from five animals.
Analysis of PEC by flow cytometry
Aliquots of 106 PEC were washed twice in PBS containing 2% FCS (Life Technologies) and 0.02% sodium azide (FACS buffer) and stained for 45 min on ice in a total volume of 50 µl using PE-anti-CD4 Ab (PharMingen, San Diego, CA), FITC-anti-CD8 Ab (PharMingen), FITC-anti-CD25 Ab (PharMingen), biotinylated anti-B220 Ab (PharMingen), biotinylated anti-F4/80 Ab (Serotec, Kidlington, U.K.), FITC-anti-I-Ab,d Ab (PharMingen), PE-anti-CD11c Ab (PharMingen), FITC-anti-B7.1 Ab (PharMingen), FITC-anti-B7.2 Ab (PharMingen), and rat anti-mouse whole neutrophil Ab (Serotec). Second-step reagents were PE-rabbit anti-rat IgG (Serotec), PE-streptavidin (Vector, Oxford, U.K.), or FITC-streptavidin (Vector). Stained PEC were analyzed on a Becton Dickinson FACScan flow cytometer (Becton Dickinson, Mountain View, CA). Data from 10,000 cells were acquired and analyzed using LYSIS II software (Becton Dickinson).
Histological examination of PEC following cytospins
Cytospins were prepared of aliquots of 5 x 105 PEC spun onto glass microscope slides at 200 rpm for 5 min using a cytocentrifuge (Shandon, Runcorn, U.K.). Cytospins were fixed in alcohol and stained with Geimsa (Sigma). The number of mast cells per field was counted at x400 magnification using a Nikon Labophot microscope (Nikon, Melville, NY). Average mast cell numbers per field were calculated after counting 10 fields. The total mast cell number per mouse was calculated using the following formula: total mast cells = (average mast cells per field/PEC per field) x total PEC per mouse.
Measurement of inflammatory cytokine production in vitro
PEC (4 x 106) were cultured in vitro in
24-well plates (Costar, Northumbria Biologicals, Cramlington, U.K.) in
1 ml of RPMI 1640 containing 10% heat-inactivated FCS, 100 U/ml
penicillin, 100 µg/ml streptomycin, 50 µg/ml fungizone (all from
Life Technologies), and 0.05 µM 2-ME (Sigma, Poole, U.K.) in the
presence or the absence of 10 µg/ml LPS (Salmonella
enteritidis, Sigma) plus 20 ng/ml recombinant murine IFN-
(PharMingen) or with 10 µg/ml Con A. Supernatants were removed after
48 h of culture at 37°C in 5% CO2 in air,
centrifuged for 5 min at 13,000 x g, and frozen at
-20°C until required. Cytokine levels in PEC culture supernatants
were assayed by sandwich ELISA in 96-well Immulon 4 ELISA plates
(Dynatech, Billingshurst, U.K.) using appropriate anti-cytokine Ab
pairs for IL-6, IFN- (both from PharMingen), or IL-1
(Genzyme,
Cambridge, MA). To measure immunoreactive p40 and p70 IL-12 levels, an
mAb specific for the p40 subunit (C15.6, Genzyme) was used as the
capture Ab, and a polyclonal rabbit anti-mouse IL-12 Ab raised
against and specific for the mouse p70 heterodimer (a gift from Dr.
F.-P. Huang, Department of Immunology, University of Glasgow) was used
as the detecting Ab (19, 20). The concentration of cytokine present in
supernatants was determined spectrophotometrically at 630 nm on an
MR5000 microtiter plate reader (Dynatech) after incubation with the
chromophore 3,3,5,5'-tetramethylbenzidine (TMB) and
H2O2 (Dynatech). Samples were assayed in
triplicate, and the cytokine concentration was determined by
extrapolation from a standard curve generated by serial dilution of the
appropriate recombinant murine cytokine (PharMingen).
Measurement of NO and reactive oxygen intermediates
The concentration of NO present in PEC culture supernatants was determined by detection of nitrite, a stable metabolite of NO, by the Griess reaction. Briefly, the Griess reagent was made by mixing equal volumes of 5% orthophosphoric acid containing 1% sulfanilamide (Sigma) and 0.1% napthylethylenediamine (Sigma) and was kept at 4°C until required. Triplicates of 50 µl of culture supernatant or a standard curve of serially diluted sodium nitrite were added to 96-well Immulon 4 microtiter plates (Dynatech), and 50 µl of the Griess reagent was then added. Any color generated was measured spectrophotometrically at 570 nm using a MR5000 microtiter plate reader (Dynatech). Reactive oxygen intermediates (ROI) released during the respiratory burst were measured by a chemiluminescence assay (21). Briefly, PEC were washed in HBSS (Life Technologies) without phenol red and warmed to 37°C in a water bath. One hundred microliters of 0.1 mM 5-amino-2,3-dihydro-1,4-phthalazinedione solution (Luminol; Sigma) was added to 400 µl of the warmed HBSS containing 2 x 106 PEC, and the cells were stimulated by addition of 1 µg/ml PMA (Sigma). The suspensions were mixed at 37°C in the water bath, and any chemiluminescence was measured using a 1250 Luminometer (LKB Wallac, Turku, Finland) every 10 s for a total of 10 min. All measurements were expressed as millivolts.
Assessment of Ag-specific immune responses
OVA-specific immune responses were assessed in mice following
s.c. immunization with OVA ISCOMS into a rear footpad or given i.p. as
described previously (2). DTH responses were measured after s.c.
challenge in the opposite rear footpad with 10 µg of heat-aggregated
OVA 7 days after s.c. immunization. OVA-specific proliferative
responses were assessed in parallel in the draining popliteal lymph
nodes. Briefly, 2 x 105 splenocytes were cultured in
96-well microtiter plates (Costar) in the presence or the absence of 1
mg/ml OVA for 5 days at 37°C and 5% CO2 in air. Cultures
were pulsed with 1 µCi/well [3H]thymidine (West of
Scotland Radionucleotide Dispensary, Western Infirmary, Glasgow, U.K.)
and harvested onto glass-fiber filter mats (Wallac) 16 h later,
[3H]thymidine incorporation into DNA was measured using a
Betaplate counter (Wallac). OVA-specific CTL activity was determined by
the ability of spleen cells taken 7 days following i.p. immunization to
lyse OVA-bearing EG7.OVA target cells as described previously (2).
Briefly, effector spleen cells from immunized animals were restimulated
in vitro with mitomycin C-treated EG7.OVA cells in culture medium
containing 5% NCTC 135 (Life Technologies) and were cultured for 5
days at 37°C. After culture, the effector cells were washed
thoroughly and incubated with 51Cr-labeled EG7.OVA or
non-OVA-expressing EL4 target cells at various E:T cell ratios in a
volume of 200 µl of RPMI 1640 and 5% FCS in V-bottom microtiter
plates (Costar) for 4 h at 37°C. One hundred microliters of
supernatant was then removed and assayed for the presence of
51Cr. OVA-specific CTL activity was calculated by the
following formula:
 |
Statistics
All results are expressed as the mean ± 1 SD and were compared using Student’s t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To study the ability of ISCOMS to stimulate the innate immune
response, we examined the recruitment of inflammatory cells after i.p.
injection of a dose of ISCOMS containing 5 µg of OVA and 0.5 µg of
Quil A. We chose this approach because the peritoneum offers an
anatomically defined site that is readily accessible and because our
previous studies had used this protocol to induce Ag-specific immune
responses in vivo. Within 3.5 h of administration of
ISCOMS, there was a 50% decrease in the total number of PEC compared
with that in saline-injected controls, and the numbers of PEC remained
depressed until 48 h after ISCOMS injection, at which time there
was a marked and significant increase in the number of PEC to levels 3
times those in control mice. The number of PEC was still significantly
elevated by 72 h (Fig. 1
a).
|
b). The
decrease in total PEC early after the injection of ISCOMS was
accompanied by an increase in the number of polymorphonuclear
leukocytes at 3.5 h (data not shown) and 7 h (Fig. 1c). Polymorphs were still present at 24 h (data not
shown), but by 48 h polymorphonuclear cell numbers had returned to
control levels, and there were now increased numbers of large
mononuclear cells, many with the appearance of activated macrophages
(Fig. 1d). In parallel, there were increased numbers of
lymphoid cells, many of which were lymphoblastoid in appearance (Fig. 1d). One further aspect of the inflammatory
infiltrates was an increase in the number of mast cells after the
injection of ISCOMS. This was seen as early as 3.5 h, when a
substantial number of these cells had degranulated (Fig. 2). A second rise in mast cell number was
found at 48 h, although no degranulation had occurred at this
time.
|
To analyze the inflammatory infiltrate in more detail, PEC were
stained for expression of markers specific for neutrophils, macrophages
(M
), dendritic cells (DC), and T and B lymphocytes. Consistent with
the histological appearances described above, the decrease in PEC
numbers early after injection of ISCOMS was accompanied by a large
increase in the absolute numbers and proportions of neutrophils, which
peaked at 3.5 h and declined rapidly thereafter (Fig. 3a). Only low numbers of M,
as identified by expression of the F4/80 marker, were found in control
PEC, but their numbers began to increase by 7 h, reaching a peak
at 48 h and then declining (Fig. 3a). Again this
confirmed the microscopic analysis of cytospins. CD4+
lymphocytes were the predominant cell type in normal PEC, but their
numbers declined rapidly over the first 7 h after ISCOMS injection
before rising back toward control levels, reaching a peak at 48 h
(Fig. 3b). The numbers of CD8+ T cells and
B220+ B cells followed a similar pattern, with a rapid
early decrease followed by recovery to levels that were eventually
double the control values (Fig. 3, b and c). This
peaked at 48 h for B cells and at 72 h for CD8+ T
cells (Fig. 3b). The numbers of DC, as determined by
expression of CD11c surface marker, were low in control PEC, but after
a further drop at early time points they increased to a peak at 72
h (Fig. 3d). These CD11c+ cells were virtually
all MHC class II+, which is consistent with their
identification as mature DC (data not shown).
|
, respectively. Small numbers of MHC class II+ cells
were found in control PEC, the majority of which were B cells (data not
shown). By 7 h after the injection of ISCOMS, there were virtually
no class II+ PEC, presumably reflecting the decline in the
numbers of B cells and M at this time. A rapid increase in MHC class
II+ PEC then occurred, reaching a peak at 48 h (data
not shown). This was paralleled by the reappearance of B cells and the
de novo recruitment of MHC class II+ F4/80+
M, whose numbers increased dramatically from 7 until 48 h, when
25% of M were MHC class II+ (Fig. 3E). By
72 h few MHC class II+ M remained. Very few
CD25+ cells were found in control PEC, and their numbers
decreased further immediately after injecting ISCOMS. At 48 h the
numbers of CD25+ PEC were twice the control values (Fig. 3D), a pattern that was paralleled by the appearance of
substantial numbers of CD25-expressing CD4+ T cells.
At 48 h 26% of CD4+ T cells were CD25+
(Fig. 3F). Similar results were obtained for CD25 expression
on CD8+ T cells, although the numbers were much smaller
(data not shown). Thus, injection of ISCOMS induces a rapid, local
recruitment of inflammatory cells, consisting initially of neutrophils
and mast cells, followed by M and DC, and later by both T and B
lymphocytes. Once recruited, many of these cells become activated, as
indicated by their increased expression of MHC class II and CD25. Induction of local cytokine production by administration of ISCOMS
We next assessed the functional consequences of the cellular
recruitment stimulated by an i.p. injection of ISCOMS. We chose to
examine IL-1, IL-6, and IL-12 as products of activated M and IFN-
as a lymphocyte product. PEC from control mice showed some spontaneous
production of IL-6, but no IL-1 and very low amounts of
immunoreactive IL-12 (Fig. 4). The
spontaneous production of IL-1 was significantly enhanced 48 h
after ISCOMS injection, when PEC produced >20 times that of controls,
and then fell to undetectable levels (Fig. 4A). The
spontaneous production of IL-6 became undetectable early after the
injection of ISCOMS, but increased rapidly after 24 h, reaching
levels more than double the control values at 48 and 72 h (Fig. 4B). A similar pattern of ISCOMS-induced enhancement of
IL-1 and IL-6 production was seen when PEC were stimulated with
IFN- and LPS in vitro (data not shown). No TNF- production was
found in any PEC preparation, either spontaneously or after stimulation
with IFN- and LPS (data not shown). Spontaneous IL-12 production
rose gradually from 24 h after administration of ISCOMS,
plateauing after 48 h at levels markedly above those of control
PEC (Fig. 4C). After stimulation in vitro with IFN- plus
LPS, production of IL-12 declined rapidly after injection of ISCOMS,
but rose sharply to a large peak at 72 h, when levels were twice
the control values (Fig. 4C).
|
from control PEC, but
low levels were detected after stimulation with Con A (Fig. 4D). Significant levels of spontaneous IFN- production
were detected 24 h after the injection of ISCOMS but not
thereafter (data not shown). Con A-stimulated PEC showed a small, but
significant, increase in IFN- production 3.5 h following
injection with ISCOMS. This fell to the control level at 7 h, but
increased again to a peak at 24 h, when IFN- production was
twice that of control PEC (Fig. 4C). The production of all
cytokines was dependent on the presence of M, as removal of adherent
cells from PEC cultures completely ablated IL-12, IL-1, and IL-6
production (Table I) and significantly
reduced IFN- production (data not shown). Thus, the local
infiltration by M and T cells with activated phenotypes is
accompanied by the production of cytokines associated with these cells.
|
In view of the marked recruitment of activated phagocytic cells
such as M and neutrophils, we next examined the production of two
other characteristic mediators of such cells, NO and ROI. There
was very little spontaneous NO production by control PEC, but NO
production was stimulated by culture with IFN- and LPS (Fig. 5A). Immediately after
injection with ISCOMS, IFN-- and LPS-stimulated production of NO
fell, but recovered from 7 h onward, reaching a peak at 48 h
that was significantly above control levels (Fig. 5A). In
addition, spontaneous production of NO became detectable 24 h
after the injection of ISCOMS, and substantial amounts were still
present at 48 h (Fig. 5A).
|
B). Thereafter, there was a decline in ROI production, and
by 72 h after injection of ISCOMS, ROI production had returned to
control levels. Thus, the recruitment of neutrophils and M by ISCOMS
is accompanied by increased production of ROI and NO, respectively. Role of ISCOMS-induced inflammatory mediators in Ag-specific immunity
The experiments described above show that ISCOMS recruit a wide range of inflammatory cells and mediators to the site of injection. We therefore went on to explore whether any of these played an essential role in the induction of Ag-specific T cell responses by ISCOMS-associated Ag. To do this we immunized appropriate gene-targeted KO mice using immunization regimens that we had previously shown to be optimal (11). Systemic DTH and OVA-specific CTL activity were used as measures of CD4+ and CD8+ T cell function in vivo, respectively, while Ag-specific proliferation was used to determine immune responsiveness in vitro.
IL-6KO mice (Fig. 6) and iNOSKO mice (Fig. 7) generated normal immune responses after immunization
with OVA ISCOMS, with DTH responses (Figs. 6A and
7A), OVA-specific proliferation (Figs. 6B and
7B), and CTL activity (Fig. 6C) identical with
those in appropriate wild-type control animals. However, p40 IL-12KO
mice had impaired induction of Ag-specific immune responses both in
vivo and in vitro (Fig. 8). These mice had a dramatic
reduction in their DTH responses after challenge with Ag, with
virtually no response above the background (Fig. 8A). In
addition, both Ag-specific proliferative responses (Fig. 8B)
and OVA-specific CTL activity (Fig. 8C) were significantly
reduced compared with those in wild-type mice, with responses on the
order of 50–60% those of normal animals.
|
|
|
Absence of innate responses in IL-12KO mice
To investigate the relationship between the decreased specific immunity in IL-12KO mice and the innate immune response, we assessed the ability of an i.p. injection of ISCOMS to stimulate peritoneal infiltration in these animals.
As we found previously, i.p. injection of ISCOMS into normal B6 mice
lead to recruitment of PEC, with a peak in PEC numbers occurring at
48 h and returning to control levels by 72 h after injection
(Fig. 9). In contrast to the decrease in
PEC numbers that occurred at early times in wild-type mice, there was a
small increase in the numbers of PEC in ISCOMS-injected IL-12KO mice at
3.5 h after injection. However, PEC numbers then decreased and
remained around those seen in control mice at all subsequent time
points (Fig. 9). In addition, PEC from IL-12KO mice did not show the
increase in production of inflammatory mediators observed in wild-type
mice, with IL-1, IL-6, IFN-, and NO levels all remaining at
control levels throughout the experiment (data not shown.)
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and lymphocytes. Many of these cells are activated, as evidenced
by the expression of surface activation markers and the production of
cytokines and other mediators. However, the majority of these factors
were not essential for the immunogenicity of ISCOMS in vivo, with only
the production of immunoreactive IL-12 appearing to be of major
importance.
The local inflammatory response induced by ISCOMS consisted of an
initial emigration of neutrophils, followed by M and DC, and
eventually lymphoid cells. Neutrophils are not normally present in the
peritoneum, but following the administration of ISCOMS, a large
increase in the numbers of polymorphonuclear cells in PEC was observed
histologically, and phenotypic studies confirmed that these were mostly
neutrophils. Our current studies extend previous observations that
neutrophils are recruited to a site of immunization with ISCOMS (15, 22) by showing that these cells are probably activated in situ. This
was shown by the fact that the peak infiltration by neutrophils
coincided with a peak in ROI production by PEC at a time when other
potential ROI-producing cells were virtually absent. The neutrophilia
was accompanied by recruitment of mast cells and morphological evidence
of mast cell degranulation. Mast cells accumulate early in other forms
of acute inflammation, and mast cell products have been implicated as
being important for the induction of several cell-mediated immune
responses in vivo, such as DTH and contact sensitivity (23). The role
of mast cells in the induction of specific immunity by ISCOMS remains
to be clarified, although they could act as an important source of
vasoactive mediators, including histamine or proinflammatory cytokines
such as preformed TNF- (24). The factors that stimulate the
recruitment and activation of mast cells by ISCOMS are also unknown,
but the rapidity of the response in naive animals clearly precludes a
role for Ag-specific IgE-mediated effects.
M and DC were next to appear in the recruited PEC, with many of the
new M being activated, as judged by their morphological appearance
and their expression of MHC class II Ags. In addition, there was
increased production of a number of inflammatory mediators
characteristic of activated M, including IL-1, IL-6, immunoreactive
IL-12, and NO. The recruited DC were also class II MHC+,
confirming their maturity, but their activation status remains to be
determined. Around the time these potential APCs appeared in PEC, there
was infiltration by B and T cells of both CD4+ and
CD8+ subsets. Substantial numbers of the T cells were
activated, as shown by their expression of the IL-2R molecule, and
the recruitment of CD8+ T cells in response to ISCOMS is
especially interesting in the light of the unusual ability of ISCOMS to
prime MHC class I-restricted Ag-specific CTL in vivo (2, 11, 12, 13, 14).
Together, these results show that ISCOMS recruit many components of the
innate and adaptive immune responses, including several populations of
potential APC, such as DC, B cells, and activated M. In preliminary
studies we have found that phagocytic M are important for induction
of class I MHC-restricted CTL responses by ISCOMS (25), a finding
consistent with recent evidence that the phagocytic uptake of Ag may be
selective for the class I processing pathway (26). In addition, DC have
recently been associated with class I MHC-restricted presentation of
Ag, particularly after ingestion of apoptosed M (27). It will be
interesting to determine the interaction among all these cell types in
the induction of class I and II MHC-restricted Ag-specific T cells by
ISCOMS.
The production of cytokines and other inflammatory mediators by
inflammatory cells is now recognized as being critical for the
induction of an effective primary immune response, and the presence of
these mediators can determine the type of Ag-specific effector response
that is generated (28). ISCOMS stimulated the production of a wide
range of inflammatory mediators, including IL-1, IL-6, IL-12, and
IFN-. The appearance of the proinflammatory cytokines IL-1 and IL-6
is consistent with the acute inflammatory reaction provoked by ISCOMS
and confirms previous reports of stimulation of both IL-1 and IL-6
production by ISCOMS (16, 17). However, previous studies were unable to
detect IL-1 production in vivo (29), and the IL-6 production
demonstrated in vivo was not examined at the cellular level (17). Both
these cytokines are important components of the innate immune response
and are often found together with TNF- during the early stages of
inflammatory reactions to, for example, bacterial infection (30).
However, as reported by others (17), we have been unable to detect the
presence of TNF- in supernatants of ISCOMS-stimulated PEC. This is
somewhat surprising given the high levels of the other mediators, and
we are currently examining TNF- production at even earlier time
points using more sensitive molecular techniques.
IL-6KO and iNOSKO mice had normal T cell-mediated immune responses to
OVA ISCOMS despite the known role of these mediators in other models of
immunity (31, 32, 33) and the increased levels of their production after
injection of ISCOMS. However, IL-12 appeared to be critically important
for the adjuvant effects of ISCOMS. The production of IL-12 by
ISCOMS-recruited PEC was greatly enhanced, extending recent findings of
elevated serum IL-12 levels after the injection of ISCOMS (18). In
addition, Ag-specific immune responses to ISCOMS were greatly reduced
in IL-12KO mice, and there was no recruitment of PEC or induced
secretion of inflammatory mediators by PEC from these mice. This
complete lack of cellular recruitment indicates a critical role for
IL-12 in initiating the inflammatory cascade, perhaps reflecting an
absence of chemotactic factors such as IL-8, whose production may be
dependent on IL-12 (34). The defective DTH and CTL responses to ISCOMS
in IL-12KO mice were not particularly surprising in view of previous
studies using these animals and of the known effects of IL-12 on the
production of IFN- (35) and differentiation of CTL (36). Ag-specific
proliferative responses were also reduced in ISCOMS-immunized IL-12KO
mice, but as IL-12 is known to regulate the expression of the IL-2R
and ß subunits (37, 38), and we have found that IL-12KO mice do not
develop CD25+ CD4 lymphocytes following administration of
ISCOMS (data not shown), the reduced Ag-specific proliferation in
IL-12KO mice could simply be due in part to IL-2 unresponsiveness in
OVA-specific CD4+ lymphocytes. More recent work indicates
that most other aspects of mucosal and systemic immunity are also
defective in these animals (D.G., et al., manuscript in preparation).
These findings extend recent studies showing diminished Ag-specific
serum IgG Ab responses following administration of neutralizing
anti-IL-12 p70 Ab (18), indicating that IL-12 plays a more
generalized role in the adjuvant effects of ISCOMS. The basis of this
remains to be determined, but it does not appear to reflect its role in
directing IFN- production, as we have shown previously that IFN-
receptor KO mice respond normally to immunization with OVA ISCOMS (39).
Although this seems paradoxical in view of our findings that IFN-
production by PEC is increased by ISCOMS, it is consistent with other
systems in which undefined IFN--independent effects of IL-12 have
been described (40). The source of ISCOMS-induced IL-12 remains to be
determined, but its production required the presence of adherent cells
and correlated with the recruitment of M and DC, cell types that
both can produce this cytokine (35). It would be of interest to
determine whether this is a direct effect of ISCOMS on the relevant
accessory cells or if it is secondary to CD40-CD40 ligand interactions
known to be important for the induction of IL-12 in other systems. It
should be noted that ISCOMS adjuvant shows a selective association with
and dependence on IL-12 production. It is generally accepted that
administration of Ag alone in vivo does not stimulate IL-12 production,
while cholera toxin, an agent with mucosal and systemic adjuvant
properties similar to those of ISCOMS, does not induce IL-12 readily,
and immune responses to OVA and cholera toxin are entirely normal in
IL-12KO mice (D.G. et al., manuscript in preparation). The unusual
propensity for ISCOMS to stimulate IL-12-dependent immune responses is
not limited to predictable responses such as IFN- production and CTL
activity, as our current studies indicate a more global defect in Ab
and cytokine production in IL-12KO mice (unpublished observations). The
basis for these remains to be determined. Interestingly, the peak of
IL-12 production was proceeded by an early spike in IFN- levels,
supporting other recent evidence that IFN- may be required for
initiation of significant levels of IL-12 production (41).
Nevertheless, it is important to note that immune responses were only
partially reduced in IL-12KO mice. The innate responses measured in
these mice were absent, potentially indicating a role for additional
factors in generating the residual immunity. Whether this reflects an
ability of some of the factors investigated here to compensate for the
absence of IL-12 or a role for novel factors such as IL-18 (42, 43)
remains to be determined.
The ways in which ISCOMS can stimulate the innate immune system are currently being explored, but we consider that this reflects delivery of the adjuvant effects of the adjuvant Quil A to proinflammatory cells in the form of a lipophyllic particle. Free Quil A reproduces patterns of specific and nonspecific effects similar to those found with intact ISCOMS, including the induction of mucosal and systemic responses as well as the recruitment and activation of local inflammatory cells (44). However, we and others have shown that 10- to 20-fold more free Quil A is required to induce equivalent responses to those found using Quil A in ISCOMS (45, 46). Thus, we propose that the particulate nature of ISCOMS targets Quil efficiently to cells of the immune system. The chemical basis of these effects of Quil A and ISCOMS is unclear, but should now be examined using individual fractions of Quil A that differ in their ability to stimulate inflammation and specific immunity (17).
In conclusion, our results show that the ability of ISCOMS to induce a wide range of Ag-specific immune responses is paralleled by the activation of a cascade of innate immune responses. This is consistent with other evidence that the best adjuvants are those that mimic the ability of pathogens to activate the innate immune system. However, our study also reveals the complexity of the resulting nonspecific signals that are generated, with many overlapping and redundant mechanisms employed, only a few of which may play an essential role in the development of an Ag-specific immune response. Elucidating and targeting these mechanisms will be important in the design of a successful vaccine.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Allan McI. Mowat, Department of Immunology, University of Glasgow, Western Infirmary, Glasgow, U.K. G11 6NT. E-mail address:
3 Abbreviations used in this paper: ISCOMS, immune-stimulating complexes; DTH, delayed type hypersensitivity; KO, knockout; IL-12KO, IL-12 deficient; NO, nitric oxide; iNOS, inducible NO synthase; iNOSKO, iNOS deficient; PEC, peritoneal exudate cells; ROI, reactive oxygen intermediates; M, macrophage; DC, dendritic cell.
Received for publication August 4, 1998. Accepted for publication February 16, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Immunology 93:556.[Medline]
, mice develop unimpaired IL-12 responses to Toxoplasma gondii while failing to control acute infection. J. Immunol. 157:4045.[Abstract]
(IFN-) regulates production of IL-10 and IL-12 in human herpes virus-6 (HHV-6)-infected monocyte/macrophage lineage. Clin. Exp. Immunol. 109:421.[Medline]
-inducing factor enhances T-helper-1 cytokine production by stimulated human T-cells: synergism with interleukin-12 for interferon- production. Eur. J. Immunol. 26:1647.[Medline]
interferon by natural killer cells. Infect. Immun. 65:3594.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  M. P. Longhi, C. Trumpfheller, J. Idoyaga, M. Caskey, I. Matos, C. Kluger, A. M. Salazar, M. Colonna, and R. M. Steinman Dendritic cells require a systemic type I interferon response to mature and induce CD4+ Th1 immunity with poly IC as adjuvant J. Exp. Med., July 6, 2009; 206(7): 1589 - 1602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Barfod, M. A. Nielsen, L. Turner, M. Dahlback, A. T. R. Jensen, L. Hviid, T. G. Theander, and A. Salanti Baculovirus-Expressed Constructs Induce Immunoglobulin G That Recognizes VAR2CSA on Plasmodium falciparum- Infected Erythrocytes Infect. Immun., July 1, 2006; 74(7): 4357 - 4360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Spadaro, E. Ambrosino, M. Iezzi, E. Di Carlo, P. Sacchetti, C. Curcio, A. Amici, W.-Z. Wei, P. Musiani, P.-L. Lollini, et al. Cure of Mammary Carcinomas in Her-2 Transgenic Mice through Sequential Stimulation of Innate (Neoadjuvant Interleukin-12) and Adaptive (DNA Vaccine Electroporation) Immunity Clin. Cancer Res., March 1, 2005; 11(5): 1941 - 1952. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Beacock-Sharp, A. M. Donachie, N. C. Robson, and A. M. Mowat A role for dendritic cells in the priming of antigen-specific CD4+ and CD8+ T lymphocytes by immune-stimulating complexes in vivo Int. Immunol., June 1, 2003; 15(6): 711 - 720. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Chen, K.-F. Hu, B. Rozell, C. Orvell, B. Morein, and P. Liljestrom Vaccination with Recombinant Alphavirus or Immune-Stimulating Complex Antigen Against Respiratory Syncytial Virus J. Immunol., September 15, 2002; 169(6): 3208 - 3216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. I. Mowat, A. M. Donachie, S. Jagewall, K. Schon, B. Lowenadler, K. Dalsgaard, P. Kaastrup, and N. Lycke CTA1-DD-Immune Stimulating Complexes: a Novel, Rationally Designed Combined Mucosal Vaccine Adjuvant Effective with Nanogram Doses of Antigen J. Immunol., September 15, 2001; 167(6): 3398 - 3405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Krishnan, S. Sad, G. B. Patel, and G. D. Sprott The Potent Adjuvant Activity of Archaeosomes Correlates to the Recruitment and Activation of Macrophages and Dendritic Cells In Vivo J. Immunol., February 1, 2001; 166(3): 1885 - 1893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Simmons, P. Mastroeni, R. Fowler, M. Ghaem-maghami, N. Lycke, M. Pizza, R. Rappuoli, and G. Dougan MHC Class I-Restricted Cytotoxic Lymphocyte Responses Induced by Enterotoxin-Based Mucosal Adjuvants J. Immunol., December 15, 1999; 163(12): 6502 - 6510. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
