0022-5347/01/1661-0329/0

THE JOURNAL OF UROLOGY® Vol. 166, 329 –334, July 2001

Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.® Printed in U.S.A.

CHARACTERIZATION OF THE SPONTANEOUS ELECTRICAL AND

CONTRACTILE ACTIVITY OF SMOOTH MUSCLE CELLS IN THE RAT
UPPER URINARY TRACT
R. J. LANG, H. TAKANO, M. E. DAVIDSON, H. SUZUKI AND M. F. KLEMM
From the Departments of Physiology, Monash University, Clayton Victoria, Australia, and Nagoya City University, Nagoya, Japan

ABSTRACT

Purpose: We morphologically and electrophysiologically identified the cells that generate the
electrical activity underlying the peristaltic contractions of the rat upper urinary tract.
Materials and Methods: Electron microscopy and tension recording techniques were used to
characterize the smooth muscle cells underlying spontaneous contractions in the wall of the rat
ureter, and proximal and distal renal pelvis. Intracellular microelectrodes, containing 4% neu-
robiotin were used to record data from the cells of the renal pelvis, which were later viewed on
a confocal microscope.
Results: Spontaneous myogenic contractions (average 22.3 ⫾ 2.2 minutes⫺1) originated in the
proximal renal pelvis and propagated into the distal renal pelvis and ureter in 6 preparations.
Smooth muscle cells in the renal pelvis and ureter were typical in appearance with greater than
85% of their sectional area containing clumped contractile filaments. In contrast, contractile
fibrils occupied only 65% of the sectional area of the smooth muscle cells within the most
proximal region of the renal pelvis (pelvicaliceal junction). In strips of the renal pelvis spindle
shaped cells 83 to 200 ␮m. long fired spontaneous action potentials (6 minutes⫺1) consisting of an
initial spike, a quiescent plateau phase and abrupt hyperpolarization to a peak diastolic potential
of ⫺60 mV. Other spindle shaped cells 94 to 112 ␮m. long displayed small membrane transients
(15 minutes⫺1) 9 to 19 mV. in amplitude, firing from a diastolic potential of ⫺40 mV.
Conclusions: It is likely that the spontaneous contractile activity of the rat upper urinary tract
arises from the discharge of action potentials in typical smooth muscle cells of the proximal renal
pelvis that are directly driven by the spontaneous membrane oscillations of atypical smooth
muscle cells.
KEY WORDS: peristalsis; action potentials; rats, Wistar; urinary tract; muscle, smooth

Spontaneous contractions occur in the mammalian upper
urinary tract to propel urine from the renal pelvis through to
the ureter and then into the bladder, where it is stored until
micturition. Strips cut along the circumference of the renal
pelvis display spontaneous contractions with an intrinsic
frequency that decreases as strips are obtained from regions
more distal to the renal calix. The isolated ureter is quiescent
in most mammals except in humans and pigs.1, 2 In the
mammalian upper urinary tract spontaneous myogenic con-
tractile activity has been associated with the presence of
histologically and morphologically distinct atypical smooth
muscle cells. In unicaliceal mammals, including the dog, cat,
rat, rabbit and guinea pig, an outer layer of darker staining
atypical smooth muscle cells extends from the region of pelvic
attachment to the renal parenchyma and to the pelviureteral
junction, which is absent in the ureter.3, 4 In mammals with
multicaliceal kidneys (humans and pigs) a layer of these
atypical cells is also found on the inner aspect of the muscle
wall, interconnecting the minor calices to each other and to
the renal parenchyma.5, 6
Although the rat upper urinary tract has been extensively
studied in terms of its morphology, sensory neuropeptide
distribution7 and afferent renal nerve function,8 there is
surprisingly little known about the mechanisms underlying
pyeloureteral motility. Recently we reported a comparison of
the effects of specific inhibitors of the 2 isoforms of cycloox-
ygenase, constitutive cyclooxygenase-1 and inducible
Accepted for publication January 5, 2001.
Supported by the Australian National Health and Medical Re-
search Council.
329
cyclooxygenase-2, on the motility of whole mount prepara-
tions of the upper urinary tract in the guinea pig and rat, and
described a number of striking differences in these 2 species.9
Spontaneous contractions in the rat upper urinary tract in
control saline at 30C occurred at a frequency 3-fold larger
than in the guinea pig at 35C and increased in amplitude with
time. Also, exposure to capsaicin had little effect on muscle wall
motility of the rat upper urinary tract, while contraction fre-
quency and amplitude in the guinea pig preparation were re-
duced.9, 10 Furthermore, selective cyclooxygenase-1 inhibition
increased the amplitude of contractions in the rat upper urinary
tract but had little effect on the frequency or amplitude of the
contractions recorded in the guinea pig. In contrast, selective
cyclooxygenase-2 blockade had little effect on the motility of the
rat upper urinary tract but reduced motility of the guinea pig in
a concentration dependent manner.9 Cyclooxygenase-2 inhibi-
tion has also recently been shown to reduce the increases in the
activity of the rat afferent renal nerve and prostaglandin E2
production induced by increased renal pelvic pressure, suggest-
ing that prostaglandins contribute to the stimulation of mech-
anosensitive neurons in the pelvic wall.8 In this report we made
a detailed examination of the contractile, electrical and morpho-
logical properties of the smooth muscle cells within the wall of
the rat upper urinary tract. As in the guinea pig, contractions
originated in the proximal renal pelvis, correlating with an
increased presence of atypical smooth muscles in the pelvical-
iceal junction. Contractions arose from the firing of driven ac-
tion potentials, which were of a much simpler time course
compared with driven cells in the guinea pig. 11, 12
330 SPONTANEOUS ELECTRICAL AND CONTRACTILE ACTIVITY IN RAT RENAL PELVIS
MATERIALS AND METHODS
Tension experiments. Wistar rats weighing 200 to 500 gm.
were anesthetized with chloroform and then decapitated. The
kidneys were removed through a midline incision and imme-
diately placed in physiological saline, as described. The upper
urinary tract was dissected free from the surrounding paren-
chyma and a longitudinal cut was made extending from the
proximal renal pelvis through to the mid region of the ureter,
enabling the whole preparation to be laid flat. Cuts directed
around the circumference and extending to the midline were
then made. The first cut divided the proximal portion of the
renal pelvis from the distal renal pelvis, while 2 smaller cuts
were made in the ureter approximately 1 cm. from the pel-
viureteral junction to create a short circumferential strip.
Threads were then tied around the 3 separated regions and
the preparation was pinned into a 1.3 ml. organ bath per-
fused with physiological saline and maintained at a temper-
ature of 30C or 35C. The threads were attached to isometric
force transducers, which were connected to a MacLab/4s
analog-to-digital converter driven by a Macintosh LC (Apple,
Sunnyvale, California), thus, allowing the simultaneous re-
cording of tension developed in the 3 regions of the upper
urinary tract. Approximately 1 mN. of tension was placed on
each region and the tissue was then left to equilibrate for 30
to 60 minutes, after which all 3 regions generated contrac-
tions that were regular in amplitude and frequency.9, 10
Electron microscopy. Three female rats weighing 150 to
180 gm. were anesthetized with 60 mg./kg. pentobarbitone
given intraperitoneally. The abdomen was opened and a can-
nula was inserted into the dorsal aorta. The animals were
initially perfused with saline (0.9% NaCl containing 1% so-
dium nitrite) for 1 minute at 80 mm. Hg, followed by fixative
for 2 minutes at 80 mm. Hg and 6 to 8 minutes at 50 mm. Hg.
The fixative consisted of 2% glutaraldehyde and 2% parafor-
maldehyde in 0.1 M. sodium cacodylate buffer, pH 7.2. After
perfusion the kidneys and ureters were gently removed and
placed in containers of fresh fixative. While immersed in
fixative, the renal pelvis and ureters were separated from the
kidney parenchyma and small pieces were cut from the pel-
vicaliceal junction, an area immediately distal to the attach-
ment of the renal pelvis to the kidney parenchyma as well as
from the renal pelvis and ureter about 1 to 2 cm. from the
pelviureteral junction. The small tissue pieces were incu-
bated in fresh fixative for a further 2 hours, rinsed 3 times for
15 minutes each time in 0.1 M. sodium cacodylate buffer, pH
7.2, fixed in buffered 1% osmium tetroxide for 2 hours before
dehydration in ethanol and epoxy propane, and embedding in
EPON. Sections were cut from randomly selected blocks from
each region using a diamond knife on a Reichert Ultracut
ultramicrotome (Leica Microsystems, Vienna, Austria),
mounted on copper grids and then examined using a Jeol
TEM100 transmission electron microscope (Joel, Tokyo, Ja-
pan).12 Grid squares containing smooth muscle cells were
photographed at 5,000⫻ magnification. Photographs were
enlarged to approximately 15,000⫻ magnification and the
outlines of a minimum of 30 smooth muscle cells per animal
were traced onto acetate sheets and digitized into a com-
puter. The areas of the smooth muscle cell profile occupied by
mitochondria, clumps of contractile filaments and the cell
nucleus (if present) were outlined. Areas of close apposition
between cells were marked and grouped as 50 to 80 nm.
appositions where the apposing cell membranes were sepa-
rated by intervening basal lamina, 20 nm. junctions where
the cell membranes were closely opposed and the basal lam-
ina was absent, and adherence-type junctions, where closely
apposed cell membranes were identified by the presence of
membrane specializations. Sectional area and cell length
were calculated and expressed as a percent of the total sec-
tional area of each cell or as a percent of the sectional perim-
eter, respectively.
Electrophysiological experiments. Recordings from the
proximal renal pelvis were made from 5 ⫻ 10 mm. circum-
ferential strips of renal pelvis dissected free of the rat kidney
and pinned to the bottom of a 1 ml. organ bath. The organ
bath was then mounted on an inverted microscope (Olympus
Optical Corp., Melville, New York) and perfused with bicar-
bonate buffered physiological saline, as described, at 3 to 4
ml. per minute at 34 to 35C. Electrophysiological recordings
were made using standard intracellular microelectrode tech-
niques and glass microelectrodes with resistances of 50 to 80
m⍀ when filled with 0.5 to 2 M. KCl. Changes in membrane
potential were recorded with a standard unity-gain pream-
plifier and stored digitally using a Labmaster TL-1 (Axon,
Union City, California) analog-to-digital converter at a sam-
ple rate of 400 to 1,000 Hz., a personal computer and Axotape
1.1 software (Axon). Various parameters of the spontaneous
potentials were measured, including the frequency of the
action potential discharge, amplitude of the initial spike
when present, absolute potential during the plateau, dura-
tion of the action potential measured from the time the initial
spike was half maximal and peak negative diastolic mem-
brane potential reached during the after-hyperpolarization
period. In each experiment the parameters of 3 or 4 action
potentials were averaged. A number of similar experiments
were then averaged, as indicated. One-way analysis of vari-
ance or the paired Student t test was done to test for signif-
icance with p ⬍0.05 considered statistically significant.11, 12
Confocal microscopy and solutions. Neurobiotin (4%) (Vec-
tor Laboratories, Burlingame California), a commonly used
cell tracer, was dissolved in the microelectrode saline. After
successful impalement the tracer was injected into the cell
for more than 1 minute. The tissue was removed from the
bath, re-pinned onto a small piece of dental wax and fixed
overnight at 4C in 4% paraformaldehyde in 0.1 M. phosphate
buffer, pH 7.2. The tissues were rinsed 3 ⫻ 15 minutes in
phosphate buffered saline (PBS) containing 0.5% Triton
X-100 (PBS/Triton), unpinned and incubated in streptavidin-
CY3 (Jackson Immunoresearch, West Grove, Pennsylvania)
and diluted 1:1,000 in PBS/Triton for 24 to 48 hours. After 3
final 10-minute rinses in PBS tissues were mounted on glass
slides using Dako mounting medium (Dako Corp., Carpinte-
ria, California) and examined using a TCS NT confocal mi-
croscope (Leica, Leica Mikroskope und Systeme GmbH, Wet-
zlar, Germany) equipped with 16{times], 40⫻ and 63⫻ oil
immersion lenses. Specimens were excited using the 568 nm.
line of the Kr:Ar laser and emissions were detected through
a 665 nm. pass filter. Preparations were scanned and XYZ
stacks were collected, so that the whole volume of neurobi-
otin filled cells was imaged. Stacks of images were saved to a
computer hard disk and 3-dimensional reconstructions were
made using Voxblast software (Vaytek, Fairfield, Iowa).
Physiological saline composition was 120 mM. NaCl, 5 mM.
KCl, 2.5 mM. CaCl2, 1 mM. MgCl2, 1 mM. NaH2PO4, 25 mM.
NaHCO3 and 11 mM. glucose, pH 7.3 to 7.4 after bubbling
with a 95% O2-5% CO2 gas mixture.
RESULTS
General morphology. Low magnification electron micros-
copy and immunostaining for ␣-smooth muscle actin revealed
the general arrangement of the cells that composed the wall
of the rat upper urinary tract (figs. 1 and 2). A single layer of
closely packed epithelial cells lined the upper urinary tract of
the rat (fig. 1, A). Lamina propria, which contained blood
vessels, axon bundles and fibroblast-like cells, separated the
epithelial cells from the main smooth muscle wall. In the
ureter smooth muscle cells were arranged into an inner cir-
cular and an outer longitudinal layer. In both muscle layers
the smooth muscle cells were tightly packed, as evidenced by
the dense immunoreactivity for ␣-smooth muscle actin ob-
served under confocal microscopy (fig. 2, C). In the renal
 SPONTANEOUS ELECTRICAL AND CONTRACTILE ACTIVITY IN RAT RENAL PELVIS
FIG. 1. Electron micrographs shows arrangement of smooth mus-
cle cells and other elements in rat. A, pelvicaliceal region. Bar rep-
resents 5 ␮m. Inset reveals 3 adherence-type junctions at higher
magnification B, renal pelvis. Arrows indicate typical 20 nm. junc-
tion. smc, smooth muscle cells. Bar represents 2 ␮m. Inset shows
typical 20 nm. junction at higher magnification. C, ureter. Arrows
indicate 50 to 80 nm. appositions filled with basal lamina. Bar
represents 1 ␮m.
FIG. 2. Confocal images of sections of rat tissue stained to show
relative immunoreactivity for ␣-smooth muscle actin. A, pelvicaliceal
region. B, renal pelvis. C, ureter. Bars represent 50 ␮m.
pelvis smooth muscle cells were arranged in smaller, ran-
domly oriented bundles, which were connected with each
other to form a plexiform layer and separated by connective
tissue (fig. 1, B). This looser packaging of muscle cells was
also evidenced by the reduced immunoreactivity for ␣-smooth
muscle actin (fig. 2, B).
Smooth muscle cell bundling within the renal pelvis and
ureter was reflected by the large percentage of cell perime-
ters in close apposition with a neighboring cell in 133 and 184
cells examined (mean 29.4% ⫾ 1.6%, range 0% to 84% and
24.2% ⫾ 1.2%, range 0% to 75%, respectively), and separated
by a 50 to 80 nm. gap filled with basal lamina (fig. 1, A and
B). Muscle cells of the upper urinary tract also formed more
tightly coupled junctions of 20 nm. with absent basal lamina
(fig. 1, B, inset). In the renal pelvis and ureter these 20 nm.
junctions occupied an average of 2.8% ⫾ 0.4% and 4.08% ⫾
0.71% of the perimeters of 20 and 37 cells, respectively.
Specialized adherence-type junctions occupying an average
of 2.2% ⫾ 0.3% of the perimeter of 11 cells were also occa-
sionally observed in the renal pelvis and ureter (fig. 1, A,
inset).
When the smooth muscle cells of the renal pelvis and
ureter were analyzed at higher magnification, they were
mostly filled with clumped contractile fibrils (fig. 3, B). The
331
FIG. 3. High magnification electron micrographs of rat renal pel-
vis. A, examples of nerve bundles surrounded by Schwann cell. B,
typical and atypical smooth muscle cells. C, ICC-like fibroblasts
containing several caveolae, discontinuous basal laminar material
and no contractile filaments. Bars represent 1 ␮m.
average sectional area of 222 and 151 individual cells con-
taining contractile fibrils was 85% ⫾ 1% (range 55% to 100%)
in the ureter and 86% ⫾ 1% (range 54% to 100%) in the renal
pelvis. In the renal pelvis and ureter mitochondrial profiles
occupied approximately an average of 4% (4% ⫾ 0.3% and
4.6% ⫾ 0.2%, respectively) of the cell sectional area (fig. 1, B
and C, and fig. 3, B). Bundles of unmyelinated nerve axons
and varicosities often surrounded by a Schwann cell were
observed within the lamina propria and muscle layers of the
ureter (fig. 1, C), renal pelvis (fig. 3, A) and pelvicaliceal
junction.
In the most proximal region of the renal pelvis (pelvical-
iceal junction) smooth muscle cells were arranged in an open
network of individual cells (fig. 1, A) with little ␣-smooth
muscle actin immunostaining (fig. 2, A). Only 3, 50 to 80 nm.
junctions were observed in this region, reflecting this sparse
bundling of cells. Pelvicaliceal muscle cells were widest at the
nucleus and had many long narrow processes that formed
specialized junctions with other similar cells (fig. 1, A and 3,
B). These junctions were of the 20 nm. and the adherence
type, occupying an average of 7.3% ⫾ 1.2% and 1.18% ⫾
0.22% of the perimeter of 29 and 17 cells, respectively. A total
of 76 smooth muscle cells of the pelvicaliceal region also had
significantly less average sectional area occupied by mito-
chondria compared with smooth muscle cells of the ureter
and renal pelvis (2.6% ⫾ 0.38%, Student’s unpaired t test p
⬍0.05). In 76 cells examined contractile fibrils in the muscle
cells of the pelvicaliceal junction occurred in only small
clumps, separated by large areas of cytoplasm rich in endo-
plasmic reticulum. The average sectional area of 76 cells
occupied by contractile fibrils was 66.4% ⫾ 3.4%, which was
significantly smaller than the sectional areas of muscle cells
in the renal pelvis or ureter (unpaired t test p ⬍0.05).
In the guinea pig upper urinary tract smooth muscle cells
have been divided into a typical and an atypical type based
on the skewed distribution frequency profiles of the percent
of sectional area of cells filled with contractile fibrils.12 In our
experiments the frequency distributions of the percent of
sectional area of cells filled with contractile fibrils in the rat
renal pelvis and ureter were such that 99% of the cells had
greater than 60% of the sectioned area occupied by contrac-
tile filaments and were termed typical smooth muscle cells.
Atypical smooth muscle cells with less than 40% of the sec-
tional area occupied by contractile filaments12 represented
only 22% of the smooth muscle cells examined in the pelvi-
caliceal junction (fig. 3, B), while the remaining 78% had
typical morphology.
332 SPONTANEOUS ELECTRICAL AND CONTRACTILE ACTIVITY IN RAT RENAL PELVIS
A third population of 7 cells was also occasionally identi-
fied within the lamina propria, pelvicaliceal junction and
renal pelvis but never in the ureter. These fibroblast-like
cells displayed numerous caveolae and a discontinuous basal
lamina material, but no visible bundles of contractile fila-
ments. These cells also had an oval shaped nuclear region
and many narrow extended processes. However, these cells
were not immunoreactive for ␣-smooth muscle actin or c-Kit,
the proto-oncogene for tyrosine kinase and specific marker
for interstitial cells of Cajal (ICC) in the intestine.12
Spontaneous contractile activity. At 35C contractions of the
whole mount preparations of the rat upper urinary tract
originated in the proximal renal pelvis and propagated dis-
tally into the distal renal pelvis and then into the ureter.
Moreover, these contractions often occurred from an oscillat-
ing baseline (fig. 4, B). Contraction amplitudes decreased
distally along the rat upper urinary tract. For example, the
average contraction amplitude in the proximal and distal
renal pelvis of 6 rats was 0.45 ⫾ 0.09 mN. and 0.3 ⫾ 0.07, and
in the ureter of 3 it was 0.18 ⫾ 0.11 mN. Average contraction
frequency also decreased distally in the proximal and distal
renal pelvis of 6 rats, and in the ureter of 3 (22.3 ⫾ 2.2
minutes⫺1 and 14.1 ⫾ 2.6, respectively). In contrast, contrac-
tions in these 3 regions of the upper urinary of 9 guinea pigs
tract recorded under the same conditions at 35C were gen-
erally 2 to 5-fold larger in amplitude and occurred at a mean
of 7.1 ⫾ 0.4 minutes⫺1 in all regions.9, 10
When the temperature was reduced to 30C, the contrac-
tions of the rat upper urinary tract became more regular in
amplitude and frequency, allowing more reliable meas-
urements of parameters during experiments into the effects
of applied excitatory or inhibitory agents.9 In the proximal
and distal renal pelvis this 5C decrease in the bathing tem-
perature to 30C significantly increased contraction ampli-
tudes by 20% and 33%, respectively (fig. 4, C). However, the
frequency of these contractions at 30C was decreased signif-
icantly to 59% of the control frequency at 35C (p ⬍0.05, fig. 4,
C). When these changes in contraction parameters in the
proximal and distal renal pelvis were expressed as a motility
index using the formula, amplitude ⫻ frequency, it was ob-
served that the calculated motility index at 30C was reduced
to 68% and 76%, respectively, of the control motility index (p
⬍0.0.5, fig. 4, C). Comparable experiments in 9 guinea pigs
revealed a similar 20% to 80% increase above the control
amplitude and decrease in frequency to 45% to 85% of the
FIG. 4. Typical example of spontaneous contractile activity re-
corded simultaneously. A, in proximal and distal renal pelvis, and
ureter at 30C. B, in proximal and distal renal pelvis, and ureter at
35C. C, average amplitude and frequency of contractions in proximal
and distal renal pelvis were measured when tissues were at 30C or
35C and motility index was calculated using formula, amplitude ⫻
frequency. Asterisk indicates statistical significance of changing
temperature.
control value of contractions at 30C in all 3 regions of the
upper urinary tract (data not shown).
Electrical recordings in the rat renal pelvis. We made some
preliminary recordings of the electrical activity underlying
the spontaneous contractions in the rat renal pelvis using
intracellular microelectrodes filled with a high KCl saline
containing 4% of the intracellular cell marker neurobiotin.
Electrical recordings were generally made from cells for 1 to
10 minutes. Tissues were the removed from the bath, fixed,
processed for immunostaining for neurobiotin and then
viewed on a confocal microscope. XYZ stacks of images (5 to
15 slices at 1 to 4 ␮m. thick) of fluorescent cells were captured
and reconstructed on a computer.
Spontaneous membrane potential discharges were of 2
general types. In 2 cells small transient oscillations of mem-
brane potential 9 to 19 mV. in amplitude and 816 to 1020 ms.
in duration were recorded at a frequency of 15 minutes⫺1 (fig.
5, A) These potential transients were separated by a period of
membrane hyperpolarization with a peak negative (diastolic)
potential of ⫺40 mV. Examination of these tissues under the
confocal microscope revealed that these transient potentials
had been recorded in spindle-shaped immunoreactive cells 94
to 112 ␮m. long (fig. 5, B).
In 6 cells a distinct action potential discharge at an aver-
age of 6.3 ⫾ 1.2 minutes⫺1 consisted of an average initial
spike of 49.2 ⫾ 4.2 mV. in amplitude, followed by a long
quiescent plateau with an average half-amplitude duration
of 981 ⫾ 83 ms. at an average absolute potential of ⫺18.1 ⫾
1.3 mV. The plateau phase in the 6 cells was terminated by
abrupt repolarization, which was followed by after-
hyperpolarization to an average peak diastolic potential of
⫺61.4 ⫾ 4.2 mV. In 3 cells in which action potentials dis-
charged at higher frequencies (10 to 14 minutes⫺1) after-
hyperpolarization showed distinct, slowly developing repo-
larization before the discharge of the next action potential
(fig. 5, C and E). In contrast, 3 cells in which action potential
discharge occurred at lower frequencies (4 minutes⫺1) the
after-hyperpolarization was relatively transient and followed
by a relatively quiescent membrane potential before the next
discharge (fig. 6, A and C).
Confocal micrographs of neurobiotin filled cells in which
recordings of action potential discharge were relatively short
(less than 5 minutes) usually consisted of 1 to 3 dimly fluo-
FIG. 5. Morphological properties of 3 neurobiotin filled cells gen-
erating spontaneous electrical activity recorded in same preparation
of rat renal pelvis. A, C and E, spontaneous electrical activity re-
corded for 1 to 2 minutes using intracellular microelectrodes contain-
ing 4% neurobiotin. B, D and F, after fixation and immunostaining
for neurobiotin, cell shapes were viewed on confocal microscope.
Calibration bar represents 50 ␮m.
 SPONTANEOUS ELECTRICAL AND CONTRACTILE ACTIVITY IN RAT RENAL PELVIS
FIG. 6. Typical examples of neurobiotin filled driven cells in rat
renal pelvis. A, driven action potentials were recorded for 1 to 2
minutes. B, after fixation and immunostaining for neurobiotin cell
shapes, confocal microscopy reveals relative spread of cell marker to
neighboring cells. C, driven action potentials were recorded for more
than 5 minutes. D, after fixation and immunostaining for neurobi-
otin cell shapes, confocal microscopy reveals relative spread of cell
marker to neighboring cells. Bar represents 50 ␮m.
rescent immunoreactive cells in close proximity to each other
(fig. 5, C and E and fig. 6, A). When recordings lasted more
than 5 minutes, immunoreactive neurobiotin was far more
intense near the point of microelectrode insertion. Moreover,
this immunoreactivity was observed to have spread consid-
erable distances into neighboring cells or cell bundles (fig. 6,
D). These 6 cells also appeared spindle-shaped, 5 to 10 ␮m.
wide at the nucleus and an average of 147 ⫾ 21 ␮m. long. In
addition, complex action potentials consisting of an initial
spike and a plateau phase with many potential oscillations
superimposed, which have been routinely recorded in the
renal pelvis (greater than 75% of cells) and ureter (100% of
cells) of the guinea pig,11, 12 were never recorded in the rat
renal pelvis.
DISCUSSION
When viewed by electron microscopy, the general arrange-
ment of the smooth muscle cells within the wall of the rat
upper urinary tract progressed from tightly packed bundles
in the ureter to loosely packed bundles in the renal pelvis and
mostly individual cells in the pelvicaliceal junction. In the
pelvicaliceal junction the average cross-sectional area (65%)
of the smooth muscle cells occupied by contractile filaments
was also significantly smaller than cells in the renal pelvis
(86%) and ureter (85%). These results are consistent with the
relatively little immunoreactivity for ␣-smooth muscle actin
observed in the pelvicaliceal junction when examined by con-
focal microscopy (fig. 2) and 3, 50 to 80 nm. junctions when
viewed by electron microscopy.
It has often been postulated that the spontaneous contrac-
tions underlying pyeloureteral motility in mammals arise
from the proximal regions of the renal pelvis, where there is
a relatively high number of atypical smooth muscles.3, 4, 12
Moreover, these propagating contractions are little affected
by tetrodotoxin or blockers of sympathetic and parasympa-
thetic motor nerves, indicating that these contractions are
myogenic in origin.9, 10, 13, 14 In the rat and guinea pig pelvi-
caliceal junction 22% and 80%, respectively, of the smooth
muscle cells were deemed to be atypical. However, the fre-
quency of contraction in the proximal renal pelvis of the rat
was significantly higher than in the guinea pig (22 minutes⫺1
versus 8, unpaired t test p ⬍0.05). Thus, although the origin
of spontaneous contractile activity is likely to be determined
333
by the relative number of atypical smooth muscle cells in the
pelvicaliceal junction, the frequency of these contractions is
not directly related to the absolute number of atypical cells
present.
Investigations into the electrical origin of pyeloureteral
motility first involved the use of extracellular sucrose gap
techniques in strips of the guinea pig renal pelvis and indi-
cated that the spontaneous contractions were associated with
simple oscillations in the membrane potential. In contrast,
electrically evoked contractions in the distal renal pelvis and
ureter were associated with action potentials consisting of a
rapid rising phase and a long plateau.15–17 More recently,
intracellular microelectrode recordings from different re-
gions of the guinea pig renal pelvis and ureter have revealed
3 types of cells characterized in terms of their electrical
behavior and morphology, as revealed by electron microscopy
or by neurobiotin injections and confocal microscopy. Simple
pacemaker oscillations of the membrane potential (8 min-
utes⫺1) were recorded in spindle-shaped cells in the pelvical-
iceal junction (85% of cells)18 and the proximal renal pelvis
(15% of cells)11 but never in the distal renal pelvis or ure-
ter.12 This finding correlated closely with the relative num-
ber of atypical smooth muscle cells. Most muscle cells of the
ureter (100%), and the proximal (83%) and distal (97.5%)
renal pelvis were typical in appearance. Driven action poten-
tials consisting of an initial spike followed by various num-
bers of potential oscillations superimposed on a plateau
phase were recorded in these regions with a similar relative
distribution. After filling with neurobiotin cells discharging
driven action potentials were also seen to be spindle-shaped
and typically 250 ␮m. long. A third class of intermediate
action potentials with a single spike followed by a quiescent
plateau and abrupt repolarization was recorded in the pelvi-
caliceal junction, and in the proximal and distal renal pelvis
in 11% to 17% of cells. These action potentials were recorded
in irregular shaped cells with an oval-shaped nucleus and
many branching interconnecting processes, which were not
immunoreactive for ␣-smooth muscle actin or c-Kit, the se-
lective marker of ICC.12
The smaller frequency of contraction in the proximal re-
gions of the guinea pig renal pelvis may perhaps correlate
with the relatively higher number of ICC-like cells present
compared with the rat renal pelvis. We previously speculated
that guinea pig ICC-like cells were electrically activated by 1
or more pacemaker atypical smooth cell bundles. Moreover,
not every pacemaker potential invading an ICC-like cell was
successful in triggering an intermediate action potential due
to their longer refractory periods. This refractory process was
repeated when intermediate action potentials invaded typi-
cal smooth muscles, which fire the driven action potentials
responsible for muscle contraction.19 This refractoriness
among cell types may well explain the distally decreasing
frequency of driven action potential discharge and contrac-
tion along the renal pelvis and ureter as the cell profile
changes in each region.12 In the rat only 7 ICC-like cells were
observed and they did not appear to form the complex net-
work of cells observed in the guinea pig. In the rat we believe
that the pacemaker potentials (15 minutes⫺1) of a number of
atypical smooth muscle cells propagate directly to the typical
smooth muscle cells and combine to initiate the action poten-
tial discharge. Thus, the absence of the integrating function
of the ICC-like cells in the rat may well give rise to the
observed higher frequency of contraction.
In the rat renal pelvis the majority of recorded action
potentials came from spindle shaped cells 150 ␮m. long.
These cells are presumably the smooth muscle cells charac-
terized as typical by electron microscopy. Short injections of
neurobiotin resulted in only dimly immunoreactive cells
when viewed by confocal microscopy. Longer injection peri-
ods revealed that neurobiotin had spread to neighboring cells
and bundles. The time course of the action potentials in the
334 SPONTANEOUS ELECTRICAL AND CONTRACTILE ACTIVITY IN RAT RENAL PELVIS
typical smooth muscle cells was far simpler than that of the
action potentials recorded in similar regions of the guinea
pig. The peak diastolic and resting membrane potential be-
fore the initiation of an action potential recorded in the rat
proximal renal pelvis was ⫺60 mV., while in the guinea pig it
was between ⫺40 and ⫺50 mV. After an initial spike of 50
mV. in amplitude the action potential in the rat proximal
renal pelvis displayed a quiescent plateau phase 1 second in
duration. In the guinea pig this phase was usually about 2
seconds in duration and often displayed oscillations of poten-
tial 20 mV. in amplitude, which lasted throughout the pla-
teau.11, 12 Although to our knowledge it has not been directly
examined, it is likely that the driven action potentials re-
corded in the rat renal pelvis arise from the activation of
membrane channels similar to those identified in the guinea
pig. If it was the case, the initial spike would arise from the
opening of dihydropyridine sensitive Ca2⫹ channels. The ces-
sation of the plateau phase would be determined by the
opening of K⫹ channels, including those voltage and Ca2⫹
activated, and sensitive to 4-aminopyridine and charybdo-
toxin or apamin, respectively.11 In contrast, the refractory
period of these action potentials would be sensitive to Ba2⫹
and charybdotoxin, which are blockers of inward rectifier and
large conductance Ca2⫹ activated K⫹ channels, respectively,
and insensitive to glibenclamide, the specific blocker of aden-
osine triphosphate dependent K channels.11, 19
CONCLUSIONS
We have demonstrated that, as in the guinea pig, it is
likely that atypical smooth muscles in the rat pelvicaliceal
junction are acting as the pacemaker cells that initiate the
propagating contractions underlying pyeloureteral motility.
Action potentials recorded in typical smooth muscle cells of
the rat renal pelvis had a simpler time course than that
recorded in the guinea pig. However, it is likely that they
arise from the opening and closing of Ca2⫹ and K⫹ channels
similar to those established in the guinea pig. Because there
were relatively few ICC-like cells present in the rat, it is
unlikely that they have a pacemaker role, as in the mamma-
lian gastrointestinal tract. In fact, the relatively longer re-
fractive periods of ICC-like versus pacemaker cells in the
guinea pig led us to suggest that ICC-like cells may well have
an integrative role, combining the influences of 1 or more
pacemaker sites before discharging an action potential. Thus,
the absence of these ICC-like cells in the rat renal pelvis
allows the pacemaker cells to drive directly the typical
smooth muscle cells, resulting in the initiation of spontane-
ous contractions at a frequency 2 to 3-fold greater than in the
guinea pig renal pelvis, which is richer in these ICC-like
cells.
REFERENCES
1. Constantinou, C. E., Neubarth, J. L. and Mensah-Dwumah, M.:
Frequency gradient in the autorhythmicity of the py-
eloureteral pacemaker system. Experientia, 34: 614, 1978
2. Constantinou, C. E.: Velocity gradient and contraction frequency
of the pyeloureteral system. Experientia, 35: 791, 1979
3. Gosling, J. A. and Dixon, J. S.: Morphologic evidence that the
renal calyx and pelvis control ureteric activity in the rabbit.
Am J Anat, 130: 393, 1971
4. Gosling, J. A. and Dixon, J. S.: Species variation in the location
of upper urinary tract pacemaker cells. Invest Urol, 11: 418,
1974
5. Dixon, J. S. and Gosling, J. A.: The fine structure of pacemaker
cells in the pig renal calices. Anat Rec, 175: 139, 1973
6. Dixon, J. S. and Gosling, J. A.: The musculature of the human
renal calices, pelvis and upper ureter. J Anat, 135: 129, 1982
7. Su, H. C., Wharton, J., Polak, J. M. et al: Calcitonin gene-related
peptide immunoreactivity in afferent neurons supplying the
urinary tract: combined retrograde tracing and immunohisto-
chemistry. Neuroscience, 18: 727, 1986
8. Kopp, U. C., Cicha, M. Z., Smith, L. A. et al: Cyclooxygenase-2
involved in stimulation of renal mechanosensitive neurons.
Hypertension, 35: 373, 2000
9. Davidson, M. E. and Lang, R. J.: Effects of selective inhibitors of
cyclo-oxygenase-1 (COX-1) and cyclo- oxygenase-2 (COX-2) on
the spontaneous myogenic contractions in the upper urinary
tract of the guinea-pig and rat. Br J Pharmacol, 129: 661, 2000
10. Teele, M. E. and Lang, R. J.: Stretch-evoked inhibition of spon-
taneous migrating contractions in a whole mount preparation
of the guinea-pig upper urinary tract. Br J Pharmacol, 123:
1143, 1998
11. Lang, R. J. and Zhang, Y.: The effects of K⫹ channel blockers on
the spontaneous electrical and contractile activity in the prox-
imal renal pelvis of the guinea pig. J Urol, 155: 332, 1996
12. Klemm, M. F., Exintaris, B. and Lang, R. J.: Identification of the
cells underlying pacemaker activity in the guinea- pig upper
urinary tract. J Physiol, 519: 867, 1999
13. Maggi, C. A. and Giuliani, S.: The neurotransmitter role of
calcitonin gene-related peptide in the rat and guinea-pig ure-
ter: effect of a calcitonin gene-related peptide antagonist and
species-related differences in the action of omega conotoxin on
calcitonin gene-related peptide release from primary afferents.
Neuroscience, 43: 261, 1991
14. Maggi, C. A., Theodorsson, E., Santicioli, P. et al: Tachykinins
and calcitonin gene-related peptide as co-transmitters in local
motor responses produced by sensory nerve activation in the
guinea-pig isolated renal pelvis. Neuroscience, 46: 549, 1992
15. Zawalinski, V. C., Constantinou, C. E. and Burnstock, G.: Ure-
teral pacemaker potentials recorded with the sucrose gap tech-
nique. Experientia, 31: 931, 1975
16. Constantinou, C. E., Silvert, M. A. and Gosling, J.: Pacemaker
system in the control of ureteral peristaltic rate in the multi-
calyceal kidney of the pig. Invest Urol, 14: 440, 1977
17. Santicioli, P. and Maggi, C. A.: Pharmacological modulation of
electromechanical coupling in the proximal and distal regions
of the guinea-pig renal pelvis. J Auton Pharmacol, 17: 43,
1997
18. Tsuchida, S. and Suzuki, T.: Pacemaker activity of the pelvica-
lyceal border recorded by an intracellular glass microelec-
trode. Urol Int, 48: 121, 1992
19. Exintaris, B. and Lang, R. J.: K⫹ channel blocker modulation of
the refractory period in spontaneously active guinea-pig ure-
ters. Urol Res, 27: 319, 1999