Inactivation of Cdc42 Is Necessary for Depolymerization of Phagosomal F-Actin and Subsequent Phagosomal Maturation

This information is current as Maria Lerm, Veronika Patcha Brodin, Iida Ruishalme, Olle of September 27, 2021. Stendahl and Eva Särndahl J Immunol 2007; 178:7357-7365; ; doi: 10.4049/jimmunol.178.11.7357 http://www.jimmunol.org/content/178/11/7357 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2008/03/13/178.11.7357.DC1 Material

References This article cites 54 articles, 22 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/178/11/7357.full#ref-list-1

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021

• Fast Publication! 4 weeks from acceptance to publication

*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 © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Inactivation of Cdc42 Is Necessary for Depolymerization of Phagosomal F-Actin and Subsequent Phagosomal Maturation1

Maria Lerm,2* Veronika Patcha Brodin,2† Iida Ruishalme,* Olle Stendahl,* and Eva Sa¨rndahl3*†

Phagocytosis is a complex process involving the activation of various signaling pathways, such as the Rho GTPases, and the subsequent reorganization of the actin cytoskeleton. In neutrophils, Rac and Cdc42 are activated during phagocytosis but less is known about the involvement of these GTPases during the different stages of the phagocytic process. The aim of this study was to elucidate the role of Cdc42 in phagocytosis and the subsequent phagosomal maturation. Using a TAT-based protein transduc- tion technique, we introduced dominant negative and constitutively active forms of Cdc42 into neutrophil-like HL60 (human leukemia) cells that were allowed to phagocytose IgG-opsonized yeast particles. Staining of cellular F-actin in cells transduced with constitutively active Cdc42 revealed that the activation of Cdc42 induced sustained accumulation of periphagosomal actin. More- Downloaded from over, the fusion of azurophilic granules with the phagosomal membrane was prevented by the accumulated F-actin. In contrast, introducing dominant negative Cdc42 impaired the translocation per se of azurophilic granules to the periphagosomal area. These results show that efficient phagosomal maturation and the subsequent eradication of ingested microbes in human neutrophils is dependent on a strictly regulated Cdc42. To induce granule translocation, Cdc42 must be in its active state but has to be inactivated to allow depolymerization of the F-actin cage around the phagosome, a process essential for phagolysosome formation. The Journal of Immunology, 2007, 178: 7357–7365. http://www.jimmunol.org/

hagocytosis of microorganisms requires a complex ma- IgG-coated particles requires Rac and Cdc42 (4). The activation of chinery of membrane reconstitution and rearrangements of Rac and Cdc42 induces the formation of membrane extensions, P the actin cytoskeleton (1, 2). The identification of the Rho i.e., pseudopodia, whereas phagocytosis of C3b/C3bi-opsonized family of small GTPases and their functions has largely improved particles, which does not require membrane protrusions, involves the understanding of the mechanisms by which the actin cytoskel- activation of RhoA (4). Looking more closely at the involvement eton is regulated. Rho GTPases, including Rho, Rac, and Cdc42, of Rac and Cdc42 during IgG-mediated phagocytosis in macro- act as molecular switches by cycling between GDP- and GTP- phages, Hoppe and Swanson (5) showed that activated Cdc42 is by guest on September 27, 2021 bound states (3). In their active GTP-bound state, Rho GTPases restricted to the leading edge of the cell and to the tip of the phago- regulate cellular functions through interaction with downstream cytic cup, whereas activated Rac1 is found evenly distributed effectors, many of which are involved in the dynamic rearrange- throughout the phagocytic cup. Neither protein in its activated ment of actin filaments. form is localized around the phagosome. In neutrophils, Rac and In phagocytic cells such as neutrophils and macrophages, two Cdc42 are activated during both CR3- and Fc␥R-mediated phago- main routes of phagocytosis of opsonized particles have been de- cytosis (6). However, the role of these Rho GTPases during the scribed (4). Particles coated with C3b/C3bi fragments of the com- different stages of phagocytosis in neutrophils is, to date, not es- plement system are recognized by the complement receptors CR1 tablished (6–9). For phagolysosome formation to occur, the actin ␤ and CR3 (also called 2 integrin), whereas particles coated with network coating the phagocytic cup and the developing phagosome ␥ IgG are recognized by Fc R. In macrophages, phagocytosis of must be reduced to allow proper apposition and the subsequent fusion of granules with the phagosome (10, 11). In fact, certain pathogens, *Division of Medical Microbiology, Institute for Molecular and Clinical Medicine e.g., Leishmania and Salmonella, build up periphagosomal F-actin as † and Division of Cell Biology, Institute for Biomedicine and Surgery, Faculty of a strategy for intracellular survival (12–14). Health Sciences, Linko¨ping University, Linko¨ping, Sweden The present study investigates how Cdc42 modulates F-actin Received for publication January 22, 2007. Accepted for publication March 7, 2007. around IgG phagosomes during phagosomal maturation in human 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 neutrophils. Using the TAT transduction technique, we show that with 18 U.S.C. Section 1734 solely to indicate this fact. constitutively active Cdc42 markedly enhances the amount of F- 1 This work was supported by grants from the King Gustaf V Memorial Foundation, actin surrounding phagosomes containing IgG-opsonized yeast. the Swedish Research Council (13103 and 5968), the Swedish Medical Society, and Furthermore, we find that Cdc42-mediated accumulation of F-actin the Medical County Council of O¨ stergo¨tland. M.L. was supported by a postdoctoral fellowship from the Swedish Society for Medical Research and an assisting professor at the phagosome results in the prevention of phagolysosome fu- fellowship from the Swedish Research Council (529-2003-5994), V.P.D. by a Ph.D. sion. Previous studies in neutrophils have shown the activation of student fellowship from the Network for Inflammation Research funded by the Swed- ish Foundation for Strategic Research, and E.S. by an assisting professor fellowship Rho GTPases during phagocytosis but have not clarified the from the Swedish Medical Research Council (12725). role of active/inactive Rho GTPases during the formation of the 2 M.L. and V.P.B. contributed equally to this work. phagosome and the subsequent phagolysosome. This is the first 3 Address correspondence and reprint requests to Dr. Eva Sa¨rndahl, Division of Med- study showing a role for Cdc42 in the remodeling of peripha- ical Microbiology, Faculty of Health Sciences, Linko¨ping University, SE-58185 gosomal F-actin, and our results suggest that a tightly regulated Linko¨ping, Sweden. E-mail address: [email protected] Cdc42 is required for an efficient phagolysosome formation in Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 human neutrophils. www.jimmunol.org 7358 Cdc42 MODULATES PHAGOLYSOSOME FUSION

Materials and Methods TAT-Cdc42 transduction Materials Transduction of TAT-Cdc42 was generally performed in KRG without 2ϩ 2ϩ The Escherichia coli strain B121 (DE3) was purchased from Novagen. Ca to avoid activation of the cells. Ca depletion did not reduce the amount of TAT-Cdc42 found in the cell lysates. Neutrophils and HL60 Ni-NTA agarose was from Qiagen. RPMI 1640, FCS, L-glutamine, peni- cells were incubated at 37°C with 200 nM PMB-treated TAT fusion protein cillin, streptomycin, and trypan blue came from Invitrogen Life Technol- for the indicated times (Fig. 1). To test whether the proteins were success- ogies. Ampicillin, formaldehyde, glutaraldehyde, glycerol, DMSO, PIPES, fully transduced, the cells were collected by centrifugation (200 ϫ g for 10 HEPES, EDTA, EGTA, isopropyl ␤-D-thiogalactoside, polymyxin B min at room temperature), washed three times, and lysed in sample buffer. (PMB),4 saponin, lysophosphatidylcholine, and Bradford reagent were pur- The samples were treated with Benzonase to cleave DNA and subjected to chased from Sigma-Aldrich, and Benzonase and phenol red were from SDS-PAGE. The proteins were transferred onto nitrocellulose, and the Merck. BSA, aprotinin, leupeptin, Pefabloc, and the FITC-conjugated hem- TAT-GTPases were detected by means of ECL using a rabbit polyclonal agglutinin (HA) Ab came from Roche. The HA Ab was purchased from Ab directed toward the HA epitope and a secondary HRP-conjugated anti- Santa Cruz Biotechnology. The HRP-conjugated goat anti-rabbit IgG Ab rabbit Ab. To analyze the efficiency of the TAT transduction, FITC-con- and the fluorescent mounting medium were obtained from DakoCytomation. jugated anti-HA Fabs were incubated (on ice for 10 min) together with the Alexa 594-conjugated phalloidin, Bodipy-conjugated phalloidin, and TAT-fusion protein and PMB before being given to the cells (37°C). After LysoTracker Red DND-99 came from Molecular Probes. The mouse anti- washing, the fluorescence was determined by microscopy. Nontransduced Cdc42 Ab came from BD Transduction Laboratories. Phosho-p38 MAPK cells, i.e., cells treated as described above but not subjected to TAT pro- Ab was obtained from Cell Signaling Technologies, and p38 MAPK Ab came teins, were used as control and showed no intracellular fluorescence. from New England Biolabs. PD-10 columns and an ECL system were ob- tained from Amersham Biosciences. The Limulus assay was from Coatech. Phagocytosis The CD63 Ab was kindly provided by Dr. Roos, (University of Amsterdam, Amsterdam, The Netherlands), and the pTAT-HA-Cdc42 vector was a gift After TAT transduction, the cells were washed twice and resuspended in from Dr. S. Dowdy (University of California, San Diego, La Jolla, CA). KRG. Phagocytosis on glass slides was performed as previously described Downloaded from (22). In short, TAT-transduced cells were seeded on glass slides covered Purification of TAT proteins with IgG-opsonized yeast, and incubated for 15 min at 37°C. Fixing the cells in ice-cold 3% paraformaldehyde (PFA) with 0.05% glutaraldehyde The pTAT-HA vector carrying dominant negative (N17) and constitutively terminated the phagocytosis. The cells were permeabilized with lysophos- active (V12) mutants of Cdc42 (N17Cdc42 and V12Cdc42, respectively) phatidylcholine (100 ␮g/ml) and stained with Alexa 594-conjugated phal- was transformed into the E. coli strain Bl21 (DE3), and well-expressing loidin (1/500 in PBS) to visualize F-actin. Fluorescence microscopy images clones were identified. Large-scale purifications were performed in Luria-

were captured with a Nikon Eclipse 800 fluorescence microscope equipped http://www.jimmunol.org/ Bertani broth containing 50 ␮g/ml ampicillin and 0.1 mM isopropyl ␤-D- with Easy Image software (Tekno Optik). For each set of experiments the thiogalactoside. The 6His-tagged TAT fusion proteins were purified from images were taken with identical camera settings. To ensure that the majority the bacterial lysates on Ni-NTA-agarose columns as previously described of the particles were fully phagocytosed under these conditions, the fluores- (15). Briefly, the bacteria were collected by centrifugation (5,000 ϫ g for cence quenching method (23) was used in experiments run in parallel. 10 min at 4°C), lysed in buffer Z (8 M urea and 10 mM imidazole), son- To evaluate the degree of phagocytosis, the TAT-transduced cells were icated, and centrifuged (18,000 ϫ g for 10 min at 4°C). The supernatants either kept in suspension (6) or were allowed to adhere to glass slides (22). were loaded onto Ni-NTA-agarose for affinity purification of the 6His- FITC-labeled yeast, prepared as previously described (24), was added to tagged TAT-Cdc42. The columns were washed with 25 bed volumes of the cells (ratio 5:1 yeast/cell) and incubated at 37°C. Phagocytosis was buffer Z before the proteins were eluted with an increasing concentration of stopped by the addition of ice-cold PBS, after which the cells were kept at imidazole (0.1, 0.25, 0.5, and 1.0 M) in buffer Z. The eluted fractions were 4°C until counted. The cells were resuspended in an equal volume of trypan checked for protein concentration using Bradford reagent, pooled, and de- blue to quench the extracellular fluorescence and thereby distinguish be- by guest on September 27, 2021 salted on PD-10 columns equilibrated with buffer (50 mM Tris (pH 7.4) tween intracellular and extracellular yeast particles (23). Phagocytosis was and 10% glycerol). After elution in the Tris-buffer, the protein preparations quantified microscopically in coded samples by counting ingested yeast were aliquoted and stored at Ϫ70°C. The amount of protein in each prep- particles per 100 cells for each treatment. The statistical significance was aration was estimated on a Coomassie gel. The eluted protein, which mi- determined by Student’s t test (two-sample, equal variance). grated as a single band with a molecular mass of ϳ30 kDa, was checked by Western blotting using anti-HA Abs (rabbit polyclonal) and anti-Cdc42 Translocation of CD63 to the phagosomal membrane Abs (mouse monoclonal). The amount of contaminating LPS, which was determined by means of a Limulus assay, ranged between 5 and 20 ng/ml The phagocytosis experiment was performed on the substratum as de- (final concentration). To prevent unwanted effects due to LPS, 500 ␮g/ml scribed above, but the cells were fixed in ice-cold 4% PFA before perme- PMB was added to the protein preparations before use (final concentration abilization with saponin (0.1%; w/v). Nonspecific binding was blocked of PMB with the cells was 10 ␮g/ml) (16). Because PMB did not affect the with BSA (2%; w/v) and goat serum (10%; v/v). The preparations were efficiency of the TAT transduction, the survival of the cells, the cell mor- incubated with a mouse anti-CD63 Ab (1/500), washed, and incubated with phology, or phagocytosis, it was used in all samples throughout the study. Alexa 594-conjugated goat anti-mouse Ab (1/400) and/or Bodipy-phalloi- din (1/500). Finally, the preparations were washed and mounted with Preparation of human neutrophils DakoCytomation fluorescent mounting medium. Imaging was performed in a Sarastro 2000 confocal microscope (Molecular Dynamics) equipped with Peripheral human blood was drawn from healthy volunteers and collected dual detectors, a Nikon microscope, and a ϫ60 oil immersion objective in heparin-containing Vacutainer tubes, and neutrophils were isolated by (numerical aperture 1.40; Nikon). Excitation of the Alexa 594 fluorophore means of sucrose gradient centrifugation as previously described (17–19). was achieved with an argon ion laser filtered through a dichroic mirror The obtained cell suspension, consisting of Ͼ95% neutrophils, was resus- (cutoff wavelength of 535 nm). For detection, a 600-nm long-pass emission pended in ice-cold Krebs-Ringer phosphate buffer (KRG; 120 mM NaCl, filter was used. For all samples, blinded analysis was performed and only 1.2 mM MgSO ϫ 7HO, 1.7 mM KH PO , 8.3 mM Na HPO ϫ 2HO, 4 2 2 4 2 4 2 cells with single, well-ingested phagosomes were analyzed. At least 50 1 mM CaCl , 4.9 mM KCl, and 10 mM glucose (pH 7.3)). 2 phagosomes per treatment in three independent experiments were scanned. Cell culture For each set of experiments the images were taken with identical camera settings. The images were evaluated with respect to fluorescent staining at The promyelocytic human leukemia HL60 cells were grown in suspension the phagosomal membrane and classified as positive or negative for CD63 culture in RPMI 1640 supplemented with 10% heat-inactivated FCS, L- staining. For the costained preparations, the Alexa 594 and Bodipy fluoro- glutamine (2 mM), penicillin (100 U/ml), and streptomycin (1 mg/ml), as phores were excited as above and detected with a 600-nm long-pass emis- previously described (20). The cells were maintained in a humidified at- sion and a 545DF30-nm filter, respectively. This filter setup ensured no 2 mosphere of 5% CO2 at 37°C and given fresh medium every third day. The detection of the red signal in the green channel or vice versa. ␹ analysis cells were induced to differentiate by the addition of DMSO (1.3%; v/v) was used to determine the significance in reduction of phagosomal matu- and harvested after 6 or 7 days (21). By then, Ͼ75% of the cells had ration observed with transduced vs nontransduced cells. differentiated to granulocyte-like cells as measured by NBT reduction. Preparations for scanning electron microscopy

4 Abbreviations used in this paper: PMB, polymyxin B; HA, hemagglutinin; KRG, Krebs- TAT-transduced cells were incubated with phalloidin (0.004% v/v) for 5 Ringer phosphate buffer; N17Cdc42, dominant negative form of Cdc42; PFA, parafor- min on ice to stabilize the F-actin cytoskeleton. The cells were extracted by maldehyde; RSD, relative SD; V12Cdc42, constitutively active form of Cdc42. the addition of Triton X-100 (0.5%; v/v) in a microtubule stabilizing buffer The Journal of Immunology 7359

(1 mM EGTA, 100 mM PIPES without Ca2ϩ, 4% (w/v) polyethylene gly- total area of all image regions in the cell. This measure is dimensionless col (pH 6.9), and phenol red) (25) for 5 min at room temperature. The and independent of individual cell sizes and is depicted in Equation 1, concentration of phalloidin was maintained at 0.004% during the extraction procedure. Cells were washed once in PBS before fixation for 30 min at max areai ʦ ͓ ͔ room temperature in 2.5% glutaraldehyde in 0.15 M sodium cacodylate i 1,...N rs ϭ (1) buffer. For dehydration, the samples were washed once in PBS and then N placed in 50% ethanol and the alcohol content of the solution was gradually ͸ areai ϳ increased to 99.5% during an 1-h period. The samples were placed inside i ϭ 1 an E3000 critical point drying apparatus (Polaron). Ethanol was drained from the samples and exchanged for carbon dioxide. After ϳ1.5 h, when where areai is the area of a region and N is the number of image regions the exchange was complete, the temperature was raised from 10°C to 40°C in the cell.

and the CO2 was slowly let out. The dried and water-free samples on glass The number of granule populations. Another way to estimate the degree coverslips were mounted on metal stubs and sputter-coated with 12-nm of of accumulation of granules is to measure the number of granule popula- tungsten. Scanning micrographs were taken in a JEOL 840 scanning elec- tions by assuming that a low number of granule populations corresponds to tron microscope. To evaluate the morphology of the cytoskeleton of the a high accumulation of granules. The number of granule populations was cells, images from coded samples were made without tilting the stubs with established by counting the identified image regions. ϫ the same magnification ( 650) from the proximity of the center of the The “closeness” to the phagosome of the granule populations. The dis- glass coverslip and, using the same images, the area of 140 (nontrans- tance between each granule population and the phagosome within the same duced) and 70 (TAT-V12Cdc42 and TAT-N17Cdc42-transduced, respec- cell was estimated as the measure “closeness”. Thus, a high value of “close- tively) randomly chosen cells per treatment was measured with Scion Im- ness” corresponds to a small average distance between the granule populations age software. The areas were compared groupwise with the nonparametric and the phagosome. The average “closeness” to the phagosome (C)ofall Mann-Whitney U test based on ranks, which was chosen to avoid assump- image regions in the cell was established as a measure where the “closeness” tions of the distribution of the data. of an individual image region is defined as the area of the image region Downloaded from divided by the square of its distance to the phagosome center. This Following lysosomes with LysoTracker measure is dimensionless and independent of individual cell size and is shown in Equation 2, To allow observation of granule movement in living cells, the granules were prelabeled with LysoTracker Red DND-99 as follows: The cells, N ͸ areai nontransduced or transduced with TAT-V12-Cdc42 and N17-Cdc42, were 2 di allowed to adhere to the glass bottom of a Delta T dish (Bioptechs) for 15 i ϭ 1 http://www.jimmunol.org/ C ϭ d ϭ ͱ͑x Ϫ c ͒2 ϩ ͑y Ϫ c ͒2 (2) min at 37°C, after which LysoTracker (50 nM) was added for an additional N i i x i y 15-min incubation at 37°C. The labeled cells were washed twice and the dish was mounted in a Confocal microscope (Bio-Rad Radiance 2100) where areai is the area of a region, di is the Euclidean distance from a equipped with a temperature control set at 37°C. IgG-opsonized yeast par- region’s “center of mass” (xi, yi) to the phagosome “center of mass” (cx, cy), ticles were added to the cells and the coded samples were evaluated by and N is the number of regions in the cell. time-resolved microscopy using the ϫ60 oil immersion objective (numer- The spatial distribution of the granule populations. To measure the ical aperture 1.40; Nikon). Each sample was scanned every fourth second spatial distribution of granules in the cell, the statistical dispersion of the using the 514-nm line of the argon laser and a 600-nm long-pass filter for distances between granule populations within each cell was established. the emitted signal. At least 15 images were taken to create a movie. The statistical dispersion of the intracellular spatial distribution of image regions was expressed by the relative SD (RSD) of all interregional dis- by guest on September 27, 2021 tances (d ), as shown in Equation 3, Quantitative image analysis ij ϭ ͱ͑ Ϫ ͒2 ϩ ͑ Ϫ ͒2 ʦ ͕ Ϫ ͖ ʦ ͕ ϩ ͖ To perform objective quantifications of granule movement in living cells, dij xi xj yi yj i 1,..N 1 j i 1,..N (3) representative confocal microscope images were subjected to quantitative where dij is the Euclidean distance from one region’s “center of mass” (xi, image analysis using the ImageJ software package (26). In the original yi) to another region’s “center of mass” (xj, yj) and N is the number of red-green-blue color microscope images, LysoTracker-dyed granules ap- regions in the cell, RSD is a dimensionless measure and is independent of peared in the red channel only. The center of the phagosome and the cell cell size. A high RSD value represents a more scattered granule boundary were manually identified in green and blue, respectively. This distribution. allowed for all further analyses to be performed automatically by using a specially developed ImageJ plugin. The red, green, and blue channels of Phosphorylation of p38 MAPK the image were thereafter extracted into separate 8-bit grey scale images. The red image was thresholded (minimum of 50) and a binary image of After TAT transduction, the cells were washed twice and resuspended in ϫ 6 LysoTracker-dyed areas depicted in black on a white background was cre- KRG. The cells (2 10 /sample) were preincubated for 1 min at 37°C and ated. Using ImageJ, the blue image was converted into a binary mask stimulated with IgG-opsonized yeast particles (ratio 1:5). Phagocytosis was representing the cell area. Combining the mask with the binary red images stopped after 15 min by the addition of ice-cold PBS supplemented with ϫ created a new image containing only the LysoTracker-dyed areas of the Na3VO4 (1 mM). The cells were pelleted (3,500 g for 30 s at 4°C) and studied cell. To find and measure individual regions, the “Analyze Parti- resuspended in lysis buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 1 mM ␮ cles” function in ImageJ was applied, resulting in “area” and “center of EDTA, 2 mM EGTA, 10% glycerol, 1% Triton X-100, 1 g/ml aprotinin, ␮ mass” statistics of each identified region in the image. Each separate region 2 g/ml leupeptin, 1 mM Na3VO4, and 1 mM Pefabloc). After lysis (15 ϫ was termed a “granule population,” and this term was applied to popula- min at 4°C), the samples were centrifuged (10,000 g for 10 min at 4°C) tions regardless of the number of granules present in the population. The and the supernatant was diluted with sample buffer. The samples were green image was used to automatically obtain the spatial coordinates of the subjected to SDS-PAGE and transferred to a nitrocellulose membrane. The “center of mass” of the phagosome region. membrane was blocked with 4% BSA overnight. Phosphorylation of p38 For each cell, the degree of accumulation of granules around the phago- MAPK was detected by means of ECL using an anti-phosho-p38 MAPK some as well as the cellular distribution of granules within the cell was Ab (1/1,000) and a secondary HRP-conjugated anti-rabbit Ab. To visualize established. This was done by determining the following: 1) the relative total amounts of p38 MAPK proteins, the membrane was stripped and size of granule populations in each cell; 2) the number of granule popu- reblotted with a rabbit polyclonal anti-p38 MAPK Ab. The blots were lations in each cell; 3) the “closeness” to the phagosome of the granule scanned and the relative intensity of the bands was quantified using Scion populations within the cell; and 4) the spatial distribution of the granule Image Alpha version 4.0.3.2. The statistical significance was determined populations within the cell. Statistical analysis of the data using GraphPad by the nonparametric Mann-Whitney U test. InStat version 3.06 (GraphPad Software) was performed with the nonpara- metric Mann-Whitney U test. Results The relative size of granule populations. We estimated the degree of TAT-Cdc42 transduction in human neutrophils and HL60 cells accumulation of granules by measuring the area of the granule populations ␥ with the assumption that the area corresponds to the number of accumu- To study the role of Cdc42 during Fc R-mediated phagocytosis, human neutrophils or DMSO-differentiated promyelocytic human lated granules. The relative size (rs) of granule populations in the cell was measured as the ratio between the area of the largest image region and the leukemia HL60 cells were transduced with constitutively active 7360 Cdc42 MODULATES PHAGOLYSOSOME FUSION

Table I. The effect of TAT-Cdc42 on the cell area of HL60 cells

Mean Cell Maximum Samplea Sizeb Ϯ SEM Cell Sizeb

Nontransducedc 3,150 Ϯ 130 9,000 TAT-V12Cdc42d 4,400e Ϯ 370 22,450 TAT-N17Cdc42d 3,600f Ϯ 260 10,800

a Adherent HL60 cells, nontransduced or transduced with 200 nM TAT- V12Cdc42 or 200 nM TAT-N17Cdc42, were extracted by Triton X-100, fixed in 2.5% glutaraldehyde, and prepared for scanning electron microscopy by dehydration, critical point drying, and tungsten coating. Scanning electron micrographs were taken in a JEOL 840 scanning electron microscope. b Expressed in pixels. c The cell area of 140 cells per treatment was measured. d The cell area of 70 cells per treatment was measured. e p Ͻ 0.05 (compared with nontransduced cells). f p ϭ 0.16, NS (compared with nontransduced cells).

trypan blue staining (not shown). Western blot analysis revealed Downloaded from that only traces of TAT-V12Cdc42 or TAT-N17Cdc42 remained in human neutrophils after short transduction times (5–10 min) and that longer treatment further reduced the amount of protein (data not shown). A HA-tagged peptide of lower molecular mass could be detected after 30 min of TAT-transduction (not shown), indi- cating that the TAT-Cdc42 protein was subject to proteolytic deg- http://www.jimmunol.org/ radation in these cells. Previous studies have shown that the pre- treatment of cells with chloroquine, a drug that elevates lysosomal pH and modulates degranulation, protects TAT proteins from deg- radation (27, 28). However, treatment of neutrophils with 100 ␮M chloroquine before transduction with TAT-Cdc42 did not protect the protein from degradation (not shown). FIGURE 1. Structure, intracellular stability, and effect of TAT-Cdc42 on the morphology of HL60 cells. A, Structure of the TAT-Cdc42. At its N ter- In HL60 cells, by contrast, TAT-V12Cdc42 and TAT- minus the construct carries the TAT protein transduction domain, which is N17Cdc42 were stable during an extended incubation after trans-

composed of 11 aa followed by the 6His tag used for purification of the re- duction (15, 30, and 60 min) (Fig. 1B). The efficacy of the TAT- by guest on September 27, 2021 combinant protein. For efficient detection of the protein in cell lysates, a HA transduction was found to be 65% or more as determined by epitope is inserted adjacent to Cdc42. In this study, TAT-V12Cdc42 (consti- intracellular staining due to the transduction of FITC-conjugated tutively active) and TAT-N17Cdc42 (dominant negative) were used. B, HL60 anti-HA Abs bound to the HA-tagged TAT-proteins. DMSO-dif- cells resuspended in KRG without calcium were incubated with buffer, ferentiated HL60 cells were therefore used throughout the study. i.e., nontransduced (control), 200 nM TAT-V12Cdc42, or 200 nM TAT- N17Cdc42. The cells were washed twice, resuspended in KRG, and incubated for additional time as indicated at room temperature before lysis. All samples were subjected to SDS-PAGE and Western blotting and the TAT proteins Table II. Phagocytosis of IgG-opsonized yeast particles in TAT-Cdc42 were detected by means of ECL using an Ab directed toward the HA epitope transduced HL60 cells on the TAT-Cdc42 proteins. Purified TAT-V12Cdc42 (0.4 ␮g) was run in parallel (far right lane). The blot is a representative of three separate experi- b c ments. C, HL60 cells were treated with buffer (Con), 200 nM TAT-V12Cdc42 Cells in Suspension Adherent Cells (V12), or 200 nM TAT-N17Cdc42 (N17) for 30 min. The cells were allowed Associated Ingested Associated Ingested to settle on glass slides for 15 min, fixed in 4% PFA, and stained with Alexa Yeast Yeast Yeast Yeast 594-conjugated phalloidin. Fluorescence microscopy images of migrating, Samplea Particles Particles Particles Particles nonphagocytosing cells were captured with a Nikon Eclipse 800 fluorescence Nontransduced 211 Ϯ 23 104 Ϯ 18 68 62 microscope equipped with Easy Image software. The images are representa- TAT-V12Cdc42 212 Ϯ 29d 119 Ϯ 34e 58 54 ␮ tives of three separate experiments. Scale bar, 10 m. D, HL60 cells treated TAT-N17Cdc42 173 Ϯ 18f 84 Ϯ 11g 65 61 with buffer (Con), 200 nM TAT-V12Cdc42 (V12), or 200 nM TAT- a N17Cdc42 (N17) were allowed to settle on human serum albumin-coated glass Differentiated HL60 cells were incubated with a buffer or transduced with 200 nM TAT-V12Cdc42 or TAT-N17Cdc42 for 30 minutes at 37°C. The cells were kept coverslips for 30 min. The cells were extracted by Triton X-100, fixed in 2.5% in suspension or allowed to adhere on glass slides. FITC-conjugated, IgG-opsonized glutaraldehyde, and prepared for scanning electron microscopy by dehydra- yeast particles were added to the cells (ratio 5:1). Phagocytosis was stopped by ice- tion, critical point drying, and tungsten coating. Scanning electron micrographs cold PBS after 15 minutes of incubation at 37°C for the adherent cells and after 30 were taken in a JEOL 840 scanning electron microscope. The cell area minutes for cells in suspension. The extracellular fluorescence was quenched by trypan blue. For each experiment, bound and ingested yeast particles were counted in of 140 (Con) and 70 (V12 and N17) cells per treatment was measured, 100 cells per treatment. and the mean cell area for each treatment group was assessed. The cells b Expressed as mean Ϯ SEM, n ϭ 6. shown are representatives for the size of the mean cell area of each c Expressed as mean, n ϭ 2, performed in duplicate. d group. Scale bar represents 1 ␮m. p ϭ 0.98, NS (compared with associated yeast particles in nontransduced cells in suspension). e p ϭ 0.38, NS (compared with ingested yeast particles in nontransduced cells in suspension). (V12) or dominant negative (N17) forms of Cdc42 fused to TAT f p ϭ 0.18, NS (compared with associated yeast particles in nontransduced cells in suspension). (Fig. 1A) (15). The TAT transduction did not influence the viabil- g p ϭ 0.14, NS (compared with ingested yeast particles in nontransduced cells in ity of either human neutrophils or HL60 cells as determined by suspension). The Journal of Immunology 7361

FIGURE 2. F-actin staining of TAT-Cdc42-transduced HL60 cells. HL60 cells were treated (30 min) as follows. Buffer, i.e., nontransduced (A), 200 nM TAT-V12Cdc42 (B), or 200 nM TAT-N17Cdc42 (C). After washing, HL60 cells were resuspended in KRG and seeded on glass slides precovered with IgG-opsonized yeast particles. Phagocytosis was terminated after 15 min by fixation in 4% PFA, and staining with Alexa 594-conjugated phalloidin. Fluorescence microscopy images were captured with a Nikon Eclipse 800 fluorescence microscope equipped with Easy Image software. The images shown are representative of three independent experiments. Scale bar, 10 ␮m.

Cytoskeleton rearrangements in TAT-Cdc42-transduced HL60 cells nontransduced cells ( p Ͻ 0.05; Table I) and displayed a rich mesh Downloaded from TAT-Cdc42-treated HL60 cells were allowed to adhere, migrate of cytoskeletal filaments (Fig. 1D). Cells transduced with TAT- ϭ toward, and engulf IgG-opsonized yeast particles on glass slides as N17Cdc42 were of the same size as nontransduced cells ( p described previously (22). Initially, the effect of Cdc42 on F-actin 0.16, NS; Table I), but with a less abundant peripheral cytoskeletal reorganization was analyzed in nonphagocytosing migrating cells network (Fig. 1D). (Fig. 1C). It is well established that Cdc42 controls cell polarity

Effects of Cdc42 on IgG-mediated phagocytosis http://www.jimmunol.org/ and the remodeling of F-actin at the leading edge of moving cells (for review, see Refs, 29 and 30). In line with this, TAT- The reorganization of actin into F-actin-rich pseudopodia and ␥ V12Cdc42-transduced HL60 cells displayed a pronounced accu- phagocytic cups is a prerequisite for Fc R-mediated phagocytosis mulation of F-actin at the leading edge, whereas cells treated with in human neutrophils (24, 31). Cdc42 has, in other types of cells, TAT-N17Cdc42 displayed lower levels of cortical and cytoplas- been suggested to play a role in this process (32, 33). The effect of mic F-actin compared with nontransduced cells (i.e., control) (Fig. TAT-V12Cdc42 and TAT-N17Cdc42 on IgG-mediated phagocy- 1C). Scanning electron micrographs supported these findings by tosis was assayed both in cells in suspension and in adherent cells revealing altered cytoskeletal organization and significant differ-

ences in cell size between nontransduced and TAT-V12Cdc42- by guest on September 27, 2021 transduced HL60 cells. The TAT-V12Cdc42-transduced cells were found to be flattened, spread out, and larger in size compared with

FIGURE 3. Effect of TAT-Cdc42 on the translocation of CD63 to phagosomes in HL60 cells. HL60 cells were incubated for 30 min with buffer, i.e., nontransduced (Con) or transduced with 200 nM TAT- V12Cdc42 (V12) or 200 nM TAT-N17Cdc42 (N17). After washing, HL60 cells were resuspended in KRG and seeded on glass slides precovered with IgG-opsonized yeast. Phagocytosis was terminated by fixation and the dis- FIGURE 4. Costaining of F-actin and CD63 in HL60 cells containing tribution of CD63 was visualized by immunofluorescence. The phago- phagosomes. HL60 cells were incubated for 30 min with buffer, i.e., non- somes were identified by the presence of intracellular yeast particles, which transduced (Con) or transduced with 200 nM TAT-V12Cdc42 (V12) or are autofluorescent when excited with light passed through a 488-nm filter 200 nM TAT-N17Cdc42 (N17). After washing, HL60 cells were resus- before confocal images of the CD63-staining were captured. At least 50 pended in KRG and seeded on glass slides precovered with IgG-opsonized phagosomes per preparation were examined for each separate experiment yeast. Phagocytosis was terminated by fixing the cells. The cells were and the number of phagosomes positive for CD63-staining was determined. stained with an Ab directed toward CD63 followed by a secondary Al- The data from three independent experiments (Exp1, Exp2, and Exp3) are exa594-conjugated Ab and Bodipy-phalloidin. Images were captured by expressed as the percentage of phagosomes positive for CD63 staining. ␹2 confocal laser scanning microscopy. Shown are representative images from analysis was used to determine the significance for V12 and N17, respec- one experiment of three independent experiments. The phagosome is in- tively, compared with control. dicated with an arrow. Scale bar, 10 ␮m. 7362 Cdc42 MODULATES PHAGOLYSOSOME FUSION Downloaded from http://www.jimmunol.org/ FIGURE 5. The effect of TAT-Cdc42 on lysosome mobility in LysoTracker-loaded HL60 cells. HL60 cells were incubated for 30 min with buffer, i.e., nontransduced (I) or transduced with 200 nM TAT-V12Cdc42 (II) or 200 nM TAT-N17Cdc42 (III) while adhering to the bottoms of Delta T dishes. During the last 15 min of incubation LysoTracker Red DND-99 was present. The cells were washed and equilibrated with KRG at 37°C after mounting the dish in the temperature setting unit of a confocal microscope. After adding yeast particles, phagocytosing cells were identified and images were captured every fourth second. The position of each yeast particle was localized by phase contrast microscopy before and after scanning the series. A, Shown are representative images from 20 recorded cells per treatment obtained at two independent experiments. Phagosomes are indicated with an asterisk. Scale bar, 10 ␮m. Movies showing the whole sequences are available as supplementary material (Video 1–3). B, The images were subjected to quantitative image analysis (ImageJ software) to objectively determine the degree of accumulation of granules around the phagosome. For each cell, the size of granule

population was established as the area of the maximum population of LysoTracker-dyed lysosomes in relation to the total area of granules. The distance by guest on September 27, 2021 between the granules and the phagosome was established as the “closeness” to the phagosome of the populations of LysoTracker-dyed lysosomes within the cell. Thus, a high value of “closeness” corresponds to small distances between granules and the phagosome. The data shown are from representative images of six recorded cells per treatment. C, The images were subjected to quantitative image analysis to objectively determine the cellular distribution of granules. For each cell, the number of populations of LysoTracker-dyed lysosomes was established. The spatial distribution of granules was established as the RSD of LysoTracker-dyed lysosome distances within the cell. Thus, a high RSD value corresponds to a more scattered granule distribution. The data shown are from representative images of six recorded cells per treatment. using fluorescent yeast particles opsonized with IgG (Table II). No Effect of Cdc42 on phagosomal maturation significant changes in phagocytosis were observed in TAT- The disassembly of periphagosomal F-actin, which occurs shortly V12Cdc42- or TAT-N17Cdc42-transduced cells when compared after ingestion, is crucial for phagosomal maturation (10, 11, 13, with nontransduced cells (Table II). 14). Thus, the sustained activity of Cdc42, resulting in accumu- Accumulation of F-actin around phagosomes in TAT-Cdc42- lated F-actin, may influence the phagosomal maturation process. transduced HL60 cells We investigated the effect of Cdc42 on phagosomal maturation by studying the translocation of the azurophilic granule marker CD63 To understand how Cdc42 affects cellular morphology and the to the phagosomal membrane. Phagocytosis was performed by al- distribution of F-actin around the phagosome during ingestion of lowing HL60 cells to engulf IgG-opsonized yeast particles on glass IgG-opsonized particles, cells were allowed to adhere and engulf slides before staining the cells with an anti-CD63 Ab. Fusion of the yeast particles on glass slides. In nontransduced cells, we observed azurophilic granule with the phagosome was identified as CD63 phagosomes surrounded by various amounts of F-actin (Fig. 2A), staining of the phagosomal membrane in images obtained by con- reflecting the presence of newly formed phagosomes (i.e., with a focal microscopy. In randomly chosen cells, both TAT-V12Cdc42 broad F-actin rim) as well as older phagosomes (i.e., with less or and TAT-N17Cdc42 significantly reduced the number of CD63- no F-actin accumulation) (10). Transduction of the cells with TAT- positive phagosomes (40 and 30% inhibition, respectively; Fig. 3). V12Cdc42 resulted in a strong accumulation of F-actin around the phagosomes (Fig. 2B). In fact, no phagosomes that lacked a dis- tinct ring of actin were observed in these cells. When evaluating 50 V12Cdc42 and phagosomal maturation phagosomes in TAT-V12Cdc42-transduced cells, 75% were clas- Our observation that periphagosomal F-actin accumulated upon sified as having a strong accumulation of F-actin. The distribution Cdc42 activation (Fig. 2) together with our previous studies show- of F-actin around the phagosomes in cells transduced with TAT- ing a requirement for the disassembly of periphagosomal F-actin N17Cdc42 (Fig. 2C) was comparable to that found in nontrans- during phagosomal maturation (10, 11, 13, 14), prompted us to duced cells (i.e., 28 and 23% of 50 phagosomes, respectively, were investigate whether any correlation exists between the presence of classified as having strong F-actin accumulation). a pronounced coat of F-actin and the absence of CD63 staining at The Journal of Immunology 7363 the phagosomes (Fig. 3). Fig. 4 shows confocal microscopy images of cells costained with Bodipy-phalloidin and anti-CD63 Abs. Nontransduced cells with no detectable F-actin around the phago- some displayed a pronounced translocation of CD63 to the phago- some (Fig. 4). In contrast, the accumulation of F-actin around the phagosomes of TAT-V12Cdc42-transduced cells coincided with the absence of CD63 translocation to the phagosomal membrane (Fig. 4). In these cells, CD63-positive granules were found outside but in close vicinity of the F-actin ring or accumulating at the cell periphery (Fig. 4). None of the investigated TAT-V12Cdc42-trans- duced cells displayed costaining of F-actin and CD63, nor was any staining for CD63 observed inside the F-actin rim. In cells without prey, the cortical F-actin at the leading edge effectively excluded all CD63-positive granules (data not shown). To further analyze the role of Cdc42 in regulating the mobility of granules in HL60 cells, TAT-V12Cdc42-cells were loaded with LysoTracker and movies observing movement of lysosomes dur- ing phagocytosis of IgG-opsonized yeast were obtained. Analysis of the movies revealed that, in nontransduced cells, lysosomes Downloaded from were present at the site of ingestion at an early stage and that continuous fusion of LysoTracker-positive granules with the phagosome occurred (Fig. 5AI and Video 1).5 These data were FIGURE 6. The effect of TAT-Cdc42 on the phosphorylation of p38 supported by quantitative image analysis showing that lysosomes MAPK. HL60 cells were incubated for 30 min with buffer (control) or accumulated close to each other in the vicinity of the phagosome transduced with 200 nM of TAT-V12Cdc42 or TAT-N17Cdc42. The cells (Fig. 5, B and C). The granules in V12Cdc42-transduced cells were washed, resuspended in KRG, and preincubated for 1 min at 37°C. http://www.jimmunol.org/ accumulated at the cell periphery or localized close to the ingested The cells were incubated with buffer or IgG-opsonized yeast particles (ratio particle; however, no fusion could be observed (Fig. 5AII and 1:5) for the indicated times and lysed in lysis buffer. A, Samples were Video 2). Quantitative image analysis revealed that the Lyso- analyzed by SDS-PAGE and immunoblotted with an Ab directed against Tracker-positive granules accumulated near each other and there phosphorylated p38 MAPK (upper panel) or an anti-p38 MAPK Ab, i.e., was no significant difference in relative size, number of granule total p38 MAPK (lower panel). The blot shown is a representative from populations, “closeness” to a phagosome, or spatial distribution four independent experiments. B, The blots were scanned and the relative compared with granules in nontransduced cells (Fig. 5, B and C). intensity of each band was quantified using Scion Image Alpha version 4.0.3.2. Filled bars show the results from cells incubated with buffer and

open bars show the results from cells incubated with IgG-opsonized yeast by guest on September 27, 2021 N17Cdc42 and phagosomal maturation particles. The bars represent the mean Ϯ SEM for four independent ex- In N17Cdc42-transduced cells, confocal microscopy images of periments (data shown in bars 1 and 4–6, numbered from the left), and the cells costained with Bodipy-phalloidin and anti-CD63 Abs re- mean for two independent experiments (data shown in bars 2 and 3). A ;p Ͻ 0.05 ,ء) vealed that the CD63-positive granules were evenly dispersed in Mann-Whitney U test was used to determine significance ϭ the cytoplasm with no obvious accumulation around or transloca- ns, p 0.48). tion to the phagosomal membrane (Fig. 4). This was the case even if the phagosome was lacking its rim of F-actin (not shown). The via inhibition of p38 MAPK. To test whether Cdc42 influences p38 observation of a dysregulated granule translocation in TAT- MAPK during Fc␥R-mediated phagocytosis, TAT-V12Cdc42- or N17Cdc42-transduced cells was supported by our findings from TAT-N17Cdc42-transduced cells were allowed to phagocytose the movies of granule movement in living cells. In contrast to IgG-opsonized yeast. Fig. 6 shows that phagocytosis of IgG-op- nontransduced cells, where continuous fusion of LysoTracker-pos- sonized yeast induced a marked phosphorylation of p38 MAPK itive granules with the phagosome occurred (Fig. 5AI and Video within 15 min ( p Ͻ 0.05). The phosphorylation was also enhanced 1), the granules in N17Cdc42-transduced cells remained dispersed in cells transduced with V12Cdc42. Moreover, V12Cdc42 induced in the cytoplasm and no fusion with the phagosome was detected p38 MAPK phosphorylation in the absence of IgG-opsonized (Fig. 5AIII and Video 3). Quantitative image analysis supported yeast. N17Cdc42 did not, however, prevent the IgG-induced p38 these data by showing that in N17Cdc42-transduced cells the MAPK phosphorylation and had no effect on the phosphorylation LysoTracker-positive granules did not accumulate around the of p38 MAPK per se (Fig. 6). phagosome ( p Ͻ 0.01 compared with nontransduced cells; Fig. 5B). The granules were significantly more dispersed ( p Ͻ 0.01, relative size; p Ͻ 0.05, number of granule populations; Fig. 5, B Discussion and C) and had a more scattered distribution ( p Ͻ 0.01; Fig. 5C) The present study demonstrates that the inactivation of Cdc42 at compared with granules in nontransduced cells. the phagosomal membrane is a prerequisite for the depolymeriza- The finding that dominant negative Cdc42 results in reduced tion of F-actin around the phagosome. Efficient depolymerization mobilization of granules suggests that Cdc42 is part of the signal of the periphagosomal F-actin is necessary for the subsequent pha- transduction involved in the regulation of granule trafficking. p38 gosomal maturation (10, 11, 13, 14, 35). Consistent with this, im- MAPK was recently shown to be involved in phagosomal matu- paired phagosomal maturation was observed in cells with perma- ration and degranulation (34). We therefore hypothesized that nently activated Cdc42. This inhibition was not due to impaired dominant negative Cdc42 could block the translocation of CD63 granule movement, because CD63-positive granules accumulated in close proximity to the periphagosomal rim of F-actin. The ob- servation that CD63 staining was completely excluded from areas 5 The online version of this article contains supplemental material. with F-actin staining (both cortical and periphagosomal F-actin) 7364 Cdc42 MODULATES PHAGOLYSOSOME FUSION shows that the inhibition of phagosomal maturation by active Cdc42 is due to a barrier of F-actin meshwork that prevents the granules from reaching the phagosome. In fact, F-actin and lyso- somal markers are never observed simultaneously on the phago- somal membrane in macrophages and fibroblasts (36, 37), suggest- ing that the disassembly of periphagosomal F-actin is a general requirement for granule fusion with target membranes. The im- portance of the depolymerization of periphagosomal F-actin is re- flected by the fact that multiple signaling molecules are involved in FIGURE 7. Model showing how constitutively active (V12) and dom- regulating this process, including intracellular Ca2ϩ (10, 11) and inant negative (N17) Cdc42 prevent phagosomal maturation. In the pres- protein kinase C␣ (13, 38). Depletion of Ca2ϩ or overexpression of ence of persistent activation of Cdc42, phagosomal maturation is blocked dominant negative protein kinase C␣ results in a phenotype similar by the dense meshwork of F-actin around the phagosomes that physically prevents the CD63 positive granule from fusing with the phagosomal mem- to that observed with permanently active Cdc42 in the present brane. The granules are still mobile and accumulate outside the F-actin study. cage or at the cell periphery. In cells with dominant negative Cdc42, the Activation of Cdc42 and Rac is a prerequisite for the uptake of F-actin around the phagosome can dissolve and constitutes no hindrance. particles by macrophages (39, 40). In our study, no significant These cells have, however, an impaired granule movement toward the site reduction of phagocytosis was observed in neutrophil-like HL60 of the ingested particle, and the CD63-positive granules are found evenly cells transduced with dominant negative Cdc42. In line with this dispersed in the cytoplasm. Phagosomal maturation fails due to the lack of observation, we and others have reported that in neutrophils the translocation and the accumulation of granules at the phagosomal mem- Downloaded from impaired function of Rho GTPases has no major effect on the brane. In this illustration, the nucleus is depicted as gray; red shows the phagocytosis of IgG- or C3bi-opsonized particles (6, 8, 9). An distribution of CD63 in the cell, whereas green represents F-actin. alternative explanation is that the loss of Cdc42-induced function may be compensated by endogenous Rac to restore phagocytic capacity. Cdc42 and Rac are activated with the same time kinetics center. These two notions regarding the mechanism involved in during the phagocytosis of IgG-opsonized particles in neutrophils Cdc42-regulated translocation of granules will be subject to future http://www.jimmunol.org/ (6), and both of their activities are required to coordinate F-actin studies. accumulation at nascent phagosomes in macrophages (41). Degranulation in neutrophils and macrophages is regulated by It is well known that many pathogens are able to manipulate the p38 MAPK (53, 54), and it was recently suggested that p38 MAPK functions of the actin cytoskeleton in host cells to avoid the host enhances phagosomal maturation by increasing granule mobility immune response (42–45). Internalized Salmonella reside in F- via Rab (34). We therefore speculated that p38 MAPK could be actin-coated phagosomes, thereby escaping elimination by macro- part of the signal transduction involved in the Cdc42-regulated phages (12). Moreover, the intracellular parasite Leishmania do- translocation of granules. The Fc␥R-mediated phagocytosis per se novani secures its intracellular survival by forming a coat of was found to trigger p38 MAPK phosphorylation. However, the by guest on September 27, 2021 F-actin (13, 14) in a Cdc42- and Rac1-dependent manner (46). In reduced mobility of granules observed after the inhibition of line with this, Cdc42 together with proteins required for F-actin as- Cdc42 must involve an alternative signaling pathway, because the sembly (i.e., Arp2/3, Wiskott-Aldrich syndrome protein (WASP), phosphorylation of p38 MAPK by IgG-opsonized yeast could still ␣ myosin, and -actinin) is retained on phagosomes containing virulent be observed in N17Cdc42-transduced cells. An alternative possi- L. donovani but not on the phagosomes of an attenuated strain that is bility is that Cdc42 acts downstream of p38 MAPK. The absence incapable of intracellular survival (47). It is therefore tempting to of N17Cdc42-mediated effects might also be explained by an in- speculate that the accumulation of F-actin around the phagosome by sufficient amount of dominant negative Cdc42 to significantly af- the manipulation of Cdc42 function is a common pathogenic trait fect the p38 MAPK activity, because the amount of modified pro- shared by different intracellular pathogens. teins in the cell subjected to TAT transduction is considerably In cells transduced with dominant negative Cdc42, CD63 gran- lower than that found after traditional cell transfection. ules were evenly dispersed in the cytoplasm with no obvious ac- Under normal conditions, the periphagosomal rim of F-actin is cumulation at or translocation to the phagosomal membrane. This depolymerized shortly after particle uptake (11), enabling phagoly- observation suggests that Cdc42 is part of the signal transduction sosome fusion. For phagosomal maturation to take place, it is involved in regulating granule trafficking. In line with this obser- therefore essential that F-actin remodeling is precisely regulated. vation, neutrophils with impaired Rac and Cdc42 function exhib- The present study shows a dual role of Cdc42 in regulating pha- ited a reduced ability to perform phagolysosome fusion and oxi- gosomal maturation as summarized in the model proposed in Fig. dative activation (6, 8, 9). An attractive hypothesis for our finding 7. In the presence of active Cdc42, phagosomal maturation is is that the azurophilic granules move through the cytoplasm in a blocked by the dense meshwork of F-actin around the phagosomes, Listeria-like, F-actin-dependent, and Cdc42-dependent manner. which physically obstructs the granule from fusing with the pha- This hypothesis is supported by two recent studies that reported gosomal membrane. The inhibition of Cdc42 also prevents pha- actin-dependent movement of lysosomes, endosomes, and phago- gosomal maturation by modulating the granule translocation per somes involving neural Wiskott-Aldrich syndrome protein (N- se, most probably by being part in the signal transduction involved WASP) and Arp2/3 (48, 49), which are downstream effectors of in regulating granule trafficking to the phagosomal membrane. Cdc42 (50). Further support for this theory comes from a study on fibroblasts showing that actin polymerization is required for the Acknowledgments fusion of endocytic and phagocytic pathways (36). An alternative We thank Pia Druid, Kristina Orselius, and Maria Gustavsson for excellent explanation to our findings is that the microtubule organizing cen- technical assistance, Dr. Margaretha Lindroth for advice regarding prepa- ter, which has been shown to be regulated by Cdc42 in different rations for scanning electron microscopy, and Patrik Lindblom, M.Sc. En- cell types (51, 52), is required for the efficient translocation of a gineering, for setting up and performing the quantitative image analysis. granule toward the phagosome and that impairment of Cdc42 func- Dr. Steve Dowdy is thanked for providing the pTAT-HA constructs and tion using N17Cdc42 results in a defective microtubule organizing Dr. Dirk Roos for the anti-CD63 Ab. The Journal of Immunology 7365

Disclosures 27. Han, H., M. Fuortes, and C. Nathan. 2003. Critical role of the carboxyl terminus of proline-rich tyrosine kinase (Pyk2) in the activation of human neutrophils by The authors have no financial conflict of interest. tumor necrosis factor: separation of signals for the respiratory burst and degran- ulation. J. Exp. Med. 197: 63–75. 28. Frankel, A. D., and C. O. Pabo. 1988. Cellular uptake of the tat protein from References human immunodeficiency virus. Cell 55: 1189–1193. 1. Underhill, D. M., and A. Ozinsky. 2002. Phagocytosis of microbes: complexity 29. Fukata, M., M. Nakagawa, and K. Kaibuchi. 2003. Roles of Rho-family GTPases in action. Annu. Rev. Immunol. 20: 825–852. in cell polarisation and directional migration. Curr. Opin. Cell. Biol. 15: 2. Aderem, A., and D. M. Underhill. 1999. Mechanisms of phagocytosis in mac- 590–597. rophages. Annu. Rev. Immunol. 17: 593–623. 30. Johnson, D. I. 1999. Cdc42: An essential Rho-type GTPase controlling eukary- 3. Etienne-Manneville, S., and A. Hall. 2002. Rho GTPases in cell biology. Nature otic cell polarity. Microbiol. Mol. Biol. Rev. 63: 54–105. 420: 629–635. 31. Ydrenius, L., M. Majeed, B. J. Rasmusson, O. Stendahl, and E. Sarndahl. 2000. 4. Caron, E., and A. Hall. 1998. Identification of two distinct mechanisms of phago- Activation of cAMP-dependent protein kinase is necessary for actin rearrange- cytosis controlled by different Rho GTPases. Science 282: 1717–1721. ments in human neutrophils during phagocytosis. J. Leuko.c Biol. 67: 520–528. 5. Hoppe, A. D., and J. A. Swanson. 2004. Cdc42, Rac1, and Rac2 display distinct 32. Castellano, F., P. Montcourrier, J. C. Guillemot, E. Gouin, L. Machesky, patterns of activation during phagocytosis. Mol. Biol. Cell. 15: 3509–3519. P. Cossart, and P. Chavrier. 1999. Inducible recruitment of Cdc42 or WASP to a 6. Forsberg, M., P. Druid, L. Zheng, O. Stendahl, and E. Sarndahl. 2003. Activation cell-surface receptor triggers actin polymerization and filopodium formation. of Rac2 and Cdc42 on Fc and complement receptor ligation in human neutro- Curr. Biol. 9: 351–360. phils. J. Leukocyte Biol. 74: 611–619. 33. May, R. C., E. Caron, A. Hall, and L. M. Machesky. 2000. Involvement of the 7. Koh, A. L., C. X. Sun, F. Zhu, and M. Glogauer. 2005. The role of Rac1 and Rac2 Arp2/3 complex in phagocytosis mediated by Fc␥R or CR3. Nat. Cell. Biol. 2: in bacterial killing. Cell Immunol. 235: 92–97. 246–248. 8. Ambruso, D. R., C. Knall, A. N. Abell, J. Panepinto, A. Kurkchubasche, 34. Blander, J. M., and R. Medzhitov. 2004. Regulation of phagosome maturation by G. Thurman, C. Gonzalez-Aller, A. Hiester, M. deBoer, R. J. Harbeck, et al. signals from toll-like receptors. Science 304: 1014–1018. 2000. Human neutrophil immunodeficiency syndrome is associated with an in- 35. Allen, L. H., and A. Aderem. 1995. A role for MARCKS, the ␣ isozyme of hibitory Rac2 mutation. Proc. Natl. Acad. Sci. USA 97: 4654–4659. protein kinase C and myosin I in zymosan phagocytosis by macrophages. J. Exp. 9. Williams, D. A., W. Tao, F. Yang, C. Kim, Y. Gu, P. Mansfield, J. E. Levine, Med. 182: 829–840. Downloaded from B. Petryniak, C. W. Derrow, C. Harris, et al. 2000. Dominant negative mutation 36. Arora, P. D., M. F. Manolson, G. P. Downey, J. Sodek, and C. A. McCulloch. of the hematopoietic-specific Rho GTPase, Rac2, is associated with a human 2000. A novel model system for characterization of phagosomal maturation, acid- phagocyte immunodeficiency. Blood 96: 1646–1654. ification, and intracellular collagen degradation in fibroblasts. J. Biol. Chem. 275: 10. Bengtsson, T., M. E. Jaconi, M. Gustafson, K. E. Magnusson, J. M. Theler, 35432–35441. D. P. Lew, and O. Stendahl. 1993. Actin dynamics in human neutrophils during 37. Henry, R. M., A. D. Hoppe, N. Joshi, and J. A. Swanson. 2004. The uniformity adhesion and phagocytosis is controlled by changes in intracellular free calcium. of phagosome maturation in macrophages. J. Cell Biol. 164: 185–194. Eur. J. Cell Biol. 62: 49–58. 38. Ng Yan Hing, J. D., M. Desjardins, and A. Descoteaux. 2004. Proteomic analysis ␣ 11. Jaconi, M. E., D. P. Lew, J. L. Carpentier, K. E. Magnusson, M. Sjogren, and reveals a role for protein kinase C- in phagosome maturation. Biochem. Bio- http://www.jimmunol.org/ O. Stendahl. 1990. Cytosolic free calcium elevation mediates the phagosome- phys. Res. Commun. 319: 810–816. lysosome fusion during phagocytosis in human neutrophils. J. Cell Biol. 110: 39. Massol, P., P. Montcourrier, J. C. Guillemot, and P. Chavrier. 1998. Fc receptor- 1555–1564. mediated phagocytosis requires CDC42 and Rac1. EMBO J. 17: 6219–6229. 12. Meresse, S., K. E. Unsworth, A. Habermann, G. Griffiths, F. Fang, 40. Cox, D., P. Chang, Q. Zhang, P. G. Reddy, G. M. Bokoch, and S. Greenberg. M. J. Martinez-Lorenzo, S. R. Waterman, J. P. Gorvel, and D. W. Holden. 2001. 1997. Requirements for both Rac1 and Cdc42 in membrane ruffling and phago- Remodelling of the actin cytoskeleton is essential for replication of intravacuolar cytosis in leukocytes. J. Exp. Med. 186: 1487–1494. Salmonella. Cell. Microbiol. 3: 567–577. 41. Cougoule, C., S. Hoshino, A. Dart, J. Lim, and E. Caron. 2006. Dissociation of 13. Holm, A., K. Tejle, T. Gunnarsson, K. E. Magnusson, A. Descoteaux, and recruitment and activation of the small G-protein Rac during Fc␥ receptor-me- B. Rasmusson. 2003. Role of protein kinase C␣ for uptake of unopsonized prey diated phagocytosis. J. Biol. Chem. 281: 8756–8764. and phagosomal maturation in macrophages. Biochem. Biophys. Res. Commun. 42. Fu, Y., and J. E. Galan. 1999. A salmonella protein antagonizes Rac-1 and Cdc42 302: 653–658. to mediate host-cell recovery after bacterial invasion. Nature 401: 293–297.

14. Holm, A., K. Tejle, K. E. Magnusson, A. Descoteaux, and B. Rasmusson. 2001. 43. Hardt, W. D., L. M. Chen, K. E. Schuebel, X. R. Bustelo, and J. E. Galan. 1998. by guest on September 27, 2021 Leishmania donovani lipophosphoglycan causes periphagosomal actin accumu- S. typhimurium encodes an activator of Rho GTPases that induces membrane lation: correlation with impaired translocation of PKC␣ and defective phagosome ruffling and nuclear responses in host cells. Cell 93: 815–826. maturation. Cell. Microbiol. 3: 439–447. 44. Lerm, M., G. Schmidt, and K. Aktories. 2000. Bacterial protein toxins targeting 15. Becker-Hapak, M., S. S. McAllister, and S. F. Dowdy. 2001. TAT-mediated rho GTPases. FEMS Microb. Lett. 188: 1–6. protein transduction into mammalian cells. Methods 24: 247–256. 45. Aktories, K., and H. Barth. 2004. Clostridium botulinum C2 toxin–new insights 16. Gao, B., and M. F. Tsan. 2003. Endotoxin contamination in recombinant human into the cellular up-take of the actin-ADP-ribosylating toxin. Int. J. Med. Micro- heat shock protein 70 (Hsp70) preparation is responsible for the induction of biol. 293: 557–564. tumor necrosis factor ␣ release by murine macrophages. J. Biol. Chem. 278: 46. Lerm, M., A. Holm, A. Seiron, E. Sarndahl, K. E. Magnusson, and B. Rasmusson. 174–179. 2006. Leishmania donovani requires functional Cdc42 and Rac1 to prevent pha- 17. Boyum, A. 1968. A one-stage procedure for isolation of granulocytes and lym- gosomal maturation. Infect. Immun. 74: 2613–2618. phocytes from human blood. General sedimentation properties of white blood 47. Lodge, R., and A. Descoteaux. 2005. Leishmania donovani promastigotes induce cells in a 1g gravity field. Scand. J. Clin. Lab. Invest. Suppl. 97: 51–76. periphagosomal F-actin accumulation through retention of the GTPase Cdc42. 18. Ferrante, A., and Y. H. Thong. 1980. Optimal conditions for simultaneous puri- Cell. Microbiol. 11: 1647–1658. fication of mononuclear and polymorphonuclear leucocytes from human blood by 48. Taunton, J., B. A. Rowning, M. L. Coughlin, M. Wu, R. T. Moon, T. J. the Hypaque-Ficoll method. J. Immunol. Methods 36: 109–117. Mitchison, and C. A. Larabell. 2000. Actin-dependent propulsion of endosomes 19. Lofgren, R., L. Serrander, M. Forsberg, A. Wilsson, A. Wasteson, and and lysosomes by recruitment of N-WASP. J. Cell Biol. 148: 519–530. O. Stendahl. 1999. CR3, Fc␥RIIA and Fc␥RIIIB induce activation of the respi- 49. Southwick, F. S., W. Li, F. Zhang, W. L. Zeile, and D. L. Purich. 2003. Actin- ϩ ratory burst in human neutrophils: the role of intracellular Ca2 , phospholipase based endosome and phagosome rocketing in macrophages: activation by the D and tyrosine phosphorylation. Biochim. Biophys. Acta 1452: 46–59. secretagogue antagonists lanthanum and zinc. Cell Motil. Cytoskeleton 54: 20. Patcha, V., J. Wigren, M. E. Winberg, B. Rasmusson, J. Li, and E. Sarndahl. 41–55. ␤ 2004. Differential inside-out activation of 2-integrins by leukotriene B4 and 50. Glogauer, M., J. Hartwig, and T. Stossel. 2000. Two pathways through Cdc42 fMLP in human neutrophils. Exp. Cell. Res. 300: 308–319. couple the N-formyl receptor to actin nucleation in permeabilized human neu- 21. Collins, S. J. 1987. The HL-60 promyelocytic leukemia cell line: proliferation, trophils. J. Cell Biol. 150: 785–796. differentiation, and cellular oncogene expression. Blood 70: 1233–1244. 51. Palazzo, A. F., H. L. Joseph, Y. J. Chen, D. L. Dujardin, A. S. Alberts, 22. Stendahl, O., K. H. Krause, J. Krischer, P. Jerstrom, J. M. Theler, R. A. Clark, K. K. Pfister, R. B. Vallee, and G. G. Gundersen. 2001. Cdc42, dynein, and J. L. Carpentier, and D. P. Lew. 1994. Redistribution of intracellular Ca2ϩ stores dynactin regulate MTOC reorientation independent of Rho-regulated microtubule during phagocytosis in human neutrophils. Science 265: 1439–1441. stabilization. Curr. Biol. 11: 1536–1541. 23. Stendahl, O., J. Hed, E. Kihlstrom, K. E. Magnusson, and C. Tagesson. 1977. 52. Lee, J. S., M. I. Chang, Y. Tseng, and D. Wirtz. 2005. Cdc42 mediates nucleus Phagocytic internalization and the requirement for membrane perturbation. FEBS movement and MTOC polarization in Swiss 3T3 fibroblasts under mechanical Lett. 81: 118–120. shear stress. Mol. Biol. Cell. 16: 871–880. 24. Hed, J., and O. Stendahl. 1982. Differences in the ingestion mechanisms of IgG 53. Forsberg, M., R. Lofgren, L. Zheng, and O. Stendahl. 2001. Tumour necrosis and C3b particles in phagocytosis by neutrophils. Immunology 45: 727–736. factor-␣ potentiates CR3-induced respiratory burst by activating p38 MAP kinase 25. Bell, P. B., Jr., M. Lindroth, and B. A. Fredriksson. 1988. Preparation of cy- in human neutrophils. Immunology 103: 465–472. toskeletons of cells in culture for high resolution scanning and scanning trans- 54. Mocsai, A., Z. Jakus, T. Vantus, G. Berton, C. A. Lowell, and E. Legate. 2000. mission electron microscopy. Scanning Microsc. 2: 1647–1661. Kinase pathways in chemoattractant-induced degranulation of neutrophils: the 26. Abramoff, M. D., P. J. Magelhaes, and S. J. Ram. 2004. Image processing with role of p38 mitogen-activated protein kinase activated by Src family kinases. ImageJ. Biophotonics Int. 11: 36–42. J. Immunol. 164: 4321–4331.