Dectin-1 Activation Controls Maturation of β-1,3-Glucan-containing Phagosomes

Background: Dectin-1 is able to recognize and phagocytose the fungal carbohydrate, (cid:1) -1,3-glucan, but its contribution to phagosomal maturation has not been explored. Results: Dectin-1-dependent Syk activation promotes phagolysosomal fusion and acidification. Conclusion: Dectin-1-dependent Syk-activation permits egress of early phagosomes to mature phagolysosomes. Significance: The surface recognition receptor, Dectin-1 shapes anti-fungal responses by controlling fungal phagosome maturation. receptor, critical for fungal recognition and triggering of Th17 responses, to phagosomal maturation has not been defined. We show that GFP-Dectin-1 translocates to the fungal phagosome, but its signal decays after 2 h. Inhibition of acidification results in retention of GFP-Dectin-1 to phagosome membranes highlighting the requirement for an acidic pH. Following (cid:1) -1,3-glucan recognition, GFP-Dectin-1 undergoes tyrosine phosphorylation by Src kinases with subsequent Syk activation. Our results demonstrate that Syk is activated independently of intraphagosomal pH. Inhibition of Src or Syk results in pro-longedretentionofGFP-Dectin-1tothephagosomesignifyinga link between Syk and intraphagosomal pH. (cid:1) -1,3-glucan phagosomes expressing a signaling incompetent Dectin-1 failed to mature as demonstrated by prolonged Dectin-1 retention, presence of Rab5B, failure to acquire LAMP-1 and inability to acid-ify. Phagosomes containing Candida albicans also require Dec-tin-1-dependent Syk activation for phagosomal maturation. Taken together, these results support a model where Dectin-1 not only controls internalization of (cid:1) -1,3-glucan containing cargo and triggers proinflammatory cytokines, but also acts as a master regulator for subsequent phagolysosomal maturation through Syk activation.

Elimination of fungal pathogens by phagocytes requires phagosome maturation, a process that involves the recruitment and fusion of intracellular proteins. The role of Dectin-1, a ␤-1,3glucan receptor, critical for fungal recognition and triggering of Th17 responses, to phagosomal maturation has not been defined. We show that GFP-Dectin-1 translocates to the fungal phagosome, but its signal decays after 2 h. Inhibition of acidification results in retention of GFP-Dectin-1 to phagosome membranes highlighting the requirement for an acidic pH. Following ␤-1,3-glucan recognition, GFP-Dectin-1 undergoes tyrosine phosphorylation by Src kinases with subsequent Syk activation. Our results demonstrate that Syk is activated independently of intraphagosomal pH. Inhibition of Src or Syk results in prolonged retention of GFP-Dectin-1 to the phagosome signifying a link between Syk and intraphagosomal pH. ␤-1,3-glucan phagosomes expressing a signaling incompetent Dectin-1 failed to mature as demonstrated by prolonged Dectin-1 retention, presence of Rab5B, failure to acquire LAMP-1 and inability to acidify. Phagosomes containing Candida albicans also require Dectin-1-dependent Syk activation for phagosomal maturation. Taken together, these results support a model where Dectin-1 not only controls internalization of ␤-1,3-glucan containing cargo and triggers proinflammatory cytokines, but also acts as a master regulator for subsequent phagolysosomal maturation through Syk activation.
There has been an unprecedented increase in the number of invasive fungal infections owing to the availability of more potent chemotherapeutic and biologic agents (1)(2)(3)(4). Additionally, specific genetic polymorphisms result in increased susceptibility to fungal infections (5). Our ability to define the critical molecular events required for innate immune cells to recognize, phagocytose, and elicit an adaptive immune response is essential to develop strategies for induction of sterilizing immunity (6,7).
Upon ligation, the intracellular domain of Dectin-1 is phosphorylated by Src kinases resulting in activation of spleen tyrosine kinase (Syk) 3 and an intracellular signaling cascade resulting in proinflammatory cytokine production capable of polarizing Th17 cells, essential for anti-fungal immunity (15)(16)(17)(18)(19)(20)(21)(22). Dectin-1 also triggers phagocytosis, and in innate immune cells places cargo in membrane-delineated compartments termed phagosomes. Maturation of these compartments occurs as a result of intracellular protein recruitment and vesicular fusion through protein chaperones such as Rab GTPase family members, or direct fusion with lysosomes. Recent observations from bacterial and fungal model systems demonstrate that specific signaling cascades as well as cross-presentation are influenced by phagosome maturation (23,24). Intraphagosomal sampling by pattern-recognition receptors (PRR) can result in unique signaling cascades (23,25,26). To date, it is not known whether Dectin-1 plays a role in ␤-1,3-glucan or fungal phagosome maturation.
In this study, we explore the contribution of Dectin-1 and subsequent Syk signaling to phagosomal maturation. To reduce the complexity of fungal cell wall ligands, we used a fungal-like platform that displays monodispersed, purified ␤-1,3-glucan on a polystyrene platform (␤-1,3-glucan beads) (27). Our results show that Dectin-1 translocates to ␤-1,3-glucan-containing phagosomes and is retained to the phagosomal membrane in a pH-dependent process. Through the use of chemical inhibitors or a signaling incompetent mutant of Dectin-1 incapable of activating Syk, we demonstrate that Dectin-1-dependent Syk activation is critical for acidification of ␤-1,3glucan-containing phagosomes. Furthermore, we show that Dectin-1-dependent Syk signaling is required not only for phagosomal pH control but also for egress of phagosomes from an early phagosomal stage into mature phagolysosomes. These data indicate that in addition to triggering phagocytosis and the elaboration of pro-inflammatory cytokines, Dectin-1 controls the transition of nascent phagosomes to mature phagolysosomes through a Syk-dependent mechanism.
Generation of GFP-Dectin-1⌬Y15 Mutant-Using wild-type Dectin-1 gene as a template, a tyrosine to phenylalanine substitution at amino acid position 15 within the intracytoplasmic domain (GFP-Dectin-1⌬Y15) in pMAX vector with GFP fused at the N terminus (gift from Dr. David Underhill). The intracytoplasmic substitution was performed using QuikChangeXL site-directed mutagenesis (Agilent Technologies, Santa Clara, CA) with the following primers: 5Ј-GAG AAT CTG GAT GAA GAT GGA TTT ACT CAA TTA GAC TTC AGC AC-3Ј (forward) and 5Ј-GTG CTG AAG TCT AAT TGA GTA AAT CCA TCT TCA TCC AGA TTC TC-3Ј (reverse). The final construct was confirmed by sequencing.
Immunofluorescence Staining of Phagosomes-For phago-FACS of purified phagosomes, 1 ϫ 10 5 phagosomes were immunostained for 30 min at room temperature using 0.1 g anti-LAMP-1 antibody or isotype control and 0.4 g adsorbed anti-rabbit IgG AF647 antibody (Jackson Immunoresearch, West Grove, PA). After washing, phagosomes were assessed using FACS Calibur (BD Biosciences, San Jose, CA). Flow analysis performed using FlowJo software (Tree Star Inc., Ashland, OR). Phagosome index was calculated by determining the percent positive phagosome compared with beads alone multiplied by the geometric mean fluorescence.
Confocal Microscopy-For visualization of subcellular compartments, RAW cells were plated onto 8-chambered slides (LabTek, ThermoScientific, Rochester, NY). Cells were incubated with AF647-␤-1,3-glucan beads for 30 min at 37°C. For visualizing acidic compartments, cells were pre-loaded with lysotracker red (Invitrogen) at 100 nM for 20 min. After washing, cells were mounted on a Nikon Ti-E inverted microscope equipped with a CSU-X1 confocal spinning-disk head (Yok-ogawa, Sugar Land, TX). A Coherent, 4 Watt laser (Coherent, Santa Clara, CA) was used as an excitation light source to produce wavelengths of 488, 568, and 647 nm. To acquire highquality fluorescence images, a high-magnification, high-numerical aperture objective was used (Nikon, 1003, 1.49 numerical aperture, oil immersion). A piezo stage (Prior Instruments, Rockland, MA) capable of X, Y, Z movement was used for z-stack acquisition. A halogen light source and an air condenser (0.52 numerical aperture) were used for bright field illumination. A polarizer (Nikon, MEN51941) and Wollaston prisms (Nikon, MBH76190) were used to acquire differential interference contrast (DIC) images. Emission light from the sample was collected after passage through the appropriate emission filters (Semrock, Rochester, NY). Images were acquired using an EM-CCD camera (Hamamatsu C9100-13, Bridgewater, NJ). Image acquisition was performed using MetaMorph software (Molecular Devices, Downingtown, PA). Images were then cropped using Adobe Photoshop CS5 (Adobe Systems, San Jose, CA).
Western Blot-0.5-1 ϫ 10 6 phagosomes or cell lysate in 1X loading buffer/reducing agent (Invitrogen) were heated to 95°C for 5 min and resolved using 12% SDS-PAGE. Following electrophoresis, gels were removed and methanol-activated PVDF membrane (PerkinElmer, Waltham MA) applied to the gel in transfer buffer (0.025 M Tris, 0.192 M glycine, 20% methanol). All buffer components from National Diagnostics (Atlanta, GA) or Sigma. The gel and PVDF membrane were sandwiched between transfer sponge/blotting paper and subjected to electrophoretic transfer at 100 volts for 1 h at 4°C.
For detection of specific proteins, PVDF-immobilized gel transfers were blocked with 5% milk in PBS-0.02% Tween-20 (PBS-T, Sigma). Blots were incubated with primary antibody in 1% milk in PBS-T for 1 h at room temperature unless noted otherwise. Following washes in PBS-T, rat anti-mouse HRP secondary (Cell Signaling Technology, Danvers, MA) was added at 1:20,000 in 1% milk in PBS-T for 1 h at room temperature. Following washes in PBS, the blot was visualized using Western Lighting Plus ECL chemiluminescent substrate (PerkinElmer, Waltham, MA) on Kodak BioMax XAR Film (Sigma). Films were then scanned, cropped and contrast adjusted evenly to entire image using Adobe Photoshop CS5 (Adobe Systems, San Jose, CA).
Statistical Significance-Student's t test with a value of p Ͻ 0.05 was used as a measure of statistical significance. Representative data are expressed as means ϩ S.E.
Inhibition of Syk-dependent Signaling Results in GFP-Dectin-1 Retention to ␤-1,3-Glucan-containing Phagosomes-Since the dominant intracellular signaling pathway resulting from Dectin-1 ligation is through Src/Syk activation, we sought to determine the potential contribution of Syk to Dectin-1 phagosomal retention. RAW GFP-Dectin-1 cells were pre-treated with PP2 (Src inhibitor), or vehicle and exposed to ␤-1,3-glucan beads. Phagocytosis occurred with normal kinetics in both conditions when examined by light and confocal microscopy (data not shown). More intense GFP-Dectin-1 retention was visualized to the ␤-1,3-glucan-containing phagosomal surface in the PP2 treated cells as compared with vehicle (Fig. 2, panel A). To quantify GFP-Dectin-1 retention to the ␤-1,3-glucan-containing phagosomes, RAW GFP-Dectin-1 cells were pre-treated with PP2, piceatannol (Syk inhibitor), or vehicle. Purified ␤-1,3-glucan-containing phagosomes were isolated from treated cells and GFP-Dectin-1 assessed by phagoFACS (Fig. 2, panel B). As compared with vehicle, the phagosome index shows a significant increase in surface recruitment of GFP-Dectin-1 in the PP2 and piceatannol group (Fig. 2, panel C). These data suggest a direct role for Syk in the control of intraphagosomal pH and subsequent Dectin-1 retention.
Syk Signaling-incompetent GFP-Dectin-1 Demonstrates Prolonged Retention to ␤-1,3-Glucan-containing Phagosomes-Given the potential for off-target effects of chemical kinase inhibitors, we sought to validate the effect of Syk activation to retention using a    (15). RAW GFP-Dectin-1⌬Y15 cells were capable of capturing and completing phagocytosis of ␤-1,3-glucan beads and heat-killed C. albicans though the rate of phagocytosis was marginally slower as compared with wild-type RAW GFP-Dectin-1 (data not shown). Moreover, non-phagocytic HEK 293 cells transduced with GFP-Dectin-1⌬Y15 became phagocytic for ␤-1,3-glucan beads (data not shown). Using purified phagosomes, the kinetics of GFP-Dectin-1⌬Y15 at the phagosomal surface revealed retention of up to 19 h as compared with 2 h with wild-type GFP-Dectin-1 by Western (Fig. 3, panel A). Interestingly, Western analysis shows a "step-down" pattern of GFP-Dectin-1 content from 0.5 to 2 h and another from 4 to 8 h (Fig. 3, panel A). When examined by phagoFACS, the distribution of GFP-Dectin-1⌬Y15 positive phagosomes was not uniform but exhibited a bimodal distribution with a composition of 30% GFP-Dectin-1⌬Y15 high and 70% GFP-Dectin-1⌬Y15 low population (Fig. 3, panel B). This ratio was unchanged even when lower ␤-1,3-glucan bead to cell ratios were used (data not shown) suggesting a receptor "sink" effect is unlikely. Phagosome index analysis illustrates a marginal decay rate of the GFP-Dectin-1⌬Y15 high decorated phagosomes as compared with a similar decay rate with GFP-Dectin-1 wild type or GFP-Dectin-1⌬Y15 low (Fig. 3, panel C). After confirming that Dectin-1-dependent Syk-activation is critical for GFP-Dectin-1 loss from ␤-1,3-glucan-containing phagosomes, we next sought to explore the role of pH on Syk activation.
Dectin-1-dependent Syk Phosphorylation Is Independent of Phagosomal Acidification-Given the dependence of Syk activity for Dectin-1 retention to ␤-1,3-glucan-containing phagosomes, we next investigated the role of phagosomal acidification on Syk phosphorylation. RAW GFP-Dectin-1 and RAW GFP-Dectin-1⌬Y15 cells were incubated with ␤-1,3-glucan beads for various time points followed by whole cell lysis and probed for Syk phosphorylation. Syk was rapidly phosphoryl- ated by GFP-Dectin-1 cells within 30 min of stimulation, which returned to baseline by 5 h (Fig. 4, panel A, arrowhead). In contrast, as expected, GFP-Dectin-1⌬Y15 cells were unable to phosphorylate Syk (Fig. 4, panel A). RAW GFP-Dectin-1 cells pre-incubated with BafA1 and subsequently stimulated with ␤-1,3-glucan beads or heat-killed C. albicans revealed no difference in phosphorylated Syk content compared with vehicle (Fig. 4, panel B). In contrast, PP2-treated cells resulted in significantly reduced phosphorylated Syk species (Fig. 4, panel B). Despite heat-killed C. albicans possessing a more rich complement of pattern-associated molecular patterns (PAMPs), the degree of Syk phosphorylation was significantly reduced with PP2 pre-treatment. These data demonstrate that an acidic intraphagosomal pH environment is not required for Dectin-1specific Syk activation.
Phagosomal Maturation of ␤-1,3-Glucan-containing Phagosomes Is Arrested with Inhibition of Syk or Src Kinases-We next explored the role of Dectin-1-dependent Syk activation to downstream phagosomal maturation including recruitment of Rab GTPase chaperones and lysosomal fusion. Phagosomes undergo a series of surface and biochemical changes following pathogen capture that mark early phagosomal stages such as recruiting surface Rab5B, followed by late stage markers, such as recruitment of lysosome-associated membrane protein-1 (LAMP-1) and the progressive development of intraphagosomal acidic environment. We first examined the effect of Src kinase inhibition on the ability to recruit LAMP-1. RAW GFP-Dectin-1 cells were pre-treated with PP2, BafA1, or vehicle. Purified ␤-1,3-glucan-containing phagosomes were immunostained with antibodies to LAMP-1 and assessed by two-color phagoFACS. ␤-1,3-Glucan-containing phagosomes from vehicle-treated RAW GFP-Dectin-1 cells are GFP-Dectin-1 dim and LAMP-1 high marking maturation into a late phagolysosomal phenotype. BafA1-treated ␤-1,3-glucan-containing phagosomes were GFP-Dectin-1 high , and LAMP-1 high signifying that lysosomal fusion occurred independent of phagosomal acidification. In contrast, PP2-treated cells are unable to recruit LAMP-1 remaining LAMP-1 dim and strongly GFP-Dectin-1 high (Fig. 5, panel A). To confirm lysosomal fusion, purified phagosomes were resolved by SDS-PAGE and probed for LAMP-1. As compared with vehicle, there was no difference with BafA1 treatment, which had an equivalent amount of LAMP-1. On the other hand, PP2 and piceatannol treatment both illustrate a diminished amount of LAMP-1 (Fig. 5, panel B). These results indicate that Src and Syk activation is required for egress from an early to late phagolysosomal stage.
vate Syk results in the arrest of ␤-1,3-glucan-containing phagosome maturation at an early phagosomal stage even within the same macrophage.
C. albicans-containing Phagosomes Remain at an Early Endosomal Stage with Syk Inhibition-Since ␤-1,3-glucan beads do not contain the full complement of cell wall-associated PAMPs found on the surface of C. albicans, we next determined the effect of Syk activation on maturation of C. albicanscontaining phagosomes. RAW GFP-Dectin-1 cells were incubated with heat-killed C. albicans (unlabeled, Fig. 6, panel A, white arrowheads) and pre-treated with piceatannol, BafA1, PP2, or vehicle. Lysotracker was used to demarcate acidified phagosomes under live cell confocal microscopy. Vehicletreated cells show a striking accumulation of lysotracker within C. albicans-containing phagosomes indicating the presence of an acidic pH (Fig. 6, panel A, top row). In addition, C. albicanscontaining phagosomes in the vehicle-treated group showed the normal decay of GFP-Dectin-1 signal from the phagosomal membrane. In contrast, RAW GFP-Dectin-1 cells pre-treated with BafA1 showed no phagosomal accumulation of lysotracker within C. albicans-containing phagosomes, but instead GFP-Dectin-1 retention to the phagosome membrane remained intense. Similar to ␤-1,3-glucan-containing phagosomes, C. albicans-containing phagosomes in Src or Syk inhibitor pre-treated cells were devoid of lysotracker, and retained GFP-Dectin-1 to the phagosomal membrane (Fig. 6, panel A) suggesting blockade of acidification. To confirm these findings using a molecular approach, we next explored the ability of RAW GFP-Dectin-1⌬Y15 cells to undergo phagosome maturation with C. albicans-containing phagosomes. Heat-killed C. albicans were incubated with RAW GFP-Dectin-1⌬Y15 in the presence of lysotracker and imaged by confocal microscopy. After 20 min, C. albicans-containing phagosomes that were GFP-Dectin-1⌬Y15 high had no accumulation of lysotracker, in contrast to GFP-Dectin-1⌬Y15 low , which were strongly lysotracker positive (Fig. 6, panel B). Despite using heat-killed C. albicans, which displays a more complete array of fungal ligands that trigger TLRs, these data suggest that Syk activation is a dominant signal for phagosomal transition from early to late phagolysosomes.

DISCUSSION
Fungal pathogens are recognized by their complex carbohydrate cell walls at the cell surface of innate immune cells by PRR. Surface membrane PRRs including Toll-like receptor (TLR)-2 (31, 32) and C-type lectins such as Dectin-1 are crucial for the elaboration of proinflammatory cytokines. However, there is mounting evidence that a well-coordinated immune response requires continued intraphagosomal cargo-specific sampling (23)(24)(25)(26). While Dectin-1 is essential to the recognition of pathogens, its contribution to intracellular phagosome maturation is not clear. In this study, we assessed the contribution of GFP-Dectin-1 to ␤-1,3-glucan-containing phagosomal maturation in macrophages.
Our experiments demonstrate rapid GFP-Dectin-1 translocation to the phagosome with an off-rate of two hours in contrast to previous publications with off-rate of 20 min (33). Several unique differences including the nature of the cargo may account for this observation; our studies utilize highly purified, homogenous ␤-1,3-glucan beads whereas many studies rely on whole yeast or the crude cell wall extract, zymosan. Interpretation of data from both yeast and/or zymosan poses considerable challenges in that there are several receptor/ligands occurring simultaneously. The loss of phagosomal GFP-Dectin-1 may be FIGURE 6. C. albicans-containing phagosomes are incapable of acidifying following Src or Syk blockade. Panel A, RAW GFP-Dectin-1 cells were preincubated with BafA1, piceatannol, PP2, or vehicle for 30 min. Heat-killed C. albicans (unlabeled) were incubated with RAW GFP-Dectin-1 cells (yeast to macrophage ratio of 1:1) for 20 min in the presence of lysotracker to demarcate acidified compartments. RAW GFP-Dectin-1 cells were then imaged using spinning-disc live cell confocal imaging. Panel B, RAW GFP-Dectin-1⌬Y15 cells were incubated with heat-killed C. albicans (yeast to macrophage ratio of 1:1) for 20 min in the presence of lysotracker and imaged using confocal microscopy. Arrowheads denote intracellular C. albicans. Bar indicates 5 m.
a result of loss of binding as the grooves on the extracellular domains are predicted to interact with glucan through hydrophobic and hydrogen bonding (34). These may be disrupted in a mature phagolysosome. Indeed, our data using BafA1 demonstrates that retention to the phagosome is pH-dependent. Yet, further work will be required to address if Dectin-1 loss is a result of receptor/ligand disruption and/or receptor degradation in the presence of lysosomal enzymes. Recent work shows that Dectin-1 does not recycle following translocation from the plasma surface suggesting the presence of an intracellular sink (35). Another limitation of our study includes the use of macrophage cell lines that may not faithfully reproduce primary macrophage biology. The generation of a GFP-Dectin-1 knock-in mouse may represent the best future tool to study phagosome biology in the native state.
Our results establish Syk-dependent control of fungal phagosomal pH including lysosomal fusion. The dominant signaling cascades downstream of Dectin-1-dependent Syk activation is a complex composed of MALT-1/Bcl-10/Card9 (reviewed in Refs. 36,37), and which of these is responsible for lysosomal fusion, VTPase recruitment and the phagosomal egress from early to mature phagolysosomes remains unclear. A number of alternate signaling candidates could be directed by the Dectin-1-Syk axis including Raf-1 (38), NFAT (39), protein kinase C-␦ (40), and the NLRP3 inflammasome (14). Recently, Dectin-1 has been shown to activate reactive oxygen species and the autophagic machinery including light chain 3, which correlate with enhanced MHC II presentation (41). Interestingly, 30% of GFP-Dectin-1⌬Y15 phagosomes undergo phagosomal arrest suggesting that other Src-or Syk-dependent pro-cesses may contribute to the 70% remaining phagosomes, which mature but remain susceptible to chemical inhibition of Src or Syk kinases. Complement receptor 3, and CD36 are potential Src/Syk-dependent receptors shown to bind glucan and may contribute to phagosomal maturation (8,(42)(43)(44). The difference in the GFP-Dectin-1⌬Y15 bimodal phagosome distribution awaits further proteosome-based experiments.
Through the use of a uniform monodisperse polystyrene ␤-1,3-glucan bead, we have eliminated the potential contribution of other fungal cell wall ligands. Strikingly, as fungi are phagocytosed there is considerable change in the gene and protein expression highlighting the impressive and dynamic rearrangement within phagosomes (45,46). Given the limitation imposed by antigenic heterogeneity, it will be vital to investigate the individual contribution(s) and cross talk of other major fungal cell wall components to phagosomal maturation. Interestingly, our experiments using heat-killed C. albicans-containing phagosomes continue to show limited egress from early endosomal stages suggesting that the control exerted by the ␤-1,3glucan-Dectin-1-Syk axis is likely dominant.
In this study, we sought to determine the role of Dectin-1 in the fate of phagocytosed ␤-1,3-glucan cargo. The model we propose suggests that Dectin-1-dependent Syk activation is required for subsequent lysosomal fusion and acidification of the phagosomal compartment (Fig. 7) and without this checkpoint phagosomal maturation is arrested at an early stage. Taken together, our results indicate that Dectin-1 not only captures and triggers phagocytosis, but also acts as a regulator for ␤-1,3-glucan phagosome maturation through a Dectin-1-dependent Syk mechanism. FIGURE 7. Schematic representation of phagosomal maturation for ␤-1,3-glucan-containing phagosomes. Initial recognition and ligation of GFP-Dectin-1 with ␤-1,3-glucan cargo result in phagocytosis and Syk activation. ␤-1,3-glucan-containing early phagosomes mature and undergo lysosomal fusion resulting in phagolysosome formation. Signaling-incompetent GFP-Dectin-1⌬Y15 forms early phagosomes, but is incapable of activating Syk and does not permit phagolysosomal maturation.