PGE2 receptors in detrusor muscle: Drugging the undruggable for urgency

Abstract

Overactive bladder (OAB) syndrome is a prevalent condition of the lower urinary tract that causes symptoms, such as urinary frequency, urinary urgency, urge incontinence, and nocturia, and disproportionately affects women and the elderly. Current medications for OAB merely provide symptomatic relief with considerable limitations, as they are no more than moderately effective, not to mention that they may cause substantial adverse effects. Identifying novel molecular targets to facilitate the development of new medical therapies with higher efficacy and safety for OAB is in an urgent unmet need. Although the molecular mechanisms underlying the pathophysiology of OAB largely remain elusive and are likely multifactorial, mounting evidence from preclinical studies over the past decade reveals that the pro-inflammatory pathways engaging cyclooxygenases and their prostanoid products, particularly the prostaglandin E2 (PGE₂), may play essential roles in the progression of OAB. The goals of this review are to summarize recent progresses in our knowledge on the pathogenic roles of PGE₂ in the OAB and to provide new mechanistic insights into the signaling pathways transduced by its four G-protein-coupled receptors (GPCRs), i.e., EP1–EP4, in the overactive detrusor smooth muscle. We also discuss the feasibility of targeting these GPCRs as an emerging strategy to treat OAB with better therapeutic specificity than the current medications.

Graphical Abstract

1. Introduction

The overactive bladder (OAB) syndrome is commonly featured by the detrusor overactivity and constitutes the most prevalent type of the lower urinary tract dysfunction (LUTD). According to the International Urogynecology Association (IUGA) and the International Continence Society (ICS), OAB syndrome is defined as the urinary urgency that is accompanied by frequency and nocturia, either with (wet-OAB) or without (dry-OAB) urgency urinary incontinence, and in the absence of urinary tract infection or any other palpable urinary pathology [1]. As a chronic medical condition, OAB causes a sudden and frequent desire to urinate that is often difficult to control, imposing a major impact on the quality of life in a large proportion of the population. Depending on the study, the OAB condition has been diagnosed in up to 27% of men and 43% of women worldwide [2], and women are more likely to suffer an urgent urinary incontinence [3]. Moreover, the older populations are more prone to OAB symptoms, and the number of patients with OAB dramatically increases with age. More than 40% of patients with OAB have incontinence, and most urinary incontinence is solely caused by OAB. Though it is not life-threatening, OAB negatively affects people’s personal life and causes enormous economic burdens, as the average per capital cost and the total national cost for OAB per year are projected to be $1969 and $82.6 billion, respectively, by 2020 in the US alone [4].

The recommended first-line nonpharmacologic therapies for OAB symptoms are various behavioral interventions, such as lifestyle modification, training in voiding habits, and pelvic floor muscle exercise. Pharmacotherapy is often initiated as an adjuvant to optimize the overall outcomes when conservative management fails to achieve adequate improvement [2]. A number of antimuscarinic drugs and β3 adrenergic receptor agonists are currently available to treat OAB as the second-line treatment. In particularly, the β3 agonist drugs are increasingly prescribed largely owing to their fewer side effects [5], although their primary mechanisms of action are not fully understood yet [6]. Intravesical injection of onabotulinumtoxinA is considered to be safer and more effective as a third-line treatment for adults with refractory OAB who do not sufficiently respond to or cannot tolerate the second-line medications [2]. However, very few OAB patients can gain complete relief with these current medications, as they are no more than moderately effective in the vast majority of patients. In addition, all these anti-OAB drugs can have moderate to severe side effects: antimuscarinic drugs commonly cause dry mouth, constipation, and dyspepsia; β3 adrenergic receptor agonists lead to hypertension; the most common complications after onabotulinumtoxinA injection are urinary tract infection and retention [7]. Given the significant economic impacts of the disease and the substantial limitations of current medications, it is an urgent unmet need to identify new molecular targets and develop safer and more efficient therapies for OAB.

2. COX/PGE₂ cascade in OAB

The etiology of OAB symptoms is not fully understood but is believed be associated with overactivity of the detrusor urinae muscle, which might relate to a variety of contributory factors, such as female sex, neurologic diseases, aging, bladder outlet obstruction, metabolic diseases [8, 9]. Several recent studies have identified signs of chronic inflammation in urine specimens and biopsied bladder tissues from the OAB patients without urinary tract infection [10, 11], insinuating a pathogenic role for inflammation in OAB. The upregulated inflammatory molecules found in patients with OAB include C-reactive protein (CRP), various cytokines, chemokines, and prostanoids [12, 13]. As a key pro-inflammatory mediator, prostaglandin E2 (PGE₂) in the bladder is primarily synthesized in the urothelial and muscle layers and exerts local actions such as the micturition reflex under normal physiological and pathophysiological conditions [14]. Elevated PGE₂ levels have been found in experimental animals with overactive detrusor [15, 16] as well as in patients with OAB [17, 18]. Intravesical administration of PGE₂ led to urinary symptoms and detrusor overactivity in human patients and experimental animals [19–21], indicative of the involvement of PGE₂ in the regulation of urinary smooth muscle function.

PGE₂ is synthesized through the oxidization of arachidonic acid in a two-step reaction. During the first step, arachidonic acid from the membrane-bound phospholipids is converted by cyclooxygenase (COX) to intermediate prostaglandin G2 (PGG₂) and then PGH₂ (Fig. 1). As the second step, the short-lived PGH₂ is rapidly converted to five different prostanoids, consisting of prostaglandins PGE₂, PGD₂, and PGF_2α, thromboxane TXA₂, and prostacyclin PGI₂ by tissue-specific prostanoid synthases. Prostanoids exert their physiological and pathological functions through acting on a panel of G protein-coupled receptors (GPCRs), among which are four receptors bound and activated by PGE₂, namely, EP1, EP2, EP3, and EP4 (Fig. 1). Receptor-coupled G proteins are heterotrimers consisting of α, β, and γ subunits, among which the Gα subunit plays a dominant role in the hydrolytic function of G proteins [22]. In bladder detrusor muscle, EP2, EP4 and β3-adrenergic receptors are coupled to Gα_s subunit that stimulates the cAMP signaling; M3 muscarinic and EP1 receptors are associated with Gα_q that can lead to the calcium-calmodulin and protein kinase C (PKC)-mediated pathways; EP3 mainly links with Gα_i to exert an inhibitory effect on the cAMP generation (Fig. 1).

Figure 1. PGE₂ biosynthesis and signaling.

The biosynthesis of PGE₂ begins from the cleavage of membrane-bound phospholipid into arachidonic acid, which is catalyzed by phospholipase A2 (PLA2). Subsequently, the arachidonic acid is oxidized by cyclooxygenase-1 or 2 (COX-1/2) to intermediate prostaglandin G2 (PGG₂) and then PGH₂. The short-lived PGH₂ is rapidly converted to PGE₂ by tissue-specific PGE synthase (PGES). PGE₂ exerts its physiological and pathological functions through acting on a panel of G protein-coupled receptors (GPCRs), EP1, EP2, EP3, and EP4. In bladder detrusor muscle, the activation of EP1 and EP3 are likely to induce muscle contraction, while the activation of EP4 is likely to cause muscle relaxation. The function of EP2 in the urinary bladder needs further investigation.

COX has two isozymes: constitutive COX-1 and inducible COX-2, both of which are found in human urothelium and detrusor muscle of the bladder (COX-1: https://www.ncbi.nlm.nih.gov/gene/5742; COX-2: https://www.ncbi.nlm.nih.gov/gene/5743) [23, 24]. The constitutional expression of COX-1 is responsible for the biosynthesis of prostanoids to maintain the homeostasis within the normal bladder, whereas the COX-2 can be quickly induced by various detrimental or pro-inflammatory stimuli under pathophysiological conditions [25]. Because the COX-2 induction can trigger long-lasting pro-inflammatory processes that might facilitate the disease progression, a number of selective and non-selective COX inhibitors have been tested to determine whether blocking the COX enzymes can decrease the spontaneous contractile activities of the bladder in various animal models of OAB.

Intravesical instillation of aspirin in rabbit bladders that were partially obstructed attenuated the contractility caused by the cholinergic stimulation with KCl [26]. Several other non-selective COX inhibitors, such as dexketoprofen, indomethacin, meloxicam, naproxen, paracetamol, flurbiprofen, and nimesulide, as well as COX-2 selective inhibitors including NS-398, celecoxib, rofecoxib, and L-745337, were investigated in rat urodynamic models. The results from these studies overall indicated that the COX-1 may be involved in modulating the activation threshold of the micturition reflex under normal conditions, whereas the COX-2 inhibition can prevent or reverse the urodynamic alterations triggered by bladder inflammation [27, 28]. In a clinical study involving 32 patients with urinary symptoms due to detrusor instability, the administration of indomethacin for one month was demonstrated to achieve a highly significant improvement in diurnal and nocturnal frequency [29]. Likewise, intravesical administration of ketoprofen, another non-selective COX inhibitor, for four weeks has been shown to improve the OAB symptom in 18 out of 30 female patients without any observatory side effects [30]. The success in these preclinical and clinical studies suggest that the COX inhibition might represent a promising strategy to treat OAB. However, the well-recognized severe adverse effects on gastrointestinal and cardiovascular systems by the chronic use of COX inhibitors indicate that certain COX prostanoid products might mediate some beneficial effects [31]. As such, targeting a key pro-inflammatory prostanoid receptor might represent a more specific therapeutic strategy compared to generic block of the entire COX cascade [32, 33].

3. PGE₂ receptors and OAB

COX enzymes execute pro-inflammatory effects mainly through synthesizing PGE₂. Upon COX-2 induction, the elevated PGE₂ binds and activates four receptors EP1–EP4, which have divergent engagement in phosphoinositol and cAMP-dependent pathways (Fig. 2) [34, 35]. These membrane-bound receptors are differentially expressed in the lower urinary system (EP1: https://gtexportal.org/home/gene/PTGER1; EP2: https://gtexportal.org/home/gene/PTGER2; EP3: https://gtexportal.org/home/gene/PTGER3; EP4: https://gtexportal.org/home/gene/PTGER4), and play important roles in the maintenance of normal bladder function as well as the pathogenesis of bladder dysfunctions [36]. Given that modulating PGE₂ signaling might provide better therapeutic specificity than blocking PGE₂ synthesis by COX inhibition, over the past decade these four EP receptors have been under intense investigation as potential molecular targets for new treatments of OAB. In the rest of this review, we discuss their contributions to the pathophysiology of detrusor overactivity and examine the feasibility of pharmacologically targeting these GPCRs for new pharmacotherapies of OAB (Tables 1–4).

Figure 2. PGE₂ signaling that regulates the activities of detrusor smooth muscle.

EP1 receptor couples to Gα_q and its activation stimulates phospholipase C (PLC) that hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP₂) into diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP₃). DAG further activates protein kinase C (PKC) which induces the Ca²⁺ influx current through the voltage dependent Ca²⁺ channel (VDCC). Meanwhile, PKC may also cause a direct inhibition of the ryanodine receptor (RyR) or a reduction in the sarcoplasmic reticulum (SR) Ca²⁺ load, resulting in an inhibition of the big potassium channel (BK channel) of the detrusor smooth muscle. IP₃ elevates the intracellular Ca²⁺ levels through the interaction with IP₃ sensitive Ca²⁺ channel at the SR. The intracellular Ca²⁺ of the detrusor smooth muscle binds to the calmodulin to form a calcium-calmodulin complex that activates myosin light chain kinase (MLCK). MLCK phosphorylates myosin to enable its interaction with the actin for the initiation of cross-bridge cycling and muscle contraction. Both EP2 and EP4 receptors couple to Gα_s subunit which activates the adenylate cyclase (AC) to synthesize cAMP. cAMP in turn activates protein kinase A (PKA) that subsequently inhibits the VDCC but activates Na⁺-Ca²⁺ exchanger (NCX) on the cell membrane. Activated PKA also phosphorylates both IP₃ sensitive Ca²⁺ channel and Ca²⁺-ATPase on the SR to cause Ca²⁺ sequestration, and thus downregulates the intracellular Ca²⁺ levels. PKA might also activate RyR, leading to the Ca²⁺ releasing to form Ca²⁺ sparks, which in turn activate the BK channel. Upon activation, the transient BK current leads to the hyperpolarization of the cell membrane which closes the VDCC. EP3 receptor mainly couples to Gα_i for an inhibitory effect on adenylate cyclase, thereby downregulating the intracellular cAMP levels to increase Ca²⁺ levels and facilitate the muscle contraction. Only the major pathways are indicated.

Table 1.

Recent studies on the EP1 receptor in OAB models

Drug	Drug action	Experimental model	Drug dose and route	Main therapeutic outcomes	Reference
N/A	N/A	Moderate BOO induced in female adult DBA/1LacJ mice	Genetic ablation of EP1 receptor	• Did not change the urodynamic parameters or bladder hypertrophy. • Prevented the intravesical PGE2-induced detrusor overactivity.	Schröder et al., 2004
ONO-8711	Antagonist	Acetic acid (0.1%)-induced inflammation of the urinary bladders in adult female Wistar rats	1 and 3 mg/kg, i.v.	• Reduced the inflammation-associated bladder afferent neuronal activity.	Ikeda et al., 2006
PF-2907617-02	Antagonist	Normal adult female Sprague-Dawley rats	0.1 and 1 mg/kg, i.v.	• PF-2907617-02 (1 mg/kg) increased bladder capacity, micturition volume and micturition interval but had no effect on other urodynamic parameters in normal rats.	Lee et al., 2007
PF-2907617-02	Antagonist	Moderate BOO induced in adult female Sprague-Dawley rats	1 mg/kg, i.v.	• Increased micturition interval and decreased the frequency and amplitude of nonvoiding contraction in rats with BOO. • Had no effect on bladder capacity or residual volume in BOO rats.	Lee et al., 2007
PF-2907617-02	Antagonist	Intravesical PGE2 (50 µM)-induced detrusor overactivity in adult female Sprague-Dawley rats	0.1 and 1 mg/kg, i.v.	• Decreased the intravesical PGE2 stimulated detrusor overactivity in normal rats.	Lee et al., 2007
PF-2907617	Antagonist	Bladder strips from female adult Sprague Dawley rats, male cynomolgus macaque monkeys	300 nM, bath	• Caused an inhibitory effect on the PGE2-induced contraction of bladder strips from rats, but not of those from monkeys or humans.	Root et al., 2015
ONO-8713	Antagonist	Male guinea pig urinary bladder	500 nM, bath	• Decreased the frequency of muscarinic agonist arecaidine-induced bladder contraction.	Hohnen et al., 2016

Abbreviations: BOO, bladder outlet obstruction; i.v., intravenous; OAB, overactive bladder; PGE₂, prostaglandin E2.

Table 4.

Recent OAB studies on the PGE₂ receptors in humans

Drug	Drug action	Experimental model	Drug dose and route	Main therapeutic outcomes	Reference
ONO-8539	EP1 Antagonist	435 OAB patients in a 12-week, randomized, double-blind, placebo controlled, parallel group study	30, 100 or 300 mg, twice daily	• No significant improvement compared to the placebo.	Chapple et al., 2014
PF-2907617	EP1 Antagonist	Bladder strips from male and female humans (age: 47–77 years)	300 nM, bath	• Caused an inhibitory effect on the PGE2-induced contraction of bladder strips from rats, but not of those from monkeys or humans.	Root et al., 2015
Sulprostone	EP3 Agonist	Bladder strips from male and female humans (age: 47–77 years)	300 nM, bath	• Induced potent contraction of rat bladder muscles. • Caused weak contraction in monkey and human bladder strips.	Root et al., 2015
CJ24979	EP3 Antagonist	Bladder strips from male and female humans (age: 47–77 years)	1 µM, bath	• Inhibited the PGE2-induced contraction of bladder strips from rats and monkeys, but not those from humans.	Root et al., 2015

Abbreviations: OAB, overactive bladder; PGE₂, prostaglandin E2.

3.1. EP1 receptor

Among the four EP receptors, the expression of EP1 in the mouse bladder is the highest (https://www.ncbi.nlm.nih.gov/gene/19216), suggesting some physiological functions in the lower urinary system. EP1 receptor couples to Gα_q – upon activation by PGE₂ – to stimulate phospholipase C (PLC) that hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP₂) into diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP₃) [37]. DAG in turn directly activates some isoforms of protein kinase C (PKC) that can induce the Ca²⁺ influx current through the voltage dependent Ca²⁺ channels (VDCC) (Fig. 2) [38, 39]. In the meantime, PKC may also cause a direct inhibition of the ryanodine receptors (RyRs) or a reduction in the sarcoplasmic reticulum Ca²⁺ load, resulting in an inhibition of the large conductance calcium-activated potassium channel (or big potassium, BK) channels of the detrusor smooth muscle [39]. In fact, it has been suggested that the outward spontaneous transient BK currents (TBKCs) from the activated BK channels can induce the hyperpolarization of the cell membrane and detrusor muscle relaxation [40], where a direct action of estrogen on the BK channels might be involved [41]. It is worth noting that the involvement of female gonadal hormone such as estrogen and progesterone in the regulation of BK channels might contribute to sex difference in the prevalent of urgent urinary incontinence [42–44]. As the other second messenger of Gα_q, IP₃ elevates the cytoplasmic Ca²⁺ level via activating its receptors in the sarcoplasmic reticulum that function as Ca²⁺ channels [45]. The cytoplasmic Ca²⁺ of the detrusor smooth muscle binds to the calmodulin to form a Ca²⁺-calmodulin complex that activates myosin light chain kinase (MLCK). The latter then phosphorylates myosin to enable its interaction with the actin for the initiation of cross-bridge cycling and muscle contraction (Fig. 2) [46]. In addition to its presence in the detrusor muscle and urothelium, EP1 receptor has been shown in the intramural ganglia cells of the bladder [47], suggesting its involvement of neurologic regulation of detrusor muscle contraction [48].

The genetic deletion of EP1 receptor prevented detrusor overactivity caused by PGE₂ stimulation or outlet obstruction in mice (Table 1) [49]. ONO-8711, a selective EP1 antagonist, was shown to decrease the afferent nerve activity in the rat bladder, indicating that the EP1 receptor might play an important role in the inflammation-induced primary afferent never activity that elicits the micturition reflex [50]. Similarly, intravenous administration of EP1 antagonist PF-2907617–02 increased bladder capacity, micturition volume, and micturition interval, and decreased the frequency and amplitude of non-voiding contraction in rats with partial bladder outlet obstruction (BOO) or PGE₂-induced OAB [20]. Ex vivo studies using bladder stripes also suggested that the EP1 inhibition by PF-2907617-02 or ONO-8713 was able to reduce the bladder contraction frequency in rats and guinea pigs [48, 51]. Despite the success in preclinical studies, a randomized, double-blind, placebo controlled phase II clinical trial showed that daily administration of an EP1 antagonist ONO-8539 for 12 weeks did not significantly improve the micturition behavior in human patients with nonneurogenic overactive bladder syndrome (https://clinicaltrials.gov/ct2/show/NCT00876421) (Table 4) [52].

3.2. EP2 receptor

Both EP2 and EP4 receptors couple to Gα_s subunit that activates the adenylate cyclase to generate cAMP, which has been shown to play essential roles in human detrusor smooth muscle contraction via regulating the calcium sensitization [53]. However, it has been suggested that the functional coupling to cAMP is more efficient for EP2 receptor than EP4 [35]. cAMP directly activates protein kinase A (PKA), which subsequently inhibits the voltage-dependent Ca²⁺ channel but activates the Na⁺-Ca²⁺ exchanger (NCX) on the cell membrane. Furthermore, the activated PKA also lead to the phosphorylation of both IP₃ sensitive Ca²⁺ channel and Ca²⁺-ATPase on the SR of the detrusor smooth muscle cells to cause Ca²⁺ sequestration. These combined effects by PKA cause a downregulation of the intracellular Ca²⁺ levels (Fig. 2) [54, 55]. In addition, PKA presumably can stimulate the RyRs on the sarcoplasmic reticulum to release Ca²⁺ sparks and activate the BK channels, thereby inducing the hyperpolarization of urinary bladder smooth muscle cells, which facilitates detrusor muscle relaxation [40, 56].

However, there is very little information about the role of EP2 in the physiological process of micturition, even though the expression of the receptor has been demonstrated on the urethra, detrusor muscle and interstitial ganglia cells in the bladder of guinea pigs [47, 57, 58] as well as in the adult mouse bladder (https://www.ncbi.nlm.nih.gov/gene/19217). The oral administration of ONO-8055, a dual EP2/EP3 agonist, decreased post-void residual urine, voiding pressure, bladder capacity, and urethral pressure in rats with neurogenic underactive bladder (UAB) caused by the lumbar spinal canal stenosis [59], but did not augment the urethral relaxation or bladder contractility in normal rats [60]. Interestingly, ONO-8055 increased the voided volume, average flow rate, and maximum flow rate in monkeys with underactive bladder induced by radical hysterectomy [61]. These studies together demonstrated the therapeutic potential of ONO-8055 to treat neurogenic underactive bladder syndrome. However, whether EP2 or EP3 is the therapeutic target of the compound remains to be determined, as the compound shows similar potency on EP2 and EP3 (EC₅₀: 0.67 vs. 0.7 nM) [62]. Up to date, no selective EP2 antagonist has been tested in animal models of OAB.

3.3. EP3 receptor

Although the EP3 receptor primarily couples to G_i for an inhibitory effect on adenylate cyclase to downregulate the intracellular cAMP level, it has several alternative splice variants that differ in their intracellular C-terminus and are associated with multiple different G proteins [63]. In humans, in addition to G_i, the EP3-II and EP3-IV isoforms also couple to G_s, whereas EP3-I and EP3-II couple to G_q. Likewise, there are three EP3 isoforms in mice – α, β, and γ. The EP3γ isoform appears to also couple to G_s, and all these three isoforms couple to G_12/13 to activate Rho pathway for smooth muscle contraction. In consequence, the G protein complexes associate with EP3, upon the receptor activation by PGE₂, dissociate into Gα_i, Gα_12/13, Gα_s, and Gβγ components that proceed to activate multiple signaling pathways depending on the cellular context. Thus, a differential isoform expression pattern of the receptor can provide another level of regulation of the EP3-mediated signaling in different types of cells and tissues. The expression of EP3 in the bladder is relatively low when compared to other organs and tissues, such as adrenal gland, colon, kidney, mammary gland, ovary, placenta, small intestine, and fat pad (https://www.ncbi.nlm.nih.gov/gene/19218).

The genetic ablation of EP3 in mice improved the bladder capacity by approximately 185% and largely blunted the PGE₂-induced bladder hyperactivity, whereas intravesical infusion of the selective EP3 receptor agonist GR63799X considerably reduced the bladder capacity in wild-type mice (Table 2) [64]. In line, EP3 receptor activation by sulprostone induced potent contraction of rat bladder muscles, whereas selective EP3 antagonist CJ24979 antagonized the PGE₂-induced contraction of bladder strips (Tables 2 and 4) [48]. In addition to its expression in the urothelium and bladder detrusor muscle, EP3 has been found in the interstitial cells of Cajal (ICC) of the bladder, where its activation by PGE₂ might contribute to the bladder detrusor contraction via activating the hyperpolarization-activated cyclic nucleotide-gated (HCN) channels and Ca²⁺ influx [65]. It is likely that EP3 receptor activates the HCN channels via altering the intracellular cAMP levels, generating an inward Na⁺ current that leads to the membrane depolarization and activation of T-type voltage-dependent Calcium channels [65, 66]. Intravenous administration of DG-041, a peripherally restricted EP3-selective antagonist, decreased the frequency of bladder rhythmic contraction and prevented the visceromotor reflex (VMR) response to bladder distension in rats [67]. Intraduodenal administration of selective EP3 agonist GR63799X reduced, while EP3 antagonists DG-041 and CM-9 enhanced, the bladder capacity in conscious rats, evaluated by assessing the micturition interval and volume of urine per void [68]. Similar to the EP4, the protein expression of EP3 in the bladder was induced by nearly three-fold in encephalomyelitis mice with OAB symptoms, and the intravenous injection of EP3 antagonist DG-041 was shown to decrease the micturition frequency per day and increase the void weight per void in these animals [69].

Table 2.

Recent studies on the EP3 receptor in OAB models

Drug	Drug action	Experimental model	Drug dose and route	Main therapeutic outcomes	Reference
N/A	N/A	PGE2-induced OAB in adult 129P3/J or B6 mice	Genetic ablation of EP3 receptor	• Enhanced bladder capacity under control conditions. • Blunted the intravesical PGE2 induced bladder overactivity.	McCafferty et al., 2008
GR63799X	Agonist	PGE2-induced OAB in adult 129P3/J or B6 mice	10 µM, intravesical infusion for 120 min	• Reduced bladder capacity under control conditions in wild-type but not in EP3 knock-out mice.	McCafferty et al., 2008
DG-041	Antagonist	Adult female Sprague-Dawley rat bladder rhythmic contraction model and bladder pain model	10 mg/kg, i.v.	• Reduced the frequency of bladder rhythmic contraction. • Inhibited the VMR response to bladder distension.	Su et al., 2008
GR63799X	Agonist	Female spontaneously hypertensive rats	0.001–1 mg/kg, i.d.	• Reduced bladder capacity. • Reduced micturition interval and void volume.	Jugus et al., 2009
CM-9	Antagonist	Female spontaneously hypertensive rats	10 and 30 mg/kg, i.d.	• Enhanced bladder capacity. • Increased micturition interval and void volume.	Jugus et al., 2009
DG-041	Antagonist	Female spontaneously hypertensive rats	30 mg/kg, i.d.	• Enhanced bladder capacity. • Increased micturition interval and void volume.	Jugus et al., 2009
Sulprostone	Agonist	Neurologic bladder model induced by encephalomyelitis in adult female SWXJ mice	100 µg/kg, i.v.	• Increased micturition frequency and decreased urine output per micturition over a 24-hour test period.	Xue et al., 2013
DG-041	Antagonist	Neurologic bladder model induced by encephalomyelitis in adult female SWXJ mice	100 µg/kg, i.v.	• Decreased micturition frequency and increased urine output per micturition over a 24-hr test period.	Xue et al., 2013
Sulprostone	Agonist	Bladder strips from female adult Sprague Dawley rats, male cynomolgus macaque monkeys	300 nM, bath	• Induced potent contraction of rat bladder muscles. • Caused weak contraction in monkey and human bladder strips.	Root et al., 2015
CJ24979	Antagonist	Bladder strips from female adult Sprague Dawley rats, male cynomolgus macaque monkeys	1 µM, bath	• Inhibited the PGE2-induced contraction of bladder strips from rats and monkeys, but not those from humans.	Root et al., 2015
Sulprostone	Agonist	Bladder detrusor strips from adult female C57BL/6 mice and Sprague-Dawley rats.	10 μM, bath	• Increased the calcium concentration of the ICCs. • Enhanced the contraction amplitude of the bladder detrusor.	Wu et al., 2017

Abbreviations: ICCs, interstitial cells of Cajal; i.d., intraduodenal; i.v., intravenous; OAB, overactive bladder; PGE₂, prostaglandin E2; VMR, visceromotor reflex.

3.4. EP4 receptor

As the other PGE₂ receptor coupled to Gα_s, EP4 is also expressed in the normal adult bladder at a relatively low basal lever (https://www.ncbi.nlm.nih.gov/gene/19219) but dramatically increases in obstructed bladder detrusor smooth muscle and epithelium in rats (Table 3) [70]. A selective EP4 agonist ONO-AE1–329 was able to relax KCl-induced contraction of bladder strips from rats with bladder outlet obstruction (BOO). Further, the intravesical infusion of ONO-AE1–329 significantly increased bladder capacity without altering the micturition pressure in rats with bladder outlet obstruction, whereas it had no effect in control animals [70]. In another interesting study, EP4 expression was elevated by nearly 100% in the detrusor muscle of OAB rats induced by cyclophosphamide (CYP). However, AH23848, a selective EP4 antagonist – when given intravenously – significantly extended the intercontraction interval (ICI) in CYP-treated rats without affecting the normal cohort [71]. Similarly, intravenous injection of AH23848 was also shown to decrease the micturition frequency and improve the OAB symptoms in encephalomyelitis mice, where EP4 expression was also elevated in the OAB bladder [69]. Intravenous or intravesical administration of MF-191, a more potent and selective EP4 antagonist, suppressed the CYP-induced OAB and inflammatory cell infiltration in the detrusor muscle of rats [72]. Interestingly, formalin-induced prostatic inflammation also elevated the PGE₂ level and the expression of EP4, but not other EP receptors, in both the bladder mucosa and detrusor layers. Further, intravesical application of selective EP4 antagonist ONO-AE3–208 increased the intercontraction interval in rats with prostatic inflammation but did not affect the control animals [73]. Taken together, these findings suggest that the EP4 expression is generally induced in OAB detrusor muscle and epithelium, and the EP4 activation overall contributes to the inflammation-associated OAB but might counteract against the BOO-induced OAB. The conflicting outcomes might be explained by the specificity of each OAB model and different compounds (agonists vs. antagonists) used in these studies. However, more study using genetic approach is required to fully understand the roles of EP4 receptor in OAB and might help to clarify this contradiction.

Table 3.

Recent studies on the EP4 receptor in OAB models

Drug	Drug action	Experimental model	Drug dose and route	Main therapeutic outcomes	Reference
AH-23848	Antagonist	CYP-induced OAB in adult female Sprague-Dawley rats	0.01 and 0.1 mg/kg, i.v.	• Extended intercontraction interval. • Had no effect on other urodynamic variables such as baseline pressure, pressure threshold, and contraction amplitude. • Had no significant effect in control animals.	Chuang et al., 2010
OAO-AE1–329	Agonist	Bladder strips from rats with BOO	, bath	• Relaxed KCl-induced contraction.	Beppu et al., 2011
OAO-AE1–329	Agonist	Rats with BOO	10 µM, continuously intravesical infusion	• Increased bladder capacity without altering micturition pressure. • No effect in control animals.	Beppu et al., 2011
MF-191	Antagonist	CYP-induced OAB in adult female Sprague-Dawley rats	0.1 and 1 mg/kg, i.v.	• MF-191 (only 1 mg/kg) improved intercontraction interval and reduced inflammatory cell infiltration.	Chuang et al., 2012
MF-191	Antagonist	CYP-induced OAB in adult female Sprague-Dawley rats	10 and 100 nM, continuously intravesical infusion	• MF-191 (100 nM) suppressed CYP-induced bladder overactivity.	Chuang et al., 2012
MF-191	Antagonist	Intravesical PGE2-induced bladder overactivity in rats	0.1 and 1 mg/kg, i.v.	• MF-191 (1 mg/kg) improved intercontraction interval but had no effect on baseline pressure, pressure threshold, and contraction amplitude.	Chuang et al., 2012
CAY-10598	Agonist	Neurologic bladder model induced by encephalomyelitis in adult female SWXJ mice	100 µg/kg, i.v.	• Increased micturition frequency and decreased urine output per micturition over a 24-hour test period.	Xue et al., 2013
AH-23848	Antagonist	Neurologic bladder model induced by encephalomyelitis in adult female SWXJ mice	100 µg/kg, i.v.	• Decreased micturition frequency and increased urine output per micturition over a 24-hour test period.	Xue et al., 2013
ONO-AE3–208	Antagonist	Prostatitis induced OAB in male adult Sprague-Dawley rats	30 µM, intravesical infusion	• Increased intercontraction interval. • Did not have any significant effect in control rats.	Mizoguchi et al., 2019

Abbreviations: BOO, bladder outlet obstruction; CYP, cyclophosphamide; i.v., intravenous; KCl, potassium chloride; OAB, overactive bladder; PGE₂, prostaglandin E2.

4. Conclusive remarks and perspectives

Though widely considered as a benign condition, OAB negatively affects people’s personal life and imposes enormous economic burdens on individuals, families, and the society. The current behavioral treatment combined with pharmacotherapy have considerable limitations, as they are merely able to provide moderate efficacy for only a small proportion of patients. Daily use of β3 adrenergic receptor agonists and antimuscarinic drugs for OAB substantially increases the risk of moderate to severe side effects such as dry mouth, constipation, hypertension, and even cardiovascular abnormalities, which further complicate the urinary conditions. Recent evidence from animal studies suggests that PGE₂ via its four GPCRs plays essential roles during the pathogenesis of OAB in various models (Tables 1–4). With interest waning in the use of selective and non-selective COX inhibitors for inflammation-associated diseases due to their well-recognized toxicities to the gastrointestinal and cardiovascular systems, it has been widely proposed that targeting these PGE₂ receptors might provide novel therapeutic strategies to treat OAB [74]. Given the fact that the majority of the EP-based OAB therapies were only tested in animal models, the current knowledge about their safety in humans is very limited. A recent phase II clinical study on EP1 selective antagonist ONO-8539 for OAB treatment suggests that it can cause slightly lower incidence of adverse effects when compared to antimuscarinic drug tolterodine (43% vs. 46.6%) [52]. Nevertheless, future studies are required to determine the safety of these PGE₂ receptors-targeting compounds in comparison with the current antimuscarinic drugs and β3 adrenergic receptor agonists.

4.1. The case of EP2

Among the four PGE₂ receptor subtypes, the Gα_s-coupled EP2 has been understudied in animal models of bladder dysfunction largely in that the selective antagonists for this receptor were not available until recently [75]. Using the poorly selective EP antagonist AH-6809 or EP2/EP3 dual agonist ONO-8055 simply cannot provide conclusive evidence to determine whether the EP2 receptor is involved in the pathophysiology of either underactive or overactive bladders [59, 62, 69]. Selective EP2 agonists such as butaprost and ONO-AE1–259 are also not much useful for in vivo studies due to the lack of stability of their prostanoid-like structures [75]. Considering its role in a number of chronic inflammation-associated conditions both in the CNS and periphery [32] as well as its nonnegligent expression in multiple sites of adult bladders from diverse species (https://www.ncbi.nlm.nih.gov/gene/19217) [47, 57, 58], the G_s-coupled EP2 might also contribute to induced PGE₂-promoted OAB, where elevated cAMP engaging PKA signaling mediates excitatory cellular response and facilitates the detrusor contraction (Fig. 2). Future studies using highly potent bioavailable selective EP2 antagonists, such as TG4-155, TG6-10-1, and TG6-129 [76, 77], in preclinical models of OAB are needed to determine this possibility (Fig. 3). In addition, the EP2-selective positive allosteric modulators including TG3-95-1 and TG6-270 might also provide useful tools to determine the role of EP2 in detrusor muscle; they enhance the potency of PGE₂ on EP2 receptor in various pathological inflamed tissues but usually do not alter the receptor activity in the absence of PGE₂ under normal physiological conditions [78, 79]. These chemical probes have diverse pharmacokinetic profiles for in vivo uses [80, 81], and were proven very useful tools to determine the pathophysiological roles of EP2 in a number of inflammation-associated conditions, such as seizure [82–86], Parkinson’s disease [87], retinitis pigmentosa [88], microglial activation and death [89, 90], glioma [34, 91–93], hyperalgesia [94], and brain ischemia [33, 95, 96].

Figure 3. Chemical structures of potential EP-targeting compounds for treatment of OAB.

(A) EP1 antagonists. (B) EP2 antagonists and positive allosteric modulators. (C) EP3 agonists and antagonists. (D) EP4 agonists and antagonists.

4.2. Where do we go from here?

The status (relaxation vs. contraction) of the smooth detrusor muscle of the urinary bladder fundamentally determines the storage and voiding of the urine, in which several families of K⁺ channels play central roles [40, 56]. It thus has been proposed that modulating these K⁺ channels directly or indirectly via targeting their upstream regulatory mechanisms might hold a therapeutic potential to control the urinary bladder functions. Using the guinea pig bladder strips, Parajuli et al. provided experimental evidence that PGE₂ might increase the spontaneous phasic contraction of the detrusor smooth muscle via the large conductance voltage- and Ca²⁺-activated K⁺ (BK) channels, accompanied by an elevation of intracellular Ca²⁺ levels [39]. Identifying the EP receptor subtype that is directly involved in the PGE₂-regulated spontaneous transient BK currents and the subsequent increase of intracellular Ca²⁺ would be a very interesting topic for the future studies and might provide a molecular target for novel pharmacological manipulation and therapeutic intervention for the underactivity or overactivity of the urinary detrusor.

Given the general concerns about the selectivity of pharmacological compounds, future studies should also be directed to use genetic strategies to validate the findings from pharmacological studies. Answering the question that whether the agonists or antagonists have any effects in OAB animals that lack the potentially targeted EP receptors would help to validate their target engagement, which is an essential step to their development into therapeutic agents for OAB. Moreover, all these EP receptors are also expressed in many other tissues and organ systems, where they may play some important roles under normal physiological conditions. Therefore, developing PGE₂ receptors-targeted therapeutics with higher efficacy and safety for the bladder overactivity would require a particular consideration of the undesirable effects on other important normal functions. For instance, the peripherally restricted antagonists TG6–129 for EP2 and DG-041 for EP3 might specifically modify the bladder micturition and bladder nociception via modulating the receptors on peripheral neurons or detrusor muscle cells without affecting the normal functions of the receptors in the CNS.