1. Immunology and Inflammation

Inhibition of ErbB kinase signalling promotes resolution of neutrophilic inflammation

  1. Atiqur Rahman
  2. Katherine M Henry
  3. Kimberly D Herman
  4. Alfred AR Thompson
  5. Hannah M Isles
  6. Claudia Tulotta
  7. David Sammut
  8. Julien JY Rougeot
  9. Nika Khoshaein
  10. Abigail E Reese
  11. Kathryn Higgins
  12. Caroline Tabor
  13. Ian Sabroe
  14. William J Zuercher
  15. Caroline O Savage
  16. Annemarie H Meijer
  17. Moira KB Whyte
  18. David H Dockrell
  19. Stephen A Renshaw
  20. Lynne R Prince  Is a corresponding author
  1. University of Sheffield, United Kingdom
  2. University of Dhaka, Bangladesh
  3. Leiden University, Netherlands
  4. University of North Carolina at Chapel Hill, United States
  5. GlaxoSmithKline Research and Development Ltd, United Kingdom
  6. University of Edinburgh, United Kingdom
https://doi.org/10.7554/eLife.50990
Share this article
Cite this article
  1. Atiqur Rahman
  2. Katherine M Henry
  3. Kimberly D Herman
  4. Alfred AR Thompson
  5. Hannah M Isles
  6. Claudia Tulotta
  7. David Sammut
  8. Julien JY Rougeot
  9. Nika Khoshaein
  10. Abigail E Reese
  11. Kathryn Higgins
  12. Caroline Tabor
  13. Ian Sabroe
  14. William J Zuercher
  15. Caroline O Savage
  16. Annemarie H Meijer
  17. Moira KB Whyte
  18. David H Dockrell
  19. Stephen A Renshaw
  20. Lynne R Prince
(2019)
Inhibition of ErbB kinase signalling promotes resolution of neutrophilic inflammation
eLife 8:e50990.
https://doi.org/10.7554/eLife.50990
  2. Download BibTeX
  3. Download .RIS

Abstract

Neutrophilic inflammation with prolonged neutrophil survival is common to many inflammatory conditions, including chronic obstructive pulmonary disease (COPD). There are few specific therapies that reverse neutrophilic inflammation, but uncovering mechanisms regulating neutrophil survival is likely to identify novel therapeutic targets. Screening of 367 kinase inhibitors in human neutrophils and a zebrafish tail fin injury model identified ErbBs as common targets of compounds that accelerated inflammation resolution. The ErbB inhibitors gefitinib, CP-724714, erbstatin and tyrphostin AG825 significantly accelerated apoptosis of human neutrophils, including neutrophils from people with COPD. Neutrophil apoptosis was also increased in Tyrphostin AG825 treated-zebrafish in vivo. Tyrphostin AG825 decreased peritoneal inflammation in zymosan-treated mice, and increased lung neutrophil apoptosis and macrophage efferocytosis in a murine acute lung injury model. Tyrphostin AG825 and knockdown of egfra and erbb2 by CRISPR/Cas9 reduced inflammation in zebrafish. Our work shows that inhibitors of ErbB kinases have therapeutic potential in neutrophilic inflammatory disease.

https://doi.org/10.7554/eLife.50990.001

eLife digest

Chronic obstructive pulmonary disease (or COPD) is a serious condition that causes the lungs to become inflamed for long periods of time, leading to permanent damage of the airways.

Immune cells known as neutrophils promote inflammation after an injury, or during an infection, to aid the healing process. However, if they are active for too long, they may also cause tissue damage and drive inflammatory diseases including COPD. To limit damage to the body, neutrophils usually have a very short lifespan and die by a regulated process known as apoptosis. Finding ways to stimulate apoptosis in neutrophils may be key to developing better treatments for inflammatory diseases.

Cells contain many enzymes known as kinases that control apoptosis and other cell processes. Drugs that inhibit specific kinases are effective treatments for some types of cancer and other conditions, and new kinase-inhibiting drugs are currently being developed. However, it remains unclear which kinases regulate apoptosis in neutrophils or which kinase-inhibiting drugs may have the potential to treat COPD and other inflammatory diseases.

To address these questions, Rahman et al. tested over 350 kinase-inhibiting drugs to identify ones that promote apoptosis in neutrophils. The experiments showed that human neutrophils treated with drugs that inhibit the ErbB family of kinases died by apoptosis more quickly than untreated neutrophils. Next, Rahman et al. used zebrafish with injured tail fins as models to study inflammation. Zebrafish treated with one of these drugs – known as Tyrphostin AG825 – had lower levels of inflammation and their neutrophils underwent apoptosis more frequently than untreated zebrafish. Since drugs can have off-target effects, Rahman et al. went on to show using gene-editing technology that reducing the activity of two genes that encode ErbB kinases in zebrafish also decreased the levels of inflammation in the fish.

Further experiments used mice that develop inflammation in the lungs similar to COPD in humans. As expected, neutrophils in the lungs of mice treated with Tyrphostin AG825 underwent apoptosis more frequently than those in untreated mice. These dead neutrophils were effectively cleared by other immune cells called macrophages, which also helps limit damage caused by neutrophils.

Together, these findings show that Tyrphostin AG825 and other drugs that inhibit ErbB kinases help to reduce inflammation by promoting the death of neutrophils. Since several of these drugs are already used to treat human cancers, it may be possible in the future to repurpose them for use in people with COPD and other long-term inflammatory diseases. Determining whether this is possible is an aim for future studies.

https://doi.org/10.7554/eLife.50990.002

Introduction

Neutrophilic inflammation is central to chronic inflammatory diseases such as rheumatoid arthritis and chronic obstructive pulmonary disease (COPD), which impose an increasing social and economic burden on our aging population. Treatment of COPD by next-generation combination therapy with inhaled corticosteroids and newer bronchodilators are viewed as maintenance pharmacotherapies but they do not specifically target cellular inflammation. The anti-inflammatory phosphodiesterase-4 inhibitor, roflumilast, targets systemic inflammation associated with COPD and reduces moderate to severe exacerbations in severe disease, possibly via effects on eosinophils (Martinez et al., 2018; Rabe et al., 2018). In these diseases, clearance of neutrophils by apoptosis is dysregulated particularly during exacerbations (Pletz et al., 2004; Sapey et al., 2011), but to date it has not been possible to therapeutically modify this, indeed corticosteroids can supress neutrophil apoptosis and hence perpetuate inflammation (Liles et al., 1995). Recognising the urgent need for new therapies, we interrogated neutrophil inflammation and survival pathways using an unbiased approach focusing on potentially druggable kinases. Neutrophil persistence in tissues, caused by a delay in apoptosis, can result in a destructive cellular phenotype, whereby neutrophils have greater potential to expel histotoxic factors such as proteases and oxidative molecules onto surrounding tissue. This can occur either actively (by degranulation) or passively (by secondary necrosis). In COPD, among other diseases, delayed apoptosis is considered to be a key part of the pathogenesis, occurring either as a result of pro-survival factors that are present in the lung microenvironment or an innate apoptosis defect (Brown et al., 2009; Haslett, 1999; Pletz et al., 2004; Zhang et al., 2012). Despite this mechanistic understanding, there are no effective treatment strategies in clinical use to specifically reverse this cellular mechanism.

Accelerating neutrophil apoptosis has been shown to promote the resolution of inflammation in multiple experimental models (Burgon et al., 2014; Chello et al., 2007; Ren et al., 2008; Rossi et al., 2006). A number of studies highlight the importance of protein kinases in regulating neutrophil apoptosis (Burgon et al., 2014; Rossi et al., 2006; Webb et al., 2000) and therefore reveal potential therapeutically targetable pathways for inflammatory disease. A growing class of clinically-exploited small molecule kinase inhibitors are being intensively developed (Wu et al., 2015), making this a timely investigation. Using parallel unbiased screening approaches in vitro and in vivo, we here identify inhibitors of the ErbB family of receptor tyrosine kinases (RTKs) as potential therapeutic drivers of inflammation resolution. The ErbB family consist of four RTKs with structural homology to the human epidermal growth factor receptor (EGFR/ErbB1/Her-1). In an in vivo zebrafish model of inflammation, we show that inhibition of ErbBs, pharmacologically and genetically, reduced the number of neutrophils at the site of injury. Furthermore, ErbB inhibitors reduced inflammation in a murine peritonitis model and promoted neutrophil apoptosis and clearance by macrophages in the mouse lung. This study reveals an opportunity for the use of ErbB inhibitors as a treatment for chronic neutrophilic inflammatory disease.

Results

Identifying kinases regulating the resolution of neutrophilic inflammation in vivo

Using a well-characterised transgenic zebrafish inflammation model (Henry et al., 2013; Renshaw et al., 2006), we adopted a chemical genetics approach, which has great potential for accelerated drug discovery (Jones and Bunnage, 2017). We initiated inflammation by controlled tissue injury of the zebrafish tail fin and screened a library of kinase inhibitors in order to establish which kinases could be exploited to enhance inflammation resolution in vivo (Figure 1—figure supplement 1A). We quantified the ability of a library of 367 publicly available kinase inhibitors (PKIS) (Elkins et al., 2016) to reduce neutrophil number at the site of injury during the resolution phase of inflammation. The screen identified 16 hit compounds which reduced neutrophil number at the site of injury in the zebrafish model (Figure 1A). For each compound the degree of kinase inhibition had been established (Elkins et al., 2016) (Figure 1A). A number of kinases were inhibited by the 16 compounds, with Abelson murine leukaemia viral homolog 1 (ABL1), Platelet-derived growth factor receptor (PDGFR) α, PDGFRβ, p38α and ErbB4 being the top five most frequently targeted kinases overall. In addition to frequency of target, we also interrogated selectivity of compound. The most selective compounds, that is those that strongly inhibited individual kinases or kinase families, targeted the kinases YES, ABL1, p38 and the ErbB family. Apoptosis is an important mechanism contributing to inflammation resolution; we therefore sought to identify kinases common to both inflammation resolution and neutrophil apoptosis pathways.

Figure 1 with 2 supplements see all
Download asset Open asset
A protein kinase inhibitor compound library screen identifies compounds that promote the resolution of inflammation in vivo and neutrophil apoptosis in vitro.

(A) mpx:GFP zebrafish larvae (three dpf) that had undergone tail fin transection resulting in an inflammatory response at six hpi were incubated with individual PKIS compounds [25 µM] three larvae/well for a further 6 hr. Wells were imaged and manually scored between 0–3 on the basis of GFP at the injury site in the larvae. ‘Hit’ compounds scored ≥1.5 (n = 2, three larvae per compound per experiment). Publicly available kinase profiling information was generated previously by Elkins et al. (2016) and kinase inhibition of each compound [1 µM] is shown as a gradient of blue to yellow. Hit compounds were ranked horizontally (left to right) from the most to least selective. Kinases (listed on the right) were vertically ranked from top to bottom from the most to least commonly targeted by inhibitors in PKIS. (B) PKIS compounds were incubated with primary human neutrophils for 6 hr. The entire library, at [62 µM], was screened on five separate days using five individual donors. Apoptosis was assessed by Annexin-V/TO-PRO-3 staining by flow cytometry and the percentage apoptosis calculated as Annexin-V single plus Annexin-V/TO-PRO-3 dual positive events. Data are expressed as fold change over DMSO control and each circle represents a single compound. Sixty two compounds accelerated apoptosis ≥2 fold as identified by red dotted line (n = 1). Grey dotted line represents level of apoptosis in DMSO control (i.e. no change). (C) Of the 62 compounds identified above, 38 of the most specific inhibitors were incubated with neutrophils at [10 µM] for 6 hr and apoptosis measured as above. Controls included media, DMSO, GMCSF [50 u/mL] and pyocyanin [50 µM]. Eleven compounds (white bars) accelerated apoptosis ≥2 fold over DMSO control (as identified by dotted line). Kinases targeted by the 11 compounds are shown in the inset table. Hatched bars represent data points in which ErbB inhibitors were used. Data are expressed as percentage apoptosis ± SEM, n = 3 neutrophil donors.

https://doi.org/10.7554/eLife.50990.003

Identifying kinases regulating neutrophil apoptosis in vitro

Circulating neutrophils have a short half-life in vivo (Summers et al., 2010) and undergo spontaneous apoptosis in the absence of growth factors in vitro. We re-screened PKIS library compounds in a human neutrophil apoptosis assay for their ability to accelerate apoptosis (Figure 1—figure supplement 1B). PKIS compounds were screened at 62 µM in order to maximise the chance of identifying ‘hits’ and resulted in 62 compounds that accelerated neutrophil apoptosis ≥2 fold compared to DMSO control (Figure 1B and Supplementary file 1). Secondary screening of top 38 compounds (chosen from the 62 hits based on greatest selectivity for kinase targets) was carried out at 10 µM in order to reduce false positives. This yielded 11 compounds that accelerated neutrophil apoptosis ≥2 fold over control (as indicated by dashed green line, Figure 1C). Representative flow cytometry dot plots illustrating Annexin-V and ToPro-3 profiles for these hit compounds are shown in Figure 1—figure supplement 2. Kinases targeted by these compounds included DYRK1B, KIT, EGFR, ErbB2 and ErbB4, PDGFR, CDK6 and p38 (Figure 1C, inset). The identification of known regulators of neutrophil survival (p38, PI3K) was encouraging support for the screen design and execution. We found that members of the ErbB family of RTKs were the next most frequently inhibited kinase family, being targeted by three highly selective compounds out of the 11 hits (Figure 1C, inset). Since inhibitors of the ErbB family were common hits in both zebrafish and human screens, we hypothesised that targeting ErbBs may be a potential strategy to reduce inflammation.

ErbB inhibitors accelerate neutrophil apoptosis

To address a role for ErbB antagonists in regulating neutrophil apoptosis we tested a range of clinical and non-clinical ErbB-targeting compounds. We show that among inhibitors of ErbBs that are in clinical use, the EGFR inhibitor, gefitinib, is the most effective in promoting neutrophil apoptosis, reaching significance at 50 µM (Figure 2A). The ErbB2-selective inhibitor, CP-724714 (Jani et al., 2007) also promoted neutrophil apoptosis in a dose-dependent manner (Figure 2B) as did Erbstatin and tyrphostin AG825, selective for EGFR and ErbB2 respectively (Osherov et al., 1993; Umezawa and Imoto, 1991) (Figure 2C–D). Since caspase-dependent apoptosis is an anti-inflammatory and pro-resolution form of cell death, engagement of the apoptosis programme was verified biochemically by measuring phosphatidylserine (PS) exposure by Annexin-V staining (Figure 2—figure supplement 1A–C). Furthermore, the pan-caspase inhibitor Q-VD-OPh (Wardle et al., 2011) completely abrogated Erbstatin and tyrphostin AG825-driven neutrophil apoptosis, confirming the caspase dependence of inhibitor mediated cell death (Figure 2—figure supplement 1D–E).

Figure 2 with 1 supplement see all
Download asset Open asset
Inhibition of EGFR and ErbB2 drives apoptosis of neutrophils isolated from COPD patients and healthy subjects.

Neutrophils were incubated with media or a concentration range of gefitinib (A), lapatinib (A), sapatinib (A), CP-724714 (B), erbstatin (Erb, C) or tyrphostin AG825 (Tyr, D) for 6 hr. Stars represent significant difference compared to DMSO control (indicated by ‘0’ in B-D). Neutrophils from COPD patients (open bars) and age-matched healthy control subjects (black bars) were incubated with DMSO or a concentration range of erbstatin (E) or tyrphostin AG825 (F) for 6 hr. Apoptosis was assessed by light microscopy. The data are expressed as mean percentage apoptosis ± SEM from 3 (B, D), 4 (A,C), 10 (E,F COPD), or 7 (E,F HC) independent experiments using different neutrophil donors. Statistical significances between control and inhibitor was calculated by one-way ANOVA with Dunnett’s post-test, indicated as *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

https://doi.org/10.7554/eLife.50990.006

COPD is a chronic inflammatory disease associated with functionally defective circulating neutrophils, including a resistance to undergoing apoptosis during exacerbations (Pletz et al., 2004; Sapey et al., 2011). To show ErbB inhibition is effective in driving apoptosis in subjects with systemic inflammation, we isolated neutrophils from the blood of patients with COPD and age-matched healthy control subjects. Erbstatin and tyrphostin AG825 significantly increased apoptosis of neutrophils from both COPD patients and healthy control subjects in a dose dependent manner at both 6 hr (Figure 2E–F) and 20 hr (data not shown).

ErbB inhibition overcomes neutrophil survival stimuli. Neutrophils are exposed to multiple pro-survival stimuli at sites of inflammation, which could undermine the therapeutic potential of anti-inflammatory drugs. Factors that raise intracellular cAMP concentration ([cAMP]i) are present during inflammation, and elevated [cAMP]i is known to prolong neutrophil survival via activation of cAMP-dependent protein kinases (Krakstad et al., 2004; Vaughan et al., 2007). We show that neutrophil apoptosis was reduced by the cAMP analogue and site selective activator of PKA, N6-monobutyryl-cAMP (N6-MB-cAMP), and that this was reversed by Erbstatin analog (Figure 3A) and Tyrphostin AG825 (Figure 3B). Similar effects were observed in neutrophils from patients with COPD (Figure 3C). GMCSF is a key neutrophil chemoattractant and pro-survival factor, and is closely associated with the severity of inflammation in disease (Klein et al., 2000; Wicks and Roberts, 2016). We show that erbstatin and tyrphostin AG825 prevent GMCSF-mediated survival in COPD and age-matched healthy control neutrophils (Figure 3D–E). GMCSF is known to promote neutrophil survival via the phosphatidylinositol 3-kinase (PI3K)/AKT pathway, ultimately leading to the stabilisation of the anti-apoptotic Bcl-2 family member, Mcl-1 (Derouet et al., 2004; Klein et al., 2000). To investigate potential mechanisms underpinning the ability of tyrphostin AG825 to prevent GMCSF-mediated survival, we assessed AKT-phosphorylation as a measure of PI3K activation and found that tyrphostin AG825 reduced GMCSF-induced AKT phosphorylation after 15 and 30 min of treatment becoming statistically significant at 15 mins (Figure 3F). Tyrphostin AG825 accelerated the spontaneous downregulation of Mcl-1 and also prevented GMCSF-induced stabilisation of Mcl-1 (Figure 3G). These data show ErbB inhibition engages neutrophil apoptosis even in the presence of inflammatory stimuli and therefore has the potential to drive apoptosis at inflammatory sites.

Figure 3
Download asset Open asset
Erbstatin and tyrphostin AG825 overcome pro-survival effects of N6-MB-cAMP and GMCSF.

Neutrophils were incubated with DMSO, Erbstatin [Erb, 40 µM] (A) or tyrphostin AG825 [Tyr, 50 µM] (B) in the presence of DMSO or N6-MB-cAMP [500 µM and 1 mM] for 20 hr. Neutrophils isolated from COPD patients were incubated with DMSO or tyrphostin AG825 [50 µM] in the presence of DMSO or N6-MB-cAMP [500 µM] for 20 hr (C). Neutrophils isolated from COPD patients and age-matched healthy control subjects (HC) were incubated with DMSO, erbstatin (D) [20, 40 µM] or tyrphostin AG825 (E) [25, 50 µM] in the presence or absence of GMCSF [50 u/mL] for 20 hr. Apoptosis was assessed by light microscopy. The data are expressed as mean percentage apoptosis ± SEM from 4 to 6 independent experiments. (F) Neutrophils were incubated with DMSO or tyrphostin AG825 [Tyr, 50 µM] for 60 min before the addition of GMCSF [50 u/mL] for 15 or 30 mins. (G) Neutrophils were incubated with DMSO, tyrphostin AG825 [50 µM] for 60 min before the addition of GMCSF [50 u/mL] for a further 7 hr. Cells were lysed, subjected to SDS-PAGE electrophoresis and membranes probed for p-AKT, Mcl-1 or loading controls, AKT and P38. Images are representative of 3 independent experiments. Charts show densitometric values of 3 individual immunoblots and are expressed as a ratio of target (p-AKT or Mcl-1) over loading control (AKT or P38, respectively). Statistically significant differences were calculated by one-way ANOVA with Sidak post-test (A–C, F–G) or two-way ANOVA with Sidak post-test (D–E) and indicated as *p<0.05, **p<0.01, ***p<0.001.

https://doi.org/10.7554/eLife.50990.008

Kinase microarray profiling reveals ErbB2 is phosphorylated by neutrophil survival stimuli

To explore whether ErbB family members are phosphorylated in response to survival stimuli we studied the activated kinome in human neutrophils stimulated with N6-MB-cAMP (Vaughan et al., 2007). A Kinex antibody microarray was performed to detect the phosphorylation of over 400 kinases and kinase-associated proteins and this data set was interrogated to seek evidence of activation of ErbB by N6-MB-cAMP. Of the phospho-specific antibodies, 17 yielded an increase over baseline control of ≥1.5 at 30 min and 8 at 60 min (Table 1). Among these targets, ErbB2 phosphorylation was detected at 30 min (1.94 > control) and 60 min (1.53 > control, Table 1). This suggests that ErbB is part of the neutrophil signalling response to survival stimuli. In support of this, we detected the presence of ErbB2 mRNA in human neutrophils by RT-PCR (Figure 4A) and a 60kD protein (lower molecular weight ErbB family products are well-documented (Jackson et al., 2013; Guillaudeau et al., 2012; Siegel, 1999; Ward et al., 2013) which was upregulated by GMCSF and dbcAMP (Figure 4B). ErbB3 was also detected in human neutrophils by ELISA (Figure 4C), at levels similar to those observed in other tissues in literature (Buta et al., 2016). We found ErbB3 expression was not regulated by growth factors, which may in part be due to regulation being primarily at the post-translational level.

Figure 4
Download asset Open asset
ErbB2 and ErbB3 expression and regulation in human neutrophils.

(A) ErbB2 was detected in neutrophils and the positive control cell line, BEAS-2B, by RT-PCR. Primer sequences are as follows: ErbB2 forward: ACCCAGCTCTTTGAGGACAA, reverse: ATCGTGTCCTGGTAGCAGAG and β-actin forward: ATATCGCCGCGCTCGTCGTC, reverse: TAGCCGCGCTCGGTGAGGAT. NTC – no template control. (B) Neutrophils were treated with GMCSF [50 u/mL] and dbcAMP [10 μM] for 5 hr and lysates subjected to SDS PAGE. Membranes were immunoblotted with antibodies to ErbB2 antibody or β-actin as a loading control. A 60kD band was detected which was upregulated by GMCSF and dbcAMP. The image is representative of three independent experiments. (C) ErbB3 was detected by ELISA in human neutrophils and the positive control cell line, HaCaT. Neutrophils were treated with media, dbcAMP [500 µM], GM-CSF [50 u/mL] or LPS [1 µg/mL] for 2 hr or 6 hr, after which lysates were collected and ELISA detecting total human ErbB3 was carried out (N = 4). Bars indicate mean + SEM and statistical differences between media control and treatments were measured by one-way ANOVA and Sidak post-test (C, ns).

https://doi.org/10.7554/eLife.50990.009
Table 1
Kinexus antibody microarray analysis.

Ultrapurified neutrophils were incubated with N6-MB-cAMP [100 µM] for 30 and 60 min or lysed immediately following isolation (0’). Lysates from four donors were pooled prior to Kinex antibody microarray analysis. Table shows all targets for which phospho-antibodies had Z ratios of >1.5 compared to t = 0 baseline control, at each timepoint. ErbB related antibodies are in bold.

https://doi.org/10.7554/eLife.50990.010
Target proteinZ-ratio
(30’ v 0’)		Target proteinZ-ratio
(60’ v 0’)
PDK15.69PDK14.79
ZAP70/Syk4.85PKCa/b22.73
p38a3.21Zap70/Syk2.72
PLCg13.16p38a2.32
MAP2K12.70S6Ka2.05
FKHRL12.58Rb1.96
GSK3a/b2.54PKCg1.79
Huntingtin2.29ErbB21.53
BLNK2.25
Jun1.99
Rb1.99
ErbB21.94
Btk1.92
Bad1.81
AMPKa1/21.70
Synapsin 11.69
PKBa1.64

ErbB inhibitors and genetic knockdown increase apoptosis and reduce neutrophil number at the site of inflammation in vivo

To determine the ability of ErbB inhibition to exert an effect on neutrophil number and apoptosis in vivo, we used three complementary animal models of acute inflammation. To specifically address whether tyrphostin AG825 was able to accelerate apoptosis of neutrophils in the mammalian lung, we used a murine model of LPS-induced airway inflammation (Thompson et al., 2014). C57BL/6 mice nebulised with LPS developed an acute pulmonary neutrophilia after 48 hr, to a degree seen previously (Figure 5A–B) (Thompson et al., 2014). Tyrphostin AG825 had no effect on percentage of, or absolute number of neutrophils or macrophages compared to DMSO control (Figure 5A–B). Tyrphostin AG825 significantly increased the percentage of neutrophil apoptosis, both visualised as ‘free’ apoptotic cells (closed circles) and as a summation of both free apoptotic cells and apoptotic inclusions within macrophages in order to capture those that had been efferocytosed (closed triangles, Figure 5C). Macrophage efferocytosis was also significantly elevated by tyrphostin AG825, compared to vehicle control (Figure 5D), determined by counting the number of macrophages containing apoptotic inclusions as a proportion of total macrophages (Figure 5E). We next tested the anti-inflammatory potential of tyrphostin AG825 when administered once inflammation was established, which is more representative of the clinical scenario. Mice were i.p injected with zymosan to induce peritonitis and after 4 hr were treated (i.p.) with tyrphostin AG825 or vehicle control. Total cell counts in peritoneal lavage were 2.2 × 106 in PBS vs 1.7 × 107 in zymosan treated animals at 4 hr demonstrating established inflammation at this time point (Navarro-Xavier et al., 2010). Importantly, tyrphostin AG825 does not induce leukopenia (Figure 5F), however significantly fewer inflammatory cells were found in peritoneal lavage following tyrphostin AG825 treatment (Figure 5G). The neutrophil chemoattractant and proinflammatory cytokine, KC, was reduced in tyrphostin AG825 treated mice, and concomitant with this, a trend for less IL-6 was also observed (Figure 5H). IgM, which correlates with the number and activation of peritoneal B lymphocytes (Almeida et al., 2001), is significantly reduced in tyrphostin AG825-treated mice (Figure 5I).

Figure 5
Download asset Open asset
Tyrphostin AG825 increases neutrophil apoptosis and reduces inflammation in murine models of inflammation.

C57BL/6 mice were nebulized with LPS and immediately injected intraperitoneally with either 10% DMSO (control, n = 8) or 20 mg/Kg tyrphostin AG825 (Tyr, n = 8). After 48 hr the mice were sacrificed and subjected to bronchoalveolar lavage. Percentage neutrophils (A, closed icons) and macrophages (A, open icons) and absolute numbers of neutrophils (B, closed icons) and macrophages (B, open icons) in BAL were calculated by haemocytometer and light microscopy. (C) Percentage neutrophil apoptosis (circles) and percentage neutrophil apoptosis calculated by also including numbers of apoptotic inclusions visualised within macrophages (triangles) was assessed by light microscopy. (D) Macrophages containing one or more apoptotic inclusions expressed as a percentage of all macrophages. Light microscopy image showing apoptotic inclusions within macrophages as indicated by black arrows (E). C57BL/6 mice were injected i.p. with 1 mg zymosan and 4 hr later injected i.p. with 20 mg/Kg tyrphostin AG825 (Tyr, n = 5) or 10% DMSO (Control, n = 5). At 20 hr mice were sacrificed and subjected to peritoneal lavage. (F) WBC, neutrophils and macrophages in blood were measured by a Sysmex cell counter. Total cells in peritoneal lavage were counted by flow cytometry (G) and KC, IL-6 (H) and IgM (I) measured in lavage by ELISA. At least two independent experimental replicates each processing 1–3 mice/group were performed. Statistical significance was calculated by Mann–Whitney U test (A–D and G–I) or one-way ANOVA with Sidak post-test (F) and indicated as *p<0.05, **p<0.01, ***p<0.001.

https://doi.org/10.7554/eLife.50990.011

To further extend this observation, we tested the ability of ErbB inhibitors to modulate neutrophilic inflammation resolution as a whole, in a model which encompasses multiple mechanisms of neutrophil removal including both apoptosis and reverse migration. In the mpx:GFP zebrafish tail fin injury model (Renshaw et al., 2006) (Figure 6A) we were able to show that tyrphostin AG825 (Figure 6B) and CP-724714 (Figure 6C) significantly reduced the number of neutrophils at the site of injury at 4 and 8 hpi. Simultaneous gene knockdown of egfra and erbb2 via CRISPR/Cas9 (referred to as ‘crispants’) also recapitulated this phenotype (Figure 6D). Tyrphostin AG825 did not affect total neutrophil number (Figure 6E), but egfra and erbb2 crispants had significantly fewer neutrophils (Figure 6F). As demonstrated by TSA and TUNEL double staining (Figure 6G), tyrphostin AG825 upregulated neutrophil apoptosis at both the site of injury (Figure 6H) and in the caudal hematopoietic tissue (CHT) of zebrafish (Figure 6I). CHT neutrophil counts were unchanged between conditions (data not shown). egfra and erbb2 crispants had increased numbers of apoptotic neutrophils at the site of injury, but this was not significant (Figure 6J), perhaps suggesting the presence of compensatory mechanisms. These findings show that inhibiting ErbB RTKs accelerate neutrophil apoptosis in vitro and in vivo and enhance inflammation resolution, making ErbB inhibitors an attractive therapeutic strategy for inflammatory disease.

Figure 6
Download asset Open asset
Pharmacological inhibition and genetic knockdown of egfra and erbb2 by CRISPR/Cas9 reduces neutrophil number at the site of injury in a zebrafish model of inflammation.

Tail fin transection was performed as indicated by the red line (A, upper image). Zebrafish larvae (mpx:GFP) were pre-treated at two dpf with DMSO, tyrphostin AG825 [Tyr, 10 µM] (B, minimum n = 28 larvae per condition), or CP-724714 [10 µM] (C, minimum n = 42 larvae per condition) for 16 hr followed by injury. egfra and erbb2 crispants were generated and injured at two dpf (D, minimum n = 36 larvae per condition). The number of neutrophils at the site of injury was determined at 4 and 8 hpi by counting GFP-positive neutrophils. To enumerate neutrophils across the whole body, uninjured inhibitor treated larvae (three dpf) (E, minimum n = 23 larvae per condition) or crispants (two dpf) (F, minimum n = 28 larvae per condition) were imaged by fluorescent microscopy (A, lower image). Apoptosis was measured at the site of injury after 8 hr by TSA and TUNEL double staining (G) (white arrow indicates TUNEL positive neutrophil, scale bar 10 μM) of mpx:GFP tyrphostin AG825 [Tyr, 10 µM] or CP-724714 [10 µM] treated larvae at three dpf (H, minimum n = 35 larvae per condition). Uninjured inhibitor treated larvae were assessed for neutrophil apoptosis in the CHT at three dpf (I, minimum n = 27 larvae per group). Apoptosis at the tail fin injury site of egfra erbb2 crispants at two dpf was also measured at eight hpi (J, minimum n = 26 larvae per group). All data collated from at least three independent experiments, displayed as mean ± SEM. Each icon shows one data point from one individual larvae. Statistically significant differences were calculated by two-way ANOVA with Sidak post-test (B–D) or one-way ANOVA with Dunnett’s post-test(E), Students’ t-test (F), Kruskal-Wallis test with Dunn’s post-test (H–I) or Mann-Whitney U test (J), and indicated as *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

https://doi.org/10.7554/eLife.50990.012

Discussion

Neutrophils are powerful immune cells because of their destructive anti-microbial contents. A deleterious by-product of this is their remarkable histotoxic potential to host tissue, ordinarily held in check by the onset of apoptosis. The inappropriate suppression of neutrophil apoptosis underpins a number of chronic inflammatory diseases, and we are yet to have available an effective treatment strategy that can reverse this cellular defect in clinical practice. Here we show in human, mouse and zebrafish models of inflammation and neutrophil cell death that targeting the ErbB family of RTKs regulates neutrophil survival and resolves inflammation.

Promoting neutrophil apoptosis is a desirable approach for the resolution of inflammation, since apoptosis functionally downregulates the cell, promotes rapid cell clearance by efferocytosis and engages an anti-inflammatory phenotype in phagocytosing cells (Savill et al., 1989; Whyte et al., 1999). As proof of principle, driving apoptosis experimentally promotes the resolution of inflammation across multiple disease models (Chello et al., 2007; Ren et al., 2008; Rossi et al., 2006). Several compounds targeting the ErbB family have been approved as medicines for the treatment of cancer (Singh et al., 2016). Our findings open up the possibility of repurposing well-tolerated ErbB inhibitors for patients with inflammatory disease, potentially addressing a currently unmet clinical need.

The ErbB family are critical regulators of cell proliferation and are associated with the development of many human malignancies (Roskoski, 2014). In addition to the development of cancer, ErbB members have known roles in inflammatory diseases of the airway, skin and gut (Davies et al., 1999; Finigan et al., 2011; Frey and Brent Polk, 2014; Hamilton et al., 2003; Pastore et al., 2008). In the context of lung inflammation, ErbB2 is upregulated in whole lung lysates in murine bleomycin models of lung injury and EGFR ligands are increased in BAL from acute lung injury patients receiving mechanical ventilation (Finigan et al., 2011), suggesting ErbB signalling axes may play a role in the process of airway inflammation in vivo. We show, in murine models where Tyrphostin AG825 was administered either at the time of inflammatory stimulus or once inflammation was established, an impact on cell number, proinflammatory cytokine production and neutrophil apoptosis, further validating the use of ErbB inhibitors to reduce inflammation. The benefit of EGFR inhibitors in reducing inflammation in ventilator-induced and OVA/LPS-induced lung injury rodent models is shown by others, further supporting the targetting of this pathway in inflammatory disease settings (Bierman et al., 2008; Shimizu et al., 2018; Takezawa et al., 2016).

Others have reported that neutrophils express members of the ErbB family (Lewkowicz et al., 2005), particularly ErbB2 at low levels (Petryszak et al., 2016) and we show that they are phosphorylated and regulated following exposure to inflammatory stimuli. ErbBs have known roles in suppressing apoptosis of epithelial cells and keratinocytes, but this study is the first to show a role for ErbBs in survival signalling of myeloid cells. Little is known about the roles of ErbBs in neutrophil function. Erbstatin has been shown to inhibit neutrophil ROS production (Dreiem et al., 2003; Mócsai et al., 1997; Reistad et al., 2005) and chemotactic responses (Yasui et al., 1994). Other kinase families have been found to play a role in neutrophil survival and neutrophilic inflammation, most notably the cyclin-dependent kinases (CDKs) (Rossi et al., 2006). In accordance with this, compounds targeting CDKs were identified as drivers of neutrophil apoptosis in both our primary and secondary screens. Moreover, p38 MAPK inhibitor compounds were also identified in both zebrafish and human screens, and since this kinase is known to mediate survival signals, these findings give confidence to the robustness of the screen design and execution.

The engagement of apoptosis by the ErbB inhibitors erbstatin and tyrphostin AG825 was confirmed both biochemically by phosphatidylserine exposure, and mechanistically by the caspase inhibitor Q-VD-OPh and loss of Mcl-1. This suggests that inhibiting ErbBs as a therapeutic strategy may achieve an overall anti-inflammatory effect in in vivo systems, facilitating clearance by macrophages. In support of this, we provide evidence of increased efferocytosis in vivo following tyrphostin AG825 treatment, with no evidence of secondary neutrophil necrosis due to overwhelming macrophage clearance capacity, evidenced both morphologically and by TO-PRO-3 staining.

The ability of ErbB inhibitors to promote neutrophil apoptosis even in the presence of multiple pro-survival stimuli emphasises the potential of ErbB inhibitors in the lung, at sites where inflammatory mediators are in abundance and where neutrophils are exposed to microorganisms. This is supported by the ability of tyrphostin AG825 to prevent early pro-survival signalling in response to GMCSF, including the phosphorylation of AKT. This precedes the onset of apoptosis, occurring at a time point (15 min) where apoptosis is typically less than 1%. Others have shown the ability of erbstatin to prevent GMCSF-mediated activation of PI3K in human neutrophils, although the impact on cell survival was not studied (al-Shami et al., 1997). Therefore, ErbBs may function as an early and upstream component of the survival pathway in neutrophils. Subsequent impact on Mcl-1 destabilisation by tyrphostin AG825 at 8 hr suggests a cellular mechanism by which these pro-apoptotic effects are mediated.

The effects of ErbB inhibitors in driving spontaneous apoptosis suggest that, under certain circumstances, ErbB activity might be required for constitutive neutrophil survival. It is not clear what, if anything, engages ErbB signalling in culture. The rapid phosphorylation of ErbB2 following N6-MB-cAMP treatment (30 min) suggests that perhaps a ligand is not required, or that the neutrophils can rapidly release ErbB agonists in an autocrine manner. Unlike all other ErbBs, ErbB2 monomers exist in a constitutively active conformation and can form homodimers that do not require a ligand for activation (Fan et al., 2008). ErbBs achieve great signalling diversity: in part because of the individual biochemical properties of ligands and multiple homo-heterodimer combinations, and in part because they activate multiple components including those known to be critical in neutrophil cell survival such as PI3K, MAPK and GSK-3, as well as phosphorylating the Bcl-2 protein Bad which inhibits its death-promoting activity (Yarden and Sliwkowski, 2001).

A limitation of our study is the genetically intractability of human neutrophils, meaning we cannot exclude the possibility that the inhibitors are having off target effects in this system. Mammalian models of ErbB deletion are limited by profound abnormalities in utero and during development (Britsch et al., 1998; Dackor et al., 2007; Gassmann et al., 1995; Miettinen et al., 1995; Riethmacher et al., 1997). For this reason, CRISPR/Cas9 was used to knockdown egfra and erbb2 in zebrafish, which confirmed a role for ErbBs in resolving inflammation. Targeting ErbBs genetically and pharmacologically reduces the number of neutrophils at the site of injury in zebrafish, which may reflect inhibition of a number of pathways that regulate neutrophil number in the tissue, including migration pathways (Ellett et al., 2015). However, the increase in apoptotic neutrophil count at the site of injury with ErbB inhibitor treatment suggests ErbBs may be inducing anti-apoptotic signalling pathways within this inflammatory environment, which could at least in part be causing the phenotype. The reduced neutrophil count at the injury site may also be due to the increase in apoptotic neutrophils in the CHT, which may be preventing neutrophil migration to sites of injury. The unchanged whole body neutrophil number is potentially due to compensatory upregulation of neutrophil production within the CHT. Genetic deletion, but not pharmacological inhibition, of egfra and erbb2 significantly reduced whole body neutrophil number, which may reflect crispants being without egfra and erbb2 genes from a one-cell stage. Reduced neutrophils at the injury site of crispants may be explained by their reduced whole body neutrophil number, but potentially also defects in the migratory response of these neutrophils to a site of inflammation. Murine models of inflammatory disease, where tyrphostin AG825 was administered either at the time of inflammatory stimulus or once inflammation was established, show an impact on cell number, proinflammatory cytokine production and neutrophil apoptosis, further validating the use of ErbB inhibitors to reduce inflammation.

In conclusion, we have identified a previously undefined role for ErbB RTKs in neutrophil survival pathways and a potential new use for ErbB inhibitors in accelerating inflammation resolution. These findings suggest the ErbB family of kinases may be novel targets for treatments of chronic inflammatory disease, and the potential for repurposing ErbB inhibitors currently in use for cancer may have significant clinical potential in a broader range of indications.

Materials and methods

Experimental design

Request a detailed protocol

Our objectives for this study are to identify compounds that are able to resolve neutrophilic inflammation. To do this we performed unbiased chemical screens in both human neutrophils in vitro and zebrafish models of inflammation in vivo. Results were validated in murine models of peritoneal and airway inflammation and zebrafish tail injury models. Genetic evidence was obtained by CRISPR/Cas9 genetic editing in zebrafish.

Isolation and culture of human neutrophils

View detailed protocol

Neutrophils were isolated from peripheral blood of healthy subjects and COPD patients by dextran sedimentation and discontinuous plasma-Percoll gradient centrifugation, as previously described (Haslett et al., 1985; Ward et al., 1999b) in compliance with the guidelines of the South Sheffield Research Ethics Committee (for young healthy subjects; reference number: STH13927) and the National Research Ethics Service (NRES) Committee Yorkshire and the Humber (for COPD and age-matched healthy subjects; reference number: 10/H1016/25). Informed consent was obtained after the nature and possible consequences of the study were explained. Mean age in years was 61.7 ± 2.3 (n = 10) and 66.0 ± 3.6 (n = 7) for COPD and age-matched healthy subjects respectively. Ultrapure neutrophils, for Kinexus antibody array experiments, were obtained by immunomagnetic negative selection as previously described (Sabroe et al., 2002). Neutrophils were cultured (2.5 × 106/ml) in RPMI 1640 (Gibco, Invitrogen Ltd) supplemented with 10% FCS 1% penicillin-streptomycin, in the presence or absence of the following reagents: GMCSF (PeproTech, Inc), N6-MB-cAMP (Biolog), anti-ErbB3 blocking antibody, Tyrphostin AG825 (both Sigma-Aldrich), CP-724714 (AdooQ Bioscience), Erbstatin analog (Cayman Chemicals), Pyocyanin (Usher et al., 2002) or compounds from PKIS (Published Kinase Inhibitor Set 1, GlaxoSmithKline) at concentrations as indicated.

In vitro screening of PKIS in neutrophil apoptosis assays. PKIS consists of 367 small molecule protein kinase inhibitors and is profiled with respect to target specificity (Elkins et al., 2016). In primary screen experiments, neutrophils (from five independent donors over 5 days) were incubated with each compound at 62 µM for 6 hr. Apoptosis was measured by flow cytometry (Attune, Invitrogen). Secondary screening was performed with selected compounds that accelerated neutrophil apoptosis greater than twofold in the primary screen. Compounds were incubated with neutrophils at 10 µM for 6 hr and apoptosis assessed by Attune flow cytometry.

Human neutrophil apoptosis assays

Request a detailed protocol

Neutrophil apoptosis was assessed by light microscopy and by flow cytometry. Briefly, for the assessment of apoptosis by light microscopy based on well-characterised morphological changes, neutrophils were cytocentrifuged, fixed with methanol, stained with Reastain Quick-Diff (Gentaur), and then apoptotic and non-apoptotic neutrophils were counted with an inverted, oil immersion microscope (Nikon Eclipse TE300, Japan) at 100X magnification (Savill et al., 1989). To assess apoptosis by flow cytometry, neutrophils were stained with PE conjugated Annexin-V (BD Pharminogen) and TO-PRO-3 (Thermofisher Scientific) (Savill et al., 1989; Vermes et al., 1995; Ward et al., 1999a) and sample acquisition was performed by an Attune flow cytometer (Life Technologies) and data analysed by FlowJo (FlowJo LLC).

Kinexus antibody array

Request a detailed protocol

Neutrophils were incubated with N6-MB-cAMP [100 µM] for 30 and 60 min or lysed immediately following isolation (t0). Cells were lysed in PBS containing Triton-X, 1 µM PMSF and protease inhibitor cocktail and following 2 min on ice were centrifuged at 10,000 RPM to remove insoluble material. Lysates (containing protein at 6 mg/mL) from four donors were pooled prior to Kinex antibody microarray analysis (Kinexus Bioinformatics) (Zhang and Pelech, 2012). Lysates are subjected to 812 antibodies including phospho-site specific antibodies to specifically measure phosphorylation of the target protein. Fluorescent signals from the array were corrected to background and log2 transformed and a Z score calculated by subtracting the overall average intensity of all spots within a sample, from the raw intensity for each spot, and dividing it by the standard deviations (SD) of all the measured intensities within each sample (Cheadle et al., 2003). Z ratio values are further calculated by taking the difference between the averages of the Z scores and dividing by the SD of all differences of the comparison (e.g, 30 min treated samples versus 0 min control). A Z ratio of ±1.5 is considered to be a significant change from control.

Western blotting

Request a detailed protocol

Whole cell lysates were prepared by resuspending human neutrophils (5 × 106) in 50 µl hypotonic lysis buffer (1 mM PMSF, 50 mM NaF, 10 mM Sodium orthovanadate, protease inhibitors cocktail in water), and by boiling with 50 µl 2X SDS buffer (0.1M 1,4-Dithio-DL-threitol, 4% SDS, 20% Glycerol, 0.0625M Tris-HCl pH6.8% and 0.004% Bromophenol blue). Protein samples were separated by SDS-polyacrylamide gel electrophoresis, and electrotransfer onto PVDF (polyvenylidene difluoride) membranes was performed by semi-dry blotting method. Membranes were then blocked with 5% skimmed milk in TBS-tween and probed against antibodies to p-AKT, AKT (both Cell signalling Technology), Mcl-1 (Santa Cruz Biotechnology), ErbB2 (New England Biolabs), p38 or β-actin (loading controls, StressMarq Biosciences Inc or Sigma respectively), followed by HRP-conjugated secondary antibodies and detection with chemiluminescent substrate solution ECL2 (GE Healthcare).

Fish husbandry

Request a detailed protocol

The neutrophil-specific, GFP-expressing transgenic zebrafish line, Tg(mpx:GFP)i114, (referred to as mpx:GFP) (Renshaw et al., 2006) was raised and maintained according to standard protocols (Nüsslein-Volhard and Dahm, 2002) in UK Home Office approved aquaria in the Bateson Centre at the University of Sheffield, according to institutional guidelines. Adult fish are maintained in 14 hr light and 10 hr dark cycle at 28°C.

Zebrafish tail injury model of inflammation

Request a detailed protocol

PKIS screening: Tail fin transection was performed on mpx:GFP zebrafish larvae at 3 days post-fertilisation (dpf) (Elks et al., 2011; Renshaw et al., 2006). At 6 hr post-injury (hpi), larvae that had mounted a good inflammatory response, as defined by recruitment of >15 neutrophils to the injury site, were arrayed at a density of 3 larvae per well and incubated with PKIS compounds at a final concentration of 25 µM or vehicle control for a further 6 hr. At 12 hpi, the plate was scanned using prototype PhenoSight equipment (Ash Biotech). Images were scored manually as described previously (Robertson et al., 2014). In brief, each well of three larvae was assigned a score between 0–3, corresponding to the number of larvae within the well with a reduced number of neutrophils at the site of injury. Kinase inhibitors which reduced green fluorescence at the injury site to an extent that their mean score was ≥1.5 were regarded as hit compounds.

ErbB inhibition studies: Briefly, two dpf mpx:GFP larvae were treated with Tyrphostin AG825 [10 µM] for 16 hr before undergoing tailfin transection (Elks et al., 2011; Renshaw et al., 2006). The number of neutrophils at the site of injury was determined at 4 and 8 hpi by counting GFP-positive neutrophils by fluorescent microscopy. To enumerate neutrophils across the whole body, uninjured larvae were treated with Tyrphostin AG825 [10 µM] for 24 hr and then mounted in 0.8% low-melting point agarose (Sigma-Aldrich) followed by imaging by fluorescence microscopy (Nikon Eclipse TE2000-U) at 4X magnification, followed by manual counting.

Zebrafish apoptosis assays

Request a detailed protocol

Larvae from each experimental group were pooled into 1.5 mL eppendorf tubes. TSA signal amplification of GFP-labelled neutrophils (driven by endogenous peroxidase activity) was carried out using TSA Plus Fluorescein System (Perkin Elmer). Larvae were fixed overnight in 4% paraformaldehyde at 4°C after which they were subjected to proteinase K digestion. Larvae were post-fixed in 4% paraformaldehyde, before subsequent TUNEL staining for apoptosis using ApopTag Red In Situ Apoptosis Detection Kit (Millipore). Larvae were then mounted in low-melting point agarose and images acquired and analysed using UltraVIEWVoX spinning disc confocal laser imaging system with Volocity 6.3 software (Perkin Elmer). Apoptotic neutrophil count was determined firstly by identifying cells with co-localisation of the TSA and TUNEL stains, then confirmed by accounting for apoptotic neutrophil morphology.

Generation of transient CRISPR/Cas9 zebrafish mutants

Request a detailed protocol

Transient dual knockdown of egfra and erbb2 was induced using a Cas9 nuclease (New England Biolabs) in combination with transactivating RNA (tracr) and synthetic guide RNAs specific to zebrafish egfra and erbb2 genes (Merck). The non-targetting control in these experiments was a guide RNA targetted towards tyrosinase, a gene involved in pigment formation and therefore easy to identify when mutated, and which is used by others in the field as a CRISPR/Cas9 control (O’Connor et al., 2019; Varshney et al., 2016). We have previously shown that this guide does not influence neutrophilic inflammation in the zebrafish (Evans et al., 2019; Isles et al., 2019). Guide RNAs were designed using the online tool CHOPCHOP (https://chopchop.cbu.uib.no/) with the following sequences: efgra: TGAATCTCGGAGCGCGCAGGAGG; erbb2: AACGCTTTGGACCTACACGTGGG; tyrosinase: GGACUGGAGGACUUCUGGGG. Each guide RNA was resuspended to 20 μM in nuclease-free water with 10 mM Tris-HCl (pH8). Guide RNA [20 μM], tracr [20 μM] and Cas9 protein [20 μM] were combined (in a 1:1:1 ratio). 0.5 μL phenol red was added to each injection solution for visualisation. A graticule was used to calibrate glass capillary needles to dispense 0.5 nL of injection solution, and 1 nL was injected into the yolk sac of single-cell stage mpx:GFP embryos. Tail injury assays were carried out at two dpf as described above.

Genotyping of crispant larvae

Request a detailed protocol

High-resolution melt curve analysis was used to determine the rate of egfra and erbb2 mutation in larvae at two dpf. Genomic DNA was collected from individual larvae in both the control and experimental groups, by adding 90 μL 50 mM NaOH to each larvae in a 96-well qPCR plate and incubating at 95°C for 20 min. 10 μL Tris-HCl (pH 8) was then added as a buffer. Master mixes containing either egfra or erbb2 primers (Integrated DNA Technologies) (sequences in table below) were made up, with each well containing: 0.5 μL 10 μM forward primer, 0.5 μL 10 μM reverse primer, 5 μL 2X DyNAmo Flash SYBR Green (Thermo Scientific), 3 μL milliQ water. One μL genomic DNA was added to each master mix in a 96-well qPCR plate. Melt curve analysis was performed and analysed with Bio-Rad Precision Melt Analysis software. Mutation rate was calculated by determining the percentage of egfra erbb2 larvae that showed a different melt-curve profile to the genomic DNA collected from tyrosinase fish (based on 95% confidence intervals). The average mutation rate in our experiments was 97.5% and 87.9% for egfra and erbb2, respectively.

Primer sequences used for high-resolution melt curve analysis.

GeneForward primer sequenceReverse primer sequenceProduct size
egfraCCAGCGGTTCGGTTTATTCAGCGTCTTCGCGTATTCTTGAGG100
erbb2ACAAAGAGCCCAAAAACAGGTTTATCCTTCAGTGCATACCCAGA93

Murine model of LPS induced acute lung inflammation

Request a detailed protocol

All work involving animals was performed in accordance with the Animal (Scientific procedures) Act 1986. Protocols were produced in line with PREPARE guidelines and FRAME recommendations and were reviewed by the University of Sheffield’s Animal Welfare Committee. C57BL/6 mice (female, 9–10 weeks old) were nebulised with LPS (3 mg per group of 8 mice) (Pseudomonas aeruginosa, Sigma-Aldrich) and immediately injected intraperitoneally (i.p.) with either Tyrphostin AG825 (Tocris Bioscience) at 20 mg/Kg in 10% DMSO v/v in vegetable oil (eight mice, treatment group) or an equivalent volume of 10% DMSO v/v in vegetable oil (eight mice, control group) (Kedrin et al., 2009; Roos et al., 2014). After 48 hr the mice were sacrificed by terminal anaesthesia by i.p. pentobarbitone and subjected to bronchoalveolar lavage (BAL, 4 × 1 mL of saline). BAL samples were microcentrifuged and the cellular fraction counted by a hemocytometer and cytocentrifuged. Neutrophil apoptosis and macrophage efferocytosis of apoptotic neutrophils was quantified by oil immersion light microscopy (Nikon Eclipse TE300, Japan).

Murine model of zymosan-induced peritonitis

Request a detailed protocol

C57BL/6 mice were i.p. injected with 1 mg zymosan (Sigma-Aldrich) and 4 hr later injected i.p with 20 mg/Kg Tyrphostin AG825 in 10% DMSO v/v in vegetable oil (five mice, treatment group) or an equivalent volume of 10% DMSO v/v in vegetable oil (five mice, control group). At 20 hr the mice were subjected to terminal gaseous anaesthesia (isoflurane) followed by a cardiac puncture and peritoneal lavage (4 × 1 mL of saline). WBC, neutrophils and macrophages were enumerated in blood by an automated haematology analyser (KX-21N, Sysmex, Milton Keynes, UK). Lavage samples were microcentrifuged and the cellular fraction subjected to flow cytometry and cytocentrifuged for light microscopy. IL-6, KC (Duoset ELISA kits, R and D systems) and IgM (Thermofisher Scientific) in cell free lavage were measured by ELISA as per manufacturer’s instructions.

Statistical analysis

Request a detailed protocol

Data were analysed using GraphPad Prism 8 (GraphPad Software, San Diego, CA) using one-way or two-way ANOVA (with appropriate post-test detailed in the Figure legends) for all in vitro data and appropriate in vivo experiments. Non-parametric tests (Mann-Whitney U-test or Kruskal-Wallis test) were used for selected in vivo experiments with non-Gaussian distribution. Data are expressed as mean ± SEM (standard error of mean), and significance was accepted at p<0.05.

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files.

References

    1. Davies DE
    2. Polosa R
    3. Puddicombe SM
    4. Richter A
    5. Holgate ST
    (1999)
    The epidermal growth factor receptor and its ligand family: their potential role in repair and remodelling in asthma
    Allergy 54:771–783.
    1. Haslett C
    2. Guthrie LA
    3. Kopaniak MM
    4. Johnston RB
    5. Henson PM
    (1985)
    Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacterial lipopolysaccharide
    The American Journal of Pathology 119:101–110.
  40. Book
    1. Nüsslein-Volhard C
    2. Dahm R
    (2002)
    Zebrafish - a Practical Approach
    Oxford University Press.
    1. Osherov N
    2. Gazit A
    3. Gilon C
    4. Levitzki A
    (1993)
    Selective inhibition of the epidermal growth factor and HER2/neu receptors by tyrphostins
    The Journal of Biological Chemistry 268:11134–11142.
    1. Yasui K
    2. Yamazaki M
    3. Miyabayashi M
    4. Tsuno T
    5. Komiyama A
    (1994)
    Signal transduction pathway in human polymorphonuclear leukocytes for chemotaxis induced by a chemotactic factor. Distinct from the pathway for superoxide anion production
    Journal of Immunology 152:5922–5929.

Decision letter

  1. Tadatsugu Taniguchi
    Senior Editor; Institute of Industrial Science, The University of Tokyo, Japan
  2. Jody Rosenblatt
    Reviewing Editor; King's College London, United Kingdom
  3. Lee-Ann Allen
    Reviewer; University of Iowa Health Care, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The first decision letter after peer review is shown below.]

Thank you for submitting your work entitled "Defining the functional neutrophil kinome reveals ErbB kinases as potential therapeutic targets in inflammatory disease" for consideration by eLife. Your article has been reviewed by a Senior Editor, a Reviewing Editor, and three reviewers. The following individuals involved in review of your submission have agreed to reveal their identity: Jean-Pierre Levraud.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

Within our discussion, the reviewers all felt that while the story on the ErbB2 role in inflammation was interesting, more work was needed to really conclude its specific role, especially as this is not the first study to identify a role for ErbBs in neutrophil function and inflammation. Moreover, there was a consensus that the manuscript, as it stands now, really consists of two incomplete and disconnected stories: the kinome and the inhibitor screen. They were also more puzzled that ErB2, which was the main focus of the paper, did not even come up in the kinome. This brings up questions about both approaches. Overall, the impression was that this paper would be better understood as two separate stories but that both stories needed more experiments to be convincing, as outlined in the specific comments, below. One of the reviewers points out that there is a conditional (floxed) Erb2 mutant in mouse (Ozcelik et al., 2002). If the authors use this line and a neutrophil-specific Cre line, they could add genetic evidence. They also agreed that kinase assays needed to be performed to show that the effects were specific as well as live imaging of their zebrafish experiments to determine whether the effects were on cell migration or cell death. Because we all felt that it was unlikely for you produce more convincing data in less than 2 months, the usual turnaround for eLife, any resubmission would need to be as a new paper, and we would argue without the kinome data.

Reviewer #1:

The authors report that inhibition of ErbB kinase limits neutrophil inflammation by affecting neutrophil apoptosis. The paper uses several systems: human neutrophils, zebrafish and mice to probe the effect of ErbB kinase inhibition on inflammation. The observations are intriguing and suggest that ErbB inhibition may provide a therapeutic target for inflammatory disease.

Essential revisions:

Genetic approaches to implicate ErbB kinases in these processes, especially in the zebrafish model or cell lines would better support their conclusion that ErbB kinases regulate neutrophil apoptosis and inflammation. Are there genetic mutants or knockdowns in either cell lines or zebrafish?

More information about where ErbB kinases are expressed in human/mouse/zebrafish tissues would be informative. Is this a neutrophil intrinsic effect? The in vitro analysis supports this idea however it is unclear if this could be due to toxic effects of the compounds on neutrophils versus inhibition of a specific pathway.

Cell toxicity remains a concern (in particular Figure 4). Was there specificity of the inhibitors for ErbB kinases in neutrophils? Can a kinase activity assay be performed? What was the expression of ErbB kinases in human neutrophils? Was the activity/expression regulated by pro-survival signals?

Throughout the number of replicates and total N was not always clear. This should be indicated in the figures and figure legends.

Figure 1: The kinome figure is not user friendly. In Figure 1B, it would be useful to know what fold increase or decrease was observed with GMCSF treatment (and more information about the conditions of the treatment). Figure 1 should reference the tables that correspond with this data. An alternative would be to include a table within the figure (with fold changes in expression by GMCSF).

Figure 2: The kinetics of the assay was long duration (24 hour pre-treatment). What was the effect if larvae were treated with the drug for shorter times? What happens at earlier time points?

Figure 4: Is there a change in ErbB kinase expression or activity in normal/COPD donors? A western blot would be useful.

Figures 5 and 6: Do pro-survival signals increase ErbB kinase expression or activity?

Figure 7C: It was not entirely clear what is being measured here?

Figure 8: Was there an effect on neutrophil reverse migration or just apoptosis? The authors have reported the use of photoconversion to analyze neutrophil reverse migration and this would be informative in this context.

Reviewer #2:

This manuscript is from an established group of respected neutrophil biologists and the study combines profiling of the neutrophil kinome with studies of selected kinase inhibitors as a strategy to advance understanding of neutrophil apoptosis regulation and to identify possible targets for therapeutic intervention in COPD and related diseases.

The main strengths of the study are the generation of a kinome profile of human neutrophils and evidence of significant overlap with the kinome of zebrafish.

The main weaknesses are the lack of validation of the kinome data and the results of the inhibitor screen. Thus, the kinome analsysis remains merely a list of enzymes, and the results of the associated inhibitor screen have not been directly validated.

Essential revisions:

1) Rather than characterizing the kinome itself, a library of kinase inhibitors was screened to identify compounds that accelerate neutrophil apoptosis. Based on these results, further experiments focused on inhibitors of the receptor tyrosine kinase ErbB. However, these data are incomplete. Thus, as presented, the manuscript contains two incomplete stories.

2) Regarding the screen, what accounts for the differential effects of inhibitors that target the same kinases? Why does the ErbB1/B4 inhibitor induces more apoptosis than the ErbB1/B/B4 inhibitor (Figure 3)?

3) Most experiments focus on ErbB2 as an inhibitor target of interest, yet the kinome suggests that only ErbB3 is present. What accounts for this discrepancy? Data demonstrating the abundance of all ErbB family proteins in neutrophils should be added.

4) There is no direct measurement of ErbB isoform activity or phosphorylation or analysis of relevant substrates (in the absence or presence of inhibitors). Thus, the effects could be indirect or non-specific. To address this, the activity and abundance of ErbB family members in resting, apoptotic, and cAMP- or GMCSF-stimulated cells is missing. Reliance on inhibitors without demonstration of specificity and efficacy is risky.

Reviewer #3:

Rahman et al. use a drug discovery approach, combining an in vivo screen in zebrafish and an in vitro screen of human cells, to identify kinase inhibitors that would help resolve neutrophil-mediated inflammatory diseases. They focus on ErbB inhibitors and show that they promote neutrophil apoptosis, even in the presence of pro-survival stimuli. Then, they perform experiments in mouse and again in zebrafish to establish a proof-of-concept of the usefulness of this type of drugs in inflammatory diseases.

This combination of models makes for a powerful and exciting approach. In general, experiments are performed carefully, and the statistical analysis is sound. However, I find that the authors tend to overstate the novelty of their findings.

Inhibition of neutrophil function par erbstatin is not exactly something new – it has been described first in a 1990 paper (Naccache et al., 1990). Actually, searching PubMed with keywords "neutrophil" and "erbstatin" returns 38 references, of which not one is cited in the manuscript. Not all these references are relevant, but clearly the sentence at the end of the Introduction ("This study is the first to identifiy a role for ErbBs in neutrophil function and inflammation") is an exaggeration. The relevant previous literature has to be incorporated in Introduction and Discussion section.

In the Introduction, the authors state "there are no treatment strategies in clinical use to reverse this cellular mechanism": maybe not "reverse", but roflumilast is an approved anti-inflammatory drug to prevent exacerbation in COPD. This should be discussed at the very least.

The final zebrafish experiment (Figure 8) is a disappointing, given the easy imaging of neutrophil migration and death in this system. It does not add much to the initial screen. A measurement of frequency of neutrophil death at the site of injury, in particular, would be desirable considering the general emphasis of the manuscript on induction of neutrophil apoptosis.

Table S1A: Please provide a quantitation of the expression level of each kinase in human neutrophils. Please also provide the complete list and quantitation of kinases expressed in zebrafish neutrophils (as S1B, thus moving the common set as S1C). For both human and zebrafish genes, provide unique identifiers (GenBank or Ensembl IDs), not just names, which are often ambiguous.

[Editors’ note: what now follows is the decision letter after the authors submitted for further consideration.]

Thank you for resubmitting your work entitled "Inhibition of ErbB kinase signalling promotes resolution of neutrophilic inflammation" for further consideration at eLife. Your revised article has been favorably evaluated by Tadatsugu Taniguchi (Senior Editor), a Reviewing Editor, and three reviewers.

We feel that the manuscript has greatly been improved and clarified since the last submission. We were fortunate to have two of the original reviewers and a new reviewer, as the previous third was unavailable at this time. We especially appreciate the new reviewer for comments, as they often need to understand the assessments from the previous iteration, which is no longer present. Through the post-review discussions, we felt it unnecessary to respond to the main points of reviewer #3 but they may be helpful in your final edits. We mainly would like you to address the statistical analysis and points on CRISPR controls raised by reviewer #2. The full reviewer comments are below:

Reviewer #1:

This study is a very interesting, well-designed, and important advance to the field of neutrophil biology and apoptosis regulation. The use of 3 complementary models is a key strength. The data are thorough and convincing, and all points noted in prior critiques have been adequately addressed. No further revisions requested.

Kinome data were leveraged to identify and characterize ErbB family kinases as key regulators of neutrophil longevity. These data were extended to shown that ErbB pathway inhibitors can overcome pro-survival signaling to induce apoptosis and thereby reverse pathological neutrophil accumulation in disease states, including COPD.

Reviewer #2:

The authors report that inhibition of ErbB kinase limits neutrophil inflammation by affecting neutrophil apoptosis. The paper uses several systems: human neutrophils, zebrafish and mice to probe the effect of ErbB kinase inhibition on inflammation. The observations are intriguing and suggest that ErbB inhibition may provide a pathway with therapeutic importance for inflammatory disease. The revised manuscript is significantly improved and addressed the concerns raised in the prior review. I recommend the revised manuscript for publication in eLife after these concerns are addressed.

The controls for the crispants are not clearly outlined. How were the crispants validated? How many targets were used? In general, more than one target should be used for each gene.

Statistical analysis should be clarified in the legends and text.

Reviewer #3:

This study by Rahman et al. investigated the role of ErbB kinase in neutrophil apoptosis, using multiple approaches in vitro and in vivo. They observed that: (1) ErbB family were the common targets of compounds that leading to neutrophil apoptosis in both zebrafish inflammation model and human neutrophils; (2) the ErbBs inhibitors promoted the apoptosis of neutrophils from both healthy volunteers and COPD patients; (3) GMCSG and dbcAMP, the neutrophil survival stimuli, promoted ErbB2 and ErbB3 expression in human neutrophils; (4) tyrphostin AG825 (an ErbBs inhibitor) and knockdown of egfra and erbb2 reduced inflammation in vivo. Altogether, the authors conclude that ErbB family participate in neutrophil survival and ErbB inhibitors play positive roles in accelerating inflammation resolution.

Overall, this study appears interesting, and data presented in this manuscript look solid. However, some data are confusing and do not fully support the conclusion of this study.

There are several issues that need to be addressed by the authors. Specific issues are referenced below.

Essential revisions:

1) Subsection “Identifying kinases regulating the resolution of neutrophilic inflammation in vivo”: "We quantified the ability of PKIS to reduce the number of neutrophils at the site of injury during the resolution phase of inflammation." Please provide the specific data. And how to determine the zebrafish inflammation model is in the phase of inflammatory resolution, not acute phase.

2) The authors just showed the statistic results about apoptosis assessed by flow cytometry. It would be better to show the specific flow images in different groups and compare the differences.

3) Detection apoptosis using light microscopy is not an ideal option. Maybe some apoptosis-related proteins should be detected by Western Blot, such as Bcl-2 and Bax.

4) In Figure 1B what does the x-axis represent?

5) The authors demonstrated that tyrphostin AG825-driven neutrophil apoptosis is caspase-dependent. It would be better to detect the caspase-3 expression by Western Blot.

6) The author showed that the expressions of both total and phosphorylated ErbB2 (pErbB2) were elevated. Does the increase of total protein indirectly increase the amount of phosphorylated protein? It would be better to detect the expression of phosphorylated ErbB2 by Western Blot. And the ratio pErbB2/ErbB2 should be shown.

7) The data in Figure 4 are not enough to support the conclusion. The flow cytometry images in different groups should be provided. Other methods for detecting neutrophil apoptosis in lung tissue should probably be provided, such as immunohistochemistry. Did Tyr accelerate the inflammatory resolution in lung tissue in zymosan treated mice? HE staining for lung tissue should be detected.

8) The authors didn't demonstrate the role of ErbB inhibitors in accelerating inflammatory resolution both in mice and zebrafish models. The process of inflammatory resolution is dynamic. It would be better to monitor the tissue injury dynamically.

https://doi.org/10.7554/eLife.50990.016

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

Within our discussion, the reviewers all felt that while the story on the ErbB2 role in inflammation was interesting, more work was needed to really conclude its specific role, especially as this is not the first study to identify a role for ErbBs in neutrophil function and inflammation. Moreover, there was a consensus that the manuscript, as it stands now, really consists of two incomplete and disconnected stories: the kinome and the inhibitor screen. They were also more puzzled that ErB2, which was the main focus of the paper, did not even come up in the kinome. This brings up questions about both approaches. Overall, the impression was that this paper would be better understood as two separate stories but that both stories needed more experiments to be convincing, as outlined in the specific comments, below. One of the reviewers points out that there is a conditional (floxed) Erb2 mutant in mouse (Ozcelik et al., 2002). If the authors use this line and a neutrophil-specific Cre line, they could add genetic evidence. They also agreed that kinase assays needed to be performed to show that the effects were specific as well as live imaging of their zebrafish experiments to determine whether the effects were on cell migration or cell death. Because we all felt that it was unlikely for you produce more convincing data in less than 2 months, the usual turnaround for eLife, any resubmission would need to be as a new paper, and we would argue without the kinome data.

On reflection, we agree that the description of the neutrophil kinome impaired the clarity of the subsequent experiments. We have therefore removed the kinome data (Figure 1). Comments relating to expression of ErbBs have resulted in new data in the form of a cellular ELISA and confirm ErbB3 protein expression by human neutrophils. We agree that adding genetic evidence would further support our conclusion. As a reviewer states, there is a conditional ErbB2 mutant in the mouse. However, we feel that a whole animal knockout may in fact be more informative since the use of ErbB inhibitors in an inflammatory disease setting, which is the long term aim of this work, would target multiple tissues and systems and not a single cell type. Both EGFR and ErbB2 whole organism knockouts in mice are limited by pre- and postnatal lethality. We have therefore addressed this in our zebrafish model of inflammation using CRISPR/Cas9 which recapitulate findings from the inhibitor studies. Furthermore, we show that targeting ErbBs induces neutrophil apoptosis in vivo, thus providing a potential mechanism for inflammation resolution, and this new data has been added to the manuscript as described in more detail below.

Reviewer #1:

The authors report that inhibition of ErbB kinase limits neutrophil inflammation by affecting neutrophil apoptosis. The paper uses several systems: human neutrophils, zebrafish and mice to probe the effect of ErbB kinase inhibition on inflammation. The observations are intriguing and suggest that ErbB inhibition may provide a therapeutic target for inflammatory disease.

We thank the reviewer for their positive comments and are encouraged that they found our study interesting. We are glad that they were persuaded of the potential of ErbB inhibition as a therapeutic target for inflammatory diseases such as COPD.

Essential revisions:

Genetic approaches to implicate ErbB kinases in these processes, especially in the zebrafish model or cell lines would better support their conclusion that ErbB kinases regulate neutrophil apoptosis and inflammation. Are there genetic mutants or knockdowns in either cell lines or zebrafish?

We agree with this reviewer that genetic approaches would provide better evidence for our conclusion that ErbB kinase inhibition increases apoptosis and enhances inflammation resolution. In order to directly address this, we have performed genetic knockdown in our zebrafish model using CRISPR/Cas9. This work forms new panels of novel data. We find that in zebrafish larvae with knockdown of both egfra (EGFR/ERBB1 orthologue) and erbb2 genes, there is a reduction in neutrophil number at the tail fin injury site (new Figure 5D), which recapitulates the phenotype seen in zebrafish larvae treated with ErbB inhibitors (new Figure 5B-C). Whole body neutrophil number was also reduced in these larvae (new Figures 5F and 5J).

More information about where ErbB kinases are expressed in human/mouse/zebrafish tissues would be informative. Is this a neutrophil intrinsic effect? The in vitro analysis supports this idea however it is unclear if this could be due to toxic effects of the compounds on neutrophils versus inhibition of a specific pathway.

EGFR is expressed in primary human neutrophils as confirmed by flow cytometry and CELISA (Lewkowicz et al., 2005) which is now referenced in the Discussion section. Datasets published on the online Expression Atlas (www.ebi.ac.uk) also confirm expression of ErbB family members in human and mouse neutrophils. Furthermore, we show human neutrophil expression of ErbB2 by western blot and RT-PCR (Figure 4) and have generated new data to show ErbB3 expression by ELISA (Figure 4C). Phosphorylated ErbB2 was detected in the Kinexus kinase antibody microarray (Table 1). We agree that the in vitro work suggests the effects of the inhibitors are mediated, at least in part, via intrinsic effects in neutrophils. Regarding toxic effects of inhibitors, we address this below.

Cell toxicity remains a concern (in particular Figure 4). Was there specificity of the inhibitors for ErbB kinases in neutrophils? Can a kinase activity assay be performed? What was the expression of ErbB kinases in human neutrophils? Was the activity/expression regulated by pro-survival signals?

We thank the reviewer for raising this point. We show in preliminary experiments that a 60KDa version of ErbB2 is upregulated by the survival factors GMCSF and dbcAMP (Figure 4). Unfortunately, kinase activity kits for ErbBs are not commercially available, but we have demonstrated that ErbB2 is rapidly phosphorylated in the presence of dbcAMP (Table 1). In new experimental work, we also show ErbB3 protein expression in human neutrophils by ELISA (new Figure 4C). Expression of ErbB3 did not change with the addition of survival factors dbcAMP, GMCSF or LPS. This assay however, measured total ErbB3 rather than phosphorylated ErbB3 and may suggest post-translational regulation is more important than protein levels. This would require further study and while very interesting, characterising ErbB expression is not the primary focus of this current manuscript. With respect to specificity, the inhibitors we have used are well-characterised, demonstrate selectivity for individual ErbB family members (see IC50 values below) and have been extensively used by others to study ErbBs. To add to strength to out conclusion, we have tested clinical ErbB inhibitors for their effect on neutrophil apoptosis (new Figure 2A). We agree however, that this does not preclude the possibility that they may have some off-target effects, and investigating this will form part of further studies in our laboratory. We argue that, since multiple ErbB-targeting compounds share a profound anti-inflammatory effect across several models and species, there is a strong likelihood of ErbBs being key to this process. Because human neutrophils are genetically intractable, we are unable to delete ErbB genes in these cells. However, new genetic evidence in zebrafish (as described above) further supports the conclusion that inhibition of ErbB signalling has anti-inflammatory effects.

IC50 values for selected inhibitors used in this study:

Gefitinib inhibits purified EGFR and HER-2 at IC50 values of 0.033 and ≥3.7 μmol/L, respectively (Wakeling et al., 2002)

Tyrphostin AG825 inhibits EGFR and ErbB2 at IC50 values of 19 and 0.35 μmol/L respectively (Osherov et al., 1993)

CP-724,714 inhibits EGFR and ErbB2 at IC50 values of 6.4 and 0.010 μmol/L respectively (Jani et al., 2007)

Erbstatin inhibits purified EGFR at a value of 0.77μmol/L (Umezawa et al., 1992)

Throughout the number of replicates and total N was not always clear. This should be indicated in the figures and figure legends.

We thank the reviewer for this comment and apologise for a lack of clarity. We have addressed this throughout the revised manuscript.

Figure 1: The kinome figure is not user friendly. In Figure 1B, it would be useful to know what fold increase or decrease was observed with GMCSF treatment (and more information about the conditions of the treatment). Figure 1 should reference the tables that correspond with this data. An alternative would be to include a table within the figure (with fold changes in expression by GMCSF).

This section has been removed from our manuscript to aid the overall clarity.

Figure 2: The kinetics of the assay was long duration (24 hour pre-treatment). What was the effect if larvae were treated with the drug for shorter times? What happens at earlier time points?

We apologise for this ambiguity. Figure 2 which detailed the zebrafish screen is now Figure 1A. There is no 24 hour pre-treatment in these experiments. The larvae are treated for a total of 6 hours, beginning at 6 hours after injury until 12 hours after injury. This timing is in order to capture those compounds which enhance inflammation resolution. We have added further detail to Figure 1—figure supplement 1 to clarify the timescale.

Figure 4: Is there a change in ErbB kinase expression or activity in normal/COPD donors? A western blot would be useful.

This is a very interesting point. Since the rates of either baseline or modified apoptosis did not differ between healthy controls and COPD patients, we hypothesised that ErbB expression also would not differ between these groups and therefore did not prioritise this experiment. Unfortunately, the ethical approvals for this study have now expired and we are unable to explore this further.

Figures 5 and 6: Do pro-survival signals increase ErbB kinase expression or activity?

The Kinexus antibody array (Table 1) shows upregulation of phosphorylated ErbB2 by dbcAMP at both 30 and 60 minutes. We also show both dbcAMP and GMCSF upregulate ErbB2 protein (Figure 4). Total ErbB3 expression was unaltered with dbcAMP, GMCSF or LPS (Figure 4). Surprisingly there are no well-optimised commercially available activity kits for ErbBs and since neutrophils are not transfectable we are unable to use plasmid-based reporter systems.

Figure 7C: It was not entirely clear what is being measured here?

We apologise for a lack of clarity. Figure 7C (now Figure 5C) shows the percentage neutrophil apoptosis on BALF cytocentrifuge slides from mice treated either with PBS (control) or tyrphostin (Tyr). The closed circles quantify ‘free’ (extracellular) neutrophils, i.e. not including neutrophils that have been ingested by macrophages. The closed triangles quantify both ‘free’ apoptotic neutrophils as well as those that are inside macrophages. This latter count takes into account apoptotic events (inclusions) that have been efferocytosed. Such apoptotic inclusions are highlighted by arrows in Figure 5E. We felt this was important to capture since ‘free’ apoptotic events in vivoare rare due to rapid efferocytosis. We have now made amendments to the figure legend to clarify this.

Figure 8: Was there an effect on neutrophil reverse migration or just apoptosis? The authors have reported the use of photoconversion to analyze neutrophil reverse migration and this would be informative in this context.

Figure 8 has been updated and is now a much expanded Figure 5. In addition to our original inhibitor data, we have added genetic evidence of a role of ErbB kinases in inflammation. We have also measured apoptosis in both inhibitor treated and egfra and erbb2 ‘crispant’ larvae. Briefly, we find that neutrophil apoptosis was increased both at the tail fin injury site and in the caudal haematopoietic tissue of zebrafish treated overnight with ErbB inhibitors. Since we know from experience (https://dmm.biologists.org/content/dmm/9/6/621.full.pdf) that reverse migration is unlikely to occur in parallel with apoptosis, we did not prioritise photoconversion experiments. Apoptosis was not significantly increased in egfra/erbb2 crispant larvae at the tail fin injury site, however potential explanations for this have been explored in the discussion section.

Reviewer #2:

This manuscript is from an established group of respected neutrophil biologists and the study combines profiling of the neutrophil kinome with studies of selected kinase inhibitors as a strategy to advance understanding of neutrophil apoptosis regulation and to identify possible targets for therapeutic intervention in COPD and related diseases.

The main strengths of the study are the generation of a kinome profile of human neutrophils and evidence of significant overlap with the kinome of zebrafish.

The main weaknesses are the lack of validation of the kinome data and the results of the inhibitor screen. Thus, the kinome analsysis remains merely a list of enzymes, and the results of the associated inhibitor screen have not been directly validated.

We thank the reviewer for their positive comments. The weaknesses that they identified are addressed point by point below.

Essential revisions:

1) Rather than characterizing the kinome itself, a library of kinase inhibitors was screened to identify compounds that accelerate neutrophil apoptosis. Based on these results, further experiments focused on inhibitors of the receptor tyrosine kinase ErbB. However, these data are incomplete. Thus, as presented, the manuscript contains two incomplete stories.

We agree with this reviewer that our original presentation of this work lacked clarity of the message we are trying to convey. In response to these comments, we have substantially rewritten the manuscript, removing the description of the kinome and focusing on the functional aspects by adding genetic evidence for the role of ErbB kinase in inflammation.

2) Regarding the screen, what accounts for the differential effects of inhibitors that target the same kinases? Why does the ErbB1/B4 inhibitor induces more apoptosis than the ErbB1/B/B4 inhibitor (Figure 3)?

These are interesting and insightful questions. Some of the effects are due to efficacy at the dose used, some due to cellular penetration of the drug (which we have not assessed) and other effects may relate to differential efficacy at the different receptors. Small changes in the balance of inhibition of one kinase over another might alter the final physiological effect. In addition, neutrophils are heterogenous cell populations with donor variability, so we would expect some variation in responses to treatment, even with the same compound.

3) Most experiments focus on ErbB2 as an inhibitor target of interest, yet the kinome suggests that only ErbB3 is present. What accounts for this discrepancy? Data demonstrating the abundance of all ErbB family proteins in neutrophils should be added.

We agree that our original manuscript did overly focus on ErbB2. This was primarily because the Kinexus antibody array (Table 1) identified only phosphorylated ErbB2, but indeed our inhibitor and genetic studies show roles for other ErbBs. We have now reduced the emphasis on ErbB2 in order to include reflect the potential of other ErbB family members as important targets. The kinome data used an arbitrary threshold and ErbB2 was present in those data at lower levels – these data have now been removed.

4) There is no direct measurement of ErbB isoform activity or phosphorylation or analysis of relevant substrates (in the absence or presence of inhibitors). Thus, the effects could be indirect or non-specific. To address this, the activity and abundance of ErbB family members in resting, apoptotic, and cAMP- or GMCSF-stimulated cells is missing. Reliance on inhibitors without demonstration of specificity and efficacy is risky.

We agree that we cannot rule out the possibility that some of the effects of the inhibitors may be exerted through other targets. This is difficult to circumvent, given that neutrophils are genetically intractable and therefore preclude the use of plasmid-based reporter systems. We have however, directly analysed the ErbB substrate AKT, and show that phosphorylation of AKT is reduced in the presence of tyrphostin (Figure 3F). Surprisingly and frustratingly there are no well-optimised commercially-available activity kits for ErbBs. With respect to abundance and activity, we show in the Kinexus antibody array (Table 1) that phosphorylated ErbB2 is upregulated by dbcAMP at both 30 and 60 minutes (ErbB2 is itself an ErbB substrate). We also show both dbcAMP and GMCSF upregulate ErbB2 protein (Figure 4). In new experimental work, we found that ErbB3 expression in human neutrophils was not regulated by dbcAMP, GMCSF or LPS (Figure 4, new panel C). This assay however, measured total ErbB3 rather than phosphorylated ErbB3 and may suggest post-translational regulation is more important than protein levels. Proving this would require further study and while very interesting, characterising ErbB expression is not the primary focus of this current manuscript. We argue that, since multiple structurally-unrelated ErbB-targeting compounds share a profound anti-inflammatory effect across several models and species, there is a strong likelihood of ErbBs being key to this process. This is supported by the new genetic knockdown of egfra and erbb2 in zebrafish, which also reduces neutrophilic inflammation.

Reviewer #3:

Rahman et al. use a drug discovery approach, combining an in vivo screen in zebrafish and an in vitro screen of human cells, to identify kinase inhibitors that would help resolve neutrophil-mediated inflammatory diseases. They focus on ErbB inhibitors and show that they promote neutrophil apoptosis, even in the presence of pro-survival stimuli. Then, they perform experiments in mouse and again in zebrafish to establish a proof-of-concept of the usefulness of this type of drugs in inflammatory diseases.

This combination of models makes for a powerful and exciting approach. In general, experiments are performed carefully, and the statistical analysis is sound. However, I find that the authors tend to overstate the novelty of their findings.

We thank the reviewer for their positive comments and for recognising the powerful approaches we have taken and the care with which we have performed this work. We have now addressed the issue of novelty in our revision of the manuscript, please see below.

Inhibition of neutrophil function par erbstatin is not exactly something new – it has been described first in a 1990 paper (Naccache et al., 1990). Actually, searching PubMed with keywords "neutrophil" and "erbstatin" returns 38 references, of which not one is cited in the manuscript. Not all these references are relevant, but clearly the sentence at the end of the Introduction ("This study is the first to identify a role for ErbBs in neutrophil function and inflammation") is an exaggeration. The relevant previous literature has to be incorporated in Introduction and Discussion section.

We thank the reviewer for bringing this to our attention. We also agree that not all of the “neutrophil” and “erbstatin” references are relevant or appropriate, particularly since many of them have not directly measured effects of the inhibitors on neutrophil function. We have however, added studies to show an impact of erbstatin on ROS generation, chemotaxis and cellular signalling (listed below) (Discussion section). We have also added additional citations describing the use of tyrphostin inhibitors (not AG825 specifically) in in vivo models to the discussion. Furthermore, we have amended the final sentence of the Introduction to now state “This study reveals an opportunity for the use of ErbB inhibitors as a treatment for chronic neutrophilic inflammatory disease” to reflect the existing studies using erbstatin.

We have added the following citations:

Bierman et al., 2008

Takezawa et al., 2016

Shimizu et al., 2018

Dreiem et al., 2003

Mocsai et al., 1997

Yasui et al., 1994

al-Shami et al., 1997

In the Introduction, the authors state "there are no treatment strategies in clinical use to reverse this cellular mechanism": maybe not "reverse", but roflumilast is an approved anti-inflammatory drug to prevent exacerbation in COPD. This should be discussed at the very least.

We have now added a sentence and citations to discuss the use of roflumilast in COPD (see below). The sentence follows directly on from points referring specifically to targeting neutrophil cell death as a therapeutic strategy and we feel the statement is correct as written.

We have added the following citations to the Introduction:

Rabe et al., 2018

Martinez et al., 2018

The final zebrafish experiment (Figure 8) is a disappointing, given the easy imaging of neutrophil migration and death in this system. It does not add much to the initial screen. A measurement of frequency of neutrophil death at the site of injury, in particular, would be desirable considering the general emphasis of the manuscript on induction of neutrophil apoptosis.

This is a very important point and we have specifically addressed this. We

performed apoptosis studies in our zebrafish model which show that treatment of zebrafish larvae with ErbB inhibitors results in an increase in neutrophil apoptosis both at a site of injury (Figure 5H) and within the caudal haematopoietic tissue (Figure 5I). This suggests that ErbB inhibitors are able to induce neutrophil apoptosis both within a homeostatic environment, and at a site of inflammation in vivo. Neutrophil apoptosis at the tail fin injury site of zebrafish larvae with genetic knockdown of egfra and erbb2 was also assessed (Figure 5J), however no significant differences were observed, as discussed previously.

Table S1A: Please provide a quantitation of the expression level of each kinase in human neutrophils. Please also provide the complete list and quantitation of kinases expressed in zebrafish neutrophils (as S1B, thus moving the common set as S1C). For both human and zebrafish genes, provide unique identifiers (GenBank or Ensembl IDs), not just names, which are often ambiguous.

As mentioned above, the description of the neutrophil kinome has been removed from our revised manuscript.

[Editors' note: the author responses to the re-review follow.]

We feel that the manuscript has greatly been improved and clarified since the last submission. We were fortunate to have two of the original reviewers and a new reviewer, as the previous third was unavailable at this time. We especially appreciate the new reviewer for comments, as they often need to understand the assessments from the previous iteration, which is no longer present. Through the post-review discussions, we felt it unnecessary to respond to the main points of reviewer #3 but they may be helpful in your final edits. We mainly would like you to address the statistical analysis and points on CRISPR controls raised by reviewer #2.

On behalf of all authors, we are extremely grateful for further evaluation of our manuscript, and for noting the improvements we have made. We also greatly appreciate the continuity of reviewers (as well as reviewer #3 who has taken the time to evaluate our paper), and for giving us the opportunity to respond to these comments with a revised manuscript. We take note that post-review discussions require us to mainly address the statistical analysis and points on CRISPR controls.

The full reviewer comments are below:

Reviewer #1:

This study is a very interesting, well-designed, and important advance to the field of neutrophil biology and apoptosis regulation. The use of 3 complementary models is a key strength. The data are thorough and convincing, and all points noted in prior critiques have been adequately addressed. No further revisions requested.

Kinome data were leveraged to identify and characterize ErbB family kinases as key regulators of neutrophil longevity. These data were extended to shown that ErbB pathway inhibitors can overcome pro-survival signaling to induce apoptosis and thereby reverse pathological neutrophil accumulation in disease states, including COPD.

We thank reviewer #1 for their time and expertise in evaluating our manuscript for a second time. We are very encouraged by these positive comments.

Reviewer #2:

The authors report that inhibition of ErbB kinase limits neutrophil inflammation by affecting neutrophil apoptosis. The paper uses several systems: human neutrophils, zebrafish and mice to probe the effect of ErbB kinase inhibition on inflammation. The observations are intriguing and suggest that ErbB inhibition may provide a pathway with therapeutic importance for inflammatory disease. The revised manuscript is significantly improved and addressed the concerns raised in the prior review. I recommend the revised manuscript for publication in eLife after these concerns are addressed.

We sincerely thank reviewer #2 for this second critique of our manuscript and are encouraged by the positive review. We are delighted to be able to respond to the comments in a revised manuscript.

The controls for the crispants are not clearly outlined. How were the crispants validated? How many targets were used? In general, more than one target should be used for each gene.

We apologise for this ambiguity. We have added the following to subsection “Generation of transient CRISPR/Cas9 zebrafish mutants”: “The non-targeting control in these experiments was a guide RNA targeted towards tyrosinase, a gene involved in pigment formation and therefore easy to identify when mutated, and which is used by others in the field as a CRISPR/Cas9 control (O'Connor et al., 2019; Varshney et al., 2016). We have previously shown that this guide does not influence neutrophilic inflammation in the zebrafish (Evans et al., 2019; Isles et al., 2019).” Gene mutation in crispants was validated by melt curve analysis. We have expanded on this in subsection “Genotyping of crispant larvae” by quoting the average mutation rate. We also determined that these larvae were generally healthy by observing a normal swim bladder at 5 dpf – a sensitive assay for generalised developmental defects. We have used one guide RNA per gene. Since we demonstrated an efficient mutation rate as well as a clear phenotype, we did not feel compelled to use additional guides.

Statistical analysis should be clarified in the legends and text.

Thank you for highlighting this important omission. We have now added full details of the statistical analysis to each figure legend and refer the reader to legends in subsection “Statistical Analysis”.

Reviewer #3:

This study by Rahman et al. investigated the role of ErbB kinase in neutrophil apoptosis, using multiple approaches in vitro and in vivo. They observed that: (1) ErbB family were the common targets of compounds that leading to neutrophil apoptosis in both zebrafish inflammation model and human neutrophils; (2) the ErbBs inhibitors promoted the apoptosis of neutrophils from both healthy volunteers and COPD patients; (3) GMCSG and dbcAMP, the neutrophil survival stimuli, promoted ErbB2 and ErbB3 expression in human neutrophils; (4) tyrphostin AG825 (an ErbBs inhibitor) and knockdown of egfra and erbb2 reduced inflammation in vivo. Altogether, the authors conclude that ErbB family participate in neutrophil survival and ErbB inhibitors play positive roles in accelerating inflammation resolution.

Overall, this study appears interesting, and data presented in this manuscript look solid. However, some data are confusing and do not fully support the conclusion of this study.

There are several issues that need to be addressed by the authors. Specific issues are referenced below.

We are grateful to reviewer #3 for evaluating our manuscript, and for suggesting modifications for improvement. Although post-review discussions concluded that these were not all necessary to carry out, we have made revisions based on these suggestions (detailed below).

Essential revisions:

1) Subsection “Identifying kinases regulating the resolution of neutrophilic inflammation in vivo”: "We quantified the ability of PKIS to reduce the number of neutrophils at the site of injury during the resolution phase of inflammation." Please provide the specific data. And how to determine the zebrafish inflammation model is in the phase of inflammatory resolution, not acute phase.

We have now added “recruitment phase” and “resolution phase” to Figure 1—figure supplement 1 as well as a citation which establishes the timecourse of neutrophilic inflammation in this model.

2) The authors just showed the statistic results about apoptosis assessed by flow cytometry. It would be better to show the specific flow images in different groups and compare the differences.

Thank you for this excellent suggestion. We have now added representative flow cytometry dot plots as a new figure supplement (Figure 1—figure supplement 2).

3) Detection apoptosis using light microscopy is not an ideal option. Maybe some apoptosis-related proteins should be detected by Western Blot, such as Bcl-2 and Bax.

Morphological assessment of neutrophil apoptosis is considered to be the gold standard. We have used a combination of light microscopy (a morphological assay) and flow cytometry (a biochemical assay) to assess neutrophil apoptosis in this paper, as well as showing the cell death was a caspase-dependent mechanism.

4) In Figure 1B what does the x-axis represent?

We apologise for any ambiguity. We have changed Figure 1B x-axis to “Individual protein kinase inhibitor compounds [62.5 μM]”.

5) The authors demonstrated that tyrphostin AG825-driven neutrophil apoptosis is caspase-dependent. It would be better to detect the caspase-3 expression by Western Blot.

As stated above, we feel that showing both Annexin V positivity as well as inhibition of cell death by Q-VD-OPh (Figure 2—figure supplement 1) are robust measurements of apoptosis.

6) The author showed that the expressions of both total and phosphorylated ErbB2 (pErbB2) were elevated. Does the increase of total protein indirectly increase the amount of phosphorylated protein? It would be better to detect the expression of phosphorylated ErbB2 by Western Blot. And the ratio pErbB2/ErbB2 should be shown.

We were unable to find a clean and reliable antibody to pErbB2 that worked well in neutrophil lysates by Western blot.

7) The data in Figure 4 are not enough to support the conclusion. The flow cytometry images in different groups should be provided. Other methods for detecting neutrophil apoptosis in lung tissue should probably be provided, such as immunohistochemistry. Did Tyr accelerate the inflammatory resolution in lung tissue in zymosan treated mice? HE staining for lung tissue should be detected.

We apologise for any confusion caused. The data in Figure 4A-E (now Figure 5) data from light microscopy and this is stated in the legend. Other methods for detecting neutrophil apoptosis in lung tissue should probably be provided, such as immunohistochemistry. Did Tyr accelerate the inflammatory resolution in lung tissue in zymosan treated mice? HE staining for lung tissue should be detected. Thank you for this suggestion. Neutrophil apoptosis was enumerated in BALF by light microscopy, which is considered the gold standard. Unfortunately, lungs were not obtained from the zymosan peritonitis animals.

8) The authors didn't demonstrate the role of ErbB inhibitors in accelerating inflammatory resolution both in mice and zebrafish models. The process of inflammatory resolution is dynamic. It would be better to monitor the tissue injury dynamically.

We thank the reviewer for their comment. Measuring tissue injury dynamically in the mouse is very challenging and beyond the scope of this study. Because we have studied multiple timepoints in the fish we feel this gives information on the dynamics of inflammation resolution.

https://doi.org/10.7554/eLife.50990.017

Article and author information

Author details

  1. Atiqur Rahman

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. Department of Biochemistry and Molecular Biology, Faculty of Biological Sciences, University of Dhaka, Dhaka, Bangladesh
    Contribution
    Investigation, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared

  2. Katherine M Henry

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Conceptualization, Formal analysis, Investigation, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0554-2063

  3. Kimberly D Herman

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Formal analysis, Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  4. Alfred AR Thompson

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0717-4551

  5. Hannah M Isles

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  6. Claudia Tulotta

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Formal analysis, Investigation, Reviewing and editing draft
    Competing interests
    No competing interests declared

  7. David Sammut

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Resources, Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  8. Julien JY Rougeot

    Institute of Biology, Leiden University, Leiden, Netherlands
    Present address
    Institute of Immunology and Immunotherapy, University of Birmingham, Birmingham, United Kingdom
    Contribution
    Funding acquisition, Formal analysis
    Competing interests
    No competing interests declared

  9. Nika Khoshaein

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  10. Abigail E Reese

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  11. Kathryn Higgins

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  12. Caroline Tabor

    The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Investigation, Writing—review and editing
    Competing interests
    No competing interests declared

  13. Ian Sabroe

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Conceptualization, Formal analysis, Funding acquisition, Writing—review and editing
    Competing interests
    No competing interests declared

  14. William J Zuercher

    SGC-UNC, Division of Chemical Biology and Medicinal Chemistry, UNC Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, United States
    Contribution
    Conceptualization, Resources, Funding acquisition, Writing—review and editing
    Competing interests
    No competing interests declared

  15. Caroline O Savage

    Immuno-Inflammation Therapy Area Unit, GlaxoSmithKline Research and Development Ltd, Stevenage, United Kingdom
    Present address
    Institute of Immunology and Immunotherapy, University of Birmingham, Birmingham, United Kingdom
    Contribution
    Writing—review and editing, Was key to the development of ideas that led on to this work
    Competing interests
    is an employee of GlaxoSmithKline Research and Development Ltd. The author declares no other competing interests exist.

  16. Annemarie H Meijer

    Institute of Biology, Leiden University, Leiden, Netherlands
    Contribution
    Conceptualisation, Funding acquisition, Writing—review and editing
    Competing interests
    No competing interests declared

  17. Moira KB Whyte

    MRC Centre for Inflammation Research, University of Edinburgh, Edinburgh, United Kingdom
    Contribution
    Conceptualization, Formal analysis, Writing—review and editing
    Competing interests
    No competing interests declared

  18. David H Dockrell

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. MRC Centre for Inflammation Research, University of Edinburgh, Edinburgh, United Kingdom
    Contribution
    Conceptualization, Formal analysis, Supervision, Writing—review and editing
    Competing interests
    No competing interests declared

  19. Stephen A Renshaw

    1. Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    2. The Bateson Centre, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Writing—original draft, Project administration, Writing—review and editing
    Contributed equally with
    Lynne R Prince
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1790-1641

  20. Lynne R Prince

    Department of Infection, Immunity and Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Writing—original draft, Project administration, Writing—review and editing
    Contributed equally with
    Stephen A Renshaw
    For correspondence
    L.r.prince@sheffield.ac.uk
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6133-9372

Funding

Commonwealth Scholarship Commission

  • Atiqur Rahman

Medical Research Council (MR/M004864/1)

  • Stephen A Renshaw

Medical Research Council (G0700091)

  • Stephen A Renshaw

European Commission (PITG-GA-2011-289209)

  • Julien JY Rougeot
  • Annemarie H Meijer

SGC

  • William J Zuercher

British Heart Foundation (Intermediate Clinician Fellowship FS/18/13/33281)

  • Abigail E Reese

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

General: We thank Lynne Williams, Carl Wright, Jessica Willis, Elizabeth Marsh and Catherine Loynes for help with animal experiments as well as volunteers and patients who donated blood to this study. We thank the Bateson Centre aquaria staff for their assistance with zebrafish husbandry.

Ethics

Human subjects: Peripheral blood of healthy subjects and COPD patients was taken following informed consent and in compliance with the guidelines of the South Sheffield Research Ethics Committee (for young healthy subjects; reference number: STH13927) and the National Research Ethics Service (NRES) Committee Yorkshire and the Humber (for COPD and age-matched healthy subjects; reference number: 10/H1016/25).

Animal experimentation: Zebrafish were raised and maintained according to standard protocols in UK Home Office approved aquaria in the Bateson Centre at the University of Sheffield, according to institutional guidelines. All work involving mice was performed in accordance with the Animal (Scientific procedures) Act 1986 and has been approved by the Animal welfare and ethical review body at University of Sheffield. Work was carried out under procedure project license 40/3726. All animals were checked prior to the start of experiments by competent personal licensees (PIL), and were deemed to be fit and well before the start of experiments.

Senior Editor

  1. Tadatsugu Taniguchi, Institute of Industrial Science, The University of Tokyo, Japan

Reviewing Editor

  1. Jody Rosenblatt, King's College London, United Kingdom

Reviewer

  1. Lee-Ann Allen, University of Iowa Health Care, United States

Version history

  1. Received: August 9, 2019
  2. Accepted: October 15, 2019
  3. Accepted Manuscript published: October 15, 2019 (version 1)
  4. Version of Record published: November 8, 2019 (version 2)

Copyright

© 2019, Rahman et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 3,443
    Page views
  • 496
    Downloads
  • 14
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Atiqur Rahman
  2. Katherine M Henry
  3. Kimberly D Herman
  4. Alfred AR Thompson
  5. Hannah M Isles
  6. Claudia Tulotta
  7. David Sammut
  8. Julien JY Rougeot
  9. Nika Khoshaein
  10. Abigail E Reese
  11. Kathryn Higgins
  12. Caroline Tabor
  13. Ian Sabroe
  14. William J Zuercher
  15. Caroline O Savage
  16. Annemarie H Meijer
  17. Moira KB Whyte
  18. David H Dockrell
  19. Stephen A Renshaw
  20. Lynne R Prince
(2019)
Inhibition of ErbB kinase signalling promotes resolution of neutrophilic inflammation
eLife 8:e50990.
https://doi.org/10.7554/eLife.50990

Categories and tags

Research organisms

Further reading

    1. Immunology and Inflammation

    Selective loss of CD107a TIGIT+ memory HIV-1-specific CD8+ T cells in PLWH over a decade of ART

    Oscar Blanch-Lombarte, Dan Ouchi ... Julia G Prado
    Research Article

    The co-expression of inhibitory receptors (IRs) is a hallmark of CD8+ T-cell exhaustion (Tex) in people living with HIV-1 (PLWH). Understanding alterations of IRs expression in PLWH on long-term antiretroviral treatment (ART) remains elusive but is critical to overcoming CD8+ Tex and designing novel HIV-1 cure immunotherapies. To address this, we combine high-dimensional supervised and unsupervised analysis of IRs concomitant with functional markers across the CD8+ T-cell landscape on 24 PLWH over a decade on ART. We define irreversible alterations of IRs co-expression patterns in CD8+ T cells not mitigated by ART and identify negative associations between the frequency of TIGIT+ and TIGIT+ TIM-3+ and CD4+ T-cell levels. Moreover, changes in total, SEB-activated, and HIV-1-specific CD8+ T cells delineate a complex reshaping of memory and effector-like cellular clusters on ART. Indeed, we identify a selective reduction of HIV-1 specific-CD8+ T-cell memory-like clusters sharing TIGIT expression and low CD107a that can be recovered by mAb TIGIT blockade independently of IFNγ and IL-2. Collectively, these data characterize with unprecedented detail the patterns of IRs expression and functions across the CD8+ T-cell landscape and indicate the potential of TIGIT as a target for Tex precision immunotherapies in PLWH at all ART stages.

    1. Genetics and Genomics
    2. Immunology and Inflammation

    Epigenetic signature of human immune aging in the GESTALT study

    Roshni Roy, Pei-Lun Kuo ... Luigi Ferrucci
    Research Article Updated

    Age-associated DNA methylation in blood cells convey information on health status. However, the mechanisms that drive these changes in circulating cells and their relationships to gene regulation are unknown. We identified age-associated DNA methylation sites in six purified blood-borne immune cell types (naive B, naive CD4+ and CD8+ T cells, granulocytes, monocytes, and NK cells) collected from healthy individuals interspersed over a wide age range. Of the thousands of age-associated sites, only 350 sites were differentially methylated in the same direction in all cell types and validated in an independent longitudinal cohort. Genes close to age-associated hypomethylated sites were enriched for collagen biosynthesis and complement cascade pathways, while genes close to hypermethylated sites mapped to neuronal pathways. In silico analyses showed that in most cell types, the age-associated hypo- and hypermethylated sites were enriched for ARNT (HIF1β) and REST transcription factor (TF) motifs, respectively, which are both master regulators of hypoxia response. To conclude, despite spatial heterogeneity, there is a commonality in the putative regulatory role with respect to TF motifs and histone modifications at and around these sites. These features suggest that DNA methylation changes in healthy aging may be adaptive responses to fluctuations of oxygen availability.

    1. Immunology and Inflammation
    2. Neuroscience

    Molecular consequences of peripheral Influenza A infection on cell populations in the murine hypothalamus

    René Lemcke, Christine Egebjerg ... Birgitte R Kornum
    Research Article

    Infection with Influenza A virus (IAV) causes the well-known symptoms of the flu, including fever, loss of appetite, and excessive sleepiness. These responses, mediated by the brain, will normally disappear once the virus is cleared from the system, but a severe respiratory virus infection may cause long-lasting neurological disturbances. These include encephalitis lethargica and narcolepsy. The mechanisms behind such long lasting changes are unknown. The hypothalamus is a central regulator of the homeostatic response during a viral challenge. To gain insight into the neuronal and non-neuronal molecular changes during an IAV infection, we intranasally infected mice with an H1N1 virus and extracted the brain at different time points. Using single-nucleus RNA sequencing (snRNA-seq) of the hypothalamus, we identify transcriptional effects in all identified cell populations. The snRNA-seq data showed the most pronounced transcriptional response at 3 days past infection, with a strong downregulation of genes across all cell types. General immune processes were mainly impacted in microglia, the brain resident immune cells, where we found increased numbers of cells expressing pro-inflammatory gene networks. In addition, we found that most neuronal cell populations downregulated genes contributing to the energy homeostasis in mitochondria and protein translation in the cytosol, indicating potential reduced cellular and neuronal activity. This might be a preventive mechanism in neuronal cells to avoid intracellular viral replication and attack by phagocytosing cells. The change of microglia gene activity suggest that this is complemented by a shift in microglia activity to provide increased surveillance of their surroundings.