Tamoxifen restores extracellular trap formation in neutrophils from patients of chronic granulomatous disease, in ROS-independent manner

Capsule Summary

Tamoxifen induces autophagy-dependent NET formation in CGD patient neutrophils and improves their antimicrobial activity. Given the genetic defect in their ROS-generation machinery, this study highlights repurposing of tamoxifen to potentiate host defense in CGD patients.

To the Editor:

Chronic Granulomatous Disease (CGD) is a primary immunodeficiency characterized by life-threatening acute and recurrent bacterial and fungal infections (1). Mutations in one of the several subunits of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase enzyme complex, responsible for generation of reactive oxygen species (ROS) constitute the genetic defects attributed to CGD. As ROS generation is a major antimicrobial defense mechanism of neutrophils, defective oxidative killing of pathogens underlies the frequent and severe infectious episodes in CGD patients. Therapeutic measures to correct the antimicrobial capacity of neutrophils could be an attractive strategy for treating this devastating immune disorder.

In addition to oxidative killing, a recently established antimicrobial activity of neutrophils is the formation of extracellular traps (NETs), which are decondensed chromatin fibrils coated with granular proteases and histones (2). NETs can trap and kill extracellular pathogens, without the need for phagocytosis, by placing them in close proximity to antimicrobial components. Because NET formation (NETosis) has largely been reported to be dependent on reactive oxygen species (ROS) generation by the NADPH oxidase complex, and neutrophils from CGD patients are reported to be deficient in NETosis, impairing extracellular pathogen killing (2, 3). Consequently, gene therapy to restore NADPH oxidase function for potentiation of NET formation is thought to be a viable therapy for infectious sequelae in this disease (3). However, this approach is severely limited by reduced survival of gene-corrected cells, and has not advanced beyond experimental salvage therapy. Here we report a new strategy to induce NETs by treating CGD patient neutrophils with an FDA-approved drug tamoxifen (TMX). We show that TMX does not require NADPH oxidase activity to induce NET formation and does so in a ROS-independent manner by activating autophagy.

We first compared NET formation in neutrophils freshly isolated from venous blood of healthy subjects and those from CGD patients. Purity was checked by flow cytometry and microscopy (Suppl. Fig. 1). As shown in Fig. 1A upper panel, in-vitro stimulation with LPS, an inflammatory TLR agonist from Gram-negative bacteria, live Staphylococcus aureus (SA), a common bacterial pathogen affecting CGD patients (1) or Phorbol myristate acetate (PMA, used as a positive control) generated NETs in healthy adult neutrophils within 4hrs of stimulation. These NETs, characterized as extruded DNA fibrils, were visualized using a fluorescent DNA stain Sytox Green. In contrast to healthy subjects, CGD patient neutrophils showed significantly impaired NET formation in response to all three stimuli (Fig. 1A, B &C). These observations confirm that CGD neutrophils exhibit impaired NET formation in response to inflammatory stimuli as well as bacterial infection. We then tested if TMX could restore this function in CGD neutrophils. Indeed, as shown in Fig. 1B, TMX treatment restored the NET formation in neutrophils from CGD patients upon LPS stimulation as well as bacterial infection, to the levels exhibited by healthy adult neutrophils (Fig 1C). Furthermore, TMX-induced NETs were restored in all patients tested, regardless of the antibiotic or steroid treatment that these patients were undergoing at the time of sample collection (Supplementary Table 1). TMX treatment significantly increased the extent of NET formation in healthy adult neutrophils as well. Importantly, CGD and healthy neutrophils exhibited significantly increased NET-mediated bacterial killing upon TMX treatment (Fig. 1D), albeit the effect was more pronounced in CGD. These data strongly supported the therapeutic potential of this drug. TMX-treated CGD neutrophils formed NETs despite an absent NADPH oxidase activity (confirmed by DHR test in these patients, data not shown). This suggested that TMX induces NET formation in a ROS independent manner. To confirm this, we measured cytosolic ROS in healthy human neutrophils stimulated with LPS or PMA with or without TMX treatment. In line with previous reports, PMA induced robust oxidative burst in human neutrophils, while LPS stimulation showed moderate ROS formation (Fig. 2A) (E1). TMX treatment did not induce any ROS formation by itself. Interestingly, this drug significantly reduced the cytosolic ROS induced by PMA or LPS in neutrophils (Fig. 2A). These data show that TMX- mediated NET formation does not require NADPH oxidase activity.

Figure 1.

TMX treatment restores NET-formation in neutrophils from CGD patients. (A) Healthy adults or CGD patient neutrophils were stimulated with LPS, PMA or S. aureus (SA), with or without TMX pretreatment (B) followed by detection of NETs using DNA dye Sytox Green. Bar=200μ. (C) Quantitation of NETting neutrophils upon indicated treatments. Bars are mean ± SE. n=5 for healthy adults; n= 3–8 for CGD patients. NS, no stimulation. Black asterisks indicate significant difference between healthy and CGD; green asterisks shows significant difference in healthy or CGD with or without TMX treatment. (*p < 0.05, **p < 0.01). (D) TMX-induced NETs improve antibacterial function of neutrophils. NETs were induced in neutrophils from healthy adults and CGD patients (n=6 each group) by TMX treatment followed by incubation with S. aureus (SA) in the presence of cytochalasin D with or without DNase I for 30min. Data is from 3 independent experiments (**p < 0.01).

Figure 2.

TMX induces NETs independent of ROS by autophagy activation. (A). Cytosolic ROS measurement in healthy adult neutrophils. n=3. (B). Western blots to show extent of autophagy activation by LC3 II processing in healthy adults or CGD patient neutrophils stimulated with LPS with or without TMX treatment. n=3 for each group. (C). Immunofluorescence analysis of autophagy activation depicted by LC3 II puncta (red) in neutrophils from healthy adults, upon stimulation with LPS or TMX alone or in combination, with or without ATG4 inhibitor NSC185058. Nuclei (Blue) are stained with 4′,6-diamidino-2-phenylindole (DAPI). NET formation was analyzed in all samples by Sytox Green staining (Green). Representative images from n=3 humans are shown. Bar=200μ.

We have recently reported that autophagy activation accelerates NET formation in murine neutrophils (E2) . We thus examined whether TMX treatment induced NET formation via autophagy activation in human CGD neutrophils. Aggregation of nascent autophagy protein LC3, also called ATG8 or LC3-I) into LC3-PE complex (LC3-II) serves as a marker of autophagy activation and autophagosome formation (E3). Western blot analysis of unstimulated CGD and healthy human neutrophils showed quiescent autophagy as indicated by low or no processed LC3 II (Fig. 2B, NS). Treatment with LPS or TMX individually or in combination induced robust LC3 II processing in healthy subject neutrophils, indicating autophagy activation. In contrast, neutrophils from CGD patients showed minimal LC3 II upon LPS treatment suggesting impaired autophagy activation. TMX treatment with or without LPS led to a robust increase in the amount of LC3-II in human CGD neutrophils (Fig. 2B). This shows that, in the absence of ROS, autophagy activation is sufficient to induce NET formation. Because healthy adult and CGD neutrophils behaved in a seemingly similar fashion upon TMX treatment, we next utilized healthy neutrophils to examine whether autophagy is necessary for TMX-induced NET formation. Fig. 2C depicts immunofluorescence imaging of LC3 II puncta formation in neutrophils stimulated with LPS with or without TMX treatment in the presence or absence of NSC185058, a specific inhibitor of cysteine protease ATG4B which is required for LC3 II lipidation and puncta formation (E4). As previously observed, TMX treatment alone induced robust NET formation and also increased LPS-induced NET formation (Sytox Green stained panels). This increase in NETosis correlated with enhanced numbers and size of LC3 II puncta (red stained panels) which localized to the nuclei (blue stained panels) of neutrophils stimulated with LPS and TMX. Importantly, treatment with the autophagy inhibitor NSC185058 almost completely inhibited NET formation induced by LPS or TMX, which directly correlated with absence of LC3 II puncta in these cells. Together, these data strongly suggest that autophagy activation is necessary and sufficient to induced NET formation upon TMX treatment.

In summary, we show here that Tamoxifen, an FDA approved drug in use for breast cancer treatment, rescues NET formation in neutrophils from human CGD patients, completely bypassing the requirement of NADPH oxidase-derived ROS. Our data of enhanced NET-mediated bacterial killing by CGD neutrophils demonstrates a novel therapeutic potential of TMX in this disease. Autophagy activation was found to be the underlying mechanism for TMX mediated NET formation. To the best of our knowledge, ours is the first report connecting TMX activated autophagy to its NET-inducing function in CGD. Autophagy-inducing action of TMX has been reported in cancer and is linked to the modulation of cholesterol biosynthesis by TMX (4). A detailed analysis of upstream autophagy components and their cross-talk, if any, with cholesterol biosynthesis in TMX-induced NET formation in CGD is a fascinating area of research currently being pursued in our lab. Of note, TMX-induced NET formation was inhibited by treatment with ATG4B inhibitor NSC185058 (Fig. 2B) which suppresses autophagosome generation, but not by inhibitors (wortmannin and LY294002) of class III PI3K activity essential for initiation of autophagy (data not shown). This suggests that TMX acts through ATG4 activation.

Defects in autophagy have been linked to systemic hyperinflammation and colitis frequently observed in CGD patients (5). Therefore, treatment of CGD patients with TMX could also reduce inflammatory sequelae, in addition to improving antimicrobial activity by restoring NET formation. Since TMX is already FDA approved for treatment of another disease, our studies make this drug an attractive option for use in CGD patients. The promise of TMX is underscored by ongoing efforts to repurpose this drug for treatment of glioblastoma multiforme, desmoid tumors and bipolar disorders (6–8). It is also important to note that TMX exhibits thromboembolic and uterotrophic side effects of low, but significant increase in endometrial cancer at clinical doses approved for treating breast cancer (9). However these effects are attributed to estrogen-receptor dependent function of TMX, as opposed to its ER-independent effects at higher, albeit well-tolerated, doses (6). Future studies regarding the detailed mechanism of action and controlled trials would provide valuable information regarding efficacy and safety of TMX for its clinical use in CGD patients.