SEARCH

SEARCH BY CITATION

Keywords:

  • Adult stem cells;
  • Neural;
  • Intestine;
  • Multipotent

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Little is known about postnatal enteric nervous system (ENS) development, but some reports suggest that the postnatal bowel may contain neural stem cells. Therefore, we created an in vitro model of desegregation using an enzymatic and mechanical tissue technique. This approach yielded a group of cells from the small intestine of lactating and adult mice, which ex vivo attach to the culture dish; actively proliferate; and express nestin, vimentin, and the pro-neural transcription factors neurogenin-2 (ngn-2), Sox-10, and Mash-1. In the conditions grown, double immunostains suggest that they differentiate into various cell types, particularly neurons, smooth muscle, and glia including 04 protein–positive cells. They also express the neurotrophic-protein tyrosine kinase (Trk) receptors TrkA, TrkB, and TrkC; the low-affinity neurotrophin receptor p75NTR; and the glial-derived neurotrophic factor receptors (GFR)α-1, GFRα-2, and GFRα-3. The neurons expressed several sensory and motor neurotransmitters present in the central and enteric nervous systems, including calcitonin gene-related peptide, neuropeptideY, peptideYY, substance P, vasoactive intestinal polypeptide, and galanin; along with glia, these neurons formed elaborate intercellular connections. They also express c-KIT, CD34, CD20, and CD45RO, suggesting they either have a hematogenous origin or may differentiate toward hematogenous lines. These findings suggest that these cells may be enteric neural stem cells (ENSCs); may normally be present in the small intestine; and may have the capacity to proliferate and differentiate into neurons, glia, and smooth muscle. Further identification and purification of intestinal ENSCs will provide a means to study the regulation of their differentiation and should give insight into the mechanisms involved in development and remodeling of the ENS. The possible therapeutic application of postnatal stem cells such as ENSCs needs to be evaluated, including their use for transplantation in the central nervous system.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Stem cells are unique in their capacity for self-renewal and the capability of generating different cell types [13]. Because of the potential these cells may have in treating a wide variety of human disease, great interest has been generated in their identification, isolation, proliferation, and differentiation [47]. Due to ethical issues, there is an emphasis on developing techniques for the possible clinical use of stem cells derived from postnatal tissue rather than embryonic tissue [3, 5, 8].

Multipotent stem cells have been previously isolated and cultured postnatally from human and rodent brain, pancreas, liver, bone marrow [916], and, recently, intestine [17]. In the bowel, enteric neural stem cells (ENSCs) are reported to express the neurotrophin receptor p75NTR and are believed to be derived from the neural crest [17]. They persist throughout life and are capable of differentiating to neurons, glia, and other cell types. The pattern of differentiation of ENSCs and the signal molecules that direct their replication, survival, and differentiation have been little explored. This information is valuable to further understand enteric nervous system (ENS) development, particularly its maintenance in postnatal life, which may shed further light on the pathophysiology of ENS disease.

In addition, the bowel is an organ easily accessible by biopsy and has a very rich innervation supplied by the ENS, which although considered part of the peripheral nervous system (PNS), shares many similarities with the central nervous system (CNS). This opens the possibility of using ENSCs to treat either CNS or PNS pathology in cell-transplant therapy. We therefore sought to investigate the timing and pattern of differentiation of ENSCs, as well as the expression of selected cytoskeletal proteins, neurotransmitters, pro-neural transcription factors, and neurotrophic protein receptors. This information is vital in understanding ENSCs and their manipulation for possible therapeutic use.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Tissue Dissociation

For each of the culture procedures, the entire small intestine of five to eight lactating BALB-c mice (3–5 days old) was used. For each of the experiments in adult mice, the entire small intestine of one adult animal (6–12 months old) was used. The animals were anesthetized with an i.p. dose of ketamine. The entire small intestine was removed, washed thoroughly in a Hank's Balanced Salt Solution (HBSS), and finely diced. Tissue fragments were digested in a combination of dispase type I (0.1 mg/ml; Boehringer Ingelheim, Ingelheim am Rhein, Germany, http://www.boehringer-ingelheim.com), crude collagenase type XI (300 U/ml; Sigma Chemical Corp., St. Louis, http://www.sigma-aldrich.com), and DNase I (10 μg/ml; Boehringer Ingelheim) for 30 minutes at 37°C. Adult mice tissue was processed in the same way as the lactating mice; however, the enzyme concentration (collagenase-dispase-DNase) was double that used for the neonatal rodent intestinal culture.

Once the tissue was digested, the cells were dispersed by trituration, using progressively narrower pipettes, for approximately 30 minutes. Large tissue fragments were removed and discarded by differential sedimentation in 2% sorbitol-containing medium and then via passage across a sterile, 200-μ pore nylon mesh. Approximately 20,000 cellular clusters were obtained from the lactating rodents. The pellet was resuspended in growth media (Dulbecco's Modified Eagle's Media high glucose, 2.5% fetal bovine serum, 10 ng/ml epidermal growth factor, penicillin-streptomycin-glutamine mix at a final concentration of penicillin 10,000 units, streptomycin 10 mg, glutamine 29.2 mg/ml [Invitrogen Life Sciences, Carlsbad, CA, http://www.invitrogen.com]) and 0.25 U/ml insulin (Sigma)], and plated at a density of approximately 50–100 cell fragments per 25 mm2 of culture surface.

Cell Culture

The cells were plated onto sterile Labtek eight-chamber slides (Nalge Nunc International, Naperville, IL, http://www.nalgenunc.com) or Petri dishes, then grown in culture for various periods of time at 370°C with 5% CO2, changing 50% of the media the day after plating and then every other day.

Immunostaining and Cell Quantification

At different time points, cells were fixed in 4% para-formaldehyde (at 40°C) for 20 minutes. Cells were pre-blocked in CAS-Block (Zymed, San Francisco, California; http://www.zymed.com) for 30 minutes at room temperature. Primary antibody incubations were held overnight at 4°C; secondary incubations with anti-mouse, anti-goat, or anti-rabbit antibodies conjugated with fluorescein isothio-cyanate (FITC) or indocarbocyanine (Cy3; Jackson Immuno Research, West Grove, PA, http://www.jacksonimmuno.com/) were done for 2 hours at room temperature. Immunohistochemistry was performed on cell cultures and on paraffin-embedded sections of mice intestine using the DAKO LSAB System (Dako; Carpinteria, CA, http://www.dakousa.com). The primary antibodies, the concentration used, and the manufacturer were as follows: nestin (1:100; Chemicon, Temecula, CA, http://www.chemicon.com); bromodeoxyuridine (BrdU; 1:50, Zymed); neurogenin-2 (ngn-2; 1:100; Chemicon); Sox-10 (1:100; CeMines; Evergreen, CO, http://www.cemines.com); Mash-1 (1:100; CeMines); PGP 9.5 (1:500; CeMines); S-100 (ready to use; Zymed, South San Francisco, CA, http://www.zymed.com); glial fibrillary acidic protein (GFAP; ready to use; Zymed); protein 04 (1:100; Chemicon); b-tubulin Type III (1:200; Zymed); tau (1:200; Dako); smooth muscle actin (SMA; 1:100; Dako); neurofilament 160/200 kDa (NF-M+H; 1:100; Zymed); neuronal nitric oxide synthase (nNOS; 1:25; Zymed); acetyl-cholinesterase (AChE) (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com); EAAC1 (1:200; Zymed); calcitonin-gene related peptide (CGRP; 1:50; Peninsula Laboratories, San Carlos, CA, http://www.penlabs.com); neuropeptide Y (NPY; 1:50; Peninsula Laboratories); peptide YY (1:50; Peninsula Laboratories); substance P (1:50, Peninsula Laboratories); vasoactive intestinal peptide (VIP; 1:50; Peninsula Laboratories); galanin (1:50; Peninsula Laboratories); GDNF family receptor alpha 1 (GFRα-1; 1:100); GFRα-2 (1:100); GFRα-3 (1:100); tyrosine kinase (Trk)A (H-190) (1:100); TrkB (H-181) (1:100); TrkC (798 and C-14) (1:100; all Trk samples from Santa Cruz Biotechnology); Synaptophysin (ready to use; Zymed); CD34 (1:100; Biomeda, Foster City, CA, http://www.biomeda.com); CD20 (1:100; Santa Cruz Biotechnology); c-KIT (1:100; Santa Cruz Biotechnology); CD45RO (1:100; Santa Cruz Biotechnology).

Each immunostain was performed at least three different times. All cells in each chamber of the Labtek chamber slides (stained and nonstained) were counted to calculate the percentage of cells that express each protein. If this was not possible due to a very large number of cells, a minimum of 1,000 cells was counted for each experiment.

BrdU Assay

Cell cultures were maintained for 24 hours with 1 μM 5-bromo-2′-deoxyuridine added at the beginning of the culture period. Cells were fixed as described, then processed according to the Zymed BrdU Staining Kit instructions (Zymed).

Image Processing and Analysis

Fluorescent samples were evaluated using a Bio-Rad (Hercules, CA) MRC600 confocal imaging system fitted to a Zeiss Axioskop microscope, which allows for simultaneous evaluation of two separate fluorophores. Peroxidase and alkaline phosphatase staining were visualized using a Leica DMLS microscope and photographed with a Pixera digital camera. Images obtained and reproduced in this article are representative of what was observed in the entire culture chamber.

RNA Isolation and RT-PCR

Total RNA was isolated from cell cultures, using TRIzol reagent (Invitrogen Life Sciences); and reverse transcription polymerase chain reaction (RT-PCR) was performed by standard methodology [18] by using 2 μg of total RNA and the following DNA primers: Nes-F, 5′-CAA GAA CCA CTG GGG TCT GT-3′ (forward) and Nes-R, 5′-ACA TCC TCC CAC CTC TGT TG-3′ (reverse); Tau-F, 5′-GGT GGC AAG GTG CAA ATA GT-3′ (forward) and Tau-R, 5′-GCC AAG GAA GCA GAC ACT TC-3′ (reverse); GFAP-F, 5′-CAC GAA CGA GTC CCT AGA GC-3′ (forward) and GFAP-R, 5′-CCT TCT GAC ACG GAT TTG GT-3′ (reverse); GAPDH-F, 5′-GAC CCC TTC ATT GAC CTC AAC TAC-3′ (forward) and GAPDH-R, 5′-TGG TGG TGC AGG ATG CAT TGC TGA-3′ (reverse). SMA-F, 5′-CTG ACA GAG GCA CCA CTG AA-3′ (forward) SMA-R, 5′-GAA GGA ATA GCC ACG CTC AG-3′ (reverse).

PCR products were separated by base pair size on agarose gels by using standard protocols [18].

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Establishment of Culture, Attachment, and Development of Enteric Neural Cells (ENCs)

In neonatal rodent tissue, cells originating from approximately 5%–10% of plated tissue fragments attached and survived. In the adult rodent, although the yield was much lower than for the neonate, cells originating from about 1% of clusters attached and survived within the first 24 hours. During the next 24 hours, cells emerged and migrated from the attached cellular clumps. Elongated processes originated and extended from the cell clusters as early as 24 hours after plating. Over the next few days, the cells proliferated, migrated, attached, and formed elaborate connections with each other, bridging the areas between nests of cells. After 7 days, neonatal rodent cultures were approximately 75% confluent, and adult rodent cultures were approximately 10%–15% confluent.

During the initial 2–5 days of culture, BrdU staining revealed nearly 100% positivity in the central areas of each attached cell cluster. As the cells migrated peripherally and differentiated, the percentage of BrdU-positive cells dropped to approximately 50% in more peripheral areas (Fig. 1A). Double immunostains revealed that some of the BrdU+ cells in the peripheral areas of the cell nests expressed synaptophysin as early as day 2 of culture (Fig. 1B). This suggests that some of the proliferating cells may be neuronal progenitors or stem cells. Despite the fact that these cells were still dividing by day 2 or 3, some had established elaborate inter-cellular processes (Fig. 1C).

thumbnail image

Figure Figure 1.. Immunostains for BrdU, nestin, and vimentin. (A): BrdU staining, culture day number 2. There are nearly 100% positive cells in the center of the cluster (dark nucleus). This drops to approximately 40%–50% positive cells in the more peripheral areas of the cluster. (B): Cells positive to BrdU (dark periphery of nucleus) and also to synaptophysin (cells look red, arrows). (C): BrdU+ cells forming elaborate intercellular connections (cell marked by blue arrow); those negative to BrdU are not forming connections (right lower corner of image). (D): On culture day 1, a few nestin+ cells in the cell cluster are seen, with elongated processes (arrows). (E): On culture day 2 to 7, nestin+ cells increase in number and migrate peripherally, leaving the cluster and acquiring morphology suggestive of glia (arrowhead) or neurons (arrow) to form (F) extensive cell-to-cell networks. (G): Vimentin expression of enteric neural cells. Abbreviation: BrdU, bromodeoxyuridine. See online version for color figures.

Download figure to PowerPoint

Nestin, an intermediate filament type VI, has been extensively used to identify neural stem cells. As early as culture day 1 or 2 (Fig. 1D), approximately 85% of the proliferating cells (about 30% of total cells in culture) were positive to a specific monoclonal anti-nestin antibody [19]. By day 21, still close to 30% of total cells in culture were nestin+. As nestin+ cells migrated peripherally from the core of the cell nests, their cytoplasm became more compact, and they produced elongated processes suggestive of neural or glial cells (Fig. 1E). These processes grew in size, branched, and formed complex networks of cells (Fig. 1F).

Nearly all nestin+ cells were positive for vimentin (Fig. 1G). Vimentin is expressed in stem cells originating from other organs as well [20]. The dramatic morphological changes that enteric neural cells (ENCs) undergo in their differentiation to neurons are shown in Figure 2(A–F). Figure 2F shows many nestin+/BrdU+ cells in various stages of replication and differentiation.

thumbnail image

Figure Figure 2.. Differentiation of stem cells into neurons. (A): Nestin (red) and bromodeoxyuridine (dark periphery of nuclei)–positive cells (ENCs), culture day 1: diffuse cytoplasm, jagged edge, undergoing nuclear division. (B): Day 2: the cell body is more compact, cytoplasm is denser, cell borders are better defined, elongated processes suggestive of neurites appear (arrows), and nuclear division continues. (C): Day 3: cytoplasm is denser, the cell is more elongated, cell borders are better defined, processes are more elongated (arrows), and two nuclei are seen. (D): Day 3: neurites are specifically directed toward other stem cells in proximity (arrows). (E): Day 4: the cell is more elongated and appears to now have a long process suggestive of being an axon (arrow) and smaller ones, possibly dendrites (arrowheads). Two nuclei are still present. (F): Day 6: neurons (arrows) with single nuclei, forming cell-to-cell connections. The ENCs continue to proliferate and differentiate, and thus the cell network contains ENCs at various stages of differentiation. Abbreviation: ENC, enteric neural cell. See online version for color figures.

Download figure to PowerPoint

Differentiation of ENCs and Expression of Neurotransmitters by Immunostains

All cell colonies originating from the cells that attached contained nestin-positive cells; after day 4, they also contained glial and neuronal cells. Double immunofluorescence studies, performed daily, revealed that ENCs and glia were initially positive to nestin, which tended to disappear as the cells differentiated into glia (Fig. 3A, 3B) or neurons (Fig. 3C–E). ENCs also appeared differentiated to a smooth muscle–like phenotype when we used anti-SMA antibodies in the double immunostains (Fig. 3F). Similarly to neurons and glia, nestin expression tended to disappear as α-actin expression increased. In some of the clusters, we also observed a small population of cells positive for 04 (Fig. 3G), a protein expressed in Schwann cells and immature oligodendrocytes. Schwann cells are found in the PNS and oligodendrocytes in the CNS; neither are usually found in the ENS [21].

thumbnail image

Figure Figure 3.. Immunostains for glial and neuronal specific proteins and smooth muscle actin (culture day 7). (A): Glial fibrillary acidic protein (red) and nestin (green, coexpression yellow) expression of ENCs. Nestin expression has almost disappeared as the ENCs differentiate into glia. (B): S-100 expression of ENC-differentiated glia. (C, D): Tau (red) and nestin (green, coexpression yellow) expression of ENCs. Nestin expression is seen initially, when tau is expressed. (E): β-tubulin type III expression (red) of ENCs that have differentiated into neurons (nestin expression green). In later stages of neuronal differentiation, nestin expression disappears (one nestin+ cell persists in the center of photograph). (F): Smooth muscle actin expression (green) and nestin (red) from ENC-derived smooth muscle cells. (G): Protein 04 expression of ENC-derived glia. (H, I): Nestin-expressing ENC (red) elongates and form intercellular connections with neurofilament (160/200 kDa)–expressing cells (brown). (J): PGP 9.5 expression in ENC-derived neurons. (K): Expression of glutamate transporter EAAC1 (green) and of neuronal nitric oxide synthase (red, coexpression yellow) in ENCs. (L): Acetylcholinesterase and (M) synaptophysin expression of ENC-derived cells. Abbreviation: ENC, enteric neural cell.

Download figure to PowerPoint

Cell processes from nestin+ ENCs and from tau- or neuro-filament-positive cells spatially interrelated with each other, suggesting that there was cell-to-cell interaction between the ENCs and the cells expressing neuronal proteins (Fig. 3H, 3I). In addition, many pure nestin+ cells, which were negative to glial and neuronal proteins, remained in the culture during the entire culture period.

After culture day 4, all colonies were abundant for neuronal cells positive for tau (Fig. 3C, 3D), β-tubulin type III [22] (Fig. 3E), and 160/200 kDa neurofilaments (Fig. 3H, 3I), as well as other neuronal cytoskeletal proteins, synaptic proteins, and enzymes including PGP 9.5 (Fig. 3J), the glutamate transporter EAAC1 (Fig. 3K), acetylcholinesterase (Fig. 3L), synaptophysin (Fig. 3M), and nNOS (Fig. 3K). The cells also expressed several neurotransmitters including calcitonin gene-related peptide (Fig. 4A), neuropeptide Y (Fig. 4B), peptide YY (Fig. 4C), substance P (Fig. 4D), vasoactive intestinal polypeptide (Fig. 4E), and galanin (Fig. 4F) [2325], all of which have been described in both the central and enteric nervous systems.

thumbnail image

Figure Figure 4.. Immunostains for neurotransmitters (culture day 7). (A): Immunohistochemistry using anti-calcitonin-gene related peptide antibodies. (B): Immunohistochemistry using anti-NPY antibodies. (C): Immunofluorescence using anti-NPY antibodies. (D): Immunohistochemistry using anti–substance P antibodies. (E): Immunohistochemistry using anti–vasoactive intestinal peptide antibodies. (F): Immunohistochemistry using anti-galanin antibodies. Abbreviation: NPY, neuropeptide Y.

Download figure to PowerPoint

Analysis of mRNA Expression by RT-PCR

We used RT-PCR to further analyze the pattern of differentiation we had seen by immunostains and to verify that the results we observed were not due to the possible presence of previously differentiated neurons that had detached from the tissue digestion and had survived in culture. For this, specific oligonucleotides for the gene, which encodes for the same specific-cell type proteins we had examined in the immunostains, were designed and employed. RNA was extracted from cultured cells at different time points, starting 24 hours after they had been plated.

We found no expression of tau on culture day number 1 or 2 (Fig. 5B). This suggests that no neurons had attached to the culture surface and survived under our culture conditions. Already on culture day 1, however, there was expression of nestin (Fig. 5A). Nestin expression increases during the first 4 days of culture, then persists in a stable manner throughout the 21 days of culture. Tau expression appears on day 4 and continues to increase up to culture day 16, then slightly decreases at day 21. This is consistent with what we observed in the immunostains, where we observed neurons appearing on day 4 and increasing in number as the culture progressed. Also, nestin+ cells were abundant from very early stages of the culture and increased in number, so that nestin expression levels remained very high during the 21 days of culture. This is also consistent with the BrdU studies, which showed that cells continue to replicate. GFAP and α-smooth muscle actin are already expressed in culture day 1 and subsequently increase their level of expression (Fig. 5C, 5D). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is included as a control of endogenous gene expression (Fig. 5E).

thumbnail image

Figure Figure 5.. Timing of differentiation of enteric neural cells in culture by reverse transcription polymerase chain reaction for nestin, tau, GFAP, and SMA mRNA expression. The horizontal axis depicts the days of culture at which the RNA was obtained: (A) nestin, (B)tau, (C) GFAP, (D) SMA, and (E) glyceralde-hyde-3-phosphate dehydrogenase. There is no tau expression on culture days 1 and 2; it begins on day 4 and continues to increase up to days 16 to 21. Expression of nestin, GFAP, and SMA is already present on day 1 and subsequently increases. Abbreviations: GFAP, glial fibrillary acidic protein; SMA, smooth muscle actin.

Download figure to PowerPoint

Pro-neural Transcription Factor Expression

Starting on culture day number 4, we found that in the nestin+ cells that differentiated toward neuron or glia there was a predominance of neurons over glia (approximately 70% neurons, 30% glia), and this persisted during the first week of culture. To explore if this was due to the level of expression of pro-neural transcription factors, we examined ngn-2, Sox-10, and Mash-1, three important pro-neural transcription factors in CNS and neural crest development [26, 27].

Starting on culture day 2, more than 90% of nestin+ cells expressed the pro-neural transcription factor ngn-2 (Fig. 6A), a pro-neural basic helix-loop-helix (bHLH) transcription factor and inducer of neuronal differentiation process [26]. Equally as abundant was the expression of the pro-neural transcription factor Sox-10 (Fig. 6B). A third pro-neural transcription factor we found expressed, but to a lesser degree, was Mash-1, also a proneural bHLH transcription factor expressed in precursors of sympathetic, parasympathetic, and enteric neurons [27]. In our cultures, Mash-1 was expressed in only 6%–9% of nestin-positive cells (Fig. 6C). In the CNS, like ngn-2, Mash-1 is involved in lineage restriction of cortical progenitors, promoting the acquisition of the neuronal fate and inhibiting the astrocytic fate [26].

thumbnail image

Figure Figure 6.. Immunostains for transcription factors and neurotrophic protein receptor expression (culture day 7). (A): ngn-2 (red, nuclear) and nestin (green, cytoplasmic) expression of ENCs. More than 90% of neural stem cells express the transcription factor ngn-2. (B): Sox-10 (red, nuclear) and nestin (green, cytoplasmic) expression of ENCs. More than 90% of ENCs express the transcription factor Sox-10. (C): Mash-1 expression (nuclear) of ENCs. (D): TrkA expression (red, membrane) of ENCs. (E): TrkB expression (red, membrane) of ENCs. (F): TrkC expression (red, membrane) and nestin (green, cytoplasmic) of ENCs. (G): Immunohistochemistry for GDNF receptor GFRα1 expression. (H): GDNF receptor GFRα2 expression. (I): GDNF receptor GFRα3 expression. (J): p75NTR expression. Abbreviations: ENC, enteric neural cell; GDNF, glial line–derived neurotrophic factor; ngn-2, neurogenin-2; Trk, tyrosine kinase. See online version for color figures.

Download figure to PowerPoint

Neurotrophic Protein Receptor Expression and Hematogenous Lineage Expression

Neural stem cells are regulated by signaling proteins from their microenvironment. To understand which signaling proteins may be involved in regulating behavior of ENCs, we examined receptor expression for several neurotrophic proteins that are important in ENS and CNS development and maintenance. Unless otherwise noted, these immunostains were performed in cultures that were 7 days old.

We found that ENCs express neurotrophin receptors TrkA, TrkB, and TrkC (Figure 6D–F) and the low-affinity neurotrophin receptor p75NTR (Fig. 6J), as well as the glial line–derived neurotrophic factor (GDNF) receptors GFRα-1, GFRα-2, and GFRα-3 (Fig. 6G–I); brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3), the principal known ligands for these receptors, have important roles in ENS development and differentiation [2830]. p75NTR is expressed during neural crest–derived cells in embryonic development [31], and its expression has been used by other groups to isolate ENSCs [17]. In our experience, however, we found that only approximately half of ENCs expressed p75NTR (Fig. 6J), even in the initial steps of the culture (days 2–7), a finding contrary to previous reports of 100% expression of p75NTR in ENSCs [32]. This may suggest that nestin+, p75NTR+ cells may be a different cell type than the nestin-positive, p75NTR–ENSCs. By immunostains, expression of nestin in sections of rodent small bowel localized abundantly to the mucosa, particularly around blood vessels (Fig. 7A). Although some of the nestin activity observed may be due to endothelial cells that may express nestin [33], this finding encouraged us to look for hematogenous marker expression in the ENCs to investigate the possibility of a hematogenous origin of some of the ENCs. We found abundant expression of c-KIT, CD34, CD20, and CD45RO (Figure 7c–7f), suggesting that there may be more than one population of ENCs present in the bowel and that the origin of at least some of the ENCs may be hematogenous.

thumbnail image

Figure Figure 7.. Immunostains for nestin in sections of rodent intestine and expression of CD20, CD34, CD45RO, and c-KIT in ENC cultures (culture day 7). (A): Nestin expression in sections of healthy rodent small bowel. The nestin expression is especially strong around blood vessels. (B): Negative control. Immunohistochemistries in ENC cultures for (C) CD20, (D) CD34, (E) CD45RO, (F) and c-KIT (C–F: phase contrast microscopy, ×10). Abbreviation: ENC, enteric neural cell. See online version for color figures.

Download figure to PowerPoint

Subsequent Passages of ENC Cultures

In separate experiments, the cells were detached 7 days after the initial culture, divided, and then re-plated in new Petri dishes at a density of approximately 4,000 cells/cm2. The re-plated cells attached and continued proliferating. In these secondary cultures, more than 60%–70% of cells were positive for nestin (Fig. 8A, 8B). Similar to primary cultures, in immunostains as cells differentiated toward neurons, glia, or smooth muscle, a loss of nestin expression was noted. Once the secondary cultures were at 70%–80% confluence(approximately day 7), the cells were again detached and re-plated. This process was repeated at least five times, resulting in a successful selection of proliferating ENCs, which differentiated into neurons, glia, and smooth muscle (Figure 8c and 8d). Thus, based on double immunostains, these cells derived from the intestinal tissue seemed to exhibit all the typical in vitro properties in previously described neural stem cells, including self-renewal and generation of different cell types, such as neurons, glia, and smooth muscle cells.

thumbnail image

Figure Figure 8.. Immunohistochemistry for nestin expression in secondary culture and neurons in a culture that had been split and re-plated five times. (A, B): Secondary ENC cultures (culture day 5). Note how abundant nestin+ cells are (A, ×20; B, ×10). (C, D): Neurons originated in ENC culture that had been split and re-plated five consecutive times (culture day 7; phase contrast ×20). Abbreviation: enteric neural cell. See online version for color figures.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In this work, we describe an in vitro culture system where we observed the presence, replication, and differentiation of nestin+ cells in the postnatal intestine. The replication and differentiation pattern of these cells suggested that they are ENSCs. To accomplish this, we used a very different approach than that used by Kruger et al. [17], who used an isolation technique based on selecting intestinal p75NTR+ cells using a cell sorter. The p75NTR+ isolated by this group also displayed an in vitro replication and differentiation pattern that suggests they may have been ENSCs. Our technique allowed intestinal desegregated cells to attach to a cell culture dish, a method similar to that used to isolate bone marrow–derived stromal stem cells [34]. In our experience based on double immunostains, we found that some of the cells that spontaneously attached and proliferated were nestin+ and coexpressed neuronal and glial proteins and α-actin. Subsequent stains suggested that as the neural, glial, and smooth muscle proteins increase the level of expression, nestin expression decreases. Nestin expression by neuroepithelial cells has been reported to sharply downregulate at the transition from proliferating stem cell to postmitotic neurons [35].

This system provides the possibility of culturing a large population of nestin+ ENCs and to study them using immunostaining techniques. However, it is a co-culture that does contain other mature cell types which normally express nestin, as, for example, endothelial cells, and may also contain other contaminating mature cells.

The RT-PCR data we present in this work suggest that no previously differentiated neurons detached from the intestines and survived past the first hours of culture. Thus, apparently, mature neurons did not contaminate our cultures. Since nestin is not expressed in neurons [36], the RT-PCR and immunostains suggest that the neurons produced in this system emerged from the nestin+ cells. Because glia and smooth muscle cells—both cells with the ability to proliferate—are already present in culture on day 1, we cannot exclude glial and smooth muscle contamination. Nestin may be expressed in subsets of enteric glial cells, but we did not find it to be expressed in mature intestinal smooth muscle cells before they were isolated (data not shown); this is also reported by others [37]. Thus, the double immunostains with cells positive for nestin and α-actin suggest that the nestin+ cells may also give rise to smooth muscle cells. Conversely, some smooth muscle cells may begin to express nestin in these conditions. However, similar to the neurons, nestin expression tended to disappear as α-actin expression increased. These data suggest that a subset of nestin+ cells in the intestine self-replicates and may give rise to diverse cell types, including neurons, smooth muscle cells, and possibly glia. Although neural stem cells have been isolated successfully from embryonic intestine, reports of postnatal intestinal neural stem cells are only recently emerging. We were able to culture them from the adult mouse, thus, possibly analogous to the brain, neural stem cells appear to be present into adulthood in the intestine as well.

The origin of ENSCs is unknown. In the mammalian intestinal tract, the intestinal mucosa is known to have mucosal stem cells in each crypt [3839]. To date, there are no specific markers for this mucosal stem cell [40], and it has not been shown to generate neural lineages; thus ENSCs and the mucosal stem cells do not appear to be the same cells. We cannot exclude transdifferentiation, which has been described in the literature [41, 42], although the data are controversial [43, 44]. Stem cells have been previously isolated from the liver [4547] and pancreas [12, 15, 48], which have a common embryonic origin with the intestine. Conversely, the ENSCs may originate from the complex enteric nervous system, which embryologically derives from neural crest cell migration. The enteric nervous system is also involved in pancreatic [49] and hepatic [50] innervation.

In our model, only approximately 50% of the ENCs we studied expressed p75NTR, which is found in neural crest–derived cells, but they were all positive to nestin. In whole gut, we found nestin expression mostly around mucosal blood vessels. Additionally, many of the ENCs in this model express c-KIT, CD34, CD20, and CD45RO, all markers associated with hematogenous lineages, which may suggest that there may be more than one origin of ENCs and that a percentage of these may be hematogenously acquired. Another explanation may be that some of the ENCs differentiate toward hematogenous lines. More studies are required to understand the origin of intestinal neural stem cells.

Of the ENCs that differentiated toward neurons or glia, we observed a predominance of neurons over glia (approximately 70% neurons, 30% glia). This may be linked to the high number of cells expressing ngn-2 and Sox-10, pro-neural bHLH transcription factors that are important inducers in the neuronal differentiation process [26, 27, 5155]. Mash-1, also a pro-neural bHLH transcription factor, was expressed in fewer than 10% of the ENCs.

We also observed that a small population of ENCs differentiated into cells positive for 04, a protein expressed in Schwann cells and immature oligodendrocytes. Neither Schwann cells nor oligodendrocytes normally exist in the ENS [21]. One possible explanation for this phenomenon may be linked to the widespread expression of the transcription factor Sox-10 [56], which is involved in differentiation of myelin-forming oligodendrocytes in the CNS. Additionally, selected neurotrophic-protein receptor expression studies showed that some ENCs express TrkA, TrkB, TrkC, the low-affinity neurotrophin receptor p75NTR, and GDNF receptors (GFRα-1, GFRα-2, and GFRα-3). Thus GDNF, BDNF, NT-3, and other ligands for these receptors may have a role in ENC signaling. Finally, we observed that the neurons derived from ENCs express several types of neurotransmitters found in both the CNS and ENS.

Overall, the data suggest that the postnatal intestine of mice, including the adult murine intestine, can give rise to multipotent nestin, ngn-2, Sox-10, and Mash-1–positive stem cells. Further identification and purification of intestinal ENCs should provide a powerful means to study the regulation of their differentiation and should give insight into the mechanisms involved in development and remodeling of the ENS. The pattern of neuronal and glial differentiation, together with the fact that the ENCs are highly prolific, suggest that they should be further studied and compared with other stem cells derived from different sources, in hopes for potential therapeutic use not only in enteric diseases but possibly in CNS and PNS disease.

The origin of ENCs should be further studied. If, as these studies suggest, there is more than one source of ENSCs (neural crest and hematogenous), novel therapies for intestinal neural disorders may be developed by various stem cell recruitment and differentiation techniques.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors thank Xochitl Alvarado for her valuable assistance with confocal microscopy analysis, as well as Angelica Rivas Hernández and Idanelli Barrios Jacobo for their dedicated laboratory technical assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References