The Membrane Phospholipid Binding A2 Promotes Phagocytosis and Nonlytic of Cryptococcus neoformans and Impacts Survival in Fungal This information is current as Infection of September 23, 2021. Sabriya Stukes, Carolina Coelho, Johanna Rivera, Anne E. Jedlicka, Katherine A. Hajjar and Arturo Casadevall J Immunol published online 1 July 2016 http://www.jimmunol.org/content/early/2016/07/01/jimmun Downloaded from ol.1501855

Supplementary http://www.jimmunol.org/content/suppl/2016/07/01/jimmunol.150185 http://www.jimmunol.org/ Material 5.DCSupplemental

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 23, 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 © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published July 1, 2016, doi:10.4049/jimmunol.1501855 The Journal of Immunology

The Membrane Phospholipid Binding Protein Annexin A2 Promotes Phagocytosis and Nonlytic Exocytosis of Cryptococcus neoformans and Impacts Survival in Fungal Infection

Sabriya Stukes,* Carolina Coelho,*,† Johanna Rivera,* Anne E. Jedlicka,† Katherine A. Hajjar,‡,x and Arturo Casadevall*,†

Cryptococcus neoformans is a fungal pathogen with a unique intracellular pathogenic strategy that includes nonlytic exocytosis, a phenomenon whereby fungal cells are expunged from macrophages without lysing the host cell. The exact mechanism and specific

involved in this process have yet to be completely defined. Using murine macrophages deficient in the membrane Downloaded from phospholipid binding protein, annexin A2 (ANXA2), we observed a significant decrease in both phagocytosis of yeast cells and the frequency of nonlytic exocytosis. Cryptococcal cells isolated from Anxa2-deficient (Anxa22/2) bone marrow–derived macro- phages and lung parenchyma displayed significantly larger capsules than those isolated from wild-type macrophages and tissues. Concomitantly, we observed significant differences in the amount of reactive oxygen species produced between Anxa22/2 and Anxa2+/+ macrophages. Despite comparable fungal burden, Anxa22/2 mice died more rapidly than wild-type mice when infected 2/2

with C. neoformans, and Anxa2 mice exhibited enhanced inflammatory responses, suggesting that the reduced survival http://www.jimmunol.org/ reflected greater immune-mediated damage. Together, these findings suggest a role for ANXA2 in the control of cryptococcal infection, macrophage function, and fungal morphology. The Journal of Immunology, 2016, 197: 000–000.

ryptococcus neoformans is an environmental fungus with and laccase, which deactivate microbicidal compounds; and 3) the a complex intracellular pathogenic strategy that is inti- ability of the fungus to damage the phagosomal membrane allowing C mately related to its capacity for virulence (1–3). Fungal efflux of phagosomal contents into the cytoplasm (1, 8, 10–13). infection results from the inhalation of aerosolized infectious However, perhaps the most unusual aspect of the intracellular strat- particles (4). Unlike other intracellular pathogens, C. neoformans egy of this fungal pathogen is its ability to escape the macrophage

can survive and replicate inside an acidic phagosome that un- in a process known as nonlytic exocytosis (14, 15). by guest on September 23, 2021 dergoes normal maturation (5). The outcome of the interaction Nonlytic exocytosis has been described in vitro for different cell between this fungal pathogen and macrophages appears to be a types such as J774 murine macrophage-like cells, primary mouse and critical determinant of virulence (3, 6–9). human macrophages, Drosophila S2 cells, and amoebae (4, 16–19). The mechanisms responsible for fungal intracellular survival Nonlytic exocytosis was shown to occur in vivo and could provide a involve a combination of factors that include: 1) residence in a large mechanism by which yeast cells migrating from the lung in Trojan phagosome where lysosomal contents are diluted; 2) the presence of horse macrophages are released into the circulation to infect the powerful antioxidant mechanisms including a large polysaccharide brain by transcytosis of the blood–brain barrier (20, 21). This in- capsule, cell wall–associated melanin that can absorb lysosome- teraction appears to be a pathogen-driven phenomenon because it generated oxidants and enzymes, such as superoxide dismutase requires cryptococcal cell viability and is selectively confined to fungal cells when macrophages ingest both live yeast cells and *Department of Microbiology and Immunology, Albert Einstein College of Medi- FITC-labeled beads (15). The phagosome appears to play a crucial cine, Bronx, NY 10461; †Johns Hopkins Bloomberg School of Public Health, Balti- role during this process, because phagosome labeling experiments more, MD 21205; ‡Department of Pediatrics, Weill Cornell Medicine, New York, NY x suggest that during nonlytic exocytosis, the entire organelle is ex- 10065; and Department of Cell and Developmental Biology, Weill Cornell Medi- cine, New York, NY 10065 punged, allowing for the release of fungal cells into the extracellular ORCIDs: 0000-0002-7523-3031 (C.C.); 0000-0001-9332-4861 (J.R.); 0000-0003- environment (22). Nonlytic exocytosis is under the control of both 3977-4356 (K.A.H.); 0000-0002-9402-9167 (A.C.). host and fungal factors, including host actin (22), phagosomal pH Received for publication August 18, 2015. Accepted for publication June 2, 2016. (20), fungal SEC14 (23), cytokine stimulation (24), host autophagy This work was supported by National Institutes of Health Grants 5R01A1033774, (25), and the presence of a capsule on fungal cells (15). 5R37AI033142, and 5T32A107506, and Clinical and Translational Science Awards In this study, we investigated the role of the membrane binding Grants 1 ULI TR001073-01, 1 TLI 1 TR001072-01, and 1 KL2 TR001071 from the National Center for Advancing Translational Sciences. protein annexin A2 (ANXA2) during in vitro fungal interactions with macrophages and in cryptococcal pathogenesis. are a Address correspondence and reprint requests to Dr. Arturo Casadevall, Johns Hopkins Bloomberg School of Public Health, 615 North Wolfe Street, Baltimore, MD 21205. family of calcium binding proteins that function by bringing cellular E-mail address: [email protected] membranes in close contact with each other to promote fusion (26). The online version of this article contains supplemental material. They function in a wide variety of biological processes, some of Abbreviations used in this article: ANXA2, annexin A2; BMDM, bone marrow– which include membrane trafficking, phagocytosis, the endocytic derived macrophage; ROS, reactive oxygen species; SNAP, soluble N-ethylmaleimide pathway, and of interest to us, exocytosis of secretory vesicles (27– sensitive fusion attachment protein. 29). Specifically, we focused on ANXA2 because this protein me- Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 diates endosomal membrane–membrane fusion and plays a role in

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1501855 2 ANNEXIN A2 DEPLETION IN MACROPHAGES IMPACTS FUNGAL INFECTION the docking mechanism needed for vesicles to adhere to cellular Casadevall laboratory for several years. Unless otherwise specified, yeast membranes. ANXA2 is also hypothesized to function in the cells were washed three times in sterile PBS, counted, and used at an E:T membrane fusion events leading to the release of vesicles in chro- ratio of 5:1 for all in vitro experiments unless otherwise noted. maffin cells, exocytosis of lamellar bodies in alveolar epithelial C57BL/6J and Anxa22/2 mouse strains cells, and regulated exocytosis of von Willebrand factor pack- 2/2 Homozygous breeding pairs of Anxa2 knockout (Anxa2 ) mice from a aged in Weibel-Palade bodies in bovine endothelial cells (30– C57BL/6J background were obtained from the Hajjar laboratory (New 32). In addition, ANXA2 mediates secretion of vesical-bound York, NY) (36) and were maintained at the Animal Institute of Albert macromolecular collagen VI from bronchial epithelial cells (33). Einstein College of Medicine. Male and female C57BL/6J (8–10 wk old) There is minimal information linking ANXA2 and C. neoformans, mice were obtained from the National Cancer Institute (Fredrick, MD). In all experiments, mice were used at 8–10 wk of age. Mice were housed in and what has been shown mostly focuses on the interaction of fungal sterile microisolator cages in a barrier environment and were maintained cells and endothelial cells. Vu et al. showed that ANXA2 and in a specific pathogen-free barrier facility in microisolator cages, fed irra- were both upregulated upon contact and ingestion of diated rodent food, provided with autoclaved bedding, and routinely moni- fungal cells by brain endothelial cells. This upregulation translated tored for serological evidence of exposure to common murine pathogens. into a decrease in the integrity of the cellular membrane, suggesting Isolation of bone marrow–derived macrophages and peritoneal that fungal cells enhanced their transmigration into the brain by macrophages compromising the underlying structure because these 2 2 C57BL/6J and Anxa2 / mice were used for all primary macrophage binding proteins are known to interact with actin (34, 35). experiments. Femurs and tibias from both strains of mice were dissected, Because of its functionality and previous research showing a role and all muscle tissues were removed. Intact bones were disinfected by during regulated exocytosis, we hypothesized that this cellular immersion in 70% ethanol for 3 min and rinsed in PBS. Both ends of each Downloaded from membrane binding protein could be part of the machinery that bone were cut off and the cells were flushed out using cold DMEM enables nonlytic exocytosis of C. neoformans. Our results dem- (Corning Cellgro) using a 25-g needle and collected into tubes. The cell suspension was centrifuged for 5 min at 650 3 g at room temperature and onstrate that ANXA2 is involved in multiple cellular processes washed once with cold DMEM. Cell clumps were disrupted by passing the during the intracellular control of C. neoformans infection, in- solution through a 70-mm cell strainer. RBCs were removed by incubation cluding internalization of fungal cells and nonlytic exocytosis of the cell suspension in ACK lysis solution (150 mM NH4CL, 10 mM KHCO3, 0.1 mM EDTA pH 7.4) for 10 min at room temperature and from macrophages. http://www.jimmunol.org/ centrifuged to pellet cells for 5 min at 650 3 g. Cells were plated on tissue culture–treated petri dishes in growth media (DMEM supplemented with Materials and Methods 20% L929 growth conditioning media, 10% FCS [Atlanta Biologicals], Growth conditions of Cryptococcus neoformans 10% NCTC-109 [Invitrogen], 1% HEPES [Corning Cellgro], GlutaMAX [Invitrogen], penicillin-streptomycin [Life Technologies], nonessential A streak from a single colony of C. neoformans strain H99 (serotype A) amino acids [Corning Cellgro], and 0.1% 2-ME [Invitrogen]). Fresh media was grown in Sabouraud dextrose broth (Difco) for 24 h at 30˚C with were added after 3 d of growth, and media were again removed and constant agitation. This strain of C. neoformans was originally obtained replaced on day 6 of culture. Bone marrow–derived macrophages from Dr. John Perfect (Durham, NC) and has been maintained in the (BMDMs) were collected on day 7 and used for subsequent experiments. by guest on September 23, 2021

FIGURE 1. Efficiency of phagocytosis by Anxa2+/+ and Anxa22/2 murine macrophages. (A) Macrophages were stained with propidium iodide (red, nuclear stain). Fungal cells were stained with Uvitex2B (cyan, chitin binding for the fungal cell well) and analyzed as described previously (4, 16, 17, 31). Uninfected macrophages (white arrow) could be readily distinguished from macrophages that had internalized fungal cells (black arrow). Original magnification 340. (B) Percent phagocytosis was determined by calculating the number of macrophages that had internalized C. neoformans divided by the total number of macrophages counted. The data are shown as mean 6 SEM with the statistical significance indicated by ***p , 0.0001 or **p , 0.001 using a Student two-tailed t test. (C and D) Number of C. neoformans cells ingested by macrophages at (C) 30 min and (D) 2 h. Each bar represents 25 fields of cells counted in 5 separate wells with an average of 1000 cells counted per well. Each experiment was performed three times. The data are shown as mean 6 SEM with the statistical significance indicated by ****p , 0.0001 or **p , 0.001 using Student two-tailed t test. The Journal of Immunology 3

fungal cells were added at an E:T ratio of 2:1. Cells were allowed to sit on ice for 15 min to allow settling of cryptococcal cells. Warm media was then added and phagocytosis was allowed to occur at two time in- tervals,0.5and2h,at37˚Cin10%CO2.Ateachtimeinterval,extra- cellular fungal cells were removed and macrophages were washed three times with PBS, fixed with cold methanol for 30 min at 220˚C, and washed two times with sterile PBS before adding appropriate dyes as detailed previously (38). In brief, wheat germ agglutinin conjugated to Alexa 633 (Invitrogen) was added at 10 mg/ml and incubated overnight at 4˚C. Uvitex 2B (Polysciences, Warrington, PA) was then added at 0.1 mg/ml and incubated for 1 min at room temperature. Subsequently, propidium iodide (Sigma-Aldrich, St. Louis, MO) was added at 5 mg/ml for a total volume of 400 ml/well and analyzed in the propidium iodide solution. Nonlytic exocytosis assay Nonlytic exocytosis assays were carried out as previously described (16). In brief, primary BMDMs were plated on MatTek glass petri dishes and allowed to adhere overnight in growth media supplemented with 10 mgof LPS and 50 U/ml IFN-g. The next day, cells were washed, and live cryptococcal cells opsonized with 10 mg/ml mAb 18b7 were added to each FIGURE 2. Lytic exocytosis is increased, whereas nonlytic exocytosis is dish. After a coinfection period of 2 h, cells were washed thoroughly with Downloaded from decreased in Anxa22/2 macrophages. Primary macrophages were infected PBS and fresh activating media without mAb 18B7 was added to the dish. with fungal cells for 2 h, after which extracellular fungal cells were washed A time-lapse video was set up to acquire images every 4 min for 24 h using away and video acquisition was initiated with images taken at 4-min in- a Zeiss Axiovert 200 M inverted microscope in an enclosed chamber with conditions at 5% CO at 37˚C. tervals for a total of 24 h. Events were placed into two categories: cellular 2 lysis and nonlytic exocytosis. Approximately 100–300 events were counted Survival studies , from five experiments with the statistical significance indicated by *p 0.04 2/2 , C57BL/6J and Anxa2 mice aged 8–10 wk were infected intratracheally or **p 0.008 using Student two-tailed t test. http://www.jimmunol.org/ as previously described (39) with 105 or 106 C. neoformans. Mice were monitored daily for signs of stress and deterioration of health throughout To induce peritoneal macrophage recruitment, we injected mice with 3% the experiment. sodium thioglycolate 3 d before isolation. Peritoneal macrophages were isolated by creating a small incision into the peritoneal cavity and aspirating Capsule growth measurement in vitro and in vivo with 10 ml cold PBS supplemented with EDTA (1 mM) and 1% penicillin/ For in vitro experiments, C57BL/6J and Anxa22/2 BMDMs were plated streptomycin (Life Technologies). Cell suspensions were kept cold and in 100-mm petri dishes, infected with opsonized C. neoformans as de- 3 g pelleted via centrifugation at 650 for 10 min. Macrophages were plated scribed for nonlytic exocytosis assays, and incubated at 37˚C for 24 h. at a density of 106 cells/ml and allowed to adhere overnight in DMEM Extracellular fungal cells were washed away, macrophages lysed with (Corning Cellgro) supplemented with 10% FCS (Atlanta Biologicals), 10% water, and cryptococcal cells collected by centrifugation. Capsule size NCTC-109 (Invitrogen), 1% penicillin-streptomycin (Life Technologies), wasvisualizedusingIndiainkandanalyzedwithanOlympusAX70 by guest on September 23, 2021 and nonessential amino acids (Corning Cellgro). microscope at 403 magnification. Individual cells were measured using Phagocytosis assay ImageJ software. For each cell, the diameter of the cell body was sub- tracted from the diameter of the entire cell to determine the capsule BMDMs were plated in triplicate on glass-bottom 96-well Matrical plate radial dimension. For in vivo measurements, mice were infected as de- at a density of 5 3 104 cells/ ml and allowed to adhere overnight in 20% scribed earlier and sacrificed at 7, 14, 21, and 28 d, and their right lung L929 feeding media containing 50 U/ml recombinant murine IFN-g and lobes were removed. Lung tissue was homogenized in 1 ml of sterile 10 ng/ml LPS. A cold mixture of fungal cells and 10 mg/ml mAb 18B7 PBS, and capsule dimensions were measured as described earlier. Left (IgG1) was prepared and placed on ice. This mAb binds to capsular lung lobes were fixed in formaldehyde, and sections were stained with polysaccharide and is opsonic (37). BMDMs were placed on ice and cold H&E for histological studies.

FIGURE 3. C. neoformans cells coincubated in vitro with Anxa22/2 macrophages for 24 h show enlarged capsule sizes. (A) Microscopic analysis of C. neoformans capsule size revealed smaller capsule size in yeast cells associated with Anxa2+/+ macrophages compared with yeast cells associated with Anxa22/2 macro- phages. All images were taken at original magnifi- cation 363. Scale bar, 10 mm. (B) C. neoformans capsule size is smaller in yeast cells isolated from Anxa2+/+ macrophages. Infected macrophages were lysed, and capsule size calculated by from the diameter of the whole cell minus the diameter of the inner cell body (n = 200). Statistical significance is represented by ****p , 0.0001 and calculated using Student two-tailed t test. (C) Distribution of capsule sizes for each genotype. 4 ANNEXIN A2 DEPLETION IN MACROPHAGES IMPACTS FUNGAL INFECTION

Immunofluorescence microscopy Results 2/2 BMDM were plated as described for nonlytic exocytosis assays. Macro- Anxa2 macrophages are less efficient at phagocytosis phage monolayers were infected with C. neoformans at an E:T ratio of 2:1. To determine the cellular processes affected by the absence of ANXA2, After a 2-h incubation the extracellular fungal cells were washed away. At specific time intervals thereafter, samples were permeabilized and fixed we examined one of the first steps of an intracellular infection, namely, with methanol at 220˚C for 20 min. Uvitex 2B was used to highlight internalization of fungal cells. Using BMDMs in conjunction with a internalized fungal cells, and an LC3 rabbit polyclonal Ab (Santa Cruz) previously described staining protocol optimized for light scanning was used to determine localization. Cells were visualized using a Zeiss cytometry (38), we could readily distinguish macrophages that had 3 Axio Observer CLEM at 63 objective. internalized yeast cells from those without fungi (Fig. 1A). We ob- 2 2 Reactive oxygen species measurement served a small but significant decrease in phagocytosis by Anxa2 / macrophages at both time intervals compared with Anxa2+/+ mac- Peritoneal macrophages were infected with C. neoformans and incubated 2/2 for various time intervals. Extracellular fungal cells were washed away, rophages (Fig. 1B). At 30 min, 30% of Anxa2 macrophages and the macrophage monolayer was stained with media supplemented with remained uninfected, whereas only 20% of wild-type macrophages CellRox Deep Red Reagent (Life Technologies) for 30 min at 37˚C. The contained no fungal cells (Fig. 1C). At 2 h, the difference decreased medium was removed, and macrophages were washed three times with with 20% of the Anxa22/2 macrophages containing no fungal cells PBS. Fluorescence was measured using a Becton Dickinson LSRII (BD 2/2 Biosciences). (Fig. 1D). Therefore, Anxa2 macrophages could phagocytose IgG- opsonized C. neoformans cells, but Anxa22/2 macrophages had Quantitative RT-PCR phagocytic defects compared with wild-type macrophages during +/+ 2/2 early stages of internalization.

Anax2 and Anxa2 mice infected with C. neoformans were sacrificed at Downloaded from days 14 and 21 postinfection. Lung and brain were removed and immedi- Absence of ANXA2 in BMDMs reduces nonlytic exocytosis ately placed in RNAlater (Ambion, Austin, TX). RNA from 10 mg of tissue was extracted with RNeasy Mini Plus kit (Qiagen, Hilden, Germany). RNA We then analyzed whether the absence of ANXA2 had an effect concentration and integrity were checked by High Sensitivity Screen Tape on on the frequency of nonlytic exocytosis. As previously described Agilent TapeStation 2200 and Nanodrop. RNA Integrity Number for brain was .7.5 for all samples, whereas for lung samples it ranged from 2 to 8. For cDNA preparation, 250 mg of RNA in a total volume of 20 ml was added to cDNA EcoDry Premix with Random Hexamers (Clontech, Mountain http://www.jimmunol.org/ View, CA). Primers (Supplemental Table I) were designed to span an exon– exon junction, and quantitative PCR was performed with Sybr Green Master Mix (ABI, Warrington, U.K.) in a StepOne instrument (ABI) by performing a cycle of 95˚C for 10 min, followed by 40 cycles of 15 s at 95˚C, and 1 min at annealing temperature. Analysis was performed in StepOne Software v2.3 with automatic threshold and Cq determination. Cq values were exported to Prism, and the sample target level was calculated by normalizing ex- pression level to the arithmetic mean of the levels of Actin and TataBox Binding Protein as reference genes. Fold change was calculated using d14 WT1 as reference. Statistical significance was calculated from expression levels using an unpaired two-way ANOVA with Sidak postcorrection on by guest on September 23, 2021 GraphPad Prism for MacOS X v6.0c. Cytokine studies Anxa2+/+ and Anxa22/2 mice (n = 6 per group) were infected with C. neoformans as described previously and sacrificed at 14 and 21 d post- infection. The lungs were homogenized in 2 ml of PBS in the presence of inhibitors (Complete Mini; Roche Applied Science). Homoge- nates were centrifuged at 2000 3 g for 10 min to remove cell debris, and the supernatant was frozen at 280˚C until tested. Supernatants were assayed using mouse cytokine protein array C3 (Ray Biotech, Norcross, GA), per the manufacturer’s instructions. In brief, membranes were blocked with 13 blocking buffer, washed three times, and then incubated with samples. Membranes were washed again and incubated for 1 h with biotin-conjugated cytokines, which were detected by incubation with HRP- conjugated streptavidin. Unbound reagents were removed by washing and the membranes developed. All incubations were done at 37˚C for 1 h. Histology Lung tissue sections from infected Anxa2+/+ and Anxa22/2 mice were stained for arginase. Sections were deparaffinized in xylene followed by graded alcohols. Ag retrieval was performed by incubating sections in 10 mM sodium citrate buffer (pH 6.0) and heated to 96˚C for 20 min. Endogenous peroxidase activity was quenched using 3% hydrogen per- oxide in PBS for 30 min at room temperature. Sections were blocked with 5% BSA in PBS for 1 h. The primary Ab to arginase (Abcam, Cambridge, MA) was used at 1:50 for 1 h at room temperature. Sections were stained by routine immunohistochemistry methods using goat anti-mouse HRP con- 2/2 jugate (Southern Tech) to localize the Ab bound to Ag with diaminobenzidine FIGURE 4. Fungal capsule is enlarged in Anxa2 lung tissue on days 7 2 2 as the final chromogen. All immunostained sections were lightly counter- and 14. Anxa2+/+ and Anxa2 / mice were infected intratracheally and sac- stained with hematoxylin. rificed at days 7 and 14. (A) Fungal cells with varying capsule sizes can be seen at day 14 for Anxa2+/+ macrophages, whereas Anxa22/2 macrophages Statistical analysis show cells that have a larger, more uniform size. All images were taken at Statistical analyses were performed using GraphPad Prism V6. Tests in- original magnification 340. Scale bar, 2.5 mm. (B) Capsule thickness was cluded Student two-tailed t test and unpaired two-way ANOVAwith Sidak quantified for both time points, and a significant difference between the two postcorrection. mouse strains was observed. ***p , 0.0002, using Student two-tailed t test. The Journal of Immunology 5

2 2 FIGURE 5. Anxa2 / mice show increased inflammation and contain fungal cells with large capsule sizes post intratracheal infection with C. neo- Downloaded from formans. Anxa2+/+ mice have a granulomatous inflammatory response to yeast cells with infiltration of macrophages and eosinophils. By day 21, the response is more organized and more lymphocytes are present. In Anxa22/2 mice, some macrophages and eosinophils are present, along with perivascular cuffing at day 7, but granuloma formation is not as organized as in Anxa2+/+ mice. By days 21 and 28, the lungs are filled with yeast cells, and lung architecture is significantly damaged, in both genotypes. Original magnification 340.

+/+ ,

(16), time-lapse movies of multiple infected macrophages were recovered from Anxa2 macrophages (p 0.0001) (Fig. 3B). http://www.jimmunol.org/ analyzed to determine the outcome of the C. neoformans– When individual capsule sizes were categorized into groups macrophage interaction. For each video, all the cells in a field were according to their individual capsule size, those associated with categorized into three groups: 1) no change, indicating that the wild-type macrophages were predominantly in the range of 0.5 to 2 2 cryptococcal cells remained in the macrophage and there was no 1.5 mM, whereas those associated with Anxa2 / macrophages release of the fungal burden (data not shown); 2) cell lysis, in were predominantly 1–3 mM (Fig. 3C). To ascertain whether this which the macrophages burst releasing the cryptococcal cells into the extracellular environment effectively killing the host cell; or 3) nonlytic exocytosis, whereby cryptococcal cells were released from the confines of the macrophage leaving both host by guest on September 23, 2021 and pathogen alive. Anxa22/2 macrophages were eight times more likely to lyse during exocytosis of cryptococcal cells than wild-type macrophages and had an average of 20% fewer non- lytic exocytosis events (p , 0.008) (Fig. 2). This phenomenon was specific to macrophages containing live fungal cells, be- cause neither uninfected Anxa22/2 macrophages nor macro- phages infected with beads or heat-killed C. neoformans underwent lysis (data not shown). Our previous work categorized the process of nonlytic exocytosis into three subcategories based on both the number of fungal cells released and the manner in which they were extruded (16). In type I release, complete emptying of fungal cells from the macrophage occurred, whereas in type II release, partial emptying of fungal cells from the macrophage was noted. In type III release, transfer of fungal cells between individual macrophages was observed. Anxa22/2 macrophages displayed all three types of nonlytic exo- cytosis at rates and frequencies that were similar to those observed in Anxa2+/+ macrophages, indicating that, although ANXA2 had an effect on the overall process, it was not specific for any single type of nonlytic exocytosis (data not shown).

Cryptococcus neoformans capsule is enlarged in Anxa22/2 macrophages In our in vitro microscopy experiments, we observed that after 24 h FIGURE 6. ROS production in infected peritoneal macrophages. ROS yeast cells inside bone marrow–derived Anxa22/2 macrophages had produced by thioglycolate-induced peritoneal macrophages were measured for both uninfected and infected macrophages using CellRox Deep Red Anxa2+/+ larger capsules than yeast cells found within macrophages Reagent. (A) Macrophages of the two genotypes displayed similar amounts (Fig. 3A). To explore this phenomenon, we infected BMDMs from of ROS at baseline. (B) Peritoneal macrophages were infected with 2/2 wild-type and Anxa2 mice in vitro for 24 h and then lysed them opsonized C. neoformans, which led to an increase in ROS. Values rep- to recover intracellular organisms. Fungal cells recovered from resent combined results from two independent experiments. *p , 0.05, 2 2 Anxa2 / macrophages had significantly larger capsules than those **p , 0.01, using a two-tailed Student t test. 6 ANNEXIN A2 DEPLETION IN MACROPHAGES IMPACTS FUNGAL INFECTION phenomenon also occurred in vivo, we infected mice intratracheally with time. Examination of the lungs of Anxa2+/+ and Anxa22/2 and analyzed lung fungal burden 7 and 14 d postinfection. Lungs mice at day 28 revealed large collections of yeast cells within the were excised and homogenized to isolate and measure capsule size. alveolar space with significant damage to the lung. This phenotype We observed a significant increase (p , 0.0002) in capsule thickness did not translate into differences in fungal burden, because no at days 7 and 14 (Fig. 4A), which translated into an increase in statistical difference was observed when CFUs were quantified capsule volume (data not shown). The largest difference between the from the lung, brain, and liver of infected mice at all time intervals two murine environments occurred at day 7 (Fig. 4B). C. neofor- (Supplemental Fig. 1). mans capsule sizes in Anxa22/2 mice were comparable between days 7 and 14, suggesting that the capsule size enlargement Reactive oxygen species production and LC3 recruitment to occurred earlier in Anxa22/2 than in wild-type mice. phagosomes We then evaluated the lung histology of Anxa22/2 and Anxa2+/+ Given that free radicals can shave C. neoformans capsules (40) and mice infected with C. neoformans. Lung histology at 7, 14, 21, that ANXA2 also functions as a cellular redox regulatory protein and 28 d showed a striking difference in the capsule size for cells (41), we evaluated the production of oxidative bursts in Anxa22/2 recovered from the two strains of mice. The lungs of Anxa22/2 and Anxa2+/+ peritoneal macrophages. Reactive oxygen species mice harbored large collections of extracellular organisms with (ROS) production by these macrophages did not differ dramati- large capsules starting at day 14 and continuing until day 28. In cally throughout the infection in either genotype, showing only a contrast, yeast cells in lungs of Anxa2+/+ mice showed enlarged small statistical difference for uninfected cells (Fig. 6A). How- capsules only at later time points in the infection (Fig. 5). At day ever, when quantifying the total amount of ROS produced, we 2 2 14, Anxa2+/+ mice exhibited granulomatous infiltrates composed observed that Anxa2 / macrophages had reduced production Downloaded from of macrophages with some polymorphonuclear cells, primarily compared with Anxa2+/+ macrophages (Fig. 6B). Overall, these eosinophils. By day 21, the response was more organized and results show that the yield of ROS from Anxa22/2 peritoneal more lymphocytes were present. At day 7, Anxa22/2 mice exhibit macrophages is lower than wild-type macrophages. minimal inflammation with large collections of C. neoformans In addition, we examined infected phagosomes of BMDMs at cells within alveolar spaces. By day 14, there was an increase in different time intervals using two different markers, the phago-

inflammatory cell infiltrate composed of macrophages and somal marker LAMP-1 and the autophagy marker LC3-II, as well http://www.jimmunol.org/ polymorphonuclear cells. Perivascular cuffing was present, but not as the acidification dye LysoTracker. We observed no difference as organized compared with Anxa2+/+ mice and did not improve between LAMP-1 localization or the degree of phagosomal by guest on September 23, 2021

FIGURE 7. LC3 localization in C. neofor- mans phagosomes is decreased in Anxa22/2 macrophages. (A) Primary murine macrophages were infected with C. neoformans stained with Uvitex2B (shown in blue). After 2 h of phagocy- tosis, extracellular yeast cells were washed away and macrophage monolayers were stained for LC3 localization (shown in green) at the indicated time intervals. Original magnification 340. (B)Fluo- rescence of internalized fungal cells with LC3 lo- calization was quantified using ImageJ software. The experiment was repeated three times, and statistical significance was assessed for each time point using Student two-tailed t test: *p = 0.0103, **p = 0.0032, ****p , 0.0001. The Journal of Immunology 7 acidification between the two genotypes (data not shown). How- ever, we did observe that fewer phagosomes accumulated LC3 at later times in the infection in Anxa22/2 macrophages (Fig. 7A). Based upon LC3 fluorescence around individual internalized fungal cells, we observed a significant reduction in LC3 recruit- ment for yeast cells within Anxa22/2 macrophages (Fig. 7B). Mice deficient in ANXA2 show decreased survival when infected with C. neoformans It is known that the capsule of C. neoformans exerts immuno- modulatory effects on host cells. Given the histological differ- ences observed between Anxa2+/+ and Anxa22/2 mice, we evaluated survival after fungal infection in the two genotypes. When infected with 106 C. neoformans/mouse, Anxa22/2 mice exhibited a non- significant trend toward susceptibility to infection (p , 0.17) (Fig. 8A). When we repeated this experiment at lower inocula (105/mouse), Anxa22/2 mice showed increased susceptibility to cryptococcal infection compared with wild-type controls (p ,

0.05) (Fig. 8B). Given the various processes affected by the Downloaded from absence of ANXA2, our results suggest a working model that implicates a multifunctional role for ANXA2 during murine in- fection with C. neoformans (Fig. 9). To investigate the contribution of macrophage ANXA2 to de- creased host survival, we performed lung and brain quantitative FIGURE 8. Survival of Anxa22/2 mice is decreased when infected with

PCR and arginase immunostaining (Supplemental Fig. 1), as well http://www.jimmunol.org/ as cytokine arrays (Supplemental Fig. 2). We found that neither C. neoformans. Mice (n = 10) were infected intratracheally with varying doses of C. neoformans.(A) Survival when mice were infected with 106 total lung nor CFU counts changed significantly 5 +/+ 2/2 C. neoformans. (B) Survival when mice were infected with 10 C. neoformans between Anxa2 and Anxa2 animals. However, we did ob- , 2/2 2/2 (p 0.05). The experiment was terminated when the last Anxa2 serve that Anxa2 mice demonstrated a widespread increase in mouse died. inflammatory cytokines such as IL12p40/p70 and XCL-1, without a discernible shift of Th1 versus Th2 polarization when compared +/+ with Anxa2 animals. We found that the Arg1 expression in the the absence of ANXA2 could inhibit the ability of the macrophage 2/2 lungs of infected Anxa2 mice was delayed in comparison with to repair the physical disruption that may occur between mem- +/+ Anxa2 animals. Therefore, the presence or absence of ANXA2 branes during exocytosis leading to lysis of host cells. by guest on September 23, 2021 does not alter mouse control of yeast burden, but instead affects The molecular mechanisms that drive nonlytic exocytosis are cytokine signaling. Its absence results in inflammatory dysregu- unknown, but they could be similar to those known to underlie lation relative to what is observed in the wild-type mice. some elements of regulated exocytosis, including vesicle docking, membrane fusion, and subsequent release of secretory vesicles. Discussion Even though annexins are not fusogenic proteins, they interact In this study, we show that several specific aspects of the host with soluble N-ethylmaleimide sensitive fusion attachment pro- response to fungal infection with Cryptococcus neoformans are teins (SNAPs), membrane binding proteins that merge membranes deficient in macrophages lacking ANXA2, a phospholipid binding together to facilitate fusion. The phagosome formation and mat- protein that promotes intracellular membrane fusion. A role for uration process is regulated by SNAP-23, and the interaction be- ANXA2 during phagocytosis was previously suggested based on tween SNAP-23 and ANXA2 is crucial for exocytosis of lung the observation that this protein localized to the phagocytic cup surfactant from alveolar epithelial cells (45, 46), as well as se- during ingestion of photoreceptor outer segments in retinal pig- cretion of large, multimeric, membrane-enclosed proteins such as ment cells (42). In our system, the absence of this protein caused a collagen VI and von Willebrand factor (33, 47). Hence it is con- subtle defect in the number of fungal cells that macrophages are ceivable that ANXA2 forms a complex with SNAP-23 located on able to phagocytose in the early stages of infection. However, fungal phagosomes, bringing them in close contact with the cel- because overall phagocytosis by Anxa22/2 macrophages was lular membrane. In our system, ANXA2 appears critical for comparable with wild-type macrophages, Anxa22/2 macrophages controlling C. neoformans infection, because the number of may compensate for this subtle phagocytic defect by other macrophages that lysed was increased greatly over the period mechanisms, possibly including the recruitment of other annexins. studied. Because we also observed a decrease in nonlytic exocy- Because deficiency of ANXA2 did not abrogate the release of tosis, this could suggest that the mechanism of nonlytic exocytosis fungal cells into the extracellular environment, other annexins may is protective for the macrophage in the sense that the host cell is have provided redundancy for the missing protein, because mac- more likely to survive the interaction with C. neoformans.In rophages contain multiple annexins with similar structures and addition, interference with autophagy is associated with reduced functions. In contrast with wild-type macrophages, infected nonlytic exocytosis, and the reduction in LC3 recruitment ob- Anxa22/2 macrophages were more likely to undergo lytic exo- served in Anxa22/2 macrophages could be an additional mecha- cytosis. Annexins are also involved in the cellular membrane re- nism contributing to this effect (25). pair pathway, specifically during Listeria monocytogenes cell–cell An unexpected finding was the observation that coincubation of infection of primary murine macrophages, as ANXA2 is recruited fungal cells with Anxa22/2 macrophages was associated with a to bacterial vacuoles filled with phosphatidylserine to promote significant increase in capsule size both in vitro and in vivo. This membrane repair (43, 44). Based on this insight, it is possible that finding is consistent with, and complementary to, the report that 8 ANNEXIN A2 DEPLETION IN MACROPHAGES IMPACTS FUNGAL INFECTION

FIGURE 9. A multifunctional role of ANXA2 during a fungal infection: a model. The depletion of this membrane phospholipid binding protein affects several aspects of host cell function when controlling a fungal infection. The absence of

ANXA2 causes an initial decrease in the number Downloaded from of fungal cells internalized. Once ingested, fungal capsule size is increased within these phagosomes, as the time of infection progresses. Knockout macrophages also exhibit differences in their phagosomal environment, as there is a reduction in both LC3 recruitment and ROS pro- duction. Overall, there is a decrease in nonlytic http://www.jimmunol.org/ exocytosis and an increase in cellular lysis due to the inability of the host cell to contain the fungal cells. by guest on September 23, 2021

when ANXA2’s subcellular ligand, S100A10, is downregulated are presumably at an increased risk for cellular damage that, over in murine brain endothelial cells, there is both a decrease in time, could translate into impaired phagolysosomal function. phagocytosis and a significant increase in the capsule size of Variations in both the cytosolic and the phagosomal environments cryptococcal cells (48). The mechanism for the enlarged capsule could allow for the dramatic increase in fungal capsular size, and phenotype is unknown, but it is conceivable that it is related to therefore facilitate its damage to the host cell. Presumably, if yeast lower ROS in Anxa22/2 macrophages and deficient recruitment of cells with large capsules are released into the extracellular envi- LC3 to the phagosome when compared with wild-type macro- ronment, they would also be less likely to be taken up by other phages. Phagocytosis of fungal cells is accompanied by an oxi- phagocytes as the larger capsules reduce the efficiency of phago- dative burst, and free radicals can shave outer parts of the cytosis (52). Therefore, it is possible that ANXA2 and S100A10 cryptococcal capsule and reduce its apparent size (40). ANXA2 are involved in a pathway that influences C. neoformans capsule has also been reported to play a role in modulating cellular oxi- growth. dative damage (49). In this scenario, the larger capsules in Our initial hypothesis was based on the assumption that ANXA2 ANXA2- and S100A10-deficient cells could reflect the fact that might function to bring the phagosomal membrane in close there is reduced damage to fungal cells from weaker oxidative proximity to the cellular membrane to initiate the key steps of fluxes in Anxa22/2 phagolysosomes, thus allowing them to ex- nonlytic exocytosis. Although that hypothesis could still hold true, press larger capsules. An increase in capsule size could also offer our results hint at another nonexclusive possibility for a reduction protection by making fungal cells less susceptible to killing, by in nonlytic exocytosis: depletion of the protein causes an en- buffering ROS used by macrophages to kill intracellular fungal largement of the polysaccharide capsule that hinders the release of cells (50). Furthermore, these data combined with the recent the fungal cells and could cause cellular stress such that the discovery that ANXA2-deficient dendritic cells show a reduction macrophage lyses. The amount of capsule needed for nonlytic in autophagic flux (51) would suggest that Anxa22/2 macrophages exocytosis to occur could be thought of as a “Goldilocks volume,” The Journal of Immunology 9 whereby too much capsule hinders nonlytic exocytosis, as does 9. Sabiiti, W., E. Robertson, M. A. Beale, S. A. Johnston, A. E. Brouwer, A. Loyse, J. N. Jarvis, A. S. Gilbert, M. C. Fisher, T. S. Harrison, et al. 2014. Efficient too little because acapsular fungal cells residing within macro- phagocytosis and laccase activity affect the outcome of HIV-associated crypto- phages are not readily exocytosed. coccosis. J. Clin. Invest. 124: 2000–2008. Our in vivo data provide further support for the notion that 10. Go´mez, B. L., and J. D. Nosanchuk. 2003. Melanin and fungi. Curr. Opin. Infect. Dis. 16: 91–96. ANXA2 plays an important role in controlling a fungal infection 11. Zaragoza, O., M. L. Rodrigues, M. De Jesus, S. Frases, E. Dadachova, and 2/2 because Anxa2 mice showed reduced survival postinfection. A. Casadevall. 2009. The capsule of the fungal pathogen Cryptococcus neo- We did not observe a noticeable difference in the types of in- formans. Adv. Appl. Microbiol. 68: 133–216. 12. McFadden, D. C., B. C. Fries, F. Wang, and A. Casadevall. 2007. Capsule flammatory cells recruited to fungal granulomas but did see structural heterogeneity and antigenic variation in Cryptococcus neoformans. changes in the structural organization of the granulomas. It is Eukaryot. Cell 6: 1464–1473. possible that having enlarged capsules in lung tissues could affect 13. Tucker, S. C., and A. Casadevall. 2002. Replication of Cryptococcus neoformans in macrophages is accompanied by phagosomal permeabilization and accumu- granuloma formation and translate into reduced mouse survival lation of vesicles containing polysaccharide in the cytoplasm. Proc. Natl. Acad. because cryptococcal polysaccharide causes numerous delete- Sci. USA 99: 3165–3170. rious effects on host immunity (53). Anxa22/2 mice infected with 14. Ma, H., J. E. Croudace, D. A. Lammas, and R. C. May. 2006. Expulsion of live pathogenic yeast by macrophages. Curr. Biol. 16: 2156–2160. C. neoformans displayed an overall increase in cytokine levels in 15. Alvarez, M., and A. Casadevall. 2006. Phagosome extrusion and host-cell sur- the tissue and a delayed activation of arginase but had comparable vival after Cryptococcus neoformans phagocytosis by macrophages. Curr. Biol. 16: 2161–2165. fungal tissue burden in both the lung and the brain. We hypoth- 16. Stukes, S. A., H. W. Cohen, and A. Casadevall. 2014. Temporal kinetics and 2/2 esize that the increased susceptibility of Anxa2 mice reflects an quantitative analysis of Cryptococcus neoformans nonlytic exocytosis. Infect. overwhelming or dysregulated inflammatory response, which Immun. 82: 2059–2067. 17. Alvarez, M., T. Burn, Y. Luo, L.-A. Pirofski, and A. Casadevall. 2009. The translates into increased tissue damage and reduced survival, as outcome of Cryptococcus neoformans intracellular pathogenesis in human Downloaded from posited by the damage-response framework (54). It is most likely monocytes. BMC Microbiol. 9: 51. that ANXA2 plays a role in the macrophage–C. neoformans in- 18. Qin, Q.-M., J. Luo, X. Lin, J. Pei, L. Li, T. A. Ficht, and P. de Figueiredo. 2011. Functional analysis of host factors that mediate the intracellular lifestyle of teraction, possibly at the macrophage lysosome level and the brain Cryptococcus neoformans. PLoS Pathog. 7: e1002078. endothelial transmigration level. However, it may concomitantly 19. Chrisman, C. J., M. Alvarez, and A. Casadevall. 2010. Phagocytosis of Cryp- affect the biology of other immune cells. In this regard, and tococcus neoformans by, and nonlytic exocytosis from, Acanthamoeba cas- tellanii. Appl. Environ. Microbiol. 76: 6056–6062. consistent with our observations, ANXA2 has been shown to be 20. Nicola, A. M., E. J. Robertson, P. Albuquerque, Lda. S. Derengowski, and http://www.jimmunol.org/ involved in the regulation of cytokine responses (55, 56). Overall, A. Casadevall. 2011. Nonlytic exocytosis of Cryptococcus neoformans from macrophages occurs in vivo and is influenced by phagosomal pH. MBio 2: 2. our data indicate that ANXA2 and associated proteins are im- 21. Casadevall, A. 2010. Cryptococci at the brain gate: break and enter or use a portant molecules used in defense against fungal infections and Trojan horse? J. Clin. Invest. 120: 1389–1392. that further experiments need to be performed. 22. Johnston, S. A., and R. C. May. 2010. The human fungal pathogen Cryptococcus neoformans escapes macrophages by a phagosome emptying mechanism that is In summary, our results implicate a role for the membrane fusion inhibited by Arp2/3 complex-mediated actin polymerisation. PLoS Pathog. 6: protein, ANXA2, in nonlytic exocytosis and other processes that e1001041. occur in the macrophage during fungal infection. ANXA2 con- 23. Chayakulkeeree, M., S. A. Johnston, J. B. Oei, S. Lev, P. R. Williamson, C. F. Wilson, X. Zuo, A. L. Leal, M. H. Vainstein, W. Meyer, et al. 2011. SEC14 tributes to phagocytic efficiency, and its absence could reduce the is a specific requirement for secretion of phospholipase B1 and pathogenicity of ability of the macrophage to control a fungal infection. We show Cryptococcus neoformans. Mol. Microbiol. 80: 1088–1101. by guest on September 23, 2021 that the presence and absence of ANXA2 is associated with dif- 24. Voelz, K., D. A. Lammas, and R. C. May. 2009. Cytokine signaling regulates the outcome of intracellular macrophage parasitism by Cryptococcus neoformans. ferences in fungal capsule size within the confines of the macro- Infect. Immun. 77: 3450–3457. phage, a phenomenon that could impact nonlytic exocytosis. Given 25. Nicola, A. M., P. Albuquerque, L. R. Martinez, R. A. Dal-Rosso, C. Saylor, M. De Jesus, J. D. Nosanchuk, and A. Casadevall. 2012. Macrophage autophagy that annexins are expressed in a wide variety of cells and that in immunity to Cryptococcus neoformans and Candida albicans. Infect. Immun. nonlytic exocytosis has been observed in different cell types and for 80: 3065–3076. various pathogens, it is likely that this protein has protean roles in 26. Swairjo, M. A., and B. A. Seaton. 1994. Annexin structure and membrane in- teractions: a molecular perspective. Annu. Rev. Biophys. Biomol. Struct. 23: 193– host–fungal interactions in different settings. 213. 27. Gruenberg, J., and N. Emans. 1993. Annexins in membrane traffic. Trends Cell Disclosures Biol. 3: 224–227. 28. Creutz, C. E. 1992. The annexins and exocytosis. Science 258: 924–931. The authors have no financial conflicts of interest. 29. Rescher, U., and V. Gerke. 2004. Annexins–unique membrane binding proteins with diverse functions. J. Cell Sci. 117: 2631–2639. 30. Liu, L., M. Wang, A. B. Fisher, and U. J. Zimmerman. 1996. Involvement of References annexin II in exocytosis of lamellar bodies from alveolar epithelial type II cells. Am. J. Physiol. 270: L668–L676. 1. Feldmesser, M., Y. Kress, P. Novikoff, and A. Casadevall. 2000. Cryptococcus 31. Pollard, H. B., E. Rojas, and A. L. Burns. 1987. Synexin and chromaffin granule neoformans is a facultative intracellular pathogen in murine pulmonary infec- membrane fusion. A novel “hydrophobic bridge” hypothesis for the driving and tion. Infect. Immun. 68: 4225–4237. directing of the fusion process. Ann. N. Y. Acad. Sci. 493: 524–541. 2. Coelho, C., A. L. Bocca, and A. Casadevall. 2014. The tools for virulence of 32. Knop, M., E. Aareskjold, G. Bode, and V. Gerke. 2004. Rab3D and annexin A2 Cryptococcus neoformans. Adv. Appl. Microbiol. 87: 1–41. play a role in regulated secretion of vWF, but not tPA, from endothelial cells. Cryptococcus 3. Alanio, A., M. Desnos-Ollivier, and F. Dromer. 2011. Dynamics of EMBO J. 23: 2982–2992. neoformans-macrophage interactions reveal that fungal background influences 33. Dassah, M., D. Almeida, R. Hahn, P. Bonaldo, S. Worgall, and K. A. Hajjar. outcome during cryptococcal meningoencephalitis in humans. MBio 2: 2. 2014. Annexin A2 mediates secretion of collagen VI, pulmonary elasticity and 4. Velagapudi, R., Y.-P. Hsueh, S. Geunes-Boyer, J. R. Wright, and J. Heitman. apoptosis of bronchial epithelial cells. J. Cell Sci. 127: 828–844. 2009. Spores as infectious propagules of Cryptococcus neoformans. Infect. 34. Hayes, M. J., U. Rescher, V. Gerke, and S. E. Moss. 2004. Annexin-actin in- Immun. 77: 4345–4355. teractions. Traffic 5: 571–576. 5. Coelho, C., A. L. Bocca, and A. Casadevall. 2014. The intracellular life of 35. Vu, K., R. A. Eigenheer, B. S. Phinney, and A. Gelli. 2013. Cryptococcus Cryptococcus neoformans. Annu. Rev. Pathol. 9: 219–238. neoformans promotes its transmigration into the central nervous system by in- 6. Mansour, M. K., J. M. Vyas, and S. M. Levitz. 2011. Dynamic virulence: real- ducing molecular and cellular changes in brain endothelial cells. Infect. Immun. time assessment of intracellular pathogenesis links Cryptococcus neoformans 81: 3139–3147. phenotype with clinical outcome. MBio 2: 2. 36. Ling, Q., A. T. Jacovina, A. Deora, M. Febbraio, R. Simantov, R. L. Silverstein, 7. Alanio, A., F. Vernel-Pauillac, A. Sturny-Lecle`re, and F. Dromer. 2015. Cryp- B. Hempstead, W. H. Mark, and K. A. Hajjar. 2004. Annexin II regulates fibrin tococcus neoformans host adaptation: toward biological evidence of dormancy. homeostasis and neoangiogenesis in vivo. J. Clin. Invest. 113: 38–48. MBio 6: 6. 37. Larsen, R. A., P. G. Pappas, J. Perfect, J. A. Aberg, A. Casadevall, G. A. Cloud, 8. Davis, M. J., A. J. Eastman, Y. Qiu, B. Gregorka, T. R. Kozel, J. J. Osterholzer, R. James, S. Filler, and W. E. Dismukes. 2005. Phase I evaluation of the safety J. L. Curtis, J. A. Swanson, and M. A. Olszewski. 2015. Cryptococcus neofor- and pharmacokinetics of murine-derived anticryptococcal antibody 18B7 in mans-induced macrophage lysosome damage crucially contributes to fungal subjects with treated cryptococcal meningitis. Antimicrob. Agents Chemother. virulence. J. Immunol. 194: 2219–2231. 49: 952–958. 10 ANNEXIN A2 DEPLETION IN MACROPHAGES IMPACTS FUNGAL INFECTION

38. Coelho, C., L. Tesfa, J. Zhang, J. Rivera, T. Gonc¸alves, and A. Casadevall. 2012. 48. Chen, Y., J. Chen, H. Wen, P. Gao, J. Wang, Z. Zheng, and J. Gu. 2011. S100A10 Analysis of cell cycle and replication of mouse macrophages after in vivo and downregulation inhibits the phagocytosis of Cryptococcus neoformans by mu- in vitro Cryptococcus neoformans infection using laser scanning cytometry. rine brain microvascular endothelial cells. Microb. Pathog. 51: 96–100. Infect. Immun. 80: 1467–1478. 49. Madureira, P. A., R. Hill, V. A. Miller, C. Giacomantonio, P. W. K. Lee, and 39. Feldmesser, M., and A. Casadevall. 1997. Effect of serum IgG1 to Cryptococcus D. M. Waisman. 2011. Annexin A2 is a novel cellular redox regulatory protein neoformans glucuronoxylomannan on murine pulmonary infection. J. Immunol. involved in tumorigenesis. Oncotarget 2: 1075–1093. 158: 790–799. 50. Zaragoza, O., C. J. Chrisman, M. V. Castelli, S. Frases, M. Cuenca-Estrella, 40. Bryan, R. A., O. Zaragoza, T. Zhang, G. Ortiz, A. Casadevall, and E. Dadachova. J. L. Rodrı´guez-Tudela, and A. Casadevall. 2008. Capsule enlargement in 2005. Radiological studies reveal radial differences in the architecture of the Cryptococcus neoformans confers resistance to oxidative stress suggesting a polysaccharide capsule of Cryptococcus neoformans. Eukaryot. Cell 4: 465–475. mechanism for intracellular survival. Cell. Microbiol. 10: 2043–2057. 41. Madureira, P. A., and D. M. Waisman. 2013. Annexin A2: the importance of 51. Morozova, K., S. Sridhar, V. Zolla, C. C. Clement, B. Scharf, Z. Verzani, being redox sensitive. Int. J. Mol. Sci. 14: 3568–3594. A. Diaz, J. N. Larocca, K. A. Hajjar, A. M. Cuervo, and L. Santambrogio. 2015. 42. Law, A.-L., Q. Ling, K. A. Hajjar, C. E. Futter, J. Greenwood, P. Adamson, S. T. Wavre- Annexin A2 promotes phagophore assembly by enhancing Atg16L⁺ vesicle Shapton, S. E. Moss, and M. J. Hayes. 2009. Annexin A2 regulates phagocytosis of biogenesis and homotypic fusion. Nat. Commun. 6: 5856. photoreceptor outer segments in the mouse retina. Mol. Biol. Cell 20: 3896–3904. 52. Zaragoza, O., C. P. Taborda, and A. Casadevall. 2003. The efficacy of 43. Draeger, A., K. Monastyrskaya, and E. B. Babiychuk. 2011. Plasma membrane complement-mediated phagocytosis of Cryptococcus neoformans is dependent repair and cellular damage control: the annexin survival kit. Biochem. Phar- on the location of C3 in the polysaccharide capsule and involves both direct and macol. 81: 703–712. indirect C3-mediated interactions. Eur. J. Immunol. 33: 1957–1967. 44. Czuczman, M. A., R. Fattouh, J. M. van Rijn, V. Canadien, S. Osborne, A. M. Muise, 53. Monari, C., F. Bistoni, and A. Vecchiarelli. 2006. Glucuronoxylomannan ex- V. K. Kuchroo, D. E. Higgins, and J. H. Brumell. 2014. Listeria monocytogenes hibits potent immunosuppressive properties. FEMS Yeast Res. 6: 537–542. exploits efferocytosis to promote cell-to-cell spread. Nature 509: 230–234. 54. Casadevall, A., and L.-A. Pirofski. 2003. The damage-response framework of 45. Sakurai, C., H. Hashimoto, H. Nakanishi, S. Arai, Y. Wada, G.-H. Sun-Wada, microbial pathogenesis. Nat. Rev. Microbiol. 1: 17–24. I. Wada, and K. Hatsuzawa. 2012. SNAP-23 regulates phagosome formation and 55. Schuliga, M., S. Langenbach, Y. C. Xia, C. Qin, J. S. L. Mok, T. Harris, maturation in macrophages. Mol. Biol. Cell 23: 4849–4863. G. A. Mackay, R. L. Medcalf, and A. G. Stewart. 2013. Plasminogen-stimulated 46. Wang, P., N. R. Chintagari, D. Gou, L. Su, and L. Liu. 2007. Physical and inflammatory cytokine production by airway smooth muscle cells is regulated by functional interactions of SNAP-23 with annexin A2. Am. J. Respir. Cell Mol. annexin A2. Am. J. Respir. Cell Mol. Biol. 49: 751–758. Downloaded from Biol. 37: 467–476. 56.Zhang,S.,M.Yu,Q.Guo,R.Li,G.Li,S.Tan,X.Li,Y.Wei,andM.Wu. 47. Pulido, I. R., R. Jahn, and V. Gerke. 2011. VAMP3 is associated with endothelial 2015. Annexin A2 binds to and negatively regulates TLR4- weibel-palade bodies and participates in their Ca(2+)-dependent exocytosis. triggered inflammatory responses via the TRAM-TRIF pathway. Sci. Rep. 5: Biochim. Biophys. Acta 1813: 1038–1044. 15859. http://www.jimmunol.org/ by guest on September 23, 2021 A B

Lung qPCR Annexin A2 +/+ Annexin A2-/-

CCL2 iNOS Ym1 Arg1 Fizz1 p40 60 Day 7

40

20

Fold change Fold change 0

-20 WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO Day 14 D14 D21 D14 D21 D14 D21 D14 D21 D14 D21 D14 D21

Brain qPCR

CCL2 iNOS Ym1 Arg1 Fizz1 p40 Day 21

20 * *

10 Fold change Fold change 0

WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO WT KO Day 28 D14 D21 D14 D21 D14 D21 D14 D21 D14 D21 D14 D21

Supplemental Figure 1. Annexin A2-deficient mice show no difference in expression of selected immune genes. (A) Total RNA was extracted from lungs and brains of infected Anxa2+/+ and Anxa2-/- mice at the indicated timempoints post infection. Whole tissue gene expression was measured by RT-qPCR. Each circle corresponds to an individual animal with mean and SD also indicated. * = p<0.05 for unpaired two-way ANOVA with Sidak post-correction. n= 6 mice per group B. (B) At days 7, 14, 21, Anxa2+/+ mice exhibited increased arginase expression with a subsequent reduction in expression by day 28. In Anxa2-/- mice on day 7, no arginase staining was detected, whereas staining began to appear on days 14 and 21, with a significant reduction by day 28. Original magnification = 40x. A B

CFU Lung 106

105

104

103 CFU 102

101

100

CFU Brain 106

105

104

103 CFU 102

101

100