Additional supporting information may be found in the online - - PDF document

2832

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

DOI: 10.1002/eji.201343600

Carbon monoxide decreases endosome–lysosome fusion and inhibits soluble antigen presentation by dendritic cells to T cells

Virginie Tardif∗1,2,3, Sebasti´ an A. Riquelme∗4, S´ everine Remy1,2,3, Leandro J. Carre˜ no4, Claudia M. Cort´ es5, Thomas Simon1,2,3, Marcelo Hill1,2,3, C´ edric Louvet1,2,3, Claudia A. Riedel5, Philippe Blancou1,2,3, Jean-Marie Bach6, Christine Chauveau1,2,3, Susan M. Bueno4, Ignacio Anegon∗∗1,2,3 and Alexis M. Kalergis∗∗1,2,3,4,7

1 INSERM, UMR 1064, Nantes, France 2 CHU Nantes, ITUN, Nantes, France 3 Facult´

e de M´ edecine, Universit´ e de Nantes, Nantes, France

4 Millenium Institute of Immunology and Immunotherapy, Facultad de Ciencias Biol´

• gicas, Pontificia Universidad

Cat´

• lica de Chile, Santiago, Chile

5 Millenium Institute of Immunology and Immunotherapy, Facultad de Ciencias Biol´

• gicas, Universidad Andr´

es Bello, Santiago, Chile

6 ONIRIS, Nantes, France 7 Departamento de Reumatolog´

ıa, Facultad de Medicina, Pontificia Universidad Cat´

• lica de Chile, Santiago, Chile

Heme oxygenase-1 (HO-1) inhibits immune responses and inflammatory reactions via the catabolism of heme into carbon monoxide (CO), Fe2+, and biliverdin. We have pre- viously shown that either induction of HO-1 or treatment with exogenous CO inhibits LPS-induced maturation of dendritic cells (DCs) and protects in vivo and in vitro antigen- specific inflammation. Here, we evaluated the capacity of HO-1 and CO to regulate antigen presentation on MHC class I and MHC class II molecules by LPS-treated DCs. We observed that HO-1 and CO treatment significantly inhibited the capacity of DCs to present soluble antigens to T cells. Inhibition was restricted to soluble OVA protein, as no inhibition was

• bserved for antigenic OVA-derived peptides, bead-bound OVA protein, or OVA as an

endogenous antigen. Inhibition of soluble antigen presentation was not due to reduced antigen uptake by DCs, as endocytosis remained functional after HO-1 induction and CO

• treatment. On the contrary, CO significantly reduced the efficiency of fusion between late

endosomes and lysosomes and not by phagosomes and lysosomes. These data suggest that HO-1 and CO can inhibit the ability of LPS-treated DCs to present exogenous soluble antigens to na¨ ıve T cells by blocking antigen trafficking at the level of late endosome– lysosome fusion. Keywords: Antigen processing · Carbon monoxide · Cross-presentation · Endocytosis · Heme

• xygenase-1
• Additional supporting information may be found in the online version of this article at the

publisher’s web-site

Correspondence: Dr. Alexis M. Kalergis e-mail: akalergis@bio.puc.cl

∗These authors contributed equally to this work. ∗∗These authors share senior co-authorship. C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2833 Introduction

Heme oxygenase-1 (HO-1) is one of the three isoforms of the HO enzyme [1,2] and catabolyzes heme into carbon monoxide (CO), Fe2+, and biliverdin [1]. HO-1 expression is induced by agents involved in oxidative stress, such as oxygen-derived free radicals, pro-inflammatory cytokines, and inflammatory stimuli [3], hav- ing a protective effect in a variety of experimental inflammatory models [3–8]. Consistent with this notion, it has been shown that both induc- tion of HO-1 expression by drugs, such as cobalt protoporphyrin (CoPP) and hemin, or the overexpression of HO-1 by gene transfer can contribute to reducing inflammatory damage during disorders involving detrimental immune responses, such as organ transplan- tation and autoimmunity, which usually arise after dendritic cell (DC) activation [5,7,9–14]. Interestingly, several studies have suggested that the function

• f DCs can be modulated by HO-1 activity [3, 6, 15]. Because

DCs are key players in regulating adaptive immunity and T-cell activation, the effect of HO-1 on regulating their function can be highly relevant for modulating the adaptive immune response. We have previously shown that immature human, rat, and mouse DCs express HO-1 and that their expression drastically decreases as a result of DC maturation [10,15]. Also, we and others have shown that overexpression of HO-1 by rat and humans DCs inhibits LPS- induced maturation and the pro-inflammatory function of these cells [6,16,17]. Interestingly, in several models, CO mimics the effects of HO-1 [7, 10, 15, 18], indicating that HO-1 acts via generation of CO. We have recently shown that the principal mediator of the effects

• f HO-1 induction in DC maturation in vivo and in vitro is CO

[15,18]. Thus, the exposure of DCs to CO seems to be an appro- priate approach to mimic the immunomodulatory effects of HO-1 [18]. However, despite the effects of CO on DC maturation and inflammation, whether CO can directly affect antigen presentation by DCs to T cells remains unknown. Here, we examined the effect of HO-1 and CO on the ability of LPS-treated DCs to present protein-derived antigens on class I and II MHC molecules to CD8+ and CD4+ T cells. We found that HO-1 and CO treatment inhibited the ability of DCs to activate CD4+ and CD8+ T cells in response to soluble antigens, both OVA and the Ag85B antigen from Mycobacterium tuberculosis. However, HO-1 and CO failed to inhibit presentation to T cells when DCs were loaded either with antigens as large particles or small peptides. Remarkably, CO did not block the activation of CD8+ T cells when antigen was endogenously expressed by DCs, suggesting that CO was specifically impairing cross-presentation of endocytosed anti-

• gens. As antigen uptake as endocytosis and phagocytosis were not

impaired by HO-1 and CO, the data suggest that inhibition tar- geted the processing/trafficking of soluble antigens internalized

• nly by endocytosis. Because an efficient late endosome–lysosome

fusion is required to generate both class I and class II MHC epitopes from soluble antigens [19–22], we evaluated whether CO could suppress this process. Consistently, we observed that Rab7+-late endosome-Lamp1+ lysosomes fusion was significantly inhibited by CO when DCs internalized soluble antigens, causing an accu- mulation of OVA in degradative late endosomal compartments. In summary, our data suggest that CO inhibits presentation of exoge- nous soluble antigens to CD8+ and CD4+ na¨ ıve T cells by blocking normal antigen trafficking in LPS-treated DCs.

Results

HO-1 and CO prevent DCs from presenting soluble antigens to T cells To evaluate whether CO can modulate the presentation of solu- ble antigens on MHC class I (MHC-I) molecules to CD8+ T cells, DCs were treated with tricarbonyldichlororuthenium (II) dimer (Ru(CO)3Cl2)2 (CORM2) or CO gas to increase intracellular CO levels, washed, and then pulsed either with soluble OVA protein, particulate OVA (OVA adsorbed to 3 µm polystyrene beads), or OVA peptide SIINFEKL (OVA257–264, which binds to H-2Kb to con- stitute the cognate ligand for the TCR expressed by OT-I CD8+ T cells). Next, treated DCs were used to prime na¨ ıve OT-I CD8+ T cells. As shown in Fig. 1A and B, an increase of CO levels induced by CORM2 treatment led to a significant reduction in the capacity

• f DCs to cross-present soluble OVA protein to OT-I T cells, as

compared with untreated DCs (even at different DC:T cell ratios, Supporting Information Fig. 1A). HO-1 induction in DCs by CoPP treatment also led to a reduction on antigen presentation equiv- alent to the CO-driven one (Fig. 1A and B). We confirmed the effects of the CORM2-mediated CO by treating DCs with CO gas either in the presence or absence of soluble OVA and LPS (Fig. 1C, left panel). Remarkably, CO did not inhibit cross-presentation when the antigen was internalized by phagocytosis as latex bead-bound OVA (Fig. 1F and G), at various OVA and BSA ratios bound to the beads. These results suggest that CO blocked cross-presentation only when DCs internalize antigens via endocytosis, but not through phagocytosis. Importantly, CORM2 and CO gas treatment failed to block acti- vation of OT-I T cells in response to DCs pulsed with the antigenic OVA peptide at various concentrations (Fig. 1C (right panel), D, and E) even at different DC:T cell ratios (Supporting Information

• Fig. 1B). These data rule out an unspecific CO inhibitory effect on

the antigen presentation capacity of DCs. Presentation of cytoplasmatic antigens on MHC-I is not inhibited by CO We tested whether CO can modulate MHC-I-restricted presen- tation of endogenous antigens, which also involves proteasomal activity [23,24]. DCs were transfected with 2 µg mRNA encoding OVA SIINFEKL and treated with CORM2 to promote CO produc-

• tion. One hour after transfection, DCs were induced to mature with

LPS and used to stimulate OT-I T cells. As shown in Fig. 1H and I, CORM2 treatment did not diminish the capacity of DCs to present

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2834

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

Figure 1. CO inhibits presentation of endocytosed (cross-presentation) but not endogenous (presentation) OVA to CD8+ T cells. (A) Immature DCs were treated with LPS, CoPP, CORM2, or iCORM2 and then pulsed with 50 µg/mL of OVA. After 16 h, DCs were washed extensively and cocultured with OT-I CD8+ T cells (1/2 DC/T-cell ratio) for 16 h. CD8+ T-cell activation was evaluated by IL-2 secretion in culture supernatants. Data are shown as mean ± SEM of six samples pooled from five independent experiments. (B) CD69 expression by T cells was analyzed by flow cytometry as an activation parameter on the CD8+ cells. (C) Cross-presentation of soluble OVA (left) and peptide OVA (right) to CD8+ T cells after CO gas treatment was measured by IL-2 release. (D) IL-2 secretion by CD8+ T cells after stimulation with CORM2-treated DCs pulsed with various concentrations of antigenic OT-I cognate-peptide OVA257–264 was measured by ELISA. (E) CD69 upregulation by CD8+ T cells after stimulation with CORM2-treated DCs pulsed with various concentrations of antigenic OT-I cognate-peptide OVA257–264 was measured by flow cytometry. (B–E) Data are shown as mean ± SEM of five samples pooled from five experiments performed. (F) DCs pulsed with beads coated with different amounts of OVA (10–0.625 mg/mL OVA, ∼67 beads/cell) were cocultured with OT-I CD8+ T cells. As an antigen-specific control, a group of cells was treated with 10 mg/mL BSA-coated

• beads. Data are shown as mean ± SEM of five samples pooled from five independent experiments. (G) CD69 was measured as an activation marker

after DCs were treated as in (F). (H) CD8+ T-cell activation was measured as IL-2 secretion by OVA mRNA-transfected DCs. (I) T-cell activation was measured as CD69 upregulation on the CD8+ cells by OVA mRNA-loaded DCs. (H-I) Data are shown as mean + SEM of five samples pooled from five independent experiments. Data were analyzed by one-way ANOVA with a posteriori Bonferroni test. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant.

endogenous OVA on MHC-I molecules to OT-I T cells. These results suggest that molecular mechanisms responsible for pro- cessing endogenous antigens are not affected by CO, showing that this molecule interferes specifically with the cross-presentation of endocytosed soluble antigens and not with the overall capacity of DCs to present antigens to T cells. While the expression of costimulatory molecules on murine DCs was not significantly altered by CORM2 (in contrast to rat and humans DCs [6, 15]), CO drastically reduces the production

• f pro-inflammatory cytokines, such as IL-12 but not of IL-10

[15]. Therefore, we tested whether the lack of IL-12 and other DC-derived soluble molecules could impair the capacity of DCs

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2835

Figure 2. CO inhibits MHC class II antigen presentation of endocytosed OVA to CD4+ T cells. (A) Immature DCs were treated with LPS, CoPP, CORM2, or iCORM2 and then pulsed with 50 µg/mL of OVA. After 16 h, DCs were washed extensively and cocultured with OT-II CD4+ T cells (1/2 +DC/T-cell ratio). CD4+ T-cell activation was evaluated by IL-2 secretion in culture supernatants. Data are shown as mean + SEM of 15 samples pooled from five independent experiments. (B) CD69 expression by T cells treated as in (A) was analyzed as an activation parameter on the CD4+ cells by flow cytometry. (C) DCs were pulsed with the OT-II cognate-peptide OVA323–339 at the indicated concentrations. IL-2 secretion by T cells was measured and shown as mean ± SEM of three samples pooled from three independent experiments. (D) CD69 upregulation was measured as an activation parameter of T cells treated as in (C). (E) DCs pulsed with beads coated with 10 mg/mL OVA (OVA-beads) were cocultured with CD4+ T cells and IL-2 production was measured. Data are shown as mean + SEM of six samples pooled from six independent experiments. (F) Activation

• f CD4+ T cells by DCs loaded with OVA-beads and stimulated with LPS was measured by CD69 expression. (B, D–F) Data are shown as mean

+ SEM of six samples pooled from six experiments performed. Data were analyzed by one-way ANOVA with a posteriori Bonferroni test. *p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant.

to cross-present antigens and activate T cells. As shown in Sup- porting Information Fig. 3, neither IL-12 at various concentrations nor conditioned medium from LPS-treated DCs could restore the capacity of CO-treated DCs to cross-present antigens and induce T-cell activation. These results further support the notion that CO does not cause a general impairment of antigen presentation, but rather an intracellular effect on the proper processing of soluble antigens in the cross-presentation pathway. CO inhibits MHC class II restricted presentation of endocytosed soluble antigens It has been shown that once reaching the intracellular space, antigens targeted to class I MHC also can be targeted for class II (MHC-II) restricted presentation in the same lysosomal com- partment [19, 20, 22, 23]. Given this, we next explored whether CO could modulate DC capacity to present internalized soluble

• r particulate antigens to CD4+ T cells, on MHC-II molecules.

CORM2-, CoPP-, or iCORM2-treated DCs were pulsed either with soluble OVA protein, bead-bound OVA, or OVA323–339 pep- tide (cognate ligand for the OT-II TCR when bound to I-Ab). DCs were used to stimulate OVA-specific CD4+ T cells (OT-II T cells). As shown in Fig. 2A and B, both CORM2 and CoPP treatments inhibited significantly the OVA-treated DC capability to activate OT-II T cells. Furthermore, the inhibitory capacity of both CO and CoPP over antigen presentation was also evidenced in a different TCR transgenic model specific for the p25 peptide derived from the BCG Ag85B protein bound to I-Ab MHC (Supporting Information

• Fig. 4).

On the contrary, no significant inhibition of either OT-II or Ag85B-specific T-cell activation was caused by CORM2 when DCs were pulsed with either bead-bound OVA (Fig. 2E and F)/ OVA323–339 peptide (Fig. 2C and D) or with Ag85B-peptide (Supporting Information Fig. 4C and D), respectively. These data suggest that CO can also inhibit MHC-II-restricted presentation to CD4+ T cells of soluble antigens internalized by

• endocytosis. On the contrary, CO failed to suppress activation
• f CD4+ T cells when DCs internalize particulate antigens via

phagocytosis or when they are pulsed with antigenic peptides that bind directly to MHC-II.

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2836

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

Figure 3. CO inhibits neither endocytosis/phagocytosis of antigens nor LPS-mediated upregulation of MHCs molecules by DCs. (A) Immature DCs were treated or not with LPS and either with 100 µM of CORM2 or 100 µM of inactive CORM2 and then pulsed with Alexa488-OVA. Then, cells were analyzed by confocal microscopy. Staining for OVA-AF488 (left) and merge between green and transmission channels (right) are shown. One representative experiment of three performed is shown. Images were captured at 60× original magnification. (B, C) Antigen capture (50 µg/mL OVA-AF488) in time either by CORM2- or CoPP-treated DCs is shown as OVA MFI increment along time progression of the experiment with respect to its initial OVA MFI value (t0 = 15 min) (fold increase). (B) Data are shown as mean ± SEM of three samples pooled from three independent

• experiments. (C) Data are shown as mean ± SEM of three samples from two independent assays. (D) Representative flow cytometry OVA-AF488

histograms for CO gas-, CoPP-, or CORM2-treated DCs are shown. Gray histograms represent fluorescence of cells incubated with OVA-AF488 at 4◦C. (E) Phagocytosis by CO-treated DCs as the percentage of cells that have captured beads (PE-beads+-CD11c+). Data are shown as mean ± SEM of three samples pooled from three independent experiments. (F, G) Immature DCs were treated either with 100 µM of CORM2 or 100 µM of inactive CORM2 and stimulated with LPS for 18 h and analyzed for the expression of MHC-I (H-2Kb) and MHC-II (I-Ab) by flow cytometry. The MFI for MHC-I and MHC-II expression is shown as mean + SEM of three samples pooled from three experiments.

Antigen uptake or MHC expression by DCs is not impaired by CO treatment To determine whether inhibition of antigen presentation on MHC-I and -II by CO was due to reduced antigen uptake, capture

• f either soluble OVA or beads by CO-treated DCs and control DCs

was measured. No significant differences on soluble OVA uptake by either CORM2 treated or CO gas treated DCs were observed (Fig. 3A, B, and D). Also, CoPP was not able to decrease antigen capture (Fig. 3C and D). Similarly, CO also failed to significantly reduce DC phagocytosis of phycoerythrin-coated beads (Fig. 3E). Furthermore, CO treatment did not significantly alter the surface expression level of H-2Kb or I-Ab MHC molecules on the cell sur- face of DCs (Fig. 3F and G). These data suggest that inhibition of

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2837

antigen presentation by CO was not due to a general impairment

• f antigen internalization or to a deficient expression of MHC by

DCs. CO impairs lysosomal destination but not degradation

• f endocytosed antigens in DCs

Given that DCs specialize in antigen trafficking and presentation in response to inflammatory stimuli [25, 26] and the effects of CO were observed only for LPS-treated DCs (Supporting Infor- mation Fig. 3) [15], we studied intracellular routes only for LPS- treated DCs. Dq-OVA assays were performed to evaluate whether reduced antigen presentation to T cells was due to less degrada- tion of intracellular endocytosed OVA by DCs treated with CO. As shown in Supporting Information Fig. 5, no significant reduction

• f OVA degradation in CO-treated DCs was observed after antigen

uptake. To fully understand the process for antigen processing, we analyzed whether CO could interfere with the destination of sol- uble OVA to different intracellular compartments that contribute to the trafficking and degradation of endocytosed antigens. DCs were treated or not with CORM2 and then pulsed with soluble OVA-AF488. One hour after OVA pulsing, we determined the rate of OVA destination to early endosomes (Rab5+) [20, 27] and observed that CO did not cause a significant impairment of OVA targeting of early Rab5+ endosomes (Supporting Information

• Fig. 6).

We next evaluated whether the next step in endosome traffick- ing could be modulated by CO administration by staining OVA- pulsed DCs for Rab7+, a marker for late endosomes, which are capable of degrading protein antigens [27–29]. Importantly, an increased rate of OVA-Rab7 colocalization was observed for CO- treated DCs (Fig. 4A and E). To rule out that the colocalization increase was due to a possible selective driving of OVA to intracel- lular locations with more Rab7+ endosomes by CO, we measured spatial distribution of both OVA and Rab7 along z-axis within each vehicle-, CORM2-, or iCORM2-treated DCs. No significant differ- ences for both colors in Z-distributions were observed (Fig. 4B–D), suggesting that increased OVA-Rab7 colocalization is an end result

• f an accumulation of OVA-containing late endosomes inside the

cell. In contrast, when antigen was internalized by phagocytosis as 3 µm fluorescent beads, CO did not modify the acquisition of Rab7 on the surface of bead-containing vesicles (Supporting Infor- mation Fig. 7), which is consistent with a normal intracellular phagosome trafficking and posterior antigen processing presenta- tion [27] (Fig. 1F and G and 2E and F). In addition, we analyzed whether CO could interfere with the destination of intracellular soluble OVA to lysosomes inside DCs (Fig. 5). DCs were analyzed by confocal microscopy to measure the colocalization rate of internalized soluble OVA with the lysosomal marker Lamp1. As shown in Figure 5A, a significant reduction of the OVA intracellular fraction targeted to lysosomes was observed for CORM2-treated DCs at 1 h after antigen endocytosis when compared with controls (Fig. 5A and C). To evaluate whether decreased fusion between OVA and lyso- somes could be due to a mis-location of Lamp1 inside each CO- treated DCs, we evaluated horizontal (XY) and vertical (Z) distri- butions for this marker in each treatment (Supporting Informa- tion Fig. 8A–D). As shown in Supporting Information Fig. 8A–D, no significant differences were observed between treatments. In addition, total Lamp1 expression was not reduced in CO-treated DCs, which could also be a plausible cause for low OVA–lysosome colocalization (Supporting Information Fig. 8E). Interestingly, in addition to DCs, CO also decreased the targeting of intracellular OVA to Lamp1 compartments in macrophages (Supporting Infor- mation Fig. 9). All together, these data support the notion that CO decreases OVA–lysosome fusion in DCs and other phagocytic cells. Finally, CORM2 treatment failed to significantly decrease the acquisition of Lamp1 over the surface of vesicles containing phago- cytosed beads, independently of the number of intracellular beads (Fig. 5B and D). These data are consistent with the observation that CO treatment failed to block the capacity of DCs loaded with OVA-coated beads to activate T cells (Fig. 1F and G).

Discussion

CO produced by HO-1 in mature DCs has been associated in vivo and in vitro with anti-inflammatory properties [1,15,18], but the mechanism by which it exerts these effects still remains unknown. In murine DCs, HO-1 induction/CO-treatment only suppresses the LPS-induced secretion of pro-inflammatory cytokines, with no sig- nificant inhibition of the expression for maturation surface mark- ers or IL-10 production [15]. Taken together, these observations raise the question whether CO could impair antigen presentation to na¨ ıve T cells. At date, this question remains unresolved. For soluble extracellular antigens, presentation to CD8+ T cells (cross-presentation) by DCs can be achieved when the antigen is degraded either by the proteasome or the lysosome [19,20,23]. On the other hand, the presentation of endogenous antigens to CD8+ T cells by DCs mainly requires proteasomal activity [23,30]. The proteasomal pathway for extracellular soluble anti- gens requires that endosomal cargo reach either early Rab5+- endosomes or the endoplasmic reticulum (ER). The proteasome then degrades antigenic proteins into peptides, which are loaded

• nto MHC-I molecules [23, 31]. Endogenously derived antigenic

peptides are directly transported to the ER and loaded onto MHC-I molecules [23, 30]. In this context, we observed that CO was unable to block the targeting of OVA to Rab5+ early endo- somes or the presentation of endogenous antigens, suggesting that the proteasomal-depending class I presentation was not targeted by CO. MHC-I cross-presentation also depends on antigen transport to lysosomes [19, 22, 32–34]. DCs unable to load OVA-derived peptides on MHC-I in lysosomes show a 50% reduction in their capacity to cross-prime T cells in vivo and in vitro [19]. This sug- gests that destination of either intact or already degraded antigens to MHC I containing lysosomes is a key event that accounts almost

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2838

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

Figure 4. CO promotes destination of endocytosed OVA to late endosomal compartments in DCs. (A) DCs were treated with CORM2, LPS, and soluble OVA-AF488. Cells were then chased at 1 h postchallenge and stained for the late endosomal marker Rab7. Images were obtained by confocal microscopy at 60× magnification. Staining for OVA-AF488 (left), Rab7-AF555 (middle), and merged images (right) are shown. (B) Fold increase of the MFI for each confocal plane (Z-stack) is shown for vehicle-treated DCs. Each MFI for each Z-stack was standardized against the maximum MFI value (MFImax) displayed along the cell. Gaussian distribution graph shows MFI fold increases for both Rab7-AF555 and OVA-AF488. Z-stack progression was standardized by the total amount of stacks analyzed and organized from 0% (bottom part of the DC) to 100% (upper part of the DC). (C) CORM2-treated DCs were analyzed as in (B). (D) iCORM2-treated DCs were analyzed as in (B). (B–D) Data are shown as mean ± SEM of ten random cells analyzed per treatment from two independent experiments. (E) The percentage of OVA-Rab7 colocalization standardized by total intracellular Rab7 fluorescence intensity (Rab7 FI (AU)) to express the rate of colocalization for late endosomal population available within each

• DC. Data are shown as mean + SEM of 60 cells pooled from two experiments performed. Data were analyzed by one-way ANOVA with a posteriori

Bonferroni test. *p < 0.05; ns, not significant.

for half of the cross-priming capacity of DCs. In this context, oth- ers also have shown that normal trafficking of antigens and MHC-I molecules from endosomes to lysosomes is a key process for cross- presentation [22]. Here, we report that CO impairs the targeting

• f OVA to lysosomes thus decreasing the priming of OT-I CD8+

T cells almost in the same proportion (50%) as previously shown for DCs unable to load peptides on MHC-I in lysosomes [19]. The decreased fusion of OVA with lysosomes is a consequence of a stalling in antigen trafficking in late Rab7+-compartments. These compartments have been previously shown to be fully capable of degrading OVA [29]. However, in these compartments MHC-I and MHC-II molecules are not strongly expressed [21, 35]. Further- more, the remaining activation capacity of CD8+ T cells displayed by DCs may result from the class I proteasomal route, which we

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2839

Figure 5. CO impairs endosome–lysosomes fusion. (A) DCs were treated with CORM2, LPS, and soluble OVA-AF488. Cells were then stained for Lamp1 at 1 h postpulse. Staining for OVA-AF488 (left), Lamp1-AF555 (middle), and merged images (right) are shown. (B) DCs were treated either with vehicle, 100 µM CORM2 or 100 µM iCORM2, and then pulsed with fluorescent 3 µm diameter beads in the presence of 10 µg/mL LPS. Cells were then immune stained for Lamp1 at 3 h postpulse. Immune staining with YG-beads (left), Lamp1-AF555 (middle), and merged images (right) are shown. Representative DCs with low and high numbers of intracellular beads are shown. (C) The percentage of intracellular OVA-AF488 colocalizing with Lamp1-AF555 standardized by total intracellular Lamp1 fluorescence intensity (Lamp1 FI (AU)) to express the rate of colocalization for lysosomal population available within each DC is shown. At least 100 cells per treatment were analyzed in each independent experiment. Data are shown as mean + SEM of the analysis of at least eight fields pooled from two experiments. Data were analyzed by one-way ANOVA with a posteriori Bonferroni test. **p < 0.01; ***p < 0.001; ns, not significant. (D) The percentage of Lamp1-coated intracellular fluorescent beads for DCs with different amounts of intracellular cargo (intervals) is shown. Data are shown as mean ± SEM of at least 100 cells analyzed pooled from two experiments. Images were captured at 60× original magnification.

show is not targeted by CO (Fig. 1). Thus, a possible explanation for the normal degradation of Dq-OVA in CO-treated DCs could be that OVA trafficking is stalled in Rab7+ late endosomes where it is degraded, and, together with proteasomal OVA processing, can compensate for the reduced lysosomal destination and degrada-

• tion. However, it is thought that antigenic peptides generated in

Rab7+ vesicles are no longer able to reach class I and class II MHC molecules, which mainly reside in lysosomes for peptide loading during LPS stimulation [22,25,26,36–38] (Fig. 6). Presentation of extracellular antigens on MHC-II molecules to CD4+ T cells requires antigen capture by pinocytosis and traf- ficking to MHC-II+-lysosomes [20, 39, 40]. Here, we provide evi- dence that the interference in the fusion between late endosomes and lysosomes can be the cause for reduced presentation on class MHC-II molecules. In agreement with these data, it has been pre- viously shown that the accumulation of pathogens in Rab7+ vesi- cles decreases late phagosome-to-lysosome fusion, thus reducing the efficiency of antigen presentation [41,42] and that decreased destination of pathogens to Lamp1+ compartments drastically decrease antigen presentation in both class I and class II MHCs [39, 40, 43, 44] because lysosomes have the full repertoire for MHCs [21,35] and peptide mounting during endo/lysosomal traf- ficking mainly occurs in these late acidic vesicles [22,36]. Because all the experiments involving CO-mediated inhibi- tion of OVA presentation were conducted in the context of LPS- stimulated DCs, it is also possible that this gas could be impairing TLR4-driven intracellular signaling. It has been recently shown that Rab7b, a Rab7 family member, can negatively modulate mem- brane TLR4 expression by promoting receptor internalization and lysosomal degradation [45, 46]. As a result, there is an impair- ment of the activation of key factors such as IRF3 and a reduced efficiency of LPS-dependent endosome maturation. Along these lines, it has been shown that Rab7b expression decreases in LPS- treated DCs to diminish its anti-inflammatory effects [47]. We have observed that in CO-treated DCs, some intracellular path- ways are impaired (such as IRF3 activation), suggesting that LPS- driven TLR4 signaling and endosome trafficking could also be decreased [15]. Thus, our data suggest that CO treatment could resemble Rab7b activation. Because we found that OVA+Rab7+ compartments are increased in CO-treated DCs as compared with control DCs, it is likely that this gas could act by modulating the function of proteins from the Rab7 family with a detrimental

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2840

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

Figure 6. Model for impairment of antigen trafficking by CO in DCs. Extracellular antigen (OVA) enters DCs via pinocytosis or mannose receptor mediated endocytosis. Pinocytosis leads OVA to sequentially fuse with early (Rab5+) and late (Rab7+) endosomes, then late endo- somes fuse with lysosomes (Lamp1+) where class I and class II MHC molecules reside. Our data suggest that CO impairs late endosome-to- lysosome fusion, accumulating antigens in Rab7+ compartments that also have the capacity to degrade OVA. On the other hand, mannose receptor mediated capture leads OVA to fuse with early endosomes and to be further processed by the proteasome. Antigen-derived pep- tides reach class I MHC molecules. Endogenous antigens reach class I MHC molecules by the same proteasomal route. CO does not impair the proteasomal route.

• utcome for endosome maturation during antigen trafficking. Fur-

ther analyses are needed to address this model. For cross-presentation of antigen-bound beads, it has been described that early phagosomes rapidly fuse with the ER (ERgosome) [22, 48–51]. In the ERgosome, antigens are rapidly processed by the proteasome and then loaded onto MHC-I molecules even before the ERgosome fuses with lysosomes [48]. Since we observed that CO did not impact proteasomal-dependent exogenous antigen processing, we propose that an early fusion between OVA-bead-containing phagosome and the ER causes a rapid cross-presentation bypassing a possible posterior interfer- ence in fusion with lysosomes (as observed for OVA–endosomes). Such an early fusion has been described in the model for ER- mediated phagocytosis, which is consistent with the size of the particles used in this study [52]. By using this model, it has been recently described that phagocytosed particles reach an MHC-I+ Calnexin+ DiOC6+(ER-proteins) MHC-II+ Lamp1+ compartment that has been suggested to be able to effectively trigger, in an LPS-independent way, both MHC-I cross-presentation and MHC-II presentation [52,53]. Despite being still controversial, phagosome maturation of big particles, as the ones used here, seems to be an LPS-independent process, which could explain why CO does not inhibit the intracellular processing of OVA-bead during LPS stim- ulation in DCs [54,55]. These data provide a possible explanation for as to why CO fails to impair both class I and II MHC presenta- tion for OVA-bound beads, despite having the capacity to suppress intracellular LPS-mediated signaling in DCs [15]. Interestingly, the mechanism by which the ERgosome matu- rates is different from that of endosomes. It has been shown that ERgosomes express ER proteins on their surface whereas endo- somes do not [48, 56]. These ER-derived proteins are acquired after the ER-mediated phagocytosis [52]. Because the proteomic map on the surface of each phagosome drives maturation [27,48], it is likely that the protein composition of ERgosomes renders them resistant of the impairment of maturation caused by CO resulting in acquisition of Rab7 and Lamp1 markers. However, additional studies are required to validate these explanations. In conclusion, the present study shows that CO, a gas molecule produced in cells by HO-1, inhibits antigen presentation of soluble antigens in DCs by decreasing the fusion of endosomes with lyso- somes and thus disrupting OVA trafficking and preventing effec- tive antigen loading on lysosomal MHC-I and MHC-II molecules. This phenomenon can have major implications on the regulation

• f adaptive immunity and supports CO as a potential candidate to

be exploited for the treatment of diseases related to self- or foreign- antigen presentation that generate inflammation and deleterious immune responses.

Materials and methods

Mice and cells C57BL/6, OVA-specific CD8+ (OT-I), and CD4+ (OT-II) TCR trans- genic mice were originally purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and maintained at the germ-free animal facility of the INSERM UMR 1064 or the Pontificia Universidad Cat´

• lica de Chile. OT-I CD8+ or OT-II CD4+ T cells obtained from

spleen and LNs were purified by negative selection using mag- netic beads (Miltenyi Biotech). P25 TCR transgenic mice, in which CD4+ T cells express a T-cell receptor specific for peptide 25 of

• M. tuberculosis antigen 85B (Ag85B) bound to I-Ab, were kindly

provided by Dr. Steven A. Porcelli. BM-derived DCs were pro- duced from C57BL/6 mice as previously described [15]. Briefly, bone marrow cells were flushed from femurs and tibias of mice and then cultured in RPMI 1640 medium containing 10% of fetal bovine serum (FBS), 2 mM glutamine, 100 U/mL penicillin, 0.1 mg/mL streptomycin, 50 µM β-mercaptoethanol (complete culture medium) supplemented with 10 ng/mL of recombinant murine GM-CSF (Prepotech). Fresh complete culture medium plus 10 ng/mL GM-CSF was added at day 3. Nonadherent immature DCs were harvested at day 6, and CD11c+ population was enriched by cell sorting using magnetic beads (purity >95%) (Miltenyi Biotech). Treatment of DCs Immature DCs were treated with CoPP (Frontier Scientific, Carnforth, UK), an inducer of HO-1, as previously described [6].

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2841

Briefly, immature DCs were pulsed for 2 h with 50 µM CoPP, then washed twice, and cultured for 1 h before the addition

• f LPS.

For CO treatment, immature DCs were incubated 5 min with 100 µM CORM2 (Sigma Aldrich, France) and its inactive form, iCORM2 (CORM2 inactivated by leaving it for 48 h at 37◦C in a 5% CO2-humidified atmosphere to liberate CO) as a control. For CO gas treatment, immature DCs were pulsed 10 min on ice with a gaseous mixture containing 450 ppm of CO and 5% of CO2 with a debit of 0.5 L/min. Maturation of treated DCs was then induced by 16 h treatment with LPS (0.5 µg/mL) (Escherichia coli 0111:B4; Invivogen). Supernatants and cells were subsequently harvested for cytokine and phenotypic analysis, respectively. The viability

• f DCs was analyzed in every experiment by flow cytometry of

DAPI-stained cells (Supporting Information Fig. 2) (Molecular Probes). Transfection of DC with OVA mRNA Two million of CO-treated and untreated DCs, resuspended in 200 µL of PBS, were transfected by electroporation (150 µF, 300 V, pulse time of 6 ms) with 2 µg in vitro transcribed RNA encoding a fusion protein including the peptide OVA257–264 SIINFEKL (N-terminal) and the GFP protein (C-terminal). Then, cells were immediately resuspended in prewarmed complete medium plus GM-CSF, plated in 96-well plate, and further incu- bated at 37◦C in a 5% CO2-humidified incubator. One hour after electroporation, DCs were matured with 0.5 µg/mL LPS for 16 h. Antigen internalization determination Endocytosis and phagocytosis were determined using Alexa488- OVA (Invitrogen, France) (OVA-AF488) and fluorescent red, amine-modified, latex beads (3 µm) (Sigma, France), respectively. Either CORM2- or CoPP-treated cells were incubated for different time points at 37◦C. Then, cells were extensively washed with cold PBS and immediately analyzed by FACS. A 4◦C control was included at each incubation time to evaluate the nonspecific endo- cytosis or phagocytosis percentage. In addition, treated DCs were pulsed with Alexa488-OVA and capture was evaluated by confocal microscopy at 1 h postpulse. OVA degradation assays CD11c+ DC cells were treated with 50 µM CoPP, 100 µM CORM2,

• r 100 µM iCORM2. Treated CD11c+ DC cells were then incubated

15 min with 50 µg/mL of DQ-OVA (Invitrogen) at 37◦C in the presence or not of 1 µg/mL LPS. Then, cells were extensively washed with PBS-5% FBS and incubated at 37◦C at different times (in the presence of LPS when it was necessary). We controlled the degradatory process using 4◦C as degradation control. T-cell activation assays DCs treated with CoPP, CORM2, or iCORM2 were incubated with

• r without LPS and soluble EndoOVA protein (endotoxin-free

OVA, Hyglos, Germany) (50 µg/mL), 3 µm EndoOVA-coated latex beads (OVA-beads) (∼67 beads per DC), OVA mRNA trans- fection (2 µg/mL) or with the specific OVA peptides, SIINFEKL (OVA257–264) (OT-I cognate peptide; 10–0.01 nM), or OVA323–339 (OTII; 1 µg/mL to 62.5 ng/mL) (Neosystems, Polypeptide lab-

• ratories, Strasbourg, France) for 16 h. OVA-coated beads were

prepared by attaching OVA (10–0.625 mg/mL) to 3 µm latex beads by passive adsorption in PBS at 4◦C overnight, following by extensive washing in PBS. After incubation with OVA, OVA-beads, OVA mRNA, or peptides, DCs were washed extensively and cocul- tured with OT-I (2.5 × 104 DCs and 2.5 × 104/ 5 × 104 / 10 × 104 OT-I cells) or OT-II (2.5 × 104 DCs and 5 × 104 OT-II T cells) in 96-well round-bottom plates, triplicate. After 20 h of coculture, IL-2 release was measured in the super- natants of the cocultures (in each individual well) by ELISA (BD OptEIA set, BD Pharmingen). IL-2 production induced by OVA- pulsed LPS-treated DC was considered as 100% (absolute num- bers of produced IL-2 for CD8+ and CD4+ T cells ranged between 250–1629 pg/mL and 2090–5809 pg/mL, respectively). For DCs that were transfected with 2 µg OVA mRNA, induction of CD8+ T cells IL-2 production by OVA mRNA and LPS-treated DCs was considered as a 100% (absolute numbers of produced IL-2 ranged between 1316 and 4212 pg/ml). For CD4+ T cells stimulated with OVA beads treated DCs, IL-2 production induced by OVA-beads LPS-treated DCs was considered as a 100% (absolute numbers of produced IL-2 ranged between 1988 and 11 500 pg/mL). Alternatively, cells were harvested (pool of triplicate) and stained with anti-CD8α or anti-CD4 and anti-CD69 antibodies (BD PharMingen) and analyzed by FACS. CD69 expression induced by OVA LPS-treated DC was considered as 100% (absolute MFI for CD8+ and CD4+ T cells ranged between 1833–52 434 and 1058– 21 152, respectively). For DCs that were transfected with 2 µg OVA mRNA, CD69 expression induced by OVA and LPS-treated DCs was considered as 100% (absolute MFI ranged between 1107 and 6563). For activation of CD4+ T cells by OVA-beads loaded LPS- stimulated DCs, CD69 expression induced by OVA and LPS-treated DCs was considered as 100% (absolute MFI ranged between 890 and 15 013). Analyses were performed over either CD4+ or CD8+ gated cells. For the Ag85B antigen, same protocols were used. However, DC-T cell cocultures were performed for 48 h. In replacement for pOVAs, we used the Ag85B-derived peptide p25, as previously described [57,58]. Confocal analyses to determine endosome–lysosome

• r phagosome–lysosome fusion

BMDCs were produced as described above. CD11c+ cells were purified and then attached to poly-L-lysine-activated 12 mm cov- erslips in 24-well plates. Then, cells were treated either with

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2842

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

vehicle, 100 µM CORM2, or 100 µM iCORM2 for 10 min. For endocytic studies, cells were pulsed with 50 µg/mL of OVA-AF488 (Invitrogen, France) for 1 h in the presence of 0.5 µg/mL of

• LPS. For phagocytic studies, cells were pulsed with 1.67 × 107

YG-beads (Polysciences) for 1 or 3 h also in the presence of LPS. Then, cells were washed and fixed with 2% paraformaldehyde for 15 min at 4◦C. Then, cells were washed and permeabilized with 0.05% Saponin-PBS for 15 min at 4◦C. Next, cells were washed again and blocked with 10% FBS–PBS for 30 min in darkness at room temperature. Coverslips were passed to a cold chamber and stained either with 1/200 rabbit-anti-mouse Lamp1 (Abcam), 1/500 rabbit-anti-Rab5 (Cell Signaling), or 1/500 rabbit-anti- Rab7 (Cell Signaling) primary antibody already dissolved in 0.05% Saponin-PBS and incubated overnight at 4◦C. Then, cov- erslips were washed and stained with 1/200 goat-anti-rabbit- AF555 (Invitrogen) secondary antibody dissolved in permeabi- lization buffer for 2.5 h at 4◦C. Next, cells were washed, dried, and mounted for confocal microscopy using DABCO as mounting

• medium. Cells were analyzed in a Confocal Spectral Nikon Eclipse

C2si (using Plan Apo VC60X OIL DIC N2, NA:1.4) and each field was recorded by using NIS element AR V3.2 software. Colocaliza- tion for OVA-AF488/Rab5-AF555, OVA-AF488/Rab7-AF555, and OVA-AF488/Lamp1-AF555 was measured using ImageJ’s plugin Colocalization Highlighter, which gives a binary image. Briefly, for each cell, channels AF488 and AF555 were separated and converted into 8-bit, background subtracted, and analyzed for colocalization using Colocalization Highlighter plugin [59]. To set threshold values, patterns of colocalization for each cell were based on PDM images from correlation intensity analysis, also from ImageJ software [60]. Colocalization was analyzed for mul- tiple Z-stacks in each cell. After the quantification of fluorescence intensity of pixels for both intracellular OVA-AF488 and intra- cellular colocalized OVA-AF488, the percentage of intracellular OVA-AF488 colocalized with either Rab5-AF555, Rab7-AF555, or Lamp1-AF555 was calculated. To avoid artifacts in colocalization measurements by variations in cell-to-cell fluorescence intensities for each endosome/lysosome marker, we normalized the percent- age of colocalization by the total fluorescence intensity for each marker in each analyzed DC. This resembles the rate of destina- tion of intracellular OVA to each specific compartment (shown as percentage of colocalized OVA per FI endosome/lysosome marker per DC). For both YG-beads/Rab7-AF555 and YG-beads/Lamp1- AF555 associations, each cell was analyzed for percentage of intra- cellular YG-beads fully coated with either Rab7 or Lamp1-AF555. This was achieved analyzing different Z-stacks. For both experi- ments, extracellular OVA-AF488 and YG-beads were discarded by Z-stack analysis. Statistical analysis All statistical analyses were performed using GraphPad Prism 5.0

• Software. Statistical significance was assessed using the one-way

ANOVA test with a posteriori Bonferroni test. Differences were considered significant when p < 0.05. Acknowledgements: This work was supported by funding from La R´ egion Pays de la Loire through the “Chaire d’excellence program” for A.M.K. and by the IMBIO program for I.A. Other funding sources include l’Agence de la Biom´ edecine, Minist` ere de la Recherche, Fondation CENTAURE, Fondation Progreffe for I.A., Grant “Nouvelles Equipes-nouvelles th´ ematiques” from the La R´ egion Pays De La Loire, INSERM CDD grant, the ECOS France-Chile grant (to I.A. and A.M.K.), and the Millennium Insti- tute on Immunology and Immunotherapy from Chile (P09/016-F for A.M.K.). L.J.C. is a Pew Latin American fellow. S.A.R. is a CONICYT-Chile fellow. We are grateful to Dr. Pascale Jeannin for helpful discussions as part of the thesis committee of V.T., and to Mar´ ıa Olga Bargsted for manuscript proofreading. Conflicts of interest: The authors declare no financial or com- mercial conflict of interest.

References

1 Ryter, S. W., Alam, J. and Choi, A. M., Heme oxygenase-1/carbon monox- ide: from basic science to therapeutic applications. Physiol. Rev. 2006. 86: 583–650. 2 Tsuchihashi, S., Fondevila, C. and Kupiec-Weglinski, J. W., Heme oxy- genase system in ischemia and reperfusion injury. Ann. Transplant. 2004. 9: 84–87. 3 Blancou, P., Tardif, V., Simon, T., R´ emy, S., Carreno, L. J., Kalergis, A.

• M. and Anegon, I., Immunoregulatory properties of heme oxygenase-1.

Methods Mol. Biol. 2011. 677: 247–68. 4 Otterbein, L. E., Otterbein, S. L., Ifedigbo, E., Liu, F., Morse, D. E., Fearns, C., Ulevitch, R. J. et al., MKK3 mitogen-activated protein kinase pathway mediates carbon monoxide-induced protection against oxidant-induced lung injury. Am. J. Pathol. 2003. 163: 2555–2563. 5 Bouche, D., Chauveau, C., Roussel, J. C., Mathieu, P., Braudeau, C., Tesson, L., Soulillou, J. P. et al., Inhibition of graft arteriosclerosis devel-

• pment in rat aortas following heme oxygenase-1 gene transfer. Transpl.
• Immunol. 2002. 9: 235–238.

6 Chauveau, C., Remy, S., Royer, P. J., Hill, M., Tanguy-Royer, S., Hubert, F. X., Tesson, L. et al., Heme oxygenase-1 expression inhibits dendritic cell maturation and proinflammatory function but conserves IL-10 expres-

• sion. Blood 2005. 106: 1694–1702.

7 Chora, A. A., Fontoura, P., Cunha, A., Pais, T. F., Cardoso, S., Ho, P. P., Lee, L. Y. et al., Heme oxygenase-1 and carbon monoxide suppress autoimmune neuroinflammation. J. Clin. Invest. 2007. 117: 438–447. 8 Xia, Z. W., Zhong, W. W., Xu, L. Q., Sun, J. L., Shen, Q. X., Wang, J. G., Shao, J. et al., Heme oxygenase-1-mediated CD4+CD25 high regulatory T cells suppress allergic airway inflammation. J. Immunol. 2006. 177: 5936– 5945. 9 Braudeau, C., Bouchet, D., Tesson, L., Iyer, S., Remy, S., Buelow, R., Ane- gon, I. et al., Induction of long-term cardiac allograft survival by heme

• xygenase-1 gene transfer. Gene Ther. 2004. 11: 701–710.

10 Chauveau, C., Bouchet, D., Roussel, J. C., Mathieu, P., Braudeau, C., Renaudin, K., Tesson, L. et al., Gene transfer of heme oxygenase-1 and

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

• Eur. J. Immunol. 2013. 43: 2832–2844

Antigen processing

2843

carbon monoxide delivery inhibit chronic rejection. Am. J. Transplant.

• 2002. 2: 581–592.

11 Clarke, H. M., Shrivastava, S., Motterlini, R., Sawle, P., Chen, D. and Dorling, A., Donor HO-1 expression inhibits intimal hyperplasia in unmanipulated graft recipients: a potential role for CD8+ T-cell mod- ulation by carbon monoxide. Transplantation 2009. 88: 653–661. 12 Li, M., Peterson, S., Husney, D., Inaba, M., Guo, K., Kappas, A., Ikehara,

• S. et al., Long-lasting expression of HO-1 delays progression of type I

diabetes in NOD mice. Cell Cycle 2007. 6: 567–571. 13 Soares, M. P., Lin, Y., Anrather, J., Csizmadia, E., Takigami, K., Sato, K., Grey, S. T. et al., Expression of heme oxygenase-1 can determine cardiac xenograft survival. Nat. Med. 1998. 4: 1073–1077. 14 Herrada, A. A., Llanos, C., Mackern-Oberti, J. P., Carreno, L. J., Henriquez, C., Gomez, R. S., Gutierrez, M. A. et al., Heme oxygenase 1 expression is altered in monocytes from patients with systemic lupus erythematosus. Immunology 2012. 136: 414–424. 15 Remy, S., Blancou, P., Tesson, L., Tardif, V., Brion, R., Royer, P. J., Motterlini, R. et al., Carbon monoxide inhibits TLR-induced dendritic cell

• immunogenicity. J. Immunol. 2009. 182: 1877–1884.

16 Kotsch, K., Martins, P. N., Klemz, R., Janssen, U., Gerstmayer, B., Dernier, A., Reutzel-Selke, A. et al., Heme oxygenase-1 ameliorates ischemia/reperfusion injury by targeting dendritic cell maturation and

• migration. Antioxid. Redox Signal 2007. 9: 2049–2063.

17 Listopad, J., Asadullah, K., Sievers, C., Ritter, T., Meisel, C., Sabat, R. and Docke, W. D., Heme oxygenase-1 inhibits T cell-dependent skin inflam- mation and differentiation and function of antigen-presenting cells. Exp.

• Dermatol. 2007. 16: 661–670.

18 Simon, T., Pogu, S., Tardif, V., Rigaud, K., Remy, S., Piaggio, E., Bach, J. M. et al., Carbon monoxide-treated dendritic cells decrease beta1-integrin induction on CD8(+) T cells and protect from type 1 diabetes. Eur. J.

• Immunol. 2013. 43: 209–218.

19 Basha, G., Omilusik, K., Chavez-Steenbock, A., Reinicke, A. T., Lack, N., Choi, K. B. and Jefferies, W. A., A CD74-dependent MHC class I endolysosomal cross-presentation pathway. Nat. Immunol. 2012. 13: 237–245. 20 Burgdorf, S., Kautz, A., Bohnert, V., Knolle, P. A. and Kurts, C., Distinct pathways of antigen uptake and intracellular routing in CD4 and CD8 T cell activation. Science 2007. 316: 612–616. 21 Harding, C. V. and Geuze, H. J., Immunogenic peptides bind to class II MHC molecules in an early lysosomal compartment. J. Immunol. 1993. 151: 3988–3998. 22 Basha, G., Lizee, G., Reinicke, A. T., Seipp, R. P., Omilusik, K. D. and Jef- feries, W. A., MHC class I endosomal and lysosomal trafficking coincides with exogenous antigen loading in dendritic cells. PLoS One 2008. 3: e3247. 23 Burgdorf, S., Scholz, C., Kautz, A., Tampe, R. and Kurts, C., Spatial and mechanistic separation of cross-presentation and endogenous antigen

• presentation. Nat. Immunol. 2008. 9: 558–566.

24 Neefjes, J. and Sadaka, C., Into the intracellular logistics of cross-

• presentation. Front. Immunol. 2012. 3: 1–6.

25 Chow, A., Toomre, D., Garrett, W. and Mellman, I., Dendritic cell matu- ration triggers retrograde MHC class II transport from lysosomes to the plasma membrane. Nature 2002. 418: 988–994. 26 Inaba, K., Turley, S., Iyoda, T., Yamaide, F., Shimoyama, S., Reis e Sousa, C., Germain, R. N. et al., The formation of immunogenic major histocom- patibility complex class II-peptide ligands in lysosomal compartments of dendritic cells is regulated by inflammatory stimuli. J. Exp. Med. 2000. 191: 927–936. 27 Vieira, O. V., Bucci, C., Harrison, R. E., Trimble, W. S., Lanzetti, L., Gruen- berg, J., Schreiber, A. D. et al., Modulation of Rab5 and Rab7 recruitment to phagosomes by phosphatidylinositol 3-kinase. Mol. Cell. Biol. 2003. 23: 2501–2514. 28 Wang, T., Ming, Z., Xiaochun, W. and Hong, W., Rab7: role of its pro- tein interaction cascades in endo-lysosomal traffic. Cell Signal. 2011. 23: 516–521. 29 Tjelle, T. E., Brech, A., Juvet, L. K., Griffiths, G. and Berg, T., Isolation and characterization of early endosomes, late endosomes and terminal lysosomes: their role in protein degradation. J. Cell Sci. 1996. 109(Pt 12): 2905–2914. 30 Compeer, E. B., Flinsenberg, T. W., van der Grein, S. G. and Boes, M., Antigen processing and remodeling of the endosomal pathway: requirements for antigen cross-presentation. Front. Immunol. 2012. 3: 1–111. 31 Kovacsovics-Bankowski, M. and Rock, K. L., A phagosome-to-cytosol pathway for exogenous antigens presented on MHC class I molecules. Science 1995. 267: 243–246. 32 Gromme, M., Uytdehaag, F. G., Janssen, H., Calafat, J., van Binnendijk,

• R. S., Kenter, M. J., Tulp, A. et al., Recycling MHC class I molecules

and endosomal peptide loading. Proc. Natl. Acad. Sci. USA 1999. 96: 10326–10331. 33 Lizee, G., Basha, G., Tiong, J., Julien, J. P., Tian, M., Biron, K. E. and Jefferies, W. A., Control of dendritic cell cross-presentation by the major histocompatibility complex class I cytoplasmic domain. Nat. Immunol.

• 2003. 4: 1065–1073.

34 Sugita, M. and Brenner, M. B., Association of the invariant chain with major histocompatibility complex class I molecules directs trafficking to endocytic compartments. J. Biol. Chem. 1995. 270: 1443–1448. 35 Harding, C. V. and Geuze, H. J., Class II MHC molecules are present in macrophage lysosomes and phagolysosomes that function in the phago- cytic processing of Listeria monocytogenes for presentation to T cells. J.

• Cell. Biol. 1992. 119: 531–542.

36 Dani, A., Chaudhry, A., Mukherjee, P., Rajagopal, D., Bhatia, S., George, A., Bal, V. et al., The pathway for MHC II-mediated presentation of endogenous proteins involves peptide transport to the endo-lysosomal

• compartment. J. Cell Sci. 2004. 117: 4219–4230.

37 Turley, S. J., Inaba, K., Garrett, W. S., Ebersold, M., Unternaehrer, J., Steinman, R. M. and Mellman, I., Transport

• f

peptide-MHC class II complexes in developing dendritic cells. Science 2000. 288: 522–527. 38 Trombetta, E. S., Ebersold, M., Garrett, W., Pypaert, M. and Mellman, I., Activation of lysosomal function during dendritic cell maturation. Science

• 2003. 299: 1400–1403.

39 Herrada, A. A., Contreras, F. J., Tobar, J. A., Pacheco, R. and Kalergis, A. M., Immune complex-induced enhancement of bacterial antigen presen- tation requires Fcgamma receptor III expression on dendritic cells. Proc.

• Natl. Acad. Sci. USA 2007. 104: 13402–13407.

40 Riquelme, S. A., Bueno, S. M. and Kalergis, A. M., IgG keeps virulent Salmonella from evading dendritic cell uptake. Immunology 2012. 136: 291–305. 41 Stein, M. P., Dong, J. and Wandinger-Ness, A., Rab proteins and endo- cytic trafficking: potential targets for therapeutic intervention. Adv. Drug Delivery Rev. 2003. 55: 1421–1437. 42 Clemens, D. L., Lee, B. Y. and Horwitz, M. A., Mycobacterium tuberculosis and Legionella pneumophila phagosomes exhibit arrested maturation despite acquisition

• f

Rab7. Infect. Immun. 2000. 68: 5154–5166. 43 Bueno, S. M., Riquelme, S., Riedel, C. A. and Kalergis, A. M., Mechanisms used by virulent Salmonella to impair dendritic cell function and evade adaptive immunity. Immunology 2012. 137: 28–36.

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu

2844

Virginie Tardif et al.

• Eur. J. Immunol. 2013. 43: 2832–2844

44 Riquelme, S. A., Wozniak, A., Kalergis, A. M. and Bueno, S. M., Evasion of host immunity by virulent Salmonella: implications for vaccine design.

• Curr. Med. Chem. 2011. 18: 5666–5675.

45 Wang, Y., Chen, T., Han, C., He, D., Liu, H., An, H., Cai, Z. et al., Lysosome- associated small Rab GTPase Rab7b negatively regulates TLR4 signaling in macrophages by promoting lysosomal degradation of TLR4. Blood 2007. 110: 962–971. 46 Progida, C., Cogli, L., Piro, F., De Luca, A., Bakke, O. and Bucci, C., Rab7b controls trafficking from endosomes to the TGN. J. Cell Sci. 2010. 123: 1480–1491. 47 Yang, M., Chen, T., Han, C., Li, N., Wan, T. and Cao, X., Rab7b, a novel lysosome-associated small GTPase, is involved in monocytic differentia- tion of human acute promyelocytic leukemia cells. Biochem. Biophys. Res.

• Commun. 2004. 318: 792–799.

48 Guermonprez, P., Saveanu, L., Kleijmeer, M., Davoust, J., Van Endert, P. and Amigorena, S., ER-phagosome fusion defines an MHC class I cross- presentation compartment in dendritic cells. Nature 2003. 425: 397–402. 49 Houde, M., Bertholet, S., Gagnon, E., Brunet, S., Goyette, G., Laplante, A., Princiotta, M. F. et al., Phagosomes are competent organelles for antigen cross-presentation. Nature 2003. 425: 402–406. 50 Ackerman, A. L., Kyritsis, C., Tampe, R. and Cresswell, P., Early phago- somes in dendritic cells form a cellular compartment sufficient for cross presentation of exogenous antigens. Proc. Natl. Acad. Sci. USA 2003. 100: 12889–12894. 51 Ackerman, A. L., Kyritsis, C., Tampe, R. and Cresswell, P., Access

• f soluble antigens to the endoplasmic reticulum can explain cross-

presentation by dendritic cells. Nat. Immunol. 2005. 6: 107–113. 52 Desjardins, M., ER-mediated phagocytosis: a new membrane for new

• functions. Nat. Rev. Immunol. 2003. 3: 280–291.

53 Wan, Y., Wu, Y., Zhou, J., Zou, L., Liang, Y., Zhao, J., Jia, Z. et al., Cross- presentation of phage particle antigen in MHC class II and endoplas- mic reticulum marker-positive compartments. Eur. J. Immunol. 2005. 35: 2041–2050. 54 Yates, R. M. and Russell, D. G., Phagosome maturation proceeds inde- pendently of stimulation of toll-like receptors 2 and 4. Immunity 2005. 23: 409–417. 55 Shiratsuchi, A., Watanabe, I., Takeuchi, O., Akira, S. and Nakanishi, Y., Inhibitory effect of toll-like receptor 4 on fusion between phago- somes and endosomes/lysosomes in macrophages. J. Immunol. 2004. 172: 2039–2047. 56 Guermonprez, P. and Amigorena, S., Pathways for antigen cross presentation. Springer Semin. Immunopathol. 2005. 26: 257–271. 57 Tamura, T., Ariga, H., Kinashi, T., Uehara, S., Kikuchi, T., Nakada, M., Tokunaga, T. et al., The role of antigenic peptide in CD4+ T helper phe- notype development in a T cell receptor transgenic model. Int. Immunol.

• 2004. 16: 1691–1699.

58 Venkataswamy, M. M., Baena, A., Goldberg, M. F., Bricard, G., Im,

• J. S., Chan, J., Reddington, F. et al., Incorporation of NKT cell-

activating glycolipids enhances immunogenicity and vaccine efficacy

• f Mycobacterium bovis bacillus Calmette-Guerin. J. Immunol. 2009. 183:

1644–1656. 59 Lu, A., Tebar, F., Alvarez-Moya, B., Lopez-Alcala, C., Calvo, M., Enrich, C., Agell, N. et al., A clathrin-dependent pathway leads to KRas sig- naling on late endosomes en route to lysosomes. J. Cell Biol. 2009. 184: 863–879. 60 Li, Q., Lau, A., Morris, T. J., Guo, L., Fordyce, C. B. and Stanley, E. F., A syntaxin 1, Galpha(o), and N-type calcium channel complex at a presy- naptic nerve terminal: analysis by quantitative immunocolocalization. J.

• Neurosci. 2004. 24: 4070–4081.

Abbreviations: CO: carbon monoxide · CoPP: cobalt protoporphyrin · CORM2: tricarbonyldichlororuthenium (II) dimer (Ru(CO)3Cl2)2 · HO-1: heme oxygenase-1 · MHC-I: MHC class I · MHC-II: MHC class II Full correspondence: Dr. Alexis M. Kalergis, Millenium Institute of Immunology and Immunotherapy, Facultad de Ciencias Biol´

• gicas.

Pontificia Universidad Cat´

• lica de Chile, Santiago 8331010, Chile

e-mail: akalergis@bio.puc.cl Additional correspondence:

• Dr. Ignacio Aneg´
• n, INSERM, UMR 1064,

Nantes F44093, France Received: 5/4/2013 Revised: 27/6/2013 Accepted: 11/7/2013 Accepted article online: 15/7/2013

C

2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eji-journal.eu