Mycobacterium Tuberculosis Lipoarabinomannan Activates

Mycobacterium tuberculosis Lipoarabinomannan Activates Human Neutrophils via a TLR2/1 Mechanism Distinct from Pam 3CSK4 This information is current as of October 3, 2021. Jessica S. Hook, Mou Cao, Kayson Weng, Nedha Kinnare and Jessica G. Moreland J Immunol published online 23 December 2019 http://www.jimmunol.org/content/early/2019/12/21/jimmun

ol.1900919 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2019/12/22/jimmunol.190091 Material 9.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 October 3, 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 © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published December 23, 2019, doi:10.4049/jimmunol.1900919 The Journal of Immunology

Mycobacterium tuberculosis Lipoarabinomannan Activates Human Neutrophils via a TLR2/1 Mechanism Distinct from Pam3CSK4

Jessica S. Hook,* Mou Cao,* Kayson Weng,* Nedha Kinnare,* and Jessica G. Moreland*,†

Neutrophils, polymorphonuclear (PMN) leukocytes, play an important role in the early innate immune response to Mycobacterium tuberculosis infection in the lung. Interactions between PMN and mycobacterial lipids impact the activation state of these migrated cells with consequences for the surrounding tissue in terms of resolution versus ongoing inﬂammation. We hypothesized that lipoarabinomannan from M. tuberculosis (Mtb LAM) would prime human PMN in a TLR2-dependent manner and inves-

tigated this with speciﬁc comparison with the puriﬁed synthetic TLR2 agonists, Pam3CSK4 and FSL-1. In contrast to Pam3CSK4 and FSL-1, we found Mtb LAM did not induce any of the classical PMN priming phenotypes, including enhancement of NADPH

oxidase activity, shedding of L-selectin, or mobilization of CD11b. However, exposure of PMN to Mtb LAM did elicit pro- and anti- Downloaded from inﬂammatory cytokine production and release in a TLR2/1-dependent manner, using the TLR1 single-nucleotide polymorphism rs5743618 (1805G/T) as a marker for TLR2/1 speciﬁcity. Moreover, Mtb LAM did not elicit p38 MAPK phosphorylation or

endocytosis, although these processes occurred with Pam3CSK4 stimulation, and were necessary for the early priming events to occur. Interestingly, Mtb LAM did not abrogate priming responses elicited by Pam3CSK4. Notably, subfractionation of light membranes from Pam3CSK4 versus Mtb LAM–stimulated cells demonstrated differential patterns of exocytosis. In summary,

Mtb LAM activates PMN via TLR2/1, resulting in the production of cytokines but does not elicit early PMN priming responses, as http://www.jimmunol.org/

seen with Pam3CSK4. We speculate that the inability of Mtb LAM to prime PMN may be due to differential localization of TLR2/1 signaling. The Journal of Immunology, 2020, 204: 000–000.

uberculosis (Tb) is an ongoing global health concern In more general terms, there is increasing understanding that causing an estimated 1.2 million deaths in 2018 (1). Al- PMN called to a site of infection and inflammation can cause T though the role of macrophages in the pathogenesis of significant host tissue damage by release of toxic contents including disease caused by Mycobacterium tuberculosis is very well char- proteases and reactive oxygen species (ROS) (4, 5). In contrast, acterized, there is more recent recognition of a critical role for certain signaling conditions lead to neutrophil-mediated suppres- by guest on October 3, 2021 neutrophils, polymorphonuclear (PMN) leukocytes, in the early sion of inflammation by release of anti-inflammatory cytokines (6) pulmonary inflammatory response (2). The role of PMN in the and the initiation of an apoptosis program which not only limits pathogenesis has been relatively understudied, but it has been neutrophil-mediated damage, but alters the macrophage inflam- demonstrated that PMN are recruited in large numbers to the lung matory potential as a consequence of efferocytosis (7, 8). Two tissue surrounding the forming granuloma (3). During the process major avenues for perpetuation of inflammatory damage by PMN of migration from the vascular space into the lung, PMN are likely are via activation of the NADPH oxidase 2 (NOX2) and through to encounter not only host chemokines and cytokines, but also azurophilic or primary granule exocytosis (9, 10). mycobacterial proteins and lipids. The impact of the PMN inter- The literature to date lacks clarity on how M. tuberculosis actions with these bacterial products determines the activation and some of its products impact PMN NOX2 activity. Lip- state of the migrated PMN with significant potential to further oarabinomannan (LAM) is a glycolipid in the mycobacterial cell amplify the ongoing inflammatory process. wall that plays an important role in the pathogenesis of Tb (11). The structure of LAM varies among mycobacterial strains and is known to impact mycobacterial virulence (11). LAM consists of a *Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, glycosylphosphatidyl anchor, a mannan core, and arabinan. The TX 75390; and †Department of Microbiology, University of Texas Southwestern Medical Center, Dallas, TX 75390 terminal arabinan may be modified, resulting in heterogeneity ORCIDs: 0000-0001-9614-6335 (K.W.); 0000-0003-1467-5722 (N.K.). among mycobacterial strains that contributes to diverse host immune responses and mycobacterial virulence. The addition of Received for publication August 1, 2019. Accepted for publication November 22, 2019. mannose caps results in mannosylated LAM and is common Address correspondence and reprint requests to Dr. Jessica G. Moreland, University among, but not exclusive to, pathogenic mycobacterial strains, of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX including the most commonly studied pathogenic M. tuberculosis 75390-8548. E-mail address: [email protected] strain, H37Rv (11). Little is known about the interaction of LAM The online version of this article contains supplemental material. with human PMN. In studies performed nearly two decades ago Abbreviations used in this article: DPBS, Dulbecco’s PBS; FFE, free-flow electro- using relatively insensitive measures of ROS production, incuba- phoresis; FSC-SSC, forward versus side scatter; LAM, lipoarabinomannan; LAP, latent alkaline phosphatase; LM, light membrane; Mtb LAM, LAM from M. tuberculosis; tion of PMN with purified LAM did not elicit any superoxide NET, neutrophil extracellular trap; NOX2, NADPH oxidase 2; PM, plasma membrane production directly (12), whereas incubation with certain strains vesicle; PMN, polymorphonuclear; ROS, reactive oxygen species; SNP, single- of whole live Mycobacteria elicited ROS production in a TLR2- nucleotide polymorphism; SV, secretory vesicle; Tb, tuberculosis. dependent manner (13). Although purified LAM did not Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 directly elicit a respiratory burst, it was demonstrated to induce

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1900919 2 NEUTROPHIL TLR2/1 ACTIVATION BY Mtb LAM very-low-magnitude priming of fMLF-mediated ROS production (Danvers, MA). Streptavidin-HRP, IL-6, and IL-1Ra ELISA reagents were as measured by chemiluminescence (14). We have recently shown from R&D Systems (Minneapolis, MN). Fluorescently conjugated secondary that exocytosis of the azurophilic granule subset can be primed by Abs were from Jackson ImmunoResearch Laboratories (West Grove, PA). FSL-1 and Pam3CSK4 were from InvivoGen (San Diego, CA). Purified Mtb exposure to inflammatory cytokines, with enhanced release after LAM was obtained through Biodefense and Emerging Infections Research secondary stimulation (15), a process that might be highly relevant Resources Repository, National Institute of Allergy and Infectious Diseases, in Tb pathogenesis. National Institutes of Health: M. tuberculosis, strain H37Rv, purified LAM, Neutrophil priming, or preactivation, describes a state whereby NR-14848. p38 MAPK (SB203580), and control (SB202474) inhibitors were from Cayman Chemical (Ann Arbor, MI). Dyngo-4A was from Abcam. PMN encounter a primary stimulus and undergo phenotypic Additional reagents were obtained from Thermo Fisher Scientific changes that allow markedly enhanced responsiveness to a unique (Pittsburgh, PA). All buffers and reagents were strictly endotoxin free. subsequent stimulus. Our laboratory has studied neutrophil priming Human PMN purification in response to both bacterial and host products, most recently using synthetic TLR2 ligands as priming stimuli (16–19). In this most Human PMN were isolated according to standard techniques from heparin recent study, we discovered that the TLR1 single-nucleotide anticoagulated venous blood from healthy consenting adults following polymorphism (SNP) rs5743618 (1805G/T) significantly im- written informed consent and in accordance with a protocol approved by the Institutional Review Board for Human Subjects at the University of pacted neutrophil priming in response to a purified TLR2/1 ligand. Texas Southwestern Medical Center. PMN were isolated using dextran Mycobacteria are well described to signal through TLRs, with sedimentation and Ficoll-Hypaque density-gradient separation, followed TLR2 reported to play a critical role (20–23), and thus, we hy- by hypotonic lysis of erythrocytes as previously described (30). pothesized that LAM from M. tuberculosis (Mtb LAM) would Measurement of NOX2 activity by lucigenin-ECL stimulate neutrophil functional responses in a similar fashion. Interestingly, although mycobacterial exposure and infection is Lucigenin-ECL measurement of ROS generation was performed in a Downloaded from 96-well microplate using the LUMIstar or CLARIOstar from BMG Labtech ubiquitous in many parts of the world, it is estimated that only (Cary, NC). Freshly isolated PMN were diluted in HBSS containing 1% 10% of infected individuals go on to develop active disease. Fa- HSA and 0.1% dextrose. A total of 2.0–4.5 3 105 PMN (depending on the milial occurrence of active disease suggests a genetic component instrument used for analysis) were added per well in the presence of to pathogenesis. In a population-based case–control study of 1312 lucigenin (100 mM). fMLF (1 mM) was injected at specified time points HIV negative adults with Tb disease, Ma and colleagues (24) following treatment with FSL-1 (100 ng/ml), Pam3CSK4 (1 mg/ml), Mtb LAM (1 or 10 mg/ml), or no agonist. The concentration of Mtb LAM used http://www.jimmunol.org/ demonstrate a significant correlation between Tb disease and the was selected based on a review of related literature (31–33). NOX2 activity 1805G/T SNP in TLR1 among African Americans and Hispanics. was expressed as relative light units using a kinetic assay with readings Given the global burden of Tb, the high frequency of this TLR1 every 30 s for 60 min. In a subset of experiments, readings were taken SNP, and the growing literature demonstrating an impact of TLR1 every 60 s for 390 min. Data are presented every 3–10 min for clarity (Supplemental Fig. 1). Dyngo-4A (20 mM) and SB203580 (10 mM) were SNPs on mycobacterial disease (24–29), this question seems used in a subset of assays to block dynamin-dependent endocytosis and highly pertinent. p38 MAPK activity, respectively. In the current study, we found that Mtb LAM did not stimulate Analysis of cell surface protein expression by flow cytometry any early PMN responses, including direct or primed NADPH oxidase activity, even at high concentrations. Moreover, Mtb LAM PMN were analyzed using a FACSCalibur flow cytometer from BD Bio- by guest on October 3, 2021 did not elicit priming of neutrophils by analysis of other pheno- sciences (Franklin Lakes, NJ). For assessment of surface expression of CD11b and L-selectin, PMN were incubated in HBSS containing 1% HSA typic endpoints including mobilization of CD11b, shedding of and 0.1% dextrose with or without agonist, as specified. Following stim- L-selectin, and priming of granule exocytosis. However, Mtb ulation, cells were centrifuged, resuspended in blocking buffer (DPBS with LAM–exposed PMNs displayed robust pro- and anti-inflammatory 2% nonfat dry milk and 4% normal goat serum), and incubated on ice for cytokine generation in a TLR2/1-dependent manner based on the 20 min. Primary Abs were added at a final concentration of 13 mg/ml after previously described 1805G/T SNP. The mechanisms underlying blocking and incubated on ice for 1 h. PMN were centrifuged and resuspended in FITC-conjugated secondary Ab diluted 1:1000 and incu- these differences in PMN response to the synthetic TLR2/1 ago- bated on ice for 30 min prior to washing and resuspension for analysis by nist Pam3CSK4 versus Mtb LAM may be related to both the site of flow cytometry. We gated on the neutrophil population by forward versus signal generation, with Pam3CSK4 eliciting endocytosis of TLR2, side scatter (FSC-SSC) to exclude debris and dead cells. Twenty thousand and the nature of the signalosome that is assembled. Although the events were collected for analysis. data presented focus on PMN–Mtb LAM interactions specifically, Elastase activity the findings have broad relevance to our understanding of the Freshly isolated PMN were diluted in PICMG buffer (10 mM sodium biology of TLR signaling in human PMNs. We conclude that the phosphate buffer with 2.7 mM KCl, 138 mM NaCl, 0.6 mM CaCl2, 1.0 mM breadth of TLR2/1 activation in PMN is likely much greater than MgCl2, and 0.1% dextrose) containing 7-amino-4-methylcoumarin seen in other cell types given the potential for both immediate (20 mM) and added to a 96-well microplate as described above. For release of preformed products and the capacity to be metabolically priming experiments, fMLF (1 mM) was injected at specified time active and generate novel proteins for secretion. points following stimulation, as specified. PMN were also exposed to dihydrocytochalasin B (2.5 mg/ml), a microfilament disrupting agent, prior to stimulation with fMLF as a measurement of maximum elastase release Materials and Methods from the primary granules. The CLARIOstar was used to measure fluorescence at an excitation of 360 6 20 nm and emission of 450 6 30 nm, Materials with readings every 60 s. Elicited elastase release was expressed as the HBSS and Dulbecco’s PBS (DPBS) were purchased from Mediatech change in fluorescence over a specified time period. Dyngo-4A (20 mM) (Manassas, VA), dextran from Pharmacosmos (Holbaek, Denmark), Ficoll- and SB203580 (10 mM) were used in a subset of assays to block dynamin- Paque from GE Healthcare (Piscataway, NJ), human serum albumin from dependent endocytosis and p38 MAPK activity, respectively. Talecris Biotherapeutics (Durham, NC), normal goat serum and 7-amino- Analysis of IL-1Ra, IL-6, and IL-8 by ELISA 4-methylcoumarin from MP Biomedicals (Solon, OH), and paraformal- dehyde from Electron Microscopy Sciences (Hatfield, PA). Abs included Freshly isolated PMN were stimulated with agonist, as specified, in RPMI CD11b (clone ICRF44), L-selectin (clone DREG-56), and IL-8 (clone 1640 plus 10% autologous serum for 6 h tumbling in a 37˚C/5% CO2 G265-5) from BD Pharmingen (San Diego, CA), b-actin (clone incubator. After stimulation, supernatant was collected from pelleted ACTBD11B7) and NOX2 (gp91phox clone 54.1) from Santa Cruz Bio- cells and stored at 280˚C until ELISA was performed. A 96-well Nunc technology (Dallas, TX), TLR2 and CR-1 from Abcam (Cambridge, MA), MaxiSorp microplate from Thermo Fisher Scientific (Rochester, NY) and phospho–p38 MAPK (Thr 180/Tyr 182) from Cell Signaling Technology was coated with 4 mg/ml IL-6, 2 mg/ml IL-8, or 2 mg/ml IL-1Ra Ab in The Journal of Immunology 3

carbonate buffer (100 mM NaHCO3 and 34 mM Na2CO3 [pH 9.6]) and buffer (10 mM PIPES, 100 mM KCl, 3 mM NaCl, 3.5 mM MgCl2, and incubated at room temperature overnight on a rotator. Wells were washed 1.25 mM ethyleneglycoltetraacetic acid) with 1 mM PMSF. The cells were three times in wash buffer (0.05% Tween 20 in DPBS), then blocked with placed in an ice-cold cell disruption vessel (Parr Instruments, Moline, IL) DPBS containing 1% BSA and 5% sucrose for 1 h. Samples and stan- and closed tightly. The vessel was pressurized to 380 pounds per square dards, diluted in assay diluent (0.1% BSA in DPBS), were applied and inch with nitrogen gas and allowed to equilibrate for 20 min at 4˚C. Cavitates allowed to incubate for 2 h. Next, 0.2 mg/ml biotinylated IL-6, 2 mg/ml were expelled dropwise into a tube containing ethyleneglycoltetraacetic acid biotinylated IL-8, or 0.8 mg/ml biotinylated IL-1Ra Ab was added and (1.25 mM) and then centrifuged 200 3 g for 10 min to remove unbroken allowed to incubate for 2 h. Streptavidin-HRP was added to the wells cells and nuclei. Postnuclear supernatants were placed on top of two-layer and allowed to incubate for 20 min. Color development occurred during Percoll gradients made as described (35) and centrifuged at 48,400 3 g for incubation with tetramethylbenzidine and was stopped by addition of 15 min at 4˚C in a Beckman Avanti J-25 centrifuge with low deceleration. 0.5 M H2SO4. Subcellular fractions were carefully removed from the gradient, and Per- coll was removed by centrifugation. The light membrane (LM) fraction Analysis of IL-8 by intracellular flow cytometry was used for free-flow electrophoresis (FFE), as described below. A por- Freshly isolated PMN were stimulated as described above for 2, 4, or 6 h. tion of the LM was lysed and incubated with sample buffer (62.5 mM Tris- After stimulation, supernatant was removed, and stimulated cells were fixed HCl [pH 6.8], 5% 2-ME, 2.3% SDS, and 5% glycerol) at 70˚C for 10 min and permeabilized according to the two-step protocol for intracellular prior to SDS-PAGE. (cytoplasmic) proteins developed by Thermo Fisher Scientific (Waltham, Free-flow electrophoresis MA). Anti–Ly-6G–FITC (clone 1A8) from BD Pharmingen was used to identify neutrophils. Anti–IL-8–APC (clone 8CH) from Thermo Fisher FFE was used to separate the LM fraction into component parts: secretory Scientific was used for the detection of intracellular IL-8. Isolated cells vesicles (SV) and plasma membrane vesicles (PM) (19, 36). To reduce the were 96.33 6 0.72% neutrophils, 2.42 6 0.49% eosinophils, 1.17 6 0.31% surface charge on vesicles, the LM fraction was treated with neuraminidase lymphocytes, and 0.08 6 0.08% monocytes by manual differential for 30 min at 37˚C at a final concentration of 0.2 U/ml by mixing 15 times

(n = 6 unique donors). with a 21-gauge needle. Following neuraminidase treatment, the samples Downloaded from were centrifuged at 184,000 3 g for 15 min at 4˚C and resuspended in Endocytosis assays 750 ml of cold media buffer (6 mM triethanolamine and 6 mM glacial Flow cytometry. A total of 7.5 3 105 PMN were incubated with 25 mg/ml acetic acid [pH 7.4], with 270 mM sucrose and conductivity adjusted to 520 mS/cm). The electrode buffer consisted of 50 mM triethanolamine and dextran (10,000 MW) and either FSL-1 (1 mg/ml), 10 mg/ml Pam3CSK4, or 10 mg/ml Mtb LAM for 30 min at 37˚C in HBSS containing 1% HSA 50 mM glacial acetic acid (pH 7.4). Zonal electrophoresis was performed and 0.1% dextrose. Cells were washed with ice-cold DPBS, resuspended in on a Becton-Dickinson FFE system (Sparks, MD) at 5˚C using a media ice-cold DPBS, and taken immediately to flow for analysis, as previously flow rate of 300 ml/h and a sample flow rate of 3.5 ml/h. High-voltage described (19). settings were 160 W, 1600 V, and 100 mAmp to give constant amperage. http://www.jimmunol.org/ Confocal. A total of 4.5 3 105 PMN were allowed to adhere to autologous Following separation by FFE, alkaline phosphatase activity assay was used human serum–coated glass coverslips for 15 min at 37˚C. Biotinylated to identify the orientation of LM vesicles within the 96 FFE fractions. Thus, SV (inside out), displaying only latent (after Triton X-100 detergent) Pam CSK (5 mg/ml) in HBSS containing 1% HSA and 0.1% dextrose was 3 4 can be distinguished from PM (right side out), as previously described (19, added to the coverslip and incubated for 30 min in a 37˚C tissue culture m incubator. A total of 10% formalin was used for fixation at room 36). Briefly, 100 l of each fraction collected by FFE was plated into temperature for 10 min, followed by blocking and staining for TLR2. duplicate 96-well microplates. Twenty five microliters of 2% Triton X-100 Images were acquired on a Zeiss LSM 880 confocal microscope with a was added to each well of one plate and incubated for 15 min at room Plan Apochromat 633/1.4 oil objective at room temperature with temperature to discern the latent alkaline phosphatase (LAP) activity of the p Zeiss Immersol 518 F halogen free/fluorescence free imaging medium SV (36). One hundred and fifty microliters of 2.5 mg/ml –nitrophenyl phosphate was added to each well and incubated for 30 min in the dark, and processed using ZEN software from Carl Zeiss Microscopy by guest on October 3, 2021 followed by reading on CLARIOstar V at 405 nm. To make comparisons (Thornwood, NY). between experiments, the OD was normalized by dividing the OD of each Analysis of p38 MAPK phosphorylation fraction by the lowest measured value, as the OD is a relative measurement. Using the LAP curves, the area under the curve was calculated for PMN (2 3 107/ml) were incubated in the presence of agonist for the the SV fraction and expressed as a percentage of the entire LM fraction. specified time points. After incubation, cells were pelleted and lysed in Fractions were lysed and sample buffer added as above for immunoblotting. lysis buffer (20 mM imidazole, 2 mM EGTA, 2% leupeptin/pepstatin A, 1 mM PMSF, 100 mM NaCl, and 1% TX-100) rotating for 20 min at 4˚C. Statistical analysis Lysates were centrifuged at 2600 3 g for 5 min at 4˚C and supernatant Results are expressed as means 6 SEM. Statistical analysis was performed placed in tubes with sample buffer and heated at 70˚C for 10 min. using GraphPad Prism 7 for Windows from GraphPad Software (La Jolla, Immunoblotting CA). Comparisons between groups were performed using one-way ANOVA with multiple comparison or Student t test. Results were con- Samples were resolved in Bio-Rad TGX gels (Hercules, CA) by SDS-PAGE sidered statistically significant with a p value ,0.05, and relative p values and then transferred to nitrocellulose. Immunoblots were processed using are noted in the figure legends. Abs specific for the protein(s) of interest and species-specific HRP or fluorescent-tagged secondary Abs followed by ECL (Super Signal Sub- Results strate; Pierce) or fluorescent detection. Immunoblots were scanned using Mtb LAM does not elicit neutrophil priming responses the Bio-Rad ChemiDoc, and relative abundances were quantitated using ImageLab software. The two most commonly described phenotypic endpoints of neutrophil priming are augmentation of NADPH oxidase activity Confocal microscopy and mobilization of CD11b from intracellular pools to the cell Freshly isolated PMNs were stimulated, as specified, after adhering to surface (37, 38). In view of our recent investigations demon- autologous serum-coated glass coverslips (34). Cells were then fixed in strating priming of ROS by TLR2-specific ligands (16), we hy- 10% formalin for 10 min at room temperature. Cells were then washed, stained, and mounted prior to image acquisition as described above. pothesized that Mtb LAM, described as a TLR2 agonist, would Neutrophils from five donors were analyzed for shape change in response also elicit priming of the fMLF-stimulated respiratory burst. to stimulation with Pam3CSK4 or Mtb LAM. A blinded observer rated Neutrophils from donors with a common SNP in TLR1 (1805T) the roundness of the cells. demonstrate markedly enhanced responsiveness to Pam3CSK4, Nitrogen cavitation and Percoll gradient centrifugation a specific TLR2/1 agonist, whereas neutrophils from donors without this SNP (1805G) are minimally responsive to Pam3CSK4, Freshly isolated PMN were treated with the serine protease inhibitor even up to concentrations of 10 mg/ml (16). There is no difference diisopropyl fluorophosphate (1 mM) for 20 min at room temperature and centrifuged at 200 3 g for 5 min at 4˚C. The cells were then stimu- in the response of neutrophils from 1805G or 1805T donors when lated, as specified, in a 37˚C/5% CO2 incubator, followed by centrifugation stimulated with the TLR2/6-specific agonist, FSL-1, demonstrat- at 200 3 g for 5 min at 4˚C. The cells were resuspended in relaxation ing the ability of 1805G PMN to respond to a TLR2 agonist (16). 4 NEUTROPHIL TLR2/1 ACTIVATION BY Mtb LAM

Using lucigenin-ECL to measure ROS production after stimula- release in PMN from 1805T donors, and Mtb LAM did not impact tion with fMLF, human PMN from 1805G and 1805T donors were elastase release in PMN from any of the donors (Fig. 1F). Con- incubated with purified Mtb LAM for 30 min and then stimulated sidered in combination, our data demonstrate that Mtb LAM does with fMLF. Mtb LAM did not elicit any ROS production primarily not stimulate early priming responses in human neutrophils. and did not prime cells for enhanced NADPH oxidase activity after Mtb LAM does not limit Pam3CSK4 signaling and early subsequent stimulation with fMLF, with FSL-1 and Pam3CSK4 eliciting the expected enhancement of the respiratory burst priming of neutrophils (Fig. 1A–C). Mtb LAM was unable to prime for enhanced ROS The absence of early priming responses may be the result of altered production even with a 10-fold increase in concentration (Fig. 1C) binding affinity to the receptor or distinct signaling pathways initiated. or extended priming periods up to 6 h (Supplemental Fig. 1). Next, To investigate this, we conducted competition studies to determine we studied the impact of exposure to Mtb LAM on surface CD11b whether an excess of Mtb LAM could diminish or prevent PMN re- levels and found no significant increase in CD11b levels with 30 sponses to Pam3CSK4. 1805T neutrophils pretreated with 10 mg/ml min exposure to Mtb LAM, whereas Pam3CSK4 elicited enhance- Mtb LAM for 15 min or cotreated with 10 mg/ml Mtb LAM displayed ment of CD11b levels in a TLR2/1-dependent manner (Fig. 1D), as no abrogation of Pam3CSK4-elicited priming of ROS (Fig. 2A), and previously demonstrated (16). L-Selectin shedding was also mea- cotreatment of 1805T PMN with 10 mg/ml Mtb LAM had no effect sured as an additional early marker of PMN activation in response on Pam3CSK4-elicited shedding of L-selectin (Fig. 2B). Our previous to inflammatory stimuli (39). Moreover, signaling elicited by work demonstrated a higher density of TLR1 on the surface of 1805T L-selectin shedding has been demonstrated to enhance neutrophil PMNs (16), and it is possible that the affinity of Pam3CSK4 is mark- functional outputs playing a role in microbial clearance (40). FSL- edly higher than Mtb LAM for TLR2/1. However, higher concentra- Downloaded from 1 stimulation resulted in rapid L-selectin shedding in all donor tions of Mtb LAM failed to elicit any priming responses whatsoever, PMN, whereas Pam3CSK4 stimulation caused rapid L-selectin even in 1805T PMN. PMN have the capacity for spatiotemporal shedding in 1805T PMN only. In contrast, Mtb LAM did not alter regulation of TLR signaling with the generation and release of in- surface L-selectin levels in any of the donor PMNs (Fig. 1E). We flammatory mediators over a wide range of time with some responses lengthened the incubation time and found complete shedding of occurring within seconds to minutes and some requiring transcription L-selectin at all time points in 1805T PMN stimulated with FSL-1 and translation over hours. Given the ability of M. tuberculosis to http://www.jimmunol.org/ or Pam3CSK4,butMtb LAM failed to elicit L-selectin shedding, delay and evade immune detection (42, 43), and the ability of PMN even after 6 h of stimulation (Supplemental Fig. 2). Both TLR4 to produce inflammatory mediators hours after stimulation, we rea- ligands (41) and TNF-a (15, 37) can elicit priming of azurophilic soned that Mtb LAM may elicit late neutrophil responses. granule release in response to a secondary stimulus. Given recognition of the role for elastase in inflammatory lung injury, the im- Inflammatory cytokine production elicited by Mtb LAM in a pact of Mtb LAM on azurophilic granule exocytosis was studied TLR2/1-dependent manner using secreted elastase activity as a measure of granule exocytosis. Neutrophils produce and secrete inflammatory cytokines in re- Cells were stimulated with fMLF following pre-exposure to FSL-1, sponse to various stimuli (6, 16). We previously demonstrated Pam3CSK4,orMtb LAM. FSL-1 elicited priming of elastase re- TLR2/1-dependent pro- and anti-inflammatory cytokine release in by guest on October 3, 2021 lease from all donor PMN, Pam3CSK4 elicited priming of elastase response to 24 h stimulation with Pam3CSK4 (16). We sought to test

FIGURE 1. Mtb LAM does not elicit neutrophil priming responses. (A–C) Priming of fMLF-elicited NADPH oxidase activity. Representative tracings of lucigenin-ECL from 1805G (A)– and 1805T (B)–donor PMN demonstrating no fMLF-elicited NADPH oxidase enhancement after priming with Mtb LAM.

(C) Marked increase in primed fMLF-elicited NADH oxidase activity in response to Pam3CSK4, but not Mtb LAM (n $ 5). (D–F) Neutrophil upregulation and exocytosis of granules. Flow cytometry measurement of cell surface CD11b (n $ 9) (D) and L-selectin shedding (n $ 7) (E). (F) Priming of fMLF-elicited elastase release (n $ 5). Each dot represents a different donor. Statistical analysis performed by one-way ANOVAwith multiple comparisons. *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. The Journal of Immunology 5

FIGURE 2. Mtb LAM does not prevent Pam3CSK4 activation of neutrophils. (A) No change in ROS production, as measured by lucigenin-ECL by neutrophils treated with Pam3CSK4 (1 mg/ml) in the presence or absence of pretreatment or cotreatment with Mtb LAM (10 mg/ml) (n $ 6). (B) No change in L-selectin shedding by neutrophils treated with Pam3CSK4 (1 mg/ml) in the presence or absence of Mtb LAM (10 mg/ml) (n = 3). Each dot represents a different donor. Statistical analysis performed by one-way ANOVA with multiple comparisons. ***p , 0.001. the hypothesis that Mtb LAM stimulation would elicit late PMN impacted by a common SNP in TLR1. In addition, although PMN responses including the generation of soluble mediators that may from every 1805T donor secreted IL-6 and IL-8 above the control alter and amplify the host response. Using PMN from individuals level in response to Pam3CSK4 or Mtb LAM, IL-1Ra secretion Downloaded from with (1805T) and without (1805G) the TLR1 SNP rs5743618 was at or below control levels for some 1805T-donor PMN. Mtb (1805G/T), we exposed puriﬁed PMN to either Pam3CSK4 or Mtb LAM elicited signiﬁcantly less IL-Ra and IL-6 from 1805T-donor LAMfor6hexvivo.MtbLAMexposureelicitedmarkedpro- PMN than Pam3CSK4. Taken together with the lack of early duction of IL-1Ra, IL-6, and IL-8 in PMNs from donors with the classical priming neutrophil responses and the inability to block 1805T genotype, whereas 1805G PMN had minimal cytokine se- Pam3CSK4-elicited responses, Mtb LAM and Pam3CSK4 signal-

cretion above no agonist (control) levels (Fig. 3A–C). Although ing via TLR2/1 occurs through different pathways. TLR4 elicits http://www.jimmunol.org/ neutrophils are the most numerous leukocyte in circulation, cell for differential signaling from the PM versus the endosome with cell, they produce smaller quantities of cytokines than monocytes. TIRAP/Mal and MyD88 being necessary for signaling from the For this reason, it is critical to ensure neutrophil purity in cytokine PM and TRIF and TRAM being necessary for signaling from the studies. To confirm the source of the cytokines generated, we endosome (44, 45). These differential signaling pathways have not employed intracellular staining of IL-8 using flow cytometry to been elucidated as clearly for TLR2-dependent activation of neu- gate on the neutrophil population using both FSC-SSC and Ly-6G trophils. In view of the stimulus-dependent responses (Pam3CSK4 positivity. The results by intracellular flow were consistent with versus Mtb LAM), we next sought to investigate the mechanisms what was measured by ELISA of secreted cytokines (Fig. 3D). governing TLR2/1 signaling of PMN responses.

Interestingly, Pam3CSK4 and Mtb LAM elicited intracellular IL-8 by guest on October 3, 2021 above control levels within 2 h of stimulation (data not shown). Pam3CSK4 elicits endocytosis via TLR2/1 and signals from an These data provide deﬁnitive evidence that Mtb LAM signals via endosomal compartment TLR2/1 under certain exposure conditions, and, importantly, that Neutrophil priming by the TLR4 ligand, endotoxin, requires en- cytokine generation in response to Mtb LAM is profoundly docytosis (19), and there are some reports of TLR2-based

FIGURE 3. Mtb LAM elicits neutrophil cytokine production in a TLR2/1-dependent manner. (A) IL-1Ra, (B), IL-6, and (C) IL-8 secretion after 6 h of stimulation with no agonist (control), Pam3CSK4 (1 mg/ml), or Mtb LAM (10 mg/ml) as measured by ELISA (n $ 8). *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. (D) Neu- trophil intracellular IL-8 after 4 h of stimulation with no agonist (control), Pam3CSK4 (1 mg/ml), or Mtb LAM (10 mg/ml), as determined by flow cytometry of purified neutrophils identified by FSC-SSC and Ly-6G positivity (n = 6). Each dot represents a different donor. Statistical analysis performed by one-way ANOVA with multiple comparisons. *p , 0.05, **p , 0.01, ***p , 0.001. 6 NEUTROPHIL TLR2/1 ACTIVATION BY Mtb LAM endosomal signaling in a human monocyte cell line (44–46). Based changes in shape to crawl through tissues and phagocytose on our findings that Mtb LAM does not elicit neutrophil priming, bacteria (47). Within minutes of stimulation, neutrophils undergo we reasoned that binding and signaling via TLR2/1 elicited by Mtb cytoskeletal alterations that facilitate movement of proteins, LAM might not be identical to the interaction between Pam3CSK4 granules, and endosomes within the cell. Given the ability of M. and TLR2/1. We postulated that Pam3CSK4 would elicit endocy- tuberculosis to evade immune detection (8, 48, 49) and to in- tosis and Mtb LAM would not. Using a flow cytometry-based assay, terfere with maturation of the phagolysosome (50), we investi- Pam3CSK4 elicited endocytosis in a TLR2/1-dependent manner. gated the ability of M. tuberculosis to elicit shape change in Of note, PMN from 1805G donors that were stimulated with neutrophils. Neutrophils stimulated with Pam3CSK4 demonstrate Pam3CSK4 did demonstrate enhanced endocytosis as compared dramatic changes in shape with the presence of lamellipodia and with control, but it was significantly lower than the amount elicited other projections (Fig. 6B). In contrast, unstimulated (control) by PMN from 1805T donors (Fig. 4A). Interestingly, Mtb LAM or neutrophils and neutrophils stimulated with Mtb LAM remain FSL-1 stimulation did not increase endocytosis above basal levels in predominantly round and show no physical evidence of early any donor (Fig. 4A) providing evidence for differences in TLR2/1 activation (Fig. 6A, 6C). Control or Mtb LAM stimulated neu- and TLR2/6 signaling from the PM. Given that the flow-based trophils were significantly rounder than neutrophils stimulated assay for endocytosis does not provide specific evidence of the with Pam3CSK4 (Fig. 6D). Given that Mtb LAM did not elicit nature of the endosomes that are formed in response to Pam3CSK4 shape change, endocytosis, or the early PMN responses seen stimulation, confocal microscopy was performed. PMNs stimu- with exposure to Pam3CSK4, we postulated that spatial regula- lated with biotinylated Pam3CSK4 for 30 min displayed intracel- tion of signaling complexes might be distinct in response to these lular vesicular localization of the Pam3CSK4 ligand, which was two agonists. noted to colocalize with TLR2 staining (Fig. 4B–D). Downloaded from Mtb LAM elicits partial exocytosis of SV Pam CSK -elicited neutrophil priming of ROS is p38 MAPK 3 4 Although neutrophil priming has predominantly been characterized and endocytosis dependent in terms of rapid onset of functional outputs from PMN, bio- We have previously demonstrated a specific role for p38 MAPK chemical correlates of PMN preactivation or priming have been signaling in TLR2-elicited neutrophil priming (16), thus, we an- described in response to certain stimuli, including partial assembly alyzed activation of the p38 MAPK pathway in response to short of the NADPH oxidase complex (38, 51). The secretory vesicle http://www.jimmunol.org/ term exposure to Pam3CSK4 or Mtb LAM. Given the lack of (SV) compartment, which contains stores of PM proteins, but only neutrophil priming in response to Mtb LAM, we hypothesized that albumin within the matrix, is mobilized to the PM fraction in Mtb LAM would not elicit early phosphorylation of p38 MAPK. response to priming with endotoxin (19). This allows for enhanced There was no evidence of enhanced phosphorylation of p38 cell surface density of the membrane components of the NADPH MAPK in response to exposure to Mtb LAM for 10–30 min, again oxidase and is one mechanism for increased and more rapid in contrast to the Pam3CSK4-elicited activation of p38 MAPK ROS production after priming. Given our findings with Mtb LAM (Fig. 5A). To further investigate the requirement for p38 MAPK versus Pam3CSK4, we studied the composition of the subcellular signaling and endocytosis in neutrophil priming, we used the fractions of neutrophils under control conditions in comparison by guest on October 3, 2021 p38 MAPK inhibitor, SB203580, and the endocytosis inhibitor, with PMN incubated with either Pam3CSK4 or Mtb LAM for dyngo-4A. Inhibition of p38 MAPK activity prevented Pam3CSK4 30 min. Nitrogen cavitation and Percoll density centrifugation was priming of ROS generation (Fig. 5B) and elastase release (Fig. 5D). used to separate the LM fraction (containing both the SV and PM) In addition, inhibition of endocytosis blocked Pam3CSK4-elicited from the primary and secondary granule fractions (35). Whereas priming of ROS (Fig. 5C) and elastase release (Fig. 5D). Taken in unstimulated PMNs .70% of the NOX2 protein (gp91phox)is together, these data provide further evidence for differential stored in the secondary granule fraction with the remainder in the TLR2/1 signaling pathways in neutrophils stimulated with LM fraction, immunoblotting of these fractions demonstrated that Pam3CSK4 versus Mtb LAM and complement our findings that stimulation with either Pam3CSK4 or Mtb LAM led to significant Mtb LAM does not elicit classic priming of neutrophils. mobilization of NOX2 from the secondary granule fraction into the LM compartment (data not shown). Subsequently, the LM Pam CSK priming of neutrophils corresponds with rapid 3 4 were further studied by FFE allowing further separation by size shape change not seen in Mtb LAM–stimulated cells and charge. The PM fraction which retains a right side out ori- Dynamin-dependent endocytosis requires cytoskeletal rearrange- entation thus displaying alkaline phosphatase without detergent ment, and neutrophils possess the ability to undergo dramatic permeabilization, can be readily distinguished from the SV

FIGURE 4. Pam3CSK4 elicits endocytosis. (A) Flow cytometry detection of dextran uptake in neutrophils from 1805G and 1805T donors stimulated with no agonist (control), FSL-1 (1 mg/ml), Pam3CSK4 (10 mg/ml), or Mtb LAM (10 mg/ml) for 30 min (n $ 5). **p , 0.01, ****p , 0.0001. (B–D) Representative image of 1805T-donor neutrophils stimulated with biotinylated Pam3CSK4.(B) Biotinylated Pam3CSK4.(C) TLR2. (D) Merge with DAPI (blue) nuclear stain. Each dot represents a different donor. Images were acquired at 633 magniﬁcation. Statistical analysis performed by one-way ANOVA with multiple comparisons. The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/

FIGURE 5. Priming of NADPH oxidase–derived ROS and elastase release are p38 MAPK and endocytosis dependent. (A) Pam3CSK4, but not Mtb LAM, elicits rapid phosphorylation of p38 MAPK by immunoblotting (n = 3). (B and C) ROS production, as measured by lucigenin-ECL, is p38 MAPK dependent (B) and endocytosis dependent (C). SB203580 and Dyngo-4A were used to block p38 MAPK activity and endocytosis, respectively (n = 3). (D) Elastase release is p38 MAPK and endocytosis dependent (n $ 4). Each dot represents a different donor. Statistical analysis performed by one-way ANOVA with multiple comparisons. **p , 0.01, ****p , 0.0001. fraction, which has an inside out orientation with alkaline phos- stimulation elicited significant mobilization of both NOX2 and phatase only on the intravesicular surface, requiring detergent to TLR2 proteins to the PM fraction from the SV with no shift in by guest on October 3, 2021 expose, using the LAP assay as previously described (19, 35, 36). response to Mtb LAM (Fig. 8). Pam3CSK4 and Mtb LAM stim- Analyses of these samples demonstrated two distinct peaks in ulation elicited movement of CR-1 to the PM fraction from the SV control unstimulated PMNs consistent with SV and PM sub- with greater movement in the Pam3CSK4 stimulated cells fractions as previously seen (Fig. 7A). Pam3CSK4 stimulation led (Fig. 8C). Taken together, these data demonstrate, for the first to full mobilization of the SV fraction (no change in the alkaline time to our knowledge, that Mtb LAM stimulates neutrophils in a phosphatase assay after Triton X-100) (Fig. 7B), whereas an in- TLR2/1-dependent manner and that Mtb LAM–elicited signaling termediate level of mobilization was noted in the Mtb LAM is not identical to that seen in response to Pam3CSK4, includ- treated PMNs (Fig. 7C). This pattern was reproducible across ing differential exocytosis. The use of 1805G versus 1805T PMN multiple donors and FFE experiments (Fig. 7D), and quantitation provide a powerful tool for further investigation of the molecular of the SV fraction as a component of the total LM fractions mechanisms of Tb pathogenesis and TLR2 biology. confirmed the significance of these findings (Fig. 7E). Subsequently, biochemical evaluation of the SV and PM sub- Discussion fractions for NOX2 (gp91phox), CR-1 and TLR2 were performed to Given the ongoing global health care burden caused by Tb, per- complement the findings seen by the LAP assay. Pam3CSK4 sistent investigation of pathophysiological mechanisms underlying

FIGURE 6. Pam3CSK4 elicits early shape change in PMN, Mtb LAM does not. Representative images of control (A)–, Pam3CSK4 (10 mg/ml) (B)–, and Mtb LAM (10 mg/ml) (C)–treated neutrophils stained with TLR2 (red) and DAPI (blue). Images were acquired at 633 magniﬁcation. (D) Quantitation of cell shape (n = 5 donors). Each dot represents 100 cells per donor. Statistical analysis performed by Student t test. ****p , 0.0001. 8 NEUTROPHIL TLR2/1 ACTIVATION BY Mtb LAM Downloaded from

FIGURE 7. Pam3CSK4 elicits complete mobilization of the SV compartment to the PM, with only partial exocytosis seen in response to Mtb LAM. Representative alkaline phosphatase assay performed on FFE fractions from unstimulated (control) (A)–, Pam3CSK4 (5 mg/ml) (B)–, and Mtb LAM (10 mg/ml) (C)–treated PMNs after 30 min exposure, demonstrating differential exocytosis. (D) Combined LAP activity curves (after addition of Triton E X-100 to permeabilize SV) from four separate FFE experiments. ( ) Quantitation of the SV area under the curve as a component of the entire LM (SV and http://www.jimmunol.org/

PM) fraction (n = 3 [control], n = 4 [Pam3CSK4 and Mtb LAM]). Each dot represents a different donor. Statistical analysis performed by one-way ANOVA with multiple comparisons. **p , 0.01. the perpetuation of the disease are warranted. A more intensive near immediate release of preformed or rapidly generated neu- interest in the role of neutrophils in M. tuberculosis–related lung trophil products and a late response requiring transcription and disease has occurred in recent years, whereas these cells were translation of new proteins. Neutrophil outputs from the early largely ignored previously. A unique feature of pathogen-elicited phase responses, including ROS production and protease release, activation of human neutrophils (in comparison with most other can clearly contribute to both pathogen killing and host tissue leukocytes) is the existence of a biphasic response with early or damage. The role of neutrophils in mediation of host defense by guest on October 3, 2021

FIGURE 8. Mtb LAM elicits differential pattern of mobilization of proteins to the PM, as compared with Pam3CSK4.(A) Representative immunoblot phox demonstrating CR-1, TLR2, and NOX2 (gp91 ) proteins in the SV versus PM fractions of PMNs stimulated with Pam3CSK4 (5 mg/ml) versus Mtb LAM (10 mg/ml) after FFE. (B–D) Quantitation of these proteins in each subfraction for multiple unique donors (n = 3 [control], n = 4 [Pam3CSK4 and Mtb LAM]). Each dot represents a different donor. Statistical analysis performed by one-way ANOVA with multiple comparisons. *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. The Journal of Immunology 9 versus perpetuation of the host inflammatory response is a primary activation responses. Our data do not explain how Mtb LAM can question for Tb infection pathophysiology (2). It is evident that signal through TLR2/1 yet avoid eliciting these initial events that understanding the impact of mycobacterial products on both the might create a toxic extracellular environment for perpetuation of neutrophil early and late cytokine/chemokine release needs to be the pathogen, but these questions are currently under investigation. studied in reductionist systems and must be investigated using Indeed, it has been previously published that mannosylated LAM primary human PMNs (52, 53). has very low affinity binding to TLR2 and limited capacity to The current study provides evidence for three novel findings. generate signaling, but these studies have only been performed First, to our knowledge, this is the first demonstration that Mtb in murine bone marrow–derived macrophages using knockout LAM specifically activates human neutrophils via TLR2/1 to elicit technology (33). Assembly of TLR2 signalosome on an endo- generation of cytokines. Given the known frequency of the TLR1 somal structure has been previously reported in murine macro- 1805G/T SNP and the obvious clinical relevance of this SNP to phages (65). In these studies, endolysosomal TLR2-signaled type I M. tuberculosis susceptibility (26, 29, 54), these data also have IFN responses in an MyD88-dependent fashion, whereas cytokine important implications for better understanding the host response. expression did not arise from endosomal TLR signaling. It was The second novel finding presented in this study is that the specific subsequently demonstrated that, in addition to MyD88, both TLR2/1 ligand, Pam3CSK4, stimulates endocytosis in human Mal/TIRAP and TRAM were required for TLR2 activation of type neutrophils and is taken into a TLR2/1 containing endosome I IFN responses (but not cytokine generation), with TRAM acting which is necessary for early TLR2/1 signaling to occur. These as a sorting adaptor for TLR2-mediated responses (66). Specific neutrophil responses appear to be signaled via p38 MAPK as we requirements for each of the adaptor proteins in TLR2 signal- previously demonstrated with LPS and TNF-a (15, 18), and do not ing has varied based on the cell type under investigation. In mu- occur in cells stimulated with Mtb LAM. Finally, we demonstrate rine synovial fibroblasts, for example, TRAM is necessary for Downloaded from biochemically that both Pam3CSK4 and Mtb LAM impact the TLR2-dependent cytokine generation in response to lipoteichoic exocytic processes of human PMNs with differential outcomes. acid (67). Considered in combination, the existing literature point Neutrophils contain several subtypes of preformed granules out the diversity of signaling based on not only ligand and cell loaded with granule-specific cargo that are exocytosed in a hier- type, but on the expression of TLR2 heterodimer partners TLR1, archical and highly programmed fashion. Azurophilic or primary 6, and even TLR10, as well as a number of coreceptors (68). granules are the last granule type to be mobilized and contain the In fact, the a3b1 integrin has been specifically demonstrated http://www.jimmunol.org/ most toxic contents in the neutrophil arsenal, including several to function as a regulator/coreceptor for TLR2/1 in human mac- proteases. Although there has been extensive investigation of the rophage endosomal signaling (46). Notably, Mtb LAM does not mechanisms governing primary granule fusion with the phagosome elicit measurable increases in endocytosis, suggesting that the after neutrophil ingestion of a pathogen, only recently has the signalosome for this agonist is assembled at the PM. Our data do regulation of primary granule fusion with the PM been studied not exclude the possibility that other receptors (scavenger recep- (55, 56). M. tuberculosis stimulation of granule exocytosis also tors and integrins) are involved in the signaling cascade leading appears to occur in a strain-specific manner in response to whole to robust generation of specific cytokines that may perpetuate bacteria (9); however, there is minimal data on the impact of the inflammatory process by recruiting further leukocytes. by guest on October 3, 2021 purified LAM on primary granule exocytosis. In the current study, A specific role for neutrophils in the amplification of ongoing we found that Pam3CSK4 significantly enhanced primary granule inflammation is underscored by the relatively recent recognition release at the PM, whereas Mtb LAM did not have any apparent that neutrophils possess the capacity to become metabolically impact on primary granule machinery. Elastase release has been active and initiate transcription and translation of new proteins (6). implicated as a mechanism of tissue damage in the pathogenesis The transcriptome of human neutrophils generated in response to of acute lung injury in the setting of sepsis with SIRS (57); interaction with pathogenic bacteria is under intense scrutiny. Our however, the pathology and cadence of pulmonary infection with findings represent novel evidence of human neutrophil cytokine M. tuberculosis is completely different from that seen in acute production specifically elicited by a mycobacterial product via lung injury following SIRS. Regulated exocytosis of this granule TLR2/1 activation. Given that it is now recognized that the life- subset appears to be distinct from release of the serine protease span of both tissue-migrated and circulating neutrophils is likely elastase as a component of neutrophil extracellular traps (NETs). much longer than initially appreciated (69), the ongoing genera- This topic is particularly relevant and merits further investigation tion of cytokine mediators is highly relevant to disease pro- given the recent focus on the role of NETs in M. tuberculosis gression. Mycobacterial products can initiate the production of pathogenesis (10), including a suggested role in limiting extrap- proinflammatory cytokines in macrophages and that production is ulmonary dissemination of mycobacteria (58, 59). Although there potentially modulated by both the level of acylation of the my- have been significant advances in our understanding of the cellular cobacterial component and the involvement of other macrophage machinery required for primary granule exocytosis in neutrophils receptors (70, 71). As demonstrated by the donor-specific (1805G (56, 60), there is much more limited understanding of the mech- versus 1805T) neutrophil cytokine production in response to Mtb anisms controlling NET release. Recently, it has been appreciated LAM exposure (Fig. 3), host factors likely play a large role in that there are numerous unique pathways leading to NET forma- Tb pathogenesis. tion, some requiring NOX2-derived ROS and others described as Despite the high numbers of Tb infections, active disease is NOX2 independent (61, 62). Spatiotemporal regulation of ROS limited to around 10% of infected individuals (72). Pathogen- signaling is one potential mechanism to explain the differences associated factors that impact virulence and the ability of Tb to seen in responsiveness to Mtb LAM versus Pam3CSK4. survive the initial interaction with host cells and delay immune The concept of spatial regulation of signaling pathways is detection have been widely studied. In addition, pathogenesis is not unique to TLR2/1. There has been extensive evaluation of the strongly correlated with the accumulation of neutrophils in the TLR signalosome generated in response to various ligands at lung in both murine and human studies (52, 73–78). The impor- the PM and/or the endosomal membrane (63, 64). In this study, we tance of host signaling via TLR2/1 in mycobacterial pathogenesis show that p38 MAPK has a specific and critical role downstream is highlighted by studies revealing differential susceptibility and of the TLR2/1 receptor in initiation of rapid or early neutrophil outcomes in individuals with various SNPs in TLR1 (24, 27, 29). 10 NEUTROPHIL TLR2/1 ACTIVATION BY Mtb LAM

We, and others, have explored the impact of a specific TLR1 SNP 12. Fietta, A., C. Francioli, and G. Gialdroni Grassi. 2000. Mycobacterial lipoarabinomannan affects human polymorphonuclear and mononuclear phago- rs5743618 (1805G/T) in leukocyte function and have demon- cyte functions differently. Haematologica 85: 11–18. strated that the presence of the SNP (1805T) is associated with 13. Romero, M. M., J. I. Basile, B. Lo´pez, V. Ritacco, L. Barrera, M. C. Sasiain, and greater surface expression of TLR1 and enhanced inflammatory M. Alemań. 2014. Outbreaks of Mycobacterium tuberculosis MDR strains dif- ferentially induce neutrophil respiratory burst involving lipid rafts, p38 MAPK potential of neutrophils (16) and augmented monocyte generation and Syk. BMC Infect. Dis. 14: 262. of TNF-a upon stimulation with Pam3CSK4 (27). The literature 14. Fa¨ldt, J., C. Dahlgren, M. Ridell, and A. Karlsson. 2001. Priming of human lacks clarity on the effect of this particular SNP and mycobacterial neutrophils by mycobacterial lipoarabinomannans: role of granule mobilisation. Microbes Infect. 3: 1101–1109. susceptibility. Although 1805T correlates with an increased inci- 15. Potera, R. M., M. J. Jensen, B. M. Hilkin, G. K. South, J. S. Hook, E. A. Gross, dence of leprosy (27), the balance of information suggests a de- and J. G. Moreland. 2016. Neutrophil azurophilic granule exocytosis is primed creased incidence of M. tuberculosis infection (24, 29). SNPs in by TNF-a and partially regulated by NADPH oxidase. Innate Immun. 22: 635–646. TLR6 and TLR10 are also associated with M. tuberculosis dis- 16. Whitmore, L. C., J. S. Hook, A. R. Philiph, B. M. Hilkin, X. Bing, C. Ahn, ease, suggesting that signaling via TLR2/6 and possibly TLR2/10 H. R. Wong, P. J. Ferguson, and J. G. Moreland. 2016. A common genetic variant are possible when neutrophils encounter the whole bacteria and in TLR1 enhances human neutrophil priming and impacts length of intensive care stay in pediatric sepsis. J. Immunol. 196: 1376–1386. other mycobacterial products (24). 17. Volk, A. P., B. M. Barber, K. L. Goss, J. G. Ruff, C. K. Heise, J. S. Hook, and In summary, we demonstrate that a known M. tuberculosis J. G. Moreland. 2011. Priming of neutrophils and differentiated PLB-985 cells by pathophysiological concentrations of TNF-a is partially oxygen dependent. virulence factor, LAM, fails to elicit PMN-priming responses that J. Innate Immun. 3: 298–314. are critical for killing of other pathogens but does elicit cytokine 18. Moreland, J. G., A. P. Davis, J. J. Matsuda, J. S. Hook, G. Bailey, W. M. Nauseef, production via TLR2/1. Although individuals with the TLR1 SNP and F. S. Lamb. 2007. Endotoxin priming of neutrophils requires NADPH oxidase-generated oxidants and is regulated by the anion transporter ClC-3. rs5743618 (1805G/T) experience a hyperinflammatory neutrophil J. Biol. Chem. 282: 33958–33967. Downloaded from phenotype and inflammatory complications in response to many 19. Lamb, F. S., J. S. Hook, B. M. Hilkin, J. N. Huber, A. P. Volk, and J. G. Moreland. simple bacterial infections (79), the protective role of this SNP in 2012. Endotoxin priming of neutrophils requires endocytosis and NADPH oxidase- dependent endosomal reactive oxygen species. J. Biol. Chem. 287: 12395–12404. M. tuberculosis infection is likely one critical factor maintaining 20. Basu, J., D. M. Shin, and E. K. Jo. 2012. Mycobacterial signaling through toll- its prevalence in the worldwide population. The current study like receptors. Front. Cell. Infect. Microbiol. 2: 145. provides novel insights into the basic biology of human neutrophil 21. Underhill, D. M., A. Ozinsky, K. D. Smith, and A. Aderem. 1999. Toll-like receptor-2 mediates mycobacteria-induced proinflammatory signaling in mac- interactions with Mtb LAM and significantly advances our knowl- rophages. Proc. Natl. Acad. Sci. USA 96: 14459–14463. edge of TLR2/1 signaling in neutrophils. The recognition of two 22. Means, T. K., B. W. Jones, A. B. Schromm, B. A. Shurtleff, J. A. Smith, http://www.jimmunol.org/ J. Keane, D. T. Golenbock, S. N. Vogel, and M. J. Fenton. 2001. Differential distinct phases (early versus late) of human neutrophil activation by effects of a Toll-like receptor antagonist on Mycobacterium tuberculosis-induced TLR agonists provides the opportunity to dissect the underlying macrophage responses. J. Immunol. 166: 4074–4082. mechanisms and signaling pathways. The knowledge gained will 23. Krutzik, S. R., M. T. Ochoa, P. A. Sieling, S. Uematsu, Y. W. Ng, A. Legaspi, P. T. Liu, S. T. Cole, P. J. Godowski, Y. Maeda, et al. 2003. Activation and allow for the development of targeted host-specific therapies. regulation of Toll-like receptors 2 and 1 in human leprosy. Nat. Med. 9: 525–532. 24. Ma, X., Y. Liu, B. B. Gowen, E. A. Graviss, A. G. Clark, and J. M. Musser. 2007. Full-exon resequencing reveals toll-like receptor variants contribute to human Acknowledgments susceptibility to tuberculosis disease. PLoS One 2: e1318. We thank Dr. Karen Dobos laboratory for insightful discussions. We thank 25. Zhang, Y., T. Jiang, X. Yang, Y. Xue, C. Wang, J. Liu, X. Zhang, Z. Chen, M. Zhao, and J. C. Li. 2013. Toll-like receptor -1, -2, and -6 polymorphisms and the University of Texas Southwestern Live Cell Imaging Facility and the by guest on October 3, 2021 pulmonary tuberculosis susceptibility: a systematic review and meta-analysis. University of Texas Southwestern Flow Cytometry Core for guidance and PLoS One 8: e63357. support with the microscopy and flow cytometry studies. 26. Dittrich, N., L. C. Berrocal-Almanza, S. Thada, S. Goyal, H. Slevogt, G. Sumanlatha, A. Hussain, S. Sur, S. Burkert, D. Y. Oh, et al. 2015. Toll-like receptor 1 variations influence susceptibility and immune response to Mycobacterium tuberculosis. Disclosures Tuberculosis (Edinb.) 95: 328–335. The authors have no financial conflicts of interest. 27. Johnson, C. M., E. A. Lyle, K. O. Omueti, V. A. Stepensky, O. Yegin, E. Alpsoy, L. Hamann, R. R. Schumann, and R. I. Tapping. 2007. Cutting edge: a common polymorphism impairs cell surface trafficking and functional responses of TLR1 but protects against leprosy. J. Immunol. 178: 7520–7524. References 28. Omueti, K. O., D. J. Mazur, K. S. Thompson, E. A. Lyle, and R. I. Tapping. 2007. 1. World Health Organization. 2019. Global Tuberculosis Report. World Health The polymorphism P315L of human toll-like receptor 1 impairs innate immune Organization, Geneva, Switzerland. Available at: https://www.who.int/tb/ sensing of microbial cell wall components. J. Immunol. 178: 6387–6394. publications/global_report/en/. 29. Uciechowski, P., H. Imhoff, C. Lange, C. G. Meyer, E. N. Browne, D. K. Kirsten, 2. Dallenga, T., and U. E. Schaible. 2016. Neutrophils in tuberculosis--first line of A. K. Schro¨der, B. Schaaf, A. Al-Lahham, R. R. Reinert, et al. 2011. Suscep- defence or booster of disease and targets for host-directed therapy? Pathog. Dis. tibility to tuberculosis is associated with TLR1 polymorphisms resulting in a 74: ftw012. lack of TLR1 cell surface expression. J. Leukoc. Biol. 90: 377–388. 3. Sia, J. K., M. Georgieva, and J. Rengarajan. 2015. Innate immune defenses in 30. Nauseef, W. M. 2007. Isolation of human neutrophils from venous blood. human tuberculosis: an overview of the interactions between Mycobacterium Methods Mol. Biol. 412: 15–20. tuberculosis and innate immune cells. J. Immunol. Res. 2015: 747543. 31. Tapping, R. I., and P. S. Tobias. 2003. Mycobacterial lipoarabinomannan 4. Mantovani, A., M. A. Cassatella, C. Costantini, and S. Jaillon. 2011. Neutrophils mediates physical interactions between TLR1 and TLR2 to induce signaling. in the activation and regulation of innate and adaptive immunity. Nat. Rev. J. Endotoxin Res. 9: 264–268. Immunol. 11: 519–531. 32. Rivie`re, M., A. Moisand, A. Lopez, and G. Puzo. 2004. Highly ordered supra- 5. Korkmaz, B., T. Moreau, and F. Gauthier. 2008. Neutrophil elastase, proteinase 3 molecular organization of the mycobacterial lipoarabinomannans in solution. and cathepsin G: physicochemical properties, activity and physiopathological Evidence of a relationship between supra-molecular organization and biological functions. Biochimie 90: 227–242. activity. J. Mol. Biol. 344: 907–918. 6. Schro¨der, A. K., M. von der Ohe, D. Fleischer, L. Rink, and P. Uciechowski. 33. Shukla, S., E. T. Richardson, M. G. Drage, W. H. Boom, and C. V. Harding. 2005. Differential synthesis of two interleukin-1 receptor antagonist variants and 2018. Mycobacterium tuberculosis lipoprotein and lipoglycan binding to toll-like interleukin-8 by peripheral blood neutrophils. Cytokine 32: 246–253. receptor 2 correlates with agonist activity and functional outcomes. Infect. 7.Duffin,R.,A.E.Leitch,S.Fox,C.Haslett,andA.G.Rossi.2010.Targetinggran- Immun. 86: e00450. ulocyte apoptosis: mechanisms, models, and therapies. Immunol. Rev. 236: 28–40. 34. Allen, L. A. 2014. Immunofluorescence and confocal microscopy of neutrophils. 8. Prusa, J., D. X. Zhu, and C. L. Stallings. 2018. The stringent response and Methods Mol. Biol. 1124: 251–268. Mycobacterium tuberculosis pathogenesis. Pathog. Dis. 76: fty054. 35. Clemmensen, S. N., L. Udby, and N. Borregaard. 2014. Subcellular fractionation 9. Faurschou, M., and N. Borregaard. 2003. Neutrophil granules and secretory of human neutrophils and analysis of subcellular markers. Methods Mol. Biol. vesicles in inflammation. Microbes Infect. 5: 1317–1327. 1124: 53–76. 10. Mittal, M., M. R. Siddiqui, K. Tran, S. P. Reddy, and A. B. Malik. 2014. Reactive 36. Sengeløv, H., M. H. Nielsen, and N. Borregaard. 1992. Separation of human oxygen species in inflammation and tissue injury. Antioxid. Redox Signal. 20: neutrophil plasma membrane from intracellular vesicles containing alkaline 1126–1167. phosphatase and NADPH oxidase activity by free flow electrophoresis. J. Biol. 11. Chatterjee, D., and K. H. Khoo. 1998. Mycobacterial lipoarabinomannan: an Chem. 267: 14912–14917. extraordinary lipoheteroglycan with profound physiological effects. Glycobiology 37. Miralda, I., S. M. Uriarte, and K. R. McLeish. 2017. Multiple phenotypic 8: 113–120. changes define neutrophil priming. Front. Cell. Infect. Microbiol. 7: 217. The Journal of Immunology 11

38. El-Benna, J., M. Hurtado-Nedelec, V. Marzaioli, J. C. Marie, M. A. Gougerot-Pocidalo, 61. Papayannopoulos, V. 2018. Neutrophil extracellular traps in immunity and dis- and P. M. Dang. 2016. Priming of the neutrophil respiratory burst: role in host ease. Nat. Rev. Immunol. 18: 134–147. defense and inflammation. Immunol. Rev. 273: 180–193. 62. Tatsiy, O., and P. P. McDonald. 2018. Physiological stimuli induce PAD4- 39. Ivetic, A. 2018. A head-to-tail view of L-selectin and its impact on neutrophil dependent, ROS-independent NETosis, with early and late events controlled behaviour. Cell Tissue Res. 371: 437–453. by discrete signaling pathways. Front. Immunol. 9: 2036. 40. Cappenberg, A., A. Margraf, K. Thomas, B. Bardel, D. A. McCreedy, V. Van 63. Balka, K. R., and D. De Nardo. 2019. Understanding early TLR signaling through Marck, A. Mellmann, C. A. Lowell, and A. Zarbock. 2019. L-selectin shedding the Myddosome. J. Leukoc. Biol. 105: 339–351. affects bacterial clearance in the lung: a new regulatory pathway for integrin 64. Satoh, T., and S. Akira. 2016. Toll-like receptor signaling and its inducible outside-in signaling. Blood 134: 1445–1457. proteins. Microbiol. Spectr. 4. 41. Johnson, J. L., H. Hong, J. Monfregola, W. B. Kiosses, and S. D. Catz. 2011. 65. Dietrich, N., S. Lienenklaus, S. Weiss, and N. O. Gekara. 2010. Murine toll-like Munc13-4 restricts motility of Rab27a-expressing vesicles to facilitate receptor 2 activation induces type I interferon responses from endolysosomal lipopolysaccharide-induced priming of exocytosis in neutrophils. J. Biol. compartments. PLoS One 5: e10250. Chem. 286: 5647–5656. 66. Stack, J., S. L. Doyle, D. J. Connolly, L. S. Reinert, K. M. O’Keeffe, R. M. McLoughlin, 42. Lerner, T. R., S. Borel, and M. G. Gutierrez. 2015. The innate immune response S. R. Paludan, and A. G. Bowie. 2014. TRAM is required for TLR2 endosomal sig- in human tuberculosis. Cell. Microbiol. 17: 1277–1285. naling to type I IFN induction. J. Immunol. 193: 6090–6102. 43. Schorey, J. S., and L. S. Schlesinger. 2016. Innate immune responses to tuber- 67. Sacre, S. M., A. M. Lundberg, E. Andreakos, C. Taylor, M. Feldmann, and culosis. Microbiol. Spectr. DOI: 10.1128/microbiolspec.TBTB2-0010-2016. B. M. Foxwell. 2007. Selective use of TRAM in lipopolysaccharide (LPS) and 44. Li, J., D. S. Lee, and J. Madrenas. 2013. Evolving bacterial envelopes and lipoteichoic acid (LTA) induced NF-kappaB activation and cytokine production plasticity of TLR2-dependent responses: basic research and translational op- in primary human cells: TRAM is an adaptor for LPS and LTA signaling. portunities. Front. Immunol. 4: 347. J. Immunol. 178: 2148–2154. 45. Kagan, J. C., T. Su, T. Horng, A. Chow, S. Akira, and R. Medzhitov. 2008. 68. van Bergenhenegouwen, J., T. S. Plantinga, L. A. Joosten, M. G. Netea, TRAM couples endocytosis of Toll-like receptor 4 to the induction of interferon- G.Folkerts,A.D.Kraneveld,J.Garssen,andA.P.Vos.2013.TLR2&Co:a beta. Nat. Immunol. 9: 361–368. critical analysis of the complex interactions between TLR2 and coreceptors. 46. Marre, M. L., T. Petnicki-Ocwieja, A. S. DeFrancesco, C. T. Darcy, and L. T. Hu. J. Leukoc. Biol. 94: 885–902. 2010. Human integrin a(3)b(1) regulates TLR2 recognition of lipopeptides from 69. Hidalgo, A., E. R. Chilvers, C. Summers, and L. Koenderman. 2019. The neu- endosomal compartments. PLoS One 5: e12871. trophil life cycle. Trends Immunol. 40: 584–597. 47. Roberts, R. E., and M. B. Hallett. 2019. Neutrophil cell shape change: mecha- 70. Doz, E., S. Rose, J. Nigou, M. Gilleron, G. Puzo, F. Erard, B. Ryffel, and Downloaded from nism and signalling during cell spreading and phagocytosis. Int. J. Mol. Sci. V. F. Quesniaux. 2007. Acylation determines the toll-like receptor (TLR)-dependent DOI: 10.3390/ijms20061383. positive versus TLR2-, mannose receptor-, and SIGNR1-independent negative 48. Flentie, K., A. L. Garner, and C. L. Stallings. 2016. Mycobacterium tuberculosis regulation of pro-inflammatory cytokines by mycobacterial lipomannan. J. Biol. transcription machinery: ready to respond to host attacks. J. Bacteriol. 198: 1360–1373. Chem. 282: 26014–26025. 49. Goldberg, M. F., N. K. Saini, and S. A. Porcelli. 2014. Evasion of innate and 71. Quesniaux, V. J., D. M. Nicolle, D. Torres, L. Kremer, Y. Gue´rardel, J. Nigou, adaptive immunity by Mycobacterium tuberculosis. Microbiol. Spectr. 2. G. Puzo, F. Erard, and B. Ryffel. 2004. Toll-like receptor 2 (TLR2)-dependent- 50. Uribe-Querol, E., and C. Rosales. 2017. Control of phagocytosis by microbial positive and TLR2-independent-negative regulation of proinflammatory cyto-

pathogens. Front. Immunol. 8: 1368. kines by mycobacterial lipomannans. J. Immunol. 172: 4425–4434. http://www.jimmunol.org/ 51. DeLeo, F. R., J. Renee, S. McCormick, M. Nakamura, M. Apicella, J. P. Weiss, 72. Zumla, A., A. George, V. Sharma, R. H. Herbert, A. Oxley, and M. Oliver, and W. M. Nauseef. 1998. Neutrophils exposed to bacterial lipopolysaccharide Baroness Masham of Ilton. 2015. The WHO 2014 global tuberculosis report-- upregulate NADPH oxidase assembly. J. Clin. Invest. 101: 455–463. further to go. Lancet Glob. Health 3: e10-2. 52. Nandi, B., and S. M. Behar. 2011. Regulation of neutrophils by interferon-g limits 73.Kimmey,J.M.,J.P.Huynh,L.A.Weiss,S.Park,A.Kambal,J.Debnath, lung inﬂammation during tuberculosis infection. J. Exp. Med. 208: 2251–2262. H. W. Virgin, and C. L. Stallings. 2015. Unique role for ATG5 in neutrophil- 53. Singh, A. K., and U. D. Gupta. 2018. Animal models of tuberculosis: lesson mediated immunopathology during M. tuberculosis infection. Nature 528: learnt. Indian J. Med. Res. 147: 456–463. 565–569. 54. Schurz, H., M. Daya, M. Mo¨ller, E. G. Hoal, and M. Salie. 2015. TLR1, 2, 4, 6 74. Niazi, M. K., N. Dhulekar, D. Schmidt, S. Major, R. Cooper, C. Abeijon, and 9 variants associated with tuberculosis susceptibility: a systematic review D. M. Gatti, I. Kramnik, B. Yener, M. Gurcan, and G. Beamer. 2015. Lung and meta-analysis. PLoS One 10: e0139711. necrosis and neutrophils reﬂect common pathways of susceptibility to 55. Ramadass, M., and S. D. Catz. 2016. Molecular mechanisms regulating secretory Mycobacterium tuberculosis in genetically diverse, immune-competent mice.

organelles and endosomes in neutrophils and their implications for inflammation. Dis. Model. Mech. 8: 1141–1153. by guest on October 3, 2021 Immunol. Rev. 273: 249–265. 75.Berry,M.P.,C.M.Graham,F.W.McNab,Z.Xu,S.A.Bloch,T.Oni, 56. Johnson, J. L., J. He, M. Ramadass, K. Pestonjamasp, W. B. Kiosses, J. Zhang, K. A. Wilkinson, R. Banchereau, J. Skinner, R. J. Wilkinson, et al. 2010. An and S. D. Catz. 2016. Munc13-4 is a Rab11-binding protein that regulates interferon-inducible neutrophil-driven blood transcriptional signature in hu- Rab11-positive vesicle trafficking and docking at the plasma membrane. J. Biol. man tuberculosis. Nature 466: 973–977. Chem. 291: 3423–3438. 76. Eum, S. Y., J. H. Kong, M. S. Hong, Y. J. Lee, J. H. Kim, S. H. Hwang, 57. Tamakuma, S., M. Ogawa, N. Aikawa, T. Kubota, H. Hirasawa, A. Ishizaka, S. N. Cho, L. E. Via, and C. E. Barry, III. 2010. Neutrophils are the predominant N. Taenaka, C. Hamada, S. Matsuoka, and T. Abiru. 2004. Relationship between infected phagocytic cells in the airways of patients with active pulmonary TB. neutrophil elastase and acute lung injury in humans. Pulm. Pharmacol. Ther. 17: Chest 137: 122–128. 271–279. 77. Eruslanov, E. B., I. V. Lyadova, T. K. Kondratieva, K. B. Majorov, I. V. Scheglov, 58. Braian, C., V. Hogea, and O. Stendahl. 2013. Mycobacterium tuberculosis- M. O. Orlova, and A. S. Apt. 2005. Neutrophil responses to Mycobacterium induced neutrophil extracellular traps activate human macrophages. J. Innate tuberculosis infection in genetically susceptible and resistant mice. Infect. Immun. 5: 591–602. Immun. 73: 1744–1753. 59.Kalsum,S.,C.Braian,V.A.C.M.Koeken,J.Raffetseder,M.Lindroth, 78. Mishra, B. B., R. R. Lovewell, A. J. Olive, G. Zhang, W. Wang, E. Eugenin, R. van Crevel, and M. Lerm. 2017. The cording phenotype of Mycobacterium C. M. Smith, J. Y. Phuah, J. E. Long, M. L. Dubuke, et al. 2017. Nitric oxide tuberculosis induces the formation of extracellular traps in human macro- prevents a pathogen-permissive granulocytic inflammation during tuberculosis. phages. Front. Cell. Infect. Microbiol. 7: 278. Nat. Microbiol. 2: 17072. 60. Ramadass, M., J. L. Johnson, A. Marki, J. Zhang, D. Wolf, W. B. Kiosses, 79. Pino-Yanes, M., A. Corrales, M. Casula, J. Blanco, A. Muriel, E. Espinosa, K. Pestonjamasp, K. Ley, and S. D. Catz. 2019. The trafficking protein JFC1 M. Garcıá-Bello, A. Torres, M. Ferrer, E. Zavala, et al; GRECIA and GEN-SEP regulates Rac1-GTP localization at the uropod controlling neutrophil chemotaxis Groups. 2010. Common variants of TLR1 associate with organ dysfunction and and in vivo migration. J. Leukoc. Biol. 105: 1209–1224. sustained pro-inflammatory responses during sepsis. PLoS One 5: e13759.