The Role of CD38 in Fcγ Receptor (FcγR)-mediated Phagocytosis in Murine Macrophages*

Background: Ca2+ signaling in FcγR-mediated phagocytosis is unclear. Results: We show that FcγR-mediated phagocytosis requires the production of cyclic ADP-ribose by CD38. Conclusion: CD38 plays a crucial role for Ca2+ signaling in FcγR-mediated phagocytosis. Significance: This study provides new perspectives in immune defense and can help shed light on developing novel methods or drugs for manipulating bacterial infections. Phagocytosis is a crucial event in the immune system that allows cells to engulf and eliminate pathogens. This is mediated through the action of immunoglobulin (IgG)-opsonized microbes acting on Fcγ receptors (FcγR) on macrophages, which results in sustained levels of intracellular Ca2+ through the mobilization of Ca2+ second messengers. It is known that the ADP-ribosyl cyclase is responsible for the rise in Ca2+ levels after FcγR activation. However, it is unclear whether and how CD38 is involved in FcγR-mediated phagocytosis. Here we show that CD38 is recruited to the forming phagosomes during phagocytosis of IgG-opsonized particles and produces cyclic-ADP-ribose, which acts on ER Ca2+ stores, thus allowing an increase in FcγR activation-mediated phagocytosis. Ca2+ data show that pretreatment of J774A.1 macrophages with 8-bromo-cADPR, ryanodine, blebbistatin, and various store-operated Ca2+ inhibitors prevented the long-lasting Ca2+ signal, which significantly reduced the number of ingested opsonized particles. Ex vivo data with macrophages extracted from CD38−/− mice also shows a reduced Ca2+ signaling and phagocytic index. Furthermore, a significantly reduced phagocytic index of Mycobacterium bovis BCG was shown in macrophages from CD38−/− mice in vivo. This study suggests a crucial role of CD38 in FcγR-mediated phagocytosis through its recruitment to the phagosome and mobilization of cADPR-induced intracellular Ca2+ and store-operated extracellular Ca2+ influx.

The catalysis of the substrate NAD ϩ to Ca 2ϩ second messenger, cyclic ADP-ribose (cADPR), 4 is made possible through the action of the type II transmembrane glycoprotein CD38 due to its ADP-ribosyl (ADPR) cyclase activity (1,2). Once production is initiated, cyclic ADP-ribose (cADPR) can then act on ryanodine receptors (RyR) located on endoplasmic reticulum/sarcoplasmic reticulum (ER/SR) stores, which will ultimately lead to an increase in intracellular Ca 2ϩ levels ([Ca 2ϩ ] i ) via Ca 2ϩ release in many different types of cells (3)(4)(5). This rise in [Ca 2ϩ ] i is responsible for various mechanisms in immune cells, and it has been implicated in chemotaxis (6), cell adhesion (7), and cytokine secretion (8).
CD38 can also generate another Ca 2ϩ signaling messenger nicotinic acid adenine dinucleotide phosphate (NAADP) depending on the cell system (9,10). Like cADPR, NAADP can be very potent in eliciting a rise in [Ca 2ϩ ] i by acting on receptors of specific Ca 2ϩ stores that are insensitive to thapsigargin (11)(12)(13). It has been suggested that these stores are lysosome-related acidic organelles that can provide a long-lasting [Ca 2ϩ ] i increase similar to ER stores (14 -16).
One of the intriguing aspects of CD38 is what is known as the "topological paradox," where the active site of CD38 is located outside of the cellular membrane (17). This begs to ask the question of how CD38 can catalyze the production of its messengers cADPR and NAADP because the substrates NAD/ NADP as well as the targets for cADPR/NAADP are present intracellularly. It has been suggested that connexin 43 hemichannels, a component of the gap junction, mediate cADPR generation in the extracellular space or intracellular vesicles and its approach to ryanodine receptor by NAD/cADPR transport (18). It has also been suggested that CD38 is internalized once activated via endocytosis, thus allowing its catalytic site to interact with intracellular substrates (19 -21). This internalizing event has been observed in many different cellular responses where cADPR production is shifted from the surface to inside the cell (22,23). It has been suggested that the internalization mechanism is mediated by non-muscle myosin heavy chain IIA (MHCIIA), where both CD38 and MHCIIA were found to be associated in activated lymphokine-activated killer cells (24). Phagocytosis is the mechanism of internalization used by phagocytes to internalize and degrade microorganisms, cell debris, and various particles (25). We have previously reported the possible role of CD38 in Fc␥ receptors (Fc␥R)stimulated phagocytosis where extracellular NAD can help regulate this event in the J774A.1 cell line (26). Thus, there is a possibility that CD38 internalization is related to Fc␥R-mediated phagocytosis, but there currently have been no studies on CD38 internalization in Fc␥R-mediated phagocytosis.
Early studies have shown that the accumulation of [Ca 2ϩ ] i may even be responsible for conducting phagocytosis by controlling many different phenomenon such as phagosomal maturation (26 -28), cytoskeletal rearrangements (29 -31), and phagosome-lysosome fusion (32). There are many different types of receptors on the surface of macrophages that can initiate phagocytosis, such as complement receptors (33,34), mannose receptors (35,36), Sp-A receptors (37), scavenger receptors (38), and subfamilies of Fc␥R. It has been hypothesized that the binding of immunoglobulin-opsonized pathogens with Fc␥R on the plasma membrane is a major factor in mediating this Ca 2ϩ response (39). This was first seen when Ca 2ϩ signals were detected during phagocytosis of opsonized targets in a variety of immune cells (40 -43).
Within the Fc␥R family, there are four different classes of Fc␥Rs, Fc␥RI, Fc␥RII, Fc␥RIII, and Fc␥RIV, where macrophages are known to express all four classes (44,45). Once initiated, the Fc␥Rs will cluster on the outer membrane of macrophages and commence the phosphorylation of immunoreceptor tyrosine-based activation motifs by Src family tyrosine kinases. The phosphorylated immunoreceptor tyrosinebased activation motifs will then gather a variety of signaling enzymes and complexes, which will start a signaling cascade that ultimately leads to phagocytosis (46,47). At this point Syk and PI3K kinases can be activated. Syk in particular can then phosphorylate phospholipase C␥, which then cleaves membrane phospholipid phosphatidylinositol 4,5-diphosphate into inositol trisphosphate (IP 3 ) and diacylglycerol (46 -48), where the former binds to Ca 2ϩ channels on the ER that allows Ca 2ϩ release into the cytosol (49). Any additional Ca 2ϩ signaling that ispossiblyinvolvedafterwardatthispointinmacrophagephagocytosis has not yet been explored.
In this study we show that CD38 is recruited and internalized to the phagosome containing IgG-opsonized particles and induces cADPR production, thereby resulting in intracellular Ca 2ϩ increase and Fc␥R-stimulated phagocytosis enhancement. In addition, we show that CD38 knock-out reduces Fc␥R-stimulated phagocytosis in murine macrophages and inhibits phagocytosis of Mycobacterium bovis BCG in mice. Our results suggest that CD38 is involved in the host's defense to bacterial infection.
Animals-C57BL/6 mice were purchased from OrientBio (Sungnam, South Korea). CD38 knock-out mice were purchased from The Jackson Laboratory. Mice were bred and kept in animal housing facilities at Chonbuk National University Medical School under specific pathogen free (SPF) conditions. All experimental animals were used under a protocol approved by the institutional animal care and use committee of the Chonbuk National University Medical School. Standard guidelines for laboratory animal care were followed.
Cell Culture-The murine macrophage cell line J774A.1 (obtained from ATCC) and macrophages prepped from wildtype C57BL/6 and CD38 Ϫ/Ϫ mice were maintained at 37°C, 5% CO 2 in RPMI supplemented with 10% heat-inactivated FBS, 100 units/ml penicillin, and 100 g/ml streptomycin. The cells were passaged weekly, and cells older than 15 passages were not used.
IgG-opsonized Latex Bead Preparation-Polystyrene latex beads (3.0 m) were washed repeatedly with Hanks' balanced salt solution (HBSS) (1.5 mM CaCl 2 , 145 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 5 mM D-glucose, 20 mM HEPES, pH 7.3) and spun down at 1,700 rpm at 4°C for 10 min. After washing, mouse IgG was added at a concentration of 3 mg/ml. The beads were then incubated at 4°C on a rotator for 8 h to prevent any bead settling and to allow proper binding. After incubation, the beads were resuspended and washed repeatedly with cold buffer to remove any unbound IgG and kept on ice for immediate use.
TRITC IgG-opsonized Latex Bead Preparation-Soluble IgG was dialyzed with PBS at 4°C for 2 h. The buffer was then replaced with 100 mM sodium carbonate buffer (10 mM Na 2 CO 3 , 90 mM NaHCO 3 , pH 8.8), and the IgG solution was again dialyzed at 4°C for an additional 2 h. TRITC was then added to the IgG solution and mixed thoroughly overnight at room temperature. For further purification and to separate any unbound TRITC, the IgGϩTRITC solution was passed through a Sephadex G-75 column and collected. The amount of recovered TRITC-bound IgG was measured with a spectrophotometer at a wavelength of 280 nm. The appropriate amount of TRITC-bound IgG was opsonized to latex beads as mentioned in the above protocol. Polystyrene latex beads were washed repeatedly with HBSS and spun down at 1,700 rpm at 4°C for 10 min. After washing, TRITC-bound IgG was added at a concentration of 3 mg/ml. The beads were then incubated at 4°C on a rotator for 8 h. After incubation, the beads were resuspended and washed repeatedly with cold buffer to remove any unbound TRITC-bound IgG and kept on ice for immediate use.
Preparation of Polyclonal Antibody against M. bovis BCG-Rabbits were given an initial 1-ml subcutaneous injection of a crude sonically treated incomplete Freund's adjuvant (Sigma) mixture. Injections were given at five separate sites. Similarly prepared injections of 1 ml were repeated once after 1 month. Blood was obtained after 1 month, and the sera was isolated and designated as anti-BCG.
Immunohistochemistry-J774A.1 cells were plated on 24-well plates with coverslips at a density of 2.5 ϫ 10 5 cells. Phagocytosis was initiated with TRITC-bound IgG-opsonized latex beads for 30 min at 37°C in 5% CO 2 incubator. Cells were then immediately washed repeatedly with HBSS to remove any unbound beads and fixed with 3.7% paraformaldehyde solution for 1 h at 4°C. The cells were then washed with PBS to remove excess paraformaldehyde and permeabilized with target retrieval solution according to the manufacturer's protocol. Cells were then washed with PBS and blocked with a filtered blocking buffer (1% BSA, 0.1% Triton X-100, 0.02% NaN 3 in PBS) at 4°C for 1 h. Primary FITC-conjugated rat anti-mouse CD38 antibody or rabbit polyclonal MYH9 antibody were diluted with an antibody diluent Preparation of Mouse Macrophage-Briefly, each mouse was injected intraperitoneally with 2 ml of 3% thioglycollate medium. After 5 days the mice were sacrificed, and the thioglycollate-induced macrophages were extracted via peritoneal lavage with DMEM. The macrophages were centrifuged at 1500 rpm at 4°C for 3 min and then resuspended at a cell concentration of 2 ϫ 10 6 cells/ml with RPMI. Macrophages were plated in a 100-mm culture dish and kept in a 5% CO 2 incubator at 37°C for 2 h. Non-adherent cells were then washed off with Dulbecco's phosphate-buffered saline (2.67 mM KCl, 1.47 mM KH 2 PO 4 , 137.93 mM NaCl, 8.06 mM Na 2 HPO 4 ⅐7H 2 O), and the attached macrophages were kept in RPMI in a 5% CO 2 incubator at 37°C. Isolated macrophages were used within 24 h.
Determination of ADPR Cyclase Activity-ADPR cyclase activity was determined fluorometrically using nicotinamide guanine dinucleotide (NGD ϩ ) as a substrate (50). Phagocytosis was initiated in 2 ϫ 10 6 cells per microtube at various time points with IgG-opsonized latex beads. The microtube-containing cells were incubated with 200 M NGD ϩ , pH 7.2, at 4°C for 30 min. The reaction was then stopped with 10% TCA, and the cells were pelleted by centrifugation. Fluorescence of cGDPR produced was determined at excitation/emission wavelengths of 297/410 nm (Hitachi F-2000).
Phagocytic Assay-J774A.1 cells were plated at a density of 1 ϫ 10 6 cells/well in a 6-well plate overnight. The cells were pretreated with various nucleotides for 30 min at 37°C in 5% CO 2 incubator. For phagocytosis, the media was removed and replaced with RPMI containing non-opsonized or IgG-opsonized latex beads for 30 min at 37°C in 5% CO 2 incubator. The cells were then washed twice and fixed with 3.7% paraformaldehyde solution. Phagocytosis was assessed by light microscopy. Images were acquired by using a Nikon Eclipse TS100. Images acquired were captured using AVT-ActiveCam Viewer V1.1.0. The phagocytic index was calculated as follows: phagocytic index ϭ number of latex beads internalized by 100 J774A.1 cells counted in 10 random fields.
Fluorometric Determination of Intracellular Ca 2ϩ Concentration-J774A.1 cells or macrophages extracted from wild-type and CD38 Ϫ/Ϫ C57BL/6 mice were washed twice with HBSS. Cells were then incubated with 5 M Fluo-3 AM in HBSS at 37°C for 30 min. The cells were washed three times with HBSS. Changes in fluorescence in the cells were determined at 488-nm excitation/ 530-nm emission by an air-cooled argon laser system. The emitted fluorescence at 530 nm was collected using a photomultiplier. One image was scanned every 4 s for 10 min using a confocal microscope (Nikon, Japan). For the calculation of [Ca 2ϩ ] i , the method of Tsien et al. (51) where K d is 488 nM for Fluo-3, and F is the observed fluorescence level. Each tracing was calibrated for the maximal intensity (F max ) by the addition of ionomycin (8 M) and for the minimal intensity (F min ) by the addition of EGTA (50 mM) at the end of each measurement.
Isolation of Phagosome and Analysis-Cells were incubated with IgG-opsonized latex bead for 1 h with a multiplicity of infection at 50:1 and washed with PBS. After cell disruption using 30-gauge needle syringes, the latex bead containing phagosomes were isolated on a sucrose gradient (52,53). The isolated phagosome was washed and resuspended with serum-free media for further use in experiments. For the identification of CD38 localization on the phagosome surface, latex bead-containing phagosomes were permeabilized with 0.1% Triton X-100 in serum-free media at 4°C for 20 min and fixed with 4% paraformaldehyde. Thereafter, CD38 and MHCIIA were stained with specific antibodies. For immunoblotting, phagosome pellets were dissolved in 4% SDS in 50 mM Tris-HCl buffer, pH 8.0, by sonication. Solubilized proteins were obtained from the supernatant after centrifugation at 13,300 rpm for 20 min at 4°C and were used for immunoblotting.
Measurement of Intracellular cADPR Concentration ([cADPR] i )-cADPR was measured by some modification of the cycling method described previously (54). Briefly, after IgGopsonized latex bead treatment to 2 ϫ 10 6 cells/microtube at various time intervals, cells were treated with 0.6 M perchloric acid under sonication. Precipitates were removed by centrifugation at 14,000 rpm for 10 min at 4°C. Perchloric acid was removed by mixing the aqueous sample with 3 parts 2 M KHCO 3 with vortex until the top aqueous layer containing cADPR cleared. The samples were then centrifuged at 14,000 rpm for 10 min at 4°C. The top-most aqueous layer was col-lected and neutralized with 0.1 M sodium phosphate, pH 8.0, mixed slightly by hand, and then let on ice. To remove all contaminating nucleotides including NAD but not cADPR, the samples were incubated with the following hydrolytic enzymes overnight at 37°C: 0.44 units/ml nucleotide pyrophosphatase, 12.5 units/ml alkaline phosphatase, 0.0625 units/ml NADase, and 2.5 mM MgCl 2 . Enzymes were removed by filtration using a centrifugal filter unit from Millipore (Billerica, MA) at 14,000 ϫ g for 90 min. The filtrate was then collected for measurement of cADPR. To convert cADPR to ␤-NAD ϩ , the samples were incubated at room temperature for 1 h with the cycling reagent at a 2:1 ratio containing 0.3 g/ml ADPR cyclase, 30 mM nicotinamide, and 20 mM sodium phosphate, pH 8.0. The samples were further incubated with a cycling reagent at 1:1 ratio containing 2% ethanol, 100 g/ml alcohol dehydrogenase, 20 M resazurin, 10 g/ml diaphorase, 10 mM nicotinamide, 0.1 mg/ml BSA, 100 mM sodium phosphate pH 8.0, and 10 M FMN. An increase in the resorufin fluorescence was measured at an excitation of 544 nm and an emission of 590 nm using a SpectraMax Gemini fluorescence plate reader (Molecular Devices Corp.) every 30 min for 3-4 h. Various known concentrations of cADPR were also included in the cycling reaction to generate a standard curve.

Measurement of Intracellular NAADP Concentration ([NAADP] i )-
The level of NAADP was measured using a cyclic enzymatic assay as described previously (55). Briefly, after IgGopsonized latex bead treatment at various time intervals, cells were treated with 0.6 M perchloric acid under sonication. Precipitates were removed by centrifugation at 14,000 rpm for 10 min at 4°C. Perchloric acid was removed by mixing the aqueous sample with 3 parts 2 M KHCO 3 with vortex until the top aqueous layer containing cADPR cleared. The samples were then centrifuged at 14,000 rpm for 10 min at 4°C. The top-most aqueous layer was collected and neutralized with 0.1 M sodium phosphate, pH 8.0, mixed slightly by hand, and then let on ice. To remove all contaminating nucleotides, the samples were incubated with the following hydrolytic enzymes overnight at 37°C: 2.5 units/ml apyrase, 0.125 units/ml NADase, 2 mM MgCl 2 , 1 mM NaF, 0.1 mM PPi, and 0.16 mg/ml NMN-AT. Enzymes were removed by filtration using a centrifugal filter unit at 14,000 ϫ g for 90 min. The filtrate was then collected for measurement of NAADP. For conversion of NAADP to NAAD, samples were incubated with 10 units/ml alkaline phosphatase overnight at 37°C. Enzymes were then removed again by filtration. For conversion of NAAD to NAD, the samples were incubated at room temperature for 1 h with a cycling reagent at a 1:1 ratio containing 0.2 mg/ml NMN-AT, 0.2 mM NMN, 0.5 mM PPi, 10 nM nicotinamide, 2.0 mM NaF, 2.0 mM MgCl 2 , and 100 mM Tris/HCl, pH 8.0. The samples were further incubated with a cycling reagent at 1:1 ratio containing 2% ethanol, 100 g/ml alcohol dehydrogenase, 20 M resazurin, 10 g/ml diaphorase, 10 M FMN, 0.1 mg/ml BSA, and 100 mM sodium phosphate, pH 8.0. An increase in the resorufin fluorescence was measured at an excitation of 544 nm and an emission of 590 nm using a SpectraMax Gemini fluorescence plate reader every 30 min for 3-4 h and once more overnight. Various known concentrations of NAADP were also included in the cycling reaction to generate a standard curve.
Statistical Analysis-Data represent the means Ϯ S.E. of at least three separate experiments. Statistical analysis was performed using Student's t test. A value of p Ͻ 0.05 was considered significant.

IgG-opsonized Latex Beads Induce Fc␥R-mediated Phagocytosis via ADPR Cyclase Activation in J774A.1 Murine
Macrophages-We first wanted to prove the possible presence and role of the transmembrane protein CD38 in J774A.1 macrophages during phagocytosis. We ascertained its expression using a NGD ϩ assay that measures the formation of the fluorescent compound, cyclic GDP-ribose (cGDPR), as a result of its ADPR cyclase activity on the surface of the cells (Fig. 1A). We examined whether the enzyme activity was reduced by the internalization of CD38 through Fc␥R-mediated phagocytosis when adding IgG-opsonized 3.0-m latex beads. This particular size of latex bead was chosen because it was found to have high levels of phagocytic cup formation and no lag in delivery to lysosomes (56). Almost immediately, after the addition of the opsonized beads, we found that the rate of formation of cGDPR significantly decreased over time with a greatest loss in activity at 15 s (Fig. 1A), but we did not observe reducing effect of cGDPR formation activity by non-opsonized latex beads, possibly due to the Fc␥R-mediated internalization of the ADPR cyclase from the plasma membrane for its activation. To visually show that particles can be internalized by Fc␥R-mediated phagocytosis, we treated J774A.1 macrophages with latex beads with or without opsonization of IgG by incubation for 30 min at 37°C, and we saw that the IgG-opsonized beads had a dramatically higher rate of particle ingestion than the unopsonized beads (Fig. 1B). The phagocytosized ratio of non-opsonized LB was very low in comparison in J774.
For further confirmation we carried out an immunohistochemistry study where we stained for CD38 and MHCIIA (Fig.  2). To also verify that the ingested particles were indeed IgGopsonized latex beads, we coated IgG with TRITC before opsonization. Not only did this make it easier to detect the phagocytosed bead, but it also confirmed that IgG was properly opsonized to the target. In the cell normal state, both CD38 and MHCIIA were found to be located mostly along the plasma membrane (Fig. 2, A and B). Once Fc␥R-mediated phagocytosis was initiated with IgG-opsonized latex beads, we found that both CD38 and MHCIIA seemed to internalize along with the opsonized bead (Fig. 2, A and B), confirming previous observations that CD38 is activated through internalization. We also observed the internalization of CD38 independently of the phagosome as well while also confirming that the IgG-opsonized latex beads themselves do not emit autofluorescence that could interfere with the immunohistochemistry interpretation (Fig.  2C). To determine the direction of the C terminus of CD38, which is the active site, we performed immunostaining with a C-terminal-specific antibody against CD38 (M-19, Santa Cruz) and anti-MHCIIA by permeabilization methods with a mild concentration of detergent. As a result, CD38 was not stained in non-permeabilized phagosomes and co-localized with the IgGcoated surface of the latex beads, whereas MHCIIA was stained in both groups with a similar pattern and co-localized with the IgG-coated surface of the latex beads. CD38 on the phagosomes was distinguished by immunoblotting with a biotinylated anti-CD38 antibody and streptavidin-HRP. These results indicate that IgG-opsonized latex beads can elicit Fc␥R-mediated phagocytosis, which seems to be mediated by the ADPR cyclase CD38 via association with MHCIIA, and the C terminus of CD38 is localized in the lumen of phagosomes.
IgG-coated Latex Beads Induce Ca 2ϩ Signaling in J774A.1 Macrophages-After witnessing the possible involvement of an ADPR cyclase, we delved deeper into its action by performing Ca 2ϩ signaling experiments and measuring the changes in intracellular Ca 2ϩ . The Ca 2ϩ -sensitive fluorescent dye Fluo-3 AM was used with or without pretreatment of a variety of clas-sical Ca 2ϩ signaling inhibitors in J774A.1 macrophages before introducing IgG-opsonized latex beads. The addition of IgGopsonized latex beads alone without any inhibitors was able to generate typical long-lasting Ca 2ϩ signals (Fig. 3A). To see if IP 3 is utilized in the Ca 2ϩ signaling pathway of Fc␥R-mediated phagocytosis, we pretreated the cells with 2 M xestospongin C, an IP 3 receptor blocker (57), and found that this had no affect on the Ca 2ϩ signal (Fig. 3B), suggesting that IP 3 is not involved. We previously reported that extracellular cADPR enhanced Fc␥R-mediated phagocytosis (25). To corroborate our previous finding that cADPR is involved in Fc␥R-mediated phagocytosis, we pretreated our cells with a 100 M concentration of an antagonistic analog of cADPR, 8-Br-cADPR (58), and we observed that preincubation with this inhibitor was able to  . Two or three-dimensional fluorescence images of CD38 and myosin heavy chain IIA in J774A.1 murine macrophages. CD38 and MHCIIA internalization via Fc␥R-mediated phagocytosis was visualized through immunohistochemistry with a primary FITC-conjugated rat antimouse CD38 or a rabbit polyclonal MYH9 antibody, respectively, as described under "Experimental Procedures." A, CD38 (green) was found to be situated along the plasma membrane in its resting state (control). Initiation of Fc␥Rmediated phagocytosis with TRITC IgG-opsonized latex beads (LB, red) resulted in CD38 internalization where it co-localized with the phagosome. DIC, differential interference contrast. B, MHCIIA (green) was also found to be situated along the plasma membrane in its resting state (control). Initiation of Fc␥R-mediated phagocytosis with TRITC IgG-opsonized latex beads (red) resulted in MHCIIA internalization, where it also colocalized with the phagosome. C, CD38 (green) also internalizes independently of the phagosome after initiating Fc␥R-mediated phagocytosis with IgG-opsonized latex beads. The blue line shows the position of the phagosome on the z axis. The x (green line), y (red line) image is the confocal z axis slice corresponding to the position of the blue line (i.e. through the center of the phagosome). The phagosome is stably positioned within the cell. D, shown are co-localized images of CD38 or MHCIIA with phagocytosed-latex bead. Latex bead-containing phagosomes were isolated as described under "Experimental Procedures" and stained with anti-CD38 and anti-MHCIIA antibodies with or without permeabilization. E, shown is a Western blot of CD38 and IgG with biotinylated anti-CD38 antibody and anti-mouse IgG antibody on latex beads (LB), latex bead-containing phagosome (phagosome LB), and whole lysate of J774.A1 cells.
abolish the long-lasting Ca 2ϩ signal (Fig. 3C). cADPR is also known to act on ryanodine receptor for Ca 2ϩ release (59). For additional confirmation we used 20 M ryanodine, as it antagonistically binds to ryanodine receptor on ER/SR Ca 2ϩ stores (60). We discovered that this provided very similar results as with 8-Br-cADPR, thus providing some consistency (Fig. 3D). Because NAADP is known to be synthesized by CD38, we wanted to see its potential role by preincubating our cells with 600 nM bafilomycin A1, a vacuolar H ϩ -ATPase inhibitor that is required for maintaining acidity of NAADP-sensitive Ca 2ϩ  APRIL (61). Surprisingly, this was not able to abrogate the longlasting phase of the Ca 2ϩ signal (Fig. 3E). CD38 is known to be activated via its internalization from the plasma membrane through its association with phosphorylated MHCIIA by protein kinase G (23). To verify this notion, we treated J774A.1 macrophages with a specific MHCIIA inhibitor, blebbistatin (62), which completely blocked the IgG-opsonized latex beadinduced Ca 2ϩ signaling (Fig. 3F). These findings reveal that Fc␥R-mediated phagocytosis by way of IgG-opsonized latex beads induce Ca 2ϩ signaling that is mediated by CD38/cADPR. Ca 2ϩ Signaling Inhibitors Greatly Reduce Fc␥R-mediated Phagocytosis in J774A.1 Murine Macrophages-Based on the Ca 2ϩ signaling data, we proceeded to further quantify our results by calculating the phagocytic index of J774A.1 macrophages incubated under varying conditions for 30 min at 37°C before inducing phagocytosis with IgG-opsonized latex beads (Fig. 4). Through microscopic imaging, we found that preincubation with bafilomycin A1 and xestospongin C did not significantly affect phagocytic capability, where full ingestion of IgGopsonized latex beads was seen as in the untreated J774A.1 cells. This is concordant with our previous Ca 2ϩ measurement data where these two inhibitors had no affect on the signaling. On the other hand, we observed a notable 46% decrease of phagocytosis when cells were preincubated with 8-Br-cADPR, where IgG-opsonized latex beads were seen mostly attached to the outer periphery of macrophages, and full internalization of the beads was decreased. Similar results were also obtained when macrophages were preincubated with the MHCIIA inhibitor, blebbistatin, where we found a 49% decrease. These observations affirm that cADPR is involved in Fc␥R-mediated phagocytosis in J774A.1 macrophages, and NAADP seems to not be a part of this particular signaling system.

CD38 in Fc␥R-mediated Phagocytosis
J774A.1 Macrophages Utilize Store-operated Ca 2ϩ Entry-It has been reported that store-operated Ca 2ϩ entry (SOCE) may be involved in macrophage phagocytosis (63)(64)(65)(66)(67). To see which intracellular Ca 2ϩ stores were present within J774A.1 cells, Ca 2ϩ measurement was done on thapsigargin-induced Ca 2ϩ release in a Ca 2ϩ free buffer, where thapsigargin is a sarco/ endoplasmic reticulum Ca 2ϩ -ATPase (SERCA) inhibitor (67). This was able to cause a rapid release of Ca 2ϩ from ER/SR stores, indicating their presence in the intracellular space (Fig.  5A). The addition of extracellular Ca 2ϩ was also able to elicit a rapid influx of Ca 2ϩ , suggesting the presence of Ca 2ϩ channels. To see the contribution of acidic-like organelles, we performed Ca 2ϩ measurements on GPN-induced Ca 2ϩ release in a Ca 2ϩfree buffer as well (Fig. 5B). Unlike the thapsigargin-induced Ca 2ϩ release, we were not able to detect any noticeable releases of intracellular Ca 2ϩ , and the addition of extracellular Ca 2ϩ did not cause any influx. This indicates a lack of role of acidic-like organelles as an intracellular Ca 2ϩ store, which remains consistent with our earlier Ca 2ϩ measurement, where bafilomycin A1 had no affect on Fc␥R-mediated phagocytosis.
To see which Ca 2ϩ channels were involved, we performed Ca 2ϩ measurement experiments using different Ca 2ϩ channel inhibitors. We first preincubated J774A.1 macrophages with 10 M nifedipine, which is a voltage-dependent L-type Ca 2ϩ channel blocker (68). Surprisingly, this was not able to abrogate any increases of [Ca 2ϩ ] i in the cells (Fig. 5C). As a result, we used 50 M concentration of another Ca 2ϩ channel blocker, SK&F 96365, which also inhibits SOCE (69). Interestingly, this was able to abolish the Ca 2ϩ signal; however, we were able to detect an initial sharp rise in Ca 2ϩ (Fig. 5D). This similar pattern was also observed when the cells were treated in a Ca 2ϩ -free buffer with 0.5 mM EGTA and was completely inhibited when treated in a Ca 2ϩ -free buffer with 100 M Fura-2AM (Fig. 5E), suggesting that the main method of Ca 2ϩ influx from the extracellular environment may be mediated through the emptying of intracellular Ca 2ϩ stores, hinting at a store-operated mechanism. To determine whether the initial Ca 2ϩ peak is mediated by cADPR, J774A.1 cells were incubated with both SK&F 96365 and 8-Br-cADPR (Fig. 5F). Our results showed that this was able to effectively block any [Ca 2ϩ ] i rise in the cells. This outcome made us conclude that SOCE is involved during Fc␥R-mediated phagocytosis, where the initiator of extracellular Ca 2ϩ influx is cADPR-mediated Ca 2ϩ release from ER/SR stores.
To further explore the possible involvement of SOCE, various SOCE inhibitors were also preincubated with J774A.1 cells, and the phagocytic index was again measured (Fig. 5H). The SOCE channel blocker SK&F 96365 and thapsigargin were able to greatly reduce the number of internalized beads by 38 and 47%, respectively. The same was also seen when the phagocytic index was determined in a Ca 2ϩ -free buffer with a 34% decrease in bead ingestion. These results further affirm our conclusion that Fc␥R-mediated phagocytosis in J774A.1 macrophages is controlled by both the mobilization of Ca 2ϩ from intracellular Ca 2ϩ stores as well as the influx of Ca 2ϩ from the extracellular space through a Ca 2ϩ channel.

CD38 in Fc␥R-mediated Phagocytosis
Fc␥R-mediated Phagocytosis Induces cADPR Formation in J774A.1 Macrophages-Because our Ca 2ϩ signaling data exhibited that 8-Br-cADPR was able to ablate the somewhat longlasting Ca 2ϩ signal and substantially reduce the phagocytic index, these considerations propelled us to measure [cADPR] i in J774A.1 macrophages after initiating Fc␥R-mediated phagocytosis with IgG-opsonized latex beads. We measured [cADPR] i within a time course and saw a significant rise in [cADPR] i within the first 5 s after IgG-opsonized latex bead addition, where levels remained elevated until 20 s, after which [cADPR] i tapered off to resting values (Fig. 6A). Interestingly, we were able to see a rise in [cADPR] i with unopsonized latex beads as well, although this increase was very slight, suggesting the possible involvement of cADPR in non-Fc␥R-mediated phagocytosis.
The use of bafilomycin A1 had no notable effects on both Ca 2ϩ signaling and the phagocytic index of J774A.1 macro- phages. To validate these observations, we also measured [NAADP] i . Unsurprisingly, we were not able to detect any noticeable levels of [NAADP] i after evoking phagocytosis (Fig.  6B). This result remains in accordance with another finding that [NAADP] i is not detected in J774A.1 macrophages using another NAADP measurement method (70,71). These aforementioned data made us conclude that Fc␥R-mediated phagocytosis with IgG-opsonized latex beads induces cADPR, but not NAADP, production.
IgG-opsonized Latex Beads Induces Ca 2ϩ Signaling in Wildtype but Not in CD38 Ϫ/Ϫ Mice-Because we have now established the role of Ca 2ϩ signaling and a possible ADPR cyclase, such as CD38, in murine macrophage phagocytosis in an established cell line, we then tested these same principles ex vivo between wild-type and CD38 Ϫ/Ϫ mice. Ca 2ϩ signaling data using IgG-opsonized latex beads on extracted wild-type macro-phages showed a typical Ca 2ϩ response involving an initial and a somewhat long-lasting signal (Fig. 7A), very similar to what was previously noted in normal J774A.1 cells. The same Ca 2ϩ signaling measurements were then performed on extracted CD38 Ϫ/Ϫ macrophages, where we observed total ablation of any Ca 2ϩ signal (Fig. 7B). These results suggest that CD38 is a possible candidate as an ADPR cyclase that is responsible for producing the Ca 2ϩ signaling messengers necessary for allowing a rising level of [Ca 2ϩ ] i .
CD38 Ϫ/Ϫ Mouse Macrophage Has Greatly Reduced Fc␥Rmediated Phagocytosis-The results of our Ca 2ϩ signaling data in both wild-type and CD38 Ϫ/Ϫ macrophages prompted us to visually examine the differences in Fc␥R-mediated phagocytosis. Microscopic images of wild-type macrophages clearly show full engulfment of IgG-opsonized latex beads (Fig. 7C). Macrophages obtained from CD38 Ϫ/Ϫ mice, however, showed a markedly impaired ability of the cells to commence phagocytosis, where little ingestion of opsonized particles was seen (Fig.  7C). Quantification of the data reveals that CD38 Ϫ/Ϫ macrophages have a considerable decrement in the phagocytic index of more than 50% when compared with wild-type macrophages (Fig. 7D). This outcome again intimates that CD38 is the ADPR cyclase accountable for phagocytosis mediated by Fc␥R.
CD38 Deficiency Leads to Decreased Phagocytosis in Vivo-We further examined whether CD38 deficiency leads to compromising phagocytosis for live bacteria. We injected mice subcutaneously with M. bovis BCG and 3 weeks later injected  intraperitoneally with opsonized M. bovis BCG and compared phagocytic M. bovis BCG in peritoneal macrophages isolated after bacterial injection. Wild-type macrophages showed phagocytosed M. bovis BCG, whereas macrophages obtained from CD38 Ϫ/Ϫ mice showed a significantly reduced phagocytic ingestion of M. bovis BCG (Fig. 8A). Flow cytometric analysis of phagocytic M. bovis BCG with biotinylated rabbit anti-BCG polyclonal antibody also shows no detectable phagocytic signal in macrophages obtained from CD38 Ϫ/Ϫ mice compared with that in macrophages obtained from wild-type mice (Fig. 8B).

DISCUSSION
In this study our data clearly show the crucial role that CD38 has to play in regulating Ca 2ϩ signals that is necessary for Fc␥Rmediated phagocytosis of IgG-opsonized latex beads. This is made possible through its internalization where the production of cADPR, but not NAADP, is allowed to take place. Although the role of IP 3 , cADPR, and CD38 has been implied for immune cells in other works, a detailed investigation of the direct relationship between CD38 and Fc␥R-mediated phagocytosis in murine macrophages has not yet been made clear. We are also the first to show ex vivo the importance of CD38 and its necessary involvement in phagocytosis, where macrophages prepared from CD38 Ϫ/Ϫ mice showed a non-existent Ca 2ϩ signal as well as a drastically reduced uptake of IgG-opsonized latex beads as well as live bacteria. This is consistent with the phenotypic traits of CD38 Ϫ/Ϫ mice where they tend to be more susceptible to infection and thus have a weaker survival rate (72). This could be due to the reduced ability for immune cells such as macrophages to uptake foreign microbes. This is the first time to our knowledge that anyone has shown the possibility of CD38 being internalized along with the phagosome as it is endocytosed. It must be noted, however, that this was not always seen in our case, where we also observed the internalization of CD38 independently of the ingested IgG-opsonized latex bead. This internalization may be important in the physiological aspect because this will allow the rise in [Ca 2ϩ ] i to be more localized as seen in other Ca 2ϩ studies with phagocytosis where there are spatial and temporal aspects to the signaling (72,73), thus allowing for better control during phagocytic ingestion. This may be important for downstream effectors as a result of the increase in [Ca 2ϩ ] i , such as Ca 2ϩ -dependent gelsolin, an actin-binding protein, that has been documented to play a part in actin remodeling through assembly/ disassembly during phagocytosis (29, 74 -77). Whether there is a direct relationship between the Ca 2ϩ that is induced by the CD38 product cADPR and gelsolin activity is highly likely but remains to be confirmed.
CD38 being internalized along with the phagosome-containing IgG-opsonized particles may have implications in not only phagocytic ingestion but in phagosome maturation as well. In one study with monocytes, a thin rim of high [Ca 2ϩ ] i was noted surrounding the phagosome of the ingested opsonized particles (76). We also observed this more localized form of [Ca 2ϩ ] i increase, unlike the more global rises seen in other systems. This confined Ca 2ϩ around the phagosome during the later stages of phagocytosis has been implicated in oxidative enzyme-containing vesicle fusion with the phagosome (14,27,31) as well as ROS production (77)(78)(79) where blocking Ca 2ϩ at this point impairs both. During these processes it is possible that CD38 may be an important player in providing the Ca 2ϩ necessary for these events to take place, where its internalization can continue to produce cADPR at the local level.
Rapid cADPR production was seen during Fc␥R-mediated phagocytosis within a few seconds after initiation. This coincides with our NGD ϩ assay where surface ADPR cyclase activity decreased in a time-dependent manner as well as our [Ca 2ϩ ] i level measurements, where it increased. This may seem contradictory especially in the context where it has been reported that it can take anywhere from 30 s to a few minutes for the actual phagocytosis process to begin (80). However; it must be again noted that CD38 can be internalized independently from the opsonized particles, as we have observed. So while the IgGopsonized latex bead is being phagocytosed after binding to Fc␥Rs, this can send a signaling cascade to the other surround- CD38 ϩ/ϩ and CD38 Ϫ/Ϫ mice (each, n ϭ 5) were injected subcutaneously with M. bovis BCG (1 ϫ 10 7 colony-forming units/mouse). Three weeks later mice were injected intraperitoneal with opsonized M. bovis BCG (1 ϫ 10 7 colonyforming units/mouse). Peritoneal macrophages were then isolated at 24 h after bacterial injection and stained with fluorescent auramine-rhodamine. A, cells containing phagocytic M. bovis BCG in five different fields for each condition were counted. Representative immunofluorescence images of three independent replicates are shown (inset). Scale bars ϭ 10 m. The arrow indicates auramine-rhodamine-stained M. bovis BCG. B, shown is flow cytometric analysis of phagocytic M. bovis BCG with biotinylated rabbit anti-BCG polyclonal antibody. Biotinylated antibody was detected with streptavidine-cy3 and analyzed with a flow cytometer. Streptavidine-Cy3 only was stained (blue line); biotinylated anti-BCG antibody and streptavidine-Cy3 were stained (red line). Data are expressed as mean Ϯ S.D. p Ͻ 0.001 (*) versus CD38 ϩ/ϩ ) (n ϭ 5). APRIL 27, 2012 • VOLUME 287 • NUMBER 18 ing CD38 to internalize, presumably via endosomes (81). This is what may be responsible for the quick production of cADPR i so that early Ca 2ϩ signals can be generated in preparation for phagocytosis to take place. At present, we were not able to detect any distinguishable levels of NAADP i, suggesting that Ca 2ϩ generation in Fc␥R-mediated phagocytosis in murine macrophages relies solely on the action of cADPR on ER/SR Ca 2ϩ stores. SOCE seems to work synergistically with the release of intracellular Ca 2ϩ from ER/SR stores with cADPR being the initiator of this Ca 2ϩ mobilization. This is attributed to the fact that SOCE inhibitor SK&F 96365 in combination with 8-Br-cADPR effectively diminishes [Ca 2ϩ ] i levels and reduces the rate of IgG-opsonized latex bead phagocytic uptake. The usage of the L-type Ca 2ϩ channel blocker nifedipine did not affect [Ca 2ϩ ] i levels in our study, suggesting that the Ca 2ϩ channels contain characteristics that are more receptor-mediated or SOCE rather than voltage-gated. This is in accordance with another study where the usage of SK&F 96365 was able to eliminate the later rise in [Ca 2ϩ ] i levels and where another L-type Ca 2ϩ channel blocker verapamil had no effect on Listeria monocytogenesinduced phagocytosis in J774A.1 murine macrophages (81,82). The exact mechanism of intracellular Ca 2ϩ store depletioninduced Ca 2ϩ influx and the nature of the Ca 2ϩ channel present on the cellular membrane remain yet to be discovered. Other Ca 2ϩ channels such as Orai1 (83) and the ER Ca 2ϩ -sensing protein STIM1 (61) have been implied to be involved in various immune cells. Whether or not CD38 works in tandem with these proteins and channels should warrant further investigation.

CD38 in Fc␥R-mediated Phagocytosis
The main focus of this study was to investigate the role of CD38 in Fc␥R-mediated phagocytosis in the murine macrophage-like cell line J774A.1. We demonstrated that Fc␥R activation during IgG-opsonized latex bead ingestion elicits Ca 2ϩ signaling through the production of cADPR by the internalization of the transmembrane ADP ribosyl cyclase CD38. Although the involvement of Ca 2ϩ in phagocytosis is still up for debate, it should be indicated that there are countless of factors that affect macrophage phagocytosis as well as other Ca 2ϩ signaling pathways that are involved in the process that are not strictly Fc␥R-mediated (84 -88). Therefore, amid all the various works presently on Ca 2ϩ -mediated phagocytosis, this study provides new perspectives in immune defense by introducing the involvement of the ADPR cyclase CD38 and can help shed light on developing novel methods or drugs that can help contain bacterial infections.