Current address: Laboratoire de Parasitologie, Université Libre de Bruxelles, Bruxelles, Belgium
Conceived and designed the experiments: E. Muraille, P. Gounon, J. Hoebeke, N. Glaichenhaus, E. Mougneau. Performed the experiments: E. Muraille, P. Gounon, J. Cazareth, J. Hoebeke, E. Mougneau. Analyzed the data: E. Muraille, P. Gounon, J. Cazareth, J. Hoebeke, N. Glaichenhaus, E. Mougneau. Contributed reagents/materials/analysis tools: C. Lippuner, A. Davalos-Misslitz, T. Aebischer, S. Muller. Wrote the paper: N. Glaichenhaus, E. Mougneau.
The authors have declared that no competing interests exist.
Protozoa and bacteria infect various types of phagocytic cells including macrophages, monocytes, dendritic cells and eosinophils. However, it is not clear which of these cells process and present microbial antigens
Phagocytosis is a cellular process that allows the engulfment of solid particles such as bacteria or parasites by the cell membrane and leads to the formation of an intracytoplasmic vesicle, the phagosome. Cell types that are capable of phagocytosis are called phagocytes and include monocytes, macrophages, dendritic cells and eosinophils. All these cells express Major Histocompatibility Complex class II molecules, although at different levels. These molecules allow the presentation of pathogen-derived peptides to CD4+ T lymphocytes, a mechanism that alerts the immune system of the occurrence of an infectious danger. In order to visualize the intracellular compartments in which complexes between pathogen-derived peptides and Major Histocompatibility Complex class II molecules are formed, we have generated a monoclonal antibody that reacts to a peptide derived from an immunodominant antigen of the intracellular parasite
The initiation of an adaptive immune response against a pathogen relies on the loading of microbial peptides onto Major Histocompatibility Complex (MHC) molecules in Antigen Presenting Cells (APCs) and on the recognition of these peptide/MHC complexes by T lymphocytes. Therefore, identifying the cell types that present peptide/MHC to T cells and the mechanisms that lead to the formation of these complexes is critical for understanding host-pathogen interactions. In contrast to soluble antigens that enter APCs through passive or receptor-mediated endocytosis, particles that are larger than 0.5 µm in size such as bacteria and protozoa enter host cells through phagocytosis
As phagosomes are part of the endocytic pathway, many studies have investigated the role of these organelles in the processing and presentation of pathogen-derived antigens. For this purpose, bone marrow-derived macrophages or DCs, as well as macrophage cell lines were incubated in
Here, we have sought to visualize peptide/MHC complexes at the surface and in intracellular compartments of various phagocytic cells infected with L. major in vivo. To this aim, we have generated a monoclonal antibody (mAb) that reacts with high avidity to a peptide derived from the
To generate a mAb reacting to LACK156–173 bound to I-Ad, we injected I-Ad/LACK dimers into 16.2β T Cell Receptor (TCR) transgenic mice that exhibit an increased frequency of LACK-specific T cells as the result of the expression of the β chain of a LACK-specific TCR
(A) Flow cytometry analysis of DCs purified from immunized mice. Mice of the indicated strains were injected into the hind footpads with 40 µg of LACK or OVA in CpG. LN cells were harvested 2 days later and enriched for CD11c+ cells using magnetic beads. Cells were stained with 2C44 and anti-CD11c mAb. Data show representative FACS profiles in mice immunized with LACK (solid line) or OVA (filled histogram). (B) Flow cytometry analysis of peptide-pulsed lymphoma cells. A20 mouse B lymphoma cells were pulsed or not with 1 µM of the indicated peptides and analyzed by flow cytometry upon staining with 2C44 and propidium iodide (PI). Data show representative FACS profiles after gating on PI− cells.
To further characterize the specificity of 2C44, BALB/c (H-2d), C3H (H-2k), C57BL/6 (H-2b) and NOD (H-2g7) mice were immunized with LACK or OVA. LN CD11c+ cells were purified 2 days later and analyzed by flow cytometry upon staining with 2C44. While 20–30% of CD11c+ cells were 2C44+ in LACK-immunized BALB/c mice, no staining or only background staining was observed in C3H, C57BL/6 or NOD mice immunized with LACK (
BALB/c mice were injected into the hind footpads with 40 µg of LACK in CpG. LN cells were harvested 2 days later and enriched for CD11c+ cells using magnetic beads. Cells were frozen and ultrathin cryosections were analyzed by electron microscopy after immunogold labelling with 2C44 (upper panel), anti-I-A/I-E (middle panel) or control isotypic (lower panel) mAbs. Representative photographs are shown. Intracellular Compartments (IC) and Extracellular Compartments (EC) are indicated. Arrows point to dendrites.
We next generated a fluorescent parasite that could allow for the purification of amastigote-containing cells. To this aim, we constructed a recombinant parasite in which the gene coding for a DsRed variant was inserted into the ribosomal RNA locus that is active in amastigotes
Lymphocyte-depleted LN cells from 4 wk-infected BALB/c mice were analyzed by flow cytometry after staining with mAbs directed to CD11c, CD11b, Ly-6C, Ly-6G, F4/80, Siglec-F, CCR3, B220 and I-A/I-E. Data show representative FACS profiles after gating on the indicated populations. Numbers indicate mfi. (A) All cells. The percentage of DsRed+ cells in the gated population is indicated (left panel). The right panel shows the R1 and R2 gates that were used later in this study. (B) Eosinophils. (C) Dendritic cells. (D) Macrophages/monocytes. (E) Neutrophils.
DsRed+ cells were then sorted by flow cytometry, embedded with resin and analyzed by electron microscopy. In agreement with flow cytometry data, about 35% of sorted DsRed+ cells exhibited features of macrophages/monocytes such as irregular cytoplasmic projections or pseudopodia, cytoplasm with a few mitochondria, lysosomes and residual bodies (features that were not always all present on the same section), 20–25% of eosinophils (large ellipsoid granules) and 35–40% of DCs (thin and long dendrites) (
Lymphocyte-depleted LN cells from 4 wk-infected BALB/c mice were analyzed by electron microscopy. Data show representative pictures of amastigote-containing DCs (upper panel), eosinophils (middle panel), and macrophage/monocytes (lower panel). Red and white arrows point to the phagosome membrane and to the eosinophil granules respectively. N: nucleus of the cell; P: phagosome.
We next investigated whether I-Ad/LACK complexes were present in phagolysosomes and/or in other cellular compartments. BALB/c and BALB.B mice were infected with DsRed-expressing promastigotes and sacrificed 4 weeks later. LN cells were depleted of both T and B lymphocytes and amastigote-containing cells were scored and analyzed for the presence of I-Ad/LACK complexes using immunogold electron microscopy following staining with 2C44, anti-I-A/I-E, anti-LAMP-1 or control isotypic mAbs. Out of 115 scored amastigote-containing cells, 45 exhibited long dendrites but no granules indicating that they were DCs. Twenty-nine amastigote-containing cells exhibited eosinophil-specific features including large ellipsoid specific granules, with an electron-dense cristallin nucleus. Forty-one amastigote-containing cells did not show dendrites or granules but exhibited pseudopodia and a large number of vesicles suggesting that they were macrophage-related cells. In amastigote-containing DCs stained with 2C44, gold grains were readily detected on the plasma membrane, in phagosomes and on their membrane, and on much smaller cytoplasmic vesicles (
Ultrathin cryosections of DCs purified from 4 wk-infected BALB/c mice were analyzed by immunogold electron microscopy after staining with either 2C44 (A, B), anti-I-A/I-E (C), a combination of a anti-LAMP-1 mAb (10 nm gold particles) and 2C44 (15 nm gold particles) (D), or a control isotypic mAb (E). (A) Data show different magnification views of a typical micrograph showing a DC harboring one single phagosome following staining with 2C44. Upper left panel: a low magnification view showing dendrites, the nucleus and an amastigote-containing phagosome. Lower left panel: a high magnification view showing the phagosome. Upper right panel: a high magnification view (after a 90° left rotation) showing gold grains located either on the cytoplasmic membrane or on small vesicles. Lower right panel: a high magnification view showing the two double sheets of the phagosome membrane with an intense 2C44 labelling of the inner side of the phagosomal membrane. (B) High magnification view showing of a small vesicle exhibiting gold grains both on peripheral and internal membranes. (C) An amastigote-containing phagosome stained with an anti-I-A/I-E mAb showing an intense labelling of the phagosomal membrane, small vesicles and the cytoplasmic membrane. (D) A fragment of an amastigote-containing phagosomal membrane labelled with both 2C44 (white arrowheads) and an anti-LAMP-1 mAb (black arrowheads). (E) An amastigote-containing phagosome stained with a control isotypic mAb. The red and black arrows point to the phagosome membrane and dendrites respectively. N: phagosome nucleus; EC: extracellular compartment; V: vesicles.
Analysis of DsRed+ macrophages/monocytes by immunogold electron microscopy upon staining with 2C44 demonstrated the presence of gold grains in the phagosome and on the phagosome membrane, and in cytoplasmic vesicles, but not on the cell surface (
Ultrathin cryosections of LN cells purified from 4 wk-infected BALB/c mice were analyzed by immunogold electron microscopy after staining with either 2C44. (A) A representative image of macrophages/monocytes. Left panel: a low magnification view of a macrophage exhibiting one amastigote-containing phagosome and several vesicles (black arrows). The photograph shows gold grains on the phagosome membrane and in the vesicles but not at the cell surface. Right panel: a high magnification view of the phagosome. Red arrows point to the phagosomal membrane. (B) A representative image of eosinophil. Left panel: a low magnification view showing an eosinophil containing three amastigote-containing phagosomes and typical ellipsoid granules (white arrows). The photograph shows gold grains on the phagosome membrane but not at the cell surface. Right panel: a high magnification view of an eosinophil phagosome. Red arrows point to the phagosomal membrane. (C, D, E) High magnification views of the plasma membrane in a representative macrophage/monocyte (C), eosinophil (D) and DC (E). EC: extracellular compartment. PM: plasma membrane.
Unexpectedly, gold grains were also observed in amastigote-containing eosinophils upon staining with 2C44 (
Altogether, these results suggest that macrophages/monocytes and eosinophils can capture
To obtain quantitative data, we counted the number of gold grains per micron on the phagosome and the plasma membranes in different cell types upon staining with 2C44, anti-I-A/I-E or a control mAb (
Lymphocyte-depleted cells from 4 wk-infected BALB/c mice were enriched for DsRed+ cells. (A) Number of dots per micron. Cells were analyzed by immunogold electron microscopy after staining with 2C44, an anti-I-A/I-E or a control mAb. Cell types were identified according to their morphology and the number of dots per micron of membrane was counted. At least 30 sections were analyzed for each cell type and each staining. Data show the mean +/− s.e.m. (B) Flow cytometry analysis. Cells were analyzed by FACS after staining with 2C44 (filled histograms) or a control Ig (dotted line). Data show representative profiles after gating on CD11c+ cells (left panels) or CD11c− CD11b+ F4/80+ Ly6G− cells (right panels).
As another method to estimate the levels of I-Ad/LACK complexes at the cell surface, lymphocyte-depleted LN cells from 4 wk-infected BALB/c mice were analyzed by flow cytometry after staining with antibodies directed to various surface markers and 2C44 or a control isotypic Ig (
To independently estimate the amount of I-Ad/LACK complexes at the cell surface, lymphocytes-depleted LN cells from 4 wk-infected BALB/c mice were sorted into different cell types and incubated with the highly sensitive LACK-specific LMR7.5 hybridoma (
Lymphocyte-depleted cells from 4 wk-infected BALB/c mice were sorted into different cell types and the indicated numbers of cells were incubated with 105 LMR7.5 T cell hybridomas with or without 1 µM of LACK156–173. Supernatants were harvested 24 hours later and analyzed for IL-2 content by ELISA. (A) Cells were sorted into CD11c+ (filled circles) and CD11c− (empty circles) cells. Data show the results of a typical experiment (out of 5). (B) DsRed+ populations: CD11c+ (filled circles), eosinophils (grey squares), macrophages/monocytes (empty triangles) neutrophils (filled squares). Data show results of a typical experiment (out of 3). (C) CD11c+ cells were sorted into DsRed+ cells (filled squares) and DsRed− cells (filled squares). Data show results of a typical experiment (out of 4).
Altogether, our data demonstrated that I-Ad/LACK complexes were present at the surface of infected DCs as well as on the phagosome membrane. In contrast, infected eosinophils and infected macrophages/monocytes both exhibited I-Ad/LACK complexes on the phagosome membrane, but not on the plasma membrane.
Here, we have tried to visualize MHC class II molecules loaded with a pathogen-derived peptide at the surface and in the intracellular compartments of phagocytic cells that had been infected with
It remains to be determined why all DC subsets are not equally infected with
While DCs accounted for 35% of infected cells in BALB/c mice in the present study, as many as 70–75% of infected cells were identified as DCs in C57BL/6 mice 4 weeks after infection
One of the major findings of this paper is that various types of amastigote-containing phagocytes express I-Ad/LACK complexes in the phagosome and on its inner membrane but that only DCs export these complexes at their surface. Immunogold electron microscopy data revealed that in DCs, I-Ad/LACK complexes were present on the plasma membrane and evenly distributed over the whole cell surface including dendrites. I-Ad/LACK complexes were also present in amastigote-containing LAMP-1+ phagosomes and on the inner side of their membrane. Moreover, I-Ad/LACK complexes were present within amastigotes, a result that is in agreement with previous studies that have shown that amastigotes internalize MHC class II molecules of their host cells
It is tempting to speculate why some phagocytes are infected with
Are I-Ad/LACK complexes assembled in amastigote-containing phagosomes
It is of interest that I-Ad/LACK complexes could be detected at the cell surface since LACK accounts for only 0.01% of the total amount of proteins expressed by amastigotes. This phenomenon is likely to be related to the very long half time life of the I-Ad/LACK complexes. Our results may also suggest that I-Ad/LACK complexes account for a significant proportion of surface peptide/MHC complexes in amastigote-containing DCs, possibly explaining why this parasite antigen is the main target of the CD4+ T cell response directed to
Several mAbs recognizing peptides bound to MHC class II molecules have been described. A set of mAbs, all IgM, were produced against peptides from the rat myelin basic protein bound to I-As
MHC class II molecules, including I-Ad, exhibit an open groove that can accommodate antigenic peptides of different lengths
Participation of amino-terminal flanking residues in the hydrogen-bonding network stabilizing the complex has been observed for multiple peptide–class II MHC complexes from both human and mouse, and appears to be a universal feature of this interaction
Monoclonal antibodies to MHC-peptide complexes have been widely used to study either antigen presentation
Animal work has been conducted according to recommendations of the national laboratory animal use and care committee (CNREEA). Animal experiments were all performed in our animal facility that was approved by the French Ministry of Agriculture (agreement number: B 06-152-5). The protocols on live animals were approved by the local animal ethic committee for animal care and use (comité régional d'éthique pour l'expérimation animale (CREEA) Cote d'azur).
BALB/c (H-2d), BALB.B (H-2b), C57BL/6 (H-2b), NOD (H-2g7), C3H (H-2k) mice, were purchased from Harlan (Bicester, UK). Mice were housed under Specific Pathogen Free (SPF) conditions and used between 6 and 10 weeks of age. 16.2β mice were crossed to BALB/c mice for 12 generations
LACK recombinant protein was purified as described
240 µg of I-Ad/LACK dimers in 200 µl of PBS were mixed with 200 µl of RIBI (Sigma-Aldrich Chimie SARL, Lyon, France) and 50 µl of this solution was injected in the footpad of 16.2β transgenic mice on days 1 and 8. Mice were immunized in the same footpad on days 15 and 18 with 5 µg of peptide/MHC dimers in PBS. Popliteal LN cells were harvested 15 h after the last immunization and fused to AG8X63 myeloma cells according to standard protocols. Cells were plated in 15 X flat-bottomed 96 well-plates in RPMI 1640 supplemented with non essential aminoacids and 10% FCS. Supernatants were screened by flow cytometry. Briefly, I-Ad-transfected fibroblasts were pulsed for 2 h at 37°C with 20 µM of either LACK156–173 or OVA323–339. 2×105 cells were washed twice with PBS containing 1% BSA, resuspended in 100 µl of supernatants and incubated for 1 h at 4°C. Cells were washed in PBS containing 1% BSA and fibroblast-bound antibodies were revealed using FITC-conjugated goat anti-mouse Ig.
The BIACORE 3000 system, sensor chip CM5, surfactant P20, amine coupling kit containing N-hydroxysuccinimide (NHS) and 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) were from BIACORE (Uppsala, Sweden). Streptavidin was obtained from Sigma. All biosensor assays were performed with Hepes-Buffered Saline (HBS) as running buffer (10 mM Hepes, 150 mM NaCl, 3.4 mM EDTA, 0.02% surfactant P20, pH 7.4). Streptavidin was immobilized on CM5 sensor chips by amine coupling according the manufacturer's protocol. The amount of immobilized mAb was approximately 5 ng/mm2. The four biotinylated mAb were immobilized on the streptavidin chip, each on one channel, at a concentration of 10 µg/ml and at a flow rate of 10 µl/min for 1 minute. The channels were washed with 10 µl of regenerating solution (KSCN 0.51 M, citrate 0.2 M, pH = 5). The amounts of bound mAb were 2.0, 2.4, 2.1 and 1.3 ng/mm2 for 2F74, 2C44, 2X80 and 2E60, respectively. I-Ad/LACK and I-Ad/Ig dimers were used at concentrations of 50 to 400 nM and kinetic parameters were measured at 25°C. In the association phase, dimers were fluxed over the antibody chip at a flow rate of 20 µl/min for 7 minutes. In the dissociation phase, buffer alone was fluxed over the chip for 10 minutes. After each cycle, 10 µl of regenerating buffer was used to return to the start value. The kinetic parameters were calculated using the BIAeval 3.1 software. Global analysis was performed using the bivalent analyte model after subtracting the sensorgrams of the I-Ad/Ig control dimer from that of the I-Ad/LACK dimer. The bivalent analyte model was used since the model fitted best to the experimental values.
Immunization was performed by injecting 6 week-old BALB/c mice in both footpads with a truncated version of LACK (40–50 µg per footpad) together with 25 µg of CpG. Cells from the draining lymph nodes were prepared 40 hours later, labeled with fluorescent antibodies and analyzed by flow cytometry. Infection was performed by injecting 6 wk-old BALB/c mice with 106 purified metacyclic promastigotes in 25 µl per footpad. Mice were sacrificed 4 wk later.
To prepare DC, LN were digested with a cocktail of DNase I fraction IX (Roche, diagnostics) (100 µg/ml) and 1.6 mg/ml of collagenase (400 Mandl U/ml) at 37°C for 45 min. DCs were positively selected by MACS using N418 (anti-CD11c) magnetic beads (Miltenyi Biotec, Germany) in the presence of 10% mouse serum and 5 mM EDTA according to the manufacturer instructions. Positively selected cells were >95% pure as determined by flow cytometry following staining with anti-CD11c mAb. To prepare phagocytes, LN cells were depleted of T and B cells. To this aim, cells were incubated with anti-CD3 and anti-CD19 mAb, and CD3+ and CD19+ cells were depleted by negative selection using anti-rat IgG magnetic Dynabeads (Invitrogen, France) resulting in a cell population containing 12% and 13% of B and T lymphocytes respectively (Supplementary
The indicated numbers of cells were incubated with 105 LMR7.5 hybridomas with or without 1 µM of LACK156–173 in DMEM supplemented with 2 mM L-glutamine, 10% heat-inactivated FCS, 5×10−5 M 2-mercaptoethanol, 100 µg/ml penicillin and 100 U/ml streptomycin in the wells of U-bottomed 96 well-plates. Supernatants were harvested 24 h later, and IL-2 contents were measured by ELISA.
Cells were analyzed by flow cytometry using a FACSCanto or a LSR II with the FACS Diva software (Becton Dickinson, San Jose, CA). Cell sorting was carried out using a FACS Aria (Becton Dickinson, San Jose, CA).
Flow-sorted cells were primary fixed with 1.6% glutaraldehyde in 0.1 M phosphate buffer pH 7.5, then at 4°C with 1% osmium tetroxide in 100 mM cacodylate buffer pH 7.5. Pellets were washed five times in water, dehydrated in increasing acetone series and embedded in epoxy resin. 70 µM thin sections were prepared and contrasted for observation with a Philips CM12 or Jeol 1400 electron microscope.
Immunogold labeling was performed on ultrathin cryo-sections of FACS-purified DsRed+ cells. Cells were fixed for 1 h at 4°C in 0.1 M phosphate buffer pH 7.5 containing 4% paraformaldehyde and 0.2% glutaraldehyde. Cells were then washed with phosphate buffer (PBS) containing 50 mM NH4Cl. Small blocks were infiltrated with 2.3 M sucrose, frozen in liquid nitrogen and then cut on a dry diamond knife at −120°C. Ultrathin 70 nm sections were then collected on formvar-coated nickel grids and processed for immunochemistry. Grids were deposited face down on the top of small drops of the following solutions: PBS containing 50 mM NH4Cl for 10 min, PBS containing 1% BSA for 5 min, PBS containing the relevant mAbs in 1% BSA for 1 h, PBS containing 1% BSA for 10 min, PBS containing 0.1% BSA for 5 min, PBS containing 1% BSA and either 10 or 15 nm colloidal gold conjugated anti-mouse IgG antibodies for 30 min, PBS containing 0.1% BSA for 5 min, PBS for 5 min twice, PBS containing 1% glutaraldehyde for 5 min and distilled water for 5 min. Grids were then embedded in methylcellulose as described elsewhere
Flow cytometry analysis of LN cells from immunized mice stained with different mAbs reacting to I-Ad/LACK complexes. BALB/c mice were immunized or not with either LACK or OVA in CpG. LN cells were purified 2 days later, enriched for CD11c+ cells using magnetic beads and stained with either 2C44, 2F74, 2E60 or 2X8. Data show representative FACS profiles after gating on CD11c+ cells.
(0.06 MB TIF)
Lymphocyte-depleted cells from 4 wk-infected BALB/c mice were analyzed by multicolor flow cytometry. Data show representative profiles after gating successively on live cells (upper left panel), CD11bhigh CD11c− cells (G1, upper right panel), Ly-6Gint Ly-6Cint (G2, lower left panel) and Siglec-Fhigh CCR3+ cells (G3, lower right panel). The frequency of DsRed+ cells in the gated population is indicated.
(0.31 MB TIF)
Flow cytometry analysis of DCs, macrophages/monocytes and neutrophils. Lymphocyte-depleted cells from 4 wk-infected BALB/c mice were analyzed by multicolor flow cytometry after gating out eosinophils (R2 gate defined in
(0.46 MB TIF)
High-resolution electron microscopy analysis of amastigote-containing phagosomes. Data show enlargments of DC phagosomes. The left photograph shows 2 phagosomes with the parasite nucleus, membrane and cortical microtubules. The right photograph shows a high magnification view of the phagosomal and amastigote membranes. P: phagosome; C: cytoplasm.
(2.70 MB TIF)
Flow cytometry analysis of LN cells before and after depletion of lymphocytes. Lymphocyte-depleted cells from 4 wk-infected BALB/c mice were analyzed by flow cytometry before (A) or after depletion of CD3+ and CD19+ lymphocytes using streptavidin-coupled beads (B) or anti-rat Ig Dynabeads (C). Data show representative FACS profiles after staining with anti-B220 (left panels) or anti-CD3 (right panels) mAbs. Of note, two different clones of anti-CD3 mAb were used for depletion (145-2C11) and staining (C363.29.B). The frequency of cells that are stained with the indicated mAb is shown.
(0.15 MB TIF)
Kinetics parameters of the binding of different mAbs to I-Ad/LACK dimer. The indicated mAbs were biotinylated and immobilized on a streptavidin chip that was fluxed with either I-Ad/LACK dimers or I-Ad/Ig control dimers. The kinetics parameters were calculated using the BIAeval 3.1 software. Global analysis was performed using the bivalent analyte model after subtracting the sensorgrams of the I-Ad/Ig control dimer from that of the I-Ad/LACK dimer.
(0.03 MB DOC)