|Home | About | Journals | Submit | Contact Us | Français|
Despite over 50 years of inhaled beta-agonists and corticosteroids as the default management or rescue drugs for asthma, recent research suggests that new therapeutic options are likely to emerge. This belief stems from both an improved understanding of what causes and regulates airway smooth muscle (ASM) contraction, and the identification of new targets whose inhibition or activation can relax ASM. In this review we discuss the recent findings that provide new insight into ASM contractile regulation, a revolution in pharmacology that identifies new ways to “tune” G protein-coupled receptors to improve therapeutic efficacy, and the discovery of several novel targets/approaches capable of effecting bronchoprotection or bronchodilation.
Asthma is a prevalent disease. The World Health Organization estimates 235 million people worldwide suffer from asthma, which is the most common noncommunicable disease among children (WHO, 2013). Unfortunately, multiple studies suggest up to 55% of asthmatics have suboptimal control of their disease (Peters, et al., 2007), and concerns of asthma drug safety have persisted for decades (Donohue, 2008). These problems represent an obvious clarion call for the development of new, better and safer anti-asthma drugs. Although recent studies suggest certain asthmatic subpopulations may benefit from new antibody-based therapies that target IgE, specific cytokines, or their receptors (reviewed in (Mitchell, et al., 2016) these represent expensive and difficult-to-manage options for niche populations. For the vast majority of asthmatics, treatment options have changed little over the last 2 decades. Other than anti-cysteinyl leukotriene receptor antagonists, which have modest efficacy for most asthmatics, there have only been refinements of the mainstream asthma drugs- beta-agonists and anti-cholinergics- to improve their specificity and duration on action. Truly novel therapeutic approaches that overcome the limitations of current drugs (including efficacy, tolerance, ease of use, and safety) for obstructive airway diseases have not emerged.
Fortunately, recent advances in multiple fields of science now provide new direction that dramatically changes the prospects for asthma therapy. Most prominent among these advances are those in pharmacology, including increased efforts into the development of allosteric modulators, the introduction of “bi-functional” drugs that represent a fusion of two established drugs in one molecule, and a newfound appreciation of qualitative signaling that has stimulated the identification and development of “biased” ligands.
Moreover, recent studies which provide a better understanding of what dictates and controls airway smooth muscle (ASM) contraction have helped create an expanded list of therapeutic targets, including previously unappreciated GPCRs on ASM. This knowledge, coupled with drug discovery approaches that no longer focus solely on orthosteric ligands, has us poised to break the long drought in delivering new asthma therapeutics.
Asthma management has historically focused on: 1) directly inhibiting ASM contraction to either prevent (bronchoprotection) or reverse bronchoconstriction (bronchorelaxation); and 2) controlling airway inflammation in order to limit the stimuli in the airway that cause ASM contraction. A secondary yet important benefit of controlling inflammation is the limiting of other factors, independent of ASM contractile state, that increase impedance to airflow (mucus, airway edema).
Corticosteroids, either inhaled or oral, have been the drug of choice for managing airway inflammation (most commonly caused by inhaled allergen). Inhaled corticosteroids are the recommended choice for managing mild asthma. However, lack of direct bronchodilator activity and the time lag between inhalation of drug and resolution of inflammation does not permit them to be a rescue drug when rapid reversal of acute bronchospasm is required. Currently, the best-selling asthma drug is a combined inhaled formulation of the corticosteroid fluticasone and the long-acting beta-agonist salmeterol. However, due the slow onset of action of salmeterol, this drug is not suitable for rescue. Faster-acting beta-agonists, as either monotherapy or when combined with corticosteroids (formoterol/budesonide), are the drugs or choice for rescue from acute bronchospasm.
Other anti-inflammatory drugs used in asthma management include cromolyn sodium, prescribed for treatment of mild persistent asthma and presumed to work by stabilizing mast cells (Lane & Lee, 1996). Anti-IgE antibody treatment has recently emerged for the management of severe, corticosteroid-resistant asthma (Humbert, et al., 2014). Other drugs that target specific inflammatory agents include anti-IL-13 antibodies which have shown promise in recent clinical trials (De Boever, et al., 2014; Hodsman, et al., 2013; Noonan, et al., 2013), whereas results from trials testing anti-IL-5 antibodies have been mixed (Powell, et al., 2015). Among those drugs with direct bronchodilator properties (discussed below), cysteinyl leukotriene type 1 receptor (CysLT1R) antagonists exhibit anti-inflammatory properties, which can be complementary to the anti-inflammatory effects of corticosteroids, in asthmatics (Currie & Lipworth, 2002; Davis, et al., 2009; Laitinen, et al., 2005; Maeba, et al., 2005; Pizzichini, et al., 1999; Ramsay, et al., 2009; Sandrini, et al., 2003). Anti-inflammatory effects of M3 mAChR antagonists have also been shown in animal models of asthma (Bos, et al., 2007; Gosens, et al., 2005; Ohta, et al., 2010). The only study to address this phenomenon in humans paradoxically found that clinical benefits of tiotropium were associated with increased inflammation markers in sputum (Powrie, et al., 2007). However, the potential concentrating effect of tiotropium on airway mucus (Powrie, et al., 2007) as well as other limitations (Soriano & Barnes, 2007) may be confounding these findings. Lastly phosphodiesterase inhibitors, currently proposed as adjuncts or add-on therapy in asthma, have demonstrated anti-inflammatory properties in clinical studies (Fan, 2006).
The scope of this review precludes an in-depth analysis of the class of corticosteroid drugs; the reader is referred to (Barnes, 2005; Boardman, et al., 2014; Ito, et al., 2006a). Those corticosteroids currently prescribed vary modestly in their ability to control airway inflammation as assessed by clinical studies measuring various indices of inflammation (Westergaard, et al., 2015). A wealth of literature exists detailing mechanisms by which corticosteroids suppress inflammation (Boardman, et al., 2014; Newton, 2000); a topic well beyond the scope of this review. Since the first corticosteroid introduced to treat asthma, the refinement of these drugs has focused on improving potency and mode of delivery to reduce systemic side effects (Colice, 2006). Current efforts to develop better corticosteroids for asthma include attempts at dissociating the transactivating (gene transcription-inducing) from the transrepressive (transcription-suppressing) properties of the drug. This approach is based on the belief that the therapeutic benefit of corticosteroids is derived primarily from their ability to suppress, via either inhibition of transactivating transcription factors or direct biding to negative glucocorticoid response elements in gene promoters, transcription of pro-inflammatory genes (Westergaard, et al., 2015). Whether this simplistic rationale will result in an improved corticosteroid is unclear. Of note, anti-inflammatory effects of GR-induced genes have been identified (King, et al., 2013; Newton & Holden, 2007), suggesting selective GR transactivation has therapeutic benefit and a more refined approach to gene regulation is desirable.
Most bronchodilators act through G protein-coupled receptor (GPCR) activation (Gs pathways) or inhibition (Gq pathways) at the receptor locus. We have previously espoused the concept of a dynamic or competitive balance of GPCR signaling in ASM as an important determinant of ASM contractile state, and thus, airway resistance in asthma and COPD (Humbert, et al., 2014; Lane & Lee, 1996; Peters, et al., 2007). Although airway resistance is admittedly influenced by additional factors including airway remodeling, airway architecture and tissue mechanics, and other phenomena contributing to impedance, agents that contract and relax airway smooth muscle have a dominant effect on airflow and largely work through ASM GPCRs. Allergic airway inflammation is typically associated with increased levels of numerous agonists for different pro-contractile (typically Gq-coupled) ASM GPCRs causing ASM contraction and increased airway resistance. One means of preventing or reversing this effect is to deliver to the airway a selective small molecule antagonist of a GPCR causing ASM contraction (for example M3 mAChR or CysLT1R antagonists). However, this strategy can be of limited effect, especially if multiple different receptors are contributing to bronchoconstriction. Alternatively, delivery of beta-agonists to the airway promotes broncho-protection/relaxation by antagonizing signaling from pro-contractile ASM GPCRs at multiple junctures including transmembrane receptor signaling, calcium flux through membrane channels and intracellular stores, and distal elements regulating myosin light chain phosphorylation (Figure 1) through myosin chain kinase regulation or myosin light chain phosphatase regulation (calcium sensitization). Thus, beta-agonists, as well as other agents (e.g., phosphodiesterase inhibitors and agonists of other Gs-coupled receptors discussed below and noted in Fig. 1) have the ability to broncho –relax/protect an airway exposed to numerous and multiple pro-contractile GPCR agonists.
Not surprisingly, the evolution of GPCR ligands as therapeutic bronchodilators has involved improving specificity of the ligands for the critical GPCR subtype target. These refinements have led to significant reductions in off-target / side effects that severely limited the usefulness of early drugs. For the M3 mAChR, progressive modification of the nonspecific mAChR antagonist atropine has produced drugs with fewer side effects, especially when delivered by inhalation. The short-acting mAChR antagonist ipratropium was one such drug used for asthma (albeit more frequently for COPD) management for almost 30 years, ultimately supplanted by the long-acting mAChR antagonist tiotropium, which possesses fast/slow dissociation kinetics for the M2 mAChR and M3 mAChR respectively, and exhibits a superior clinical profile to that of previous mAChR antagonists (Haddad, et al., 1994). Current CysLT1R antagonists have not undergone refinement since their initial introduction as asthma therapies.
Beta-agonists as asthma therapy have a long and illustrious history. Since the introduction of epinephrine use for asthma in the early 1900’s and the later use of isoporoterenol in the 1940’s, beta-agonists have undergone continuous refinement in terms of increased specificity for the beta-2 (β2) -adrenoceptor (AR) (versus β1AR and αAR) subtype, and development of increasingly longer acting compounds (Walker, et al., 2011). Albuterol (salbutamol), introduced in 1969, was the first widely used inhaled β2AR-selective drug, and remains the drug of choice today for rescue from acute bronchospasm.
Beta-agonists confer bronchoprotection and cause bronchodilation primarily via activation of the cAMP-dependent protein kinase aka protein kinase A (PKA). This principal mechanism of action was the long-held dogma based not on direct empirical evidence, but on the observation that most agents known to stimulate intracellular cAMP accumulation caused relaxation. In 2011 Zieba et al. (Zieba, et al., 2011) challenged this notion asserting Epac as the cAMP effector mediating the relaxant effect of cAMP-inducing agents in ASM, providing evidence that Epac-selective activators were sufficient to cause ASM relaxation. However, our group recently demonstrated that direct inhibition of PKA in human ASM cells or murine airways ex vivo inhibited the vast majority of the relaxant effect of beta-agonist, demonstrating that PKA is indeed the principal effector of beta-agonist-mediated ASM relaxation (Morgan, et al., 2014).
Although asthma drugs undoubtedly save thousands of lives each year, they are not perfect. Numerous clinical studies assessing various measures of clinical efficacy report a high percentage (in one study up to 55%) of asthmatics have suboptimal control (Joyce & McIvor, 1999). Because asthma is a syndrome with multiple contributing pathogenic mechanisms that can vary among sufferers, this fact is not too surprising, particularly when one considers the ever present problem of treatment adherence. Beyond efficacy issues, related safety concerns for certain asthma drugs have also existed for years. Below we will discuss the various issues related to the limitations of several asthma drugs.
Despite inhaled short acting beta-agonists being the drug of choice for relief from acute asthmatic attacks while long acting beta-agonists (LABAs) combined with corticosteroids are the most frequently prescribed asthma control medication, beta-agonist efficacy and safety have been the subject of ongoing debate for decades. Several studies have linked chronic beta-agonist use with adverse patient outcomes such as functional β2AR tachyphylaxis (Lands, et al., 1967; Waldeck, 2002), deterioration of asthma control (Nelson, et al., 2006; Salpeter, et al., 2006), and death (Salpeter, et al., 2006). Although chronic beta-agonist treatment is not always associated with adverse events (Walters, et al., 2007), there is clearly a dearth of mechanistic understanding of the effects of beta-agonists on airway physiology and asthma pathology.
Drug tolerance or tachyphylaxis is a well-appreciated concept that applies to many drugs. Multiple clinical studies have documented a loss of the bronchoprotective effect of inhaled beta-agonist that occurs with repeated beta-agonist use (Cates & Cates, 2008; Salpeter, et al., 2006). Compelling data do not exist to support a loss of the bronchodilatory effect of beta-agonists, as assessed by the reversal of the drop in FEV1 after methacholine challenge (Rosenthal, et al., 1999). However, the safety concerns associated with long-acting beta-agonist (LABA) use (discussed below) suggest that under certain conditions beta-agonist-mediated bronchodilation might be compromised.
Initial safety concerns over beta-agonist treatment of asthma related to the nonselective nature of early β-agonists such as epinephrine (α- and β adrenoceptor- selective) and isoproterenol (β adrenoceptor-selective) which resulted in numerous cardiovascular-related side effects including tachychardia, arrhythmia, tremor, and headache. However, a progressive understanding of adrenoceptor subtypes (Ahlquist, 1948; Lands, et al., 1967) facilitated drug discovery efforts that led to the development of the β2AR -selective albuterol and terbutaline (Waldeck, 2002). Despite the relative β2AR-selectivity of the widely used short-acting beta-agonists (SABAs) and LABAs in the treatment of asthma , there is considerable variability among these drugs, and patient sensitivity for experiencing cardiovascular side effects varies as well.
The more pressing beta-agonist safety issue for over two decades now relates to mortality concerns. A history of the various “epidemics” associated with use of SABAs and LABAs (not in combination with corticosteroids) as asthma drugs is detailed in Ortega and Peters (Ortega & Peters, 2010). Increasing concern over beta-agonist safety reached a crescendo following termination in 2003 of the Salmeterol Multicentre Asthma Research Trial (SMART), a prospective study of salmeterol initiated by Glaxo Smith Kline in 1996. Termination of the trial was prompted when an interim analysis demonstrated a 4.4-fold increase in death in asthmatics receiving salmeterol versus placebo (Nelson, et al., 2006). A subsequent meta-analysis of multiple LABA trials reported significantly increased odds ratios for life-threatening exacerbations and asthma-related deaths (Salpeter, et al., 2006). Shortly after termination of the SMART trial the FDA issued a black box warning issued for LABA (including salmeterol and formoterol) use in the United States, which has since been updated multiple times.
Although considerable debate continues over the interpretation of the SMART study and the Salpeter meta-analysis, and of the appropriateness of the FDA black box warning, concerns of LABA safety with asthma are a reality that currently impacts asthma management, basic and clinical research in beta-agonist efficacy and pharmacology, and current drug discovery efforts.
Although the basis for the loss of the clinical effect of beta-agonist and the current safety concerns is unknown, desensitization of the beta-2-adrenoceptor, the transducer of beta-agonist effects on airway smooth muscle, has been assumed to underlie this loss of functional effect (Billington & Penn, 2003; Deshpande & Penn, 2006; Walker, et al., 2011).
Mechanisms effecting β2AR desensitization, including those specifically in ASM, have been reviewed elsewhere (Billington & Penn, 2003; Penn, et al., 2000; Reiter & Lefkowitz, 2006), but will be briefly summarized here. Activation of the β2AR in any cell sets in motion both rapid and more slowly induced feedback mechanisms that shut down signaling. Rapid mechanisms involve receptor phosphorylation by kinases that serve to uncouple the β2AR from the Gs protein. The agonist-occupied receptor is rapidly phosphorylated by GPCR kinases (GRKs) which not only diminish the ability of the β2AR to couple to Gs, but also promote the association of arrestin molecules with the receptor; this arrestin-β2AR association serves many purposes (discussed below), one of which is to sterically inhibit β2AR-Gs association. Other kinases, including PKA and PKC can also phosphorylate the β2AR to promote uncoupling, without the β2AR necessarily being occupied by agonist. More extended mechanisms of β2AR desensitization include receptor internalization (mediated by arrestins for the agonist-occupied receptor) which sequesters the receptor from G protein, and receptor downregulation (loss of total cellular β2AR protein) which can occur over the course of hours when internalized receptors traffic to lysosomes for degradation.
Another important rapid feedback mechanism is the rapid activation of phosphodiesterases (PDEs) that occurs as a result of intracellular PKA activation. PDEs break down cAMP thus removing the stimuli for PKA activation; PDE inhibitors sustain β2AR-mediated (or that by any other GPCR or agent that can activate adenylyl cyclase) induction of cellular cAMP levels, resulting in sustained intracellular PKA activity.
We have demonstrated the relative importance of the different β2AR negative feedback mechanisms in ASM, and have established their (negative) impact on the ability of beta-agonists to relax ASM in vitro, ex vivo, and in vivo. Knockdown of GRK2/3 or β-arrestin-2 augments beta-agonist-stimulated cAMP accumulation and PKA activation in human ASM cells, and knockout of β-arrestin-2 also augments β2AR signaling and the ability of beta-agonists to relax methacholine-stimulated murine airways ex vivo and bronchoconstriction in vivo (Deshpande, et al., 2008; Pera, et al., 2015). Inhibition of PKA via heterologous expression of a PKI inhibitory peptide also significantly increases beta-agonist-stimulated cAMP accumulation, but inhibits beta-agonist-mediated relaxation of airways ex vivo due to the role of PKA as the principal effector of β2AR-mediated relaxation of ASM (Morgan, et al., 2014). Collectively, these studies strongly suggest that β2AR desensitization does occur in ASM and does limit the therapeutic utility of beta-agonist in obstructive lung diseases.
Also likely contributing to the limitations of beta-agonists as asthma drugs is the increasing realization that despite providing symptomatic relief, beta-agonists do little to address the underlying pathogenic mechanisms of asthma. Increasing evidence now supports beta-agonists having no anti-inflammatory effect, and perhaps possessing a pro-inflammatory effect (Callaerts-Vegh, et al., 2004; Loza, et al., 2007; Nguyen, et al., 2009; Thanawala, et al., 2013). Despite some early studies suggesting beta-agonists can suppress pro-inflammatory properties of certain cells (Ammit, et al., 2002; Hallsworth, et al., 2001), beta-agonists do not appear capable of suppressing airway/lung inflammation, and appear to be important in promoting airway mucus production (Thanawala, et al., 2013). Indeed, editorials published shortly after the SMART trial controversy suggested the symptomatic relief provided by beta-agonists has a masking effect on uncontrolled airway inflammation (Salpeter, et al., 2006). Although this interpretation at first seemed to lose traction due to a lack of supportive research, recent studies suggest it may indeed be fairly accurate (see below).
Additionally, beta-agonists (as well as corticosteroids) fail to address the pathogenic progression of airway remodeling (including increases in ASM/mesenchymal cell mass, matrix deposition, basement membrane thickening) that occurs in many asthmatics to increase the severity of their disease with age (and likely contributes to the prevalence and severity of asthma in the elderly (Hanania, et al., 2011)). Although beta-agonists have been shown to inhibit mitogen-stimulated proliferation of ASM cells in culture (Yan, et al., 2011), the effect is rather modest, and there is no evidence from animal or clinical studies suggesting beta-agonists deter airway remodeling in vivo. The apparent inability of current asthma drugs to address airway remodeling has emerged as an important topic, yet resolving this problem will continue to be a significant challenge until regulatory and economic forces become favorable for drugs that target slow-developing (and age-related) mechanisms of disease.
Other bronchodilators drugs do not suffer from the significant safety concerns plaguing beta-agonists, but do have limitations with respect to efficacy, and side effects. Muscarinic receptor antagonists are not considered as efficacious in bronchodilating as beta-agonists (particularly for those asthmatics in which non-cholinergic factors contribute to bronchoconstriction) (Baigelman & Chodosh, 1977; Rebuck, et al., 1983), and can produce dry mouth in some patients (Kesten, et al., 2006; Tashkin, et al., 2008). CysLT1R agonists can also have limited efficacy relative to beta-agonists (Ducharme, et al., 2006). Limitations of PDE inhibitors are discussed below.
Although corticosteroids tend to be extremely effective drugs in suppressing allergic lung inflammation, they suffer from limitations with respect to: 1) side effects; and 2) limited efficacy in specific asthma populations or with specific asthma triggers (Cavkaytar, et al., 2015; Cooper, et al., 2015; Fuhlbrigge & Kelly, 2014).
Chronic corticosteroid therapy is associated with significant and serious local and systemic side effects. Systemic side effects of oral corticosteroid use include suppressed hypothalamic-pituitary-adrenal (HPA)-axis function, reduced mineral bone density and retarded growth velocity in children (Xu, et al., 2009). Although systemic side effects are reduced by local delivery of corticosteroids by inhalation, the drug retains systemic bioavailability through ingestion of particles which do not reach the lung, as well as absorption through the pulmonary circulation. High dose inhaled corticosteroid use has been linked to transient reduction in growth velocity, which can negatively influence treatment compliance in children (Boulet, 1998; Xu, et al., 2009). The local side effects are caused by the deposition of corticosteroids in the mouth and throat during inhalation and include pharyngitis, cough and oropharyngeal infections (Dahl, 2006).
Although inhaled corticosteroids are effective in most asthmatics, there are subgroups of patients which do not respond sufficiently in what is referred to as corticosteroid resistant asthma. In particular, severe asthma, presence of a COPD-like asthma phenotype and current cigarette smoker status are associated with steroid resistance (Nyenhuis, et al., 2010). There are several potential mechanisms of steroid insensitivity. Under inflammatory conditions, in particular through the actions of the cytokines IL-2, IL-4 and IL-13 the GRα may undergo increased phosphorylation by p38 MAP kinase (Irusen, et al., 2002), rendering it unable to bind GRE promoter sites in the nucleus. In addition, proinflammatory cytokines including IL-2, IL-4 and IL-13 as well as TNFα and IL-1β upregulate the expression of GRβ (Bamberger, et al., 1995; Webster, et al., 2001). GRβ is a homolog of GRα which does not bind corticosteroids or promote GRE-mediated gene transcription, while retaining its ability to bind GRE sequences and thus has been proposed to act as an endogenous inhibitor of GR-signaling. Importantly, increased numbers of inflammatory cells (primarily T-cells) expressing GRβ have been found in steroid resistant and fatal asthma (Christodoulopoulos, et al., 2000; Leung, et al., 1997). Other mechanisms include activation of transcription factors (Adcock, et al., 1995) AP1 and IRF-1 which inhibit GRE binding either by direct interaction with GRα or by scavenging the GR transcriptional co-activator GR interacting protein 1 (GRIP-1) (Bhandare, et al., 2010; Tliba, et al., 2008), respectively. Decreased HDAC2 expression in severe asthma and in smokers with asthma may also contribute to steroid insensitivity, as GR-mediated increase in HDAC2 expression is a mechanism by which steroids inhibit proinflammatory gene expression (Ito, et al., 2000; Ito, et al., 2006b). For additional detail regarding mechanisms mediating corticosteroid resistance, the reader is referred to the following reviews (Adcock & Barnes, 2008; Boardman, et al., 2014)
In addition to their ability to inhibit contraction mediated by CysLT1Rs on ASM, CysLT1R antagonists inhibit cysteinyl leukotriene-mediated inflammatory events effected by various inflammatory cells. Because these events are but a subset of the inflammatory gestalt in the airway, the effectiveness of CysLT1R antagonists can be limited, and those patients effectively managed on CysLT1R antagonists may have inflammation (or bronchoconstriction) that is largely cysteinyl leukotriene-mediated. Similarly, the efficacy of other asthma drugs with potential anti-inflammatory properties (e.g., M3 mAChR antagonists, cromolyn), is limited due to their restricted modes of action.
With respect to other treatments options intended to address lung inflammation, the most currently discussed option is anti-IgE therapy. However, the high cost and (high maintenance) of anti-IgE therapy renders it an option primarily for poorly- or un- controlled severe asthmatics. The strengths and limitations of other anti-inflammatory treatments for asthma are reviewed elsewhere (Chung, 2015; Durham, et al., 2015; Humbert, et al., 2014; Olin & Wechsler, 2014; Schuligoi, et al., 2010).
Aided by the increasing power of molecular biology and genetic approaches, screening strategies, and animal models of asthma, the last decade has witnessed the discovery of numerous players that function as asthma mediators and/or therapeutic targets. Below we describe both previously unappreciated GPCRs and non-GPCRs that have potential as useful therapeutic targets, focusing on those capable of direct effects on ASM contractile state.
Initial studies by the Liggett lab identified multiple mRNA species of the BTR family in a transcriptome analysis of human ASM cultures. Deshpande et al. (Deshpande, et al., 2010). subsequently demonstrated that various BTR ligands, including saccharin, chloroquine, and quinine stimulated a calcium flux in ASM cells. Despite promoting calcium mobilization, the bitter tastants decrease ASM cell stiffness, relax contracted ASM tissue from mice and humans, and decrease airway resistance in a murine model of asthma. The exact mechanism for this relaxant effect has been subject of debate (Deshpande, et al., 2010; Tan & Sanderson, 2014; Zhang, et al., 2013b), but bitter tastant receptors have nevertheless emerged as intriguing new therapeutic targets, with considerable efforts underway to synthesize new, more potent TAS2R agonists (Behrens & Meyerhof, 2013; Brockhoff, et al., 2010; Levit, et al., 2014).
OGR1 aka GPR68 was a poorly understood orphan receptor until the discovery in Ludwig et al. that lowering extracellular pH (i.e. acidifying cell culture media), with the proton as the presumed “agonist”, could stimulate the receptor (Ludwig, et al., 2003). This discovery contributed the characterization of a GPCR subfamily of “proton-sensing” receptors, comprised of OGR1, GPR4, TDAG8 (GPR65) and G2A (GPR132), with diverse functions including the regulation of cell/survival growth under oncogenic conditions (Seuwen, et al., 2006). OGR1 was found to be expressed in vascular smooth muscle (Tomura, et al., 2005) and later in ASM (Ichimonji, et al., 2010; Saxena, et al., 2012). Modest reductions in extracellular pH (acidification; pH 8.0 – pH 6.8) contract ASM cells and tissue, and multiple OGR1-dependent signals including calcium mobilization, cAMP accumulation and PKA activation, and p42/p44 MAPK activation were elicited in response to reduced extracellular pH (Saxena, et al., 2012). Interestingly, a significant proportion of the OGR1-mediated PKA activation was Gq-independent, indicating OGR1 has the capacity to couple to both Gq and Gs. This raises the possibility of exploiting qualitative signaling properties of OGR1 to bias it towards a Gs-mediated, pro-relaxant function (see below).
However, given the nature of its apparent cognate ligand- the proton- OGR1 appears to be an intractable target or at least an extremely difficult receptor to study and manipulate. Fortunately, a large collaborative effort headed by the Roth lab recently identified a subset of benzodiazepines that function as allosteric modulators of OGR1, (Huang, et al., 2015) thus providing a critical tool for characterizing the properties of this receptor.
Huang et al. recently demonstrated that certain desmethyl benzodiazepines can stimulate OGR1 to promote calcium mobilization, cAMP accumulation, and elicit the same phospho-protein induction caused by decreased extracellular pH (Huang, et al., 2015). Moreover, insight provided by structure of OGR1 enabled the synthesis of novel OGR1 ligand capable with demonstrable signaling and function effects in a behavioral model (OGR1 is highly expressed in brain). Future studies assessing the effects of these benzodiazepines on ASM function will be interesting, and offer the hope that OGR1 can be manipulated to function primarily as a Gs-coupled GPCR mediating bronchodilation.
Although we tend to focus on intracellular calcium as the bad boy causing ASM contraction and bronchoconstriction, extracellular calcium is now known to effect ASM contraction by acting as a ligand for CaSRs expressed on ASM. CaSRs are best known for being expressed on the parathyroid gland, serving as a homeostatic sensor of blood calcium levels by controlling the release of parathyroid hormone. In ASM, the Gq-coupled CaSR tranduces the extracellular calcium signal into an intracellular calcium signal to regulate ASM contractile state (Yarova, et al., 2015). Interestingly, the CaSR also responds to polyvalent cations, certain amino acids, and virus elements all of which can be present in the airway under different pathological conditions. Yarova et al. (Yarova, et al., 2015) recently demonstrated the relevance of multiple CaSR activators as stimuli for ASM contraction, and demonstrated the contributory roles of CaSRs in the development of pathology in a murine model of asthma. Inhaled small molecule CaSR antagonists termed calcilytics were effective in reversing both AHR as well as lung inflammation. Interestingly, CaSR expression in ASM appears greater in asthmatic versus non-asthmatic humans (Yarova, et al., 2015). The potential for calcilytics as asthma therapy (calcilytics are currently in trials for treatment of autosomal dominant hypocalcemia ("A Study to Determine the Effects of NPSP795 on the Calcium-sensing Receptor in Subjects With Autosomal Dominant Hypocalcemia as Measured by PTH Levels and Blood Calcium Concentrations https://clinicaltrials.gov/ct2/show/NCT02204579?term=npsp795&rank=1),") has been recently discussed (Penn, 2015).
Several potential asthma therapy targets have been recognized for years but for various reasons, effective drugs targeting them were either never developed or never advanced through clinical trials. Recent studies providing clarity in their role in asthma pathology or how they might be better targeted, or simply improved medicinal chemistry have brought these targets back into play. Some of these are discussed below.
Although recognized as regulators of ASM contractile state for some time, interest in targeting EP receptors as asthma therapy waned once initial clinical trials identified substernal burning and cough as significant side effects of inhaling the endogenous EP receptor ligand PGE2 (Walters & Davies, 1982). However, recent studies have shed light on the specific actions of the different EP receptor subtypes in various resident and infiltrating lung cell types, and suggest that recently-developed subtype selective ligands may limit such side effects and possibly serve as effective bronchodilators. The Belvisi and Birrell labs recently demonstrated the EP3 receptor being the EP receptor that mediates cough (Maher, et al., 2009). The same groups also recently demonstrated striking species differences in the roles of various EP receptor subtypes in mediating relaxation/contraction of ASM. Using subtype-selective ligands in human airway tissue, the EP4 receptor was shown to mediate the relaxant effect of PGE2. In contrast, the EP2 receptor mediated relaxation of murine and guinea pig smooth muscle. Alternatively, the EP3 receptor has been shown to mediate contractions and pro-contractile signaling in both murine and human ASM (Maher, et al., 2009; Tilley, et al., 2003) (also Penn, unpublished observations). With respect to EP receptor function in the inflammation process, studies of EP receptor knockout mice are somewhat conflicting and attribute a predominate anti-inflammatory role to either the EP4 (Birrell, et al., 2015) or EP2 (Zasłona, et al., 2013) receptor. Our understanding of EP receptor subtype function in human inflammatory cells and (human) allergic inflammation is relatively poor. Studies from Feng (Feng, et al., 2006) and colleagues implicate the EP3 receptor as pro-inflammatory and the EP2 receptor as anti-inflammatory in human mast cells, and both EP2 and EP4 inhibiting the hyperosmolarity-induced histamine release in human mast cells.
Collectively, the data point to EP4 and possibly EP2/4 agonists as putative asthma therapeutics, for both their direct effects on ASM and their anti-inflammatory potential.
Protease-activated receptors are activated via an irreversible mechanism in which proteases cleave the receptor’s extracellular N-terminus to create a new N-terminus that functions as a tethered ligand activating the receptors. There are 4 members of PAR family (PAR1–4), coupled primarily to Gq and Gi, and different complements of the receptor subtypes are expressed in multiple cell types including endothelium, epithelium, various immune cells and ASM. PAR expression in cells important in allergic lung inflammation and ASM contraction has been known for some time (Knight, et al., 2001) but like the EP receptor family, targeting PARs in asthma has been problematic given the promiscuity (i.e. ability to engage a variety of G proteins) of numerous receptor subtypes and an abundance of agents (thrombin, trypsin, tryptase, and various proteolytic allergens/fungi) capable of activating them. However, recent studies have demonstrated an important role for PAR2 in mediating both allergic lung inflammation and associated AHR (Asaduzzaman, et al., 2015; Ebeling, et al., 2005; Ramachandran, et al., 2012; Schmidlin, et al., 2002; Takizawa, et al., 2005). Studies using PAR2 knockout mice have demonstrated a requirement for PAR2 in allergen-induced leukocyte recruitment (and allergen-challenged PAR2 knockout mice have a significantly attenuated asthma phenotype), whereas transgenic mice overexpressing PAR2 have an exaggerated inflammatory phenotype (Schmidlin, et al., 2002). Administration of a PAR-2 activating peptide (Ebeling, et al., 2005; McGuire, et al., 2004) enhanced allergen-induced AHR and airway inflammation in mice, while an anti-PAR(2) blocking antibody delivered during the sensitization phase to cockroach extract inhibited airway inflammation and decreased AHR (Arizmendi, et al., 2011). Collectively, these and other studies (Cocks & Moffatt, 2001) implicate PAR2 in the development of allergic inflammation and the asthma phenotype (de Boer, et al., 2014).
Yet not all actions of PAR2 are pro-asthmatic/inflammatory, due in part to actions of COX-derived PGE2 induced via PAR2 activation. Administration of PAR2 agonists can promote prostanoid-induced relaxation of mouse, rat, guinea-pig and human airways, and bronchoconstriction is elevated in PAR2 −/− mice (Cocks & Moffatt, 2001). COX-dependent PGE2 production is also known to inhibit eosinophil migration and degranulation (Cocks, et al., 1999; Ebeling, et al., 2007; Gatti, et al., 2006; Moffatt, et al., 2002; Nichols, et al., 2012; Schmidlin, et al., 2001).
These mixed actions of PAR2 have the potential to limit the therapeutic utility of any strategy for antagonizing PAR2. However, Nichols et al. recently demonstrated that the disparate effects of PAR2 could be separated into arrestin –dependent and –independent actions of the receptor (Nichols, et al., 2012). Using an ovalbumin sensitization/challenge model, repeated insufflation of a PAR2 agonist exacerbated airway inflammation and mucin production in WT mice, but not in β-arrestin2 knockout mice, suggesting that the pro-inflammatory effect of PAR2 activation requires β-arrestin2. However, PAR2-mediated ASM relaxation ex vivo as well as bronchodilation in vivo was similar in WT and β-arrestin2 knockout mice, as were BAL levels of PGE2. PAR2-dependent PGE2 production and release were dependent upon nuclear p42/p44 MAPK activation and PI3K activity, both of which have been shown to be inhibited by chelation of intracellular Ca2+, Gαq knockdown or pharmacological inhibition of PLCbeta. Thus, it is likely that the protective effects of PAR2 in the airway are mediated downstream of Gαq. In contrast, PAR2-dependent activation of β-arrestin-dependent events, such as activation of actin cytoskeletal events and cell migration, was independent of Gαq, Gαi and Gα12 signaling, consistent with the hypothesis that the pro-inflammatory effects of PAR2 in the airway are independent of G-protein coupling. Collectively, these findings posit PAR2 as an intriguing GPCR for which biased ligand pharmacology or biasing strategies might enable optimal therapeutic targeting of the receptor. Towards that end, ongoing efforts to develop novel PAR2 agonists and antagonists have been made. Recently a number of high affinity “tethered” PAR2 peptidomimetic agonists and antagonists, effective both in vivo and in vitro, have been developed (Boitano, et al., 2014). Further investigation of the potential bias of these compounds may lead to the identification of PAR2 targeted peptides with sufficient pathway specificity to inhibit the inflammatory but not the protective effects of PAR2 in the airway (Boitano, et al., 2015; Flynn, et al., 2013; Sherwood, et al., 2014).
The role of GABA receptors in neurotransmission has been long appreciated, yet their function in non-neuronal tissues remains poorly understood. Early studies discounted a role for GABA in the direct modulation of airway smooth muscle function, in part because of confounding effects of GABA receptor function in neural control of ASM, but also due to a failure to appreciate the antithetical effects of GABAA and GABAB receptors on ASM.
GABAB receptors are Gi-coupled GPCRs and have recently been shown to be expressed on ASM and represent one of the few Gi-coupled GPCRs capable of promoting ASM contraction (Mizuta, et al., 2008). Moreover, GABAB receptor activation augments pro-contractile signaling and contraction stimulated by ASM Gq-coupled receptors (Osawa, et al., 2006). Alternatively, GABAA receptors are ion channels, composed of multiple subunits, that conduct inward chloride currents that serve to hyperpolarize the cell and limit depolarization. GABAA channels in the brainstem regulate cholinergic transmisson to the lung (Haxhui, et al., 1986; Moore, et al., 2004), and both GABAA channels and GABAB receptors reside on lung postganglionic parasympathetic nerves of the lung and serve to regulate cholinergic nerve activity (Chapman, et al., 1991; Tamaoki, et al., 1987; Tohda, et al., 1998).
The Emala lab has demonstrated the GABAA receptors are expressed in ASM (Mizuta, et al., 2008), that selective GABAA activation with muscimol can relax contracted ASM (Gleason, et al., 2009), and also potentiate the relaxant effect of beta-agonists on ASM (Gallos, et al., 2008). Because ASM expresses primarily the α4 and α5 subunits of the GABAA, medicinal chemistry efforts to develop specific α4 or α5 activators have been undertaken. Most recently, the Emala and Cook laboratories have demonstrated that specific targeting of α5 GABAA subunits can relax contracted airway ex vivo as well as lung slices, and a specific a4 allosteric modulator of the GABAA channel relaxed human ASM ex vivo and reduced respiratory resistance on a murine asthma model (Gallos, et al., 2012a; Gallos, et al., 2015; Yocum, et al., 2015). These latter studies suggest that precise targeting of ASM GABAA subunits represents a promising means of regulating bronchomotor tone while avoiding potentially confounding off-target effects (in epithelia, or the nervous system, for example) of nonspecific GABAA activation.
In addition to GABAA channels, other types of chloride channels exist including the glycine channel (Yim, et al., 2011) and calcium-activated chloride channels (CaCCs) (Huang, et al., 2012; Zhang, et al., 2013a), both of which have been shown to be important in ASM relaxation. The effects of chloride flux in or out of ASM can be complicated, and likely relate to the difference is the subcellular localization of chloride channels in ASM cells and their effects on: 1) membrane potential; or 2) the regulation of other ion fluxes, including calcium fluxes, at either the plasma membrane or the sarcoplasmatic reticulum (Gallos, et al., 2012b). Whereas GABAA channel activation appears to relax ASM via an inward chloride current that hyperpolarizes the plasma membrane, the glycine channel and CaCCs can mediate a chloride efflux, which contributes to membrane depolarization during calcium increases (Huang, et al., 2009). Numerous studies have demonstrated that various chloride channel blockers can inhibit agonist- or depolarization- mediated contraction of multiple specifies of ASM ex vivo or in vivo, with the efficacy of different blockers dependent upon the nature of induced contraction (Danielsson, et al., 2015; Gallos, et al., 2013). Only recently has the TMEM 16 family been identified as the true molecular identity of calcium-activated chloride channels expressed on ASM (Huang, et al., 2009; Rock, et al., 2008) and selective TMEM16 antagonists have been shown to inhibit contraction of guinea pig (Gallos, et al., 2012a), mouse and human ASM (Huang, et al., 2012), and reduce airway hyperactivity in a murine asthma model (Huang, et al., 2012; Zhang, et al., 2013a). Collectively, these studies identify chloride channels as an intriguing therapeutic target for asthma. Moreover, results from these studies shed a new (positive) light into the regulatory role of membrane depolarization in ASM contraction. Early studies demonstrating a lack of effect of the L-type calcium channel blocker verapamil (Fish, 1984) led to an abandoning of strategies for manipulating membrane potential to control ASM contraction. Yet with the complexity of ion channel function and its role in calcium handling in ASM starting to be revealed, such strategies are clearly back in play and offer the promise of completely novel approaches to regulating bronchomotor tone.
Various Gq-coupled GPCRs activate Rho kinase in ASM to augment contractile responsiveness of ASM tissues. Rho kinase augments contraction by inhibiting myosin light chain phosphatase (MLCP), a process referred to as “calcium sensitization”. The role of Rho kinase in ASM contraction has been well documented (Gosens, et al., 2006). Rho kinase inhibitors have been shown to inhibit ASM contraction both ex vivo (including human tissues) (Gosens, et al., 2004a; Gosens, et al., 2004b; Iizuka, et al., 1999; Schaafsma, et al., 2005; Yoshii, et al., 1999) and in vivo (animals) (Iizuka, et al., 2000; Tokuyama, et al., 2002). Of note, the contribution of Rho kinase to ASM contraction is increased in various animal models of allergen sensitization (Chiba, et al., 2001; Hashimoto, et al., 2002; Schaafsma, et al., 2004), indicating an important role for this enzyme under disease conditions. However, despite a plethora of encouraging findings from animal studies, including effective bronchodilatation by inhaled Rho kinase inhibitors (Schaafsma, et al., 2008), inhalation studies in human subjects have not been published to date. The only currently clinically available Rho kinase inhibitor, Fasudil, has been found to effectively lower pulmonary arterial pressure when administered to PAH patients by inhalation (Fujita, et al., 2010), suggesting that inhalation is a viable method of delivery for this drug. Given the profound efficacy of Rho kinase inhibitors to relax contracted ASM under a wide range of conditions and systems, interest in these drugs remains high, and ultimate clinical application may depend on overcoming the hypotension associated with their use (Kuroda, et al., 2015), perhaps through development of improved (airway) targeting strategies.
The nonspecific PDE inhibitor theophylline has been used to treat COPD for decades (Giembycz & Newton, 2015; Matera, et al., 2014; Page, 2014). Its use has been constrained however, by its narrow therapeutic window, and significant side effects including gastrointestinal stress. Xanthine PDE inhibitors, other than theophylline, are not used to treat asthma; current interest in testing existing or newly developed PDE4 (BinMahfouz, et al., 2015) subtype inhibitors as asthma drugs appear focused on using such drugs as adjuncts, in combination with glucocorticoids or direct bronchodilators. Although both nonspecific PDE and PDE4 inhibition in human airway smooth muscle cells can increase intracellular cAMP (Billington, et al., 2007; Horvat, et al., 2012), nonselective PDE or PDE4 inhibitors are poor direct bronchodilators, and do not reverse acute bronchoconstriction. PDE concentrations required for ASM relaxation by far exceed those required for anti-inflammatory effects and would lead to intolerable adverse effects (Nicholson, et al., 1995). Recent studies suggest the PDE4 inhibitors as monotherapy can improve lung function in COPD (Boswell-Smith & Spina, 2007) and recently, in asthma (Bateman, et al., 2015), largely due to anti-inflammatory actions rather than direct bronchodilator properties.
A reasonable conceptual basis exists for supporting PDE inhibitors as part of a combination therapy. PDE inhibition appears to produce some beneficial effects that are non-overlapping with those of corticosteroids, and by increasing cAMP/PKA activity should similarly induce mechanisms of cooperativity demonstrated between beta-agonists and corticosteroids (Rider, et al., 2011), including synergistic effects on gene transcription (BinMahfouz, et al., 2015; Holden, et al., 2011). Moreover, cAMP/PKA signaling also appears to augment glucocorticoid receptor nuclear translocation and signaling (Eickelberg, et al., 1999; Usmani, et al., 2005) through mechanisms discussed below. Finally, an interesting property of theophylline appears to be its ability to regulate histone acetylation which is believed important in steroid resistance (Ito, et al., 2000; Ito, et al., 2006b).
The idea of combination therapies is not new, but new combinations of various drugs are being considered and coming to the market. Two types of combination therapies exist, one being the combined delivery of 2 drugs, and the other being “bi-functional” molecules.
The rationale underlying combination therapy is that 2 (or possibly more) drugs are better than one; equal or greater therapeutic efficacy might be achieved at lower drug doses due to additive or synergistic effects between the two components. Particularly if the drugs address non-overlapping features of the disease, simply inhibiting more mechanisms of disease is likely to better mitigate symptoms and manage the disease. For the combination therapy of inhaled long-acting beta-agonist (LABA) and corticosteroids, these drugs provide both direct bronchodilation and inflammation control; enabling both acute symptomatic relief while addressing underlying mechanism translates into better asthma control. Moreover, an argument can be made that bronchochonstriction per se plays a pathogenic role (Gosens & Grainge, 2015; Grainge, et al., 2011), thus beta-agonist-mediated bronchorelaxation also addresses, to some degree, mechanism of disease. For combinations involving two direct bronchodilator drugs (e.g., a muscarinic receptor antagonist plus β2AR agonist such as Combivent), two distinct mechanisms effecting bronchodilation are targeted.
Beyond the concept of addressing multiple disease mechanisms (as beneficial), some studies have also suggested that beta-agonists and corticosteroids cooperate at a cellular/molecular level to augment each’s effect (Giembycz & Newton, 2015; Holden, et al., 2011). In addition to studies demonstrating beta-agonists can induce ligand-independent translocation of the glucocorticoid receptor in cell-based assays (Eickelberg, et al., 1999; Roth, et al., 2002). Usmani et al. (Usmani, et al., 2005) reported that combination therapy (inhaled salmeterol plus fluticasone propionate) in seven asthmatics augmented glucocorticoid receptor translocation in vivo to greater extent than either salmeterol or fluticasone treatment alone. On the flip side, steroids are also believed to enhance responsiveness to beta-agonists by upregulating β2AR expression which has been shown in the lung (Mano, et al., 1979), presumably by GR transactivation of the ADBR gene (Malbon & Hadcock, 1988). Moreover, dexamethasone was shown to inhibit beta-agonist-induced β2AR downregulation in rat lung (Mak, et al., 1995). The Panetierri group also reported that acute (1 hour) pre-treatment of precision cut human lung slices with dexamethasone completely prevented the loss of albuterol-mediated ASM relaxation; an effect not attributed to changes in β2AR expression (Cooper & Panettieri Jr, 2008). Thus, multiple mechanisms appear to exist that cause positive cross-talk between GR and β2AR signaling to create the observed therapeutic cooperativity of beta-agonists and corticosteroids.
Currently several LAMA/LABA combinations including tiotropium/olodaterol, glycopyrronium/indacaterol, umeclidinium/vilanterol and aclidinium/formoterol, as well as the LABA/ICS vilanterol/fluticasone are being evaluated as treatments for COPD (Bateman, et al., 2014; Ramadan, et al., 2015), but currently not for asthma.
A new approach to combination therapy has been the development of bifunctional molecules; two pharmacophores linked covalently so as to form a single molecule with the ability to engage two therapeutic targets (dual pharmacology). One example of such a bifunctional molecule is GS-5759, a molecule with a (long-acting) beta agonist and PDE4 inhibitor moiety, designed for inhaled delivery (Tannheimer, et al., 2014). The compound retains both bronchodilator and anti-inflammatory properties in vitro (Tannheimer, et al., 2014) and in vivo in animals (Salmon, et al., 2014) suggesting a use for this approach as treatment of disease. In addition to the potential benefits of potentiation of beta agonist effectiveness by concomitant inhibition of phosphodiesterases and the resultant increase in cAMP levels (Seldon, et al., 2005; Tannheimer, et al., 2012a; Tannheimer, et al., 2012b), a number of advantages to using bifunctional molecules (as opposed to two separate molecular entities formulated together) has been discussed (Matera, et al., 2011). These include simplified formulation for inhaled delivery, a single pharmacokinetic profile and an easier route to regulatory approval as opposed to two combined pharmacological agents. However, for all these perceived potential benefits of bifunctional agonists, their success will ultimately depend on their effectiveness in patients.
Other bifunctional molecule approaches include the combined muscarinic antagonism/β2-agonism (MABA) activity such as the GSK961081 (Bateman, et al., 2013). These types of compounds would also facilitate the use of triple therapy (LAMA/LABA/ICS) as combining three drugs in one inhaler may be technically challenging and could pose an issue with variable pharmacokinetic profiles of the different components. Tiotropium as an add-on to LABA/ICS has been shown to be more effective than LABA/ICS alone in both COPD (van Noord, et al., 2010), and severe asthma (Kerstjens, et al., 2011; Kerstjens, et al., 2012) indicating an added benefit of triple therapy. An additional benefit of tiotropium as an add-on to LAB/ICS is its potential corticosteroid-sparing effect in severe asthma (Fardon, et al., 2007); lowering the ICS dose lowers the risk of ICS related side-effects.
The philosophy guiding most drug development over the last several decades has focused on improving existing drugs to make them: 1) more selective and possibly more potent for their intended target, with fewer off-target side effects; and 2) longer-acting, which would presumably improve patient adherence. Related to this latter goal are those attempts to improve adherence by making drugs easier to take- for example, by making a drug in an oral formulation patient adherence to their drug regimen is more likely. Yet these goals can be competing, given oral drugs provide systemic delivery which increases the likelihood of off-target effects, whereas although inhaled drugs improve targeting specificity to the lung, effective drug delivery as well as adherence can suffer.
An underlying premise of drug discovery efforts to date is that GPCRs have simplistic linear signaling properties that promote or thwart disease, and that GPCR ligands either turn their target receptor “on” or “off”, that antagonists serve to turn them off, agonists turn them on, and the goal is to make drugs that are best at these functions. This premise is reflected in the high throughput screening strategies for many candidate GPCR ligands, which often entail multi-well (96, 384, or more) plate assays for assessing the ability of candidates to stimulate or inhibit an intracellular signal (Ca2+ or cAMP) in cells (often artificial expression systems) expressing the receptor of interest. Later the “passed” drug candidates are tested in more integrative, physiologically-relevant systems.
The emerging concept of biased agonism, also referred to as functional selectivity or qualitative signaling, suggests that most GPCRs have a much more diverse signaling capacity than previously appreciated, which effectively renders traditional definitions of “agonism” and “antagonism” if not obsolete, certainly oversimplistic. Although many receptors are known to be able to couple to more than one G protein, and thereby stimulate more than one signaling pathway, it is now known that the diversity of GPCR signaling extends much further, and includes the ability to signal through G protein-independent pathways. Due to such pleiotropic capabilities of the receptor, it is clear that the characterization of any ligand or modulator must take into account its ability to regulate the multiple signals produced. Indeed, certain ligands have been shown to engage the same receptor to act as an agonist of one signaling pathway and inverse agonist or antagonist of another. For example, the 5-HT2CR ligand SB 242084 functions as an agonist for 5-HT2cR-mediated PLC activation (Cussac, et al., 2002) yet is an inverse agonist for 5-HT2c R-stimulated arachidonic acid (AA) release by phospholipase A2 (PLA2) (De Deurwaerdere, et al., 2004). In addition, ligands exhibit differences in their ability to induce receptor internalization or desensitization that is dissociated from their signaling efficacy (Urban, et al., 2007).
G protein-independent signaling by GPCRs has been frequently attributed to arrestin proteins. Arrestins, were originally characterized as important regulators of GPCR internalization and desensitization. It is now clear they can also can bind to GPCRs and serve as scaffolds for the binding of various proteins which initiate signaling independent of heterotrimeric G proteins (Lefkowitz & Shenoy, 2005). Numerous studies demonstrate the capacity of ligands to promote arrestin-dependent p42/p44 MAPK activation that is independent of their efficacy in activating G proteins. Azzi et al. (Azzi, et al., 2003) demonstrated that ICI 118,551 and propanolol, which function as inverse agonists for β2AR -mediated Gs activation, are partial agonists for β2AR-mediated, arrestin-dependent p42/p44 MAPK activation. Unlike with the β2AR agonist isoproterenol, ICI 118,551- and propranolol- stimulated p42/p44 MAPK activation occurred in the absence of either Gαs or Gαi. Similar findings of dissociated G protein- and arrestin- dependent signaling for other GPCRs are plentiful (Penn, 2008).
Collectively, these findings suggest that ligands can differ in their ability to induce various receptor conformations that preferentially activate one pathway over another. The diversity with which ligands can induce different receptor conformations linked to different events is portrayed in Wisler et al. (Wisler, et al., 2007), in which an analysis of 16 different βAR ligands revealed a significant range of efficacies for β2AR-mediated Gs- and arrestin- dependent signaling, receptor phosphorylation, and receptor internalization. Carvedilol- traditionally known as a “beta-blocker” used in the treatment of heart failure, demonstrated unique properties in its:1) negative efficacy for Gs-mediated adenylyl cyclase activation; 2) modest ability to promote β2AR phosphorylation and internalization; 3) ability to recruit arrestin; and 4) stimulation of arrestin-dependent, G protein-independent p42/p44 MAPK activity.
If a receptor is capable of diverse signaling events that are associated with distinct physiological consequences, and ligands can “bias” signaling and therefore function, the opportunity exists to “tune” the receptor and optimize the most therapeutically beneficial signaling profile. In heart failure, a belief that Gs/cAMP signaling in cardiac myocytes is pathological while arrestin-dependent signaling is beneficial has been used to explain the efficacy of carvedilol relative to other “beta-blockers”. In asthma, the roles of these two signaling pathways appear reversed. We have demonstrated that the Gs/cAMP/PKA pathway mediates the relaxant effect of beta-agonists on ASM, whereas beta-arrestin 2 serves to not only constrain the efficacy of beta-agonists but is also critical to the development of inflammation induced by allergen (Walker, et al., 2003). These findings suggest that beta-agonists stimulate both pathogenic and therapeutic signals through β2ARs in the lung (Figure 2). A series of studies by the Bond lab supports this notion, and the profile of effects of various beta-agonists suggests that the bias properties of these β2AR ligands influence the asthma phenotype. Early studies demonstrated that either ADBR gene ablation or systemic delivery of the inverse agonist nadolol greatly reduced both allergen-induced inflammation and AHR in murine model, suggesting the sum of beta-agonist effects in the lung is deleterious (Nguyen, et al., 2009). Yet certain “beta-blockers” worked better in preventing the allergen-induced asthma phenotype, and these differences could not be explained by the inverse agonism properties as initially proposed (Wisler, et al., 2007). As the bias properties of numerous β2AR ligands were revealed in multiple studies (Walker, et al., 2011), a more consistent interpretation emerged in which β2AR-mediated arrestin signaling drove pathogenic features (mucus production, inflammation) of asthma and the capacity of any β2AR ligand to stimulate or block arrestin-dependent signaling was critical to its effect. Thus, ligands such as nadolol which blocked epinephrine-stimulated arrestin signaling, and lacked the ability to stimulate arrestin-dependent signaling on their own, were effective in inhibiting allergic lung inflammation despite failing to stimulate (while also blocking epinephrine-stimulated) cAMP/PKA signaling. Alternatively, “beta-blockers” such as carvedilol which can stimulate β2AR-mediated arrestin signaling, failed to antagonize allergic lung inflammation in wild type mice and promoted inflammation and AHR in mice lacking endogenous epinephrine (Thanawala, et al., 2015). Similarly, the failure of propranolol to improve the asthma phenotype in a recent human trial can be attributed to its ability to stimulate arrestin-dependent signaling (Penn, 2014).
The impressive pre-clinical results of nadolol have prompted the current clinical trial sponsored by the National Institute of Allergy and Infectious Diseases ("https://clinicaltrials.gov/ct2/show/NCT01804218?term=NCT01804218&rank=1,") assessing the effect of nadolol on airway hyperresponsiveness, airway inflammation, and asthma control in mild asthmatics. Should nadolol prove effective, the possibility remains that an even more effective biased β2AR ligand (capable of both blocking barrestin-signaling and stimulating canonical Gs/cAMP/PKA signaling) could emerge. Alternatively, a therapeutic approach designed to skew signaling toward the Gs/cAMP/PKA pathway (e.g., augmenting cAMP/PKA with PDE4 inhibitors while maintaining inhibition of arrestin signaling with nadolol) also represents a potentially superior strategy.
Of course, the application of biased ligand pharmacology to asthma is not limited to β2AR ligands. As noted above, PAR2 signaling in the airway can be parsed into (beneficial) G protein-dependent and (pathogenic) arrestin-dependent signaling and functional consequences. Strategies to bias the signaling of PAR2, be they biased orthosteric ligands or allosteric modulators, have the potential to overcome past difficulties in targeting this receptor for asthma.
For a more in-depth analyses of functional selectivity and its impact on models of GPCR activation and drug discovery, the reader is referred to Penn (Penn, et al., 2014) and references therein.
Forces impacting asthma drug development and drug approval/application today include not only the research findings in asthma, airway cell biology, structural biology, and pharmacology, but also the economic forces that dictate priorities and areas of investment.
Although the research detailed in this review suggests multiple, exciting opportunities for new asthma drugs, the reality is that it is extremely difficult to advance any conceptually appealing and empirically supported therapy into an approved and used asthma drug. Given the current significant cost of new drug development as well as regulatory hurdles in getting any new drug approved (Barnes, et al., 2015), it is more likely that drugs that represent modest variations of existing drugs, or new combinations of existing drugs, are the most likely to supplant those drugs used currently. The reduced cost of development and preclinical/clinical validation of repurposed drugs, combined with less onerous regulatory hurdles, also renders repurposed drugs more likely to advance to the market. Indeed, this reality likely prompted recent programmatic efforts of the NIH to employ a crowdsourcing strategy to establish collaboration between industry and academia supported by specific funding mechanism for research into repurposed drugs and their application (Colvis & Austin, 2014), in addition to the past and ongoing efforts of the National Center for Advancing Translational Sciences (https://ncats.nih.gov).
However, one would hope that basic, preclinical, and clinical research efforts will prevail despite current funding levels and economic forces, and accurately identify those drugs that are either superior, or at least non-inferior, to currently prescribed asthma drugs. Layer in the global interest in personalized medicine and existing asthma clinical network efforts to categorize specific asthma populations based on their underlying pathology (Levy, et al., 2015), and it is not difficult to envision an expanded repertoire of asthma drugs, or drug combinations tailored to better manage specific asthmatic subpopulations. Perhaps most encouraging (for the treatment of all diseases) is the promise of biased ligand pharmacology, which now has a strong foundation built on multiple basic science disciplines and will likely find wide spread clinical application once industry expands its platforms and strategies to accommodate the reality that ligands have much greater capacity to manipulate GPCRs than previously thought.
The authors would like to thank Terence Murphy and Brian Tiegs for their assistance with manuscript preparation. The work in Dr. Penn’s lab is supported by NIH grants R01 HL58506, R01 AI110007 and P01 HL114471.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest statement
The authors declare that there are no conflicts of interest.