The functions of the lower urinary tract (LUT) to store and periodically release urine are dependent upon neural circuits located in the brain, spinal cord and peripheral ganglia (see Morrison et al., 2005). This dependence on central nervous control distinguishes the LUT from many other visceral structures (e.g., the gastrointestinal tract and cardiovascular system) that maintain a certain level of activity even after elimination of extrinsic neural input. The LUT is also unusual in regard to its pattern of activity and the complexity of its neural regulation. For example, the urinary bladder has two principal modes of operation: storage and elimination. Thus many of the neural circuits controlling the bladder exhibit switch-like or phasic patterns of activity in contrast to tonic patterns occurring in autonomic pathways to cardiovascular organs. In addition, micturition is under voluntary control and depends upon learned behaviour that develops during maturation of the nervous system, whereas many other visceral functions are regulated involuntarily. Micturition also depends on the integration of autonomic and somatic efferent mechanisms within the lumbosacral spinal cord (see Morrison et al., 2005). This is necessary to coordinate the activity of visceral organs (the bladder and urethra) with that of urethral striated muscles. This paper will review the peripheral and central neural mechanisms controlling the LUT and the disruption of this control by neural injury.

The storage and periodic elimination of urine is dependent upon the activity of two functional units in the LUT: (1) a reservoir (the urinary bladder) and (2) an outlet, consisting of bladder neck, urethra and striated muscles of the urethral sphincter (see Fry et al., 2005; Morrison et al., 2005). These structures are, in turn, controlled by three sets of peripheral nerves: sacral parasympathetic (pelvic nerves), thoracolumbar sympathetic nerves (hypogastric nerves and sympathetic chain) and sacral somatic nerves (pudendal nerves) (Figure 1) (see Morrison et al., 2005).

Figure 1 Diagram showing the sympathetic, parasympathetic and somatic innervation of the urogenital tract of the male cat. Sympathetic preganglionic pathways emerge from the lumbar spinal cord and pass to the sympathetic chain ganglia (SCG) and then via the inferior (more ...)

Table 1 Receptors for putative transmitters in the lower urinary tract

The parasympathetic ganglion cells in turn excite bladder smooth muscle via the release of cholinergic (acetylcholine) and nonadrenergic–noncholinergic transmitters. Cholinergic excitatory transmission in the bladder is mediated by muscarinic receptors, which are blocked by atropine (see Andersson, 1993; Andersson & Arner, 2004; Morrison et al., 2005), whereas noncholinergic excitatory transmission is mediated by adenosine triphosphate (ATP), acting on P2X purinergic receptors (Table 1) (Ralevic & Burnstock, 1998; Burnstock, 2001). Inhibitory input to the urethral smooth muscle is mediated by nitric oxide (NO) released by parasympathetic nerves (Andersson, 1993; Andersson & Arner, 2004). Both M₂ and M₃ muscarinic receptor subtypes are expressed in bladder smooth muscle; however, examination of subtype selective muscarinic receptor antagonists and studies of muscarinic receptor knockout mice have revealed that the M₃ subtype is the principal receptor involved in excitatory transmission (Matsui et al., 2000; 2002). Muscarinic receptors are also present prejunctionally on parasympathetic nerve terminals. Activation of these receptors by acetylcholine can enhance (M₁ receptors) or suppress (M₄ receptors) transmitter release, depending upon the intensity of neural firing (Somogyi et al., 1996; 1998; 2003; D'Agostino et al., 1997; see de Groat & Yoshimura, 2001). Postganglionic neurones innervating the bladder also contain neuropeptides, such as vasoactive intestinal polypeptide (VIP) and neuropeptide Y (NPY) (Keast & de Groat, 1989). These substances are co-released with acetylcholine or ATP and may function as modulators of neuroeffector transmission (Tran et al., 1994). It has been proposed that frequency-dependent synaptic modulatory mechanisms in parasympathetic ganglia and at postganglionic nerve terminals perform gating functions that suppress excitatory input to the bladder during urine storage when the parasympathetic preganglionic outflow from the spinal cord is low and amplify the input to the bladder during voiding when preganglionic nerve activity is high, thereby contributing to efficient bladder emptying (de Groat & Booth, 1993; Tran et al., 1994; Somogyi et al., 1996).

C-fibre afferents are sensitive to the neurotoxins, capsaicin and resiniferatoxin as well as to many other substances including tachykinins, NO, ATP, prostaglandins, endothelins and neurotrophic factors released in the bladder by afferent nerves, urothelial cells and inflammatory cells (Maggi, 1993; Chuang et al., 2001; Rong et al., 2002; Morrison et al., 2005). These substances can modulate afferent nerve excitability and change the response of afferents to mechanical stimulation. Intravesical administration of ATP enhances the firing of bladder afferent nerves (Rong et al., 2002; Morrison et al., 2005), presumably by acting on P2X₃ or P2X_2/3 receptors on afferent terminals within or adjacent to the urothelium. In addition, ATP applied to afferent nerves on the surface of the rat bladder enhances the firing induced by bladder distension and reduces the threshold for electrical stimulation of A-δ and C-fibre afferent axons (Yu & de Groat, 2004). These data suggest that purinergic receptors are located on the axons of afferent nerves as well as at the nerve terminals. This raises the possibility that afferent axons may be sensitive to purinergic excitation at any site along the path of the axon as it passes through the bladder wall. Sensitivity of vagal and sural nerve C-fibre afferent axons to ATP has also been reported (Inrich et al., 2001; 2002; Lang et al., 2005).

Axonal tracing studies have revealed that a small percentage of lumbosacral afferent neurons innervate multiple pelvic organs. For example, 3–15% of dorsal root ganglion neurons were double labelled following injections of different tracers into the colon and bladder (Keast & de Groat, 1992; Christianson et al., 2004) The double labelling occurred more frequently in rostral lumbar (L1–L2) than in caudal lumbosacral (L6–S1) dorsal root ganglia, which provide the major innervation to the bladder and colon. It has been speculated that double labelling is due to dichotomizing afferents that send axonal branches to different target organs. This unusual anatomical arrangement has been put forward as a possible mechanism for neural cross-talk and bidirectional crosssensitization between pelvic organs in which chemical irritation of the colon leads to enhancement of reflex bladder activity or irritation of the bladder leads to enhancement of colonic reflexes (Pezzone et al., 2005).

The properties of lumbosacral dorsal root ganglion cells innervating the bladder, urethra and external urethral sphincter in the rat and cat have been studied with patch-clamp recording techniques in combination with axonal tracing methods to identify the different populations of neurons (Yoshimura et al., 1996; 2003; Yoshimura & de Groat, 1997; 1999; Sculptoreanu et al., 2005a, 2005b). Based on responsiveness to capsaicin, it is estimated that approximately 70% of bladder afferent neurons in the rat are of the C-fibre type. These neurons exhibit high threshold tetrodotoxin-resistant sodium channels and action potentials and phasic firing (one to two spikes) in response to prolonged depolarizing current pulses. Approximately 90% of the bladder afferent neurons are also excited by ATP, which induces a depolarization and firing by activating P2X₃ or P2X_2/3 receptors (Zhong et al., 2003). Bladder afferent nerves near the urothelium express P2X₃ and P2Y₄ purinergic receptors (Birder et al., 2004). A-fibre afferent neurons are resistant to capsaicin, have low threshold tetrodotoxin-sensitive sodium channels and action potentials and tonic firing (multiple spikes) to depolarizing current pulses. C-fibre bladder afferent neurons also express a slowly decaying A-type K⁺ current that controls spike threshold and firing frequency (Yoshimura et al., 1996, 2003). Suppression of this K⁺ current by drugs or chronic bladder inflammation induces hyperexcitability of the afferent neurons (Yoshimura & de Groat, 1999). Conversely, enhancement of A-type K⁺ currents with an experimental drug (KW-7158) suppresses the excitability of cultured dorsal root ganglion neurons (Sculptoreanu et al., 2004) and decreases bladder hyperexcitability induced by chemical irritation of the bladder in vivo (Lu et al., 2002).

A large percentage of bladder afferent neurons contain peptides: calcitonin-gene-related peptide, vasoactive intestinal polypeptide, pituitary-adenyl cyclase activating polypeptide (PACAP), tachykinins, galanin and opioid peptides (Maggi, 1993; Morrison et al., 2005). Nerves containing these peptides are common in the bladder, in the submucosal and epithelial layers, and around blood vessels. Peptidergic bladder afferent neurons in the rat also express TrkA, a high-affinity receptor for nerve growth factor (NGF) and receptors for tachykinins (NK₂ and NK₃ receptors) (Sculptoreanu & de Groat, 2003; Morrison et al., 2005) and endothelins (Ogawa et al., 2004). Peptidergic afferent axons project into the lumbosacral parasympathetic nucleus in the spinal cord and application of various neuropeptides to the spinal cord influences bladder activity. These findings suggest that the neuropeptides may be important transmitters in the afferent pathways from the LUT.

Recent studies have revealed that the urothelium, which has been traditionally viewed as a passive barrier at the bladder luminal surface (Lewis, 2000; Apodaca, 2004), also has specialized sensory and signalling properties that allow urothelial cells to respond to their chemical and physical environment and to engage in reciprocal chemical communication with neighbouring nerves in the bladder wall (Ferguson et al., 1997; Birder et al., 1998; 2001; 2002a, 2002b; de Groat, 2004; Stein et al., 2004; Beckel et al., 2005a, 2005b; Chopra et al., 2005). These properties include: (1) expression of nicotinic, muscarinic, tachykinin, adrenergic, bradykinin and transient receptor potential receptors (TRPV1, TRPV2, TRPV4, TRPM8 and ANKTM1), (2) responsiveness to transmitters released from sensory nerves, (3) close physical association with afferent nerves and (4) ability to release chemical mediators such as ATP, ACh and NO that can regulate the activity of adjacent nerves and thereby trigger local vascular changes and/or reflex bladder contractions.

The role of ATP in urothelial-afferent communication has attracted considerable attention because bladder distension releases ATP from the urothelium (Ferguson et al., 1997; Sun et al., 2001; Birder et al., 2003) and intravesical administration of ATP induces bladder hyperactivity, an effect blocked by administration of P2X purinergic receptor antagonists that suppress the excitatory action of ATP on bladder afferent neurons (Morrison et al., 2005). Mice in which the P2X₃ receptor was knocked out exhibited hypoactive bladder activity and inefficient voiding (Cockayne et al., 2000), suggesting that activation of P2X₃ receptors on bladder afferent nerves by ATP released from the urothelium is essential for normal bladder function. In humans and cats with interstitial cystitis, a painful bladder condition, ATP release from urothelial cells is enhanced (Sun et al., 2001; Birder et al., 2003). Higher levels of ATP may induce abnormal afferent nerve firing and pain.

Botulinum toxin A (BTX-A), which is injected into the bladder wall to reduce neurogenic detrusor overactivity in patients (Smith & Chancellor, 2004; Schurch et al., 2005), not only suppresses the release of acetylcholine and norepinehrine from autonomic nerves in the rat bladder (Smith et al., 2003a) and inhibits neurally evoked bladder contractions (Smith et al., 2003b) but also reduces the release of ATP into the bladder lumen of chronic spinal cord-injured rats (Khera et al., 2004). BTX-A also blocks the stretch-evoked or capsaicin-evoked release of ATP from cultured urothelial cells (Barrick et al., 2004) and reduces the activation of afferent nerves by bladder irritation (Chuang et al., 2004; Vemulakonda et al., 2005). Thus, the clinical efficacy of BTX-A in the treatment of bladder dysfunction may be related to its action on urothelial sensory mechanisms as well as to its effects on neurotransmitter release from efferent nerves.

NO released from the urothelium (Birder et al., 1998) has been implicated in an inhibitory modulatory mechanism (Ozawa et al., 1999). Exogenous NO inhibits Ca²⁺ channels in dissociated lumbosacral dorsal ganglion neurons innervating the urinary bladder (Yoshimura et al., 2001). In addition, intravesical administration of NO donors suppresses bladder hyperactivity in cyclophosphamide-induced cystitis (Ozawa et al., 1999), whereas intravesical administration of oxyhaemoglobin, an NO scavenger, produces bladder hyperactivity in normal rats (Pandita et al., 2000). These data indicate that NO released from the urothelium can suppress the excitability of adjacent afferent nerves. However, NO may also contribute to bladder overactivity. NO production and nNOS expression are increased in the urothelium of cats with IC (Birder et al., 2005) and high levels of NO can disrupt the urothelial passive barrier. Thus, NO produced in the urothelium may also enhance sensory mechanisms in the bladder.

The presence of muscarinic and nicotinic receptors in the urothelium has attracted interest in the role of acetylcholine as a chemical mediator of neural–urothelial interactions (Hawthorn et al., 2000; Templeman et al., 2002; Beckel et al., 2004; 2005a; de Groat, 2004). Cholinergic nerves staining for vesicle acetylcholine transporter (VAT) have been detected in close proximity to the urothelial cells in the rat bladder (Beckel et al., 2005b). Exogenous muscarinic and nicotinic cholinergic agonists applied to cultured urothelial cells can elicit an increase in intracellular Ca²⁺ concentration and evoke the release of NO and ATP (Birder et al., 2003; Beckel et al., 2005a). In bladder strips or whole bladder preparations, muscarinic agonists also stimulate the release of a smooth muscle inhibitory factor from the urothelium (Hawthorn et al., 2000). Electrical stimulation of the pelvic nerve or reflex activation of the autonomic nervous system by spinal cord injury (SCI) (Apodaca et al., 2003; Birder, 2005) can elicit changes in urothelial permeability as well as changes in the morphology of the urothelium in the rat raising the possibility that autonomic or sensory nerves make ‘synaptic connections' with the urothelial cells. Further studies are needed to determine if acetylcholine is involved in these connections.

The function of cholinoceptors in the urothelium has also been evaluated by testing the effects of intravesically administered cholinergic agonists and antagonists on voiding function in cats and rats (Beckel et al., 2004; de Groat et al., 2004; Kim et al., 2004; Ungerer et al., 2005). Intravesical application of nicotine in the rat elicits two effects: a decrease in the frequency of reflex micturition in low concentrations and an increase in frequency in high concentrations. The inhibitory effect was blocked by methyllycaconitine, an antagonist of α7 nicotinic receptors, whereas the facilitatory effect was blocked by hexamethonium, an antagonist of α3-type nicotinic receptors. Methyllycaconitine alone did not alter reflex bladder activity, whereas hexamethonium alone decreased reflex bladder activity suggesting the existence of a tonically active nicotinic facilitatory mechanism. Nicotine also increased intracellular Ca²⁺ in cultured urothelial cells by activating hexamethonium-sensitive receptors. These data coupled with the results of real-time polymerase chain reaction (RT PCR) experiments that revealed the expression of multiple subtypes of nicotinic receptors in rat urothelial cells (α3, α5, α7, β3 and β4) raise the possibility that sensory mechanisms in the urothelium are modulated by complex nicotinic mechanisms (Beckel et al., 2005a).

In chronic spinal cord-injured cats, intravesical infusion of carbachol, a muscarinic-nicotinic agonist, as well as oxotremorine methiodide, a quaternary muscarinic agonist that should have a relatively low ability to penetrate the urothelial barrier, decreased bladder capacity and enhanced the number of premicturition contractions (Ungerer et al., 2005) during cystometrograms, but did not alter the amplitude of micturition contractions. These effects were blocked by intravesical administration of atropine sulphate or the quaternary analogue, atropine methylnitrate. Intravesical administration of neostigmine methylsulphate, a quaternary anticholinesterase agent, mimicked the facilitatory effects of muscarinc agonists. The effects of neostigmine were blocked by atropine. These results indicate that activation of muscarinic receptors in the urothelium or in suburothelial afferent nerves facilitates the spinal micturition reflex mediated by C-fibre afferent nerves (Cheng et al., 1999). In the rat, a similar facilitation of bladder activity induced by intravesically administered muscarinic agonists has been reported (de Groat et al., 2004; Kim et al., 2004).

Urothelial cells express the various proteins necessary for the synthesis and storage of acetylcholine including the plasma membrane choline transporter, choline acetyltransferase and the VAT as well as the enzyme responsible for the metabolism of acetylcholine (acetylcholinesterase) (Beckel et al., 2005b). In addition, there are reports that acetylcholine is released from the bladder urothelium in rats (Klapproth et al., 1997; Beckel et al., 2004) and humans (Yoshida et al., 2004). Synthesis and release of acetylcholine has also been reported in epithelial cells in the lung (Proskocil et al., 2004). Thus, acetylcholine released from urothelial cells may function as an autocrine factor that acts on cholinoceptors on the urothelial cells to release other transmitters or to modify urothelial cell functions. Alternatively, because afferent nerves express cholinoceptors, the acetylcholine released from urothelial cells may act to alter afferent nerve excitability (de Groat, 2004). The clinical effect of antimuscarinic agents to decrease sensory symptoms in overactive bladder may be related, in part, to a block of muscarinic receptors in the urothelium or afferent nerves.

The neural pathways controlling LUT function are organized as simple on–off switching circuits (Figure 2) that maintain a reciprocal relationship between the urinary bladder and urethral outlet. The principal reflex components of these switching circuits are listed in Table 2 and illustrated in Figure 3. Intravesical pressure measurements during bladder filling in both humans and animals reveal low and relatively constant bladder pressures when bladder volume is below the threshold for inducing voiding (Figure 2a) (Morrison et al., 2005). The accommodation of the bladder to increasing volumes of urine is primarily a passive phenomenon dependent upon the intrinsic properties of the vesical smooth muscle and quiescence of the parasympathetic efferent pathway. In addition, in some species, urine storage is also facilitated by sympathetic reflexes that mediate an inhibition of bladder activity, closure of the bladder neck and contraction of the proximal urethra (Table 2, Figure 3a). During bladder filling, the activity of the urethral sphincter electromyogram (EMG) also increases (Figure 2b), reflecting an increase in efferent firing in the pudendal nerve and an increase in outlet resistance that contributes to the maintenance of urinary continence.

Figure 2 Combined cystometrograms and sphincter electromyograms (EMG) comparing reflex voiding responses in an infant (a) and in a paraplegic patient (c) with a voluntary voiding response in an adult (b). The abscissa in all records represents bladder volume in (more ...)

Table 2 Reflexes to the lower urinary tract

Figure 3 Diagram showing neural circuits controlling continence and micturition. (a) Urine storage reflexes. During the storage of urine, distention of the bladder produces low-level vesical afferent firing, which in turn stimulates (1) the sympathetic outflow (more ...)

The storage phase of the urinary bladder can be switched to the voiding phase either involuntarily (reflexly) or voluntarily (Figure 2). The former is readily demonstrated in the human infant (Figure 2a) when the volume of urine exceeds the micturition threshold. At this point, increased afferent firing from tension receptors in the bladder produces firing in the sacral parasympathetic pathways and inhibition of sympathetic and somatic pathways. The expulsion phase consists of an initial relaxation of the urethral sphincter (Figure 2a) followed by a contraction of the bladder, an increase in bladder pressure, and flow of urine. Relaxation of the urethral outlet is mediated by activation of a parasympathetic reflex pathway to the urethra (Table 2) that triggers the release of NO, an inhibitory transmitter (Andersson, 1993), as well as by removal of adrenergic and somatic excitatory inputs to the urethra.

The reflex circuitry controlling micturition consists of four basic components: spinal efferent neurones, spinal interneurones, primary afferent neurones and neurones in the brain that modulate spinal reflex pathways. New research methods, including transneuronal virus tracing (Figures 4 and 5) (Nadelhaft et al., 1992; Vizzard et al., 1995), measurements of gene expression (Figure 5b) (Birder & de Groat, 1993; Birder et al., 1999) and patch-clamp recording in spinal cord slice preparations (Araki & de Groat, 1996; 1997), have recently provided new insights into the morphological and electrophysiological properties of these reflex components.

Figure 4 Transneuronal virus tracing of the central pathways controlling the urinary bladder of the rat. Injection of PRV into the wall of the urinary bladder leads to retrograde transport of virus (dashed arrows) and sequential infection of postganglionic neurones, (more ...)

Figure 5 Comparison of the distribution of bladder afferent projections to the L6 spinal cord of the rat (a), with the distribution of c-fos-positive cells in the L6 spinal segment following chemical irritation of the LUT of the rat (b), and the distribution of (more ...)

Efferent neurons Parasympathetic preganglionic neurones are located in the intermediolateral grey matter (laminae V–VII) in the sacral segments of the spinal cord (Figure 4), whereas sympathetic preganglionic neurones are located in medial (lamina X) and lateral sites (laminae V–VII) in the rostral lumbar spinal cord. EUS motoneurones are located in lamina IX in Onuf's nucleus (Thor et al., 1989; de Groat et al., 2001; Morrison et al., 2005). Parasympathetic preganglionic neurones and EUS motoneurones send dendrites to similar regions of the spinal cord (laminae I, V–VII and X) indicating that these sites contain important pathways for coordinating bladder and sphincter function.

Afferent projections in the spinal cord Afferent pathways from the LUT project to discrete regions of the dorsal horn that contain interneurones and efferent neurones innervating the LUT (Morgan et al., 1981; Steers et al., 1991). Afferent pathways from the urinary bladder project into Lissauer's tract at the apex of the dorsal horn and then pass rostrocaudally, giving off collaterals that extend laterally and medially through the superficial layer of the dorsal horn (lamina I) into the deeper layers (laminae V–VII and X) at the base of the dorsal horn (Figure 5a). The lateral pathway terminates in the region of the sacral parasympathetic nucleus and also sends some axons to the dorsal commissure (Figure 5a). Pudendal afferent pathways from the urethra and urethral sphincter exhibit a similar pattern of termination in the sacral spinal cord (Thor et al., 1989; de Groat et al., 2001). The overlap of bladder and urethral afferents in the lateral dorsal horn and dorsal commissure indicates that these regions are likely to be important sites of viscerosomatic integration and involved in coordinating bladder and sphincter activity.

Spinal interneurones As shown in Figures 4 and 5, interneurones retrogradely labelled by injection of pseudorabies virus (PRV) into the urinary bladder or urethra of the rat are located in regions of the spinal cord receiving afferent input from the bladder (Nadelhaft et al., 1992; Vizzard et al., 1995). Large populations of interneurones are located just dorsal and medial to the preganglionic neurones as well as in the dorsal commissure and lamina I (Figure 5c).

The spinal neurones involved in processing afferent input from the LUT have been identified by the expression of the immediate early gene, c-fos (Figure 5b). In the rat, stimulation of the bladder and urethra increases the levels of Fos protein primarily in the dorsal commissure, the superficial dorsal horn and in the area of the sacral parasympathetic nucleus (Figure 5b) (Birder & de Groat, 1993; Birder et al., 1999). Some of these interneurones send long projections to the brain (Birder et al., 1999), whereas others make local connections in the spinal cord and participate in segmental spinal reflexes (Araki & de Groat, 1996). The former are involved in transmitting sensory input to supraspinal centres for subsequent relay to the cerebral cortex or to micturition reflex circuits in the brainstem. Ascending axons terminate in several areas, including the periaqueductal grey (PAG) (Blok et al., 1995) and the gracile nucleus. It is believed that neurones in the PAG relay information to the pontine micturition centre (PMC) and initiate the micturition reflex (Figure 3b). Projections to the gracile nucleus carry nociceptive signals which are eventually routed to the thalamus and cortex.

Patch-clamp recordings from parasympathetic preganglionic neurones in the neonatal rat spinal slice preparation have revealed that interneurones located immediately dorsal and medial to the parasympathetic nucleus make direct monosynaptic connections with the preganglionic neurones. Microstimulation of interneurones in both locations elicits glutamatergic, N-methyl-D-aspartate (NMDA), and non-NMDA excitatory postsynaptic currents in preganglionic neurones (Araki & de Groat, 1996; 1997). Stimulation of a subpopulation of medial interneurones elicits GABAergic and glycinergic inhibitory postsynaptic currents (Araki, 1994). Stimulation of neurons in the dorsal commissure also elicits monosynaptic and polysynaptic glutamatergic excitatory inputs to the preganglionic neurons (Miura et al., 2003). Thus, local interneurones are likely to play an important role in both excitatory and inhibitory reflex pathways controlling the preganglionic outflow to the LUT.

Other anatomical studies in which anterograde tracer substances were injected into brain areas and then identified in terminals in the spinal cord are consistent with the virus tracing data. Tracer injected into the paraventricular nucleus of the hypothalamus labelled terminals in the sacral parasympathetic nucleus as well as the sphincter motor nucleus (Holstege & Mouton, 2003). On the other hand, neurones in the anterior hypothalamus project to the PMC. Neurones in the PMC in turn project primarily in the lateral funiculus to the sacral parasympathetic nucleus and the lateral edge of the dorsal horn and the dorsal commissure, areas containing dendritic projections from preganglionic neurones, sphincter motoneurones and afferent inputs from the bladder. Patch-clamp studies revealed that lumbosacral preganglionic neurons in the neonatal rat spinal cord receive monosynaptic and polysynaptic glutamatergic excitatory inputs from axons in the lateral funiculus (Miura et al., 2001). Conversely, projections from neurones in the lateral pons terminate rather selectively in the sphincter motor nucleus. Thus, the sites of termination of descending projections from the PMC are optimally located to regulate reflex mechanisms at the spinal level.

Sympathetic reflex activity is elicited by a sacrolumbar intersegmental spinal reflex pathway that is triggered by vesical afferent activity in the pelvic nerves (Figure 3a) (de Groat & Lalley, 1972; de Groat & Theobald, 1976). The reflex pathway is inhibited when bladder pressure is raised to the threshold for producing micturition. This inhibitory response is abolished by transection of the spinal cord at the lower thoracic level, indicating that it originates at a supraspinal site, possibly the PMC. Thus, the vesicosympathetic reflex represents a negative feedback mechanism that allows the bladder to accommodate larger volumes (Figure 3).

Sphincter-to-bladder reflexes may also contribute to urine storage because afferent activity arising in the striated sphincter muscles during contractions can suppress reflex bladder activity and in turn increase bladder capacity. Studies in cats and monkeys revealed that direct electrical stimulation of the sphincter muscles or electrical stimulation of motor pathways to induce a contraction of the sphincters suppresses reflex bladder activity (McGuire et al., 1983; de Groat et al., 2001). Similar inhibitory responses are elicited by electrical stimulation of afferent axons in the pudendal nerve, some of which must arise in the sphincter muscles. Electrophysiological studies in cats showed that activation of pudendal afferent input from various sites including penis, vagina, anal canal/rectum and sphincters inhibited micturition by suppressing interneuronal pathways in the spinal cord as well as by directly inhibiting parasympathetic preganglionic neurons (de Groat, 1978; de Groat et al., 1981). The inhibitory responses occur at low frequencies of stimulation (1–5Hz), whereas excitatory responses, which are due in part to activation of perineal cutaneous afferent axons, are elicited by high frequencies of stimulation (20–40Hz) (Tai et al., 2005).

Figure 6 Diagram showing the organization of the parasympathetic excitatory reflex pathway to the detrusor muscle. Scheme is based on electrophysiologic studies in cats. In animals with an intact spinal cord, micturition is initiated by a supraspinal reflex pathway (more ...)

Chronic spinal injury in humans also causes the emergence of an unusual bladder reflex that is elicited in response to infusion of cold water into the bladder (Geirsson et al., 1993). The response to cold water does not occur in normal adults but does occur in (1) infants, (2) patients with multiple sclerosis and Parkinson's disease and (3) elderly patients with hyperactive bladders. Studies in animals indicate that cold temperature activates receptors in bladder C-fibre afferents and urothelial cells (Figure 6) (Fall et al., 1990; Stein et al., 2004). The presence of the cold reflex in infants, its disappearance with maturation of the nervous system and its re-emergence under conditions in which higher brain functions are disrupted suggests that it may reflect a primitive spinal involuntary voiding reflex activated by C-fibre afferents. Evidence of the contribution of C-fibre bladder afferents to bladder hyperactivity and involuntary voiding has been obtained in clinical studies in which capsaicin or resiniferatoxin, C-fibre afferent neurotoxins, was administered intravesically to patients with multiple sclerosis or SCIs and hyperreflexic bladders. In these patients, capsaicin increased bladder capacity and reduced the frequency of incontinence (Chancellor & de Groat, 1999; Szallasi & Fowler, 2002).

The emergence of C-fibre bladder reflexes seems to be mediated by several mechanisms including changes in central synaptic connections and alterations in the properties of the peripheral afferent receptors that lead to sensitization of the ‘silent' C fibres and the unmasking of responses to mechanical stimuli (de Groat, 1995; de Groat & Yoshimura, 2005). In rats, it has been shown that bladder afferent neurons undergo both morphologic (neuronal hypertrophy) (Kruse et al., 1995) and physiologic changes (upregulation of tetrodotoxin (TTX)-sensitive Na⁺ channels and downregulation of TTX-resistant Na⁺ channels) following SCI (Yoshimura & de Groat, 1997). It has been speculated that this neuroplasticity is mediated by the actions of neurotrophic factors such as NGF released within the spinal cord or the urinary bladder. The production of neurotrophic factors including NGF increases in the bladder after SCI (Vizzard, 2005) and chronic administration of NGF into the bladder of rats induced bladder hyperactivity and increased the firing frequency of dissociated bladder afferent neurons. On the other hand, intrathecal application of NGF antibodies to neutralize NGF in the spinal cord suppressed detrusor hyper-reflexia and detrusor-sphincter dyssynergia in spinal cord-injured rats (Seki et al., 2002; 2004).

The recent development of duloxetine (Dmochowski et al., 2003; Millard et al., 2004), a serotonin-norepinephrine reuptake inhibitor, for the treatment of stress urinary incontinence has focused attention on the role of serotonin and norepinephrine in the control of LUT function. Nerves containing these transmitters are localized in sympathetic and parasympathetic nuclei in the lumbosacral spinal cord as well as in Onuf's nucleus indicating that they are involved in the control of the bladder and the urethral sphincter (de Groat, 2002). Studies in cats in which the bladder was irritated with intravesical infusion of acetic acid indicate that duloxetine inhibits reflex bladder activity and enhances external urethral sphincter activity (Thor & Katofiasc, 1995). The excitatory effects on the sphincter seem to be mediated by 5-HT₂ receptors and α₁ adrenoceptors, whereas the inhibitory effects on the bladder seem to be mediated by 5-HT₁ receptors.

The role of 5-HT₁ receptors in inhibiting bladder activity was confirmed by administration of 8-hydroxy-2-(di-n-propylamino)-tetralin (8-OH-DPAT), a 5-HT_1A receptor agonist. This agent increased bladder capacity in chloralose anesthetized cats in which the bladder was irritated with acetic acid but had only moderate effects on bladder activity in the absence of irritation (Thor et al., 2002). The drug had a facilitator effect on activity of the external urethral sphincter. 8-OH-DPAT also inhibited reflex bladder activity in awake or chloralose-anesthetized, chronic spinal cord-injured (SCI) cats, but did not alter the somato-bladder excitatory reflex induced in SCI cats by tactile stimulation of the perigenital region (Miscik et al., 2003; Gu et al., 2004). The effects of 8-OH-DPAT were blocked by WAY 100635, a 5-HT_1A receptor antagonist which alone had no effect. These results indicate that 8-OH-DPAT acts in the spinal cord to inhibit the micturition reflex triggered by C-fibre bladder afferent axons and has much less effect on the spinobulbospinal reflex elicited by A-δ afferents. It seems likely that inhibition occurs at a proximal site on the reflex pathway at the primary afferent terminals or at an interneuronal level rather than on the efferent limb of the reflex (i.e., preganglionic and postganglionic neurons) because the efferent limb should be common to both the perigenital stimulation-evoked and bladder distension-evoked reflexes and only the latter was inhibited by 8-OH-DPAT.

In rats the modulatory effects of 5-HT on LUT function are different than those in cats (de Groat, 2002). Intravenous, intrathecal or intracerebroventricular administration of 8-OH-DPAT facilitates the micturition reflex and intravenous administration enhances spontaneous and reflex activity of the external urethral sphincter (Lecci et al., 1992; Conley et al., 2001; Chang et al., 2004). On the other hand, WAY 100635, a 5-HT_1A receptor antagonist administered via various routes depressed reflex bladder and sphincter activity (Testa et al., 1999; Conley et al., 2001; Kakizaki et al., 2001). It has been speculated that WAY 100635 blocks 5-HT_1A autoinhibitory receptors in raphe neurons in the brain stem and enhances raphe neuron firing which in turn increases release of 5-HT in the spinal cord (Testa et al., 1999; Kakizaki et al., 2001). In the spinal cord it is thought that 5-HT released from raphe bulbospinal axons activates 5-HT_2C receptors on inhibitory neurons to suppress the micturition reflex. NAD-299, another selective 5-HT_1A receptor antagonist, had a similar inhibitory response on micturition in the rat (Pehrson et al., 2002). The intracerebroventricular injection of a 5-HT₇ receptor antagonist also increased the volume threshold for triggering micturition in the anesthetized rat (Read et al., 2003). Thus, micturition in the rat is sensitive to both excitatory and inhibitory serotonergic mechanisms, whereas in the cat serotonin appears to act primarily to promote urine storage by enhancing sphincter activity and suppressing bladder activity. The similarities in the effects of duloxetine on sphincter activity in cat and human indicate that micturition in the cat may be a useful model for developing centrally acting serotonergic agents for the treatment of LUT dysfunction (Thor et al., 2002; de Groat, 2002; Burgard et al., 2003).

Activation of cholinoceptors in the rat brain also elicits mixed effects. Nicotinic agonists administered intracerebroventricularly suppress voiding in awake or anesthetized rats (Lee et al., 2003), whereas activation of muscarinic receptors stimulates bladder activity during bladder filling but suppresses voluntary voiding (Ishiura et al., 2001; Nakamura et al., 2003). Atropine blocked both the inhibitory and excitatory effects of muscarinic agonists. Intracerebroventricular administration of atropine alone increased bladder capacity and reduced voiding efficiency indicating that muscarinic excitatory mechanisms in the brain are tonically active.

The effects of dopamine on the central control of micturition are complex. Inhibitory effects of dopamine are mediated by D₁-like (D₁ and D₅), and the facilitatory effects are mediated by D₂-like (D₂, D₃ and D₄) receptor subtypes (Yoshimura et al., 1993; 1998; Yokoyama et al., 1999; 2001; 2002; de Groat & Yoshimura, 2001; Seki et al., 2001). Loss of forebrain dopaminergic mechanisms in patients with idiopathic Parkinson's disease is associated with bladder hyperactivity (de Groat & Yoshimura, 2001). Interactions between dopamine and NMDA glutamatergic facilitatory mechanisms also are important in the emergence of bladder overactivity after middle cerebral artery occlusion in the rat (Yokoyama et al., 2002).

The functions of the LUT to store and periodically eliminate urine are regulated by a complex neural control system that performs like a simple switching circuit to maintain a reciprocal relationship between the bladder and urethral outlet. The switching circuit is modulated by several neurotransmitter systems and is therefore sensitive to a variety of drugs and neurologic diseases. A more complete understanding of the neural mechanisms involved in bladder and urethral control will no doubt facilitate the development of new diagnostic methods and therapies for LUT dysfunction.