|Home | About | Journals | Submit | Contact Us | Français|
Platelets are potent regulators of neutrophil accumulation in septic lung damage. We hypothesized that platelet‐derived CXCL4 might support pulmonary neutrophilia in a murine model of abdominal sepsis.
Polymicrobial sepsis was triggered by coecal ligation and puncture (CLP) in C57BL/6 mice. Platelet secretion of CXCL4 was studied by using confocal microscopy. Plasma and lung levels of CXCL4, CXCL1 and CXCL2 were determined by elisa. Flow cytometry was used to examine surface expression of Mac‐1 on neutrophils.
CLP increased CXCL4 levels in plasma, and platelet depletion reduced plasma levels of CXCL4 in septic animals. Rac1 inhibitor NSC23766 decreased the CLP‐enhanced CXCL4 in plasma by 77%. NSC23766 also abolished PAR4 agonist‐induced secretion of CXCL4 from isolated platelets. Inhibition of CXCL4 reduced CLP‐evoked neutrophil recruitment, oedema formation and tissue damage in the lung. However, immunoneutralization of CXCL4 had no effect on CLP‐induced expression of Mac‐1 on neutrophils. Targeting CXCL4 attenuated plasma and lung levels of CXCL1 and CXCL2 in septic mice. CXCL4 had no effect on neutrophil chemotaxis in vitro, indicating it has an indirect effect on pulmonary neutrophilia. Intratracheal CXCL4 enhanced infiltration of neutrophils and formation of CXCL2 in the lung. CXCR2 antagonist SB225002 markedly reduced CXCL4‐provoked neutrophil accumulation in the lung. CXCL4 caused secretion of CXCL2 from isolated alveolar macrophages.
Rac1 controls platelet secretion of CXCL4 and CXCL4 is a potent stimulator of neutrophil accumulation in septic lungs via generation of CXCL2 in alveolar macrophages. Platelet‐derived CXCL4 plays an important role in lung inflammation and tissue damage in polymicrobial sepsis.
|GPCRs a||Catalytic receptors b|
|CXCR2||Mac‐1 (integrin, alpha M subunit)|
These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (a,b,cAlexander et al., 2013a,2013b,2013c).
Management of patients with sepsis poses a significant challenge to clinicians. One reason is related to the lack of a comprehensive understanding of the underlying pathophysiology in sepsis. It is well known that the lung is a very sensitive and critical target organ in sepsis (Parrillo, 1993). Although neutrophil recruitment constitutes a central component in the host response to bacterial infections (Gorbach and Bartlett, 1974; Reutershan et al., 2005), excessive neutrophil responses are known to cause organ damage in sepsis (Issekutz and Issekutz, 1992; Reutershan et al., 2005; Basit et al., 2006; Asaduzzaman et al., 2008). Based on experiments blocking neutrophil recruitment, several studies have shown that neutrophil accumulation constitutes a rate‐limiting step in septic lung damage (Asaduzzaman et al., 2008, 2009b; Hasan et al., 2011). It is interesting to note that accumulating data implicates platelets in the development of septic lung injury. For example, several studies have shown that platelets promote neutrophil activation and recruitment to the lung in sepsis (Zarbock et al., 2007; Asaduzzaman et al., 2009a). One study has shown that platelet‐derived CD40L is a potent inducer of neutrophil infiltration in septic lung injury (Rahman et al., 2009). However, platelets harbour a wide spectrum of different pro‐inflammatory compounds, such as chemokines, which could mediate platelet‐dependent accumulation of neutrophils in septic lung damage.
CXCL4 is one of the most abundant chemokines in platelets and belongs to the CXC chemokine family although it lacks an ELR sequence needed for binding to chemotactic CXCRs on neutrophils (Clark‐Lewis et al., 1993). The literature on the chemotactic activity of CXCL4 is complex and contradictory. For example, one early study reported that CXCL4 exerts chemotactic activity toward neutrophils (Deuel et al., 1981). However, these findings could not be confirmed in later studies, and the early findings may have been caused by contamination with other chemokines (Petersen et al., 1996). The studies on the role of CXCL4 have also been complicated due to the absence of clearly defined CXCL4 receptors. CXCL4 binds avidly to glycosaminoglycans, such as a chondroitin sulphate proteoglycan on neutrophils (Petersen et al., 1998). In addition, CXCL4 has been shown to bind to CXCR3B, which is a splice variant of CXCR3 expressed in humans but not in mice (Lasagni et al., 2003; Kowalska et al., 2010). Another study has reported that CXCL4 also can bind to CXCR3A on human T‐lymphocytes (Mueller et al., 2008). Nonetheless, experimental evidence has shown that CXCL4 plays a role in regulating neutrophil recruitment and tissue damage in complex inflammatory disease models, such as liver fibrosis (Zaldivar et al., 2010) and intestinal reperfusion injury (Lapchak et al., 2012). One study reported that plasma levels of CXCL4 are elevated in patients with sepsis (Lorenz and Brauer, 1988), raising the question of whether CXCL4 can regulate neutrophil‐dependent tissue damage in abdominal sepsis.
The intracellular signalling cascades triggering platelet secretion of CXCL4 are not well understood. We recently reported that Rac1, a member of the Ras‐homologus (Rho) family, not only plays a key function septic lung damage (Hwaiz et al., 2013) but also controls platelet secretion of CD40L in sepsis (Hwaiz et al., 2014). Other studies have identified an essential role of Rac1 in lamellipodia formation, phospholipase Cγ2 activation, granule secretion and clot retraction in platelets (McCarty et al., 2005; Akbar et al., 2007; Pleines et al., 2009). Thus, we hypothesized that Rac1 might be involved in the secretion of CXCL4 from platelets in the present study.
Thus, one aim of the present study was to examine the function of Rac1 in regulating platelet secretion of CXCL4. Moreover, we investigated the role of CXCL4 in regulating pulmonary neutrophilia and tissue injury in polymicrobial sepsis.
All experiments were performed using male C57BL/6 mice (Taconic Bioscience, Lille Skensved, Denmark) 8–9weeks old (20–25g) in accordance with the legislation on the protection of animals and were approved by the Regional Ethical Committee for Animal Experimentation at Lund University, Sweden. Animals were kept in a pathogen‐free facility on a 12–12h light–dark cycle and had free access to food (R3 breeding food for rats and mice; Lantmännen, Kemstad, Sweden) and tap water. Mice were housed for 1week before use in the experiments. Animals were kept at a maximum of seven mice per cage with environment enrichment, such as a little house and toys. Animals were anaesthetized by i.p. administration of 75mg ketamine hydrochloride (Hoffman‐La Roche, Basel, Switzerland) and 25mg xylazine (Janssen Pharmaceutica, Beerse, Belgium) ·kg−1 body weight. Animals were evaluated every 6h during the experiments. At each inspection, mice that appeared moribund, such as lack of response to stimulation or no movement, and animals with respiratory distress or agonal breathing were killed by CO2 inhalation and cervical dislocation. The ARRIVE guidelines (Kilkenny et al., 2010) and an editorial on the application of ARRIVE guidelines to pharmacological studies (McGrath et al., 2010) were consulted for all in vivo studies involving animals.
Abdominal sepsis was induced in anaesthetized mice by puncture of the coecum as previously described in detail (Wichterman et al., 1980). First, the abdomen was opened, and the coecum was filled with faeces by milking stool backward from the ascending colon. Then a ligature was placed below the ileocoecal valve, and the coecum was soaked with PBS (pH7.4). The coecum was punctured twice with a 21‐gauge needle, and a small amount of bowel contents was extruded. The coecum was then returned into the peritoneal cavity, and the abdominal wall was closed. Animals were treated i.p. with vehicle (dH2O) or with 5mg·kg−1 of the Rac1 inhibitor, NSC23766 (N6‐[2‐[[4‐(Diethylamino)‐1‐methylbutyl] amino]‐6‐methyl‐ 4‐pyrimidinyl]‐2 methyl‐4, 6‐quinolinediamine trihydrochloride, Tocris Bioscience, Bristol, UK). This selection of dose of NSC23766 was based on previous publications (Hwaiz et al., 2013, 2014). NSC23766 is a specific and reversible Rac1 inhibitor that competitively inhibits the interaction between Rac1 and Rac‐specific guanine nucleotide exchange factors. NSC23766 does not affect the activity of related Rho GTPases, including Cdc42 or RhoA in vitro or in vivo (Gao et al., 2004; Akbar et al., 2006). A control antibody (clone 141945 R&D Systems, Minneapolis, MN, USA) or a monoclonal antibody against murine CXCL4 (clone 140910, R&D Systems) was injected i.p. (10µg per mouse) 30min before coecal ligation and puncture (CLP) induction. In order to study the role of platelets, a platelet‐depleting antibody directed against murine CD42b (GP1bα, 1.0mg·kg−1, Emfret Analytics GmbH & Co. KG, Wurzburg, Germany) or an isotype control antibody (clone R3–34, 1.0mg·kg−1, BD Biosciences Pharmingen, San Jose, CA, USA) was given i.p. 2h before the CLP. Sham mice underwent the same surgical procedures, that is, laparotomy and resuscitation, but the coecum was neither ligated nor punctured. The animals were then returned to their cages and provided food and water ad libitum. Animals were re‐anaesthetized 6 and 24h after CLP induction. The left lung was ligated and excised after lung perfusion for oedema measurement. The right lung was used for collecting bronchoalveolar lavage fluid (BALF) in which neutrophils were counted. Next, the lung was perfused via the heart with PBS, and one part of the right lung was fixed in formaldehyde for histology, and the remaining lung tissue was snap‐frozen in liquid nitrogen and stored at −80°C for later elisa and myeloperoxidase (MPO) assays as described subsequently. Anaesthetized animals were killed by cervical dislocation.
Lung tissue was thawed and homogenized in 0.02M phosphate buffer (pH7.4). Supernatant was discarded after centrifugation for 10min at 8770g at room temperature, and then the pellet was dissolved by adding 1mL of 0.5% hexadecyltrimethylammonium bromide. Samples were stored at −20°C overnight and then thawed and kept at in a 60°C water bath for 2h followed by 90s sonication. Supernatant was collected after 5min centrifugation at 8770g, and the MPO activity in the supernatant was determined spectrophotometrically as the MPO‐catalysed change in absorbance in the redox reaction of H2O2 (450nm, with a reference filter 540nm, 25°C) as described previously (Krawisz et al., 1984). Values were expressed as MPO U·g−1 tissue.
Animals were placed supine, and the trachea was exposed by dissection. A catheter was inserted into the trachea. BALF was collected by five washes of 1mL of PBS containing 5mM EDTA. The number of polymorphonuclear leukocytes (PMNLs) was counted in a Burker chamber. The left lung was excised and then weighed. The left lung was then dried at 60°C for 72h and re‐weighed. The change in the ratio of wet weight to dry weight was used as indicator of lung oedema formation.
Intratracheal administration of CXCL4 was performed via a skin incision over the trachea and injection of murine CXCL4 (1µg, PeproTech Nordic, Stockholm, Sweden) into the trachea with a 27G needle in anaesthetized animals. Negative control mice underwent the same surgical procedures but received only PBS intratracheally. Fourhours after CXCL4 challenge, BALF was collected as described previously for analysis of CXCL2 levels and number of neutrophils. In certain experiments, a CXCR2 antagonist (SB225002, 4mg·kg−1, Calbiochem, Merck, Darmstadt, Germany) or vehicle (PBS) was administered i.p. 10min before intratracheal injection of CXCL4.
Lung tissue was fixed in 4% formaldehyde phosphate buffer overnight and then dehydrated and paraffin embedded. Six micrometre sections were stained with haematoxylin and eosin. Lung injury was quantified in a blinded manner by use of a modified scoring system (Carraway et al., 2003; Borzone et al., 2007), including size of alveoli, thickness of alveolar septae, alveolar fibrin deposition and neutrophil scoring system infiltration graded on a zero (absent) to four (extensive) scale. In each tissue sample, five random areas were scored, and the mean value was calculated. The histology score is the sum of all four parameters.
For analysis of surface expression of Mac‐1 on circulating neutrophils, blood was collected (1:10 acid citrate dextrose) 6h after CLP induction and incubated (10min at room temperature) with an anti‐CD16/CD32 antibody‐blocking Fcγ III/II receptors to reduce non‐specific labelling and then incubated with phycoerythrin (PE)‐conjugated anti‐Ly6G (clone RB6‐8C5, rat IgG2b, eBioscience, Frankfurt, Germany) and FITC‐conjugated anti‐Mac‐1 (clone M1/70, integrin αM China, rat IgG2b κ, BD Biosciences Pharmingen) antibodies. Alveolar macrophages were isolated as described later and incubated with an anti‐CD16/CD32 antibody‐blocking Fcγ III/II receptors and a peridininchlorophyll protein‐Cy5.5‐conjugated anti‐mouse F4/80 antibody (clone BM8, eBioscience, Frankfurt, Germany) and a FITC‐conjugated anti‐Ly6G (clone 1A8, BD Biosciences) antibody. Cells were fixed, and erythrocytes were lysed; neutrophils were recovered following centrifugation. Flow‐cytometric analysis was performed according to standard settings on a FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA, USA) and analysed with Cell‐Quest Pro software (BD Biosciences).
CXCL1, CXCL2 and CXCL4 levels in lung tissue and plasma were analysed by use of double antibody Quantikine elisa kits (R&D Systems, Europe, Abingdon, Oxon, UK) using recombinant murine CXCL1, CXCL2 and CXCL4 as standards.
Neutrophil chemotaxis was performed as previously described (Smith et al., 1991). Neutrophils were isolated from bone marrow by use of Ficoll‐Paque™ as previously described in detail (Swamydas and Lionakis, 2013). Briefly, 1.5 × 106 neutrophils were placed in the upper chamber of the transwell inserts (5µm pore size; Corning Costar, Corning, NY, USA). Inserts were placed in wells containing medium alone (control) or medium plus murine CXCL2 (0.1µg·mL−1, R&D Systems) or murine CXCL4 (0.1, 0.5 and 1µg·mL−1, R&D Systems). After 120min, inserts were removed, and migrated neutrophils were stained with Turks solution. Chemotaxis was determined by counting the number of migrated neutrophils in a Burker chamber.
Bronchoalveolar lavage fluid was collected as described previously from healthy mice. Cells were isolated by centrifugation (450g, 10min) and resuspended in RPMI1644. Next, cells were washed three times and then allowed to rest and adhere for 4h. Adherent cells were resuspended in RPMI1644 media, and purity of alveolar macrophages was determined by staining cells with anti‐F4/80 and anti‐Ly6G antibodies as described previously. Macrophages were defined as F4/80+/Ly6G− cells. The purity of alveolar macrophages was more than 97% (not shown). Alveolar macrophages were co‐incubated with recombinant murine CXCL4 (0.2, 0.3, 0.4 and 0.5µg·mL−1) for 4h (37°C) and then CXCL2 levels were measured by elisa.
Blood collected in syringes containing 0.1mL of acid–citrate–dextrose was immediately diluted with equal volumes of modified Tyrode solution (1µg·mL−1 PGE1 and 0.1U·mL−1 apyrase) and centrifuged (200g, 5min). Platelet‐rich plasma was collected and centrifuged (800g, 15min), and pellets were resuspended in modified Tyrode solution. After being washed once more (10000g, 5min), 0.5 × 106 platelets were seeded on a chamber slide coated with fibrinogen (20µg·mL−1). Adherent platelets were stimulated with the PAR4 agonist (AYPGKF, 200μM, 37°C) with and without NSC23766 (100μM). Platelets were fixed with 2% paraformaldehyde for 5min and washed and blocked with 1% goat serum for 45min. Platelets were then permeabilized with 0.15% Triton X‐100 for 15min followed by washing and incubation with an anti‐CD16/CD32 antibody (10min) to block Fcγ III/II receptors and reduce non‐specific labelling and a rabbit polyclonal primary antibody against CXCL4 (bs‐2548R, Bioss, Boston, MA, USA) for 2h. Chamber slides were washed and incubated with FITC‐conjugated anti‐rabbit secondary antibody (Cell Signaling Technology, Beverly, MA, USA) and platelet‐specific PE‐conjugated anti‐CD41 (clone MWReg30, eBioscience, San Diego, CA, USA) for 1h. Chamber slides were washed three times, and confocal microscopy was performed using Meta 510 confocal microscopy (Carl Zeiss, Jena, Germany). FITC and PE were excited by 488 and 543nm laser lines, and corresponding emission wavelengths of FITC and PE were collected by the filters of 500–530 and 560–590nm respectively. The pinhole was ~1 airy unit, and the scanning frame was 512 × 512pixels. The fluorescent intensity was calculated by use of ZEN2009 software.
Rac1 activity was determined in platelets from sham and CLP mice pretreated with vehicle (dH2O) or NSC23766 by active Rac1 pull‐down and detection kit using the protein‐binding domain of GST‐PAK1, which binds with the GTP‐bound form of Rac1 (Pierce Biotechnology, Rockford, IL, USA). Briefly, platelets were suspended in lysis buffer on ice and centrifuged (16000g, 15min). Ten microlitres from each lysate were removed to measure protein content using Pierce BCA Protein Assay Reagent (Pierce Biotechnology), and the rest was used for the pull‐down assay. Supernatant containing equal amount of proteins was then diluted with 2× SDS sample buffer and boiled for 5min. Proteins were separated using SDS‐PAGE (10–12% gel). After being transferred to a nitrocellulose membrane (Bio‐Rad, Hercules, CA, USA), blots were blocked with Tris buffer saline/Tween20 containing 3% BSA at room temperature for 1h, followed by incubation with an anti‐Rac1 antibody (1:1000) at 4°C overnight. Binding of the antibody was detected using peroxidase‐conjugated anti‐mouse antibody (1:100000, Pierce Biotechnology) at room temperature for 2h and developed by Immun‐Star WesternC Chemiluminescence Kit (Bio‐Rad). Total Rac1 and β‐actin were used as loading controls.
Sample size calculations were based on a previous study (Asaduzzaman et al., 2009a) with an effect size of 40% (mean value 100 in sham and 60 in treated animals) with SD of 20. This, together with a power of 0.8 and a P‐value of 0.05, required a sample size of at least four animals per group. Data were presented as mean values ± SEM. Statistical evaluations were performed using Kruskal–Wallis one‐way ANOVA on ranks followed by multiple comparisons versus control group (Dunnett's method). P < 0.05 was considered significant, and n represents the total number of mice in each group. Statistical analysis was performed by using SigmaPlot® 10.0 software (Systat Software, Chicago, IL, USA).
Plasma levels of CXCL4 were low but detectable in sham animals (Figure 1A). CLP increased CXCL4 levels in plasma by 41‐fold (Figure 1A). To deplete animals of platelets, an anti‐CD41 antibody was administered before CLP induction. It was observed that platelet depletion reduced plasma levels of CXCL4 by 87% in septic animals (Figure 1A), indicating that platelets are the dominating source of CXCL4 in abdominal sepsis. Treatment with the Rac1 inhibitor NSC23766 had no effect on CXCL4 levels in the plasma of control animals (Figure 1A). However, administration of NSC23766 reduced plasma levels of CXCL4 by 77% in CLP animals (Figure 1A). Moreover, we found that CLP increased Rac1‐GTP levels in platelets, indicating that Rac1 is activated in platelets in septic animals (Figure 1B and and1C).1C). Administration of NSC23766 abolished CLP‐evoked Rac1 activation in platelets (Figure 1B and and1C),1C), demonstrating that NSC23766 is an effective inhibitor of Rac1 activation. We stimulated isolated platelets with a PAR4 agonist (AYPGKF) in order to determine the direct role of Rac1 in regulating platelet secretion of CXCL4. It was found that CXCL4 was present in unstimulated platelets and that co‐incubation with the PAR4 agonist reduced intracellular levels of CXCL4 in platelets (Figure 2A and B). Notably, co‐incubation with NSC23766 markedly decreased the PAR4 agonist‐induced secretion of CXCL4 from the platelets (Figure 2A and and22B).
It was found that CLP increased lung oedema and that immunoneutralization of CXCL4 reduced oedema formation in the lung by 65% in septic animals (Figure 3A). The lung injury in septic mice was characterized by severe destruction of tissue microarchitecture, extensive oedema of and massive infiltration of neutrophils (Figure 3B). Inhibition of CXCL4 attenuated CLP‐induced tissue damage and neutrophil accumulation in the lung (Figure 3B). Quantification of the morphological damage revealed that CLP significantly enhanced lung damage score and that immunoneutralization of CXCL4 decreased CLP‐evoked tissue injury in the lung by 67% (Figure 3C).
CLP increased pulmonary levels of MPO by 24‐fold, and immunoneutralization of CXCL4 reduced MPO activity in the lung by 57% in septic mice (Figure 4A). CLP increased the number of alveolar neutrophils by 19‐fold, and administration of the anti‐CXCL4 antibody decreased the number of alveolar neutrophils by 42% in the inflamed lungs (Figure 4B). Neutrophil expression of Mac‐1 was increased in septic mice (Figure 4C). Inhibition of CXCL4 function had no effect on neutrophil expression of Mac‐1 (Figure 4C). In vitro CXCL2 triggered robust migration of isolated neutrophils, whereas CXCL4 exerted no chemotactic effect on neutrophils (Figure 4D), suggesting that the CXCL4‐dependent accumulation of neutrophils is not a direct action on neutrophils but rather an indirect effect of CXCL4.
Plasma and lung levels of CXCL1 and CXCL2 were low but detectable in sham mice (Figure 5). We found that CXCL1 and CXCL2 levels in the plasma increased by 1183‐fold and 1153‐fold, respectively, in septic animals (Figure 5A and and5B).5B). Inhibition of CXCL4 decreased CLP‐induced plasma levels of CXCL1 by 94% and CXCL2 by 95% (Figure 5A and and5B).5B). Moreover, CLP increased lung levels of CXCL1 by 144‐fold and CXCL2 by 581‐fold (Figure 5C and and5D).5D). Immunoneutralization of CXCL4 decreased CLP‐induced increases of CXCL1 and CXCL2 by 85% and 95% respectively (Figure 5C and and5D).5D). We next studied direct effects of CXCL4 on neutrophil accumulation in the lung. Intratracheal administration of CXCL4 enhanced pulmonary levels of CXCL2 (Figure 6A) and the number of alveolar neutrophils (Figure 6B). In addition, treatment with the CXCR2 antagonist SB225002 reduced CXCL4‐triggered recruitment of neutrophils by 82% in the lung (Figure 6C). We next isolated alveolar macrophages and co‐incubated them with CXCL4 and found that CXCL4 increased CXCL2 formation (Figure 6D). This finding was repeated in RAW264.7 cells showing that CXCL4 enhanced secretion of CXCL2 from macrophages (not shown).
Our present findings indicate an important function of Rac1‐mediated secretion of CXCL4 from platelets in sepsis. In addition, this study reveals key mechanisms controlling CXCL4‐dependent pulmonary accumulation of neutrophils in abdominal sepsis. These results point to an important role of platelets in sepsis and suggest that inhibition of Rac1 signalling and/or CXCL4 function might be useful strategies to ameliorate septic lung damage.
Platelets are not only critical in wound healing and thrombosis but also exert numerous pro‐inflammatory functions in the host response to bacterial invasion (Rahman et al., 2009, 2013; Hwaiz et al., 2014). For example, data have shown that platelets regulate numerous aspects of leukocyte responses to severe infections (Asaduzzaman et al., 2008, 2009a). One such key aspect of the inflammatory response is neutrophil activation and accumulation at sites of microbial invasion. For example, platelet‐derived CD40L has been reported to regulate sepsis‐evoked neutrophil activation and tissue recruitment (Rahman et al., 2009, 2013). Notably, platelets harbour several other pro‐inflammatory substances, including a plethora of different chemokines (Yan et al., 1994; Flad and Brandt, 2010). The most abundant chemokines in platelets are CCL5 and CXCL4, but these chemokines are poor stimulators of neutrophil migration (Petersen et al., 1996; Hartl et al., 2008). Instead, CCL5 and CXCL4 are potent activators lymphocytes, macrophages and eosinophils (McColl et al., 1993; Hartl et al., 2008). Furthermore, studies in the literature have reported that targeting CXCL4 can decrease neutrophil recruitment in models of inflammatory diseases (Zaldivar et al., 2010; Grommes et al., 2012; Lapchak et al., 2012). Interestingly, we found that platelet depletion greatly reduced plasma levels of CXCL4 in CLP animals, suggesting that platelets are a dominating source of circulating CXCL4 in polymicrobial sepsis. Because the mechanisms of platelet secretion of CXCL4 are not known, we wanted first to examine the signalling mechanisms controlling platelet secretion of CXCL4. We observed that platelets from septic mice exhibited increased Rac1 activity. In addition, treatment with the Rac1 inhibitor NSC23766 prior to CLP induction markedly decreased plasma levels of CXCL4 in septic animals, indicating that Rac1 signalling is a critical component in regulating circulating levels of CXCL4 in abdominal sepsis. Next, we wanted to determine the direct role of Rac1 in controlling platelet secretion of CXCL4. It was found that co‐incubation of platelets with NSC23766 abolished the PAR4 agonist‐induced platelet secretion of CXCL4 in vitro, showing for the first time that Rac1 regulates CXCL4 secretion from platelets. In this context, it is interesting to note that Rac1 was demonstrated to be involved in agonist‐evoked mobilization of P‐selectin and platelet‐derived growth factor in platelets (Akbar et al., 2007; Dwivedi et al., 2010), and considering that P‐selectin, platelet‐derived growth factor and CXCL4 are localized in platelet α‐granules (Akbar et al., 2007; Galkina and Ley, 2007; Gleissner et al., 2008; Blair and Flaumenhaft, 2009), these present findings suggest that Rac1 is a potential regulator of α‐granule secretion in platelets.
Activation of innate immune cells is a landmark in the host response to severe infections and is an important cause of acute lung injury characterized by disturbed gaseous exchange (Remick, 2007; Asaduzzaman et al., 2009a). In the present study, we demonstrate that inhibition of CXCL4 reduces oedema formation and tissue damage in the lungs of septic mice, suggesting that CXCL4 exerts a causative role in the development of acute lung injury in polymicrobial sepsis. In this context, it should be noted that one study reported that overexpression of CXCL4 protects against endotoxin‐induced acute lung injury (Kowalska et al., 2007). In contrast, another study showed that inhibition of CXCL4 protects against endotoxin‐induced acute lung injury (Grommes et al., 2012). The reason for these discrepant findings is not known, but it is well known that administration of endotoxin alone is not an optimal substitute for studying polymicrobial sepsis. Nonetheless, our present findings suggest that CXCL4 plays a pro‐inflammatory role in abdominal sepsis. This notion is in line with recent observations reporting a pro‐inflammatory role of CXCL4 in complex diseases, such as atherosclerosis and liver fibrosis (Zaldivar et al., 2010; Karshovska et al., 2014). However, administration of the anti‐CXCL4 antibody 2h after induction of CLP had no effect on inflammation and tissue damage in the septic lung (not shown). Our data showed that immunoneutralization of CXCL4 reduced pulmonary MPO activity, a marker of neutrophil accumulation, by 57% in septic animals, which correlated well with the 42% reduction of neutrophil numbers in the bronchoalveolar space. These findings suggest that CXCL4 is an important regulator of pulmonary neutrophilia in polymicrobial sepsis. This notion is also supported by a recent study showing that neutrophil accumulation in the lung in a model of mesenteric ischemia and reperfusion is decreased in mice lacking CXCL4 (Lapchak et al., 2012). Knowing that neutrophil infiltration is a rate‐limiting step in septic lung injury, our data suggest that the protective effect of inhibiting CXCL4 is related to the reduction in neutrophil recruitment in the lung.
Neutrophil accumulation at extravascular sites of tissue injury and inflammation is a multistep process mediated by specific adhesion molecules on neutrophils, such as CD162 and Mac‐1 (Asaduzzaman et al., 2008; Zhang et al., 2013). Herein, we studied whether targeting of CXCL4 could reduce neutrophil up‐regulation of Mac‐1. However, inhibition of CXCL4 had no effect on neutrophil expression of Mac‐1 in septic mice, indicating that CXCL4 is not involved in Mac‐1 expression on neutrophils. Tissue trafficking of neutrophils is coordinated by secreted CXC chemokines, including CXCL1 and CXCL2 (Tekamp‐Olson et al., 1990). Previous reports have shown that CXC chemokines play an important role in septic lung injury (Hasan et al., 2013; Hwaiz et al., 2013). Herein, it was observed that immunoneutralization of CXCL4 abolished plasma and lung levels of CXC chemokines in CLP animals. These results indicate that CXCL4 might control neutrophil accumulation indirectly via generation of CXC chemokines in the septic lung. This conclusion is also in line with our present results demonstrating that local intratracheal challenge with CXCL4 enhanced generation of CXCL2 and neutrophil recruitment in the lung. That local injection of CXCL4 can provoke neutrophil infiltration in the lung is supported by a previous study showing that local administration of CXCL4 triggers accumulation of neutrophils in the skin (Sharpe et al., 1991). Moreover, we observed that blocking CXCR2 function markedly decreased pulmonary neutrophilia induced by intratracheal administration of CXCL4, which further supports the conclusion that CXCL4 triggers neutrophil accumulation in the lung via generation of CXC chemokines in the inflamed lung. We next asked if alveolar macrophages might be a target cell of CXCL4. We found that co‐incubation of isolated alveolar macrophages with CXCL4 dose‐dependently increased CXCL2 secretion, suggesting that CXCL4 is a potent inducer of CXCL2 secretion in alveolar macrophages. In fact, this is the first time that CXCL4 has been demonstrated to promote macrophage secretion of CXCL2 although this observation is in line with findings showing that CXCL4 stimulates macrophage phagocytosis and oxidative burst (Pervushina et al., 2004). It is important to note that although these findings point to a role of alveolar macrophages in mediating CXCL4‐induced pulmonary accumulation of neutrophils in sepsis, our results do not exclude that CXCL4 also targets other cells, such as endothelial or epithelial cells in the lung.
Figure 7 shows a schematic representation of how platelet‐derived CXCL4 causes neutrophil accumulation in septic lung injury. Thus, our findings suggest that Rac1 signalling is enhanced in platelets and regulates platelet secretion of CXCL4 in polymicrobial sepsis. Moreover, our results indicate that CXCL4 controls neutrophil accumulation via secretion of CXCL2 from alveolar macrophages in septic lung injury. Thus, these findings not only delineate complex mechanisms of neutrophil trafficking in sepsis but also suggest that platelet‐derived CXCL4 plays a key role in regulating inflammation and tissue damage in septic lung injury.
R. H., M. R. and E. Z. performed the experiments, analysed data and wrote the manuscript. H. T. supervised the project, designed the experiments and wrote the manuscript.
Authors have no financial conflicts of interest.
This work was supported by grants from the Swedish Medical Research Council (2012‐3685), Crafoord Foundation, Einar and Inga Nilssons Foundation, Harald and Greta Jaenssons Foundation, Greta och Johan Kocks Foundation, Agnes Nilssons Foundation, Franke and Margareta Bergqvists Foundation for promotion of cancer research, Magnus Bergvalls Foundation, Mossfelts Foundation, Nanna Svartz Foundation, Ruth och Richard Julins Foundation, Swedish Medical Society fund, Malmö General Hospital Foundation for combating cancer, MAS fund, Malmö University Hospital and Lund University. R.H. is supported by Hawler Medical University and Kurdistan Regional Government. E.Z.'s position receives funding from the EU Integrated project BetaBat and a collaborative SRC grant, as well as the Swedish Strategic Research EXODIAB. This work was performed using imaging equipment financed by the Knut and Alice Wallenberg Foundation.
Hwaiz R., Rahman M., Zhang E., and Thorlacius H. (2015) Platelet secretion of CXCL4 is Rac1‐dependent and regulates neutrophil infiltration and tissue damage in septic lung damage. British Journal of Pharmacology, 172: 5347–5359. doi: 10.1111/bph.13325.