Introduction
Asthma is a heterogeneous chronic respiratory disease, characterized by airway hyperresponsiveness (AHR) and airway inflammation [
1]. There are four main inflammatory phenotypes in asthma based on the proportion of granulocytes in induced sputum: neutrophilic asthma, eosinophilic asthma, paucigranulocytic asthma, and mixed granulocytic asthma [
2]. Neutrophilic asthma is one of the main types of severe asthma [
3], which exhibits worse lung function, more severe airway inflammation, and worse treatment response. Therefore, it is crucial to investigation the pathogenesis of neutrophilic asthma.
Experimental animal models are essential to advance asthma pathophysiological research, and mice are the main subjects. There are classical animal models of eosinophilic asthma, such as ovalbumin (OVA)/ aluminum (Alum)-sensitized and OVA-challenged model of asthma, house dust mite (HDM) extract-induced model of asthma and so on [
4‐
7]. However, to the best of our knowledge, there is no animal model of neutrophilic asthma to be universally accepted. Currently, there are some animal studies of neutrophilic asthma, for example, OVA/complete Freund’s adjuvant (CFA)-sensitized and OVA-challenged model of asthma, OVA/ lipopolysaccharide (LPS)-sensitized and OVA-challenged model of asthma [
8‐
10]. While these models recapitulate some of the features of human neutrophilic asthma, each of these models had certain limitations.
Neutrophils are the main effector cells of inflammation and tissue infection [
11]. The formation of neutrophil extracellular traps (NETs) is a function of neutrophils [
12]. Neutrophils increase when inflammation occurs, and can undergo a process of NETosis, releasing NETs outside the cells [
13]. However, whether NETs are generated and function in neutrophilic asthma have not been completely elucidated.
Our study demonstrated that the OVA/CFA/LPS-induced murine model is suitable for study as a model of neutrophilic asthma. This kind of model has massive neutrophilic inflammation and severe airway hyperresponsiveness. In addition, we also found that a large number of NETs were produced and correlated with the severity of airway inflammation. Therefore, it is reasonable to hypothesize that NETs play an important functional role in neutrophilic asthma.
Materials and methods
Animals
BALB/c mice (female, 6-8 weeks) were purchased from Hangzhou Medical College (Hangzhou, China). We maintained all the mice in specific pathogen-free (SPF) conditions with a 12 light/12 dark cycle in the Experimental Animal Center, the First Affiliated Hospital, School of Medicine, Zhejiang University.
Mouse allergen sensitization and challenge
Mouse models were induced using modified protocols as previously reported [
7‐
9]. Mice were randomly grouped and sensitized on day 1 and day 8. For the OVA/Alum group of asthma, 25 µg OVA (A5503, Sigma Aldrich, St. Louis, MO, USA) dissolved in 100 µl 0.9% saline was mixed 1:1 with Imject Alum (Pierce, Rockford, IL, USA) by intraperitoneal (IP) injection. For the OVA/CFA group of asthma, 25 µg OVA dissolved in 100 µl 0.9% saline was mixed 1:1 with CFA (F5881, Sigma Aldrich, St. Louis, MO, USA) by IP. injection. For the OVA/LPS group of asthma, mice were lightly anesthetized with isoflurane, and intratracheally injected using a combination of 25 µg OVA with 0.1 or 10 µg LPS (L2630, Sigma Aldrich, St. Louis, MO, USA) in a total volume of 40 µl, with 0.9% saline as the diluent. For the OVA/CFA/0.1 LPS group, the model of sensitization is a combination of the OVA/CFA group and OVA/LPS group. After sensitization, the above-mentioned groups were challenged with aerosolized 1% OVA for 30 min on days 15-17. For the OVA/CFA/0.1 LPS + DNase I group or the OVA/CFA/0.1 LPS + CI-amidine group, on the basis of the OVA/CFA/0.1 LPS group, intravenous injection of DNase I (5 mg per kg body weight) or intraperitoneal injection of Cl-amidine (10 mg per kg body weight) was performed 1 h before each challenge. Mice sensitized and challenged with 0.9% saline were used as controls.
Airway hyperresponsiveness measurement
Airway responsiveness was assessed 24 h after the last OVA challenge as previously described [
14]. Briefly, mice were placed in a plethysmograph chamber (EMKA Technologies, Paris, France) for at least 10 min to adaption. Baseline pulmonary parameters were first determined, and then mice were sequentially challenged with aerosolized PBS and methacholine (Mch, A2251, Sigma Aldrich, St. Louis, MO, USA) at increased concentrations (3.125, 6.25, 12.5, 25 and 50 mg/ml in PBS). Enhanced Pause (Penh), an indirect estimate of airway resistance, was used to measure airway resistance to methacholine. After each nebulization, record the Penh value for 3 min and take the average of three consecutive values.
Bronchoalveolar lavage
Mice were euthanized 48 h after the last OVA challenge, serum, bronchoalveolar lavage fluid (BALF), bone marrow and lung tissues were collected for further experiments. As described previously [
14], BALF was collected on whole lungs by infusing 1 ml of PBS via a tracheal cannula. Then, the BALF was centrifuged, and the supernatants were stored at-80 °C until analysis, the cell pellet was resuspended in PBS for differential cell counting or flow cytometry analysis.
Cytokine analysis
Expression of interleukin (IL)-4 (431,104), IL-17 A (432,504), IL-6 (431,304), and IL-1β (432,604) in BALF were quantitated by enzyme-linked immunosorbent assay (ELISA) kit (Biolegend, San Diego, CA, USA) according to the manufacturers’ instructions.
Lung histopathology staining
After the BALF was collected, the right main bronchus was ligated, and the left lung was perfused and fixed with 4% paraformaldehyde for 24 h, followed by paraffin embedded, sectioned, stained with hematoxylin and eosin (H&E) or paraffin acid-Schiff (PAS) staining. Sections were performed in a blinded fashion. The degree of inflammation on H&E-stained lung sections was scored as described previously [
15]: 0, normal; 1, few inflammatory cells; 2, a ring of inflammatory cells 1 cell layer deep; 3, a ring of inflammatory cells 2 to 4 cells deep; 4, a ring of inflammatory cells greater than 4 cells deep. Mucus-containing goblet cells on PAS-stained lung sections was also scored previously described [
16]: 0, no PAS-positive cells; 1, less than 25%; 2, 25 to 50% PAS-positive cells; 3, 50 to 80% PAS-positive cells; 4, greater than 80% PAS-positive cells.
Flow cytometry analysis
Flow cytometry of leukocytes in BALF, lung tissue, peripheral blood and bone marrow. The BALF were collected according to the above procedure. Lung single-cell suspensions were obtained as previously described [
17]. The peripheral blood was collected from mice by removed their eyeball and placed in tubes containing EDTA. Red blood cells (RBCs) were lysed with RBC lysis buffer (Biolegend, 420,301, San Diego, CA, USA). The bone marrow cells were harvested from the femora and tibiae by flushing with PBS, then cells filtered through 40 μm strainers (352,340, BD Biosciences, San Diego, CA, USA) to obtain single cell suspension. After a washing step, all single-cell suspensions were incubated with Fixable Viability Stain 780 (FVS780, 565,388, BD Biosciences, San Diego, CA, USA, 1:1000) according to manufacturer’s instructions to gate out dead cells. Subsequently, samples were stained with antibodies against CD45-FITC (157,214, Biolegend, San Diego, CA, USA, 1:100), CD11b-PE (101,207, Biolegend, San Diego, CA, USA, 1:100) and Ly6G-PE/Cy7 (560,601, BD Biosciences, San Diego, CA, USA, 1:100) for 30 min on ice, then cells were analyzed on a Cytoflex LX flow cytometer (Beckman Coulter, CA, USA). All above antibodies were diluted in PBS. Analyses were performed using FlowJo software v10.6.2. Neutrophils were identified as CD45(+)CD11b(+)Ly6G(+).
Immunostaining and confocal microscopy
Bone marrow cells from mice were harvested as describe above. Subsequently, neutrophils from bone marrow were isolated by density gradients according to the mouse neutrophil isolation kit instructions (TBD2013NM, TBD Science, China). The neutrophils (2 × 10
5 cells per well in serum free RPMI 1640) were plated on poly-lysine-coated round coverslips and placed in 24-well culture plates. Then, the cells were stimulated with 100 nM Phorbol 12-myristate 13-acetate (PMA, P3681, Sigma Aldrich, St. Louis, MO, USA) or RPMI 1640 (Ctrl) for 4 h. Subsequently, cells were fixed in 4% paraformaldehyde (PFA) for 30 min and permeabilized by incubation in 0.5% Triton X-100 at room temperature (RT) for 1 min. After that some samples were stained with propidium iodide (PI, 5 µg/ml, P4170, Sigma Aldrich, St. Louis, MO, USA) for 30 min, and washed in PBS before confocal microscopic observation. To quantify NET formation, NETosis was defined as neutrophils with flattened nuclei, decondensed chromatin and expulsion of extracellular neutrophils under fluorescence microscopy. Two independent researchers analyzed 200 neutrophils in each sample [
18]. The other samples were blocked by blocking solution (phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) and 0.1% Tween-20) for 1 h at RT, samples were stained with rabbit anti-citrullinated histone 3 (CitH3) (ab5103, Abcam, 1:500) and mouse anti- Myeloperoxidase (MPO) (AF3667, R&D, 1:50) overnight at 4 °C. The next day, samples were washed in PBST and incubated with secondary antibodies CoraLite 594-conjugated Goat Anti-Rabbit IgG (H + L) (Proteintech, SA00013-4, 1:400) and CoraLite 488-conjugated Affinipure Goat Anti-Mouse IgG (H + L) (Proteintech, SA00013-1, 1:400). All above antibodies were diluted in blocking solution. The nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI, 422,801, Biolegend 1:10000). The NETs were identified by the colocalization of antibodies (MPO, CitH3 and DAPI) and quantified using Fiji software v2.1.0. To determine the colocalization, we used a ratio by normalizing the percentage CitH3 signal to the percentage MPO signal, analysis of the CitH3/MPO percentage values from four regions was performed [
19].
Western blot
Protein lysates of lung tissue were prepared. Then, the proteins were separated by 10-12% sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE), transferred onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked in 5% skim milk for 1 h RT, follow by incubated with primary antibodies overnight at 4 ℃. The following day, membranes were washed and incubated with HRP-conjugated secondary antibodies for 1 h and proteins were visualized using the ECL reagent. Rabbit anti-CitH3 (ab5103, 1:1000) was purchased from Abcam; Mouse anti-MPO (AF3667, 1:400) was purchased from R&D; beta (β)-Tubulin (66,240–1-Ig, 1:5000) was purchased from Proteintech Technology. All primary antibodies were diluted in primary antibody dilution buffer (Beyotime Biotechnology, China), and secondary antibodies were diluted in 5% skim milk.
Statistical analysis
All statistics and graphs were performed using GraphPad Prism software v8.4.0 (GraphPad Software Inc., San Diego, CA, USA). In this study, a one-way ANOVA with Tukey posttests was used for multiple comparisons. Data are presented as means ± standard errors of measurement (SEM). Significant differences are shown as *P < 0.05, **P < 0.01, and ***P < 0.001.
Discussion
Asthma is a heterogeneous disease with multiple phenotypes and endotypes [
24,
25]. Approximately 5-25% of asthmatics suffer from severe asthma and do not respond to existing medical treatment, those people account for 50-80% of all asthma-related health care costs [
26]. Neutrophilic airway inflammation is one of the main features of severe asthma. As of today, although there are many studies on neutrophilic asthma, the pathogenesis of neutrophilic asthma is poorly understood, and there are still many difficulties in treatment.
OVA is a classic allergen for establishing allergic asthmatic mouse models, which usually requires co-sensitization with appropriate adjuvants [
27]. The type of OVA-induced airway inflammation varies depending on the adjuvant. OVA/CFA or OVA/LPS is often used to develop neutrophilic asthma mouse models by investigators. CFA is known as a powerful inflammatory stimulus and widely used experimental models of arthritis [
28]. In recent years, many studies have successfully established neutrophilic asthma model using CFA combined with OVA [
8,
10,
29]. This model have higher degrees of airway neutrophilic inflammation, however airway responsiveness is not significantly increased [
30]. This could be explained by the fact that despite the association between AHR and airway inflammation, there is evidence that AHR can occur independently of inflammation [
31‐
33]. A similar phenomenon was also found in our study, although Penh in the CFA group increased compared with the control group, it was lower than other neutrophil groups or even the eosinophil group.
LPS is a major component of the outer membrane of gram-negative bacteria, which stimulates the innate immune response to inflammation [
34]. The use of LPS alone cannot establish a mouse model of asthma, and it acts as an adjuvant in the asthma model. Neutrophilic airway inflammation and airway hyperresponsiveness can be induced when LPS is used in combination with OVA [
9,
35]. However, in asthma models, the role of LPS remains controversial. LPS is present at high levels in air and dust, some studies indicated that the extent of LPS exposure negatively correlates with the risk of developing asthma [
36], while others held the opposite opinion that LPS in the environment probably plays an important role in the occurrence of asthma exacerbations and insensitive responses to corticosteroids in humans [
37]. According to Eisenbarth et al. [
38], low dose LPS promote Th2 immunity, whereas high doses promote Th1 responses. This view is also supported by a recent study showing that exposure to low-dose LPS (0.1 µg) in BALB/c mice enhanced type 2 allergic asthma, whereas starting with a higher dose of LPS (10 µg) had no significant effect [
39]. However, there is also study using 50 µg OVA with 0.1 µg LPS sensitization combined with OVA challenge to successfully establish Th17-dependent neutrophilic airway inflammation [
9]. We speculate that different effects of LPS in various studies in asthma are due to differences in time and dose. Following the studies described above, we established the OVA/low-dose (0.1 µg) LPS and OVA/high-dose (10 µg) LPS models. Our work revealed that both the 0.1LPS group and 10LPS group showed significant airway hyperresponsiveness, much higher than the CFA group, while the inflammatory cell infiltration in the 0.1LPS group and 10LPS group was lower than that in the CFA group. Based on the above conclusions, with consideration of high doses of LPS may cause acute lung injury, we combined OVA/CFA/0.1LPS to establish a mouse model and subsequent experimental validations. The results showed that the 0.1LPS + CFA group exhibited marked airway hyperresponsiveness and airway inflammation.
NETs are composed of histones, neutrophil elastase (NE), MPO and double-stranded (ds) DNA [
40,
41]. NETs are important in antibacterial defense, helping to limit systemic infection and maintain host defenses against fungal pathogens [
42]. When the body was stimulated by pathogenic or chemical stimuli, the neutrophils use degranulation, phagocytosis, and the production of NETs to control initial infections [
43]. However, with our knowledge about NETs have been greatly expanded in recent years, accumulating evidence indicated that NETs are considered to be a double-edged sword in lung disease. Infectious and noninfectious pulmonary diseases led to large-scale neutrophil infiltration into the lungs, and activated neutrophils release substantial amounts of NETs [
44‐
46]. However, excessive NETs generation in noninfectious settings could be damaging for the tissue/organ, excessive NETs formation increases mucus viscosity and fills the lungs, impairing lung function [
44,
47]. In the neutrophilic asthma models of our study, the group with more severe airway inflammation also produced more NETs. In the pathological staining of the 0.1LPS + CFA group, we could even see a large amount of mucus occluded the lumen. We speculate that this is related to the oversecretion of NETs. It has been shown that BALF from patients with severe asthma had detectable NETs that were positively correlated with IL-17 levels [
48]. IL-1β is a proinflammatory cytokine central to the inflammatory response driven by the IL-6 signaling pathway [
49]. Proinflammatory cytokines such as IL-1β, IL-6 are up-regulated when neutrophils undergo NETosis [
50]. Also in our work, we observed increased expression of IL17A, IL-1β and IL-6 in the neutrophilic asthma models.
We validated our results by increasing NETs degradation or decreasing NETs generation to determine that the presence of NETs exacerbated disease in neutrophilic asthma model. In our previous experiments, we found that airway mucus secretion and airway responsiveness were higher in the 0.1LPS + CFA group than that in the other groups. After treated with DNase I, the airway hyperreactivity and mucus production of the 0.1LPS + CFA group, however, were almost completely eliminated. It has also previously been shown that airway mucus embolism in asthmatic patients can achieve complete lysis within minutes after administration of recombinant human DNase [
51]. It now appears that recombinant human DNase can relieve airway mucus embolism, possibly by degrading airway NETs. Activation of PAD4 is likely a major driver of NETosis, and histone citrullination catalyzed by PAD4 appears to be a critical step in NETosis [
22,
52]. Similarly, we demonstrate that a PAD4 inhibitor, CI-amidine, can significantly alleviate airway inflammation and airway hyperresponsiveness in neutrophilic asthma model. Both above-mentioned results confirmed NETs is critical to the pathogenesis of neutrophilic asthma model. In addition, we found that DNase I and CI-amidine had no significant impact on AHR of EOS mice (Fig S7). This can be explained as follows: DNase I and CI-amidine play a role in neutrophilic asthma model mainly because they can increase NETs degradation or decrease NETs generation. For NETs that are not highly expressed in the EOS group, DNase I and CI-amidine did not work.
However, the current study did have some shortcomings as well. How NETs is regulated in neutrophilic asthma remains unknown. Meanwhile, we also have recognized our study lacks the support of clinical correlative data. The issues mentioned above will require further study.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.