Background
Stable allergic asthma is characterized by airways inflammation, pulmonary eosinophilia, airways hyperreactivity (AHR) and mucus hypersecretion. In contrast, exacerbations of asthma involve worsening AHR and are associated with pulmonary neutrophilia [
1]. In inflammation, mediators including histamine, leukotrienes, certain prostaglandins and possibly acetylcholine, are produced by multiple cell types, including neutrophils, epithelial cells, airways smooth muscle (ASM) and nerves. By acting on G protein-coupled receptors (GPCRs) that signal through the heterotrimeric G protein, Gq, and which are present on ASM, these mediators may promote bronchoconstriction [
2]. However, agonists at Gq-coupled GPCRs can also enhance the expression of inflammatory cytokines and could therefore contribute towards increased inflammation [
3,
4]. Similarly, many allergens, for example house dust mite (HDM) or cockroach allergen, are major triggers of asthma and contain proteolytic activities that activate a family of GPCRs known as protease-activated receptors (PARs) [
5,
6]. PARs may signal via multiple transducers, which include Gq, as well as Gi, G12/13, and likely β-arrestin, and may contribute to various inflammatory responses [
5‐
7] Thus, HDM extracts produce profound airway inflammation and AHR in mouse models of asthma [
8‐
10]. Indeed, many proteases, including those in HDM or cockroach allergen, elicit inflammatory responses and, in vivo, may act via PAR2, which can couple via Gq and other transducers [
7,
11], to induce inflammation and reduce lung function [
12‐
15]. Similarly, activation of coagulation cascades during inflammation can activate PAR1 to up-regulate inflammatory cytokine expression and PAR2 deficiency produced increased expression of the C-X-C motif chemokine ligand (CXCL), CXCL1, following intratracheal instillation of the Gram-negative bacterial well-wall component, lipopolysaccharide (LPS) [
16‐
18]. Likewise, neutrophil recruitment and activation leads to release of elastase, and other proteases, that can activate both PAR1 and PAR2 to promote cytokine expression and inflammation [
19‐
21]. Indeed, inhibition of neutrophil elastase improved AHR and inflammation in a mouse model of airways inflammation, suggesting that neutrophil-derived proteolytic activity can be a major driver of inflammation [
22].
The regulator of G protein signaling (RGS) family of proteins interact with active GTP-bound, Gα, to promote intrinsic GTPase activity and GTP hydrolysis [
23]. This returns the heterotrimeric G protein to an inactive heterotrimeric (αβγ) GDP-bound state and switches off GPCR signaling. Pertinent to the current study are the R4 sub-family of RGS proteins as these have selectivity for Gαq (and Gαi) [
23,
24]. While many R4 family members, including RGS2-5 are expressed at the mRNA level in human ASM [
25], loss of
Rgs2,
Rgs4 and
Rgs5 in the mouse produce AHR and/or enhanced ASM contractility [
25‐
30]. Of relevance to therapeutics used in asthma is the finding that RGS2 mRNA is induced in vivo in human airways following budesonide inhalation [
31]. Furthermore, RGS2 mRNA and protein are increased by β
2-adrenoceptor agonists in a manner that is synergistically enhanced by glucocorticoids in human ASM [
26]. Similar effects also occur in primary bronchial epithelial cells, where RGS2 is more glucocorticoid-inducible [
32].
While RGS2 is bronchoprotective and its expression may be enhanced by commonly used asthma therapeutics, expression of RGS2 in non-contractile cells suggests additional roles in the airways. Indeed, in airway epithelial cells, signaling and cytokine expression induced by agonists of Gq-coupled GPCRs, including the histamine H
1, muscarinic M
3 and thromboxane receptors was reduced by RGS2 [
32]. As PARs are Gq-coupled GPCRs present on the airway epithelium and promote inflammatory cytokine expression [
33,
34], their targeting by RGS2 could be therapeutically beneficial. Indeed, a PAR1 antagonist markedly reduced neutrophil influx into the mouse lung 4 and 24 h post infection with
S. pneumonia [
35]. Given that PARs mediate inflammatory responses following their cleavage by proteases released by neutrophils and other inflammation-activated processes [
7,
36], we used a mouse, LPS-driven, model of airway inflammation and AHR to mimic the effects of acute bacterial infection. This allows effects of
Rgs2 deficiency to be explored in an acute neutrophilic setting [
37], and which may be relevant to exacerbations of airway diseases, including asthma [
1].
Discussion
As may occur with acute bacterial infections, the current LPS exposure model produced a rapid onset neutrophilic inflammation of the airways that was associated with transient AHR. Robust neutrophil recruitment into the lungs and BAL fluid was evident 3 h post LPS-exposure, presumably due to the rapid production of pro-neutrophilic/pro-inflammatory chemokines and cytokines observed 3 h following LPS exposure. These findings are consistent with prior studies using aerosolized or intra-nasal LPS administration [
37,
41‐
45]. The current analysis found no evidence of PAS staining in the lungs, suggesting that mucus hypersecretion was not a feature of the acute response to LPS. There was also no evidence of LPS-induced airway remodelling. This is consistent with the acute nature of the model, but, as reported elsewhere, may occur at later times post-exposure [
45]. LPS-induced AHR was transient, with marked increases in MCh-induced reactivity observed 3 h following LPS exposure, being largely resolved at 24 h. This result was slightly unexpected as the rapid accumulation of neutrophils into the BAL fluid at 3 h was maintained at both 6 and 24 h post-LPS exposure. Nevertheless, overall lung inflammation, as evidenced by H&E staining, was maximal around 3 h and by 6 h, and certainly 24 h, post-LPS exposure was resolving in the wild type animals. This is consistent with the expression of major inflammatory cytokines (CSF2, CSF3, IL6, TNF) and key neutrophil chemoattractants, including CXCL1 and CXCL2. These were all highly expressed in the lung and BAF fluid at 3 h, but by 24 h post-LPS exposure, expression had declined. Of relevance are possible effects of anesthetics, including pentobarbital and ketamine. These may directly, and/or indirectly, impact on inflammation [
46]. While used prior to lung function analysis and/or lung removal, the time available to modulate existing LPS-induced inflammation is short and, therefore, unlikely to materially effect outcomes. Use of these compounds is also common to each study arm and therefore controlled within the study design.
In the absence of inflammatory stimulus,
Rgs2 loss produced a marked AHR apparent on MCh challenge [
26,
28]. The current study extends these data by showing that LPS-induced reductions in lung function were further exacerbated in animals lacking
Rgs2. Thus, 3 h post-LPS exposure, airways resistance induced by MCh was significantly greater in
Rgs2 knock-out animals. Possibly more striking was that 24 h post-LPS exposure, lung function in the wild type animals was recovering, whereas in the
Rgs2−/− animals, MCh-induced resistance remained elevated. Furthermore, at 24 h,
Rgs2 knock-out animals displayed a markedly and significantly reduced compliance compared to wild type animals. Such data, together with those of prior studies, confirm that the
Rgs2 gene is not only bronchoprotective in the absence of airway inflammation, but is also bronchoprotective in chronic airway inflammation induced by HDM, ovalbumin, IL13 and, as now reported, in acute neutrophilic inflammation [
26‐
28,
47]. Interestingly, not only was RGS2 bronchoprotective, but the effect of
Rgs2 loss on lung compliance at 24 h post-LPS exposure suggests roles for RGS2 in assisting with resolution, or protection, post-exposure. What these mechanisms could be are unclear, but may relate to
Rgs2 loss enhancing the profibrotic and remodelling effects of receptors, including PARs [
16,
48‐
50]. Such effects may become more relevant in chronic neutrophilic inflammatory diseases, as caused by cigarette and smoke inhalation [
51]. However, as noted, no effects of
Rgs2 deficiency were apparent on remodelling and this would require investigation in longer-term models.
In terms of roles for
Rgs2, in the absence of inflammatory stimulus, prior studies show no effect of
Rgs2 gene loss on lung inflammation or the recruitment of inflammatory cells into the BAL, when compared to wild type animals [
26,
28]. In the current study, an effect of
Rgs2 deficiency on LPS-induced inflammation was hypothesized. However, when compared to wild type, no major change in the inflammatory response to LPS was noted in animals lacking
Rgs2. In each case, lung inflammation and inflammatory cell recruitment, including neutrophils, to the BAL fluid were similar. Furthermore, expression of the most highly expressed cytokines and chemokines was similar in
Rgs2−/− and wild type animals. Such data do not support a major role, protective or otherwise, for RGS2 in LPS-induced airway inflammation and is consistent with a lack of the effect of
Rgs2 deletion on inflammatory indices in HDM- and IL13-induced inflammation [
28,
47]. In this regard, LPS acts via TLR4 to induce inflammation [
52]. As this pathway does not directly involve Gq-coupled GPCRs and primarily promotes expression of inflammatory mediators, such as TNF, via the activation of NF-κB and MAPKs [
53], direct effects of RGS2 were not anticipated. Nevertheless, our original hypothesis was that neutrophilic inflammation, induced by LPS inhalation, may involve GPCRs that are regulated by RGS2. Numerous neutrophil-derived mediators act on GPCRs, many of which do couple via Gq [
54]. The release such mediators is induced by LPS and roles for Gq-coupled GPCRs in accentuating, or further promoting, inflammatory responses were therefore expected. Similarly, numerous proteases are produced by neutrophils, and other inflammatory processes, and may cleave and activate PARs [
55]. Both PAR1 and PAR2 are present on multiple cell types in the airways, in particular structural cells, including the epithelium and ASM, and can also be activated by neutrophil proteases, including neutrophil elastase [
7,
55]. Indeed, PAR activation enhances expression of various inflammatory mediators [
16,
19‐
21,
56,
57]. Similarly, PARs are implicated in various pro-inflammatory effects in vivo [
7,
17,
18,
36,
55]. Thus, the current data showing no clear effect of
Rgs2 loss on inflammation, even 24 h post-exposure, requires explanation. One possibility may be that PARs do play a role in neutrophilic inflammation, but that these receptors, or their relevant downstream signaling pathways, for example. Gi, G12/13 or β-arrestin, unlike Gq, are not targeted by RGS2 and are therefore unaffected by
Rgs2 deficiency. However, while understanding of the selectivity of different RGS proteins for specific GPCRs is incomplete [
58,
59], there is little suggestion of Gq-coupled GPCR selectivity for RGS2 and multiple studies indicate that, at least, PAR1 signalling is targeted by RGS2 [
60‐
62].
Further explanations for the above data may come from a more detailed consideration of the roles for Gq-coupled GPCRs, such as PAR1 and PAR2, in inflammation. These are not simply “pro-inflammatory”, both pro- and anti-inflammatory effects are attributed to PARs [
55,
63]. For example, agonism at Gq-coupled GPCRs leads to phospholipase A
2 activation to induce prostaglandin (PG) production in epithelial cells [
64]. PAR activation certainly promotes PGE
2 release, which, in turn, can dampen expression of inflammatory cytokines and provide bronchoprotection [
34,
63,
65‐
67]. Indeed, the PAR2-PGE
2 axis was anti-inflammatory in an ovalbumin sensitized model of allergic airway inflammation [
68]. Thus, RGS2 may well modulate Gq-coupled GPCR activity, but the effects of this could be both pro- and anti-inflammatory leading to little, or no, net effect due to
Rgs2 deletion.
Finally, while there was no clear effect of
Rgs2 knockout on the expression of highly expressed cytokines and chemokines, including CCL4, CCL4, CSF2, CSF3, CXCL1, CXCL2, IL6 and TNF,
Rgs2 loss significantly increased IL12B expression. Given a central role for IL12 in the development of Th1 responses, these data raise the possibility that RGS2 may play a regulatory role in the modulation of Th polarization. Indeed, neutrophil elastase promotes IL12 generation from LPS-stimulated macrophage via an apparently PAR2-dependent mechanism [
69]. This could be enhanced in the absence of RGS2 or, possibly, abrogated by increased RGS2 expression. Similar effects may occur on dendritic cells, a known source of IL12 [
70], which, like macrophage, are present in the lungs. Since, (1) IL12 and other Th1 gene polymorphisms are associated with severe asthma [
71,
72]; and (2) in the context of allergen-induced inflammation and AHR, IL12 reduces Th2 responses, including AHR [
70], further analysis of roles for RGS2 in T-helper cell programming/function is appropriate.