NF-Κb Activation Controls Phagolysosome Fusion-Mediated Killing of Mycobacteria by Macrophages

Home , ASAH2, COPE (gene), NEU2, RABEP1, STX11, STX6

NF-κB Activation Controls Phagolysosome Fusion-Mediated Killing of Mycobacteria by Macrophages

This information is current as Maximiliano Gabriel Gutierrez, Bibhuti B. Mishra, Luisa of September 27, 2021. Jordao, Edith Elliott, Elsa Anes and Gareth Griffiths J Immunol 2008; 181:2651-2663; ; doi: 10.4049/jimmunol.181.4.2651 http://www.jimmunol.org/content/181/4/2651 Downloaded from

References This article cites 49 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/181/4/2651.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 27, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

NF-␬B Activation Controls Phagolysosome Fusion-Mediated Killing of Mycobacteria by Macrophages1

Maximiliano Gabriel Gutierrez,2,3* Bibhuti B. Mishra,2† Luisa Jordao,† Edith Elliott,‡ Elsa Anes,† and Gareth Grifﬁths*

Macrophages can potentially kill all mycobacteria by poorly understood mechanisms. In this study, we explore the role of NF-␬B in the innate immune response of macrophages against Mycobacterium smegmatis, a nonpathogenic mycobacterium efficiently killed by macrophages, and Mycobacterium avium which survives within macrophages. We show that infection of macrophages with M. smegmatis induces an activation of NF-␬B that is essential for maturation of mycobacterial phagosomes and bacterial killing. In contrast, the pathogenic M. avium partially represses NF-␬B activation. Using microarray analysis, we identified many lysosomal enzymes and membrane-trafficking regulators, including cathepsins, LAMP-2 and Rab34, were regulated by NF-␬B during infection. Our results argue that NF-␬B activation increases the synthesis of membrane trafficking molecules, which may Downloaded from be rate limiting for regulating phagolysosome fusion during infection. The direct consequence of NF-␬B inhibition is the impaired delivery of lysosomal enzymes to M. smegmatis phagosomes and reduced killing. Thus, the established role of NF-␬B in the innate immune response can now be expanded to include regulation of membrane trafficking during infection. The Journal of Immu- nology, 2008, 181: 2651–2663.

ycobacterium tuberculosis is the causative agent of tu- unidentified components (4). In the latter study, endocytosis of a http://www.jimmunol.org/ berculosis that infects more than one-third of the mixture of lysosomal enzyme inhibitors significantly retarded M. M world’s population (1). Only 5–10% of these infected smegmatis killing, providing functional evidence that lysosomal individuals will develop tuberculosis at some point in their life (2). enzymes delivered to the phagosome contribute to this process. In A major factor in determining whether the disease is contained or contrast, pathogenic mycobacteria such as M. tuberculosis and progresses is the macrophage, which hosts mycobacteria in mem- Mycobacterium avium block phagolysosome fusion and NO re- brane-enclosed phagosomes. When macrophages are activated, lease and consequently survive and grow within phagosomes phagosomes enclosing the pathogen can fuse with lysosomes and (5). However, when the proinflammatory response of macro- the bacteria are killed. However, under steady-state conditions, the phages infected with pathogenic mycobacteria is stimulated by by guest on September 27, 2021 phagosome becomes programmed to prevent its maturation. As a some specific lipids, even pathogenic mycobacteria may be ef- consequence the bacteria survive and may grow. The mechanisms ficiently killed (6). by which lysosomal factors kill mycobacteria are poorly under- The family of NF-␬B transcription factors plays a crucial role in stood (3). In this study, we focused on the nonpathogenic Myco- the regulation of the inflammatory process (7). This family of pro- bacterium smegmatis as a model system and addressed the intra- teins is composed of two subfamilies: the NF-␬B proteins and the cellular mechanisms that are responsible for killing mycobacteria Rel proteins. NF-␬B transcription factors are present in the cyto- in mature phagosomes. plasm as heterodimers, most commonly of p65 and p50 subunits in M. smegmatis is phagocytosed by macrophages and is killed a complex with an inhibitor, I␬B (8). When proinflammatory sig- within phagosomes by a combination of factors delivered from naling occurs via activation of cell surface receptors (e.g., TLRs), lysosomes, reactive nitrogen intermediates, and likely other still the I␬B becomes phosphorylated, leading to its degradation by the proteosome system. This allows the active subunits to enter the nucleus where they up-regulate the transcription of two to three † *European Molecular Biology Laboratory, Heidelberg, Germany; Unidade de Ret- hundred genes (9). rovı´rus e Infecço˜es Associadas-Centro de Patogeńese Molecular Faculty of Pharmacy, University of Lisbon, Lisbon, Portugal; and ‡School of Biochemistry, Genetics, Mi- The products characteristic of early events of the early immune crobiology and Plant Pathology, University of KwaZulu-Natal, Pietermaritzburg, response mediated by NF-␬B include the release of ILs and in- South Africa ducible NO synthase-mediated NO production (10). However, re- Received for publication February 28, 2008. Accepted for publication May 28, 2008. active nitrogen intermediate production only plays a partial role in The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance mycobacterial clearance (6, 11). It is clear that some ILs such us with 18 U.S.C. Section 1734 solely to indicate this fact. IFN-␥ and TNF-␣ increase the maturation of mycobacterial phago- 1 M.G.G. was supported by a Research Fellowship from Alexander von Humboldt somes as well as killing of mycobacteria. However, the intracel- Foundation and is currently funded by an European Molecular Biology Organization lular mechanisms by which ILs boost mycobacterial killing are Fellowship. E.A. was supported by Fundaçaõ para a Cieˆncia e a tecnologia Grant POCI/BIA-BCM/55327/2004 with coparticipation of the European Union fund poorly understood (12–14). FEDER Programme POCI2010. The fusion of phagosomes with late endocytic organelles (often 2 M.G.G. and B.B.M. contributed equally to this work. abbreviated as lysosomes) is usually considered to be one, perhaps 3 Address correspondence and reprint requests to Dr. Maximiliano Gabriel Gutierrez, the main effector of mycobacterial killing. This is based on exten- European Molecular Biology Laboratory, Postfach 102209, 69117 Heidelberg, Ger- sive correlative data associating killing with the late fusion events many. E-mail address: [email protected] that deliver the proton ATPase and lysosomal enzymes to phago- Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 somes (15–17). For example, mutants of both M. tuberculosis and www.jimmunol.org 2652 NF-␬B CONTROLS PHAGOLYSOSOME FUSION

bacillus Calmette-Gue´rin that fail to block phagosome maturation ilized water. Quantitative cultures for M. smegmatis or M. avium were are killed more effectively than wild-type bacteria (18, 19). As performed by 10-fold serial dilutions inoculated on 7H10 agar plates. Five already mentioned, a role for lysosomal enzymes in killing is sup- microliters was plated by triplicate and the number of colonies was counted after 48 h and referred as number of colonies (CFU) per milliliter. ported by our earlier studies (4). Phagocytosis and inflammation are strongly correlated (20). Indirect immunofluorescence Binding of mycobacteria to macrophage surface receptors, especially TLR2 and TLR4, initiates a rapid proinflammatory response Cells were fixed with 3% paraformaldehyde solution in PBS for 10 min and which is more robust for nonpathogenic than for pathogenic my- quenched by incubating with PBS/50 mM NH4Cl. Subsequently, cells were permeabilized with 0.05% saponin in PBS containing 0.2% BSA and then cobacteria (21). This proinflammatory response is mediated by incubated with the primary and secondary Abs. Cells were mounted with MAPKs and by proinflammatory transcription factors, including Permafluor mounting medium (DAKO) and analyzed by confocal micros- NF-␬B (21). In addition, the signaling from TLR has also been copy (Zeiss LSM510). implicated in the regulation of phagosome maturation and bacterial killing (22). ELISA ␬ The data regarding the activation of NF- B by pathogenic my- The p65 activation was assayed using a multiwell assay TransAM NF-␬B cobacteria are conflicting. A number of studies have described Kit (Active Motif). Briefly, J774-infected macrophages were scraped and inhibition of NF-␬B activation by M. tuberculosis components (23, nuclear extracts were isolated. Five micrograms of total protein from each 24). Other studies showed an initial, transient NF-␬B activation sample was incubated in 96-well plates coated with NF-␬B consensus oli- Ј Ј but afterward the system is inhibited (25, 26). Other studies re- gonucleotide sequence (5 - AGTTGAGGGGACTTTCCCAGGC-3 ) for 1 h and then with primary anti-NF-␬B Ab and subsequently with secondary ␬ Downloaded from ported that M. avium and M. tuberculosis activate NF- B under HRP-conjugated Ab. After a chemiluminescent reaction, the luminescence some conditions (27, 28). produced was recorded. Competition experiments were conducted with the In this study, we focus on the role of NF-␬B during the proin- 22-bp dsDNA, either wild type (see above) or mutated: 5Ј- flammatory response to M. smegmatis infection. Our results show AGTTGAGCTCACTTTCCCAGGC. that the NF-␬B pathway is required for mycobacterial killing since blocking this pathway prevents killing and can even allow the Western blot

bacteria to grow in macrophages. In addition, we show that the Cells under the different conditions were washed with PBS, scraped in lysis http://www.jimmunol.org/ pathogen M. avium is indeed able to initiate the activation of buffer, and processed as described before (4). Briefly, 50 ␮g of protein NF-␬B in the first minutes after infection, but thereafter the patho- extracts was subjected to electrophoresis in 10% SDS-PAGE gels, trans- gen strongly represses NF-␬B activation. Activation of NF-␬B ferred to a nitrocellulose membrane, and blocked with 0.1% Tween 20/5% of milk TBS. The nitrocellulose membrane was then incubated with pri- during infection leads to the synthesis of a set of lysosomal en- mary Abs, washed, and incubated with secondary HRP-conjugated Abs. zymes as well as a large number of membrane-trafficking regula- The bands were visualized with a chemiluminescent reagent (Amersham tors, whose increased synthesis is likely important for efficient Biosciences). phagosome maturation and to allow the macrophages to kill M. smegmatis. Small interfering RNA (siRNA)4 knockdown by guest on September 27, 2021 RAW264.7 macrophages were grown to a final density of 40,000 cells/well Materials and Methods in 24-well plates. Cells were transfected using HiPerFect (Qiagen) with 25 Reagents nM siRNA against p65 (Qiagen) according to the manufacturer’s instruc- tions. Cells were left for 48 h before analysis of knockout phenotype by SC-514, SN-50, NF-␬B activation inhibitor (6-amino-4-(4-phenoxyphenyl- Western blot. Scrambled siRNA were used as nonsilencing control. ethylamino)quinazoline) and MG-132 were obtained from Calbiochem. The following Abs were used: polyclonal rabbit anti-p65, polyclonal rabbit anti-I␬B, polyclonal goat Ab anti-cathepsin (Cts) Z (Santa Cruz Biotech- RNA extraction nology). Rat monoclonal anti-LAMP-2 was purchased from the Iowa Hy- Total RNAs were isolated from J774 (uninfected and infected with M. bridoma Bank. The polyclonal rabbit anti-v-ATPase Ab was previously smegmatis) at 1 and 4 h postinfection (with and without NF-␬B inhibitors described (29). Rabbit polyclonal Ab anti-CtsH was raised as described SC-514 and SN-50) using a Qiagen RNeasy kit. previously (30). Rabbit affinity-purified Ab anti-Rab34 was purchased from Genemed Synthesis and generated using a specific peptide. Secondary Abs were conjugated with Alexa Fluor 555, Alexa Fluor 488 (Molecular Microarray analysis Probes), or HRP (Amersham Biosciences). Sample preparation and hybridization to CodeLink Mouse Whole Genome Cell lines and bacterial culture conditions Bioarray (ϳ36 000 mouse gene targets) were performed at the European Molecular Biology Laboratory (EMBL) Genecore Facility. The complete The mouse macrophage cell lines J774A.1 and RAW264.7 were cultured data set has been submitted to the GEO database as accession number as described previously (4). M. smegmatis mc2155 harboring a p19 (long- GSE8999. lived)-EGFP plasmid and M. avium MAC101 were grown as previously RNA samples isolated as described in Materials and Methods were sub- described (6). jected to quality control tests using a Bioanalyzer (Agilent Technologies) at the Genomics Core Facility, EMBL-Heidelberg. Samples with a 28S/ Macrophage infection 18S rRNA Ͼ 1 were used for microarray template preparation. Briefly, 2 ␮g of total RNA diluted bacterial mRNA controls and T7 oligo(dT) prim- Bacterial cultures in exponential growth phase were pelleted, washed in ers were incubated for 10 min at 70°C. All of the components were mixed PBS, and resuspended in DMEM to reach a multiplicity of infection of 10. and incubated for2hat42°C for first-strand cDNA synthesis. The first- Clumps of bacteria were removed by ultrasonic treatment of bacteria sus- strand components were then added to second-strand reaction mix and pensions in an ultrasonic water bath for 15 min followed by a low-speed incubated for2hat16°C. The double-stranded cDNAs were purified and centrifugation for 2 min. Cells were seeded onto 24-well tissue culture concentrated. This was then used for the in vitro transcription reaction in plates at 70% confluence. In each experiment, after1hofinfection, cells a mix of biotinylated UTP, dNTPs, and 10ϫ T7 enzyme mix at 37°C for were washed and medium plus gentamicin (10 ␮g/ml) were added to kill 14 h. The cRNAs were purified and assessed for quality using a extracellular bacteria. NF-␬B inhibitors were added concomitant with the Bioanalyzer. infection (t0) or after1hofinfection (t1). CFU assay 4 Abbreviations used in this paper: siRNA, small interfering RNA; qRT-PCR, quan- Macrophages were plated in 24-well plates and infected with M. smegmatis titative real-time PCR; BMM, bone marrow macrophage; Cts, cathepsin; Gla, at different time points. Cells were washed with PBS and lysed with ster- ␣-galactosidase. The Journal of Immunology 2653 Downloaded from http://www.jimmunol.org/

FIGURE 1. M. smegmatis infection activates NF-␬B in macrophages. A, J774 macrophages were infected with GFP-M. smegmatis or rhodamine-stained M. avium MAC101 (depicted in red) at the indicated time points and subjected to indirect immunoﬂuorescence against p65 (green). Pictures show by guest on September 27, 2021 representative confocal images. N, Nucleus. B, Quantitative kinetic analysis of p65 translocation in macrophages. Cells were infected with the indicated strains or incubated with LPS (10 ␮g/ml). Data represent the means Ϯ SEM from at least three independent experiments. At least 100 cells were counted .p Յ 0.001 relative to control in the same conditions. MS, M. smegmatis; MS-HK, M. smegmatis heat killed; MAC101, M. avium ,ءء .in each experiment C, J774 cells were infected with M. smegmatis or M. avium MAC101 at the indicated time points. Uninfected cells were used as a control. Nuclear fractions were isolated and a DNA-binding assay of induced NF-␬B activation was determined. NF-␬B-binding activity was effectively competed for by the wild-type p Յ 0.001 relative to control in the same conditions. D, Western blot analysis of I␬B ,ءء .consensus oligonucleotide but not mutated oligonucleotide degradation in cytoplasm. J774 cells were infected with M. smegmatis at the indicated time points and cytoplasmic fractions were isolated and I␬B was detected by Western blotting. A loading control detecting tubulin is shown.

Array hybridization Statistical analysis and calculation of fold change cRNA probes were subjected to fragmentation to generate 40–60 bp long The GenePix-scanned files were analyzed using the CodeLink Expre- oligonucleotide RNAs in a standard fragmentation reaction. These oligo- ssion Analysis software Codelink EXP version 4.0 (http://www4. nucleotide RNAs were then used for hybridization onto a Mouse CodeLink gelifesciences.com/aptrix/upp01077.nsf/Content/codelink_soft). The sig- Mouse Whole Genome Bioarray (ϳ36 000 mouse gene targets). Briefly, 10 nal and background fluorescence intensities were calculated for each spot ␮g of cRNAs and 5 ␮lof5ϫ fragmentation buffer were incubated at 94°C to average the intensities of every pixel inside the target region. The in- for 20 min. Ten micrograms of fragmented cRNA, 78 ␮l of hybridization tensity of each spot was calculated as the difference between mean signal buffer component A, 130 ␮l of component B to a final volume of 260 ␮l intensity and mean local background intensity. Each time point/condition were incubated at 90°C for 5 min to denature the samples followed by tested was repeated two to three times as biological replicates. The output immediate cooling on ice for 30 min for annealing. Two hundred fifty Code Link files were analyzed by using GeneSpring GX 7.3 Expression microliters of hybridization reaction mixture was loaded onto the array analysis software (Agilent Technologies), which filters those genes input port and sealed with strips to avoid dehydration. The slides were which are absent in all of the samples and we considered those genes incubated for 20 h at 37°C on a shaker at 300 rpm. Then bioarrays were which exhibit a signal intensity above the background level in at least transferred to a bioarray rack containing 0.75ϫ TNT (0.10 M Tris-HCl, one of the samples using the “Filtering on Flags” tool. All of the fold 0.15 M NaCl, 0.05% Tween 20) followed by the transfer of the racks to changes in signal intensity of the test samples were calculated by sub- a preheated 0.75ϫ TNT-containing chamber. It was then incubated at tracting the mean of the signal intensity of the controls (uninfected 46°C for 1 h. Each bioarray was then transferred to a chamber contain- macrophages) using the “Filtering on Fold Changes” tool. Genes show- ing 3.4 ml of streptavidin-Cy5 solution and incubated at room temper- ing 1.5-fold higher or lower expressions against the controls were con- ature for 30 min. The arrays were then washed four times (5 min each) sidered significant. All analyses of the fold changes were performed with 1ϫ TNT followed by a brief rinsing of the same with 0.1ϫ SSC using a p value cutoff Ͻ0.01. (0.05% Tween 20) for 30 s. Bioarrays were dried by centrifugation at Sample preparation and hybridization to CodeLink Mouse Whole 9000 rpm for 10 min. Then they were scanned with GenePix 4000B Genome Bioarray (ϳ36 000 mouse gene targets) were performed at the (Genomics Core Facility, EMBL-Heidelberg) and scanned files were EMBL Genecore Facility. The complete data set has been submitted to the analyzed by CodeLink Expression Analysis software. GEO database as accession number GSE8999. 2654 NF-␬B CONTROLS PHAGOLYSOSOME FUSION qPCR A total of 1 ␮g of RNA was used for random hexamer-primed cDNA synthesis (Superscript II reverse transcriptase; Invitrogen) according to the manufacturer’s protocol. Real-time PCR was performed on an Ap- plied Biosystems 7500 Real-Time PCR System using a SYBR Green PCR master mix (Applied Biosystems) and different sets of primers (Qiagen) at a final concentration of 0.3 ␮M, with a slight modification of the PCR steps: 95°C for 15 min for 1 cycle, 95°C for 15 s, 55°C for 30 s, and 60°C for 32 s for 40 cycles. Fluorescence data were collected at the amplification step. The mRNA expression profiles were normalized with respect to ␤-actin. Fold increase of each gene was calculated using the Ϫ2Ϫ⌬⌬Ct method.

Statistical analysis Data are presented as means Ϯ SEM of at least three independent experiments; p values (two-way ANOVA) are relative to the control.

Results Downloaded from M. smegmatis infection induces NF-␬B activation in macrophages To monitor NF-␬B activation during infection, we ﬁrst used ﬂuorescence microscopy to evaluate the translocation to the nucleus of p65 upon infection of J774 cells with M. smegmatis-

GFP or rhodamine-M. avium. In uninfected cells, we detected a http://www.jimmunol.org/ low basal level of translocation (Fig. 1B), in agreement with the observations of constitutive p65 shuttling in resting cells (31). In contrast, with LPS (10 ␮g/ml), a classical activator of NF- ␬B, the number of cells with positive nuclear staining increased steadily, reaching 80% of the cells by 24 h. Similar results were obtained with heat-killed M. smegmatis (Fig. 1B). When macrophages were infected with live M. smegmatis, p65 was trans- FIGURE 2. NF-␬B is required for killing of M. smegmatis by mac- located to the nucleus in a fraction (20%) of the cells already at rophages. A, J774 cells were infected with M. smegmatis in the presence by guest on September 27, 2021 30 min postinfection, reaching 34% by 1 h after infection (Fig. or absence of the NF-␬B inhibitor SN-50 at the indicated times after 1, A and B). Surprisingly, this translocation was transient, peak- infection and CFU were determined. B, Survival of M. smegmatis in the ing at 50 min postinfection and switching to a cytoplasmic dis- presence of several inhibitors of the NF-␬B pathway added concomitant tribution of p65 after1hofinfection that reached basal levels with the infection. C, RAW264.7 cells were knocked down for p65 by 24 h (Fig. 1, A and B). To confirm that these results are not using specific siRNA or scrambled RNA as a control. After 2 days, cells restricted to J774 macrophages, the same set of experiments were infected with M. smegmatis and CFU were determined. Inset was performed in bone marrow macrophages (BMM). In BMM, shows a representative knockdown of p65 using siRNA p65 compared the same pattern of p65 activation in response to the different with treatment with scrambled (scram) RNA (scrRNA). Tubulin was used as an internal control. D, J774 cells were infected with M. avium conditions was observed (data not shown). ␬ ␬ MAC101 in the presence or absence of the NF- B inhibitor SN-50 at When cells were infected with M. avium, we observed NF- B the indicated times after infection and CFU were determined. E, Sur- ϳ translocation in 30% of the cells after 30 min of infection; this vival of M. avium in the presence of several inhibitors of the NF-␬B level lowered to 5% by 1 h and 3% by 24 h after infection (Fig. 1, pathway added concomitant with the infection. F, RAW264.7 cells were A and B). Similar results were obtained in BMM (data not shown). knocked down for p65 using specific siRNA or scrambled RNA as a These data argue that, although NF-␬B is activated in a fraction of control. After 2 days, cells were infected with M. avium and CFU were M. smegmatis and, to a lesser extent, in M. avium-infected cells, in determined. Inset shows a representative knockdown of p65 using both cases the live bacteria are able to suppress the activation by siRNA p65 compared with treatment with scrambled RNA (scrRNA). p Յ 0.001 relative to ,ءء .min of infection. In agreement with previous reports, M. Tubulin was used as an internal control 60–30 avium induced a less potent and more transient activation of control in the same conditions. NF-␬B in macrophages (26). To corroborate these observations, we used ELISA to monitor the active form of NF-␬B in nuclear extracts of M. smegmatis-infected cells. For this we used an Ab that recognizes an competitor for NF-␬B binding inhibited the binding and a mu- epitope on p65 that is accessible only when NF-␬B is activated tated oligonucleotide, unable to bind p65, had no effect on and bound to its target DNA. The phosphorylated p65 can be NF-␬B binding activity (Fig. 1C); this provides strong evidence detected in isolated nuclei from infected cells at 10 min after of specificity of the positive reaction. In addition, Western blot infection (Fig. 1C). This confirmed that p65 was translocated to analysis showed an increase in degradation of I␬B in cytoplasm the nucleus in its active form. In agreement with the immuno- during the first hour of infection (Fig. 1D). fluorescence data, the signal after 60 min was lower than at 30 Taken together, these data show clearly that both live and killed min. Moreover, infection with M. avium induced a less potent M. smegmatis induced an early proinflammatory response medi- activation of NF-␬B compared with M. smegmatis (Fig. 1C). As ated by NF-␬B. However, the live bacteria are able to repress the expected, the wild-type consensus oligonucleotide used as a activation of NF-␬B after the first hour of infection. The Journal of Immunology 2655

Table I. Representative genes regulated during M. smegmatis infection in the presence or absence of NF-␬B inhibitors (INH)

Accession No. Common Name Gene Name 1 h 1-h INH 4 h 4-h INH

ILs NM_031168 Interleukin 6 IL-6 4.576 Ϫ3.030 282.100 22.620 NM_008361 Interleukin 1␤ IL1␤ 18.380 Ϫ2.400 109.200 7.580 NM_010548 Interleukin 10 IL-10 6.505 Ϫ4.780 31.340 NS NM_010554 Interleukin 1␣ IL-1␣ NS NS 13.210 NS AF453945 Interleukin 19 IL-19 NS NS 2.039 1.980 NM_177396 Interleukin 28 IL-28 Ϫ5.780 NS Ϫ2.140 NS NM_010551 Interleukin 16 IL-16 Ϫ3.060 NS Ϫ2.810 NS NM_013693 Tumor necrosis factor Tnf 16.020 NS 26.130 3.750 Matrix metalloproteinases NM_008605 Matrix metallo proteinase 12 Mmp12 2.320 NS 7.426 3.170 NM_019471 Matrix metallo proteinase 10 Mmp10 1.820 2.760 6.340 3.850 NM_008607 Matrix metallo proteinase 13 Mmp13 3.060 4.270 13.550 14.450 NM_010809 Matrix metalloproteinase 3 Mmp3 NS NS 2.249 NS NM_152944 Matrix metalloproteinase 21 Mmp21 NS NS 2.219 NS NM_011846 Matrix metalloproteinase 17 Mmp17 NS NS 1.723 NS TLRs NM_030682 Toll like receptor 1 Tlr1 NS Ϫ4.010 NS Ϫ3.950 NM_011905 Toll-like receptor 2 Tlr2 1.964 NS 1.711 NS Downloaded from NM_016928 Toll-like receptor 5 Tlr5 2.342 NS 4.192 NS NM_011604 Toll like receptor 6 Tlr6 NS Ϫ2.280 NS Ϫ2.790 NM_133211 Toll like receptor 7 Tlr7 NS Ϫ2.100 Ϫ2.700 NS Caspases NM_007610 Caspase 2 Casp2 NS NS Ϫ2.780 NS NM_007609 Caspase 4 Casp4 NS NS NS Ϫ2.240

NM_009811 Caspase 6 Casp6 NS NS Ϫ2.030 NS http://www.jimmunol.org/ NM_009812 Caspase 8 Casp8 NS NS Ϫ2.25 NS NM_015733 Caspase 9 Casp9 NS NS Ϫ2.11 NS NF-␬B related NM_009045 V-rel reticuloendotheliosis viral oncogene RelA 2.942 NS 2.597 NS homolog A (avian) NM_010907 Nuclear factor of ␬ L chain gene enhancer in Nfkbia(I␬B␣) 2.749 NS 2.185 NS B cell inhibitor, ␣ NM_010908 Nuclear factor of ␬ L chain gene enhancer in Nfkbib(I␬B␤) 1.714 NS 2.473 NS B cell inhibitor, ␤a NM_019777 Inhibitor of ␬B kinase, ⑀ IKK⑀ 1.984 Ϫ2.1 3.458 Ϫ2.6 NM_174989 I-␬B kinase IKKb NS NS 2.209 NS by guest on September 27, 2021 NM_099044 Reticulo-endothelial oncogene, RelA Rel 4.57 1.89 8.4 2.6

Activation of the transcription factor NF-␬B is essential for phagocytic uptake process for M. smegmatis or latex beads macrophages to kill the nonpathogen M. smegmatis (data not shown). In agreement, in BMM-treated with either NF-␬B is activated during the first hour after infection with M. SN-50 or SC-514, we also observed an increase in the number smegmatis. In the subsequent 3 h, the bacteria are subjected to a of viable intracellular bacteria only when these inhibitors were robust killing attack by the macrophage and the majority are killed added coincident with infection (data not shown). In addition, by 4 h (4). To investigate the role of NF-␬B in the intracellular there was no significant effect on overall macrophage viability killing of M. smegmatis, J774 macrophages were infected in the under any conditions using these drugs, as monitored by trypan presence or absence of inhibitors of NF-␬B activation. For this, blue staining (data not shown). bacteria were isolated from infected cells at different time points To complement these data using chemical inhibitors, we used and CFU were determined. The addition of SN-50, an inhibitor of interference RNA. The siRNA treatment led to an almost complete NF-␬B transport to the nucleus (32), significantly increased the removal of p65 from macrophages, as revealed by Western blot survival of M. smegmatis when it was added at the start of the analysis (Fig. 2C, inset). In agreement with the previous results, infection (Fig. 2A). However, when the inhibitor was added after knock-down of p65 in macrophages also increased the survival of 1 h of infection, bacterial killing was not affected, arguing that the M. smegmatis (Fig. 2C). After the treatment of cells with siRNA, killing potential of the NF-␬B system is restricted to the first hour some killing still occurred, possibly due to a residual pool of p65 of infection. being synthesized. Similar results were seen with SC-514, a highly selective inhib- We next performed the same set of experiments using inhibitor of IKK-2 (Fig. 2B). This compound specifically blocks NF- itors of NF-␬B activation and siRNA against p65 to evaluate the ␬B-dependent gene expression but not MAPK pathways (33). survival of M. avium. In contrast to what we observed with M. Moreover, the addition of NF-␬B activation inhibitor, a cell-per- smegmatis, there were no significant differences in survival of meable compound that acts as an inhibitor of NF-␬B transcrip- M. avium when NF-␬B activation was blocked or p65 knocked tional activation, also increased bacterial survival (Fig. 2B). down (Fig. 2, D–F). Altogether, these results argue strongly Since NF-␬B activation requires proteosome function, we also that proteins regulated by the NF-␬B system induced relatively observed that MG-132, a reversible and cell-permeable protea- early in infection, especially during the first hour, are required some inhibitor, completely blocked bacterial killing (Fig. 2B). for the process by which macrophages kill nonpathogenic in- Treatment of cells with different inhibitors had no effect on the tracellular mycobacteria. In contrast, although an early transient 2656 NF-␬B CONTROLS PHAGOLYSOSOME FUSION

Table II. Lysosomal hydrolases regulated during M. smegmatis infection in the presence or absence of NF-␬B inhibitors (INH)

Accession No. Common Name Gene Name 1 h 1-h INH 4 h 4-h INH

NM_010421 Hexosaminidase A hexA 1.746 NS 2.656 NS NM_011179 Prosaposin psap 5.940 NS 5.931 NS NM_031843 Dipeptidylpeptidase 7 dpp7 3.354 NS 3.937 NS NM_015750 Neuraminidase 2 Neu2 2.413 NS 2.602 NS AA592597 Mannosidase 1,␣ Man1␣ 4.900 Ϫ2.020 4.160 Ϫ2.220 NM_013463 Galactosidase, ␣ Gl␣ 1.700 NS 2.370 Ϫ2.110 AK082407 Dipeptidylpeptidase 10 dpp10 2.021 NS NS NS NM_026656 Mucolipin 2 Mcoln2 NS NS 2.462 NS NM_172827 Leucyl/cystinyl aminopeptidase Lnpep NS NS 2.071 NS NM_011421 Sphingomyelin phosphodiesterase 1, acid lysosomal Smpd1 NS NS Ϫ2.240 NS NM_007754 Carboxypeptidase D Cpd NS NS 1.700 Ϫ1.930 BQ938912 Lipase, hepatic Lipc NS NS 1.870 NS NM_008509 Lipoprotein lipase Lpl NS NS Ϫ2.030 NS NM_009982 Cathepsin C CtsC NS Ϫ2.290 NS Ϫ1.850 AK082407 Dipeptidylpeptidase 10 Dpp10 NS Ϫ2.410 NS Ϫ2.820 NM_013590 P-Lysozyme structural Lzp-s NS Ϫ2.460 NS Ϫ3.130 NM_007798 Cathepsin B CtsB NS NS NS Ϫ1.870 NM_007801 Cathepsin H CtsH NS NS NS Ϫ4.150 NM_022325 Cathepsin Z CtsZ NS NS NS Ϫ1.900 Downloaded from BQ555380 Dipeptidylpetidase 6 Dpp6 NS NS NS Ϫ1.840 NM_010764 Mannosidase 2, ␣ B1 Man2b1 NS NS NS Ϫ2.370 BQ557908 Monoglyceride lipase Mgll NS NS NS Ϫ1.960 NM_017372 Lysozyme Lyzs NS NS NS Ϫ2.130 NM_020561 Sphingomyelin phosphodiesterase, acid- like 3A Smpdl3a NS NS NS Ϫ1.810 NM_133676 O-sialoglycoprotein endopeptidase Osgep NS NS NS Ϫ1.810

NM_018830 N-acylsphingosine amidohydrolase 2 Asah2 NS NS NS Ϫ1.980 http://www.jimmunol.org/

activation of NF-␬B was seen with M. avium, this activation Although the functional classes of proteins described above was not sufficiently robust to have any significant effect on the were not unexpected, two sets of genes associated with lysosome- killing of this pathogen. mediated killing were more surprising. These two sets comprise those for lysosomal enzymes and vesicular trafficking regulators. M. smegmatis infection modulates the expression After 4 h, the expression of several lysosomal enzymes such as of membrane-trafficking regulators and several hexosaminidase A, prosaposin, dipeptidilpeptidases 7 and 10, lysosomal enzymes mannosidase 1␣, and ␣-galactosidase (Gla) was up-regulated (Ta- by guest on September 27, 2021 We next focused on the identification of factors that were regulated ble II). However, not all lysosomal enzyme genes were up-regu- during infection that could contribute to killing. For this, global lated. Smpd1 and Lpl were both down-regulated after 4 h of in- gene expression analysis was performed. We compared the gene fection (Table II). Rab and syntaxin proteins are regulators of expression profile in noninfected vs M. smegmatis-infected mac- vesicle budding, vesicle delivery, tethering, and fusion (34). Many rophages at 1 and 4 h. We scored a significant up- or down-reg- genes from both families were highly up-regulated after infection, ulation when the degree of difference was at least 1.5-fold and including syntaxin 6, syntaxin 7, Rab5b, Rab20, Rab24, and Rab12 statistically significant ( p Յ 0.01). (Table III). Overall, the analysis showed that after1hofinfection, when ␬ infected cells were compared with uninfected cells, 1145 genes Inhibition of NF- B leads to a change in the expression of were up-regulated and 1435 down-regulated. We categorized this genes connected to lysosomal enzymes and intracellular set of genes into different subsets of genes of interest. Numerous trafficking IL genes were up-regulated upon1hofinfection such as IL-6, The above list could now provide us with a reference for asking IL-1␤, IL-10, and TNF-␣. From the MAPK signaling pathway, which genes up-regulated by M. smegmatis infection are under the MAPK13 was 2.2-fold up-regulated. The expression of some met- control of NF-␬B. For this, we conducted an expression analysis in alloproteinases (MMP) was elevated, such as MMP10, MMP12, the presence and absence of NF-␬B inhibitors. Those genes which and MMP13. The mRNA levels for TLRs TLR2 and TLR5 were display differences in their activation when the NF-␬B system was also higher. As expected from our data, proteins associated with blocked would be considered to be potentially under the control the NF-␬B signaling pathway were also up-regulated, such as (directly or indirectly) of NF-␬B (Fig. 3A). From the set of genes RelA and IKK␧ (Table I). that were differentially regulated when the NF-␬B system is Since the most intense phase of killing occurs between 1 and 4 h blocked, some new interesting targets of lysosomal-mediated kill- after infection, we next monitored gene expression in live M. ing were identified. In the presence of the NF-␬B inhibitor, some smegmatis-infected cells at 4 h after infection. IL-6, IL-1␤, IL-10, lysosomal enzymes were down-regulated at 1 h postinfection such and TNF-␣ were also up-regulated at this time but the mRNA as pro-CtsC, mannosidase 1␣, and lysozyme. A larger set of hy- levels were considerably higher compared with1hofinfection. In drolytic enzymes were down-regulated at 4 h after infection in the contrast to the 1-h infected cells, other ILs were additionally up- presence of the inhibitor; including pro-CtsB, pro-CtsC, pro-CtsH, regulated, such us IL-1␣ and IL-19. Besides MMP10, the other pro-CtsZ, lysozyme, and Gla (Table II and Fig. 3B). related metalloproteinases MMP3 and MMP12, MMP13, MMP17, When NF-␬B activation was blocked, a number of genes in- and MMP21 were also higher at 4 h after infection. TLR2 and volved in the regulation of intracellular trafficking processes were TLR5 were up-regulated to a higher level at 4 h than at 1 h after significantly down-regulated, including Rab5a, Rab10, Rab20, and infection (Table I). Rab34 (Table III and Fig. 3C). Moreover, the expression of The Journal of Immunology 2657

Table III. Membrane-trafﬁcking proteins regulated during M. smegmatis infection in the presence or absence of NF-␬B inhibitors (INH)

Accession No. Common Name Gene Name 1 h 1-h INH 4 h 4-h INH

AW555412 Syntaxin 6 Stx6 6.773 NS 3.791 NS BE457571 Rab6-interacting protein 2 Rab6ip2 6.522 NS 4.408 NS NM_019972 Sortilin 1 Sort1 5.281 NS 10.750 NS NM_009295 Syntaxin-binding protein 1 Stxbp1 4.406 NS 5.016 NS AI048075 Lysosomal membrane glycoprotein 1 Lamp1 3.070 NS 2.090 NS NM_011229 Rab5B Rab5b 2.471 NS 2.133 NS NM_024448 Rab12 Rab12 2.387 NS 3.435 NS AK030769 Rab11 family-interacting protein 1 Rab11ﬁp1 2.149 NS 3.619 NS NM_009000 Rab24 Rab24 2.139 NS NS NS NM_011227 Rab20 Rab20 1.974 Ϫ3.950 7.942 Ϫ4.200 NM_177620 Ras and Rab interactor 3 Rin3 1.785 NS NS NS NM_016797 Syntaxin 7 Stx7 1.725 NS 1.786 NS AK017897 Syntaxin 11 Stx11 NS NS 5.327 6.130 NM_139128 Coronin, actin-binding protein 6 Coro6 NS NS 2.081 NS AI316712 Early endosome antigen 1 Eea1 NS NS 1.889 NS NM_019400 Rabaptin, RAB GTPase-binding effector protein 1 Rabep1 NS Ϫ12.590 1.800 Ϫ1.900 NM_019983 Rab guanine nucleotide exchange factor (GEF) 1 Rabgef1 NS Ϫ2.280 NS Ϫ2.050 NM_033475 Rab34, member of RAS oncogene family Rab34 NS Ϫ3.710 NS Ϫ3.780 NM_019490 Vesicle-docking protein Vdp NS NS NS Ϫ1.820 Downloaded from NM_008111 Guanosine diphosphate (GDP) dissociation inhibitor 2 Gdi2 NS NS NS Ϫ1.900 NM_025887 Rab5A, member RAS oncogene family Rab5a NS NS NS Ϫ1.920 NM_007479 ADP-ribosylation factor 4 Arf4 NS NS NS Ϫ1.930 NM_021518 Rab2, member RAS oncogene family Rab2 NS NS NS Ϫ2.000 NM_016676 Rab10, member RAS oncogene family Rab10 NS NS NS Ϫ2.030 NM_009003 Rab4A, member RAS oncogene family Rab4a NS NS NS Ϫ2.090

NM_177356 Lysosomal-associated membrane protein 3 Lamp3 NS NS NS Ϫ2.300 http://www.jimmunol.org/ NM_010685 Lysosomal membrane glycoprotein 2 Lamp2 NS NS NS Ϫ2.510

LAMP-2, a protein implicated recently in phagolysosome fusion (35), also decreased. These data argue that the NF-␬B system regulates gene expression of both lysosomal enzymes and proteins that control membrane trafﬁcking during M. smegmatis infection. These ﬁndings

also suggest that the activation of NF-␬B could provide a link by guest on September 27, 2021 between the proinﬂammatory response and lysosome-mediated killing of intracellular bacteria.

M. smegmatis infection up-regulates genes for lysosomal enzymes and membrane-trafficking regulators in an NF-␬B-dependent manner Dpp6, Dpp10, Manla, Man2b1, Mg11, Lzp-s, Lyzs, Smpd13a, Osgep, Asah2, CtsB, CtsC, CtsH, CtsZ We tested some selected genes of interest by quantitative real-time PCR (qRT-PCR) after 1 and4hofinfection. We were especially interested in selected lysosomal enzymes and membrane-traffick- Smpd1, Lp1 ing regulators. To address the lysosomal enzymes, we selected pro-CtsB, pro-CtsH, pro-CtsZ, and Gla. At the level of transcripts, none of the genes was significantly altered after1hofinfection hexA, psap, dpp7, Neu2, Mcoln2, Dpp10, Lipc, Lnpep, Cpd, Gla (Fig. 4A). In addition, CtsZ, CtsB, CtsH, and Gla were markedly up-regulated at 4 h after infection. However, the expression of CtsZ, CtsH, and Gla was markedly down-regulated in M. smegmatis-infected cells treated with the NF-␬B inhibitor. This is con- ␬ Stx6, Rab6ip2, Sort1, Stxbp1, Lamp-1, Rab5b sistent with these genes being normally activated by NF- B. In Rab12, Rab11fip1, Rab24, Rab20, Rin3, Stx7 contrast, CtsB was up-regulated at 1h but no differences relative to Coro6, Eeal uninfected controls were observed at 4 h (Fig. 4A). These results suggest that NF-␬B activation is a negative regulator of the ex- Stx11 pression of CtsB. We next focused on four interesting membrane-trafficking genes, Rab10, Rab20, Rab34, and LAMP-2. These genes were all Rabep1, Vdp, Gdi2, Rab5a, Arf4, Rab2, Rab10 Rabgef1, Rab4a, Lamp-3, Lamp-2, Rab34 significantly up-regulated during infection with M. smegmatis, compared with noninfected cells (Fig. 4B). The pattern of regula- FIGURE 3. Identification of genes associated to lysosomal-mediated killing by microarray. A, Classification of the gene expression data accord- tion at the transcriptional level during infection was different for ing to function. ϩ, Up regulated; Ϫ, down regulated. B, Lysosomal en- these proteins. In the case of Rab10, Rab34, and LAMP-2, the zymes regulated during infection with M. smegmatis in a NF-␬B dependent levels of transcripts were higher at 4 h after infection but not sig- manner. C, Membrane- trafficking regulators regulated during infection nificantly different from the uninfected cells at 1 h after infection. with M. smegmatis in a NF-␬B dependent manner. In contrast, Rab20 was highly up-regulated already from 1 h after 2658 NF-␬B CONTROLS PHAGOLYSOSOME FUSION

FIGURE 4. Analysis of the expression pattern of genes connected to intracellular trafﬁcking and lysosomal enzymes during M. smegmatis infection. A and B, qRT-PCR of the indicated genes were measured by SYBR Green incorporation. Results are shown as the average absolute gene regulation Ϯ SD of at least three in- p Յ 0.001 ,ءء .dependent experiments relative to control in the same condi-

tions. C and D, J774 cells were in- Downloaded from fected for 1 and4hinthepresence or absence of SC-514. Western blot analysis of CtsH and CtsZ (C) or LAMP-2 and Rab34 (D) in the treated cell lysates. Right panel of each ﬁgure indicates the representative quantitation of

one of three experiments that showed http://www.jimmunol.org/ similar results. Results are represented as normalized to the untreated control levels, arbitrarily put as 1. by guest on September 27, 2021

infection. Strikingly, the up-regulation of these four genes was increased synthesis may contribute to later stages of killing strongly NF-␬B dependent (Fig. 4B). Some discrepancies were (data not shown and Ref. 4). observed between the microarray and qRT-PCR data. The microar- The delivery of lysosomal enzymes to the phagosome depends ray approach was not able to detect the up-regulation of Rab34, on vesicular trafficking machinery. According to our data, some of LAMP-2, Rab10, and cathepsins during infection observed by the proteins that control trafficking are up-regulated upon infec- qRT-PCR. The lack of detection of the mRNA-relative signals for tion. Therefore, we performed experiments to evaluate the expres- these proteins was probably because of the lower sensitivity of this sion of membrane-trafficking proteins. Rab34 has been associated technique in detecting low-abundance messages, as previously re- to lysosome positioning within the cell (37). However, the role of ported (36). Altogether, these results indicate that the expression of this protein in phagosome maturation is unknown. The role of many genes involved in hydrolytic activity and intracellular trans- LAMP-2 in phagolysosome fusion and organelle motility has port is up-regulated during M. smegmatis infection by a NF-␬B- been recently shown. Cells lacking both LAMP-1 and LAMP-2 dependent process. showed impaired phagolysosome fusion without loss of lyso- Expression of Rab34, LAMP-2 but not mature cathepsins is somal membrane integrity (35). We therefore selected Rab34 regulated by NF-␬B during the first4hofM. smegmatis and LAMP-2 for further analysis. Western blot analysis showed infection that M. smegmatis infection up-regulates the level of both After infection, the activation of NF-␬B led to increased mRNA LAMP-2 and Rab34 proteins at 1 and 4 h after infection (Fig. levels for some lysosomal enzymes. However, Western blot 4D). However, both proteins were markedly down-regulated ␬ analysis showed that the total levels of mature Cts H and Z were when NF- B was blocked (Fig. 4D, see quantitation, right pan- not significantly affected at 1 and 4 h after infection (Fig. 4C). els). Altogether, these data show that NF-␬B-regulated expres- Since lysosomal enzymes are known to be long-lived proteins, sion of CtsH and CtsZ does not significantly contribute to the it is likely that the newly made RNAs for procathepsins will overall amounts of these proteins up to4hofinfection with M. result in higher amounts of the active enzymes at later times of smegmatis. In contrast, the elevated expression of LAMP-2 and infection. Actually, we observed an increase in the levels of Rab34 proteins in macrophages is strongly dependent on NF-␬B these enzymes after 12 and 24 h of infection; therefore, the activation early during infection. The Journal of Immunology 2659

FIGURE 5. Analysis of the expression pattern of genes connected to intracellular trafﬁcking and lysosomal enzymes during M. avium infection. A and B, qRT-PCR of the indicated genes were measured by SYBR Green incorporation. Results are shown as the average absolute gene regulation Ϯ SD of at least three indepen- -p Յ 0.001 rela ,ءء .dent experiments tive to control in the same conditions.

C and D, J774 cells were infected for Downloaded from 1and4hinthepresence or absence of SC-514. Western blot analysis of CtsH and CtsZ (C) or LAMP-2 and Rab34 (D) in the treated cell lysates. Right panel of each ﬁgure indicates the representative quantitation of one of three experiments that showed sim- http://www.jimmunol.org/ ilar results. Results are represented as normalized relative to the untreated control levels, arbitrarily put as 1. by guest on September 27, 2021

Regulation of the expression of genes for lysosomal confirmed by Western blot analysis (Fig. 5D). Taken together, this enzymes and membrane-trafficking regulators during argues that in the first hour of infection Rab34 and LAMP-2 are M. avium infection negatively regulated by NF-␬BinM. avium-infected cells. How- We next analyzed the same set of selected genes in the context of ever, no difference in the expression of these proteins was ob- M. avium infection. Focusing on CtsB, CtsH, CtsZ, and Gla1 ex- served at 4 h after infection. pression, we observed an up-regulation of CtsB but not CtsH, CtsZ, and Gla1 between 1 and 4 h after infection that were similar to Block of NF-␬B activation alters the normal fusion of uninfected cells (Fig. 5A). These findings with CtsH and CtsZ were mycobacterial phagosomes with lysosomes/late endosomes supported by Western blot (Fig. 5C). Surprisingly, two of the enzymes, CtsB and Gla1, were significantly up-regulated when the Our results collectively argue that infection with M. smegmatis cells were infected in the presence of SC-514; CtsB at both 1 and results in a significant NF-␬B-dependent regulation of membrane- 4 h, whereas up-regulation of Gla1 was only seen at 1 h. As for trafficking molecules (e.g., Rab34 and LAMP-2). Since M. smeg- CtsB during M. smegmatis infection, this provides further evidence matis is effectively killed by macrophages when NF-␬B is acti- that NF-␬B activation represses CtsB expression. A similar mech- vated, this led to the hypothesis that some of these molecules are anism can be invoked for Gla1 in M. avium-infected cells. necessary for trafficking events that allow phagosomes to fuse with When we investigated the expression of Rab10, Rab20, Rab34, late endosomes/lysosomes. To address this hypothesis, we moni- and LAMP-2 in M. avium-infected cells, no easily discernable pat- tored by microscopy the acquisition of different markers of late tern could be seen that was general for all of these mRNAs (Fig. endosomes/lysosomes by M. smegmatis phagosomes. 5B). The levels of Rab10 were not altered by the infection with or We first focused on the proton pumping v-ATPase. As in our without inhibitor. The levels of Rab20 were up-regulated by 1 and recent study (4), this marker localized to 15–30% of live M. smeg- 4 h after infection but not when NF-␬B was blocked. This pattern matis phagosomes between 1 and 4 h after infection. Treatment of is similar to that seen during M. smegmatis infection. Rab34 was macrophages with the NF-␬B inhibitor SC-514 significantly de- up-regulated after1hinpresence of the NF-␬B inhibitor but no creased the fraction of mycobacterial phagosomes labeled for this differences in expression were observed at 4 h of infection. A marker at 4 h after infection (Fig. 6A). When heat-killed myco- similar increase in the expression of LAMP-2 was seen in the pres- bacteria were internalized, the v-ATPase-labeled fraction was ence of the inhibitor at the 1-h time point. These observations were higher than with live bacteria. In this case, inhibition of NF-␬B 2660 NF-␬B CONTROLS PHAGOLYSOSOME FUSION Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 6. NF-␬B inhibition impairs fusion of M. smegmatis phagosomes with lysosomes/late endosomes. A, J774 cells were infected with live M. smegmatis in the presence or absence of SC-514. After4hofinfection, cells were fixed and LAMP-2 localization was determined. B, Quantitative analysis of the colocalization under different conditions. C, Cells were infected and processed as in A and v-ATPase was detected by indirect immunofluorescence. D, Quantitative analysis of the percentage of colocalization of v-ATPase with M. smegmatis phagosomes. Results are shown as means Ϯ SEM of at least .p Յ 0.001 relative to control (CTRL) in the same conditions. At least 100 phagosomes were counted in each experiment ,ءء .three independent experiments Bars, 10 ␮m. after 1 and4hofinternalization also led to a decrease in colocal- was significantly decreased after treatment with SC-514 (Fig. 7, ization with this marker (Fig. 6B). Similar results were obtained A–C) and SN-50 (data not shown). When macrophages were using another NF-␬B inhibitor, SN-50 (data not shown). incubated with heat-killed bacteria, the fraction of phagosomes As shown previously (4), live M. smegmatis phagosomes ac- that labeled for CtsH and CtsZ was also significantly reduced in quired LAMP-2 such that by 4 h the majority were positive for this the presence of the NF-␬B inhibitor (Fig. 7, B–D). This argues marker (Fig. 6C). In contrast, J774 cell treated with SC-514 that NF-␬B regulates processes that control the delivery of CtsH showed a significant reduction in the fraction of LAMP-2-positive and CtsZ to phagosomes enclosing both live and killed M. phagosomes. In agreement, the number of phagosomes positive for smegmatis. LAMP-2 was lower in BMM treated with SC-514 compared with To test whether this effect is general to all lysosomal enzymes, the control (data not shown). Thus, when the NF-␬B activation is we analyzed CtsS. The expression of this enzyme was not altered blocked, the efficiency of phagosome maturation is significantly in our microarray data. No difference in colocalization of M. smeg- retarded. matis phagosomes with CtsS were observed in the presence or NF-␬B had no effect on the overall macrophage protein con- absence of the NF-␬B inhibitor. Indeed, the percentage of colo- tent of CtsZ and CtsH during the first4hofinfection (Fig. 4C). calization for this marker was low from the beginning (data not The observation that blocking NF-␬B activation inhibited shown). Since the total levels of CtsZ and CtsH were not affected phagolysosomal fusion suggested that the delivery of these en- at4hofinfection, it seems that NF-␬B is more important for the zymes to phagosomes might be lowered in the absence of regulation of late endosome/lysosome delivery of these enzymes to NF-␬B activation. To investigate this, we monitored the frac- M. smegmatis phagosomes than for up-regulating the levels of ly- tion of M. smegmatis phagosomes that could be labeled for sosomal enzymes. Taken together, these results confirm that these markers. At 4 h after infection, the colocalization of live NF-␬B is required for processes leading to fusion of “lysosomes” M. smegmatis-containing phagosomes with both CtsH and CtsZ and M. smegmatis phagosomes. The Journal of Immunology 2661 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 7. NF-␬B inhibition impairs the delivery of CtsH and CtsZ to M. smegmatis phagosomes. A, J774 cells were infected with M. smegmatis in the presence or absence of SC-514. After 4 h of infection, cells were ﬁxed and subjected to indirect immunoﬂuorescence or CtsH. B, Quantitative analysis of the percentage of colocalization under different conditions. C, Cells were infected and processed as in A for CtsZ. D, Quantitative analysis of the ,ءء .percentage of colocalization of CtsZ with M. smegmatis phagosomes. Results are shown as means Ϯ SEM of at least three independent experiments p Յ 0.001 relative to control in the same conditions. At least 100 phagosomes were counted in each experiment. Bars, 10 ␮m.

Discussion The infection with live M. avium also initiated the activation of Pathogenic mycobacteria such as M. tuberculosis or M. avium sur- NF-␬B with the same kinetics as LPS or all other mycobacterial vive within macrophages, whereas nonpathogenic mycobacteria conditions. However, after 30 min, the system was blocked. Our such as M. smegmatis are killed. This means that pathogenic my- results argue that this small transient peak of activation is not ro- cobacteria block essential mechanisms of killing that are activated bust enough to activate genes that lead to the killing of M. avium by nonpathogenic mycobacteria. We rationalized that by studying since inhibition of NF-␬B had no effect on M. avium survival. how nonpathogenic mycobacteria are killed, it would not only fa- Thus, although both live M. avium and M. smegmatis allow an cilitate the identification of the general macrophage mechanisms initial activation of NF-␬B, the permissive period was more sig- involved in mycobacterial killing but it could also help to identify nificant for M. smegmatis. key targets that pathogenic mycobacteria subvert to survive. To identify which genes regulated by NF-␬B were important for In agreement with previous observations, we showed that M. killing, we performed RNA microarray analysis. Many expected smegmatis induced a potent activation of NF-␬B after infection NF-␬B-dependent genes were up-regulated during the infection (26). In this study, we demonstrated that this NF-␬B activation is including ILs and proteins involved in the known NF-␬B signaling essential for killing of M. smegmatis since several inhibitors and pathways. However, two classes of genes were identified that had siRNA treatment targeting NF-␬B prevent killing and even al- previously not been associated either with the classical NF-␬B lowed M. smegmatis to grow in macrophages. For this effect, it pathway or with mycobacterial infection. The first class includes a was important that the inhibitors were added at the start of infec- large set of lysosomal enzymes while the second pinpoints several tion. However, when they were added after1hofinfection, killing regulators of membrane trafficking whose expression is under the proceeded normally. This argues that a transient activation of control of NF-␬B. NF-␬B during the first hour of infection is essential for an efficient It has been shown that NF-␬B regulates the production of lyso- killing. somal enzymes in cancer cells (38). In some types of cancer, a 2662 NF-␬B CONTROLS PHAGOLYSOSOME FUSION continuous activation of NF-␬B induces the overproduction of ly- a proposed direct transport pathway from the Golgi to latex bead sosomal enzymes that are secreted (39). Our results show that this phagosomes, but it fails to reach phagosomes enclosing M. bovis phenomenon is more general since NF-␬B controls the expression bacillus Calmette-Gue´rin (40). of several lysosomal enzymes in M. smegmatis-infected It is still open whether the proteins we identified are directly or macrophages. indirectly regulated by NF-␬B. Using MatInspector (Genomatrix) Our results show that CtsH and CtsZ are up-regulated after 4 h software, we identified a putative NF-␬B binding site in the first of infection and down-regulated when NF-␬B was blocked. In con- 2-kb promoter region of Rab34 (data not shown). However, it is trast, Western blot analysis indicates that the total cellular content expected that some of the proteins identified here that regulate the of some of the active enzymes was not affected by the 4-h time fusion of mycobacterial phagosomes with lysosomal compart- point. However, later during the infection, the observed up-regu- ments may be indirectly regulated by NF-␬B (M. G. Gutierrez et lation resulted in a higher amount of cathepsins that are potentially al., unpublished data). In fact, 4 h after infection is sufficient time involved in later killing stages (data not shown). for other transcription factors to be activated in response to the Nevertheless, we clearly observed a reduction in the fraction of direct activation of NF-␬B. Indeed, NF-␬B- activated gene prod- phagosomes that are labeled by CtsZ and CtsH in mycobacterial ucts may themselves be transcription factors (45). phagosomes when NF-␬B activation was blocked. Moreover, the Whether or not pathogenic mycobacteria activate NF-␬Bisnot fraction of phagosomes that acquired LAMP-2 and v-ATPase was clearly established, with different groups reporting conflicting re- also reduced in the presence of NF-␬B inhibitors. This argues that sults (23, 24, 28, 46). Using confocal microscopy and ELISA, we proteins induced by the NF-␬B system regulate the overall pro- observed a low and transient activation of NF-␬BbyM. avium. cesses leading to fusion of phagosomes with late endosome-lyso- Similar results were reported before (26). In contrast, others ob- Downloaded from somes, which deliver lysosomal enzymes into the phagosome lu- served that high and low virulence strains of the M. avium-intra- men. These observations could be rationalized by the finding that cellular complex (MAC) activated NF-␬B (46). There are similar Ͼ20 known membrane-trafficking molecules are mostly up-regu- reports of NF-␬B activation in macrophages infected with M. tu- lated in a NF-␬B-dependent fashion in infected macrophages. berculosis (27, 28). However, many studies argue that the patho- Particularly interesting here were LAMP-2 and Rab34, two reg- gens strongly block the proinflammatory response in general (47,

ulators of lysosome motility and function. The synthesis of these 48). http://www.jimmunol.org/ proteins was significantly increased in the first4hofinfection of Based on our observations, we suggest the model in which control macrophages but not when NF-␬B was blocked. LAMP-2 pathogenic mycobacteria induce NF-␬B activation upon contact has recently been shown to mediate lysosome motility that is im- with the macrophage. As soon as the mycobacteria start the syn- portant for phagolysosome fusion (35). Our results suggest that the thesis and secretion of virulence effectors, the system may be re- macrophage protein level of LAMP-2 is rate limiting for this pro- pressed. This model is in agreement with previous studies of gene cess and that its synthesis must be significantly enhanced to cope expression in M. avium-infected cells (49). A hitherto unappreci- with the increased membrane trafficking to the phagosome in the ated benefit to the pathogen in inhibiting the NF-␬B system is the infected macrophages. consequence that the phagolysosome fusion events would also be Manipulation of Rab and SNARE proteins is a key component blocked. by guest on September 27, 2021 of the survival strategy of Mycobacterium (40–42). Moreover, ex- In summary, our results pinpoint a novel role of the NF-␬B pression of Rabs is down-regulated in patients with tuberculosis system to allow (directly or indirectly) the synthesis of molecules compared with healthy M. tuberculosis-infected donors (43).We involved in intracellular trafficking. This regulation is linked to found that the synthesis of Rab34 was elevated in a NF-␬B de- phagolysosome fusion, which facilitates killing of intracellular pendent manner in both M. smegmatis- and M. avium-infected mycobacteria. In addition, our microarray screen identified a large macrophages. However, the regulation of this protein was the op- number of potential new regulators of phagolysosome fusion; posite in the two systems. In M. smegmatis-infected cells, NF-␬B NF-␬B is thus a key regulator of factors that facilitate the intra- was a positive regulator, whereas in M. avium-infected cells it cellular killing of mycobacteria. behaved as a negative regulator. Although the mechanism of this difference is open, these data suggest that M. avium has the ability Acknowledgments to switch the transcriptional response linked to NF-␬B to a state We are grateful to Dr. Luis Mayorga for his critical reading of this manu- more conducive for its survival. Rab34 has been shown to regulate script. We also thank Vladimir Benes and Tomi Ivacevic at the Genecore lysosome positioning within the cell (37). Since lysosome posi- Facility and Stefan Terjung at Advance Light Microscopy Facility (EMBL- tioning may be required for mycobacterial killing, we are testing Heidelberg) for excellent technical assistance and advice. the hypothesis that the regulation of Rab34 expression controls the Disclosures ability of phagosomes to be positioned close to lysosomes (M. G. We have no financial conflict of interest. Gutierrez, W. Hong, and G. Griffiths, unpublished data). Indeed, we observed that the compartments positive for LAMP-2 are dis- References tributed more in the peripheral region in macrophages treated with 1. Frieden, T. R., T. R. Sterling, S. S. Munsiff, C. J. Watt, and C. Dye. 2003. the NF-␬B inhibitor SC-514 (Fig. 6C). Tuberculosis. Lancet 362: 887–899. ␬ 2. Kaufmann, S. H., and A. J. McMichael. 2005. Annulling a dangerous liaison: Rab20 was also up-regulated in a NF- B-dependent manner vaccination strategies against AIDS and tuberculosis. Nat. Med. 11: S33–S44. during both M. smegmatis and M. avium infection. This Rab was 3. Kusner, D. J. 2005. Mechanisms of mycobacterial persistence in tuberculosis. recently shown to be up-regulated in response to phagocytosis of Clin. Immunol. 114: 239–247. ␬ 4. Anes, E., P. Peyron, L. Staali, L. Jordao, M. G. Gutierrez, H. Kress, M. Hagedorn, Aspergillus fumigatus, a pathogen that also activates NF- B (44). I. Maridonneau-Parini, M. A. Skinner, A. G. Wildeman, et al. 2006. Dynamic life However, the role of Rab20 in the host response against intracel- and death interactions between Mycobacterium smegmatis and J774 macro- lular pathogens has not been investigated. Syntaxins are another phages. Cell. Microbiol. 8: 939–960. 5. Russell, D. G. 2001. Mycobacterium tuberculosis: here today, and here tomor- group of membrane-trafficking regulators that are up-regulated row. Nat. Rev. 2: 569–577. during the response against M. smegmatis infection. Stx6 is espe- 6. Anes, E., M. P. Kuhnel, E. Bos, J. Moniz-Pereira, A. Habermann, and G. Griffiths. 2003. Selected lipids activate phagosome actin assembly and mat- cially highly up-regulated at 1 and 4 h after infection but not when uration resulting in killing of pathogenic mycobacteria. Nat. Cell. Biol. 5: NF-␬B is blocked (Table II). Syntaxin 6 has been associated with 793–802. The Journal of Immunology 2663

7. Caamano, J., and C. A. Hunter. 2002. NF-␬B family of transcription factors: 30. Coetzer, T. H., E. Elliott, P. H. Fortgens, R. N. Pike, and C. Dennison. 1991. central regulators of innate and adaptive immune functions. Clin. Microbiol. Rev. Anti-peptide antibodies to cathepsins B, L and D and type IV collagenase: spe- 15: 414–429. cific recognition and inhibition of the enzymes. J. Immunol. Methods 136: 8. Delhalle, S., R. Blasius, M. Dicato, and M. Diederich. 2004. A beginner’s guide 199–210. to NF-␬B signaling pathways. Ann. NY Acad. Sci. 1030: 1–13. 31. Nelson, D. E., A. E. Ihekwaba, M. Elliott, J. R. Johnson, C. A. Gibney, 9. Natoli, G., S. Saccani, D. Bosisio, and I. Marazzi. 2005. Interactions of NF-␬B B. E. Foreman, G. Nelson, V. See, C. A. Horton, D. G. Spiller, et al. 2004. with chromatin: the art of being at the right place at the right time. Nat. Immunol. Oscillations in NF-␬B signaling control the dynamics of gene expression. Science 6: 439–445. 306: 704–708. 10. Hayden, M. S., A. P. West, and S. Ghosh. 2006. NF-␬B and the immune re- 32. Lin, Y. Z., S. Y. Yao, R. A. Veach, T. R. Torgerson, and J. Hawiger. 1995. sponse. Oncogene 25: 6758–6780. Inhibition of nuclear translocation of transcription factor NF-␬B by a synthetic 11. Shiloh, M. U., J. D. MacMicking, S. Nicholson, J. E. Brause, S. Potter, peptide containing a cell membrane-permeable motif and nuclear localization M. Marino, F. Fang, M. Dinauer, and C. Nathan. 1999. Phenotype of mice and sequence. J. Biol. Chem. 270: 14255–14258. macrophages deficient in both phagocyte oxidase and inducible nitric oxide syn- 33. Kishore, N., C. Sommers, S. Mathialagan, J. Guzova, M. Yao, S. Hauser, thase. Immunity 10: 29–38. K. Huynh, S. Bonar, C. Mielke, L. Albee, et al. 2003. A selective IKK-2 inhibitor 12. Flesch, I. E., and S. H. Kaufmann. 1990. Activation of tuberculostatic macro- blocks NF-␬B-dependent gene expression in interleukin-1␤-stimulated synovial ␥ phage functions by interferon, interleukin-4, and tumor necrosis factor. Infect. fibroblasts. J. Biol. Chem. 278: 32861–32871. Immun. 58: 2675–2677. 34. Zerial, M., and H. McBride. 2001. Rab proteins as membrane organizers. Nat. 13. Schaible, U. E., S. Sturgill-Koszycki, P. H. Schlesinger, and D. G. Russell. 1998. Rev. 2: 107–117. Cytokine activation leads to acidification and increases maturation of Mycobac- 35. Huynh, K. K., E. L. Eskelinen, C. C. Scott, A. Malevanets, P. Saftig, and terium avium-containing phagosomes in murine macrophages. J. Immunol. 160: S. Grinstein. 2007. LAMP proteins are required for fusion of lysosomes with 1290–1296. phagosomes. EMBO J. 26: 313–324. 14. Via, L. E., R. A. Fratti, M. McFalone, E. Pagan-Ramos, D. Deretic, and V. Deretic. 1998. Effects of cytokines on mycobacterial phagosome maturation. 36. Volpe, E., G. Cappelli, M. Grassi, A. Martino, A. Serafino, V. Colizzi, J. Cell Sci. 111: 897–905. N. Sanarico, and F. Mariani. 2006. Gene expression profiling of human macro- 15. Clemens, D. L., and M. A. Horwitz. 1995. Characterization of the Mycobacterium phages at late time of infection with Mycobacterium tuberculosis. Immunology tuberculosis phagosome and evidence that phagosomal maturation is inhibited. 118: 449–460. Downloaded from J. Exp. Med. 181: 257–270. 37. Wang, T., and W. Hong. 2002. Interorganellar regulation of lysosome positioning 16. Sturgill-Koszycki, S., P. H. Schlesinger, P. Chakraborty, P. L. Haddix, by the Golgi apparatus through Rab34 interaction with Rab-interacting lysosomal H. L. Collins, A. K. Fok, R. D. Allen, S. L. Gluck, J. Heuser, and D. G. Russell. protein. Mol. Biol. Cell 13: 4317–4332. 1994. Lack of acidification in Mycobacterium phagosomes produced by exclu- 38. Bien, S., C. A. Ritter, M. Gratz, B. Sperker, J. Sonnemann, J. F. Beck, and sion of the vesicular proton-ATPase. Science 263: 678–681. H. K. Kroemer. 2004. Nuclear factor-␬B mediates up-regulation of cathepsin B 17. Ullrich, H. J., W. L. Beatty, and D. G. Russell. 1999. Direct delivery of pro- by doxorubicin in tumor cells. Mol. Pharmacol. 65: 1092–1102. cathepsin D to phagosomes: implications for phagosome biogenesis and parasit- 39. Benes, P., A. Vashishta, S. Saraswat-Ohri, M. Fusek, S. Pospisilova, B. Tichy,

ism by Mycobacterium. Eur. J. Cell Biol. 78: 739–748. and V. Vetvicka. 2006. Effect of procathepsin D activation peptide on gene ex- http://www.jimmunol.org/ 18. Pethe, K., D. L. Swenson, S. Alonso, J. Anderson, C. Wang, and D. G. Russell. pression of breast cancer cells. Cancer Lett. 239: 46–54. 2004. Isolation of Mycobacterium tuberculosis mutants defective in the arrest of 40. Fratti, R. A., J. Chua, I. Vergne, and V. Deretic. 2003. Mycobacterium tubercu- phagosome maturation. Proc. Natl. Acad. Sci. USA 101: 13642–13647. losis glycosylated phosphatidylinositol causes phagosome maturation arrest. 19. Stewart, G. R., J. Patel, B. D. Robertson, A. Rae, and D. B. Young. 2005. My- Proc. Natl. Acad. Sci. USA 100: 5437–5442. cobacterial mutants with defective control of phagosomal acidification. PLoS 41. Kyei, G. B., I. Vergne, J. Chua, E. Roberts, J. Harris, J. R. Junutula, and Pathog. 1: 269–278. V. Deretic. 2006. Rab14 is critical for maintenance of Mycobacterium tubercu- 20. Aderem, A. 2003. Phagocytosis and the inflammatory response. J. Infect. Dis. losis phagosome maturation arrest. EMBO J. 25: 5250–5259. 187(Suppl. 2): S340–S345. 42. Roberts, E. A., J. Chua, G. B. Kyei, and V. Deretic. 2006. Higher order Rab 21. Schorey, J. S., and A. M. Cooper. 2003. Macrophage signalling upon mycobac- programming in phagolysosome biogenesis. J. Cell Biol. 174: 923–929. terial infection: the MAP kinases lead the way. Cell. Microbiol. 5: 133–142. 43. Jacobsen, M., D. Repsilber, A. Gutschmidt, A. Neher, K. Feldmann, 22. Blander, J. M., and R. Medzhitov. 2004. Regulation of phagosome maturation by H. J. Mollenkopf, A. Ziegler, and S. H. Kaufmann. 2005. Ras-associated small signals from Toll-like receptors. Science 304: 1014–1018. GTPase 33A, a novel T cell factor, is down-regulated in patients with tubercu- by guest on September 27, 2021 23. Pathak, S. K., S. Basu, K. K. Basu, A. Banerjee, S. Pathak, A. Bhattacharyya, losis. J. Infect. Dis. 192: 1211–1218. T. Kaisho, M. Kundu, and J. Basu. 2007. Direct extracellular interaction between 44. Cortez, K. J., C. A. Lyman, S. Kottilil, H. S. Kim, E. Roilides, J. Yang, the early secreted antigen ESAT-6 of Mycobacterium tuberculosis and TLR2 B. Fullmer, R. Lempicki, and T. J. Walsh. 2006. Functional genomics of innate inhibits TLR signaling in macrophages. Nat. Immunol. 8: 610–618. host defense molecules in normal human monocytes in response to Aspergillus 24. Pathak, S. K., S. Basu, A. Bhattacharyya, S. Pathak, M. Kundu, and J. Basu. fumigatus. Infect. Immun. 74: 2353–2365. 2005. Mycobacterium tuberculosis lipoarabinomannan-mediated IRAK-M induc- 45. Schreiber, J., R. G. Jenner, H. L. Murray, G. K. Gerber, D. K. Gifford, and tion negatively regulates Toll-like receptor-dependent interleukin-12 p40 produc- ␬ tion in macrophages. J. Biol. Chem. 280: 42794–42800. R. A. Young. 2006. Coordinated binding of NF- B family members in the re- 25. Toossi, Z., B. D. Hamilton, M. H. Phillips, L. E. Averill, J. J. Ellner, and sponse of human cells to lipopolysaccharide. Proc. Natl. Acad. Sci. USA 103: A. Salvekar. 1997. Regulation of nuclear factor-␬B and its inhibitor I␬B-␣/ 5899–5904. MAD-3 in monocytes by Mycobacterium tuberculosis and during human tuber- 46. Giri, D. K., R. T. Mehta, R. G. Kansal, and B. B. Aggarwal. 1998. Mycobacte- ␬ culosis. J. Immunol. 159: 4109–4116. rium avium-intracellular complex activates nuclear transcription factor- B in dif- 26. Lee, S. B., and J. S. Schorey. 2005. Activation and mitogen-activated protein ferent cell types through reactive oxygen intermediates. J. Immunol. 161: kinase regulation of transcription factors Ets and NF-␬BinMycobacterium-in- 4834–4841. fected macrophages and role of these factors in tumor necrosis factor ␣ and nitric 47. Beltan, E., L. Horgen, and N. Rastogi. 2000. Secretion of cytokines by human oxide synthase 2 promoter function. Infect. Immun. 73: 6499–6507. macrophages upon infection by pathogenic and non-pathogenic mycobacteria. 27. Dhiman, R., M. Raje, and S. Majumdar. 2007. Differential expression of NF-␬B Microb. Pathog. 28: 313–318. in mycobacteria infected THP-1 affects apoptosis. Biochim. Biophys. Acta 1770: 48. Falcone, V., E. B. Bassey, A. Toniolo, P. G. Conaldi, and F. M. Collins. 1994. 649–658. Differential release of tumor necrosis factor-␣ from murine peritoneal macro- 28. Loeuillet, C., F. Martinon, C. Perez, M. Munoz, M. Thome, and P. R. Meylan. phages stimulated with virulent and avirulent species of mycobacteria. FEMS 2006. Mycobacterium tuberculosis subverts innate immunity to evade specific Immunol. Med. Microbiol. 8: 225–232. effectors. J. Immunol. 177: 6245–6255. 49. Greenwell-Wild, T., N. Vazquez, D. Sim, M. Schito, D. Chatterjee, 29. Skinner, M. A., and A. G. Wildeman. 2001. Suppression of tumor-related gly- J. M. Orenstein, and S. M. Wahl. 2002. Mycobacterium avium infection and cosylation of cell surface receptors by the 16-kDa membrane subunit of vacuolar modulation of human macrophage gene expression. J. Immunol. 169: Hϩ-ATPase. J. Biol. Chem. 276: 48451–48457. 6286–6297.