Rab27a regulates GM‐CSF‐dependent priming of neutrophil exocytosis

Short abstract

Identification of Rab27a as the molecular regulator of neutrophil exocytosis in response to GM‐CSF‐priming.

Abstract

Neutrophil secretory proteins are mediators of systemic inflammation in infection, trauma, and cancer. In response to specific inflammatory mediators, neutrophil granules are mobilized and cargo proteins released to modulate the microenvironment of inflammatory sites and tumors. In particular, GM‐CSF, a cytokine secreted by several immune, nonimmune, and tumor cells, regulates neutrophil priming and exocytosis. Whereas a comprehensive understanding of this process is necessary to design appropriate anti‐inflammatory therapies, the molecular effectors regulating GM‐CSF‐dependent priming of neutrophil exocytosis are currently unknown. With the use of neutrophils deficient in the small GTPase Rab27a or its effector Munc13‐4, we show that although both of these secretory factors control matrix metalloproteinase‐9 (MMP‐9) and myeloperoxidase (MPO) exocytosis in response to GM‐CSF, their involvement in exocytosis after GM‐CSF priming is very different. Whereas GM‐CSF priming‐induced exocytosis is abolished in the absence of Rab27a for all secondary stimuli tested, including TLR7, TLR9, and formyl peptide receptor 1 (Fpr1) ligands, cells lacking Munc13‐4 showed a significant exocytic response to GM‐CSF priming. The mobilization of CD11b was independent of both Rab27a and Munc13‐4 in GM‐CSF‐primed cells unless the cells were stimulated with nucleic acid‐sensing TLR ligand, thus highlighting a role for both Rab27a and Munc13‐4 in endocytic TLR maturation. Finally, the observation that the absence of Rab27a expression impairs the exocytosis of MMP‐9 and MPO under both primed and unprimed conditions suggests that Rab27a is a possible target for intervention in inflammatory processes in which GM‐CSF‐dependent neutrophil priming is involved.

Abbreviations

Fpr

formyl peptide receptor

MMP‐9

matrix metalloproteinase 9

MPO

myeloperoxidase

Munc13‐4‐KO

Munc13‐4^−/− knockout

ODN

oligodeoxynucleotide

Rab27a‐KO

Rab27a^−/− knockout

ROS

reactive oxygen species

TAN

tumor‐associated neutrophil

VEGF

vascular endothelial growth factor

WT

wild‐type

Introduction

GM‐CSF is a pluripotent cytokine secreted by several different cell types, including T cells, macrophages, fibroblasts, smooth muscle cells, bronchial epithelial cells, endothelial cells, and monocytes, and it is also secreted by many types of tumor cells. The biologic effects of GM‐CSF are mediated through the GM‐CSFR, which is widely expressed in hematopoietic cells, and on some nonhematopoietic cells, such as endothelial cells [1, 2]. GM‐CSF is known to regulate neutrophil differentiation, survival, and priming [3, 4]. Pretreatment of neutrophils with GM‐CSF enhances phagocytosis and their oxidative responses to various stimuli, such as C5a, leukotriene B4, and fMLF [4, 5].

GM‐CSF is a proinflammatory cytokine, and a significant increase in GM‐CSF levels has been found to be associated with many inflammatory disorders [1]. In addition, tumor‐derived GM‐CSF is an important regulator of inflammation and immune suppression within the tumor microenvironment [6, 7–8]. Accordingly, the targeting of GM‐CSF by using inhibiting antibodies or by depletion has been shown to suppress pathologic processes in a wide range of inflammatory and autoimmune diseases, such as arthritis, experimental autoimmune encephalomyelitis, asthma and lung inflammation, psoriasis, and cancer [2, 6, 9, 10–11]. Tumor‐derived GM‐CSF has been shown to drive the development of Gr‐1⁺ CD11b⁺ cells that suppress the anti‐tumor T cell response in a model of pancreatic cancer [6]. GM‐CSF is also known to increase the invasive potential of human breast cancer cells by inducing VEGF secretion mediated by neutrophils [12]. Further supporting a role for neutrophils in cancer, neutrophils are observed in vivo in close association with tumor cells and their vasculature and make up a significant portion of the inflammatory cell infiltrate in a variety of mouse models and human cancers [13, 14]. The tumor cells themselves are capable of recruiting these neutrophils, frequently referred to as TANs, by secreting CXC chemokines [15]. Importantly, clinical studies have shown that the presence of neutrophils in tumors is associated with poor prognosis, although some studies indicate that neutrophils may be beneficial at early stages in some types of cancer [16, 17, 18, 19, 20, 21–22]. Importantly, the relationship between GM‐CSF and neutrophil secretory proteins in cancer progression is further suggested by studies showing that GM‐CSF enhancement of tumor invasion is mediated by elevated levels of MMPs [23].

Neutrophils contain different types of granules in which they house an arsenal of antimicrobial peptides (α‐defensins and cathelicidins), pro‐oxidant enzymes (MPO), hydrolytic enzymes (lysozyme, sialidase, and collagenase), proteases (MMPs, cathepsin G, azurocidin, and elastase), cationic phospholipase, and metal chelators (lactoferrin) that are released upon stimulation [24]. The granule proteins MMP‐9 and MPO have been associated with several disease states in cancer. MMP‐9 (stored in specific and tertiary granules) and MPO (stored in azurophilic granules) from TANs have been attributed protumorigenic roles. Tissue‐infiltrating neutrophils are the major source of proangiogenic MMP‐9 in the tumor microenvironment and are known to be involved in tumor progression by facilitating angiogenesis and the progression to invasive cancer [25, 26]. Thus, MMP‐9‐expressing myeloid cells were shown to restore tumor vasculature and allow tumor growth in irradiated tissues in experimental mammary carcinoma [27]. MMP‐9 also regulates angiogenesis by promoting the release of extracellular matrix‐bound or cell‐surface‐bound cytokines, such as VEGF [28]. In addition to its effects on the matrix, it has been shown that MMP‐9 secreted from neutrophils prevents apoptosis of tumor cells in the lung [29]. TANs have also been implicated in promoting the initiation and progression of chronic colitis‐associated colon carcinogenesis by secreting MMP‐9 and neutrophil elastase [30].

MPO is a heme peroxidase enzyme secreted by neutrophils that generates ROS and reactive nitrogen species as a mechanism for pathogen removal. It has been suggested that extended exposure to ROS as a result of chronic inflammation results in DNA damage and genomic instability, leading to malignant cell transformation [31]. Genetic studies have shown that a single nucleotide polymorphism in the human MPO gene promoter region is associated with reduced MPO levels and a lower risk for developing breast cancer [32, 33], hepatoblastoma [34], laryngeal cancer [35], esophageal cancer [36], ovarian cancer [37], gastric cancer [38], and possibly lung cancer [39, 40, 41–42]. Furthermore, patients suffering from chronic obstructive pulmonary disease and lung cancer exhibit increased levels of MPO in serum and bronchoalveolar lavage fluid [43]. Studies using mice genetically deficient in MPO show reduced tumor burden in a Lewis lung carcinoma model, and the same study also showed that treating mice with a tripeptide inhibitor of MPO activity reduced tumor burden [44].

Given the associated roles of GM‐CSF and neutrophil secretory proteins in tumor progression, it is important to understand how GM‐CSF‐mediated neutrophil functions are regulated. In particular, the process of neutrophil vesicular trafficking mediated by GM‐CSF‐induced activation is, as yet, unknown. Previous studies showed the involvement of the small GTPase Rab27a and its effector Munc13‐4 in the regulation of various aspects of neutrophil vesicular trafficking and exocytosis [45, 46]. Rab27a is a small GTPase, which together with its effectors, regulates vesicular trafficking and exocytosis in nonhematopoietic and hematopoietic cells, such as CTLs, NK cells, mast cells, and neutrophils. The mechanisms regulated by Rab27a are cell specific, and Rab27a acts in concert with various effector molecules to regulate different stages of the exocytic process [46]. Rab27a, along with its effector Munc13‐4, is known to regulate azurophilic and gelatinase/specific granule exocytosis in neutrophils in response to fMLF and to the TLR4 agonist LPS [47]. However, little is known about the role of Rab27a or Munc13‐4 in neutrophil exocytosis induced by other stimuli and their function in GM‐CSF priming‐mediated secretion in neutrophils. In this study, we characterize the mechanisms regulating the exocytosis of the different sets of neutrophil granules by GM‐CSF‐mediated priming focusing on Rab27a and Munc13‐4 as regulators of secretion. Our data identify novel mechanisms regulating GM‐CSF‐activated neutrophils and suggest important differences in the mechanisms of control of GM‐CSF priming‐induced neutrophil exocytosis by small GTPases and their effector molecules.

MATERIALS AND METHODS

Animals

Our experiments use C57BL/6 Rab27a‐KO, C57BL/6 Munc13‐4‐KO, and their parental strain, C57BL/6 WT mice. Mice (6–12 wk old) were maintained in a pathogen‐free environment and had access to food and water ad libitum. All animal studies were performed in compliance with the U.S. Department of Health and Human Services Guide for the Care and Use of Laboratory Animals. All studies were conducted according to U.S. National Institutes of Health and institutional guidelines and with approval from the Animal Review Board at The Scripps Research Institute.

Materials

The stimuli used for treating cells included the following: GM‐CSF (200‐15; Shenandoah Biotechnology, Warwick, PA, USA), CpG ODN 1826 (tlrl‐1826; InvivoGen, San Diego, CA, USA), CL097 (tlrl‐c97; InvivoGen), fMLF (Sigma, St. Louis, MO, USA), and PMA (Sigma). The antibodies used for flow cytometry staining are as follows: FITC‐conjugated anti‐Ly6G (clone 1A8; BD Biosciences, San Jose, CA, USA) and Alexa Fluor 647‐conjugated anti‐mouse CD11b (clone M1/70; BD Biosciences).

Mouse neutrophil isolation

Bone marrow‐derived neutrophils were isolated using a Percoll gradient fractionation system, as described [48]. A 3‐layer Percoll gradient was used (52, 64, and 72%), and neutrophils were isolated from the 64% to 72% interface, washed, and used in the assays.

Neutrophil stimulation

Purified mouse neutrophils (1 × 10⁶) were resuspended in phenol red‐free RPMI and either stimulated with GM‐CSF (10 ng/ml) or left untreated for 30 min at 37°C, at which point the cells were stimulated with 5 µM CpG ODN 1826 for 1 h, 10 µg/ml CL097 for 1 h, 1 µM fMLF for 10 min, or 100 ng/ml PMA for 30 min at 37°C. Where indicated, neutrophils were washed after incubation with GM‐CSF or vehicle by spinning down the cells and resuspending the cells in medium, prewarmed at 37°C. This approach was used to eliminate the priming agent and the cargoes secreted during priming before treatment with the second stimulus. Following the treatments, the cells were spun down at 800 g for 10 min, the cell‐free supernatant was collected for ELISA analysis, and the cells were stained for flow cytometry analysis.

Flow cytometry

To stain for cell‐surface markers, cells were blocked in ice‐cold PBS containing 1% BSA and stained with FITC‐conjugated anti‐Ly6G (clone 1A8; BD Biosciences) and Alexa Fluor 647‐conjugated anti‐mouse CD11b (clone M1/70; BD Biosciences). The cells were then washed and fixed in 1.5% paraformaldehyde in PBS. The samples were analyzed using a BD LSR II flow cytometer (BD Biosciences), and the data were analyzed using FlowJo software.

ELISA

MMP‐9 levels in the supernatant were assessed using the Mouse Pro‐MMP‐9 ELISA kit (DY909; R&D Systems, Minneapolis, MN, USA), per the manufacturer’s protocol. MPO levels in the supernatant were assessed using the Mouse Myeloperoxidase DuoSet ELISA (DY3667; R&D Systems), per the manufacturer’s protocol.

Statistical analysis

Data are presented as means, and error bars correspond to sem. Statistical significance was determined using the unpaired or paired Studentˈs t test using GraphPad Prism (version 6), and graphs were made using GraphPad Prism (version 6) software (GraphPad Software, La Jolla, CA, USA). Grubb’s test was used to determine statistical outliers.

RESULTS AND DISCUSSION

GM‐CSF is largely known to regulate neutrophil priming and differentiation [3, 4]. Recent reports have identified GM‐CSF as a determinant for the development of an immunosuppressive type of granulocytes associated with cancer [49, 50]. Here, to identify the vesicular trafficking proteins associated with the mechanism of GM‐CSF‐mediated neutrophil exocytosis and priming of secretion, we focused on the small GTPase Rab27a and the tethering factor Munc13‐4.

Neutrophil exocytosis is stimulated by a variety of agonists, both pathogen‐derived and endogenous ligands. In this work, we focused on the analysis of endocytic TLR ligands, as these stimuli induce neutrophil activation but at the same time, are frequently proposed as potential cancer treatment [51]. We use unmethylated DNA, an agonist of endosomal‐associated TLR9, which is activated by both bacterial DNA and endogenous DNA‐protein complexes. Cells were also vetted against the TLR7 agonist, CL097. The experiments comparatively explore endocytic TLR and nonendocytic ligands under both GM‐CSF‐primed and unprimed conditions. As nonendocytic ligands, we use formylated peptides, synthetic agonists of the plasma membrane receptor Fpr1 that mimics both mitochondrial peptides released during tissue damage and pathogen‐associated molecular patterns, whereas the PKC ligand, phorbol ester (PMA), was used as control.

Firstly, we analyzed the secretion of MMP‐9 (gelatinase B), which is stored in both secondary (specific) and tertiary (gelatinase) granules in neutrophils. In priming studies, the cells were incubated in the presence of GM‐CSF for 30 min before the addition of a second stimulus. MMP‐9 secretion was weakly but significantly increased by stimulation of neutrophils with the TLR9 agonist CpG and the TLR7 ligand CL097 ( Fig. 1 ). fMLF and GM‐CSF were also significant stimulators of MMP‐9 secretion in WT cells (Fig. 1). Preincubation of neutrophils with GM‐CSF significantly amplified the subsequent exocytic response to all stimuli evaluated, suggesting that the growth factor acts as an efficient sensitizer of neutrophil secretion, independent of the nature and the subcellular localization of the receptor for the second agonist.

Figure 1.

Neutrophil secretion of MMP‐9 is induced by endocytic TLR ligands and formylated peptides, and the responses are enhanced upon GM‐CSF treatment. Mouse bone marrow neutrophils were either left unprimed or were primed with GM‐CSF for 30 min before incubation with different stimuli for different times (CpG and CL097 for 1 h, fMLF for 10 min, and PMA for 30 min). The supernatant was collected and tested for secreted MMP‐9 levels by ELISA. Data represent means ± sem from a minimum of 6 mice. **P < 0.01, ***P < 0.001, ****P < 0.0001. The statistical P values are compared with the basal unstimulated sample (−).

Next, we performed experiments to rule out the possibility that the increased secretion of MMP‐9 by neutrophils incubated with GM‐CSF before the stimulation with second agonists is the result of simply additive effects and not priming. In these assays, neutrophils were incubated with GM‐CSF or vehicle and subsequently washed to eliminate the priming agent. The washing step also eliminates the proteins secreted during GM‐CSF treatment and any basal secretion during the first 30 min. The cells were then incubated with the second stimulus, and the secretion by GM‐CSF‐pretreated cells was compared with that observed in the absence of the priming agent. In Fig. 2 , we demonstrate that GM‐CSF efficiently and significantly induces priming of gelatinase (MMP‐9) granule exocytosis in cells washed after priming and subsequently, stimulated with the TLR9 agonist CpG or with fMLF (Fig. 2A and D and C and F, respectively). A similar effect was observed for the TLR7 agonist CL097, although the effect was less marked (Fig. 2B and E). No difference in the percentage of increase in MMP‐9 secretion upon GM‐CSF priming was observed in samples with or without wash post‐GM‐CSF treatment (Fig. 2G). Altogether, our data highlight GM‐CSF as a priming agent of exocytosis for neutrophils stimulated through both plasma membrane‐associated receptors and endocytic receptor ligands.

Figure 2.

Increased MMP‐9 secretion in response to GM‐CSF pretreatment is a result of priming. Mouse bone marrow neutrophils from WT mice were either left unprimed or were primed with GM‐CSF for 30 min at 37°C. Following the treatment, 1 set of samples was washed to remove the supernatant (with wash, D–F), whereas the other set of samples was left unwashed (No wash, A–C). This was followed by addition of either CpG for 1 h (A and D), CL097 for 1 h (B and E), or fMLF for 10 min (C and F). The supernatant was collected and tested for secreted MMP‐9 levels by ELISA. (G) The percentage increase in MMP‐9 secretion upon GM‐CSF pretreatment, followed by the secondary stimulus, compared with treatment with the secondary stimulus alone. Data represent means ± sem from a minimum of 4 mice. *P < 0.05, **P < 0.01, paired t test.

The small GTPase Rab27a has been shown to regulate exocytosis of gelatinase granules. In particular, we previously showed that neutrophils from Rab27a‐KO mice stimulated with the TLR4 agonist LPS and fMLF show decreased MMP‐9 secretion [46]; however, nothing is known about the role of Rab27a in the regulation of MMP‐9 secretion induced by other stimuli. Furthermore, the potential regulation of GM‐CSF‐induced MMP‐9 exocytosis by this small GTPase is currently unknown.

In Fig. 3 , we show that MMP‐9 secretion is significantly impaired in the absence of Rab27a in response to GM‐CSF (Fig. 3A), CpG (Fig. 3B), CL097 (Fig. 3C), fMLF (Fig. 3D), and PMA (Fig. 3E). GM‐CSF itself failed to induce MMP‐9 secretion in the absence of Rab27a (Fig. 3A), thus highlighting the role of this GTPase in the regulation of GM‐CSF‐induced exocytosis. Furthermore, the secretion of MMP‐9 in response to GM‐CSF priming was significantly inhibited in the absence of Rab27a expression in neutrophils (Fig. 3B–D). Taken together, our data suggest that Rab27a not only regulates MMP‐9 secretion in response to several stimuli but also controls the effect mediated by GM‐CSF while priming for MMP‐9 secretion in neutrophils.

Figure 3.

MMP‐9 secretion and GM‐CSF priming of exocytosis are severely impaired in the absence of Rab27a. Mouse bone marrow neutrophils from WT or Rab27a‐KO (Rab27a^−/−) mice were either left unprimed or were primed with GM‐CSF for 30 min before incubation with no secondary stimulus (A), CpG for 1 h (B), CL097 for 1 h (C), fMLF for 10 min (D), or PMA alone for 30 min (E). The supernatant was collected and tested for secreted MMP‐9 levels by ELISA. Data represent means ± sem from a minimum of 7 mice. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

As many of the Rab27a‐dependent secretory mechanisms are modulated by the tethering factor Munc13‐4, we tested how the absence of Munc13‐4 affects gelatinase granule exocytosis. Compared with Rab27a‐KO cells, Munc13‐4‐deficient neutrophils showed a more selective inhibitory pattern. Thus, differently from that observed in Rab27a‐KO cells, GM‐CSF was able to induce MMP‐9 secretion in the absence of Munc13‐4, although the response was significantly decreased compared with that in WT cells ( Fig. 4A ). Cells lacking Munc13‐4 responded poorly to the endocytic agonists CpG and CL097 (Fig. 4B and C), and a significantly reduced but detectable response was also observed for fMLF‐stimulated Munc13‐4‐KO cells (Fig. 4D). Furthermore, Munc13‐4‐KO neutrophils showed a significant GM‐CSF‐mediated priming of MMP‐9 exocytosis when stimulated with CpG, CL097, or fMLF (Fig. 4B–D), suggesting that different from that observed in Rab27a‐KO neutrophils, Munc13‐4 is not essential for GM‐CSF priming of MMP‐9 secretion. Altogether, our data highlight important differences between Rab27a and Munc13‐4 function in the regulation of gelatinase granule exocytosis upon GM‐CSF‐induced priming. Based on these data, we propose that GM‐CSF priming of gelatinase granule exocytosis in response to both plasma membrane receptor and endocytic ligands is a Rab27a‐dependent mechanism, with minimal participation of Munc13‐4 function.

Figure 4.

MMP‐9 secretion is significantly impaired in the absence of Munc13‐4, whereas GM‐CSF priming‐mediated exocytosis is not. Mouse bone marrow neutrophils from WT or Munc13‐4‐KO (Munc13‐4^−/−) mice were either left unprimed or were primed with GM‐CSF for 30 min before incubation with either no secondary stimulus (A), CpG for 1 h (B), CL097 for 1 h (C), fMLF for 10 min (D), or PMA alone for 30 min (E). The supernatant was collected and tested for secreted MMP‐9 levels by ELISA. Data represent means ± sem from n = 9 mice. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. #Statistical outlier determined by Grubb’s test.

Next, we analyzed exocytosis of the most toxic neutrophil granule cargo proteins that are stored in the azurophilic granules. To this end, we measured the release of MPO in response to the same set of stimuli used for MMP‐9 release. MPO secretion by WT cells was significantly up‐regulated under all experimental conditions except when stimulated with the TLR9 agonist CpG alone ( Fig. 5 ). Importantly, pretreatment with GM‐CSF amplified the secretory response to all the stimuli tested, including CL097, the TLR9 agonist CpG, and fMLF. GM‐CSF was also an efficient stimulus of MPO secretion when neutrophils were treated with the CSF in the absence of other stimuli (Fig. 5). Importantly, similar to that observed for MMP‐9, washing‐after‐priming experiments clearly demonstrate that GM‐CSF is indeed a priming agent for MPO secretion for all stimuli evaluated, although the degree of GM‐CSF‐mediated priming was less manifested in CL097‐treated cells ( Fig. 6A–G ).

Figure 5.

Neutrophil secretion of MPO is induced by TLR7 ligands and formylated peptides but not CpG, and the responses are enhanced upon GM‐CSF priming. Mouse bone marrow neutrophils were either left unprimed or were primed with GM‐CSF for 30 min before incubation with different stimuli for different times (CpG and CL097 for 1 h, fMLF for 10 min, and PMA for 30 min). The supernatant was collected and tested for secreted MPO levels by ELISA. Data represent means ± sem from n = 3–6 mice. The statistical P values are compared with the basal unstimulated sample (−). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6.

Increased MPO secretion in response to GM‐CSF pretreatment is a result of priming. Mouse bone marrow neutrophils from WT mice were either left unprimed or were primed with GM‐CSF for 30 min at 37°C. Following the treatment, 1 set of samples was washed to remove the supernatant (with wash, D–F), whereas the other set of samples was left unwashed (No wash, A–C). This was followed by addition of either CpG for 1 h (A and D), CL097 for 1 h (B and E), or fMLF for 10 min (C and F). The supernatant was collected and tested for secreted MPO levels by ELISA. (G) The percentage increase in MPO secretion upon GM‐CSF pretreatment, followed by the secondary stimulus, compared with treatment with the secondary stimulus alone. Data represent means ± sem from a minimum of 4 mice. *P < 0.05, **P < 0.01, ***P < 0.001, paired t test.

To determine the stimuli‐specific regulatory mechanisms of azurophilic granule exocytosis, neutrophils from Rab27a‐KO and Munc13‐4‐KO mice were challenged with TLR7, TLR9, and Fpr1 agonists under both unprimed and primed conditions. The data presented in Figs. 7 and 8 for Rab27a‐KO and Munc3‐4‐KO neutrophils, respectively, suggest that although both trafficking molecules are important for azurophilic granule exocytosis, independently of the stimulus used, the regulation of GM‐CSF‐mediated priming was different. Thus, whereas GM‐CSF‐mediated priming of MPO exocytosis was completely abolished in Rab27a‐KO cells (Fig. 7), Munc13‐4‐KO cells showed a significant enhancement of exocytic response to priming by the CSF (Fig. 8). Altogether, our data support the idea that Rab27a and Munc13‐4 play important roles in the regulation of azurophilic granule exocytosis. In addition, GM‐CSF‐mediated priming of azurophilic granule exocytosis is dependent on Rab27a but can partially circumvent the need for Munc13‐4 in the process of azurophilic granule secretion.

Figure 7.

MPO secretion and GM‐CSF priming of exocytosis are severely impaired in the absence of Rab27a. Mouse bone marrow neutrophils from WT or Rab27a‐KO (Rab27a^−/−) mice were either left unprimed or were primed with GM‐CSF for 30 min before incubation with either no secondary stimulus (A), CpG for 1 h (B), CL097 for 1 h (C), fMLF for 10 min (D), or PMA alone for 30 min (E). The supernatant was collected and tested for secreted MPO levels by ELISA. Data represent means ± sem from n = 3–6 mice. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 8.

MPO secretion is significantly impaired in the absence of Munc13‐4, whereas GM‐CSF priming‐mediated exocytosis is not. Mouse bone marrow neutrophils from WT or Munc13‐4‐KO (Munc13‐4^−/−) mice were either left unprimed or were primed with GM‐CSF for 30 min before incubation with either no secondary stimulus (A), CpG for 1 h (B), CL097 for 1 h (C), fMLF for 10 min (D), or PMA alone for 30 min (E). The supernatant was collected and tested for secreted MPO levels by ELISA. Data represent means ± sem from n = 9 mice. *P < 0.05, **P < 0.01, ***P < 0.001.

CD11b is a β₂‐integrin subunit that plays a major role in inflammation. Recent studies have linked CD11b to cancer progression by showing that interference with CD11b by means of specific antibodies increases the tumor response to radiotherapy by decreasing myeloid cell presence in tumors [52]. These data suggest that the effect of locally produced GM‐CSF in the tumor environment on CD11b expression is important in the context of tumor growth. Supporting this, recent studies established that abrogation of tumor‐derived GM‐CSF inhibited the recruitment of Gr‐1⁺ CD11b⁺ cells to the tumor microenvironment and blocked tumor development and suggested that tumor‐derived GM‐CSF is an important regulator of inflammation and immune suppression within the tumor microenvironment [6]. In this context, a better understanding of the regulation of CD11b expression at the plasma membrane of neutrophils in response to GM‐CSF is of great importance. Here, we evaluated the role of Rab27a and Munc13‐4 in CD11b expression in experimental conditions that mimic GM‐CSF‐dependent priming and activation. In Fig. 9 , we show that all agonists tested significantly induced CD11b up‐regulation in WT cells, with fMLF as the more efficient, single physiologic stimulus. The cells also responded relatively strongly to GM‐CSF, whereas the endocytic TLR ligands CL097 and CpG also elicited significant CD11b up‐regulation. GM‐CSF pretreatment triggered a strong priming response for all stimuli evaluated in WT cells. The responses to all stimuli after GM‐CSF prestimulation were stronger than the addition of individual responses, supporting that mechanisms of amplification, rather than mobilization of CD11b from different intracellular pools, operate when cells are exposed to GM‐CSF. The up‐regulation of CD11b was largely independent of Rab27a or Munc13‐4 function, with the exception of the response to the TLR9 agonist CpG. The defect in up‐regulation of CD11b in response to CpG is associated with a defect in the endosomal maturation, which is necessary for CpG‐mediated signaling. This is in agreement with previous results from our laboratory, showing that Munc13‐4 is necessary for late endosomal maturation and TLR9 signaling in a mechanism that involves calcium‐dependent interaction between Munc13‐4 and the late endosomal soluble N‐ethylmaleimide‐sensitive factor attachment protein receptor syntaxin 7 [53]. Neither Rab27a‐ nor Munc13‐4‐deficient neutrophils showed any defect in GM‐CSF priming‐mediated CD11b up‐regulation on the cell surface.

Figure 9.

Up‐regulation of CD11b is largely independent of Rab27a and Munc13‐4 with the exception of the response to the TLR9 agonist. Mouse bone marrow neutrophils were either left unprimed or were primed with GM‐CSF for 30 min before incubation with different stimuli for different times (CpG and CL097 for 1 h, fMLF for 10 min, and PMA for 30 min). The cells were then stained for CD11b using a CD11b‐488 antibody and analyzed by flow cytometry. MFI, Mean fluorescence intensity. Data represent means ± sem from n = 3–6 mice. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

A role for neutrophil function in cancer has emerged in recent years, with many studies supporting that neutrophil secretory proteins induce immunosuppression associated with tumor progression. In particular, MMP‐9, a metalloproteinase abundant in neutrophil secondary and tertiary granules, and MPO, an oxidant‐producing enzyme, have been largely associated with neovascularization, angiogenesis, and tumor progression in cancers [25, 26, 44]. Furthermore, GM‐CSF, a proinflammatory cytokine released by a variety of immune and cancer cells, has been shown to play a central role in the regulation of inflammation and immune suppression within the tumor microenvironment [6, 7–8, 12, 49, 54]. In particular, GM‐CSF is a determinant for the development of an immunosuppressive type of granulocytes associated with cancer [6, 49, 55]. In this context, the data presented here, showing that GM‐CSF enhances the secretion of MMP‐9 and MPO in a Rab27a‐dependent manner, are highly significant.

Rab27a, a small GTPase that regulates exocytosis of neutrophils and other cells of hematopoietic origin, has recently emerged as a factor associated with cancer progression and poor survival. Recent reports have indicated that enhanced expression of Rab27a is associated with breast and pancreatic cancer progression [56, 57]. Furthermore, differential expression of Rab27a and Rab27b correlates with clinical outcome in hepatocellular carcinoma, and disrupting Rab27a expression reduces exosome release, tumor growth, and metastasis in melanomas [58, 59]. Although there is a correlation between Rab27a expression and cancer, the molecular mechanism is not well understood. Thus, the observation that Rab27a is a central regulator of GM‐CSF‐induced, neutrophil‐mediated release of protumorigenic cargos is highly relevant. Furthermore, in this work, we have established that the Rab27a effector Munc13‐4 plays a secondary role in this mechanism, suggesting that whereas Rab27a is potentially a good therapeutic target to prevent GM‐CSF‐ and neutrophil‐dependent tumor progression, the most successful approaches should also consider the Rab27a interaction with other effectors. Slp1/JFC1 (synaptotagmin‐like protein 1), a Rab27a effector that together with Rab27a, is a central regulator of neutrophil exocytosis, is potentially a good candidate.

Altogether, our data suggest that Rab27a is essential for the regulation of the neutrophil secretory response to GM‐CSF, both as a direct stimulus and as a priming agent, whereas Munc13‐4 plays a less significant role in GM‐CSF priming‐induced exocytosis of tertiary and azurophilic granules. As GM‐CSF‐induced cargo secretion from specific and azurophilic granules is directly involved in inflammation and tumor progression, our data highlight neutrophil Rab27a as a possible target for novel therapies to treat systemic inflammation and cancer.

AUTHORSHIP

M.R. performed experiments, analyzed data, and contributed ideas, comments, and to manuscript writing. J.L.J. contributed ideas, comments, and input for manuscript writing. S.D.C. conceived of the manuscript, contributed ideas and comments, and wrote the manuscript with input from M.R. and J.L.J.

DISCLOSURES

The authors declare that they have no conflicts of interest with the contents of this article.