Excitatory bladder efferent innervation consists of pre- and post-ganglionic parasympathetic nerves, and an intermediate ganglion. Frequency-dependent properties of post-ganglionic fibres have been characterised; raising the firing rate between 1-40 Hz increases detrusor contraction force.  At low frequencies (1-10 Hz) ATP release at the nerve-muscle junction is important, but at high frequencies acetylcholine (ACh) increasingly dominates.  In the normal human bladder, ACh is the only functional transmitter, but with benign pathologies (and also in most non-human animals) ATP also has a role.  Strategies to selectively minimise nerve-related ATP release may therefore have the potential to reduce overactive bladder contractions.  The frequency domain of pre-ganglionic nerves is less well-characterised, although it may be up to 15 Hz [1].  However, it is unclear if the intermediate ganglia act as frequency-dependent filters, i.e. do they manipulate selectively low or high-frequency stimulation of post-ganglionic nerves.  This is important in the field of neuromodulation for conditions such as pelvic pain where pre-ganglionic nerve activity can be regulated by external sources [2], and selective low frequency block of post-ganglionic nerves should be useful.  The null hypothesis under test, using rat and pig models, was that intermediate ganglia do not offer frequency-dependent modulation of parasympathetic detrusor muscle activation.  This was tested in two animal models, pigs and rats, by comparing the frequency-dependence of detrusor contraction when stimulating pre- or post-ganglionic nerves.

Data were from adult pigs (Sus scrofa domestica, 6-months) obtained from a local abattoir, immediately after slaughter, and rats (Wistar, female, 200-240 g). Pig bladders were perfused with Tyrode’s solution (24mM HCO3-/5% CO2, pH 7.4) via the abdominal aorta with free venous outflow and housed in a water-jacketted chamber with a drainage siphon sufficient to maintain the bladder in a humidified environment [1]. The bladder was filled with 150 ml Tyrode’s solution. Intravesical pressure was measured by a 6-Ch catheter introduced through the left ureter and the right ureter tied off.  Interventions were introduced into the arterial perfusate.  Pre-ganglionic stimulation was by Pt-Ir cuff-electrodes around the pelvic nerve. Mixed pre-and post-ganglionic stimulation was by titanium mesh electrodes, with a Pt coating, sutured to the lateral wall of the dome, and changes to intravesical pressure recorded.  After these experiments a bladder wall section from the mid-dome was dissected, the mucosa removed and a detrusor strip (<1 mm diam.) tied to an isometric force transducer to record post-ganglionic nerve-mediated isometric contractions.  The same stimulation protocol was used in all preparations: a 3-s train of 0.1-ms square waves, frequency 1-40 Hz, every 90 s.
Rats were anaesthetised with urethane (1.4 g.kg-1, i.p.); the right femoral artery cannulated to record blood pressure and heart rate; the right femoral vein used for fluid infusion; the trachea cannulated to ensure airway patency; core temperature was maintained at 37°C.  Through a midline laparotomy, the pelvic nerve was gently separated from the uterine wall and a cuff electrode placed round the distal part.  Intravesical pressure was monitored via a suprapubic penetration with a 25G needle tip.  The same stimulation protocol was used as with pig experiments.  At the end of the experiment the animal was killed (Na pentobarbital (200 mg.kg-1, i.p.), the bladder immediately removed and a detrusor strip, after removal of mucosa, mounted as above to record post-ganglionic responses.
Data sets are mean±SD; n=number of animals.  Tension (T)- or pressure (P)-frequency relations were parameterised using the formulation: T(P) = (T(P)max.f^n)/(f1/2^n+f^n), where Tmax (Pmax) is the maximum value at high frequencies (f) and f1/2 the frequency required to attain Tmax/2 (Pmax/2); n is a constant that raises f and f1/2 to this power.  Differences between data sets were tested with ANOVA and post-hoc parametric t-tests.  The null hypothesis was rejected at *p<0.05.

Pig experiments.  The frequency-dependence of responses to pelvic nerve, bladder wall and isolated detrusor strip stimulation were not significantly different (figure 1A), with f1/2 values (frequency to attain half maximal responses) of 6.4±0.3 (n=7), 7.1±0.5 (n=7) and 6.5±0.7 (n=10), respectively. Lidocaine-HCl (7.4 mM) abolished responses in all three preparations indicating they were all nerve-mediated.  The nicotinic receptor antagonist hexamethonium (300 µM) completely abolished responses to pre-ganglionic pelvic nerve stimulation (n=6).  In contrast, there was no significant effect on post-ganglionic detrusor strip stimulation (102.8±6.1% control, n=6).  However, there was an intermediate effect on responses to bladder wall stimulation (pre- and post-ganglionic stimulation; 32.6±11.2% control, n=6). The action of the PDE5 inhibitor sildenafil was also tested.  With all preparations, responses were reduced (figure 1B); with this effect on pelvic nerve-stimulation > bladder wall stimulation > detrusor strip stimulation. 
Rat experiments.  Similar experiments with pelvic nerve and isolated detrusor strip stimulation also revealed similar frequency-response curves (figure 2A); the inset shows paired f1/2 values with the two methods of stimulation and no significant difference of mean values.  Hexamethonium (300 µM) had no effect on post-ganglionic detrusor-evoked responses and partially reduced pre-ganglionic pelvic nerve-evoked responses (30 mg.kg-1; figure 2B).

The similarity of frequency-response curves with pre-and post-ganglionic stimulation was shown in two animal models - pigs and rats, the former physiologically similar to the human bladder.  An interpretation is that, under these experimental conditions, the frequency-dependence of contractions is not modulated by the intermediate ganglion under these experimental conditions, unlike similar experiments in rabbits and cats [1].  The significance is that at postganglionic nerve-muscle junctions two transmitters (ATP and ACh) are released, both of which initiate contraction, but whose release may be separately modulated with agents such as the PDE5 inhibitor, sildenafil [3]. Selective blockade of low-frequency responses would reduce disproportionate dependence on ATP, a transmitter associated with human overactive bladder syndromes. Thus, neuromodulation of pre-ganglionic efferent firing would be reflected in post-ganglionic rates.  Moreover, the greater suppression by sildenafil of pre-ganglionic, compared to post-ganglionic, responses is consistent with the hypothesis that there is also dual transmitter release at the nerve-ganglion junction which offers a potential selective targeted drug model.

Data from a rat and a pig animal model show that the parasympathetic ganglion to the bladder does not offer a frequency-dependent filter of excitatory efferent firing.  However, sildenafil experiments show that the PDE5 inhibitor may modulate both pre-and post-ganglionic neurotransmitter release.

de Groat WC, Saum WR.  J Physiol 1976; 256: 137-158.
Roy H, Offiah O, Dua A. Brain Sci 2018; 8: 180.
Chakrabarty B, Ito H, Ximenes M, Nishikawa N, Vahabi B, Kanai AJ, Pickering AE, Drake MJ, Fry CH.  Br J Pharmacol. 2019; 176: 2227-2237