Requisite Role of the Cholinergic Α7 Nicotinic Acetylcholine Receptor

Home , Cholinergic

Requisite Role of the Cholinergic α7 Nicotinic Acetylcholine Receptor Pathway in Suppressing Gram-Negative Sepsis-Induced Acute Lung Inflammatory Injury This information is current as of September 28, 2021. Xiao Su, Michael A. Matthay and Asrar B. Malik J Immunol 2010; 184:401-410; Prepublished online 30 November 2009; doi: 10.4049/jimmunol.0901808 http://www.jimmunol.org/content/184/1/401 Downloaded from

References This article cites 34 articles, 6 of which you can access for free at:

http://www.jimmunol.org/content/184/1/401.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 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 © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Requisite Role of the Cholinergic a7 Nicotinic Acetylcholine Receptor Pathway in Suppressing Gram-Negative Sepsis-Induced Acute Lung Inﬂammatory Injury

Xiao Su,*,† Michael A. Matthay,† and Asrar B. Malik*

Although activation of the a7 nicotinic acetylcholine receptor (a7 nAChR) modulates the response to sepsis, the role of this pathway in the development of sepsis-induced acute lung injury (ALI) is not known. In this study, we addressed the contribution of a7 nAChR in mediating endotoxin- and live Escherichia coli–induced ALI in mice. Because we found that a7 nAChR+ alveolar macrophages and neutrophils were present in bronchoalveolar lavage and injured lungs of mice, we tested whether acetylcholine released by lung vagal innervation stimulated these effector cells and thereby down-regulated proinﬂammatory chemokine/ cytokine generation. Administration of a7 nAChR agonists reduced bronchoalveolar lavage MIP-2 production and transalveolar Downloaded from neutrophil migration and reduced mortality in E. coli pneumonia mice, whereas vagal denervation increased MIP-2 production and airway neutrophil accumulation and increased mortality. In addition, a7 nAChR2/2 mice developed severe lung injury and had higher mortality compared with a7 nAChR+/+ mice. The immunomodulatory cholinergic a7 nAChR pathway of alveolar macrophages and neutrophils blocked LPS- and E. coli–induced ALI by reducing chemokine production and transalveolar neutrophil migration, suggesting that activation of a7 nAChR may be a promising strategy for treatment of sepsis-induced ALI. The Journal of Immunology, 2010, 184: 401–410. http://www.jimmunol.org/

he cholinergic antiinflammatory pathway (1) has been also indicated a possible role of specific a7 nAChR agonists GTS- described in a series of experiments showing that vagus 21 (13) and choline (14) or cholinesterase inhibitors (15) for im- T nerve stimulation attenuated the systemic inflammatory proving outcome in experimental sepsis. response to endotoxin (1–3). The a7 nicotinic acetylcholine re- Sepsis-induced acute lung injury (ALI) causes acute respiratory ceptor (a7 nAChR) expressed in macrophages regulates this failure in critically ill patients and has a mortality rate of 40% (16, pathway during inflammation (4) such that cholinergic activation 17). The most common causes of ALI are pneumonia and sepsis of these cells dampens the inflammatory response. The studies (17, 18). Although activation of the a7 nAChR modulates the

showed that activation of a7 nAChR by agonists (e.g., nicotine) response to sepsis, the role of this pathway in the development by guest on September 28, 2021 downregulated expression of HMGB1 protein (a late mediator of of sepsis-induced ALI is not known. Our objectives here are to sepsis) and improved survival (5, 6). Stimulation of the cholin- determine the role of activation of alveolar macrophages and ergic antiinflammatory pathway also appears to play a protective neutrophils expressing a7 nAChR on the production of proin- role in other inflammatory models, peritonitis (7, 8), and renal flammatory cytokines and chemokines, whether a7 nAChR affects ischemia and reperfusion injury (9). the ALI response induced by endotoxin and Escherichia coli Activation of a7 nAChR in macrophages and monocytes may pneumonia, and whether antagonism or deficiency of a7 nAChR downregulate production of proinflammatory cytokines and at- activity and vagus denervation has the opposite effects of lung tenuate the inflammatory response by several possible but poorly inflammation and injury and survival in mouse models. understood mechanisms: 1) suppression of NF-kB translocation k and I- B phosphorylation (5, 10), 2) activation of Jak2-STAT3 Materials and Methods signaling (11), and 3) inhibition of expression of LPS receptors Reagents and binding proteins CD14 and TLR4 (12). Some studies have (-)-nicotine, acetylcholine (ACh), methyllycaconitine (MLA), and LPS were purchased from Sigma-Aldrich (St. Louis, MO). Dimethylamino- *Center for Lung and Vascular Biology and Department of Pharmacology, University benzaldehyde (DMAB), PNU 282987 (PNU), and PHA 568487 (specific of Illinois at Chicago, Chicago, IL 60612; and †Cardiovascular Research Institute, agonists of a7 nAChR) were purchased from Tocris Bioscience (Ellisville, University of California, San Francisco, CA 94143 MO) and dissolved in 0.9% saline before each experiment. H-302, an anti- Received for publication June 8, 2009. Accepted for publication October 23, 2009. a7 nAChR Ab used to detect a7 nAChR of mouse and human origin, was This work was supported by National Heart, Lung, and Blood Institute HL-51854 purchased from Santa Cruz Biotechnology (Santa Cruz, CA). PE anti- and HL-51856 (to M.M.); R01-HL-045638 (to A.M.); and a Parker B. Francis Award mouse CD11, Ly-6G (Gr-1, Gr1), and corresponding isotype Abs were (to X.S.). purchased from eBioscience (San Diego, CA). Address correspondence and reprint requests to Dr. Xiao Su, Center for Lung and Vascular Biology, Department of Pharmacology, University of Illinois at Animals Chicago, E403 MSB, 835 S. Wolcott Ave, Chicago, IL 60612. E-mail address: [email protected] Most experiments were done with CD1 mice (purchased from Charles River Laboratories, Wilmington, MA). a7 nAChR–deficient mice (C57BL/6 Abbreviations used in this paper: ACh, acetylcholine; AChE, acetylcholinesterase; background, B6.129S7-Chrna7tm1Bay, number 003232) and wild type ALI, acute lung injury; BAL, bronchoalveolar lavage; DMAB, dimethylaminoben- zaldehyde; ELW, extravascular lung water; EPE, extravascular plasma equivalent; IT, (WT) littermates (C57BL/6J, 8 wk old) were purchased from The Jackson intratracheal; MLA, methyllycaconitine; a7 nAChR, a7 nicotinic acetylcholine re- Laboratory (Bar Harbor, ME) (4). Anesthesia was induced with an i.p. in- ceptor; PNU, PNU 282987; WT, wild type. jection of a mixture of ketamine (90 mg/kg) and xylazine (10 mg/kg). The Committees on Animal Research of the University of Illinois at Chicago and Copyright Ó 2009 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 the University of California, San Francisco approved the protocols. www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901808 402 ACTIVATION OF a7 nAChR PREVENTS ACUTE LUNG INJURY

Isolation and culture of alveolar macrophages Unilateral vagotomy Alveolar macrophages were isolated by a bronchoalveolar lavage (BAL) Right or sham cervical vagotomy was performed with the animals under an- (19, 20). Greater than 90% alveolar macrophage purity was confirmed with esthesia. The procedure involved a longitudinal midline incision in the ventral a cytospin preparation (Cytospin 3, Thermo Electron, Milford, MA) and region of the neck. Using blunt dissection, the overlying muscles and fascia Hema 3 staining (Fisher Scientific, Kalamazoo, MI). Alveolar macro- were separated until the right vagus and carotid artery were visible. The vagus phages were cultured in RPMI 1640 medium (2.5 3 105 /ml). a7 nAChR was carefully stripped away from carotid artery and lightly cut off in the va- agonists or an antagonist (MLA) were added 20 min before LPS stimu- gotomygroup.Thevaguswaskeptintactinshamgroup.Thewoundwasclosed lation (3 mM). PBS was used as a negative control. The media was col- and sutured. The respiration rhythm was not affected by unilateral vagotomy. lected after 12 h incubation for TNF-a and MIP-2, and after 20 h for HMGB1 measurements. BAL and plasma cytokine and protein concentration measurements Neutrophil isolation and culture BAL and plasma samples were obtained from mice at designated time points. Supernatant was used to measure protein and cytokines. TNF-a and As described (20), mice were euthanized by cervical dislocation, and the MIP-2 were measured with ELISA kits (R&D Systems, Minneapolis, bone marrow from the femurs and tibias was flushed with PBS using a 25- MN). Protein concentration was measured by a Bio-Rad protein assay kit gauge needle. The whole bone marrow was centrifuged and washed in (Hercules, CA). PBS, and red blood cells were hypotonically lysed with 0.2% NaCl. This solution was resumed to isotonicity with 1.2% NaCl and then filtered over Measurement of leukocyte, neutrophil, and monocyte counts in a 70-mm nylon cell strainer (BD Discovery Labware, Bedford, MA). The BAL solution was centrifuged and resuspended in PBS and then delicately applied over a 62% Percoll gradient. The Percoll solution was centrifuged BAL leukocytes, neutrophils, and monocytes were measured by Hemavet for 30 min at 1500 3 g. The neutrophil pellet was then isolated, washed, 950FS (Drew Scientific, Dallas, TX). Morphology of BAL cells was de- Downloaded from and centrifuged twice, and counted with a Coulter counter (Z1 series; termined by BAL cytospin and Hema 3 staining. Beckman Coulter, Fullerton, CA). Greater than 90% neutrophil purity was confirmed with the cytospin preparation and Hema 3 staining. Neutrophils Determination of lung myeloperoxidase activity 6 a were suspended in RPMI 1640 medium (10 /ml). 7 nAChR agonists or As described (20), supernatants of lung homogenate were mixed with o- an antagonist (MLA) were added 20 min before LPS stimulation (3 mM). Dianisidine HCl (Sigma-Aldrich) and H2O2 to measure optical density by PBS was used as a negative control. The media was collected after 12 h a spectrophotometer at 405 nm. incubation for measuring MIP-2 or TNF-a concentration. http://www.jimmunol.org/ HMGB1 enzyme-linked immunosorbent assay LPS-induced ALI mouse model To measure HMGB1 using ELISA in this study, we used a capture Ab (clone HAP46.5, anti-HMGB1 monoclonal Ab; Abcam, Cambridge, MA), which As described, mice were intratracheally (IT) instilled with LPS (5 mg/kg) by was diluted in coating buffer and added to a 96-well plate overnight at 4˚C. a direct visualized instillation method (21). Mice were monitored for After blocking and washing, standard curve was made by adding re- 24 and 48 h, and killed to perform BAL or measure extravascular lung combinant human HMGB1 (H4652; Sigma-Aldrich; 500, 250, 125, 62.5, water (ELW). 31, 16, and 8 ng/ml) to each well and incubated at room temperature for 2 h. Detection Ab (anti-mouse RAGE Ab labeled with HRP; SureLINK Acute E. coli pneumonia mouse model and survival study HRP Conjugation Kit; Gaithersburg, MD) was added and incubated at

room temperature for 2 h. Color was developed by adding substrate A and by guest on September 28, 2021 Live E. coli were obtained from American Type Culture Collection, B mixture (R&D Systems, Minneapolis, MN), and the plate was read at Manassas, VA (ATCC 25992) (22). E. coli (107 CFU) was instilled into the 450 nm in a spectrophotometer. air spaces of lungs. Immediately before exposure to E. coli, mice received 0.05 mCi [125I]-albumin via the right jugular vein. Mice were monitored Measurement of acetylcholinesterase activity and choline levels for 4 h and killed to measure ELW and lung endothelial permeability to Acetylcholinesterase (AChE) activity of BAL cells was measured as pre- protein or to perform BAL. viously described (23). Briefly, 105 BAL cells (sonicated after removal of We had found a correlation between the degree of lung injury and red blood cells) were mixed with 5,59-dithiobis-2-nitrobenzoic acid mortality and the dose of IT E. coli. Therefore, different doses were used (Sigma-Aldrich) to measure OD [A], and then acetylthiocholine io- depending on the experimental objectives: 1) E. coli (107 CFU) for early 405nm dide (Sigma-Aldrich) to measure OD [B] at 30 min. AChE activity experiments at 4 h to ensure that the lungs were substantially injured; 2) E. 405nm (DOD ) was calculated by OD [B] – OD405[A]. Choline in BAL coli (2.5 3 106 CFU) for longer experiments (24 h) to ensure that there 405nm 405 was measured by a Choline/Acetylcholine Quantification kit (BioVision was no death in both control and treated groups; 3) E. coli (5 3 106 CFU) Research Products, Mountain View, CA). for the survival study, to be certain that some mice died of substantial lung injury within 24 h, which facilitated observing the difference between the Detection of a7 nAChR expression in alveolar macrophages control and treated groups; and 4) E. coli (4 3 106 CFU) for the survival 2 2 and neutrophils by immunofluorescence study with vagotomized or a7 nAChR / mice, because they were more susceptible to the E. coli infection compared with the WT mice. As described (19), using cytospin, alveolar macrophages, or neutrophils were prepared on slides and fixed in 4% paraformaldehyde. The smear was permeabilized in 0.25% triton and incubated with 1:67 rabbit anti-mouse a7 ELW and lung extravascular plasma equivalent nAChR Ab (H-302). Then, goat-anti-rabbit fluorescein isothiocyanate As described (19), homogenate and supernatant of lung, and blood were (FITC)-labeled secondary Ab (Santa Cruz Biotechnology) was added on the weighed and then desiccated in an oven (60˚C for 24 h). ELW was cal- smear and incubated for 1 h. After washing with hypertonic PBS (2.7% culated by standard formula: NaCl), the slides were mounted with Vectashield Mounting Medium (Vector ÂÀ Á Laboratories, Burlingame, CA). For immunofluorescence in frozen lung ELW ¼ Q =Q 3 Q sections, lungs were collected at endpoints of experiments and embedded in À W exp d exp d exp ÁÃ OCT (Sakura FineTek, Torrance, CA), and frozen on dry ice. Frozen sections m − QW control=Qd control 3 Qd control (5 m) were prepared in a cryostat (Microm, Walldorf, Germany). The lung À Á sections was permeablized in 0.2% triton and incubated with the following 3 1000 ml Abs: 1:50 rabbit anti-mouse a7 nicotinic ACh receptor Ab (H302; Santa Cruz Biotechnology); 1:50 rat anti-mouse CD11b monoclonal Ab (M1/70; where QW exp equals water volume of the lung in the experimental group; BD Pharmingen, San Diego, CA), or 1:50 rat anti-mouse Gr1 Ab (LEW; BD Qd exp equals dry weight of lung in the experimental group. The controls Pharmingen). Then, anti-rabbit or rat Alexa Fluor 594 or 488 labeled sec- were the normal mice with the same age as the experimental group. ondary Abs (Invitrogen, Carlsbad, CA) was added on the smear and in- Lung extravascular plasma equivalents (EPEs; index of lung vascular cubated for 1 h. After washing with hypertonic PBS (2.7% NaCl), the slides permeability to protein) were calculated as the counts of [125I]-albumin were mounted with ProLong Gold antifade reagent (Molecular Probes). The in the blood free lung tissue divided by the counts of [125I]-albumin in fluorescence was visualized and analyzed using a digitized fluorescence the plasma. microscope (Carl Zeiss, Thornton, NY). The Journal of Immunology 403

Flow cytometric analysis olar macrophages were pretreated separately with PBS, 1027 M 26 25 Monoclonal Abs PE anti-mouse CD11b (M1/70), Gr1 (RB6-8C5), and IgG nicotine or ACh, 10 M nicotine or ACh, or 10 M nicotine or isotype controls were obtained from eBioscience. Rabbit anti-mouse a7 ACh. At 20 min, the cells from these four groups were stimulated nAChR and control Abs were obtained from Santa Cruz Biotechnology. with LPS (3 mM) for 20 h to measure HMGB1 in the media. All samples were pretreated with Fc receptor blocking reagent to prevent Untreated cells with PBS challenge were used as controls. LPS nonspecific binding. After BAL, red blood cells were lysed. The BAL cell stimulation increased HMGB1 production in alveolar macrophages pellets were washed with PBS (2.5% FCS), fixed in 2% paraformaldehyde, 6 6 permeabilized with 0.2% Triton, and then labeled with corresponding Abs. (3998 204 versus 28 6 pg/ml in the PBS stimulated group. 26 25 Fluorescent cells were analyzed after exclusion of debris and aggregates Nicotine reduced HMGB1 by 25 and 32% at 10 and 10 M, with CyAn ADP or MoFlo (DakoCytomation, Carpinteria, CA). Data were respectively (Fig. 1M). ACh suppressed HMGB1 by 30% at 1027 to analyzed by Summit 4.3 software (DakoCytomation). 1025 M (Fig. 1N). a7 nAChR Western blotting of BAL samples a As described (19), denatured proteins (20 mg) from BAL cells were loaded Neutrophils express 7 nAChR and run on a 4–12% gradient Bis-Tris gel (Invitrogen, Carlsbad, CA). The To determine whether neutrophils express a7nAChR,wecol- proteins were then transferred to a nitrocellulose membrane and incubated 3 6 with anti-a7 nAChR Ab (H-302). Membranes were exposed to HRP- lected BAL cells from E. coli pneumonia (2.5 10 CFU, IT, labeled Ab and developed with an ECL kit (Amersham, Piscataway, NJ). sacrificed at 12 h) to perform immunofluorescence and Western blotting. The smear of BAL cells by Hema 3 staining demon- Statistical analysis strated that the number of neutrophils (Fig. 2A) was increased in

Statistical analysis was done with SPSS software (SPSS, Chicago, IL). An BAL from E. coli pneumonia compared with normal BAL cells, Downloaded from unpaired t test was used unless there were multiple comparisons, in which which are mainly alveolar macrophages (Fig. 1A). By Western case we used ANOVAwith post hoc Bonferoni test (significance level was blotting, we observed that the expression of a7 nAChR in BAL set at p , 0.05). The log-rank test was used for comparing survival data by GraphPad Prism software (GraphPad, San Diego, CA). The results are cells from E. coli pneumonia (alveolar macrophages mixed with shown as mean 6 SD. neutrophils) was markedly increased compared with normal BAL cells (Fig. 2B). Immunofluorescence demonstrated that BAL neutrophils from pneumonia expressed a7 nAChR in the cell Results http://www.jimmunol.org/ Alveolar macrophages express a7 nAChR membrane and cytoplasm (Fig. 2C,2D, segmented neutrophils). To confirm that neutrophils expressed a7 nAChR, BAL cell pel- In normal BAL cells, alveolar macrophages are dormant or resting lets (from pneumonia mice at 12 h) were double stained with anti- a cells (Fig. 1A). 7 nAChR immunoreactivity was found in the cell mouse Gr1 (a surface marker of neutrophils) and a7 nAChR Abs membrane and cytoplasm (Fig. 1B,1C). Western blotting dem- to perform flow cytometry. We observed that 43.6 6 1.8% of BAL onstrated that normal BAL cells (mainly alveolar macrophages) cells coexpressed Gr1 and a7 nAChR (Fig. 2E,2F), confirming a expressed 7 nAChR with a band at 55 kDA, using PC12 cells and the finding that neutrophils express a7 nAChR. To investigate rat brain extract as positive controls (Fig. 1D). To study alveolar coexpression of a7 nAChR and Gr1 in E. coli–injured lungs, lung macrophage cell surface expression of a7 nAChR, normal BAL sections were obtained from pneumonia mice at 12 h after by guest on September 28, 2021 cell pellets were double stained with anti-mouse CD11b (Mac-1, IT challenge of E. coli to perform immunostaining. Coexpression of a a surface marker of macrophages) and 7 nAChR Abs to perform a7 nAChR and Gr1 was found in injured lung sections (Fig. 2G–J). flow cytometry. We observed that 14.7 6 4.8% of normal BAL cells coexpressed CD11b and a7 nAChR (Fig. 1E,1F). To in- a7 nAChR activation downregulates TNF-a and MIP-2 vestigate coexpression of a7 nAChR and CD11b in E. coli– production in neutrophils injured lungs, lung sections were obtained from pneumonia mice To investigate whether activation of a7 nAChR agonists by nic- at 12 h after IT challenge of E. coli for immunostaining. Coex- otine, DMAB (a novel partial a7 nAChR agonist), or PNU (a pression of a7 nAChR and CD11b was found in injured lung highly specific a7 nAChR agonist, K 27 nM) affect MIP-2 pro- sections (Fig. 1G–J). i duction in neutrophils, neutrophils were pretreated separately with Effects of nicotine and MLA on TNF-a and MIP-2 production PBS; 1027 M nicotine, DMAB or PNU; 1026 M nicotine, DMAB in alveolar macrophages or PNU; or 1025 M nicotine, DMAB or PNU. At 20 min, cells m To test whether activation of a7 nAChR by nicotine in alveolar from these four groups were stimulated with LPS (3 M) for 12 h macrophages alters proinflammatory cytokine production and to measure MIP-2 concentration in the media. Untreated cells with whether MLA (a specific a7 nAChR antagonist) reverses the PBS challenge were used as controls. The MIP-2 concentration in 6 effects of nicotine, alveolar macrophages were pretreated sepa- LPS-stimulated neutrophils was increased (1124 84 pg/ml) 2 2 2 6 rately with PBS, 10 6 M nicotine, or 10 6 M + MLA 10 6 M compared with 11 3 pg/ml in PBS group. Nicotine reduced MIP-2 by 23–40% at 1027 to 1026 M, and 50% at 1025 M (Fig. nicotine. After 20 min, cells from these three groups were stim- 27 25 ulated with LPS (3 mM) for 12 h to measure TNF-a and MIP-2 in 2K). DMAB inhibited MIP-2 by 51–53% at 10 to 10 M (Fig. 2L), and PNU suppressed MIP-2 by 49% at 1027 to 1026 M and the media. Untreated cells with PBS challenge were used 25 as controls. LPS stimulation increased MIP-2 (545 6 24 versus 30% at 10 M (Fig. 2M). 31 6 18 pg/ml in PBS group) and TNF-a (704 6 6 versus 16 6 a7 nAChR-deficient neutrophils propagate proinflammatory 7 pg/ml in PBS group) production in alveolar macrophages. cytokine production Nicotine reduced LPS-induced MIP-2 and TNF-a production. MLA counteracted this effect of nicotine on TNF-a and MIP-2 To test whether deficiency of a7 nAChR in neutrophils facilitates production (Fig. 1K,1L). proinflammatory cytokine production, the isolated bone marrow neutrophils from a7 nAChR+/+ and a7 nAChR2/2 mice were Effects of nicotine and ACh on HMGB1 production in alveolar stimulated with LPS (3 mM). The media was collected at 12 h. macrophages After LPS stimulation, TNF-a and MIP-2 concentrations in the To test whether activation of a7 nAChR agonists by nicotine and media from a7 nAChR2/2 neutrophils were increased compared ACh affects HMGB1 production in alveolar macrophages, alve- with a7 nAChR+/+ neutrophils (Fig. 2N,2O). 404 ACTIVATION OF a7 nAChR PREVENTS ACUTE LUNG INJURY Downloaded from http://www.jimmunol.org/

FIGURE 1. Alveolar macrophages express a7 nAChR. A, Morphology of normal BAL cells. B and C, a7 nAChR immunofluorescence in normal BAL cells (mainly alveolar macrophages). B, Control a7 nAChR Ab. C, Rabbit anti-mouse a7 nAChR Ab (3 200; second Ab labeled with Alex Fluor-488). D, Western blot for a7 nAChR in normal BAL cells. PC12 cells and rat brain extract were used as positive controls. E and F, a7 nAChR and CD11b are coexpressed in normal BAL cells identified by flow cytometry. E, Control Abs. F, CD11b and a7 nAChR Abs (repeated 4 times). G–J. a7 nAChR and CD11b coexpression in the lung sections from E. coli pneumonia at 12 h. G, Control Abs followed by DAPI staining. H–J, Anti-CD11b (H) and anti-a7 nAChR (I) Abs staining followed by DAPI (3 200; second Ab labeled with Alexa Fluor-594 and Alexa Fluor-488). K and L, Inhibitory effects of nicotine on proinflammatory cytokines in LPS-stimulated alveolar macrophages (isolated from normal BAL cells), which is reversed by MLA (an a7 nAChR by guest on September 28, 2021 antagonist). K, MIP-2. L, TNF-a. Cells were treated with PBS, nicotine, or nicotine + MLA, then stimulated with LPS. pp , 0.05 versus LPS only; n = 3–5 in each group. M and N, HMGB1 levels in the media of cultured alveolar macrophages, which were treated with nicotine or ACh and then stimulated with LPS for 20 h. M, Effect of nicotine. N, Effect of ACh. pp , 0.05, ppp , 0.01 versus LPS only; n = 3–5 in each group. Data are mean 6 SD. a7 nAChR activation attenuates LPS-induced ALI ceived saline IT and saline i.p. In the LPS + saline group, more TostudywhetheradministrationofnicotinesuppressesLPS-induced consolidation, hemorrhage, neutrophil infiltration, and interstitial ALI, mice were administrated nicotine (0.4 mg/kg, i.p., every 6 h). thickness (Fig. 3H) were found compared with the control group (Fig. The first dose was given 15 min before LPS (5 mg/kg, IT), and the 3G). Administration of nicotine, DMAB, or PNU markedly reduced control group received the same volume of saline (5) for 24 h. Lungs the consolidation, hemorrhage, neutrophil infiltration, and interstitial thickness compared with the LPS + saline group (Fig. 3I–K). were lavaged to measure protein, and then homogenized to measure HMGB1 and blood was collected to measure plasma MIP-2. The a7 nAChR modulates lung inflammation and edema by BAL protein (an index of permeability of the lung vascular and suppressing inflammatory mediator production in local milieu epithelial barriers) was decreased in the nicotine-treated group (Fig. To test whether activation of a7 nAChR reduces ELWand lung EPE 3A). HMGB1 levels in lung homogenate were also reduced by in E. coli pneumonia, we divided mice into four groups: 1) saline + nicotine therapy (Fig. 3B). MIP-2 generation in the plasma (a sur- saline group, in which the mice were pretreated with saline and then rogate of the systemic inflammatory response) was abolished by IT saline; 2) nicotine + saline group, in which the mice were treated nicotine treatment (Fig. 3C). with nicotine (3.5 mg/kg, i.v.) (19) 5–10 min before IT saline; 3) To compare the effects of three different a7 nAChR agonists saline + E. coli (107 CFU; the lungs were substantially injured at 4 h (nicotine, DMAB, and PNU) on LPS-induced ALI, three agonists at this dose) group, in which the mice were pretreated with saline and were separately given (0.4 mg/kg, i.p., every 6 h). The first dose was then IT E. coli; and 4) nicotine + E. coli (107 CFU) group, in which given 15 min before IT challenge with LPS (5 mg/kg). The control mice were pretreated with nicotine and then IT E. coli. At 4 h, blood group received the same volume of saline. At 24 h, body weight loss was withdrawn and the lungs were removed to measure ELW and in the agonist-treated group was reduced compared with the saline lung EPE. ELW and lung EPE in the saline + E. coli group were group (Fig. 3D). At 48 h, myeloperoxidase activity in lung homo- increased eightfold to ninefold compared with the saline + saline genates in the nicotine, DMAB, and PNU groups was lower than in group. ELW and lung EPE in the nicotine + E. coli group were re- the saline group (Fig. 3E). ELW in the agonist-treated groups was duced compared with the saline + E. coli group at 4 h (Fig. 4A,4B). also reduced compared with the saline-treated group (Fig. 3F). We also used the same experimental design to determine BALTNF- To test for histologic differences, LPS-challenged mice (5 mg/kg, a and MIP-2 levels and neutrophil numbers. Nicotine reduced the IT)weregivensaline,nicotine,DMAB,orPNU(0.4mg/kg,i.p.,every BAL neutrophil counts and TNF-a and MIP-2 levels compared with 6 h) separately and then euthanized after 24 h. The control mice re- the saline group at 4 h (Fig. 4C–E). The Journal of Immunology 405 Downloaded from http://www.jimmunol.org/

FIGURE 2. Neutrophils express a7 nAChR. A, Cytologic change of BAL cells (enriched in neutrophils) from E. coli pneumonia mice at 12 h. B, Western blot for a7 nAChR in BAL cells from normal and E. coli pneumonia at 4 h (equal quantity loading in protein). C and D, a7 nAChR immunofluorescence in pneumonia BAL cells (segmented neutrophils, arrow heads). C, Control a7 nAChR Ab. D, Rabbit anti-mouse a7 nAChR Ab (3 200; second Ab labeled with Alex Fluor-488). E and F, Neutrophils (Gr1+ cells) coexpress a7 nAChR in BAL cells from pneumonia at 12 h confirmed by flow cytometry (repeated 3 times). E, Control Abs. F, BAL cells were labeled with anti-a7 nAChR and Gr1 Abs. G–J. a7 nAChR and Gr1 (a neutrophil marker) coexpression in the by guest on September 28, 2021 lung sections from E. coli pneumonia at 12 h. G, Control Abs followed by DAPI staining. H–J. Anti-Gr1 (H) and anti-a7 nAChR (I) Abs staining followed by DAPI (3 200; second Ab labeled with Alexa Fluor-594 and Fluor-488). K–M. Effect of nicotine, DMAB, and PNU (specific a7 nAChR agonist) on production of MIP-2 in LPS-stimulated neutrophils. K, Nicotine. L, DMAB. M, PNU. pp , 0.05, ppp , 0.01 versus LPS only; n = 5–6 in each group. N and O, Deficiency of a7 nAChR in neutrophils increases production of proinflammatory cytokines under LPS-stimulation. N, TNF-a. O, MIP-2. ppp , 0.01 versus WT; n = 8 in each group. Data are mean 6 SD.

To address whether activation of a7 nAChR by nicotine affects BAL collected. BAL from normal mice was used as a control. E. coli pneumonia at a 24-h time frame, mice were treated with AChE activity in BAL cells and choline levels in the BAL were either nicotine (0.4 mg/kg, i.p.) or saline 15 min before IT chal- increased in E. coli pneumonia mice at 12 h (Fig. 4I,4J) compared lenge with 2.5 3 106 CFU E. coli (at this dose, there was no death with normal mice. There was no difference in BAL ACh levels in both control and treated groups within 24 h of the experiment). between control (2.1 6 2.4 nmol/ml) and E. coli pneumonia The same therapy was repeated every 6 h (the total dose 2.4 mg/kg groups (3.2 6 2.1 nmol/ml) at 12 h. within 24 h) (5). At 24 h, ELW in pneumonia treated with nicotine a7 nAChR pathway regulates proinflammatory cell was decreased compared with pneumonia treated with saline (Fig. transmigration and transalveolar permeability 4F). Because HMGB1 is a later mediator of sepsis and in- a flammation, lung HMGB1 levels were measured. HMGB1 levels To study the effects of nicotine (a nonselective agonist of 7 nAChR) in lung homogenates after nicotine therapy were reduced com- on transmigration of proinflammatory cells into the airspaces as well pared with saline therapy (Fig. 4G), indicating that activation of a7 as lung epithelial and endothelial permeability, mice were pretreated nAChR attenuates lung inflammation and injury by reducing with either nicotine (0.4 mg/kg, i.p.) or saline 15 min before IT 3 6 HMGB1 locally, also in the late time frame. To address whether challenge with E. coli (2.5 10 CFU). The therapy was repeated alackofa7 nAChR worsens pulmonary edema, a7nAChR+/+ and every 6 h. The mice were killed at 4, 12, and 24 h. Normal mice were a7 nAChR2/2 mice were instilled ITwith E. coli (2.5 3 106 CFU) used as controls. BAL was collected to measure numbers of leu- and sacrificed at 12 h. ELW in a7nAChR2/2 mice was 2.3-fold kocytes, neutrophils, and monocytes. BAL protein (an index of lung higher compared with a7 nAChR+/+ mice (Fig. 4H). epithelial and endothelial permeability) was also analyzed. At 4 h, BAL leukocyte (1.15 6 0.34 k/ml versus 0.62 6 0.14 k/ml) and AChE activity and choline concentration in BAL in E. coli neutrophil counts (0.125 6 0.017 versus 0.065 6 0.017 k/ml) were pneumonia reduced (Fig. 5A,5B) with nicotine therapy. At 12 h, BAL leukocyte, To study AChE activity in alveolar proinflammatory cells and BAL neutrophil, and monocyte counts showed a trend in reduction in the choline concentrations in E. coli pneumonia, mice were chal- nicotine-treated group (Fig. 5A–C). At 24 h, there were no differ- lenged IT with E. coli (2.5 3 106 CFU) and sacrificed at 12 h with ences in leukocyte, neutrophil, and monocyte counts (Fig. 5A–C) 406 ACTIVATION OF a7 nAChR PREVENTS ACUTE LUNG INJURY Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 3. Activation of a7 nAChR by nicotine protects against LPS-induced ALI at 24 and 48 h. At 24 h, nicotine therapy reduced BAL protein (A), HMGB1 in the lung homogenate (B), and plasma MIP-2 (C). pp , 0.05 versus saline; n = 5 in each group. Data were pooled from 2 different experiments. D–F. Activation of a7 nAChR by its agonists nicotine, DMAB, or PNU attenuates LPS-induced ALI. Body weight loss at 24 h (D); pp , 0.05 versus saline + LPS. Lung myeloperoxidase activity at 48 h (E); ppp , 0.01 versus saline only; #p , 0.05 versus saline + LPS. ELW at 48 h (F); pp , 0.05 versus saline + LPS; n = 5 in each group. Data were pooled from 4 different experiments. Data are mean 6 SD. G–K. Representative photomicrographs of lung histology in LPS-induced ALI at 24 h treated with a7 nAChR agonists nicotine, DMAB, or PNU. Magnification 3400; scale bar = 50 mm; n = 2–3 in each group. compared with nicotine- and saline-treated groups. BAL protein epithelial migration of proinflammatory cells and lung epithelial levels were diminished at 4, 12, and 24 h compared with nicotine- and vascular permeability. Therefore, mice with denervation of and saline-treated groups (Fig. 5D). vagus were used to test this hypothesis. Mice were first surgically To test the effects of PHA 568487 (a specific agonist of a7 prepared with a unilateral cervical vagotomy (right side) or sham nAChR) on transmigration of proinflammatory cells into the air- operation and then instilled IT with E. coli (2.5 3 106 CFU). The space and lung epithelial and endothelial permeability, mice were mice were killed at 4, 12, and 24 h. The right lungs were lavaged. pretreated with either PHA 568487 (0.4 mg/kg, i.p.) or saline 15 Normal mice were used as controls. BAL was collected to mea- min before IT challenge with E. coli (2.5 3 106 CFU). The sure the numbers of leukocytes, neutrophils, and monocytes, and therapy was repeated every 6 h. The mice were killed at 4, 12, and BAL protein levels. BAL leukocyte, neutrophil, and monocyte 24 h. Normal mice were used as controls. BAL was collected to counts were increased at 12 and 24 h in the vagotomized group measure numbers of leukocytes, neutrophils, and monocytes, and (Fig. 5I–K). BAL protein levels in vagotomized group were also BAL protein levels. At 4 h, BAL leukocyte (0.32 6 0.02 versus 0.16 increased compared with the sham group at 4 and 12 h (Fig. 5F). 6 0.04 k/ml) and neutrophil counts (0.09 6 0.05 verus 0.01 6 0k/ a ml) were reduced (Fig. 5E,5F) with PHA 568487 therapy. At 12 h, 7 nAChR pathway negatively regulates MIP-2 production BAL neutrophil and monocyte counts showed a trend in reduction in To study the effects of nicotine (a nonselective agonist of a7 the PHA 568487-treated group (Fig. 5F,5G). At 24 h, BAL neu- nAChR) on MIP-2 production in BAL and circulation, mice were trophil count was lower in the PHA 568487-treated group (Fig. 5F). pretreated with either nicotine (0.4 mg/kg, i.p.) or saline 15 min BAL protein levels were lower at 4, 12, and 24 h in the PHA 568487- before IT challenge with E. coli (2.5 3 106 CFU). The therapy was treated groups (Fig. 5H). repeated every 6 h. The mice were killed at 4, 12, and 24 h. Normal Based on the above data, we hypothesized that the vagus nerve– mice were used as controls. BAL and plasma were collected to mediated ACh-a7 nAChR pathway regulated transendothelial or measure MIP-2 levels by ELISA. The MIP-2 levels in the BAL were The Journal of Immunology 407 Downloaded from http://www.jimmunol.org/

FIGURE 4. a7 nAChR is a key regulator of lung inflammation and injury in E. coli pneumonia. Administration of nicotine reduces pulmonary edema (A) and lung vascular permeability (B)inE. coli pneumonia at 4 h. Mice were pretreated with either saline or nicotine (3.5 mg/kg, i.v.), then instilled IT with E. coli (107 CFU), and killed at 4 h. ppp , 0.01 versus saline + saline and nicotine + saline; #p , 0.05 versus saline + E. coli; n = 4–6 in each group. Data were pooled from three different experiments. Administration of nicotine reduces protein concentration TNF-a (C), MIP-2 (D), and neutrophil counts (E) in the BAL in E.coli pneumonia at 4 h. pp , 0.05; ppp , 0.01 versus saline controls; n = 4–6 in each group. Data were pooled from three different experiments. F and G, Activation of a7 nAChR by nicotine attenuates E. coli pneumonia at 24 h. F,ELW.G, Levels of HMGB1 in the lung homogenate. pp , 0.05; n =4– 6 in each group. H, Deficiency of a7 nAChR worsens pulmonary edema in E. coli pneumonia. pp , 0.05 for a7 nAChR+/+ versus a7 nAChR2/2 mice; n = 3 in each group. Data are mean 6 SD. I and J, Changes of AChE and choline in the BAL at 12 h after E. coli pneumonia. I, AChE activity in BAL cells from normal and E. coli pneumonia mice. J, Choline levels in the BAL. ppp , 0.01 versus normal mice; n = 6 in each group. Data are mean 6 SD. by guest on September 28, 2021 reduced at 4 and 12 h in the nicotine-treated group compared with a ALZET [Cupertino, CA] osmotic pump implanted under skin) or the saline group (Fig. 6A). MIP-2 levels in the plasma in the nicotine- saline. Mice were followed for 24 h, and survival was increased in the treated group also showed a lower trend at 4 h (Fig. 6B). nicotine therapy group compared with saline (Fig. 7A). To address Tostudythe effects of PHA 568487 (a specific agonist of a7nAChR) whether MLA, a specific a7 nAChR antagonist, counteracts the ef- on MIP-2 production in BAL and circulation, mice were pretreated fect of nicotine, mice were treated with either nicotine (2.4 mg/kg, with either PHA 568487 (0.4 mg/kg, i.p.) or saline 15 min before IT via osmotic pump) or nicotine + MLA (4 mg/kg, i.p. at 8 and 16 h). At challenge of E. coli 2.5 3 106 CFU, and then the same therapy was 24 h, survival in MLA treated mice was reduced compared with the repeated every 6 h. The mice were killed at 4, 12, and 24 h. Normal nicotine-only group (Fig. 7B). mice were used as controls. BAL and plasma were collected to mea- Cervical vagotomy and a7 nAChR deficiency reduces survival sure MIP-2 levels by ELISA. MIP-2 levels in the BAL were reduced at in E. coli pneumonia 4 h and showed a lower trend at 12 and 24 h in the PHA 568487-treated group (Fig. 6C). The MIP-2 levels in the plasma in the PHA 568487- To study whether vagus denervation affects survival in E. coli treatedgroupwerealsoreducedat4h(Fig.6D). Because vagus nerve pneumonia, unilateral cervical vagotomy in mice (right side) or 6 signaling via the ACh–a7 nAChR pathway may limit MIP-2 pro- a sham operation was performed; E. coli (4 3 10 CFU) was duction, we studied the effects of denervation of vagus on the response. instilled IT, and mice were followed for 48 h. Survival in vago- First, unilateral cervical vagotomy (right side) was done and then mice tomized mice was reduced compared with the sham group (Fig. were instilled ITwith E. coli (2.5 3 106 CFU). The mice were killed at 7C). To determine whether the deficiency of a7 nAChR affects +/+ 2/2 4, 12, and 24 h, and the right lungs were lavaged. Normal sham-op- survival from E. coli pneumonia, a7 nAChR and a7 nAChR 6 erated mice were used as controls. BAL and plasma were collected to mice were instilled IT with E. coli (4 3 10 CFU) and followed for 2/2 measure MIP-2 levels by ELISA. MIP-2 levels in the BAL were in- 12 h. Survival in a7 nAChR mice was reduced compared with +/+ creased at 4 and 12 h in the vagotomized group (Fig. 6E), and MIP-2 a7 nAChR mice (Fig. 7D). levels in the plasma in vagotomized group were also increased com- paredwiththeshamgroupat4and12h(Fig.6F). Discussion The current study demonstrates for the first time that activation of a Nicotine activation of 7 nAChR increases survival in E. coli a7 nAChR in alveolar macrophages and neutrophils is a critical pneumonia mechanism that decreases the lung inflammatory response to ei- To test whether activation of a7 nAChR by nicotine affects survival ther E. coli pneumonia or LPS challenge in mice. Activation of a7 in E. coli pneumonia, mice were challenged ITwith E. coli (5 3 106 nAChR in these cells reduced the production of proinflammatory CFU) and treated with either nicotine (2.4 mg/kg, delivered by cytokines and chemokines (especially MIP-2) and neutrophil 408 ACTIVATION OF a7 nAChR PREVENTS ACUTE LUNG INJURY Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 5. Activation of a7 nAChR reduces BAL proinflammatory cells and protein in E. coli pneumonia. A–D, Effect of nicotine. A, BAL leukocytes. B, BAL neutrophils. C, BAL monocytes. D, Protein levels. pp , 0.05 for nicotine versus saline; n = 3–5 in each group. Data were pooled from three different experiments. E–H, Effect of PHA 568487. E, BAL leukocytes. F, BAL neutrophils. G, BAL monocytes. H, Protein levels. pp , 0.05 for PHA 568487 versus saline; n = 3–5 in each group. Data were pooled from three different experiments. I–L, Unilateral cervical vagotomy (right side) increases BAL proinflammatory cells and protein levels in E. coli pneumonia. I, BAL leukocytes. J, BAL neutrophils. K, BAL monocytes. L, Protein levels. pp , 0.05 for vagotomy versus sham group at 4, 12, or 24 h after IT E. coli challenge; n = 4–5 in each group. Data were pooled from three different experiments. Data are mean 6 SD. transmigration, and thereby reduced lung injury and mortality. and a7 nAChR+Gr1+ cells are activated in the local milieu at sites There was a causal relationship between a7 nAChR activation and of vagal innervation. Vagus denervation failed to activate a7 the reduction in lung inflammation, because lung MIP-2 pro- nAChR+CD11b+ and a7 nAChR+Gr1+, resulting in persistent duction, lung neutrophil infiltration, and mortality were increased MIP-2 production and neutrophil infiltration, and worsened lung in a7 nAChR2/2 and vagus-denervated mice. inflammation and injury compared with control animals. a7 nAChR+CD11b+ alveolar macrophages and a7 nAChR+ In LPS and E. coli pneumonia ALI mouse models, alveolar Gr1+ neutrophils activated by a7 nAChR agonists induced a de- macrophages are the central effector cells in the production of creased production of MIP-2, TNF-a, and HMGB1 consistent proinflammatory cytokines (25, 26), which initiate and amplify with an important role of a7 nAChR activation in mediating the neutrophil transmigration into the lungs to mediate inflammation inactivation of inflammatory mediator production. The question and injury (27–29). Infiltrated neutrophils may function in a feed- arises as to the mechanisms of vagal innervation-induced pro- forward manner to generate MIP-2 and promote further neutrophil tection mediated by a7 nAChR receptors in alveolar macrophages transmigration (30, 31). Using exogenous a7 nAChR agonists to and neutrophils. It is known that airway epithelia and peri- stimulate a7 nAChR+CD11b+ and a7 nAChR+Gr1+ cells in injured epithelial tissue are vagally innervated (6), thus release of ACh in lungs disrupted this feed-forward inflammatory loop. In this sense, airways may activate a7 nAChR+ inflammatory cells accumulat- activation of a7 nAChR by endogenous ACh represents a homeo- ing in the airways. The ACh concentration, however, did not in- static negative feedback mechanism that probably fine-tunes the crease in BAL in our models consistent with rapid hydrolysis of lung–host defense system by dampening neutrophil transmigration, ACh (half-life, 2 min) by AChE occurring after ACh release from and thereby mitigating lung inflammation and injury. nerve endings (24). This possibility was assessed by measuring E. coli pneumonia shares similar pathogenic mechanisms with AChE activity in proinflammatory cells and BAL choline con- LPS-induced ALI, including activation of alveolar macrophages, centration (50 nmol/ml); thus, it is likely that a7 nAChR+CD11b+ increased production of early and late proinflammatory mediators, The Journal of Immunology 409

FIGURE 7. Activation of a7 nAChR enhances survival in E. coli pneumonia. A, Effects of nicotine. Mice were treated with saline or nicotine (2.4 mg/kg, delivered by an osmotic pump) and followed up for 24 h. pp , 0.05 Downloaded from for nicotine versus saline treated mice; n = 12 in each group. Data were pooled from two different experiments. B, Antagonism of a7 nAChR by its specific antagonist MLA counteracts the protective effects of nicotine on survival in E. coli pneumonia. Mice were treated with nicotine or nicotine + MLA (2.4 mg/kg, delivered by an osmotic pump; MLA was given i.p.) and followed up for 24 h. pp , 0.05 for nicotine (n = 9) versus nicotine + MLA (n = 5) treated mice. Data were pooled from two different experiments. http://www.jimmunol.org/ C, Unilateral vagotomy reduces survival in E. coli pneumonia. Sham and vagotomized mice were challenged with E. coli and followed for 48 h. pp , 0.05 for vagotomy (n = 30) versus sham (n = 30). Data were pooled from 5 different experiments. D, Deficiency of a7 nAChR worsens reduces survival in E. coli pneumonia; n =5a7 nAChR+/+ mice; n =4a7 nAChR2/2 mice; pp , 0.05 for a7 nAChR+/+ versus a7 nAChR2/2 mice. FIGURE 6. Activation of a7 nAChR reduces MIP-2 generation in E. coli pneumonia. A and B, Effect of nicotine. A, BAL MIP-2. B, Plasma and the previous observations that activation of a7 nAChR pro- pp , n MIP-2. 0.05 for nicotine versus saline; = 3–5 in each group. Data tected against Gram-negative sepsis (1, 5, 12–15, 33). A reason for by guest on September 28, 2021 were pooled from three different experiments. C and D, Effect of PHA this apparent difference in results with live bacteria may be that 568487. C, BAL MIP-2. D, Plasma MIP-2. pp , 0.05 for PHA 568487 versus saline; n = 3–5 in each group. Data were pooled from three different TLR4 recognizes the Gram-negative product LPS, whereas TLR2 experiments. E and F, Unilateral cervical vagotomy (right side) increases recognizes Gram-positive components (34); therefore, the protective MIP-2 levels in BAL from isolateral lung and plasma in E. coli pneumonia. effects by a7 nAChR may occur downstream of TLR4 activation. In E, BAL MIP-2. F, Plasma MIP-2. pp , 0.05 versus sham group at 4 and 12 the sepsis models (endotoxemia and cecal ligation and puncture), the h after IT E. coli challenge; n = 4–5 in each group. Data were pooled from protective effects of a7nAChRactivationweremediatedbyreducing three different experiments. Data are mean 6 SD. HMGB1, a late mediator of inflammation (5). In this study, we found that acute lung inflammation and injury induced by Gram-negative and neutrophil transmigration (17, 24, 29). Recently, studies have bacteria can be regulated by the cholinergic antiinflammatory path- shown that nicotine (nonspecific), DMAB (partial), PNU (spe- way. a7nAChR+CD11b+ and a7nAChR+Gr1+ cells might be im- cific), and PHA 568487 (specific) are a7 nAChR agonists, but portant players in this pathway, which deserves further study. these chemical preparations are different in molecular structure In summary, activation of a7 nAChR inhibited production of and specificity. To compare the effects of a7 nAChR agonists on proinflammatory cytokines (MIP-2, TNF-a, HMGB1) in alveolar proinflammatory cytokine production and lung inflammation, we ad- macrophages and neutrophils. a7 nAChR agonists reduced pul- ministrated a7 nAChR agonists (nicotine, DMAB, PNU, or PHA monary edema and lung inflammation in both LPS-induced ALI 568487) to disrupt the propagation of inflammation by reducing MIP-2 and E. coli pneumonia; it also increased survival from E. coli production in a7nAChR+CD11b+ and a7nAChR+Gr1+ cells, de- pneumonia. Thus, activation of a7 nAChR may be useful as creasing pulmonary edema and neutrophil infiltration in the early time a novel antiinflammatory strategy to treat ALI. frame(4and12h),andtherebyenhancing the survival of pneumonia in the nicotine treated mice. The role of vagal signaling was established Acknowledgments by the vagus denervation study, which increased MIP-2 production in We thank Jae Woo Lee, Sandra Brady, and Xiaopei Gao for suggestions and BAL and lung neutrophil infiltration at these early time points. These technical assistance. findings demonstrate that regulation of proinflammatory responses (e. g., MIP-2 production) by a7 nAChR-cholinergic antiinflammatory pathway plays an important role in the early stage of lung inflammation Disclosures induced by E. coli pneumonia or LPS challenge in mice. The authors have no financial conflicts of interest. Studies of Streptococcus pneumoniae–induced pneumonia showed that nicotine treatment increased bacterial burden and References worsened lung inflammation at 24 h (32), results that are different 1. Borovikova, L. V., S. Ivanova, M. Zhang, H. Yang, G. I. Botchkina, from the lung antiinflammatory effects of activating a7 nAChR L. R. Watkins, H. Wang, N. Abumrad, J. W. Eaton, and K. J. Tracey. 2000. Vagus 410 ACTIVATION OF a7 nAChR PREVENTS ACUTE LUNG INJURY

nerve stimulation attenuates the systemic inflammatory response to endotoxin. work. 2001. Efficacy of low tidal volume ventilation in patients with different Nature 405: 458–462. clinical risk factors for acute lung injury and the acute respiratory distress 2. Bernik, T. R., S. G. Friedman, M. Ochani, R. DiRaimo, S. Susarla, C. J. Czura, syndrome. Am. J. Respir. Crit. Care Med. 164: 231–236. and K. J. Tracey. 2002. Cholinergic antiinflammatory pathway inhibition of 19. Su, X., J. W. Lee, Z. A. Matthay, G. Mednick, T. Uchida, X. Fang, N. Gupta, and tumor necrosis factor during ischemia reperfusion. J. Vasc. Surg. 36: 1231–1236. M. A. Matthay. 2007. Activation of the alpha7 nAChR reduces acid-induced 3. Bernik, T. R., S. G. Friedman, M. Ochani, R. DiRaimo, L. Ulloa, H. Yang,S. Sudan, acute lung injury in mice and rats. Am. J. Respir. Cell Mol. Biol. 37: 186–192. C. J. Czura, S. M. Ivanova, and K. J. Tracey. 2002. Pharmacological stimulation of 20. Su, X., M. Johansen, M. R. Looney, E. J. Brown, and M. A. Matthay. 2008. the cholinergic antiinflammatory pathway. J. Exp. Med. 195: 781–788. CD47 deficiency protects mice from lipopolysaccharide-induced acute lung in- 4. Wang, H., M. Yu, M. Ochani, C. A. Amella, M. Tanovic, S. Susarla, J. H. Li, jury and Escherichia coli pneumonia. J. Immunol. 180: 6947–6953. H. Wang, H. Yang, L. Ulloa, et al. 2003. Nicotinic acetylcholine receptor alpha7 21. Su, X., M. Looney, L. Robriquet, X. Fang, and M. A. Matthay. 2004. Direct subunit is an essential regulator of inflammation. Nature 421: 384–388. visual instillation as a method for efficient delivery of fluid into the distal air- 5. Wang, H., H. Liao, M. Ochani, M. Justiniani, X. Lin, L. Yang, Y. Al-Abed, spaces of anesthetized mice. Exp. Lung Res. 30: 479–493. H. Wang, C. Metz, E. J. Miller, et al. 2004. Cholinergic agonists inhibit HMGB1 22. Brovkovych, V., X. P. Gao, E. Ong, S. Brovkovych, M. L. Brennan, X. Su, release and improve survival in experimental sepsis. Nat. Med. 10: 1216–1221. S. L. Hazen, A. B. Malik, and R. A. Skidgel. 2008. Augmented inducible nitric 6. Matthay, M. A., and L. B. Ware. 2004. Can nicotine treat sepsis? Nat. Med. 10: oxide synthase expression and increased NO production reduce sepsis-induced 1161–1162. lung injury and mortality in myeloperoxidase-null mice. Am. J. Physiol. Lung 7. van Westerloo, D. J., I. A. Giebelen, S. Florquin, J. Daalhuisen, M. J. Bruno, A. F. de Vos, K. J. Tracey, and T. van der Poll. 2005. The cholinergic anti- Cell. Mol. Physiol. 295: L96–L103. inflammatory pathway regulates the host response during septic peritonitis. J. 23. Matsumura-Takeda, K., S. Sogo, Y. Isakari, Y. Harada, K. Nishioka, Infect. Dis. 191: 2138–2148. T. Kawakami, T. Ono, and T. Taki. 2007. CD41+/CD45+ cells without acetyl- 8. vanWesterloo,D.J.,I.A.Giebelen,S.Florquin,M.J.Bruno,G.J.Larosa,L.Ulloa, cholinesterase activity are immature and a major megakaryocytic population in K. J. Tracey, and T. van der Poll. 2006. The vagus nerve and nicotinic receptors mod- murine bone marrow. Stem Cells 25: 862–870. ulate experimental pancreatitis severity in mice. Gastroenterology 130: 1822–1830. 24. Yamada, H., M. Otsuka, K. Fujimoto, K. Kawashima, and M. Yoshida. 1996. 9. Sadis, C., G. Teske, G. Stokman, C. Kubjak, N. Claessen, F. Moore, P. Loi, Determination of acetylcholine concentration in cerebrospinal fluid of patients B. Diallo, L. Barvais, M. Goldman, et al. 2007. Nicotine protects kidney from with neurologic diseases. Acta Neurol. Scand. 93: 76–78. renal ischemia/reperfusion injury through the cholinergic anti-inflammatory 25. Beck-Schimmer, B., R. Schwendener, T. Pasch, L. Reyes, C. Booy, and Downloaded from pathway. PLoS One 2: e469. R. C. Schimmer. 2005. Alveolar macrophages regulate neutrophil recruitment in 10. Yoshikawa, H., M. Kurokawa, N. Ozaki, K. Nara, K. Atou, E. Takada, endotoxin-induced lung injury. Respir. Res. 6: 61. H. Kamochi, and N. Suzuki. 2006. Nicotine inhibits the production of proin- 26. Lentsch, A. B., B. J. Czermak, N. M. Bless, N. Van Rooijen, and P. A. Ward. flammatory mediators in human monocytes by suppression of I-kappaB phos- 1999. Essential role of alveolar macrophages in intrapulmonary activation of phorylation and nuclear factor-kappaB transcriptional activity through nicotinic NF-kappaB. Am. J. Respir. Cell Mol. Biol. 20: 692–698. acetylcholine receptor alpha7. Clin. Exp. Immunol. 146: 116–123. 27. Mizgerd, J. P., M. R. Spieker, and C. M. Doerschuk. 2001. Early response cy- 11. de Jonge, W. J., E. P. van der Zanden, F. O. The, M. F. Bijlsma, D. J. van tokines and innate immunity: essential roles for TNF receptor 1 and type I IL-1

Westerloo, R. J. Bennink, H. R. Berthoud, S. Uematsu, S. Akira, R. M. van den receptor during Escherichia coli pneumonia in mice. J. Immunol. 166: 4042– http://www.jimmunol.org/ Wijngaard, and G. E. Boeckxstaens. 2005. Stimulation of the vagus nerve at- 4048. tenuates macrophage activation by activating the Jak2-STAT3 signaling pathway. 28. Folkesson, H. G., M. A. Matthay, C. A. He´bert, and V. C. Broaddus. 1995. Acid Nat. Immunol. 6: 844–851. aspiration-induced lung injury in rabbits is mediated by interleukin-8-dependent 12. Hamano, R., H. K. Takahashi, H. Iwagaki, T. Yoshino, M. Nishibori, and N. Tanaka. mechanisms. J. Clin. Invest. 96: 107–116. 2006. Stimulation of alpha7 nicotinic acetylcholine receptor inhibits CD14 and the 29. Yoshida, K., R. Kondo, Q. Wang, and C. M. Doerschuk. 2006. Neutrophil cy- toll-like receptor 4 expression in human monocytes. Shock 26: 358–364. toskeletal rearrangements during capillary sequestration in bacterial pneumonia 13. Pavlov, V. A., M. Ochani, L. H. Yang, M. Gallowitsch-Puerta, K. Ochani, X. Lin, in rats. Am. J. Respir. Crit. Care Med. 174: 689–698. J. Levi, W. R. Parrish, M. Rosas-Ballina, C. J. Czura, et al. 2007. Selective al- 30. Armstrong, D. A., J. A. Major, A. Chudyk, and T. A. Hamilton. 2004. Neutrophil pha7-nicotinic acetylcholine receptor agonist GTS-21 improves survival in chemoattractant genes KC and MIP-2 are expressed in different cell populations murine endotoxemia and severe sepsis. Crit. Care Med. 35: 1139–1144. at sites of surgical injury. J. Leukoc. Biol. 75: 641–648. 14. Parrish, W. R., M. Rosas-Ballina, M. Gallowitsch-Puerta, M. Ochani, K. Ochani, 31. Matzer, S. P., T. Baumann, N. W. Lukacs, M. Ro¨llinghoff, and H. U. Beuscher.

L. H. Yang, L. Hudson, X. Lin, N. Patel, S. M. Johnson, et al. 2008. Modulation by guest on September 28, 2021 2001. Constitutive expression of macrophage-inflammatory protein 2 (MIP-2) of TNF release by choline requires alpha7 subunit nicotinic acetylcholine receptor-mediated signaling. Mol. Med. 14: 567–574. mRNA in bone marrow gives rise to peripheral neutrophils with preformed MIP- 15. Hofer, S., C. Eisenbach, I. K. Lukic, L. Schneider, K. Bode, M. Brueckmann, 2 protein. J. Immunol. 167: 4635–4643. S. Mautner, M. N. Wente, J. Encke, J. Werner, et al. 2008. Pharmacologic 32. Giebelen, I. A., M. Leendertse, S. Florquin, and T. van der Poll. 2009. Stimu- cholinesterase inhibition improves survival in experimental sepsis. Crit. Care lation of acetylcholine receptors impairs host defence during pneumococcal Med. 36: 404–408. pneumonia. Eur. Respir. J. 33: 375–381. 16. Rubenfeld, G. D., E. Caldwell, E. Peabody, J. Weaver, D. P. Martin, M. Neff, 33. Giebelen, I. A., D. J. van Westerloo, G. J. LaRosa, A. F. de Vos, and T. van der E. J. Stern, and L. D. Hudson. 2005. Incidence and outcomes of acute lung Poll. 2007. Stimulation of a 7 cholinergic receptors inhibits lipopolysaccharide- injury. N. Engl. J. Med. 353: 1685–1693. induced neutrophil recruitment by a tumor necrosis factor a-independent 17. Ware, L. B., and M. A. Matthay. 2000. The acute respiratory distress syndrome. mechanism. Shock 27: 443–447. N. Engl. J. Med. 342: 1334–1349. 34. Takeuchi, O., K. Hoshino, and S. Akira. 2000. Cutting edge: TLR2-deficient and 18. Eisner, M. D., T. Thompson, L. D. Hudson, J. M. Luce, D. Hayden, MyD88-deficient mice are highly susceptible to Staphylococcus aureus in- D. Schoenfeld, and M. A. Matthay; Acute Respiratory Distress Syndrome Net- fection. J. Immunol. 165: 5392–5396.