Annexin A2 Regulates Autophagy in Pseudomonas aeruginosa Infection through the Akt1−mTOR−ULK1/2 Signaling Pathway

This information is current as Rongpeng Li, Shirui Tan, Min Yu, Michael C. Jundt, of September 28, 2021. Shuang Zhang and Min Wu J Immunol published online 14 September 2015 http://www.jimmunol.org/content/early/2015/09/13/jimmun ol.1500967 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2015/09/13/jimmunol.150096 Material 7.DCSupplemental http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 28, 2021 *average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published September 14, 2015, doi:10.4049/jimmunol.1500967 The Journal of Immunology

Annexin A2 Regulates Autophagy in Pseudomonas aeruginosa Infection through the Akt1–mTOR–ULK1/2 Signaling Pathway

Rongpeng Li,*,† Shirui Tan,*,‡ Min Yu,*,x Michael C. Jundt,* Shuang Zhang,*,{ and Min Wu*

Earlier studies reported that a cell membrane protein, Annexin A2 (AnxA2), plays multiple roles in the development, invasion, and metastasis of cancer. Recent studies demonstrated that AnxA2 also functions in immunity against infection, but the underlying mechanism remains largely elusive. Using a mouse infection model, we reveal a crucial role for AnxA2 in host defense against Pseudomonas aeruginosa,asanxa22/2 mice manifested severe lung injury, systemic dissemination, and increased mortality 2/2 compared with wild-type littermates. In addition, anxa2 mice exhibited elevated inflammatory cytokines (TNF-a, IL-6, IL- Downloaded from 1b, and IFN-g), decreased bacterial clearance by macrophages, and increased superoxide release in the lung. We further identified an unexpected molecular interaction between AnxA2 and Fam13A, which activated Rho GTPase. P. aeruginosa infection induced autophagosome formation by inhibiting Akt1 and mTOR. Our results indicate that AnxA2 regulates autophagy, thereby contrib- uting to host immunity against bacteria through the Akt1–mTOR–ULK1/2 signaling pathway. The Journal of Immunology, 2015, 195: 000–000. http://www.jimmunol.org/ nnexin A2 (AnxA2), a member of the annexin family, is carcinoma, chronic obstructive pulmonary disease (COPD), and expressed in various human cells, such as endothelial chronic inflammatory diseases (4); thus, AnxA2 may be a novel A cells, mononuclear cells, macrophages, marrow cells, diagnostic serum biomarker for lung cancer and other diseases (5). and some tumor cells (1). Accumulating evidence indicates that For instance, AnxA2 is involved in p53-induced apoptosis in lung AnxA2 plays multiple roles in signal transduction, cell prolifera- cancer (6). Increasing secretion of collagen VI and pulmonary tion, differentiation, apoptosis, endocytosis, exocytosis, and in- elasticity of bronchial epithelial cells (7), along with increased flammation (1, 2). AnxA2 downregulation is associated with the AnxA2 on lung epithelial cell surface, were recognized by severe occurrence, invasion, and metastasis of cancer, whereas its up- acute respiratory syndrome–associated coronavirus spike domain regulation is related to the development, invasion, metastasis, and 2 Abs (8). An epithelial cell surface AnxA2/p11 complex is also by guest on September 28, 2021 drug resistance of hepatocellular carcinoma, colorectal cancer, required for efficient invasion of Salmonella typhimurium (9). breast cancer, acute promyelocytic leukemia, and renal cell car- Evidence also suggests that AnxA2 directly interacts with Rab14 cinoma (3). In addition, AnxA2 was shown to play diverse roles in in alveolar type II epithelial cells (10). In addition, AnxA2 inducer pulmonary diseases, such as non-small cell lung cancer, adeno- or inhibitor dramatically modulates AnxA2-related cell functions. Investigators demonstrated that cationic lipid-guided AnxA2 short hairpin RNA attenuates tumor growth and metastasis in a mouse *Department of Biomedical Sciences, University of North Dakota, Grand Forks, ND 58203; †College of Biotechnology and Pharmaceutical Engineering, Nanjing Uni- lung cancer stem cell model (11), whereas IFN-g stimulates AnxA2 versity of Technology, Nanjing 211800, People’s Republic of China; ‡College of expression in lung epithelial cells to enhance apoptosis (8). Agriculture, Yunnan University, Kunming 650091, People’s Republic of China; x Pseudomonas aeruginosa is an opportunistic bacterial pathogen Department of Thoracic Oncology, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, People’s Republic of China; and {State Key Laboratory causing acute and chronic pulmonary infection in immunocom- of Biotherapy and Cancer Center/Collaborative Innovation Center of Biotherapy, promised people (12), such as patients with cystic fibrosis and West China Hospital, Sichuan University, Chengdu 610041, People’s Republic of China COPD, and most of the morbidity and pathophysiology associated with these diseases is due to innate hypersusceptibility to bacterial Received for publication April 27, 2015. Accepted for publication August 6, 2015. infection (13). Innate immunity, primarily through inflammatory This work was supported by Flight Attendant Medical Research Institute Grant 103007 and National Institutes of Health Grants AI109317-01A1, AI101973-01, cytokine production, cellular recruitment, and phagocytic clear- and AI097532-01A1. ance by neutrophils and macrophages, is key to host control of Address correspondence and reprint requests to Dr. Min Wu, Department of Bio- P. aeruginosa infection (14). Recently, macroautophagy (hereafter medical Sciences, University of North Dakota, 501 North Columbia Road, P.O. referred to as “autophagy”) in macrophages was found to be in- Box 9037, Grand Forks, ND 58203-9037. E-mail address: [email protected] volved in host defense, and it directly impacts immunity and the The online version of this article contains supplemental material. inflammatory response (15). Autophagy participates in the elimi- Abbreviations used in this article: AM, alveolar macrophage; AnxA2, Annexin A2; nation of invasive bacteria through autolysosome by functioning BALF, bronchoalveolar lavage fluid; CLSM, confocal laser scanning microscopy; COPD, chronic obstructive pulmonary disease; CTB, cholera toxin B chain; as downstream factors of pattern recognition receptors (such as H2DCF, dihydrodichlorofluorescein diacetate; IP, immunoprecipitation; JC-1, TLRs) (16) and pathogen-associated molecular patterns (17). 5,59,6,69-tetrachloro-1,193,3-tetraethyl-benzimidazolyl-carbocyanine iodide; LB, ly- sogeny broth; 3-MA, 3-methyladenine autophagy inhibitor; MOI, multiplicity of Our previous work indicated that autophagy plays a role in infection; PMN, polymorphonuclear neutrophil; ROS, reactive oxygen species; siNC, immune defense against P. aeruginosa invasion (15). However, the scrambled small interfering RNA; siRNA, small interfering RNA; Tre, Trehalose; roles of AnxA2 in host defense against P. aeruginosa infection WT, wild-type. have not been defined. Investigations indicated that cell surface Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 AnxA2 serves as a receptor for P. aeruginosa in mammalian

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1500967 2 AnxA2 REGULATES PSEUDOMONAS-INDUCED AUTOPHAGY IN MACROPHAGES epithelial cells for bacterial internalization into host cells (18). Rad, Hercules, CA) and gene-specific primers (Supplemental Table I, Evidence also suggests that AnxA2 participates in endocytosis and synthesized by Integrated DNA Technologies, Coralville, IA) in a CFX Connect Real-Time PCR Detection System (Bio-Rad). Relative transcript EGFR-mediated signal transduction in macrophages (19), because 2DD levels were obtained using 2 Ct threshold methodology following nor- suppression of AnxA2 impairs the phagocytic ability of peritoneal malization with GAPDH (31). macrophages (20). In this study, we hypothesized that AnxA2 Plasmid construction plays roles in host defense against P. aeruginosa infection. To this 2 2 end, we used wild-type (WT) and anxa2 / mice to investigate Fam13A gene was amplified from MH-S genome DNA with specific pri- the role and underlying mechanism of AnxA2 in P. aeruginosa mers (Supplemental Table I, synthesized by Integrated DNA Technologies) infection, particularly the participation of various autophagy- by PCR and cloned into the BamHI and HindIII sites of the pcDNA3.1 vector (Addgene, Cambridge, MA). Constructed plasmids were electro- related factors. We demonstrate that AnxA2 regulates autopha- porated into DH5a using an Electroporator 2510 system (settings: 25 mF, gosome formation after PAO1 infection through the Akt1–mTOR– 200 V, 2.5 kV; Eppendorf, Hauppauge, NY). Transformants were selected ULK1/2 signaling pathway. and maintained in LB medium containing 100 mM ampicillin (Sigma- Aldrich, St. Louis, MO). All of the nuclease, , and used in molecular cloning were purchased from New England Biolabs. Materials and Methods Transfection of small interfering RNA, plasmids, activators, Mice and inhibitors C57BL/6J female mice (6–8 wk) were obtained from the Jackson Labo- ratory (21), and anxa22/2 mice that are constructed based on C57BL/6J AnxA2, Fam13A, Atg5, Atg7, Beclin1, and scrambled small interfering mice were kindly provided by Dr. K. Hajjar (Cornell University) (22). RNAs (siNCs) were obtained from Santa Cruz Biotechnology (Santa Cruz, Exons 3 and 4 of anxa2 were disrupted with a cassette containing neo- CA). MH-S cells were transfected with small interfering RNA (siRNA; Downloaded from mycin driven by the phosphoglucokinase promoter to 5 pM), LC3-RFP G120A, pcDNA3.1, and pcDNA3.1-Fam13A (100 ng) 2/2 plasmids using Lipofectamine 2000 (Life Technologies) for 24 h, fol- generate anxa2 mice (22). Animals were kept in a specific pathogen– 2/2 free facility at the University of North Dakota (23). All animal studies lowing the manufacturer’s instructions. In some cases, anxa2 AMs were approved by the University of North Dakota Institutional Animal were treated with 5 mM autophagy activator Trehalose (Tre; Sigma- Care and Use Committee and performed in accordance with the animal Aldrich), and WT AMs were treated with 5 mM 3-methyladenine auto- care and institutional guidelines (IACUC approval #1204-4). The animal phagy inhibitor (3-MA; Sigma-Aldrich) for 2 h before and during PAO1 infection, as indicated.

experimental procedures, including treatment, care, and end point choice, http://www.jimmunol.org/ followed Animal Research: Reporting In Vivo Experiment guidelines. Bacterial burden assay Primary cells and cell lines AMs from BALF and ground lung, spleen, liver, and kidney tissues were homogenized with PBS and spread on LB dishes to enumerate bacteria. Mice were sacrificed, and the thoracic cavity and trachea were dissected. A The dishes were cultured in a 37˚C incubator overnight, and colonies were small incision was made in the trachea via a 1-ml syringe with an angiocath counted. Duplicates were performed for each sample and control (32). (BD Biosciences, Franklin Lakes, NJ). The lungs were lavaged three times with 1 ml PBS containing 1% FBS (Life Technologies, Grand Island, NY). NBT assay The retained bronchoalveolar lavage fluid (BALF) was centrifuged at 600 3 g for 5 min at 4˚C. The cell pellets were resuspended in RPMI 1640 This assay is based on the color change of NBT dye upon reduction by medium (Life Technologies) supplemented with 10% FBS and incubated released superoxide. AMs from BALF were grown in a 96-well plate in by guest on September 28, 2021 on a culture plate for 1 h at 37˚C/5% CO2 in an incubator to allow at- serum-containing medium at 37˚C for 4 h, and NBT (1 mg/ml) dye (Sigma- tachment of macrophages. Nonadherent cells were removed by washing Aldrich) was added to each well. Cells were incubated for an additional with normal saline. Murine MLE-12 lung type II epithelial cells and MH-S hour or until color developed. The dye is yellow and gives a blue-colored alveolar macrophages (AMs) were obtained from American Type Culture formazan upon reduction by superoxide. The reduction was ter- Collection (Manassas, VA) and cultured following the manufacturer’s minated by adding 100 ml stop solution (10% DMSO; 10% SDS in 50 mM instructions (24). HEPES buffer). The plate was kept at room temperature overnight for complete dissolution of formazan, and absorbance at 560 nm was recorded Bacteria preparation and infection experiments using a multiscan plate reader to quantify the concentration of superoxide The P. aeruginosa WT strain, PAO1, was kindly provided by Dr. S. Lory anion (24). Triplicates were performed for each sample and control. (Harvard University) (25). PAK and PAO1-EGFP were obtained from Dihydrodichlorofluorescein diacetate assay Dr. G. Pier (Harvard University) (26). PAO1 Xen-41 was obtained from PerkinElmer-Caliper (Waltham, MA); Klebsiella pneumoniae was pro- Dihydrodichlorofluorescein diacetate (H2DCF) dye (Life Technologies) vided by Dr. V. Miller (University of North Carolina) (27); and Escherichia does not normally fluoresce, but it emits green fluorescence upon reaction coli DH5-a was obtained from New England Biolabs (Ipswich, MA). with superoxide inside of cells. AMs were treated as above, and an equal Bacteria were grown for ∼16 h in lysogeny broth (LB) at 37˚C at 220 rpm amount of dye was added. After 10 min of incubation, fluorescence was with shaking and pelleted by centrifugation at 5000 3 g. Various mam- measured using a fluorometer at 485 nm excitation with a 528-nm emission malian cells were changed to antibiotic-free medium and infected with filter (33). bacteria at a multiplicity of infection (MOI) of 20. To control for LC3 degradation, 5 mM NH4Cl was added to the medium for 2 h of bacterial MTT assay infection in in vitro LC3 immunoblotting experiments (28). This short This assay measures the color change in MTT (Sigma-Aldrich) upon re- duration of culture with a low NH Cl concentration did not cause signif- 4 duction by to assess the viability of cells. Cells were treated as icant cellular alterations. Mice were anesthetized with 45 mg/kg ketamine above, and an equal amount of dye was added. After 1 h of incubation, the and instilled intranasally with 0.5 3 107 CFU PAO1 in 50 ml PBS (six reaction was stopped by stop solution and left at room temperature overnight mice/group) (29). Mice were monitored for symptoms and euthanized for complete dissolution of formazan. Absorbance at 560 nm was recorded when they were moribund. using a multiscan plate reader to quantify the concentration of superoxide In vivo imaging anion (34). At various time points postinfection, P. aeruginosa Xen-41–infected mice Mitochondrial potential assay were imaged with an IVIS XRII system (PerkinElmer, Waltham, MA), A JC-1 Mitochondrial Membrane Potential Assay Kit (Cayman Chemical, following the user guides provided by PerkinElmer-Caliper (30). Ann Arbor, MI) was used for this assay, following the manufacturer’s RNA isolation and quantitative real-time PCR instructions. The cytofluorimetric lipophilic cationic dye, 5,59,6,69-tetra- chloro-1,193,3-tetraethyl-benzimidazolyl-carbocyanine iodide (JC-1), can RNA was isolated from primary AMs. A total of 50-ng DNA-free RNA selectively enter mitochondria and reversibly change the color from green was subjected to first-strand cDNA synthesis using the SuperScript III First- to red as the membrane potential increases. AMs were treated as above, Strand Synthesis System (Life Technologies). Quantitative real-time and an equal amount of dye was added. After 30 min of incubation, PCR was performed using iTaq Universal SYBR Green Supermix (Bio- fluorescence was measured using a fluorometer at 560-nm excitation and The Journal of Immunology 3 with a 595-nm emission filter to detect healthy cells and at 485-nm exci- inhibitors (1:50; Thermo Fisher Scientific, Waltham, MA). Then total-cell tation and with a 535-nm emission filter to detect dead cells (35). lysates were mixed with immunoprecipitation (IP) Abs, which were coupled to agarose beads (50:50; Thermo Fisher Scientific). Immunoprecipitates Histological analysis were separated by SDS-PAGE and transferred to nitrocellulose transfer Lung tissues of three independent mice were fixed in 10% formalin (Sigma- membranes. Membranes were incubated with the primary Abs overnight. Aldrich) for 24 h and then embedded in paraffin using a routine histologic Labeling of the first Abs was detected using relevant secondary Abs procedure. Four-micrometer sections were cut, stained with standard H&E, conjugated to HRP and detected using ECL reagents (23). and examined for differences in morphology postinfection (36). LC3 puncta observation Inflammatory cytokine profiling AMs and MH-S cells were retransfected with LC3-RFP G120A plasmids Cytokine concentrations of TNF-a, IL-6, IL-1b, and IFN-g were measured, for 24 h. Cells were infected with PAO1-EGFP at an MOI of 20 for 2 h. using ELISA kits (eBioscience, San Diego, CA), in BALF samples col- Cells were observed under an LSM 510 Meta Confocal Microscope. LC3 lected at the indicated times postinfection. BALF was collected, and 100-ml puncta values were derived from 100 cells/sample (33). aliquots were added to the coated microtiter wells. The cytokine concentrations Immunostaining were determined with corresponding detection HRP-conjugated Abs. The values were read at 450 nm (36). AMs were isolated from WT mice 24 h post-PAO1 infection, as well as from control mice. Cells were incubated individually with primary anti-AnxA2 CFU count assay Ab and anti-Fam13A Ab, and then with the secondary FITC-conjugated Phagocytosis and intracellular killing were assessed with a CFU assay, as Abs, as described (31). Colocalization was observed under a Zeiss LSM previously described (37). AMs were challenged with the indicated bac- 510 Meta Confocal Microscope. teria at a 20:1 ratio. The numbers of internalized and killed bacteria were

Rho GTPase activity assay Downloaded from assessed after 1 or 2 h of incubation. After 1 h, gentamicin (300 mg/ml) was added to the medium for 30 min to kill extracellular bacteria. Cells Rho GTPase activity in AMs and MH-S cells was detected with a Pierce were counted using a hemocytometer (Sigma-Aldrich); subsequently, active Rho GTPase Pull-down and Detection Kit (Life Technologies), cells were lysed, and bacterial CFU were determined. Phagocytosed bac- following the manufacturer’s instructions. Briefly, whole-cell lysates were teria were as calculated as CFU (1 h)/cell number. A second series of prepared from AMs using the lysis buffer provided in the kit. Lysate internalization assays was run in parallel to determine the number of viable (containing active and inactive Rho GTPase) was incubated with the GST- bacteria following 2 h of incubation. After the same treatment to remove protein binding domain fusion protein from the respective downstream

extracellular bacteria, cells were incubated for an additional hour and lysed effector protein and glutathione resin. Unbound lysate proteins, including http://www.jimmunol.org/ for analysis of intracellular bacterial CFU. The killing efficiency was inactive or GDP-bound GTPase, were removed using spin columns. The calculated as [CFU (1 h) 2 CFU (2 h)]/CFU (1 h) and normalized to the active GTPase population was recovered from the glutathione resin using control group (37). SDS-PAGE loading buffer and analyzed by immunoblotting (39). Phagocytosis assay Statistical analysis AMs were challenged with PAO1 EGFP at an MOI of 20 for 1 h and then Most experiments were conducted in triplicate. Differences between two stained with cholera toxin B chain (CTB; Sigma-Aldrich) to track mem- groups were compared by one-way ANOVA (Tukey post hoc test) using brane sphingolipid-rich lipid rafts for 30 min. Cells were observed under GraphPad Prism 5 software, whereas survival rates were calculated using an LSM 510 Meta Confocal Microscope (Carl Zeiss Micro Imaging, a Kaplan–Meier curve (40). Thornwood, NY). After observation, cells were washed three times with PBS, by guest on September 28, 2021 centrifuged to remove extracellular bacteria, and lysed. Intracellular bacteria were incubated on LB dishes and analyzed using an IVIS XRII system (30). Results anxa22/2 mice are highly susceptible to PAO1 infection Autophagy-based PCR primer assay To investigate the role of AnxA2 in the P. aeruginosa infection RNA was isolated from AnxA2-silenced or negative MH-S cells 4 h post- process, we instilled PAO1 (0.5 3 107 CFU/mouse) intranasally in PAO1 infection. DNA-free RNA (100-ng) was used for first-strand cDNA 2/2 synthesis using the SuperScript III First-Strand Synthesis System (Life anxa2 and WT mice (with otherwise similar genetic back- Technologies). A PCR primer assay was performed using iTaq Universal grounds) to establish an acute pneumonia model and compared SYBR Green Supermix and gene-specific primers that attached to the the survival rates of these two groups of mice (six mice/group). As bottom of the Autophagy M384 predesigned 384 well panel in a CFX shown in Fig. 1A, anxa22/2 mice exhibited increased lethality Connect Real-Time PCR Detection System (both from Bio-Rad). PCR primer (50% died within 38 h postinfection). By 68 h, all anxa22/2 mice assay data were analyzed using PrimePCR analysis software (Bio-Rad). had died, whereas 80% of WT mice remained alive. This result is Immunoblotting shown in Kaplan–Meier survival curves (p = 0.021, log-rank test). Mouse mAbs against LC3, b-actin, and Akt1 and rabbit polyclonal Abs To directly examine the in vivo infection process, we also instilled against AnxA2, mTOR, ULK1, p-Akt1, and p-mTOR were obtained from mice with PAO1 Xen-41 (an engineered bacterium emitting bio- Santa Cruz Biotechnology. Rabbit polyclonal Abs against Fam13A was luminescence for imaging) at the same dose/mouse and observed obtained from Proteintech Group (Chicago, IL), whereas Phospho-ULK1 the pattern of bacterial dissemination. We found that anxa22/2 (Ser317) was obtained from Cell Signaling Technology (Danvers, MA). The samples derived from cells and lung homogenates were lysed in RIPA mice displayed a much broader distribution of bioluminescence in buffer, separated by electrophoresis on 12% SDS-PAGE gels, and trans- the thoracic cavity area at 2 h postinfection, and it continuously ferred to nitrocellulose transfer membranes (GE Amersham Biosciences, expanded, compared with WT mice (Fig. 1B, 1C). These results Pittsburgh, PA). Proteins were detected by Western blotting using primary suggest that AnxA2 contributes to the host defense against P. Abs at a concentration of 1/200 (Santa Cruz Biotechnology) or 1/1000 aeruginosa in pneumonia models by slowing bacterial dissemi- (Cell Signaling Technology) and were incubated overnight. Binding of the first Abs was determined using corresponding secondary Abs conju- nation. gated to HRP (Santa Cruz Biotechnology) and signals were developed AnxA2 loss results in increased oxidation, lung injury, and using the ECL reagent (Santa Cruz Biotechnology) (38). Gel bands were quantified by Quantity One software (Bio-Rad), and data are presented as inflammatory response 6 2 2 means SEM from three independent immunoblotting assays (38). To further analyze the cause of infection lethality in anxa2 / Phosphorylated and total protein levels were determined and quantified by three independent successive immunoblotting membranes. mice, we examined bacterial burdens in the lung, liver, spleen, kidney, BALF, and blood from PAO1-infected mice. Bacterial Immunoprecipitation CFU increased significantly in the organs of anxa22/2 mice To obtain whole-cell lysates, AMs and MH-S cells were homogenized compared with the same organs in WT mice (Fig. 2A, 2B, in lysis buffer containing phosphatase inhibitor (1:10000) and protease Supplemental Fig. 1A–D), demonstrating severe lung injury and 4 AnxA2 REGULATES PSEUDOMONAS-INDUCED AUTOPHAGY IN MACROPHAGES

FIGURE 1. anxa22/2 mice showed increased sus- ceptibility and mortality against P. aeruginosa infec- tion. (A) anxa22/2 and WT mice were challenged intranasally with 0.5 3 107 CFU PAO1/mouse and then observed up to 96 h. Kaplan–Meier survival curves were obtained. p = 0.021, log-rank test; 95% confidence interval: 16.9–68.8 h. (B and C) Whole-animal imaging of bioluminescence was obtained using an IVIS XRII system at different time points after anxa22/2 and WT mice were challenged with 0.5 3 107 CFU PAO1 Xen- 41/mouse (arrows indicate regions of PAO1 spread). Data (mean 6 SEM) are representative of three inde- pendent experiments. *p # 0.05, **p # 0.005, one-way ANOVA with Tukey post hoc test. Downloaded from

pneumonia associated with AnxA2 deficiency. Increased recruit- capacity in anxa22/2 mice was substantially dampened. To con- ment of polymorphonuclear neutrophils (PMNs) into the lung, firm this, we determined the survival of AMs, which is essential

which is necessary for bacterial clearance, also may contribute to for maintaining bacterial clearance, 24 h post-PAO1 infection. The http://www.jimmunol.org/ severe lung injury and systemic bacterial infection (41). To test this results showed a significant decrease in the viability of AMs in idea, we quantified PMN infiltration and observed increased PMNs anxa22/2 mice (Fig. 3A); consistent with this reduced survival in the BALF of anxa22/2 mice compared with WT mice (Fig. 2C). rate, the capacity for bacterial clearance, as quantified by CFU Phagocyte-released reactive oxygen species (ROS) are crucial for assays, also was impaired (Fig. 3B). We then transfected the host defense against bacterial infection by either augmenting anti- model murine lung macrophages (MH-S cells) and lung epithelial biotic activity or directly participating in bacterial killing in lyso- cells (MLE-12 cells) with an siRNA against AnxA2 to inhibit somes (42). However, increased levels of ROS may damage cellular AnxA2 expression, as well as with siNC as a negative control. organelles and, thereby, cause tissue injury (43). To measure oxi- Two hours post-PAO1 infection, similar decreases in survival and dative stress, anxa22/2 and WT mice were infected with PAO1, and bacterial clearance capacity were detected in AnxA2-silenced by guest on September 28, 2021 AMs were isolated for analysis of oxidation levels. As determined MH-S cells and negative controls (Fig. 3C, 3D). Note that the by NBT assay (Fig. 2D), AMs of anxa22/2 mice showed an ∼2-fold survival of AnxA2-silenced MLE-12 cells showed no significant increase in oxidative stress at 24 h postinfection compared with WT difference from controls (Fig. 3C), indicating that AnxA2 may mice. This result was confirmed using an H2DCF assay (Fig. 2E), play more dominant roles in macrophages than in epithelial cells. a sensitive fluorescence method for quantifying superoxide (44). We To determine whether the role of AnxA2 is limited to P. aerugi- measured mitochondrial membrane potential to determine the via- nosa or it plays a general role in bacterial defense, other bacte- bility of cells postinfection; the data showed a decreased mito- rial strains and species were investigated. When infected with chondrial membrane potential in AMs of anxa22/2 mice, as P. aeruginosa strains PAK and PA14, as well as K. pneumoniae determined using a JC-1 fluorescence assay (Fig. 2F). Thus, our and E. coli for 2 h, both the survival and bacterial clearance ca- data indicate that increased oxidation resulted in apoptosis of AMs. pacity were significantly decreased in AnxA2-silenced MH-S cells Lung histology was examined 24 h post-PAO1 infection as (Supplemental Fig. 1E, 1F), suggesting that the role of AnxA2 a direct indicator of lung injury. Although both anxa22/2 and WT may extend to a broad range of Gram-negative bacteria. mice showed signs of pneumonia, significant histological alter- A previous study reported that human epithelial cell surface ations were only observed in the lungs of anxa22/2 mice, indi- AnxA2 could recognize PAO1, indicating that AnxA2 may be cating more severe lung injury in these animals (Fig. 2G, insets). required to allow internalization of P. aeruginosa (18). However, The insets demonstrate typical regions with serious tissue damage whether AnxA2 facilitates phagocytosis in AMs remains unclear. with broken structure and inflammatory response due to increased In this study, CFU assay showed that, 2 h post-PAO1 infection, PMN penetration. We further measured cytokine concentrations in the number of PAO1 phagocytized by anxa22/2 AMs is similar to BALF 24 h post-PAO1 infection to gauge the inflammatory re- that of WT AMs (Fig. 3E). Confocal laser scanning microscopy sponse. TNF-a, IL-6, IL-1b, and IFN-g levels were significantly (CLSM) also showed a close colocalization of PAO1 EGFP with elevated in BALF from anxa22/2 mice compared with WT mice CTB stained from WT and anxa22/2 AMs (Fig. 3F). In addition, (Fig. 2H), indicating that lung macrophages might be the main we lysed these infected AMs to measure phagocytized PAO1 on source of increased cytokines. Collectively, these data suggest that LB dishes using an IVIS imaging system. Phagocytosis of PAO1 the severe lung injury in anxa22/2 mice is caused by excessive ROS EGFP by anxa22/2 AMs was similar to that of WT AMs (Fig. 3G). accumulation and an overly active inflammatory response. Collectively, these data suggest that AnxA2 is not required for bacterial phagocytosis in AMs. AnxA2 loss results in impaired PAO1-induced autophagy in In contrast, autophagy plays an essential role in the clearance of AMs PAO1 by macrophages (15). We speculated that AnxA2 is involved Bacteria burdens in BALF from anxa22/2 mice were much higher in autophagy in AMs to facilitate bacterial clearance. To test this than in WT mice (Fig. 2B), indicating that the bacterial clearance hypothesis, we used immunoblotting to determine the expression The Journal of Immunology 5 Downloaded from

FIGURE 2. Increased bacterial burdens, PMN, oxidation injury, and intense inflammatory response in lung of anxa22/2 mice. anxa22/2 and WT mice 3 7 A B were infected with 0.5 10 CFU PAO1/mouse for 24 h. ( ) Bacterial burdens in the lungs after homogenization in PBS. ( ) Bacterial loads in BALF http://www.jimmunol.org/ derived from infected mice. (C) PMN cell percentages evaluated in BALF versus total nuclear cells using HEMA-3 staining. Superoxide production in AMs was detected by an NBT assay at a wavelength of 560 nm (D) and an H2DCF assay at a wavelength of 488 nm (E). (F) Mitochondrial potential of AMs measured by JC-1 fluorescence assay at a wavelength of 532 nm. (G) Lung injury and inflammation assessed by histology. Lungs from indicated mice were embedded in formalin. Sections were analyzed by H&E staining (arrows indicate the region of insets showing typical tissue injury and inflammatory influx). Original magnification 3200, insets, 31000). (H) Inflammatory cytokines in BALF were assessed by ELISA. Data (mean 6 SEM) are representative of three independent experiments. *p # 0.05, **p # 0.005. patterns of LC-3, which is a marker of microtubule-associated whereas Akt1 and mTOR are interconnected junctions within the protein in the autophagosomal membrane, in mouse AMs. We autophagy regulation pathway (46). Among the downregulated by guest on September 28, 2021 found that endogenous conversion of LC3-I to LC3-II was dra- factors, ULK1 and ULK2 were the most significantly decreased matically increased in AMs of WT mice 24 h post-PAO1 infection factors that directly participate in autophagosome formation (47). (Fig. 3H). However, this conversion was not detected in AMs of Previous studies suggest that mTOR phosphorylates ULK1/2 and anxa22/2 mice (Fig. 3H), indicating that loss of AnxA2 inhibited Atg13, thereby inhibiting the translocation of ULK1/2–Atg13– PAO1-induced autophagy in AMs. Similar results were observed FIP200 complex to preautophagosomal membrane to induce in vitro using AnxA2-silenced MH-S cells 2 h post-PAO1 infec- autophagy (46). According to data from the autophagy array- tion (Fig. 3I). A significant increase in LC3-II was detected in based quantitative real-time PCR assay, we hypothesized that MH-S control cells, whereas the level of LC3-II remained un- AnxA2 functions in PAO1-induced autophagy through an Akt1– changed in AnxA2-silenced MH-S cells (Fig. 3I). We also trans- mTOR–ULK1/2 signaling pathway. fected an RFP-LC3 plasmid into anxa22/2 and WT AMs and To validate the results of the autophagy array-based quantitative infected them with PAO1 EGFP. PAO1 infection induced signifi- real-time PCR assay and confirm our proposed roles for AnxA2 in cant LC3 puncta in AMs from WT mice, but AnxA2 knockdown PAO1-induced autophagy, we examined the expression levels of resulted in an ∼80% decrease in RFP-LC3 puncta (Fig. 3J). In predicted factors in AMs of PAO1-infected anxa22/2 and WT addition, CLSM showed colocalization of PAO1 GFP and LC3 mice by immunoblotting using related Abs. In anxa22/2 AMs, (Fig. 3J), indicating that PAO1 clearance is actually due to auto- protein expression of Akt1 and mTOR was highly increased; in phagy. contrast, ULK1 expression was significantly decreased in anxa22/2 AMs compared with WT AMs (Fig. 4B). We also observed that AnxA2 loss results in impaired PAO1-induced autophagy phosphorylation of Akt1 and mTOR was increased in anxa22/2 through the Akt1–mTOR–ULK1/2 pathway AMs (Fig. 4B), indicating that, in addition to increased protein To further investigate the role of AnxA2 in autophagy, we assessed levels, AnxA2 deficiency led to enhanced Akt1 and mTOR acti- expression patterns of autophagy-related factors in AnxA2- vation. In WT AMs, phosphorylation of Akt1 remained at the silenced MH-S cells and negative controls, using an autophagy normal level compared with AMs without PAO1 infection, array-based real-time PCR primer assay. Postinfection with PAO1, whereas phosphorylation of mTOR seemed to be inhibited the quantitative real-time PCR primer array assay revealed that (Fig. 4B). Phosphorylation of Ser317 of ULK1 is an independent three genes were significantly (.2-fold) upregulated, whereas 73 marker for ULK1/2-Atg13-FIP200 and mTOR (48). We found genes were downregulated (Fig. 4A, Supplemental Table II), in- that, after PAO1 infection, ULK1 Ser317 was dramatically phos- dicating that AnxA2 repression may profoundly impair expression phorylated in WT AMs but was not phosphorylated in anxa22/2 of autophagy-related factors. Among the upregulated factors, NF- AMs (Fig. 4B), indicating that AnxA2 deficiency inhibited the trans- kB is a key transcription regulator that controls diverse gene ex- location of ULK1/2–Atg13–FIP200 complex to preautophagosomal pression related to host defense against PAO1 infection (45), membrane. 6 AnxA2 REGULATES PSEUDOMONAS-INDUCED AUTOPHAGY IN MACROPHAGES Downloaded from http://www.jimmunol.org/ FIGURE 3. AnxA2 deficiency blunted PAO1-induced autophagy in AMs. anxa22/2 and WT mice were infected with 0.5 3 107 CFU PAO1/mouse for 24 h. (A) Viability of the indicated AMs was determined by MTT assay at a wavelength of 570 nm. (B) Bacterial killing by AMs was assessed with a CFU assay. (C) AnxA2-silenced MH-S and MLE-12 cells were infected with PAO1 at an MOI of 20 for 2 h. Viability of the indicated MH-S and MLE-12 cells was determined by an MTT assay at a wavelength of 570 nm. (D) Bacterial killing by MH-S cells was detected using a CFU assay. (E) Phagocytosis of anxa22/2 and WT AMs infected with PAO1 was assessed by CFU count. (F) CTB-rhodamine–stained anxa22/2 and WT AMs (red) were challenged with PAO1 EGFP at an MOI of 20 for 1 h. Uptake of bacteria was observed using an immunofluorescence confocal microscope. (G) Phagocytized PAO1 EGFP was cultured on LB dishes and detected using an IVIS XRII imaging system. (H) PAO1-infected anxa22/2 and WT AMs were lysed to measure LC3 levels by immunoblotting. (I) LC3 levels in AnxA2-silenced and MH-S control cells were determined by immunoblotting. During PAO1 infection, 5 mM NH4Cl was added to exclude LC3 degradation. (J) Purified anxa22/2 and WT AMs were transfected with RFP-LC3 G120A plasmids for 24 h and then infected by guest on September 28, 2021 with PAO1-EGFP for 1 h (MOI = 20). Cells with apparent puncta ($10/cell) were considered LC3-RFP+ cells (arrows indicating colocalization of LC3 puncta and PAO1 EGFP). Values are derived from 100 cells/sample. Data (mean 6 SEM) are representative of three independent experiments. *p # 0.05, **p # 0.005, one-way ANOVA with Tukey post hoc test.

To further confirm this result, we prepared total-cell lysates of it remains unknown whether they are required for host defense anxa22/2 and WT AMs to perform co-IP analysis. At 24 post- against bacterial infection in macrophages. We randomly selected PAO1 infection, we used anti-mTOR or anti-ULK1 Ab to pull- 15 of these factors and measured their transcripts levels in AMs down the target protein and determined ULK1 and mTOR inter- from WT and anxa22/2 mice 24 h post-PAO1 infection, by action by immunoblotting using the appropriate Ab. Our results quantitative real-time PCR using specific primers (Supplemental showed that, in the absence of PAO1 infection, mTOR and ULK1 Table I). After PAO1 infection, transcripts of APOB, Brd4, Cd1d, showed a stable association in anxa22/2 and WT AMs (Fig. 4C). Fam13A, Malt1, NSF, ScgB, Sirt1, and Tob1 were significantly However, after PAO1 infection, this association was disrupted in increased, whereas transcript levels of Sirga were reduced (Fig. WT AMs but not in anxa22/2 AMs (Fig. 4C). This result strongly 5A). Among these 10 proteins, only transcripts of Fam13A and suggests that AnxA2 plays a role in mTOR–ULK1 association, Malt1 showed significant differences between anxa22/2 and con- which impacts autophagy initiation. trol AMs (Fig. 5A), indicating that they played roles associated To demonstrate that AnxA2 specifically acts by regulating the with AnxA2 in AMs during PAO1 infection. We focused further Akt1–mTOR–ULK1/2 signaling pathway, we treated WT AMs on Fam13A, because the change in the levels of its expression in with 3-MA to inhibit the Akt1-mTOR signaling pathway and si- anxa22/2 AMs was more significant compared with Malt1 (Fig. multaneously treated anxa22/2 AMs with Tre to activate the 5A). Akt1-mTOR signaling pathway, to mediate autophagy (Fig. 4D). As reported, Fam13A is involved in COPD and lung cancer, but Two hours post-PAO1 infection, the viability of 3-MA–treated WT little is known about its role in cystic fibrosis and bacterial defense AMs decreased to a level similar to that seen in anxa22/2 AMs, (49). We next detected Fam13A protein levels in AMs by im- whereas the survival of Tre-treated anxa22/2 AMs increased to munoblotting. We observed a significant induction of Fam13A in equal the level in WT AMs (Fig. 4E). Similar results were ob- PAO1-infected WT AMs, but not in anxa22/2 AMs (Fig. 5B), served in bacterial clearance ability detection in these indicated again indicating that Fam13A plays a role in AMs during PAO1 AMs (Fig. 4F). infection. To further confirm our hypothesis, we introduced Fam13A-interfering RNA into MH-S cells to block Fam13A ex- Fam13A is required in PAO1-induced autophagy pression, while using siNC as a negative control (Fig. 5C). Our Previous studies revealed several new factors that are involved in results showed that the survival and bacterial clearance capacity of diverse pulmonary diseases, including COPD and lung cancer, but Fam13A-silenced MH-S cells were decreased after PAO1 infection The Journal of Immunology 7 Downloaded from

FIGURE 4. AnxA2 deficiency impaired PAO1-induced autophagy through the Akt1–mTOR–ULK1/2 pathway. (A) Quantitative real-time PCR primer array assay of autophagy-related factor expression in AnxA2-silenced MH-S cells versus MH-S control cells after PAO1 infection for 4 h. Genes with .2- http://www.jimmunol.org/ fold changes among subsets are shown (n = 2 biological replicates). Upper left dots of double lines indicate increased proteins, and lower right dots indicate decreased proteins. (B) Autophagy-related factors and cell signaling protein levels of anxa22/2 and WT AMs were determined by immunoblotting 24 h postinfection with PAO1. Gel data were quantified from three independent experiments using densitometry with Quantity One. (C) Association of mTOR and ULK1 in AMs from anxa22/2 and WT mice was determined by immunoblotting with the indicated Ab for IP and detection. (D) Purified anxa22/2 AMs were treated with 5 mM Tre and WT AMs were treated with 5 mM 3-MA for 4 h before and during PAO1 infection. During PAO1 infection, 5 mM

NH4Cl was added to exclude LC3 degradation. LC3 levels of indicated AMs were measured by immunoblotting. (E) Viability of the indicated AMs was determined by MTT assay at a wavelength of 570 nm. (F) Bacterial killing by indicated AMs was tested with a CFU assay. Data (mean 6 SEM) are representative of three independent experiments. *p # 0.05, **p # 0.005, one-way ANOVA with Tukey post hoc test.

(MOI=20for2h)comparedwithMH-Scontrolcells(Fig.5D, mice at 24 h post-PAO1 infection and performed a co-IP assay with by guest on September 28, 2021 5E), indicating that Fam13A is also required for bacterial clear- anti-AnxA2 and anti-Fam13A Abs. A stable interaction between ance in MH-S cells, similar to AnxA2. AnxA2 and Fam13A was detected (Fig. 6A). CLSM also showed Because Fam13A expression was consistently reduced in anxa22/2 a significant colocalization of AnxA2 and Fam13A in AMs iso- AMs after PAO1 infection (Fig. 5A, 5B), we postulated that lated from WT mice 24 h post-PAO1 infection (Fig. 6B), strongly Fam13A was also involved in PAO1-induced autophagy. Therefore, indicating that AnxA2 and Fam13A are associated in AMs and we compared autophagosome formation between Fam13A-silenced likely act together in host defense as a single complex. This and MH-S control cells using RFP-LC3 plasmid transfection. PAO1 protein–protein interaction was also detected in siNC-transfected infection induced significant LC3 puncta in MH-S control cells, but MH-S cells after PAO1 infection (Fig. 6C), but it was abolished these LC3 puncta were diminished in Fam13A-silenced MH-S cells when either AnxA2 or Fam13A was repressed by RNA interfer- (Fig. 5F), suggesting that Fam13A was required for PAO1-induced ence (Fig. 6C), and it could not be rescued with Tre treatment in autophagy. We again analyzed the expression of autophagy-related AnxA2-inhibited MH-S cells (Fig. 6C). In addition, we saw re- factors of Fam13A-silenced MH-S cells after PAO1 infection, using duced macrophage viability and bacterial killing in autophagy- the autophagy-based quantitative real-time PCR primer assay. The inhibited MH-S cells by directly silencing some key autophagy results showed that, in Fam13A-silenced MH-S cells, Akt1, mTOR, factors, including Atg5, Atg7, and Beclin1 (Fig. 6D, Supplemental and NF-kB were significantly induced, and many factors, including Fig. 2A, 2B), again indicating that autophagy plays key roles in ULK1 and ULK2, were downregulated (Fig. 5G, Supplemental bacterial defense in macrophages. The association of AnxA2 with Table II). To validate these array data, we also examined expres- Fam13A was stable in all of these autophagy-inhibited MH-S sion levels of predicted factors in Fam13A-silenced and MH-S cells after PAO1 infection (Fig. 6E). Collectively, our results sug- control cells and found that phosphorylation of Akt1 and mTOR gested that the AnxA2–Fam13A complex, rather than canonical was significantly increased in Fam13A-silenced MH-S cells, whereas autophagy pathways, may be a critical regulator of autophagy. phosphorylation of Ser317 in ULK1 was dramatically inhibited, Protein function prediction analysis suggests that Fam13A con- compared with MH-S control cells (Fig. 5H). This pattern was ex- tains two conserved Rho GTPase-activating domains (www.ebi. tremely similar to that of AnxA2-silenced MH-S cells (Fig. 4A, 4B, ac.uk/interpro/protein/O94988). We suspected that Fam13A reg- Supplemental Table II), revealing an extremely critical function for ulates Rho GTPase activity in the defense against P. aeruginosa.To Fam13A in macrophage autophagy during PAO1 infection. confirm this, we analyzed Rho GTPase activities in WT and Fam13A-silenced MH-S cells, with or without PAO1 infection. We Fam13A binds directly to AnxA2 and activates Rho GTPase to found that Fam13A deficiency directly reduced Rho GTPase acti- facilitate autophagosome formation during PAO1 infection vation in MH-S cells, even without PAO1 infection (Fig. 6F, To test the potential association between Fam13A and AnxA2 Supplemental Fig. 2E). After PAO1 infection, MH-S control during PAO1 infection, we lysed primary AMs isolated from WT samples exhibited dramatically increased levels of activated 8 AnxA2 REGULATES PSEUDOMONAS-INDUCED AUTOPHAGY IN MACROPHAGES Downloaded from http://www.jimmunol.org/

FIGURE 5. Fam13A is associated with AnxA2 in PAO1-induced autophagy. (A) Transcripts of 15 indicated factors in anxa22/2 and WT mice AMs, at 24 h post-PAO1 infection, compared with control AMs. (B) Fam13A expression levels in AMs of PAO1-infected anxa22/2 and WT mice, as determined by immunoblotting. (C) Fam13A levels in Fam13A-silenced and siNC-treated MH-S control cells were determined by immunoblotting. Viability (D) and bacterial killing (E) of Fam13A-silenced MH-S cells was determined by an MTT assay at a wavelength of 570 nm and by a CFU assay, respectively. (F) Fam13A-silenced and MH-S control cells were transfected with RFP-LC3 G120A plasmids for 24 h and then infected with PAO1 EGFP for 1 h (MOI = 20). Cells with apparent puncta ($10/cell) were considered LC3-RFP+ cells (arrows indicate colocalization of LC3 puncta and PAO1 EGFP). Values in the bar graph are derived from 100 cells/sample. (G) Quantitative real-time PCR primer array assay of autophagy-related factor expression in Fam13A-silenced MH-S cells versus MH-S control cells after PAO1 infection for 4 h. Genes with .2-fold changes among subsets are shown (n = 2 biological replicates). 317 Upper left dots of double lines indicate increased proteins, and lower right dots indicate decreased proteins. (H) p-Akt1, p-mTOR, and p-ULK1 (Ser )of by guest on September 28, 2021 Fam13A-silenced and MH-S control cells were determined by immunoblotting 2 h postinfection with PAO1. Gel data were quantified from three inde- pendent experiments using densitometry with Quantity One. Data (mean 6 SEM) are representative of three independent experiments. *p # 0.05, **p # 0.005, one-way ANOVA with Tukey post hoc test.

GTPase compared with control cells (Fig. 6F, Supplemental Fig. a critical contributor to bacterial clearance. Importantly, additional 2E), but the level of activated Rho GTPase in Fam13A-silenced data demonstrate that AnxA2 plays an essential role in PAO1- cells was still low (Fig. 6F, Supplemental Fig. 2E). We further induced autophagy in AMs (Fig. 6I). Furthermore, we identified cloned Fam13A gene into the pcDNA3.1 vector and transfected the underlying mechanism by which AnxA2 impacted autophagy this plasmid into MH-S cells to overexpress Fam13A; MH-S cells during infection: by directly binding to Fam13A and activating transfected with empty pcDNA3.1 vector were used as negative Rho GTPase to regulate the Akt1–mTOR–ULK1/2 signaling axis. control (Fig. 6D). High expression of Fam13A did not change Similar phenotypes (increased cell death and decreased bacterial macrophage viability or bacterial killing in MH-S cells clearance) were observed in AnxA2-silenced MH-S cells infected (Supplemental Fig. 2C, 2D) after PAO1 infection, but it directly with different bacterial strains or species, such as P. aeruginosa induced Rho GTPase activation, even in the absence of PAO1 strains PA14 and PAK, K. pneumoniae, and E. coli (Supplemental infection (Fig. 6G, Supplemental Fig. 2F). Together, these data Fig. 1E, 1F). Thus, AnxA2 may render a critical immune defense indicate that Fam13A regulates PAO1-induced Rho GTPase acti- against a wide range of microorganisms. vation and contributes to host immunity in bacterial defense. The Our study established that AnxA2 is involved in host inflam- impaired GTPase activation also was detected in anxa22/2 AMs matory response regulation during P. aeruginosa infection. As after PAO1 infection compared with WT AMs (Fig. 6H, reported previously, AnxA2 plays multiple roles in host defense Supplemental Fig. 2G), suggesting that AnxA2 plays a role in the against bacteria. For instance, AnxA2 is predominantly an anti- activation of GTPase via Fam13A. inflammatory agent, largely because of its similarities to Annexin A1, which was universally demonstrated to play such a role (50). Discussion Existing data support a role for AnxA2 in the proinflammatory In this study, we identified a critical role for AnxA2 in host defense process; AnxA2 elicits activation of MAPK and nuclear translo- against P. aeruginosa infection. We observed a severe disease cation of NF-kB, resulting in inflammatory cytokine production, phenotype in anxa22/2 mice in our experiments, which included as well as chemokine production. This function was reflected in decreased survival, increased inflammatory response, and more enhanced macrophage effector function, because AnxA2 dramat- severe lung injury compared with WT mice (Figs. 1, 2, ically facilitated phagocytosis of E. coli bacteria (51). Other Supplemental Fig. 1A–D). The increased bacterial burdens in the studies also illustrated the role for AnxA2 in regulating the lung and other organs of anxa22/2 mice indicate that AnxA2 is inflammatory response (52). Our results are in line with this The Journal of Immunology 9 Downloaded from http://www.jimmunol.org/

FIGURE 6. Fam13A binds AnxA2 and activates Rho GTPase to facilitate autophagy during PAO1 infection. (A) Association of AnxA2 and Fam13A in AMs from WT mice was determined by immunoblotting with the indicated Ab for IP and detection 24 h postinfection with PAO1. (B) AMs were purified from control WT mice and PAO1-infected mice 24 h postinfection. CLSM showed the colocalization of AnxA2 and Fam13A (arrows) after PAO1 infection by immune staining. (C) Association of AnxA2 and Fam13A in MH-S cells was determined by immunoblotting with AnxA2 Ab for IP and detection 2 h postinfection with PAO1. Before infection, MH-S cells were transfected with siNC, siRNA against AnxA2 (with and without Tre), or siRNA against by guest on September 28, 2021 Fam13A for 24 h. (D) MH-S cells were transfected with siAtg5, siAtg7, siBeclin1, and pcDNA3.1-Fam13A for 24 h. Expression of Atg5, Atg7, Beclin1, and Fam13A in MH-S cells was determined by immunoblotting. (E) Association of AnxA2 and Fam13A in siNC-, siAtg5-, siAtg7-, and siBeclin1- transfected MH-S cells was determined by immunoblotting with AnxA2 Ab for IP and detection 2 h postinfection with PAO1. Total Rho GTPases and activated Rho GTPases in Fam13A-silenced MH-S cells (F), Fam13A-overexpressed MH-S cells (G), anxa22/2 AMs (H), and control cells were deter- mined by immunoblotting, with or without PAO1 infection for 2 h. (I) Diagram delineating a pathway in AnxA2-deficient macrophages against PAO1 infection. After PAO1 infection, AnxA2 binds Fam13A to activate Rho GTPase and subsequently inactivates the Akt1–mTOR–ULK1/2 signaling pathway to induce autophagosome formation. siAtg5, siAtg7, siBeclin1, and siFam13A indicate siRNA specifically against Atg5, Atg7, Beclin1, and Fam13A, respectively. Data are representative of three independent experiments. observation, because significantly enhanced inflammatory re- specifically in macrophages rather than epithelial cells. Autophagy sponse, lung injury, ROS accumulation, and cytokine production array-based quantitative real-time PCR assay and immunoblotting were found in anxa22/2 mice following PAO1 infection (Fig. 2), showed increased expression and activation of Akt1 and mTOR in and the autophagy array-based quantitative real-time PCR assay AnxA2-silenced MH-S cells (Fig. 4A, 4B), suggesting that AnxA2 showed that AnxA2 negatively regulates the host inflammatory regulates PAO1-induced autophagy via an Akt1-mTOR pathway. response by NF-kB (Fig. 4A). Autophagy is normally inhibited in metabolically repressed cells AnxA2 was reported to serve as a receptor for P. aeruginosa in by the metabolic checkpoint mTOR (54); how the Akt1- mammalian epithelial cells and is involved in its internalization mTOR signaling pathway induces autophagy upon bacterial in- (18). Whether AnxA2 is required for phagocytosis of PAO1 by fection remains unclear. A previous study determined that infec- AMs is unknown. Our data showed that phagocytized PAO1 by tion of epithelial cells with Shigella and Salmonella triggers acute anxa22/2 AMs is similar to that of WT AMs, indicating that intracellular amino acid starvation as the result of host membrane AnxA2 does not play a critical role in bacterial uptake by AMs damage, and pathogen-induced amino acid starvation was shown (Fig. 3E–G). Instead, we observed that inhibition of AnxA2 re- to cause downregulation of mTOR activity, resulting in the in- sulted in suppression of autophagosome formation in AMs after duction of autophagy (55). Our data provide evidence that AnxA2 PAO1 infection (Fig. 3H–J). No study has reported a mechanistic regulates pathogen-induced host autophagy through the Akt1- role for AnxA2 in regulating autophagy and bacterial clearance. mTOR signaling pathway. Generally, PAO1 infection stimulates The only previous study demonstrated that AnxA2 expression AnxA2 to interact with free Fam13A in the cytoplasm and acti- increased along with autophagy-regulated genes, such as vates Rho GTPase. Activated Rho GTPase inactivates Akt1 and GABARAP, LC3B, and ATG3, in HUVECs after repeated treat- mTOR; thus, ULK1/2 is released to form the autophagosome (46). ment with resveratrol (53). We found that the reduction in survival In contrast, AnxA2 deficiency impaired Rho GTPase activation; occurred in AnxA2-silenced MH-S cells, and not in MLE-12 cells hence, activated mTOR strongly binds ULK1/2 and inhibits the (Fig. 3C), indicating that AnxA2 might play important roles formation of the autophagosome (Fig. 6I). However, AnxA2 10 AnxA2 REGULATES PSEUDOMONAS-INDUCED AUTOPHAGY IN MACROPHAGES overexpression was strongly associated with rapid recurrence to release ULK1/2 to form autophagosomes during P. aeruginosa after gemcitabine-adjuvant chemotherapy in patients with resected infection. We supposed that AnxA2 affects autophagy-mediated pancreatic cancer. In GEM-MIA PaCa-2 cells, in which AnxA2 is bacterial clearance during P. aeruginosa infection, as well as some highly expressed, the levels of p-Akt and p-mTOR were signifi- other pathways, because autophagy plays diverse important roles, cantly increased (56). Thus, AnxA2 acts as a dual regulator in the such as inflammasome activation (67, 68). Overall, our findings Akt-mTOR signaling pathway that may be independent of cell provide insight into the role of AnxA2 in the regulation of auto- nutrition. phagy against P. aeruginosa and may identify a novel therapeutic Another important finding in this work is that Fam13A is in- approach to combat bacterial infection. volved in the regulation of PAO1-induced autophagy by activat- ing Rho GTPase. Previously, variants in Fam13A were found Acknowledgments in genome-wide association studies to be associated with lung We thank S. Rolling (University of North Dakota Imaging Core Facility) functions in the general population, as well as in several common for help with confocal imaging. We also thank Dr. K. Hajjar for kindly chronic lung diseases, such as COPD, asthma, and idiopathic in- providing anxa22/2 mice and Dr. Xiao-min Yin (Indiana University) for terstitial pneumonias (57). Sequence analysis indicated the pres- providing LC3-RFP G120A plasmids. ence of a Rho GTPase–activating protein domain on exons 2–5 (58). We observed an induction of Fam13A and consistent in- Disclosures duction of activated Rho GTPase in AMs after PAO1 infection, The authors have no financial conflicts of interest. suggesting that Fam13A may mediate Rho GTPase activation instead of inactivation (Figs. 5B, 6H, Supplemental Fig. 2G). Downloaded from Further data determined that overexpression of Fam13A could References directly induce Rho GTPase activation; on the contrary, Fam13A 1. Gerke, V., C. E. Creutz, and S. E. Moss. 2005. Annexins: linking Ca2+ signalling deficiency inhibited Rho GTPase activation in AMs, in the pres- to membrane dynamics. Nat. Rev. Mol. Cell Biol. 6: 449–461. 2. Moss, S. E., and R. O. Morgan. 2004. The annexins. Genome Biol. 5: 219. ence or absence of PAO1 infection (Fig. 6F, 6G, Supplemental 3. Zhang, X., S. Liu, C. Guo, J. Zong, and M. Z. Sun. 2012. The association of Fig. 2E, 2F), indicating that Fam13A directly mediates Rho annexin A2 and cancers. Clin. Transl. Oncol. 14: 634–640. 4. Cui, J. W., and Y. L. Wang. 2013. Expression and function of Annexin II in lung GTPase activation. In addition, we observed a stable association http://www.jimmunol.org/ cancer tissue. Asian Pac. J. Trop. Med. 6: 150–152. between AnxA2 and Fam13A after PAO1 infection (Fig. 6A, 6B), 5. Yang, J., F. Yang, J. Nie, X. Zou, H. Tian, Y. Qin, and C. Liu. 2015. Evaluation of and expression of activated Rho GTPase did not increase in Annexin A2 as a novel diagnostic serum biomarker for lung cancer. Cancer AnxA2-depleted AMs (Fig. 6H). These results indicated that Rho Biomark. 15: 205–211. 6. Huang, Y., Y. Jin, C. H. Yan, Y. Yu, J. Bai, F. Chen, Y. Z. Zhao, and S. B. Fu. GTPase regulator function of Fam13A may be strictly limited to 2008. Involvement of Annexin A2 in p53 induced apoptosis in lung cancer. Mol. AnxA2. Cell. Biochem. 309: 117–123. 7. Dassah, M., D. Almeida, R. Hahn, P. Bonaldo, S. Worgall, and K. A. Hajjar. Rho GTPases are normally identified as key regulators of 2014. Annexin A2 mediates secretion of collagen VI, pulmonary elasticity and cytoskeletal dynamics (59), and they affect several vascular apoptosis of bronchial epithelial cells. J. Cell Sci. 127: 828–844. processes, such as endothelial permeability, cell motility, 8. Fang, Y. T., C. F. Lin, P. C. Liao, Y. M. Kuo, S. Wang, T. M. Yeh, C. C. Shieh, I. J. Su, H. Y. Lei, and Y. S. Lin. 2010. Annexin A2 on lung epithelial cell surface by guest on September 28, 2021 angiogenesis, NO production, smooth muscle contractility, cell is recognized by severe acute respiratory syndrome-associated coronavirus spike proliferation, differentiation, and apoptosis (60, 61). Previous domain 2 antibodies. Mol. Immunol. 47: 1000–1009. studies also indicated that Rho GTPases play roles in the defense 9. Jolly, C., S. Winfree, B. Hansen, and O. Steele-Mortimer. 2014. The Annexin A2/p11 complex is required for efficient invasion of Salmonella typhimurium in against bacterial pathogen invasion. For example, Rho GTPases epithelial cells. Cell. Microbiol. 16: 64–77. are involved in the regulation of pulmonary vascular barrier 10. Gou, D., A. Mishra, T. Weng, L. Su, N. R. Chintagari, Z. Wang, H. Zhang, L. Gao, P. Wang, H. M. Stricker, and L. Liu. 2008. Annexin A2 interactions with function (62) and macropinosome formation and maturation (63), Rab14 in alveolar type II cells. J. Biol. Chem. 283: 13156–13164. and they act as regulators of immunity in plants and animals (64). 11. Andey, T., S. Marepally, A. Patel, T. Jackson, S. Sarkar, M. O’Connell, However, the role of Rho GTPases in the regulation of autophagy, R. C. Reddy, S. Chellappan, P. Singh, and M. Singh. 2014. Cationic lipid guided short-hairpin RNA interference of annexin A2 attenuates tumor growth and especially bacteria-induced autophagy, is still unclear. The only metastasis in a mouse lung cancer stem cell model. J. Control. Release 184: 67– report described that Rho GTPase RhoA positively regulates 78. starvation-induced autophagy through a mechanism that is likely 12. Lyczak, J. B., C. L. Cannon, and G. B. Pier. 2000. Establishment of Pseudo- monas aeruginosa infection: lessons from a versatile opportunist. Microbes In- dependent on actin, whereas Rac1 has the opposite effect (65). We fect. 2: 1051–1060. determined that Rho GTPases are involved in regulation of PAO1- 13. Sousa, A. M., and M. O. Pereira. 2014. Pseudomonas aeruginosa Diversification during Infection Development in Cystic Fibrosis Lungs-A Review. Pathogens 3: induced autophagy, because reduced Rho GTPase activation is 680–703. correlated with dampened autophagosome formation (Fig. 6C, 14. Lovewell, R. R., Y. R. Patankar, and B. Berwin. 2014. Mechanisms of phago- 6F). In the process of autophagosome formation induced by cytosis and host clearance of Pseudomonas aeruginosa. Am. J. Physiol. Lung Cell. Mol. Physiol. 306: L591–L603. starvation, previous studies established that RhoA, a Rho GTPase, 15. Yuan, K., C. Huang, J. Fox, D. Laturnus, E. Carlson, B. Zhang, Q. Yin, H. Gao, triggers a series of signals leading to actin polymerization (65). and M. Wu. 2012. Autophagy plays an essential role in the clearance of Pseu- However, our data demonstrate that, as potential upstream signals, domonas aeruginosa by alveolar macrophages. J. Cell Sci. 125: 507–515. 16. Saitoh, T., N. Fujita, M. H. Jang, S. Uematsu, B. G. Yang, T. Satoh, H. Omori, Rho GTPases regulate autophagosome formation via the Akt- T. Noda, N. Yamamoto, M. Komatsu, et al. 2008. Loss of the autophagy protein mTOR signaling pathway. Gordon et al. (66) supported a model Atg16L1 enhances endotoxin-induced IL-1beta production. Nature 456: 264– 268. in which RhoA GTPase represses mTOR signaling upstream of 17. Delgado, M. A., R. A. Elmaoued, A. S. Davis, G. Kyei, and V. Deretic. 2008. Akt1, which is consistent with our observation. Toll-like receptors control autophagy. EMBO J. 27: 1110–1121. In conclusion, we demonstrated a typical phenotype of P. aer- 18. Kirschnek, S., C. Adams, and E. Gulbins. 2005. Annexin II is a novel receptor 2/2 for Pseudomonas aeruginosa. Biochem. Biophys. Res. Commun. 327: 900–906. uginosa infection in anxa2 mice, suggesting an important role 19. Wos, M., and J. Bandorowicz-Pikuła. 2014. [Participation of annexins in endo- for AnxA2 in bacterial defense in mice. AnxA2 deficiency im- cytosis and EGFR-mediated signal transduction]. Postepy Biochem. 60: 55–61. paired autophagosome formation and bacterial clearance capacity, 20. Wu, M. H., P. C. Chuang, Y. J. Lin, and S. J. Tsai. 2013. Suppression of annexin A2 by prostaglandin E₂ impairs phagocytic ability of peritoneal macrophages in as well as intensified inflammatory responses. Importantly, we women with endometriosis. Hum. Reprod. 28: 1045–1053. revealed a previously unidentified mechanistic role for AnxA2, 21. Drab, M., P. Verkade, M. Elger, M. Kasper, M. Lohn, B. Lauterbach, J. Menne, C. Lindschau, F. Mende, F. C. Luft, et al. 2001. Loss of caveolae, vascular in association with Fam13A, in the activation of Rho GTPases. dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science These activated Rho GTPases, in turn, inactivate Akt1 and mTOR 293: 2449–2452. The Journal of Immunology 11

22. Ling, Q., A. T. Jacovina, A. Deora, M. Febbraio, R. Simantov, R. L. Silverstein, 46. Dunlop, E. A., and A. R. Tee. 2014. mTOR and autophagy: a dynamic re- B. Hempstead, W. H. Mark, and K. A. Hajjar. 2004. Annexin II regulates fibrin lationship governed by nutrients and energy. Semin. Cell Dev. Biol. 36: 121– homeostasis and neoangiogenesis in vivo. J. Clin. Invest. 113: 38–48. 129. 23. Tan, S., C. Gan, R. Li, Y. Ye, S. Zhang, X. Wu, Y. Y. Yang, W. Fan, and M. Wu. 47. Wong, P. M., C. Puente, I. G. Ganley, and X. Jiang. 2013. The ULK1 com- 2015. A novel chemosynthetic peptide with b-sheet motif efficiently kills plex: sensing nutrient signals for autophagy activation. Autophagy 9: 124– Klebsiella pneumoniae in a mouse model. Int. J. Nanomedicine 10: 1045–1059. 137. 24. Kannan, S., H. Huang, D. Seeger, A. Audet, Y. Chen, C. Huang, H. Gao, S. Li, 48. Jung, C. H., C. B. Jun, S. H. Ro, Y. M. Kim, N. M. Otto, J. Cao, M. Kundu, and and M. Wu. 2009. Alveolar epithelial type II cells activate alveolar macrophages D. H. Kim. 2009. ULK-Atg13-FIP200 complexes mediate mTOR signaling to and mitigate P. Aeruginosa infection. PLoS One 4: e4891. the autophagy machinery. Mol. Biol. Cell 20: 1992–2003. 25. Kulasakara, H., V. Lee, A. Brencic, N. Liberati, J. Urbach, S. Miyata, D. G. Lee, 49. Young, R. P., R. J. Hopkins, B. A. Hay, C. F. Whittington, M. J. Epton, and A. N. Neely, M. Hyodo, Y. Hayakawa, et al. 2006. Analysis of Pseudomonas G. D. Gamble. 2011. FAM13A locus in COPD is independently associated with aeruginosa diguanylate cyclases and phosphodiesterases reveals a role for bis- lung cancer - evidence of a molecular genetic link between COPD and lung (39-59)-cyclic-GMP in virulence. Proc. Natl. Acad. Sci. USA 103: 2839–2844. cancer. Appl. Clin. Genet. 4: 1–10. 26. Priebe, G. P., M. M. Brinig, K. Hatano, M. Grout, F. T. Coleman, G. B. Pier, and 50. Gerke, V., and S. E. Moss. 2002. Annexins: from structure to function. Physiol. J. B. Goldberg. 2002. Construction and characterization of a live, attenuated Rev. 82: 331–371. aroA deletion mutant of Pseudomonas aeruginosa as a candidate intranasal 51. Swisher, J. F., U. Khatri, and G. M. Feldman. 2007. Annexin A2 is a soluble vaccine. Infect. Immun. 70: 1507–1517. mediator of macrophage activation. J. Leukoc. Biol. 82: 1174–1184. 27. Lawlor, M. S., J. Hsu, P. D. Rick, and V. L. Miller. 2005. Identification of 52. Woodham, A. W., A. B. Raff, L. M. Raff, D. M. Da Silva, L. Yan, J. G. Skeate, Klebsiella pneumoniae virulence determinants using an intranasal infection M. K. Wong, Y. G. Lin, and W. M. Kast. 2014. Inhibition of Langerhans cell model. Mol. Microbiol. 58: 1054–1073. maturation by human papillomavirus type 16: a novel role for the annexin A2 28. Hart, P. D., and M. R. Young. 1991. Ammonium chloride, an inhibitor of heterotetramer in immune suppression. J. Immunol. 192: 4748–4757. phagosome-lysosome fusion in macrophages, concurrently induces phagosome- 53. Takizawa, Y., Y. Kosuge, H. Awaji, E. Tamura, A. Takai, T. Yanai, R. Yamamoto, endosome fusion, and opens a novel pathway: studies of a pathogenic myco- K. Kokame, T. Miyata, R. Nakata, and H. Inoue. 2013. Up-regulation of endo- bacterium and a nonpathogenic yeast. J. Exp. Med. 174: 881–889. thelial nitric oxide synthase (eNOS), silent mating type information regulation 2 29. Kannan, S., A. Audet, H. Huang, L. J. Chen, and M. Wu. 2008. Cholesterol-rich homologue 1 (SIRT1) and autophagy-related genes by repeated treatments with membrane rafts and Lyn are involved in phagocytosis during Pseudomonas resveratrol in human umbilical vein endothelial cells. Br.J.Nutr.110: 2150– Downloaded from aeruginosa infection. J. Immunol. 180: 2396–2408. 2155. 30. Li, X., Y. Ye, X. Zhou, C. Huang, and M. Wu. 2015. Atg7 enhances host defense 54. Wullschleger, S., R. Loewith, and M. N. Hall. 2006. TOR signaling in growth against infection via downregulation of superoxide but upregulation of nitric and metabolism. Cell 124: 471–484. oxide. J. Immunol. 194: 1112–1121. 55. Tattoli, I., M. T. Sorbara, D. Vuckovic, A. Ling, F. Soares, L. A. Carneiro, 31. Zhou, X., X. Li, Y. Ye, K. Zhao, Y. Zhuang, Y. Li, Y. Wei, and M. Wu. 2014. C. Yang, A. Emili, D. J. Philpott, and S. E. Girardin. 2012. Amino acid starvation MicroRNA-302b augments host defense to bacteria by regulating inflammatory induced by invasive bacterial pathogens triggers an innate host defense program. responses via feedback to TLR/IRAK4 circuits. Nat. Commun. 5: 3619. Cell Host Microbe 11: 563–575.

32. Liang, H., X. Deng, X. Li, Y. Ye, and M. Wu. 2014. Molecular mechanisms of 56. Kagawa, S., S. Takano, H. Yoshitomi, F. Kimura, M. Satoh, H. Shimizu, http://www.jimmunol.org/ master regulator VqsM mediating quorum-sensing and antibiotic resistance in H. Yoshidome, M. Ohtsuka, A. Kato, K. Furukawa, et al. 2012. Akt/mTOR Pseudomonas aeruginosa. Nucleic Acids Res. 42: 10307–10320. signaling pathway is crucial for gemcitabine resistance induced by Annexin II 33. Wu, M., H. Huang, W. Zhang, S. Kannan, A. Weaver, M. McKibben, in pancreatic cancer cells. J. Surg. Res. 178: 758–767. D. Herington, H. Zeng, and H. Gao. 2011. Host DNA repair proteins in response 57. Li, X., G. A. Hawkins, E. J. Ampleford, W. C. Moore, H. Li, A. T. Hastie, to Pseudomonas aeruginosa in lung epithelial cells and in mice. Infect. Immun. T. D. Howard, H. A. Boushey, W. W. Busse, W. J. Calhoun, et al. 2013. Genome- 79: 75–87. wide association study identifies TH1 pathway genes associated with lung 34. Wu, M., S. L. Tang, R. J. Zang, and H. Yu. 1990. Selective killing of tumor cells function in asthmatic patients. J. Allergy Clin. Immunol. 132: 313–320.e15. in vitro by an immunotoxin composed of ricin and monoclonal antibody against 58. Cohen, M., M. Reichenstein, A. Everts-van der Wind, J. Heon-Lee, M. Shani, Ia antigen. Int. J. Immunopharmacol. 12: 235–239. H. A. Lewin, J. I. Weller, M. Ron, and E. Seroussi. 2004. Cloning and charac- 35. Li, G., K. Yuan, C. Yan, J. Fox, III, M. Gaid, W. Breitwieser, A. K. Bansal, terization of FAM13A1–a gene near a milk protein QTL on BTA6: evidence for H. Zeng, H. Gao, and M. Wu. 2012. 8-Oxoguanine-DNA glycosylase 1 defi- population-wide linkage disequilibrium in Israeli Holsteins. Genomics 84: 374–

ciency modifies allergic airway inflammation by regulating STAT6 and IL-4 in 383. by guest on September 28, 2021 cells and in mice. Free Radic. Biol. Med. 52: 392–401. 59. Hall, A., and G. Lalli. 2010. Rho and Ras GTPases in axon growth, guidance, 36. Yan, J., X. Liu, Y. Wang, X. Jiang, H. Liu, M. Wang, X. Zhu, M. Wu, and P. Tien. and branching. Cold Spring Harb. Perspect. Biol. 2: a001818. 2007. Enhancing the potency of HBV DNA vaccines using fusion genes of HBV- 60. Vega, F. M., and A. J. Ridley. 2008. Rho GTPases in cancer cell biology. FEBS specific antigens and the N-terminal fragment of gp96. J. Gene Med. 9: 107–121. Lett. 582: 2093–2101. 37. Yuan, K., C. Huang, J. Fox, M. Gaid, A. Weaver, G. Li, B. B. Singh, H. Gao, and 61. Pedersen, E., and C. Brakebusch. 2012. Rho GTPase function in development: M. Wu. 2011. Elevated inflammatory response in caveolin-1-deficient mice with how in vivo models change our view. Exp. Cell Res. 318: 1779–1787. Pseudomonas aeruginosa infection is mediated by STAT3 protein and nuclear 62. Duluc, L., and B. Wojciak-Stothard. 2014. Rho GTPases in the regulation of factor kappaB (NF-kappaB). J. Biol. Chem. 286: 21814–21825. pulmonary vascular barrier function. Cell Tissue Res. 355: 675–685. 38. Li, G., J. Fox, III, Z. Liu, J. Liu, G. F. Gao, Y. Jin, H. Gao, and M. Wu. 2013. Lyn 63. Egami, Y., T. Taguchi, M. Maekawa, H. Arai, and N. Araki. 2014. Small mitigates mouse airway remodeling by downregulating the TGF-b3 isoform in GTPases and phosphoinositides in the regulatory mechanisms of macropinosome house dust mite models. J. Immunol. 191: 5359–5370. formation and maturation. Front. Physiol. 5: 374. 39. Williams, D. A., Y. Zheng, and J. A. Cancelas. 2008. Rho GTPases and regu- 64. Kawano, Y., T. Kaneko-Kawano, and K. Shimamoto. 2014. Rho family GTPase- lation of hematopoietic stem cell localization. Methods Enzymol. 439: 365–393. dependent immunity in plants and animals. Front. Plant Sci. 5: 522. 40. Kannan, S., A. Audet, J. Knittel, S. Mullegama, G. F. Gao, and M. Wu. 2006. Src 65. Aguilera, M. O., W. Bero´n, and M. I. Colombo. 2012. The actin cytoskeleton kinase Lyn is crucial for Pseudomonas aeruginosa internalization into lung cells. participates in the early events of autophagosome formation upon starvation Eur. J. Immunol. 36: 1739–1752. induced autophagy. Autophagy 8: 1590–1603. 41. Langereis, J. D. 2013. Neutrophil integrin affinity regulation in adhesion, mi- 66. Gordon, B. S., A. A. Kazi, C. S. Coleman, M. D. Dennis, V. Chau, gration, and bacterial clearance. Cell Adhes. Migr. 7: 476–481. L. S. Jefferson, and S. R. Kimball. 2014. RhoA modulates signaling through the 42. Heid, M. E., P. A. Keyel, C. Kamga, S. Shiva, S. C. Watkins, and R. D. Salter. mechanistic target of rapamycin complex 1 (mTORC1) in mammalian cells. 2013. Mitochondrial reactive oxygen species induces NLRP3-dependent lyso- Cell. Signal. 26: 461–467. somal damage and inflammasome activation. J. Immunol. 191: 5230–5238. 67. Jabir, M. S., N. D. Ritchie, D. Li, H. K. Bayes, P. Tourlomousis, D. Puleston, 43. Sunday, M. E. 2014. Oxygen, gastrin-releasing Peptide, and pediatric lung dis- A. Lupton, L. Hopkins, A. K. Simon, C. Bryant, and T. J. Evans. 2014. Caspase- ease: life in the balance. Front Pediatr. 2: 72. 1 cleavage of the TLR adaptor TRIF inhibits autophagy and b-interferon pro- 44. Roychowdhury, S., A. Luthe, G. Keilhoff, G. Wolf, and T. F. Horn. 2002. Oxi- duction during Pseudomonas aeruginosa infection. Cell Host Microbe 15: 214– dative stress in glial cultures: detection by DAF-2 fluorescence used as a tool to 227. measure peroxynitrite rather than nitric oxide. Glia 38: 103–114. 68. Jabir, M. S., L. Hopkins, N. D. Ritchie, I. Ullah, H. K. Bayes, D. Li, 45. Chai, W., J. Zhang, Y. Duan, D. Pan, W. Liu, Y. Li, X. Yan, and B. Chen. 2014. P. Tourlomousis, A. Lupton, D. Puleston, A. K. Simon, et al. 2015. Mitochon- Pseudomonas pyocyanin stimulates IL-8 expression through MAPK and NF-kB drial damage contributes to Pseudomonas aeruginosa activation of the inflam- pathways in differentiated U937 cells. BMC Microbiol. 14: 26. masome and is downregulated by autophagy. Autophagy 11: 166–182.

SI Figure 1. (A-D) Bacterial burdens in blood (A), liver (B), spleen (C) and kidney

(D) were significantly increased after 24 h post-PAO1 infection in anxa2-/- mice compared with WT mice. (E) Viabilities of siNC- and siAnxA2-transfected MH-S cells after 2 h post-infection of PAK, PA14, KP and E. coli, by an MTT assay at a wavelength of 570 nm. (F) Bacterial killing in siNC- and siAnxA2-transfected MH-S cells after 2 h post-infection of PAK, PA14, KP and E. coli, were determined by plate count. Data are representative of three reproducible experiments expressed as means ±

SEM (one-way ANOVA with Tukey’s post hoc; *p ≤ 0.05; **p ≤ 0.005).

SI Figure 2. (A) Viabilities of siNC-, siAtg5-, siAtg7- and siBeclin1-transfected

MH-S cells after 2 h post-infection of PAO1, by an MTT assay at a wavelength of 570 nm. (B) Bacterial killing in siNC-, siAtg5-, siAtg7- and siBeclin1-transfected MH-S cells after 2 h post-infection of PAO1, were determined by plate count. (C) Viabilities of pcDNA3.1-Fam13a- and empty pcDNA3.1- transfected MH-S cells after 2 h post-infection of PAO1, by an MTT assay at a wavelength of 570 nm. (D) Bacterial killing in pcDNA3.1-Fam13a- and empty pcDNA3.1- transfected MH-S cells after 2 h post-infection of PAO1, were determined by plate count. (E-G) Densitometric quantification of the immunoblotting gel data presented in Fig. 6 (in text) using

Quantity One software, respectively for Fig. 6F (E), Fig. 6G (F) and Fig. 6H (G). Data are representative of three reproducible experiments expressed as means ± SEM

(one-way ANOVA with Tukey’s post hoc; *p ≤ 0.05; **p ≤ 0.005).

Table S1. Primers used in this study. Name Sequence APOB F CGGTCTTCAGTGAGGCTACA APOB R GCTCCCATGTGGTGTAGATG Brd4 F AGTGTCCCACCTGTTCTTCC Brd4 R TTAAGGAGAGAAGCCCTCCA Cd1d F TTTGCAAACAGAAGCTGGTC Cd1d R TGCTGGTTACTCAACTTGCC EYA4 F TTCGCTCTCCTTGAACACAC EYA4 R AAGACAGGCAGACACAATGC Fam13A F GCCAGGAAGCAGAAAGATTC Fam13A R TGAGACTTTCCATGCTCAGG Malt1 F GACACTGTGGAGGAGAAGCA Malt1 R AGATGTGGAGCTGTGGTACG Ndusf4 F TTCAGTGCCAAAGAAGATGC Ndusf4 R AGACTTGGACTTGGGTTTCG NSF F GGGTCATAAGTTGCCCAGAT NSF R GGCCTCTGACTGGATGAGTT PINK1 F TGGCTCTGTGCTCCAGTTAC PINK1 R CCCTCTACTCCAGCTTGTCC RBPJ F TGTGCCACCTTTGATTCATT RBPJ R CAATACAGCTCGGTCCTGAA ScgB F TTCTTCATGGACTCATTGGC ScgB R GGCCAAGTGGCTTAATGGTA Sgk1 F GTTGCCAGCTGACAGAACAT Sgk1 R GAGAGGAGGGTGTGCTCTTC Sirga F CACAAGCACTTCAGAGGCAT Sirga R CCAGGTTCTGTTCCAGTGTG Sirt1 F TCCTCTAGTTCCTGTGGCAGT Sirt1 R GGCCTCTCCGTATCATCTTC Tob1 F AATGCCAGCTTTATGGAACC Tob1 R ACCACTGTCAAGCTCTCCCT CCAAGCTTATGGCTTGTGAAATCATG pFam13a F (Hind III) CGGGATCCAATCTCTTTTGCTGATGAG pFam13a R (BamH I)

Table S2. Significant change of host autophagy-related factors in AnxA2 or Fam13A deficient MH-S cells (over 5-fold). siAnxA2 siFam13A vs vs CTRL Gene name Gene description CTRL Fold Fold change change Actb Actin, beta 1.01171 -11.63180 Akt1 Thymoma viral proto-oncogene 1 20.09237 13.13340 Atg10 Autophagy-related 10 (yeast) -3.21075 -5.81637 Atg7 Autophagy-related 7 (yeast) -3.06084 -37.80196 Bcl2 B cell leukemia/lymphoma 2 -1.90966 -7.92708 Cdkn1b Cyclin-dependent kinase inhibitor 1B -2.45685 -6.07543 Cxcr4 Chemokine (C-X-C motif) receptor 4 -2.03963 -15.25398 Dapk1 Death associated 1 -1.38929 -29.69519 DNA-damage regulated autophagy Dram2 -2.57016 -5.72730 modulator 2 Eukaryotic translation initiation factor 2 Eif2ak3 -3.06101 -5.43009 alpha kinase 3 Esr1 Estrogen receptor 1 (alpha) -1.15217 -12.39905 Hdac1 Histone deacetylase 1 -5.33757 -1.98670 Igf1 Insulin-like growth factor 1 -5.44704 -2.66451 Microtubule-associated protein 1 light Map1lc3b -3.51274 -6.02464 chain 3 beta Mapk14 Mitogen-activated protein kinase 14 -2.40791 -5.12917 Mechanistic target of rapamycin Mtor 17.33539 12.79797 (serine/threonine kinase) Nuclear factor of kappa light polypeptide Nfkb1 5.75894 32.52907 gene enhancer in B cells 1, p105 Npc1 Niemann Pick type C1 -3.16719 6.42205 Rb1 Retinoblastoma 1 -2.04819 -5.07060 Snca Synuclein, alpha 1.57903 -40.75448 Ulk1 Unc-51 like kinase 1 (C. elegans) -14.10357 -96.45157 Ulk2 Unc-51 like kinase 2 (C. elegans) -25.53407 -157.82231