2504 Research Article

IntroductionMacrophages ingest microbes and deliver them to an acidic digestivecompartment, the phagosome, where the internalized particles aredigested. The phagocytic digestion has an essential physiologicalrole in morphogenesis and homeostasis by clearing degeneratedcellular and non-cellular components in tissues. After formation onthe cell surface, nascent phagosomes lose plasma membranemoieties by recycling them back to the plasma membrane, andprogressively acquire digestive characteristics. This maturation ofphagosomes involves regulated interaction with the other membraneorganelles, including recycling endosomes, late endosomes andlysosomes (Vieira et al., 2002).

The phagosomes establish an acidic milieu inside during thematuration. This phagosomal acidification is thought to be aprerequisite for the killing function: first, most lysosomal digestiveenzymes prefer the acidic pH for their optimal activity, and second,production of germicidal hydrogen peroxide and other reactiveoxygen species requires large amounts of protons (Vieira et al.,2002). Third, acidification is also required for the maturation ofphagosomes themselves. Although the underlying mechanistic linkbetween the membrane fusion and acidification is not fullyunderstood, many observations suggest that the proper acidificationinside the endocytic organelles is required for the correct progressionof the endocytic pathway (Hurtado-Lorenzo et al., 2006), and thisalso seems to be true in the maturation process of phagosomes.

The phagosomal acidification is established by the vacuolar-typeproton transporting ATPase (V-ATPase), which is distributed tovarious exocytic and endocytic organelles. Although the presenceof V-ATPase in the phagosomal membranes is well established(Lukacs et al., 1990), the mechanism by which this proton pumparrives at phagosomes remains obscure. Phagosomal acidificationis considered to be a result of fusion between phagosomes andlysosomes where the V-ATPase and proteinases are enriched.

However, V-ATPase-mediated phagosomal acidification begins ata very early stage of phagosome formation, well before thephagosomes acquire markers of lysosomes (Lukacs et al., 1992;Yates et al., 2005). V-ATPase and the other lysosomal proteinases(cathepsins) are recruited to phagosomes with different kinetics(Tsang et al., 2000; Yates et al., 2007). Mycobacterium-containingphagosomes acquire the lysosomal membrane protein lamp1, butV-ATPase is absent on their membrane (Deretic et al., 2006;Drecktrah et al., 2007; Sturgill-Koszycki et al., 1994). Theseobservations, collectively, suggest that V-ATPase arrives at thephagosomes via a distinctive pathway from those used by the lampproteins and cathepsin proteases. Phagosomal V-ATPase isconsidered to be derived from lysosomes; however, this has yet tobe formally proven. Because V-ATPases are distributed in variousmembranes including the Golgi complex and secretory granules(Sun-Wada et al., 2006; Toyomura et al., 2000), it is probable thatthese membranes provide the acidification machinery to maturingphagosomes, as suggested from the effects of brefeldin A on V-ATPase trafficking (Fratti et al., 2003). In highly specialized cellssuch as osteoclasts or renal epithelial cells, V-ATPase resides onthe plasma membranes and participates in acidification of cellexterior (Oka et al., 2001; Toyomura et al., 2003). Another possiblescenario thus emerges that if macrophages possess cell surface V-ATPase, this could be internalized into nascent phagosomes andcontribute to the early phase acidification.

In this study, we showed that the V-ATPase a3 subunit, whichis accumulated in late endocytic compartments, is localized to thelimiting membranes of phagosomes. The acidification inphagosomes and bacteria-killing function were impaired inmacrophages lacking the a3 subunit. These findings support theview that the phagosomes acquire the proton pump from lateendocytic organelles, such as lysosomes. We modified the a3 subunitlocus to express a fluorescent a3 subunit (a3-GFP), and visualized

The nascent phagosome progressively establishes an acidicmilieu by acquiring a proton pump, the vacuolar-type ATPase(V-ATPase). However, the origin of phagosomal V-ATPaseremains poorly understood. We found that phagosomes wereenriched with the V-ATPase a3 subunit, which also accumulatedin late endosomes and lysosomes. We modified the mouse Tcirg1locus encoding subunit a3, to express an a3-GFP fusion protein.Live-cell imaging and immunofluorescence microscopy revealedthat nascent phagosomes received the a3-GFP from tubularstructures extending from lysosomes located in the perinuclearregion. Macrophages from a3-deficient mice exhibited impaired

acidification of phagosomes and delayed digestion of bacteria.These results show that lysosomal V-ATPase is recruited directlyto the phagosomes via tubular lysosomes to establish the acidicenvironment hostile to pathogens.

Supplementary material available online athttp://jcs.biologists.org/cgi/content/full/122/14/2504/DC1

Key words: Acidification, Macrophage, Microtubules, Tubularlysosome, V-ATPase

Summary

Direct recruitment of H+-ATPase from lysosomes forphagosomal acidificationGe-Hong Sun-Wada1, Hiroyuki Tabata1, Nobuyuki Kawamura1, Minako Aoyama2 and Yoh Wada2,*1Department of Biochemistry, Faculty of Pharmaceutical Sciences, Doshisha Women’s College, Kohdo, Kyotanabe, Kyoto 610-0395, Japan2Division of Biological Sciences, Institute of Scientific and Industrial Research, Osaka University, Mihogaoka 8-1, Ibaraki, Osaka 567-0047, Japan*Author for correspondence (e-mail: yohwada@sanken.osaka-u.ac.jp)

Accepted 17 April 2009Journal of Cell Science 122, 2504-2513 Published by The Company of Biologists 2009doi:10.1242/jcs.050443

Jour

nal o

f Cel

l Sci

ence

2505V-ATPase during phagosome maturation

how phagosomes acquire the V-ATPase in live macrophages. Directobservation of the fluorescently tagged V-ATPase showed thatlysosomal V-ATPase was transferred to maturing phagosomes viatubular structures. These data provide compelling evidence thatphagosome acidification is carried out by the V-ATPase of lysosomalorigin, and phagosome-lysosome interaction is mediated by thetubular extensions from lysosomes.

ResultsV-ATPase a3 subunit is localized to phagosomesV-ATPase a-subunit isoforms are components of the membranesector and localize to specific subcellular compartments: a1 and a2accumulate in the Golgi complex and early endosomes (Hurtado-Lorenzo et al., 2006; Toyomura et al., 2000), whereas the a3 subunitis the major component for the proton pumps associated with lateendosomes and lysosomes. The a3 subunit is also found on thevarious lysosome-related organelles, including pancreatic insulingranules, hormone-containing granules and melanosomes (Sun-Wada et al., 2007; Sun-Wada et al., 2006; Tabata et al., 2008).

V-ATPases are distributed in various organelles consisting of theexocytic and endocytic pathways. We examined subcellularlocalization of the subunits in mouse peritoneal macrophages byimmunofluorescence microscopy using the antibodies specific forthe a1-a3 subunits. In peritoneal macrophages, the a3 subunit washighly accumulated in organelles localized to perinuclear region(Fig. 1A). The a1 subunit, which was first identified as a componentof clathrin-coated vesicles (Perin et al., 1991), appeared as numerousdots concentrated at the perinuclear region and in macrophages (Fig.1G). The a2 subunit was highly accumulated in the perinuclearregion (Fig. 1I), suggesting that the a2 subunit is a component ofV-ATPase of the Golgi complex, as shown previously in the murinemacrophage-like cell line RAW264.7 (Toyomura et al., 2003).

Next, we observed the subcellular localization of the subunits inmacrophages internalizing bacteria. The macrophages werecocultured with Escherichia coli expressing red fluorescent protein,then fixed and stained with the subunit-specific antibodies.Phagosomes containing bacteria were decorated with the a3 subunit(Fig. 1B-F), whereas they were essentially free from the a1 and a2subunits (Fig. 1H,J), showing that the phagosomes possessed theV-ATPase with the a3 subunit as the proton pump. The distributionsof a1 and a2 subunits remained unchanged before and after bacterialingestion (Fig. 1H,J), suggesting that either the a1 or a2 subunit isnot recruited to the phagosomal membranes.

Creation of mice expressing a3-GFP fusion proteinThe a3 subunit changed its appearance upon internalization ofbacteria, as shown in Fig. 1B-E. We reported previously that thea3 subunit exhibits dynamic changes in its cellular localizationduring osteoclast differentiation (Toyomura et al., 2003; Toyomuraet al., 2000). We attempted to examine these dynamic processes bylive cell imaging using a fluorescent a3 subunit, as successfullyused before in Dictyostelium (Clarke et al., 2002). However, severaltrials to express the a3 subunit from cDNA-based constructs invarious cell lines were unsuccessful, probably because of the toxicityof deregulated episomal expression of the membrane intrinsicsubunit of the pump complex. Therefore, we modified the mouseTcirg1 locus, encoding the a3 subunit, to produce an a3-GFP fusionprotein via embryonic stem (ES)-cell-mediated gene targeting (Fig.2A). We obtained mice homozygous for Tcirg1floxGFP with theexpected ratio (33% Tcirg1WT/WT, 47% Tcirg1WT/floxGFP and 20%Tcirg1floxGFP/floxGFP, n=49) from crosses of Tcirg1WT/floxGFP mice.

These Tcirg1floxGFP/floxGFP mice (referred to as a3-GFP micehereafter) were indistinguishable from wild-type littermates inappearance and they were fertile.

We performed immunoblot analysis of lysates prepared frommouse livers. As expected, the ~110 kDa a3 protein was onlydetected in Tcirg1WT/WT and Tcirg1WT/floxGFP mice, whereas the~140 kDa product, which was also detected with an anti-GFPantibody, was apparent in Tcirg1floxGFP/floxGFP and Tcirg1WT/floxGFP

mice (Fig. 2B). In macrophages prepared from the a3-GFP mice,GFP and immunofluorescence signals for a3 colocalized well (Fig.3). The GFP signal partially overlapped those of syntaxin 7 (Stx7)(Fig. 3U-X) and Lamp2 (Fig. 3I-L), markers for late endosomesand lysosomes, respectively, whereas the early endosomal markers,sorting nexin 1 (Snx1) and Rab5 (Fig. 3M-T) were not colocalizedwith the GFP signals. In addition, the staining patterns of the variousorganelle markers in cells expressing the a3-GFP were similar tothose observed in wild-type cells (Fig. 3). These observations,collectively, indicated that the a3-GFP was associated with the lateendocytic compartments in macrophages, and the overall cellularand endosome/lysosome morphology was not affected by theexpression of the a3-GFP from the Tcirg1 loci.

The V-ATPase complex undergoes assembly and disassembly ofthe catalytic sector (V1) and the membrane embedded sector (Vo)responding to physiological stimuli and cellular conditions

Fig. 1. Subcellular localization of V-ATPase subunits in macrophages.Peritoneal macrophages cultured without (A,G,I) or with (B-F,H,J) E. coliexpressing DsRed monomer. After washing out the bacteria and chasing forthe indicated periods, macrophages were fixed with paraformaldehyde andreacted with specific antibodies against the V-ATPase a3 (A-F), a1 (G,H) anda2 (I,J) subunits and FITC-conjugated secondary antibodies. Highermagnification view of phagosomes in the boxed area in E is shown in F.Images were acquired on a laser confocal microscope LSM510 (Zeiss). Scalebars: 10 μm (A-E, G-J) and 2 μm (F).Jo

urna

l of C

ell S

cien

ce

2506

(Beyenbach and Wieczorek, 2006; Breton and Brown, 2007; Kaneand Parra, 2000). We asked whether the a3-GFP was assembledinto V-ATPase holoenzyme by immunofluorescence microscopy(Fig. 3Y-β). The a3-GFP signals overlapped with signals for the B2subunit, a component of the V1 sector, suggesting that most a3-GFP existed as a component of the V1-Vo holoenzyme.

Live-cell imaging of the a3 subunit in macrophagesWe then examined V-ATPase dynamics during phagosomeformation and maturation. In live cells, a3-GFP was associated withtubular structures extending from the perinuclear region toward thecell periphery, in addition to the perinuclear region (Fig. 4A,B).The peripheral terminals of the extensions frequently movedforward and backward as if moving on a rail (supplementarymaterial Movie 1), and treatment with nocodazole, a microtubule-depolymerizing reagent, disrupted the tubular extensions (Fig. 4C-H). The tubular extensions positive for a3-GFP were labeled witha fluid-phase marker, TMR-dextran, after a prolonged chase period(Fig. 4I-L). By contrast, Texas-red-conjugated transferrin, a markerfor early and/or recycling endosomes was essentially absent in thetubular structures positive for a3-GFP (Fig. 4M-P). Theseobservations suggested that the tubular extensions are located distalto the endocytic pathway rather than in the intermediate endosomalcompartments.

When macrophages expressing a3-GFP were exposed to IgG-opsonized yeast cells, the a3-GFP-positive tubular extensionsmoving in the cytoplasm seemed to be trapped by the phagocytosed

Journal of Cell Science 122 (14)

yeast cells (Fig. 5; supplementary material Movie 1), and the GFP-positive tubules embraced the phagosomes within 1 minute (Fig.5I-H). These images showed that the phagosomes acquired V-ATPase directly from tubular lysosomes during the maturationprocess. Furthermore, this deposition of V-ATPase subunit occurredat an early stage of phagosomal maturation, around 5-10 minutesafter incubation with phagocytic substrates.

Recruitment of the a3 subunit in the early phase ofphagosome maturationWe then examined the sequence of phagosomal maturation (Fig.6). In the early stages of phagosome maturation (t<10 minutes),latex beads positive for F-actin (but negative for human IgG signal,see legend to Fig. 6) were negative for the GFP signal, indicatingthat the V-ATPase did not associate with the earliest stage ofphagosomes. At 5 minutes, most latex beads were positive for Snx1(an early endosome marker) and Rab7 (a late endosome marker),and approximately half of them were already positive for the a3-GFP signals, implying that the a3 subunit is recruited to the nascentphagosomes soon after they have lost F-actin (Fig. 6). Thecolocalization of Rab7 and a3-GFP on the latex beads becameobvious at 10 minutes (Fig. 6E,F), but that of Snx1 and the a3-GFPsignal was decreased at 10 minutes and was completely absent at30 minutes (Fig. 6C,D), reflecting the fact that the internalized beadsgradually lost early endocytic characteristics. In parallel, the a3-GFP signal seemed to become stronger (Fig. 6G). Lamp2 (thelysosome protein) seemed to be recruited to the internalized beads

Fig. 2. Modification of Tcirg1 locus encoding the V-ATPase a3 subunit. (A) Tcirg1 alleles created in thisstudy. A Tcirg1targeting allele, which has an FRT-loxP-Neo-FRT-loxP cassette in intron 13-14 and a modifiedlast exon from which the a3-GFP fusion protein isexpressed, was constructed and introduced into mousegenome by ES-cell-mediated gene targeting. TheTcirg1targeting mouse was crossed with FLPe-expressingmice to generate Tcirg1WT/floxGFP mice. TheTcirg1WT/floxGFP mice were further crossed with Cre-recombinase-expressing mice to generate Tcirg1WT/Δ

animals. (B,C) Expression of the a3 subunit in micewith modified Tcirg1 alleles. Liver lysates wereobtained from mice with various genotypes andsubjected to immunoblot analyses using antibodiesagainst the V-ATPase a3 subunit, GFP or actin.

Jour

nal o

f Cel

l Sci

ence

2507V-ATPase during phagosome maturation

after they had accumulated the V-ATPase, because at an early stageof maturation (5 minutes), no a3-positive beads exhibited detectablelevels of the Lamp2 signal (Fig. 6H).

Phagosomal acidification and functionThe dynamic subcellular localization of the a3 subunit implies thatthe proton pump with this subunit is responsible for the phagosomalacidification, a hallmark of macrophage function. We then examinedthe phenotypic consequence of genetic inactivation of the a3subunit. We introduced loxP sites into intron 13-14 and the non-coding region in exon 20 (the last exon) of Tcirg1, so that themodified Tcirg1 allele, Tcirg1floxGFP could be inactivated throughCre recombinase-mediated recombination (Fig. 2A, Tcirg1Δ).Heterozygous crossing of Tcirg1WT/Δ yielded Tcirg1Δ/Δ pups (25%

Tcirg1WT/WT, 56% Tcirg1WT/Δ and 19% Tcirg1Δ/Δ, n=154), and thusTcirg1Δ/Δ mice (referred to as a3 mutants, hereafter) develop to termand are delivered. The absence of the a3 subunit in the a3 mutantswas confirmed by immunoblot analysis (Fig. 2C). The a3 mutantswere indistinguishable at birth, but they were smaller than normallittermates by 2 weeks of age. They did not develop teeth, and couldnot survive for more than 1 month, as reported for the targeted orinduced mutation of Tcirg1 (Li et al., 1999; Scimeca et al., 2000).

We analyzed phagosomal acidification by ratiometricmeasurement of fluorescence emitted by internalized FITC-labeledbeads. At the earliest point of measurement (4-10 minutes), therewas a slight difference between the wild-type (pH 7.5±0.8) andmutant (pH 7.7±0.3) phagosomes. After 1 hour (recorded at 60-70minutes after feeding the beads), the wild-type phagosomes wereacidified (pH 5.9±1.0), whereas the mutant pH remained near neutral(pH 7.0±0.4). The prominent decrease in phagosomal pH in thewild-type macrophages as well as the limited acidification in themutant cells were depends on the V-ATPase function, because bothacidifications were completely inhibited by the addition of

Fig. 3. Distribution of organelle marker proteins and the a3-GFP subunit inmacrophages. Peritoneal macrophages obtained from Tcirg1floxGFP (A-C, E-G,I-K, M-O, Q-S, U-W and X-α) and wild-type (D, H, L, P, T, X and β) micewere fixed with paraformaldehyde, and then probed with antibodies against a3(A-C), Rab7 (E-G), Lamp2 (I-K), Snx1 (M-P), Rab5 (Q-T) and Stx7 (U-W),B2 subunit of V-ATPase (X-β) and Cy3-conjugated secondary antibodies. Theimages were acquired under illumination for GFP and Cy3 on a laser confocalmicroscope (LSM510). Scale bar: 10 μm.

Fig. 4. a3-GFP is associated with tubular compartments. (A-H) Macrophageshomozygous for Tcirg1floxGFP were cultured on glass-bottomed dishes placedon a microscope stage in a climate chamber (5% CO2, 37°C), and viewed withNomarski optics (A). The GFP signal was recorded (B), then the medium wasreplaced with medium containing 10 nM nocodazole (C-G). The addition ofnocodazole was made at time 0. After approximately 90 minutes of treatmentwith nocodazole, the cells were washed extensively with drug-free mediumand further incubated for 1 hour (H). Arrowheads indicate the tubularstructures. (I-L) a3-GFP-positive tubular extension accumulates endocyticmarkers. Macrophages expressing a3-GFP were labeled with 1 mg/ml TMR-dextran (molecular weight 7000) for 1 hour, then chased for 6 hours indextran-free medium for endocytic labeling of lysosomal compartments (I-L),or they were incubated with 10 mg/ml Texas-red-labeled transferrin for 10minutes, and chased for 12 minutes in DMEM plus FBS for labeling of earlyand recycling endosomes (M-N). Cells were viewed on a microscope withillumination for GFP (I,M) and Texas red (J,N) epifluorescence. Scale bars:10 μm.

Jour

nal o

f Cel

l Sci

ence

2508

bafilomycin A1, a potent and specific inhibitor of V-ATPase (Fig.7A). These results indicated that genetic loss of the a3 subunit resultsin severe impairment in the phagosome acidification, and the V-ATPases with the other a subunits were able to rescue this defectonly partially, if at all.

The wild-type macrophages efficiently inactivated the internalizedbacteria within 24 hours; however, the mutant macrophages exhibitedimpaired killing activity (Fig. 7B). In the wild-type macrophages,the phagocytosed bacteria became round and eventually becamefragmented 3 hours after infection, and most signals had disappearedby 24 hours (Fig. 7C,D). By contrast, in the a3 mutant macrophages,a significant portion of the bacteria exhibited a rod-like shape evenafter a 1-day chase period, indicating that their phagosomes are lessable to decompose the internalized materials (Fig. 7D). This defectin bacterial killing was not due to impaired phagocytosis of bacteria,because both wild-type and mutant macrophages engulfed E. coliwith similar efficiencies (Fig. 7B,D).

Journal of Cell Science 122 (14)

The mutant macrophages accumulated mature forms of thelysosomal proteinases cathepsin-B, -D and -L (Dragonetti et al.,2000; Kominami et al., 1988; Yeyeodu et al., 2000), implying thatthe steady-state levels of active lysosomal enzymes were notseverely changed by the loss of the a3 subunit (Fig. 8). We examinedthe time-course of Snx1, Rab7 and Lamp2 to phagosomes in thea3 mutant macrophages, and found that the mutant phagosomeslost Snx1 within 30 minutes, and became positive for Rab7transiently, then acquired Lamp2 after 60 minutes (Fig. 9). Thisobservation showed that the nascent phagosomes could acquire thelysosomal characteristics, even in the absence of the a3 subunit andprominent acidification.

DiscussionWe found that the a3 subunit, a component of late endosomal andlysosomal V-ATPase complex, localizes to phagosomes. We alsoshowed that loss of the a3 subunit causes impaired phagosomalacidification and bacteriocidal function. These findings indicate that

Fig. 5. Time-lapse live imaging ofa3-GFP recruited onto phagosomalmembranes. a3-GFP macrophageswere incubated with aformaldehyde-fixed vam5 mutantof Saccharomyces cerevisiae(Wada et al., 1990) on a wide-fieldfluorescence microscope (LeicaASMDW), and viewed with 490nm illumination (B-S) or Nomarskioptics (A,T). A phagocytosed yeastcell (arrows) was free from theGFP signal until 6 minutes (B-H),when a GFP-positive tubule(arrowhead) interacted with theyeast cell (6 minutes 10 seconds;I), and GFP encircled the yeast cellcompletely within a few minutes(M-O). Insets show highermagnified, contrast-enhancedimages. The time sequence of thisevent is shown in a movie format(supplementary material Movie 1).

Table 1. Oligonucleotides used in this studyPrimer Sequence

a3GFP-S1 5 -CAGGCACCGGCTACAAGCTCAGCCCCTTCACCTTCACTGTGGACAGTGACGCCGCTCTAGAACTAGTGGA-3a3GFP-A1 5 -GCCAGCTCTTTATTCCTGTCCCACTTCGGAGAAGCAGGGATTGCAGTGGGTTTCTGTCTCTTATACACAT-3a3GFPDT-L 5 -AGGAGTAGGCTCCTCGGAAATGGTACCCACCCTTCAATCCCAGAGCAAAGGAAGCTTCCGCGGTCCGTTTAAACAGCTC-3a3GFPDT-R 5 -GGAGTTTGAAGAAAGGCCATGTAAGACTGTGCTGGCGTGCAGCATTCTTGAAGCTTCGGCCGGCCCGGGCGATCCCCCGA-3GFP-S5 5 -CCGTGCTGCTGCCCGACAACCACTACCTGA-3HR-Rv1 5 -AGCTGCAGCTCTTCCTGTTGCTCTGACCAACC-3

For oligonucleotides a3GFP-S1 and a3GFP-A1, sequences corresponding to the a3 region and EGFP-loxP-Zeo-loxP cassette are shown in italics and romantext, respectively. For a3GFPDT-L and a3GFPDT-R, sequences corresponding to the a3 region and pBSDT-AII plasmid are indicated in italics and roman text,respectively.

Jour

nal o

f Cel

l Sci

ence

2509V-ATPase during phagosome maturation

this particular proton pump is responsible for phagosomalacidification and function. Although targeted mutations, as well asnatural mutations in mouse Tcirg1 have been characterizedpreviously (Li et al., 1999; Scimeca et al., 2000; Utku et al., 2004),the defects in macrophage function have not been described.Tcirg1, the current official nomenclature for the structural gene ofV-ATPase a3 subunit, stands for ‘T-cell, immune regulator 1’,reflecting the fact that this molecule was identified during studieson a T-cell activation signaling (Utku et al., 1998). Mutations inTCIRG1 are responsible for human infantile malignant osteopetrosis

(OMIM: *604592), which is characterized by impaired boneresorption because of osteoclast dysfunction. The patients suffermultiple infections and often develop life-threatening sepsis andosteomyelitis (Del Fattore et al., 2007; Wilson and Vellodi, 2000).Although the etiology is considered to be defective bone marrowdevelopment (Blin-Wakkach et al., 2004), it is also plausible thatdefective macrophage functions cause severe infections.

V-ATPase a3 subunit is transferred to nascent phagosomespositive for the early endosome marker Snx1, whereas we observedthat F-actin and a3 are rarely colocalized in the same phagosomes.

Fig. 6. Recruitment of the V-ATPase a3 subunitduring phagosome maturation. a3-GFPmacrophages were fed with IgG-opsonizedbeads and chased for the indicated period, andstained with the organelle markers. Typicalimages for each marker staining (5 minutes and60 minutes) are shown in A,C,E and G. Externalbeads were identified by staining with Cy5-labeled anti-human IgG antibodies. Scale bar:10 μm. (B,D,F,H) Because the human-IgG-positive beads did not show the signal of a3-GFP (none out of over 1000 beads), weexamined the association of marker moleculesin human-IgG-negative completely internalizedbeads. At least 100 beads were analyzed perexperimental point. The total beads analyzed ateach time point was set at 100%, and thepercentage of beads with markers is shown.

Jour

nal o

f Cel

l Sci

ence

2510

Therefore, soon after losing the filamentous actin, the prematurephagosomes acquire the proton pump from the late endocyticcompartments (Fig. 7E). This time-course agrees with theacidification kinetics reported previously (Lukacs et al., 1990;Steinberg et al., 2007; Yates et al., 2007). The early recruitment ofa3-GFP to phagosomes occurs via tubular structures extending fromthe perinuclear lysosomes along microtubules. Nascent phagosomesbecome acidic by acquiring the V-ATPase through a microtubule-dependent mechanism (Desjardins et al., 1994; Harrison et al.,2003). Given the nascent or early phagosomes are localized moreperipherally, whereas the late endosomes/lysosomes are centrallynear the microtubule-organizing centre (MTOC), the V-ATPase witha3 subunit is directed in a centrifugal direction in macrophages (Fig.

Journal of Cell Science 122 (14)

7D). In addition to this centrifugal movement, imaging of Rab7and its effecter (RILP) revealed a membrane flow in the opposite,centripetal direction: these markers appear on tubules extendingfrom maturing phagosomes toward the MTOC, where lysosomesare accumulated (Harrison et al., 2003). Taken together with thefact that phagosome moves bidirectionally along microtubules(Blocker et al., 1997), both centripetal and centrifugal membranedynamics seem to be involved in the phagosomal maturation. Itshould be noted that the experimental system might enhance onedirection preferentially. Because RILP regulates the microtubule-mediated movement of Rab7-positive compartments, its ectopicexpression could shift the balance of membrane movement towardcentripetal direction, because overexpression of RILP results in

Fig. 7. Defective acidification and impairedbacteriocidal activity of phagosomes lacking thea3 subunit. (A) Acidification of phagosomes.Macrophages were obtained from a3 mutantsand their wild-type littermates at 2 weeks afterbirth. The wild-type and mutant macrophageswere fed with FITC-labeled, IgG opsonizedlatex beads (t=0) and the phagosomal pH wasdetermined by fluorescence ratiometry underillumination with pH-insensitive (440 nm) andpH-sensitive (490 nm) excitation. In severalmeasurements, bafilomycin A1 (BafA1, 1 μM)was added at 35 minutes, and pH wasdetermined at 55-65 minutes. (B) Bacteriocidalactivity of macrophages. Peritonealmacrophages were exposed to E. coli for 1 hourand then cultured in the presence of gentamicin.Viable intracellular bacteria were quantified bygentle lysis of the macrophages at the indicatedtime points and subsequent plating on LB agarplates. (C,D) Digestion of bacteria inmacrophages. After exposure to E. coliexpressing DsRed monomer for 30 minutes, themacrophages were fixed at the indicated timesand images were recorded with Zeiss LSM510.The cells containing undigested bacteria werecounted (n>500 at each time point) and areplotted in C. Typical images at the indicatedtimes are shown in D. Experiments wereperformed at least three times. (E) A tubularlysosome provides V-ATPase to a nascentphagosome. The phagosome acquires thelysosomal proton pump from tubular lysosomesextending along microtubules. This transferoccurs soon after F-actin filaments have becomedetached from the nascent phagosomes orphagocytic cups, and thus the acidification bythe V-ATPase with the a3 subunit begins at anearly stage of phagosome maturation. Results inA and C are means ± s.d.

Jour

nal o

f Cel

l Sci

ence

2511V-ATPase during phagosome maturation

clustering of the lysosomes near the MTOC (Cantalupo et al., 2001).The centrifugal membrane dynamics might not be restricted tomacrophages. Dendritic cells, another class of bone-marrow-derivedcell with an antigen-presenting function, also exhibit tubularlysosomes upon stimulation by LPS (Vyas et al., 2007). Osteoclasts,bone-resorbing bone-marrow-derived cells, acquire acid-secretingactivity by relocating the V-ATPase to specialized areas of theplasma membrane called bone-resorbing lacuna, through a process

involving centrifugal translocation of the lysosome compartmentstoward the cell surface (Toyomura et al., 2003).

The lysosomal membrane protein Lamp2 and V-ATPase reachthe maturing phagosomes at different times. Biochemical studiesshow that V-ATPase is fractionated in detergent-resistant membrane(DRM) subdomains, whereas Lamp proteins are associated withnon-DRM subdomains (Dermine et al., 2001; Lafourcade et al.,2008; Sprenger et al., 2006). During maturation, lipid microdomainsare formed on phagosomes, where the membrane fusion machineryis enriched (Dermine et al., 2001). We considered whether the kineticdifferences in the arrival times of Lamp2 and V-ATPase involveslipid microdomain segregation. However, we could not detectcolocalization of a3-GFP with any DRM markers, including choleratoxin B-subunit and flotillin-1, in live or fixed macrophages,respectively (data not shown). Furthermore, the a3-GFP expressionpattern was not changed by addition of filipin, nystatin or bytreatment with U18666A (data not shown).

The phagosome-lysosome interaction occurs through a complexprocess in which multiple phases of membrane interactions areinvolved. This is an important aspect of phagosome dynamics,because various pathogens overcome the innate immune defenseby modifying these processes (Haas, 2007). Indeed, mycobacterianeutralize the bacteriocidal activity by interfering with V-ATPaserecruitment to bacterium-containing phagosomes (Sturgill-Koszyckiet al., 1994). In this regard, the imaging of V-ATPase dynamics inpathogen-infected macrophages will provide an insight into howdistinctive pathogens affect the different stages of phagosomematuration.

Materials and MethodsAntibodiesAntibodies against V-ATPase a subunits, B2 subunit, Snx1, Stx7 and Rab7 weredescribed previously (Nakamura et al., 2001; Nakamura et al., 2000; Sun-Wada etal., 2003; Toyomura et al., 2000). Rat anti-Lamp2 monoclonal antibodies wereprovided by Developmental Studies Hybridoma Bank. Anti-cathepsin-D (rabbitpolyclonal) was from Wako Pure Chemicals, and rat monoclonal antibodies againstcathepsin L (MAB9521) and B (MAB965) were from R&D Systems. Anti-GFP andanti-actin antibodies were purchased from Clontech and Abcam. Fluorescent dye-or enzyme-linked secondary antibodies were obtained from JacksonImmunoResearch. Tetramethylrhodamine-conjugated dextran (TMR-dextran, MW.7000, lysine-fixable) and Texas-red-conjugated human transferrin were purchasedfrom Invitrogen.

Creation of a modified Tcirg1 alleleA loxP-EM7-ZeoR-loxP cassette (Aoyama et al., 2005) was placed in a modifiedversion of pEGFP-C1 plasmid (Clontech). Using this construct as a template, PCRwas performed with primers a3GFP-S1 and a3GFP-A1 [exhibiting homology to a3(shown in italics in Table 1) and the EGFP-loxP-Zeo-loxP cassette (in roman scriptin Table 1)]. A bacterial artificial chromosome (BAC) containing the Tcirg1 locus(clone 206L24, Incyte) was modified with the PCR product in E. coli DY380 (Leeet al., 2001). This strategy created a gene encoding an a3-GFP fusion protein, inwhich the C-terminal of a3 was joined to the N-terminal of EGFP by a linker sequence(AALELVDPPGSIAT). The modified a3 sequence (from a part of intron 9-10 to 1.4-kb downstream of the 3� untranslated region) was retrieved in pBSDTA-II plasmid(Aoyama et al., 2005) in the DY380 strain by transforming a DNA fragment amplifiedwith a3GFPDT-L and a3GFPDT-R primers (Table 1). Then loxP-Zeo-loxP wasconverted a single loxP site in a Cre-expressing bacterium. A cassette encoding theneo gene with FRT and loxP sequences was inserted by means of an in vitro reaction(Aoyama et al., 2005) to generate a collection of targeting vectors. We selected aclone having the FRT-loxP-neo cassette in intron 13-14 (Fig. 2A).

The R1 embryonic stem cell line (Nagy et al., 1993) was transfected with the gene-targeting construct. Homologous recombinants were identified by PCR screening usinga pair of oligonucleotides, GFP-S5 and HR-Rv1 (Table 1). Chimeric animals weregenerated by injecting the homologous recombinants into blastocoels of C57Bl/6embryos (Hogan et al., 1994). A Tcirg1targeting mouse was crossed with B6;SJL-Tg(ACTFLPe)9205Dym/J (Rodriguez et al., 2000) to remove the Neo segment, andmice of Tcirg1WT/floxGFP genotype were obtained. The Tcirg1WT/floxGFP mice werecrossed to generate mice homozygous for Tcirg1floxGFP, which turned out to be healthyand fertile. Tcirg1WT/floxGFP mice were further crossed with the FvB/N-Tg(EIIa-

Fig. 8. Maturation of lysosomal cathepsins in a3 mutant macrophages. Celllysates were prepared from wild-type and mutant macrophages and thensubjected to immunoblot analysis using anti-cathepsin D, L, or B. pro,precursors of cathepsin; i, 47 kDa immature intermediate of cathepsin D; msc,43 kDa mature single chain of cathepsin D; lc and sc, mature large and smallchains, respectively. Single chain and heavy chain are the mature forms ofcathepsin L. m, mature form of cathepsin B.

Fig. 9. Phagosome maturation in a3 mutant macrophages. Macrophagesobtained from a3 mutant animals were fed with latex beads and then processedfor immunofluorescence analyses as described in Fig. 3. The mutantmacrophages lost F-actin and Snx1, and were transiently labeled with Rab7,and in parallel, they accumulated Lamp2 with kinetics essentially the same asseen in wild-type macrophages. Each panel represents the distribution in 200-400 beads. The total beads analyzed at each time point was set at 100%, andthe percentage of beads with markers is shown.

Jour

nal o

f Cel

l Sci

ence

2512 Journal of Cell Science 122 (14)

cre)C5379Lmgd/J strain expressing Cre recombinase at preimplantationdevelopmental stages (Lakso et al., 1996) to remove exons 15-20 and to generateTcirg1WT/Δ animals. Tcirg1WT/Δ were further crossed with wild-type C57Bl/6 animalsto remove the Cre transgene, and a cohort of Tcirg1WT/Δ animals were established.All animals were handled according to the guidelines of institution and governmentalpolicies.

MicroscopyLive cells were observed on ASMDW microscope equipped with a 63� glycerol-immersed objective of 1.3 NA (Leica) and a Roper CoolSNAP HQ camera. Fixedspecimens were observed under a laser confocal microscope LSM510 (Zeiss)equipped with oil-immersed objective of 1.4 NA. The images were captured withLeica software and processed for presentation with Adobe Photoshop and NIH ImageJsoftware.

Macrophage culture and microscopyMacrophages were obtained from mice intraperitoneally injected with 1 mlthioglycollate medium 4 days before harvest, and cultured on glass bottom dishes inDMEM plus 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen). Themacrophages were starved for 1-4 hours in DMEM plus 0.1% FBS, and then theywere fed with phagocytic markers at t=0, spun briefly in a table-top centrifuge at250 g for 1 minute to synchronize phagocytosis. Cells were washed extensively withassay medium (DMEM plus 0.1% FBS, or Dulbecco’s PBS supplemented with glucoseand pyruvate). For live-cell imaging, phenol red was omitted from the medium. Thecells were fixed with freshly prepared 4% paraformaldehyde in Dulbecco’s PBS, andprocessed for immunocytochemistry as described elsewhere (Sun-Wada et al., 2006;Tabata et al., 2008). Immunofluorescence images were acquired either on a confocallaser-scanning microscope (Zeiss LSM510 equipped with Plan-Apochromat 63� 1.4NA oil-immersed objective) or on a wide-field fluorescence microscope (LeicaASMDW, 63� glycerol-immersed objective of 1.3 NA. The images were capturedby Zeiss or Leica software and processed for presentation by Adobe Photoshop andNIH ImageJ software.

Phagocytosis assay and pulse-chase protocola3-GFP macrophages were overlaid with the opsonized latex beads (Bangslaboratories) which were opsonized with 1 mg/ml human IgG (Sigma). Aftercentrifugation at 250 g for 1 minute (t=0) and standing for 1 minute at 37°C, unboundbeads were washed out with prewarmed medium. The cells were fixed afterincubating for the indicated periods. The samples were incubated with Cy5-conjugated donkey anti-human IgG (Jackson ImmunoResearch) to decorate externalbeads, then the cells were permeabilized and processed for immunofluorescencestaining.

Macrophage functionMacrophages were incubated with E. coli XL1-Blue in a bacteria:macrophages ratioof 25:1 for 1 hour at 37°C to allow phagocytosis to occur. After 1 hour, gentamicin(100 μg/ml) was added to the medium to kill extracellular bacteria. Then, the mediumwas replaced with fresh medium containing 10 μg/ml gentamicin. At 2, 6, and 24hours, cells were lysed with 0.5% Triton X-100 in PBS, and the lysates were platedon LB plates and colonies were counted after overnight incubation at 37°C.

The phagosomal pH was measured by single-particle fluorescence ratiometry(Steinberg et al., 2007). Macrophages fed with FITC beads opsonized with humanIgG were placed under a Leica ASMDW microscope. They were illuminated at440±10 nm and 490±10 nm excitation with a monochromator (Leica), and imageswere captured using a 527±15 nm band-pass filter (Leica L5 cube without an excitationfilter). For some measurements, V-ATPase inhibitor bafilomycin A1 (1 μM) was addedat 35 minutes, and images were recorded at 55-65 minutes. The backgroundsubtracted fluorescence ratios were determined using NIH ImageJ software, and thepH of each phagosome was calculated from a standard curve generated by ratio-imaging the latex beads incubated in buffer series of pH 4-8, 25 mM Mes-Tris or 50mM potassium acetate containing 180 mM KCl, 10 μg/ml nigericin and 1 μg/mlvalinomycin.

Statistical analysesThe results are presented as means ± s.d. Student’s t-test (two-tailed) was used toevaluate the data. Values of P<0.05 were considered significant.

We thank Andras Nagy (Mount Sinai Hospital, Toronto, Canada),Hisato Kondoh and Yujiro Higashi (Osaka University, Osaka, Japan)for the mouse ES cell line, and Neal Copeland (National CancerInstitute, Frederick, MD) for the E. coli strains and plasmids. We thankAyaka Kuze (Doshisha Women’s College, Kyoto, Japan) for genotypeanalyses and mouse handling. This work was supported in part byGrants-in-Aid from the MEXT of Japan.

ReferencesAoyama, M., Agari, K., Sun-Wada, G. H., Futai, M. and Wada, Y. (2005). Simple and

straightforward construction of a mouse gene targeting vector using in vitro transpositionreactions. Nucleic Acids Res. 33, e52.

Beyenbach, K. W. and Wieczorek, H. (2006). The V-type H+ ATPase: molecular structureand function, physiological roles and regulation. J. Exp. Biol. 209, 577-589.

Blin-Wakkach, C., Wakkach, A., Sexton, P. M., Rochet, N. and Carle, G. F. (2004).Hematological defects in the oc/oc mouse, a model of infantile malignant osteopetrosis.Leukemia 18, 1505-1511.

Blocker, A., Severin, F. F., Burkhardt, J. K., Bingham, J. B., Yu, H., Olivo, J. C.,Schroer, T. A., Hyman, A. A. and Griffiths, G. (1997). Molecular requirements forbi-directional movement of phagosomes along microtubules. J. Cell Biol. 137, 113-129.

Breton, S. and Brown, D. (2007). New insights into the regulation of V-ATPase-dependentproton secretion. Am. J. Physiol. Renal Physiol. 292, F1-10.

Cantalupo, G., Alifano, P., Roberti, V., Bruni, C. B. and Bucci, C. (2001). Rab-interactinglysosomal protein (RILP): the Rab7 effector required for transport to lysosomes. EMBOJ. 20, 683-693.

Clarke, M., Kohler, J., Arana, Q., Liu, T., Heuser, J. and Gerisch, G. (2002). Dynamicsof the vacuolar H+-ATPase in the contractile vacuole complex and the endosomal pathwayof Dictyostelium cells. J. Cell Sci. 115, 2893-2905.

Del Fattore, A., Cappariello, A. and Teti, A. (2007). Genetics, pathogenesis andcomplications of osteopetrosis. Bone 42, 19-29.

Deretic, V., Singh, S., Master, S., Harris, J., Roberts, E., Kyei, G., Davis, A., de Haro,S., Naylor, J., Lee, H. H. et al. (2006). Mycobacterium tuberculosis inhibition ofphagolysosome biogenesis and autophagy as a host defence mechanism. Cell Microbiol.8, 719-727.

Dermine, J. F., Duclos, S., Garin, J., St-Louis, F., Rea, S., Parton, R. G. and Desjardins,M. (2001). Flotillin-1-enriched lipid raft domains accumulate on maturing phagosomes.J. Biol. Chem. 276, 18507-18512.

Desjardins, M., Huber, L. A., Parton, R. G. and Griffiths, G. (1994). Biogenesis ofphagolysosomes proceeds through a sequential series of interactions with the endocyticapparatus. J. Cell Biol. 124, 677-688.

Dragonetti, A., Baldassarre, M., Castino, R., Demoz, M., Luini, A., Buccione, R. andIsidoro, C. (2000). The lysosomal protease cathepsin D is efficiently sorted to andsecreted from regulated secretory compartments in the rat basophilic/mast cell line RBL.J. Cell Sci. 113, 3289-3298.

Drecktrah, D., Knodler, L. A., Howe, D. and Steele-Mortimer, O. (2007). Salmonellatrafficking is defined by continuous dynamic interactions with the endolysosomal system.Traffic 8, 212-225.

Fratti, R. A., Chua, J., Vergne, I. and Deretic, V. (2003). Mycobacterium tuberculosisglycosylated phosphatidylinositol causes phagosome maturation arrest. Proc. Natl. Acad.Sci. USA 100, 5437-5442.

Haas, A. (2007). The phagosome: compartment with a license to kill. Traffic 8, 311-330.Harrison, R. E., Bucci, C., Vieira, O. V., Schroer, T. A. and Grinstein, S. (2003).

Phagosomes fuse with late endosomes and/or lysosomes by extension of membraneprotrusions along microtubules: role of Rab7 and RILP. Mol. Cell. Biol. 23, 6494-6506.

Hogan, B., Beddington, R., Constantini, F. and Lacy, E. (1994). Manipulating MouseEmbryo: A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring HarborLaboratory Press.

Hurtado-Lorenzo, A., Skinner, M., El Annan, J., Futai, M., Sun-Wada, G. H., Bourgoin,S., Casanova, J., Wildeman, A., Bechoua, S., Ausiello, D. A. et al. (2006). V-ATPaseinteracts with ARNO and Arf6 in early endosomes and regulates the protein degradativepathway. Nat. Cell Biol. 8, 124-136.

Kane, P. M. and Parra, K. J. (2000). Assembly and regulation of the yeast vacuolar H+-ATPase. J. Exp. Biol. 203, 81-87.

Kominami, E., Tsukahara, T., Hara, K. and Katunuma, N. (1988). Biosyntheses andprocessing of lysosomal cysteine proteinases in rat macrophages. FEBS Lett. 231, 225-228.

Lafourcade, C., Sobo, K., Kieffer-Jaquinod, S., Garin, J. and van der Goot, F. G.(2008). Regulation of the V-ATPase along the endocytic pathway occurs throughreversible subunit association and membrane localization. PLoS ONE 3, e2758.

Lakso, M., Pichel, J. G., Gorman, J. R., Sauer, B., Okamoto, Y., Lee, E., Alt, F. W.and Westphal, H. (1996). Efficient in vivo manipulation of mouse genomic sequencesat the zygote stage. Proc. Natl. Acad. Sci. USA 93, 5860-5865.

Lee, E. C., Yu, D., Martinez de Velasco, J., Tessarollo, L., Swing, D. A., Court, D. L.,Jenkins, N. A. and Copeland, N. G. (2001). A highly efficient Escherichia coli-basedchromosome engineering system adapted for recombinogenic targeting and subcloningof BAC DNA. Genomics 73, 56-65.

Li, Y. P., Chen, W., Liang, Y., Li, E. and Stashenko, P. (1999). Atp6i-deficient miceexhibit severe osteopetrosis due to loss of osteoclast-mediated extracellular acidification.Nat. Genet. 23, 447-451.

Lukacs, G. L., Chang, X. B., Kartner, N., Rotstein, O. D., Riordan, J. R. and Grinstein,S. (1992). The cystic fibrosis transmembrane regulator is present and functional inendosomes. Role as a determinant of endosomal pH. J. Biol. Chem. 267, 14568-14572.

Lukacs, G. L., Rotstein, O. D. and Grinstein, S. (1990). Phagosomal acidification ismediated by a vacuolar-type H+-ATPase in murine macrophages. J. Biol. Chem. 265,21099-21107.

Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W. and Roder, J. C. (1993).Derivation of completely cell culture-derived mice from early-passage embryonic stemcells. Proc. Natl. Acad. Sci. USA 90, 8424-8428.

Nakamura, N., Yamamoto, A., Wada, Y. and Futai, M. (2000). Syntaxin 7 mediatesendocytic trafficking to late endosomes. J. Biol. Chem. 275, 6523-6529.

Nakamura, N., Sun-Wada, G. H., Yamamoto, A., Wada, Y. and Futai, M. (2001).Association of mouse sorting nexin 1 with early endosomes. J. Biochem. 130, 765-771.

Jour

nal o

f Cel

l Sci

ence

2513V-ATPase during phagosome maturation

Oka, T., Murata, Y., Namba, M., Yoshimizu, T., Toyomura, T., Yamamoto, A., Sun-Wada, G. H., Hamasaki, N., Wada, Y. and Futai, M. (2001). a4, a unique kidney-specific isoform of mouse vacuolar H+-ATPase subunit a. J. Biol. Chem. 276, 40050-40054.

Perin, M., Fried, V., Stone, D., Xie, X. and Sudhof, T. (1991). Structure of the 116-kDapolypeptide of the clathrin-coated vesicle/synaptic vesicle proton pump. J. Biol. Chem.266, 3877-3881.

Rodriguez, C. I., Buchholz, F., Galloway, J., Sequerra, R., Kasper, J., Ayala, R., Stewart,A. F. and Dymecki, S. M. (2000). High-efficiency deleter mice show that FLPe is analternative to Cre-loxP. Nat. Genet. 25, 139-140.

Scimeca, J. C., Franchi, A., Trojani, C., Parrinello, H., Grosgeorge, J., Robert, C.,Jaillon, O., Poirier, C., Gaudray, P. and Carle, G. F. (2000). The gene encoding themouse homologue of the human osteoclast-specific 116-kDa V-ATPase subunit bears adeletion in osteosclerotic (oc/oc) mutants. Bone 26, 207-213.

Sprenger, R. R., Fontijn, R. D., van Marle, J., Pannekoek, H. and Horrevoets, A. J.(2006). Spatial segregation of transport and signalling functions between humanendothelial caveolae and lipid raft proteomes. Biochem. J. 400, 401-410.

Steinberg, B. E., Touret, N., Vargas-Caballero, M. and Grinstein, S. (2007). In situmeasurement of the electrical potential across the phagosomal membrane using FRET andits contribution to the proton-motive force. Proc. Natl. Acad. Sci. USA 104, 9523-9528.

Sturgill-Koszycki, S., Schlesinger, P. H., Chakraborty, P., Haddix, P. L., Collins, H.L., Fok, A. K., Allen, R. D., Gluck, S. L., Heuser, J. and Russell, D. G. (1994). Lackof acidification in Mycobacterium phagosomes produced by exclusion of the vesicularproton-ATPase. Science 263, 678-681.

Sun-Wada, G. H., Murata, Y., Namba, M., Yamamoto, A., Wada, Y. and Futai, M.(2003). Mouse proton pump ATPase C subunit isoforms (C2-a and C2-b) specificallyexpressed in kidney and lung. J. Biol. Chem. 278, 44843-44851.

Sun-Wada, G. H., Toyomura, T., Murata, Y., Yamamoto, A., Futai, M. and Wada, Y.(2006). The a3 isoform of V-ATPase regulates insulin secretion from pancreatic β-cells.J. Cell Sci. 119, 4531-4540.

Sun-Wada, G. H., Tabata, H., Kawamura, N., Futai, M. and Wada, Y. (2007).Differential expression of a subunit isoforms of the vacuolar-type proton pump ATPasein mouse endocrine tissues. Cell Tissue Res. 329, 239-248.

Tabata, H., Kawamura, N., Sun-Wada, G. H. and Wada, Y. (2008). Vacuolar-type H+-ATPase with the a3 isoform is the proton pump on premature melanosomes. Cell TissueRes. 332, 447-460.

Toyomura, T., Oka, T., Yamaguchi, C., Wada, Y. and Futai, M. (2000). Three subunita isoforms of mouse vacuolar H+-ATPase. Preferential expression of the a3 isoformduring osteoclast differentiation. J. Biol. Chem. 275, 8760-8765.

Toyomura, T., Murata, Y., Yamamoto, A., Oka, T., Sun-Wada, G. H., Wada, Y. andFutai, M. (2003). From lysosomes to plasma membrane: localization of vacuolar typeH+-ATPase with the a3 isoform during osteoclast differentiation. J. Biol. Chem. 278,22023-22030.

Tsang, A. W., Oestergaard, K., Myers, J. T. and Swanson, J. A. (2000). Altered membranetrafficking in activated bone marrow-derived macrophages. J. Leukoc. Biol. 68, 487-494.

Utku, N., Heinemann, T., Tullius, S. G., Bulwin, G. C., Beinke, S., Blumberg, R. S.,Beato, F., Randall, J., Kojima, R., Busconi, L. et al. (1998). Prevention of acute allograftrejection by antibody targeting of TIRC7, a novel T cell membrane protein. Immunity9, 509-518.

Utku, N., Boerner, A., Tomschegg, A., Bennai-Sanfourche, F., Bulwin, G. C.,Heinemann, T., Loehler, J., Blumberg, R. S. and Volk, H. D. (2004). TIRC7 deficiencycauses in vitro and in vivo augmentation of T and B cell activation and cytokine response.J. Immunol. 173, 2342-2352.

Vieira, O. V., Botelho, R. J. and Grinstein, S. (2002). Phagosome maturation: aginggracefully. Biochem. J. 366, 689-704.

Vyas, J. M., Kim, Y. M., Artavanis-Tsakonas, K., Love, J. C., Van der Veen, A. G. andPloegh, H. L. (2007). Tubulation of class II MHC compartments is microtubule dependentand involves multiple endolysosomal membrane proteins in primary dendritic cells. J.Immunol. 178, 7199-7210.

Wada, Y., Kitamoto, K., Kanbe, T., Tanaka, K. and Anraku, Y. (1990). The SLP1 geneof Saccharomyces cerevisiae is essential for vacuolar morphogenesis and function. Mol.Cell. Biol. 10, 2214-2223.

Wilson, C. J. and Vellodi, A. (2000). Autosomal recessive osteopetrosis: diagnosis,management, and outcome. Arch. Dis. Child 83, 449-452.

Yates, R. M., Hermetter, A. and Russell, D. G. (2005). The kinetics of phagosomematuration as a function of phagosome/lysosome fusion and acquisition of hydrolyticactivity. Traffic 6, 413-420.

Yates, R. M., Hermetter, A., Taylor, G. A. and Russell, D. G. (2007). Macrophageactivation downregulates the degradative capacity of the phagosome. Traffic 8, 241-250.

Yeyeodu, S., Ahn, K., Madden, V., Chapman, R., Song, L. and Erickson, A. H. (2000).Procathepsin L self-association as a mechanism for selective secretion. Traffic 1, 724-737.

Jour

nal o

f Cel

l Sci

ence