Novel drug targets for the pharmacotherapy of benign prostatic hyperplasia (BPH)

Abstract

Benign prostatic hyperplasia (BPH) is the major cause of lower urinary tract symptoms in men aged 50 or older. Symptoms are not normally life threatening, but often drastically affect the quality of life. The number of men seeking treatment for BPH is expected to grow in the next few years as a result of the ageing male population. Estimates of annual pharmaceutical sales of BPH therapies range from $US 3 to 10 billion, yet this market is dominated by two drug classes. Current drugs are only effective in treating mild to moderate symptoms, yet despite this, no emerging contenders appear to be on the horizon. This is remarkable given the increasing number of patients with severe symptoms who are required to undergo invasive and unpleasant surgery. This review provides a brief background on prostate function and the pathophysiology of BPH, followed by a brief description of BPH epidemiology, the burden it places on society, and the current surgical and pharmaceutical therapies. The recent literature on emerging contenders to current therapies and novel drug targets is then reviewed, focusing on drug targets which are able to relax prostatic smooth muscle in a similar way to the α₁-adrenoceptor antagonists, as this appears to be the most effective mechanism of action. Other mechanisms which may be of benefit are also discussed. It is concluded that recent basic research has revealed a number of novel drug targets such as muscarinic receptor or P2X-purinoceptor antagonists, which have the potential to produce more effective and safer drug treatments.

Background: normal prostate growth, development and gland function

The human prostate gland undergoes two androgen-dependent major growth stages. The first occurs during foetal development, when the gland arises from outgrowths of the urethral epithelium. The second phase occurs at puberty when the gland reaches a weight of approximately 20 g (McConnel, 1995). The adult prostate gland is a sponge-like male accessory structure which indirectly facilitates fertilization by increasing sperm motility and providing nourishment by the expulsion of fluid into the urethra. The columnar epithelial cells which line the prostatic ducts continually produce secretions. These secretions are stored in the ducts until the smooth muscle surrounding the ducts contracts under autonomic control, to expel the prostatic fluid just prior to ejaculation. The major constituents of prostatic fluid are polyamines, plasminogen, proteolytic enzymes, acid phosphatase, electrolytes, zinc, glucose and citric acid (Fair and Cordonnier, 1978; Zaichick et al., 1996; Weisser et al., 1997). Prostatic fluid, together with the fluid expelled from the seminal vesicle makes up the bulk of semen in which sperm is suspended to make up the ejaculate.

Background: benign prostatic hyperplasia (BPH)

With increasing age, alterations in hormone levels often result in the serious disorder known as benign prostatic hyperplasia (BPH) which is caused by an imbalance between cellular proliferation and apoptosis within the prostate. BPH initially manifests itself as microscopic nodules which progressively proliferate and enlarge to increase the mass of both the glandular and stromal prostatic tissue (Wilson, 1980). This abnormal third phase of prostate growth leads to an increase in smooth muscle tone and/or size of the gland. Because of the prostate's anatomical location surrounding the urethra, this growth is the major cause of bladder outlet and urethral obstruction in ageing men. Despite gross morphological and anatomical differences between prostate glands in humans and rodents, there appears to be many similarities with regard to their pharmacology and histochemistry making most laboratory animals of use as investigative models of prostate function.

Symptoms of BPH begin to appear in some men as early as age 40 and include a reduced urinary stream, nocturia, urinary retention, urgency, frequency and dribbling (Scarpa, 2001). These lower urinary tract symptoms and pathological BPH continue to worsen with age (Tsukamoto et al., 2007). Diagnosis of BPH is generally by a combination of physical examination by a physician and patient self-assessment to determine a score of urinary symptom severity.

Because of the unique capsular anatomy of the human prostate, BPH is almost exclusively unique to humans. The only experimental animals to spontaneously develop BPH are dogs (no prostatic capsule) and chimpanzees (prostatic capsule), with only the latter developing the symptoms associated with BPH (Wilson, 1980). Nearly every man at some point in his life will be affected by BPH, with 90% of men over the age of 85 affected by the disease (Gallegos and Frazee, 2008). A recent study of 5990 Australian men reported that 16% of men over the age of 40 complained of moderate to severe lower urinary tract symptoms, with 19% reporting symptoms of nocturia which necessitated two or more voids per night. Above the age of 70 years, 38% reported prostate disease and 41% reported nocturia. Twenty-nine per cent of men over the age of 70 complained of moderate to severe symptoms (Holden et al., 2005).

Background: burden on society of BPH

While BPH may be asymptomatic and undiagnosed in some men, obstruction of the urethra and deformation of the base of the bladder by the hyperplasic prostate will cause painful and debilitating lower urinary tract symptoms in most men as they age. Although rarely life threatening, the symptoms caused by BPH can have an extremely detrimental effect on the quality of life of both the men suffering from the disease as well as their partners. In the worst cases, sufferers may be forced to wake up to 10 times per night to urinate. Often, sufferers will also wake their partners as well. Such interrupted sleep patterns severely impact the ability of people to function normally during the day. Thus, BPH can reduce quality of life and drastically reduce the productivity of people affected by the disease. Because of the age-related nature of BPH progression, its prevalence will continue to increase in the short term as the population of western society ages.

Market research data for 2009 freely available on the Internet lists worldwide prostatic pharmacotherapy sales of $US 4.8 billion (Table 1). Nevertheless, pharmacological therapies are only effective for treating mild to moderate symptomatic cases of BPH. At present, patients experiencing severe symptoms can only be treated with the common but very invasive surgical procedures available such as transurethral resection of the prostate (TURP). More than 25 000 TURPs are performed in Australia each year at a cost of approximately $US 70 million (including public and private systems). The public system component alone represents 0.2% of total Medicare benefits paid (data obtained from National Hospital Morbidity Collection, Australian Institute of Health and Welfare). Similarly, in the United Kingdom, the economic burden of BPH to the National Health System (NHS) and Department of Social Security represents 0.4% of total NHS expenditure.

Table 1.

Top 4 marketed pharmacotherapy products for prostatic therapies 2009

				Annual worldwide sales $US (million)			Group share (%)
Rank	Product	Generic name	Company	2007	2008	2009	2007	2008	2009
1	Flomax/Alna/Harnal	Tamsulosin HCl	Boehringer Ingelheim / Astellas Pharma	2171	2279	2692	48	48	50
2	Avodart/Avolve	Dutasteride	GlaxoSmithKline	571	739	830	14	17	17
3	Xatral/Uroxatral	Alfuzosin HCl	Sanofi-Aventis	457	486	413	11	11	9
4	Proscar	Finasteride	Merck & Co	411	324	291	10	8	6
Other				424	486	554	11	11	12
Total				4033	4313	4779	100	100	100

Source: EvaluatePharma® (http://www.evaluatepharma.com).

Current therapies: surgical treatments for BPH

Current treatment options for BPH include surgical and pharmacological therapies. In severe cases, potentially life-threatening complications may occur, such as acute urinary retention, recurrent infections and renal failure. In such cases, surgery is the only truly effective treatment. As well as TURP, other surgical procedures include transurethral microwave thermotherapy (TUMT) and transurethral needle ablation (TUNA) (Miano et al., 2008). TURP is the most common surgical procedure and provides the greatest symptomatic relief, although it is a highly invasive procedure and side effects such as sexual dysfunction and incontinence are common. TUMT and TUNA are less invasive procedures and have fewer side effects, but are not as effective (Thorpe and Neal, 2003). For bothersome and enlarged prostates, open radical prostatectomy is also an option. However, nowadays this extremely invasive procedure is almost exclusively reserved for patients diagnosed with prostate cancer, which is beyond the scope of this review.

Current therapies: 5α-reductase inhibitor pharmacotherapy

Several recent reviews address the topic of future targets for the pharmacological treatment of BPH (Auffenberg et al., 2009; Stamatiou, 2009; Hashim and Abrams, 2010). This review broadens the list of potential targets by summarizing in more detail the literature on prostate function in not only humans but also laboratory animals.

Current pharmacotherapy targets either the static (an increase in the physical size of the prostate) or dynamic (an increase in the tone of the smooth muscle) component of BPH. Drugs targeting the static component of BPH act by inhibiting the proliferative action of androgens. These drugs include finasteride (Proscar®, Merck, Whitehouse Station, NJ, USA) and dutasteride (Avodart®, GlaxoSmithKline, Brentford, Middlesex, UK) which make up 23% ($US 1.1 billion) of the prostate pharmacotherapy worldwide market (Table 1). This class of drugs inhibit the action of the 5α-reductase enzyme which catalyses the conversion of testosterone to the more potent androgen dihydrotestosterone (Figure 1). Urethral obstruction is therefore relieved as the physical size of the prostate is decreased (Carson and Rittmaster, 2003; Tarter and Vaughan, 2006). The main complaints relating to patient use of 5α-reductase inhibitors are that they are slow-acting, often taking up to 6 months before effective relief of symptoms, and are associated with sexual side effects including impotence, decreased libido and abnormal ejaculation.

Figure 1.

Chemical structure and metabolism pathway showing the steroid conversion of testosterone into dihydrotestosterone and estradiol, via 5α-reductase and aromatase respectively.

Current therapies: α₁-adrenoceptor antagonist pharmacotherapy

The most effective and fastest acting treatments for BPH are drugs targeting the dynamic component (Miano et al., 2008). These include α₁-adrenoceptor antagonists such as tamsulosin (Flomax®, Yamanouchi, Tokyo, Japan; CSL, Parkville, VIC, Australia; Alna®, Boehringer Ingelheim am Rhein, Germany; Harnal®, Astellas, Tokyo, Japan) and alfuzosin (Uroxatral®, Sanofi-Aventis, Paris, France) which make up 65% ($US 3.1 billion) of the prostate therapy worldwide market (Table 1). These drugs decrease the stromal smooth muscle tone by blocking prostatic α₁-adrenoceptors (Figures 2 and 3), thus relieving urethral obstruction and the associated troublesome voiding symptoms (Lepor, 2007).

Figure 2.

Schematic diagram showing the postjunctional site of action of the α₁-adrenoceptor antagonists as smooth muscle relaxants in the prostate gland.

Figure 3.

Schematic diagram showing the likely locations on various cell types of membrane bound receptors and extracellular messengers which play a role in the control of contractility of prostatic smooth muscle. The diagrammatic representation is derived from experimental data obtained in a number of species. ACh, acetylcholine; ATP, adenosine 5′-triphosphate; CGRP, calcitonin gene-related peptide; ET, endothelin; NA, noradrenaline; NO, nitric oxide; SP, substance P.

There is a direct correlation between urethral obstruction and the amount of prostatic smooth muscle (Shapiro et al., 1992). In addition, 5α-reductase inhibitors are less effective than the α₁-adrenoceptor antagonists in the reduction of symptoms (Rigatti et al., 2003; Hasan et al., 2007; Lepor, 2007), indicating a greater importance for the dynamic component in the pathology of BPH. The relative importance of the dynamic component is further highlighted by observations that the severity of symptoms is not related to prostate size (Eckhardt et al., 2001a,b,c). α₁-Adrenoceptor antagonists produce mainly vasodilator side effects, but selective blockade of the α_1A-adrenoceptor subtype is also associated with some abnormal ejaculatory effects (Rokosh and Simpson, 2002).

Emerging contenders to current therapies

α₁-Adrenoceptor antagonists and 5α-reductase inhibitors remain the major targets of new drugs for the treatment of BPH. Most pharmaceutical research centres on making new compounds which will more selectively act at these targets and reduce adverse side effects, thereby improving patient adherence. Despite the effectiveness of current BPH treatments, there still remains the opportunity for new therapies that can stop disease progression and the subsequent need for surgery. The therapeutic potential for a combination therapy of 5α-reductase inhibitors and α₁-adrenoceptor antagonists is also very apparent. Combined use leads to a decreased progression of BPH, as well as mild symptomatic relief (Schulman, 2003). Indeed, in June 2010, a single-capsule formulation of dutasteride and tamsulsion (Jalyn®, GSK) was approved by the US Food and Drug Administration for use in symptomatic BPH in men with an enlarged prostate.

Gonadotrophin modulators, in particular, luteinizing hormone releasing hormone (LHRH) receptor antagonists such as cetrorelix have been extensively trialled for the treatment of BPH. The rationale for their use is based on inhibiting LHRH receptors in the pituitary, thereby blocking the signal for testosterone production in the testes and consequently inhibiting prostate growth. Clinical studies have shown that cetrorelix decreases prostate volume and improves lower urinary tract symptoms in patients suffering from BPH (Gonzalez-Barcena et al., 1994; Comaru-Schally et al., 1998; Debruyne et al., 2008, 2010). However, despite mild long-term symptomatic improvement, serum testosterone levels decrease only during dosing, and return to baseline levels in a matter of weeks (Comaru-Schally et al., 1998). This suggests that LHRH antagonists may act directly on the prostate to inhibit growth. On the contrary, LHRH agonists have been shown to have anti-proliferative effects in both androgen-dependent and independent prostate cancer cell lines (Limonta et al., 1992; 1999; Dondi et al., 1994). However, in the non-tumourgenic epithelial line BPH-1 (derived from prostate tissue of a 68-year-old BPH patient), the LHRH antagonist, cetrorelix, inhibits cell proliferation and reduces expression of mitogenic growth factors (Siejka et al., 2010). In a rat model of BPH, cetrorelix has also been shown to reduce prostate size presumably by inhibiting pro-inflammatory cytokines and growth factors (Rick et al., 2011). Because of either their direct or indirect effects on prostate growth, LHRH receptors represent an emerging drug target for the treatment of BPH. However, their efficacy appears only mild, and the amount of symptomatic relief that they can provide remains to be seen. Furthermore, they are associated with generally unacceptable anti-androgenic side effects such as impotence, which has led to a number of pharmaceutical companies abandoning this approach to treat BPH.

Novel vitamin D3 analogues such as BXL-628 have also shown efficacy in arresting prostate growth in patients with BPH in recent clinical trials (Colli et al., 2006). In this study, there was a small percentage decrease of approximately 7% in prostate size compared with placebo over 3 months of treatment. However, this decrease in prostate size did not translate to an improvement in urinary symptom score. In vitro studies have also shown BXL-628 to inhibit RhoA/Rho kinase signalling, which is an important intracellular pathway in smooth muscle cell contractility (Figure 4) (Morelli et al., 2007). However, the observed delay in carbachol-induced smooth muscle contraction was only minimal, and the lack of effect on patient urinary symptom score indicates that BXL-628 may not have the efficacy to compete with currently available treatments. On the other hand, carbachol is a non-selective muscarinic receptor agonist which suggests that it may only partially produce its effect by interaction with the RhoA/Rho kinase pathway. The merits of the RhoA/Rho kinase signalling pathway as a therapeutic target for BPH will be further discussed later in this review.

Figure 4.

Schematic diagram of the likely intracellular signalling pathways at play in controlling contractility of prostatic smooth muscle cells. ACh, acetylcholine; Adr, adrenaline; AMP, adenosine monophosphate; ATP, adenosine 5′-triphosphate; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanosine monophosphate; DAG, diacylglycerol; GMP, guanosine monophosphate; GPCR, G-protein coupled receptor; IP₃, inositol triphosphate; MLC, myosin light chain; NO, nitric oxide; PDE, phosphodiesterase; PIP₂, phosphatidylinositol bisphosphate; PK, protein kinase; PL, phospholipase.

Novel targets: β-adrenoceptor agonists

There are surprisingly few studies investigating the effects of β-adrenoceptors on human prostate contractility, given their seemingly large efficacy (Table 3). Early studies, attempting to characterize β-adrenoceptor activity in human prostate tissues met with little success (Caine et al., 1975). As the pharmacological tools developed, further attempts to classify prostatic β-adrenoceptors arose. β-Adrenoceptor-mediated elevations of adenylate cyclase were seen in human prostate ‘stromal’ and ‘epithelial’ tissues (Purvis et al., 1986). Later, it was shown that the non-selective antagonist, propranolol enhanced responses to both transmural stimulation and exogenously applied noradrenaline, and that this effect was more pronounced in tissue from non-hyperplastic tissues, which also showed a reduction in dihydroalprenolol binding sites (Tsujii et al., 1992). In the mid-1990s, β₃-adrenoceptor mRNA was identified in human prostatic tissue (Berkowitz et al., 1995), while more radioligand binding studies showed a dominance of the β₂- over the β₁-adrenoceptor subtype (Goepel et al., 1997). Subsequently, a β₃-adrenoceptor antibody appeared to localize the β₃-adrenoceptor to a subpopulation of prostatic stromal cells (Chamberlain et al., 1999). Functionally, the β₃-adrenoceptor appears to be a better target for pharmaceutical agents as its distribution appears more restricted than that of the β₁- and β₂-adrenoceptors. In prostatic stromal cells derived from human prostatic tissue, β₃- and possibly β₁-adrenoceptor activation reduces α₁-adrenoceptor-mediated contractions (Haynes and Hill, 1997), although animal studies show a predominant role for β₂- and/or β₁-adrenoceptors (Haynes and Hill, 1997; Kalodimos and Ventura, 2001). The most significant indication that β-adrenoceptors may prove a suitable therapeutic target is data suggesting that β blocker therapy increases the risk of developing BPH (Meigs et al., 2001).

Table 3.

Relative efficacies of various drug classes in relaxing isolated prostatic smooth muscle preparations precontracted by electrical stimulation or exogenous application of adrenoceptor agonists (where indicated by an asterisk*)

Drug class	Contraction remaining (%)	Species	References
Adenosine receptor agonist	∼20	Rat	Preston et al. (2000)
CGRP	∼60*	Rat	Watts and Cohen (1991)
	∼45	Rat	Ventura et al. (2000b)
α-adrenoceptor antagonist	∼65	Mouse	Gray and Ventura (2005)
	∼50	Mouse	White et al. (2010)
	∼40	Rat	Najbar-Kaszkiel et al. (1997)
	∼45	Rat	Ventura et al. (2003)
	∼50	Guinea-pig	Buljubasich and Ventura (2004)
	∼40	Guinea-pig	Najbar-Kaszkiel et al. (1997)
	∼50	Rabbit	Najbar-Kaszkiel et al. (1997)
	∼90	Pig	Najbar-Kaszkiel et al. (1997)
β-adrenoceptor agonist	∼60	Rat	Kalodimos and Ventura (2001)
	∼60*	Guinea-pig	Haynes and Hill (1997)
PGE2	∼65	Rat	Tokanovic et al. (2007)
	∼65	Rat	Tokanovic et al. (2010)
Nitric oxide donors	100*	Rat	Najbar-Kaszkiel et al. (1997)
	∼70*	Guinea-pig	Najbar-Kaszkiel et al. (1997)
	∼75*	Rabbit	Najbar-Kaszkiel et al. (1997)
	∼70*	Pig	Najbar-Kaszkiel et al. (1997)
	∼45*	Human	Hedlund et al. (1997)
CB-cannabinoid receptor agonist	∼75	Rat	Tokanovic et al. (2007)
	<73*	Human	Gratzke et al. (2010)
P2X-purinoceptor blocker	100	Mouse	Gray and Ventura (2005)
	100	Mouse	White et al. (2010)
	∼70	Rat	Ventura et al. (2003)
	∼70	Guinea-pig	Buljubasich and Ventura (2004)
Muscarinic receptor antagonist	∼80	Mouse	Gray and Ventura (2005)
	∼75	Mouse	White et al. (2010)
	100	Rat	Najbar-Kaszkiel et al. (1997)
	∼85	Guinea-pig	Najbar-Kaszkiel et al. (1997)
	∼80	Rabbit	Najbar-Kaszkiel et al. (1997)
	100	Pig	Najbar-Kaszkiel et al. (1997)
ET-endothelin receptor antagonist	100	Rat	Salamoussa et al. (2000)
	100	Guinea-pig	Lau et al. (1999)
Histamine receptor antagonist	100	Guinea-pig	Kerr (2006)
NK-tachykinin receptor antagonist	100	Guinea-pig	Buljubasich et al. (1999)

Estimates of efficacy are derived from published experimental data (often estimated from figures) in prostates taken from various species using various drugs.

CGRP, calcitonin gene-related peptide; PGE₂, prostaglandin E₂.

Novel targets: muscarinic receptor antagonists

Muscarinic receptor antagonists are not really novel treatments for BPH, but their use is limited by theoretical concerns that such drugs may lead to an increase in post-void residual, voiding difficulties and acute urinary retention due to detrusor inhibition. Recent reviews of clinical trials using muscarinic receptor antagonists (e.g. tolterodine, solifenacin, propiverine, oxybutynin or darifenacin) for the treatment of lower urinary tract symptoms, usually in combination with an adrenoceptor antagonist (e.g. doxazosin, tamsulosin or terazosin), show varying improvement in storage symptoms without worsening of voiding symptoms or significant development of acute urinary retention (Athanasopoulos, 2010; Chapple, 2010). This suggests that these drugs do indeed act by inhibition of detrusor instability in the bladder. Whether or not these antagonists also act in the prostate to relieve bladder outlet obstruction remains unclear.

Early studies found that compared with adrenoceptor-mediated contraction, muscarinic receptor agonists elicited only modest contractile responses from the human prostatic capsule. These responses were absent from the prostatic adenoma and did not enhance or inhibit adrenoceptor-mediated contraction (Caine et al., 1975; Gup et al., 1989; Kester et al., 2003). A more recent study observed that carbachol enhanced the contractile response to phenylephrine (Roosen et al., 2009). Furthermore, phenylephrine significantly reduced the pEC₅₀ for carbachol-mediated contractions. The authors suggested that a significant adreno-muscarinic receptor synergy in the prostate exists which may contribute to bladder outflow resistance.

A recent study in the mouse prostate showed that after inhibition of the adrenergic contractile response, a large cholinergic component remained (White et al., 2010). Moreover, a single randomized clinical trial investigating the safety and tolerability of tolterodine for the treatment of overactive bladder in patients with BPH indicated that there was a significant reduction in the bladder outlet obstruction index; however, the study provided no discussion of this result (Abrams et al., 2006).

Finally, it should be noted that muscarinic receptor agonists can induce proliferation in a number of prostate cancer cell lines (Witte et al., 2008; Avellar et al., 2009). Therefore, as well as inhibiting detrusor overactivity, muscarinic receptor antagonists may also relieve voiding symptoms of BPH by decreasing the size of the prostate and decreasing prostatic smooth muscle tone.

Novel targets: P2X₁-purinoceptor antagonists

Adenosine 5′-triphosphate (ATP) has been shown to be an excitatory co-transmitter with noradrenaline from the sympathetic nerves innervating the prostate of both the rat (Ventura et al., 2003) and guinea-pig (Buljubasich and Ventura, 2004). Despite the presence of several P2X-purinoceptor subtypes in the prostatic smooth muscle of mice (Gray and Ventura, 2005), rats (Lee et al., 2000) and guinea-pigs (Buljubasich and Ventura, 2004), both studies also used pharmacological methods to demonstrate that the P2X-purinoceptor responsible for prostatic contraction was of the P2X₁-purinoceptor subtype. In both studies, blockade of P2X₁-purinoceptors with either suramin or α, β-methylene ATP was able to inhibit electrically evoked nerve-mediated contractions in addition to the inhibition produced by prazosin. The relevance of these findings involving ATP to human prostate contractility is questionable, but ecto 5′-nucleotidase, an enzyme responsible for the catabolism of ATP, has been shown to be present in the human prostate (Konrad et al., 1998). Furthermore, P2X₁-purinoceptors are also expressed in human prostate (Longhurst et al., 1996).

Functional studies in human prostate have shown that the contractile response to electrical field stimulation is almost completely suppressed by α-adrenoceptor antagonists (Hedlund et al., 1985; Guh et al., 1995; Chueh et al., 1996). These studies were limited in that they used only high frequency stimulation (>2 Hz). Studies using prostates from laboratory animals have shown that purinergic neurotransmission is frequency dependent and particularly important at lower concentrations (Ventura et al., 2003; Buljubasich and Ventura, 2004). Furthermore, the human studies had the limitation of tissue being solely obtained at surgery. Many prostate surgery techniques cause trauma to the nerves which causes compensatory phenotype changes. Nerves containing ATP are thought to be particularly susceptible to such plasticity of purinoceptor expression (Burnstock and Knight, 2004) which may dampen their involvement in human tissue studies.

The importance of ATP as a neurotransmitter in male genitourinary function can not be underestimated as P2X₁-purinoceptor knockout mice are known to be infertile (Mulryan et al., 2000). ATP has recently been found to be involved in neurotransmission in the human vas deferens (Banks et al., 2006). If ATP is similarly involved in neurotransmission in the human prostate, the development of selective inhibitors at the P2X₁-purinoceptor may provide an additional target. When used in combination with a α_1A-adrenoceptor antagonist, a P2X₁-purinoceptor antagonist could provide greater prostatic smooth muscle relaxation and therefore, greater relief from symptoms. There is also convincing evidence for a role of ATP at genitourinary P2X-purinoceptors in the human bladder, particularly in those displaying pathological contractile overactivity (Fry et al., 2010). Therefore, it is likely that drugs affecting this mechanism would also act at the level of the bladder.

Novel targets: adenosine receptor agonists

Evidence exists that adenosine may modulate prostate smooth muscle contractility at two sites (Figure 3). In isolated rat prostates, activation of prejunctional A₁ adenosine receptors attenuates electrical field stimulation-induced contractile responses by inhibiting noradrenaline release (Preston et al., 2000). In human prostatic stromal cells, activation of post-junctional A_2A adenosine receptors inhibits α₁ adrenoceptor-mediated responses via stimulation of adenylate cyclase and subsequent accumulation of cyclic adenosine monophosphate (cAMP) (Preston et al., 2004). The physiological relevance of these observations is questionable, as gene disruption of the A_2A adenosine receptor in mice also produces a small reduction in prostate contractility (Gray et al., 2008b). The major effects that adenosine has on the cardiovascular system also places doubt on whether adenosine receptors would be of any clinical value as a target for the treatment of BPH.

Novel targets: cannabinoids

Stimulation of CB₁ cannabinoid receptors have been shown to inhibit contractions of the rat prostate gland (Tokanovic et al., 2007) (Tables 2 and 3). This action appeared to be indirect, and the authors used immunohistochemical techniques to localize the CB₁ cannabinoid receptors to the prostatic epithelium. CB₁ cannabinoid receptor expression has also been shown in the epithelial layer of the human prostate (Ruiz-Llorente et al., 2003). Furthermore, an anandamide uptake transporter and the fatty acid amidohydrolase enzyme, which degrades endocannabinoids, are also expressed in the human prostate (Ruiz-Llorente et al., 2004). More recently, CB₁ and CB₂ cannabinoid receptors have also been localized on sensory nerves innervating the human prostatic stroma and have been implicated in mediating inhibition of contraction (Gratzke et al., 2010). This indicates that cannabinoid receptors may also be a possible therapeutic target for the treatment of BPH. However, their inhibitory effects on prostate contractility appear to be only modest (Table 3) (Tokanovic et al., 2007; Gratzke et al., 2010), and the effects of cannabinoid receptor agonist treatment on the central nervous system would also need to be considered.

Table 2.

Receptor mechanisms mediating contractile and relaxant effects of prostatic stroma

Receptor	Agonists	Antagonists	Species	References
α1-adrenoceptor	↑	↓	Mouse	Gray and Ventura (2005)
				Gray and Ventura (2006)
			Rat	Najbar-Kaszkiel et al. (1997)
			Guinea-pig	Najbar-Kaszkiel et al. (1997)
				Pennefather et al. (1999)
			Rabbit	Couldwell et al. (1993)
				Testa et al. (1993)
				Yazawa and Honda (1993)
				Hiraoka et al. (1995)
				Delaflotte et al. (1996)
				Najbar-Kaszkiel et al. (1997)
			Dog	Ohmura et al. (1993)
				Normandin and Lodge (1996)
			Human	Testa et al. (1993); Testa et al. (1996)
				Chapple et al. (1994)
				Muramatsu et al. (1994)
				Teng et al. (1994)
				Eckert et al. (1995)
				Guh et al. (1995)
				Marshall et al. (1995)
				Tseng-Crank et al. (1995)*
β2-adrenoceptor	↓	–	Rat	Kalodimos and Ventura (2001)
			Guinea-pig	Haynes and Hill (1997)
			Dog	Normandin and Lodge (1996)
			Human	Tsujii et al. (1992)
Muscarinic receptor	↑	↓	Mouse	White et al. (2010)
			Rat	Najbar-Kaszkiel et al. (1997)
				Lau and Pennefather (1998)
				Lau et al. (1998)
			Guinea-pig	Najbar-Kaszkiel et al. (1997)
				Lau et al. (1998)
				Lau et al. (2000)
			Rabbit	Najbar-Kaszkiel et al. (1997)
			Dog	Normandin and Lodge (1996)
				Fernandez et al. (1998)
P2X1-purinoceptor	↑	↓	Rat	Ventura et al. (2003)
			Guinea-pig	Buljubasich and Ventura (2004)
A1-adenosine receptor	↓	–	Rat	Preston et al. (2000)
NK-tachykinin receptor	↑	–	Guinea-pig	Buljubasich et al. (1999)
			Human	Palea et al. (1996)
Calcitonin gene-related peptide receptor	↓	–	Rat	Watts and Cohen (1991)
				Ventura et al. (2000b)
ETA-endothelin receptor	↑	–	Rat	Salamoussa et al. (2000)
			Guinea-pig	Lau et al. (1999)
			Dog	Normandin and Lodge (1996)
				Langenstroer et al. (1997)
				Imajo et al. (1997)
			Human	Langenstroer et al. (1993)
				Webb et al. (1995)*
				Moriyama et al. (1996)
				Imajo et al. (1997)
				Raschack et al. (1998)
CB-cannabinoid receptors	↓	–	Rat	Tokanovic et al. (2007)
			Human	Gratzke et al. (2010)
EP2-prostanoid receptor	↓	–	Rat	Tokanovic et al. (2010)
Histamine receptors	↓	–	Guinea-pig	Kerr (2006)
			Dog	Normandin and Lodge (1996)

Upward arrows indicate an overall excitatory effect on contractility indicated by either an increase in basal tone or an enhancement of electrically evoked contractions.

Downward arrows indicate an overall inhibitory effect on contractility indicated by a relaxation of precontracted tissue or inhibition of electrically evoked contraction.

Dashes indicate no effect on contractility elicited by electrical stimulation or exogenous addition of α-adrenoceptor agonists.

All cited studies used direct force measurements of isolated prostate tissue preparations unless indicated by an asterisk*.

Novel targets: prostaglandin E₂

Prostaglandin E₂ has been shown to inhibit contractions of the rat prostate gland through action at a prostanoid receptor of the EP₂ subtype (Tokanovic et al., 2010) (Tables 2 and 3). Interestingly, prostaglandins were first isolated from semen and are so named because they were believed to be derived from the prostate gland (Bergstrom et al., 1968). The effects of prostaglandins on the contractility of the prostate gland are not well characterized. Early studies indicated that some of the prostaglandins caused an excitatory contractile response (Kitada and Kumazawa, 1987; Najbar-Kaszkiel et al., 1997; Sudoh et al., 1997). Nevertheless, there is now mounting evidence that inflammatory roles may play a role in the development of BPH (Kramer et al., 2007; Sciarra et al., 2007; 2008). Despite this, prostanoids have very widespread effects throughout the body (Narumiya et al., 1999) which may limit their use in the treatment of BPH.

Novel targets: histamine

Histamine has been shown to contract dog prostate (Normandin and Lodge, 1996) as well as have excitatory effects on Ca²⁺ mobilization in prostate cancer cells (Wasilenko et al., 1997). A later study in guinea pig prostate showed that histamine was not able to cause contraction on its own but potentiated contractions in response to electrical nerve stimulation and the exogenous application of ATP, noradrenaline or acetylcholine (Kerr, 2006). This potentiation appeared to be mediated by H₁ histamine receptors. The clinical significance of these findings for the treatment of BPH is unclear as the H₁ histamine receptor antagonist mepyramine on its own appeared to have no effect on contractions mediated by electrical field stimulation (Tables 2 and 3).

Novel targets: peptides

A number of peptides have been shown to have varying effects on the contractility of the prostate gland in a number of species (Table 2). Tachykinins have been demonstrated to be able to produce tonic contractions of isolated preparations of human prostate via the stimulation of NK₂ tachykinin receptors (Palea et al., 1996). Similar effects on smooth muscle have not been reported in rodent models, but tachykinins potentiate electrical field stimulation induced contractile responses of prostates taken from guinea pigs through stimulation of NK₁ tachykinin receptors (Buljubasich et al., 1999; Ventura et al., 2000a). In stark contrast, tachykinins have no effect on prostates taken from rats despite the presence of nerves immunoreactive for substance P and neurokinin A (Buljubasich et al., 1999).

Conversely, the sensory neuropeptide calcitonin gene-related peptide (CGRP) has been shown to inhibit contractions of the rat prostate elicited by exogenous administration of phenylephrine (Watts and Cohen, 1991) or electrical field stimulation (Ventura et al., 2000a,b) (Tables 2 and 3). In further contrast to the tachykinins, CGRP has no effect on contractility of prostates obtained from guinea-pigs despite the presence of CGRP immunoreactive nerves (Ventura et al., 2000b). The effects of CGRP are yet to be tested in human prostate despite numerous reports of CGRP immunoreactivity (Chapple et al., 1991; Jen and Dixon, 1995; Tainio, 1995; Hedlund et al., 1997). If similar smooth muscle relaxant effects were seen to CGRP, then this peptide may be potentially of some use in the treatment of BPH.

Similar to the sensory neuropeptides, endothelins also show species biodiversity in prostates taken from laboratory animals. In guinea-pigs, endothelins potentiate nerve-mediated contractions without affecting basal smooth muscle tone (Lau et al., 1999), whereas in rat prostates endothelins raise smooth muscle tone without affecting nerve mediated contractions (Salamoussa et al., 2000). In contrast to the neuropeptides, endothelins have been widely studied in human prostate. Endothelin-1 immunoreactivity has been demonstrated in the glandular epithelium (Langenstroer et al., 1993), and cultured cells derived from human prostate epithelium have been shown to secrete endothelin-1 (Walden et al., 1998). Endothelin-1 has also been shown to have contractile effects in human prostate (Langenstroer et al., 1993; Moriyama et al., 1996; Imajo et al., 1997). As in all animal species studied (Table 2), contractile effects of endothelins in the normal human prostate are mediated via ET_A endothelin receptors which are found throughout the human prostatic stroma (Kobayashi et al., 1994). However, in BPH, the subtype of endothelin receptor which mediates contraction may change to ET_B (Webb et al., 1995), as the number of endothelin receptors increases dramatically (Kondo et al., 1994; 1995; Moriyama et al., 1996). Endothelins also promote growth of human cultured smooth muscle prostatic cells through both ET_A and ET_B subtypes of endothelin receptor (Saita et al., 1998). This suggests that endothelins may play an excitatory role in proliferation as well as contractility in the human prostate. Such a combination of actions would theoretically be advantageous in a BPH drug target; however, antagonists of endothelin receptors have been shown to have little or no effect on contractility on their own (Tables 2 and 3).

Novel targets: α-adrenoceptor interacting proteins

Recent work investigating the changed pharmacology of the α_1A-adrenoceptor in the prostate gland has led to the possibility that an interacting protein is associated with this receptor in this tissue. With the use of genetically modified knockout mice, it has been shown in our laboratory that the α_1L-adrenoceptor pharmacological phenotype observed in the prostate arises from the α_1A-adrenoceptor gene (Gray et al., 2008a) and this has since been confirmed by other researchers (Muramatsu et al., 2008). Initially, it was thought that a polymorphism or splice variant of the α_1A-adrenoceptor gene may lead to the pharmacological expression of the α_1L-adrenoceptor functional phenotype, but this was shown not to be the case by a number of different groups (Shibata et al., 1996; Suzuki et al., 2000; Ramsay et al., 2004). Using whole segments as well as homogenates of prostate tissue in radioligand binding experiments to support their theory, it has now been postulated that an interacting protein may be associated with the α_1A-adrenoceptor in tissues where α_1L-adrenoceptor pharmacology is exhibited (Muramatsu et al., 2005; Nishimune et al., 2010a). This may explain why α_1L-adrenoceptor pharmacology is only seen in functional studies using intact tissue or whole cells and not in radioligand binding experiments where prostate tissue homogenates have been used (Muramatsu et al., 2005; Nishimune et al., 2010a).

A yeast two-hybrid approach to screen the human prostate cDNA library using the full open reading frame of the human α_1A-adrenoceptor gene as bait was recently used to identify cysteine-rich with epidermal growth factor-like domain 1α (CRELD1α) as a protein that interacts with the α_1A-adrenoceptor, and may account for the emergence of α_1L-adrenoceptor pharmacology in certain cell types (Nishimune et al., 2010b). Co-expression of CRELD1α with the α_1A-adrenoceptor generates a larger proportion of receptors displaying α_1L pharmacology in Chinese hamster ovary cells (Nishimune et al., 2010b). However, the apparent abundance of adrenoceptors also changed markedly, so it is not clear whether CRELD1α is able to change the properties of this adrenoceptor by modifying its conformation (as seen with receptor activity modifying proteins), simply reducing its expression or changing its localization within the cell.

The possibility of a protein which interacts with the α_1A-adrenoceptor to change its pharmacology is plausible and would be a logical explanation for the diverse α_1A-adrenoceptor pharmacology seen in the prostate gland. CRELD1α may or may not be such a protein; however, identification of a α_1A-adrenoceptor interacting protein or a α_1A-adrenoceptor heteromer complex with another receptor which changed its pharmacology would be a significant advance in developing a more prostate-specific α_1A-adrenoceptor antagonist.

Novel targets: phosphodiesterase (PDE) inhibitors

The major intracellular signalling pathways involved in prostatic smooth muscle cell contraction are shown in Figure 4. Of these, phosphodiesterase (PDE) would appear to be the most appropriate drug target for the treatment of BPH. PDEs play an important role in terminating signalling through the breakdown of cyclic nucleotide monophosphates [cAMP and cyclic guanosine monophosphate (cGMP); Figure 4]. Early studies showed that both direct or indirect activators of the cyclic nucleotide stimulating pathway relaxed prostate tissue (Drescher et al., 1994; Haynes and Cook, 2006; Kedia et al., 2006; Oger et al., 2009), an effect possibly mediated through the activation of potassium channels (Kurokawa et al., 1998; Cook et al., 2002; Haynes and Cook, 2006). Cyclic nucleotides and PDE inhibitors also modulate human prostatic cell proliferation (Guh et al., 1998; Adolfsson et al., 2002; Cook and Haynes, 2004; Fibbi et al., 2010) and fibroblast-to-myofibroblast transdifferentiation (Zenzmaier et al., 2010).

Prostatic tissue is reported to have at least 14 isozymes (Stacey et al., 1998; Uckert et al., 2001), but the major isozymes appear to be the PDE4A, 4B, 5A and 11A4 (Fawcett et al., 2000; Yuasa et al., 2000) which have varied, and occasionally opposing, distributions throughout the glandular and stromal compartments (Uckert et al., 2006; Fibbi et al., 2010; Waldkirch et al., 2010).

Largely, as an indirect result of current treatments for erectile dysfunction, there is clinical data suggesting that PDE inhibitors may be of benefit in treating lower urinary tract symptoms. For example, the administration of two PDE5 inhibitors, adalafil or udenafil, increases prostatic cAMP and cGMP (Zhao et al., 2011). More importantly, the PDE5 inhibitors, tadalafil, sildenafil and vardenafil improve prostate symptom scores (McVary et al., 2007a,b; 2008; Roehrborn et al., 2008; Stief et al., 2008; Dmochowski et al., 2010; Tuncel et al., 2010). Whether PDE5 inhibitor therapies provide more relief from lower urinary tract symptoms alone, than in combination with other therapies, is still equivocal (Kaplan and Gonzalez, 2007; Kaplan et al., 2007; Kaplan and Hatzichristou, 2007; Tuncel et al., 2010).

Novel targets: RhoA/Rho kinase

The RhoA/Rho kinase pathway has received much attention in recent years due to its role in regulating smooth muscle contractility. RhoA is a G-protein coupled to excitatory receptors (e.g. α₁-adrenoceptors), which activates Rho kinase, which in turn maintains actin filaments in a contracted state through inhibition of myosin light chain phosphatase (Figure 4). RhoA and Rho kinase are expressed in smooth muscle cells of the human prostate, and inhibitors of Rho kinase decrease contractile responses to electrical field stimulation, noradrenaline, phenylephrine and endothelin in human prostate tissue (Takahashi et al., 2007) or cultured prostatic stromal cells (Rees et al., 2003). Interestingly, inhibitors of Rho kinase have also been found to decrease cell proliferation (Rees et al., 2003). Although studies on the efficacy of Rho kinase inhibitors are yet to be carried out in a clinical setting, the RhoA/Rho kinase pathway shows potential as a therapeutic target for controlling both the increased growth and contractility associated with BPH. Indeed, this pathway has been shown to be the mechanism of action of a number of novel compounds which show potential in BPH treatment, such as the vitamin D receptor agonist BXL-628 (Morelli et al., 2007), KMUP-1 (Liu et al., 2007) and isoflavones (Seok et al., 2008).

Novel targets: nitric oxide

The human prostate stroma receives dense nitrergic innervation which has been found to be reduced in BPH patients (Bloch et al., 1997). Studies have shown that nitric oxide donors inhibit noradrenaline-mediated contractions and induce smooth muscle relaxation in human prostate tissue (Hedlund et al., 1997) (Table 3). Similarly, inhibitors of nitric oxide synthesis have been found to decrease electrically induced relaxations (Takeda et al., 1995; Hedlund et al., 1997). Thus, nitric oxide has been postulated to play a significant role in regulating prostate smooth muscle relaxation through stimulation of cGMP production (Figure 4) and, in turn, activation of K_ATP channels (Cook et al., 2002). Targeting this pathway has been investigated in clinical trials which show that lower urinary tract symptoms associated with BPH improve under the influence of agents that enhance nitric oxide activity such as direct nitric oxide donors (Klotz et al., 1999) or PDE5 inhibitors such as sildenafil (Sairam et al., 2002).

Novel targets: interstitial pacemaker cells

Spontaneous contractions in the guinea pig prostate have been reported to be associated with the firing of ‘slow waves’ recorded in the smooth muscle stroma using electrophysiological techniques (Exintaris et al., 2002). This slow wave activity is myogenic in origin as it is unaffected by blockers of neural propagation or transmission, such as tetrodotoxin, guanethidine, atropine, capsaicin or indomethacin (Exintaris et al., 2002). Further characterization of this spontaneous electrical activity showed that it is dependent on an external source of Ca²⁺ (Exintaris et al., 2002). More recently, it was established that slow waves recorded in the guinea pig prostate are also dependent on the cycling of Ca²⁺ from inositol triphosphate-dependent Ca²⁺ stores and the buffering of Ca²⁺ by mitochondria (Exintaris et al., 2009). In contrast, voltage-activated 4-amino-pyridine-sensitive K⁺ channels, as well as iberiotoxin-sensitive large conductance Ca²⁺-activated K⁺ (BK_Ca) channels regulate the time-course and frequency of the spontaneous electrical events as demonstrated by patch-clamping freshly digested guinea pig prostatic cells (Oh et al., 2003; Lang et al., 2004) or human prostatic smooth muscle cells (Sui et al., 2004).

A discrete population of prostatic interstitial cells is thought to generate this electrical activity. In the guinea pig prostate gland, these cells lie between the glandular and stromal smooth muscle layers of the individual acini (Exintaris et al., 2002). Similar cells have been previously described in the rat prostate (Aumuller et al., 1987) and more recently in the human prostate (Van der Aa et al., 2003; Shafik et al., 2005). Electron microscopic examination has revealed that these prostatic interstitial cells lack myosin filaments, possess many mitochondria and have an incomplete basal lamina. In contrast, prostatic smooth muscle cells comprise a continuous basal lamina, relatively few mitochondria and an abundance of contractile filaments (Exintaris et al., 2002). By analogy with the gut, it is likely that prostatic interstitial cells provide the depolarizing pulse to neighbouring smooth muscle cells, initiating slow wave activity and subsequent contractility. It is also likely that prostatic interstitial cells are modulated by intrinsic nerves (Exintaris et al., 2006; Dey et al., 2009).

It is conceivable that pathological changes to prostatic interstitial cells or disruption to prostatic interstitial cell/nerve/smooth muscle networks may play a role in disorders of the prostate gland (Exintaris et al., 2006). An understanding of the cellular mechanisms underlying the prostatic rhythmic contractile effects of nerve-mediated agents could lead to the development of therapeutic agents to treat prostate-specific conditions such as BPH. However, whether a drug can be tailored to target this type of cell is yet to be elucidated.

Concluding remarks

It would be highly desirable to have a BPH therapy that affects the progression of disease rather than only symptom relief. Nevertheless, drugs which relax prostatic smooth muscle are faster acting, more effective and better tolerated than those which shrink or reduce prostate volume when it comes to treating BPH (Hutchison et al., 2007). Furthermore, there is poor correlation between the severity of lower urinary tract symptoms caused by BPH and prostate size (Eckhardt et al., 2001a,b,c). Thus, while an effective agent for BPH that affects disease progression remains elusive, it would seem that the best strategy to improve drugs used in the treatment of BPH would be to concentrate on finding mechanisms which can produce prostatic relaxation greater than that produced by the currently available α₁-adrenoceptor antagonists.

α₁-Adrenoceptor blockade is only able to block approximately half of the contractile response to nerve stimulation in most species studied (Table 3). This implies that inhibiting the nonadrenergic residual response may provide a target, which when used in combination with α₁-adrenoceptor antagonists will provide greater relaxation of prostatic smooth muscle and therefore, greater relief from symptoms. In moderate to severe cases, improved efficacy of drugs may be sufficient for some patients to avoid the need for surgery. Of the possible contenders described in this review, and summarized in Figures 3 and 4, as well as other possible targets not listed here, it would appear that muscarinic receptor antagonists are the most likely candidates. Among all the rest, PDE inhibitors, nitric oxide donors and perhaps P2X₁-purinoceptor antagonists also offer some hope.

Glossary

Abbreviations

ATP

adenosine 5′-triphosphate

BPH

benign prostatic hyperplasia

cAMP

cyclic adenosine monophosphate

cGMP

cyclic guanosine monophosphate

CRELD

cysteine-rich with epidermal growth factor (EGF)-like domain

CGRP

calcitonin gene-related peptide

CHO

Chinese hamster ovary

DHT

dihydrotestosterone

FAAH

fatty acid amidohydrolase

FDA

Food and Drug Administration

IP₃

inositol triphosphate

LHRH

luteinizing hormone releasing hormone

NHS

National Health System

PDE

phosphodiesterase

RAMP

receptor activity modifying protein

TUMT

transurethral microwave therapy

TUNA

transurethral needle ablation

TURP

transurethral resection of the prostate

Conflict of interest

The authors state no conflicts of interest.

Supporting Information

Teaching Materials; Figs 1–4 as PowerPoint slide.