SEARCH

SEARCH BY CITATION

Keywords:

  • neural crest;
  • enteric nervous system;
  • adult stem cells;
  • primary cultures;
  • neurospheres

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

An increasing body of evidence has accumulated in recent years supporting the existence of neural stem cells in the adult gut. There are at least three groups that have obtained them using different methodologies and have described them in vitro. There is a growing amount of knowledge on their biology, but many questions are yet unanswered. Among these questions is whether these cells are part of a permanent undifferentiated pool or are recruited in a regular basis; in addition, the factors and genes involved in their survival, proliferation, migration, and differentiation are largely unknown. Finally, with between 10 and 20% of adults suffering from diseases involving the enteric nervous system, most notably irritable bowel syndrome and gastroesophageal reflux, what is the possible role of enteric nervous stem cells in health and disease? Developmental Dynamics 236:20–32, 2007. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

In the central nervous system (CNS) of adult mammals, two neurogenic zones are known (Lledo et al.,2006): the subgranular zone (SGZ) of the hippocampal dentate gyrus, and the subventricular zone (SVZ). Neurogenesis in these zones occurs on a regular basis, for instance, in rodents, the SVZ produces 30,000 neuroblasts per day (Alvarez-Buylla et al.,2001).

That the enteric nervous system (ENS) acquires its complete set of neuron types at different developmental stages and that some of these types are scarcely present at birth (Vanucchi and Faussone-Pellegrini,1996) suffices to suggest that, after birth, there must be a pool of neural crest (NC) -derived neural stem cells within the intestine. Judging for the scarcity of the non-neural nonglial cells (between 1.7 and 4.7%, depending on the region of the intestine) found in intestinal ganglia of mice at the day of birth (Young et al.,2003), and knowing that these proportions only reduce with maturity, it would seem a discouraging task to search for stem cells in such a tissue. Nevertheless, solid evidence supporting the existence of postnatal enteric neuron stem cells (P-nENSC) has been published and is concisely presented and discussed in the following sections.

ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

The NC is heterogeneous because populations of cells with different potential coexist. There are stem cells with a great differentiation potential, because they can give rise to at least six cell types: sensory neuroblasts, autonomic neuroblasts, melanocytes, smooth muscle cells, chondrocytes, and connective tissue cells (Sieber-Blum,2000). There are also cells with a more restricted potential or already committed cells (such as melanogenic, smooth myogenic, sensory neurons). Differences are also patent within the migratory populations, as well as according to the axis level that originates them. Cells from the NC give rise to a peculiar nervous system intimately associated to the gastrointestinal tract (GIT). In this section we will focus on the origin of the ENS.

The ENS is originated from vagal, rostrotruncal, and sacral levels of the NC (Le Douarin and Teillet,1973; Gariepy,2001). Migration of NC cells to the foregut and hindgut starts at embryonic day (E) 1.5, E8.5, and E21 in bird, mouse, and human, respectively (Farlie et al.,2004). Most of the neurons colonizing the GIT are derived from the vagal region of the NC (adjacent to the first seven somites). Cells issued from this region colonize the entire intestine, whereas cells coming from the rostrotruncal region of the NC colonize only esophagus and stomach. The colonization ability of NC-derived cells is the same whether they are located at the leading edge of the migrating wave or behind it (Sidebotham et al.,2002). In birds, mice, and human, the postumbilical intestine is originated from the sacral division of the NC (Pomeranz et al.,1991; Serbedzija et al.,1991; Burns et al.,2000). In zebrafish, though, the contribution of the sacral NC is not known (Elworthy et al.,2005).

Some cells reach the gut in a multi- or pluripotent state, expressing Sox10 (which is encoded by a gene that lies upstream of Ret); and then, a complex program of genes, such as p75, Nestin, Phox2b, Ret, Mash1, Hu, TuJ1, PGP9.5, α4 integrin, and S-100, drive differentiation into neurons or glia in the gut wall (Lo et al.,1997; Gershon,1998; Pachnis et al.,1998; Young et al.,2000; Elworthy et al.,2005). Some of those genes are shown in Figure 1 at the embryonic age at which they have been detected.

thumbnail image

Figure 1. Markers of undifferentiated prenatal mouse neural crest (NC) -derived cells, shown on an embryo time scale. Vertical arrow indicates entry of NC-derived cells into foregut. Information for the elaboration of this figure has been obtained from Lo and Anderson,1995; Anderson et al.,2006; Srinivasan et al.,2005; Chalazonitis,2004; Sidebotham et al.,2001; Young et al.,1998, 1999, 2003; Bixby et al.,2002.

Download figure to PowerPoint

Glia-Derived Neurotrophic Factor

Survival of NC-derived cells depends upon glia-derived neurotrophic factor (GDNF), a member of the transforming growth factor-beta family (TGFβ), cloned by Lin et al. (1993). A defective protein or its total absence are reflected in developmental renal and enteric innervation defects (Ivanchuck et al.,1996; Sánchez et al.,1996; Pichel et al.,1996; Moore et al.,1996). In cultures of quail's enteric neural precursors (E4.7 and E7), GDNF induced proliferation rate and induced an increase in the number of neural precursors (Hearn,1998). In mice, GDNF stimulated proliferation of nestin-positive cells isolated from rostral foregut at E12. Furthermore, in these cells, as well as in cells from E14–E16, both the number of peripherin-expressing neurons and general TrkC expression were induced. On the other hand, GDNF inhibited S-100 expression on E14–E16 cells, thus favoring neural but not glial development (Chalazonitis et al.,1998; Gianino et al.,2003).

The chemoattractant properties of GDNF on enteric NC-derived cells have been described by Young et al. (2001) in E11.5–E12.5 esophagus, midgut, and hindgut mice explants. Although the effects of GDNF extend onto postnatal enteric neurons and glia, they are limited to neuronal survival and neurite elongation (Schafer and Mestres,1999).

The effects that GDNF exerts can take different pathways. The first one, through the RET receptor, as demonstrated by phosphorylation of tyrosines of the intracellular catalytic domain after binding (Vega et al.,1996). The second possible route, activation of the Ras or mitogen-activated protein kinase pathways after dimerization or autophosphorylation, these pathways induce survival or proliferation. The third route induces either cell motility (Farlie et al.,2004) or proliferation of the enteric neuroblasts (Focke et al.,2001; Srinivasan et al.,2005) by activating the PI3K/Akt/Forkhead pathway.

It has also been shown that GFRα1 is a component of the GDNF's receptor complex. It is as necessary as GDNF and RET for the survival of NC-derived cells and can be expressed by the same cells that express RET (Cacalano et al.,1998; Pachnis et al.,1998). The restricted expression pattern of GFRα1 at later times of embryogenesis on NC-derived cells determines that the effect of GDNF on neural precursors shift from mitogenic to trophic (Worley et al.,2000).

ET-3

Endothelin-3 (ET-3) plays not just an important role during ganglia development throughout the intestine, but also contributes in a specific way to gangliogenesis in the distal colon, as this is the only aganglionic region in ET-3 or endothelin-B (ET-B) knockouts (Kruger et al.,2003). Naturally occurring examples of mutations in ET-3 result in the lethal spotting (ls/ls) phenotype, characterized by pigmentation defects in the skin, and enteric aganglionosis. Mutations in the gene encoding for ET-B produce the piebald lethal (s1/s1) phenotype. These manifestations are attributed to defects in migration of melanoblasts and neural precursors from the NC toward the skin or gut, respectively (Baynash et al.,1994; Kruger et al.,2003). Expression of the EDNRB transgene, under the control of dopamine β hydroxylase, compensate for deficiencies of the endogenous receptor ET-B in rat, thus preventing developmental defects (Gariepy et al.,1998). Inhibition of early neuron differentiation seems to be another function of ET-3 and ET-B. This inhibition is suggested to occur by means of a direct action of ET-3 over smooth muscle cells, which would progressively decrease the concentration of laminin-1 in the surrounding environment. The alfa-1 subunit of laminin (Laminin-1) binds to its specific receptor LBP110, this binding stimulates neuronal development in gut. In the absence of ET-3 and ET-B, an increase in the concentration of laminin-1 is favored, stimulating the rate of neuronal differentiation, thus leading to an exit from the neuronal cell cycle. Migration stops as soon as the cell leaves the cell cycle, leading to ectopic ganglia, and aganglionosis in the most distal parts of the colon (Gershon,1999). It has been proposed that EDNRB enhances the proliferating response of progenitors to RET signaling but antagonizes its effect on migration (Barlow et al.,2003). In general, nerve cell adhesion molecule (NCAM) molecules are found in gut containing ganglionic cells (both in normal individuals and in those suffering from Hirschsprung's disease [HD]), whereas these NCAM-expressing neurons are missing in aganglionic segments (Furness and Costa,1980).

Other Factors

Intervention of other molecules is also important at different developmental stages of enteric neural crest-derived cells. For instance, neurotrophin 3 (NT-3) influences precursor cells, both directly or in combination with GDNF and ciliary neurotrophic factor (CNTF), affecting differentiation and maintenance of late-developing enteric neurons. Signaling is regulated by bone morphogenetic protein-4 (BMP-4) by changing the expression of TrkC, the transducing receptor for NT-3. When this signaling is disrupted, a loss of the TrkC-expressing neurons is observed (Chalazonitis,2004). Sonic hedgehog (Shh) is expressed in the gut endoderm and shows opposite effects to GDNF, regarding proliferation and differentiation, on neural crest-derived progenitors. It apparently contributes to the formation of the myenteric plexus or “secondary migration” from the intestinal lumen to its external layer, as opposed to the rostrocaudal migration or “primary migration” conveyed by GDNF (Fu et al.,2004).

Aganglionosis of the distal colon results from two probable mechanisms: (1) there is a small number of differentiated cells deriving from inadequate GDNF-Ret signaling. As described above, the complex GDNF/Ret is involved in proliferation and migration of NC-derived precursors and eventually colonization of the whole GIT below rostral foregut. Therefore, partial or complete deficiency of GDNF or Ret produces an inadequate colonization of gut by enteric NC-derived precursors. (2) Migration of NC-derived cells is stopped prematurely because progenitors differentiate too early due to a missing ET-3/ET-B system. This system is involved in the inhibition of early neuronal differentiation. Under normal conditions, such inhibition contributes to the maintenance of the enteric NC-derived precursors in an undifferentiated state until they reach their final destination. A faulty expression of the ET-3/ET-B system results in an arrested migration of neural precursors, which differentiate before they reach their final destination, thus producing aganglionosis of the distal parts of the gut (Gershon,1998). Migration of NC-derived cells depends not only on extrinsic regulating factors like GDNF or endothelins, cellular interactions within the front of the migrating wave as well as pressure exerted from the cells behind them are also important (Tsai and Gariepy,2005).

Enteric neural precursors from a different source than NC have been described (Sohal et al.,2002). These cells are called ventrally emigrating neural tube, and derive from the ventral hindbrain. They colonize the esophagus and stomach, can be localized in the submucosal and myenteric plexus, and give rise to neurons, glia, and interstitial cells of Cajal (ICC).

Evidence suggesting that multipotential cells persist during postnatal stages has been reported and is discussed in the following sections. The first line of evidence consists in the successful establishment of in vitro cultures. Below, the most common culture methods are presented.

CULTURE OF P-nENSC

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

Culture conditions of P-nENSC described so far are summarized in Table 1.

Table 1. Culture Conditions for the Proliferation of P-nENSCa
SpeciesAgeTissueDispersion methodCultureReference
MediumModality
  • a

    EDTA, ethylenediaminetetraacetic acid; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; GDNF, glia-derived neurotrophic factor; BDNF, brain-derived neurotrophic factor; bFGF, basic fibroblast growth factor; IGF, insulin-like growth factor; NLB, neurosphere-like bodies; FBS, fetal bovine serum; EGF, endothelial growth factor; NCSC, neural crest stem cells; CNTF, ciliary neurotrophic factor.

Rat1, 5, 10 and 15 days after birthSmall intestineCollagenase, Trypsin-EDTADMEM 10% FCS x 30 min, Serum free culture medium + N2 supplement, + 3% FCS alone or + GDNF, BDNF, bFGF x 48 hrDispersed cellsLintz et al.,1999
RatAdultNonspecifiedTrypsin/EDTA and collagenase; trituration and filtration through nylon screenStandard medium for 6 days: DMEM-low plus 15% chick embryo extract (CEE), bFGF, N2 supplement, IGF-1Dispersed cellsKruger et al.,2002
Mouse3-5 days after birthSmall bowelCollagenase, dispase, DNAse I, trituration, elimination of large fragments by sorbitol selection, filtration through nylon meshDMEM-high glucose 2.5% FBS and EGFDispersed cellsSuárez-Rodríguez and Belkind-Gerson,2004
Mouse2 to 14 days after birthSmall bowelCollagenaseNCSC 15% CEE, bFGF and after NLBs, EGFNLBsBondurand et al.,2003
Human5 days to 5 years after birthSmall and large bowelTrypsin and collagenaseDMEM/F12 N1 supplemented plus bFGF, GDNF, Neurturin and CNTFNLBsRauch et al.,2006a

Primary Cultures

Myenteric plexus was isolated from rats at several postnatal stages. Cells with neuronal characteristics after 48 hr of culture were positive to protein gene product 9.5 (PGP9.5) immunostain and showed neurite outgrowth (Lintz et al.,1999). Cells obtained from lactating or adult mice attached to the culture plate and expressed nestin, vimentin; proneural transcription factors; neural, glial, and smooth muscle markers; Trk, p75, and GFRα receptors; several neurotransmitters as encountered in central and enteric nervous system such as calcitonin gene-related peptide (CGRP), neuropeptide Y (NPY), peptide YY, substance P, vasoactive intestinal peptide (VIP), and galanin; and hematopoietic markers such as c-KIT, CD20, CD34, and CD45RO, which suggested that these cells can differentiate toward hematopoietic phenotype or are originated from hematopoietic precursors (Suárez-Rodríguez and Belkind-Gerson,2004). Neural crest-derived stem cells have been described in the postnatal gut. Approximately 60–70% of p75+ and alpha-4-integrin cells isolated by fluorescence-activated cell sorting from primary cell cultures of postnatal rat gut formed neuron, glia, and myofibroblast colonies and showed self-renewing capacity. The derived neurons expressed neurotransmitters such as VIP, NPY, and nitric oxide (NO; Kruger et al.,2002).

Neurospheres

P-nENSC have also been isolated from neurospheres obtained from fetal, and postnatal bowel of surgical or postmortem specimens in humans, then cultured in medium supplemented with basic fibroblast growth factor, GDNF, Neurturin, and CNTF. Precursors obtained in this way were transplanted in vitro into human bowel segments (Rauch et al.,2006a). Enteric neural progenitors have also been isolated from neurospheres originated from fetal and postnatal mice gut cultures. These cells become glia and different neuronal subtypes as indicated by the expression of several neural markers including tyrosine hydroxylase (TH), VIP, NPY, and CGRP (Bondurand et al.,2003).

Cultures of P-nENSC carried so far have yielded comparable cellular diversities. The main challenge consists on being able to maintain cultures of proliferating undifferentiated cells, once this is achieved, further characterization can take place. Coupling optimized culture conditions to an effective purification protocol, can lead to significant progress in the field of ENS development.

It is interesting that P-nENSC can be obtained and cultured using such diverse techniques as has been described above. Nonetheless, to better characterize them, it is essential to purify the desired cell population(s) as best as possible. To date, this has probably been better achieved using immunosorting. Because there are no surface proteins that are known to be 100% specific for P-nENSC, future attempts will probably be more successful in purifying very select populations, as a better characterization will undoubtedly be obtained using cutting edge technology such as proteomics and microarray analysis.

MARKERS OF P-nENSC

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

Stem cells are identified by their behavior, as there is no “perfect marker.” However both, in vitro and in vivo intestinal neural stem cells, express several key proteins that are valuable for identification, and their pattern of expression has been correlated with functional aspects (migration, proliferation, differentiation, etc.). This section will review expression patterns and function of markers identified in undifferentiated P-nENSC. Although it is not the scope of this review, the evolutionary marker–panorama of prenatal ENSC will be briefly mentioned.

The use of surface markers indicative of the presence of stem cells is a common practice in neurobiology (Uchida et al.,2000). Although, as stated by Lo and Anderson, back in 1995, there were no markers that uniquely identified neural crest stem cells in vitro; a great deal of information is now available suggesting which markers most likely define P-nENSC. As NC-derived cells migrate, guided toward the embryonic gut with the aid of versican (Dutt et al.,2006) and other proteins, different markers are identified. A summary of such markers, characteristic of undifferentiated NC-derived cells, identified in prenatal gut, is shown in Figure 1. Detailed information is available regarding changes in the expression of these markers as embryonic development progresses. Such changes are recorded not only with regard to the age of the embryo (Young et al.,2003) but also according to the position of the cells with respect to the front of the migrating wave (Young et al.,1998, 1999), which correlates with the developmental condition of the cells. An analysis of the most undifferentiated NC-derived enteric cells from E10.5, E11.5, E12.5, and E13.5 mice, showed predominant positive staining with p75 (Young et al.,1999), strong nuclear Phox2b staining, and strong Ret staining (Fig. 1). Cells in positions predicted to be less undifferentiated showed positive staining to Phox2a, and little or not p75 immunoreactivity, suggesting that they have already given a step toward neural maturation. According to Young et al. (1999), undifferentiated prenatal ENSC were shown to be p75+/Phox2b+/Ret+. Neuron differentiation markers (Phox2a, nitric oxide synthase [NOS], CGRP) were identified in more mature cells, as well as down-regulated Ret expression, associated with glial differentiation.

Persistence of some of these markers is expected if a pool of stem cells is conserved after birth. Finding those markers should promote research on the biology of these potentially important cells; therefore, this section describes the markers that have been used to identify undifferentiated NC cells in cultures of postnatal intestinal tissue. Figure 2 shows characteristic gene products reported to be expressed in undifferentiated, committed and fully differentiated P-nENSC, respectively.

thumbnail image

Figure 2. Simplified chart showing gene products and direction of the differentiation flow seen in postnatal enteric neural stem cells (ENSC) at the undifferentiated, committed, and differentiated stages. The question mark interrupting the differentiation line from committed cells to myofibroblast reflects lack of details in this step.

Download figure to PowerPoint

SOX10

The earlier neural crest marker to be found in the embryonic gut is SOX10 (see Fig. 1 for embryo time scale), and is encoded by SRY (sex determining region-Y) box 10. This gene belongs to the high-mobility group gene family that plays a critical role in the formation of tissues and organs during embryonic development, and is widely distributed in the animal kingdom. The protein product of this gene has DNA binding domains; it works as a transcriptional regulator and is active in NC-derived cells (Mollaaghababa and Pavan,2003). SOX10 is expressed in undifferentiated enteric neural stem cells; it inhibits neuronal and glial differentiation of multipotent enteric neural stem cells (Bondurand et al.,2006) and seems to be crucial for the maintenance of the stem cell state (Kim et al.,2003).

Selecting for proliferating cells, Bondurand et al. (2003), identified SOX10-positive P-nENSC that were clonogenic, multipotential, and organized themselves in neurospheres. These cells were also GDNF-responsive, indicating not only that their source is the neural crest but also that the elemental signaling cues of the developing enteric neural system are conserved. Expression of SOX10 is down-regulated as soon as neural lineage definition commences (Young et al.,2003); therefore, it seems like a good marker for undifferentiated cells.

p75

This low affinity neurotrophin receptor (p75), a transmembrane glycoprotein, is considered a standard marker for neural crest cells (Lo and Anderson,1995; Bixby et al.,2002). Expressed in the surface of the membrane, it binds, although with different affinities, the neurotrophin nerve growth factor, brain-derived neurotrophic factor (BDNF), NT-3, and neurotrophins 4/5 (NT-4/5; see Friedman and Greene,1999, for a review on neurotrophin signaling). In postnatal intestine, this marker localizes in the submucosal and myenteric plexus (Bixby et al.,2002).

Using a cell sorter and the fraction of the population with the strongest p75 expression (ca. 2% of total cells), it was possible to identify multipotential P-nENSC (Kruger et al.,2002). Cells p75 sorted and cultured were isolated from postnatal animals ranging in postnatal age from 5 to 110 days. This procedure allowed for an enrichment of P-nENSC and selected for multilineage forming cells. P75-positive cells only represented 1–2% of the cultures, thus, the population of multipotent cells is small. Care has to be taken regarding the proportion of the cellular population that stains positive for p75, because the expression of this receptor is down-regulated with time in culture.

Ret

Among the most important genes involved in the development of the enteric nervous system is Ret, which encodes for the RET tyrosine kinase receptor, which binds to ligands of the GDNF family. Cultures derived from intestines of homozygous null mutant (Ret−/−) mice did not produce multilineage cells (Bondurand et al.,2003), indicating that Ret is necessary for the development of ENSC. Upon birth, Ret does not identify undifferentiated cells anymore, because in cultures of postnatal mice intestine it is coexpressed along with the pan-neuronal marker PGP9.5, which is expressed in fully differentiated neurons (Young et al.,2003).

Nestin

This designation refers to the intermediate filament type VI expressed in neuroepithelial stem cells. Upon maturation of the stem cells, expression of this marker is lost (Ehrmann et al.,2005). Even though this marker has been used for identification of postnatal enteric neural stem cells in mice (Suárez-Rodríguez and Belkind-Gerson,2004) and human (Rauch et al.,2006b), it cannot be used as an exclusive marker because other cell types, such as endothelial cells, CD34+ fibroblast-like cells, and pericytes, are also positive for nestin (Ehrmann et al.,2005; Rauch et al.,2006b). Pericytes are perivascular cells that envelop the surface of the vascular tube; they seem to be undifferentiated mesenchymal cells recruited by the endothelium (Hirschi and D'Amore,1996). Such abundant nestin expression in the GIT certainly obscures the significance of its presence, as already pointed out by Vanderwinden et al. (2002).

We began this section stating that there is no perfect marker for stem cells; their presence can be confirmed by the expression of gene products known to pertain to early developmental stages. Assisted with proteomic tools, a screening of P-nENSC expressing any of the above-described proteins can identify characteristic membrane-associated proteins that could be used in combination with current markers, thus facilitating the identification of multipotential cells.

PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

There are differences in the proliferative ability of P-nENSC compared with embryonic ENSC. The first observation indicating that P-nENSC were less proliferative was provided by Kruger et al. (2002). On the basis of clonal analysis, they reported that colonies resulting from cultures of P-nENSC were smaller and contained 44% fewer cells than the cells contained in colonies derived from prenatal ENSC. Bondurand et al. (2003) noticed that neurosphere-like bodies derived from intestines of mice (2 to 14 days old) were generally smaller than the structures formed with embryonic tissue and contained almost half of the proliferating cells contained in neurospheres from embryonic intestines. Moreover, plating efficiencies decreased with the age of the tissue of origin. Indeed, Kruger et al. (2002) showed that almost half of the p75-sorted cells isolated from intestinal tissue of postnatal mice aged from 5 to 22 days old, attached and formed colonies. Whereas, when tissue from older animals was used, scarce cells attached to the culture plates (ca. 9%). A similar inverse correlation between plating efficiency and age of the donor animal was observed by Bondurand et al. (2003) and by Suárez-Rodríguez and Belkind-Gerson (2004).

In CNS, mitotic activity of adult neural stem cells is reduced with respect to their embryonic counterpart (Lledo et al.,2006), and it holds true also for P-nENSC. A cell cycle analysis of p75+ cells performed by Kruger et al. (2002) showed that, in cultures derived from postnatal day (P) 15 mice, only few of the cells were mitotically active compared with cultures from E14.5 mice. Whereas in a 20-hr period, more than 80% of p75+ cells from E14.5 cultures had undergone at least one cell division, in P15 cultures, only 13% of p75+ cells had (Kruger et al.,2002). Bondurand et al. (2003) also recorded reduced mitotic activity from P-nENSC using an antibody and a G2/M-specific histone. Such labeling identified 50% fewer mitotic cells in P-nENSC neurospheres.

There is a common agreement that stemness involves proliferation, self-renewal of multipotential (nondifferentiated) cells, and differentiation toward specialized phenotypes (NIH,2001). Although in a recent review, Zipori (2005) states that self-renewal is not necessarily a characteristic of stem cells. He cites as an example the not always rewarding efforts put toward in vitro expansion of hematopoietic stem cells. From the therapeutic point of view, plasticity is the most wanted feature encoded by a cell. And the more potency encoded by the cells, the more interesting they are. Pluripotency in adult NC-derived cells has already been reported (Sieber-Blum et al.,2004). Indeed, by isolating and culturing cells from the bulge of adult mice hair follicles, different cell lineages (neuron, muscle, glia, chondrocyte) were identified by immunocytochemistry. A summary cartoon showing different gene products, characteristic of each step (from undifferentiated to fully developed cell) and experimentally documented, is represented in Figure 2. At least two reports evidence multipotentiality of P-nENSC (Kruger et al.,2002; Suárez-Rodríguez and Belkind-Gerson,2004). In Kruger et al. (2002), P-nENSC were cultured at clonal density after being selected with an antibody against p75. After 14 days in culture, resulting colonies showed positive reaction to antibodies against peripherin (neuron), glial fibrillary acidic protein (GFAP, glia), and smooth muscle actin (SMA, myofibroblast). Cultures developed by Suárez-Rodríguez and Belkind-Gerson (2004) were established as monolayer adherent cultures from dissociated postnatal murine gut, and although care was taken to monitor the cultures for already differentiated cells from the beginning, the presence of stem cells from epithelia, muscle, or blood cannot be completely ruled out.

Bipotential P-nENSC were described by Bondurand et al. (2003) after 10 days as neurosphere cultures. Such cultures of progenitor GFP-labeled cells gave rise to colonies containing both neurons and glia.

Multipotency of P-nENSC is also compromised with age; it is reduced nearly 3.5-fold in cells from P65–P110 mice compared with P5–P22 (Kruger et al.,2002). In the ENS, the order of differentiation as occurs in the CNS is conserved, that is, neurons develop before glial cells do (Young et al.,2003).

NEURONAL CELLS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

Neural crest-derived cells that have arrived to the developing gut are expected to become neurons; commitment to such differentiation is characterized by the expression of MASH1, Phox2b, and ngn-2, as shown in Figure 2.

MASH1

Mammalian achaete–scute homologue 1 gene, encodes for a basic helix–loop–helix (bHLH) transcription factor that plays a role in the early development of the nervous system (Lo et al.,1991). In neuron-committed progenitors, it activates a series of neuron-specific genes leading to neuronal differentiation. In the CNS, it plays a central role in directing (Parras et al.,2002) the specific type of neurons (McNay et al.,2006). Its role in the development of the esophageal portion of the ENS has been reported (Sang et al.,1999).

Phox2b

Expression of Phox-2b at birth cannot be used as criteria for identification of undifferentiated NC-derived cells, because its expression seems to be down-regulated postnatally (Young et al.,2003). Phox-2b is described as a homeodomain transcription factor that regulates differentiation toward neurons, although it is necessary specifically for motoneurons. It has a negative effect on inhibitors of neurogenesis (Dubreuil et al.,2002). The protein Phox-2b plays an important role in the differentiation of noradrenergic neurons in the peripheral nervous system. As a matter of fact, cultured mouse or chicken neural crest cells overexpressing Phox-2b express TH and protein gene product 9.5 (PGP9.5; Young et al.,2003). Such labeling pattern clearly indicates neuronal nature. Therefore, the use of anti-Phox-2b antibodies should not be used when screening for P-nENSC.

Ngn-2

This gene encodes for neurogenin-2, a bHLH transcription factor required for the generation of neural progenitors in both the peripheral nervous system (PNS) and CNS, and its expression defines one of the initial steps in neurogenesis. It is expressed in different but complementary cells than Mash1 (Parras et al.,2002) and has a permissive role rather than a decisive role in neuron-type specification. The expression of Ngn-2 is said to be complementary to that of Mash-1, because the induction of the neuronal phenotype depends on Mash-1, regardless of the presence of Ngn-2. Whereas the decision of the resulting type of neuron is only taken by cells that already express Mash-1. Ngn-2 takes part in the definition of the type of neuron; the final phenotype depends on the instructive factors that are coexpressed in the population of precursor cells (Parras et al.,2002).

Full neuronal differentiation of P-nENSC in adherent cultures is seen as early as 4 days after plating, as indicated by the expression of neuron-specific markers (Kruger et al.,2002; Suárez-Rodríguez and Belkind-Gerson,2004). A summary of the different specific neuronal markers, which have been used for characterizing differentiated cells issued from P-nENSC cultures, is shown in Table 2.

Table 2. Neuronal Markers Identified in Cultures of P-nENSC
NameRecognizesTested onaReference
  • a

    Immunocytochemistry performed on (A) colonies derived from neurosphere-like bodies, 15 days post plating; (B) adherent cultures 4-7 days postplating; (C) p75+-sorted cells seeded at clonal density and stained 14 days postplating; (D) postnatal rat intestines cultured in noradrenergic stimulating medium. P-nENSC, postnatal enteric neuron stem cells; PGP9.5, protein gene product 9.5; CNS, central nervous system; PNS, peripheral nervous system; GIT, gastrointestinal tract; NOS, nitric oxide synthase.

Tuj1Neuron-specific beta tubulin III(A) (B)Bondurand et al.,2003; Suárez-Rodríguez and Belkind-Gerson,2004
PGP9.5Neuron-specific 27-kDa intracellular C-terminal ubiquitinilated hydrolase(A)Bondurand et al.,2003
TauMicrotubule-associated protein in neurons(B)Suárez-Rodríguez and Belkind-Gerson,2004
160/200-kDa NF160- and 200-kDa proteins of human neurofilament(B)Suárez-Rodríguez and Belkind-Gerson,2004
Glutamate transporter EACC1EAAC1 glutamate transporter, sodium-dependent(B)Suárez-Rodríguez and Belkind-Gerson,2004
SynaptophysinA 38-kDa glycoprotein of presynaptic vesicles of almost all neurons(B)Suárez-Rodríguez and Belkind-Gerson,2004
VIPVasoactive intestinal peptide, structurally related to secretin, wide distribution in CNS and PNS(B)Suárez-Rodríguez and Belkind-Gerson,2004; Bixby et al.,2002
NPYNeuropeptide Y, the most abundant neuropeptide in the brain(B),(C)Suárez-Rodríguez and Belkind-Gerson,2004; Bixby et al.,2002; Kruger et al.,2002
Peptide YYAgonist of the neuropeptide Y receptor, secreted by GIT(B),(C)Suárez-Rodríguez and Belkind-Gerson,2004; Kruger et al.,2002
Peptide PNeuromodulator and neurotransmitter form peripheral receptors to CNS(B)Suárez-Rodríguez and Belkind-Gerson,2004
GalaninInhibits secretion of transmitters or hormones in nervous and endocrine systems.(B)Suárez-Rodríguez and Belkind-Gerson,2004
CGRPCalcitonin gene-related peptide, present in central and peripheral nerves(B)Suárez-Rodríguez and Belkind-Gerson,2004
nNOSSynthesizes NO, which is a gaseous free-radical that carries messages between cells(B),(C)Bixby et al.,2002; Kruger et al.,2002
Catecholaminergic neurons
THTyrosine hydroxylase, produces a precursor for adrenaline and dopamine(A)Bondurand et al.,2003
DβHDopamine-β-hydroxylase(D)Kruger et al.,2002
Cholinergic neurons
AChEEnzyme that hydrolyzes acetylcholine in synaptic junctions(B)Suárez-Rodríguez and Belkind-Gerson,2004

Among those markers, neural ultrastructure is revealed by Tuj1. This antibody recognizes with high affinity the neuron-specific beta tubulin III. In colonies from P-nENSC, these postmitotic neurons were identified after 4 days (Suárez-Rodríguez and Belkind-Gerson,2004) or 10 days (Bondurand et al.,2003) in culture. Other markers for neurofilaments (such as Tau, or 160/200 kDa) were first evident in P-nENSC after 7 days in culture (Suárez-Rodríguez and Belkind-Gerson,2004).

Neuron-specific proteins (glutamate transporter, synaptic proteins, neuromodulators, and neurotransmitters) indicative of potentially functional neurons were also revealed in vitro with antibodies (see Table 2). For instance, the PGP9.5, a 27-kDa, neuron-specific, intracellular C-terminal ubiquitinilated hydrolase, has been shown to be present in premigratory NC-derived cells (costaining with p75; Sidebotham et al.2001), as well as in enteric neurons. In human postnatal tissue, Rauch et al. (2006b) showed coexpression of this pan-neuronal marker with nestin. Whereas Bondurand et al. (2003) found it, in cells that expressed other neuron commitment markers such as Mash1 or RET; or along with markers of fully differentiated neurons (Tuj1) or glia (GFAP).

The variety of neurons obtained in clonogenic assays of P-nENSC cultures after 15 days (Bondurand et al.,2003) and in monolayer cultures after 4 days (Suárez-Rodríguez and Belkind-Gerson,2004) corresponds to the variety found in the mammalian enteric system, see Table 2 for a list of the neuron-specific markers that have been identified in cultures of P-nENSC.

Such specialization is suggested by the expression of enzymes such as TH, which is the enzyme catalyzing the conversion of L-tyrosine to dihydroxyphenylalanine, a precursor for dopamine and adrenaline. It is a characteristic marker for catecholaminergic (adrenergic and dopaminergic) neurons. In murine embryos, expression of TH is transitory by cells identified as transiently catecholaminergic (Baetge et al.,1990; Young et al.,1999). Neurons derived from P-nENSC cultured as neurospheres, expressed TH, although it was not specified whether it was accompanied by the expression of markers of fully differentiated neurons (Bondurand et al.,2003), such was the case observed for prenatal cultures. In adult mouse stomach, ileum, duodenum, and colon, the expression of TH was confirmed by reverse transcriptase-polymerase chain reaction (RT-PCR; Li et al.,2004), suggesting the existence of intrinsic enteric dopaminergic neurons in adult gut. This finding was further confirmed by the identification of transcripts encoding dopamine and dopamine transporter.

According to Kruger et al. (2002), serotonergic neurons could not be originated by P-nENSC derived from P15 mice, probably due to a reduced response to the neurogenic signals of BMP.

In vivo assays have also indicated that postnatal enteric neural stem cells have the potential to become neurons or glia (Bixby et al.,2002). This feature will be discussed further in the animal model section below.

GLIAL CELLS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

The second phenotype normally expected from NC-derived cells is glia. Postnatal ENSC are significantly more inclined to gliogenesis than to neurogenesis (Kruger et al.,2002) due to an increased responsiveness to gliogenic factors such as Notch and neuregulin.

During the development of the bowel in mouse, glial precursors can be identified by their expression of brain fatty acid binding protein (B-FABP). Among the functions of fatty acids are enzyme and membrane function, gene expression, growth, and differentiation (Veerkamp and Zimmerman,2001). It is, therefore, necessary to have proteins for uptake, transport, and targeting of the fatty acids. B-FABP belongs to a family of four cytosolic fatty acid binding proteins (Veerkamp and Zimmerman,2001), it is not expressed in neural crest cells but it is found in glial cells and their precursors. Young et al. (2003) found this protein expressed only in intestinal cells negative for markers of neural lineage. Expression of this protein occurs even before glia differentiation; this is illustrated by the observation of Young et al. (2003) who recorded B-FABP in E11.5 mice, 3 days before S100b expression (see below). This marker can, therefore, be considered as a commitment marker for glial lineage. Based on the temporal expression of B-FABP, compared with the expression of neuronal markers, Young et al. (2003) suggested that, in the ENS, glial cells do not participate in the migration of neurons. The following are markers of differentiated glia: S100 calcium binding protein, beta chain, also called neurite extension factor, is a calcium-regulated protein produced by glial cells. Expression of this factor has not been studied in differentiated cells that derive from cultures of presumptive P-nENSC; however, its expression in ENS has been documented (Gershon and Rothman,1991; Young et al.,2003).

After 10 days in vitro, some Sox10+ P-nENSC showed positive reaction to GFAP antibody (Bondurand et al.,2003). This marker identifies the intermediate filament found in cells of the astroglial lineage. Along with the GFAP+ cells, neurons in various developmental stages were identified (expressing PGP9.5, Tuj1, Mash1, RET), but no other cell lineage was suggested.

SMOOTH MUSCLE CELLS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

Culturing mice P-nENSC as neurospheres (Bondurand et al.,2003), or as adherent cultures, either at clonal density (Kruger et al.,2002) or not (Suárez-Rodríguez and Belkind-Gerson,2004) revealed a small amount of cells (1–5%) with muscular phenotype, because they express SMA and whose mRNA was detected in RT-PCR assays in increasing concentrations throughout the culture period (Sánchez-Rodríguez and Belkind-Gerson,2004). In the latter case, despite contaminating muscle cells not being completely ruled out, the authors suggested the possibility that a subset of NC-derived cells (thus nestin+) express alfa actin. Such a phenotype has already been described in NC-derived cells as a response to growth factors (Shah et al.,1996). Namely, rat neural crest stem cells cultured in the presence of members of the TGFβ growth factor family (BMP2 or 4), expressed SMA. An interesting possibility is raised, which is that multipotential stem cells also reside in the intestine, in very low numbers, because this cell lineage was only evident before any selection. On the other hand, the possibility of these stem cells being non–NC-derived also exists, as has been suggested for non-neuron nonglial cells identified by Young et al. (2003).

ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

In the past 15–20 years, we have learned a great deal about ENS prenatal development and several genetic factors are now known to play an important role in the pathogenesis of conditions such as HD, where ENS dysfunction may begin at birth. However, in addition to genetic insults, throughout life, the gut is exposed to a great number of pathogens, allergens, toxins, and other potential insults, which may result in ENS damage and, as in the case of postinfectious irritable bowel syndrome, there is increasing evidence to suggest that ENS dysfunction may also be acquired (Parry and Forgacs,2005). Because the ENS function primarily depends on postmitotic neurons, which like neurons in the CNS do not have the ability to replicate, innovative future treatments may be based on better understanding of migration and repair.

In a provocative experiment, Hanani et al. (2003) tested the regenerative capacity of the ENS. They ablated the myenteric plexus of the mouse colon with the detergent benzalkonium chloride (BAC) and followed the ensuing morphological changes for up to 60 days by light and electron microscopy. They found that, upon injury, the colon lost all neural elements. By day 7, however, nerve fibers and new neurons had returned to the area of injury. After 30–60 days, colonic innervation was completely recovered both structurally and functionally, suggesting a much greater regenerative capacity than previously thought. The authors commented: “We obtained electron microscopic evidence that, two weeks and later after treatment with BAC, there were undifferentiated cells in the transition zone as well as differentiating ones. The latter were identified as cells belonging to the neuronal lineage by their ultrastructural features.” The authors postulated that these cells could represent neural stem cells. Axonal degeneration with subsequent reestablishment of innervation has also been documented in another inflammatory model of intestinal injury, the TNBS (trinitrobenzene sulphonic acid) model of colitis in the rat (Poli et al.,2001).

Postnatal stem cells have been found in several tissues, including brain, bone marrow, peripheral blood and skin in rodents and humans (Wagers and Weissman,2004). Enteric neural stem cells (ENSC) have been isolated and grown ex vivo from rodent adult gut (Kruger et al.,2002; Suárez-Rodríguez and Belkind-Gerson,2004). These cells do not appear to be the same cell-type as the intestinal mucosal stem cells that differentiate into intestinal mucosal cell types and are not known to differentiate into neural lineages (Brittan and Wright,2004). Techniques to isolate ENSC have included cell sorting using anti P-75 antibody and alpha-4-integrin antibody (Kruger et al.,2002) and harvesting neurospheres from dissociated tissue cultures (in a manner analogous to the CNS neural stem cell; Schafer et al.,2003) and an in vitro culture model that yields a population of ENSC, which attach to the culture dish and are capable of proliferating (Suárez-Rodríguez and Belkind-Gerson,2004).

The role of ENSC in normal intestinal physiology and in states of disease or injury is unknown, yet some reports suggest that spontaneous neural regeneration is possible in the postnatal intestine, for example, Meyrat and Laurini (2003) described three cases of children with intestinal neuronal dysplasia related to HD, 9–18 months after a diverting colostomy, these children were rebiopsied and the intestinal neuronal dysplasia had resolved. Intestinal neuronal dysplasia is an ill-defined entity characterized by abnormal ganglia in size and/or number and associated with intestinal motility abnormalities. This disease does not usually improve. Meyrat and Laurini's report (2003) suggests that, in some cases, there may be a greater regenerative ability of the ENS than previously suspected.

In experiments in which enteric neural stem cells have been grafted include RET+ cells collected from dissociated embryonic mouse gut that were introduced into embryonic bowel explants from wild-type or an aganglionic HD mouse model, the implanted cells gave rise to neurons and glia (Natarajan et al.,1999). Also using bowel explants, Bondurand et al. (2003) transplanted progenitors capable of colonizing wild-type and aganglionic gut in organ culture and had the potential to generate a differentiated progeny that localized within the intrinsic ganglionic plexus. Similar cells expressing high levels of p75 and alpha-4-integrin isolated from P15 fetal mouse gut were implanted into two hindlimb bud somites of eight chick embryos (stage 17–18), four of which showed engraftment of both neurons and glia, two engrafted only glia, and two did not engraft at all (Kruger et al.,2002). This same group implanted p75- and alpha-4-integrin–positive cells into wild-type or HD mouse model intestine, and these also gave rise to neurons and glia (Kruger et al.,2003). Because ENS cells have several similarities with those of the CNS, the use of CNS-derived stem cells for transplantation into the GIT has also been performed. Micci and Learish (2001) implanted CNS-derived stem cells into the pylorus of adult mice, these differentiated into NOS-expressing neurons. There are yet few studies, however, that show that these cells not only survive but that they are functional. Micci et al. (2005) have transplanted neural stem cells, not of enteric origin, but from the subventricular region (SVZ cells) of embryonic GFP mice into the gastric antrum of a NOS-deficient mouse model, known to have delayed gastric emptying. The implantation of embryonic SVZ cells was associated with an improvement of the gastric emptying (46.67% vs. 35.09% in the vehicle-injected controls P < 0.01). The NOS inhibitor Ng-nitro-L-arginine methyl esther and the neuronal blocker tetrodotoxin blocked electrical field stimulation-induced relaxation of the muscle, suggesting an NO-mediated effect. One week after implantation, cells were seen to be viable, had undergone neuronal differentiation and expressed NOS (Micci et al.,2005). Whether the actual direct release of this neurotransmitter by the implanted cells was the cause of improved gastric emptying is still unknown. Longer follow-up periods are also warranted. The result from this study raises yet more questions about neural stem cells in intestinal disease, among them: What is the best source of the stem cells for the gut? What are the mechanisms that have led to the improved motility? And is the improvement stable over time?

The ENS is involved in and regulates different aspects of intestinal function in addition to motility, including sensory information; electrolyte, mucus, and fluid secretion; and intestinal blood flow. More recent findings suggest that the ENS is also involved in other gastrointestinal functions, including absorption of nutrients and fluid, enteroendocrine and paracrine secretion, and immune modulation (Genton and Kudsk,2003). Unfortunately, enteric neuropathies can be found throughout the GIT from the esophagus to the rectum (De Giorgio et al.,2004). Because of the ENS involvement in all the noted aspects of gastrointestinal function, it is not surprising that abnormalities of the ENS may result in significant gastrointestinal dysfunction, although, clinically, the most evident are those in which intestinal motility or sensory functions are altered. How will ENSC therapy aid in performing the different functions of the ENS? If ENSC are to be used clinically, there are two basic approaches: the first is to increase the availability of the endogenous cells and, thus, stimulate neurogenesis, the second is to implant ENSC from either autologous tissue (obtained from another area of the GIT from the same individual), avoiding the need for immunosuppression or, as several types of stem cells (bone marrow, CNS, embryonic, and others) have been shown to be able to differentiate into neurons, the source of the cells to implant may be nongastrointestinal. In this instance, the work of Yamada et al. (2002) is worth mentioning as they describe the formation of a three-dimensional gut-like organ from mouse embryonic stem cells. Such structure contains a central lumen, neurons, ICC, and smooth muscle cell layers, which are able to elicit a coordinated peristalsis. Despite these and other encouraging observations, the optimum source, purification, pretransplant treatment (or not), and implantation techniques are yet largely unknown and several important questions remain unanswered. Furthermore, if the endogenous pool of postnatal ENSC is to be taken advantage of in clinical situations, there are still little data available on the signaling proteins and genes involved in their proliferation, survival, and differentiation; therefore, much research is still needed. Despite this finding, there is growing evidence that ENSC are present throughout our life and that we conserve some degree of postnatal enteric neural plasticity. Both of these findings taken together suggest a therapeutic opportunity that will undoubtedly be further explored in the coming years.

The similarities and differences between ENSC, CNS, and PNS stem cells are mostly unknown, yet this information is important as attempts to use stem cells from CNS, in the GIT, have been reported (Micci et al.,2005). In the next few years, studies focusing on the comparative characteristics of all types of neural stem cells, both in vitro and in vivo may make it clear whether these are truly interchangeable in a clinical setting.

PERSPECTIVES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES

The field of enteric neural stem cells should benefit from the knowledge generated in CNS and PNS as well as from knowledge generated culturing other NC-derived cells, with proven multipotential capabilities. Isolation and culture of P-nENSC is relatively easy and should allow further characterization of these adult stem cells. Existing culture and identification techniques will undoubtedly continue to improve in the near future, for example, with the use of retroviral vectors for labeling stem cells (Geraerts et al.,2006). Moreover, the use of new markers coupled with flow cytometry could lead to a more precise identification of undifferentiated cells, such as reported for the isolation of CNS neural stem cells with the phenotype CD133+/CD45/CD34 (Uchida et al.,2000). Continued understanding of the biology of P-nENSC will also attract more attention to this newly emerged niche.

Several lines of research derive from unanswered questions implicit in this review such as: Why proliferative abilities of ENSC are so different upon birth? Is there a P-nENSC niche, such as it has been described for CNS (Doetsch,2003)? Is there only one population of P-nENSC? By selecting for proliferating cells, as did Bondurand et al. (2003), only two types of specialized cells were found: neurons and glia. Whereas by sorting a population of cells based on the expression of a surface marker, as did Kruger et al. (2002) selecting p75-positive cells, or just by culturing adherent dissociated intestinal cells (Suárez-Rodríguez and Belkind-Gerson,2004), a third phenotype—myofibroblasts—was identified. It is worth noting that Bondurand et al (2003) found SMA-positive cells (1–5%) within the neurosphere, before selecting for proliferating cells. Could these observations be taken as a suggestion of the existence of two different populations with different potentialities? A thorough analysis of each one of the populations is required to answer this question.

It is suggested that P-nENSC follow the same differentiation routes as fetal ENSC. It is known that they are capable of colonizing aganglionic portions of the intestine. However, their neurogenic potential seems to be compromised upon birth, because differentiation potential of P-nENSC becomes basically gliogenic. Can the original potential be recovered? Evidence detailed in this review show that P-nENSC can be multipotential. However, further research should be carried to find out the best way to ensure that the population of enteric stem cells that gets selected is the most versatile. Such an achievement could be very valuable in clinical therapy. Having pluripotency in mind, can new culture methods and signaling proteins be tested?

Studies should be considered to identify any existing relationship between carbohydrates and enteric neural stem cells. For instance, in CNS, neural stem cells were selected from adult mouse brain using peanut agglutinin (Rietze et al.,2001). Furthermore, proliferation of adult neural stem cells seems to be mediated by galectin-1, a carbohydrate binding protein (Sakaguchi et al.,2006).

The discovery of NC-derived stem cells in postnatal enteric nervous system, along with the demonstration of their differentiation potential, constitutes a model to further characterize adult stem cells. Comparative studies between the enteric and other adult tissue-derived cells may yield answers to long-standing questions on the plasticity, regeneration, and aging process of our bodies, in particular to neural tissue.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ORIGIN AND EARLY SIGNALING OF ENTERIC NEURAL STEM CELLS
  5. CULTURE OF P-nENSC
  6. MARKERS OF P-nENSC
  7. PROLIFERATIVE ABILITIES AND DIFFERENTIATION POTENTIAL OF P-nENSC
  8. NEURONAL CELLS
  9. GLIAL CELLS
  10. SMOOTH MUSCLE CELLS
  11. ANIMAL MODELS AND ENTERIC NEURAL STEM CELLS: EVIDENCE ON THEIR ROLE IN HEALTH AND DISEASE
  12. PERSPECTIVES
  13. REFERENCES