Am J Physiol Cell Physiol 286: C1344 –C1352, 2004.
First published February 11, 2004; 10.1152/ajpcell.00392.2003.
Characterization of in vitro gutlike organ formed from mouse
embryonic stem cells
Tadao Ishikawa,1,3 Shinsuke Nakayama,2 Tadashi Nakagawa,1 Kazuhide Horiguchi,4 Hiromi Misawa,1
Makoto Kadowaki,1 Akimasa Nakao,3 Soichiro Inoue,3 Terumasa Komuro,4 and Miyako Takaki1
1
Department of Physiology II, Nara Medical University, Kashihara, Nara 634-8521; Departments of 2Physiology I
and 3Surgery II, Nagoya University Graduate School of Medicine, Tsurumai, Nagoya 466-8550; and 4Department
of Basic Human Sciences, School of Human Sciences, Waseda University, Tokorozawa, Saitama 359-1192, Japan
Submitted 15 September 2003; accepted in final form 10 February 2004
One of the characteristics of GI smooth muscle is the
generation of rhythmic spontaneous contractions. Recent investigations have demonstrated that the ICC network in the
musculature of the GI tract is responsible for the generation of
electrical pacemaker activity and also controls the frequency
and propagation characteristics of GI motility (6, 19, 25, 29).
ICC comprise a cell population that is unique to the GI tract,
and its pacemaker activity results in rhythmic oscillations of
smooth muscle membrane potential, called slow waves (2, 23).
Enteric neurons also innervate smooth muscle and are essential
for peristalsis in GI motility (1). Thus ICC and/or enteric
neurons could coordinate GI motility (8, 9, 14, 21, 22). ES cells
have a pluripotent ability to differentiate into a wide range of
cell types, and thus the types of ES clusters are heterogeneous.
Characterization of comprehensive physiological and morphological properties of spontaneously differentiated ES clusters in
the absence of exogenously added factors that could influence
differentiation has not fully been performed.
The aim of the present study was to characterize physiological and morphological properties of the contracting ES gut on
approximately day 21 of outgrowth culture. We performed
morphological characterization by immunohistochemistry and
electron microscopy and performed physiological characterization by analyses of spontaneous rhythmic contractions, intracellular Ca2⫹ movements, and electrical activities. Our results
should provide the basis for developing appropriate models to
study the origin of the rhythmicity in the mammalian GI tract
(20, 25, 27).
intracellular calcium concentration oscillation; interstitial cells of
Cajal; peristalsis
MATERIALS AND METHODS
ES Cell Culture
RECENTLY, EMBRYONIC STEM
(ES) cells have been shown to
spontaneously give rise to a functional organlike unit, the “gut”
(ES gut), which undergoes rhythmic contractions and is topographically composed of enteric derivatives of all three embryonic germ layers: epithelial cells (endoderm), smooth muscle cells and interstitial cells of Cajal (ICC) (mesoderm), and
enteric neurons (ectoderm) (28). On approximately day 21 of
outgrowth culture, the ES gut shows distinct, highly coordinated, rhythmic contraction patterns that are composed of
periodic contractions and relaxations. Moreover, the ES gut
motor patterns are reminiscent of gastrointestinal (GI) motilities, i.e., peristalsis and segmentation (14).
Undifferentiated ES cells (EB3) were maintained on gelatin-coated
dishes without feeder cells in Dulbecco’s modified Eagle’s medium
(DMEM; Sigma, St. Louis, MO) supplemented with 10% fetal bovine
serum (FBS; GIBCO-BRL, Grand Island, NY), 0.1 mM 2-mercaptoethanol (Sigma), 0.1 mM nonessential amino acids (GIBCO-BRL), 1
mM sodium pyruvate (Sigma), and 1,000 U/ml leukemia inhibitory
factor (LIF; GIBCO-BRL). The EB3 cells (a kind gift from Dr.
Hitoshi Niwa, Center for Developmental Biology, RIKEN, Kobe,
Japan) carried the blasticidin S-resistant selection marker gene driven
by the Oct-3/4 promoter (active under undifferentiated status) and
were maintained in medium containing 10 ␮g/ml blasticidin S to
eliminate differentiated cells (17). To induce embryoid body (EB)
formation, we cultured dissociated ES cells in hanging drops (10, 18)
Address for reprint requests and other correspondence: M. Takaki, Dept. of
Physiology II, Nara Medical Univ., 840 Shijo-cho, Kashihara, Nara 634-8521,
Japan (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
C1344
0363-6143/04 $5.00 Copyright © 2004 the American Physiological Society
http://www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Ishikawa, Tadao, Shinsuke Nakayama, Tadashi Nakagawa,
Kazuhide Horiguchi, Hiromi Misawa, Makoto Kadowaki, Akimasa Nakao, Soichiro Inoue, Terumasa Komuro, and Miyako
Takaki. Characterization of in vitro gutlike organ formed from mouse
embryonic stem cells. Am J Physiol Cell Physiol 286: C1344 –C1352,
2004. First published February 11, 2004; 10.1152/ajpcell.00392.
2003.—Using an embryoid body (EB) culture system, we have made
a functional organlike cluster: the “gut” from embryonic stem (ES)
cells (ES gut). There are many types of ES clusters, because ES cells
have a pluripotent ability to develop into a wide range of cell types.
Before inducing specific differentiation by exogenously added factors,
we characterized comprehensive physiological and morphological
properties of ES guts. Each ES gut has a hemispherical (or cystic)
structure and exhibits spontaneous contractions [mean frequency:
13.5 ⫾ 8.8 cycles per min (cpm)]. A dense distribution of interstitial
cells of Cajal (ICC) was identified by c-Kit immunoreactivity, and
specific subcellular structures of ICC and smooth muscle cells were
identified with electron microscopy. ICC frequently formed close
contacts with the neighboring smooth muscle cells and occasionally
formed gap junctions with other ICC. Widely propagating intracellular Ca2⫹ concentration oscillations were generated in the ES gut from
the aggregates of c-Kit immunopositive cells. Plateau potentials,
possibly pacemaker potentials in ICC, and electrical slow waves were
recorded for the first time. These events were nifedipine insensitive, as
in the mouse gut. Our present results indicate that the rhythmic
pacemaker activity generated in ICC efficiently spreads to smooth
muscle cells and drives spontaneous rhythmic contractions of the ES
gut. The present characterization of physiological and morphological
properties of ES gut paves the way for making appropriate models to
investigate the origin of rhythmicity in the gut.
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
C1345
the intracellular Ca2⫹ concentration ([Ca2⫹]i). The ES guts were
illuminated at 488 nm, and fluorescent emissions of 515–565 nm were
recorded at an intensity of fluo 3. Digital Ca2⫹ images (328 ⫻ 247
pixels) were normally collected at 300-ms intervals. The temporal
fluorescence intensity of the dye (Ft) was normalized by the fluorescence intensity at the start (F0). These relative values represent
integrated [Ca2⫹]i. In some ES guts, after the fluorescence intensity
was recorded, the distribution of ICC was examined by staining with
an anti-c-Kit antibody. During Ca2⫹ imaging, the temperature was
kept at 30°C. Although [Ca2⫹]i oscillation is temperature dependent,
[Ca2⫹]i oscillation could be observed at this temperature, but the
motility in all ES guts was moderately depressed. This is advantageous to Ca2⫹ imaging, because high motility would disturb the
detection of real [Ca2⫹]i signal changes.
Motility Analysis of Video Images
Electrophysiological Studies
We monitored and recorded video images of ES guts with a
microscope video recording system (Olympus IX-70 and Victor
cassette recorder BR-S605B; Tokyo, Japan). We counted the number
of spontaneous contractions during a 5-min period at least three times
from the reproduced videotape. The temperature of the dish was kept
at 35°C using a micro-warm plate system (U HP-100; Kitazato,
Tokyo, Japan).
The culture dish was perfused with warmed (35°C) Tyrode solution
bubbled with 100% oxygen gas at a constant flow rate of ⬃2.0
ml/min. The temperature of the dish was kept at 35°C using the U
HP-100 micro-warm-plate system. Conventional microelectrode techniques were used to record electrical responses of single cells from
each ES cluster under an inverted microscope (Olympus IX-70). Glass
capillary microelectrodes (1.0- to 1.2-mm outer diameter) filled with
3 M KCl had tip resistances ranging between 50 and 80 M⍀. The
intracellular potentials thus recorded were displayed on a cathode ray
oscilloscope (SS-7602; Iwatsu, Tokyo, Japan). The data were also
acquired by a personal computer (Fujitsu, Tokyo, Japan) through an
analog-to-digital converter (Axon Instruments, Foster City, CA) at
500 Hz, filtered at 100 Hz, and analyzed with AxoScope 7 (Axon
Instruments, Foster City, CA). Maximal amplitude and rate of rise of
electrical activities were calculated on the basis of digital data with
Clampfit 8.1 in AxoScope 7.
Immunohistochemistry
For Kit and connexin43 immunohistochemistry, the whole mount
preparations of ES guts were fixed in acetone (4°C, 5 min). After
fixation, preparations were washed for 30 min in PBS (0.1 M, pH 7.4).
Nonspecific antibody binding was reduced by incubation for 2 h in
10% normal goat serum in PBS containing 0.3% (vol/vol) Triton-X
100 at room temperature. Tissues were incubated overnight at 4°C
with a rat monoclonal antibody raised against c-Kit protein (ACK2, 5
␮g/ml in PBS; eBioscience, San Diego, CA) and with a rabbit
polyclonal antibody raised against mouse connexin43 (Cx43, gap
junction ␣1 protein, 5 ␮g/ml in PBS; Chemicon International, Temecula, CA). Immunoreactivity for Kit was detected using Alexa
Flour 488-conjugated secondary antibody (Alexa Flour 488 goat
anti-rat; Molecular Probes, Eugene, OR) diluted 1:200 in PBS for 2 h
in the dark at room temperature. Immunoreactivity for Cx43 was
detected using Alexa Flour 546-conjugated secondary antibody
(Alexa Flour 546 goat anti-rabbit; Molecular Probes) diluted 1:200 in
PBS for 2 h in the dark at room temperature. Tissues were examined
with a Bio-Rad MRC 600 (Hercules, CA) confocal microscope.
Confocal micrographs are digital composites of Z-series scans of
10 –15 optical sections through a depth of 10 –20 ␮m. Final images
were constructed with Comos software (Bio-Rad).
Electron Microscopy
Tissues were fixed with 4% paraformaldehyde and 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.3) at room temperature. After
being rinsed in the same buffer, tissues were postfixed in 1% osmium
tetroxide for 0.5 h at 4°C. Tissues were rinsed in distilled water,
block-stained with 3% uranyl acetate solution for 3 h, dehydrated in
a graded series of ethyl alcohols, and embedded in Epon epoxy resin.
Ultrathin sections were cut with a Reichert ultramicrotome, doublestained with 3% uranyl acetate and lead citrate, and observed with a
JEM1200 EXII electron microscope (JEOL, Tokyo, Japan).
Ca2⫹ Imaging
The ES guts were incubated for 4 h at room temperature in a
modified Krebs solution containing 10 ␮M fluo 3-acetoxymethyl ester
(Dojindo, Kumamoto, Japan) and detergents [0.02% Pluronic F-127
(Dojindo) and 0.02% Cremophor EL (Sigma)]. A digital imaging
system (Argus HiSCA; Hamamatsu Photonics, Shizuoka, Japan) combined with an inverted microscope was used to monitor oscillation of
AJP-Cell Physiol • VOL
Drugs
Nifedipine, tetrodotoxin (TTX), nickel ions (Ni2⫹), and ryanodine
were purchased from Sigma (St. Louis, MO). 2-Aminoethoxydiphenyl
borate (2-APB) was generously donated by Dr. K. Mikoshiba (University of Tokyo, Tokyo, Japan). Nifedipine, ryanodine, and 2-APB
were dissolved in DMSO at concentrations of 5–10 mM. Other
chemicals were dissolved in distilled water as a stock solution and
diluted further with Tyrode or Krebs solution to the desired concentrations (ratios of dilution were ⬎1:1,000).
RESULTS
Contracting ES clusters were evaluated by recording serial
video images at various differentiation stages, and we determined that ES guts typically exhibited rhythmic contractions
on day 21 of EB outgrowth culture. As shown in Fig. 1, A–C,
a variety of motility patterns could be detected in ES guts. Each
mechanical activity was composed of periodic contraction
(arrows) and relaxation (not shown). A tubular (cystic structure) ES gut exhibited distinct patterns of highly coordinated
peristalsis-like (but not peristalsis) contractions: the contraction ring propagated from the base to the top of the cyst (Fig.
1A), and a hemispherical ES gut showed local contractions; the
left side of the ES gut contracted toward the right side, thereby
acquiring a round shape (Fig. 1B). A twin hemispherical ES gut
exhibited segmentation-like contractions on the boundary of
each gut; there was thickening between the two spheres and a
reduced size of the entire body (Fig. 1C).
Motility Analysis of Video Images
On day 21 of EB outgrowth culture, at 35°C, the mean
frequency of spontaneous motility in ES guts was 13.5 ⫾ 8.8
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
with minor modifications. The cell density of one drop was 500 cells
per 15 ␮l of ES cell medium in the absence of LIF. After 6 days in a
hanging drop culture, the resulting EBs were plated onto plastic
100-mm gelatin-coated dishes and allowed to attach for the outgrowth
culture (4, 17, 28). The culturing in hanging drops was the most
important process for differentiating ES gut. Each contracting cluster
underwent a dramatic transformation into a cystlike structure, with a
cavity containing fluid and solids. On approximately day 14, these
clusters proliferated to form more prominent three-dimensional structures with lumens and began rhythmic contractions that were not
necessarily frequent or regular. On approximately day 21, the clusters
(ES gut) showed coordinated contraction patterns with relatively
regular rhythms, although there were some incomplete cystlike structures even on day 21.
C1346
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
cycles per minute (cpm) (n ⫽ 45 ES guts). Motility was also
evaluated at a lower temperature because pacemaker activity in
ICC is temperature dependent (19, 24). At 24°C, the mean
frequency was reduced to 1.0 ⫾ 0.9 cpm (15.2 ⫾ 13.9% of
control; n ⫽ 14). The voltage-gated L-type Ca2⫹ channel
blocker nifedipine (0.1–10 ␮M) decreased the mean frequency
of motility to 0.8 ⫾ 1.3 cpm (n ⫽ 20) (Fig. 1D). At a
concentration of 10 ␮M, nifedipine abolished spontaneous
motility in 8 of 11 ES guts, but motility persisted in the
remaining 3 guts.
Immunohistochemistry
When immunohistochemistry was used as a means to identify and localize ICC in ES guts, c-Kit-immunoreactive cells
were found to be abundant in these preparations. Figure 2A
shows a hemispherical domelike cyst (ES gut on day 21 of EB
culture) that exhibited a large number of c-Kit-positive (cKit⫹) cells on the wall of the domelike structure surrounding
the lumen. c-Kit⫹ cells did not form a single layer but were
scattered throughout the muscle layer (Fig. 2A). In general,
most of the c-Kit⫹ cells were multipolar, and they formed a
distinct and dense network (Fig. 2, A, b and B). The network
structure of c-Kit⫹ cells was similar to that of ICC at the level
of the myenteric plexus in the pylorus, small intestine, and
colon of a murine embryo or neonate, as described previously
AJP-Cell Physiol • VOL
(25). Immunoreactivity for Cx43, a useful marker for gap
junction (15), was also identified on the wall of the domelike
structure in another ES gut that had potent, rhythmic spontaneous contractions (Fig. 2C).
Ultrastructure of ICC Within the Musculature of Contracting
ES Guts
As previously reported, the walls of the ES guts evaluated in
this investigation were composed of three layers: epithelium,
submucosa, and musculature (28). The musculature consisted
primarily of smooth muscle cells. Although the musculature
was not well organized, several smooth muscle cells were
oriented in the same direction and the musculature was divided
into two or more layers (Fig. 2D, a). Cells showing some
ultrastructural features of ICC were observed within the musculature, as previously reported (28). These cells were characterized by electron-dense cytoplasm and the presence of many
mitochondria (Fig. 2D, a and b), and the Golgi apparatus as
well as smooth and rough endoplasmic reticulum (ER) were
well detected in the perinuclear cytoplasm (Fig. 2D, b and c).
Caveolae were also observed along the cell membrane (Fig.
2D, d). ICC frequently formed close contacts with the neighboring smooth muscle cells (Fig. 2D, c). Gap junctions between the very thin cytoplasmic processes of unidentified cells,
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 1. Three series of video images show clusters (from their left to right sides) of embryonic stem cells (“ES guts”) periodically
contracting at 2- (A), 3- (B), and 1.5-s intervals (C) on day 21 of embryoid body (EB) outgrowth culture. A: a tubular (cystic
structure) ES gut exhibited distinct patterns of highly coordinated peristalsis-like contractions. As shown at left upper corner in each
frame, a contraction ring propagates from the base to the top of the cystic ES gut. B: a single hemispherical ES gut exhibited local
contractions at its left side, resulting in a round shape. C: twin hemispherical ES guts exhibited segmentation-like contractions on
the boundary of each gut; there was a thickening between the 2 spheres and reduced size of the entire body. Arrows indicate
directions of contractions. Scale bar, 0.5 mm. D: effects of nifedipine (0.1–10 ␮M) at 35°C and low temperature (24°C) on mean
frequency of spontaneous motility in contracting ES guts on day 21 of EB culture. The concentration of 10 ␮M is still appropriate
for blocking L-type Ca2⫹ channel because of ES gut properties with less permeability for drugs. *P ⬍ 0.05 vs. control (by ANOVA
and post hoc test). N, number of ES guts.
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
C1347
probably ICC, were occasionally observed (Fig. 2D, c, inset).
Enteric neurons were not detected in this moving ES gut.
Ca2⫹ Oscillation in ES Guts
In Ca2⫹ imaging studies, spontaneous waves of elevated
[Ca2⫹]i were observed originating at the upper pole in the ES
gut. Figure 3 shows an example of one such spontaneous
[Ca2⫹]i oscillation, referred to as a global wave, originating at
the upper area in the ES gut and propagating widely to the
middle region of the ES gut (Fig. 3A, b and c). Many c-Kit⫹
cells aggregated in the upper pole of the body of this same
preparation (Fig. 3A, a). The aggregates of c-Kit⫹ cells in the
upper pole are thought to be ICC and may be responsible for
initiation of the global wave. Figure 3A, d– g, shows a series of
pseudocolor ratio images, demonstrating the global wave from
the upper pole to the middle area of this ES gut.
On the other hand, the local and intense [Ca2⫹]i oscillation (local wave) that originated at another local area (including point 2) in the same ES gut was also observed. A
small number of scattering c-Kit⫹ cells were detected in
AJP-Cell Physiol • VOL
this area (Fig. 3B, a), indicating that these cells generated
the local wave (Fig. 3B, b–f).
This local wave was not synchronous with the global wave
over the body in this ES gut. The frequency of the global wave
recorded at point 1 was 1.5 cpm and was quite different from
that (13 cpm) of the local wave recorded at point 2 (Fig. 3C, d).
These results suggest that the global and local [Ca2⫹]i oscillations originate independently.
In another ES gut, the widely propagated [Ca2⫹]i oscillation
(global wave) originated at the same site (Fig. 4B, b and i) and
propagated over the body even after the treatment with 10 ␮M
nifedipine (Fig. 4B, c and j), indicating that nifedipine does not
affect the generation or propagation of this global wave.
Additional treatment with 40 ␮M Ni2⫹ abolished this global
wave, although 40 ␮M Ni2⫹ alone did not abolish it (data not
shown).
Figure 4C, a shows that [Ca2⫹]i oscillations (global waves)
recorded at two neighboring sites (⬃75 ␮m distant) were
synchronous in the same ES gut shown in Fig. 4, A and B. In
the presence of nifedipine, the frequency of global waves was
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 2. Immunohistochemistry for interstitial cells of Cajal (ICC; A and B), gap junctions (C), and ultrastructures of the musculature
(D) of 3 different contracting ES guts on day 21 of EB culture. A: a hemispherical domelike cyst showed a large number of
c-Kit-immunopositive (c-Kit⫹) cells that were identified on the wall of the domelike structure surrounding the lumen. The c-Kit⫹
cells were mainly multipolar-type ICC, and they did not form a single layer but were scattered throughout the muscle layer.
B: enlargement of area b in A. c-Kit⫹ cells formed a distinct and dense network. C: immunoreactivities for connexin43 (Cx43),
a marker for gap junctions (arrows), were scattered on the wall in the body of another contracting ES gut on day 21 of EB culture.
D, a: electron micrograph showing the muscle layer consisting of thick inner (IM) and thin outer muscle layers (OM), which was
covered by the outermost cell layer similar to the serous membrane. ICC (IC) were observed between muscle layers of different
orientation. Scale bar, 1 ␮m. b: micrograph showing the perinuclear cytoplasm of ICC containing numerous mitochondria,
a well-differentiated Golgi apparatus (G), and smooth endoplasmic reticulum. Scale bar, 1 ␮m. c: micrograph showing the close
contact (arrow) between ICC and a neighboring smooth muscle cell (SM). Scale bar, 0.5 ␮m. Inset: gap junction between
the cytoplasmic processes of probable ICC. Scale bar, 0.1 ␮m. d: caveolae (arrows) along the cell membrane of ICC. Scale bar,
0.1 ␮m.
C1348
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
not changed, indicating that nifedipine does not affect the
generation or propagation of the global waves, as mentioned
above. Some [Ca2⫹]i oscillations that originated at point 1,
where the global waves originated, did not propagate to the
very close but different site, point 2, resulting in local waves
(Fig. 4C, b). Therefore, nifedipine may suppress the propagation of these [Ca2⫹]i oscillations.
In three other ES guts, similar global waves were observed,
although the frequency was variable. The mean frequency of
the global waves of four ES guts was 11.8 ⫾ 4.3 (7–18.5) cpm.
In comparison, no significant difference was observed in the
mean frequency of the global waves in the presence of nifedipine [11.3 ⫾ 4.0 (5–18.5) cpm]. Taken together, these results
suggest that nifedipine does not affect the generation or propagation of the global waves in domelike ES guts, although
2–10 ␮M nifedipine abolishes global waves in sheetlike ES
clusters (n ⫽ 2) (data not shown).
AJP-Cell Physiol • VOL
Electrophysiological Studies
We investigated the electrical activity of the contracting ES
clusters at various differentiation stages by using a conventional intracellular recording technique at 35°C, and we focused on ES guts with rhythmic contractions present on day 21
of EB outgrowth culture.
Various types of electrical activities. Electrical slow waves
were frequently recorded in these preparations (Fig. 5A). The
mean frequency and amplitude of these events were 8.0 ⫾ 4.7
(1–18) cpm and 12.3 ⫾ 6.0 (5.7–21.2) mV, respectively (n ⫽
17). The shape of the electrical slow waves was very similar to
that recorded in cultured ICC of mouse jejunum (27).
In the current study, we recorded, for the first time,
plateau potentials in ES guts. These events were possibly
pacemaker potentials generated in ICC of a twin hemispherical ES gut (Fig. 5C; the same one shown in Fig. 1C). The
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 3. Intracellular Ca2⫹ concentration ([Ca2⫹]i) oscillation was observed in a domelike ES gut. A, a: c-Kit⫹ cells stained by
ACK2 after fluorescence intensity was recorded. b and c: from the upper area of the ES gut (enclosed with dotted curve and
indicated by arrow in A, a) where c-Kit⫹ cells aggregate, [Ca2⫹]i oscillation was initiated and propagated (arrow) to the middle
area including point 1 (green) of this ES gut (enclosed by dotted curve). d– g: a series of pseudocolor ratio images, at 300-ms
intervals of the same [Ca2⫹]i oscillation, demonstrating propagation from the upper pole to the middle area of the same ES gut.
B, a: a small number of c-Kit⫹ cells were scattered on the local area indicated (arrow). b–f: another [Ca2⫹]i oscillation appeared
at the local area including point 2 (arrow in c). A series of pseudocolor ratio images at 300-ms intervals demonstrated that [Ca2⫹]i
oscillation propagated locally rather than widely. C: time-course changes in fluorescence intensity recorded at points 1 and 2 were
demonstrated in the same ES gut as shown in A and B. Each of the pseudocolor ratio images a, b, and c corresponds to a, b, and
c in d. d: widely propagated, intense [Ca2⫹]i oscillations (global waves) were recorded at 1.5 cycles per minute (cpm), and locally
propagated, weak [Ca2⫹]i oscillations (local waves) were recorded at 5 cpm at point 1 (green curve). Intense and moderate local
[Ca2⫹]i oscillations (local waves) were recorded at 13 cpm at point 2 (red curve). These [Ca2⫹]i oscillations recorded at points 1
and 2 were not synchronous.
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
C1349
mean amplitude of plateau potentials was very large (47.1 ⫾
1.46 mV; n ⫽ 9), and the mean rate of rise of plateau
potentials was 29.1 ⫾ 6.7 mV/s (n ⫽ 9). The plateau
potentials were followed by transient potentials that exhibited a large mean amplitude (40.4 ⫾ 4.7 mV; n ⫽ 9) and a
mean rate of rise of 31.3 ⫾ 2.4 mV/s (n ⫽ 9). There were
no differences in amplitude and rate of rise between plateau
potentials and transient potentials. Therefore, these two
potentials do not appear to be distinct events. The shape of
the plateau potentials recorded in this cell type is very like
that of electrical potentials recorded in an isolated ICC in
short-term culture (5 days) after enzymatic dissociation
from mouse ileum (13).
In general, the regularity and noise level of the electrical
waves in individual ICC and smooth muscle cells of the ES gut
were not homogeneous (Fig. 5, A–C). This might be the result
of a pluripotent ability to develop into a wide range of cell
types and a different degree of differentiation without exogenously added growth factor even on the same day 21 of EB
outgrowth culture.
Effects of the L-type Ca2⫹ channel blocker nifedipine. Electrical slow wave is considered to be generated by pacemaker
current from ICC and to be insensitive to L-type Ca2⫹
channel blockers (19, 20, 27). Pacemaking in ICC is thought
to involve oscillations in [Ca2⫹]i involving ER and mitochondria (27). To test the sensitivity of ES gut slow waves
to L-type Ca2⫹ channel blockers, we applied nifedipine (5
␮M) to ES guts. The slow waves persisted in the presence of
nifedipine (Fig. 5B). Similar effects were obtained in two
different ES guts.
After treatment with nifedipine (1–5 ␮M), plateau potentials
and transient potentials did not change significantly, though the
AJP-Cell Physiol • VOL
amplitude of the plateau potentials slightly decreased (Fig. 5C).
In another two ES guts, plateau potentials similar to this
plateau potential in shape were recorded. Mean frequency and
rate of rise were 4.8 cpm and 24.1 ⫾ 7.1 mV/s (n ⫽ 11). These
potentials were also unaffected by nifedipine (5 ␮M).
Resting membrane potentials including spontaneously active
and inactive cells were ⫺26.2 ⫾ 6.5 (n ⫽ 12 ES guts in 12- to
17-day outgrowth culture), ⫺35.1 ⫾ 6.7 (n ⫽ 24 ES guts in 18to 24-day outgrowth culture), and ⫺37.2 ⫾ 6.7 mV (n ⫽ 13 ES
guts in 25-day outgrowth culture), indicating that the resting
membrane potential becomes progressively more hyperpolarized with time in culture.
Additional Observations in Video Images and
[Ca2⫹]i Oscillations
To assess whether intracellular Ca2⫹ release channels are
involved in spontaneous contractions, we examined the effect
of ryanodine. Ryanodine (0.1–10 ␮M) caused a concentrationdependent decrease in the mean frequency of periodical contractions in 18 ES guts, although the effects of ryanodine in
individual ES guts were variable. Ryanodine (1 ␮M) significantly decreased mean frequency to 3.4 ⫾ 1.7 cpm (70.2 ⫾
37.1% of control) in 10 ES guts (in 6 ES guts: 45.8 ⫾ 16.9%
of control, P ⬍ 0.05; in 4 ES guts: 106.8 ⫾ 28.1% of control).
Ryanodine (10 ␮M) further significantly decreased contraction
frequency to 2.9 ⫾ 1.7 cpm (64.3 ⫾ 32.1% of control) in 5 ES
guts (in 2 ES guts: 15.0 ⫾ 1.7% of control, P ⬍ 0.05; in 3 ES
guts: 84.5 ⫾ 25.9% of control) (Fig. 5D).
In contrast, the inositol 1,4,5-trisphosphate (IP3) receptor
blocker 2-APB (1–10 ␮M) only slightly decreased the fre-
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 4. Comparison of widely propagated [Ca2⫹]i
oscillations between guts treated without (A) and
with 10 ␮M nifedipine (B). Each series of pseudocolor ratio images was presented at 300-ms intervals. A, a–f: from the small area indicated (arrow at
point 1 in b), [Ca2⫹]i oscillation was initiated and
propagated over the ES gut (c). B, a–j: the first
[Ca2⫹]i oscillation was initiated and propagated over
the ES gut (c and d) from point 1 (arrow in b). The
second [Ca2⫹]i oscillation was initiated and propagated over the ES gut (j) from the same point 1
(arrow in i), although the propagated pattern was
slightly different from that of the first oscillation.
C: time-course changes in the fluorescence ratio
Ft/F0 recorded at points 1 and 2 at 300-ms intervals
were demonstrated in the same ES gut as shown in
A and B. a: synchronous [Ca2⫹]i oscillations were
recorded at points 1 and 2 without nifedipine. Six
global waves at points 1 and 2 and 1 local wave at
point 1 were observed. b: [Ca2⫹]i oscillations at
points 1 and 2 were almost synchronous after treatment with 10 ␮M nifedipine. However, when the
Ft/F0 ratio at point 1 was less intense (⬍2.0; thick
arrows), [Ca2⫹]i oscillation at point 2 was not recorded. Five and one-half global waves at points 1
and 2 and local waves at point 1 were observed.
C1350
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
quency from 6.0 ⫾ 4.5 to 5.4 ⫾ 3.6 cpm (89.8 ⫾ 23.2% of
control) in all 6 ES guts (Fig. 5E).
To assess whether intracellular Ca2⫹ release channels are
involved in [Ca2⫹]i oscillation, we examined the effect of
ryanodine. Final application of 10 ␮M ryanodine significantly
decreased the frequency of [Ca2⫹]i oscillation in 13 ES guts,
indicating the contribution of intracellular Ca2⫹ release channels to [Ca2⫹]i oscillation. TTX (0.1–1 ␮M), which inhibits
neural activity via Na⫹ channel blockade, did not affect the
frequency of spontaneous periodic contractions in 12 ES guts.
Characteristics of each phenomenon in domelike ES guts are
summarized in Table 1.
AJP-Cell Physiol • VOL
DISCUSSION
We previously demonstrated that ES cells could give rise to
a functional organlike unit, the ES gut, which consists of a
broad spectrum of enteric derivatives of all three embryonic
germ layers including various kinds of epithelial cells, smooth
muscle cells, ICC, and neurons (28). Morphological and physiological characterizations of the ES gut, however, have not
been previously accomplished (28).
ES guts have similar spontaneous rhythmic contractions,
although they have a variety of structures. In the present study,
we demonstrate for the first time that the voltage-dependent
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 5. Representative electrical activities (A–C) and summarized effects of ryanodine and 2-aminoethoxydiphenyl borate (2-APB)
on mean frequency of spontaneous motility (D and E) of contracting ES guts on day 21 of EB outgrowth culture. A: typical slow
waves with a regular rhythm of 6 cpm were recorded in a single cell of a hemispheric ES gut. B: slow waves with an irregular
rhythm of 6 cpm and a variety of amplitudes were recorded in a single cell of another hemispheric ES gut in the absence of
nifedipine (left). Slow waves with a similar rhythm of 6 cpm and rather constant amplitude were recorded in the same cell of the
same ES gut in the presence of nifedipine (5 ␮M) (right). C: control plateau potentials were recorded in a single cell of the ES gut
on day 21 of EB outgrowth culture (top). Nifedipine at 1 (bottom left) and 5 ␮M (bottom right) did not abolish these plateau
potentials. Compare each plateau potential (1–3) in each trace. The amplitude of the plateau potentials slightly decreased after
treatment with 5 ␮M nifedipine. This decrease is not definite, because immediately after this recording, the electrode slipped out
from the cell. D and E: effects of ryanodine (D) and 2-APB (E) on mean frequency of spontaneous motility in the contracting ES
gut on day 21 of EB culture. Ryanodine (1 ␮M) greatly reduced the frequency in 6 ES guts (45.8 ⫾ 16.9% of control) but did not
affect the frequency in 4 ES guts (106.8 ⫾ 28.1% of control). Ryanodine (10 ␮M) almost abolished the motility in 2 ES guts
(15.0 ⫾ 1.7% of control) and slightly reduced the frequency in 3 ES guts (84.5 ⫾ 25.9% of control). *P ⬍ 0.05 vs. control (by
ANOVA and post hoc test). #P ⬍ 0.05 (by ANOVA and post hoc test).
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
Table 1. Characteristics of spontaneous motility, electrical
activity, and [Ca2⫹]i oscillation in domelike ES guts
[Ca2⫹]i
Oscillation
Characteristics
Motility
Electrical Activity
(Slow Waves,
Plateau Potentials)
Temperature dependence
Ryanodine sensitivity
2-APB sensitivity
Nifedipine sensitivity
Ni2⫹ sensitivity
⫹
⫹, ⫺
⫺
⫹
0
⫹
0
0
⫺
0
⫹
⫹
0
⫺
⫹
⫹
⫹
0
⫺
⫹
c-Kit⫹ dependence
Gap junction dependence
Mutual dependence
⫹
⫹
⫹
⫹
⫹
⫹
⫺
⫹
⫹
⫺
3
Global
Local
L-type Ca2⫹ channel contribution to these spontaneous contractions is similar to that previously described in the mature
mouse gut. This finding indicates that functional voltagedependent L-type Ca2⫹ channels are probably expressed in
smooth muscle cells of the ES gut and that the muscle contraction is dependent on activation of L-type Ca2⫹ channels.
The most striking finding of the present study is that widely
propagated Ca2⫹ oscillations (global waves) generated at an
aggregate of many c-Kit⫹ cells are frequently observed in the
ES gut even in the presence of nifedipine. This finding indicates that the network between ICC and smooth muscle cells is
well differentiated in the ES gut. Electron microscopic analysis
confirmed that ICC frequently formed close contacts with the
neighboring smooth muscle cells and occasionally formed gap
junctions between what were probably ICC. Furthermore, the
presence of immunoreactivity for Cx43 (gap junction ␣1 protein) scattered on the wall in the body of the ES gut supports
the possibility that the network between ICC and smooth
muscles is well differentiated in the ES gut. Ca2⫹ oscillations
that were generated from small clusters of c-Kit⫹ cells varied
in frequency and did not propagate widely.
L-type Ca2⫹ channels are known to scarcely affect electrical
slow waves recorded in the mouse small intestinal ICC (11)
and smooth muscle cells (26). Furthermore, it has been confirmed (23) that a single cell in the mouse small intestine,
identified as an ICC by light microscopy, electron microscopy,
and expression of Kit mRNA, generates a rhythmic, inward
current that is insensitive to L-type Ca2⫹ channel blockers.
Plateau potentials and electrical slow waves evaluated in the
present study were hardly affected by nifedipine, indicating
that these events are generated by the pacemaker current in
ICC and propagated electrotonically to smooth muscles, respectively.
Periodic activation of plasmalemmal ion channels of ICC to
generate pacemaker current triggers intracellular Ca2⫹ release
from the ER, although Ca2⫹ release is mediated through IP3
type 1 receptor in the ER and subsequent Ca2⫹ entry into
mitochondria (20, 27). These processes are very similar to
those in cultured cell clusters isolated from mouse intestine
(24). Several candidates such as nonselective cation channels,
including transient receptor potential channels (16, 24), Cl⫺
channels (7, 12), and/or Ca2⫹-activated K⫹ channels (3), have
been reported (5, 6).
AJP-Cell Physiol • VOL
In the present study, 2-APB did not affect the frequency of
spontaneous motility in ES guts, but ryanodine significantly
decreased the frequency of both Ca2⫹ oscillations and spontaneous motility in ES guts. However, ryanodine did not affect
the frequency of spontaneous motility in a subset of ES guts.
These findings suggest that ryanodine receptors differentiate
and contribute to intracellular Ca2⫹ movements in the majority
of ES guts on 21-day outgrowth, which is different from
previous findings in mouse stomach (20) and in cultured mouse
myenteric ICC (27). These previous studies indicate that Ca2⫹
release from IP3 receptor-operated stores (but not ryanodine
receptor-operated stores) is linked to initiation of pacemaker
current (20, 27). Although intracellular Ca2⫹ movements certainly contribute to generate pacemaker current among sarcolemma, ER, and mitochondria in ICC of ES guts, what regulates the periodicity of the firing of pacemaker activity in ICC
and what initiates and terminates the firing remain to be
elucidated.
The present results reveal physiological and morphological
characterization of ES guts spontaneously formed from mouse
ES cells in the absence of exogenously added growth factors:
pacemaker activity generated by intracellular Ca2⫹ movements
in ICC propagates to the smooth muscle layer through gap
junctions over the body and drives spontaneous rhythmic
contractions of domelike ES guts (Table 1). Because ES cells
have a pluripotent ability to develop into a wide range of cell
types in the present experimental conditions (without exogenously added growth factors), various ES guts could have
been differentiated showing heterogeneous physiological and
morphological properties such as various structures (dome,
twin dome, tubular cyst, or sheet), variable sensitivities to
ryanodine, and various regularity and noise levels of electrical
activities. For example, in domelike ES guts, nifedipine did not
abolish, whereas additional Ni2⫹ did abolish, [Ca2⫹]i oscillations (Table 1); however, in sheetlike ES clusters, nifedipine
did abolish [Ca2⫹]i oscillations.
We have revealed that Ca2⫹ influx via L-type Ca2⫹ channels
contributes to smooth muscle contraction in the domelike ES
guts. On the other hand, it is possible that either different types
of Ca2⫹ channels or intracellular Ca2⫹ movements contribute
to [Ca2⫹]i oscillations in ICC. [Ca2⫹]i oscillations in smooth
muscle cells are driven indirectly by electrical slow waves of
ICC. Electrical slow waves in ICC are not dependent on L-type
Ca2⫹ channels but are dependent on intracellular Ca2⫹ movements, and they electrotonically propagate over the smooth
muscle layers. Such ICC were well differentiated and were
distributed densely throughout the small body of the domelike
ES guts presently studied, suggesting that electrical slow
waves in ICC could propagate over the body with less decline.
Therefore, we believe that this ICC network generates the
[Ca2⫹]i oscillations that persist in the presence of nifedipine.
Furthermore, application of Ni2⫹ abolished [Ca2⫹]i oscillations in some domelike ES guts, indicating that T-type Ca2⫹
channels might have a role on the origination and/or propagation of [Ca2⫹]i oscillations of a subpopulation of the ES guts.
Finally, at 30°C for Ca2⫹ imaging, the motility in the ES
guts (of smaller size) was moderately depressed, as mentioned
in MATERIALS AND METHODS. At 35°C, the motility of the ES guts
(of larger size) was normal. This difference of 5°C in temperature would seem unlikely to affect the contribution of L-type
Ca2⫹ channels to [Ca2⫹]i oscillations if the temperature sen-
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
⫹, Positive; ⫺, negative; 0, unknown. [Ca ]i, intracellular Ca2⫹ concentration; ES gut, an organlike cluster from embryonic stem cells; 2-APB,
2-aminoethoxydiphenyl borate; c-Kit⫹, c-Kit-positive cells.
2⫹
C1351
C1352
PHYSIOLOGICAL AND MORPHOLOGICAL PROPERTIES OF ES GUT
sitivity in these ES guts is not higher than in the welldeveloped mouse gut.
In conclusion, the information obtained from the present
study should provide the basis for future studies of the development of the elements of the gut that underlie GI motility.
Improved technology (EB culture), combined with various
exogenously added factors, can now be developed to make
appropriate models from ES cells. This approach could facilitate significant advances in studies on gut organ physiological
functions such as spontaneous rhythmicity.
ACKNOWLEDGMENTS
We thank Prof. Gary M. Mawe (Dept. of Anatomy and Neurobiology,
University of Vermont) for critical reading of this manuscript.
This work was supported by Grants-in-Aid for Scientific Research
14370189 and 14657311 (to M. Takaki) and 15300134 (to S. Nakayama) from
the Ministry of Education, Science, Sports and Culture of Japan.
REFERENCES
1. Costa M, Hennig GW, and Brookes SJ. Intestinal peristalsis: a mammalian motor pattern controlled by enteric neural circuits. Ann NY Acad
Sci 860: 464 – 466, 1998.
2. Der-Silaphet T, Malysz J, Hagel S, Arsenault AL, and Huizinga JD.
Interstitial cells of Cajal direct normal propulsive contractile activity in the
mouse small intestine. Gastroenterology 114: 724 –736, 1998.
3. Fujita A, Takeuchi T, Saitoh N, Hanai J, and Hata F. Expression of
Ca2⫹-activated K⫹ channels, SK3, in the interstitial cells of Cajal in the
gastrointestinal tract. Am J Physiol Cell Physiol 281: C1727–C1733, 2001.
4. Hooper M, Hardy K, Handyside A, Hunter S, and Monk M. HPRTdeficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature 326: 292–295, 1987.
5. Horowitz B, Ward SM, and Sanders KM. Cellular and molecular basis
for electrical rhythmicity in gastrointestinal muscles. Annu Rev Physiol 61:
19 – 43, 1999.
6. Huizinga JD, Thuneberg L, Klüppel M, Malysz J, Mikkelsen HB, and
Bernstein A. The W/kit gene required for interstitial cells of Cajal and for
intestinal pacemaker activity. Nature 373: 347–349, 1995.
7. Huizinga JD, Zhu Y, Ye J, and Molleman A. High conductance chloride
channels generate pacemaker currents in interstitial cells of Cajal. Gastroenterology 123: 1627–1636, 2002.
8. Hukuhara T, Kotani S, and Sato G. Comparative studies on the motility
of the normal, denervated and aganlionic THIRY-loops. Jpn J Physiol 12:
348 –356, 1962.
9. Hukuhara T, Sumi T, and Kotani S. The role of the ganglionic cells in
the small intestine taken in the intestinal intrinsic reflex. Jpn J Physiol 11:
281–288, 1961.
10. Keller GM. In vitro differentiation of embryonic stem cells. Curr Opin
Cell Biol 7: 862– 869, 1995.
11. Koh SD, Sanders KM, and Ward SM. Spontaneous electrical rhythmicity in cultured interstitial cells of Cajal from the murine small intestine.
J Physiol 168: 131–139, 1998.
AJP-Cell Physiol • VOL
286 • JUNE 2004 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 15, 2017
GRANTS
12. Kölbel CB, Holtmann G, Mcroberts JA, Schöler S, Aengenvoordt P,
Singer MV, and Mayer EA. Involvement of chloride channels in the
receptor mediated activation of longitudinal colonic muscle. Neurogastroenterol Motil 10: 489 – 498, 1998.
13. Lee JCF, Thuneberg L, Berezin I, and Huizinga JD. Generation of slow
waves in membrane potential is an intrinsic property of interstitial cells of
Cajal. Am J Physiol Gastrointest Liver Physiol 277: G409 –G423, 1999.
14. Nakagawa T, Yamanouchi M, Ueshima S, Fujii H, Nakajima Y, and
Takaki M. Intestinal spontaneous electrical and mechanical activities in
W/Wv mutant mice (Abstract). Neurogastroenterol Motil 15: 224, 2003.
15. Nakamura K and Shibata Y. Connexin43 expression in network-forming cells at the submucosal-muscular border of guinea pig and dog colon.
Cell Tissues Organs 165: 16 –21, 1999.
16. Nakayama S and Torihashi S. Spontaneous rhythmicity in cultured cell
clusters isolated from mouse small intestine. Jpn J Physiol 52: 217–227,
2002.
17. Niwa H, Miyazaki J, and Smith AG. Quantitative expression of Oct-3/4
defines differentiation, dedifferentiation or self-renewal of ES cells. Nat
Genet 24: 372–376, 2000.
18. Robertson EJ. Embryo-derived stem cell lines. In: Teratocarcinomas and
Embryonic Stem Cells: A Practical Approach (1st ed.), edited by Robertson EJ. Washington, DC: IRL, 1987, p. 71–112.
19. Sanders KM, Ördög T, Koh SD, Torihashi S, and Ward SM. Development and plasticity of interstitial cells of Cajal. Neurogastroenterol
Motil 11: 311–338, 1999.
20. Suzuki H, Takano H, Yamamoto Y, Komuro T, Saito M, Kato K, and
Mikoshiba K. Properties of gastric smooth muscles obtained from mice
which lack inositol trisphosphate receptor. J Physiol 525: 105–111, 2000.
21. Takaki M and Yoneda S. The interrelation between spontaneous electrical activities and spontaneous longitudinal and circular muscle contractions in the mouse proximal colon (Abstract). Neurogastroenterol Motil
15: 225, 2003.
22. Takaki M, Yoneda S, Nakagawa T, and Ishikawa T. Role of interstitial
cells of Cajal and enteric neurons on gut spontaneous motility (Abstract).
Auton Neurosci 106: 41, 2003.
23. Thomsen L, Robinson TL, Lee JC, Farraway LA, Hughes MJ, Andrews DW, and Huizinga JD. Interstitial cells of Cajal generate a
rhythmic pacemaker current. Nat Med 4: 848 – 851, 1998.
24. Torihashi S, Fujimoto T, Trost C, and Nakayama S. Calcium oscillation linked to pacemaking of interstitial cells of Cajal. J Biol Chem 277:
19191–19197, 2002.
25. Torihashi S, Ward SM, and Sanders KM. Development of c-Kitpositive cells and the onset of electrical rhythmicity in murine small
intestine. Gastroenterology 112: 144 –155, 1997.
26. Ward SM, Burns AJ, Torihashi S, and Sanders KM. Mutation of the
proto-oncogene c-kit blocks development of interstitial cells and electrical
rhythmicity in murine intestine. J Physiol 480: 91–97, 1994.
27. Ward SM, Ördög T, Koh SD, Baker SA, Jun JY, Amberg G, Monaghan K, and Sanders KM. Pacemaking in interstitial cells of Cajal
depends upon calcium handling by endoplasmic reticulum and mitochondria. J Physiol 525: 355–361, 2000.
28. Yamada T, Yoshikawa M, Takaki M, Torihashi S, Kato Y, Nakajima
Y, Ishizaka S, and Tsunoda Y. In vitro function gut-like organ formation
from mouse embryonic stem cells. Stem Cells 20: 41– 49, 2002.
29. Yoneda S, Takano H, Takaki M, and Suzuki H. Properties of spontaneously active cells distributed in the submucosal layer of mouse proximal
colon. J Physiol 542: 887– 897, 2002.