Inhibitory effects of phosphodiesterase 5 inhibitor, tadalafil, on mechanosensitive bladder afferent nerve activities of the rat, and on acrolein-induced hyperactivity of these nerves


Jean-Jacques Wyndaele, Department of Urology, Faculty of Medicine, University of Antwerp, S-4.50, Universiteitsplein 1, B-2610, Wilrijk, Belgium. e-mail:


What's known on the subject? and What does the study add?

Tadalafil, a phosphodiesterase type 5 inhibitor, might be effective for not only erectile dysfunction but also lower urinary tract symptoms (LUTS).

One of the mechanisms of tadalafil on LUTS is added in the study.


  • • To determine if tadalafil, a phosphodiesterase type 5 inhibitor, decreases afferent nerve activity from the bladder with and without chemical stimulation by intravesical acrolein instillation.


  • • Female Sprague-Dawley rats were used. Under urethane anaesthesia, single afferent fibres of the nerve primarily originating from the bladder were identified by electrical stimulation of the pelvic nerve and by bladder distension, and classified by conduction velocity as Aδ- or C-fibres. After measuring the baseline single afferent activities (SAA) during constant filling, two experiments were performed.
  • • First, tadalafil was administrated intravenously (i.v.) at three doses, 0.01, 0.03 and 0.1 mg/kg cumulatively and SAA were repeatedly studied after each administration.
  • • Second, in the presence of vehicle or tadalafil (0.1 mg/kg) i.v., the effect of intravesical instillation of acrolein (0.003%) was studied.


  • • In all, 39 single units were isolated (Aδ-fibres 21; C-fibres, 18) in 25 rats.
  • • Tadalafil dose-dependently decreased SAA of both Aδ- and C-fibres during saline instillation. Intravesical acrolein facilitated SAA of both fibres after vehicle administration.
  • • Pretreatment with tadalafil significantly inhibited the acrolein-induced hyperactivity of both fibres.


  • • Our study shows, using selective unifibre potential measurement, that systemic administration of tadalafil reduces mechanosensitive afferent activities of both Aδ- and C-fibres elicited by bladder distension in the rat, and also that tadalafil has an inhibitory effect on the increased activities of both fibres induced by intravesical acrolein instillation.

cyclic adenosine monophosphate


cyclic guanosine monophosphate


conduction velocity






erectile dysfunction


nitric oxide






single afferent activity


transient receptor potential ankyrin


LUTS are common among ageing men and the prevalence of LUTS in men aged >50 years has been estimated to be at least 50% [1]. LUTS is often comorbid with erectile dysfunction (ED) [2]. In general, α1-adrenergic receptor antagonists (α1-blockers) and/or 5α-reductase inhibitors are used to treat LUTS in men with an enlarged prostate [3]. Although effective, they can have several side-effects, e.g. dizziness, hypotension, and sexual dysfunction (especially, ejaculatory dysfunction) [4].

Phosphodiesterases (PDEs) are enzymes that metabolise second messenger molecules, e.g. cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP). The 21 human PDE genes are divided into 11 families based on their protein sequences, catalytic and regulatory considerations, and sensitivity to inhibitors, as well as their cAMP and/or cGMP affinity [5]. The mechanisms involved in improving LUTS by PDE5 inhibitors remain unclear, but these drugs have shown beneficial effects on smooth muscle relaxation, smooth muscle and endothelial cell proliferation, nerve activity, and tissue perfusion [6]. Several clinical studies of tadalafil and other PDE5 inhibitors widely used for treating ED have reported significant symptom reduction of LUTS in men with and without ED [7–9]. Caremel et al. [10] reported that the nitric oxide (NO)-cGMP signalling pathway is involved in the regulation of the micturition reflex, with an action that seems more predominant on the sensory rather than on the motor component in a rat model of bladder overactivity associated with C-fibre activation.

Cyclophosphamide (CYP)-induced haemorrhagic cystitis has been used as a model to study bladder-related pain and inflammation, and acrolein is identified as the causative metabolite of CYP [11–13]. We have recently shown that l-arginine, an NO synthase substrate, can inhibit both Aδ- and C- mechanosensitive afferent fibres of the bladder and also the activated responses of both fibres to intravesical acrolein in the rat [14]. In the present study, we investigated the effects of tadalafil on mechanosensitve single afferent activity (SAA) originating from the bladder, as well as on acrolein-induced afferent overactivity.


Female Sprague-Dawley rats (203–287 g) were used. The rats were maintained under standard laboratory conditions with a 12:12-h light:dark cycle, and free access to food pellets and tap water. The rats were anaesthetised with urethane (1.5 g/kg, i.p.). Body temperature was maintained by a heated blanket at 38 °C. After the experiments, the rats were killed with an overdose of urethane. All experimental protocols were approved by the Animal Ethics Committee of the University of Antwerp Faculty of Medicine (2007-44).


Single afferent fibre measurements were performed as described before [14]. In brief, the pelvic structures were exposed by a left lower abdominal incision. Both ureters were ligated close to the bladder. The left pelvic nerve was separated from surrounding tissue proximal to the major pelvic ganglion. A pair of teflon-coated silver electrodes were placed around the pelvic nerve and sealed with Wacker Silgel (Wacker Chemie, Munich, Germany). A polyethylene (PE) catheter (Clay-Adams PE-50, NJ, USA) was inserted into the bladder through the dome, and secured by a 5-0 surgical silk suture. The intravesical catheter was attached to a pressure transducer (DX-100, Nihon Kohden, Tokyo, Japan) and filling pump (NE-1000, ProSense, The Netherlands) for recording intravesical bladder pressure and infusing saline or drugs into the bladder. The lumbosacral spinal cord was exposed by a laminectomy and the dura mater was opened. The dorsal skin was tied up to make a pool, which was filled with body warm paraffin oil (Fisher Scientific, Loughborough, UK). Both L6 dorsal roots were cut and fine filaments were dissected from the left L6 dorsal root and placed across shielded bipolar silver electrodes

Recorded nerve activity was pre-amplified with a low noise AC differential amplifier (×10) and filtered (60–5000 Hz). A final amplification (×10 000) was used before the activity was displayed on an oscilloscope (TDS-310, Tektronix, Inc., Beaverton, OR, USA). Afferent fibres originating from the bladder were identified by electrical stimulation (0.5-ms square-wave pulses, SEN-3301, Nihon Kohden) of the pelvic nerve and bladder distension with saline. The nerve filaments were teased until a maximum of three clearly different unitary action potentials was evoked by electrical stimulation. These action potentials were discriminated by the Spike2 (CED, Cambridge, UK) impulse shape recognition program. Nerve activity was sampled at 20 kHz, bladder pressure at 100 Hz with the data acquisition program. Conduction velocity (CV) was calculated from the latency of response to electrical stimulation and the distance between stimulation and recording sites. Fibres were grouped based on CV. Those with a CV of <2.5 m/s were considered to correspond with unmyelinated C-fibres and those with a CV of ≥2.5 m/s to myelinated Aδ-fibres [15].

SAA was recorded during constant filling cystometry with 0.08 mL/min saline at room temperature. Filling continued until an intravesical pressure of 30 cmH2O was reached and then the saline in the bladder was evacuated by opening the catheter. During filling, the urethra was clamped to prevent leakage. The afferent activity caused by pelvic nerve stimulation was also recorded before and after bladder filling and confirmed to correspond with that caused by bladder filling. For i.v. administration of drugs, a PE-50 catheter was placed in the right femoral vein.

At the beginning of the experiments, recording was repeated consecutively three times at 5-min intervals to evaluate the reproducibility. The third recording served as the baseline value [16]. After measuring the baselines of SAA during constant filling cystometry, the following two experiments were performed.

First, tadalafil was administrated i.v. at three doses, 0.01, 0.03 and 0.1 mg/kg cumulatively. At 3 min after each dose administration of tadalafil, consecutive bladder fillings were performed with saline.

Second, 0.003% of acrolein was instilled into the bladder before i.v. administration of vehicle or tadalafil (0.1 mg/kg). Three cycles of investigation were performed after each administration. To facilitate permeability of the bladder urothelium for drugs, protamine sulphate solution (10 mg/mL, 0.3 mL) was instilled intravesically and kept in the bladder for 60 min just before the baseline measurement. It has been previously confirmed that protamine sulphate itself at this dose had no significant effects on SAA [17]. We reconfirmed that protamine sulphate alone did not affect significantly the afferent activity in a separated part of this study (data not shown).

The relationship of SAA to pressure or volume was established by comparing nerve activity and intravesical pressure at 1-s intervals. These values were then averaged at 5 cmH2O intervals of pressure or equally divided into five parts (0–20, 20–40, 40–60, 60–80, and 80–100%) of volume in the filling phase, and average unitary activity was totalled as a function of intravesical pressure or volume [14,16]. SAA are expressed as a percentage of baseline activity, integrated for the whole filling phase.


Tadalafil was a gift from Lilly Co. Ltd. (Eli Lilly and Company, Indianapolis, USA). Acrolein and N,N-Dimethylacetamide (DMA) were purchased from Sigma Chemical Company (St. Louis, MO, USA), and protamine sulphate from LEO Pharma N.V./S.A. (Wilrijk, Belgium). To obtain the adequate concentration of tadalafil, 1 mg of tadalafil was dissolved with 1 mL 40% DMA just before use, and then it was diluted with saline at concentrations of 0.01, 0.03 and 0.1 mg/mL. As a pilot experiment, we evaluated the effect of i.v. administration of 4% DMA solution, which corresponded to the highest concentration of the solvent used in the present study, on SAA and bladder compliance, and confirmed there were no significant effects (n= 3, data not shown). Thus, the concentration of DMA solution was used as a vehicle.


Statistical analyses were performed with Stat Mate® III III (Atoms Ltd., Tokyo, Japan). All data are expressed as mean (sem). Results were analysed using two-way anova followed by Tukey's test for comparison before and after vehicle or tadalafil administration. Unpaired Student's t-test was applied for comparison between groups. In all analyses P < 0.05 was considered to indicate statistical significance.


In all, 39 single-unit afferent fibres were isolated from 25 rats, of which 21 units corresponded to criteria for myelinated Aδ-fibres (mean [sem] CV of 9.35 [2.19] m/s), and 18 for unmyelinated C-fibres (mean [sem] CV of 1.45 [0.11] m/s).

After i.v. tadalafil or vehicle administration, neither numerical value nor percentage of baseline value of bladder compliance changed significantly when the bladder was filled with either saline or acrolein (Table 1).

Table 1.  Changes in bladder compliance after i.v. tadalafil during intravesical saline or acrolein instillation
Mean (sem) bladder compliance, mL/cmH2O; % (sem)
(a) Saline instillationBase0.01 mg/kg0.03 mg/kg0.1 mg/kg n
Tadalafil administration, mL/cmH2O;0.0212 (0.0012);0.0208 (0.0021);0.0231 (0.0025);0.0253 (0.0029);5
%10097.89 (4.16)107.02 (5.43)115.57 (6.06) 
(b) Acrolein instillationBaseAFTER 1after 2After 3 n
  1. Bladder compliance is expressed as an absolute value (millilitres per centimetre of water) calculated between start and end (intravesical pressure of 30 cmH2O) of the filling phase, and the percentage of the base-line value. The values are indicated as mean plus or minus sem. There is no statistically significant difference between before and after administration of vehicle or tadalafil.

Vehicle administration, mL/cmH2O;0.0211 (0.0012);0.0216 (0.0012);0.0226 (0.0012);0.0234 (0.0012);11
%100104.82 (5.91)107.39 (6.20)109.85 (7.01) 
Tadalafil administration, mL/cmH2O;0.0151 (0.0012);0.0140 (0.0010);0.0156 (0.0011);0.0157 (0.0012);9
%100102.34 (6.17)108.73 (4.81)105.07 (4.52) 

After tadalafil administration, the SAAs of both Aδ- and C-fibres in response to saline instillation decreased in a dose-dependent manner (Fig. 1A,B). The SAAs of all fibres investigated, except one out of six Aδ-fibres based on pressure, were decreased from the baseline after the highest dose (0.1 mg/kg) of tadalafil administration (Fig. 1C–F). The decreases in SAA were statistically significant in both Aδ- and C-fibres (Fig. 2).

Figure 1.

Representative recordings of the bladder pressure and firing rate of an Aδ-fibre (A) and a C-fibre (B) during bladder filling with saline before (base) and after i.v. tadalafil administration (0.01, 0.03, and 0.1 mg/kg). Individual responses of the Aδ-fibres (C and D) and C-fibres (E and F) integrated during the whole filling phase based on pressure (C and E) and volume (D and F) with instillation of saline before (base) and after tadalafil-administration (0.01, 0.03, 0.1 mg/kg).

Figure 2.

Responses of the Aδ-fibres (A) and C-fibres (B) integrated during the whole filling phase based on pressure (green bars) and volume (red bars) with instillation of saline before (base) and after tadalafil-administration (0.01, 0.03, 0.1 mg/kg). *P < 0.05, **P < 0.01: significant differences from base (two-way anova followed by Tukey's test).

After vehicle administration, the SAA of seven out of eight Aδ-fibres and all six C-fibres increased from the baseline during acrolein instillation (Fig. 3A,C,E–H). The increases in SAA induced by acrolein instillation were statistically significant in both Aδ- and C-fibres (Fig. 4). When only pretreated with tadalafil, the SAA did not change significantly during acrolein instillation in either Aδ- or C-fibres (Fig. 4). After tadalafil administration the SAA of two out of six Aδ-fibres and one out of six C-fibres slightly increased during acrolein instillation (Fig. 3B,D,I–L). There were significant differences in the responses of SAA during acrolein instillation between the vehicle- and tadalafil-pretreated groups in both Aδ- and C-fibres (Fig. 4).

Figure 3.

Representative recordings of the bladder pressure and firing rate of Aδ-fibres during bladder filling with acrolein before (base) and after vehicle (A) or tadalafil administration (B), and those of C-fibres during bladder filling with acrolein before (base) and after vehicle (C) or tadalafil-administration (D). Three cycles of investigation were performed after each administration (after 1, 2, and 3). Individual responses of nerve activity with instillation of acrolein before (base) and after vehicle (E, F, G, and H) or tadalafil administration (I, J, K, and L). Responses of the Aδ-fibres (E, F, I and J) and C-fibres (G, H, K and L) were shown and integrated during the whole filling phase based on pressure (E, G, I and K) and volume (F, H, J, and L). Three cycles of investigation were performed after each administration (after 1, 2, and 3).

Figure 4.

Responses of the Aδ-fibres (A, B) and C-fibres (C, D) integrated during the whole filling phase based on pressure (A, C) and volume (B, D) with instillation of acrolein before (base) and after vehicle (green bars) or tadalafil administration (red bars). Three cycles of investigation were performed after each administration (after 1, 2, and 3). *P < 0.05, **P < 0.01: significant differences from base (two-way anova followed by Tukey's test). #P < 0.05, ##P < 0.01: significant differences between vehicle and tadalafil administration (unpaired Student's t-test).


In the present study, i.v. administration of tadalafil significantly decreased the SAA of both Aδ- and C-fibres in response to bladder filling in a dose-dependent manner in anaesthetised rats. Individual results confirmed this finding. Neither the numerical value nor the percentage of the baseline of bladder compliance changed after tadalafil administration at the doses used. These results indicate that tadalafil inhibits both mechanosensitive Aδ- and C-fibre activities of the rat, without affecting the bladder tone.

We confirmed the dose-dependent effect of tadalafil by using three different doses, but a log dose-response study could not be made. The time to evaluate the SAAs was limited because the L6 nerve fibres put on the electrode were cut and torn in fine filaments to measure single afferent fibre activities. Although we confirmed the stability of SAAs with three measurements after the baseline measurement, to evaluate the effect of three doses in a fibre, it proved to take too long to obtain the results of a log dose-response study and keep the intact fibres viable.

Protamine sulphate was used to remove the glycocalix layer of urinary bladder and increase the permeability of the urothelium for drugs. Lavelle et al. [18] reported that only epithelial cells are damaged immediately after the protamine sulphate exposure, which we used, and that the underlying layers appear normal.

It has been reported that vardenafil, a PDE5 inhibitor, can inhibit bladder afferent nerve activity in unanaesthetised, decerebrate, spinal cord-injured rat. This complies with the present findings with tadalafil selectively in urethane-anesthetised rats [19]. We confirmed the inhibitory effect of tadalafil with single unit afferent measurement. From the present results, it seems likely that tadalafil can directly influence the mechanosensitive bladder primary afferent pathway via PDE5 inhibition.

Previous studies have identified an essential role of the PDE5 enzyme in the relaxation of penile erectile tissue through the NO–cGMP signaling pathway [20]. Expression of PDE5, as evidenced by immunochemical staining, was reported not only in the penile erectile tissue but also in the vascular smooth muscle and endothelial cells of the human prostate and smooth muscle cells of the prostatic urethra and bladder neck [21–23]. Animal and clinical studies have indicated that PDE5 inhibitors may act by multiple mechanisms to suppress LUTS: upregulating NO–cGMP signalling pathway activity; down-regulating RhoA kinase pathway activity; inhibiting autonomic overactivity; modulating bladder and prostate afferent activity; increasing pelvic blood perfusion and reducing chronic pelvic ischaemia [6]. NO has an inhibitory effect on ion channels in afferent neurones and on afferent nerve activity. The inhibition of Ca2+ channels in bladder afferent neurones by NO has been shown to be mediated by cGMP [24]. Aizawa et al. [14] reported an inhibitory effect of a NO synthase substrate (l-arginine) and a stimulatory effect of a NO synthase inhibitor (Nω-nitro-L-arginine methyl ester hydrochloride [L-NAME]) on primary mechanosensitive Aδ- and C-fibre afferents from the urinary bladder in the rat. So it seems possible that the reduction of storage LUTS by PDE5 inhibitors is due in part to an enhancement of the inhibitory action of NO on afferent nerve activity [6]. However, in chronic treatment prevention of ischaemic changes and remodelling of the bladder may also contribute to the effect of PDE5 inhibitors on LUTS. In an animal study, Matsumoto et al. [25] reported that chronic treatment with a high dose of vardenafil protected the rat bladder from contractile dysfunction to carbachol, and attenuated increase of bladder weight induced by BOO.

We used acrolein to induce chemical cystitis and consequently an increase in afferent nerve activity from the bladder. It has been shown that intravesically instilled acrolein facilitates mechanosensitve bladder SAAs of both Aδ- and C-fibres in the rat, and pretreatment with l-arginine blocks the facilitatory effect of acrolein [14]. In the present study, we confirmed that acrolein has a stimulatory effect on SAAs. And we showed that tadalafil has a similar suppressive effect on both fibre types as l-arginine [14]. These data indicate that accumulation of cGMP, via PDE5 inhibition, has a similar effect as endogenously increased NO, and can inhibit pathologically increased afferent activities induced by acrolein. Acrolein sensitises the mammalian transient receptor potential ankyrin (TRPA) 1 ion channels, which are mostly found in association with transient receptor potential vanilloid 1 (TRPV1) channels on C-fibres [26,27]. Streng et al. [26] reported that stimulation of TRPA1 ion channels induced detrusor overactivity of rat urinary bladder. However, acrolein is not a selective TRPA1 stimulator and it may increase the SAAs of both Aδ- and C-fibres by inducing inflammation of the bladder. In the present study, we showed that this SAA increase can be prevented by tadalafil administrated before acrolein.

In the present study, we also found individual differences in the response of SAAs of both Aδ- and C-fibres to tadalafil as well as acrolein, although most of them responded well. Previous studies showed that there are different types of mechanosensitive and chemosensitive fibres in rat bladder primary afferents [28,29]. It is conceivable that some of the mechanosensitive fibres did not respond to tadalafil or acrolein.

In conclusion, the results of the present study showed using direct afferent nerve activity measurement that systemic administration of tadalafil reduces mechanosensitive afferent activities of both Aδ- and C-fibres elicited by bladder distension in the rat, and also that tadalafil has an inhibitory effect on the increased activities of both types of fibres induced by intravesical acrolein instillation.


Tadalafil was kindly provided by Lilly Co. Ltd. without any financial support for this study. The present study was sponsored by Training Urology Scientists (TRUST) and EC Marie Curie project.


Jean-Jacques Wyndaele was an Advisor to Lilly Co. Ltd in 2010. Source of funding: Urological Research Fund UA, with no finances from Lilly Co. Ltd.