Syk facilitates phagosome-lysosome fusion by regulating actin-remodeling in complement-mediated phagocytosis

Effective phagocytosis is crucial for host defense against pathogens. Macrophages entrap pathogens into a phagosome and subsequently acidic lysosomes fuse to the phagosome. Previous studies showed the pivotal role of actin-remodeling mediated by phosphoinositide-related signaling in phagosome formation, but the mechanisms of phagosome-lysosome fusion remain unexplored. Here we show that in complement-mediated phagocytosis, phagosome-lysosome fusion requires the disappearance of F-actin structure surrounding the phagosome and a tyrosine kinase Syk plays a key role in this process. Using macrophage-like differentiated HL60 and Syk-knockout (Syk-KO) HL60 cells, we found that Syk-KO cells showed insufficient phagosome acidification caused by impaired fusion with lysosomes and permitted the survival of Candida albicans in complement-mediated phagocytosis. Phagosome tracking analysis showed that during phagosome internalization process, F-actin surrounding phagosomes disappeared in both parental and Syk-KO cells but this structure was reconstructed immediately only in Syk-KO cells. In addition, F-actin-stabilizing agent induced a similar impairment of phagosome-lysosome fusion. Collectively, Syk-derived signaling facilitates phagosome-lysosome fusion by regulating actin-remodeling.

Efficient phagocytosis of pathogens is crucial for host defense mechanisms, and macrophages act as professional phagocytes in concert with neutrophils and dendritic cells. Macrophages express a variety of phagocytic receptors, such as the Fc receptor, C-type lectin receptors, and complement receptor 3 (CR3, CD11b/CD18, integrin αMβ2). Phagocytosis is triggered by an association between ligands on the surface of pathogens and receptors on the membrane of phagocytes. Complement-mediated phagocytosis is initiated by binding of complement component C3bi to CR3. Complement system is comprised of cascade reactions that converge on the formation of complement component C3b. C3b and its inactive fragment C3bi are recognized by CR3 on phagocytes. Particularly, C3bi is a strongest inducer of complement-mediated phagocytosis.
Phagocytosis requires dynamic and coordinated reconstruction of the membrane and underlying cytoskeleton. Pathogens are entrapped into a phagosome followed by internalization and subsequently acidic lysosomes fuse to the phagosome 17 . Previous studies showed pivotal roles of phosphoinositide-mediated signaling and actin-remodeling in phagosome formation process but the mechanisms of phagosome-lysosome fusion remain unexplored [18][19][20] . cation, we used two types of fluorescence-labeled zymosan particles, one was fluorescein isothiocyanate (FITC)labeled particles whose fluorescence intensity is reduced with the decrease of pH, and the other was Texas Redlabeled particles whose intensity is conserved independent of pH change (Fig. 2a). Using flow cytometry, we analyzed the ratio of fluorescence intensity of two types of fluorescence-labeled zymosan particles in various pH conditions among pH 3.5-8.0 and determined the standard curve for estimating phagosome pH (Fig. 2b). Then, we quantified the phagosome pH of macrophage-like differentiated HL60 and Syk-KO cells by performing phagocytosis assay using both FITC-and Texas Red-labeled zymosan particles. Flow cytometric analysis showed that pH value of phagosome in parental HL60 cells decreased gradually in a time-dependent fashion and reached below pH 5.5 at 2 h after incubation, but the pH value of phagosomes in Syk-KO cells remained above pH 6. After 5 h, the pH value of phagosomes of parental cells showed pH 5 but that of Syk-KO was around pH 6 ( Fig. 2c), indicating the acidification defect of Syk-KO phagosomes.
Next, we examined the phagosome acidification microscopically using acidic organelle selective fluorescent probe, LysoTracker Green. At 2 h after initiation of phagocytosis, most phagosomes of parental HL60 cells were fluorescence-positive but few phagosomes of Syk-KO cells were positive (Fig. 2d,e). This result also indicates that Syk-KO inhibits acidification of phagosome. These results showed that Syk plays a significant role in facilitating phagosome acidification in complement-mediated phagocytosis.
To ascertain if the acidification of phagosomes is brought about by the function of proton pump, we treated bafilomycin A1, a vacuolar-type H + -ATPase (V-ATPase)-specific inhibitor. Treatment with bafilomycin A1 completely abrogated phagosome acidification of both parental and Syk-KO cells and phagosomal pH was retained around pH 6.5 (Fig. 2f).
These data suggest that phagosome acidification initiated by complement-mediated phagocytosis is due to the function of lysosome-derived proton pump. How Syk facilitates the acidification of the phagosomes? We (e) C. albicans stained with Calcofluor White that was phagocytosed by macrophagelike differentiated parental or Syk-KO HL60 cells for 15 min, followed by wash and incubation for another 8 h.
(f) C. albicans incubated for 3 h in the 50 mM citrate (pH 5.5) or phosphate (pH 6.5) buffer containing 10% FCS. In (e) and (f), DIC indicates differential interference images. Syk facilitates fusion of phagosomes with lysosomes. To examine whether Syk-KO affects the function of proton pump in lysosomes, we treated macrophage-like differentiated cells with pH-sensitive fluorescent probe, FITC-labeled dextran, which is endocytosed and accumulated in the lysosomal compartment. Flow cytometric analysis showed no apparent difference in fluorescence intensity between parental and Syk-KO HL60 cells (Fig. 3a). These results suggest that Syk-KO cells have functional V-ATPase in lysosomes and the defect of phagosome acidification in Syk-KO cells is not due to dysfunction of the proton pump, V-ATPase in lysosomes. Then, we presumed that defect of phagosome acidification in Syk-KO cells is caused by impaired fusion of phagosome with lysosome. We generated two hypotheses as follows. One was that lysosomes failed to reach phagosomes by the defect of intracellular transport and the other was that the final stage of phagosome-lysosome fusion is directly inhibited by an unknown molecular mechanism.
To determine whether lysosome transport into phagosomes is impaired in Syk-KO cells, we investigated the acetylation of α-tubulin at lysine 40 (K40) focusing on intracellular distribution and the amount of modification, because trafficking of lysosomes is reported to be affected by this post-translational modification 21 . Both microscopic analysis and immunoblotting analysis showed little difference in the distribution and in the amount of acetylation of α-tubulin (K40), between Syk-KO and parental HL60 cells (Fig. 3b,c, Fig. S2a-c).
Next, we tracked phagosome dynamics paying attention to the interaction with lysosomes. Imaging analysis revealed that in parental HL60 cells lysosomes fused to phagosomes sequentially, whereas in Syk-KO cells, lysosomes approached close to phagosomes but failed to fuse to phagosomes ( Fig. 3d and Supplementary Video 1, 2).
These results suggest that the lysosomal transport toward phagosomes is intact but subsequent lysosomephagosome fusion is impaired in Syk-KO cells.
Syk promotes phagosome-lysosome fusion by interrupting the formation of F-actin structure around phagosomes. We previously reported that Syk accelerates the engulfment of pathogen by participation in actin dynamics in complement-mediated phagocytosis 10 . Then, we speculated that Syk also contributes to actin dynamics in the regulation of phagosome-lysosome fusion. To test the hypothesis, we analyzed the distribution of F-actin surrounding phagosomes using fluorescence-labeled phalloidin, an F-actin specific indicator. At 15 min after addition of complement-opsonized zymosan particles, in both parental and Syk-KO HL60 cells, almost all phagosomes were surrounded by F-actin but they were not deeply internalized into the cell. At 2 h after further incubation, most of phagosomes were internalized and became free from F-actin in HL60 cells but in Syk-KO cells about 70% of the phagosomes still remained surrounded by F-actin (Fig. 4a,b). To confirm the effect of Syk-KO on F-actin around phagosomes, we utilized Staphylococcus aureus (S. aureus) as another phagocytic target. A similar difference of F-actin surrounding phagosomes was observed between HL60 cells and Syk-KO cells (Fig. S3a, b). Transmission electron microscopy (TEM) imaging also confirmed that phagosomes were surrounded by thick F-actin-like structure only in Syk-KO cells (Fig. 4c). These results suggest that Syk is committed to the F-actin dynamics around phagosomes in relation to internalization process.
To further analyze F-actin dynamics in detail, we tracked phagosomes containing heat-killed C. albicans in the presence of SiR-actin, a fluorescent probe for live-cell imaging of F-actin and LysoTracker Green. As a result, just after phagosome internalization, F-actin surrounding phagosomes clearly disappeared in parental HL60 cells and such phagosomes became free from F-actin. At 2 h after the initiation of phagocytosis, fluorescence-labeled SiR-actin surrounding phagosomes attenuated and instead the fluorescence of LysoTracker Green was detected in the phagosomes of parental HL60 cells (Fig. 5a,b and Supplementary Video 3). In contrast, in Syk-KO HL60 cells, F-actin surrounding phagosomes was once decreased during internalization process, but immediately after the process, F-actin newly appeared and surrounded the phagosomes again and fluorescence of LysoTracker Green was invisible (Fig. 5a,b and Supplementary Video 4). Tracking analysis of the individual phagosomes demonstrated that F-actin surrounding phagosomes disappeared and remained invisible in parental HL60 cells, whereas in Syk-KO cells, F-actin structure was eventually re-constructed around phagosomes after internalization  www.nature.com/scientificreports/ www.nature.com/scientificreports/ even though the extent and kinetics of F-actin reduction/re-acquisition varied among phagosomes (Fig. 5c, and Supplementary Video 3, 4).
To confirm that the disappearance of F-actin is essential for phagosome-lysosome fusion, we treated parental HL60 cells with jasplakinolide, a stabilizing agent of F-actin and performed complement-mediated phagocytosis assay. To avoid any effect of jasplakinolide on the early stage of phagocytosis, we added the drug at 15 min after the initiation of phagocytosis. Fluorescence imaging revealed that treatment with jasplakinolide retained F-actin structure and inhibited fusion with acidic lysosomes (Fig. 5d,e). In addition, we treated human peripheral bloodderived monocytes with jasplakinolide, performed phagocytosis assay and confirmed almost the same effect on F-actin structure of monocytes as the case of HL60 (Fig. 5f,g).
These results suggest that F-actin structure surrounding phagosome interrupted the fusion with lysosomes. In other words, disappearance of F-actin structure surrounding phagosome is essential for the fusion with lysosomes and Syk plays a critical role in this process.
What kind of signaling pathways work at the downstream of Syk in actin remodeling? We analyzed the distribution of phospholipase Cγ2 (PLCγ2) during phagosome formation process, because this protein has been known to act as a regulator of F-actin via binding of actin-binding proteins. PLCγ2 accumulated around phagosomes in parental HL60 cells but in Syk-KO cells the accumulation reduced to about a half (accumulation-positive phagosomes: 58% in HL60 cells, 33% in Syk-KO HL60 cells), although the expression of PLCγ2 was unchanged between parental and Syk-KO HL60 cells by immunoblotting analysis (Fig. 6a,b). Similarly, the expression of actin-regulating proteins, cofilin and vinculin was also unchanged between these cells (Fig. 6b, Fig. S4a-d). Next, to examine the role of phospholipase C activity in F-actin structure surrounding phagosomes, we treated parental and Syk-KO HL60 cells with a phospholipase C inhibitor, U-73122 and performed complement-mediated phagocytosis assay. To avoid any effect of U-73122 on the early stage of phagocytosis, we added this inhibitor at 15 min after the initiation of phagocytosis. Fluorescence imaging revealed that the ratio of phagosomes surrounded by F-actin was increased in parental HL60 cells but not in Syk-KO cells by the treatment with this inhibitor (Fig. 6c,d). Furthermore, we showed that functional Syk (phosphorylated at Tyr525/526) and PLCγ2 were located around both F-actin-surrounded and F-actin-free phagosomes (Fig. S5). These results suggested that the PLCγ2 is one of the possible candidates which regulate F-actin dynamics at the downstream of Syk and determine lysosome fusion.

Discussion
In the present study, we demonstrate that in complement-mediated phagocytosis, phagosome-lysosome fusion requires disappearance of F-actin structure surrounding phagosomes to permit the contact of lysosomes to phagosomes and Syk plays a critical role in facilitating the fusion by regulating actin-remodeling.
We firstly found that larger number of C. albicans phagocytosed by Syk-KO HL60 cells survived than those phagocytosed by parental HL60 cells and found that phagosome acidification is impaired in Syk-KO cells (Figs. 1e,f, 2). Next, we showed that the defect of phagosome acidification in Syk-KO cells is caused by impaired phagosome-lysosome fusion, neither by dysfunction of the proton pump in lysosomes nor by the lysosomal transport toward phagosomes (Fig. 3).
Why Syk-KO leads to impaired phagosome-lysosome fusion? We focused on the effect of Syk on actinremodeling and found that phagosomes in Syk-KO cells were surrounded by thick F-actin structure after internalization into the cells (Fig. 4). Tracking analysis of the individual phagosomes using a F-actin fluorescent probe for live-cell imaging clearly showed that F-actin surrounding phagosomes disappeared and remained invisible in parental HL60 cells, but in Syk-KO cells F-actin structure was re-constructed around phagosomes after internalization (Fig. 5c, and Supplementary Video 3, 4). In addition, treatment with F-actin stabilizing agent, jasplakinolide brought a similar fusion impairment in both parental HL60 cells and human peripheral blood-derived monocytes ( Fig. 5d-g). Collectively, disappearance of F-actin structure surrounding phagosomes is a critical process in the phagosome-lysosome fusion and Syk plays a key role in this process.
What kind of signaling pathways work at the downstream of Syk in actin-remodeling? As a candidate molecule, we pay attention to PLCγ2, because it is known to regulate the amount of PtdIns(4,5)P 2 and act as a regulator of F-actin in phagocytosis 20 . Previously, Syk-PLCγ2 signaling has been reported in immune and hematopoietic cells, B cell receptor-signaling, integrin signaling in neutrophils and megakaryocytes [22][23][24][25] . Our results indicated that at 15 min after initiation of phagocytosis, accumulation of PLCγ2 around phagosomes was reduced in Syk-KO cells (Fig. 6a). Furthermore, treatment with a phospholipase C inhibitor, U-73122 increased the ratio of phagosomes surrounded by F-actin only in parental HL60 cells but not in Syk-KO cells (Fig. 6c,d). Considering both these previous data and our results, the hypothesis might be raised as follows: during the earlystage of complement-mediated phagocytosis, PtdIns(4,5)P 2 produced by PtdIns(4)P-5-kinases accelerates actin polymerization and F-actin structure surrounds phagosomes. After phagosome internalization into the deep cytoplasm, Syk activates PLCγ2 and hydrolyze PtdIns(4,5)P 2 , and its decrease leads to actin de-polymerization. Consequently, Syk might promote actin de-polymerization process via activation of PLCγ2. A previous report showed that phagosome maturation resulting from fusion with lysosome was delayed by transient assembly of F-actin 26 . More recently, it was showed that WHAMM initiates autolysosome tubulation by promoting actin polymerization 27 . Lysosome dynamics might be affected by F-actin. Rho family proteins-related phagosome modulation of F-actin remodeling has also been reported and the involvement of Syk in this pathway might be www.nature.com/scientificreports/ suspected 28 . Our study further illustrates the critical role of Syk-derived signaling that accelerates phagosomelysosome fusion by regulating actin dynamics surrounding phagosomes (Fig. 7).
In the current study, we addressed the role of Syk in phagosome-lysosome fusion in complement-mediated phagocytosis and found that Syk-derived signaling interrupts the reconstruction of F-actin structure around phagosomes and accelerates phagosome-lysosome fusion. Since phagosome-lysosome fusion determines the fate of phagocytozed C. albicans, further studies of this process via Syk will be warranted for host defense against pathogens.
Human peripheral blood mononuclear cells (PBMCs) of healthy volunteers were isolated by density-gradient centrifugation using Ficoll-Paque (Pharmacia Biotech AB, Uppsala, Sweden), and CD14 + monocytes were obtained using the magnetic cell sorting (MACS) system and microbeads conjugated with monoclonal mouse anti-human CD14 antibodies (Cat No. 130-050-201) purchased from Miltenyi Biotec (Bergisch Gladbach, Germany). Isolated monocytes were differentiated into macrophages on 3-well glass-base dishes at a density of 1 ×   In (b, e, g), data show the means ± SD derived from three independent experiments. Results were compared using one-way ANOVA. **p < 0.01; *p < 0.05. Immunoblotting analyses. HL60 cells were lysed with lysis buffer (62.5 mM Tris-HCl, pH 6.8; 2% SDS; 5% glycerol, 5% 2ME, BPB) and the lysates were separated in SDS-PAGE and transferred to PVDF membranes (Merck Millipore, MA, Cat No. IPVH00010). The membrane was blocked with 5% skim milk in T-TBS (10 mM Tris-HCl, pH 7.5; 100 mM NaCl; 0.1% Tween 20) and then incubated with corresponding antibodies in T-TBS for 45 min at room temperature (RT) or overnight at 4 °C. After incubation with HRP conjugated goat antirabbit or goat anti-mouse IgG antibodies for 30 min at RT, specific proteins were detected using an enhanced chemiluminescence immunoblotting system and LAS-3000 Lumino-Image Analyzer (Fuji Film; Tokyo, Japan).

Preparation of C. albicans. C. albicans (NBRC 1594 strain) was obtained from Biological Resource Center
of National Institute of Technology and Evaluation (Tokyo, Japan). C. albicans yeasts were incubated overnight at 37 °C on Sabouraud dextrose agar plate (Nissui Pharmaceutical Co. Ltd., Tokyo, Japan). Heat-killed C. albicans yeasts were obtained by water-bath treatment at 60 °C for 30 min. Killing of yeasts were confirmed by the absence of colonies on Sabouraud dextrose agar. For FCS-induced hyphal formation, C. albicans yeasts (2 × 10 5 ) were suspended in 50 mM citrate buffer (pH 5.5) or 50 mM phosphate buffer (pH 6.5) containing 10% FCS and incubated for 3 h at 37 °C in 3-well glass-base dish. After incubation in each pH condition, the yeasts were fixed in 4% formalin at 4 °C overnight and DIC images were acquired with a laser-scanning confocal microscope (LSM510 META; Carl Zeiss, Oberkochen, Germany).

Analysis of LysoTracker localization in phagosomes.
Macrophage-like differentiated HL60 cells on 3-well glass-base dishes (2.5 × 10 4 cells/200 μl/well) were preincubated with 2 μM LysoTracker Green for 15 min, and further incubated with serum-opsonized Texas Red-labeled zymosan (1 × 10 6 particles/well) for 15 min at 37 °C. After removing unbound zymosan, the cells were incubated in the presence of LysoTracker Green for indicated chase intervals at 37 °C, and microscopic images were acquired with a laser-scanning confocal microscope. For live cell imaging, Incubator PMS (PeCon GmbH, Erbach, Germany) was mounted on the microscope stage and Temperature Module S1 and CO 2 Module S1 were used to maintain the cells at culture condition. For three-dimensional (3D) live cell imaging, a series of Z-stacks were recorded at 0.54 μm intervals, and the images were recorded at 31.43 s intervals for time-lapse live cell imaging. Electron microscopy. Macrophage-like differentiated HL60 cells on 3-well glass-base dishes were incubated with serum-opsonized FITC-labeled zymosan as described above for 2 h. The cells were fixed with 2% paraformaldehyde/2.5% glutaraldehyde/1% tannic acid in PBS, followed by post-fixation in 2% osmium tetroxide, ethanol dehydration and embedding in Epon resin. Thin sections by ultramicrotomy were observed with a transmission electron microscopy.  www.nature.com/scientificreports/ Statistical analysis. In some experiments, statistical significance was determined by the Student's twotailed t-test and one-way analysis of variance (ANOVA).
Ethics approval. This study included the experiments using human subjects. Therefore, this study was approved by the ethics committee of Himeji Dokkyo University based on the Declaration of Helsinki and the ethical guidelines for medical and health research involving human subjects by Ministry of Education, Culture, Sports, Science and Technology of Japan. Primary monocytes and serum were obtained from the peripheral blood of healthy volunteers after informed consent.