Pyeloureteric peristalsis: role of atypical smooth muscle cells and interstitial cells of Cajal-like cells as pacemakers

Abstract

Pyeloureteric peristalsis has long been considered to be triggered by pacemaker atypical smooth muscle cells (SMC) located in the proximal regions of the renal pelvis. However, interstitial cells with many of the morphological features and c-Kit immuno-reactivity of interstitial cells of Cajal (ICC), the established pacemaker cells in the intestine, have been demonstrated to be present in small numbers within the ureteropelvic junction (UPJ) of many mammals. Freshly isolated ICC-like cells (ICC-LC) of the mouse UPJ also display autorhyhmicity. This review discusses the notion that ureteric peristalsis depends on the presence of both atypical SMC and ICC-LC which form separate but interconnected networks that drive electrically quiescent typical SMC. In contrast to the intestine or prostate, all regenerative potential discharge in the mouse UPJ is abolished by the L-type Ca²⁺ channel blocker nifedipine revealing a fundamental pacemaker signal. Whether these pacemaker transients arise from atypical SMC or ICC-LC or both has yet to be established. We speculate that the presence of spontaneously active ICC-LC in the distal regions of the UPJ maintains rudimentary peristaltic waves and movement of urine towards the bladder after pyeloureteral obstruction or pyeloplasty and disconnection from the proximal pacemaker drive.

In the upper urinary tract, the mechanisms by which urine is transported from the kidney to the bladder remain little understood. For the last 35 years it has been thought that pyeloureteric autorhythmicity arises in specialized electrically active atypical smooth muscle cells (SMC) that have many of the morphological and electrical characteristics of cardiac sino-atrial cells and are located predominantly in the proximal regions of the ureteropelvic junction (UPJ). However, increasing evidence indicates that ICC-like cells (ICC-LC), displaying many of the morphological features of intestinal ICC and immuno-reactivity to antibodies raised against the c-Kit proto-oncogene, are present in the UPJ of a number of mammals (Lang & Klemm, 2005). c-Kit immuno-reactivity also appears developmentally at the same time as coordinated unidirectional peristaltic contractions in mouse embryonic ureters in organ culture. Moreover, the development of the ureteric structure and peristaltic contractions in these explants can be prevented upon exposure to the c-Kit antibody, ACK45 (David et al. 2005), suggesting that ICC-LC have an important role in pyeloureteric peristalsis.

This review examines a new model in which atypical SMC and ICC-LC both play a pacemaker role in the initiation and propagation of pyeloureteric peristalsis. It discusses the distributions and electrical properties of these two pacemaker populations. In contrast with the prostate (Exintaris et al. 2002; Lang et al. 2006) and rabbit urethra (McHale et al. 2006), but in common with the guinea pig bladder (Bramich & Brading, 1996) and urethra (Hashitani & Edwards, 1999), almost all spontaneous regenerative electrical activity in the UPJ recorded with intracellular microelectrodes is abolished in the presence of the L-type Ca²⁺ channel blocker nifedipine (Lang et al. 1995, 1998). Preliminary evidence suggests that the expression profile of voltage- and Ca²⁺-activated ion channels in freshly isolated ICC-LC (Lang et al. 2007) differs from the channel profiles in ICC-LC of other visceral smooth muscle organs and that Ca²⁺ mobilization from IP₃-dependent stores (Lang et al. 2002b) and the opening of cation-selective channels are involved in autorhythmicity (Lang et al. 2007).

The ability of the human ureter to maintain rudimentary peristaltic waves and movement of urine from the papilla to the bladder after pyeloplasty, uretero-ureteral anastomosis or ureteral obstruction (Djurhuus et al. 1976; Djurhuus & Constantinou, 1982; Tillig et al. 2004) in the absence of a proximal pacemaker drive may well arise from the recruitment of ICC-LC (Schwentner et al. 2005) as pacemakers in regions distant from the papilla rather than the activation of the SMC layer itself (Santicioli & Maggi, 1998). As such, pyeloureteric ICC-LC with their unique pacemaker mechanisms could provide a selective pharmacological target when considering non-surgical interventions to alleviate hydronephrosis arising from UPJ remodelling during and after ureteric blockade or pyeloplasty.

Structure of the mammalian upper urinary tract

In most mammals, including mouse, guinea pig, rat, sheep, dog and cat, the kidney contains a single papilla which is surrounded by a funnel-shaped calyx or renal pelvis. Bundles of long spindle-shaped, contractile filament-rich ‘typical’ SMC form a ‘basket weave’ inner layer which originates near the base of the papilla and extends through to the ureter (Gosling & Dixon, 1974; Dixon & Gosling, 1982). In contrast, the outer layer consists of relatively few morphologically distinct atypical SMC, the space between cells being filled with collagen-rich connective tissue and axon bundles (Gosling & Dixon, 1974; Dixon & Gosling, 1990; Klemm et al. 1999) (Fig. 1Ba). Atypical SMC form a diffuse net that wraps around the most proximal regions of the papilla and terminates at the UPJ (Gosling & Dixon, 1974). In human and pig, the kidney is multipapillate so that a number of major and minor calyces fuse to form a separate renal pelvis which extends into the ureter. Atypical SMC alone form the muscle coat of the minor calyces (Dixon & Gosling, 1973, 1982, 1990; Gosling & Dixon, 1974) which extends distally to create an inner layer within the wall of the major calcyces and renal pelvis, but not the UPJ (Dixon & Gosling, 1973, 1982, 1990). In both uni- and multipapillate kidneys, the number of typical SMC increases with distance from the papilla base so that there is a thickening of the pelvic wall as these SMC are increasingly more tightly packed into closely associated bundles (Fig. 1Ca) (Dixon & Gosling, 1982, 1990; Klemm et al. 1999).

Figure 1. Summary of current knowledge on the morphology and electrophysiology of the upper urinary tract of uni-papillate mammals.

Frequency of contractions (A) has been correlated with the relative distribution of atypical (Ba) and typical (Ca) smooth muscle cells (SMC). In the guinea pig, transient potentials (Bb) and driven action potentials (Cb) are recorded in short (Bc) and long (Cc) spindle-shaped SMC, respectively. D, ICC-like cells (ICC-LC) (Da), c-Kit-positive cells and intermediate action potentials (Db) occur in small numbers along the length of the ureteropelvic junction. Intermediate action potentials (Db) are recorded in fusiform cells (Dc). Scale bars: Ba, Ca, Da: 2 μm; Bc and Cc: 50 μm; Dc: 10 μm. PCJ, pelvi-calyceal junction; RP, renal pelvis.

Pelviureteric peristalsis

Ureteric peristaltic contractions in vivo are little affected by blockers of parasympathetic and sympathetic innervation (Golenhofen & Hannappel, 1973), but reduced in vitro by blockers of sensory nerve function or prostaglandin production (Maggi & Giuliani, 1992; Teele & Lang, 1998; Santicioli & Maggi, 1998; Davidson & Lang, 2000; Lang et al. 2002a). Contractile activity originates at the border where the calyces attach to the base of the papilla (pelvi-calyceal junction, PCJ) and propagate distally to the renal pelvis. In both uni-calyceal and multicalyceal kidneys, circumferentially cut strips of muscle wall, dissected from different regions equally distant from the papilla base, contract at the same frequency (Constantinou & Yamaguchi, 1981; Yamaguchi & Constantinou, 1989). In contrast, contraction frequency decreases as strips are dissected from sites increasingly distant from the papilla base. (Hannappel & Lutzeyer, 1978; Hannappel et al. 1982; Potjer et al. 1992; Patacchini et al. 1998; Lang et al. 1998, 2002a) The ureter in the multipapillate kidney contracts spontaneously in vitro; while spontaneous contractility in the ureter of uni-papillate kidneys is only observed in vitro if the proximal renal pelvis remains connected (Golenhofen & Hannappel, 1973; Hannappel & Golenhofen, 1974; Teele & Lang, 1998; Davidson & Lang, 2000).

In both uni- and multicalyceal kidneys, a single pacemaker region on the pelvi-calyceal border is responsible for every wave of activation which conducts radially across the pelvis to form a crescent-shaped wave that conducts distally towards the UPJ. In uni-papillate kidneys, this pacemaker region shifts spontaneously along the pelvi-calyceal border (Shimizu, 1978; Lammers et al. 1996). In multipapillate kidneys this ‘primary’ pacemaker shifts spontaneously between calyces (Yamaguchi & Constantinou, 1989). Block or delay of the axial conduction of the wave of excitation can occur at any point or time resulting in a pathway of conduction that is neither constant, nor direct; but appears to meander throughout the renal pelvis to the UPJ (Lammers et al. 1996). The mechanisms underlying this plasticity remain little understood but represent potential points of pharmacological intervention during clinical conditions such as kidney stones, vesicoureteric reflux or infection.

Pacemaking in the UPJ

In vivo (Edmond & Ross, 1970) and in vitro (Constantinou, 1974; Djurhuus et al. 1977; Morita et al. 1981) pressure recordings and intracellular and extracellular electrophysiological investigations (Kuriyama & Tomita, 1971; Hannappel & Golenhofen, 1974; Zawalinski et al. 1975) have established that every peristaltic contraction of the renal pelvis and ureter is preceded by a complex ‘ureteric’ or ‘driven’ action potential (Fig. 3Aa). Driven action potentials have a time course consisting of an initial spike followed by a long plateau phase (Fig. 3Ab) which, in guinea pig also displays a variable number of additional spikes (Fig. 1Cb) (Zawalinski et al. 1975; Zhang & Lang, 1994; Lang et al. 1995; Lang & Zhang, 1996). In the guinea pig UPJ, driven action potentials (discharge frequency 3–6 min⁻¹) are always recorded in long (150–400 μm) typical SMC (Fig. 1Cc). The proportion of cells displaying driven action potentials increases with distance from the papilla base, from 75% of cells in the proximal renal pelvis to 98–100% of cells in the distal renal pelvis and ureter (Fig. 1C) (Klemm et al. 1999).

Figure 3. Typical intracellular microelectrode recordings from the mouse UPJ: effects of caffeine.

A, in all tissues muscle contractions were associated with regenerative action potential discharge consisting of an initial spike followed by a quiescent plateau and a small afterhyperpolarization 500–1000 ms in duration. The membrane potential between action potentials was either relatively quiescent (Aa) or displayed additional spontaneous transient depolarizations (STDs) (Ab). Single STDs seldom triggered muscle contraction but were observed to often occur near-synchronously, sum and trigger a regenerative action potential that caused muscle contraction. B, caffeine (1 mm) causes a reversible reduction in the amplitude and frequency of the spontaneous regenerative electrical events in strips of UPJ bathed in 1 μm nifedipine. Ca–c, sections of trace indicated by a–c in B displayed on an expanded time base for better comparison.

Extensive recordings from single typical SMC enzymatically isolated from the renal pelvis and ureter of guinea pig (Imaizumi et al. 1989, 1990; Lang, 1989, 1990), rat (Smith et al. 2002) and mouse (Lang et al. 2007) have established the presence of a small slowly inactivating, nifedipine-sensitive ‘L-type’ Ca²⁺ channel current (I_Ca), a distinctive voltage-activated 4-aminopyridine-sensitive K⁺ channel current (I_KV) and spontaneous transient outward currents (STOCs) arising from the opening of iberiotoxin-sensitive large conductance Ca²⁺-activated K⁺ (BK_Ca) channels. SMC of rat ureter also display a delayed rectifier K⁺ current and a Ca²⁺-activated Cl⁻ current which are not evident in guinea pig ureteric myocytes (Smith et al. 2002).

Recently it has been demonstrated that the hyperpolarizing action of increased STOC discharge in the ureter is the main mechanism underlying the relatively long refractory periods observed after action potential discharge (Burdyga & Wray, 2005). Burdgyga & Wray suggest that Ca²⁺ entry during an action potential increases the internal Ca²⁺ concentration ([Ca²⁺]_i) which loads Ca²⁺ stores within the sarcoplasmic reticulum. This rise in stored Ca²⁺ promotes an increase in the frequency of localized release of Ca²⁺ (Ca²⁺ sparks) which activates BK_Ca channels simultaneously in the form of STOCs that keep the ureteric SMC from threshold. Ureteric myocytes do not display any spontaneous inward currents that could be ascribed as representing pacemaker activity.

Atypical SMC

The decreasing presence of ‘atypical’ SMC with distance from the papillate base has been correlated with the decreasing frequency of contraction (Fig. 1A and B) to suggest that atypical SMC are the essential pacemaker cells of pelviureteric peristalsis (Dixon & Gosling, 1973, 1982; Gosling & Dixon, 1974; Morita et al. 1981). Investigators have envisaged that autorhythmicity within the upper urinary tract involves a ‘chain of coupled linear oscillators’ with the most proximal oscillator firing at the highest frequency (Constantinou & Yamaguchi, 1981). Diuresis is suggested to distend the muscle wall which increases, in an unknown manner, the ‘coupling’ of the pacemaker regions in the longitudinal direction until the highest frequency oscillator in the proximal regions entrains the oscillators in the distal regions of the renal pelvis (Hannappel et al. 1982). This model implies that the intrinsic frequency of pacemaking atypical SMC decreases with distance from the papilla base which would require morphologically similar cells having different expression profiles of voltage- and Ca²⁺-activated ion channels and pacemaking apparatus as their location increases with distance from the papilla. In 1999 we proposed a simpler explanation, that the frequency gradient of contraction of the renal pelvis arises from an increasing ratio of typical: atypical SMC with distance from the papilla base (Klemm et al. 1999). We envisaged that each SMC type displays a distinct expression profile of ion channels and that the increasing number of typical SMC with their BK_Ca and K_V channels along the pelvic wall of the UPJ explains the increasing negativity of the membrane potential (Lang et al. 1995). The increasing influence of the intrinsic refractoriness of typical SMC as their relative number increases, in the presence of a constant frequency pacemaker drive from a diminishing number of atypical SMC, reduces the muscle wall excitability and therefore the frequency of action potential discharge and muscle contraction with distance from the papilla base (Klemm et al. 1999).

In addition to ascribing a pacemaker role in driving pyeloureteric peristalsis, the rhythmic contractile activity of the atypical SMC network has been suggested to play a ‘milking’ role, aiding the secretion of urine from the renal medulla. Dwyer & Schmidt-Nielsen (2003) suggest that contractions of the muscle wall at the papilla base of the hamster kidney create oscillating hydrostatic and osmotic gradients that are essential in the initial emptying of the papilla and the movement of fluids through the loops of Henle and collecting ducts. How atypical SMC communicate between themselves, with neighbouring ICC-LC and typical SMC to differentiate and coordinate this ‘milking’ action while promoting the movement of urine towards the bladder has yet to be established.

There have been very few studies of the electrical activity of atypical SMC. Figure 1A and B illustrates examples of contractions and electrical discharge recorded in the finger-like tissue septa of the PCJ that penetrate into the renal substance. These strips contract (at a frequency of 8 min⁻¹) with an amplitude 1/5 of that recorded in strips of proximal renal pelvis. Intracellular microelectrode impalements of these strips reveal that high frequency (8–12 min⁻¹) transient potentials (Tsuchida & Suzuki, 1992; Lang et al. 1995) of a simple waveform (Fig. 1Bb) are recorded in short (90–230 μm) spindle-shaped SMC (Fig. 1Bc) (Klemm et al. 1999). These transient potentials are recorded less often in the proximal renal pelvis and never in the distal renal pelvis and ureter (Fig. 1B) (Lang et al. 1995, 1998; Klemm et al. 1999). To date, there have not been any reports of atypical SMC at the single cell or ion channel level. However, collagenase treatment of the most proximal regions of the PCJ, particularly in the rat (Lang et al. 2001), may well produce a suspension of single cells that are mostly atypical SMC suitable for experimentation.

ICC-LC in the UPJ

Over the last 40 years several reports have described the presence of a secondary pacemaker region in distal regions of the UPJ (Weiss et al. 1967; Hannappel et al. 1982). In 1999 we described a population of electrically active cells in the renal pelvis which may well be responsible for this additional autorythmicity (Klemm et al. 1999). Under the electron microscope, these cells displayed many of the morphological features repeatedly used to distinguish ICC from fibroblasts in the intestinal tract (Fig. 1Da) (Lang & Klemm, 2005). In guinea pig, these ICC-LC lie within in the lamina propria of the PCJ and the renal pelvis, but not in the ureter. These ICC-LC form close appositions with themselves and with neighbouring typical and atypical SMC, suggesting electrical connectivity and conduction (Klemm et al. 1999). In contrast, only a few ICC-LC are present in the rat renal pelvis (Lang et al. 2001).

Staining of fixed and unfixed preparations with antibodies raised against c-Kit has become a standard means of recognizing the presence of ICC in the gastrointestinal tract and a number of other visceral organs (Lang & Klemm, 2005). In our hands, immuno-histochemical analysis of acetone-fixed preparations of the UPJ from the guinea pig failed to display any c-Kit staining that was not attributed to the presence of mast cells (Klemm et al. 1999). More recently, a number of investigators have demonstrated the presence of a sparsely distributed network of bipolar and stellate shaped c-Kit-positive cells within the lamina propria and muscle layers of the renal pelvis and proximal ureter in a number of species including human (van der Aa et al. 2004; Metzger et al. 2004), mouse (Fig. 2) (Pezzone et al. 2003; Lang & Klemm, 2005), swine, cow, dog, cat, rabbit and rat (Metzger et al. 2005). Double labelling experiments have discounted the possibility that these c-Kit-positive cells are mast cells, fibroblasts or macrophages (Fig. 2) (Metzger et al. 2004). It is becoming clear that ICC and ICC-LC in various urogenital and intestinal tissues can also be divided into a number of subpopulations on the basis of their immuno-reactivity or -negativity to c-Kit, vimentin, actin filaments, ion channel populations (Klemm & Lang, 2002), receptors (van der Aa et al. 2004) and gap junction subunits. It has yet to be determined to what extent these subpopulations are present in the UPJ, or whether they have distinct or differing pacemaker roles in pyeloureteric peristalsis.

Figure 2. Fluorescence micrographs of c-Kit-positive cells in the proximal (A) and distal (B) regions of unfixed whole-mount preparations of mouse UPJ exposed to the extracellularly binding c-Kit antibody H-300 (1: 100 dilution).

Tissues were previously exposed to FITC-dextran FD-70S (1.3 mg ml⁻¹), to label macrophages, and to the vital nuclear stain Hoechst Blue. Note the punctate red immuno-staining on stellate-shaped cells (*), some of which also displayed the green fluorescence of FITC-dextran, particularly in the proximal renal pelvis (RP, arrows). The number and orientation of Hoechst Blue-stained urothelial and SMC nuclei indicates the likely position of c-Kit-positive cells within the UPJ wall. Scale bar, 50 μm.

In guinea pig UPJ, a constant proportion (< 20%) of cells fire action potentials that consist of a single spike followed by a quiescent plateau (Fig. 1Db). These cells have membrane potentials that are more negative than atypical SMC but more positive than typical SMC (Lang et al. 1995; Klemm et al. 1999). When filled with Neurobiotin, cells firing these ‘intermediate’ action potentials were often stellate-shaped, displaying a similar cell-to-cell connectivity and located in the same regions as ICC-LC. In the mouse UPJ, all regenerative action potentials consist of a single spike and quiescent plateau preventing any electrical distinction between intermediate and driven action potentials. Lucifer yellow-filled cells firing these action potentials are occasionally fusiform, but more often bipolar in shape, suggesting that most recordings were made in SMC.

In both guinea pig and mouse UPJ, the membrane potential between action potential discharge in many cells is not electrically quiescent (Fig. 3Aa), as seen when recording driven action potentials in the distal renal pelvis or ureter (Lang et al. 1995). Spontaneous transient depolarizations (STDs) occur at a high frequency (> 30 min⁻¹) and often sum and trigger the discharge of an action potential (Fig. 3Ab and Ca). In guinea pig, we originally envisaged that transient potentials arising in atypical SMC propagate to neighbouring typical SMC and c-Kit-negative ICC-LC, triggering driven and intermediate action potentials, respectively. We suggested that the network of ICC-LC formed as a network of integration, providing a pathway of rapid electrical communication throughout the renal pelvis to drive neighbouring typical SMC bundles (Lang et al. 1998; Klemm et al. 1999).

This model may well be a little too simplistic as we have recently established that single ICC-LC of mouse UPJ have their own autorhythmicity (Fig. 4B). Stellate-shaped ICC-LC of the UPJ (Fig. 4Aa) are readily distinguished from typical SMC (Fig. 4Ab) by the lack of any outward current arising from the opening of iberiotoxin-sensitive BK_Ca channels or 4-aminopyridine-sensitive K_V channels and by the presence of a unique slowly inactivating K⁺ current (Fig. 4Aa) that is only blocked by high concentrations (> 20 mm) of tetraethylammonium (Lang et al. 2007).

Figure 4. Voltage clamp recordings from enzymatically isolated cells of mouse UPJ.

A, comparison of whole-cell membrane currents in single ICC-LC (Aa) and SMC (Ab) of the mouse UPJ. Currents evoked during step depolarizations to potentials every 10 mV between −86 and +46 mV (see inset) from a conditioning potential of −96 mV. B, spontaneous electrical activity in ICC-LC, under voltage clamp, consisted of high frequency small spontaneous transient inward currents (STICs), which summed to form large inward currents (LICs) that lasted for many seconds before termination. Hyperpolarizing steps (−20 mV) before, during and after these LICs revealed that these inward currents arose from a 10–1000 fold increase in the membrane conductance. C, a reversal potential near −10 mV for these spontaneous inward currents (Cc) can be obtained upon subtraction of the membrane current evoked in response to a ramped depolarization (between −95 and +60 mV over 0.5 s) in the absence (Ca and Cb: a and c) from the ramp current in the presence (Ca and Cb: b) of a LIC. Pipette filled with 130 mm potassium gluconate. D, effects of 1 mm caffeine on STICs and LICs in single ICC-LC. Ea–c, sections of trace indicated by a–c in D displayed on an expanded time base for better comparison.

Under current clamp (I = 0), ICC-LC, bathed in 1 μm nifedipine, display STDs of varying amplitudes as well as the occasional long-lasting sustained membrane depolarization (Lang et al. 2007). Under voltage clamp, it is clear that the STDs arise from the discharge of small spontaneous transient inward currents (STICs) which occur in bursts while the sustained depolarizations arise from long-lasting large inward currents (LICs) resulting from the summation of a number of near-synchronous STICs (Fig. 4B) (Lang et al. 2007). LICs have a reversal potential between −20 and −10 mV (Fig. 4C) which is shifted only 10 mV in the positive direction when potassium gluconate (130 mm) in the pipette solution is replaced with KCl. In addition, STICs are little affected by the Cl⁻ channel blockers, 4,4′-diisothiocyanostilbene-2,2′-disulphonic acid (DIDS) or niflumic acid, or the capacitative Ca²⁺ entry blocker LaCl₃ (100 μm) (Lang et al. 2007). Altogether these results suggest that stellate-shaped ICC-LC of the mouse UPJ can generate spontaneous inward currents that arise from the opening of cationic-selective channels. We speculate that these inward currents could well provide a pacemaker drive for neighbouring typical SMC bundles, particularly in the absence of a pacemaker drive from proximal atypical SMC.

Role of Ca²⁺

In contrast to slow waves in the gastrointestinal tract (Hirst & Ward, 2003) and prostate (Exintaris et al. 2002; Lang et al. 2006), both the amplitude and duration of the spontaneous electrical activity in the UPJ are greatly reduced upon blockade of Ca²⁺ entry through L-type Ca²⁺ channels with nifedipine. In guinea pig UPJ, transient potentials and both driven and intermediate action potentials are all blocked by 1 μm nifedipine (Lang et al. 1995; Exintaris & Lang, 1999; Takano et al. 2000). However, I_Ca in single ureteric SMC (Lang, 1990) and the propagated electrical and mechanical response recorded upon electrical stimulation of the intact ureter (Santicioli & Maggi, 1998) are only completely blocked by 3–10 μm nifedipine. In mouse UPJ, both STD and the regenerative action potentials are substantially reduced in amplitude and duration in 1 μm nifedipine (> 10–60 min). Any regenerative electrical and associated contractile activities not blocked by 1 μm nifedipine are blocked by 3–10 μm nifedipine (> 10 min), such that only small (< 10 mV) STDs remain. These residual STDs are reversibly blocked upon prevention of the accumulation of Ca²⁺ into internal stores using cyclopiazonic acid (CPA, 10 μm), the selective blocker of sarcoplasmic/endoplasmic Ca²⁺-ATPase (SERCA) or upon the removal of Ca²⁺ from the external bathing solution (M. A. Tonta, H. Hashitani, H. C. Parkington & R. J. Lang, unpublished observations).

Although it hasn't yet been determined whether STDs are arising from ICC-LC or atypical SMC, or both, it is tempting to suggest that these fundamental events underlie pacemaking in the UPJ. STD discharge appears not to be dependent on Ca²⁺ entry through L-type Ca²⁺ channels, but dependent on the release of Ca²⁺ from internal stores in a manner sustained by an entry of external Ca²⁺ through non-nifedipine-sensitive pathways. However, entrainment or recruitment of neighbouring regions of a particular pacemaker, or neighbouring pacemaker cells, into an effective electrical signal that drives neighbouring typical SMC bundles to generate contraction appears to be sensitive to nifedipine.

Pacemaker mechanisms in the UPJ

The mechanisms involved in the generation and propagation of pacemaker potentials in smooth muscle display considerable variation between tissues and species. In the small intestine, release of Ca²⁺ from IP₃-dependent stores is thought to lead to the activation of mitochondrial Ca²⁺ uptake, the lowering of localized [Ca²⁺]_i near the ICC plasmalemma and the activation of cationic-selective channels to generate the pacemaker potential (Hirst & Ward, 2003). In contrast, in the stomach (Hirst & Ward, 2003) and prostate (Lang et al. 2006) pacemaker potentials arise from an increase in the frequency and summation of unitary potentials (STDs) generated upon the release of Ca²⁺ from IP₃-dependent Ca²⁺ stores and the opening of Ca²⁺-activated Cl⁻ channels. In rabbit urethra ICC-LC, pacemaker potentials are generated by Ca²⁺-sensitive Cl⁻ channels opened when Ca²⁺ is released from ryanodine-sensitive Ca²⁺ stores (McHale et al. 2006).

In guinea pig renal pelvis, the amplitude and frequency of the action potentials underlying muscle contraction are reduced upon blockade of the release of stored Ca²⁺ using blockers of IP₃ formation (2-APB, neomycin) or receptor binding (xestospongin C) and transiently increased and then slowly reduced upon blockade of SERCA with CPA (Santicioli & Maggi, 1998; Lang et al. 2002b). Caffeine (1 mm) reduces the rising and plateau phases of the majority of the spontaneous action potentials recorded in the guinea pig renal pelvis and completely blocks electrical discharge in the remaining (35%) cells (Lang et al. 2002b). In the mouse UPJ, caffeine (1 mm) blocks regenerative action potential discharge in the absence or presence (Fig. 3B and C) of 1 μm nifedipine in a manner associated with a membrane hyperpolarization and a reduction in the amplitude and discharge frequency of STDs. When single freshly isolated ICC-LC of mouse UPJ (bathed in 1 μm nifedipine) (Fig. 4D) are exposed to caffeine (1 mm for 5–10 min, n = 6 cells) LICs are readily abolished and STICs discharge is only partially reduced in a manner that is reversible upon washout (Fig. 4D and E).

Previously, researchers have attributed caffeine's action to (i) an inhibition of IP₃-mediated Ca²⁺ release not involving a depletion of the internal Ca²⁺ stores (Hirst & Ward, 2003), (ii) an increased opening of ryanodine Ca²⁺-release channels, and/or (iii) a decrease in Ca²⁺ entry through L-type Ca²⁺ channels via the inhibition of phosphodiesterases and the elevation of cGMP levels (Lang et al. 2002b). It is clear that the mechanisms of pacemaking generated by ICC-LC and atypical SMC of the UPJ remains little understood. However, this preferential action of caffeine on LIC over STIC discharge in ICC-LC is an interesting portent of future investigation.

Conclusions

The pacemaker-generating mechanisms underlying the spontaneous electrical activity in the UPJ have been confounded by a number of unique properties not encountered during similar studies of spontaneous slow waves and STDs in intestine, urethra or prostate. Previous intracellular microelectrode examinations into renal pelvic pacemaking have assumed the presence of only one pacemaker signal generated by atypical SMC and that the pacemaker mechanisms were likely to be voltage dependent as in the heart (Lang et al. 1998). However, except for the very proximal regions of the PCJ and distal ureteric regions of the UPJ system, there is a real possibility that two populations of pacemaker cells, atypical SMC and ICC-LC, are present in any portion of UPJ under study and that the drive of each pacemaker system on any typical SMC bundle will vary as their relative number and coupling changes with distance from the papilla base, particularly during pathological conditions. Importantly there aren't any pharmacological agents at present that will selectively identify, block or activate the drive of either pacemaker system. Although c-Kit antibody binding to unfixed tissues allows for the identification and selective recording from c-Kit-positive ICC-LC, there are no vital stains selective for atypical SMC or c-Kit-negative ICC-LC to allow a similar targeted examination of their influence on muscle wall contractility.

Blockade of regenerative electrical activity in the intact UPJ with 1–10 μm nifedipine has also complicated attempts to elucidate UPJ pacemaking. It has not been possible, as it has in the intestine and prostate, to assume that the regenerative electrical events recorded in nifedipine (1 μm) with an intracellular microelectrode reflect directly the intrinsic pacemaker activity of ICC-LC or atypical SMC electrically close to the point of recording. The influx of Ca²⁺ through L-type Ca²⁺ channels appears to be essential to entrain the small fundamental pacemaker signals into an electrical response that can be recorded with an intracellular microelectrode. It is not yet clear whether these Ca²⁺ channels are located in typical SMC coordinating the pacemaker signals from neighbouring pacemaker cells or within the plasmalemma of the pacemaker cells themselves. Thus to fully understand UPJ pacemaking there is an urgent need to examine the properties of atypical SMC and ICC-LC in isolation or in culture, after selection and sorting.

A complete electrophysiological and pharmacological profile of c-Kit- positive/negative ICC-LC and atypical SMC at the single cell level and the establishment of any species differences or changes in properties arising upon disconnection from their syncytium will identify selective blockers/activators that can be directly applied to the intact UPJ in vitro or in vivo. These agents can then be applied to portions of the UPJ that have been carefully dissected to maximize the number and therefore the pacemaker drive of atypical SMC (PCJ of the rat) and c-Kit-positive (distal UPJ of the mouse and human) and c-Kit-negative (distal UPJ of the guinea pig) ICC-LC, respectively, to examine their action in situ. This information is crucial before the development of a complete model of UPJ autorhythmicity or any pharmacological interventions to relieve hydronephrosis in both infants and adults that might aid or replace current surgical therapies.