EMBRYONIC STEM CELLS

This information is current as of April 11, 2008

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.StemCells.com/cgi/content/full/24/6/1414

STEM CELLS®, an international peer-reviewed journal, covers all aspects of stem cell research: embryonic stem cells; tissue-specific stem cells; cancer stem cells; the stem cell niche; stem cell genetics and genomics; translational and clinical research; technology development. STEM CELLS® is a monthly publication, it has been published continuously since 1983. The Journal is owned, published, and trademarked by AlphaMed Press, 318 Blackwell Street, Suite 260, Durham, North Carolina, 27701. © 2006 by AlphaMed Press, all rights reserved. Print ISSN: 1066-5099. Online ISSN: 1549-4918.

Downloaded from www.StemCells.com by on April 11, 2008

In Vitro Formation of Enteric Neural Network Structure in a Gut-Like Organ Differentiated from Mouse Embryonic Stem Cells Miyako Takaki, Shinsuke Nakayama, Hiromi Misawa, Tadashi Nakagawa and Hiroki Kuniyasu Stem Cells 2006;24;1414-1422; originally published online Mar 9, 2006; DOI: 10.1634/stemcells.2005-0394

EMBRYONIC STEM CELLS In Vitro Formation of Enteric Neural Network Structure in a Gut-Like Organ Differentiated from Mouse Embryonic Stem Cells MIYAKO TAKAKI,a SHINSUKE NAKAYAMA,b HIROMI MISAWA,a TADASHI NAKAGAWA,a HIROKI KUNIYASUc a

Department of Physiology II, Nara Medical University, School of Medicine, Kashihara, Nara, Japan; bDepartment of Physiology I, Nagoya University Graduate School of Medicine, Tsurumai, Nagoya, Japan; cDepartment of Molecular Pathology, Nara Medical University, School of Medicine, Kashihara, Nara, Japan

ABSTRACT Using an embryoid body (EB) culture system, we developed a functional organ-like cluster—a “gut”—from mouse embryonic stem (ES) cells (ES gut). Each ES gut exhibited spontaneous contractions but did not exhibit distinct peristalsis-like movements. In these spontaneously contracting ES guts, dense distributions of interstitial cells of Cajal (c-kit [a transmembrane receptor that has tyrosine kinase activity]-positive cells; gut pacemaker cells) and smooth muscle cells were discernibly identified; however, enteric neural ganglia were absent in the spontaneously differentiated ES gut. By adding brain-derived neurotrophic factor (BDNF) only during EB formation, we for the first time succeeded in in vitro formation of enteric neural ganglia with connecting nerve fiber tracts (enteric nervous system [ENS]) in the ES gut. The ES gut with ENS exhibited strong peristalsis-like movements.

During EB culture in BDNFⴙ medium, we detected each immunoreactivity associated with the trk proto-oncogenes (trkB; BDNF receptors) and neural crest marker, protooncogene tyrosine-protein kinase receptor ret precursor (c-ret), p75, or sox9. These results indicated that the present ENS is differentiated from enteric neural crestderived cells. Moreover, focal stimulation of ES guts with ENS elicited propagated increases in intracellular Ca2ⴙ concentration ([Ca2ⴙ]i) at single or multiple sites that were attenuated by atropine or abolished by tetrodotoxin. These results suggest in vitro formation of physiologically functioning enteric cholinergic excitatory neurons. We for the first time succeeded in the differentiation of functional neurons in ENS by exogenously adding BDNF in the ES gut, resulting in generation of distinct peristalsis-like movements. STEM CELLS 2006;24:1414 –1422

INTRODUCTION

within the musculature of the GI tract is responsible for the generation of electrical pacemaker activity and thus for the generation of rhythmic oscillations of the smooth muscle membrane potential called slow waves [3– 8]. These oscillations in turn control the frequency and the propagation characteristics of GI motility [9 –13]. Enteric neurons are also present within the GI tract; enteric neurons innervate the smooth muscle and are essential for peristalsis [14 –16]. This suggests that enteric neurons might act in concert with ICCs to mediate the various patterns of GI motility [17–20]. Although ES cells have a pluripotent ability to differentiate into a wide range of cell types, we have been able to differentiate only few enteric neurons within the ES gut in the absence of exogenously added neurotrophic factors [1, 2]. This

Recently, embryonic stem (ES) cells were shown to spontaneously give rise to a functional organ-like unit, the “ES gut,” which undergoes rhythmic contractions and is topographically comprised of enteric derivatives of all three embryonic germ layers: epithelial cells (endoderm), smooth muscle cells and interstitial cells of Cajal (ICCs; c-kit⫹ cells) (mesoderm), and a small number of diffusely distributed enteric neurons (ectoderm) [1]. On approximately day 21 of outgrowth culture, the ES gut begins to exhibit highly coordinated patterns of rhythmic motor activity comprised of periodic contractions and relaxations (phasic contraction); however, the motor patterns of those ES guts are not identical to gastrointestinal (GI) peristalsis [2]. Recent investigations have demonstrated that the ICC network

Correspondence: Miyako Takaki, Ph.D., Department of Physiology II, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8521, Japan. Telephone: ⫹81-744-23-4696; Fax: ⫹81-744-23-4696; e-mail: [email protected] Received on August 16, 2005; accepted for publication on March 7, 2006; first published online in STEM CELLS EXPRESS March 9, 2006. ©AlphaMed Press 1066-5099/2006/$20.00/0 doi: 10.1634/stemcells.2005-0394

STEM CELLS 2006;24:1414 –1422 www.StemCells.com


Key Words. Brain-derived neurotrophic factor • c-kit • Proto-oncogene tyrosine-protein kinase receptor ret precursor Interstitial cells of Cajal • Neurofilament • Protein gene product 9.5 • Tyrosine kinase B receptor

Takaki, Nakayama, Misawa et al.

MATERIALS

AND

METHODS

ES Cell Culture Undifferentiated ES cells (EB3) were maintained on gelatincoated dishes without feeder cells in Dulbecco’s modified Eagle’s medium (Sigma, St. Louis, http://www.sigmaaldrich.com) supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, http://www.invitrogen.com), 0.1 mM 2-mercaptoethanol (Wako Chemical, Osaka, Japan, http://www. wako-chem.co.jp/english), 0.1 mM nonessential amino acids (Gibco), 1 mM sodium pyruvate (BioWhittaker Molecular Applications, Rockland, ME, http://www.bmaproducts.com), and 1,000 U/ml of LIF (Chemicon, Temecula, CA, http://www. chemicon.com) [14, 29]. The EB3 cells (a kind gift from Dr. Hitoshi Niwa, Center for Developmental Biology, Riken, Kobe, Japan, http://www.riken.go.jp/engn) carried the blasticidin Sresistant selection marker gene driven by the Oct-3/4 promoter (active in undifferentiated cells) and were maintained in medium containing 10 ␮g/ml blasticidin S to eliminate differentiated cells [30]. To induce embryoid body (EB) formation, www.StemCells.com

Figure 1. Neurofilament immunoreactivity in embryonic stem cell guts differentiated from the embryoid body (GDNF⫹ or NT-3⫹) reveals a small number of nerve cells (A) and a network-like structure (B) but not any enteric neural ganglia. Abbreviations: GDNF, glial cell line-derived neurotrophic factor; NT-3, neurotrophin-3.

dissociated ES cells were cultured in hanging drops [31, 32] as previously described, with minor modifications [1, 2]. The cell density of one drop was 500 cells per 15 ␮l of ES cell-medium in the absence of LIF and the presence of BDNF (recombinant protein expressed in Escherichia coli, 10⫺9 to approximately 5 ⫻ 10⫺8 g/ml; Upstate, Lake Placid, NY, http://www.upstate. com) [24, 26, 33] or the endothelin B (ETB) receptor agonist IRL-1620 [Suc(Glu9, Ala11,15)-Endothelin-1 (8 –21)], 10⫺9 to approximately 2 ⫻ 10⫺9 M; Alexis, Lausen, Switzerland, http:// www.alexiscorp.com] [34, 35]. For experiments, BDNF was used at 10⫺8 to approximately 10⫺9 g/ml; BDNF at 10⫺9 g/ml was unable to induce differentiation of enteric neural ganglia, but at 10⫺8 g/ml it was able to do so. No additional effects were obtained at higher than 5 ⫻ 10⫺8 g/ml. NT-3 (recombinant protein expressed in E. coli, 10⫺9 to approximately 10⫺8 g/ml; PeproTech EC Ltd., London, http://www.peprotechec.com) and GDNF (recombinant protein expressed in E. coli, 10⫺9 to approximately 10⫺8 g/ml; PeproTech EC Ltd.) were also used to induce differentiation of enteric neural ganglia. After 6 –7 days in a hanging-drop culture, the EBs were formed. These formed EBs were placed in outgrowth cultures on plastic 100-mm gelatin-coated dishes and allowed to attach [1, 2, 30, 36]. In each outgrowth culture, cell clusters underwent a dramatic transformation into cyst-like structures, 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 to spontaneously contract. On approximately day 21, the structures (ES guts) showed coordinated patterns of contraction with relatively regular rhythms, although there were some incomplete cyst-like structures even on day 21 [1, 2].

Analysis of Gut Motility We monitored and recorded video images of ES guts using a microscope-video recording system (Olympus IX70 and Victor cassette recorder BR-S605B, Tokyo, http://www. olympus-global.com). In the videotaped images, we counted the number of spontaneous contractions per 5-minute period over the course of at least three 5-minute periods, while maintaining the temperature of the dish at 35°C using a micro-warm plate system (U HP-100; Kitazato Supply, Tokyo, http://www.kitazato-supply.com) [2].

Immunohistochemistry For immunohistochemical detection of c-kit, a transmembrane receptor that has tyrosine kinase activity, whole-mount


result motivated us to apply neurotrophic factors for differentiation of abundant, extensively distributed enteric neurons. One such factor is brain-derived neurotrophic factor (BDNF), which regulates a wide range of cellular functions in the nervous system [21], including dendritic and axonal growth [22], neuronal survival, synaptic transmission, and long-lasting potentiation in the hippocampus [23]. BDNF also supports the survival and outgrowth of a variety of central (“first” brain) and neural crest-derived neurons. Notably, strong BDNF immunoreactivity is also seen within enteric ganglion cells during human enteric nervous system (ENS; “second” brain) development [24, 25]. Neurotrophin-3 (NT-3) and glial cell line-derived neurotrophic factor (GDNF) also specifically promote the differentiation of enteric crest-derived cells into a neuron or glia and may thus play a role in the development and/or maintenance of the ENS [26, 27]. The ENS is an independent nervous system that structurally resembles the “first” brain [28]. Like the central nervous system (CNS), it shows the phenotypic diversity of its component neurons and includes every class of neurotransmitter found in the CNS; moreover, the number of neurons in the ENS rivals that in the spinal cord [28]. The neurons and glia of the ENS, which regulate GI peristalsis and fluid absorption/secretion in an autonomous fashion, are derived from precursor cells formed in the neural crest. Early during fetal gestational life, primitive neural crest cells migrate ventrally from their origin in the neural tube along the length of the developing intestine and ultimately form enteric ganglion cells and supportive glia [24]. Our first aim in the present study was to differentiate the enteric neural network structure (ENS) within the ES gut by applying the neurotrophic factor BDNF at the appropriate time. Our second aim was to evaluate whether the ENS differentiated by BDNF was able to function physiologically. To do this, first, we characterized the spontaneous motility pattern in the ES gut with the formed ENS. Second, we evaluated the response of increases in the intracellular calcium concentration ([Ca2⫹]i) elicited by focal stimulation of enteric nervous circuits. Improved technology presented here could facilitate a significant advance in the formation of a complete gut-like organ for regenerative medicine.

1415

Formation of Enteric Nervous System in ES Gut

1416

Table 1. Immunohistochemistry of trkB, c-ret, NF, c-kit, sox9, and p75 BDNFⴙ

Hanging-drop culture 2 days 4 days 6–7 days (embryoid body) Outgrowth culture, 2 weeks Outgrowth culture, 3 weeks

BDNFⴚ

trkB

c-ret

NF

c-kit

sox9

p75

trkB

c-ret

NF

c-kit

sox9

p75

⫹ ⫹ ⫹ ⫹ ⫹

⫹ ⫹ ⫹ ⫹ ⫺

⫺ ⫺ ⫺ ⫹ ⫹

⫺ ⫺ ⫺ ⫹ ⫹

⫹ ⫹ ⫺ ⫺ ⫺

⫹ ⫹ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫹ ⫹

⫺ ⫺ ⫺ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺

⫹, positive; ⫺, negative. Abbreviation: BDNF, brain-derived neurotrophic factor.

conjugated secondary antibody diluted to 0.5 ␮g/ml (Medical & Biotechnological Laboratories Co., Ltd., Nagoya, Japan, http:// www.mbl.co.jp). All sections were then rinsed with PBS, colordeveloped using diaminobenzidine (DAB) solution (DakoCytomation, Inc.), washed in water, and counterstained with Meyer’s hematoxylin (Sigma). Care was taken to ensure that the antibody reaction and DAB exposure were the same for all specimens. The antibodies and working concentrations used in the primary reaction were anti-c-ret (proto-oncogene tyrosine-protein kinase receptor ret precursor) (N-term, 1 ␮g/ml; Abgent, San Diego, http://www.abgent.com), anti-mouse c-kit (2 ␮g/ml; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, http://www.scbt.com), antitrkB (tyrosine kinase B receptor) (clone H-181, 1 ␮g/ml; Santa Cruz Biotechnology, Inc.), anti-NF (clone 2F11, reacting with 70, 160 and 200 kDa proteins, 0.5 ␮g/ml; DakoCytomation, Inc.), anti-sox9 (clone H-90, 0.5 ␮g/ml; Santa Cruz Biotechnology, Inc.), and anti-p75 (intracellular domain, 0.5 ␮g/ml; Upstate).

Ca2ⴙ Imaging ES guts were incubated for 3– 4 hours at room temperature in modified Krebs solution containing 10 ␮M fluo-3 acetoxymethyl ester (Dojindo, Kumamoto, Japan, http://www.dojindo. com) and detergents (0.02% Pluronic F-127 [Dojindo] and 0.02% Cremophor EL [Sigma]), after which changes in [Ca2⫹]i were monitored using a digital imaging system (Argus HiSCA; Hamamatsu Photonics, Shizuoka, Japan, http://www.hamamatsu.com/) mounted on an inverted microscope [37, 38]. The fluo-3-containing ES guts were illuminated at 488 nm, and the intensity of the fluorescent emission from the indicator at 515– 565 nm was recorded [2]. Digital Ca2⫹ images (328 ⫻ 247 pixels) were normally collected at 300-ms intervals, and the intensity of the fluorescence at a given time (Ft) was usually normalized to the fluorescence intensity at the start (F0), yielding relative values representative of the integrated [Ca2⫹]i. In some ES guts, after recording the fluorescence intensity, the distribution of ICCs or enteric neurons was examined by staining with an anti-c-Kit, anti-PGP 9.5, or anti-NF antibody. During Ca2⫹ imaging, the temperature was kept at 30°C. At that temperature, the motility of ES guts was moderately depressed, which was advantageous because too much movement would interfere with our ability to accurately record changes in the [Ca2⫹]i signal. At this temperature, we were readily able to detect increases in [Ca2⫹]i induced by focal stimulation (5–10 Hz, 100 ␮seconds, 10 –20 pulses, variable voltage) applied through the Ag-wire inserted in a fine glass electrode; an Agplate served as the indifferent electrode.


preparations of ES gut were fixed in acetone (4°C, 5 minutes) [14], and for detection of neurofilament (NF) or protein gene product (PGP), the preparations were fixed in 4% paraformaldehyde (4°C, 10 minutes). After fixation, the preparations were washed for 30 minutes in phosphate-buffered saline (PBS) (0.01 M, pH 7.4) and then incubated for 3 hours at room temperature in 10% normal goat serum in PBS containing 0.3% (vol/vol) Triton X-100 (PBS-TX) to reduce non-specific antibody binding. Tissues were then incubated either for 2 days at 4°C with a rat monoclonal antibody raised against c-kit protein (ACK2, 5 ␮g/ml in PBS; eBioscience, San Diego, http://www. ebioscience.com) or for 1 day at 4°C with a rabbit polyclonal antibody raised against human brain PGP 9.5 (1:3,000 in PBS; Chemicon) or a rabbit polyclonal antiserum cocktail against primate and bovine low- (68 –70 kDa), medium- (150 kDa), and high-molecular-weight NF (200 –210 kDa) (1:500 in PBS; Affiniti Research Products Ltd, Devon, U.K., http://www.affiniti-res. com). This antiserum cocktail reacts with neuronal cell bodies, dendrites, and thick and thin axons. Kit immunoreactivity was detected using an Alexa Flour 594-conjugated secondary antibody (Alexa Flour 594 goat anti-rat; Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com; 1:200 in PBS for 2 hours in the dark at room temperature), whereas PGP 9.5 and NF immunoreactivity was detected using a Texas Red-conjugated secondary antibody (Texas Red-goat anti-rabbit; MP Biomedicals, Inc., Aurora, OH, http://www.mpbio.com; 1:100 in PBS for 1 hour in the dark at room temperature). Tissues were then examined using a Bio-Rad MRC 600 (Hercules, CA, http:// www.bio-rad.com) confocal microscope, which yielded confocal micrographs that were digital composites of Z-series scans of 10 to 15 optical sections through a depth of 10 –20 ␮m. Final images were constructed using Comos software (Bio-Rad) [2]. For immunohistochemistry of thin sections, ES gut was fixed with 4% paraformaldehyde at 4°C and then scraped from the culture dish and embedded into 0.6% agarose-PBS, which was subsequently embedded in a paraffin block in the usual manner. Consecutive 4-␮m sections were cut from each block, subjected to antigen retrieval with pepsin (DakoCytomation, Inc., Carpinteria, CA, http://www.dakousa.com) for 20 minutes at room temperature, and immunostained using the immunoperoxidase technique. After blockade of endogeneous peroxidase activity by incubation for 15 minutes in 3% H2O2-methanol, the sections were rinsed with PBS and incubated with a primary antibody diluted with Washing Solution (BioGenex, San Ramon, CA, http://www.biogenex.com) for 2 hours at room temperature. Thereafter, they were rinsed again with PBS and incubated for 1 hour at room temperature with a peroxidase-


1417

Immunohistochemistry

Figure 2. Immunohistochemical detection of trkB, c-ret, NF, c-kit, sox9, and p75 after 4 days and 6 –7 days (an EB) in hanging-drop culture and after 2–3 weeks in outgrowth culture. (A): Immunoreactivity associated with trkB, c-ret, sox9, and p75, but no NF or c-kit immunoreactivity, was identified after 4 days, treated with BDNF (BDNF⫹). (B): No trkB, c-ret, NF, c-kit, sox9, or p75 immunoreactivity was seen after 4 days, not treated with BDNF (BDNF⫺). (C): Immunoreactivity associated with trkB and c-ret, but no NF, c-kit, sox9, or p75 immunoreactivity, was identified in an EB treated with BDNF (BDNF⫹). (D): No trkB, c-ret, NF c-kit, sox9, or p75 immunoreactivity was seen in another EB not treated with BDNF (BDNF⫺). (E): c-ret immunoreactivity was apparent in an ES gut differentiated from EB (BDNF⫹) after 2 weeks in outgrowth culture but was absent after 3 weeks (top panels). No c-ret immunoreactivity was detected in ES guts differentiated from EB (BDNF⫺) after 2–3 weeks (bottom panels) in outgrowth culture. (F): trkB, c-ret, NF, and c-kit immunoreactivity in a BDNF⫹ ES gut after 2 weeks in outgrowth culture. The specimen shown is the same as in the left top panel of (E). Abbreviations: BDNF, brain-derived neurotrophic factor; EB, embryoid body; ES, embryonic stem; NF, neurofilament.

Among the drugs used, tetrodotoxin (TTX), atropine, and hexamethonium were purchased from Sigma. All chemicals were dissolved in distilled water as stock solutions and further diluted as needed with Tyrode or Krebs solution to the desired concentrations. (The ratios of the dilution were always greater than 1:1,000.)

RESULTS ES Cell Culture To induce differentiation, we placed ES cells in hanging-drop cultures with medium containing BDNF (10⫺9 to ⬃5 ⫻ 10⫺8 g/ml) [24, 31, 32] or the ETB-receptor agonist IRL-1620 (2 ⫻ www.StemCells.com

The timing and BDNF-dependence of the expression of tyrosine kinase B receptors encoded by the trk proto-oncogenes (trkB; BDNF receptors), proto-oncogene tyrosine-protein kinase receptor ret precursor (c-ret) [39, 40], migrating neural crest cell markers, sox9 [41] and p75 [39, 40, 42], NF, and c-kit are summarized in Table 1. We examined the formation process of EBs from ES cells during the hanging-drop culture in the presence of BDNF. Throughout the hanging-drop culture, we detected immunoreactivity associated with trkB and c-ret, both of which are expressed during in vivo development of enteric neurons during mouse embryogenesis [24, 26, 39, 43]. By contrast, we detected sox9 and p75 immunoreactivity only after 2 and 4 days in the hanging-drop culture, but we detected none of c-kit, NF, sox9, and p75 immunoreactivity in the EBs (after 6 –7 days in the hanging-drop culture) (Fig. 2A, 2C; Table 1). In addition, no expression of trkB, c-ret, NF or PGP 9.5, sox9, and p75 was detected throughout the hanging-drop culture in the absence of BDNF (Fig. 2B, 2D; Table 1). Regardless of BDNF, after 6- to 7-day hanging-drop culture, the formation into the EB dramatically occurred (Fig. 2C, 2D). Simultaneously, migrating neural crest cell markers sox9 and p75 disappeared in the EB formed in the presence of BDNF (Fig. 2C; Table 1). We also detected c-ret immunoreactivity in ES guts differentiated from EBs treated with BDNF (BDNF⫹) after 2 weeks in outgrowth culture but not after 3 weeks (Fig. 2E; Table 1). On the other hand, trkB, NF, and c-kit immunoreactivity was detected after 2 and 3 weeks in outgrowth culture (Fig. 2F). Within ES guts, NF immunoreactive (⫹) cells formed a ganglion within the wall of the dome-like structure surrounding the lumen (Fig. 2F). No trkB, c-ret (Fig. 2E), or NF immunoreactivity was detected at any time in ES guts differentiated from EBs formed in the absence of BDNF (BDNF⫺), whereas c-kit immunoreactivity was identified regardless of BDNF treatment (Table 1). In whole-mount preparations of ES guts differentiated from BDNF⫹ EBs, a group of NF⫹ cells formed several ganglia within the walls of the dome-like structure that surrounds the lumen (Fig. 3A, 3B). Moreover, these ganglia connected by


10⫺9 M) [34, 35], but without LIF. We found that a BDNF concentration of 10⫺8 g/ml was the most appropriate for differentiating enteric neural ganglia without hyperplasia. After 6 days in hanging-drop cultures, the EBs that formed were placed in outgrowth cultures [1, 2, 30, 36] on 100-mm gelatin-coated dishes and allowed to attach in the absence of BDNF. Apparently, addition of BDNF to the hanging-drop cultures was the key step leading to differentiation of ES guts containing enteric neural ganglia, because additional BDNF applied to outgrowth cultures had no effect on formation of enteric neural ganglia. We confirmed that NT-3 and GDNF at 10⫺9 to approximately 10⫺8 g/ml were also able to differentiate enteric nerve fibers and a few or a moderate number of cell bodies, but we did not confirm the differentiation of a single or multiple enteric neural ganglia (Fig. 1). Migration of neural crest stem cells to the gut is dependent on transient expression of ETB receptors [29]. However, differentiation of enteric neural ganglia, identified by NF and PGP 9.5 immunoreactivity, did not occur when an ETB receptor agonist, IRL-1620 (10⫺9 to ⬃2 ⫻ 10⫺9 M), was added to the hangingdrop cultures, the outgrowth cultures, or both.


1418

Table 2. Differences in motility pattern of ES guts BDNF⫺ ES guts BDNF⫹ ES guts

Motility pattern Frequency (times/minute) Motility pattern Frequency (times/minute)

Rhythmic contraction 8.6 ⫾ 5.1 (n ⫽ 5) Rhythmic contraction 12.4 ⫾ 6.4 (n ⫽ 27)

Phasic contraction 5.1 ⫾ 2.1 (n ⫽ 4) Peristalsis-like contraction 6.3 ⫾ 2.7 (n ⫽ 40)a

p ⬍ .0001 versus rhythmic contraction by nonpaired t test. Abbreviations: BDNF, brain-derived neurotrophic factor; ES, embryonic stem.

a

fiber tracts form an enteric neural network. Also, c-kit⫹ cells were present. These mainly multipolar cells did not form a distinct layer but were instead scattered throughout the muscle layer (Fig. 3C, 3D), where they formed a dense network. As previously reported [2], the muscle layers of the ES gut were not organized into longitudinal and circular ones, but the structure of the c-kit⫹ cell network was nevertheless similar to that of the ICC network at the level of the myenteric plexus in the murine small intestine (Fig. 3C) [12]. And in the same ES gut, we detected some c-kit⫹ cells that were similar to ICCs at the level of the deep muscular plexus in the mouse small intestine (Fig. 3D) [17]. Taken together, the data presented here suggest that BDNF specifically differentiates enteric neurons within ES guts from enteric neural crest-derived cells in the EB during the 2–3 weeks of outgrowth culture.

Analysis of ES Gut Motility By recording serial video images at various stages of differentiation, we were able to determine that ES guts typically begin to exhibit spontaneous contractions on day 21 of EB outgrowth culture. As shown in Figure 3E, a tubular (cystic structure) ES gut differentiated from a BDNF⫹ EB exhibited distinct patterns of highly coordinated peristalsis-like contractions that differed from the phasic contractions seen in ES guts that spontaneously

differentiated from a BDNF⫺ EB. These peristalsis-like contractions occurred significantly less frequently than the rhythmic, often local, contractions that were also seen (Table 2).

The Response of Increases in [Ca2ⴙ]i Induced by Focal Stimulation in ES Guts

We found that focal stimulation (5–10 Hz, 100 ␮seconds, 10 pulses) of ES guts differentiated from BDNF⫹ EBs elicited the response of increases in [Ca2⫹]i that propagated widely from the site of stimulation (Figs. 4A, 5A, 6A) and that these evoked [Ca2⫹]i signals could be nearly abolished by TTX (Figs. 5B, 6C). This indicated that voltage-gated Na⫹ channels actually expressed in ES guts and were of particular significance to neural signaling [44]. We did not find any increases in [Ca2⫹]i evoked by focal simulation at multiple sites of six ES guts differentiated from BDNF⫺ EBs (Fig. 4B). This suggested that no neurons existed in the ES guts differentiated from BDNF⫺ EBs. We also detected the response of increase in [Ca2⫹]i as reliable evidence for the existence of an enteric neural network in the ES gut (Fig. 5C, 5D), which included a large NF⫹ enteric neural ganglion that was approximately 80 ␮m in diameter (Fig. 5C) and extensively branching nerve processes (Fig. 5D). In fact, we were able to identify multiple ganglia in ES guts based on NF (Fig. 3A) or PGP 9.5 immunoreactivity (data not shown). Similar observations were made in 17 other ES guts.


Figure 3. NF (A, B) and c-kit (C, D) immunoreactivity in different contracting ES guts on day 21 of EB (BDNF⫹) culture. Arrows in (A) indicate enteric neural ganglia. (E): Series of video images showing a highly coordinated peristaltic contraction of an ES gut on day 21 of EB (BDNF⫹) culture. Abbreviations: BDNF, brain-derived neurotrophic factor; EB, embryoid body; ES, embryonic stem; NF, neurofilament.


1419

Figure 5. Increases in [Ca2⫹]i elicited by focal stimulation in a dome-like ES gut (A). The response originated in the wall in an ES gut differentiated from the EB (BDNF⫹) and propagated widely along its edge (from points 1–2 and 3 in (A). The time courses of the changes in fluo-3 fluorescence (expressed as Ft/F0) at points 1, 2, and 3 are shown in the lower traces. The response was almost completely abolished by TTX (B). NF immunoreactivity in the same ES gut reveals a large enteric neural ganglion (C) with neural cell processes (D). Abbreviations: BDNF, brainderived neurotrophic factor; EB, embryoid body; ES, embryonic stem; NF, neurofilament; TTX, tetrodotoxin.

Consistency of physiological results with immunohistochemical results strongly supported the existence of a physiologically functioning enteric neural network in the ES gut. We were also able to elicit the response of increases in [Ca2⫹]i at multiple sites within each ES gut (Fig. 7A). In this preparation, the fluo-3 fluorescence increased by more than 10% to approximately 30% from the pre-stimulation level for at least 30 seconds. To determine which neurotransmitters contribute to elicit the response of increases in [Ca2⫹]i, we analyzed the effects of a nonselective blocker of muscarinic receptors and a www.StemCells.com

nonselective blocker of nicotinic receptors. The [Ca2⫹]i response was almost completely abolished by pretreatment with 10 ␮M atropine (Fig. 7B), which had a similar inhibitory effect in three other ES guts. Likewise, the [Ca2⫹]i response in four ES guts was largely attenuated by hexamethonium (100 ␮M), which had no effect in one ES gut (Fig. 6B).

DISCUSSION We previously showed that ES cells can give rise to a functional organ-like unit, the ES gut, which consists of a broad array of


Figure 4. Increases in [Ca2⫹]i by focal stimulation in an ES gut on day 21 of EB (BDNF⫹) culture (A) and no increases in [Ca2⫹]i in an ES gut on day 21 of EB (BDNF⫺) culture (B). (A): Increases in [Ca2⫹]i were elicited by focal stimulation in an ES gut on day 21 of EB (BDNF⫹) culture. i, ii, and iii: Three successive [Ca2⫹]i images obtained after focal stimulation. The response originated in the wall of the ES gut and then propagated widely along its edge (from points 1–2 and 3). The time courses of the changes in fluo-3 fluorescence at points 1, 2, and 3 are shown in the right traces. iv: Corresponding bright-field image showing the site of the electrode used for focal stimulation of this ES gut. (B): Increases in [Ca2⫹]i were not elicited by focal stimulation close to points 1 and 3 in an ES gut on day 21 of EB (BDNF⫺) culture. i, ii, and iii: Three successive [Ca2⫹]i images obtained after focal stimulation. The time courses of the changes in fluo-3 fluorescence at points 1, 2, and 3 are shown in the right traces. iv: Corresponding bright-field image showing the site of the electrode used for focal stimulation of this ES gut. Abbreviations: BDNF, brain-derived neurotrophic factor; EB, embryoid body; ES, embryonic stem; NF, neurofilament.

1420


Figure 6. Increases in [Ca2⫹]i elicited by focal stimulation (A) were largely attenuated by hexamethonium (B), and the small remaining increases were abolished by TTX (C). Abbreviation: TTX, tetrodotoxin.

enteric derivatives from all three embryonic germ layers, including various kinds of epithelial cells, smooth muscle cells, and ICCs [1]. The morphological and physiological characteristics of ES guts that differentiate spontaneously from EBs in the absence of exogenous neurotrophic factors have been fully documented [2]; in those preparations, only few, if any, enteric neurons are observed [1, 2]. Thus, we describe here the first successful effort to stimulate differentiation of an ES gut containing enteric neural ganglia; the resultant organ-like cluster exhibited highly coordinated peristalsis-like movements. During early hanging-drop culture in the presence of BDNF, trkB, c-ret, sox9, and p75 were expressed. In BDNF⫹ EBs (in late hanging-drop culture), we detected trkB and c-ret, both of which are expressed during in vivo development of enteric neurons during mouse embryogenesis [24, 39, 43], but not NF or c-kit. After 2 weeks in outgrowth culture, however, we detected trkB, c-ret, NF, and c-kit. Although we did not detect c-ret after 3 weeks in outgrowth culture, we did detect both NF (or PGP 9.5) and c-kit, suggesting that enteric neurons, fibers, and ICCs are differentiated in the ES gut during at least 2–3 weeks in outgrowth culture. It is predictable that formed enteric neurons by BDNF might act in concert with ICCs to mediate the various

patterns of GI motility, including peristalsis-like motor activity [17–20, 45, 46]. Other c-ret ligands, NT-3, and GDNF showed qualitatively similar effects, but those effects on differentiation of enteric neural network structure in the ES gut seemed to be less potent than those of BDNF, although further studies at higher concentrations than the present one are needed. Taken together, these findings suggest a scenario in which BDNF binds to trkB, leading to formation of enteric neural crest cells, which in turn intensely differentiate into enteric neural ganglion cells that form a network able to mediate peristalsis-like motor activity. Notably, this in vitro differentiation process seems reminiscent of ENS development during mouse embryogenesis, when neural crest cells migrate into the GI tract and develop into enteric neurons [39, 40, 43]. The scenario summarized above is supported by our observation that focal electrical stimulation of ES guts elicited distinct increases in [Ca2⫹]i which were widely propagated within the structure. This implies the presence of a physiologically functioning enteric neural network comprised of cell somas, dendrites, and axons and capability of synaptic transmission within ES guts differentiated from BDNF⫹ EBs. Furthermore,


Figure 7. Increases in [Ca2⫹]i elicited by focal stimulation at multiple areas (A) were almost completely abolished by atropine (B).


1421

REFERENCES 1

Yamada T, Yoshikawa M, Takaki M et al. In vitro function gut-like organ formation from mouse embryonic stem cells. STEM CELLS 2002; 20:41– 49.

2

Ishikawa T, Nakayama S, Nakagawa T et al. Characterization of in vitro gutlike organ formed from mouse embryonic stem cells. Am J Physiol Cell Physiol 2004;286:C1344 –C1352.

3

Der-Silaphet T, Malysz J, Hagel S et al. Interstitial cells of Cajal direct normal propulsive contractile activity in the mouse small intestine. Gastroenterology 1998;114:724 –736.

4

5

2⫹

⫹

Fujita A, Takeuchi T, Saitoh N et al. Expression of Ca -activated K channels, SK3, in the interstitial cells of Cajal in the gastrointestinal tract. Am J Physiol Cell Physiol 2001;281:C1727–C1733. Huizinga JD, Zhu Y, Ye J et al. High conductance chloride channels generate pacemaker currents in interstitial cells of Cajal. Gastroenterology 2002;123:1627–1636.

6

Lee JCF, Thuneberg L, Berezin I et al. Generation of slow waves in membrane potential is an intrinsic property of interstitial cells of Cajal. Am J Physiol Gastrointest Liver Physiol 1999;277:G409 –G423.

7

Suzuki H, Takano H, Yamamoto Y et al. Properties of gastric smooth muscles obtained from mice which lack inositol triphosphate receptor. J Physiol (Lond) 2000;525:105–111.

8

Thomsen L, Robinson TL, Lee JC et al. Interstitial cells of Cajal generate a rhythmic pacemaker current. Nat Med 1998;4:848 – 851.

9

Huizinga JD, Thuneberg L, Klu¨ppel M et al. The W/kit gene required for interstitial cells of Cajal and for intestinal pacemaker activity. Nature 1995;373:347–349.

10 Nakamura K, Shibata Y. Connexin43 expression in network-forming cells at the submucosal-muscular border of guinea pig and dog colon. Cells Tissues Organs 1999;165:16 –21.

earlier report that endothelin-3 acts via ETB receptors to inhibit in vitro development of enteric neurons [35].

CONCLUSION We for the first time have succeeded in combining improved technology (EB culture) with appropriate addition of an exogenous neurotrophic factor to specifically induce differentiation of an ES gut equipped with a functional enteric neuronal network structure (“second brain”) mediated via neural-crest formation. This new approach could facilitate significant advances in studies of the integrative physiological functions of the gut and in the development of a complete gut-like organ for regenerative medicine.

ACKNOWLEDGMENTS This work was supported by grants-in-aid for Scientific Research (14657311, 16650090, and 17300130 for M.T. and 15300134 for S.N.) from the Ministry of Education, Science, Sports and Culture of Japan.

DISCLOSURES The authors indicate no potential conflicts of interest.

14 Costa M, Hennig GW, Brookes SJ. Intestinal peristalsis: A mammalian motor pattern controlled by enteric neural circuits. Ann N Y Acad Sci 1998;860:464 – 466. 15 Hukuhara T, Kotani S, Sato G. Comparative studies on the motility of the normal, denervated and aganlionic THIRY-loops. Jpn J Physiol 1962; 12:348 –356. 16 Hukuhara T, Sumi T, Kotani S. The role of the ganglionic cells in the small intestine taken in he intestinal intrinsic reflex. Jpn J Physiol 1961;11:281–288. 17 Nakagawa T, Ueshima S, Fujii H et al. Different modulation of spontaneous activities by nitrergic inhibitory nerves between ileum and jejunum in W/Wv mutant mice. Auton Neurosci 2005;119:25–35. 18 Nakagawa T, Misawa H, Nakajima Y et al. Absence of peristalsis in the ileum of W/Wv mutant mice that are selectively deficient in myenteric interstitial cells of Cajal. J Smooth Muscle Res 2005;41:141–151. 19 Yoneda S, Fukui H, Takaki M. Pacemaker activity from submucosal interstitial cells of Cajal drives high-frequent and low-amplitude circular muscle contractions in the mouse proximal colon. Neurogastroenterol Mot 2004;16:621– 627. 20 Takaki M, Yoneda S, Nakagawa T et al. Role of interstitial cells of Cajal and enteric neurons on gut spontaneous motility. Auton Neurosci 2003; 106:41. 21 Schratt GM, Nigh EA, Chen WG et al. BDNF regulates the translocation of a select group of mRNAs by a mammalian target of rapamycinphosphatidylinositol 3-kinase-dependent pathway during neuronal development. J Neurosci 2004;24:7366 –7377. 22 Cohen-Cory S, Fraser SE. Effects of brain-derived neurotophic factor on optic axon branching and remodeling in vivo. Nature 1995;378:192–196. 23 Schuman EM. Neurotrophin regulation of synaptic transmission. Curr Opin Neurobiol 1999;9:105–109.

¨ rdo¨g T, Koh SD et al. Development and plasticity of 11 Sanders KM, O interstitial cells of Cajal. Neurogastroenterol Mot 1999;11:311–338.

24 Hoehner JC, Wester T, Påhlman S et al. Localization of neurotrophins and their high-affinity receptors during human enteric nervous system development. Gastroenterology 1996;110:756 –767.

12 Torihashi S, Ward SM, Sanders KM. Development of c-Kit-positive cells and the onset of electrical rhythmicity in murine small intestine. Gastroenterology 1997;112:144 –155.

25 Lucini C, Maruccio L, de Girolamo P et al. Localisation of neurotrophincontaining cells in higher vertebrate intestine. Anat Embryol 2002;205: 135–140.

13 Yoneda S, Takano H, Takaki M et al. Properties of spontaneously active cells distributed in the submucosal layer of mouse proximal colon. J Physiol (Lond) 2002;542:887– 897.

26 Chalazonitis A, Rothman TP, Chen J et al. Neurotophin-3 induces neural crest-derived cells from fetal rat gut to develop in vitro as neurons or glia. J Neurosci 1994;14:6571– 6584.

www.StemCells.com


the finding that [Ca2⫹]i signaling was blocked by atropine or hexamethonium suggests that excitatory cholinergic enteric neurons contribute to the [Ca2⫹]i signaling mediated via nicotinic receptors, although subtypes of these muscarinic and nicotinic receptors were not determined. We also identified an ICC network in the ES gut, making it plausible that interactions among enteric neurons, ICCs, and smooth muscle cells led to the highly coordinated peristalsis-like movements exhibited by the ES guts differentiated from BDNF⫹ EBs like in murine small intestine [18]. Such peristaltic movements are rarely seen in spontaneously differentiated ES gut, which lacks enteric neurons (Table 2), supporting the postulate that both the ICC network and ENS are essential for generation of highly coordinated peristalsis in the ES gut [18]. Highly coordinated peristalsis is a physiologically essential function to transport intraluminal contents in the GI tract, although the present ES gut is not able to transport the intraluminal contents because of the lack of an orifice (cystic structure) [1, 2]. Although ETB receptors are briefly required to mediate migration of gut neural crest stem cells [29], the ETB receptor agonist IRL-1620 did not stimulate differentiation of enteric neural ganglia in the ES gut. This result is consistent with an


1422

27 Burns AJ, Pasricha PJ, Young HM. Enteric neural crest-derived cells and neural stem cells: Biology and therapeutic potential. Neurogastroenterol Motil 2004;16(suppl 1):3–7.

37 Nakayama S, Torihash S. Spontaneous rhythmicity in cultured cell clusters isolated from mouse small intestine. Jpn J Physiol 2002;52:217– 227.

28 Gershon MD. Genes and lineages in the formation of the enteric nervous system. Curr Opin Neurobiol 1997;7:101–109.

38 Torihashi S, Fujimoto T, Trost C et al. Calcium oscillation linked to pacemaking of interstitial cells of Cajal. J Biol Chem 2002;277:19191– 19197.

29 Kruger GM, Mosher JT, Tsai YH et al. Temporally distinct requirements for endothelin receptor B in the generation and migration of gut neural crest stem cells. Neuron 2003;40:917–929. 30 Niwa H, Miyazaki J, Smith AG. Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells. Nat Genet 2000;24:372–376. 31 Keller GM. In vitro differentiation of embryonic stem cells. Curr Opin Cell Biol 1995;7:862– 869. 32 Robertson EJ. In: Robertson EJ, ed. Teratocarcinomas and Embryonic Stem Cells: A Practical Approach. 1st ed. Washington, DC: IRL Press, 1987:71–112. 33 Kume T, Nishikawa H, Tomioka H et al. p75-mediated neuroprotection by NGF against glutamate cytotoxicity in cortical cultures. Brain Res 2000;852:279 –289.

40 Young HM, Turner KN, Berger AJ. The location and phenotype of proliferating neural-crest-derived cells in the developing mouse gut. Cell Tissue Res 2005;320:1–9. 41 Hong C-S, Saint-Jeannet J-P. Sox proteins and neural crest development. Semin Cell Dev Biol 2005;16:694 –703. 42 Trainor PA. Specification of neural crest cell formation and migration in mouse embryos. Semin Cell Dev Biol 2005;16:683– 693. 43 Gershon MD. Lessons from genetically engineered animal models. II. Disorders of enteric neuronal development: Insights from transgenic mice. Am J Physiol Gastrointest Liver Physiol 1999;277:G262–G267. 44 Bains RA, Pym ECG. Determinants of electrical properties in developing neurons. Semin Cell Dev Biol 2006;17:12–19.

35 Wu JJ, Chen J-X, Rothman TP et al. Inhibition of in vitro enteric neuronal development by endothelin-3: Mediation by endothelin B receptors. Development 1999;126:1161–1173.

45 Ward SM, Beckett EA, Wang X et al. Interstitial cells of Cajal mediate cholinergic neurotransmission from enteric motor neurons. J Neurosci 2000;20:1393–1403.

36 Hooper M, Hardy K, Handyside A et al. HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature 1987;326:292–295.

46 Ward SM, Sanders KM, Hirst GDS. Role of interstitial cells of Cajal in neural control of gastrointestinal smooth muscles. Neurogastroenterol Motil 2004;16 (suppl 1):112–117.

See www.StemCells.com for supplemental material available online.


34 Takai M, Umemura I, Yamasaki K et al. A potent and specific agonist, Suc-[Glu9, Ala11,15]-endothelin-1(8 –21), IRL 1620, for the ETB receptor. Biochem Biophys Res Commun 1992;184:953–959.

39 Young HM, Hearn CJ, Ciampoli D et al. A single rostrocaudal colonization of the rodent intestine by enteric neuron precursors is revealed by the expression of Phox2b, Ret, and P75 and by explants grown under the kidney capsule or in organ culture. Dev Biol 1998;202:67– 84.

In Vitro Formation of Enteric Neural Network Structure in a Gut-Like Organ Differentiated from Mouse Embryonic Stem Cells Miyako Takaki, Shinsuke Nakayama, Hiromi Misawa, Tadashi Nakagawa and Hiroki Kuniyasu Stem Cells 2006;24;1414-1422; originally published online Mar 9, 2006; DOI: 10.1634/stemcells.2005-0394 This information is current as of April 11, 2008 Updated Information & Services

including high-resolution figures, can be found at: http://www.StemCells.com/cgi/content/full/24/6/1414

Supplementary Material

Supplementary material can be found at: http://www.StemCells.com/cgi/content/full/24/6/1414/DC1
