| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*
Laboratoire d’Immunologie des Tumeurs, Université Paris-Sud, Chatenay Malabry, France; and
Laboratoire d’Immunologie Cellulaire, Institut Gustave-Roussy, Villejuif, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The lymphocyte activation gene-3 (LAG-3)3 molecule is related to CD4 at the gene and protein levels (1, 2). LAG-3 is expressed in activated CD4+ and CD8+ T lymphocytes where it is associated with the CD3/TCR complex at the cell surface (3, 4). LAG-3, like CD4 (5), may oligomerize at the cell surface to interact more efficiently with MHC class II molecules, as shown by the finding of three dominant negative mutations in the LAG-3 domain 1 that were able to inhibit the binding of wild-type LAG-3 molecules to class II molecules in a cell-cell adhesion assay (6).
As a soluble fusion protein, LAG-3 has been shown to bind MHC class II
molecules with a much higher avidity than CD4 (7) and also
to increase the capacity of phagocytic cells (MHC class
II+ macrophages or immature dendritic cells
(DCs)) to induce T cell responses in vitro (8). The
changes brought about in the APCs during such an activation of the MHC
class II signaling pathway are not fully understood but probably
involve a combination of improved Ag processing, increased expression
of costimulatory and adhesion molecules, and an up-regulation in the
production of cytokines such as IL-12 and TNF-
(8). In
vivo, tumor regression and tumor cell-specific
CD8+ T cell responses have been shown to be
induced by LAG-3-transfected tumor cells or by a soluble LAG-3 fusion
protein together with irradiated tumor cells (9). However,
the potential of LAG-3 as an adjuvant for both humoral and
cell-mediated immune responses to a defined protein Ag has not been
explored in previous studies.
In this report, we have evaluated the adjuvant effect of mLAG-3Ig (a fusion protein between murine LAG-3 and the Fc fraction of a murine IgG2a mAb) in two different strains of mice (BALB/c and C57BL/6) for the induction of both CD4 and CD8 T cell responses in two well-described Ag systems, the particulate hepatitis B surface Ag (HBsAg), which elicits a CTL response without adjuvant, and the soluble OVA Ag, which could not induce any CTL response in the absence of an adjuvant. A significant increase in both the humoral and cellular responses was obtained with mLAG-3Ig. The stimulation of both arms of the immune response, induced by LAG-3 and probably mediated via MHC class II signaling into professional APCs, may have important implications for the immunotherapy of infectious diseases and cancer and provides an interesting focus for future therapeutic reagents.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female C57BL/6 (H-2b) mice, 6–8 wk old, were purchased from IFFA-CREDO Laboratories (Lyon, France). Female BALB/c (H-2d) mice, 4–8 wk old, were purchased from Janvier Laboratories (Le Genest St. Isle, France). All of these mouse strains were raised in specific pathogen-free conditions.
Antigens
Purified HBsAg particles (ay subtype) were purchased from Fitzgerald (Concord, MA), and purified chicken OVA (grade IV) was from Sigma (St. Louis, MO). Ags were dissolved in PBS and stored at -20°C.
mLAG-3Ig, peptides, and aluminium hydroxide (alum)
Recombinant soluble mLAG-3 molecules were generated by fusing the extracellular domain of mLAG-3 to a murine IgG2a Fc portion (9, 10). The resulting recombinant protein, mLAG-3Ig, was produced in Chinese hamster ovary cells and purified on protein-A columns (Dr. M. Subramanyam and M. Tepper, Ares Advanced Technology, Randolph, MA). The total protein purity was >95% sLAG-3Ig by Coomassie blue SDS-PAGE densitometry. Potential contamination of the purified protein with bacterial endotoxin (LPS) was determined using the chromogenic Limulus amoebocyte lysate assay (BioWhittaker, Walkerville, MD), which measures the activity of a L. amoebocyte protease after activation by LPS. A calibration curve based on enzymatic activity vs LPS was constructed to determine endotoxin units in the test sample, and values of less than 1 EU/mg were obtained for mLAG-3Ig.
The hepatitis B virus (HBV) S peptide (aa 28–39, IPQSLDSWWTSL, H-2d restricted) and the OVA peptide (aa 257–264, SIINFKEL, H-2b restricted) were synthesized by solid-phase technique and purified by reverse-phase HPLC (Institut Gustave-Roussy, Villejuif, France). Peptides were dissolved in PBS and stored at -20°C. Alum was purchased as Alu-Gel-S (Serva, Heidelberg, Germany) and stored at room temperature.
Immunization protocols
Immunization with HBsAg was conducted on BALB/c mice (five mice per group). Each mouse received two injections s.c. on the ventral surface on days 0 and 21 of 0.01 µg or 1 µg HBsAg in 200 µl PBS with or without 1 µg of isotype-matched murine control IgG2a (Southern Biotechnology Associates, Birmingham, AL) or 1 µg of mLAG-3Ig.
In other experiments, C57BL/6 mice (five mice per group) received three injections on days 0, 21, and 28 of 50 µg native OVA alone (or adsorbed onto alum) in 200 µl PBS or with 1 µg mLAG-3Ig or a control IgG2a molecule. No side effects were ever observed with mLAG-3Ig (even at 10 µg/site) using these immunization protocols.
Evaluation of CTL response
Spleens were recovered under sterile conditions 1 wk after the
last immunization. Cell suspensions were prepared
(107 cells/well) individually from each spleen
and suspended in 2 ml of -MEM tissue culture medium supplemented
with 10 mM HEPES buffer, 1 mM Na pyruvate, nonessential amino acids,
5 x 10-5 M 2-ME, antibiotics, and 10% FCS
(Life Technologies, Cergy-Pontoise, France) in 24-well plates.
Responder splenocytes from BALB/c or C57BL/6 mice were stimulated with
10-6 M HBV S peptide or
10-6 M OVA peptide, respectively. After 5 days
in culture, half of the medium was replaced with fresh medium and the
cells were then used as effectors in a standard 4-h chromium release
assay performed 2 or 3 days later. Targets were P815 or EL4 cells
labeled with 51Cr (3.7
MBq/106 cells; Amersham Life Sciences, Bucks,
U.K.). After a 4-h incubation at 37°C, 50 µl of supernatant was
removed from each well and counted on a beta counter (Topcount;
Packard, Meriden, CT). The percentage specific release was calculated
as: (experimental release - spontaneous release)/(total
release - spontaneous release) x 100. Total release was
measured by resuspending the target cells in lysis buffer. Spontaneous
release was obtained from targets incubated with medium alone and was
less than 15% of the total release.
In blocking assays, P815 cells pulsed with the HBV S peptide were preincubated with CD4- (clone RM4-4), CD8- (clone 53-6.7), MHC class I-specific (clone 34-2-12), or MHC class II-specific (clone 34-5-3) mAb (Becton Dickinson, Le Pont de Claix, France) before addition of effector cells. Final concentration of mAb was 10 µg/ml.
T cell proliferation assay
Spleen cell suspensions from immunized mice were cultured in triplicate using 96-well round-bottom plates at 5 x 106 cells/ml in 200 µl RPMI 1640 medium containing 2% mouse sera with different concentrations of HBsAg particles or OVA Ag. Splenocytes were stimulated for 96 or 120 h, [3H]thymidine (Amersham Life Sciences) was added (1 µCi/well), and the [3H]thymidine incorporation into the DNA was measured after an additional 18 h.
Cytokine assay
In vivo primed spleen cells were cocultured with Ags in parallel
to the proliferation assays. Culture supernatants were collected at 48
and 72 h, and the concentrations of IL-4, IL-10, IFN-
, and
TNF- were determined by ELISA using commercial kits (R&D Systems,
Abingdon, U.K.). In blocking assays, splenocytes were incubated for
48 h with CD4-, CD8-, MHC class I-, or MHC class II-specific mAb
(10 µg/ml).
Serology
Blood was collected by retrobulbar puncture using heparinized glass pipettes, and mouse anti-HBs or anti-OVA Abs were detected using an ELISA on HBsAg- or OVA-coated wells (1 µg/ml). Bound Abs were detected by anti-mouse biotinylated Abs and then Streptavidin-HRP (Amersham Life Sciences) staining.
Statistics
Data were analyzed by the nonparametric Mann-Whitney U rank test, and differences with p < 0.01 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We first examined spleen cells from BALB/c mice primed with
various HBsAg formulations for the development of HBsAg-specific CTL
responses. The surface HBV protein (S-Ag) self-assembles into 22-nm
particles and is an efficient immunogen for inducing CTL responses in
its native, particulate form. Spleen cells from immunized mice were
cultured with the HBV S (aa 28–39) peptide (without IL-2) for 7 days
and then assayed for cytolytic activity (11, 12). Spleen
cells from mice immunized with 1 µg HBsAg developed CTLs that lysed
syngeneic H-2d mastocytoma P815 cells in the
presence of the HBV S peptide (18% lysis vs 6% without peptide at a
100/1 E:T ratio; Fig. 1
A).
Spleen cells from mice immunized with a lower dose (0.01 µg) of HBsAg
also recognized the HBV S peptide-loaded P815 cells (16% lysis vs 3%
without peptide at a 100/1 E:T ratio; data not shown).
|
A) or 0.01 µg (not shown) HBsAg plus 1 µg mLAG-3Ig, a
strong CTL response was obtained. Mice injected with the same dose of
HBsAg plus 1 µg of the isotype-matched murine IgG2a control molecule
did not show any evidence of increased cytolytic activity. Similar
results were obtained when testing these peptide-induced CTL lines on
HBsAg-transfected P815 cells (P815/S) vs untransfected P815 cells
(11, 12) (data not shown). These latter results suggest
that the vaccination Ag is capable in the presence of LAG-3 of priming
a population of CTLs that can be restimulated during a natural
infection, i.e., HBsAg was used as the exogenous Ag that primed BALB/c
mice to recognize target cells that had processed the Ag
endogenously.
To confirm that the lysis of peptide-pulsed P815 target cells could be
attributed to MHC class I-restricted CTLs, we performed blocking assays
with anti-CD4, -CD8, -MHC class I and -MHC class II mAbs (Fig. 1B). Lysis was inhibited in the presence of 10 µg/ml CD8-
or MHC class I-specific and not CD4- (not shown) or MHC class
II-specific mAbs, showing that the amplified response of splenocytes to
the HBV S epitope in mice immunized with mLAG-3Ig as an adjuvant was
mediated by MHC class I-restricted CD8+ T
cells.
mLAG-3Ig was also evaluated as an adjuvant for inducing cytotoxic T
lymphocyte responses to a soluble Ag. The well-described OVA system was
chosen as a model to determine whether soluble exogenous proteins can
prime CTLs when delivered with a low dose (1 µg) of
mLAG-3Ig. The EL4 mouse T lymphoma cells were used as target cells
in the presence or absence of the OVA257–264
peptide. Fig. 2A shows that
splenocytes from C57BL/6 mice immunized three times with OVA (50 µg)
and mLAG-3Ig (1 µg) had high Ag-specific cytolytic activity (87% and
70% lysis vs 8% and 7% without peptide at a 100/1 and 30/1 E:T
ratios, respectively), whereas mice injected with OVA alone or OVA plus
1 µg of an isotype-matched control Ab (IgG2a) had little detectable
cytolytic activity. In a second experiment (Fig. 2B), OVA
was adsorbed onto alum, and a significant CTL response was observed
(36% lysis vs 11% without peptide at a 100/1 E:T ratio). This was
probably due to the alum providing a depot that could enhance the
uptake of OVA by macrophages leading to some CTL priming in vivo. Under
these conditions, which are close to the vaccination formulations used
in humans, a low dose of mLAG-3Ig (1 µg) further increased the
priming of CTLs (60% and 34% lysis vs 10% and 7% without peptide at
a 100/1 and 30/1 E:T ratio, respectively; Fig. 2B).
|
We examined spleen cells from mice primed with the various
HBsAg (Fig. 3) or OVA (Fig. 4) formulations for the development of
proliferative HBsAg- or OVA-specific T cell responses. Spleen cells
from primed mice were tested for their capacity to proliferate at days
4 and 5 in response to various doses of either HBsAg or OVA. Mice
immunized with Ag plus 1 µg mLAG-3Ig had stronger Ag-specific
proliferative responses than the mice immunized with either the
particulate or soluble Ag alone. Fig. 3 describes the proliferative
response obtained at day 4 in mice vaccinated with 1 µg HBsAg plus 1
µg mLAG-3Ig, with mice injected with HBsAg plus 1 µg IgG2a as a
control group. Similar results were obtained with a low HBsAg dose
(0.01 µg/mice). Fig. 4A shows that splenocytes of mice
vaccinated with OVA (50 µg/mouse) together with 1 µg mLAG-3Ig had a
stronger proliferative response at day 4 as well as at day 5 (not
shown) than mice injected with Ag alone or with Ag plus an
isotype-matched control Ab (IgG2a). This increased response was also
observed in groups of mice receiving OVA adsorbed onto alum (Fig. 4B).
|
|
Spleen cells were also tested 1 wk after the last injection for
their capacity to produce cytokines at 48 or 72 h in response to
various doses of HBsAg or OVA. Splenocytes of mice vaccinated with
either 1 µg (Fig. 5) or 0.01 µg (data
not shown) HBsAg plus mLAG-3Ig exhibited significantly greater IFN-
and TNF- production in response to 3 µg/ml HBsAg than did the
splenocytes of mice immunized with Ag alone or with Ag plus the control
IgG2a molecule. A similar increase in Th1-type cytokine response was
also observed with a low dose (0.3 µg/ml) of HBsAg (data not shown).
In contrast, no significant increase of IL-4 or IL-10 (Fig. 5, C and D) Th2-type cytokine production was
observed in mice immunized with either of the two doses of HBsAg.
|
by
splenocytes stimulated for 48 h with 3 µg HBsAg. As shown in
Fig. 6, CD4- and MHC class II-specific
and not CD8- and MHC class I-specific mAbs significantly inhibited
IFN- production, indicating that the amplified response of
splenocytes to the nanoparticles in mice immunized with mLAG-3Ig as an
adjuvant was mediated by Th1-polarized MHC class II-restricted
CD4+ T cells.
|
). Only Th1 responses were induced
when using mLAG-3Ig as an adjuvant (IFN- (A) and TNF-
(B)). No induction was observed for IL-4 (C) or
IL-10 (D). As expected, coadministration of an
isotype-matched control Ab (IgG2a) did not affect the amount of
secreted cytokines (data not shown). Similar results were observed in
groups of mice immunized with OVA adsorbed onto alum (Fig. 7).
|
At day 27 after the initial immunization with 1 µg HBsAg alone
or with 1 µg mLAG-3Ig or the control IgG2a molecule, Ab levels in
BALB/c mice were determined in an ELISA on HBsAg-coated wells. Specific
IgG levels were about 100-fold higher in those mice immunized with Ag
plus mLAG-3Ig as shown in Fig. 8A. Similar results were
obtained with mice immunized with a lower dose of HBsAg (0.01 µg;
data not shown).
|
B). Because subclasses of Abs may be important in vaccine efficacy, we also examined the Ab subclasses in HBsAg-immunized mice. IgG2a and IgG1 differ in function (for example, they differ in their ability to fix complement), and they have been used as indicators for the induction of Th1 and Th2 responses, respectively. Both HBsAg-specific IgG1 and IgG2a levels were 10- to 100-fold higher in mice immunized with 1 µg HBsAg plus mLAG-3Ig (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both HBsAg and OVA are model protein Ags that have been widely used due to their immunogenic efficacy having been well established. Immunization with soluble protein Ag preferentially stimulates CD4+ T cells as proteins in the extracellular fluid are processed through the exogenous processing pathway and are degraded into peptides that bind to MHC class II molecules. CD8+ CTLs are stimulated by proteins degraded in an alternative endogenous processing pathway where MHC class I-binding peptides are derived from cytosolic proteins by partial proteolytic degradation. Nonetheless, under certain conditions exogenously acquired Ags, including those from cancer cells, can be cross-primed into the endogenous pathway (13, 14). Cross-priming is particularly efficient when Ags, instead of being obtained free in solution, are engulfed as part of an apoptotic tumor cell or a nanoparticle. In particular, an instance in which the entry of such an exogenous protein into the endogenous class I pathway might be of physiological relevance occurs with HBsAg. Indeed, the S protein that self-assembles into 22-nm particles is an efficient immunogen in its particulate form, as shown in the present study in which mice immunized twice with 0.01 µg HBsAg developed CTLs. In contrast, immunization with soluble, nonparticulate proteins such as OVA rarely induces specific CTL responses.
Inefficient CTL priming by protein Ags adjuvanted in alum or mineral oil to induce humoral responses is a general feature of this type of Ag administration (12). Nonetheless, some protein Ags, such as OVA emulsified in CFA (15) or adsorbed onto alum (the present study), may elicit CD8+ MHC class I-restricted CTLs. Innovative adjuvant formulations, such as immune-stimulating complexes (16), liposomes (17, 18), or saponins (19, 20, 21), that facilitate in vivo priming of CD8+ CTLs by protein Ags are now being developed. When testing the new formulations, it is of increasing importance to define which compartments of the immune system the different adjuvant preparations selectively enhance and/or suppress. The advantages of our system are that efficient in vivo priming of CTLs was achieved without denaturing the soluble protein Ag (i.e., sparing B cell responses that in fact were increased) and that Th1-type CD4 T cell response was increased.
The development of subunit vaccines requires the use of an adjuvant that acts by stimulating components of the innate immune responses. Our results suggest that signaling APCs via MHC class II molecules is an efficient method of inducing both humoral and cellular responses. These MHC class II molecules may be perceived as "danger signals" to the innate immune system, as are the unmethylated CpG motifs in bacterial DNA (22) and exogenous human hsp60 (23). The precise mechanisms underlying the ability of LAG-3 proteins to promote the induction of CTL activity or to shift toward a Th1 differentiation pattern have yet to be elucidated. DCs, potent APCs located in the skin and lymphoid organs, are likely to be the MHC class II+ target cells of soluble LAG-3 proteins used as biological adjuvants because these cells play a major role in promoting the immune responses generated by most vaccine formulations. Cross-linking MHC class II molecules expressed on APCs appeared to contribute not only to APC activation but also to the commitment of T cells toward the Th1/Tc1 type in vivo. In accordance with the latter observation, immature human DCs stimulated by hLAG-3Ig have been shown to secrete more IL-12 (a known Th1 inducer) in vitro without the need of a CD40 ligand signal (8). Although no Th2-type cytokine responses were detected in mice injected with mLAG-3Ig, a net increase in serum Ig titers was observed in all mice, including both IgG2 (Th1) and IgG1 (Th2) subclasses. The mechanism underlying the adjuvant role of LAG-3 in eliciting both a cell-mediated and a humoral response in vivo is still unclear, but our data in vitro with the blocking mAb indicate that both MHC class I-restricted CD8+ and MHC class II-restricted CD4+ T cells are reacting to the LAG-3 effect.
In any case, the demonstration that a soluble protein Ag (i.e., OVA) coinjected with mLAG-3Ig can induce both strong humoral and cellular responses when adsorbed onto alum is certainly of importance for vaccine development using recombinant soluble Ags. Because alum is approved for use in human vaccination and soluble LAG-3 molecules are unlikely to be toxic when given locally in low levels, the combined use of these two agents may be clinically practical. More generally, the ability of soluble LAG-3 proteins to augment both Ab and cell-mediated immune responses suggests that this adjuvant could be a valuable component of subunit vaccines.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Frédéric Triebel, Laboratoire d’Immunologie Cellulaire, Institut Gustave-Roussy, 39 rue Camille Desmoulins, 94805 Villejuif Cedex, France.
3 Abbreviations used in this paper: LAG-3, lymphocyte activation gene-3; DC, dendritic cell; mLAG-3Ig, fusion protein between murine LAG-3 and the Fc fraction of a murine IgG2a mAb; HBsAg, hepatitis B surface Ag; HBV, hepatitis B virus; alum, aluminium hydroxide.
Received for publication August 16, 1999. Accepted for publication March 13, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-12 production by monocytes and dendritic cells. J. Immunol. 162:2748.This article has been cited by other articles:
|
|
 |
|
  B. Li, M. VanRoey, F. Triebel, and K. Jooss Lymphocyte Activation Gene-3 Fusion Protein Increases the Potency of a Granulocyte Macrophage Colony-Stimulating Factor-Secreting Tumor Cell Immunotherapy Clin. Cancer Res., June 1, 2008; 14(11): 3545 - 3554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Casati, C. Camisaschi, L. Novellino, A. Mazzocchi, F. Triebel, L. Rivoltini, G. Parmiani, and C. Castelli Human Lymphocyte Activation Gene-3 Molecules Expressed by Activated T Cells Deliver Costimulation Signal for Dendritic Cell Activation J. Immunol., March 15, 2008; 180(6): 3782 - 3788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Brignone, C. Grygar, M. Marcu, K. Schakel, and F. Triebel A Soluble Form of Lymphocyte Activation Gene-3 (IMP321) Induces Activation of a Large Range of Human Effector Cytotoxic Cells J. Immunol., September 15, 2007; 179(6): 4202 - 4211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Casati, C. Camisaschi, F. Rini, F. Arienti, L. Rivoltini, F. Triebel, G. Parmiani, and C. Castelli Soluble Human LAG-3 Molecule Amplifies the In vitro Generation of Type 1 Tumor-Specific Immunity. Cancer Res., April 15, 2006; 66(8): 4450 - 4460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Andreae, S. Buisson, and F. Triebel MHC class II signal transduction in human dendritic cells induced by a natural ligand, the LAG-3 protein (CD223) Blood, September 15, 2003; 102(6): 2130 - 2137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Cappello, F. Triebel, M. Iezzi, C. Caorsi, E. Quaglino, P.-L. Lollini, A. Amici, E. Di Carlo, P. Musiani, M. Giovarelli, et al. LAG-3 Enables DNA Vaccination to Persistently Prevent Mammary Carcinogenesis in HER-2/neu Transgenic BALB/c Mice Cancer Res., May 15, 2003; 63(10): 2518 - 2525. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Andreae, F. Piras, N. Burdin, and F. Triebel Maturation and Activation of Dendritic Cells Induced by Lymphocyte Activation Gene-3 (CD223) J. Immunol., April 15, 2002; 168(8): 3874 - 3880. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
