-inulin is mediated by C3 fragments deposited on antigen-presenting cells
Research Group of the Hungarian Academy of Sciences
* Department of Immunology, Eötvös Loránd University, Göd, Hungary
Division of Immunology and Cell Biology, John Curtin School of Medical Research, The Australian National University, Canberra, Australia
Correspondence: Anna Erdei, Department of Immunology, Eötvös Loránd University, H-2131 Göd, Jávorka S.u.14., Hungary.
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-inulin, a strong activator of the
alternative complement pathway, is well-known, but its exact mechanism
is not revealed yet. Here, we show that macrophages, isolated from the
peritoneal cavity of -inulin-injected mice and used as
antigen-presenting cells, enhance the proliferation of antigen-specific
T-cells up to 2.5-fold when compared with macrophages of nontreated
animals. This effect is abrogated by the presence of anti-C3
F(ab')2 fragments and by prior decomplementation of the
donor animals with CVF. It is demonstrated that treatment of mice with
the adjuvant results in deposition of C3-fragments onto the surface of
peritoneal macrophages, as does in vitro incubation of the
cells with -inulin in the presence of fresh autologous serum. Prior
incubation of macrophages with -inulin plus serum in
vitro enhances subsequent C3 production. Because it has been
shown earlier that CR1/2 expressed on activated T-cells and interacting
with covalently bound C3-fragments plays an important role in the
augmentation of the adaptive response, our present results reveal a
mechanism that contributes to the adjuvant effect of -inulin and
point to a further link between innate and adaptive immunity.
Key Words: adjuvanticity • complement activation • C3 deposition • T-cell response
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-IN), the storage carbohydrate of
Compositae, is known to activate the alternative complement
pathway [1
], and its adjuvant effect has been
demonstrated in several experiments [2
]. Minimal doses
of -IN, which activate the alternative complement pathway in
vivo (e.g., 50 µg/mouse), were shown to increase immunoglobulin
(Ig)M and IgA responses to keyhole limpet hemocyanin (KLH) four- to
sixfold, and the amount of antigen-specific IgG increased to 28-fold.
Moreover, delayed hypersensitivity was increased some tenfold as a
consequence of -IN injection. The role of the complement system, particularly that of the third component C3 and receptors interacting with its activation fragments, has been shown to influence specific immune responses by various mechanisms [3 4 ]. The process of antigen presentation and the antibody response were shown to be influenced by C3b bound to the antigen [5 6 ]. The attachment of C3b to tetanus toxin has been demonstrated to induce the redistribution of peptide-major histocompatibility complex (MHC) complexes [7 ]. iC3b/C3dg bound to immune complexes generated with natural antibodies, and a primary antigen was found to promote antigen uptake and the expression of costimulatory molecules [8 ]. C3d has been described as a "molecular adjuvant" by its capacity to augment primary and secondary antibody responses in vivo [9 ]. Recently, we have shown that C3 fragments, deposited in vitro onto the surface of murine antigen-presenting cells (APC) as a consequence of alternative pathway activation, enhance strongly the proliferation of antigen-specific T-cells [10 ], confirming our earlier results obtained using cells of normal and C3-deficient guinea pigs [11 ].
In our present investigations, the effect of in vivo
complement activation on the response of T-cells to ovalbumin (OA) was
studied, using macrophages as APC isolated from mice injected
intraperitoneally with -IN. We demonstrate that as a consequence of
treatment of the animals with the adjuvant, C3 split-products are
deposited onto the surface of APC. These cell-bound C3 fragments are
shown to play an important role in the enhancement of T-cell
proliferation, because the effect is abrogated by anti-C3
F(ab')2 fragments and by prior decomplementation of the
animals. These data suggest that APC-bound C3 fragments may contribute
to the adjuvant effect of complement activating substances.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-IN is described elsewhere
[12
]. Fluorescein isothiocyanate (FITC)-labeled goat
F(ab')2 fragments of antimouse C3 were obtained
from Cappel Labs (Cochranville, PA). Monoclonal antibodies (mAbs)
recognizing mouse CR1 (7E9) were kindly provided by Dr. T. Kinoshita
(Osaka University, Japan). The antibody to CR3 (CD11b/CD18), M1/70-PE,
was purchased from Boehringer Mannheim Biochemica GmbH (Mannheim,
Germany). FITC-labeled mAbs reacting with murine MHCII, Fc receptor for
IgG (FCR)II, B7-2, intercellular adhesion molecule-1 (ICAM-1),
lymphocyte function-associated antigen-1 (LFA-1), and the rabbit
antimouse IgM antibody preparation were kindly provided by Dr. G.
László (Eötvös University, Göd, Hungary).
OA and concanavalin A (Con A) were purchased from Sigma-Aldrich Ltd.
(Budapest, Hungary). Cobra venom factor (CVF) was purchased from Cordis
Laboratories (Miami, FL). 3H-thymidine (spec. act. 2
Ci/mmol) was purchased from Amersham (Gödölló',
Hungary). Serum collected from Balb/c mice was freshly aliquoted and
kept at -70°C until use. Treatment of mouse serum by methylamine was
carried out as described [10
].
Cells
Macrophages were isolated from the peritoneum of 6- to
8-week-old Balb/c mice three days after i.p. injection of 100 µg Con
A. For treatment with -IN, the adjuvant was injected i.p. at a dose
of 100 µg/animal, together with Con A, and peritoneal macrophages
were isolated three days later. Cells were cultured in RPMI medium
containing 5% fetal calf serum (FCS). For CVF-treatment, -IN was
injected three days after the administration of the venom factor. For
the antigen presentation assay, T-cells were isolated from the lymph
nodes of Balb/c mice injected with OA (100 µg/animal) 10 days before
the experiment. In vitro treatment of the cells with serum
was carried out as described earlier [10
].
Cytofluorimetry
Samples of cells at a concentration of 2 x 107
cells/ml were incubated with the relevant fluorochrome-labeled antibody
and analyzed using a FACScan (Becton Dickinson, Rutherford, NJ)
instrument. Isotype-matched antibodies were used as control.
CVF treatment
Animals were injected i.p. with 50 units of CVF, as described by
Cooper and Carter [13
].
Antigen-presentation assay
For the presentation of OA by peritoneal macrophages, cells
isolated from -IN-treated and control Balb/c mice were distributed
in 96-well TC plates at a density of 5 x 104
cells/well. Various dilutions of the antigen were added to samples
prepared in triplicates, followed by the addition of antigen-specific
T-cells (2x105 cells/culture) isolated from the lymph
nodes of OA-injected animals. After culturing for four days, samples
were pulsed with 3H-thymidine (0.5 µCi/0.2 ml culture)
for 16 h and then harvested and measured.
Determination of C3
Peritoneal macrophages treated in vitro as indicated
were cultured at the density of 5 x 104 cells/well.
C3 produced by the cells was assessed after 24 h in a sandwich
enzyme-linked immunosorbent assay (ELISA) using F(ab')2
fragments of goat antimouse C3 (Cappel) as capture antibody and
peroxidase-conjugated goat antimouse C3 (Cappel) as the second
antibody. Isolated mouse C3 was used as standard to estimate the amount
of cell-derived C3.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-IN induces C3 deposition onto peritoneal
macrophages in vivo-IN is a potent activator
of the alternative pathway of complement [1
]. Because
generated complement fragments fixed to various cells have been shown
to modulate immune responses [3
4
10
11
], we wished
to know whether i.p. injection of -IN induces C3 deposition onto the
surface of peritoneal macrophages as a consequence of complement
activation. Membrane-bound C3 was monitored by cytofluorimetry, using
FITC-labeled F(ab')2 of antimouse C3. Figure 1
shows that 
40% of the CD11b-positive peritoneal macrophages
isolated from -IN-injected animals bear C3 fragments on the cell
membrane. It is interesting that 20% of the macrophages isolated
from the peritoneal cavity of nontreated animals were also stained with
the anti-C3 antibody, indicating that C3 fragments might have bound to
the cells under normal physiological conditions. A similar finding was
made with cells of human origin [14
].
 View larger version (58K): [in a new window] |
Figure 1. Deposition of C3 on peritoneal macrophages derived from -IN-treated
animals. (A) Macrophages were isolated from the peritoneal cavity of
mice injected i.p. with -IN three days earlier. Cells were stained
with FITC-labeled F(ab')2 fragment of anti-C3 and
phycoerythrin-conjugated antibody to Mac1. As control, Con A-elicited
cells were used (B). Data shown are representative of four
experiments.
|
-IN in the presence of fresh,
autologous serum in vitro results in deposition of C3
split-products onto the cell membrane-IN results in C3 deposition onto the cell membrane of
macrophages also. To this end, cells isolated from the peritoneum of
Balb/c mice were incubated with RPMI medium containing 10% fresh
autologous serum in the presence of -IN (100 µg/ml).
Time-dependent deposition of C3 was assessed by cytofluorimetry. In
agreement with our earlier data [10
], serum treatment is
shown to result in C3 deposition onto the cell membrane as a
consequence of alternative pathway activation (Fig. 2
). For macrophages incubated with a minimal dose of -IN in the
presence of fresh mouse serum, however, the amount of cell-bound C3
fragments is even higher. These data demonstrate that in addition to
complement activation by macrophages, -IN added to the cells induces
C3 deposition further onto the cell membrane. Regarding the time course
of this process, it is shown that cell-bound C3 fragments are already
high after 15 min of incubation. Although the amount bound decreased
after the first h, C3 fragments from serum and -IN treatment do not
disappear from the cell surface even after 4 h (Fig. 2)
, in
agreement with earlier results [10
].
 View larger version (17K): [in a new window] |
Figure 2. Deposition of C3 on the cell membrane of macrophages treated with
-IN in vitro. Cells were incubated with -IN
(•—•), with -IN in the presence of fresh autologous serum
diluted tenfold with RPMI medium ( —), with fresh autologous
serum diluted tenfold with RPMI medium only ( —), and with medium
only ( —). At intervals as indicated, cells were washed and
incubated with FITC-labeled F(ab')2 fragments of anti-C3.
 -Mean fluorescence values of one representative experiment are
shown.
|
-IN and fresh, autologous serum-IN is able also to induce
the production of C3, peritoneal macrophages were incubated with the
carbohydrate on its own. To mimic in vivo conditions, in
separate samples macrophages were incubated with -IN in the presence
of fresh, autologous serum. Adherent cells were washed extensively, and
subsequent C3 production was assessed in the supernatant fluid by ELISA
after 24 h. As shown in Figure 3
, macrophages activated with -IN in the presence of fresh serum
produced up to two times more C3 than control cells. These data suggest
that serum components activated by the carbohydrate in the presence of
the cells play a role in the induction of C3 production. To test if
complement activation and C3 fixation are involved in this process,
cells were incubated with -IN in the presence of methylamine-treated
serum. Because methylamine is known to destroy the covalent-binding
capacity of C3, data shown in Figure 3
suggest that covalently attached
C3 fragments play a role in the augmentation of C3 production.
Moreover, because C3 molecules released by the activated cells may bind
to the producing cell also [18
19
], complement
production by macrophages may contribute to an increase in the number
of C3 fragment-bearing cells seen after in vivo treatment
with the adjuvant (see Fig. 1
). The enhancement of C3 production shown
in Figure 3
could be abrogated by inhibitors of protein synthesis,
cycloheximide, and emetin (unpublished results).
 View larger version (43K): [in a new window] |
Figure 3. C3 production by macrophages activated by -IN. Peritoneal
macrophages were treated in vitro at 37°C for 30 min with
-IN alone, in the presence of fresh autologous serum diluted tenfold
with RPMI medium, or with -IN + serum. Mouse serum inactivated
by methylamine (MA-serum) was included for comparison. Cells were
washed, and C3 production was measured in serum-free supernatants by
ELISA after 24 h. Data shown are representative of three
experiments.
|
-IN-treated mice induce enhanced T-cell response
to exogenous antigen-IN is a powerful adjuvant
of the antibody response [2
]. Because APC bearing
in vitro-deposited C3 fragments were shown to enhance the
response of antigen-specific T-cells to exogenous antigens [10
11
], we aimed to study here whether in
vivo-deposited C3 fragments (see Fig. 1
) act similarly.
Macrophages isolated from the peritoneum of -IN-injected mice were
used as APC, and the proliferation of OA-specific TH cells
was measured using a wide concentration range of antigen. As shown in
Figure 4
, macrophages from -IN-treated animals present OA more
efficiently than cells from nontreated animals. This effect is more
pronounced with suboptimal antigen doses, namely 1.25 and 2.5 µg/ml
OA.
 View larger version (23K): [in a new window] |
Figure 4. Enhancement of the proliferation of antigen-specific T-cells by
peritoneal macrophages of -IN-treated mice. Con A-elicited
peritoneal macrophages isolated from -IN-treated mice were used to
present the antigen as described in Materials and Methods (solid bars).
OA was used at concentrations indicated, and the proliferation of
T-cells was assessed by the uptake of 3H-thymidine. As
control, Con A-elicited macrophages were taken from normal mice (shaded
bars). Data shown are representative of two experiments.
|
-IN-induced augmentation of TH-cell
responses might involve the enhanced expression of MHCII proteins and
costimulatory and accessory molecules on APC also, the appearance of
the following cell membrane proteins was measured: MHCII, B7-2, ICAM-1,
CR1, CR3, LFA-1, and FcRII. As Table 1
shows, the expression of none of these molecules is influenced by
in vivo injection of -IN, further supporting the
assumption that C3 split-products fixed to APC are the active
molecules. |
View this table: [in a new window] |
Table 1. The Expression of Surface Molecules on Peritoneal Macrophages Derived
from -IN-Treated Animals
|
-IN.
 View larger version (29K): [in a new window] |
Figure 5. Effect of anti-C3 F(ab')2 on the enhancement of T-cell
proliferation. Peritoneal macrophages isolated from -IN-treated mice
were used to present OA (2 µg/ml) to T-cells isolated from the lymph
nodes of OA-injected mice without (shaded bars) or in the presence of
anti-C3 F(ab')2 (solid bars). Activation of T-lymphocytes
was assessed as described in Figure 4
. As control, Con A-elicited
macrophages from normal mice were used in similar experiments. Data
shown are representative of three experiments.
|
-IN-IN, experiments were carried out
using peritoneal macrophages of mice injected with -IN after
treatment with CVF. Figure 6
shows that the -IN-enhanced response of T-cells is abrogated
when APC are isolated from previously decomplemented animals. We
injected the animals with -IN three days after decomplementation
with CVF, when serum C3 level diminished to <6–8% of the normal
value and used 125 µg/ml OA.
 View larger version (29K): [in a new window] |
Figure 6. Effect of CVF treatment on the antigen-presenting capacity of
macrophages derived from -IN-treated animals. Mice were injected
with CVF (5 units/animal) three days before treatment with -IN.
After three days, peritoneal macrophages were isolated and used for the
antigen-presentation assay, as described in Materials and Methods.
Activation of T-lymphocytes was assessed as described in Figure 4
,
using 125 µg/ml OA. As control, the antigen-presenting capacity of
macrophages from mice treated with -IN or CVF alone was measured.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-IN, are known to have
immune modulator effects. All these substances have a vaccine adjuvant
effect also that is active if given before, with, or after the antigen.
Minimal doses of -IN that activate the alternative pathway
systemically in mice were shown to increase secondary IgG response to
KLH five- to 28-fold, and memory recall at more than 80 days post
primary injection was increased four- to tenfold [21
]. It has long been known that the complement system, particularly component C3 and receptors reacting with various activation fragments, plays an important role in the maintenance of immunological memory and in the modulation of immune responses [3 4 5 6 7 8 9 10 11 22 ]. T-cell independent and -dependent responses had been shown to be influenced strongly by C3 and its receptors expressed by various cells, particularly in conditions when the antigen is present in suboptimal doses. Recently, increasing numbers of studies reveal the role of various elements of innate immunity—including the complement system—not only in influencing but also in directing adaptive immune responses [23 ].
The mechanism, by which adjuvants induce, augment, and modulate
immune responses, is still not clarified in detail. Because
several adjuvants are efficient activators of the alternative
complement pathway also, the assumption that C3 split-products are
involved in the triggering and development of immune responses is
plausible. Here, we demonstrate that the adjuvant effect of the
complement-activating polysaccharide -IN is at least partially
mediated by C3-fragments generated upon the activation of the
complement cascade, as follows.
Injection of -IN i.p. into normal mice results in C3
deposition onto the surface of peritoneal macrophages, and as a
consequence of "opsonization," the antigen-presenting capacity of
these cells strongly increases (Fig. 4)
. Because activated T-cells have
been shown to express CR1/2, which interact with covalently fixed C3
split-products, it is suggested strongly that a C3-mediated contact
between APC and TH cells is involved; -IN, whether or
not opsonized, is not of itself mitogenic for T-cells
[24
]. The lack of detectable change after -IN
treatment in the availability of other molecules involved in the
cellular contact between APC and the responding T-cells—such as MHCII,
B7-2, ICAM-1, CR1, CR3, LFA-1, FcRII—supports this mechanism
further. The covalent interaction of C3 with various cells in the
absence of membrane-bound antibodies has been described by several
authors [10
11
14
19
25
]. For human B-cells, it had
been demonstrated that CR2 is involved in the initiation of the
alternative pathway activation, and it had also been demonstrated that
this receptor is the main acceptor molecule reacting with activated C3
[14
25 ]. However, in the case of macrophages, which fix
C3 but do not express CR2, the cell-membrane molecule interacting
covalently with the complement protein is not identified still and
needs further investigations. Another source for the augmentation of
antigen presentation of -IN-treated mice is the induction of
inflammatory cytokines interleukin (IL)-1, tumor necrosis factor 
(TNF-), and IL-6 by their macrophages. This possibility, however,
can be excluded, because earlier experiments showed that -IN is not
pyrogenic in rabbits and does not induce TNF in mice
[21
].
If injected 2–3 days before antigen, the adjuvanticity of
-IN for antibody production is four- to sixfold higher than if
injected with antigen [24
]. An interesting
prediction of the present study is that this effect is likely to
depend, inter alia, on "priming" of a subset of APC by
C3-fragment opsonization, which is amplified by increased C3 production
by these cells and ultimately facilitates the interaction with helper
T-cells when exposed to antigen.
Our data presented in this paper provide further support for the role of complement C3 in the induction/augmentation of adaptive immune response. Based on these results, we suggest that the mechanism described here might apply in all conditions when complement activating antigens/adjuvants are used.
Received April 11, 2000; revised July 12, 2000; accepted July 14, 2000.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-receptors by the covalently fixed C3 fragments Mol. Immunol. 25,295-303[Medline]This article has been cited by other articles:
 |
|
 |
|
  P. S. Heeger, P. N. Lalli, F. Lin, A. Valujskikh, J. Liu, N. Muqim, Y. Xu, and M. E. Medof Decay-accelerating factor modulates induction of T cell immunity J. Exp. Med., May 16, 2005; 201(10): 1523 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Suresh, H. Molina, M. S. Salvato, D. Mastellos, J. D. Lambris, and M. Sandor Complement Component 3 Is Required for Optimal Expansion of CD8 T Cells During a Systemic Viral Infection J. Immunol., January 15, 2003; 170(2): 788 - 794. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
